Mass Transfer in Bioreactors: Linda Gonzalez Ruiz Reyes Mayra

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/221910731
Mass Transfer in Bioreactors
Chapter · February 2011

DOI: 10.5772/15876 · Source: InTech
CITATIONS READS
3 9,682
6 authors, including:
Linda Gonzalez Ruiz Reyes Mayra

Centro De Investigación Y Desarrollo Tecnológico En Electroquí… Benemérita Universidad Autónoma de Puebla
33 PUBLICATIONS 577 CITATIONS 5 PUBLICATIONS 29 CITATIONS
SEE PROFILE SEE PROFILE
Oscar M. Hernández-Calderón Eleazar M. Escamilla-Silva

Universidad Autónoma de Sinaloa Instituto Tecnológico de Celaya
23 PUBLICATIONS 220 CITATIONS 57 PUBLICATIONS 1,258 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Production of Taxol by endophytic fungus isolated from Taxodium mucronatum View project
Extraction of Oleoresin from Jalapeno and other Mexican Peppers varieties (Capsicum annuum) using Supercritical CO2:
Determination of Global Yield Isotherms. View project
All content following this page was uploaded by Oscar M. Hernández-Calderón on 19 June 2014.
The user has requested enhancement of the downloaded file.

30
Mass Transfer in Bioreactors

Ma. del Carmen Chávez1, Linda V. González2, Mayra Ruiz3, Ma. de la Luz
X. Negrete4, Oscar Martín Hernández5 and Eleazar M. Escamilla6
1Facultad de Ingeniería Química, Universidad Michoacana de San Nicolás de Hidalgo,
Francisco J. Mújica s/n, Col. Felicitas del Río, 58060, Morelia, Michoacán.
2Centro de Investigación y Desarrollo Tecnológico en Electroquímica, Parque
Tecnológico Querétaro Sanfandila, 76703 Sanfandila, Pedro Escobedo, Qro.,

3Facultad de Ingeniería Química, Benemérita Universidad Autónoma de Puebla.
4 sur 104 centro histórico C.P. 72000, Puebla.,

4Departamento de Ingeniería Ambiental, Instituto Tecnológico de Celaya, Ave.
Tecnológico y Antonio García Cubas S/N, Celaya, Gto., C.P. 38010,

5Universidad Autónoma de Sinaloa. Facultad de Ciencias Químico Biológicas.
Ciudad Universitaria, C.p. 80090, Culiacán, Sinaloa.

6Instituto Tecnológico de Celaya, Departamento de Ingeniería Química, Ave.
Tecnológico y Antonio García Cubas S/N, Celaya, Gto., C.P. 38010,Sinaloa.

México
1. Introduction
The study of transport in biological systems is complicated for two reasons: 1. because each
system is different, we cannot generalize it and 2. Because always take place in more than
one phase. If we talk about microorganism, there is a range of them with physicochemical
and biological characteristics very different, and certain microorganisms can be filamentous
and can grow branched or dispersed, in some the viscosity and density increases with time.
In some times their maximum growth rate is achieved in two hours while others in 15 days.
Some are affected by the light, others agitation rate, others require air for developing others
not. If we talk about production of plants by tissue culture systems have become more
complex, that the transport properties are affected by agitation rate, type of agitation, the
growth of tissues. To design the bioreactors of these biological systems requires knowledge
of the nature of what is to be produced, the dynamics of transport, rheology, to decide what
type of reactor we can used. Biological fluids such reactors behave as highly non-Newtonian
systems and as such require special treatment. This paper will discuss three types of
reactors: air-lift, packed column and fluidized bed and stirred tank, where case studies are
applied to biological systems. 1. Production of Gibberellic acid and Bikaverin 2.
Biodegradation of azodyes in textile industry and 3. Gibberellins Production. It is intended
that in these three cases brought to appreciate as engineering parameters are evaluated
where they involve the transport mass balances and the type of bioreactor and feature you
in l fluid. On the other hand show a combination of experimental results and simulations
with mathematical models developed to strengthen the knowledge of chemical engineering
applied to biological systems.
718 Mass Transfer in Multiphase Systems and its Applications
2. Case I. Hydrodynamics, mass transfer and rheological studies of

gibberellic acid production in an airlift bioreactor
2.1 Introduction
Gibberellic acid is an endogenous hormone in higher plants, belonging to the group of
gibberellins, and also a product of the secondary metabolism in certain fungi. Approximately
126 gibberellins have been characterized (Tudzynski 1999; Shukla et al. 2003) but only a few
are commercially available. Gibberellic acid is the most important and its effects on higher
plants are: marked stem elongation, reversal of dwarfism, promotion of fruit setting, breaking
of dormancy, acceleration of seed fermentation, among others (Bru¨ ckner and Blechschmidt
1991; Tudzynski 1999). Currently, gibberellic acid is microbiologically produced in a
submerged culture (SmF) fashion but another fermentation techniques such as solid sate
fermentation or with immobilized mycelium are also reported (Heinrich and Rehm 1981; Jones
and Pharis 1987; Kumar and Lonsane 1987, 1988; Nava Saucedo et al. 1989; Escamilla et al.
2000; Gelmi et al. 2000, 2002). Nevertheless stirred tank bioreactors with or without a fed-batch
scheme have been the most employed in gibberellic acid production. Other geometries and
type of bioreactors have also been reported. Only Chavez (2005) has described gibberellic acid
production employing an airlift bioreactor. Airlift bioreactors are pneumatically agitated and
circulation takes place in a defined cyclic pattern through a loop, which divides the reactor
into two zones: a flow-upward and a flow-downward zone. The gas-sparged zone or the riser
has higher gas holdup than the relatively gas-free zone, the downcomer, where the flow is
downward (Gouveia et al. 2003). Practical application of airlift bioreactors depends on the
ability to achieve the required rates of momentum; heat and mass transfer at acceptable capital
and operating costs. The technical and economic feasibility of using airlift devices has been
conclusively established for a number of processes and these bioreactors find increasing use in
aerobic fermentations, in treatment of wastewater and other similar operations. The simplicity
of their design and construction, better defined flow patterns, low power input, low shear
fields, good mixing and extended aseptic operation, made possible by the absence of stirrer
shafts, seals and bearings, are important advantages of airlift bioreactors in fermentation
applications (Chisti 1989).
Even though gibberellic acid has been produced on an industrial scale since the last century,
hydrodynamics, mass transfer and rheological studies are sparse. Flow regime, bubble size
distribution, and coalescence characteristics, gas holdup, interfacial mass transfer coefficients,
gas–liquid interfacial area, dispersion coefficients and heat transfer coefficients are important
design parameters for airlift bioreactors. A thorough knowledge of these interdependent
parameters is also necessary for a proper scale-up of these bioreactors (Shah et al. 1982).
Besides hydrodynamics and mass transfer studies, rheological studies are important since in
many chemical process industries, the design and performance of operations involving fluid
handling like mixing, heat transfer, chemical reactions and fermentations are dependent on the
rheological properties of the processed media (Brito-De la Fuente et al. 1998). Mycelial
fermentation broths present challenging problems in the design and operation of bioreactors
since the system tends to have highly non-Newtonian flow behaviour and this has a very
significant effect on mixing and mass transfer within the bioreactor.
The main objective of this work was to study hydrodynamic, mass transfer and rheological
aspects of gibberellic acid production by Gibberella fujikuroi in an airlift reactor.
2.2 Materials and methods

Microorganism and inoculum preparation Gibberella fujikuroi (Sawada) strain CDBB H-984
maintained on potato dextrose agar slants at 4_C and sub-cultured every 2 months was used
Mass Transfer in Bioreactors 719
in the present work (Culture collection of the Department of Biotechnology and

Bioengineering, CINVESTAV-IPN, Mexico). Fully developed mycelia materials from a slant
were removed by adding an isotonic solution (0.9% NaCl). The removed mycelium was
used to inoculate 300 ml of fresh culture medium contained in an Erlenmeyer flask. The
flask was placed in a radial shaker (200 rev min–1) for 38 h at 29 ± 1_C. Subsequent to this
time; the contents of the flask were used to inoculate the culture medium contained in the
airlift bioreactor. The culture medium employed for the inoculum preparation is reported by
Barrow et al. (1960).
Batch culture in the airlift bioreactor
An airlift bioreactor (Applikon, Netherlands, working volume, 3.5 l) was employed in the
present work. It consists of two concentric tubes of 4.0 and 5.0 cm of internal diameter with
a settler. The air enters the bioreactor through the inner tube. A jacket filled with water
allowing temperature control surrounds the bioreactor. It is also equipped with sensors of
pH and dissolved oxygen to control these variables. Moreover it allows feed or retiring
material from the bioreactor employing peristaltic pumps. Typical culture medium
contained glucose (50 g l–1), NH4Cl (0.75 g l–1) or NH4NO3 (1.08 g l–1), KH2PO4 (5 g l–1),
MgSO4. 7 H2O (1 g l–1) and trace elements (2 ml l–1). A stock solution of the trace elements
used contained (g l–1) 1.0 Fe SO4. 7 H2O, 0.15 CuSO4. 5 H2O, 1.0 ZnSO4. 7 H2O, 0.1 MnSO4. 7
H2O, 0.1 NaMoO4, 3.0 EDTA (Na2 salt) 1 l of distilled water, and hydrochloric acid sufficient
to clarify the solution (Barrow et al. 1960). During the fermentation period, the pH was
controlled to 3.0, temperature to 29°C and aeration rate to 1.6 vvm. These conditions
promoted gibberellic acid production with the studied strain but they are not optimized
values. About 30 ml subsamples were withdrawn from the bioreactor at different times and
were used to perform rheological studies. Biomass concentration was quantified by the dry
weight method.
2.3 Hydrodynamics and mass transfer studies

Gas holdup was determined in the actual culture medium using an inverted U-tube
manometer as described by Chisti (1989). Liquid velocities in the riser were determined
measuring the time required for the liquid to travel through the riser by means of a pulse of
concentrated sulphuric acid using phenolphthalein as an indicator; the same was done for
the downcomer. The mixing time was calculated as the time required obtaining a pH
variation within 5% of the final pH value. For doing this, pH variation was followed after
injection of a pulse of a concentrated solution of ammonium hydroxide. The volumetric
mass transfer coefficient was determined employing the gassing-out method as described
elsewhere (Quintero 1981).
2.4 Rheological studies

Rheological studies of fermentation broth were performed in a rotational rheometer (Haake,
Model CV20N) equipped with a helical impeller to perform torque measurements. This type
of geometry is appropriate when dealing with complex fluids and the measurement
methodology is reported by Brito-de la Fuente et al. (1998). Rheological results, like
hydrodynamics and mass transfer, are given as the average of two replicates for each
sample. All the experiments were carried out in triplicate and the results that are presented
are an average.
2.5 Results and discussion

Gas holdup
The importance of gas holdup is multifold. The gas holdup determines the residence time of
the gas in the liquid and, in combination with the bubble size, influences the gas–liquid
interfacial area available for mass transfer. The gas holdup impacts upon the bioreactor
design because the total design volume of the bioreactor for any range of operating
conditions depends on the maximum gas holdup that must be accommodated (Chisti 1989).
Figure 1 shows the gas holdup (ε) variation with superficial gas velocity in the riser (vgr).
Experimental data were fitted to a correlation of the type of Eq. 1.
B
F = A vgr (1)
Where F could be the gas holdup (ε), the liquid velocity in the riser (vlr), liquid velocity in
the downcomer (vld) or the volumetric mass transfer coefficient (kLa). This type of
correlation has been applied by many investigators (Shah et al. 1982; Godbole et al. 1984;
Chisti 1989; Gravilescu and Tudose 1998; Abashar et al. 1998) and was derived empirically.
Chisti (1989) presented an analysis for Newtonian and non-Newtonian fluids where shows
the theoretical basis of Eq. 1 (for the gas holdup case). He found that parameters A and B
were dependent on the flow regime and on the flow behaviour index of the fluid. Moreover,
parameter A is dependent on the consistency index of the fluid, on the fluid densities and on
the gravitational field. Equation 2 was obtained from fitting experimental data.
ε = 0.7980 v1.0303
gr (2)
Fig. 1. Gas holdup variation with superficial gas velocity in the riser.
• Experimental data ––– Equation 2 --- Equation 12
An increase in superficial gas velocity in the riser implies an increase in the quantity of gas
present in the riser, that is, an increase of gas fraction in the riser (Chisti 1989; Gravilescu
and Tudose 1998). Chisti (1989) reports a correlation that calculates the value of B in Eq. 1
(for the gas holdup case). The value obtained employing this correlation is 1.2537.
Gravilescu and Tudose (1998) present a similar correlation, which predicts a value of 0.8434
for B. The B value obtained in the present work is between the B values obtained from these
correlations that employ the flow behaviour index obtained from rheological studies. Shah
et al. (1982) reported that B values in Eq. 1 oscillate between 0.7 and 1.2.
Liquid velocity
The liquid circulation in airlift bioreactors originates from the difference in bulk densities of
the fluids in the riser and the downcomer. The liquid velocity, while itself controlled by the
gas holdups in the riser and the downcomer, in turn affects these gas holdups by either
enhancing or reducing the velocity of bubble rise. In addition, liquid velocity affects
turbulence, the fluidreactor wall heat transfer coefficients, the gas–liquid mass transfer and
the shear forces to which the microorganism are exposed. Figure 2 shows liquid velocities
variation in the riser and the downcomer as a function of superficial gas velocity in the riser.
Liquid velocities in the riser (vlr) and in the downcomer (vld) were fitted to correlations of
the type of Eq. 1 and Eqs. 3 and 4 were obtained.
vlr = 1.3335 v0.3503

gr (3)
vld = 0.8716 v0.2970

gr (4)
Fig. 2. Liquid velocities as a function of superficial gas velocity in the riser.

• Experimental data ––– Equation 3 or 4
The B value in Eq. 1 must be close to 0.3333 as was reported by Freitas and Teixeira (1998) for
the liquid velocity in the riser, Kawase (1989) theoretically derived this value. The B value
obtained in the present work is closer to 0.3333. Freitas and Teixeira (1998) also showed that
the B values for the liquid velocity in the downcomer were lower than the B value for the
liquid velocity in the riser, which agrees with the results obtained in this work. Liquid
velocities in the riser and in the downcomer increase with an increase in gas velocity in the
riser due to an increase in the density difference of the fluids in the riser and the downcomer.
Mixing time
Mixing in airlift bioreactors may be considered to have two contributing components: back
mixing due to recirculation and axial dispersion in the riser and
downcomer due to turbulence and differential velocities of the gas and liquid phases (Choi
et al. 1996).
Mixing time is used as a basis for comparing various reactors as well as a parameter for
scaling up (Gravilescu and Tudose 1999). Figure 3 shows the mixing time variation with the
superficial gas velocity in the riser. Once again, the mixing time variation was fitted to a
correlation of the type of Eq. 1 and Eq. 5 was obtained.
tm = 5.0684 v −gr0.3628 (5)
Choi et al. (1996) reported a B value in Eq. 5 of –0.36 while Freitas and Teixeira (1998)
reported a B value equal to –0.417. The B value obtained in this work is similar to the value
reported by Choi et al. (1996). The mixing time decreases with an increase in superficial gas
velocity in the riser since the fluid moves more often to the degassing zone where most of
the mixing phenomenon takes place, due to the ring vortices formed above the draught tube
(Freitas and Teixeira 1998).
Volumetric mass transfer coefficient
One of the major reasons that oxygen transfer can play an important role in many biological
processes is certainly the limited oxygen capacity of the fermentation broth due to the low
solubility of oxygen. The volumetric mass transfer coefficient (kLa) is the parameter that
characterizes gas-liquid oxygen transfer in bioreactors. One of the commonest employed
scale-up criteria is constant kLa. The influences of various design (i.e., bioreactor type and
geometry), system (i.e., fluid properties) and operation (i.e., liquid and gas velocities)
variables on kLa must be evaluated so that design and operation are carried out to optimize
kLa (Chisti, 1989).
Fig. 3. Mixing time as a function of superficial gas velocity in the riser.

The value of the volumetric mass transfer coefficient determined for a microbial system can
differ substantially from those obtained for the oxygen absorption in water or in simple
aqueous solutions, i.e., in static systems with an invariable composition of the liquid media
along the time. Hence kLa should be determined in bioreactors which involve the actual
media and microbial population (Tobajas and García-Calvo, 2000). Figure 4 shows the
volumetric mass transfer coefficient variation with the superficial gas velocity in the riser.
Experimental data shown in Figure 4 were fitted to a correlation of the type of Equation 1
and Equation 6 was obtained.
kL a = 0.4337 v 1.2398
gr (6)
Barboza et al., (2000) report a B value in Equation 6 equal to 1.33 and Schügerl et al., (1977)
report a value of 1.58. The value of 1.2398, obtained in this work, is close to these last values.
Fig. 4. Effect of the superficial gas velocity in the riser on kLa.

Volumetric mass transfer coefficient (kLa) increases with an increase in superficial gas velocity
in the riser due to an increase in gas holdup which increases the available area for oxygen
transfer. Moreover an increase in the superficial gas velocity in the riser increases the liquid
velocity which decreases the thickness of the gas-liquid boundary layer decreasing the mass
transfer resistance. Figure 5 shows the evolution of kLa through fermentation course employing
two different nitrogen sources. The kLa decreases in the first hours of fermentation and reaches
a minimum value at about 24 hours. After this time the kLa starts to increase and after 48 hours
of fermentation it reaches a more or less constant value which remains till the end of
fermentation process. This behaviour is similar irrespective of the nitrogen source and will be
discussed with the rheological results evidence.
Fig. 5. kLa through fermentation time in the airlift bioreactor.

Figure 6 shows the relation between gas holdup and kLa. McManamey and Wase (1986)
point out that the volumetric mass transfer coefficient is dependent on gas holdup in
pneumatically agitated systems. The later was experimentally determined in bubble
columns by Akita and Yoshida (1973) and Prokop et al., (1983). Shah et al., (1982) mention
that this was expectable since both the volumetric mass transfer coefficient and the gas
holdup present similar correlations with the superficial gas velocity. McManamey and Wase
(1996) proposed a correlation similar to Equation 1 to relate volumetric mass transfer
coefficient with gas holdup. Equation 7 presents the obtained result.
kL a = 0.2883 ε 0.9562 (7)
Akita and Yoshida (1973) and Prokop et al. (1983) found that the exponent in Equation 7
oscillates between 0.8 and 1.1.
⎛ k ⎞ ε
ln kL a = ln ⎜ 6 L ⎟ + ln (8)
⎝ dB ⎠ (1 − ε )
Fig. 6. kLa vs. gas holdup in the airlift bioreactor, unit slope.
It is well known (Chisti, 1989) that logarithmic scale plots of kLa vs. ε/(1- ε) for any particular
data set should have a unit slope according to Equation 8. Where kL is the mass transfer
coefficient and dB is the bubble diameter. Even though the later is a generally known fact,
few investigators determined these slopes for their data to ascertain the validity of their
experimental results. Figure 6 shows this analysis for the experimental data of the present
work obtaining a slope of 1.034. Chisti (1989) shows the same analysis for two different data
set and obtained slopes of 1.020 and 1.056.
A rearrangement of Equation 8 leads to Equation 9 which results are shown in Figure 7. As
is showed in the Figure 7 the gas superficial velocity practically did not affect the kL/dB
values, therefore it can be taken as a value average and constant to slant the superficial
velocity changes.
kL kL a ( 1 − ε )
= (9)
dB 6ε
The average value of kL/dB obtained in the present work is 0.050 s-1. Chisti (1989) performed a
similar analysis for 97 data points obtained from several different reactors and found an
average value of 0.053 s-1. The foregoing observations have important scale-up implications. In
large industrial fermenters the kLa determination is not only difficult, but there is uncertainty
as to whether the measured results reflect the real kLa or not. The gas holdup measurements on
these reactors are relatively easy to carry out, however. Thus, Equation 9 can help to estimate
kLa in these reactors once gas holdup measurements have been made (Chisti, 1989).
Fig. 7. The kL/dB ratio as a function of superficial gas velocity.
2.6 Rheology
Rheological parameters such as the flow index (n) and the consistency index (K) depend on
such factors as the concentration of solids in the broth, the morphology (length, diameter,
degree of branching, shape) of the particles, the growth conditions (flexibility of cell wall
and particle), the microbial species and the osmotic pressure of the suspending liquid,
among others possible factors. For the case of mycelial cultures, as the biomass
concentration increases the broth becomes more viscous and non-Newtonian; leading to
substantial decreases in oxygen transfer rates. This effect is often important since for many
aerobic processes involving viscous non-Newtonian broths oxygen supply is the limiting
factor determining bioreactor productivity (Moo-Young et al., 1987). Apparent viscosity is a
widely used design parameter which correlates mass transfer and hydrodynamic
parameters for viscous non-Newtonian systems (Al-Masry and Dukkan, 1998).
It is worth to mention that the present work uses impeller viscometry for performing
rheological studies avoiding the use of other geometries, i.e., concentric tubes or cone and
plate, overcoming associated problems with these geometries such sedimentation, solids
compacting and jamming between measuring surfaces or pellet destruction (Metz et al.,
1979). Impeller viscometry was used to obtain torque data at different velocities of the
impeller, these data were transformed to shear stress (τ) and shear rate (γ) data and typical
results are shown in Figure 8. As can be seen in Figure 8, the experimental data follow a
straight line and can be represented by the Ostwald-de Waele model (Equation 10).
τ = Kγ n (10)
Fig. 8. Typical rheogram employing impeller viscometry

Rheograms obtained from fermentations employing different nitrogen source show a
pseudo plastic behaviour for the culture medium during the fermentation period since the
exponent, n, in Equation 10 is always lower than unity. Figure 9 shows the results of
consistency and flow indexes for the different fermentations, employing ammonium
chloride or ammonium nitrate as nitrogen source, where similar results were obtained.
2 ,0 0 ,6
1 ,6 0 ,5
K 0 ,4
K, N sn m-2
1 ,2
n, -
0 ,3
0 ,8
0 ,2
0 ,4
n 0 ,1
0 ,0 0 ,0
0 50 100 150 200 250
T im e , h
Fig. 9. K and n through fermentation time in the airlift bioreactor • K for ammonium nitrate
▲ n for ammonium nitrate K for ammonium chloride ¡ n for ammonium chloride.
Fig. 10. Growth kinetics employing ammonium chloride ( ) or ammonium nitrate (•) as
nitrogen source.
Figure 10 shows the growth kinetics of Gibberella fujikuroi obtained during different
fermentations. As can be seen in Figure 10, the growth kinetics is similar irrespective of the
employed nitrogen source. Experimental data where fitted to two-parameter Gompertz
model proposed by Chavez-Parga et al., (2005).As can be seen in Figure 10, there is no lag
phase and exponential growth of mycelia starts immediately and ceases during the first 24
hours of fermentation. The later causes the medium viscosity to increase (K and n increase in
Figure 9) which causes a kLa decrease in Figure 5. After 24 hours of fermentation, the
formation of pellets by the fungus starts to occur reflected in a decrease of medium viscosity
(K and n start to decrease in Figure 9) and hence an increase in kLa value in Figure 5. After 72
hours of fermentation the medium viscosity was practically unchanged (K and n remain
constant in Figure 9) because the stationary growth phase is reached by the fungus reflected
in practically constant values of medium viscosity and kLa. Also, after 72 hours of
fermentation, the pellet formation process by the fungus stops.
Figure 11 shows the correlation between consistency and flow indexes with biomass
concentration. Experimental data were fitted to Equations 11 and 12 proposed in the present
work. Optimized values for constants in Equations 11 and 12 are summarized in Table 1.
c1
K= 2
(11)
⎛ c2 ⎞ ⎛ x ⎞
⎜1 + ⎟ + ⎜ ⎟
⎝ x ⎠ ⎝ c3 ⎠
c1
n= 2
(12)
⎛ c2 ⎞ ⎛ x ⎞
⎜1 + ⎟ + ⎜ ⎟
⎝ x ⎠ ⎝ c3 ⎠
2.0 0.6
1.8
0.5
1.6
0.4
1.4
K, Nsnm-2
n, -
1.2 0.3
1.0
0.2
0.8
0.1
0.6
0.4 0.0
0 2 4 6 8 10 12
Biomass, g/L
Fig. 11. K and n as a function of biomass concentration in the airlift bioreactor.

• K for ammonium nitrate ▲ n for ammonium nitrate K for ammonium chloride
¡ n for ammonium chloride.
Consistency index
Nitrogen source c1 c2 c3
Ammonium nitrate 6.31 6.55 4.69
Ammonium chloride 3.43 2.05 6.67
Flow index
Nitrogen source c1 c2 c3
Ammonium nitrate 7.64 106.49 1.27
Ammonium chloride 7.63 80.91 1.14
Table 1. Optimized values found for constants of Equations 11 and 12.
With the aid of rheological studies is possible to use correlations of the type of Equation 13
to relate gas holdup and volumetric mass transfer coefficient with fermentation medium
viscosity (Godbole et al., 1984; Halard et al., 1989; Al-Masry and Dukkan, 1998; Barboza et al.,
2000) to obtain Equations 14 and 15.
F = Av Bgr μ app
C
(13)
−0.5488
kL a = 0.0036 v0.3775
gr μ app (14)
−0.5703
ε = 0.0072 v0.2381
gr μ app (15)
Figures 1 and 4 show experimental data fitting for gas holdup and kLa, respectively. As it
was expectable, Equations 14 and 15 present a better fit to experimental data than that
obtained with the aid of Equations 2 and 3 due to the existence of an extra adjustable
parameter.
2.7 Conclusions
In the present work preliminary hydrodynamics, mass transfer and rheological studies of
gibberellic acid production in an airlift bioreactor were achieved and basic correlations
between gas holdup, liquid velocity in the riser, and liquid velocity in the downcomer,
mixing time and volumetric mass transfer coefficient with superficial gas velocity in the
riser were obtained. Adjustable parameters calculated for each variable were compared with
literature reported values and a good agreement was obtained. Gassing out method was
successfully applied in determining volumetric mass transfer through fermentation time
employing two different nitrogen sources. Irrespective of the nitrogen source the volumetric
mass transfer behaviour was similar and it was explained in terms of the fungus growth and
changes in its morphology which affect the culture medium rheology. Pellet formation by
the fungus was used to explain the increase of kLa or the decrease of medium viscosity. In
both fermentations, kLa decreases as exponential growth of the fungus occurs and reaches an
asymptotic value once the stationary growth phase is reached. A helical impeller was
employed successfully for rheological studies, avoiding problems of settling, jamming or
pellet destruction, finding that the culture medium behaves as a pseudoplastic fluid.
Rheological measurements were used to correlate gas holdup and kLa with apparent culture
medium viscosity. Once again, for both fermentations, apparent viscosity increases as
exponential growth of the fungus occurs and reaches an asymptotic value once the
stationary growth phase is reached. A satisfactory validation of experimental data for gas
holdup and volumetric mass transfer coefficient was performed which allows to employ
these data in scale-up strategies.
3. Case 2. Dynamic transport and reaction model for the removal of azo dye
in a UAFB reactor
3.1 Introduction
Azo dye degradation from textile effluents has been the objective of research for some years
due to the pollution problem they generate. For the removal of these compounds different
processes have been applied: physicochemical, advanced oxidation, and biological.
However there is a continuous search for an efficient, low cost and low environmental
impact process to eliminate this problem. In particular, Reactive dyes are highly water
soluble due to the sulphonated groups in their molecule so it cannot be reduced under the
ordinary wastewater treatment processes (Beydilli, 2005). Anaerobic bioreactors have an
important role in the treatment process of hazardous wastes, besides they can treat higher
organic loads than aerobic reactors. Fixed bed reactors can be immerse, usually upflow, or
trickle bed, downflow, the main characteristic is that the biomass is forming a biofilm
covering a material that works as a support or carrier for the growth and maintenance of the
microorganisms; in this way, the reactor efficiency is improved because the substrate-
biomass contact is increased (effective surface area), and the process is more stable. The use
of a carrier in the reactor is to improve the mechanical properties of the biomass and cell
retention; in addition, the carrier may participate in the degradation process (Van der Zee, et
al. 2003). A biofilm usually do not grow in a homogeneous way on the support, but rather
forms clusters on the surface; the way in which a biofilm is grown and their internal
structure is formed depends on the superficial velocity of the flow through the reactor, it is
also affected by the mass transfer velocity and microorganism activity (Beyenal, 2002). The
degree of biomass buildup affects the hydrodynamic behavior of the reactor. In this work,
an Upflow Anaerobic Fixed Bed (UAFB) bioreactor with activated carbon (AC) as the carrier
was used to remove azo dye from the effluent. It has been proved that AC possess good
properties for biofilm growth and to remove diverse pollutants (Fan, et al. 1987; Fan, et al.
1990; Herzberg and Dosoretz, 2003; McCarty and Meyer,2005), moreover, AC could
accelerate azo dye degradation due to its redox mediator function through the chemical
groups on its surface (Van der Zee, et al. 2003) Di Iaconi et al (2005) proposed a mechanism
for biofilm growth: 1) formation of a thin film covering the support by the microorganisms,
2) increment of the biofilm thickness, 3) the break of the added biofilm clusters and release
of particles (biomass due to the excess of growth) and 4) small pellet formation by detached
particles. In UAFB reactors it is common to have the bioparticles (carrier plus biofilm), some
free cells and biomass pellets as a function of the superficial velocity on the reactor; the
water flowing through the bioreactor can carry out the drag of small biomass pellets. The
mass transport through this bioparticles occurs on three stages: diffusion of the dye
molecule from the solution to the biofilm, diffusion through the biofilm, adsorption-
diffusion through the carbon surface and reaction. One disadvantage of using upflow fixed
bed reactors is that the liquid flow is non-ideal and dispersion, backmixing and bypassing
flow are considerable (Iliuta, et al. 1996), therefore it is important to carry out the hydraulic
characterization of the reactor through tracer test, although it is common to consider plug
flow to model the reactor. The reasons of modelling a reactor of this kind are to estimate all
the important parameters in its function, to optimize the efficiency and to predict its
behaviour, besides its future scale-up. However, scaling a reactor from laboratory models is
often difficult, since some factors which are negligible when modelling small reactors have
to be included in real reactor models, such as the transport between static and dynamic
zones. Therefore, the main objective of this paper is to propose a dynamic mathematical
model for an UAFB bioreactor with AC as carrier, to attach microorganisms and enhance
biodegradation, in the removal of the azo dye reactive red 272 (Fig. 1).
Fig. 1. Reactive red 272

The presented mathematical model includes all the transport phenomena: convection,
dispersion, diffusion and mass transfer from one phase to another, along the reactor and
through the bioparticle, as well as the reaction of dye reduction. The balance equations are
coupled and solve together as a system. We try to include in the model all the possible
phenomena that take place in the reactor in order to describe it and obtain enough
information about it.
3.2 Materials and methods

3.2.1 Reactor assembling.
It was started up to work an anaerobic upflow reactor of the kind of the UAFB, made from
Pyrex glass, with a fixed bed of AC of 42% of its operation volume, equivalent to 1.244 L and
541.17 g of AC. The reactor is outlined in Figure 2 and its characteristics are shown in Table 1.
Fig. 2. Upflow Anaerobic Fixed Bed (UAFB) Reactor.

Work volume, L 3.3

Inside diameter, cm 6
Inside diameter of the settle, cm 9.5
Total lon gi tude, c m 10 5. 5
Initial and steady state porosity of the bed 0.53, 0.19
Fixed bed volume, L 1 . 24 4
Fixed bed longitude, cm 48
Superficial velocity (average), cm/min 0.52
Volumetric flow (average), mL/min 18
RTm (average), min 2 06 . 2 5
Table 1. UAFB Reactor Characteristics
At the beginning, there was an adsorption stage to saturate the AC in the reactor with dye
and do not attribute the removal efficiency to simply adsorption on to AC. Afterwards, the
rector was inoculated by recirculating water with 10% v/v of adapted sludge for a period of
15 days; this was a consortium of microorganisms adapted to azo dye reduction using textile
wastewater enriched with reactive red 272. In this stage, 11.586 mg biomass/g AC was
adsorbed, forming a biofilm on the AC surface. The reactor was operated using synthetic
wastewater, containing different azo dye concentration, from 100 to 500 mg/L, and 1 g/L of
dextrose and yeast extract as carbon and nitrogen source to the microorganism.
3.2.2 Residence time distribution.

A lithium chloride solution was used as a tracer in order to determine the hydraulic
characteristics of the reactor and to obtain the residence time distribution. Smith and Elliot
(1996) used LiCl as a tracer and recommend a concentration of 5 mg Li+/L to avoid toxicity
problems. In this case it was applied a one minute pulse of a 2000 mg/L LiCl solution.
Dextrose and yeast extract with a concentration of 1 g/L were used as the substrates during
the test. Samples were taken in the reactor effluent every 30 min during approximately 3
times the half residence time (RTm), in this case, during 10 hr. Lithium concentration was
analyzed in an atomic adsorption spectrophotometer (Perkin Elmer model 2280; USA).
The hydraulic residence time (HRT) was calculated as:
‘ HRT =
∫0 t(C − C0 )dt (1)
∞
∫0 (C − C0 )dt
Were C is the tracer concentration at a time t and C0 is the tracer concentration at t=0. The
parameters and non-dimensional numbers necessary to describe the reactor as well as the axial
dispersion and mass transfer coefficients were calculated according to the next equations.
Dispersion number (d) and Péclet (Pe). These numbers indicate the dispersion grade in the
reactor. A Pe above 1 indicates that convection is the leading factor in the mass transport,
and if it is lesser than 1, the leading factor is the dispersion. The numbers are calculated as
(Levenspiel, 2004):
1 σ Δ2C D
d= = (2)
2 RTm 2 uL
uL 1
Pe = = (3)
D d
Where u is the superficial velocity in the reactor, L is the longitude and D is the axial
dispersion coefficient.
Dispersion coefficient (D). It can be calculated by the dispersion number or by other
correlations as the presented through the Reynolds number.
D = d u L = 1.01ν N Re 0.875 (4)
Here, ν is the cinematic viscosity of the water in the reactor (Levenspiel, 2004).
Sherwood number (Sh) and mass transfer coefficient (km). It was calculated by the Frössling
correlation (Fogler, 1999), which is applied to the mass transfer or flux around a spherical
Sh = 2 + 0.6 N Re 1/2Sc 1/3 (5)
particle. Supposing this the following equation was used:

And the mass transfer coefficient of the dye was estimated by the equation:
Where dp is the average particle diameter of the carbon particles and biomass in the bed. For
this analysis it was taken the dp values of the carbon particles at the beginning of the study,
1.03 mm.
Def Sh
km = (6)
dP
3.2.3 Kinetic model

The applied kinetic model to represent the dye biodegradation (reduction) was derived
according to experimental observations, after fitting kinetic data at dye concentrations from
100 to 500 mg/l. The model expresses a change in the reaction order since it was noticed that
the reaction in the system is a function of dye concentration and occur in two stages: first
order, the dye is adsorbed by the bioparticle and reduced, and second, the enzymatic reactions
take place to degrade the dye to certain extent. This is shown by Equation 7, here: CA0 and CA
are the initial and every moment dye concentration, k1 and k2 are 1st and 2nd order specific
reaction rate, (h-1, L/mg⋅h). The deduction is explained in another paper (in revision).
dC A
rA = − = k1C A − k2C A (C A0 − C A ) (7)
dt
3.2.4 Model dimensionless numbers

From the dimensionless analysis of the model, the dimensionless numbers that explain the
transport process in the reactor were obtained. These were: Biot’s number (Bi), that relates
mass transfer with diffusivity, Fourier’s number (Fo), that relates the diffusivity in the
reaction area in the reaction time, Wagner’s module (Φ2), by means of which it is obtained
the Thiele’s number (Φ) that indicates if diffusion modifies the reaction rate; from this
number, the Effectiveness factor (η) is calculated, which relates the real reaction rate with
the reaction rate without diffusion resistance, in other words, it expresses the influence of
the diffusion on the reaction rate. Thiele number is calculated by Equation 8 (according the
proposed kinetic model there is a Thiele number for the first order term and other for the
second order term). The Effectiveness factor for the reduction rate of the dye by volume unit
of bioparticle was calculated using Equation 9, according to the definition of volume
average (Escamilla-Silva et al, 2001) and using the proposed kinetic model expressed in
Equation 7. Here, RA is the average reaction rate in the biofilm and RA ξ = 1 is the reaction
rate in the bioparticle surface in the liquid boundary; Fob is the characteristic Fourier number
for the biofilm defined in Equation 19, in the next section.
k1 kC
Φ1 = δ ; Φ 2 = δ 2 A0 (8)
Deb Deb
1
3 ∫0 ⎡⎣Φ 12 Fobωb − Φ 22 Fobωb (ωL − ωb ) ⎤⎦ ξ 2 dξ
1 1
4π ∫0 RAξ 2 dξ 3 ∫0 RAξ 2 dξ RA
η= = = = (9)
4
π RA RA ξ =1 ⎡Φ 2 Fobωb − Φ 2 Fobωb (ωL − ωb ) ⎤ RA
3 ξ =1 ⎣ 1 2 ⎦ ξ =1 ξ =1
3.3 Results and discussion.

The UAFB reactor efficiently removes the reactive red dye, from 91.35% to 98.64% and up to
56% of DQO, at inflow concentration from 100 to 500 mg/L and at a RTm from 3 to 5 hours.
Higher removal rates can be obtained at higher residence times. The difference between
colour and COD removal is because the first step in the biodegradation of the dye takes
place when the azo bond is broken, and this results in the lost of colour of the solution.
There are aromatics amines and other organic compounds in the water as products of dye
reduction, which can be degraded to a certain extent in to other low molecular weight
molecules, as carboxylic acids. The results described in this section are in regard to an
analysis of the transport and reaction phenomena inside the reactor and to obtain
predictions about its performance. The balance equations are proposed according to
theoretical principles. Some of the parameters used were calculated according to
experimental and real results and others in base to references. The model can be used and
applied to similar problems, but it will need a parameter fit. Because of this, the real
removal rate of the reactor is higher than the predicted for the model, at the highest dye
concentration used (400-500 mg/L).
3.4 Residence time distribution.

The parameters and non-dimensional numbers that describe the transport in the reactor
fixed bed are shown in Table 1. The superficial velocity was calculated as uL = Q ε Lπ Ri2 and
the porosity of the bed εL was 0.19 after equilibrium was reached. The hydraulic behaviour
of the reactor was approximated to a plug flow with axial dispersion. Figure 3 show HRT
distribution curves; it was observed that when Q was increased, the dispersion was reduced
and the reactor was closer to ideal plug flow behaviour. This is a hydrodynamic effect, but
for packed beds it is attributed to the particle size of the packing material. This result can be
attributed to the fine particles formed with time operation in the inter-particle space in the
reactor, because it reduces the bed porosity and as a result the by-pass fluxes. Kulkarni et al
(Kulkarni, 2005) established that the fine particles formed in packed bed reactors reduce the
by-pass flux because there is a better spreading of the water flow, and therefore the
dispersion is reduced. During the residence time distribution tests there was biogas
production due to the digestion of dextrose in the synthetic wastewater, however, the biogas
production with or without dye in the water is not enough to consider the reactor as a
mixed tank. Besides, the rise of biogas through the bed is very slow, this generates by-pass
flow due to the biogas bubbles trapped, and when the superficial velocity in the reactor is
increased, the bubbles are pushed and can flow better and the by-pass flux is reduced; thus
the reactor became closer to a plug flow. The HRT was from 1.6 to 1.8 times the RTm at the
less volumetric flow in the reactor, and from 1.1 to 1.3 times at high volumetric flow. The
residence time distribution achieved for all the tests was fitted by a statistical distribution of
extreme value (Fisher-Tippet) shown by Equation 10; this is a frequency distribution
function for slant peaks.
P(t ) = y0 + a * Exp [ −Exp( − f 1) − f 1 * s + 1]

t − tC (10)
f1=
w
Where P(t) is the normalized tracer concentration, y0 represents the distribution
displacement (time 0 tracer concentration), a is the amplitude of the distribution and w the
wide of the peak. These calculated parameters, the error and correlation coefficient (p<5).
Figure 3 shows the residence time distribution in the reactor for each test and the fit by the
Extreme model. Iliuta et al (1996), states that a long tail in the residence time distribution
could be caused because of axial dispersion in the dynamic flow as well as the mass transfer
between the dynamic and static zones. Also, when the reactor has porous particles in the
packed bed, as in this case, the alteration in the distribution can be attributed to internal
diffusion and reversible adsorption of the tracer, however, this is negligible, but it happens
for the dye.
The reactive red 272 dye molecule is adsorbed reversible and superficially on to biomass,
also is adsorbed onto AC surface, but poorly, and has a limited adsorption capacity due to
its molecule size and chemical groups as naphthalene disulphonic acid, a choloro-triazine
and phenyl-amine (see Fig. 1); also AC has limited active sites for adsorption. There is an
equilibrium between adsorption, reaction and desorption. In this case it is considered a
constant saturation of the bioparticle, so the adsorption dynamics will be negligible.
1,0
P(t)1 E1
0,9
P(t)2 E2
0,8
P(t)3 E3
0,7
P(t)4 E4
0,6
P(t)5 E5
P(t) 0,5
P(t)6 E6
0,4
0,3
0,2
0,1
0,0
0 100 200 300 400 500 600 700 800
time, h
Fig. 3. HRT distribution; tracer pulse applied to the bioreactor.

Parameter HRT1 HRT2 HRT3 HRT4 HRT5 HRT6

RTm (min) 57.264 66.203 66.203 42.412 42.412 42.412
uL (cm/s) 0.0735 0.0631 0.0631 0.0985 0.0985 0.0985
NReL 54.962 47.468 47.468 74.096 74.096 74.096
BoL 11029.2 9461.54 9461.54 14769.2 14769.2 14769.2
dL 2.864 2.255 2.407 2.332 1.208 0.413
DL (cm2/s) 10.026 6.828 7.287 11.022 5.709 1.953
PeL 0.349 0.443 0.415 0.429 0.828 2.420
ShL 45.403 42.246 42.246 52.283 52.283 52.283
kL (cm/s) 0.0176 0.0164 0.0164 0.0203 0.0203 0.0203
y0 0.0270 0.0470 0.0251 7.6×10-3 9.3×10-3 5.5×10-4
a 0.5697 0.6629 0.6175 0.8944 0.6959 1.0029
w 22.033 46.312 33.212 29.893 19.756 32.259
SEa 0.0255 0.0443 0.0325 0.0182 0.0226 0.0162
R2 b 0.9945 0.9823 0.9920 0.9977 0.9966 0.9980
Table 2. Hidrodinamic, mass transfer and and fitted extreme

model Parameters for the UAFB reactor (L = fixed bed).
3.5 Dynamic model: Transport and reaction in the fixed bed

After the calculus of HRT and all the necessary parameters and dimensionless numbers, a
mathematical model was deduced and executed to represent the transport and reaction of
the dye in the water upflow through the reactor (the clarifier zone is not considered here
due to complexity of the process).
The dynamic model is based in the next assumptions:
1. Radial dispersion is negligible.
2. The reactor is divided in two zones: the fixed bed and the clarifier (not considered) and
there is mass transfer between them.
3. The dispersion coefficient is constant in each zone.
4. There is mass transfer between the water flowing through the bed and the bioparticles.
5. The superficial velocity trough the bed is constant and calculated as uL = Q ε Lπ Ri2 .
6. The dye can be reversible adsorbed into bioparticles.
7. The dye is diffused, adsorbed and reacts in the biofilm and there is superficial diffusion
in the activated carbon core macropores because the molecule size (21 Å) would limit
the micropore diffusion (pore average diameter 23.3 Å).
8. The particle is spherical and with a uniform and porous biofilm.
9. The biomass on the activated carbon core grow up to certain thickness, and afterwards
the biomass excess is detached and form small granules, then the biomass attached on
the activated carbon came back to their normal or equilibrium thickness. Thus, there is a
dynamic in the biofilm thickness but will be taken as an average value supposing
equilibrium between grow and detachment. The model includes the balance in the
bioparticle divided in two zones: zone I represents the AC particle (core) and zone II the
microorganism biofilm surrounding the AC core. The reaction term is included in two
balance equations, in the zone II of the bioparticle and in the liquid balance; this
because is an extracellular process and there are free cells or small biomass granules in
the flowing liquid. Figure 4 shows a graphical scheme of the model; here CAL is the
liquid concentration of the dye, CAP is the dye concentration in the AC core, and CAb the
biofilm concentration.
CAS C Az + Δ z
convection dispersion
dC A
LC = 57.5 cm Δz
dt
convection dispersion
CAi C Az
Lf = 48 cm S
RB
CA0
I
II r
UAFB RC d C Ap d C Ab
CAL CAb CAp ;
Reactor dt dt
δ=RB-RC
II I Bioparticle
Km
Fig. 4. Transport and reaction model in the UAFB, Reactor scheme.

The governing equations are:
a) Balance for the liquid flow in the fixed bed.
∂C AL ∂ 2C AL ∂C AL
= DL − uL − K m asb (C AL − C Ab r = RB ) − k1C AL + k2C AL (C A0 − C AL ) (11)
∂t ∂Z 2 ∂Z
This equation describes the dye convection, dispersion and mass transfer from the liquid
phase to the biofilm surface and the reaction. The initial and boundary conditions are:
t=0 CAL = CA0
Z=0 CAL = CA0 (12)
∂C AL
Z=Lf =0
∂Z
b) Balance for the bioparticle.
For the AC core:
∂C Ap ⎛ 2 ∂C Ap ∂ 2C Ap ⎞
= Dep ⎜ + ⎟ (13)
∂t ⎜ r ∂r ∂r 2 ⎟⎠
⎝
This equation describes the dye convection and diffusion. The initial and boundary
conditions that expresses field equality in the interface are:
t=0 CAp = 0
∂C Ap
r=0 =0 (14)
∂r
r = RC CAp = CAb
For the biofilm (diffusion and reaction):
∂C Ab ⎛ 2 ∂C Ab ∂ 2C Ab ⎞
= Deb ⎜⎜ + ⎟ − k1C AL + k2C AL (C A0 − C AL ) (15)
∂t ⎝ r ∂r ∂r 2 ⎟⎠
This equation describes the dye convection, diffusion and reaction. The initial and boundary
conditions that expresses flux equality in the interface and mass transfer from the bioparticle
to the liquid phase are:
t=0 CAb = 0
∂C Ap ∂C Ab
r = RC −Dep = −Deb (16)
∂r ∂r
∂C Ab
r = RB −Deb = K m (C Ab − C AL )
∂r
c) Dimensionless model. The dimensionless governing equations are shown next.
For the liquid flow in the fixed bed:
∂ωL ∂ 2ωL ∂ωL

= dL − − β m (ωL − ωb ) − Φ 12 FobωL + Φ 22 FobωL ( 1 − ωL ) (17)
∂τ ∂ζ 2 ∂ζ
τ=0 ωL = 1
ζ=0 ωL = 1 (18)
∂ωL
ζ=1 =0
∂ς
Using the dimensionless numbers:
C AL Z t tuL D 1 K m asb L f
ωL = ; ζ = ; τ= = ; dL = L = ; βm =
C A0 Lf tmL Lf uL L f PeL uL
δ 2 k1 δ 2 k2C A0 Deb L f
Φ 12 = ; Φ 22 = ; Fob = (19)
Deb Deb δ 2 uL
In order to become dimensionless the two zones of the bioparticle, it was proposed a parallel
model; the AC core and the biofilm are then expressed as a single particle with a radius
from 0 to 1 (see Fig. 5):
For the AC core:
∂ωp ⎛ 2 ∂ωp ∂ 2ωp ⎞

= Fop ⎜ + ⎟ (20)
∂τ ⎜ ξ ∂ξ ∂ξ 2 ⎟⎠
⎝
τ=0 ωp = 1
ξ=0 (21)
ξ =1 ωp = ωb
Fig. 5. Parallel dimensionless model.

C Ap DepL f t tuL
ωp = ; Fop = ; τ= = (22)
C A0 RC2 uL tmL Lf
For the biofilm:
∂ωb ⎡ ∂ 2ω ⎛ 2 ⎞ ∂ωb ⎤
⎥ − Φ 1 Fobωb + Φ 2 Fobωb (ωL − ωb )
2 2
= Fob ⎢ 2b + ⎜ ⎟ (23)
∂τ ⎢⎣ ∂ξ ⎝ξ + β ⎠ ∂ξ ⎥⎦
τ=0 ωb = 0
∂ωb ∂ω p
ξ=0 =α β (24)
∂ξ ∂ξ
∂ωb
ξ=1 + Bi ωb = Bi ωL
∂ξ
C Ab t tu δ 2 k1 δ 2 k2C A0 Deb L f
ωb = ; τ= = L ; Φ 12 = ; Φ 22 = ; Fob = 2
C A0 tmL L f Deb Deb δ uL
K m RB D RC R
Bi = ; α = eb ; β = = C (25)
Deb Dep RB − RC δ
The reactor model then consist of three differential parabolic equations; to solve the model,
all the parameters are calculated and is used the finite differences with a 5th order Runge-
Kutta-Fehlberg method, programmed in Fortran language.
3.6 Model solution.

The data that we had previous to the model solving are: from the AC properties: ρp = 0.435
g/cm3 (density), εp = 0.1392 (porosity), RC = 0.0515 cm (average particle radius), and about
the reactor we have: LL = 48 cm (longitude), Ri = 3 cm (radius), εL = 0.19 (bed porosity). The
dispersion coefficient (DL), the dispersion number (dL), the Peclet number (Pe), the
superficial velocity (uL) and the half residence time (RTm), were estimated in the residence
time distribution analysis, as was explained in the methodology section. The diffusion
coefficient in the AC core was calculated by the method explained in Hines and Hines and
Maddox (1987), supposing Knudsen diffusivity, but the value was changed in two orders of
magnitud in order to fit the model (Fan et al. 1990; Lee et al.2006), comparing the values
reported for other dyes in references. The diffusivity in the biofilm was assumed to be 100
times larger than the diffusivity in the AC core, based in values reported for other dyes of
the same kind (Fan et al. 1987; Chen et al. 2003). The kinetic constants were estimated from
experimental data and fitted to model.
The mass transfer surface of the bioparticles was calculated by the Equation 20 proposed by
Iliuta and Larachi (2005)
6 6
asb = = (26)
db dp + 2γ
The parameters used to solve the mathematical model are shown in Table 4; in this case, the
Thiele number of second order is applied for an initial dye concentration of 250 mg/L, for
400 mg/L was 1.43 and for 500 mg/L was 1.60. This indicates that when the dye
concentration is augmented in the reactor influent, the mass transfer effects are increased,
specifically the resistance to the diffusion, and therefore, it decreases dye removal. Figure 6
shows the concentration profile along the reactor to different dye concentration at the
reactor inflow, and Figure 7 shows the concentration profile within the bioparticle (AC core
plus biofilm) for an initial concentration of 250 mg/L.
It can be seen in the Figure 7 that the particles close to the reactor influent (ζ = 0.045) contain
higher dye concentration than the ones in the effluent extreme, whose concentration is close
to the dye effluent concentration. The concentration profile in the bioparticle changes with
time as the bioparticle is saturated and reaches equilibrium, while a curve indicative of the
reaction in the biofilm is observed. When the bioparticle is close to the effluent, the profile is
becoming flat, and the concentration is approximately uniform within the bioparticle. This is
because at the effluent zone there is the smallest dye concentration in the reactor height, and
the kinetics depends on this.
1,0
CA0 100 mg/L
CA0 250 mg/L
0,8 CA0 400 mg/L
CA0 500 mg/L
0,6
ω
0,4
0,2
0,0
0 0,2 0,4 0,6 0,8 1
ζ
Fig. 6. Predicted concentration profile along the reactor
DL (cm2/min) 423.4 Φ 12 a 1.061
u (cm/min) 3.786 Φ 22 b 1.276
Km (cm/min) 0.984 Φ1 c 1.030
Dep (cm2/min) 2.58×10-5 Φ2 d 1.130
Deb (cm2/min) 2.58×10-3 β 2
k1 (min-1) 3.046 α 100
k2 (L/mg min) 1.47×10-2 Bi 34.27
asb (cm2/cm3) 36.81 Bm 459.22
Fop 0.091 dL 2.33

Fob 36.404
Table 4. Parameters used in model solution.

τ = 0 .4 τ = 1
1.0 1.0 ζ =0.045
0.8 0.8 ζ =0.13
ζ =0.5
0.6 0.6
ζ =1
ω
0.4 0.4
0.2 0.2
0.0 0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.00 0.02 0.04 0.06 0.08 0.10
τ = 2 τ = 5
1.0 1.0
0.8 0.8
0.6 0.6
ω
0.4 0.4
0.2 0.2
0.0 0.0
0.00 0.02 0.04 0.06 0.08 0.10 0.00 0.02 0.04 0.06 0.08 0.10
bioparticle radius bioparticle radius
Fig. 7. Predicted concentration profile in the bioparticle at different τ and ζ in the bed.
Radius in cm.
The RTm in the reactor does not have much influence on the removal of the dye according
the predicted results of the model. Figure 8 shows the concentration profile along the reactor
at different conditions of RTm (See Tab. 3), and using a dye inflow concentration of 250
mg/L; it can be seen that there is not a noticeable effect in the concentration profile as RTm is
incremented. The concentration profile in the bioparticle is affected in a different way by the
RTm; in Figure 9 we can see that as RTm is increased, the bioparticles are saturated faster, this
is because the contact time for adsorption and for reaction is augmented and the mass
transfer between the liquid phase and the bioparticles is improved.
As a result, the RTm in the reactor affects only the mass transport rate in the reactor but not
the biodegradation reaction and thus does not have an influence on the concentration profile
along the reactor and consequently, in the removal efficiency. Therefore, analyzing the
profiles expressed in Figure 6, the main factor that affects the removal efficiency is the
inflow dye concentration.
Similar results were found by other authors. Spigno et al (2004) presented a mathematical
model for the steady state degradation of phenol along a biofilter reactor and obtained a
concentration profile for phenol reduction at two different concentrations, and they found
the same profile for both conditions displaced by a concentration gradient, more reduction
at the lesser concentration. Mammarella and Rubiolo (2006) could predict the concentration
profile for lactose hydrolysis in an immobilized enzyme packed bed reactor under different
operation conditions, and they obtained an asymptotic profile for lactose conversion along
the reactor height and obtained a much higher conversion at the lesser volumetric flow (100
mL/h); on the contrary, in this work the volumetric flow does not have much influence.
These authors did not include the dynamic of the pollutant inside the bioparticle, as is
shown in this work.
1,0 RTm
226.195 min
0,8 81.429 min
75.398 min
0,6 67.859 min
ω 54.287 min
0,4
0,2
0,0
0 0,2 0,4 0,6 0,8 1
ζ
Fig. 8. Predicted concentration profile along the reactor a different RTm
RTm (min) 226.195 81.429 75.398 67.859 54.287
Q (cm3/min) 6.000 16.67 18.00 20.00 25.00
u (cm/min) 1.117 3.102 3.351 3.723 4.654
Km (cm/min) 0.695 0.913 0.939 0.978 1.083
dL 7.899 2.843 2.633 2.370 1.896
βm 1099.8 519.7 495.1 464.4 411.5
Fop 0.308 0.111 0.103 0.093 0.074
Fob 123.4 44.42 41.13 37.02 29.62
Bi 24.21 31.78 32.70 34.07 37.75
Table 5. Parameters used in model solution at different RTm

Leitão and Rodrigues (1996) presented the influence of the biofilm thickness on the removal
of a substrate when the support carrier material is an adsorbent and when it is not, and they
obtained the concentration profile within the bioparticle but in the absence of reaction; the
profiles show the saturation of the bioparticle as time increases, and the concentration
augments as the biofilm thickness increase, but it was not observed the reaction zone in the
biofilm as in the results presented in this work. Leitão and Rodrigues (1998) proposed an
intraparticle model for biofilms on to a carrier material including the convective flow inside
the particle, and they obtained the concentration profile and biofilm thickness regarding
time. They conclude that bioreactors have to be operated under conditions such that allow
the liquid movement to occur in the void space of the biofilm, in order to improve mass
transfer and as a result the efficiency of the process.
0,9
0,8
0,7
0,6
RTm
ω 0,5 226.195 min
81.429 min
0,4
75.398 min
0,3 67.859 min
54.287 min
0,2
0,00 0,02 0,04 0,06 0,08 0,10
bioparticle radius
Fig. 9. Predicted concentration profile in the bioparticle at different RTm.

Radius in cm.τ = 2 and ζ = 0.045 (close to the influent) in the bed.
In biofilms the hydrodynamics and kinetics of the system is related to the fact that most
biofilm reactions are diffusion limited, therefore the shape of the concentration profiles will
be determined by diffusivity and convection (Lewandowski, 1994). We suppose that the
biofilm is a continuous phase, but it is not, biofilms are formed by clusters, void spaces and
channels, so the convection in the system is important, and the measured diffusion
coefficients are always approximations. The effect of convection in the bioparticle is shown
in Figure 9; mass transfer in the biofilm is fast but in the core the RTm affects the mass
transfer and as a result the concentration profile.
1,0
Lf = 48 cm
Lf = 60 cm
0,8
Lf = 100 cm
Lf = 120 cm
0,6 Lf = 148 cm
ωL
0,4
0,2
0,0
0 0,2 0,4 0,6 0,8 1
ζ
Fig. 10. Predicted concentration profile along the reactor, increasing
the reactor bed height (Lf). CA0 = 250 mg/L at the inlet.
The effect of the bed height is shown in Figure 10, with a constant inlet dye concentration of
250 mg/L. Here we can see that the dimensionless dye concentration is reduced as the
longitude of the bed height (Lf) is increased, this means higher removal efficiency; a higher
Lf implies more bioparticles and time for degradation reaction, however the most of the
reduction of the dye is carried out in the lower third of the bed longitude.
3.7 Effectiveness factor.

The effectiveness factor (η) was calculated at different azo dye concentration at the reactor
influent, from 100 to 500 mg/L, and at different RTm from 54.3 to 226.2 min. Each calculation
was carried out taking the predicted dimensionless concentration value in the biofilm and in
the liquid, and solving the model for a τ = 2. The results explain that the average reaction
rate in the biofilm and at the biofilm surface is changed regarding the height of the reactor,
and in consequence the effectiveness factor; furthermore, the effectiveness factor depends
mainly on the azo dye concentration. The change of the reaction rate regarding the reactor
height is explained because there is a higher amount of active biomass at the inlet, at the
bottom of the reactor. In this zone, dye and substrate concentration (dextrose and yeast
extract) is always higher and in consequence there is always a major reaction activity;
therefore, in this zone takes place the most of the degradation of the dye, as it was also
noticed in Figure 6. The calculated effectiveness factor decreased from 0.78 to 0.47 when
dye concentration was increased from 100 to 500 mg/l, proving that, when dye
concentration is increased the diffusion of dye through biofilm has a major influence on the
reaction rate, what is reflected in a reduction of the efficiency of removal. On the contrary,
changes in RTm in the reactor at a constant inflow dye concentration did not influence the
value of the effectiveness factor; this did vary from 0.74 to 0.73 when RTm was decreased
from 226.2 to 54.3 min. These values indicate that there is a slight effect of dye diffusion on
the reaction rate when RTm is changed.
3.8 Conclusions
The proposed mathematical model for a UAFB bioreactor was able to predict the
concentration profiles along the reactor and within the bioparticle (carbon core and biofilm)
for the biodegradation of a reactive red azo dye, using a kinetic model with a change in
reaction order. The profiles at different inflow dye concentration showed an asymptotic
curve with a major activity of reaction in the lower zone of the reactor, and this
concentration fall is reduced as the inflow dye concentration is augmented. Nevertheless,
the RTm does not have much influence on the concentration profile along the reactor. In
addition, the model predicts a larger removal rate as the longitude of the bed is increased for
the same inlet dye concentration.
The profiles within the bioparticle illustrate the saturation of the particle and reflect the zone
of reaction in the biofilm; it can be seen the differences in the concentration values with
regard to the reaction zone along the reactor. The saturation rate of the bioparticles change
with the RTm, at a larger time, the mass transfer is improved and the bioparticles are
saturated faster, without affecting the reaction. The calculation of the effectiveness factor
showed that the rate of reaction is changed in regard to the position at the height of the
reactor and depends of the dye diffusion when the concentration is increased. By means of
the presented dynamic model it can be predicted the dye and COD removal rate in a UAFB
reactor, specifying its characteristics, dye inflow concentration, and residence time.
3.9 Nomenclature
asb Specific surface of the particles, cm2/cm3
C, CA Dye or tracer concentration, mg/L
C0,CA0 Initial tracer, dye concentration, mg/L
Cm Average concentration, mg/L
d Dispersion number
D Axial dispersion coefficient, cm2/s
De Effective diffusivity coefficient, cm2/s
Km Mass transfer coefficient, cm/s
k1 1st order specific reaction rate, h-1
k2 2nd order specific reaction rate, L/mg⋅h
Lf Reactor bed height, cm
Q Volumetric flow
rA Reaction rate, mg/L⋅h
RB Bioparticle radius, cm
RC AC core radius, cm
Ri Reactor internal radius, cm
S Crosswise area to the flux, cm2
HRT Hydraulic residence time
t Time
tm, RTm Half residence time
VP Pore volume
u Superficial velocity
ρ Density, g/cm3
ε Porosity
δ Biofilm thickness
ζ Dimensionless longitude
ξ Dimensionless particle radius
ω Dimensionless concentration
τ Dimensionless time
Fo Characteristic Fourier number
Bi Biot number
Φ2 Wagner number
Φ Thiele number
βm Dimensionless parameter
α Dimensionless parameter
β Dimensionless parameter
Subindex
b biofilm
L fixed bed
p AC particle
1 For the first order term
2 For the second order term
4. Case III. A method to evaluate the isothermal effectiveness factor for

dynamic of oxygen in mycelial pellets in submerged cultures
4.1 Introduction
Although a lot of research has been done into modelling microbial processes, the
applicability of these concepts to problems specific for bioreactor design and optimization of
process conditions is limited. This is partly due to the tendency to separate the two essential
factors of bioreactor modelling, i.e. physical transport processes and microbial kinetics. The
deficiencies of these models become especially evident in industrial production processes
where O2 supply is likely to become the limiting factor, e.g. production of gibberellic acid
and other organic acids (Takamatsu et al. 1981; Qian et al. 1994; Lu et al. 1995; Hollmann,et
al.1995). Important physical transport processes related to O2 supply have not always
received sufficient attention (el-Enshasy et al. 1999; Yamane and Shimizu, 1984; Wang and
Stephanopoulos, 1984; Parulekar and Lim 1985; Sharon et al. 1999; Stephanopoulos and
Tsiveriotis, 1989). Attempts to obtain optimal trajectories for various control variables in
different bioreactors, in particular at different scale of operation, are likely to cause
unrealistic conditions far from the expected optimum. Another reason to combine the
intrinsic O2 kinetics for growth and product formation with mass transfer is related to O2
gradients on production scale. Recent attempts to account for these effects in stirrer
bioreactors use different structured models and include the O2 supply in viscous mycelial
fermentation broths (Reuss et al. 1986; Nielsen and Villadsen, 1994; Sunil and Subhash, 1996;
Cui et al. 1998; Goosen, 1999; Fan et al.1996). It is important when evaluating the potential of
a fermentation process using mycelial pellets to have a mathematical model that can predict
the O2 consumption, fungal growth rates and substrate consumption.
In this study, we present a model that combines transport phenomena and kinetic
mechanisms for O2 consumption and applied it to a process in which gibberellin was
produced in a submerged fermentation with mycelia of Gibberella fujikuroi. The effectiveness
factors found in this work are compared to those reported in literature for other fungi.
4.2 Mathematical modelling

The mathematical model was developed assuming a bioreactor in which the continuous
phase was well mixed and experimentally verified. O2 concentration in the bioreactor was
measured so the model only considered O2 transport in the pellets. The method of volume-
averaging was used to simulate bulk transport in a two-phase system (Carbonell and
Whitaker, 1984; Ochoa, 1986):
∂C
+ ∇.( vC ) = D : ∇∇C + k o ρ (1)
∂t
with C the dissolved O2 concentration (kg-moles O2 m-3), t time (h), ko the specific O2 uptake
rate per unit dry mycelial weight (kg-moles O2 kg-1 of dry cell h-1) and ρ the pellet
suspension density (kg m-3).
The use of average transport equations imposes constraints on the length and
physicochemical parameters of the system (Whitaker, 1991). The following assumptions
were made to model the O2 diffusion and reaction in the pellets:
1. The pellet is an effective system, isotropic with constant thermodynamic properties.

Effective diffusivity of O2 in porous medium can be used and the scalar Deff or the
effective diffusivity coefficient of dissolved O2 in mycelial pellet (m2 h-1) replaces the
tensor D of eq 1. Effective diffusivity contains the convective effects generated inside
the pellets.
1.03 mm.
2. Deff is only a function of void fraction as proposed by different authors (e.g. Aris, 1975).
3. Mycelial pellets are spherical, and the O2 transport occurs only in the radial direction.
4. An analogous Michaelis-Menten model was used to describe the O2 consumption rate:
( k O )max C
kO = (2)
Km + C
where (ko)max is the maximum specific O2 uptake rate per unit dry mycelial weight (kg-
moles O2 kg-1 of dry cell h-1) and Km the apparent Michaelis constant for mycelia (kg-moles
m-3) and the transport equation (eq 1) can be reduced to :
∂C ⎡ ∂ ⎛ 2 ∂ C ⎞ ⎤⎥
= Deff ⎢ 12 ⎜r ⎟ − ko ρ (3)
∂t ⎢ r ∂ r ⎜⎝ ∂ r ⎟⎠ ⎥
⎣ ⎦
with r the radial distance from the centre of the mycelial pellet (m). Boundary conditions for
eq 3 were:
∂C
1 @ r=R − Deff = k P (CS - CL ) t>0 (4a)
∂r
∂C
2 @ r=0 =0 t>0 (4b)
∂r
with R the radius of the mycelial pellet (m), kP the mass-transfer coefficient for the liquid
film around cells or pellets (m h-1), CL the concentration of dissolved O2 in bulk of liquid (kg-
moles O2 m-3) and CS the concentration of dissolved O2 at the liquid-pellet interface (kg-
moles O2 m-3).
Initial conditions were:
t=0 C = Co 0≤r≤R (5)

with CO the initial concentration of dissolved O2 (kg mole O2 m-3).
The dissolved O2 concentration was monitored during the fermentation and is expressed as:
CL = Co f (t).
Using the dimensionless variables:
C r Deff t
u= ξ= τ= (6)
Co R R2
and introducing them into eqs 3 to 5 gives the dimensionless boundary value problem:
∂u 1 ∂ ⎛ 2 ∂u ⎞ 2 u
= ⎜ξ ⎟−φ (7)
∂ τ ξ2 ∂ξ ⎝ ∂ ξ ⎠ β+u
with φ the Thiele modulus and subjected to the following :
∂u
at ξ=1 − = NSh (u − u L ) (8a)
∂ξ
∂u
ξ=0 =0 (8b)
∂ξ
When τ=0 u=1 0≤ξ≤1 (9)
R 2( k O )max ρ kpR Km
Where φ2 = NSh = β= (10)
Deff COεφφ Deff CO
with NSh the Sherwood number and uL the dimensionless O2 concentration when the
external mass transfer resistance was not neglected.
From the mass balance in the pellet we obtain:
∂u 1 ∂ ⎛ 2 ∂u ⎞
= ⎜ξ ⎟−ℜ (11)
∂τ ξ2 ∂ξ ⎝ ∂ξ ⎠
with the reaction rate ( ℜ )defined by:
u
ℜ = φ2 (12)
β+u
Whitaker (Whitaker, 1984; 1991) defines the volume-averaging function ( Ψ ) as:
4π rp 1
Ψ r 2dr = 3∫0 Ψ ξ2dξ
4 3 ∫0
Ψ= (13)
πrp
3
with rp the radius of one pellet (m) and Ψ. The mean O2 concentration ( u ) in the pellet was:
1
u = 3∫0 u ξ2dξ (14)
Substituting eq 14 in eq 11 gives:
∂u ξ= 1 ∂ ⎛ 2 ∂ u ⎞
= 3∫ξ= 0 ⎜ξ ⎟−ℜ
∂τ ∂ξ ⎝ ∂ ξ ⎠
∂u ∂u ξ= 1 ∂u ∂u
= 3ξ2 ξ= 0 −ℜ = 3 ξ =1 −ℜ = (15)
∂τ ∂ξ ∂ξ ∂τ
and then the mean reaction rate ( ℜ ) is defined by :

1
ℜ = 3∫0 ℜξ2 d ξ (16)
The effectiveness factor for O2 consumption rate per unit of mycelial pellet (η) is defined as:
1 r
4π∫0 ℜξ 2dξ 4π∫0p ℜ R 2dr
η= = (17)
4 4
πℜ u L π R 3 ℜ uL
3 3
1 ⎛ β + uL ⎞
And η= ⎜ ⎟ℜ (18)
φ2 ⎝ u L ⎠
Using eq 15 and eq 18 gives:
1 ⎛ β + uL ⎞⎛ ∂ u ∂u ⎞
η= ⎜ ⎟⎜3 ξ= 1 − ⎟ (19)
φ2 ⎝ u L ⎠⎝ ∂ ξ ∂τ ⎠
By solving eq 7 to 15, ∂u/∂ξ can be determined and used to calculate the effectiveness factor
with eq 19. Eq 7 was discretized in radial direction with 13 orthogonal collocation points,
using Legendre polynomials (Finlayson, 1980). The set of ordinary differential equations
generated was solved with the Runge-Kutta-Fehlberg method with an adaptive control of
each step.
Micro-organism. Gibberella fujikuroi (Sawada) strain CDBB H-984 conserved in potato
glucose slants at 4oC and sub-cultured every two months was used in the experiment
(Culture collection of the Department of Biotechnology and Bioengineering, CINVESTAV-
IPN, Mexico).
Culture medium. The culture medium contained 100 g of glucose l-1, 3 g of NH4Cl l-1, 5 g of
KH2PO4 l-1, 1.5 g of MgSO4 l-1, 2 g of rice flour l-1.
Culture conditions and equipment. The fungus was cultured on potato dextrose agar
(PDA) slants at 29ºC for seven days. A 1000 ml Erlenmeyer flask containing 500 ml of
medium was inoculated with spores and mycelium taken from the slants and incubated on a
rotary shaker at 180 rpm and 29ºC for 36-38 h. A 30 dm3 turbine-agitated fermenter
(Chemap A.G., Zurich) containing 20 dm3 of sterilized culture medium (pH 5) was
inoculated with 5 % v/v of this culture. The aeration rate was 1 volume air volume-1 of
medium min-1 (vvm), the temperature and agitation-speed were automatically controlled at
29oC and 700 rpm, respectively. pH and dissolved O2, measured with three polarographic
electrodes (Ingold, USA) installed at different depths in the culture medium, were
monitored each h for 7 days. Every two h, 60 ml of medium was sampled and analysed for
biomass, density (ρ) and diameter of the wet and dry pellet, reductive sugars, NH4+-N and
gibberellic acid concentrations.
Pellets characterisation. The pellets in the sub-sample of the medium were filtered, washed
twice with distilled H2O and dried to constant weight at 90ºC in a vacuum-oven. The
fermented broth was centrifuged in conical graduate tubes at 3000 rpm for 20 min and
density, volume and weight of the wet pellets were determined while their diameter was
measured with a microscope (Leica, MSD) on a calibrated micrometer grid. The pellets were
vacuum dried at 90ºC for 16 h and their dry weight measured.
Volumetric mass-transfer coefficient (kLa). The gas flow rate was measured with a Brooks
Mass controller 5851E while O2 and CO2 were monitored at the in and outlet with a
paramagnetic O2 analyser (Sybron 540A) and infrared CO2 analyser (Sybron, Anatek PSA
402). The volumetric mass transfer coefficient obtained at maximum pellet concentration
(kLa) (h-1) was derived from the O2 mass balance in the bioreactor (Sano et al. 1974).
O2 uptake rate. Different concentrations of dissolved O2 in the bioreactor were obtained by
changing the compositions of the inlet air while keeping agitation speed and volumetric gas
flow rate constant. The rate of O2 uptake was determined by measuring the O2
concentrations at the in and outlet and, as such, kinetics of O2 were obtained without
disturbing the system, i.e. power supply and gas hold-up (Wang and Fewkes, 1977).
Mixing time. The model assumed perfect mixing and two methods were used to verify this.
First, the bioreactor with agitation speed of 700 rpm, a temperature of 29oC and an airflow of
1 vvm was filled with 0.1 M NaOH and phenolphthalein as a tracer. Samples were taken
every 10 to 15 s at four different depths in the bioreactor (A, B, C, D), and analysed for
absorbance at 550 nm (Figure 1). Second, a culture of G. fujikuroi in its maximum growth
phase to which dextran blue was added as a tracer, was sampled every 10-15 sec at four
different depths in the bioreactor and analysed for absorbance at 617.1 nm. Dextran blue
was used as it is not affected by pH or by oxide-reduction processes, which take place
during fermentation.
The distribution ages were determined by fitting the normalized equation (Levenspiel, 1999):
∞ ∞ A
∫0 Adt = ∫0
Q
= dt (20)
∞
To the dynamics of the tracer with Q = ∫0 Adt the area under the curve of absorbance. A is
absorbance of the tracer and t is time. The mixing grade was determined by:
⎛ A− A ∞ ⎞
m=⎜ ⎟ 100 (21)
⎝ A∞ − A0 ⎠
Fig. 1. Diagram of the bioreactor

4.3 Parameter estimation

kLa determinations. The O2 transfer rate (OTR) was derived from:
7.32 × 105 ⎡ Qi Pi yi Qo Po yo ⎤
OTR = ⎢ − ⎥ (22)
VL ⎣ Ti To ⎦
where 7.32×105 is a conversion factor (60 min h-1) [mole (22.4 dm3)-1(standard conditions of
Temperature and Pressure)] (273º K atm-1), Qi and Qo is the volumetric air flow rate at the air
in and outlet (dm3 min-1), Pi and Po is the total pressure at the bioreactor air in and outlet
(atm absolute), Ti and To is the temperature of the gases at the in and outlet (ºK), VL is the
volume of the broth contained in the vessel in dm3, and yi and yo is the mole fraction of O2 at
the in and outlet (Wang et al. 1979).
The experimental values of kLa obtained from the G. fujikuroi culture were used to determine
the volume fraction (θp) of the pellet using the empirical equation (Van Suijdam, 1982):
k La
( k L a )0
(
= 0.5 ⎡⎣1 − tanh 15θp − 7.5 ⎤⎦ ) (23)
with (kLa)o the initial volumetric mass transfer coefficient (h-1).

The liquid to pellet mass-transfer coefficient (kpap) was calculated using the Sano,
Yamaguchi and Adachi correlation (Sano et al. 1974). This correlation is based on
Kolmoghorov’s theory of local isotropic turbulence and is independent of the geometry of
the equipment or the method energy input used. The Sherwood number NSh is:
1 1
NSh = 2.0 + 0.4 ( N Re ) 4 ( NSc ) 3
(24)
where NRe is the Reynolds number and NSc the Schmidt number.
NSh is given by:
k pd p
NSh = (25)
Deff
with kp is defined by eq (4a) and dp is the diameter of the pellet (m). NRe is defined as:
∈ d 4p
NRe = (26)
ν3
where ∈ is the mean of local energy dissipation per unit mass of suspension (W kg-1) and ν
is the kinematics viscosity of the suspending medium (9.18×10-6 m2 s-1). NSc is equal to νDL-1
and approximately 3991 with DL the molecular diffusion coefficient of dissolved O2 in H2O
(m2 h-1).
∈ in the impeller jet stream can be given as a function of the distance from the impeller
shaft (ris), the stirrer speed (N), and the stirrer diameter (DR) (Van Suijdam and 1981, Metz):
0.86N3D6R
∈= (27)
ris4
∈ obtained was 140 W kg-1; acceptable for inter-medium viscosity in the region of the
impeller as the mycelial pellet suspensions showed Newtonian characteristics. The specific
surface area of the these pellets (ap) was estimated using
6θ p
ap = (28)
dp
The value for the liquid-solid mass transfer coefficient was estimated using eqs 25 to 30 with
dCS (29)
= kP aP ( C L − CS ) − k o
dt
with ko the mean O2 consumption rate per unit of mycelial pellet (kg-moles of O2 kg-1 of
dry cell h-1). Experimental radii, pellet density, maximal O2 uptake rate and the effective
diffusivity coefficient (Deff) were used to calculate the Thiele modulus (eq 10).
O2 uptake. The O2 uptake rate was derived from the measured inlet gas flow rate ( V α ),
G
volume of the broth contained in the vessel (VL), and gas compositions at the in and outlet
using the gas balance taking into account the differences in inner and outlet gas flow rates:
α ⎡ α
V ω
α ⎞⎤
⎛ 1 − YOα2 − YCO
ko = G
⎢ YO2 − YO2 ⎜⎜ ω ω
2
⎟⎟ ⎥ (30)
VL ⎣⎢ ⎝ 1 - YO2 − YCO2 ⎠ ⎦⎥
where YO2 and YCO2 are the volume fractions of O2 and carbon dioxide in gas (α = inlet, ω =
outlet).
Effective diffusivity estimation. Miura (Miura 1976) assumed that the effective diffusion
coefficient is proportional to the void fraction within the pellet
Deff = DL ε (31)
with DL being 9×10-6 m2 h-1 at 29ºC (Perry, 1997). Although eq 31 implies only the rectilinear
paths inside the particles, similar results have been obtained with other empirical equations
that consider tortuosity (Riley et al.. 1995; Riley et al. 1996) or intra-particle convection
(Sharonet al. 1999).
Void fraction (ε) was defined as:
ρv
ε =1− (31)
ρc
where ρc is the density of the dry pellet (kg m-3) and ρv is the density of the wet pellet (kg m-
3).
Both were experimentally determined.
The intrapellet Peclet number (Pein):
⎛χ⎞
Pein ≅ ⎜ ⎟ Peout (32)
⎝ε⎠
was calculated to estimate the contribution of intrapellet convection (Parulekar and

Lim,1985). The extra-Peclet number Peout is defined by:
3 (33)
⎛ NSh ⎞
Peout ≅ ⎜
⎜ 0.6245 ⎟⎟
⎝ ⎠
where the dimensionless number χ is defined as:
κ
χ= (34)
d 2p
where κ is the hydraulic permeability of the pellet (m2) and estimated through Johnson's
equation (Johnson and Kamm, 1987):
κ −1.17
= 0.31( θP ) (35)
rp2
Numerical method. To fit the experimental oxygen uptake values with the non-linear ξ with
parameters φ (involving (ko)max) and β (involving Km), a least square algorithm coupled with
the discretization of eq 7 via orthogonal collocation using Legendre polynomials and Runge-
Kutta-Fehlberg methods was used (Jiménez-Islas et al. 1999). The set of non-linear equations
derived in the minimization process, are solved with the Newton-Raphson method with LU
factorisation. The optimization sequence is shown in Figure 2.
Experimental data ko vs time Initial values of parameters φ Model given by eq (7), with
and β boundary and initial conditions
Statistical analysis for assessing Nonlinear optimization via

data confidence least squares
Discretization of radial New values of φ and β given by

coordinate (ξ) by orthogonal Newton algorithm
collocation
Least square algorithm and

formation of normal equations
Time integration by Runge-

Kutta-Felhberg method
Solution of normal equations

by Newton´s method with LU
factorization
Yes No
The minimization method
converges?
Optimized parameters
End
φ and β
Fig. 2. Flow diagram for the optimization of the parameters φ and β (eq. 7).
4.4 Results and discussion

The bioreactor was well mixed (Figure 3). G. fujikuroi grew in dispersed mycelia (10%) or in
the form of pellets (90%) within 38 h of culturing. The mean size of the pellets increased
from 39 to 60 h and remained constant thereafter (Table I). The density of the pellets
increased and gave a maximum after 82 h whereupon it decreased. O2 uptake rates were
simulated using eq 7 with a program specifically written for this purpose and the
parameters were varied to fit the experimental data (Figure 4). These results included the
resistance effects in the Michaelis-Menten equation (eq 3) not optimised before in this way.
The estimated values for (ko)max were 1.80×10-4± 3.05×10-6 kg mole kg-1 dry cell h-1 and for Km
2.49×10-5 ± 2.28×10-6 kg-moles m-3 (Table II). These values are similar to those reported for
Aspergillus niger (Miura et al.1975) and Aspergillus orizae (Kim et al. 1983) but lower than
Fig. 3. Tracer absorbance of phenolphthalein measured at 550 nm (●) and dextran blue
measured at 617.1 nm ( ) used to verify the mixing behaviour in the bioreactor.
Fermentation time Size of the pellet Pellet density

(h) (×103 m) (kg m-3)
39 1.50 (0.30)† 16.0 (0.90)
45 1.67 (0.40) 17.5 (0.75)
60 1.90 (0.32) 18.0 (0.50)
66 1.91 (0.28) 18.8 (0.23)
82 2.05 (0.51) 20.7 (0.45)
95 1.92 (0.40) 20.3 (0.30)
108 1.92 (0.30) 19.5 (0.45)
132 1.92 (0.29) 18.4 (0.23)
† values between parenthesis are standard deviations of five replicates
Table I Size and density of Gibberella fujikuroi pellets during fermentation.
those obtained for Penicillium chrysogenum (Aiba, S.; Kobayashi,1975; Kobayashi et al. 1973).
Differences between simulated and experimental data were less than 6 % and differences
can be due to:
1. O2 transfer rate in the mycelial pellet increases with agitation (Miura and Miyamoto
1977),
2. mycelial density is not uniform (Miura, 1976),
3. respiratory activity is not uniform in radial direction within the pellet (Wittler et al.
1986),
4. and internal convection (Sharon 1999).
The importance of each of these factors has not been assessed separately but they are
indistinguishable in a model using Deff and a homogeneous pellet. A summary of
experimental and estimated parameters of O2 diffusion in a bioreactor with G. fujikuroi (eq 7
to 34) is given in Table II. Deff was derived from eqs 30 and 31 and is comparable to values
reported in literature for other fungi. θp values below 30 % did not affect kLa values but they
decreased when θp values were between 40 % and 60 % (Figure 5). The calculated θp value
for pellets of G. fujikuroi was 39.8 % and allowed calculation of κ (eq 35) and Pein (eq 32). Pein
for G. fujikuroi was 1.38 and κ was 8.22×10-7 m2 (Table II). Stephanopoulos and Tsiveriotis
(Sharon et al 1999) stated that the O2 flow through the pellet does not affect the external
mass transfer when Pein was close to 1 as found in this study. A constant Deff can thus be
assumed in our model. O2 concentration derived from numerical solutions of eq 7 indicated
that φ = 1 gave an overall reaction rate of O2 lower than the diffusion rate.
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Parameter value Dimension Remarks
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
-3
dp 2.090×10 m Experimental data
Deff 4.15×10-6 m2 h-1 estimated from eq 31
∈ 139.96 W kg-1 estimated from eq 27
Km 2.49×10-5 (700 rpm) kg mol O2 m-3 fitted from experimental
data of Figure 4
-1 -1
(ko)max 1.80×10-4 (700 rpm) kg mol O2 h kg dry pellet fitted from experimental
data of Figure 4
kLa 91.93 h-1 Experimental data
-1
(kLa)o 188.92 h Experimental data
NRe 2.36×106 dimensionless estimated from eq 26
κ 8.22×10-7 m2 estimated from eq 36
Peout 5.856 dimensionless estimated from eq 34
Pein 1.38 dimensionless estimated from eq 33
R 0.95×10-3 m Experimental data
NSh 250.6 dimensionless estimated from eq 25
φ 1.12 to 2.4 dimensionless estimated from eq 10
θp 0.398 dimensionless estimated from Figure 5
ν 9.18×10-6 m2 s-1 Experimental data
ρ 18.65 kg m-3 Experimental data
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Table II Summary of experimental and estimated parameters used for the solution of eqs 7
to 26.
The O2 concentration did not change substantially in the pellet when φ > 5 and the O2
uptake was limited by diffusion and by mycelial activity. O2 is then mostly consumed in the
external core of the pellet.
1.8
(kg-moles O2 kg-1 dry weight h-1 )

1.6
1.4
O2 uptake×10-4
1.2
0.8
0.6
0.4 ko experimental
0.2 ko simulated (eq 7)
0
0 1 2 3 4
CL ×10-4 (kg-moles O2m-3)
Fig. 4. Measured (♦) and simulated (⎯) O2 uptake (kg-moles kg-1 dry weight h-1) by
Gibberella fujikuroi in function of the O2 dissolved in bulk liquid (kg-moles m-3).
1.2
1
kL a / (kLa)0
0.8 Simulated
0.6 Miura
This work
0.4
Reub
0.2 Sano
0
0 20 40 60 80
θP (%)
Fig. 5. Simulation of relationship between the dimensionless gas-liquid mass-transfer

coefficient kLa (kLa)o–1 and the volume fraction of Gibberella fujikuroi pellets.
Experimental values for φ in fermentation with G. fujikuroi varied between 1.125 to 2.4
(Figure 6). The transport within the pellet depends on both diffusion and kinetics of the O2
reaction. The mycelial activity in the inner zone of the pellet was reduced by O2 limitation.
Our model predicted that for φ<1.875, η was close to 1 (Figure 7), consistent with other
model predictions (Miura, 1976). Under these conditions, the respiratory activity is not
limited by O2 transport. For φ>1.875, η is inversely proportional to φ. The estimated φ for G.
fujikuroi indicated a small limitation of O2 diffusion into the pellet. The large agitation rates
and the small size of the pellet formed could explain this.
1.00
φ= 0.5
0.90 φ= 1.0
Dimensionless O2 concentration (u)
0.80 φ=1.13
φ = 1.78
0.70
φ= 1.40
0.60
φ = 1.88
φ= 2.4
0.50
0.40
0.30 φ= 3
0.20 φ=5 φ=7
0.10
0.00
0.00 0.20 0.40 0.60 0.80 1.00
Dimensionless radius ( ξ)
Fig. 6. O2 concentration (u) in the pellet in function of the dimensionless radius (ξ) and the
Thiele modulus (φ) with theoretical values of 0.5, 1.0, 3, 5 and 7 and other values calculated
from experimental data 1.13 (5), 1.40 (Δ), 1.78 (ο), 1.88 (•), and 2.4 ().
1.2
Effectiveness factor(η)
experimental
1 simulated
Miura
0.8 Aiba
Kobayashi
0.6 Yano
0.4
0.2
0
0 5 10 15
Thiele modulus φ
()
Fig. 7. Effect of Thiele Modulus (φ) on the effectiveness factor for mycelial pellets (η) as
measured (♦) and simulated (⎯) in this experiment for Gibberella fujikuroi and as reported
by Aiba et al. (30) (t), Kobayashi et al. (31) (Δ), Miura et al. (28) (o) and Yano (38) (•).
Data from different authors were recalculated and expressed for φ in function of η (Figure
7). The effectiveness model was used to simulate those data and only η obtained with the
data reported by Miura (Miura 1976) were comparable with those η values found in this
experiment. A possible explanation is that Miura (Miura 1976) used a Michaelis-Menten
type kinetic to calculate Km and (ko)max while the other authors used a zero and first-order
kinetic resulting in values that were unrealistically large. η was not limited by transport for
τ < 0.2 (45.7 h) (Figure 8). After that, limitation of O2 diffusion into the pellet started and a
minimum for η was found for τ 0.8 (183.1 h). η remained constant thereafter (Wittler 1986).
Fig. 8. Typical effectiveness factor through the dimensionless time for a representative
experiment (pellet size ≥2 mm, air flow rate 1 vvm, 700 rpm, 29 ºC, Thiele modulus 2.8).
4.5 Conclusions
Limitations in models simulating O2 transfer into mycelial pellets with different strains of
fungi have been reported, e.g. Sunil and Subhash, 1996; Miura, 1976; Aiba and Kobayashi,
197; Metz and Kossen, 1977; Chiam and Harris, 1981; Reuss et al. 1982; Nienow 1990.
Explanations for these shortcomings can be related to unrealistically large values for Deff, Km
and (ko)max Experimental data of O2 diffusion into pellets of G. fujikuroi were simulated
satisfactorily. The O2 reaction rate in pellets of 1.7-2.0 mm was only marginally inhibited by
diffusion constraints under the conditions tested. Pein was small enough to justify a constant
effective diffusivity and an isotropic pellet system with constant thermodynamic
characteristics. O2 transfer into the mycelial pellet can become the limiting factor in
submerged fermentation of fungi when pellets larger than 2 mm are formed in the
bioreactor. Eqs 7 and 19 allows to identify conditions critical for fermentations and to derive
values for process parameters.
4.6 Nomenclature
aP = specific surface area of pellets (m2)
C = concentration of dissolved O2 (kg-moles O2 m-3)
CO = initial concentration of dissolved O2 (kg-moles O2 m-3)
CL = concentration of dissolved O2 in bulk of liquid (kg-moles O2 m-3)
CS = concentration of dissolved O2 at liquid-pellet interface (kg-moles O2 m-3)
dP = diameter of the pellet (m)
DL = molecular diffusion coefficient of dissolved O2 in H2O (m2 h-1)
Deff = effective diffusivity coefficient of dissolved O2 in mycelial pellet (m2 h-1)
DR = stirrer diameter (m)
kO = specific O2 uptake rate per unit dry mycelial weight (kg-moles of O2 kg-1 of dry
cell h-1)
kO = mean O2 consumption rate per unit of mycelial pellet (kg-moles of O2 kg-1 of dry
cell h-1)
(kO)max = maximum specific O2 consumption rate per unit dry mycelial weight (kg-moles of
O2 kg-1 of dry cell h-1)
kP = mass-transfer coefficient for the liquid film around cells or pellets defined by eq
4a (m h-1)
kP aP = liquid to pellet mass-transfer coefficient (m2 h-1)
kLa = volumetric mass transfer coefficient obtained at maximum pellet concentration
(h-1)
(kLa)0 = initial volumetric mass transfer coefficient (h-1)
Km = apparent Michaelis constant for mycelia (kg-moles m-3)
N = stirrer speed (rpm)
NRe = Reynolds number
NSc = Schmidt number
Nsh = Sherwood number
⎛ χ ⎞
Pein = intraparticle Peclet number ⎜⎜ ⎟⎟ Peout
⎝ (ε) ⎠
3
N
Peout = extra-Peclet number ⎛⎜ Sh ⎞⎟
⎜ 0.6245 ⎟
⎝ ⎠
Pi Po = the total pressure at the bioreactor air in and outlet (atm absolute),
Qi Qo = the volumetric air flow rate at the air in and outlet (dm3 min-1)
r = radial distance from centre of mycelial pellet (m)
R = radius of mycelial pellet (m)
ris = radius from the impeller shaft (m)
rp = radius of one pellet (m)
t = time (h)
Ti To = the temperature of the gases at the in and outlet (ºK)
u = dimensionless concentration of O2 defined in eq 6
u = dimensionless mean concentration of O2 defined in eq 14
uL = dimensionless O2 concentration when the external mass transfer resistance was
not neglected defined in eq 8a
α
V = gas flow rate (m3 h-1)
G
VL = volume of broth contained in the vessel (dm3)

yi yo is the mole fraction of O2 at the in and outlet
α
YO 2
= are the volume fractions of O2 in gas (α = inlet, ω = outlet)
ω
YO 2
α
Y CO 2
= are the volume fractions of CO2 in gas (α = inlet, ω = outlet)
ω
Y CO 2
Greek
β = constant defined in eq 10 (dimensionless)
ε = void fraction (dimensionless)
∈ = mean local energy dissipation per unit mass (W kg-1)
θp = volume fraction of pellets (dimensionless)
ξ = ratio of radial distance to radius of the pellet (dimensionless)
κ = effective hydraulic permeability of the pellet (m2)
η = Effectiveness factor for O2 consumption rate per unit mycelial pellet
(dimensionless)
ρc = density of the dried pellet (kg m-3)
ρv = density of wet pellet (kg m-3)
ρ = pellet suspension density (kg m-3)
τ = dimensionless time defined in eq 6
φ = Thiele modulus (dimensionless)
ν = kinematics viscosity of the suspending medium (m2 s-1)
χ = dimensionless parameter defined in eq 35
ℜ = mean reaction rate defined in eq 16
ℜ = reaction rate defined by eq 12
Ψ = volume function
Ψ = volume averaging function defined in eq 13
umerical values for process parameters.
5. Reference
5.1 References cited in Case I
Abashar, M. E., Narsingh, U., Rouillard, A. E. and Judd, R. (1998). Hydrodynamic flow
regimes, gas holdup, and liquid circulation in airlift reactors. Ind. Eng. Chem. Res.
37: 1251-1259.
Akita, K. and Yoshida, F. (1973). Gas holdup and volumetric mass transfer coefficient in
bubble columns. Effects of liquid properties. Ind. Eng. Chem. Process Des. Develop.
12: 76-80.
Al-Masry, W. A.. and Dukkan, A. R. (1998). Hydrodynamics and mass transfer studies in a
pilot-plant airlift reactor: non-Newtonian systems. Ind. Eng. Chem. Res. 37: 41-48.
Barboza, M., Zaiat, M. and Hokka, C.O. (2000). General relationship for volumetric oxygen
transfer coefficient (kLa) prediction in tower bioreactors utilizing immobilized cells.
Bioprocess Eng. 22: 181-184.
Barrow, A., Jefferys, E. G. and Nixon, I. S. (1960). Process for the production of gibberellic
acid. ICI Patent. GB 838,032.
Brito-De la Fuente, E., Nava, J. A., López, L. M., Medina, L., Ascanio, G. and Tanguy, P. A.
(1998). Process viscometry of complex fluids and suspensions with helical ribbon
agitators. Can. J. Chem. Eng. 76: 689-695.
Brückner, B. and Blechschmidt, D. (1991). The Gibberellin Fermentation. Crit. Rev. Biotech.
11, 163-192.
Chavez Parga, M. C. (2005). “Producción de ácido giberélico en un biorreactor airlift”. Ph. D.
Thesis. Instituto Tecnológico de Celaya. Celaya, Gto., México.
Chisti, M. Y. (1989). Airlift bioreactor. London-New York: Elsevier Appl. Science.
Choi, K. H., Chisti, Y. and Moo-Young, M. (1996). Comparative evaluation of hydrodynamic
and gas-liquid mass transfer characteristics in bubble column and airlift slurry
reactors. Biochem. Eng. J. 62:223-229.
Escamilla-Silva, E. M., Dendooven, L., Magaña, I. P., Parra-Saldivar, R. and De la Torre, M.
(2000). Optimization of Gibberellic acid production by immobilized Gibberella
fujikuroi mycelium in fluidized bioreactors. J. Biotechnol. 76:147-155.
Freitas, C. and Teixeira, J. A. (1998). Hydrodynamic studies in an airlift reactor with an
enlarged degassing zone. Bioprocess Eng.18: 267-279.
Gelmi, C., Pérez-Correa, R., González, M. and Agosin, E. (2000). Solid substrate cultivation
of Gibberella fujikuroi on an inert support. Process Biochem. 35:1227-1233.
Gelmi, C., Pérez-Correa, R. and Agosin, E. (2002). Modelling Gibberella fujikuroi growth and
GA3 production in solid-state fermentation. Process Biochem. 37:1033-1040.
Godbole, S. P., Schumpe, A., Shah, T. and Carr, N. L. (1984). Hydrodynamics and mass
transfer in non-Newtonian solutions in a bubble column. AIChE J. 30: 213-220.
Gouveia, E. R., Hokka, C. O. and Badino-Jr, A. C. (2003). The effects of geometry and
operational conditions on gas holdup, liquid circulation and mass transfer in an
airlift reactor. Braz. J. Chem. Eng. 20:363-374.
Gravilescu, M. and Tudose, R. Z. (1998). Hydrodynamics of non-Newtonian liquids in
external-loop airlift bioreactor. Part I. Study of the gas holdup. Bioprocess. Eng.
18:17-26.
Gravilescu, M. and Tudose, R. Z. (1999). Modelling mixing parameters in concentric-tube
airlift bioreactors. Part I. Mixing time. Bioprocess Eng. 20:423-428.
Halard, B., Kawase, Y. and Moo-Young, M. (1989). Mass transfer in a pilot plant scale airlift
column with non-Newtonian fluids. Ind. Eng. Chem. Res. 28: 243-245.
Heinrich, M. and Rehm, H. J. (1981). Growth of Fusarium moniliforme on n-alkanes:
comparison of an immobilization method with conventional processes. Eur. J. Appl.
Microbiol. Biotechnol. 11:239.
Jones, A. and Pharis, R. P. (1987). Production of gibberellins and bikaverin by cells of
Gibberella fujikuroi immobilized in carrageenan. J. Ferment. Technol. 65:717-722.
Kawase, Y. (1989). Liquid circulation in external-loop airlift bioreactors. Biotechnol. Bioeng.
35:540-546.
Kumar, P. K. P. and Lonsane, B. K. (1987) Gibberellic acid by solid state fermentation:
consistent and improved yields. Biotechnol. Bioeng. 30:267-271.
Kumar, P. K. P. and Lonsane, B. K. (1988) Immobilized growing cells of Gibberella fujikuroi P-
3 for production of gibberellic acid and pigment in batch and semi-continuous
cultures. Appl. Microbiol. Biotechnol. 28:537-542.
Metz, B., Kossen, N. W. F. and van Suijdam, J. C. (1979). The rheology of mould suspensions.
Adv. Biochem. Eng. 11:103-156.
McManamey, W. J. and Wase, D. A. J. (1986). Relationship between the volumetric mass
transfer coefficient and gas holdup in airlift fermentors. Biotechnol. Bioeng.
28:1446-1448.
Moo-Young, M., Halard, B., Allen, D. G., Burrell, R. and Kawase, Y. (1987). Oxygen
transfer to mycelial fermentation broths in an airlift fermentor. Biotechnol. Bioeng.
30:746-753.
Nava Saucedo, J. E., Barbotin, J. N. and Thomas, D. (1989). Continuous production of
gibberellic acid in a fixed-bed reactor by immobilized mycelia of Gibberella fujikuroi
in calcium alginate beads. Appl. Microbiol. Biotechnol. 30:226-233.
Prokop, A., Janík, P., Sobotka, M. and Krumphanzi, V. (1983). Hydrodynamics, mass
transfer, and yeast culture performance of a column bioreactor with ejector.
Biotechnol. Bioeng. 25: 114-1160.
Quintero, R. R. (1981). Ingeniería bioquímica, Teoría y aplicaciones. Ed. Alambra. México.
Schügerl, K., Lücke, J. and Oels, U. (1977). Bubble column bioreactors. Adv. Biochem. Eng.
7:1-81.
Shah, Y. T., Kelkar, B. G., Godbole, S. P. and Deckwer, W. D. (1982). Design parameters
estimations for bubble column reactors. AIChE J. 28:353-379.
Shukla, R., Srivastava, A. K. and Chand, S. (2003). Bioprocess strategies and recovery
processes in gibberellic acid fermentation. Biotechnol. Bioprocess Eng. 8:269-278.
Tobajas, M. and García-Calvo, E. (2000). Comparison of experimental methods for
determination of the volumetric mass transfer coefficient in fermentation processes.
Heat and mass transfer 36: 201-207.
Tudzynski, B. (1999). Biosynthesis of gibberellins in Gibberella fujikuroi: biomolecular aspects.
Appl. Microbiol. Biotechnol. 52:298-310.
5.2 References cited in Case II

Beydilli M I, Pavlostasthis S G. (2005). Decolorization kinetics of the azo dye Reactive Red 2
under methanogenic conditions: effect of long-term culture acclimation.
Biodegradation. 16: 135–146.
Van der Zee F P, Bisschops I A E, Lettinga G. (2003). Activated carbon as an electron
acceptor and redox mediator during the anaerobic biotransformation of azo dyes.
Environ Sci Technol. 37: 402-408.
Beyenal H, Lewandowski Z. (2002). Internal and external mass transfer in biofilms grown at
various flow velocities. Biotechnol Prog. 18: 55-61.
Fan L S, Fujie K, Long T R, Tang W T. (1987). Characteristics of draft tube gas-liquid-solid
fluidized-bed bioreactor with immobilized living cells for phenol degradation.
Biotechnol Bioeng. 30: 498-504.
Fan L-S, Leyva-Ramos R, Wisecarver K D, Zehner B J. (1990). Diffusion of phenol through a
biofilm grown on activated carbon particles in a draft-tube three-phase fluidized-
bed biorreactor. Biotechnol Bioeng. 35: 279-286.
Herzberg M, Dosoretz C G, Tarre S, Green M. (2003). Patchy biofilm coverage can explain
the potential advantage of BGAC reactors. Environ Sci Technol. 37: 4274-4280.
McCarty P L, Meyer T E. (2005) Numerical model for biological fluidized-bed

reactor treatment of perchlorate-contaminated groundwater. Environ Sci Technol.
39: 850-858.
Di Iaconi C, Ramadori R, Lopez A, Pasión R. (2005). Hydraulic shear stress calculation in a
sequencing batch biofilm reactor with granular biomass. Environ Sci Techno.
l39: 889-894.
Iliuta I, Thyrion F C, Muntean O, Giot M. (1996). Residence time distribution of the liquid in
gas-liquid cocurrent upflow fixed-bed reactors. Chem Eng Sci. 51(20): 4579-4593.
Smith L C, Elliot D J, James A. (1996). Mixing in upflow anaerobic filters and its influence on
performance and scale-up. Water Res. 30(12): 3061-3073.
Levenspiel O. Chemical reaction engineering (Spanish translation). 3rd Ed. México: Limusa
Wiley; 2004. p. 668.
Fogler H S. Elements of chemical reaction engineering, 3rd Ed. New Jersey, USA: Prentice
Hall PTR; 1999. p. 967.
Escamilla-Silva E M, Gutierrez G F, Dendooven L, Jimenez-Islas H, Ochoa-Tapia J A. (2001).
A Method to Evaluate the Isothermal Effectiveness Factor for Dynamic Oxygen into
Mycelial Pellets in Submerged Cultures. Biotechnol Prog.17(1): 95-103.
Kulkarni R R, Wood J, Winterbottom J M, Stitt E H. (2005). Effect of fines and porous catalyst
on hydrodynamics of trickle bed reactors. Ind Eng Chem Res; 44(25): 9497-9501.
Hines A L, Maddox R N. Mass transfer, fundamentals and applications (spanish
translation). México: Prentice Hall Hispanoamericana; 1987. p. 600.
Lee J-W, Choi S-P, Thiruvenkatachari R, Shim W-G, Moon H. (2006). Evaluation of the
performance of adsorption and coagulation processes for the maximum removal of
reactive dyes. Dyes and Pigments. 69: 196-203.
Chen K-C, Wu J-Y, Yang W-B, Hwang S-C J. Evaluation of Effective Diffusion Coefficient
and Intrinsic Kinetic Parameters on Azo Dye Biodegradation Using PVA-
Immobilized Cell Beads. Biotechnol Bioeng 2003; 83(7): 821-832.
Iliuta I, Larachi F. (2005). Modeling simultaneous biological clogging and physical plugging
in trickle-bed bioreactors for wastewater treatment. Chem Eng Sci. 60: 1477 – 1489.
Spigno G, Zilli M, Nicolella C. (2004). Mathematical modelling and simulation of phenol
degradation in biofilters. Biochem Eng J. 19: 267–275.
Mammarella E J, Rubiolo A C. (2006). Predicting the packed-bed reactor performance with
immobilized microbial lactase. Process Biochem. 41: 1627–1636.
Leitão A, Rodriguez A. (1996). Modeling of biodegradation/adsorption combined processes
in fixed-bed biofilm reactors: effects of the intraparticle convective flow. Chem Eng
Sci. 51(20): 4595 4604.
Leitão A, Rodriguez A. Dynamic behavior of a fixed-bed biofilm reactor: analysis of the role
of the intraparticle convective flow under biofilm growth. Biochem Eng J 1998;
2: l-9.
Lewandowski Z, Stoodley P, Altobelli S, Fukushima E. (1994). Hydrodynamics and kinetics
in Biofilm systems - Recent advances and new problems. Wat Sci Tech. 29(10-11):
223-229.
5.3 References cited in Case III

Takamatsu, T.; Shioya, S.; Furuya, T. (1981). Mathematical Model for Gluconic Acid
Fermentation by Aspergillus niger. J. Chem. Tech. Biotechnol. 31, 697-704.
Qian, X.M.; du Preez, J.C.; Kilian, S. G. (1994). Factors Affecting Gibberellic Acid Production
by Fusarium moniliforme in Solid-State Cultivation on Starch. World J. Microbiol.
Biotechnol. 10, 93-99.
Lu, Z.X.; Xie, Z.C.; Kumakura, M. (1995). Production of Gibberellic Acid in Gibberella
fujikuroi Adhered onto Polymeric Fibrous Carriers. Process Biochem. 30, 661-665.
Hollmann, D.; Switalski, J.; Geipel, S.; Onken, U. (1995). Extractive Fermentation of
Gibberellic Acid by Gibberella fujikuroi. J. Ferment. And Bioeng.79, 594-600.
el-Enshasy, H., Hellmuth, K., Rinas U. (1999). Fungal morphology in submerged cultures
and its relation to glucose oxidase excretion by recombinant Aspergillus niger. Appl.
Biochem. Biotechnol., 81, 1-11.
Yamane, T.; Shimizu, S. (1984). Fed-batch techniques in microbial processes. In: Fiechter A.
Ed.: Adv. Biochem. Eng./Biotechnol., 30, 147-194.
Wang, N. S.; Stephanopoulos, G. (1984). Computer Applications for Fermentations
Processes. CRC Critical Reviews in Biotechnology. 2, 1-103.
Parulekar, S. J.; Lim, H.C. (1985). Modelling Optimization and Control of Semi-Batch
Bioreactors. In: Fiechter A.(De.): Adv. Biochem. Eng./Biotechnol. 32, 207-258.
Sharon, C., Nakazato, M., Ogawa H.I., Kato Y. (1999). Bioreactor operated production of
lipase: castor oil hydrolysis using partially-purified lipase. Indian. J. Exp. Biol. 37,
481-486.
Stephanopoulos, G.; Tsiveriotis, K. (1989). The Effect of Intraparticle Convection on Nutrient
Transport in Porous Biological Pellets. Chem. Eng. Sci. 44, 2031-2039.
Reuss, M.; Fröhlich, S.; Kramer, B.; Messerschmidt, K.; Pommerening, G. (1986). Coupling of
Microbial Kinetics and oxygen transfer for Analysis and Optimization of Gluconic
Acid Production with Aspergillus niger. 1,79-91.
Nielsen, J.; Villadsen, J. In Bioreaction Engineering Principles. Plenum Press: New York 1994.
Sunil, N.; Subhash, Ch. (1996). Mass Transfer and Biochemical Reaction in Immobilised Cell
Packed Bed Reactors: Correlation of Experimental with Theory. J. Chem. Tech.
Cui, Y.Q., van der Lans, R.G., Luyben, K.C. (1998). Effects of dissolved oxygen tension and
mechanical forces on fungal morphology in submerged fermentation. Biotechnol.
Bioeng. 57, 409-410
Goosen, M.F. (1999). Physico-chemical and mass transfer considerations in micro-
encapsulation. Ann. N. Y. Acad. Sci. 875, 84-104.
Fan, D., Shang, L., Yu, J. (1996). Research on fermentation scale-up based on the OUR
obtained from a shake flask. Chin. J. Biotechnol. 12, 177-184.
Carbonell, R. G.; Whitaker S. Heat and mass transfer in porous media. In Fundamentals of
Transport in Porous Media; Bear, J.; Corapcioglu, M. Y. Eds.; Martinus Nijhoff:
Brussels, 1984; pp 121-198.
Ochoa, J.A.; Strove P.; Whitaker, S. (1986). Diffusion and Reaction in Cellular Media. Chem.
Eng. Sci. 41, 2999-3013.
Whitaker, S. (1991). Improved Constraints for the Principle of Local Equilibrium. I & E C
Research. 29, 983-997.
Aris, R. The Mathematical Theory of Diffusion and Reaction in Permeable Catalysts. Vol.1,
The theory of the Steady State; Vol.2, Questions of Uniqueness, Stability and
Transient Behaviour; Clarendon Press: Oxford, 1975.
Whitaker, S. Flow in Porous Media I: A Theoretical Derivation of Darcy’s Law. Transport in

Porous Media. 1986, 1, 3-25.
Finlayson, B. A. Nonlinear Analysis in Chemical Engineering. McGraw-Hill Book Co.: USA,
1980.
Sano, Y.; Yamaguchi, N.; Adachi, T. (1974). Mass Transfer Coefficients for Suspended
Particles in Agitated Vessels and Bubble Columns. J. Chem. Eng. Jpn. 7, 255-261.
Wang, D. I. C.; Fewkes, R. C. Mass Transfer Studies in Fermentation Broths. J. Dev. Ind.
Microbiol. 1977, 18, 39-44.
Levenspiel O. Chemical Reaction Engineering. Wiley International, Second Edition, 1972.
Wang, D. I. C.; Cooney, C. L.; Demain, A. L.; Dunnill., P.; Humphrey, A. E.; Lilly, M. D.
Fermentation and Enzyme Technology. Wiley, N.Y. 1979.
Van Suijdam, J. C.; Hols, H.; Kosen, N. W. F. (1982). Unstructured Model for Growth of
Mycelial Pellets in Submerged Cultures. Biotechnol. Bioeng. 24, 177-191.
Van Suijdam, J. C.; Metz, B. (1981). Influence of Engineering Variables upon the Morphology
of Filamentous Molds. Biotechnol. Bioeng. 23, 111-148.
Miura, Y. (1976). Transfer of Oxygen and Scale-Up in Submerged Aerobic Fermentation.
Adv. Biochem. Eng. 4, 1-40.
Perry, H. R.; Green, Don W.; Maloney, J.O. Perry’s Chemical Engineers Handbook; Mc Graw-
Hill: New York, 1997.
Riley, M. R.; Muzzio F. J.; Buettner H. M.; Reyes, S. C. (1995). A Diffusion in Heterogeneous
Media: Applications to Immobilized Cell Systems. AIChE J. 41, 691-700.
Riley, M. R.; Muzzio, F. J.; Buettner, H. M.; Reyes S. C. (1996). A Simple Correlation for
Predicting Effective Diffusivities in Immobilized Cell Systems. Biotechnol. Bioeng.
49, 223-227.
Johnson, M. R.; Kamm, C. R.; Ethier, T. P. (1987). Scaling Laws and the Effects of
Concentration Polarization on the Permeability of Hyaluronic Acid. PhysicoChem.
Hydrody. 9, 427.
Jiménez-Islas, H., López-Isunza, F., Ochoa-Tapia, J.A. (1999). Natural convection in a
cylindrical porous cavity with internal heat source: a numerical study with
Brinkman-extended Darcy model. Int. J. Heat Mass Transfer. 42, 4185-4195.
Miura, Y.; Miyamoto, K.; Kanamori, T.; Ohira, N. (1975). Oxygen Transfer within Fungal
Pellets. J. Chem. Eng. Jpn. 8, 300-304.
Kim. J. H., Lebeault, J. M., Reuss, M. (1983). Comparative Study of Rheological Properties of
Mycelial Broth in Filamentous and Pelleted Forms. Eur. J. App. Microbiol.
Aiba, S.; Kobayashi, K. (1971). Comments on Oxygen Transfer Within a Mold Pellet.
Biotechnol. Bioeng. 15, 27-29.
Kobayashi, T.; Van Dedem, G.; Moo-Young, M. (1973) Oxygen Transfer Into Mycelial Pellets.
Miura, Y.; Miyamoto, K. (1977). Oxygen Transfer within Fungal Pellets. Biotechnol. Bioeng.
19, 1407-1409.
Wittler, R.; Baumgartl, H.; Lübbers, D.W.; Schürgerl, K.(1986). Investigations of Oxygen
Transfer into Penicillium chrysogenum Pellets by Microprobe Measurement.
Metz, B.; Kossen, N. W. F. (1977). The Growth of Molds in the Form of the Pellets A
Literature Review. Biotechnol. Bioeng. 19, 781-799.
Chiam, H. F.; Harris I. J. (1981). Microelectrode Studies of Oxygen Transfer in Trickling

Filter Slimes. Biotechnol. Bioeng. 23, 781-792.
Reuss, M.; Bajpai, R. K.; Berke, W. (1982). Effective Oxygen-Consumption Rates in
Fermentation Broths with Filamentous Organisms. J. Chem. Tech. Biotechnol. 32,
81-91.
Nienow, A. W. 1990. Agitation for Mycelial Fermentations. Trends Biotechnol. 8, 224-233.
Yano, T., Kodama, T., Yamada, K. (1961). Fundamental Studies on the Aerobic
Fermentation. Agr. Biol. Chem. 25(7), 580-584.
Huang, M.Y.; Bungay, H.R. (1973).Microprobe Measurements of Oxygen Concentration in
Mycelial Pellets. Biotechnol. Bioeng. 15, 1193-1201.
Ngian, K. F.; Lin S. H. (1976). Diffusion Coefficient of Oxygen in Microbial Aggregates.
Fan, L.S.; Leyva Ramos, R.; Wisecarver, K. D.; Zehener, B. J. (1990). Diffusion of Phenol
through a Biofilm Grown on Activated Carbon Particles in a Draft-Tube Three-
Phase Fluidized-Bed Bioreactor. Biotechnol. Bioeng. 39, 279-286.
Ross, L. W.; Updegraff, D. M. (1971). Kinetics of Diffusion-Coupled Fermentation Processes:
Conversion of Cellulose to Protein. Biotechnol. Bioeng. 13, 99-111.
View publication stats

Mass Transfer in Bioreactors: Linda Gonzalez Ruiz Reyes Mayra

Uploaded by

Copyright:

Available Formats

Mass Transfer in Bioreactors: Linda Gonzalez Ruiz Reyes Mayra

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mass Transfer in Bioreactors: Linda Gonzalez Ruiz Reyes Mayra

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Mass Transfer in Bioreactors

Chapter · February 2011

Linda Gonzalez Ruiz Reyes Mayra

SEE PROFILE SEE PROFILE

Oscar M. Hernández-Calderón Eleazar M. Escamilla-Silva

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Mass Transfer in Bioreactors

Tecnológico Querétaro Sanfandila, 76703 Sanfandila, Pedro Escobedo, Qro.,

4 sur 104 centro histórico C.P. 72000, Puebla.,

Tecnológico y Antonio García Cubas S/N, Celaya, Gto., C.P. 38010,

Ciudad Universitaria, C.p. 80090, Culiacán, Sinaloa.

Tecnológico y Antonio García Cubas S/N, Celaya, Gto., C.P. 38010,Sinaloa.

2. Case I. Hydrodynamics, mass transfer and rheological studies of

2.2 Materials and methods

in the present work (Culture collection of the Department of Biotechnology and

2.3 Hydrodynamics and mass transfer studies

2.4 Rheological studies

2.5 Results and discussion

vlr = 1.3335 v0.3503

vld = 0.8716 v0.2970

Fig. 2. Liquid velocities as a function of superficial gas velocity in the riser.

tm = 5.0684 v −gr0.3628 (5)

Fig. 3. Mixing time as a function of superficial gas velocity in the riser.

Fig. 4. Effect of the superficial gas velocity in the riser on kLa.

Fig. 5. kLa through fermentation time in the airlift bioreactor.

kL a = 0.2883 ε 0.9562 (7)

Fig. 7. The kL/dB ratio as a function of superficial gas velocity.

Fig. 8. Typical rheogram employing impeller viscometry

Fig. 11. K and n as a function of biomass concentration in the airlift bioreactor.

Fig. 1. Reactive red 272

3.2 Materials and methods

Fig. 2. Upflow Anaerobic Fixed Bed (UAFB) Reactor.

Work volume, L 3.3

3.2.2 Residence time distribution.

D = d u L = 1.01ν N Re 0.875 (4)

Sh = 2 + 0.6 N Re 1/2Sc 1/3 (5)

particle. Supposing this the following equation was used:

3.2.3 Kinetic model

3.2.4 Model dimensionless numbers

3.3 Results and discussion.

3.4 Residence time distribution.

P(t ) = y0 + a * Exp [ −Exp( − f 1) − f 1 * s + 1]

Fig. 3. HRT distribution; tracer pulse applied to the bioreactor.

Parameter HRT1 HRT2 HRT3 HRT4 HRT5 HRT6

Table 2. Hidrodinamic, mass transfer and and fitted extreme

3.5 Dynamic model: Transport and reaction in the fixed bed

Fig. 4. Transport and reaction model in the UAFB, Reactor scheme.

t=0 CAL = CA0

Z=0 CAL = CA0 (12)

∂ωL ∂ 2ωL ∂ωL

∂ωp ⎛ 2 ∂ωp ∂ 2ωp ⎞

Fig. 5. Parallel dimensionless model.

For the biofilm:

3.6 Model solution.

DL (cm2/min) 423.4 Φ 12 a 1.061

u (cm/min) 3.786 Φ 22 b 1.276