Date of Award

Spring 1-1-2011

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Chemical & Biochemical Engineering

First Advisor

Christine M Hrenya

Second Advisor

Ray A Cocco

Third Advisor

Richard D Noble

Abstract

Industrial applications harnessing fluidized bed technology span energy production, chemical synthesis, and pharmaceutical processes, among others. However, operational challenges continue to plague such processes, in part due to the lack of predictive understanding on the effects of polydispersity, which is ubiquitous. An experimental investigation into the impact of polydispersity on various flow phenomena will lead to a better appreciation of the physics involved, contribute to validating models and thereby enhance the performance of fluidized bed systems.

The overall goal of this research is to gain a better understanding how polydispersity affects fluidized bed systems, with a focus on Geldart Group B particles. Systems ranging from low-velocity bubbling bed to high-velocity circulating fluidized bed (CFB) systems have been examined experimentally. The overriding question is: how do the different categories of polydispersity – namely binary mixtures and continuous distributions – affect species segregation, elutriation, bubbles (bubbling fluidized bed) and clusters (CFB riser)?

With regards to low-velocity bubbling fluidized beds, although investigation into species segregation behavior has been fairly extensive for binary mixtures, what is still missing is the impact of the width of continuous distributions on species segregation and bubbling behavior, and the side-by-side comparisons of the two behaviors, which are addressed in this work. Results surprisingly indicate (i) a non-monotonic trend (increase then decrease) in the extent of segregation with the widening of the distribution for continuous size distributions, and (ii) the existence of a bubble-less bottom layer for continuous distributions, the thickness of which is linked to the level of species segregation, thereby providing the physical link between species segregation and bubbling behaviors.

As for the moderately dense CFB riser, the lack of comprehensive datasets for monodisperse and polydisperse materials hampers modeling efforts. Hence, this portion of the thesis aims to collect extensive riser measurements spanning the entire riser under a range of operating conditions for six different solid materials, including both monodisperse and polydisperse. Interesting new findings include (i) the reversal of the traditional core-annulus profiles for the larger Group B particles, which is linked to the associated higher Stokes number, (ii) different trends in species segregation between binary mixtures and continuous distributions, and (iii) comprehensive profiling of cluster characteristics for all (monodisperse and polydisperse) materials, thereby illustrating the impact of local riser position, material type, and operating conditions on cluster behavior.

Pertaining to the dilute CFB riser, the impact of varying compositions of binary mixtures and widths of continuous distributions on the total and species elutriation, local mass flux and species segregation are examined. Qualitative differences between the two categories of polydispersity include (i) the total elutriation flux of binary mixtures increases with the composition of fines (Ut<Us) but not so for continuous PSDs and (ii) the elutriation flux of coarse particles (Ut>Us) depends non-monotonically on fines composition for binary mixtures but monotonically for continuous PSDs. These differences are explained by the increasing size disparity of continuous PSDs as the distribution width increases, while the size disparity remains constant in binary mixtures of varying compositions. A third qualitative difference is the monotonic decrease in mass % of coarse particles with riser height observed for continuous PSDs, and a non-monotonic behavior for binary mixtures.

The comprehensive experimental datasets and physical insights in this work are expected to be valuable in the validation of kinetic-theory based models and in the design of related systems.

Share

COinS