PART 3

Engineering Principles for Bioprocesses

9
Operating Considerations
for Bioreactors
for Suspension
and Immobilized Cultures

So far we have discussed what cells are, how they work, and how to describe their growth
in simple reactors. We now begin our discussion of how to use these cells in processes.
We will explore some more complicated reactor strategies and why they might be consid-
ered for use in real processes. Chapter 10 will give more details on reactor design, and
Chapter II will detail how to recover products from these reactors. These chapters should
give the reader an understanding of how real bioprocesses can be assembled.
An important decision for constructing any process concems the configuration the
reactor system should take. The choice of reactor and operating strategy determines prod-
uct concentration, number and types of impurities, degree of substrate conversion, yields,
and whether sustainable, reliable performance can be achieved. Unlike many traditional
chemical processes, the reactor section represents a very major component (usually
:> 50%) of the total capital expenditures. Choices at the reactor level and of the biocatalyst

determine the difficulty of the separation. Thus, our choice of reactor must be made in the
context of the total process: biocatalyst, reactor, and separation and purification train.

245
 The ratio for rates of biomass formation is
9.2. CHOOSINGTHE CULTIVATlON METHOD
X
rc,opl = ln -.!!L + IlmtI (9.7)
One of the first decisions is whether to use a batch or continuous cultivation scheme. rb Xo
Although a simple batch and continuous-flow stirred-tank reactor (CFSTR) represent ex-
tremes (we wiU soon leam about other reactors with intermediate characteristics), consid- Most commercial fermentations operate with X,jXo '" 10 to 20. Thus, we would ex-
eration of these two extreme altematives will clarify some important issues in reactor pect continuous systems to always have a significant productivity advantage for primary
selection. products. For example, an E. coli fermentation with X,jXo = 20, ti = 5 h, and !lm = 1.0 h-l
First, we can consider productivity. The simplest case is for the production of cel1 would yield rc.op/rb = 8.
mass or a primary product. For a batch reactor, four distinct phases are present: lag phase, Based on this productivity advantage we might be surprised to leam that most com-
exponential growth phase, harvestí~g, and preparation for a new batch (e.g., cleaning, mercial bioprocesses are batch systems. Why? There are several answers.
sterilizing, and filling). Let us define tI as the sum of the times required for the lag phase, The first is that eq. 9.7 applies only to growth-associated products. Many secondary
harvesting, and preparation. The value for ti will vary with size of the equipment and the products are not made by growing cells; growth represses product formation. Under such
nature of the fermentation but is normally in the range of several hours (3 to 10 h). Thus, circumstances, product is made only at very low dilution rates, far below those values op-
the total time to complete a batch cycle (tc) is timal for biomass formation. For secondary products, the productivity in a batch reactor

t
c
=-IlmI In-+tIXo
Xm
may significantly exceed that in a simple chemostat.
Another primary reason for the choice of batch systems over chemostats is genetic
instability. The biocatalyst in most bioprocesses has undergone extensive selection. These
highly "bred" organisms often grow less wel1 than the parental strain. A chemostat im-
where Xm is the maximal attainable cel1 concentration and Xo is the cell concentration
poses strong selection pressure for the most rapidly growing cello Back-mutation from the
inoculation.
The total amount of cell mass produced comes from knowing the total amount at productive specialized strain to one similar to the less productive parental strain (Le., a re-
vertant) is always present. In the chemostat the less productive variant will become domi-
growth-extent-limiting nutrient present and its yield coefficient: nant, decreasing productivity. In the batch culture the number of generations available
Xm - Xo = Yx/sSo « 25 from slant cultures to a commercial-scale fermenter) for the revertant cell to out-
grow the more productive strain is limited. Cel1s at the end of the batch are not reused.
The rate of cell mass production in one batch cycle (rb) is These considerations of genetic stability are very important for cells with recombinant
Yx/sSo DNA and are discussed in detail in Chapter 14.
rb = Another consideration is operability and reliability. Batch cultures can suffer great
(l/llm)ln(Xm/XO)+tI
variability from one run to another. Variations in product quality and concentration create
As discussed in Chapter 6, the maximum productivity of a chemostat is found by differe problems in downstream processing and are undesirable. However, long-term continuous
culture can be problematic; pumps may break, control1ers may fail, and so on. Mainte-
tiating DX with respect to D and setting dDX/dD to zero. The value for D optimal wb
nance of sterility (absence of detectable foreign organisms) can be very difficult to
simple Monod kinetics apply is given by eq. 6.83, and the corresponding X can be det,
mined to be achieve for periods of months, and the consequences of a loss of sterility are more severe
than with batch culture.
Xopt = Yx/s{So + Ks -.[Ks(So +Ks)} One other factor determining reactor choice is market economics. A continuous sys-
tem forms the basis of a dedicated processing system--dedicated to a single product.
Thus, the best productivity that could be expected from a chemostat where Monod ki Many fermentation products are required in small arnounts, and demand is difficult to
ics apply is Dop' . Xopt, or project. Batch systems provide much greater flexibility. The same reactor can be used for
two months to make product A and then for the next three for product B and the rest of the
year for product C.
DoptXop, = YxlsllJI- L JKsKs+So 1[So +Ks -~Ks(So+Ks)) Most bioprocesses are based on batch reactors. Continuous systems are used to
make single-cell protein (SCP), and modified forms of continuous culture are used in
Under normal circumstances 50 » Ks, so the rate of chemostat biomass production, Waste treatrnent, in ethanol production, and for some other large-volume, growth-
approximately associated products such as latic acid.
rc.opt = DoptXopt = IlmYX/SSO Let us consider some modifications to these reactor modes.

Sec. 9.2 Choosing the Cultivation Method 247

246 Bioreactors for Suspension and Immobilized Cultures
 At steady state, dS/dt = O and

9.3. MODIFYING BATCH AND CONTlNUOUS REACTORS

D M

X1=llg Yx/s(So-S (9.11)

9.3.1. Chemostat with Recycle
Substitution of eq. 9.9 when kd = O into eq. 9.11 yields
Microbial conversions are autocatalytic, and the rate of conversion increases with cell yM (S -S)
concentration. To keep the cell concentration higher than the normal steady-state level in a X = x/s o (9.12)
chemostat, cells in the effluent can be recycled back to the reactor. cen recycle increases
I (l+a-aC)
the rate of conversion (or productivity) and also increases the stability of some systems Therefore, the steady-state cen concentration in a chemostat is increased by a factor
(e.g., waste-water treatment) by minimizing the effects of process perturbation. cens in of 1/(1 + a - aC) by cen recycle. The substrate concentration in the effluent is deter-
the effluent stream are either centrifuged, filtered, or settled in a conical tank for recy- mined from eq. 9.9 and the Monod eq. 6.30, where endogenous metabolism is neglected,
cling. .' andis
Consider the chemostat system with cen recycle as depicted in Fig. 9.1. A materill1
balance on cen (biomass) concentration around the fermenter yields the following s= K,D(l+a-aC) (9.13)
Ilm -D(I+a-aC)
equation:
dXI
Then eq. 9.12 becomes
FXo +aFCXI-(l +a)FXI + Viloe' XI = V-dt

where a is the recycle ratio based on volumetric flow rates, C is the concentration fac' XI = (l+a-aC)
Y~s [so _ IlmK,D(I+a-aC)
-D(I+a-aC) ] (9.14)
or ratio of cen concentration in the cen recycle stream to the cen concentration in the
aetor effluent, F is nutrient flow rate, V is culture volume, Xo and XI are cen concen Effluent cen concentrations and productivities in a chemostat with and without cen
tions in feed and recycle streams, and X2 is cen concentration in effluent from the recycle are compared in Fig. 9.2. cen concentrations and productivities are higher with
cen recycle, resulting in higher rates of substrate consumption. Systems with cen recycle
separator.
At steady state, and if dX/dt = O and Xo = O (that is, sterile feed); then eq. 9.8 are used extensively in waste treatment and are finding increasing use in ethanol produc-
comes tion. The application of cell recycle reactors in waste treatment is detailed in Chapter 16.
The equations differ from the case above due to the inclusion of a term for endogenous
Ilne. = (l + a - aC)D = [1+ a(l- C)]D metabolism (Le., kd). The basic concept of operation at flows above the "washout" rate ap-
plies when kd *- O.
Since C> 1 and a(1 _ C) < O, then Iloe' < D. That is, a chemostat can be operated at
tion rates higher than the specijic growth rate when cell recycle is used. Example9.1
A material balance for growth-limiting substrate around the fermenter yields ln a chemostat with cell recycle, as shown in Fig. 9.1, the feed flow rate and culture volumes
Il X dS are F = 100 mlJhand V = 1000 ml, respectively.The system is operatedunder glucose limita-
FSo+ aFS - V~Yx/s - (1+ a)FS = V-dt tion, and the yield coefficient, Yts. is 0.5 gdw cells/g substrate. Glucose concentrationin the

X2 Figure 9.2. Comparisonofbiomasscon-

centrationsandoutputratesinsteadystates
ofchemostatcultureswithandwithoutrecy-
cle.Symbols:XI = biomassconcentration in
chemostalwithoulrecycle;X2 = biomass
aF concentration in chemostalculturewithre-
cycle;R I = biomassoutpulrateperunilvol-
umewithoulrecycle;Rz = biomassoulput
raleofchemostalwithrecycle;/lm = 1.00
Figure9.1. Chemostatwith hol; Sr= 2.0 gI1; KS= 0.010 gI1; YXIS =
Thecellseparatorcouldbe a 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 0.5 g/g;concentration faclor.C = 2.0; and
tank,a centrifuge.ora micro: Dilulion rale (h-') recyclerate,a = 0.5.
device.
eX1

Modifying Batch and Continuous Reactors 249

248 Bioreaetors for Suspension and Immobilized CultureS
 stage, the inducer is added and large quantities of product are made. Cells defective in
product synthesis should not overtake the culture (at least not completely), because fresh
feed is 50= 10 g glucosell.The kinetic constants of the organisms are 11m = 0.2 h-l, Ks = I g genetieally unaltered cells are being continuously fed to the reactor. Thus, the two-stage
glucosell.The value of C is 1.5, and the recyc1eratio is a = 0.7. The systemis at steady system can allow the stable continuous production of the target protein when it would be
a Find the substrateconcentration in the recyc1estream (S). impossible in a simple chemostat.
b. Find the specific growth ra1e(Iloe~ofthe organisms.
Perhaps an easier situation to consider is the production of a secondary product
c. Fmd the cell (biomass) concentrationin the recyc1estream·
(e.g., ethanol or an antibiotic). Here we worry not so much about a mixture of subpopula-
d. Find the cell concentrationin the centrifuge effluent (Xv·
tions, but that conditions that promote growth completely repress product formation. A
Solutlon Using eq. 9.9, we deterrnineIloe"
very large scale multistage system for ethanol production is currently in use. A multistage
Ilnet= [l + <XCI - C)]D = [l + (1 - 1.5)0.7](0.1) = Ilg system of CFS1R approaches PFR behavior. A PFR rnirnics the batch system, where
=0.065h-1 spaee time (the time it takes the eulture fluid to reach a specific location in the PFR) re-
places eulture time. A multistage system is much like taking the batch growth curve and
Then ,. dividing it into sections, with each section being "frozen" in a corresponding stage of the
5=~= (1)(0.065) =0.48 g/I multistage system. As in the batch reactor, the culture's physiological state progresses
11m -Ilne< 0.2 - 0.065 from one stage to the next.
The mathematical ana1ysis of the multistage system that we present here is imper-
Xl = D(50 - 5)Y~/s = (0.1)(10- 0.48)0.5 fect. Growth in the second and subsequent stages is intrinsica1ly unbalanced growth, even
Ilg 0.065
though it is steady-state growth. New cells entering the second or subsequent stage are
= 7.3 gll continuously adapting to the new eonditions in that stage. Consequently, unstructured
models are not expected to give completely accurate predictions. However, we use un-
A biomass balance around the concentratoryieIds structured models here due to their simplicity and to illustrate at least some aspects of
(1 + a)XI = aCXI + X2 multistage systems.
A two-stage chemostat system is depicted in Fig. 9.3. Biomass and substrate bal-
X2 = (1 + a)XI - aCXI
ances on the first stage yield the following equations (ignoring endogeneous metabolism):
= (1.7)(7.3) - (0.7)(1.5)(7.3)
=4.8 gIl SI = Ks~ (9.15)
Ilm -11

Xl =~s(So-~) (9.16)
9.3.2. Multistage Chemostat Systems
The biomass balance for the second stage yields
In some fermentations, particularly for secondary metabolite production, the
product-formation steps need to be separated, since optimal conditions for each
different. Conditions such as temperature, pH, and limiting nutrients may be
FXI - FX2 + 112 VzX2 = Vz-
dX2
dt
(9.17)

each stage, resulting in different cell physiology and cellular products in m'
systems.
AD. example of a multistage system that may be beneficial is in the cult\lfe S~
F F
cally engineered cells. To improve genetic stability, a plasmid-carrying recombi ,'F'
II
F2
usually uses an inducible promoter to control production of the target protein So S, X2
ter 8). ln the uninduced state, the plasmid-containing cell grOWSat nearly the s XI • S2
the cell that loses the plasmid (a revertant), so the plasmid-free cell holds lil
advantage over the plasmid-containing cell. However, if the inducer is ad
mid-containing cells wi11 make large quantities of the desired protein S2
have greatly reduced growth rates. Thus, a single-stage chemostat would fi' ~
for the production of the target protein because of resulting problems in gene'
A multistage system can circumvent this problem. ln the fust stage, 110 indu' v, V2 Figun 9.3. 1\vo-stage chemostat system.
and the plasmid-containing cell can be maintained easily (usually an antibiotiC,
kill plasmid-free cells; -.s~e5napter 14 for a more complete discussion). In Modifying Batch and Continuous Reaetors 251

Bioreaetors for Suspension and Immobilized Cultur

250
 or
At steady state, eq. 9.17 becomes
8 =~ Xn-Xn_l (9.24t
(9.18)
n Dn Tx.n(Xn,Sn)
~2 =Dz(I-~)
1
(9.25a
where XI/X2 < 1 and ~2 < D2• T,.n =y-Tx.n(Xn,Sn)
x/s = Dn(Sn -Sn_I)
The substrate ba1ance for the limiting substrate in the second stage is
or
(9.19)
FS -FS _~2X2 V =V dS2
I 2 Y~s 2 2 dt
9n =~=
D Sn -Sn-I
-- (9.25b)
n T,.n
At steady state, eq. 9.19 becomes

S2 =
S--
I
~2

D2 Yx/s
X2
M Tp.n(Xn.Sn.···) = Dn(P" - Pn-l) (9.26a)
or
where
9 = 2.- = __1',,_-_1',,_-_1 (9.26b)
D2 = FlVz and ~2 = K,+S2
~mS2 n Dn Tp,n(Xn,SR"")

where Dn = FlVm 9n is the mean residence tÍIDe in the nth stage, and Tx,n' T',n' and Tp,n a11
Equations 9.18 and 9.20 can be solved simultaneously for X2 and S2 by substitutin" represent rates of reaction in the nth stage.
~ = ~mS/(K, + S2) in both equations or any other functiona1 farm that describes ~. The preceding set of equations lends itself to machine ca1culations. However,
When a feed stream is added to the second stage, then the design equations chan. graphica1 approaches to multistage design can a!so be used and have the advantage that
The second feed stream may contain additiona! nutrients, inducers, hormones, or the functional form of the growth or production rate need not be known, AlI that is re-
hibitors. Biomass ba!ance for the secon<Ístage in this case is quired is a batch growth curve. However, the transfer of the information from batch
growth curve to predictions of the multistage system stilI requires the assumption of ba!-
F;Xj + F'X' - (F; + F')X2 + V2~2X2 = V2 áX2
dt anced growth, Hence, the ana!ysis must be used with caution, ln at least one case (the pro-
duction of spores from Bacillus), this approach has made experimenta!ly verifiable
At steady state when X' = O, eq. 9.21 becomes predictions ofthe performance vf a six-stage system,
The graphica! approaches make use of eqs. 9.24 to 9.26, Dne approach is to use a
~2=D'2---F; XI plot of lI(dX/dt) versus X or lI(dP/dt) versus P derived from batch growth curves. This
Vz X2
corresponds to using eqs. 9.24b and 9.26b. The size of the required reactor is determined
by the area ofthe rectangle described with sides Xn - Xn_1 and lI(dX/dt) or Pn - Pn-I and
where
lI(dP/dt), The area of the rectangle is 9, and if F is known, V can be ca!culated. An a!ter-
~mS2 native approach avoids some tria!-and-error solutions that are necessary with the first ap-
D2-,_FJ+F'Vz and ~2= K,+S2 proach. This second approach requires plots of dXldt versus X and dP/dt versus P. The
intersection of the reaction curve with a line from the mass balance equation (e,g"
Substrate ba!ance for the second stage yields eq. 9.24a) determines the exit concentration of X or P, while the slope of the line deter-
mines D, and if F is known, V can be found. We iIIustrate the use of these approaches in
F.S +F'S' -(F. +F')S - V2~2X2 =V dS2 Example 9.2,
I I o I 2 Y:Js 2 dt
EJtaniple 9.2
Equations 9.22 and 9.23 need to be solved simultaneously for X2 and S2' Data for the productionof a secondarymetabolitefrom a small-scalebatch reactor are shownin
We can genera1ize these equations for a system with no additiona1 streantS Fig, 9.4, Assume that tworeactors, each with 700-1 workingvolume.are available.Youwill use
second or subsequent units. If we do a ba1ance around the nth stage on .biomass, su exactlythe same culture conditions (medium,pH. temperature.and so on) as in the batch reac-
and product, we find tor,Ifthe flowrate is 100 l/h. predict the outletconcentrationofthe product.Comparethat to the
Tx.n(Xx'Sn) = Dn(Xn - Xn_l) valuepredicted if a single 1400-1 reactor were used, Use both graphicalapproaches,
 0.8
84
(g/Il 8-
62
3 7-
7 X 5 4-
gll
X, 16- I
2-
X-
3-
611

Figure 9A. Data for Example 9.2. Data

are for the production of a secondaryP'
o 4 8 12 16
o uet in batch culture.
t (h)
o
o t.2 .4
~ P,gll
Solution The fust step in using either graphical approach is to differentiate the data tn,
batch growth curve to yield dXldt and dP/dt. The differentiation of experimenta! data Figure9.5. Solution of Example 9.2 for two-stage system, each with e = 7 h.
magnify errors present in the originál data, so the values of dXldt and dP/dt must be
preted cautiously.
For the graphical approach illustrated in Fig. 9.5, we have plotted l/(dXldt) 62=7 h=(Pz-JP--
I
d~/dt
X and lI(dP/dt) versus P, which corresponds to eqs. 9.24b and 9.26b. For 61 = 7 h (,
700 1/100 lib), we must detemune what value of Xl will satisfy
By tria! and error, we find that at P2 = 0.49 g/l

61 =(XI-XO{~/dtjLl 62 = (0.49 g/l-0.08 gll)(17 h/g/l)

=6.97 h

or which corresponds reasonably c10sely to 7 h.

In this solution the reader should note that for the first stage, only solutions that exist
for XI greater than the value of X for which 1/(dXldt) is a minimum are practically obtainable.
7 h=X{~/dtl. Washout occurs if 6 I is too small.
We can compare the result to a single-stage system with the same tota! volume as the
two-stage system (Fig. 9.6). Here the trial-and-error approach indicates for XI = 7.35 g/l that
Since a sterile feed is to be used, Xo = O.
By tria! and error, we find on the graph that Xl = 7.2 g/I corresponds to 7.35 gll . 1.9 h/g/l = 13.97 h = 14 h
0.95 h/g/I or 7.2 g/I . 0.95 h/g/l = 6.84 h.
Given the accuracy with which Fig. 9.5 can be read, this is an acceptable sól1 The value of Pl that corresponds to Xl = 7.35 g/l is 0.10 g/l. Thus, the use of the two-stage
product concentration that corresponds to Xl = 7.2 g/l is deterrnined from tM bal . system in this case increased product concentration from 0.10 to 0.49 g/l.
curve. As mustrated, Xl = 7.2 g/l is achieved at 9.4 h after inocuJation; at the sametu,j. An a!ternative graphica! approach that e1iminates the tria!-and-error aspect of the first
value for P 1 is 0.08 g/l. .. approach is shown in Fig. 9.7. Here eqs. 9.24a and 9.26a have been used. DI = 1/61 =
The effect of the second stage on the process is detemuned by using eq. 9.2'
ing that again 62 = 7 h. Thus, 255
Modifying Batch and Continuous Reaetors

Bioreaetors for Suspension and Immobilized cultures .

254
 1.5

"
6- 4-X_
8- 1.0
7- O/I
X. 3-
dX/dt
2-
gll-h

2 4 6
X gll t 8
XI
0.15

0.10
dP/dt
gll-h
o
O 2 4 6 8 O .2 .4 .6
Xl
X, O/I P, g/J
Figure 9.6. Solution ofExarnple 9,2 with a single stage, where e = 14h,
o
O
0.8
Figure 9.7. Solution to Example 92 using
0.143 h-l. The intersection of the reaction curve with the straight line determi a1temative graphica1 approach.
Dl(Xj - Xo) = DlXl is the solution to eq. 9.24a. For the second stage, we consider the
tion phase and use eq. 9.26a. The predicted values of Xl and P2 are the same as in
approach. Note thal the dP/dt-versus-P curve is displaced in time from the dXldt-
X curve. Consequently,we use the dXldt plot before using the dP/dt plot.
where 50 is the initial substrate concentration, Y~ is the yield coefficient, and Xo is the
initial biomass concentration. When biomass concentration reaches its maximum value
9.3.3. Fed-batch Operation (Xm), the substrate concentration is very low, 5 « 50' and also Xo « X. That is,
Xm '" Y~o. Suppose that at Xm == Y~50' a nutrient feed is started at a tlow rate F, with the
ln fed-batch culture, nutrients are continuously or semicontinuously fed, while efll substrate concentration 50' The total amOunt of biomass in the vessel is X' = \IX, where V
is the culture volume at time t. The rate of increase in culture volume is
removed discontinuously (Fig. 9.8). Such a system is caIled a repeatedfed-batch
dV
Fed-batch culture is usuaIly used to overcome substrate inhibition or catabolite
by intermittent feeding of the substrate. If the substrate is inhibitory, intermitten
-=F
dt
(9.28)
of the substrate improves the productivity of the fermentation by maintaining the Integration of eq. 9.28 yields
concentration low. Fed-batch operation is also called the semicontinuous
variable-volume continuous culture. Consider a batch culture where the concenl V= Va+Ft
biomass at a certain time is given by (9.29)
wbere Vo is the initial culture volume (I).
X = Xo + Yts(So -S) The biomass concentration in the vessel at any time t is
X=X'/V
(9.30)
256 Bioreactors for Suspension and Immobilized Cultures
MOdifying Batch and Continuous Reactors
257
 The balance on the rate-limiting substrate without maintenance energy is
F,So
dS' = FSo _ IlnerX' (9.35)
dt y:~

llJ Start
where S' is the tota! amount of the rate-Iimiting substrate in the cultore and So is the con-
centration of substrate in the feed stream.
At quasi-steady state, X' = VXm and essentially all the substrate is consumed, so no
significant level of substrate can accumulate. Therefore,

FSo = I! ••• X'

Y~s
(9.36)

SE Fill
Equation 9.31 at quasi-steady state with S ""Oyields

- dt = Xm -dt
dX' ("dV)
= XmF = FYxIsSo
M
(9.37)

Integration of eq. 9.37 from t = O to t with the initial amount of biomass in the reactor
being Xó yields

Figure 9.8.
X' = X~+FYftsSot (9.38)

V.' X. S, P I Harvest culture. That is, the total amount of cell in the culture increases linearly with time (which is exper-
imentally observed) in a fed-bateh cultore. Dilution rate and therefore fJ.n •• decrease with
time in a fed-bateh culture. Since !J.ne, = D at quasi-steady state, the growth rate is con-
Tbe rate of change in biomass concentration is trolled by the dilution rate. The use of unstructured models is an approximation, since fJ.n ••
is a function of time.
dX V(dX'/dt)-X'(dV/dt)
Produet profiles in a fed-batch culture can be obtained by using the definitions of
dí V2
YPIS or qp. When the product yield coefficient YPIS is constant, at quasi-steady state with

S «So
Since dX'/dt= 1J.ne,X', dV/dt= F, and FIV=D, eq. 9.31 becomes
dX
Ps YPISSO (9.39)
dí = (I1net - D)X or the potential product output is
FP"" YPlsSoF (9.40)
When the substrate is totally consumed, S "" O and X = Xm = ylfJsSo. Furthermore•
nearly all the substrate in a unit volume is consumed. then dXldt = O. This is an e When the specific rate of product formation qp is constant,
of a quasi-steady state. A fed-batch system operates at quasi-steady state when IIi dP' , (9.41)
consumption rate is nearly equal to nutrient feed rate. Since dXldt = O at quasi· -=qpX
dt
state, then
!J.net =D where P' is the tota! amount of product in culture.
Substituting X' = (Vo+ Ft)Xm into eq. 9.41 yields
If maintenance energy can be neglected, dP'
S (9.42)
-=qpXm(Vo+Ft)
dt
I1net = 11m K+S
..

then
(9.43)
Ss Kp P " = Po+qpXm ( Vo +2"
Ft) t
11m -D

Modifying Batch and Continuous Reactors 259

Bioreactors for Suspension and Immobilized Cultures'

258
 tions such as lactic acid and other plant cell and mamma!ian cell fermentations. where the
rate of product formation is maxima! at low nutrient eoncentrations.
In tenns ofproduet eoneentration, eq. 9.43 ean be written as Fed-batch culture is important for E. coli fermentations to make proteins from reeom-
binant DNA techDology. To make a high concentration of product, it is desirable to grow the
P=Po ~ +qpXm(~ + ~} culture to very high cell density before inducing production of the target protein. If E. coli
has an unlimited supply of glucose it will grow at a maxima! rate, but produce organic acids
Figure 9.9 depiets the variation of V, !J. (= D), X, 5, and P with time at quasi-steady (e.g., acetic acid) as by-products. The accumulation of these by-products inhibits growth.
state in a single eyc1e of a fed-bateh eulture. If glucose is fed at a rate that substains the growth rate at slightly less than maxima!, E. coli
In some fed-bateh operations, part of the eulture volume is removed at eertain inter· uses the glucose more efficiently, making less by-product. Very high cell densities (50 to 100
va!s, sinee the reaetor volume is limited. This operation is ea!led the repeated fed-batch gll) can be achieved. Fed-batch culture may benefit from active process contro!. For exam-
culture. The eulture volume and dilution rate (= llneJ undergo eyeliea! variations in this ple, the feed rate of glucose could be controlled by measuring glucose concentration in the
medium or the CO2 evolution rate using a feedback controller.
operation.
If the eycle time tw is eonstant and the system is a!ways at quasi-steady state, then
Example9.3
the produet coneentration at the end of eaeh eycle is given by
In a fed-batch culture operating with intennittent addition of glucose solution. values of the
Pw=rPo + qpxm(r + D~w }w following parameters are given at time t = 2 h. when the system is at quasi-steady state.

v = 1000 ml F= dV =200 m1/h

where Dw = FlVw' Vw is the eulture volume at the end of each cycle, Vo is the residua! c dt
ture volume after remova!, y is the fraetion of eulture volume remaining at eaeh cY' So = 100 g glucose/l Ilm = 0.3 h-I
(= VdVw)' and tw is the eycle time.
The eycle time is defined as Ks = 0.1g g1ucose/l Y~s = 0.5 gdw cellsl g glucose
t
Vw-Vo
== ~ = --,,--,---",---l-y
Vw-yVw
- X~ =30 g
w F F Dw

a Find Vo (the initial volume of the culture).

Substitution of eq. 9.46 into eq. 9.45 yields b. Deterrnine the concentration of growth-Iirniting substrate in the vessel at quasi-steady
state.
Pw = rPo + q;~mw (1- r2) c. Detennine the concentration and tota! amount of biomass in the vessel at t = 2 h (at
quasi-steady state).
d. If qp = 0.2 g product/g cells, Po = O,deterrninethe concentration of product in the vessel at
An example of fed-bateh eulture is its use in some antibiotie fennentations, w. t=2h.
glucose solution is intermittently added to the fennentation broth due to the repressi' Solution
pathways for the production of seeondary metabolites eaused by high initia! glueose a. V= Vo+Ft
eentrations. The fed-batch method ean be applied to other secondary metabolite fl Vo = 1000 - 200(2) = 600 ml
b. D= FIV= 0.2 h-1

KsD
S ~--- = (0.1)(0.2) - O.2 ggucose
I II
p'Ypx Ilm - D 0.3-0.2

c. X' = X~+ FY::sSot

p(qpCOllIIonII
=30+(0.2)(0.5)(100)(2)=50 g

So FiguN 9.9. (a)Variationofcul! d. P=Po V +q P Xm (Vv

Vv V'+T
Dt) t
Time (V) • specificgrowthrate (I.l). ceIl
(b) substrate(S) concentration withIÍ
x quasi-steadystate.(b)Variation
of., = 0+(0.2)(50) 1000 + ---
(600 (0.2)(2») (2)
(P) concentrationwithtimeatq . =16 g/l 2
L. statein a singlecycleofa fed-b
Time
Modifying Batch and Continuous Reactors 261

260
Bioreaetors for Suspension and Immobilized CulturElS
 9.4. IMMOBILlZED CELL SYSTEMS

9.3.4. Perfusion Systems 9.4.1. Introdu~ion

An alternative to fed-batch culture is a perfusion system. Such systems are used most
ImmobiIization of cells as biocatalysts is almost as common as enzyme immobilization.
often with animal cell cultures (see Chapter 12). The basic characteristic is constant
Immobilization is the restriction of cell mobility within a defined space. Immobilized cell
medium {low, cell retention, and in some cases selective removal of dead cells. High cel!
cultures have the following potential advantages over suspension cultures.
density can be achieved. Cell retention is usually achieved by membranes or screens or by
a centrifuge capable of se1ective cell removal. When a membrane is used, the system has
characteristics of an immobilized cell system (see Section 9.4) except the cells are usual!y 1. Immobilization provides high cell concentrations.
maintained in suspension and rni.J(ed.With a selective removalJrecycle the system ap- 2. Immobilization provides cell reuse and eliminates the costly processes of cell recov-
proaches the cell recycle reactor discussed earlier in this chapter. Figure 9.10 depicts one ery and cell recycle.
type of perfusion system. 3. Immobilization eliminates cell washout problems at high dilution rates.
The potential advantages of a perfusion system is the potential removal of cell
bris and inhibitory by-products, removal of enzymes released by dead cells that may 00' 4. The combination of high cell concentrations and high flow rates (no washout re-
strictions) allows high volumetric productivities.
stroy product, shorter exposure time of product to potentially harsh production conditio
(compared to batch or fed-batch operation), high per-unit volumetric productivity (due 5. Immobi1ization may also provide favorable microenvironmental conditions (Le.,
high cell density and metabolism), and a rather constant environment. cell-cell contact, nutrient-product gradients, pH gradients) for cells, resulting in
The primary disadvantage is that a large amount of medium is typically used better performance ofthe biocatalysts (e.g., higher product yields and rates).
the nutrients in the medium are less completely utilized than in batch or fed-batch 6. In some cases, irnmobilization improves genetic stability.
tems. High medium usage is expensive, owing not only to the high cost of raw mate 7. For some cells, protection against shear damage is important.
but also to the costs to prepare and sterilize the medium. Additionally, costs for
treatrnent increase. Typically the bioprocess engineer must consider the trade-off of .
proved product quality and reactor productivity with the extra costs associated wil The major limitation on immobilization is that the product of interest should be ex-
more complex reactor system (membranes, pumps, centrifuga! separator, etc.) creted by the cells. A further complication is that immobilization often leads to systems
creased medium usage. The best choice depends on the specific situation. for which diffusionallimitations are important. In such cases the control of microenviron-
mental conditions is difficult, owing to the resulting heterogeneity in the system. With liv-
ing cel!s, growth and gas evolution present significant problems in some systems and can
lead to significant mechanical disruption of the immobilizing matrix.
Product
In Chapter 3 we discussed enzyme immobilization. Figure 3.16 provides a useful
summary of immobilization strategies. Many of the ideas in enzyme immobilization have
a direct counterpart in whole cells. However, the maintenance of a living cell in such a
system is more complex than maintaining enzymatic activity. The primary advantage of
irnmobilized cells over immobilized enzymes is that immobi1ized cells can perform multi-
step, cofactor-requiring, biosynthetic reactions that are not practical using purified en-
Live CeU Return zyme preparations.

9.4.2. Aetive Immobilization of Cells

Supernotert ,
Woste
Active immobilization is entrapment or binding of cells by physical or chemical forces.
The two major methods of active immobilization are entrapment and binding.
a Physical entrapment within porous matrices is the most widely used method of celI
Oead Cell~ . inunobilization. Various matrices can be used for the immobilization of cells. Among
Medium
.'these are porous polymers (agar, alginate, K-carrageenan, polyacrylarnide, chitosan,
Reservoir Bioreoctor
'gelatin, collagen), porous metal screens, polyurethane, silica gel, polystyrene, and cellu-
'lOse triacetate.
Fig. 9.10. Schematic of a perfusion system with externa! centrifugation and returll
cells. Internal retention of cel1s is a!so possible. ReturD of spent medium is optional.
.9,4 Immobilized Ce" Systems 263

262 Bioreactors for Suspension and Immobilized CulturéS

 Encapsulation is another method of cell entrapment. Microcapsules are hollow,
spherical particles bound by semipermeable membranes. Cells are entrapped within the
Polymer beads should be porous enough to allow the transport of substrates and
hollow capsule volume. The transport of nutrients and products in and out of the capsule
products in and out of the bead. They are usually formed in the presence of cells and can
takes place through the capsule membrane. Microcapsules have certain advantages over
be prepared by one of the following methods:
gel beads. More cells can be packed per unit volume of support material into capsules,
1. Gelation oj polymers: Gelatin and agar beads may be prepared by mixing the and intraparticle diffusion limitations are less severe in capsules due to the presence of
liquid form of these polymers with cell suspensions and using a template to form beads. liquid cell suspension in the intracapsule space. Various polymers can be used as capsule
Reduction of temperature in the templates causes solidification of the polymers with the membranes. Among these are nylon, collodion, polystyrene, acrylate, polylysine-alginate
cells entrapped. Gel beads are usually soft and mechanically fragile. However, we can use hydrogel, cellulose acetate-ethyl cellulose, and polyester membranes. Different mem-
a hard core (glass, plastic) and a soft gelatin shell with entrapped cells to overcome some branes (composition and MW cutofi) may need to be used for different applications in
mechanical problems associated with polymer beads. Because of diffusionallimitations, order to retain some high-MW products inside capsules and provide passage to low-MW
the inner core of such beads is ofted not active, so this approach does not necessarily de- nutrients and products.
crease the amount of product made per bead. Another form of entrapment is the use of macroscopic membrane-based reactors.
. The simplest of these is the hollow-fiber reactor. This device is a mass-transfer analog of
2. Precipitation oj polymers: Cells are dispersed in a polymer solution, and by
the shell-and-tube heat exchanger in which the tubes are made of semipermeable mem-
changing the pH or the solvent, the polymer can be precipitated. The starting solution of branes. Typically, cells are inoculated on the shell side and are allowed to grow in place.
the polymer has to be prepared with an organic solvent or a water-solvent mixture. The nutrient solution is pumped through the insides of the tubes. Nutrients diffuse through
Ethanol and acetone are examples of water-miscible solvents. polymers used for this pur- .
the membrane and are utilized by the cells, and metabolic products diffuse back into the
pose are polystyrene, cellulose triacetate, and collagen. The direct contact of cells with' flowing nutrient stream. Owing to diffusional limitations, the unmodified hollow-fiber
solvents may cause inactivation and even the death of cells.
unit does not perform well with living cells. Modifications involving multiple membrane
3. lon-exchange gelation: lon-exchange gelation tak.es place when a water-solub types (for example, for gas exchange or extractive product removal) or changes to pro-
polyelectrolyte is mixed with a salt solution. Solidification occurs when the polyeI mote convective flux within the cell layer have been proposed. Several commercial reac-
trolyte reacts with the salt solution to form a solid gel. The most popular example of tors for animal cell cultivation use membrane entrapment.
kind of gelation is the formation of Ca-alginate gel by mixing Na-alginate solution wi' ln addition to entrapment or encapsulation, cells can be bound directly to a support.
CaClz solution. Some other polymers obtained by ion-exchange gelation are Al-algin Immobilization of cells on the surfaces of support materials can be achieved by physical
Ca/Al carboxymethyl cellulose, Mg pectinate, lC-carrageenan, and chitosan pol adsorption or covalent binding.
phosphate. Alginate and lC-carrageenan are the most widely used polymers for ce' Adsorption of cells on inert support surfaces has been widely used for cell immobi-
irnmobilization purposes. lonic gels can be further stabilized by covalent cross-linking. lization. The major advantage of immobilization by adsorption is direct contact between
nutrient and support materials. High cellloadings can be obtained using microporous sup-
4. Polycondensation: Epoxy resins are prepared by polycondensation and can port materials. However, porous support materials may cause intraparticle pore diffusion
used for cell immobilization. polycondensation produces covalent networks with .. limitations at high cell densities, as is also the case with polymer-entrapped cell systems.
chemical and mechanica! stability. Usually, liquid precursors are cured with a mulu Also, the control of microenvironmenta! conditions is a problem with porous support ma-
tiona! component. Functional groupS usua!ly are hydroxy, arnino, epoxy, and isoc: terials. A ratio of pore to cell diameter of 4 to 5 is recommended for the immobilization of
groups. Some examples of polymer networks obtained by polycondensation are e cells onto the inner surface of porous support particles. At small pore sizes, accessibility
polyurethane, silica gel, gelatin-glutaraldehyde, albumin-glutaraldehyde, and collag' of the nutrient into inner surfaces of pores may be the limiting factor, whereas at large
glutaraldehyde. Severe reaction conditions (high temperature, low or high pH valuesJ pore sizes the specific surface area may be the limiting factor. Therefore, there may be an
toxic functiona! groupS may adversely affect the activity of cells. optimal pore size, resulting in the maximum rate of bioconversion.
5. Polymerization: Polymeric networks can be prepared by eross-linking C' . Adsorption capacity and strength of binding are the two major factors that affect the
mers of a vinyl group containing monomers. polyacrylarnide beads are the most selection of a suitable support material. Adsorption capacity varies between 2 mg/g
used polymer beads, prepared by copolymerization of acrylarnide and bisacrylarni' (porous silic a) and 250 mg/g (wood chips). Porous glass carriers provide adsorption ca-
eral different monomers can be used for polymer formation; acrylamide, methacryl: pacities (lOS to 109 cells/g) that are less than or comparable to those of gel-entrapped cell
and 2_hydroxyethyl methacrylate are the most widely used. Cross-linking is usua1l: concentrations (109 to 1011 cellslml). The binding forces between the cell and support sur-
ated by copolymerization with a divinyl compound, such as methylenebis-acrylarni' faces may vary, depending on the surface properties of the support material and the type
Immobilization by polymerization is a simple method. The polymerizing sol of cells. Electrostatic forces are dominant when positively charged support surfaces (ion-
mixed with the cell suspension, and polymerization takes place to form a polymeric exchange resins, gelatin) are used. Cells also adhere on negatively charged surfaces by co-
which is pressed through a sieve plate to obtain regular-shaped particles. Suspen: valent binding or H bonding. The adsorption of cells on neutral polymer support surfaces
emulsion polymerization can also be used to form polymeric beads for cell entraptn'
Immobilized Ce/l Systems 265
264 Bioreactors for Suspension and Immobilized Cultures
may be mediated by chemical bonding, such as cova!ent bonding, H bonds, or van der TABLE 9.1 Examples of Celllmmobilization by Entrapment Using Different
Support Materials
Waals forces. Some specific chelating agents may be used to develop stronger celI-surface
interactions. Among the support materials used for celI adsorption are porous glass, Cells Support matrix Conversion
porous siIica, alumina, ceramics, gelatin, chitosan, activated carbon, wood chips,
S. cerevisiae
polypropylene ion-exchange resins (DEAE-Sephadex, CMC-), and Sepharose. K-Carrageenan or polyacrylamide Glucose to ethanol
E. aerogenes K-Carrageenan Glucose to 2,3-butanediol
Adsorption is a simple, inexpensive method of celI immobiIization. However, lim- E. coli K-Carrageenan Fumaric acid to aspartic acid
ited celI loadings and rather weak binding forces reduce the attractiveness of this method. Trichoderma reesei K-Carrageenan Cellulose production
Hydrodynamic shear around adsorbed cells should be very mild to avoid the removal of Z. mobilis Ca-alginate Glucose to ethanol
cells from support surfaces. Acetobacter sp. Ca-alginate Glucose to glucomc acid
Morinda citrifolia Ca-alginate
Cova!ent binding is the most widely used method for enzyme immobilization. How· Anthraquinone formation
Candida tropicalis Ca-alginate Phenol degradation
ever, it is not as widely used for celI immobiIization. Functional groups on celI and sup- Nocardia rhodocrous Polyurethane Conversion of testosterone
port materia! surfaces are not usually suitable for cova!ent binding. Binding surfaces need E. coli Polyurethane Pemcillin G to G-APA
to be specialIy treated with coupIing agents (e.g., glutara!dehyde or carbodiimide) or reac- Catharantus roseus Polyurethane
lsocitrate dehydrogenase aetivity
Rhodotorula minuta Polyurethane
tive groups for cova!ent binding. These reactive groups may be toxic to cells. A number of Menthyl succinate to menthol
inorganic carriers (metal oxides such as titanium and zirconium oxide) have been devel·
oped that provide satisfactory functional groups for cova!ent binding.
Cova!ent binding forces are stronger than adsorption forces, resulting in more stablé ln mixed-culture microbia! films, the presence of some polymer-producing organ-
binding. However, with growing celIs, large numbers of cell progeny must be lost. Sup- isms facilitates biofilm formation and enhances the stability of the biofiIms. Microenvi-
port materials with desired functional groups are rather limited. Among the support mate· ronmental conditions inside a thick biofiIm vary with position and affect the physiology
ria!s used for covalent binding are CMC plus carbodiimide; carriers with aldehyde, amine, of the cells.
epoxy, or ha!ocarbonyl groups; Zr(IY) oxide; Ti(IY) oxide; and cellulose plus cyanuric ln a stagnant biologica! film, nutrients diffuse into the biofiIm and products diffuse
chloride. Support materia!s with -OH groups are treated with CNBr, materia!s with out into liquid nutrient medium. Nutrient and product profiIes within the biofilm are im-
-NH2 are treated with glutaraldehyde, and supports with COOH groups are treated with portant factors affecting cellular physiology and metabolism. A schematic of a biofilm is
carbodiimide for covalent binding with protein groups on cell surfaces. depicted in Fig. 9.11. Biofilm cultures have a!most the same advantages as those of the
The direct cross-linking of cells by glutara!dehyde to form an insoluble aggregate . immobiIized celI systems over suspension cultures, as Iisted in the previous section.
more like cell entrapment than binding. However, some cells may be cross-linked The thickness of a biofiIm is an important factor affecting the performance ofthe bi-
adsorption onto support surfaces. Cross-Iinking by glutaraldehyde may adversely affect, otic phase. Thin biofilms wiII have low rates of conversion due to low biomass concentra-
the cell's metabolic activity and may also cause severe diffusion limitations. Physica! tion, and thick biofilms may experience diffusionally limited growth, which may or may
cross-linking may a!so be provided by using polyelectrolytes, polymers such as chitos not be beneficial depending on the cellular system and objectives. Nutrient-depleted re-
and salts [CaCI2,Al(OH)3,FeCI3]. Direct cross-linking is not widely used because of gions may also develop within the biofilm for thick biofilms. In many cases, an optima!
aforementioned disadvantages. biofiIm thickness resulting in the maximum rate of bioconversion exists and can be deter-
A good support materia! should be rigid and chernically inert, should bind ce mined. In some cases, growth under diffusion limitations may result in higher yields of
firmly, and should have high loading capacity. In the case of gel entrapment, gels shouJ, products as a result of changes in celI physiology and cell-eelI interactions. In this case,
be porous enough and particIe size should be smalI enough to avoid intraparticIe diffusio!
limitations.
TABLE 9.2 Examples of Celllmmobilization by Surface Attachment
Some examples of cell immobilization by entrapment and by surface attachm
Cells Support surface
(binding) are summarized in Tables 9.1 and 9.2, respectively. Conversion

Lactobacillus sp. Gelatin (adsorption) Glucose to lactic acid

Clostridium acetobutylicum Ion-exchange resins Glucose to acetone, butanol
9.4.3. Passive Immobilization: Biological Films Streptomyces Sephadex (adsorption) Streptomycin
Anima! cells Hormones
DEAE-sephadexlcytodex (adsorption)
E.coli
Biological films are the multilayer growth of cells on solid support surfaces. The sup Ti(lV) oxide (covaient binding)
B. subtillis Agarose-cartJodiimide (covalent
material can be inert or biologica!ly active. Biofilm formation is common in natural
binding)
industrial fermentation systems, such as biological waste-water treatment and mold Solanum aviculare
Polyphenylene oxide-glutaraidehyde Steroid glycoalkaloids formation
mentations. The interaction among cells and the binding forces between the celI and s (covalent binding)
port material may be very complicated.

266 Bioreaetors for Suspension and Immobilized Cultures Sec. 9.4 Immobilized Cell Systems 267
 Da = maximum rate of bioconversion = rmax (9.48)
maximum rate of diffusion (D/o)So

where rmax is the maximum rate of bioconversion (mg Sil h), De is the effective diffusivity
of the rate-limiting substrate, o is the thickness of diffusion path (or liquid film), and So is
the bulk substrate concentration in liquid phase. When the film-theory model applies,
D/o is the mass transfer coefficient (Le., kL = D/o).
If Da » I, the rate of bioconversion is diffusion limited; for Da « I, the rate is
limited by the rate of bioconversion; and for Da '" I, the diffusion and bioreaction rates
are comparable. It is desirable to keep Da < I to eliminate diffusion limitations when the
Figure 9.11 Schematic representation productivity of a cell population does not improve upon immobi1ization due to cell-eell
biofilm.
contact and nutrient gradients.
Diffusiona!limitations may be extemal (that is, between fluid and support surface in
adsorption and cova!ent binding), intrapartic1e (Le., inside partic1es in entrapment, encap-
improvement in reaction stoichiometry (e.g., high yield) may.overcome the reduction in su1ation, or irnmobi1ization in porous partic1es), or both. If the extema! mass transfer is
reaction rate, and it may be more beneficia! to operate the system under diffusion limita- limiting, an increase in liquid-phase turbulence should result in an increase in the reaction
tions. Usua1ly, the most sparingly soluble nutrient, such as dissolved oxygen, is the: rate. In case of intrapartic1e mass-transfer limitations, a reduction in partic1e size or an in-
rate-limiting nutrient within the biofilm. A typica! variation of dissolved oxygen concen-. crease in the porous void fraction of the support materia! should result in an increase in
tration within the biofilm is depicted in Fig. 9.12. the rate of the bioreaction.
In Chapter 3 we discussed in reasonable detail a mathematica! model of the interac-
9.4.4. Diffusional Limitations in Immobilized Cell Systems tion of diffusion and reaction for surface immobi1ized or entrapped biocatalysts. These
model s apply directly to immobi1ized cells when the kinetics of bioconversion are de-
Immobi1ization of cells may cause extra diffusiona!limitations as compared to suspensi, scribed by a Michaelis-Menten type of kinetic expression. Thus, the reader may wish to
cultures. The presence and significance of diffusiona! limitations depend on the relat" consult Chapter 3 again.
rates of bioconversion and diffusion, which can be described by the Damkohler numl Another interesting case is to consider biofilms where we allow cell growth. Models
(Da) (see eq. 3.52 a!so). for immobi1ized enzymes have no terms for biocata1yst replication, so this case presents a
new problem.
0'00 The thickness of a biofilm or the size of microbia! floc increases with time during
the growth phase. A microbial floc is an aggregation of many cells, and in some processes
sr
1
0'01 these aggregates can be more than 1 mm in diameter. However, since the rate of increase
7
e in biofi1m thickness is much slower than the rate of substrate uptake, the system can be
assumed to be at quasi-steady state for relatively short periods. The simplest case is to
16
...••
0'02 assume that the system is at quasi-steady state and a!1 the cells inside the biofilm are
0·03lJ..
in the same physiological state. In this situation we write a steady-state substrate bal-
OS li
ii
"a
ance within the biofi1m by using average kinetic constants for the biotic phase (living
cells).
0'04\
8
4
3 O'OS

0'06S
! Figure 9.12. Dissolved-oxygen pro
and oxygen gradients in a microbial : .,
bathed in flowing medium: -A-A- ox
A differential materia! ba!ance for the rate-limiting substrate within the biofilm (see
Fig. 9.11) yields at steady state

~ 2 profi1e for 20 ppm nutrient broth, 27.5'

D d2; =_1_ JlmS X (9.49)
0'07 _ -{)xygen gradient for this profi1e; -, e dy YX1S Ks +S
1 oxygen profi1e for 500 ppm nutrient bi
260C; oxygen gradient for this prolíl
o (With pennission, from H. R. Bungay . where De is the effective diffusivity (cm2/S) and YXIS is the growth yield coefficient
100 50 o 50 others, Bioteehno/. Bioeng. 1I:765, 19'
(g cells/g substrate).
Ibofttum John Wl1ey & Sons, Inc., New Yor\<.)
Dlstance (MlD)

Sec. 9.4 Immobilized Cell Systems 269

268 Bioreactors for Suspension and Immobilized Cultures
 The boundary conditions are
S=SQi aty = O

dS =0 aty = L
cly

where L is the thickness of biofilm.

If it is also assumed that the liquid nutrient phase is vigorously agitated and the liq-
uid film resistance is negligible, then So '" Soi' By defining a maximum rate of substrate
utilization as rm = Il",x/YXIS (g subs/cm' h), we rewrite eq. 9.49 as Figure 9.13. Effectiveness factor for a fiat
biofilm as a function of ~, the dimensionless
"Dd2S=~ initial substrate concentration, and ljl, the
e dl Ks+S 100 1000 Thiele modulus. (With pennission, redrawn
from B. Atkinson, Biochemica/ Reactors,
In dimensionless form, eq. 9.50 can be written as Pion Ltd., London, 1974, p. 81.)

d2S <jl2S
di = 1+~S
where
(<jl< 1) to elirninate diffusion limitations. As the biofilm grows (slowly), the value of <jl
- S will gradually
decrease increase. If shear forces cause a portion of the film to detach, then <jlwill
abruptly.
s=-S' -
y=r;'
y
~=~
o Ks The effectiveness factor (11)can be calculated as
and
11= 1 - <jl tanhOl _
tanh<jl(~ I), for 00:;;;1 (9.55)
<jl=L
YXISDeKs
IlmX _ L ~ DeKs
rm
11=001 - -<jl-
tanh<jl( tanh<jl
1 - 1,J for 00;:::1 (9.56)
Equation 9.51 can be solved numerically. An analytical solution can be derived for
lirniting cases of zero or first -order reaction rates. where 00is the modified Thiele modulus and is given by
The maximum rate of substrate flux in the absence of diffusion lirnitations is
by the following equation:

In(I+~Jr/2 (9.57)
~(1+~J Ks
00= <jl(SoIKs) [So _
N s As = -A sed
D dS I = K rmSO (LA)s
s + So
y y=O Ks J

where As is a surface area of biofilm available for substrate flux, Ns is the substrate Some cells such as molds (A. niger) form pellets in a fermentation broth, and sub-
and L is the thickness of the biofilm.
strates need to diffuse inside pellets to be available for rnicrobial consumption. Cells may
In the presence of diffusion lirnitation, the rate of substrate consumption or f1, form biofilms on spherical support particles, as depicted in Fig. 9.14. Sirnilar equations
expressed in terms of the effectiveness factor. need to be solved in spherical geometry in this case to determine the substrate profile
within thewithin
equation floc and the substrate
the microbial flocconsumption
is rate. The dimensionless substrate transport

N s---D edy
dSl y=o-11
- ( Ks+So
rmSO JL

where 11is the effectiveness factor, defined as the ratio of the rate of substrate con:
-+--=-
d2S 2 dS <jl2S
(9.58)
dr2 r dr I+S/W
tion in the presence of diffusion lirnitation to the rate of substrate consumption in where
sence of diffusion limitation. In the absence of diffusion limitations, 11 == 1, and
presence of diffusion limitations, 11< 1. The effectiveness factor is a function of <jl S=~ r
s.' r==-, w=.:&
Figure 9.13 is a plot of 11versus ~ for various values of <jl.The <jlvalue should o R
Ks

270 Bioreactors for Suspension and Immobilized Cultures Sec. 9.4

Immobilized CeIJ Systems
271
 Tbe rate of bioreaction can be approximated to zero order at values of S » Ks. Be-
b) cause Ks is often very small, the zero-order limit useful!y describes many systems of prac-
o) tical interest.
50
50
rs = ---=1:
/lmX
(9.64)
Yx/s m

Figure 9.14. (a) Microbial film on inert

spherical support particle. (b) Spherical Tbe solution to eq. 9.58 in this case is
microbial floc.
S=So --.!Í!L(K _(2) (9.65)
6D.
"
and Substrate concentration may be zero at a certain radial distance from tbe center of
tbe floc according to eq. 9.65. This distance is called tbe critical radius (rcr) and is deter-
<\>=R mined by setting S = O at rá in eq. 9.65.
I!mX =Rt
Yx1sD.Ks rm=
D.Ks

Tbe boundary conditions are R

(rcr)2 =1- 6D.So
rmR2 (9.66)
8=1, atr=1
When rcr > o-tbat is, R > (6D .501rm)'a_tben tbe concentration of tbe limiting sub-
d8
_= O atr=- O strate is zero for O < r < rcr In this case, tbe limiting substrate is consumed only in tbe
dr '
outer shell of the floc, and tbe effectiveness factor is given by

For nonspherical partic1es, a characteristic lengtb is defined as 4 3 3

L= Vp 1]= 3 =1- ..sr. (9.67)
Ap rm-1t(R
~1tR3.r-rcr) (r R )3
3 m

where Vp and Ap are tbe volume and surface area of rnicrobial pellet. or
The rate of substrate consumption by a single rnicrobial floc is

(9.68)
NsAp=-ApD dS\
'drr=R =1]~VP
Ks+So 1] = 1- (1- rmR2 J312
6D.So

The effectiveness factor (1]) is a function of <\>and ~. Variation of 1] witb <\>and ~ is

9.4.5. Bioreactor Considerations ln Immobilized
to that of Fig. 9.13. However, 1] values for spherical geometry are slightly 10W'
those of rectangu1ar geometry for intermediate values of <ji(1 < <\>< 10). An ana1yti' Cel! Systems
lution to eq. 9.58 is possible for first- and zero-order reaction kinetics.
The reaction rate can be approximated to fust order at low substrate conceh! Various reactor configurations can be used for immobilized cell systems. Since tbe sup-
"S 1:
port matrices used for cel! immobilization are often mechanical!y fragile, bioreactors witb
r =.J:!!l=-X =...!!!-S
low hydrodynamic shear, such as packed-column, fluidized-bed, or airlift reactors, are
s Yx/sKs Ks
preferred. Mechanically agitated ferrnenters can be used for some immobilized-cell sys-
tems if tbe support matrix is strong and durable. Any of these reactors can usual!y be op-
where r m = (I!JY XlS)X. The effectiveness factor in this case is given by
erated in a perfusion mode by passing nutrient solution through a column of immobilized
cells. Schemafic diagrams of immobilized-cell packed-column and fluidized-bed reactors
1[
1]=$L~-3<\>j1 11 are depicted in Fig. 9.15. Tbese reactors can be operated in batch or continuous mode.
Consider tbe reactors shown in Fig. 9.15. When tbe fluid recirculation rate is high,

where lesystem approaches CFSTR behavior. One commercial fluidized-bed, immobilized-

hnal-cel! bioreactor system requires high recirculation to maintain uniforrn conditions
tbe reactor. Tbe models we have discussed so far can be applied to such systems. Tbe
$= Ap D.
Vp ~rmIK=

,.9.4 Immobilized Cell Systems 27~

Bioreactors for Suspension and Immobilized CulturéS'

272
 For low substrate concentrations in the feed, the rate of substrate consumption is
Feed from
Reservoir r-
I
Feed from
Reservoir
fust order and eq. 9.71 has the following form:

ln ~ = _ 'f1rmLaA (9.72)
SOj ~H s
Recycle Recycle
Chomber Chomber
Substrate concentration drops exponentially with the height of tbe column in this case,
and a plot of ln So versus H results in a straight line. Equation 9.71 or 9.72 can be used as
tbe design equation for immobilized-biofilm column reactors to determine the height of
the column for a desired level of substrate conversion.

"
Example9A
Pump
t. Glueose is converted to ethanol by immobilized S, cerevisiae eens entrapped in Ca-alginate
:. Pump
beads in a packed column. The specifie rate of ethanol production is qp = 0.2 g ethanol/
g cen, h, and the average dry-weight cen coneentration in the boo is X = 25 gIl bed, Assume
Alternollve
Air lnlel that growth is negligible (Le., almost all glucose is converted to ethanol) and the bead size is
sufficiently small that TI :: l. The feed flow rate is F = 400 l/h, and glueose concentration in
the feOOis SQj= 100 g glueose/L The diameter of the eolumn is 1 m, and the produet yield co-
Figure 9.15. Sehematies of a paeked-bed and a fluidized-bed biofilm or immobilized- efficient is YPIS '" 0.49 g ethanol/g glucose.
cen bioreaetors are shown, In batch operation. only tbe sucams witb soUd Unes exist. In a. Write a material balance on the glucose concentration over a differential height of the col-
continuous operauon. tbe streams shown by dashed Unes are added, For the fluidized bed,
umn and integrate it to determine S = S(z) at steady state.
fluidizauon can be aceompUshed by Uquid reCÍfcnlation only or a ffiÍXture of liquid and
b. Determine the column height for 98% glucose eonversion at the exit of the column.
gas flows, c. Determine the ethanol concentration in the effluent.
Solution
a. A material balance on the glucose concentration over a differential height of the column
otber extreme involves some waste-treatment systems where tbe rate of fluid recircul: yields
is low or even zero. ln tbe latter case, Ihe system cannot be modeled as a CFSTR but
be treated as a PFR. To analyze such a system, eonsider a materlal balance on tbe
-FdSo=qpXdV=qpx Adz
limiting substrate over a differential element: YplS YP1S
-F tiSo =NsaA dz
lntegration yields
where So is tbe bulk liquid-phase substrate concentration (mg S/cm3) and is a funeti, s - H
height, F is tbe liquid nutrient flow rate (cm31h), N, is flux of substrate into tbe bi,
-Ff. 5",. dSo = qpX
Yp1S Ai o dz
(mg S/cm2 h), a is tbe biofilm or support partiele surface area per unit reactor
(cm2/cm3), A is tbe cross-sectional area of tbe bed (cm2), and dz is tbe differential
of an element of tbe column (cm), Substituting eq. 9.54 into eq. 9.69 yields the foJ so,.-s 0---8
_ qpX A
YPIS F
equation:

_FdSo =11 rmSo LaA This equation differs from the form of eq. 9.72 because So; is high and the reaction rate is
dz K, +So
effectively zero order.
b. So = 0,02(100) = 2 g glucoselL Substituting the given values into the above equation yields
Integration of eq. 9.70 yields
(100- 2) = (0.2)(25) (1r/4)(10/ 8
K ln~+(So' -So)= Tl'mLaA H 0.49 400
'So I F
8=49 dm=4.9 m

where SOi is the inlet bulk substrate concentration, L is tbe biofilm thicknesS or c. p,= Yp/S (So; - So) = 0.49(98) = 48 gIl.
teristic 1engtb of tbe support particle (L = Vp/Ap), and H is tbe total height of
bed.
; 9.4 .•.•..
Immobilized Cel! Systems
Bioreactors for Suspension and Immobilized Cultures
274
9.5. SOUD-STATE FERMENTATIONS
TABLE 9.3 SomeTraditional Food Fermentations
'T'L_. _
Solid-state fermentations (SSF) are fermentations of solid substrates at low moisture lev- HI~rmal processing 22
72
32
30
44
No
Y es lneubation
Time
Furtber
(oe)
(h)
Temp.
processing Yes
els or water activities. The water content of a typical submerged fermentation is more than 35(%) 36 moisture
40 Initial
95%. The water content of a solid mash in SSF often varies between 40% and 80%. Prirnary
genus
Rhizopus Common
Aspergillus 45 Temp. Tune
Product
Hamanatto
Solid-state fermentations are usually used for the fermentation of agricultural products or SoyMiso
sauce
substrate (oe) (min)
Tempeh
foods, such as rice, wheat, barley, com, and soybeans. The unique characteristic of
SoyDean,
solid-state fermentations is operation at low moisture levels, which provides a selective wheat 110 30
environment for the growth of mycelial organisms, such as molds. In fact, most solid-state Riee, 100 40
fermentations are mold fermentations producing extracellular enzymes on moist agricul- soybean
tural substrates. Since bacteria and-yeasts cannot tolerate low moisture levels (water activ- Soybean 100 30
ities), the chances of contarnination of fermentation media by bacteria or yeast are greatly Soybean,
wheat
reduced in SSF. A1though most SSFs are mold fermentations, SSFs based on bacteria and Sufu
yeast operating at relatively high moisture levels (75% to 90%) are also used. Solid-state Actinomucor Tofu 100 10 74 15 Yes
fermentations are used widely in Asia for food products, such as tempeh, rniso, or soy
With permission, from R. E. Midgett, in A. L. Demain and N. A. Solomon, eds., Manual oj Industrial Micro-
sauce fermentations, and also for enzyme production. biology and Biotechnology, ACS Publieations. Washington, O.c.. 1986.
The major advantages of SSFs over submerged fermentation systems are (I) the
small volume of fermentation mash or reactor volume, resulting in lower capital and oper·
ating costs, (2) a lower chance of contarnination due to low moisture levels, (3) easy prod"
uct separation, (4) energy efficiency, and (5) the allowing of the development of fully The major industrial use of the koji process is for the production of enzymes by
differentiated structures, which is critical in some cases to product formation. The major fungal species. Fungal amylases are produced by SSF of wheat bran by A. oryzae in a
disadvantage of SSFs is the heterogeneous nature of the media due to poor mixing cbar: rotating-drum fermenter. Wheat bran is pretreated with formaldehyde, and the initial pH
teristics, which results in control problems (pH, DO, temperature) within the fermentati()] of the bran is adjusted to pH = 3.5 to 4.0 to reduce the chance of contamination. Usually,
mash. To elirninate these control problems, fermentation media are usually rnixed eithl perforated pans, rotating drums, or packed beds with air ventilation are used. A typical
continuously or interrnittently. For large fermentation mash volumes, the concentratiOl rotary-drum type of koji fermenter is depicted in Fig. 9.16. Enzymes other than amylases,
gradients may not be elirninated at low agitation speeds, and mycelial cells may be d: such as cellulase, pectinase, protease, and lipases, can also be produced by koji fermenta-
aged at high agitation speeds. Usually, a rotating-drum fermenter is used for SSF syste. tions. bran
wheat Trichoderma viride species
in a rotary-tray have been used for the production of cellulases from
fermenter.
and the rotational speed needs to be optimized for the best performance.
Solid-substrate fermentations imply a more general method of fermentations . Some secondary metabolites, such as antibacterial agents, are produced by Rhizopus
which moisture content may not need to be low, but the substrate is in the form of s and Actinomucor species in some koji processes. Certain mycotoxins, such as aflatoxins,
merged solid particles in liquid media. Bacterial ore leaching (Le., growth and microb' were produced by SSF of rice (40% moisture) by A. parasiticus. Ochratoxins were also
oxidation on surfaces of rnineral sulfide particles) or fermentation of rice in a pac produced by Aspergillus species on wheat in a rotary-drum koji fermenter. Microbial
colurnn with circulating liquid media are examples of solid-substrate fermentati degradation of Iignocellulosics can also be accomplished by soIid-state fermentations for
Solid-state (or solid-phase) fermentations are a special form of solid-substrate ferme' waste-treatrnent purposes or in biopulping of wood chips for use in paper manufacture.
tions for which the substrate is solid and the moisture level is low. Spores from some molds have found use as insecticides. Proper spore formation is diffi-
The koji process is an SSF system that employs molds (Aspergillus, Rhizop, cult to obtain in submerged culture, and SSF must be used.
growing on grains or foods (wheat, rice, soybean). A typical SSF process involves Major process variables in SSF systems are moisture content (water activity), inocu-
stages. The first and the primary stage is an aerobic, fungal, solid-state fermentation lum density, temperature, pH, particle size, and aeration/agitation. Optirnization of these
grains cal1ed the koji. The second stage is an anaerobic submerged fermentation wi' parameters to maximize product yield and rate of product formation is the key in SSF
mixed bacterial culture called the moromi. The products listed in Table 9.3 are the systems and depends on the substrate and organism used. Most natural substrates
ucts of aerobic SSF, the koji process. Fermentation in the second stage (mororni) may (e.g., grains) require pretreatrnent to malce the physical structure of substrates more sus-
realized by using the natural flora, or, usual1y, with externally added bacteria and y' ceptible to myceIial penetration and utilization. Solid substrates are usually treated with
Some strains of Saccharomyces, Torulopsis, and Pediococcus are used as flavor produ- antimicrobial agents, such as formaldehyde, and are steamed in an autoclave. Nutrient
in soy sauce manufacture. The moromi is usually fermented for 8 to 12 months. How, media addition, pH adjustrnent, and the adjustment of moisture level are realized before
the processing time can be reduced to 6 months by temperature profi1ing. The final inoculation of the fermentation mash. Koji fermentations are usually realized in a
uct is pressed to recover the liquid soy sauce and is pasteurized, filtered, and bottled. controlled-hurnidity air environment with air ventilation and agitation. Many soIid-state
mycelial fermentations are shear sensitive due to disruption of the myceIia at high
276 Bioreaetors for Suspension and Immobilized Cultures Sec. 9.5
SOlid-State Fermentations
277
 INLET DUCT
IMNHOLE
OI1TLET DUCT
HEATEIl WHEAT BRAN
\

1·1' i Yi ",S
'i~'l\',..•••

o
::~::::~:::::::=;

AU1O'AATIC SOLlO CULTUIlE APf'ARATUS (IIOTARY PIIOCOSI ROTARY CHAMBER FOR SOLID CULTURE

Figure 9.16. Rotary-drum type of koji-making apparatus used for rice solids culture by Figure 9.17. Rotary, automatic koji-making apparatus. Tbe apparatus has a two-storied
A. oryzae. AU operations (washing, cooking, inoculation, loosening of solids, water spray- chamber. Each chamber has a large rotary tray on which wheat bran is heaped evenly.
ing, cooling, air circulation, fi1ling, and exhausting) can be done in this apparatus. (With After inoculated fungus has grown sufficiently, solid culture is transferred by a screw con-
permission, from N. Toyama, Biotechnol. Bioeng. Symp., Vol. 6, pp. 207-219, 1976, John veyor to the lower rotary-tray hopper. (With permission, from N. Toyama. Biotechnol.
Wiley & Sons, Inc., New York.) Bioeng. Symp., Vol. 6, pp. 207-219,1976, John Wiley & Sons, Inc .• NewYork.)

agitationlrotation speeds. At low agitation rates, oxygen transfer and CO2 evolution ral
become limiting, Therefore, an optimal range of agitation rate or rotation speed needs
be determined. Similarly, there is a minimum level of moisture content (-30% by wei; .. volume, low-product-value processes (e.g., waste treatrnent and fuel-grade ethanol pro-
below which microbial activity is inhibited. At high moisture levels (>60%), solid SI
duction). Multistage continuous systems improve the potential usefulness of continuous
strates become sticky and form large aggregates. Moreover, moisture level affects processes for the production of secondary metabolites and for the use of genetically unsta-
metabolic activities of cells. Optimal moisture level needs to be determined experi ble cells. The fed-batch system is widely used in commercial plants and combines the fea-
tally for each cell-substrate system. For most of the koji processes, the optimal moisl tures of continuous culture and batch that allow the manufacturer to maintain flexibility
level is about 40% ± 5%. partic1e size should be small enough to avoid any oxygen and ease of intervention. The perfusion system is another option that is particularly attrac-
tive for animal cells.
exchange or other nutrient transport limitations. Porosity of the partic1es can be imprl
by pretreatrnent to provide a larger intrapartic1e surface to volume ratio for mk lmmobilized cell systems offer a number of potential processing advantages, and
action. the commercialization of such systems is proceeding rapidly where cell culture is expen-
Most of the SSF processes are realized using a rotary-tray type of reactor in a sive and difficult (e.g., animal cell tissue culture). Physical entrapment or encapsulation is
used in most cases, although adsorption onto surfaces or covalent binding of cells to sur-
perature- and humidity-controlled chamber where controlled-humidity air is circ
through stacked beds of trays containing fermented solids. Figure 9.17 depi faces is possible.
rotary-tray chamber for koji fermentations. Rotary-drum fermenters are used le: In some cases, self-immobilization on surfaces is possible and a biofilm is formed.
quently because of the shear sensitivity of mycelial cells. Biofilm reactors can apply to tissue culture, mold, and bacterial systems. Biofilm-based
reactors are very important in waste-treatrnent applications and in natural ecosystems. The
analysis of immobilized cell reactors is analogous to that for immobiIized enzyme reac-
tors except for the feature of biocatalyst replication.
9.6. SUMMARY Solid-state fermentations share sorne characteristics with immobilized cell systems,
Bioreactors using suspended cells can be operated in many modes intermediate be but differ in that no discernible liquid is present. SSFs have found important uses in the
production of sorne traditional fermented foods and may have use in upgrading agricul-
batch reactor and a single-stage chemostat. Although a chemostat has potential p
tural or forest materials and in the production of mold products requiring full mold differ-
ity advantages for primary products, considerations of genetic instability, process entiation.
ity, and proces s reliability have greatly limited the use of chemostat units. The
recyc1e with a CSTR increases volumetric productivity and has found use.

Summary 279

278 Bioreactors for Suspension and Immobilized Cultures

 The reactor options described in this chapter are many. The best choice of reactor
systems wiU ultimately be determined by the choice of biocatalyst and the requirements
ally high value), and the yield factor YX1S = 0.5 g biomass/g substrate consumed. If normal
for product recovery and purification. operation is with a sterile feed containing 10 gII g1ucose at a rate of 100 1Ih:
a. What is the specific biomass production rate CgII-h)at steady state?

b. If recycle is usOOwith a recycle strearn of 10 IIh and a recycle biomass concentration five
SUGGESTIONS FOR FURTHER READING tion rate?
times as large as thal in the reactor exit, what would be the new specific biomass produc-

c. Explain any difference between the values found in parts a and b.

AlBA, S., A. E. HuMPHREY,ANDN. F. MILLlS, Biochemical Engineering, 2d ed., AcadeITUc
NewYork,1973. 9.2. In a two-stage chemostat system, the volumes of the fust and second reactors are V] = 500 I
ATKINSON,B., ANDF. MAVITUNA,Blóchemical Engineering and Biotechnology Handbook, 2d and V2 = 300 1, respectively. The first reactor is usOOfor biomass production and the second is
Stockton Press, NewYork, 1991.
cells.a secondary metabolite formation. The feed flow rate to the first reactor is F = 100 1Ih,
for
BAlLEY,J. E., ANDD. F. OLLlS,Biochemical Engineering Fundamentals, 2d 00., McGraw-Hill and the glucose concentration in the feed is S = 5.0 g/l. Use the fOllowing constants for the
CO., NewYork, 1986.
J.lm = 0.3 h-l, g dw cells
BLANCH,H. W., ANDD. S. CLARK,Biochemical Engineering, Marcel Dekker, Inc., New York, 199, Ks = 0.1 g/l, YXlS = 0.4 _
CHARACKLlS, W. G., R. BAKKE,ANDM. G. TRULEAR,1991, "Fundarnental Considerations ofE g glucose
Film Systems," in M. Moo-Young, 00., Comprehensive Biotechnology, Vol. 4, pp. 945-'
1985. a. Determine cell and glucose concentrations in the effluent of the fust stage.
b. Assume that growth is negligible in the second stage and the specific rate of product for-
CHlBATA,I., T. TosA, ANDT. SATO,"Methods of Cell Immobilization," in Manual of 1ndustria/
crobiology and Biotechnology, A. L. Demain and N. A. Solomon, eds., American Society mation is qp = 0.02 g PIg cell h, and YPIS = 0.6 g Plg S. Determine the product and sub-
9.3. strate concentrations in the effluent of the second reactor.
Microbiology, Washington, DC, pp. 217-229, 1986. Consider the following batch growth data:
DE GoOUER,C. D., W. A. M. BAKKER,H. H. BEEFTINK,ANDJ. 1'RAMPER,Bioreactors in Se' .
Overview of Design ProcOOures and Practical Appllcations, Enz;yme Microbiol Technol,14
202-219,1996. O
9h
17
13
15
12
16
18
10
10.5 1168 3
Tlme ---
11.4
0.021
0.005
0.130
0.042
7.4
5.1XP
7.8
8.0
dXJdt
g/l
.4
0.36
0.60
0.58
4.00.010
0.3
2.3
1.0
7.0
6.5
7.7
0.26
0.02
0.087
0.54
0.47
0.17
0.20
0.105
0.072
<0.011.3
<0.01
0.005
0.045
0.010
0.059
0.60
1.0
-O
0.55
0.30
0.060
0.025
dPldt
g/l-h
g/l-h
g/l
0.10

KARGI,F., ANDM. MOO-YoUNG, "Transport Phenomena in Bioprocesses," in M. Moo-Ya

Comprehensive Biotechnology, Vol. 2, Pergarnon Press, Elmsford, NY, pp. 5-55,1985.
KLElN,J., ANDK. D. VORLOP,"Irnmobi1ization Techniques: Cells," in M. Moo- Young, 00., C.
hensive Biotechnology, Vol. 2, Pergarnon Press, Elmsford, NY, pp. 203-334, 1985.
MERCILLE,S., M. JOHNSON,S. LAUTH!ER,A. A. KAMEN,ANDB. MASSIA,Understanding F:
LiITUtthe Productivity of Suspension-BasOO Perfusion Cultures Operated at High M
newal Rates, Biotechnology Bioengineering, 67: 435-450, 2000.
MIDGETI, R. E., "Solld State Ferrnentations," in A. L. Demain and N. A. Solomon, Manual
trial Microbiology and Technology, American Society for Microbiology, Washin:
pp. 66-83,1986.
Moo-YoUNG, M., Bioreactor 1mmobilized Enzymes and Cells: Fundamentals and APJ
Elsevier Science Publlshing, Inc., New York, 1988.
SCHROEDER, E. D., Water and Wastewater Treatment, McGraw-Hill Book Co., New York,
W ANG,D. I. C., ANDaTHERS,Fermentation and Enzyme Technology. John Wiley & Sons,
WEBSTER,I. A., M. L. SHULER,ANDP. RONY.The Whole Cell Hollow Fiber Reactor:
Factors, Biotech. Bioeng. 21: 1725-1748, 1979.

You have available three tanks of different volumes: 900, 600, and 300 I. Given a flow rate of
100 1Ih, what configuration oftanks would maxiITUzeproduct formation?
PROBLEMS
9.4. Penicillin is produced by P. chrysogenum in a fed-batch culture with the intermittent addition
of glucose solution to the culture medium. The initial culture volume at quasi-steady state is
9.1. Consider a 1000-1 CSTR in which biomass is being produced with glucose a$ Vo = 500 I, and glucose-containing nutrient solution is added with a flow rate of F = 50 1Ih.
The ITUcrobial system follows a Monod relationship with J.lm = 0.4 h-1, Ks = 1.5 G1ucose concentration in the feed solution and initial cell concentration are So = 300 gII and
Ks = 0.5
Xo 20 gII, and YX1S = 0.3The
gII, respectively. kinetic
g dw/g and yield coefficients of the organism are J.lm = 0.2 h-l,
glucose.

280 Bioreactors for Suspension and Immobilized Cultu

281