In many applications of fluidized reactors, the particles processed are cohesive. The presence of inter-particle cohesive force causes different fluidization dynamics compared to non-cohesive system. This paper studied the fluidization dynamics of cohesive Geldart B particles in a three dimensional fluidized bed. X-ray tomography and in-bed pressure fluctuation analysis were conducted to reveal bubble behaviors as well as overall fluidization status. The results show that increasing cohesive force facilitates bubble coalescence, leading to the growth of bubble size. When bubbles grow to similar dimensions to the column cross section, the particles above bubbles are lifted as a particle slug and the fluidization turns to slugging. As the cohesive force increases, the energy distribution of power spectral density of pressure signal changes from mono-modal shape to bi-modal shape where two peaks relate to slugging and normal fluidization respectively. At low static bed heights, no slugging takes place even under high cohesive forces. Therefore, to avoid slugging in the practical application, the static bed height should be kept as low as possible. With further increase in the static bed heights, bubbles keep growing by continuous coalescence and become considerable near the bed surface, which triggers slugging and causes an increase in slugging duration.

Nanoparticles form multi-stage agglomerates when they are fluidized: simple agglomerates (a few dozens of micrometres), and complex agglomerates (a few hundreds of micrometres). This paper studies nanoparticle fluidization by using adhesive CFD-DEM (Discrete Element Model) simulation, which includes an adhesive contact model having the ability to model the contact and bounce/stick between elastic-plastic and cohesive particles. The discrete elements represent the simple agglomerates of nanoparticles. The effects of particle density, fluidizing gas velocity, and van der Waals force on gas-solid flow and agglomerate behaviors are investigated. It is shown that the particles, representing simple agglomerates, form large agglomerates during fluidized process. As the agglomerates move around in the bed, they continuously break and recombine. The solid movement is evaluated by a dispersion coefficient and it decreases with increase in cohesion. By monitoring the evolution of contacts, the agglomerate breakage is visualized and compared for the different conditions. Increasing fluidizing gas velocity and reducing particle adhesion both increase agglomerate breakage rate. The structure parameters of agglomerates are further analyzed: averaged coordination number of agglomerates around 3.0, packing fraction around 0.2–0.3, and fractal dimension around 1.9–2.3, which can be used to indicate structures and formation mechanisms of agglomerates.

Nanoparticles are fluidized as agglomerates with hierarchical fractal structures. In this study, we model nanoparticle fluidization by assuming the simple agglomerates as the discrete element in an adhesive (Computational Fluid Dynamics—Discrete Element Modelling) CFD-DEM model. The simple agglomerates, which are the building blocks of the larger complex agglomerates, are represented by cohesive and plastic particles. It is shown that both the particle contact model and drag force interaction in the conventional CFD-DEM model need modification for properly simulating a fluidized bed of nanoparticle agglomerates. The model is tested for different cases, including the normal impact, angle of repose (AOR), and fluidization of nanoparticle agglomerates, represented by the particles with the equivalent material properties. It shows that increasing the particle adhesion increases the critical stick velocity, angle of repose, and leads from uniform fluidization to defluidization. The particle adhesion, bulk properties, and fluidization can be linked to each other by the current adhesive CFD-DEM model.