Session A41: Particle-Laden Flows: Particle-Resolved Simulations

Particle-Resolved DNS of heavy and neutrally-buoyant particles with the volume-filtering method

We present a novel computational method for direct numerical simulations of particle-laden with fully-resolved particles (PR-DNS). The method is based on the recently developed Volume-Filtering Immersed Boundary (VFIB) method derived by volume-filtering the transport equations. This approach is mathematically and physically rigorous, in contrast to other PR-DNS methods which rely on ad-hoc numerical schemes to impose no-slip boundary conditions on the surface of particles. With the present PR-DNS based on VFIB, we show that the ratio of filter size to particle diameter acts as a slider that controls the level of fidelity. In the limit where this ratio is very small, a well-resolved PR-DNS is obtained. Conversely, when the ratio of filter size to particle diameter is large, a classic point-particle method is obtained. The discretization of the filtered equations is discussed and compared to other PR-DNS strategies based on direct-forcing immersed boundary methods. Numerical examples with sedimenting resolved spheres show that the method is very robust, stable, and accurate, even with neutrally buoyant particles. The volume-filtering formulations enables multi-resolution simulations where, depending on the local resolution, the particles are represented as resolved spheres or Lagrangian particles.   

Particle-resolved simulations of the wall effect on the sedimentation of a single sphere in yield-stress fluids

We perform high-fidelity numerical simulations to investigate the confinement effect on the sedimentation of a single spherical particle in an otherwise quiescent yield stress fluid. The simulations are performed in the presence of finite elasticity and weak inertia. The carrier fluid is modeled utilizing the elastoviscoplastic constitutive laws proposed by Saramito (2009). The additional elastic stress tensor is fully coupled with the flow equation, while the rigid particle is represented by an immersed boundary method. The particle-resolved simulations demonstrate the faster relaxation of the fluid velocity and the progressive translation of the location of the negative wake downstream of the sphere as the bounding walls are brought closer to the particle. Furthermore, the drag force exerted from the surrounding fluid on the particle decreases by escalating the particle-wall distance. We show that the confinement ratio (ratio of the gap between rigid confining walls and the sphere radius) reaches a critical value beyond which the wall-effect on the particle and flow dynamics becomes negligible. The key finding here is that the critical confinement ratio and the maximum variation of the Stokes drag with confinement ratio are weakly dependent on the level of material elasticity and plasticity for a certain range of material parameters. Finally, we propose an expression for the Stokes drag coefficient, as a function of material plasticity and confinement ratio.   

Discrete Transport from a Tumbling Prolate Spheroid in a Simple Shear Flow

Through high-fidelity numerical simulation based on a multi-grid strategy with the lattice Boltzmann framework, an in-depth study on the heat and mass transport from a prolate spheroid suspended in a simple shear flow has been carried out. In the simulation, the temperature and mass concentration are modeled as a passive scalar released at the spheroid's surface. The interaction between the carrier fluid and the suspended spheroid, as well as the resultant scalar transport process, have been investigated. The interaction creates several particle-scale flow modes in the fluid layer surrounding the spheroid. When the advective scalar transport is weak, the passive scalar released at the spheroid surface is transported through the fluid layer to the outer passing fluid via diffusion, and then is transported downstream by the passing fluid via flow advection. When the advective scalar transport is strong, most of the passive scalar is gained by the fluid layer surrounding the spheroid. The multiple flow modes of the fluid layer and the outer passing fluid lead to multiple advective scalar transport modes, constituting a spatiotemporally discrete set of scalar transport mechanisms for the prolate spheroid in a simple shear flow. Factors such as spheroid-to-fluid density ratio, spheroid aspect ratio, and Reynolds number, affect the behaviors of the fluid and spheroid, thus have influences on the scalar transport.   

Interface-resolved simulation of particles with different shapes in sediment transport

Sediment transport is a dynamic process with significant implications for landscapes and human activities. The movement and collision of sediment particles have a profound impact on the surrounding flow field. Among various flow configurations and particle properties, the shape of particles plays a crucial role in the fluid-particle interaction process. In this work, we employ numerical simulations to investigate sedimentation transport for different particle shapes, including spherical particles, oblate particles, and their mixtures.

 

To accurately capture the complex interactions between the flow and particles, we employ the immersed boundary method. This approach allows us to resolve the intricate fluid-particle interactions that cannot be adequately represented using a point particle model. Furthermore, collisions between particles are modeled using a linear spring-dashpot system, with the Adaptive Collision Time Model employed to calculate the contact force.

 

From simulations, we observe ripples during particle transport. By comparing particle velocities and angular velocities among different cases, we gain insights into the influence of particle shape on particle movement. Additionally, we examine the mean shear stress and its components, including the viscous contribution, Reynolds shear stress, and fluid-particle interaction. The analysis provides a comprehensive understanding of how each component affects the shear stress in the flow.   

Towards a more realistic configuration to simulate the settling of finite-size particles in the dilute regime

Simulating the settling of particles in the dilute regime is a very complex problem which, thanks to the relatively recent maturity reached by immersed boundary methods, we can analyze by means of particle-resolved direct numerical simulations. The cost of these simulations is still very high and one of the assumptions typically made by the authors to reduce the complexity of the problem is the use of a triply periodic configuration. This configuration has shown to be very successful to understand the onset of clustering, but the analysis of cluster dynamics is not possible because of the strong correlation of the solution in the vertical direction. In this talk we present preliminary results of the settling of particles in a configuration where the periodicity in the vertical direction is removed. The advantages and challenges of using this configuration will be also briefly discussed. In order to exploit our previous knowledge on the settling of particles we set the Galileo number to 121 and 178 and the particle-to-fluid density ratio to 1.5 based on the work of Uhlmann and Doychev (2014). We observe a similar clustering behavior as in the triply periodic counterpart, but at a higher particle concentration.   

High Resolution Simulations of Porous Fractal Aggregates Settling through Density Gradients

We perform Particle-Resolved Direct Numerical Simulations (PR-DNS) on rigidly-bonded fractal aggregates of spherical particles settling across varying density gradients. These simulations are performed to study the impact of the complex, fractal geometry on the settling of the aggregate, particularly as it passes into a region of high density fluid while retaining less dense fluid in the pore space between individual particles. The study of these aggregates and their behaviors has applications in a broad range of problems in environmental fluid dynamics. We investigate over many simulations the velocity and rotation of the aggregates, varying the number of particles N in an aggregate, the aggregate’s fractal dimension D_f, and the associated fractal prefactor k_f, which together all characterize the overall structure of the cluster. We compare these results with the settling of spheres of comparable volume to the aggregate, to discuss the effect that the shape of the aggregate has on its motion. From these results, we intend to develop general relationships between the size of the aggregate and its fractal parameters to its overall settling velocity during sedimentation.   

Flocculation of Finite Sized Particles in Turbulent Channel Flow

We present a computational study of the flocculation dynamics and turbulence modulation in a channel flow laden with small, finite size, cohesive particles. We conduct a series of four way coupled, grain-resolved direct numerical simulations, across which the strength of the short-range inter-particle cohesive force is varied, in order to determine its effect on the flocculation process. The strength of cohesion is shown to govern the size (O(1-100) particle diameters) and number of aggregates. Additionally, we analyze specific floc properties such as size, shape, and lifespan as a function of the wall normal distance. The higher mean velocity gradient and turbulent activity near the wall is found to both facilitate aggregation by bringing particles together, while also prompting breakup. Finally, enabled by the fully coupled nature of the simulations, we quantify the effect of aggregation on the turbulence by reporting classical first and second order turbulence statistics for both the fluid and particles.   

Particle-Resolved DNS of flow modulation induced by finite-size particles dispersed in a turbulent channel

The modification of turbulent structures by finite size particles is of interest to pipeline design for oil and slurry transport. The interaction of finite size particles with turbulent flow structures is investigated in Particle-Resolved Direct Numerical Simulations of turbulent channel flow at friction Reynolds number 180 laden with spherical particles with diameter equal to 12 wall units and dispersed at the volume fraction 1%. To understand the role of particle inertia in modulating turbulence, particle-to-fluid density ratio is varied from 1, i.e., neutrally buoyant particles, up to 4. About 2320 resolved particles are tracked throughout the computational domain and represented using the recently developed volume-filtering immersed boundary method to account for stresses at the fluid solid interface. Additional turbulence statistics are collected for a particle free channel for comparison with the particle laden channels.