| Popis: |
The focus of this dissertation is the numerical analysis of confined aerosol jets used in fiber classification and dustiness measurement. Of relevance to the present work are two devices, namely, the Baron Fiber Classifier (BFC), and the Venturi Dustiness Tester (VDT). The BFC is a device used to length-separate fibers, important for toxicological research. The Flow Combination Section (FCS) of this device consists of an upstream region, where an aerosol of uncharged fibers is introduced in the form of an annular jet, in-between two sheath flows. Length-separation occurs by dielectrophoresis, downstream of the FCS in the Fiber Classification Section (FClS). In its standard operation, BFC processes only small quantities of fibers. In order to increase its throughput, higher aerosol flow rates must be considered. The goal of the present investigation is to understand the interaction of sheath and aerosol flows inside the FCS, and to identify possible limits to increasing aerosol flow rates using Computational Fluid Dynamics (CFD). Simulations involve solution of Navier-Stokes equations for axisymmetric and 3D models of the FCS for six different flow rates, and a pure aerodynamic treatment of the aerosol jet. The results show that the geometry of the FCS, and the two sheath flows, are successful in preventing the emergence of vortices in the FCS for aerosol-to-sheath ?ow inlet velocity ratios below ˜ 50. For larger aerosol-to-sheath flow inlet velocity ratios, two vortices are formed, one near the inner cylinder and one near the outer cylinder.The VDT is a novel device for measuring the dustiness of powders, relevant for dust management and controlling hazardous exposure. It uses just 10 mg of the test powder for its operation, during which the powder is aerosolized and turbulently dispersed (Re = 19,900) for 1.5s into a 5.7 liter chamber; the aerosol is then gently sampled (Re = 2050) for 240s through two ?lters located at the chamber top. Pump-driven suction at the Extraction Port, located at the chamber top, activates injection and can lead to loss of powder before the Sampling Phase begins. The goal of this work is to analyze the flow in the VDT during its operation using CFD. An Unsteady-Reynolds-Averaged Navier-Stokes formulation (Shear-Stress Transport turbulence model), along with fluid/powder tracking in a Lagrangian framework, is used for simulations. The obtained results are applicable for powder with Stokes number St << 1. It is observed that the Injection Phase results in a uniform distribution of powder inside the VDT, for both homogeneous and inhomogeneous dispersion, satisfying a necessary condition for the accurate evaluation of dustiness. The results show that ˜ 24% of fluid-tracers, injected between 0 and 0.24s of flow time, escape the VDT before the Sampling begins. Simulation results for reduced extraction flow rates showed significant non-uniformity in powder distribution, confirming the criticality of standard extraction flow rates for uniform distribution of powder. This study lends confidence on accuracy of the VDT in determining dustiness, and supports other ongoing efforts towards establishing the VDT as an ISO standard device for dustiness measurement. |
