(448al) Solids mixing in bubbling fluidized beds: CFD-based analysis of BubbleDynamics and Time Scales

Bubbling fluidized bed reactors are commonly employed for combustion and gasification applications. In these reactors, mixing of the bed material is critical because it directly affects fuel distribution/segregation and residence time as well as wall-to-surface heat transfer. Although solids mixing has been the focus of substantial effort over the past few decades, there continues to be a lack of understanding, especially at commercial-scales, because (a) most diagnostic techniques are only feasible in lab-scale setups, where the hydrodynamics are substantially different [1], and (b) the dynamics are significantly sensitive to the operating conditions (such as superficial gas velocity and particle characteristics). Thus, it is not surprising that quantitative estimates of mixing (e.g., dispersion coefficient, mixing indices) and their dependence on the operating conditions often span orders of magnitude. Nevertheless, it is well accepted that the micro-mixing and gross circulation of solid particles is driven by bubble motion through wake lift, emulsion drift and bubble eruptions. Quantifying this dependence is the focus of this study.

Solids mixing is investigated using fine-grid 3D CFD simulations. The fluidization of two distinct Geldart B particles (0.5 mm glass and 1.15 mm LLDPE) is simulated in an intermediate size 50 cm diameter fluidized bed since hydrodynamics at this scale are wall-independent and, therefore, scalable [1]. Detailed diagnostics of the computed flowfield data are performed using MS3DATA [2], a tool that we developed to detect and track bubbles in 3D, and the solids axial and lateral motion are correlated with bubble rise. Subsequently, the mixing time-scales in the fluidized bed are estimated using the spatial and size distribution of bubbles. This is one of the first studies investigating solids mixing using 3D simulations an intermediate size bubbling fluidized bed. It provides valuable insights for estimating mixing time scales in larger beds, as well as for the reactor network modeling of bubbling fluidized beds (e.g. [3]). All simulations are based on the Two-Fluid Model (TFM) framework and performed using MFiX.

References

[1] A. Bakshi, C. Altantzis, R.B. Bates and A.F. Ghoniem, Study of the effect of reactor scale on fluidization hydrodynamics using fine-grid CFD simulations based on the two-fluid model, Powder Technology, 299: 185-198, 2016

[2] A. Bakshi, C. Altantzis, R.B. Bates and A.F. Ghoniem, Multiphase-flow Statistics using 3D Detection and Tracking Algorithm (MS3DATA): Methodology and application to large-scale fluidized beds, Chemical Engineering Journal 293: 355-364, 2016

[3] A.K. Stark, C. Altantzis, R.B. Bates and A.F. Ghoniem, Towards an advanced reactor network modeling framework for fluidized bed biomass gasification: Incorporating information from detailed CFD simulations. Chemical Engineering Journal. 303: 409-424, 2016