Liquid-solid Conical fluidized bed has wide applicability in various processes including coating, crystallization, coal liquefaction, water treatment, ore cleaning and separation, mining etc. However, the fundamental understanding of liquid-solid conical fluidized bed behaviour seems far from their applications. Further, most of the above mentioned processes deal with wide range of particle size and/or density distribution which makes the system more complicated as particle size and bed density distribution plays a significant role. The fluidized bed behaviour such as bed expansion, particle segregation and mixing completely depends upon composition of the bed and process conditions. Thus, it is necessary to understand the effect of particle size and density distribution on behaviour of binary conical fluidized bed. Though some studies in the literature are reported on such system however, most of them or based on bulk measurements and pressure drop analysis. In-situ measurement of hydrodynamic flow quantities in such system is severely missing.

In the present work, hydrodynamic study of liquid-solid conical fluidized bed with binary particles is investigated through state-of-the-art experimental technique and advanced CFD modelling.  Glass beads of two different sizes (0.6 mm and 1 mm) with same density (2500 kg/m³) are used as solids and water is used as a liquid phase. Euler-Lagrangian model is used for simulating the flow in which history of each particle is tracked by using Newton’s second law of motion. Navier-Stokes equation is used for continuous phase modelling. The particle-particle interaction and collision between vessel wall and particle is modelled through spring dash pot model given by Cundall and Strack (1979). Radioactive Particle Tracking (RPT) technique is used to track the motion of both the particles individually. In RPT technique a radioactive particle emitting gamma ray is used as marker of the phase. The tracer particle is prepared in such a way that it behaves identical to the phase of interest. An array of scintillation detectors is strategically located around the column. The tracer particle emits photon during its movement which are recorded by the detectors placed around the column. The position time series data of the particle can be reconstructed by using suitable reconstruction algorithm. Thereafter, other parameters like instantaneous velocities, mean velocities, rms velocities, kinetic energy of fluctuations etc., is calculated by using suitable post processing.

Experiments and simulations are performed for different liquid inlet velocities and bed compositions. The composition of these two different sizes of particle is changed in the experiments keeping over all mass of the bed constant. The bed composition of 100:0, 25:75, 50:50, 75:25 and 0:100 are used for 0.6 and 1 mm particles respectively. The bed is operated at three different liquid velocities i.e. 2Umf, 3Umf and 4Umf. The RPT experiments are performed for each condition by tracking both the solids individually. Further simulations are conducted for the same conditions. Simulations results, in which motion of all the particles are tracked for small time are validated with the RPT experiments where motion of one particle is tracked for longer time. Therefore, if system is ‘ergodic’ both should give exactly same results. The validation is presented for all the inlet velocities and bed compositions. Further, conditions for a particular mixing and segregation pattern are determined from experimental and numerical simulations.