Novel Antifoam for Fermentation Processes: Fluorocarbon

As foaming appears as a problem in chemical and fermentation processes that inhibits .... Foam Fractionation of Protein with the Presence of Antifoam ...
0 downloads 0 Views 112KB Size
Langmuir 2005, 21, 8613-8619

8613

Articles Novel Antifoam for Fermentation Processes: Fluorocarbon-Hydrocarbon Hybrid Unsymmetrical Bolaform Surfactant Pinar C¸ alik,* Nazar Ileri, and Burak I. Erdinc¸ Department of Chemical Engineering, Industrial Biotechnology Laboratory, Middle East Technical University, 06531 Ankara, Turkey

Nihal Aydogan*,† and Muharrem Argun Department of Chemical Engineering, Hacettepe University, Beytepe Campus, 06800 Ankara, Turkey Received January 25, 2005. In Final Form: July 6, 2005 As foaming appears as a problem in chemical and fermentation processes that inhibits reactor performance, the eminence of a novel fluorocarbon-hydrocarbon unsymmetrical bolaform (FHUB: OH(CH2)11N+(C2H4)2(CH2)2(CF2)5CF3 I-) surfactant as an antifoaming agent as well as a foam-reducing agent was investigated and compared with other surfactants and a commercial antifoaming agent. The surface elasticity of FHUB was determined as 4 mN/m, indicating its high potential on thinning of the foam film. The interactions between FHUB and the microoganism were investigated in a model fermentation process related with an enzyme production by recombinant Escherichia coli, in V ) 3.0 dm3 bioreactor systems with VR ) 1.65 dm3 working volume at air inlet rate of Qo/VR ) 0.5 dm3 dm-3 min-1 and agitation rate of N ) 500 min-1 oxygen transfer conditions, at T ) 37 °C, pHo ) 7.2, and CFHUB ) 0 and 0.1 mM, in a glucose-based defined medium. As FHUB did not influence the metabolism, specific enzyme activity values obtained with and without FHUB were close to each other; however, because of the slight decrease in oxygen transfer coefficient, slightly lower volumetric enzyme activity and cell concentrations were obtained. However, when FHUB is compared with widely used silicon oil based Antifoam A, with the use of the FHUB, higher physical oxygen transfer coefficient (KLa) values are obtained. Moreover, as the amount required for the foam control is very low, minute changes in the working volume of the bioreactor were obtained indicating the high potential of the use of FHUB as an antifoaming agent as well as a foam-reducing agent.

Introduction Foams are gas-liquid dispersions (>95% gas) and are often generated during chemical and fermentation processes where intense agitation and aeration combined with the presence of surface-active species in the medium exist.1 Foam undergoes two main processes: (i) water drainage and (ii) lamella rupture. When the time scale of foam formation is considered, water drainage is the slow process compared to the lamella rupture.2 There are several factors affecting the stability of foam, such as electrostatic or steric interactions that stabilize the foam structure and van der Waals interactions which help to destabilize the foam.1,2 The studies related with the mechanism of foaming and antifoaming performance of surfactants reveal that the dynamic surface tension, in other words, the surface coverage surfactants at short time scales, and the orientation of the molecule at the air-water interface are important factors influencing the foam formation.3 More* Corresponding author of bioprocess engineering. E-mail: [email protected]. † Corresponding author of molecular design and surface science. E-mail: [email protected]. (1) Myers, D. Surfaces, Interfaces, and Colloids: Principles and Applications, 2nd ed.; Wiley: New York, 1999. (2) Pelton, R.; Flaherty, T. Polym. Int. 2003, 52, 479-485. (3) Colin, A.; Giermanska-Kahn, J.; Langevin, D.; Desbat, B. Langmuir 1997, 13, 2953-2959.

over, the surface viscoelasticity which is related with the surface excess concentration of surfactant has significant role on the stability of the foam formed.3,4 The configuration of surfactant molecules at the interface affects the area per molecule which leads to higher surface excess concentration and surface elasticity. The configuration of surfactant molecule also gives rise to longer time for the adsorption of surfactant molecules, since it requires more time for the orientation of the molecules in addition to the time required for the diffusion.4 For these reasons, it is indeed important to consider the possible configuration of surfactant molecule at the interface to obtain the desired property. Particularly in fermentation processes, most cell cultures produce foam-producing agents, that is, extracellular enzymes or proteins, carbohydrates, ol-keto acids, lipophilic biosurfactants, and extracellular pigments, as the target biomolecules that tend to transfer into stable foams in the bioreactors. Therefore, the presence of a foam layer is very common, particularly in aerobic fermentations, and foaming appears as a problem that inhibits bioreactor performance. The consequences are an increase in pressure drop and a decrease in gas flow rate, and foam escaping from the bioreactor can wet filters, which provides (4) Monteux, C.; Fuller, G. G.; Bergeron, V. J. Phys. Chem. B 2004, 108, 16473-16482.

10.1021/la050207b CCC: $30.25 © 2005 American Chemical Society Published on Web 08/18/2005

8614

C¸ alik et al.

Langmuir, Vol. 21, No. 19, 2005

avenues into the bioreactor for contaminating cells, and can cause blockage of the outlet gas lines. Moreover, a fraction of liquid phase and cell trapped in the foam represent a loss of bioreactor volume where conditions may not be favorable for metabolic activity. As absolute sterility is required in industrial production of biomolecules, foam in the bioreactor can either be controlled using a mechanical foam breaker or, more effectively, by the addition of a surface-active chemical agent. Nevertheless, foam-breaking surfactant chemicals tend to stabilize the bubble interface and alter the surface tension and, usually, lower physical oxygen transfer coefficient (KLa) values and reduce the bioreactor’s capacity to supply oxygen and can also inhibit the cell growth and the metabolism. In this context, besides the choice of an antifoam, an effective way of antifoam addition to ensure maximum distribution over the reaction surface, minimum addition, and minimum carryover is the problem enrolled in bioreactor engineering. In aerobic processes, oxygen is a key substrate, and because of its low solubility in aqueous solutions, a continuous transfer of oxygen from the gas phase to the liquid phase is critical for maintaining the oxidative metabolism of cells. Thus, to maintain the noncoalescing character of the fermentation medium and for high KLa values, new surfactants are indeed required. In this context, the present work reports on (i) the performance of a fluorocarbon-hydrocarbon unsymmetrical bolaform surfactant (FHUB: OH(CH2)11N+(C2H4)2(CH2)2(CF2)5CF3 I-) as an antifoaming agent as well as the foam-reducing agent, (ii) the comparison of FHUB’s performance with other surfactants, and (iii) the application of FHUB5 on an enzyme production, i.e., benzaldehyde lyase (BAL), by recombinant E. coli, the widely used microorganism in fermentation processes, in a defined medium and the variations in product distributions, and oxygen transfer characteristics of the bioprocess for the first time. Materials and Methods FHUB and Its Properties. 1,1,1,2,2,3,3,4,4,5,5,6,6-Tridecafluoro-8-iodooctane, 11-bromo-1-undecanol, diethylamine, trimethylamine, 11-bromoundecane, and DTAB (dodecyltrimethylammonium bromide) were purchased from Across (Belgium). The acetone, hexane, diethyl ether, and lithium bromide were purchased from Sigma (Germany). The hybrid surfactant (11-hydroxundecyl) tridecafluorooctane diethylammonium iodide (OH(CH2)11N+(C2H4)2(CH2)2(CF2)5CF3I- or FHUB)5 and HTAB (ω-hydroxyundecyltrimethyle ammonium bromide)6 were synthesized in our laboratory as described elsewhere. Equilibrium and Dynamic Surface Tensions. Aqueous surfactant solutions were prepared freshly for each experiment using water from a water purification system (Barnstead, U.S.). The equilibrium surface tensions of aqueous solutions of FHUB with and without the microorganisms inoculated into the fermentation medium were measured using a Wilhelmy plate method in a tensiometer (Kruss, Germany). Dynamic surface tension of FHUB, DTAB, and HTAB were measured using pendant drop method with a goinometer (DSA 10, Kruss, Germany). All surface tension measurements were repeated at least twice. All the glassware was cleaned in piranha solution (18 M H2SO4, 30% H2O2, 70:30 v/v). Foaming Performance. A conventional cylinder shake test was used to evaluate the foaming performance of surfactant/ antifoam mixtures. Foaming experiments were carried out in 100-mL volumetric cylinders using 15 mL of the foaming solution. The stoppered cylinder was shaken by hand 10 times, and the foam height and the defoaming time, which is the time for the (5) Aydogan, N.; Aldis, N.; Guvenir, O. Langmuir 2003, 19, 1072610731. (6) Aydogan, N.; Abbott, N. L. J. Colloid Interface Sci. 2001, 242, 411-418.

appearance of solution surface free of bubbles, were measured. Foaming performance of FHUB, HTAB, DTAB, and Antifoam A (A5758, Sigma Germany), which contains silicon oil and 10-40 µm solid particles, were investigated using this method. Surface Elasticity. Surface dilatational moduli measurements were performed using a modified version of the pendant drop method (DSA 10, Kruss, Germany). A drop is formed at the end of the tip of the needle and is kept in an environmental chamber to prevent evaporation and drop deformation. The drop deformation was controlled through a computer-controlled pumping system. The drop profile was obtained using the image analysis software which allows us to obtain the value of the surface tension as well as the drop surface area. From the periodic variation of surface tension and area with time, the dilatational surface moduli were deduced. The interfacial dilatational modulus is defined by the surface tension increase after a small increase in area of a surface element:

E)

dγ dA

where γ is the surface tension and A is the area of the surface element.4 As the surface area of the drop is oscillated periodically, the dilatational modulus exhibits two contributions: an elastic component, E′, accounting for the recoverable energy stored in the interface and a loss modulus, E′′, accounting for dissipation of energy through relaxation processes. The storage and loss moduli are the real and imaginary parts of the elasticity: ω ) E′ + iE′′.4,7 The surface elasticity measurements were performed using the oscillation frequency at 0.05 Hz which is the lower limit of our experimental system. The deformation at the area was set to 3% to prevent the nonlinear effects. Microorganism and Culture Maintenance. E. coli K-12 carrying pUC18::bal gene was used for benzaldehyde lyase (BAL: EC. 4.1.2.38) production. Cultures were maintained on LB agar slants that contained (kg m-3) yeast extract, 5.0; tryptone, 10.0; NaCl, 10.0; ampicillin, 0.1; and agar, 15. Cells from the newly prepared slants were inoculated into the preculture medium that contained (kg m-3) yeast extract, 5.0; tryptone, 10.0; NaCl, 10.0; and ampicillin, 0.1 and were incubated at 37 °C for 12 h. Batch-bioreactor experiments were conducted at V ) 3.0 dm3 at pHUC ) 7.2 and uncontrolled-pH operation at air inlet rate of QO/VR ) 0.5 vvm (Qo/VR ) 0.5 dm3 dm-3 min-1), and the agitation rate of N ) 500 min-1 in bioreactor systems (BBraun, Germany) consisted of temperature, pH, foam, air inlet, and stirring rate controls with VR )1.65 dm3 working volume using the medium containing (kg m-3) 8.0 glucose, 6.7 Na2HPO4, 3.1 KH2PO4, 0.5 NaCl, 0.5 MgSO4‚7H2O, 5 (NH4)2HPO4, and 0.1 ampicillin, and T ) 37 °C with the inoculation ratio IR ) 1/10. The cells were induced with 1 mM IPTG at t ) 4 h of the bioprocess.8 Analyses. Cell concentrations (CX) based on dry weights were determined with UV-spectrophotometer at 600 nm. After the lysis of the cell wall at f ) 10 s-1 for 10 min (Retsch, MM 200), the enzymatic activity of BAL was determined using 0.035 mM benzoin, 20 mM tris hydrochloride, 0.01 mM thiamin diphosphate (TPP), and 0.1 mM MgCl2 at 37 °C. Three milliliters of 0.035 mM benzoin was converted into benzaldehyde with 20 µL of diluted crude extract for 10 s. The absorbance was measured at 250 nm with a UV-spectrophotometer. One unit of BAL activity was defined as the activity which synthesizes one nanomole of benzaldehyde per second.8 Glucose concentration was measured with the DNS method.9 The dynamic method was applied10 to determine the oxygen uptake rate (OUR) and oxygen transfer coefficient KLa values during the cultivation times corresponding to the characteristic regions of the bioprocess within short durations in order to not effect the fermentation process. (7) Blank, M.; Lucassen, J.; van den Tempel, M. J. Colloid Interface Sci. 1970, 33, 94-100. (8) C¸ alik, P.; Yilgor, P.; Ayhan, P.; Demir, A. S. Chem. Eng. Sci. 2004, 59 (22-23), 5075-5083. (9) Miller, G. L. Anal. Chem. 1959, 31, 426-428. (10) Rainer, B. W. Chem. Biochem. Eng. 1990, 4, 185-196.

Novel Antifoam for Fermentation Processes

Langmuir, Vol. 21, No. 19, 2005 8615 Table 2. Performance of FHUB and Commercial Antifoam as an Antifoaming Agent

fermentation broth SDS

a

antifoam

foam height (mm)

none Antifoam A (0.3% v/v) FHUB (0.05 mM) none Antifoam A (0.3% v/v)a FHUB (0.01 mM)

60