Process development and modeling of fluidized bed reactor with co-immobilized biocatalyst for fuel ethanol production

Author

May Yongmei Sun

Date of Award

12-1997

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Chemical Engineering

Major Professor

Paul R. Bienkowski

Committee Members

John W. Prados, Brian Davison, Paul D. Frymier, D.C. White

Abstract

This research focuses on two steps of commercial fuel ethanol production processes: the hydrolysis starch process and the fermentation process. The goal of this research is to evaluate the performance of co-immobilized biocatalysts in a fluidized bed reactor with emphasis on economic and engineering aspects and to develop a predictive mathematical model for this system The productivity of an FBR is higher than productivity of a traditional batch reactor or CSTR. Fluidized beds offer great advantages over packed beds for immobilized cells when small particles are used or when the reactant feed contains suspended solids. Plugging problems, excessive pressure drops (and thus attrition), or crushing risks may be avoided. No mechanical stirring is required as mixing occurs due to the natural turbulence in the fluidized process. Both enzyme and microorganism are immobilized in one catalyst bead which is called co-immobilization. Inside this biocatalyst matrix, starch is hydrolyzed by the enzyme glucoamylase to form glucose and then converted to ethanol and carbon dioxide by microorganisms. Two biocatalysts were evaluated : (1) co-immobilized yeast strain Saccharomyces cerevisiae and glucoamylase. (2) co-immobilized Zymomonas mobilis,/em> and glucoamylase. A co-immobilized biocatalyst accomplishes the simultaneous saccharification and fermentation (SSF process). When compared to a two-step process involving separate saccharification and fermentation stages, the SSF process has productivity values twice that given by the pre-saccharified process when the time required for pre-saccharification (15-25 h) was taken into account. The SSF process should also save capital cost. The information about productivity, fermentation yield, concentration profiles along the bed, ethanol inhibition, et al, was obtained from the experimental data. For the yeast system, experimental results showed that: no apparent decrease of productivity occurred after two and half months, the productivity was 25-44g/L-hr (based on reactor volume), the average yield was 0.45 g ethanol/g starch, the biocatalyst retained physical integrity and contamination did not affect fermentation. For the Z. mobilis system the maximum volumetric productivity was 38 g ethanol/L-h, the average yield was 0.51 g ethanol/g starch and the FBR was successfully operated for almost one month. In order to develop, scale-up and economically evaluate this system more efficiently, a predictive mathematical model that is based on fundamental principles was developed and verified. This model includes kinetics of reactions, transport phenomena of the reactant and product by diffusion within the biocatalyst bead, and the hydrodynamics of the three phase fluidized bed. The co-immobilized biocatalyst involves a consecutive reaction mechanism. The mathematical descriptions of the effectiveness factors of reactant and the intermediate product were developed. Hydrodynamic literature correlations were used to develop the dispersion coefficient and gas, liquid, and solid holdup. The solutions of coupled non-linear second order equations for biocatalyst bead and reactor together with the boundary conditions were solved numerically. This model gives considerable information about the system, such as concentration profiles inside both the beads and column, flow rate and feed concentration influences on productivity and phase hold up, and the influence of enzyme and cell mass loading in the catalyst. This model is generic in nature such that it can be easily applied to a diverse set of applications and operating conditions.