A similar example is also discussed and analyzed in Maestri M. and Cuoci A., "Coupling CFD with detailed microkinetic modeling in heterogeneous catalysis", Submitted to Chemical Engineering Science (2013).
In order to show potentiality and reliability of the code we now extend the analysis to a randomly packed bed of Raschig rings. Details of the bed along with the operating conditions are reported in Figure 1 ant Table 1.
|IInternal diameter||10 mm|
|Total length||50 mm|
|O2 mole fraction||0.056|
|CH4 mole fraction||0.100|
|N2 mole fraction||0.844|
|Residence time||150 ms|
We use the UBI-QEP microkinetic model of (Maestri et al., 2009) to account for the surface reactivity. Since the Raschig rings are randomly arranged in the reactor we cannot exploit any particular symmetry of the geometry. Consequently, we perform a fully 3D simulation of the system. The resulting unstructured mesh consists of ~240,000 cells with ~17,000 catalytic faces. We have considered isothermal conditions at 473 K. Given the very small reactor diameter (1.5 cm), the flow turns out to be laminar (the Reynolds' number calculated on the basis of the shell diameter is equal to 50).
Figure 2 shows the computed velocity field at steady-state. The random structure of the packing results in very distorted and non-uniform streamlines and preferential by-pass zones and recirculations are present. These strong radial variations of the velocity result in a non-uniform residence time. As a consequence surface reactivity and radial concentration profiles are strongly affected by this non-uniformity. This becomes evident from the analysis of the concentration profiles and surface coverage, as better explained below.
Figure 3 shows the molar fractions of gas phase species at steady-state. It is worth noting that the by-pass zone determines a consistent break-through of oxygen and consequently also water formation and methane consumptions are affected. Figure 4 reports the steady-state mass fractions of surface spcies at the catalytic surface.