Factors affecting membrane fouling in filtration of Saccharomyces cerevisiae in an internal ceramic filter bioreactor

Abstract

Filtration of ethanol fermentation medium and broth by using symmetric and asymmetric ceramic membranes has been studied in an internal ®lter bioreactor. Factors studied included membrane structure and pore size, medium sterilization, and concentrations of glucose, yeast extract in the medium, yeast cell and protein in broth. The aim was to determine the main factors responsible for the decline in ®ltration performance during ethanol fermentation by Saccharomyces cerevisiae. Flux index (Fi) of a new concept has been developed to evaluate the degree of¯ux decline during the membrane fouling process. Fi was de®ned as the ratio of the membrane¯ux at certain ®ltration time (t t) to the initial (t A0)¯ux of pure water, not the initial (t +0)¯ux of the test¯uid. Flux with sterilized medium was approximately two-fold higher than that with unsterilized medium although the reason could not be explained clearly. Glucose, interaction between glucose and yeast extract, yeast cells, and proteins in fermentation broth were found to play an important part in membrane fouling. Fi of the symmetric membrane decreased to a less extent than that of the asymmetric membrane with increasing glucose concentration. But, the result with various yeast cell concentrations turned out to be contrary. Fouling was more serious for asymmetric membrane during the ®ltration of fermentation supernatant. This was thought to be due to different fouling mechanisms for the two types of membrane.

Studies on the interaction of fermentation and microfiltration operations: Erythromycin recovery fromSaccharopolyspora erythraea fermentation broths

“…The membrane used was a 0.2 m hydrophilic Durapore membrane. A hydrophilic membrane was chosen to minimize fouling by proteins (Zhang et al, 1998) and anti-foaming agents (Liew et al, 1997). A single-membrane plate was used with a membrane area of 6 × 10 −3 m 2 , which enabled processing of small amounts of broth (ca.…”

Modeling the influence of slurry concentration on Saccharomyces cerevisiae cake porosity and resistance during microfiltration

“…This is especially true because yeast is compressible, differ in shape, and can aggregate. Moreover, yeast properties are dependent on the cultivation conditions 2–8. Microscopic measurements show a ratio of maximal to minimal cell axis of ∼1.20–1.33 to 1.465–2.064 3, 4, 6, 7.…”

“…Other investigators used cultivated yeast cells, washed, and dispersed in isotonic saline (0.9% NaCl) solution 18, 19. Depending on the yeast technology, yeast concentrations during microfiltration vary over a wide range from 0.022 to 120 g/L5, 6, 12, 16–20 and ∼0.14 to 13 vol % 3, 8, 15. Some articles mention the use of a wide range of yeast concentrations but present results only for selected yeast concentration values thereby not allowing us to obtain a widely applicable correlation.…”

“…The highest concentration is below the critical yeast concentration of 70.4 g/L, at which, according to Klein et al,21 the slurry viscosity begins increasing exponentially thereby starting a structuring phenomenon. Moreover, Zhang et al5 investigated the factors affecting membrane fouling occurring in filtration of S. cerevisiae through a ceramic filter inside a bioreactor. The operation was conducted at a maximum cell concentration of 60 g/L, which was considered as a suitable and operable concentration in membrane‐based culture systems.…”

A continuous membrane bioreactor for ester synthesis in organic media: II. Modeling of MBR continuous operation

“…These systems are promising as they combine reaction and separation facilitating the downstream processing of the products (Vieth et al, 1976). Despite the significant number of applications that recently have been reported for membrane bioreactors (Doig et al, 1998;Hekmat et al, 1999;Kamoshita et al, 1998;Martin-Orue et al, 1999;Satory et al, 1997;Serralheiro et al, 1999;Zhang et al, 1998), little attention has been given to model these reactor systems. Moreover, the works dedicated to model membrane bioreactors focussed on kinetic modeling by transformation of Michaelis-Menten equations and assumed the presence of mass-transfer effects.…”