CBET Award Achievements
Notable Accomplishments from CBET Awards
 

GOALI: Toward a Fundamental Understanding of Elutriation in Fluidized Beds

Christine Hrenya, University of Colorado at Boulder

Backgound: Solid particulates, or granular materials, are everywhere around us. We come into direct contact with them in the foods we eat (cereals, grains, dry milk, powder formulas for infants, etc.), the medication we take (solid tablets; powder formulas), the air we breathe (solid pollutant particles), and the beaches we walk on. We also have indirect contact with them on a daily basis in the electrical energy we consume (coal combustion for power generation), the fuel we use (solid catalysts used in the cracking of petroleum), and the sunblocks we apply (titania particles as used to shield the sun's harmful rays), to name only a few. Despite their ubiquity, processes involving granular materials are not well understood, and thus often treated in a wasteful trial-and-error manner. In this work, the focus is on the fluidized bed operation, with particular emphasis on the unwanted carryover of small particles (elutriation). Hundreds of empirical correlations are available for such systems, though they have been shown to vary over 100% from one another for a given system. The goal is to develop a first-principles, mathematical model describing this phenomenon, using a combination of new theory, experiments, and simulations.

Results: As driven by an industrial process of great interest to the GOALI partner on this grant (Lyondell Chemicals), the focus has been on a binary mixture of particulates. Such mixtures are known to exhibit non-intuitive behavior, such as an increased carryover of heavy particles with the addition of light particles, which is not explained by current models. The focus here is on the incorporation of two mechanisms which require modification for binary systems: particle-fluid drag force and particle-particle collisions. In an effort to isolate these mechanisms and assess them individually, focus in the past year has been on low-velocity systems in which drag forces predominate (and collisions have a negligible influence). New theory on binary drag forces has been incorporated into a simulation framework provided in-kind by the other GOALI partner (Arena-flow, maker of computational fluid dynamics software for particulate flows). The simulations indicate that the conventional ad-hoc approach to describing drag forces in binary mixtures results in markedly different behavior than an improved theory targeted specifically at binary mixtures (See Figure below). A corresponding experiment has also been built and run, thereby serving as a validation testbed for the improved drag force model.


Christine Hrenya Image
 
Discrete-particle simulation of a fluidized bed in which segregation is observed.
Small particles (red) segregate toward bottom of bed, while a mixture of
large (blue) and small particles rise to the surface.


Credit:  Christine Hrenya, University of Colorado


Scientific Uniqueness: This work involves unique contributions to both the physics and numerics associated with the description of particulate flows. In the first regard, the work illustrates the profound impact of a first-principles approach to describing the drag force in polydisperse systems on the segregation (de-mixing) of unlike particles. From a numerics standpoint, the research involves a new hybrid approach to simulating particulate flows which allows for the simulation of millions or billions of particles, where previous methods were restricted to tens of thousands of particles. The hybrid is a cross between traditional molecular dynamics (MD) methods and conventional continuum (kinetic-theory-based) models.

Impact on Industry and/or Society: Because the theory utilized here is based on first principles, the applicability of the corresponding models goes well beyond the specific unit operation examined here. Because particles of different size and density have a tendency to segregate upon agitation, the driving forces and corresponding prediction of segregation is important both when mixing is desired (blending of binders and medication prior to tablet formation) or unwanted (separation of ores).

In addition, the physical models are being directly incorporated into a commercial software package, Arena-flow, which is currently used by major U.S. companies like Dow Chemical, ExxonMobil, etc. to model particulate flows. Hence, the technology transfer is immediate. Code modifications resulting from this project have already been incorporated into the 2005 release of Arena-flow (i.e., these modifications are already available to companies using this software package).

This work is notable because it involves significant scientific progress, namely the identification of new physics responsible for the long-observed phenomena of segregation of unlike particles. This rapid progress is possible due to the cross-fertilization of researchers from a wide range of disciplines (chemical engineering, manufacturing engineering, physics, computer science, and nuclear engineering), sectors (industrial co-PI's from Lyondell Chemicals and Arena-Flow), as well as international collaborations (Loughborough University, UK). Such partnerships have opened up a host of opportunities for US students, including internships with both industrial partners and an overseas research stay with the international collaborator.

The project is high risk since it involves a fundamental approach to describing a system which heretofore has been described only by empirical correlations.

The work is also highly multidisciplinary. Partners include industrial investigators and international collaborators with backgrounds in chemical engineering, manufacturing engineering, physics, computer science, and nuclear engineering.


     
Program Officer:   Judy Raper
     
NSF Award Number:   0318999
     
Award Title:   GOALI: Toward a Fundamental Understanding of Elutriation in Fluidized Beds
     
PI Name:   Christine Hrenya
     
Institution Name:   University of Colorado at Boulder
     
Program Element:   1415
     

Top of Page


This Nugget was Updated on 20 November 2006.