There is no doubt that oxidation involving oxygen is one of the most important catalytic reactions: [1][2][3][4][5] Oxygen is a very active species and participates in many catalytic reactions, such as CO oxidation, epoxidation and fuel reforming. Taking catalytic CO oxidation as an example, which is very important not only technologically (in car exhaust emission control, CO 2 lasers, and air purification) but also scientifically, the elementary steps of CO oxidation at low and medium pressures are believed to be as follows: 1 (i) O 2 molecules dissociate on a metal surface, resulting in O atom chemisorption; (ii) CO molecules adsorb on the metal surface and then react with the chemisorbed O atoms, forming CO 2 (the Langmuir-Hinshelwood mechanism); and (iii) the product, CO 2 , desorbs from the metal surface. Recently, Zhang, Hu, and Alavi 6 have found theoretically that there are two crucial events in CO oxidation on a Ru surface. First, O atoms must be activated from hollow sites (usually the most stable site) to bridge sites, and second, CO molecules have to approach the activated O atoms from the correct direction and at an appropriate time. The reaction barrier is predominately determined by O-metal bond breaking. 6,7 Experimental evidence 8-10 also suggests that O activation is essential. Then fundamental questions are the following: Why must O atoms be activated from hollow sites to bridge sites for CO oxidation to occur? Is it likely to be true for other oxidation reactions in heterogeneous catalysis? Obviously, these issues are fundamental for understanding catalytic oxidation and may also be important to catalysis in general. In this paper, we show that the necessity of the O activation is also true for CO oxidation on Rh(111). We aim to answer the above questions.We carried out Density Functional Theory calculations for CO oxidation on Rh(111). A generalized gradient approximation 11 was utilized in the calculations. The electronic wave functions were expanded in a plane wave basis set and the ionic cores were described with ultrasoft pseudopotentials. 12 The surface was modeled by a p(2 × 2) unit cell with a slab of three layers of Rh(111). The vacuum region between slabs was 10 Å. A cutoff energy of 300 eV and 2 × 2 × 1 k-point sampling within the surface Brillouin zone were used. Recent work shows 13-16 that this set-up provides sufficient accuracy. In all the calculations, the bottom two layers of Rh atoms were held fixed in their bulk positions, while the top layer of surface atoms was allowed to relax. Transition states (TS's) were searched with a constrained minimization technique. 6,7,16 The TS is identified when (i) the forces on the atoms vanish and (ii) the energy is a maximum along the reaction coordinate, but a minimum with respect to all remaining degrees of freedom.Two TS's for CO oxidation on Rh(111) have been identified, which are shown in Figure 1a,b. The reaction barriers from these two TS's are very similar, being 0.99 and 1.13 eV for TS's (a) and (b), respectively. These two TS's...

No abstract

In order to increase the beam blank cleanliness, the aim of this work is to analyse the flow field in the mould of beam blank continuous casting, to find the factors influencing the strand cleanliness and then to optimise the process parameters. A three-dimensional steady finite element model was developed to simulate and analyse the turbulent flow field in the mould. The volume of fluid model was used to track the free surface evolution at the meniscus. The influences of processing parameters, such as casting speed and nozzle parameters, on the molten steel flow in the strand (such as vortex location, liquid steel impact depth, velocity and fluctuation of the liquid steel at free surface) were analysed and the optimum processing parameters determined based on mass calculation. The results of this research project have been applied in actual production, and it has been shown that they are very useful and efficient for improving the steel cleanliness and controlling the surface cracks on the beam blank web.