The agglomeration of particles caused by the formation of capillary bridges has a decisive impact on the transport properties of a variety of at a first sight very different systems such as capillary suspensions, fluidized beds in chemical reactors, or even sand castles. Here, we study the connection between the microstructure of the agglomerates and the rheology of fluidized suspensions using a coupled lattice Boltzmann and discrete element method approach. We address the influence of the shear rate, the secondary fluid surface tension, and the suspending liquid viscosity. The presence of capillary interactions promotes the formation of either filaments or globular clusters, leading to an increased suspension viscosity. Unexpectedly, filaments have the opposite effect on the viscosity as compared to globular clusters, decreasing the suspension viscosity at larger capillary interaction strengths. In addition, we show that the suspending fluid viscosity also has a nontrivial influence on the effective viscosity of the suspension, a fact usually not taken into account by empirical models.

For description of the fluidization state of fluidized beds, both time-domain and frequency-domain analyses of highfrequency pressure fluctuations are established approaches. Common methods for the detection of agglomeration or defluidization in fluidized beds use the variance or the standard deviation of the pressure signal or the maximum in its frequency spectrum. These methods are used, for example, in biomass combustion or gasification. However, these approaches lack the reliability for applications as an early agglomeration warning system in industrial applications. To address this issue, the present study introduces a robust methodology by means of extracting a characteristic frequency from the power spectral density of the pressure signal. A comparison of our developed approach with the commonly used frequency maximum and standard deviation for predicting the onset of agglomeration in laboratory experiments shows promising sensitivity on agglomeration formation. In order to evaluate the general applicability of this method on an industrial scale, this work investigates dependencies of possible influences, such as gas velocity, sand quantity, and temperature, on the characteristic frequency. The results indicate that the characteristic frequency can be a promising and robust method for the early detection of the onset of agglomeration in industrial plants.

The agglomeration of particles caused by the formation of capillary bridges has a decisive impact on the transport properties of a variety of at a first sight very different systems such as capillary suspensions, fluidized beds in chemical reactors, or even sand castles. Here, we study the connection between the microstructure of the agglomerates and the rheology of fluidized suspensions using a coupled lattice Boltzmann and discrete element method approach. We address the influence of the shear rate, the secondary fluid surface tension, and the suspending liquid viscosity. The presence of capillary interactions promotes the formation of either filaments or globular clusters, leading to an increased suspension viscosity. Unexpectedly, filaments have the opposite effect on the viscosity as compared to globular clusters, decreasing the suspension viscosity at larger capillary interaction strengths. In addition, we show that the suspending fluid viscosity also has a non-trivial influence on the effective viscosity of the suspension, a fact usually not taken into account by empirical models.