Bioreactor Analysis and Operation

Chapter 7

Overview of bioreactors, Modified batch and continuous

reactors, Immobilized cell systems
 Fermentation processes
Solid state: water content: 40~ 80%, mostly mold
fermentation on agriculture products and food: rice, wheat,
barley, corn and soybean.
e.g. rotary drum fermentator
Submerged systems: water content > 95%
e.g. bacteria, yeast.

JUST Biochemical Engineering ChE

483 2Chapter 7
Department of Chemical
Overview of bioreactors for submerged
system
Classification:
Operation modes:
- Batch: stirred tank
- Continuous: chemostat, fluidized-bed
- Modified types of the above modes:
fed-batch, chemostat with recycle,
multi-stage continuous reactors
Oxygen supply:
- aerobic: airlift
- anaerobic
Form of biocatalyst:
- free cell (enzyme)
- immobilized cell (enzyme)
- packed-bed, membrane reactor
3Chapter 7
Requirements for Cultivation Methods
Biomass concentration which must remain high

Sterile conditions being maintained

Effective agitation so that the distribution of substances in the

reaction is uniform

Heat removal

Creation of the correct shear conditions - high may damage

cells, low may lead to flocculation or growth on wall and stirrer

4Chapter 7
Comparison between Batch and Continuous Cultivations
The cycle time of batch
cultivation expressed by the
cell concentration, X, and the
time t, of cultivation
t0: harvest of cells
t1: preparation of another batch culture
t2: the lag time following inoculation; tm: log growth
tc = tm + t0 + t1 + t2 = tm + tl tl : lag phase
1 Xm Xm: maximum cell concentration
tc ln tl
m X i Xi: initial cell concentration
Xm Xi
Productivity in batch P
culture: cell ,batch 1 X m (mg cells/l . h)
ln t
m X i l
5Chapter 7
In continuous operation, Pcell = D X
and , Pmax = Dopt Xopt Ks
where Dopt m (1 )
K s S0
M
X opt Y X / S ( S 0 K s ( K s S 0 ) K s )
Thus, Ks
Pmax,cell ,cont Y X / S ( S 0 K s ( K s S 0 ) K s ) m (1 )
K s S0
2
or, K s S0 Ks
Pmax,cell ,cont Y X / S m S 0
S0 S0
Assuming that Ks << S0, Pmax,cell ,cont Y X / S m S 0

Defining G as, Pcont YX / S m S 0

G
Pbatch YX / S S 0
1 Xm
ln tl
m X i
6Chapter 7
 Pcont Xm Pcont X m 0.693tl
ln mtl ln
Pbatch Xi Pbatch Xi td

Example: suppose
td = 0.5 hr, tl = 10 hr
and Xi/Xm = 0.05
G = 17
Continuous cultivation
is more efficient that
batch cultivation.
G is affected markedly
by td and tl but is less
sensitive to the value of
Xi/Xm..

7Chapter 7
Examples of Continuous Cultivation
Yeast
YX/S is not constant over
a narrow range of D.

Variation in YX/S is due to a specific physiological state at the low

dilution rates where the substrate concentration is low, and due to
the difficulties in maintaining a true steady state near wash out
at high dilution rates.

8Chapter 7
Bacteria
This example was
subject to conditions
where NH3 was growth
limiting substrate, S
Both values of X and S
were measured only
after at least 2 to 3
days at steady state at
each value of D.
Typical trends for X and S with constant yield YX/S except in
the particular range of D near wash out

9Chapter 7
Fungi
In these cases, transportation
of mycelia from vessel to
vessel is an additional
problem not encountered
when other microbes such as
yeast and bacteria are
continuously cultivated.
Two levels of sucrose
concentrations were studied

Steady states at each dilution rate were maintained for 1 to 3 days

before values of S and X were measured
YX/S was not constant and maximum out occurred near the was out.

10Chapter 7
Chemostat with Cell Recycle
- To keep the cell concentration higher than the normal steady-state
level in a chemostat.
- To increase the cell and growth-associated product yield.
- For low-product-value processes: e.g. waste treatment.
fuel ethanol

11Chapter 7
 Total balance : Fa (1 ) F and F Fe Fex
Mass balance for cell concentration in the reactor with complete
mixing and with cell recycle:
Mass concentration in reactor = cells in recycle stream cells out of reactor + cell growth in reactor
dX dX
V X xF Fa X V
dt dt growth
(It is assumed that cell maintenance and death are small and neglible)
where: Xx cell mass concentration in recycled suspension
- recycled ratio
F flow rate of recycled cell suspension
At steady state (dX/dt = 0):
1 dX Fa F X x

X dt growth V V X

12Chapter 7
 Combining above equations:
X x D 1 1 X x
1 D D
X X
A chemostat can be operated at dilution rates higher than the
specific growth rate when cell recycle is used.
Since above equation indicate that D in system with
recycle and since that
[1+(1-Xx/X)] < 1 because Xx/X > 1
D In system with recycle

Mass balance on growth limiting substrate with cell recycle:

S in feed stream + S in recycle stream S out S consumed = S accumulated
F F (1 ) F X dS
S0 S S
V V V YX / S dt

13Chapter 7
 dS F
At 0 and D
dt V
D YX / S S 0 S
X YX / S S 0 S or: X
Xx
1 1 X

Steady state cell concentration with recycle is greater by a

factor 1
Xx
1 1
X

In order to have relationship between substrate concentration

and specific growth rate, or cell concentration and specific
growth rate the Monod equation can be used, using the
expression for derived here:

14Chapter 7
 X
D 1 1 x
X
S KS KS
m X
m D 1 1 x
X
Substituting S in the expression for X in the previous slide:
X
D 1 1 x
YX / S X
X S0 K S
Xx Xx
1 1 X m D 1 1
X

15Chapter 7
m = 1.00h-1, S0 =2.0g/l, Ks = 0.01 g/l, Yx/s = 0.5 g/g,
concentration factor Xx/X = 2.0 and recycle ratio = 0.5

No recycle

16Chapter 7
 Cell mass balance around the cell separator.

(1 ) FX Fe X e Fex X x FX x

to get Xe in supernatant
often Fex = 0

The average residence time in cell separator

Vcell separator

(1 ) F

17Chapter 7
Example: Chemostat with Cell Recycle
Organisms are cultured in a chemostat with cell recycle. The
system is operated under glucose limitation.
F 100 ml/h, V 1000ml, S0 10 g glucose/L
YX/S
M
0.5gdw cells/g substrate; m 0.2h 1 ,
Ks 1g/L, X x /X 1.5, 0.7, X 0 0, K d 0

Determine specific growth rate net, S in the reactor effluent, cell

concentration in the recycle stream (Xx) and in the concentrator
effluent (Xe)
If the concentrator has a volume of 300 mL, what is the residence
time in it?

18Chapter 7
Fed-Batch
Nutrients are continuously or semi-continuously fed, while effluent
is removed discontinuously.

Overcome substrate inhibition or

catabolite repression by
intermittent feeding of substrate by
maintaining low substrate
concentration.

For production of secondary

metabolites e.g. antibiotics, lactic
acid, E. Coli making proteins from
recombinant DNA technology.

19Chapter 7
 Analysis of fed-batch with substrate continuously fed
and no output: t = 0, V0 = 0, F is constant.
dV
Volume: F V Vo Ft
dt
At quasi steady state, S added = S consumed,
X, S, P concentrations are constant.
d ( XV )
Cell mass balance: FX O V net X
dt
d ( XV ) dX dV
since V X , then
dt dt dt
dX dV
FX O V net X V X
dt dt
dV 1 dV F
V net X X net D
dt V dt V
F F Do
net
V V0 Ft 1 D0 t
m S KsD
net D if K d 0 Then S
Ks S m D

20Chapter 7
 M
where S 0
assuming X Xm
M

where Xt = Xt0 at t=0

21Chapter 7
 (g)

qp (i.e. g product/g cells-min)

at Pt = Pt0, t=0

22Chapter 7
23Chapter 7
Example: Fed- Batch
In a fed-batch culture operating with intermittent addition of
glucose, the value of V is given at time t=2hr, when the
system is at quasi-steady state.

dV
F 200 ml/h, V 1000ml, S0 100 g glucose/L
dt
M
YX/S 0.5gdw cells/g glucose; m 0.3h 1 ,
Ks 0.1g/L, X t 0 30 g cells
Determine V0.
At t = 2 hr, find S, X and Xt and P at quasi-steady state if qp= 0.2 g
product/g cells-h, P0=0.

24Chapter 7
Multi-stage Chemostat System
In some fermentations, the growth and product-formation steps
need to be separated.
e.g. secondary metabolite, culture of genetically engineered
cells.

P2
P10
Growth stage Product formation stage
At steady state, Vn, Xn,Sn,,Pn in the reactor of each stage dont change with time.

25Chapter 7
X0 = 0, Vi, i=1,2.n constant.
Stage 1: cell growth condition, kd = 0, qp = 0
dX 1
Cell mass: FX 0 FX 1 1 X 1V1 V1
dt
At steady state 1 D1 (1 X 0 )
X1
1 X 1 dS1
Limiting substrate: FS 0 FS1 V1 M V1
YX / S dt
At steady state S S 1 X 1
1 2
D1Y XM/ S

1 , 2 ... and n are net specific growth rates in Stage 1, 2,..., and n, recspectively.
When Kd 0, they are equal to the respective specific gross growth rates in each stage.

26Chapter 7
Stage 2: product formation conditions, kd = 0, F = 0
dX 2
Cell mass: FX 1 FX 2 2 X 2V2 V2
dt
At steady state 2 D2 (1 1 ) where D2 F
X
V2 is constant.
X2 V2
2 X 2 qp X 2 dS 2
Limiting substrate: FS1 FS 2 V2 M
V V2
YX / S YP / S dt
2 X 2 qp X 2
At steady state S 2 S1
D2YXM/ S D2YP / S
dP2
Product: FP1 FP2 V2 q p X 2 V2
dt
qp X 2
At steady state P2 P1
D2

27Chapter 7
Stage n: product formation conditions, kd = 0, F = 0
Similarly, equations could be obtained for nth stage.
X n 1
n Dn (1 )
Xn

n X n qp X n
S n S n 1 M

DnYX / S DnYP / S
qp X n
Pn Pn 1
Dn
If n (e.g. Monod model) and qp are known functions, Xn , Pn, and Sn
at nth stage could be determined by the above equations.

28Chapter 7
To determine the parameters (X, P) graphically when growth kinetics
cannot be expressed analytically.

With no additional streams added to the second or subsequent units, mass

balance around the nth stage on cell, substrate and product yields
dX n
FX n 1 FX n n X nVn Vn
dt
dX n n X n Dn ( X n 1 X n )
at steady state 0,
dt
Since rx ,n ( X x , S n ) n X n , rx ,n ( X x , S n ) Dn ( X n X n 1)
rx,n: cell growth rate at nth stage (g/l-h);
F: the volumetric flowrate (l/h);
Dn = F/Vn: the dilution rate at the nth stage (1/h);
Vn: the liquid volume of the nth chemostat (l)
29Chapter 7
 dPn
FPn 1 FPn rp ,nVn Vn
dt
dS n
FS n 1 FS n rs ,nVn Vn
dt
Similarly at steady state
rp ,n ( X x , S n ) Dn ( Pn P n 1 )
1
rs ,n ( X x , S n ) rx ,n ( X x , S n ) Dn ( S n S n 1)
YX / S

rp,n: product formation at nth stage (g/l-h);

rs,n: substrate consumption at nth stage (g/l-h);

30Chapter 7
To determine the parameters (X, P) graphically.
For an example, in a two-stage chemostat system, the first tank is
600 L and the second one is 900 L. The flowrate is 100L/h,
determine the concentrations of cells and product in each of the tank.
Solution: Get kinetics data (X~t, P~t) from the batch culture.

originating from (Pn-1, 0).

31Chapter 7
 To determine the parameters (X, P) graphically in multi-stage
chemostat system
Get the kinetics data of X~t and P~t from batch culture and plot

Get dX/dt~X from the kinetics data and plot

Determine X1 and P1 in the first stage:

On the graph of dX/dt~X, draw a line of y = D1 (X1 -X0)

D1 = F/V1, X0 = 0 in the feed, X1 is determined from the intersection.

On the graph of X,P~t, P1 is determined by the X1 at the respect time

Determine X2 and P2 in the second stage:

Plot the graph of dP/dt~P from the kinetics data of P,

draw a line of y = D2(P2-P1), D2 = F/V2 , P2 is determined from the intersection.

On the graph of X,P~t, X2 is determined by the P2 at the respect time. 32Chapter 7

dX/dt (g/L-h)

33Chapter 7
X (g/L)
Get data of dX/dt~X from the kinetics data in the batch culture and
plot.
rx ,n ( X x , S n )
Dn ( X n X n 1)

dX/dt (g/L-h)
7 X 1
y D1( X1 X 0) .1 6
= 0
F Y
( X1 X 0)
V1
100 l/h
( X1 0)
600 l X (g/L) x1 is 7.1g/l
0.167 X1
- Determine the x1 in the first stage of chemostat.
On the graph of dX/dt~X, draw a line of y D1( X1 X 0)
originating from (X 0 , 0). x1 can be found from the intersection.
34Chapter 7
- From the graph of kinetics data get the P1 in respect with the
X1.

X1 = 7.1 g/l

P1 = 0.09 g/l

35Chapter 7
- Determine the X2 and P2 For the second stage.
- Get dP/dt ~ P from the kinetics data in the batch culture and plot.
rp ,n ( X x , S n ...) Dn ( Pn P n 1 )
y D (P P )

dP/dt (g/L-h)
2 2 1
F 0 9)
( P2 P1 ) 11(P 2-0 .
V2 =0 . 1
y
100l / h
( P2 0.09 g / l )
900l P1=0.09 g/l P (g/L)
P2=0.53 g/l
0.111( P2 0.09 g / l )
- on the graph of dP/dt ~ P, draw a line of y D2 ( P2 P1)
P2 can be determined from the intersection.

36Chapter 7
- From the graph of kinetics data get the X2 in respect with the P2

X2 = 7.95 g/l

P2=0.53 g/l
X (g/L)

X1= 7.1 g/l

P1 = 0.09 g/l

Time (hr)

- Similarly, Xn and Pn can be determined for the nth stage of reactor.

37Chapter 7
In summary, the first tank is 600 l and the second one is 900 l.
The flowrate is 100l/h, X1=7.1 g/l, P1=0.09 g/l; X2=7.95 g/l, P2=0.53 g/l
If one stage chemostat is used with the same total working
volume i.e. 1500 l as that in the above example.
The flowrate is also 100l/h. What is the achieved concentrations
of the cells and the product? D = F/V=100/1500=0.067 (h-1)
dX/dt (g/L-h)

X (g/L)

P (g/L)
X=7.5g/l

7X
y=0.06 P=0.14g/l

X=7.5g/l
X (g/L) Time (hr)

Product concentration can be increased by using multistage chemostats

instead of single stage with the same total reactor working volume. Ex.9.2

38Chapter 7
Problems for Practice
Textbook (M.Shular)
Q: 9.3:
You are not required to determine the configuration of tanks that would
maximize product formation. Instead, determine the concentration of
the product in each tank for the following configurations using the
graphical method introduced in the lecture:
Reactor 1 (600 L), reactor 2 (900L) and reactor 3 (300 L).
Reactor 1 (900 L), reactor 2 (300L) and reactor 3 (600 L).
Which of the above configurations gives higher product concentration?
If 300 L reactor is the first reactor in series, what would happen?

Q: 9.4
Ex.9.1-9.2

39Chapter 7
Immobilized Cell System
Immobilized cells
Provide high cell concentration
Reuse cell
Eliminate washout problem at high dilution rate and cell
recovery
May provide favorable microenvironmental conditions
May prove genetic stability
Protect against shear damage
Can perform multi-step biosynthesis reactions that are not
practical purified immobilized enzyme preparation.
Disadvantages
Diffusional limitation are important.
Growth and gas evalution may lead to significant mechanical
disruption of the immobilizing matrix.
40Chapter 7
Immobilized Cell System
Immobilized methods:
- Active immobilization of cells
Entrapment and Binding
similar to immobilized enzyme except covalently binding
(toxic agents, e.g. glutaraldehyde)

- Passive immobilization of cells: biological film.

Multi-layer growth of cells on solid surface.
e.g. common in biological waste water treatment and mold
fermentation
thick biofilm: high concentration but diffusion problem
Bioreactor consideration:
- hydrodynamic shear: packed-column, fluidized-bed or airlift reactors.

41Chapter 7
Diffusional Limitations in Immobilized Cell Systems
Immobilization of cells may cause extra diffusional limitations as
compared to suspension cultures.
The presence and significance of diffusional limitations depend on
the relative rates of bioconversion and diffusion, i.e. Da
maximum rate of bioconversion rmax
Da
maximum rate of diffusion De / S0
De: effective diffusivity of the rate limiting substrate = kL
: thickness of diffusion path (or liquid film) according to
So: bulk substrate concentration in liquid phase film-theory
Da >> 1 rate of bioconversion is diffusion limited
Da << 1 rate is limited by the rate of bioconversion
Da = 1 diffusion and bioconversion rate are comparable

42Chapter 7
Diffusional limitations:
1. External: between fluid and support surface in adsorption and
covalent bonding.
2. Intraparticle: i.e. inside particles in entrapment, encapsulation
or immobilization in porous particles
3. Both
Biofilm Thickness
The thickness of a biofilm or the size of microbial
floc increases with time during the phase growth
A microbial floc is an aggregation of many cells
and in some processes can be more than 1 mm in
diameter

43Chapter 7
Quasi-steady state assumption: since the rate of increase in biofilm
thickness is much slower than the rate of substrate uptake, the
system can be assumed to be quesi-steady state for relatively short
period and that all the cells inside the biofilm are in the same
physiological state
For such case, it is possible to write a steady-state substrate balance
within the biofilm by using average kinetic constants for the biotic
phase (living cells)
A differential mathematical balance at steady-state for the rate-
limiting substrate within the biofilm yields

d 2S 1 m S S S oi at y 0
De X BCs:
dy 2 YX / S K S S dS
0 at y L
dy

44Chapter 7
Another assumption: liquid nutrient phase is vigorously agitated and
liquid film resistance is negligible
S o S oi
m X
Define maximum rate of substrate utilization: rm
YX / S
Thus we can write De d S rm S
2

dy 2 KS S
d2S 2S
In dimensionless form Solved numerically
dy
2
1 S
where S S , y
y ,
So m X rm
, L L
So L KS YX / S De K S De K S
Maximum rate of substrate flux in the absence of diffusion
limitations: rm S 0
dS LAS
N S AS AS De K S S0
dy y 0

45Chapter 7
In the presence of diffusion limitation, the rate of substrate
consumption or flux is expressed in terms of effectiveness factor:
dS rm S 0
N S De L
dy y 0 K S S0
where:
rate of substrate consumption in presence of diffusion limitation

rate of substrate consumption in absence of diffusion limitation
= 1; in the absence of diffusion limitation
< 1; in the presence of diffusion limitation

f ( , )
< 1 to eliminate diffusion limitations
As the biofilm grows (slowly), gradually

46Chapter 7
The effectiveness factor () can also be calculated as:

tanh
1 1 , for 1
tanh
1 tanh 1
1 , for 1
tanh
where is the modified Thiele modulus and is given by:
1 / 2
S0 / K S S0 S0
K ln 1 K
S0 S S
2 1
KS

47Chapter 7
 Some cells such as molds (A. niger) form pellets in a fermentation
broth, and substrate need to diffuse inside the pellets to be
available for microbial consumption.
Cells may form biofilm or spherical support particles as shown.
The dimensionless substrate transport
equation within microbial floc is:
d2S 2 dS 2S

dr
2
r d r 1 S / ' (a) Microbial film on inert
spherical support particle.
where S r So Spherical microbial floc
S , r , '
(b)
,
So R KS
and
S 1 at r 1
m X rm
R R BCs: dS
YX / S De K S De K S 0 at r 0
dr

48Chapter 7
 For nonspherical particles, a characteristic length is defined as:
VP
L
AP
where VP and AP are the volume and surface area of microbial pellets
The rate of substrate consumption by a single microbial floc is:
dS rm S 0
N S AP AP De VP
dr r R K S S0

f ( , )
In general, values for spherical
geometry are slightly lower than those
of rectangular for intermediate values of
(1 < < 10).

49Chapter 7
Approximations to first- and zero-order reaction kinetics
The reaction rate can be approximated to first order at low
substrate concentrations:
m S rm
rS X S where rm = (m/YX/S)X
YX / S K S KS
For such case, the effectiveness factor is given by:
1 1 1 VP rm / K S
where
tanh 3 3 AP De
The rate of bioreaction can be approximated to zero order at values
of S >> KS:
m X
rS rm
YX / S
In this case, the solution of the differential equation is

50Chapter 7
 rm
S S0 R2 r 2
6 De
According to this equation, substrate concentration may be zero at a
certain distance from the centre of the floc. Such distance is called
the critical radius (rcr) and is determined by setting S = 0 at rcr:
2
rcr 6 De
1
R rm R 2

When rcr > 0, i.e. R > (6DeS0/rm)1/2, then the concentration of the
limiting substrate is zero for 0 < r < rcr. In this case, the limiting
substrate is consumed only in the outer shell of the floc, and the
effectiveness factor is given by
3/ 2
rm (4 / 3) R r 3 3 rcr
3
1 1
6 De

cr
1 or
(4 / 3)R 3 rm rm R
2
R

51Chapter 7
Bioreactor Considerations in Immobilized Cell Systems

52Chapter 7