Abstract
Performance of chemical looping combustion processes can be improved drastically by enhancing the overall redox characteristics of the system through the use of binary mixtures of oxygen carriers. However, binary mixtures of oxygen carrier particles are often found to differ in both size and density and therefore have the tendency to segregate under certain operating conditions.
In this work, a numerical study was conducted to investigate the mixing and segregation behaviour of binary mixtures of particles with different sizes and densities in a bubbling fluidized bed under conditions pertinent to the fuel reactor of a cold flow model (i.e. a non-reacting replica) of a 10kWth chemical looping combustor. The motion of particles was tracked individually by discrete element model (DEM), whilst the gas flow was modelled by computational fluid dynamics (CFD). Gas–particle interactions were considered by a two-way coupling method. Further, a modified version of Lacey's method was developed to calculate the mixing index, taking into account both the heterogeneity of solids spatial distribution and particle size differences.
Results showed that the modified Lacey's method provided very consistent and stable mixing indexes, proving to be effective for an in-situ quantitative description of mixing. It was also found that as the size ratio of the binary mixture of particles reduced, the mixing index increased indicating better mixing conditions. The agreement between the DEM/CFD model predictions and the experimental data was found to be satisfactory. The optimum conditions for mixing of binary mixtures appeared to be a function of bubble size, bubble rising rate and bubbling dynamics (e.g., splitting and coalescence). Application of the DEM/CFD model for prediction of layer inversion phenomenon in gas–solid fluidized beds was also demonstrated.
A numerical study was conducted to investigate the mixing behaviour of binary mixtures with different sizes and densities. A modified Lacey's method was developed to calculate the mixing index, taking into account both the heterogeneity of solids' spatial distribution and particle size differences. The Wu–Baeyens correlation provided the closest result to the simulation data when the size ratio was changed. [Display omitted]
► Modified Lacey's method proves effective for in-situ quantification of mixing ► The model allows us to predict layer inversion in gas–solid fluidized beds ► Mixing was achieved at a lower Uexcess for particles differing in size and density