INGAR   05399
INSTITUTO DE DESARROLLO Y DISEÑO
Unidad Ejecutora - UE
artículos
Título:
Hydrodynamic aspects in fluidized bed bioreactor modeling
Autor/es:
M. FUENTES; N.J. SCENNA; P.A. AGUIRRE; M.C MUSSATI
Revista:
CHEMICAL ENGINEERING AND PROCESSING
Editorial:
Elsevier Science Ltd.
Referencias:
Año: 2008 vol. 47 p. 1530 - 1540
ISSN:
0255-2701
Resumen:
The main purpose of this paper is to analyze the adequacy of some hypotheses assumed in the literature for modeling mass transfer phenomena
and hydrodynamics in bioreactors. Four different hydrodynamic models were investigated to simulate the dynamic behavior of an anaerobic
fluidized bed reactor (AFBR). A total developed flow condition and the assumption of an incipient gas phase are some of the evaluated hypotheses.
All AFBR models simultaneously compute the dynamics of the phases and their components, including the effect of the biofilm growth in the
fluidization characteristics. From a computational point of view, ordinary and partial differential equation-based models were calculated using
gPROMS (Process System Enterprise Ltd.). Simulations based on a case study were compared. The bioreactor performance was analyzed through
the main variable profiles such as phase holdups and bed height, pH, chemical oxygen demand (COD), biofilm concentration and biogas flow
rate. In a previous paper [M. Fuentes, M.C. Mussati, N.J. Scenna, P.A. Aguirre, Global modeling and simulation of a three-phase fluidized bed
bioreactor, Comput. Chem. Eng., Ms. Ref. No.: 4281, submitted for publication], a heterogeneous model of a three-phase bioreactor system was
presented by proposing a one-dimensional (axial) dispersive model. Its results are here used to establish a reference point. For example, the fact
of considering a three-phase system with total developed flow (hydrodynamic pseudo-steady state) and complete mixture in all phases causes
deviations around 5% in predictions of biofilm concentration, and 0.5% in predictions of liquid and gas phase component concentrations, when
compared with results from the phenomenological dispersive model. However, predicted total COD removal efficiency is almost the same for both
models. Although the gas holdup is negligible when compared with the liquid and solid ones in anaerobic fluidized bed reactors, results from model
simplification assuming an incipient gas phase differ considerably from predictions based on original three-phase modeling.
© 2007 Elsevier B.V. All rights reserved.