Characterization of mixing in stirred tank bioreactors equipped with rushton turbines

Abstract

Microbial conversions involve multiphase systems with many different interactions, i.e., cellular reactions, gasliquid mass transfer, liquid mixing, and substrate diffusion into cell aggregates.6 A complete quantitative description of a bioprocess requires the fitting together of pieces of modelling, and the final model for the whole system may be very complex. In an attempt to reduce the interactions in the model, one normally studies the effect of different mechanisms separately. Thus, microbial kinetics is studied in high-performance bioreactors where effects of mass transfer and liquid mixing are as far as possible eliminated." Similarly, bioreactor performance, i.e., gas-liquid mass transfer and liquid mixing, is best studied separately using model media.The object of the present article is the characterization of mixing processes in small pilot plant bioreactors used in our laboratory for studies of the microbial kinetics of Penicillium chrysogenum and Aspergillus oryzae. The characterization is based on injection of the radioactive isotope 113mIn, followed by measurement of the radioactivity at different positions in the tank using scintillation counters placed outside the bioreactor. The bioreactor characterization is done with model media, i.e., water and a highviscosity xanthan solution, but the mixing process was also studied during a fed-batch penicillin production process. MATERIALS AND METHODS Tracer MethodAn isotope tracer technique is used for characterization of liquid mixing. The tracer-the metastable 113mIn (t,,2 = 99.5 min)-is the daughter isotope from the breakdown of Sn (t,,2 = 115 days), and it is degraded by isomeric transition to stable l131n by emission of monoenergetic y-radiation at 392 keV. The '13Tn is supplied by a 50-mCu t To whom all correspondence should be addressed. Denmark.Amersham generator consisting of a lead encapsulated ionexchange column on which the Il3Sn isotope is absorbed.113mIn is released by elution of the column with 0.04 M HC1-resulting in an aqueous solution of ionic '13?n.The method applies a traditional stimulus/response technique in which a small volume of the tracer (1-5 mL) is rapidly added to the tank, and the mixing process is continuously monitored by four scintillation detectors placed outside the tank. The y-radiation penetrates the tank walls, and the detectors are sensitive enough to permit a noninvasive measurement from outside the tank even if it is equipped with a heat-exchange jacket for temperature control. The volume "seen" by the scintillation detectors is controlled by placing these in a ring of lead. Hereby the measured radiation is from a cone, with its top point at the center of the scintillation crystal. ApparatusThe mixing experiments have been carried out in standard bioreactors with volumes ranging from 7 to 41 L. Also, many experiments were carried out in a model tank which is an in-house built copy of a commercial 15-L MBR bioreactor used as a standard reactor in our laboratory. The bioreactors are equipped with four standard baffle...

On the improvement of bacterial growth on complementary substrates by partial segregration in the broth

“…has described¯uorescence studies with a 30 dm 3 fermenter, and Namdev et al Fluid mixing is important in bioreactors because incomplete mixing generates non-uniform spatial distributions of cells, substrates and products. These non-uniformities can have profound effects on the relative growth rates of competing species, 6 the rates of oxygen uptake, 7 and the stability of the reactor.…”

Effect of tank size on kLa and mixing time in aerated stirred reactors with non‐newtonian fluids

“…Cooke et al (1988) used 0.4% CMC solutions to quantify k L a inside 4 m 3 bioreactors, and proposed a correlation to predict k L a , including a surprisingly strong effect of the viscosity: k L a ∝ µ −1 . Several studies have also been done with XG solutions (Pedersen et al, 1994; Garcia‐Ochoa and Gomez, 1998) inside small reactors of less than 25 L. Herbst et al (1992) studied XG production inside 0.2–3 m 3 reactors by oxygen mass balance and validated the correlation of Kawase and Moo‐Young (1988) at different scales. This is a very interesting study reports a strong effect of the choice of impellers but it covers a specific range of rheology and operating conditions, as it focuses on very high XG concentrations—up to 2%—and only one gas flow rate.…”

“…Fibre suspensions have also been used for their filamentous structure (Cooke et al, 1988; Dronawat et al, 1997; Benchapattarapong et al, 2005), despite possible difficulties in mimicking the real rheology of highly shear‐thinning filamentous fungi. Model fluids are also used for the study of mixing time and power consumption in viscous media (Luong and Volesky, 1979; Nienow and Elson, 1988; Henzler and Obernosterer, 1991; Pedersen et al, 1993, 1994; Nienow et al, 1996; Vlaev et al, 2002; Albaek et al, 2008). A detailed discussion on rheological measurements of mycelial broth and model media is proposed by various authors (Metz and Kossen, 1977; Metz et al, 1979; Kim et al, 1983; Allen and Robinson, 1990; Olsvik and Kristiansen, 1994; Svihla et al, 1997).…”

Experimental determination of the shear rate in a stirred tank with a non‐newtonian fluid: Carbopol

“…Much of the most important work on stirred tank hydrodynamics has been devoted to the choice of an appropriate impeller or combination of impellers to maximize mixing and mass transfer on a global scale . Because of the difficulties of studying biological media, model fluids have often been chosen as they are cheaper and easier to use.…”