Energy-minimization multiscale based mesoscale modeling and applications in gas-fluidized catalytic reactors

Abstract

<jats:title>Abstract</jats:title> <jats:p>Gas-solid fluidization is intrinsically dynamic and manifests mesoscale structures spanning a wide range of length and timescales. When involved with reactions, more complex phenomena emerge and thus pose bigger challenges for modeling. As the mesoscale is critical to understand multiphase reactive flows, which the conventional two-fluid model without mesoscale modeling may be inadequate to resolve even using extremely fine grids, this review attempts to demonstrate that the energy-minimization multiscale (EMMS) model could be a starting point to develop such mesoscale modeling. Then, the EMMS-based mesoscale modeling with emphasis on formulation of drag coefficients for different fluidization regimes, modification of mass transfer coefficient, and other extensions are discussed in an attempt to resolve the emerging challenges. Its applications with examples of development of novel fluid catalytic cracking and methanol-to-olefins processes prove that the mesoscale modeling plays a remarkable role in improving the predictions in hydrodynamic behaviors and overall reaction rate. However, the product content primarily depends on the chemical kinetic model itself, suggesting the necessity of an effective coupling between chemical kinetics and flow characteristics. The mesoscale modeling can be believed to accelerate the traditional experimental-based scale-up process with much lower cost in the future.</jats:p>

Hydrodynamic Full-Loop Simulation of an Industrial Fluid Catalytic Cracking Fluidized Bed System with Mechanical Valves

“…The gas–solid interphase drag coefficient, β, employs the EMMS-based drag model since it shows grid-insensitive characteristics and is thus suitable for coarse-grid simulations. − The EMMS-based drag coefficient is expressed by β = 3 4 C normald 0 ρ g ( 1 − ε normals ) ε s | bold-italicν⃗ normalg − bold-italicν⃗ normals | d normalp ε normalg − 2.65 H D In the above, H D is the heterogeneity index, defined by β/β 0 , and β 0 is calculated from the Wen and Yu model for homogeneous fluidization. The standard drag coefficient for an individual particle, C d0 , and the Reynolds number, Re s , are expressed by C normald 0 = { lefttrue 24 italicRe normals ( 1 + 0.15 italicRe normals 0.687 ) , Re s ≥ 1000 0.44 , Re s < 1000 , Re s = ε g ρ g d p | bold-italicν⃗ normalg − bold-italicν⃗ normals | μ normalg…”

“…Unstructured grids are used to mesh the transition section, distribution plate, cyclone, stripper, and elbow regions and structured grids for the cylindrical regions, totaling about 1.7 million grids, as shown in Figure . According to many successful coarse-grid simulations of large-scale industrial-scale reactors with EMMS-based drag models, − this grid resolution is enough to conduct coarse-grid simulations of industrial-scale reactors.…”

Multiscale Eulerian CFD of Chemical Processes: A Review

“…The resin penetration velocity (u p ) at the two tow scales was interconnected, therefore several iterations were necessary to obtain a converged K value. In gas-solid fluidization systems, several multiscale concepts for the Eulerian CFD model have been proposed: a filtered drag force (FDF) model using constitutive equations from highly resolved subgrid models [103], a mesoscale structure-based drag model, where the flow structure and stability conditions were considered to capture the clustering behavior of solid particles using the EMMS model [46,73,104,105], and a CSD drag model [87,100].…”

“…In gas-solid fluidization systems, several multiscale concepts for the Eulerian CFD model have been proposed: a filtered drag force (FDF) model using constitutive equations from highly resolved subgrid models [103], a mesoscale structure-based drag model, where the flow structure and stability conditions were considered to capture the clustering behavior of solid particles using the EMMS model [46,73,104,105], and a CSD drag model [87,100]. Lu et al (2019) and Wang (2020) [46,47] reviewed several sequential coupling strategies between the EMMS drag and Eulerian CFD models to speed up the multiscale simulation using a lookup table [105] or a fitting function [104] of the drag force.…”

“…Drikakis et al (2019) presented the application of CFD to the energy field [3], integrated together with molecular dynamics and LBMs. Lu et al (2019) [46] and Wang (2020) [47] reviewed Eulerian CFD approaches for dense gas-solid flows, providing a concise introduction to multiscale methods and highlighting the effects of mesoscale structures (gas bubbles and particle clusters) on the gas-solid Eulerian CFD model, focusing on the energy minimization multi-scale (EMMS) method.…”