Bulletin of the American Physical Society
68th Annual Meeting of the APS Division of Fluid Dynamics
Volume 60, Number 21
Sunday–Tuesday, November 22–24, 2015; Boston, Massachusetts
Session A40: CFD: Particulate Flows |
Hide Abstracts |
Chair: Sreenath Krishnan, Stanford University Room: Sheraton Back Bay D |
Sunday, November 22, 2015 8:00AM - 8:13AM |
A40.00001: Immersed boundary methods for viscoelastic particulate flows Sreenath Krishnan, Eric Shaqfeh, Gianluca Iaccarino Viscoelastic particulate suspensions play key roles in many energy applications. Our goal is to develop a simulation-based tool for engineering such suspensions. This study is concerned with fully resolved simulations, wherein all flow scales associated with the particle motion are resolved. The present effort is based on Immersed Boundary methods, in which the domain grids do not conform to particle geometry. In this approach, the conservation of momentum equations, which include both Newtonian and non-Newtonian stresses, are solved over the entire domain including the region occupied by the particles. The particles are defined on a separate Lagrangian mesh that is free to move over an underlying Eulerian grid. The development of an immersed boundary forcing technique for moving bodies within an unstructured-mesh, massively parallel, non-Newtonian flow solver is thus developed and described. The presentation will focus on the numerical algorithm and measures taken to enable efficient parallelization and transfer of information between the underlying fluid grid and the particle mesh. Several validation test cases will be presented including sedimentation under orthogonal shear – a key flow in drilling muds and fracking fluids. [Preview Abstract] |
Sunday, November 22, 2015 8:13AM - 8:26AM |
A40.00002: Influence of Non-homogeneous Particle Distributions on Drug Release in a Couette \textit{in vitro} Dissolution Device Balaji Jayaraman, James Brasseur, Yanxing Wang Drug dissolution rates from powdered formulations are commonly measured in \textit{in vitro} devices. Both measurements and models commonly assume perfect mixing of drug and particle within the device. In this study we analyze the potential importance of heterogeneity in particle concentration and distribution using CFD that incorporates physically accurate mathematical representations of hydrodynamic enhancement of mass transport from shear as applicable to drug dissolution \textit{in vivo} as well as \textit{in vitro}. We have developed a high-fidelity computational formulation using the Lattice Boltzmann Method (LBM) with the parallel particle tracking for a polydisperse collection transported by the flow. Drug release from the small (\textless 100 $\mu $m) Lagrangian `point' particles is modeled using a mathematical framework that is built on a validated first principles `quasi-steady state' approximation with correlations for shear enhancement and integrated with the coarser Eulerian LBM flow field using a subgrid formulation Our Eulerian-Lagrangian formulation takes into account spatial variations in particle `bulk' concentration from polydisperse particle distributions with specified particle distribution heterogeneities. We shall discuss the primary influences of heterogeneous bulk concentrations surrounding individual particles and non-homogeneous particle distributions in an \textit{in vitro} Couette flow device to quantify the relative influences of shear enhancement on drug dissolution \textit{in vivo} vs. \textit{in vitro} [Preview Abstract] |
Sunday, November 22, 2015 8:26AM - 8:39AM |
A40.00003: Comparison of Strongly Coupled Diffuse and Sharp Interface Fluid-Structure Interaction Approaches for Particle-Laden Flows Fazlolah Mohaghegh, H.S. Udaykumar The aim of this study is to find a proper method for the simulation of blood as a particulate flow. Since the blood cell density is almost the same as plasma, the high added mass effect necessitates implementation of a strongly coupled FSI method in the numerical simulation. Therefore, three different FSI approaches are compared, two Smoothed Profile Methods (SPM) with one and two projection steps as diffuse interface approaches and the Sharp Interface Method (SIM). Stable FSI computations can be achieved by using sub-iterations within each time step, i.e. by updating the fluid and structure and their boundary conditions at each time step multiple times to reach a desired tolerance as the convergence criteria. Various cases were used to benchmark the methods, including particles motion in a channel and particles sedimentation. The results show that the number of sub-iterations plays a key role in the efficiency. While use of SPM with two projection steps has the most expensive sub-iteration process, it has the best efficiency as it requires the lowest number of sub-iterations within each time step. Moreover, the method is more stable than SIM and the SPM with one projection. SIM is faster than SPM with one projection and it has better stability. [Preview Abstract] |
Sunday, November 22, 2015 8:39AM - 8:52AM |
A40.00004: A hierarchical Cartesian method for particle-laden flows with conjugate heat transfer Gonzalo Brito Gadeschi, Matthias Meinke, Wolfgang Schroeder We present a numerical method for simulating particle-laden flows including conjugate heat transfer between the fluid and the particle phase, where the particles are fully-resolved. This is a challenging problem since the flow field around the moving particles as well as the heat exchange across the particles surface has to be resolved at any time instant. It is also computationally expensive when particles with a diameter on the order of, or smaller than, the Kolmogorov length require a local mesh resolution which exceeds that of a Direct Numerical Simulation of the turbulent fluid phase alone. The present approach uses an efficient adaptive algorithm to couple the two numerical methods for the fluid phase and the heat conduction in the particles. The algorithm is based on hierarchical Cartesian grids and is especially suited for moving boundary problems. The fluid phase is solved with a thermal Lattice-Boltzmann method while the heat equation in the particle phase is discretized based on a finite volume method. The moving surface of the particles is tracked using a level-set method. The approach is validated against results available from the literature and its applicability is demonstrated it by studying canonical problems of suspended rigid particles in channel-like flows. [Preview Abstract] |
Sunday, November 22, 2015 8:52AM - 9:05AM |
A40.00005: Direct numerical simulation of evaporation-induced particle motion Hochan Hwang, Gihun Son A sharp-interface level-set (LS) method is presented for direct numerical simulation (DNS) of evaporation-induced particle motion. The liquid surface is tracked by the LS function, which is defined as a signed distance from the liquid-gas interface. The conservation equations of mass, momentum, energy for the liquid and gas phases and vapor mass fraction for the gas phase are solved accurately imposing the coupled temperature and vapor fraction conditions at the evaporating liquid-gas interface. A dynamic contact angle model is also incorporated into the LS method to account for the change between advancing and receding contact angles at the liquid-gas-solid contact line. The solid surface is tracked by another LS function, which is defined as a signed distance from the fluid-solid interface. The conservation equations for multiphase flows are extended to treat the solid particle as a high-viscosity non-evaporating fluid phase. The velocity inside the solid domain is modified to enforce the rigid body motion using the translational velocity and angular velocity of the particle centroid. The DNS results demonstrate the particle accumulation near the evaporating interface and the contact line pinning and stick-slip motion near the evaporating contact line. [Preview Abstract] |
Sunday, November 22, 2015 9:05AM - 9:18AM |
A40.00006: Numerical simulation of contact line motion and particle distribution in dip coating Gihun Son, Jaewon Lee A level-set method is presented for computing contact line motion and particle distribution in dip coating, which is a popular process for production of thin films and has received new attention as a simple particle deposition process for patterning microstructures. Assuming that the interface temperature is below the saturation temperature, we solve the conservation equations of mass, momentum, and energy in the liquid-gas phases, the vapor mass fraction in the gas phase, and the particle concentration in the liquid phase. To consider the case where the particle concentration reaches the maximum value (in random packing), the diffusion coefficient of particles is determined from the generalized Stokes-Einstein equation. The temperature and vapor fraction at the interface and the evaporation mass flux are simultaneously determined from the coupled equations for the mass and energy balances at the interface and the thermodynamic relation. The present computations demonstrated that the plate withdrawal velocity significantly affects the liquid film formation and particle distribution pattern. In the regime of a low plate velocity, the computed liquid-gas-solid contact line reaches a quasi-steady state and the particle accumulation is pronounced near the stationary contact line. [Preview Abstract] |
Sunday, November 22, 2015 9:18AM - 9:31AM |
A40.00007: Clustering Instability in Sedimenting Gas-Solid Suspensions and its Influence on Flow Properties. Xiaoqi Li, Xiaolong Yin, Guodong Liu It is well known that sedimentation or fluidization of solid particles through gas is unstable. Instability is usually recognized as particle clusters when the solid fraction is low, or as void `bubbles' when the solid volume fraction is high. Using particle-resolved numerical simulations, we studied cluster formation in gas-solid systems with gas-to-solid density ratio being 0.01 and 0.001. The particles are uniformly sized spheres with a terminal Re of 30. The solid fraction is 0.25. Up to 4808 particles were used such that the clustering phenomena can be adequately examined. In periodic computational domains whose lateral dimension is about eight particle diameters, nucleated particle clusters quickly coalesce and grow into traveling waves that span the entire width of the domain. Consequently, gas-solid drag is significantly increased compared to that in a homogeneous liquid-solid suspension, the lateral velocity variance is suppressed, and the particle velocity distributions are strongly non-Gaussian. When lateral dimension is increased to about thirty particle diameters, particle clusters never turn into width-spanning traveling waves. As results, the drag is similar to that in a homogeneous suspension, the lateral velocity variance is strongly enhanced and the vertical variance reduced, and particle velocity distributions are nearly Gaussian. These results suggest that the effect of particle clusters should be examined in domains with large lateral dimensions. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2019 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700