Cultivation of E. coli in single- and ten-stage tower-loop reactors

E. Coli was cultivated in batch and continuous operations in the presence of an antifoam agent in stirred-tank and in single- and ten-stage airlift tower reactors with an outer loop. The maximum specific growth rate, mu(m), the substrate yield coefficient, Y(x/s), the respiratory quotient, RQ, substrate conversion, U(s), the volumetric mass transfer coefficient, K(L)a, the specific interfacial area, a, and the specific power input, P/V(L), were measured and compared. If a medium is used with a concentration of complex substrates (extracts) 2.5 times higher than that of glucose, a spectrum of C sources is available and cell regulation influences reactor performance. Both mu(m) and Y(X/S), which were evaluated in batch reactors, cannot be used for continuous reactors or, when measured in stirred-tank reactors, cannot be employed for tower-loop reactors: mu(m) is higher in the stirred-tank batch than in the tower-loop batch reactor, mu(m) and Y(x/s) are higher in the continuous reactor than in the batch single-stage tower-loop reactor. The performance of the single-stage is better than that of the ten-stage reactor due to the inefficient trays employed. A reduction of the medium recirculation rate reduces OTR, U(s), Pr, and Y(X/S) and causes cell sedimentation and flocculation. The volumetric mass transfer coefficient is reduced with increasing cultivation time; the Sauter bubble diameter, d(s), remains constant and does not depend on operational conditions. An increase in the medium recirculation rate reduces k(L)a. The specific power input, P/V(L), for the single-stage tower loop is much lower with the same k(L)a value than for a stirred tank. The relationship k(L)a vs. P/V(L) evaluated for model media in stirred tanks, can also be used for cultivations in these reactors.     

Estimation of oxygen mass transfer coefficient in stirred tank reactors using artificial neural networks

The estimation of volumetric mass transfer coefficient, k(L)a, in stirred tank reactors using artificial neural networks has been studied. Several operational conditions (N and V(s)), properties of fluid (μ(a)) and geometrical parameters (D and T) have been taken into account. Learning sets of input-output patterns were obtained by k(L)a experimental data in stirred tank reactors of different volumes. The inclusion of prior knowledge as an approach which improves the neural network prediction has been considered. The hybrid model combining a neural network together with an empirical equation provides a better representation of the estimated parameter values. The outputs predicted by the hybrid neural network are compared with experimental data and some correlations previously proposed in the literature for tanks of different sizes.     

Characterization of oxygen transfer in miniature and lab-scale bubble column bioreactors and comparison of microbial growth performance based on constant k(L)a

This work describes the engineering characterization of miniature (2 mL) and laboratory-scale (100 mL) bubble column bioreactors useful for the cultivation of microbial cells. These bioreactors were constructed of glass and used a range of sintered glass gas diffusers with differently sized pores to disperse humidified air within the liquid biomedium. The effect of the pressure of this supplied air on the breakthrough point for gas diffusers with different pore sizes was examined and could be predicted using the Laplace-Young equation. The influence of the superficial gas velocity (u(g)) on the volumetric mass transfer coefficient (k(L)a) was determined, and values of up to 0.09 s(-1) were observed in this work. Two modeling approaches were considered in order to predict and provide comparison criteria. The first related the volumetric power consumption (P/V) to the k(L)a and a good correlation was obtained for differently sized reactors with a given pore size, but this correlation was not satisfactory for bubble columns with different gas diffusers. Values for P/V ranged from about 10 to 400 W.m(-3). Second, a model was developed predicting bubble size (d(b)), bubble rising velocity (u(b)), gas hold-up (phi), liquid side mass transfer coefficient (k(L)), and thus the k(L)a using established theory and empirical correlations. Good agreement was found with our experimental data at different scales and pore sizes. Values for d(b) varied from 0.1 to 0.6 mm, and k(L) values between 1.7 and 9.8 x 10(-4) m.s(-1) were determined. Several E. coli cultivations were performed in the miniature bubble column at low and high k(L)a values, and the results were compared to those from a conventional stirred tank operated under identical k(L)a values. Results from the two systems were similar in terms of biomass growth rate and carbon source utilization.     

Hydrodynamics and mass transfer in Aspergillus niger fermentations in bubble column and loop bioreactors

The influence of Aspergillus niger broth rheology, bioreactor geometry, and superficial gas velocity on the volumetric liquid phase oxygen transfer coefficient (k(L)a(L)), riser gas holdup (epsilon(GR)), and circulating liquid velocity (u(LR)) was studied in a bubble column (BC) and two external-circulation-loop airlift (ECLAL) bioreactors. The results are compared to those of previous studies on homogeneous fluids and in particular with a recent study on non-Newtonian carboxymethylcellulose (CMC) solutions conducted in the same contactors used for the A. niger fermentations. As expected from the CMC-based studies, in the heterogeneous broths of A. niger epsilon(GR), k(L)a(L), and u(LR) decreased with increasing broth apparent viscosity; epsilon(GR) and k(L)a(L) decreased with increasing downcomer-to-riser cross-sectional area ratio, A(d)/A(r), whereas u(LR) increased with increasing A(d)/A(r). Gas holdup data in the airlift fermentations of A. niger were well predicted by the CMC-based correlation. However, the CMC-based correlations produced conservative estimations of k(L)a(L) and overestimates of u(LR) compared to the observed values in the A. niger broths.     

Characterization and feasibility of a miniaturized stirred tank bioreactor to perform E. coli high cell density fed-batch fermentations

The use of small scale bioreactors that are mechanically and functionally similar to large scale reactors is highly desirable to accelerate bioprocess development because they enable well-defined scale translations. In this study, a 25-mL miniaturized stirred tank bioreactor (MSBR) has been characterized in terms of its power input, hydrodynamics, and volumetric oxygen transfer coefficient (k(L)a) to assess its potential to grow high cell density (HCD) cultures using adequate scale-down criteria. Engineering characterization results show scale down, based on matched specific power input (P(G)/V), is feasible from a 20-L pilot scale stirred tank bioreactor. Results from fed-batch fermentations performed using Fab' producing E. coli W3110 at matched (P(G)/V) in the MSBR and 20-L STR demonstrated that the MSBR can accurately scale down the 20-L fermentation performance in terms of growth and Fab' production. Successful implementation of a fed-batch strategy in the MSBR resulted in maximum optical density of ca. 114 and total Fab' concentration of 940 μg/mL compared with ca. 118 and 990 μg/mL in 20-L STR. Furthermore, the use of the MSBR in conjunction with primary recovery scale-down tools to assess the harvest material of both reactors showed comparable shear sensitivity and centrifugation performance. The conjoint use of the MSBR with ultra scale-down (USD) centrifugation mimics can provide a cost-efficient manner in which to design and develop bioprocesses that account for good upstream performance as well as their manufacturability downstream.     

Influence of oxygen adsorption on the dynamic K(L)a measurement in three-phase slurry reactors

To check for possible mass transfer limitations of oxygen and/or carbon dioxide in kinetic experiments on microbial desulphurization of coal, it is important to properly measure the volumetric mass transfer coefficient (k(L)a) especially at high slurry densities. Volumetric mass transfer coefficients of oxygen, at different solid hold-up values (epsilon(s) = 0 to 0.28) of coal slurries (d(par) < 100 * 10(-6) m), were measured in a lab scale fermentor and in a lab scale pachuca tank, using the dynamic gas-liquid absorption method. It was shown that serious errors could occur due to oxygen adsorption at the coal surface. Using the data of an independently measured adsorption isotherm, the real k(L)a could be calculated from the measured apparent k(L)a. The results show a k(L)a decrease of 40% to 50% at a volumetric solid hold-up of 28%. Estimation of the oxygen and carbon dioxide transfer rates, from the measured mass transfer coefficients, indicates that the stirred fermentor is suitable for kinetic experiments at high slurry densities, whereas the pachuca tank and shake flask are not. (c) 1992 John Wiley & Sons, Inc.     

CFD simulation coupled with population balance equations for aerated stirred bioreactors

Mammalian cells have been widely used to produce therapeutic proteins in stirred bioreactors in suspension culture. Local hydrodynamics can have a great impact on cell proliferation and protein synthesis, but there are few reports on spatial heterogeneity of nutrients, gas bubbles, and mass transfer coefficients. We have employed computational fluid dynamics (CFD) coupled with population balance equations to study local hydrodynamics in a 20 L stirred bioreactor. The flow patterns, energy dissipation rates, gas volume fraction, gas bubble size distribution and local mass transfer coefficient have been displayed throughout the whole bioreactor. Their implications for mammalian cell culture have been discussed. This study provides an insight into rational design and optimum operation conditions in a stirred bioreactor for mammalian cell cultivation.     

Mass transfer phenomena in an airlift reactor: effects of solids loading and temperature

Gas holdups and volumetric mass transfer coefficients were measured in a concentric tube airlift reactor designed for the microbial desulfurization of coal. The solutions studied were comprised of an acidified basal salts solution containing thirteen different weight percentages (0 to 40) of coal (74 mum Ohio #1) at three different temperatures (30, 50, and 72 degrees C). Gas holdup epsilon(G) decreased with solids loading for the entire range studied. An enhancement in the volumetric mass transfer coefficient K(L)a with respect to that in pure solution was observed from zero to approximately 5 wt % (solids volume fraction epsilon(s) = 0.035), the maximum enhancement occurring at approximately 2 wt % (epsilon(s) = 0.014). At higher solids fractions, the mass transfer coefficient decreased with further solids additions. Gas holdups and the mass transfer coefficients increased with temperature over the studied range. The K(L)a and epsilon(G) were correlated to three process variables separately and the separate correlations combined to yield generalized correlations for the mass transfer coefficient and gas holdup for this system. The correlations may be used for design, operation, and ost estimation of such systems.     

Gas holdup and overall volumetric oxygen transfer coefficient in airlift contactors

The two major types of airlift contactors, concentric-tube and external-loop, were investigated for their gas holdup (riser and downcomer) and overall mass transfer characteristics. Results obtained in batch charges of tap water and 0.15 kmol/m(3) NaCl solution are reported for external-loop airlift contactors having downcomer-to-riser cross-sectional area ratios, A(d)/A(r), ranging from 0.11 相似文献   

CFD of mixing of multi‐phase flow in a bioreactor using population balance model

Mixing in bioreactors is known to be crucial for achieving efficient mass and heat transfer, both of which thereby impact not only growth of cells but also product quality. In a typical bioreactor, the rate of transport of oxygen from air is the limiting factor. While higher impeller speeds can enhance mixing, they can also cause severe cell damage. Hence, it is crucial to understand the hydrodynamics in a bioreactor to achieve optimal performance. This article presents a novel approach involving use of computational fluid dynamics (CFD) to model the hydrodynamics of an aerated stirred bioreactor for production of a monoclonal antibody therapeutic via mammalian cell culture. This is achieved by estimating the volume averaged mass transfer coefficient (k_La) under varying conditions of the process parameters. The process parameters that have been examined include the impeller rotational speed and the flow rate of the incoming gas through the sparger inlet. To undermine the two‐phase flow and turbulence, an Eulerian‐Eulerian multiphase model and k‐ε turbulence model have been used, respectively. These have further been coupled with population balance model to incorporate the various interphase interactions that lead to coalescence and breakage of bubbles. We have successfully demonstrated the utility of CFD as a tool to predict size distribution of bubbles as a function of process parameters and an efficient approach for obtaining optimized mixing conditions in the reactor. The proposed approach is significantly time and resource efficient when compared to the hit and trial, all experimental approach that is presently used. © 2016 American Institute of Chemical Engineers Biotechnol. Prog., 32:613–628, 2016     

Pulse radiolysis studies of beta-carotene in oxygenated DMSO solution. Formation of beta-carotene radical cation

The spectroscopic and kinetic characteristics of beta-carotene radical cation (beta-carotene(.+)) were studied by pulse radiolysis in aerated DMSO solution. The buildup of beta-carotene(.+) with k(1) = (4.8 +/- 0.2) x 10(8) dm(3) mol(-1) s(-1) [lambda(max) = 942 nm, epsilon = (1.6 +/- 0.1) x 10(4) dm(3) mol(-1) cm(-1)] results from an electron transfer from beta-carotene to DMSO(.+). The beta-carotene(.+) species decays exclusively by first-order reaction, k = (2.1 +/- 0.1) x 10(3) s(-1), probably by two processes: (1) at low substrate concentration by hydrolysis and (2) at high concentrations also by formation of dimer radical cation (beta-carotene)(2)(.+). Under the experimental conditions, a small additional beta-carotene triplet-state absorption ((3)beta-carotene) in the range of 525 to 660 nm was observed. This triplet absorption is quenched by oxygen (k = 7 x 10(4) s(-1)), resulting in singlet oxygen ((1)O(2)), whose reactions can also lead to additional formation of beta-carotene(.+).     

Gas transfer characteristics of a novel membrane bioreactor

We show the design features of a membrane bioreactor based on pulsatile flow across dimpled membranes. Results show an enhanced mass transfer of air of at least five-fold magnitude as compared with flat membranes. An increased working volume form 20 mL to 120 mL reduced the k(L)A at a given Reynolds number because of axial mixing of fluid from the deoxygenated end chamber. The bioreactor was used to supply air to a hybridoma mammalian cell line, and the calculated oxygen uptake showed that high-density cultures could be maintained in a 20mL, single-dimpled cultures could be maintained in a 20 mL, single-dimpled membrane system. Indirect aeration of a 2 L continuous stirred tank reactor, by a double-membrane system, showed that air could be supplied to mammalian cells at cell densities of approximately 4 x 10(6) /mL.     

Effect of biologically produced sulfur on gas absorption in a biotechnological hydrogen sulfide removal process

Absorption of hydrogen sulfide in aqueous suspensions of biologically produced sulfur particles was studied in a batch stirred cell reactor, and in a continuous set-up, consisting of a lab-scale gas absorber column and a bioreactor. Presence of biosulfur particles was found to enhance the absorption rate of H(2)S gas in the mildly alkaline liquid. The mechanism for this enhancement was however found to depend on the type of particles used. In the gently stirred cell reactor only small hydrophilic particles were present (d(p) < 3 microm) and the enhancement of the H(2)S absorption rate can be explained from the heterogeneous reaction between dissolved H(2)S and solid elemental sulfur to polysulfide ions, S(x) (2-). Conditions favoring enhanced H(2)S absorption for these hydrophilic particles are: low liquid side mass transfer (k(L)), high sulfur content, and presence of polysulfide ions. In the set-up of gas absorber column and bioreactor, both small hydrophilic particles and larger, more hydrophobic particles were continuously produced (d(p) up to 20 microm). Here, observed enhancement could not be explained by the heterogeneous reaction between sulfide and sulfur, due to the relatively low specific particle surface area, high k(L), and low [S(x) (2-)]. A more likely explanation for enhancement here is the more hydrophobic behavior of the larger particles. A local increase of the hydrophobic sulfur particle concentration near the gas/liquid interface and specific adsorption of H(2)S at the particle surface can result in an increase in the H(2)S absorption rate.     

Scale up of a viscous fungal fermentation: application of scale-up criteria with regime analysis and operating boundary conditions

The scale up of the novel, pharmaceutically important pneumocandin (B(0)), from the filamentous fungus Glarea lozoyensis was successfully completed from pilot scale (0.07, 0.8, and 19 m(3)) to production scale (57 m(3)). This was accomplished, despite dissimilar reactor geometry, employing a combination of scale-up criteria, process sensitivity studies, and regime analysis using characteristic time constants for both oxygen mass transfer and bulk mixing. Dissolved oxygen tension, separated from the influence of agitation by gas blending at the 0.07 m(3)-scale, had a marked influence on the concentrations of pneumocandin analogs with different levels of hydroxylation, and these concentrations were used as an indicator of bulk mixing upon scale up. The profound impact of dissolved oxygen tension (DOT) (low and high levels) on analog formation dictated the use of constant DOT, at 80% air saturation, as a scale-up criterion. As a result k(L)a, Oxygen uptake rate (OUR) and hence the OTR were held constant, which were effectively conserved across the scales, while the use of other criterion such as P(g)/V(L), or mixing time were less effective. Production scale (57 m(3)) mixing times were found to be faster than those at 19 m(3) due to a difference in liquid height/tank diameter ratio (H(L)/D(T)). Regime analysis at 19 and 57 m(3) for bulk mixing (t(c)) and oxygen transfer (1/k(L)a) showed that oxygen transfer was the rate-limiting step for this highly shear thinning fermentation, providing additional support for the choice of scale-up criterion.     

Combined effect of hydrodynamic and interfacial flow parameters on lysozyme deactivation in a stirred tank bioreactor

The dynamic environment within a bioreactor and in the purification equipment is known to affect the activity and yield of enzyme production. The present research focuses on the effect of hydrodynamic flow parameters (average energy dissipation rate, maximum energy dissipation rate, average shear rate, and average normal stress) and the interfacial flow parameters (specific interfacial area and mass transfer coefficient) on the activity of lysozyme. Flow parameters were estimated using CFD simulation based on the k-epsilon approach. Enzyme deactivation was investigated in 0.1, 0.3, 0.57, and 1 m i.d. vessels. Enzyme solution was subjected to hydrodynamic stress using various types of impellers and impeller combinations over a wide range of power consumption (0.03 < P(G)/V < 7, kW/m3). The effects of tank diameter, impeller diameter, blade width, blade angle, and the number of blades on the extent of deactivation were investigated. At equal value of P(G)/V, epsilon(max), and gamma(avg), the extent of deactivation was dramatically different for different impeller types. The extent of deactivation was found to correlate well with the average turbulent normal stress and the mass transfer coefficient.     

Protein kinase C(mu) regulation of the JNK pathway is triggered via phosphoinositide-dependent kinase 1 and protein kinase C(epsilon)

The protein kinase C (PKC)-related enzyme PKC(mu)/PKD (protein kinase D) is activated by activation loop phosphorylation through PKC(eta). Here we demonstrate that PKC(mu) is activated by the direct phosphorylation of PKC(epsilon). PKC(mu) colocalizes with PKC(epsilon) in HEK293 and MCF7 cells as shown by confocal immunofluorescence analyses. PDK1, known as the upstream kinase for several PKC isozymes, associates intracellularly with PKC(epsilon) and PKC(eta). PKC(eta) is phosphorylated by PDK1 in vitro, leading to kinase activation as similarly reported for PKC(epsilon) activation by PDK1. Coexpression of PDK1, PKC(epsilon) and PKC(mu) in HEK293 cells results in PKC(mu) activation. In contrast, the coexpression of PDK1 and PKC(eta) with PKC(mu) does not activate PKC(eta) or consequently PKC(mu). PDK1/PKC(epsilon)-triggered activation of PKC(mu) inhibits JNK, a downstream effector of PKC(mu), whereas upon transient expression of PDK1, PKC(eta), and PKC(mu), JNK is not affected. These data implicate PKC(epsilon) as the biologically important upstream kinase for PKC(mu) in HEK293 cells, regulating downstream effectors. Our results further indicate a PDK1/PKC(eta)/PKC(mu) controlled negative regulation of PKC(eta) kinase activity. In this study, we show that differentially activated kinase cascades involving PDK1 and novel PKC isotypes are responsible for the regulation of PKC(mu) activity and consequently inhibit the JNK pathway.     

Site-specific 1,N6-ethenoadenylated single-stranded oligonucleotides as structural probes for the T4 gene 32 protein-ssDNA complex

Bacteriophage T4 gene 32 protein (g32P) is a DNA replication accessory protein that binds single-stranded (ss) nucleic acids nonspecifically, independent of nucleotide sequence. G32P contains 1 mol of Zn(II)/mol of protein monomer, which can be substituted with Co(II), with maintenance of the structure and activity of the molecule. The Co(II) is coordinated via approximately tetrahedral ligand symmetry by three Cys sulfur atoms and therefore exhibits intense S(-)----Co(II) ligand to metal charge-transfer (LMCT) transitions in the near ultraviolet [Giedroc, D. P., et al. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 8452-8456]. A series of fluorescent 1,N6-ethenoadenosine (epsilon A)-containing oligonucleotides conforming to the structure (5'----3') d[(Tp)m epsilon A(pT)l-m-1] where 0 less than or equal to m less than or equal to l - 1 and length (l) six or eight nucleotides have been evaluated as dynamics probes and potential fluorescence energy transfer donors to Co(II) in mapping the spatial proximity of the (fixed) intrinsic metal ion and a variably positioned epsilon A-base in a series of protein-nucleic acid complexes. We provide spectroscopic evidence that the epsilon A-oligonucleotides bind to g32P-(A + B) with a fixed polarity of the phosphodiester chain. A Trp side chain(s) makes close approach to a epsilon A base positioned toward the 3' end of a bound l = 8 oligonucleotide. Six oligonucleotides of l = 8 and m = 0, 1, 3, 5, 6, or 7 were investigated as energy transfer donors to Co(II) at 0.1 M NaCl, pH 8.1, 25 degrees C upon binding to Co(II)-substituted or Zn(II) g32P-(A + B), i.e., in the presence and absence of an energy acceptor, respectively. Detectable quenching of the epsilon A-fluorescence by the Co(II)-LMCT acceptors was found to occur in all epsilon A-oligonucleotide-protein complexes, yielding energy transfer efficiencies (E) of 0.43, 0.31, 0.26, 0.26, 0.28, and 0.41 for l = 8 and m = 0, 1, 3, 5, 6, and 7 epsilon A-oligonucleotides, respectively. The two-dimensional distances R (in A) were found to vary as follows: d[epsilon A(pT)7] (m = 0), 16.0 (15.5-16.9); d[Tp epsilon A(pT)6] (m = 1), 17.7 (16.9-19.1); d[(Tp)3 epsilon A(pT)4] (m = 3), 20.7 (19.5-22.1); d[(Tp)5 epsilon A(pT)2] (m = 5), 20.5 (19.5-21.9); d[(Tp)6 epsilon ApT] (m = 6), 19.0 (18.0-20.4); and d[(Tp)7 epsilon A] (m = 7), 18.6 (17.8-19.8).(ABSTRACT TRUNCATED AT 400 WORDS)     

In vivo mechanical properties of leukocytes during adhesion to venular endothelium

Transient deformations of leukocytes (WBCs) were studied during their saltation along post-capillary venous endothelium (EC) in mesentery of the rat. During intermittent adhesion of WBCs to EC, prevailing fluid shear stresses, tau wall, resulted in a stepwise loading of the WBC upon attachment with a transient increase in length, L(t), and reduction in height, H(t). Measurements of L(t) and H(t) from frame-by-frame analysis of video recordings were modelled as the simple shear of a standard linear viscoelastic solid to facilitate calculation of the elastic (k1, k2) and viscous (mu) elements with k1 in parallel with serial elements k2 and mu. The magnitude of tau wall was determined from measurements of red cell velocity within the venule. During the spontaneous adhesion of WBCs, a value of cell viscosity (mu) of 45 Poise was determined. Stimulating adhesion by topical application of the chemoattractant FMLP resulted in a 15-fold increase of mu to 668 Poise. Transient deformations during topical application of cytochalesin B to disrupt actin fibers within the WBC, yielded a 40% reduction in k1, compared to an 80% reduction with colchicine which disrupts the microtubule structure. Thus, colchicine treated cells appear to be twice as deformable as cells treated with cytochalesin. During adhesion stimulated by the cytokine Interleukin-1, mu increased 50% without changes in k1 and k2, possibly due to slight activation of the WBC.     

Use of computational fluid dynamics as a tool for establishing process design space for mixing in a bioreactor

The concept of "design space" plays an integral part in implementation of quality by design for pharmaceutical products. ICH Q8 defines design space as "the multidimensional combination and interaction of input variables (e.g., material attributes) and process parameters that have been demonstrated to provide assurance of quality. Working within the design space is not considered as a change. Movement out of the design space is considered to be a change and would normally initiate a regulatory post-approval change process. Design space is proposed by the applicant and is subject to regulatory assessment and approval." Computational fluid dynamics (CFD) is increasingly being used as a tool for modeling of hydrodynamics and mass transfer. In this study, a laboratory-scale aerated bioreactor is modeled using CFD. Eulerian-Eulerian multiphase model is used along with dispersed k-ε turbulent model. Population balance model is incorporated to account for bubble breakage and coalescence. Multiple reference frame model is used for the rotating region. We demonstrate the usefulness of CFD modeling for evaluating the effects of typical process parameters like impeller speed, gas flow rate, and liquid height on the mass transfer coefficient (k(L)a). Design of experiments is utilized to establish a design space for the above mentioned parameters for a given permissible range of k(L)a.     

Optimization of oxygen mass transfer in a multiphase bioreactor with perfluorodecalin as a second liquid phase

Oxygenation is an important parameter involved in the design and operation of mixing-sparging bioreactors and it can be analyzed by means of the oxygen mass transfer coefficient (k(L)a). The operational conditions of a stirred, submerged aerated 2-L bioreactor have been optimized by studying the influence of a second liquid phase with higher oxygen affinity (perfluorodecalin or olive oil) in the k(L)a. Using k(L)a measurements, the influence of the following parameters on the oxygen transfer rate was evaluated: the volume of working medium, the type of impellers and their position, the organic phase concentration, the aqueous phase composition, and the concentration of inactive biomass. This study shows that the best experimental conditions were achieved with a perfluorodecalin volume fraction of 0.20, mixing using two Rushton turbines with six vertical blades and in the presence of YPD medium as the aqueous phase, with a k(L)a value of 64.6 h(-1). The addition of 20% of perfluorodecalin in these conditions provided a k(L)a enhancement of 25% when pure water was the aqueous phase and a 230% enhancement when YPD medium was used in comparison to their respective controls (no perfluorodecalin). Furthermore it is shown that the presence of olive oil as a second liquid phase is not beneficial to the oxygen transfer rate enhancement, leading to a decrease in the k(L)a values for all the concentrations studied. It was also observed that the magnitude of the enhancement of the k(L)a values by perfluorodecalin depends on the biomass concentration present.