Sub-grid modeling of gas-solid flows in fluidized beds

• Grant number: 15/22224-3
• Support Opportunities: Regular Research Grants
• Duration: June 01, 2017 - May 31, 2019
• Field of knowledge: Engineering - Chemical Engineering - Chemical Process Industries
• Principal Investigator: Fernando Eduardo Milioli
• Grantee: Fernando Eduardo Milioli
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Fluidized bed reactors are widely used in chemical and energy industries, representing a formidable impact on world's economy. The development of such reactors is still a very much empirical science, based upon gradually scaled demonstration plants that involve extremely high costs and times of accomplishment. The current research proposition intends to contribute in the context of replacing those very expensive plants by computational simulation. More specifically, intends to contribute for the development of realistic mathematical models for accurately describing gas-particle flows in fluidized beds. Owing to the commonly huge physical volumes that are involved, the simulation of real scale fluidized beds imposes filtered formulations, which require closure models to recover filtered effects. The current effort is turned to the formulation of increasingly realistic closure models for hydrodynamics and inter-phase interactions by means of computational experiment with microscopic two-fluid modeling. Filtered two-fluid models have been widely used to simulate gas-particle flows in fluidized beds, which required closure models for filtered interface interaction forces and stresses in each phase. Such closures have been recently derived form highly resolved simulations applying microscopic two-fluid modeling. The usual highly resolved simulation is applied over periodic domains where the flow driving force is enforced through a boundary imposed gas-pressure gradient which is chosen to exactly match the gravity acting on the average gas-particle mixture. This renders a particulate flow field which is upwards in low solid concentration regions and downwards in high solid concentration regions, with Reynolds numbers varying in a very confined low range. In such a condition Reynolds number is found not to significantly affect any filtered parameter. The current research proposition will investigate the effects of Reynolds number over filtered parameters under more realistic gas flow conditions. In practice, this means to impose gas-pressure gradients in excess to that required to exactly match the gravity acting on the gas-particle mixture. Besides investigating the effects of Reynolds number, researches are also proposed to investigate the effects of particle-particle collisional/frictional interactions and gas turbulence over the meso-scale hydrodynamics. The present concern is always turned to the application of high Stokes number particulates which are typical of fluidized bed applications. Computational experiments are to be developed for ranges of solid fractions and gas flow rates covering flow regimes from very dilute to very dense. The MFIX open source code is to be used in all the simulations. The ultimate goal which is pursued is to provide new more realistic sub-grid closure models for filtered two-fluid modeling of fluidized gas-solid flows. (AU)