Fe (II) biooxidation has recently gained significant interest. It plays a key role in a number of environmental and industrial processes such as bioleaching, acid mine drainage treatment, desulphurization of sour gases, and coal desulphurization. In this work, a three-dimensional CFD model for gas-liquid flow in a lab-scale packed-bed biooxidation reactor is used. The reactor is randomly packed with spherical particles, and the particles are covered with Leptospirillum ferrooxidans biofilm for Fe (II) biooxidation. A modified Jodrey-Tory algorithm is used to generate random packing with actual porosity of 0.42, and biofilm layer with constant thickness is considered over the particles. A simplified Eulerian–Eulerian model is used to obtain detailed flow field. The concentration profile in the reactor and the conversion of Fe (II) from the present simulations are obtained and validated using experimental data reported in the literature. The results of the study indicate that about three-quarters of the conversion occurs in the upper half of the reactor and Fe (II) concentration on the biofilm surface at the lower quarter of the reactor does not exceed 5 mM (The inlet concentration is 89.6 mM). The findings reveal that rate-limiting phenomena may vary in different parts of the reactor. The results obtained through the simulations represent advantages for the design and optimization of packed-bed biofilm reactors.
The sedimentation of solid particles in stirred bioreactors such as bioleaching reactors reduces the available solid surface area, which results in a lower rate of reaction. On the other hand, excessive impeller speed, in addition to energy consumption, especially in large reactors, can negatively affect the performance of microorganisms. Therefore, the impeller must be set at the optimum speed. In this study, the standard k-ε model was employed to model turbulency and a particle tracing module was used to predict the percentage of suspended particles in different speeds and find the optimal speed. The results show at the speeds higher than 300 rpm sedimentation does not take place.
Simulating packed beds in detail using CFD technique has obtained much more attention in the recent years. Detailed simulation provides a better understanding of transport phenomena and local information in packed-bed reactors. The first step of these studies is to create their complex geometry. Various algorithms have been proposed for generating such geometries in 60 years. This study reviews common algorithms which are employed to generate sphere particles in cylindrical containers. A packed-bed is also generated and simulated. COMSOL Multiphysics was used for detailed simulation of flow behavior in a 110-sphere random packed bed. This simulation was carried out in laminar flow regime at steady state. The pressure drop calculated in simulation was in satisfaction with Carman-Kozeny equation. Reducing the diameter of spheres from 0.5 to 0.49 caused 8.2% error in pressure drop.
A simplified model for the dynamic removal of nickel (II) in a fixed bed biosorption column is presented. The key assumption of the model is that equilibrium is fast enough and mass transfer resistances do not exist for biosorption of metal ion on the biosorbent materials. The equilibrium kinetics is expressed with Langmuir isotherm. The resulting partial differential and Darcy’s law equations represent mass balance for nickel (II) ion and flow hydrodynamic in biosorption fixed bed column. These equations were solved with COMSOL Multiphysics 5.4 simultaneously. An apparent axial dispersion coefficient has been used as a main parameter of the model. The breakthrough curve and the bed concentration distribution show good agreement with experimental data. The operating parameters of the apparent axial dispersion coefficient and bed density are also examined. These parameters have major effects on the results according to parameter effects on the breakthrough curve.