No abstract

An alternative way for determining the oxygen mass transfer coef®cient, k L a, based upon the traditional dynamic method, is proposed. The oxygen material balance equation in the liquid phase is integrated after insertion of the oxygen probe response time (®rst order type), and k L a values are determined by employing Marquardt's algorithm, considering as a weighting factor the model's sensitivity with respect to k L a. Bench-scale fermentations of Aspergillus awamori, performed under different agitation (300±700 rpm) and aeration conditions (0.2±0.6 vvm), were utilized for calculating k L a values (0.0283±0.0874 s À1 ), employing three methods: two so-called traditional, the gas balancing and the dynamic methods, and the one proposed here. The latter method is shown to be as reliable as the aforementioned methods but is easier to apply when the oxygen level in the reactor is above the critical value. # 2000 Society of Chemical Industry Keywords: oxygen mass transfer coef®cient; oxygen uptake rate; electrode response time; dynamic methodDissolved oxygen concentration at instant t= t s in eqn (16) (mol m À3 ) C* Dissolved oxygen saturation concentration in liquid phase at gas±liquid interface (mol m À3 ) C 0Dissolved oxygen concentration at t= t 0 (mol m À3 ) C crit Critical dissolved oxygen concentration (mol m À3 ) C e Electrode signal (mol m À3 ) C e0Electrode signal at t= t 0 (mol m À3 ) C es Electrode signal on quasi steady state (mol m À3 ) C Ã ei ith experimental value of dependent variable C e (mol m À3 ) C ei ith calculated value of dependent variable C e obtained from eqns (9) and (10) (mol m À3 ) C s Dissolved oxygen concentration for quasi steady state (mol m À3 ) k e Electrode's sensitivity (s À1 ) k L a Oxygen mass transfer coef®cient (s À1 ) n Number of experimental values n in Inlet dry molar¯ow rate (mol s À1 ) n out Outlet dry molar¯ow rate (mol s À1 ) N Stirrer speed (rpm) Q O2 Speci®c respiration rate (mol kg À1 s À1 ) Q O2 X Molar microbial oxygen uptake rate (mol m À3 s À1 ) rpm Revolutions per minute (min À1 ) R 2 Correlation coef®cient SSR Sum of squares of residuals t Time (s) t 0 Initial time for integration of eqn (10) (s) t s Instant in which the aeration is re-started (eqn (16)) (s) vvm Air volume per liquid volume per minute (m 3 m À3 min À1 ) V Broth fermentation volume (m 3 ) Y Oin Oxygen molar fraction in the inlet gas stream Y Oout Oxygen molar fraction in the outlet gas stream t e Electrode response time (s À1 ) f Speci®c air¯ow rate (vvm or m 3 m À3 min À1

A torque meter has been developed for determining the power consumption in a bench stirred tank. The device has been bonded in the stirrer shaft inside a commercial bench fermentor, in order to avoid frictional losses in the mechanical seal. Power consumption measurements in ungassed and gassed systems were obtained at different agitation and aeration conditions, for Newtonian and non-Newtonian¯uids. Also, a``simple modi®ed sul®te method'' for volumetric oxygen transfer coef®cient (k L a) determination was developed and the experimental data were correlated with the gassed power (P g ) by using well-known correlations presented in the literature. List of symbols m kinematic viscosity of liquid (m 2 s )1 ) q liquid density (kg m )3 ) l liquid viscosity (Pa s) l app apparent viscosity of liquid (Pa s) Dt time period for the reaction of all sodium sul®te (s) D i impeller diameter (m) D t tank diameter (m) g acceleration of gravity (m s )2 ) HHenry's constant for the oxygen-water system (H = 0.8912 atm m 3 mol )1 ) K consistency index (Pa s n ) k L a volumetric oxygen transfer coef®cient (s )1 ) k L a calc calculated volumetric oxygen transfer coef®cient (s )1 ) k L a exp experimental volumetric oxygen transfer coef®cient (s )1 )power number (±) n s sodium sul®te number of moles OTR oxygen transfer rate (mol m )3 s )1 ) P Ã partial pressure of oxygen which is in equilibrium with bulk concentration in liquid phase (atm) P partial pressure of oxygen in gas phase in bulk (atm) P g power consumption of liquid agitation in gassed system (W) P o power consumption of liquid agitation in ungassed system (W) Q air¯ow rate (m 3 s )1 ) ReReynolds number (±) T torque (N m) V liquid volume (m 3 ) v s super®cial velocity of air (m s )1 ) IntroductionThe most utilized criteria for scale-up of aerobic fermentations, maintaining geometric similarity, are based on empirical or semiempirical equations which correlate the volumetric oxygen transfer coef®cient (k L a) and the gassed power per unit volume (P g /V). However, the consistency of such correlations is strongly dependent on the accuracy of both k L a and P g measurements. Several methods for power consumption measurements are proposed in the literature. In large scale plants, the use of wattmeters or ammeters allows a reasonable assessment of power uptake, but for small laboratory and even pilot plant fermentors, electrical measurements are too inaccurate. Two other adequate systems for measuring power consumption in stirred and aerated tanks are dynamometers [1] and strain-gages [2]. Oosterhuis and Kossen [3] developed a device for power measurements based on the heat balance on insulated vessels. For accurate power measurements in pilot scale, the use of a torsion dynamometer is recommended. Otherwise, on bench scale it is quite common to employ strain gage dynamometers attached on the agitator shaft [2]. Recently, Rese Ândiz et al. [4] proposed a new pneumatic bearing dynamometer for power measurements in stirred tanks aiming to reduce frictional losses in the bearings that ...

The morphology of Streptomyces olindensis (producer of retamycin, an antitumor antibiotic) grown in submerged culture was assessed by image analysis. The morphology was differentiated into four classes: pellets, clumps (or entangled filaments), branched and unbranched free filaments. Four morphological parameters were initially considered (area, convex area, perimeter, and convex perimeter) but only two parameters (perimeter and convex perimeter) were chosen to automatically classify the cells into the four morphological classes, using histogram analysis. Each morphological class was evaluated during growth carried out in liquid media in fermenter or shaker. It was found that pellets and clumps dominated in early growth stages in fermenter (due to the inoculum coming from a shaker cultivation) and that during cultivation, the breakage of pellets and clumps caused an increase in the percentage of free filaments. The criteria of morphological classification by image analysis proposed were useful to quantify the percentage of each morphological class during fermentations and may help to establish correlations between antibiotic production and microorganism morphology.

A morphologically structured model is proposed to describe trends in biomass growth, substrate consumption, and antitumoral retamycin production during batch and fed-batch cultivations of Streptomyces olindensis. Filamentous biomass is structured into three morphological compartments (apical, subapical, and hyphal), and the production of retamycin, a secondary metabolite, is assumed to take place in the subapical cell compartment. Model accounts for the effect of glucose as well as complex nitrogen source on both the biomass growth and retamycin production. Laboratory data from bench-scale batch and fed-batch fermentations were used to estimate some model parameters by nonlinear regression. The predictive capability of the model was then tested for additional fed-batch and continuous experiments not used in the previous fitting procedure. The model predictions show fair agreement to the experimental data. The proposed model can be useful for further studies on process optimization and control.