Liquid circulation in external-loop airlift bioreactors

A simple model for prediction of liquid velocity in external-loop airlift bioreactors has been developed. Theoretical correlations for friction factor of gas-non-Newtonian two-phase flows and for liquid velocity in the riser were derived using the concept of an eddy diffusivity. The predictions of the proposed model were compared with the available experimental data for the friction factor and the liquid velocity in the riser of external-loop airlift contactors. Satisfactory agreement was obtained.     

The specific interfacial area in external-loop airlift bioreactors

The specific interfacial areas in two external-loop airlift bioreactors of laboratory and pilot scale were determined, mainly by the chemical reaction method (sulphite oxidation). The parameter studied in water/salt and starch/salt solutions was greately affected by gas superficial velocity, A _D /A _R ratio, by H _R ?H _D /H _D ratio and η _ap , respectively. Correlations for the specific interfacial area in the two systems, considering the effects of the above-mentioned parameters, were proposed.     

Hydrodynamics of non-Newtonian liquids in external-loop airlift bioreactors

In order to obtain further information on the behaviour and optimal design of external-circulation-loop airlift bioreactors, the liquid circulating velocity was studied using highly viscous pseudoplastic solutions of starch and antibiotic biosynthesis liquids of Penicillium chrysogenum, Streptomyces griseus, Streptomyces erythreus, Bacillus licheniformis and Cephalosporium acremonium. Measurements of liquid circulation velocity were made in laboratory and pilot plant external-loop airlift bioreactors, under various conditions concerning gas flow rate, riser liquid height at constant downcomer height, A _D /A _R ratio, using the impulse-response technique. It has been found that these parameters had a significant effect on liquid circulation velocity together with the apparent viscosity and dry weight of the solid phase in the biosynthesis liquids. For the tested liquids, the superficial liquid velocity in the riser section of an external-loop airlift bioreactor may be described by the following equation: where the exponents and the constant c take different values depending on the liquid phase properties and flow regime.     

Hydrodynamics in external-loop airlift bioreactors with static mixers

Liquid circulation superficial velocity and gas holdup behaviours were investigated in an external-loop airlift bioreactor of 0.170?m³ liquid volume in gas-induced and forced-circulation-loop operation modes, in the presence of static mixers made of corrugated stainless steel pieces, resulting in packets with the height-to-diameter ratio equal to unity and using non-Newtonian starch solutions as liquid phase. The static mixers were disposed in the riser in three blocks, each with three mixing packets, successively turned 90° to the adjacent mixing element. It was found that in the presence of static mixers and forced-loop operation mode, liquid circulation superficial velocity in the riser section was significantly diminished, while gas holdup increased in a great measure. It was considered that static mixers split the fluid into individual streams and break up the bubbles, resulting in small bubble sizes with a relative homogeneous bubble distribution over riser cross section. They act as supplementary resistances in liquid flow, reducing riser cross sectional area, equivalent with A _D /A _R area ratio diminishing.     

Hydrodynamics of non-Newtonian liquids in external-loop airlift bioreactors

Gas holdup investigation was performed in two external-loop airlift bioreactors of laboratory (V _L =1.189·10^?3? 1.880·10^?3 m³; H _R =1.16 ? 1.56 m; H _D = 1.10 m; A _D /A _R = 0.111 ? 1.000) and pilot scale (V _L =0.157?0.170 m³; H _R =4.3?4.7 m; H _D =4.0?4.4 m;A _D /A _R =0.04?0.1225), respectively, using as liquid phase non-Newtonian starch solutions of different concentration with K=0.061?3.518 Pa sⁿ and n=0.86?0.39 and fermentation broths of P. chrysogenum, S. griseus, S. erythreus, B. licheniformis and C. acremonium at different hours since inoculation and from different batches. The influence of bioreactor geometry, liquid properties and the amount of introduced compressed air was investigated. The effect of sparger design on gas holdup was found to be negligible. It was found that gas holdup depends on the flow media index, ?_GR decreasing with the increase of liquid pseudoplasticity, A _D /A _R ratio and H _R /H _D ratio. The experimental data are in agreement with those presented in literature by Popovic and Robinson, which take into account liquid properties, geometric parameters and gas superficial velocity, with a maximum error of ±30%. It was obtained a correlation for gas holdup estimation taking into account the non-Newtonian behaviour of the fermentation broths and the dry weight of the solid phase, as well. The concordance between the experimental data and those calculated with the proposed correlation was good, with a maximum error of ±17%. Also, a dimensionless correlation for gas holdup involving superficial velocities of gas and liquid, cross sectional areas ratio, dispersion height to riser diameter ratio, as well as Froude and Morton numbers, was obtained.     

Stochastic modelling of axial dispersion in external-loop airlift bioreactors

The paper presents a model of the motion of a particle subjected to several transport processes in connection with mixing in two phase flow. A residence time distribution technique coupled with a one-dimensional dispersion model was used to obtain the axial dispersion coefficient in the liquid phase, D_ax. The proposed model of D_ax for an external-loop airlift bioreactor is based on the stochastic analysis of the two-phase flow in a cocurrent bubble column and modified for the specific flow in the airlift reactor. The model takes into account the riser gas superficial velocity, the riser liquid superficial velocity, the Sauter bubble diameter, the riser gas hold-up, the downcomer-to-riser cross sectional area ratio. The proposed model can be applied with an average error of ᆨ.     

Study of the liquid circulation velocity in external-loop airlift bioreactors

Liquid circulation velocity was studied in externalloop air-lift bioreactors of laboratory and pilot scale, respectively for different gas input rates, downcomer-to-riser cross-sectional area ratio, A _D/A_R and liquid phase apparent viscosities.It was found that, up to a gas superficial velocity in the riser v _SGR 0.04 m/s the dependency of v _SLR on v _SGR is in the following form: v _SLR = a v _SGR ^b , with the exponent b being 0.40. Over this value of v _SGR, only a small increase in liquid superficial velocity, v _SLR is produced by an increase in v _SGR. A _D/A_R ratio affects the liquid superficial velocity due to the resistance in flow and overall friction.For non-Newtonian viscous liquids, the circulation liquid velocity in the riser section of the pilot external-loop airlift bioreactor is shown to be dependent mainly on the downcomer-to-riser cross-sectional area ratio, A _D/A_R, the effective (apparent) liquid viscosity, _eff and the superficial gas velocity, v _SGR.The equation proposed by Popovic and Robinson [11] was fitted well, with an error of ± 20%.List of Symbols A _D m² downcomer cross-sectional area - A _Rm² riser cross-sectional area - a = coefficient in Eq. (7) - b = exponent in Eq. (7) - c _s m^–1 Coefficient in Eq. (3) - D _D m downcomer diameter - D _R m riser diameter - g m²/s gravitational acceleration - H _D m dispersion height - H _L m ungassed liquid height - K Pa s ⁿ consistency index - K _B = friction factor at the bioreactor bottom - K _F = friction factor - K _T = friction factor at the bioreactor top - V _L m³ liquid volume in the bioreactor - V _D m³ liquid volume in downcomer - V _R m³ liquid volume in riser - v _LDm/s downcomer linear liquid velocity - v _LR m/s riser linear liquid velocity - v _SGR m/s riser superficial liquid velocity - v _SLR m/s riser superficial liquid velocity - s^–1 shear rate - _GD = downcomer gas holdup - _GR = riser gas holdup - _eff Pa s effective (apparent) viscosity - Pa shear stress The authors wish to thank Mrs. Rodica Roman for the help in experimental data collection and to Dr. Stefanluca for the financial support.     

Gas hold-up measurements in bioreactors

The gas hold-up of a gas-liquid dispersion is an important parameter in the fermentation industry. If it is too low, or too high, productivity can be adversely affected. Gas hold-up in fermentors cannot be calculated from physico-chemical correlations and, therefore, must be measured accurately for each fermentation.This article surveys a number of methods for measuring the gas hold-up in gas-liquid dispersions, making particular note whether these methods can be applied aseptically.     

Gas hold-up in converging-diverging tube airlift fermenter

Summary Fractional gas holdup study was carried out in two airlift fermenters: one having of conventional design, the other having an asymetric riser arm. Air flow rate was varied from 1.5 to 9.0 cm/sec and gas hold-up values compared. Fractional gas holdup, _G, was strongly dependent on superficial gas velocity and initial liquid height. The modified fermenter always showed a higher gas holdup than the conventionally designed one.Symbols ALF Airlift Fermenter - CDT Convergent-divergent Tube - UT Uniform Tube - U_G Superficial gas velocity, cm/s - h_i Initial liquid height in riser, cm - H_i Dispersed liquid height in riser, cm - H_O Dispersed liquid height in downcomer, cm - K,m,n Constant - a,a Constant - A_d Riser cross sectional area, cm^z - A_r Downcomer cross sectional area, cm^z - U_b Bubble rise velocity, cm/s - g Acceleration due to gravity, cm/s^z - d_B Bubble diameter, cm - R_ev Bubble's Reynolds number, dimensionless Greek Letters _G Fractional gas holdup, dimensionless - {ITG9}{INL} Liquid density, g/cc - {IT}{INL} Liquid viscosity, poise(g/cm.s) - {ITGS}{INL} Liquid surface tension, dyne/cm - porous plate pore diameter, cm     

Effects of downcomer-to-riser cross sectional area ratio on operation behaviour of external-loop airlift bioreactors

Experiments performed in two external-loop airlift bioreactors of laboratory and pilot scale, (1.880–1.189) · 10^–3 m³ and (0.170-0.157)m³, respectively, are reported. The A _D /A _R ratio was varied between 0.111–1.000 and 0.040–0.1225 in the laboratory and pilot contractor respectively.Water and solutions of different coalescence (2-propanol 2% vol, 1 M Na (glucose 50% wt/vol) and rheological behaviour (non-Newtonian starch solutions with consistency index K=0.061–3.518 Pas ⁿ and flow behaviour index n=0.86-0.39), respectively, were used as liquid phase. Compressed air at superficial velocities v _SGR =0.016–0.178 ms^–1 in the laboratory contactor and v _SGR =0.010–0.120 ms^–1 in the pilot contactor, respectively was used as gaseous phase.The A _D /A _R ratio affect gas-holdup behaviour as a result of the influence of A _D /A _R on liquid circulation velocity.Experimental results show that A _D /A _R ratio affect circulation liquid velocity by modifying he resistence to flow and by varying the fraction of the total volume contained in downcomer and riser. A _D /A _R ratio has proven to be the main factor which determines the friction in the reactor. Mixing time increases with increasing of the reactor size and decreases with A _D /A _R decreasing.The volumetric gas-liquid mass transfer coefficient increases with A _D /A _R ratio decreasing, as a result of variations of the liquid velocity with A _D /A _R , which affect interfacial areas.Correlations applicable to the investigated contactors have been presented, together with the fit of some experimental data to existing correlation in literature.List of Symbols A _D downcomer cross sectional area (m²) - A _R riser cross sectional area (m²) - a coefficient in Eq. (9) (-) - a _L gas-liquid interfacial area per unit volume (m^–1) - b coefficient in Eq. (9) (-) - C tracer concentration (kg m^–3) - C tracer concentration at the state of complete mixing (kg m^–3) - c coefficient in Eq. (12) - c _S coefficient in Eq. (5) - D _D downcomer diameter (m) - D _R riser diameter (m) - d _B bubble size (m) - H _D downcomer height (m) - H _d dispersion height (m) - H _L gas-free liquid height (m) - H _R riser height (m) - I inhomogeneity (-) - K consistency index (Pa s ⁿ ) - k _L a volumetric gas-liquid oxygen mass transfer coefficient (s^–1) - m exponent in Eq. (12) (-) - n flow behaviour index (-) - P _G power input due to gassing (W) - t _M mixing time (s) - V _A connecting pipe volume (m³) - V _D downcomer volume (m³) - V _d volume of dispersion (m³) - V _R riser volume (m³) - V _T total reactor liquid volume (m³) - v _SGR riser gas superficial velocity (m s^–1) - _GR riser gas holdup (-) - shear rate (m s^–1) - _app apparent viscosity (Pa s) - shear stress     

Concentric-tube airlift bioreactors

Mass transfer coefficients were measured in three concentric-tube airlift reactors of different scales (RIMP, V _L =0.07 m³;RIS?1,V _L =2.50 m³;RIS?2, V _L =5.20 m³). The effects of top and bottom clearance and flow resistances at downcorner entrance were studied in water-air system. Experimental results show that h _s ,h _B and A _d /A _R ratio affect K _L a values as a result of their influence on gas holdup and liquid velocity. The gas-liquid mass-transfer coefficients for all the geometric variables were successfully correlated as Sherwood number with Froude and Galilei numbers, the bottom spatial ratio (B=h _B /D _R ), the top spatial ratio , the gas separation ratio and the downcomer flow resistance ratio (R=A _d /A _R ). The proposed empirical model satisfactorily fitted the experimental data obtained in large airlift reactors and some data presented in literature.     

Concentric-tube airlift bioreactors

Gas holdup investigations were performed in three concentric-tube airlift reactors of different scales of operation (RIMP: 0.070 m³; RIS-1: 2.5 m³; RIS-2: 5.2 m³; nominal volumes). The influences of the top and bottom clearances and the flow resistances at the downcomer entrance were studied using tap water as liquid phase and air as gaseous phase, at atmospheric pressure. It was found that the gas holdup in the individual zone of the reactor: riser, downcomer and gas-separator, as well as that in the overall reactor is affected by the analyzed geometrical parameters in different ways, depending on their effects on liquid circulation velocity. Gas holdup was satisfactorily correlated with Fr, Ga, bottom spatial ratio (B), top spatial ratio (T), gas separation ratio (Y) and downcomer flow resistance ratio (A _d /A _R ). Correlations are presented for gas holdup in riser, downcomer, gas separator and for the total gas holdup in the reactor. All the above stressed the importance of the geometry in dynamic behaviour of airlift reactors.     

Concentric-tube airlift bioreactors

Liquid circulation velocity was investigated in three concentric-tube airlift reactors of different scales (RIMP, V _L =0.07 m³; RIS-1, V _L =2.5 m³; RIS-2, V _L =5.20 m³). The effects of top and bottom clearance and resistance in flow pathway at downcomer entrance on the riser liquid superficial velocity, the circulation time, the friction coefficient and flow radial profiles of the gas holdup and the liquid superficial velocity in riser, using water-air as a biphasic system, were studied. It was found that the riser liquid superficial velocity is affected by the analyzed geometrical parameters in different ways, depending on their effects on the pressure loss. The riser liquid superficial velocity, the friction coefficient and the parameters of the drift-flux model were satisfactorily correlated with the bottom spatial ratio (B), gas separation ratio (Y) and downcomer flow resistance ratio (A _d /A _D ), resulting empirical models, with correlation coefficients greater than 0.85.     

Modelling mixing parameters in concentric-tube airlift bioreactors

Axial dispersion of the liquid phase was investigated in a concentric-tube airlift bioreactor (RIMP: V _L=0.70?m³) as a whole and in the separate zones (riser, downcomer, gas-separator) using the axial dispersion model. The axial dispersion number Bo and the axial dispersion coefficient, D _ax were determined from the output curves to an initial Dirac pulse, using the tracer response technique. They were analyzed in relation to process and geometrical parameters, such as: gas superficial velocity, ν_SGR; top clearance, h _S; bottom clearance, h _B, and resistances at downcomer entrance expressed as A _d/A _R ratio. Correlations between Bodenstein numbers in the overall bioreactor and riser and downcomer sections (Bo_T,Bo_R,Bo_D) and the geometrical and process parameters were developed, which can allow to assess the complex influence of these parameters on liquid axial dispersion.     

Gas hold-up and oxygen mass transfer in three pneumatic bioreactors operating with sugarcane bagasse suspensions

Sugarcane bagasse is a low-cost and abundant by-product generated by the bioethanol industry, and is a potential substrate for cellulolytic enzyme production. The aim of this work was to evaluate the effects of air flow rate (Q _AIR), solids loading (%_S), sugarcane bagasse type, and particle size on the gas hold-up (ε _G) and volumetric oxygen transfer coefficient (k _L a) in three different pneumatic bioreactors, using response surface methodology. Concentric tube airlift (CTA), split-cylinder airlift (SCA), and bubble column (BC) bioreactor types were tested. Q _AIR and  %_S affected oxygen mass transfer positively and negatively, respectively, while sugarcane bagasse type and particle size (within the range studied) did not influence k _L a. Using large particles of untreated sugarcane bagasse, the loop-type bioreactors (CTA and SCA) exhibited higher mass transfer, compared to the BC reactor. At higher  %_S, SCA presented a higher k _L a value (0.0448 s^?1) than CTA, and the best operational conditions in terms of oxygen mass transfer were achieved for  %_S < 10.0 g L^?1 and Q _AIR > 27.0 L min^?1. These results demonstrated that pneumatic bioreactors can provide elevated oxygen transfer in the presence of vegetal biomass, making them an excellent option for use in three-phase systems for cellulolytic enzyme production by filamentous fungi.     

Modelling mixing parameters in concentric-tube airlift bioreactors

The mixing behaviour of the liquid phase in concentric-tube airlift bioreactors of different scale (RIMP: V_L=0.070 m³; RIS-1: V_L=2.50 m³; RIS-2: V_L=5.20 m³) in terms of mixing time was investigated. This mixing parameter was determined from the output curves to an initial Dirac pulse, using the classical tracer response technique, and analyzed in relation to process and geometrical parameters, such as: gas superficial velocity, x_SGR; top clearance, h_S; bottom clearance, h_B, and ratio of the resistances at downcomer entrance, A_d/A_R. A correlation between the mixing time and the specified operating and geometrical parameters was developed, which was particularized for two flow regimes: bubbly and transition (x_SGRА.08 m/s) and churn turbulent flow (x_SGR> 0.08 m/s) respectively. The correlation was applied in bioreactors of different scale with a maximum error of ᆲ%.     

Residence time distribution of liquid phase in an external-loop airlift bioreactor

The residence time distribution analysis was used to investigated the flow behaviour in an external-loop airlift bioreactor regarded as a single unit and discriminating its different sections. The experimental results were fitted according to plug flow with superimposed axial dispersion and tank-in-series models, which have proved that it is reasonable to assume plug flow with axial dispersion in the overall reactor, in riser and downcomer sections, as well, while the gas separator should be considered as a perfectly mixed zone. Also, the whole reactor could be replaced with 105-30 zones with perfect mixing in series, while its separate zones, that is the riser with 104-27, the downcomer with 115-35 and the gas separator with 25-5 perfectly mixed zones in series, respectively, depending on gas superficial velocity, A_D/A_R ratio and the liquid feed rate.List of Symbols A _D cross sectional area of downcomer (m²) - A _R cross sectional area of riser (m²) - A ₁ A ₂ length of connecting pipes (m) - Bo Bodenstein number (Bo=vL·L/D _ax (-) - C concentration (kg m^–3) - C residence time distribution function - C ₀ coefficientEquation (12) - C _r dimensionless concentration - D _D diameter of downcomer (m) - D _R diameter of riser column (m) - D _ax axial dispersion coefficient (m²s^–1) - H _d height of gas-liquid dispersion (m) - H _L height of clear liquid (m) - i number of complete circulations - L length of path (m) - m order of moments - N _eq number of perfectly mixed zones in series - n _c circulating number - Q _c recirculating liquid flow rate (m³ s^–1) - q _F liquid feed flow rate (m³s^–1) - Q _G gas flow rate (m³s^–1) - Q _T total liquid flow rate (m³s^–1) - r recycle factor - s exponent inEquation (12) regarded as logarithmic decrement of the oscillating part of RTD curve - t time (s) - t _C circulation time (s) - t _s mean residence time (s) - t ₉₉ time necessary to remove 99% of the tracer concentration (s) - V _A volume of connecting pipes (m³) - V _D volume of downcomer (m³) - V _L liquid volume in reactor (m³) - V _R volume of riser (m³) - V _LD linear liquid velocity in downcomer (m s^–1) - V _LR linear liquid velocity in riser (m s^–1) - V _SLD superficial liquid velocity in downcomer (m s^–1) - V _SLR superficial liquid velocity in riser (m s^–1) - x independent variable inEquation (1) - ¯x mean value of x - z axial coordinate - _GR gas holdup in riser - _m(x) central moment of m order - ² variance - dimensionless time     

alpha-Amylase production by Bacillus subtilis with dregs in an external-loop airlift bioreactor

An external-loop airlift bioreactor, with a low ratio 2.9 of height-to-diameter of the riser and a ratio 6.6 of riser-to-downcomer diameter, was used to produce alpha-amylase from fermentation with dregs by Bacillus subtilis. The effects of gas flow rate and liquid volume on alpha-amylase production were investigated. After a 36-h fermentation time, an average of 432.3U/ml alpha-amylase activity was obtained under the conditions of liquid volume 8.5l and gas flow rate 1.2vvm for the first 12h of fermentation, 1.4vvm from 12 to 27h, and 1.2vvm from 27h to the end. The activity was higher than that obtained in shaking flasks (409.0U/ml) and in a mechanically stirred tank bioreactor (397.2U/ml) under optimized operating conditions. The fermentation cycle of the airlift bioreactor was shorter than the 48h required for the shaking flasks and close to the 36h of the mechanically stirred tank bioreactor. It was demonstrated that the external-loop airlift bioreactor could substitute for the traditional mechanically stirred tank bioreactor to produce alpha-amylase from fermentation by Bacillus subtilis with dregs.     

Local gas hydrodynamics in an external-loop airlift reactor: newtonian and non-newtonian fluids

Measurements of local gas phase characteristics are obtained in an external-loop airlift reactor filled with newtonian or viscous non-newtonian liquids. A double-optical fiber probe technique is used. It allows the determination of the axial and radial profiles of gas hold-up, bubbling frequency, bubble size and velocity. In the case of air-water system, the results show a strong effect of radial liquid velocity variation on the gas flow characteristics at the bottom of the riser. In the case of highly viscous non-newtonian solution, the gas flow is strongly affected by the gas distribution just above the gas sparger. This study also points out the bubble coalescence and the break-up phenomena in different liquids and levels in the reactor. Furthermore, the local measurements of bubble size and velocity allows to gain more detailed information on the dynamics of the bubble-flow and shows a tendency of large bubbles to circulate in the column center.     

Oxygen transfer and mixing in mechanically agitated airlift bioreactors

Gas holdup, mixing, liquid circulation and gas–liquid oxygen transfer were characterized in a large (∼1.5 m³) draft-tube airlift bioreactor agitated with Prochem^® hydrofoil impellers placed in the draft-tube. Measurements were made in water and in cellulose fiber slurries that resembled broths of mycelial microfungi. Use of mechanical agitation generally enhanced mixing performance and the oxygen transfer capability relative to when mechanical agitation was not used; however, the oxygen transfer efficiency was reduced by mechanical agitation. The overall volumetric gas–liquid mass transfer coefficient declined with the increasing concentration of the cellulose fiber solids; however, the mixing time in these strongly shear thinning slurries was independent of the solids contents (0–4% w/v). Surface aeration never contributed more than 12% to the total mass transfer in air–water.