MICROBIAL GROWTH AND

PRODUCT FORMATION
KINETICS
Subject :- Bioprocess Principle
1 Model classification- Model Formulation-
2 Unstructured Models
3 Phases of batch growth cycle
4 Monod Models
5 Multiple substrate models and model Inhibition,
6 Models of growth and non-growth product inhibition
7 Models for the growth of fungi, Plant cell and Animal cells,
8 Structured models
9 Models of metabolites and growth-compartmental Models
10 Models of product formation.

Biochemical Engineering Fundamentals

By Bailey and Ollis
Page no 383 - 438

2nd Edition, Biochemical Engineering

By Douglas S. Clark,
Harvey W. Blanch
Mathematical Modelling
Modelling is an iterative process
Why do Mathematical Modelling?
• Mathematical modeling has always been an important activity in
science and engineering.
• The formulation of qualitative questions about an observed
phenomenon as mathematical problems was the motivation for and an
integral part of the development of mathematics from the very
beginning.
• Computational power is particularly important in modeling chemical
engineering systems, as the physical and chemical laws governing these
processes are complex.
• Besides heat, mass, and momentum transfer, these processes may also
include chemical reactions, reaction heat, adsorption, desorption, phase
transition, multiphase flow, etc.
• This makes modeling challenging but also necessary to understand
complex interactions.
 Why do Mathematical Modelling?
• In a chemical engineering context, mathematical modeling is a prerequisite
for:
– Design and scale-up;
– Process control;
– Optimization;
– Mechanistic understanding;
– Evaluation/planning of experiments;
– Trouble shooting and diagnostics;
– Determining quantities that cannot be measured directly;
– Simulation instead of costly experiments in the development lab;
– Feasibility studies to determine potential before building prototype equipment or
devices.
 Modelling Procedure

• Each step in the modeling

process requires an
understanding of a variety
of concepts and
techniques blended with a
combination of critical
and creative thinking,
intuition and foresight,
and decision making.

Different steps in the

Model Development
• Step 1: Problem definition:
– This involves stating clear goals for the modeling, including the various elements that pertain to the
problem and its solution. (what hypothesis and which assumptions to use)
• Step 2: Formulation of conceptual model:
– Collect data and experience about the subject to be modeled.
– The main challenges are in identifying the underlying mechanisms and governing physical/chemical
principles of the problem.
• Step 3: Formulation of mathematical model:
– Each important quantity is represented by a suitable mathematical entity, e.g. a variable, a function, a
graph, etc.
– What are the variables (dependent, independent, parameters)?
– What are the constraints?
– What boundary conditions?
– What initial conditions?
• Step 4: Solution of the mathematical problem:
– Check the validity of individual mathematical relationships, and whether the relationships are mutually
consistent.
– Verify the mathematical solution, i.e. ensure that you have solved the equations correctly. This step
involves checking your solution against previously known results (analytical/numerical), simplified
limiting cases, etc.
• Step 5: Estimation of parameters:
– The parameters of the system must be evaluated and the appropriate values must be used in the model. Some
parameters can be obtained independently of the mathematical model. (e.g. gravitation constant, solubility data)
– However, it is usually not possible to evaluate all the parameters from specific experiments, and many of them
have to be estimated by taking results from the whole (or a similar system), and then using parameter-fitting
techniques to determine which set of parameter values makes the model best fit the experimental results.
– For example, a complex reaction may involve ten or more kinetic constants.
– These constants can be estimated by fitting a model to results from a laboratory reactor.

• Step 6: Evaluation/validation :
– A key step in mathematical modeling is experimental validation.
– Ideally the validation should be made using independent experimental results, i.e. not the same set as used for
parameter estimation.
– During the validation procedure it may happen that the model still has some deficiencies.
– In that case, we have to “iterate” the model and eventually modify it.
– Certain characteristics that models which have a bearing on the question of how good they are: Accuracy;
Descriptive realism; Precision; Robustness; Generality; Fruitfulness.

• Step 7: Interpretation/application:
– The validated model is then ready to be used for one or several purposes as described earlier, e.g. to enhance our
understanding, make predictions, and give information about how to control the process.
 Mathematical Model:

Use of mathematical model

Research and
Design Plant Operation
Development
• Determining chemical • Exploring the sizing • Troubleshooting
kinetics mechanisms and arrangement of control and processing
and parameters from processing equipment's problems
pilot-scale or lab-scale • Simulating start-up, • Studying the effect of
• Aiding in scale-up shut down, and and requirements for
condition emergency situations expansion projects
and procedures

Principle of formulation:
Mathematical
Solution of the Verificat
Basis Assumption consistency of
model equations ion
model
 Classification of Mathematical
Models
• Mathematical models are classified by
– Grouping into opposite pairs;
– Mathematical complexity;
– Degree of resolution.
Grouping of models into opposite
pairs
• Linear versus nonlinear;
• Steady state versus non-steady state;
• Lumped parameter versus distributed parameter;
• Continuous versus discrete variables;
• Deterministic versus stochastic;
• Interpolation versus extrapolation;
• Mechanistic versus empirical;
• Coupled versus not coupled.
Classification of modelling
techniques: Lumped parameters
model
Distributed parameter
model
Partial differentiation
model
Steady state model
Modelling
techniques
Unsteady state model
Population balance
model
Discrete event model
Stochastic
model/Probabilistic
Fuzzy model (Monad
model)
Linear versus nonlinear Steady state versus non-steady state
Interpolation versus extrapolation

Danger of extrapolation. Yield of a

chemical reactor versus time
 Empirical model Mechanistic model

C B
S A
S
Empirical model of microbial growth in sludge Mechanistic model of microbial growth in sludge

f(x)= s f(x)= A+B+C

Where ‘s’ is over all Where f(x) is the sum of individual effects
Example : Growth of micro
effect organism
of A,B and C in sludge

Growth of micro organism is given Growth of micro organism is given by

by f(x)= s f(x)= Nitrogen + Carbon+ Nutrients

Where f(x) is the sum of individual effects

Where ‘s’ is over all nutrient effect of nitrogen, carbon and nutrients
Mechanistic model:

Understand underlying concept Explain individual components

Characteristics

1
Testable and verifiable Precision in data may be
compromised
Mechanistic
model
Accurate representation of a
project

Time & cost ineffective

Understanding a problem Merits Demerits

Analyse weakness of individual

point

Help in redesigning components

Figure 4 : Overview of mechanistic model
Classification based on Mathematical complexity
 Classification of models according to scale

Macroscopic (Entire System) Microscopic (Elemental)

Bioprocess modeling
In order to improve process understanding or performance, different
automatic tools can be developed simulators, which are able to
reproduce system behaviors, software sensors which allow to obtain
an estimation of an unmeasured signal or controllers to maintain
optimal conditions. All these tools relay on representation of
biosystem, and a mathematical model.

21
Characteristics of bioprocess models
The are two new characteristics for bioprocess models. First, a
model can be structured or unstructured depending on whether it
describes intracellular characteristics of the cell (metabolic
reactions, cellular processes etc.) or (considers the cell like an
entity without internal structure).

The next class of models are segregated or unsegregated depending

on whether it considers the heterogeneity of the cellular population,
are the position in the cell cycle..

22
Such reaction rates vary with time and are usually influenced by many
physicochemical and biological environmental factors like substrate,
biomass and product concentrations as well as pH, temperature,
dissolved oxygen concentration or various microbial growth inhibitors.
 Malthus law
• The simplest cell growth model is the Malthusian model commonly known
as exponential law.
• overall rate of biomass formation is proportional to the biomass itself.
For a batch system,
dx
 x
dt

• where x is the mass of cells per unit volume, µ is the proportionality

constant known as 'specific growth rate' (h-1) and t is the time (h).

• Here as the time increases, cell mass will also increase.

• X=X0 eµt

• The demerit of this model is that it ignores the substrate needed for its
growth and that the resources are constrained.
 Quantifying Growth Kinetics
• Structured vs Unstructured -
– S - model divides the cell mass into components
– U - assumes fixed cell composition -
exponential growth phase of batch or in
continuous
• Segregated vs Nonsegregated
– S - assumes different types of cells exist
– N - all cells are the same type
 Unstructured, Nonsegregated Models
(Monod)
mS
• Assumptions

– One limiting substrate
– Semi empirical relationship K S
S
– Single enzyme system with
Michaelis-Menten kinetics is  m - maximum growth rate when S >>
responsible for the uptake of
substrate Ks
– Amount of enzyme is sufficiently • Ks - saturation constant - concentration
low to be growth limiting of the rate-limiting substrate when the
– Low population density specific rate of growth is equal to one
half of the maximum.
• Most commonly used • Ks = S when  = 1/2m
expression for growth • In general  = m for S >> Ks and
for S << Ks
 Cell Growth - Monod
Does not account
for death phase
0.7
Cell Concentration (g/L)

0.6

0.5

0.4

0.3
Does not account
0.2 for lag phase
0.1

0
0 5 10 15 20 25
Time (hr)
 Unstructured, Nonsegregated Models

• Assuming one limiting substrate

– Blackman
– Tessier
– Moser
– Contois
• Multiple Substrates are growth limiting
– Usually do not use unstructured, nonsegregated
model
Other unstructured, nonsegregated models (assuming one
limiting substrate):

Blackman equation: μ = μm if CS ≥ 2KS

μm CS
μ = if CS < 2KS
2 KS

Tessier equation: μ = μm [1 - exp(-KCS)]

μm CSn
Moser equation: μ =
KS + C Sn

μm CS
Contois equation: μ =
KSX CX + CS
Blackman equation:
μ = μm if CS ≥ 2 KS This often fits the data better than the
Monod model, but the discontinuity can be a
μm CS problem.
μ= if CS < 2 KS
2 KS
1
0.8
0.6
μ (per h)

0.4 μm = 0.9 per h

Ks = 0.7 g/L
0.2
0
0 2 4 6 8 10
M.PANDIMADEVI Cs (g/L) 30
Tessier equation:
μ = μm [1 - exp(-KCS)]

0.8

0.6

0.4
μ (per h)

0.2 μm = 0.9 per h

K = 0.7 g/L
0
0 2 4 6 8 10
Cs (g/L)
Moser equation:
μm CSn When n = 1, Moser equation describes Monod model.
μ =
KS + C Sn
1

0.8

0.6
Monod
μ (per h)

0.4
n = 0.25
0.2 n = 0.5
μm = 0.9 per h n = 0.75
0 Ks = 0.7 g/L
0 2 4 6 8 10
Cs (g/L)
Contois equation:
μmax CS Saturation constant (KSX CX ) is proportional to cell
μ = concentration
KSX CX + CS

Verlhurst model

Verlhurst kinetic model depends on the cell concentration. This model has two
kinetic constants of maximum specific growth rate (μmax) and maximum cell
concentration (xmax). Verlhurst model is expressed by the following equation
µ=µmax-(µmax /Xmax)*X

Andrew model
Andrew model is proposed by the following equation for the growth dependent
which incorporate substrate inhibition.

µ=µmax/[(1-ks/s)(1+s/Kis)
Extended Monod model:
μm (CS – CS,min) Extended Monod model includes a CS,min term,
μ= which denotes the minimal substrate
KS + CS – CS,min concentration needed for cell growth.

1
0.8
0.6
μ (per h)

0.4
0.2 μm = 0.9 per h
Ks = 0.7 g/L
0 CS,min = 0.5 g/L
0 2 4 6 8 10
Cs (g/L)
Monod model modified for substrate inhibition:
Monod model does not model substrate inhibition.
Substrate inhibition means increasing substrate concentration beyond certain value
reduces the cell growth rate.

1
0.8
μ (per h)

0.6
0.4
0.2
0
0 2 4 6 8 10
M.PANDIMADEVI Cs (g/L) 35
Monod model modified for cell growth with
noncompetitive substrate inhibition:
μm
μ=
(1 + KS/CS)(1 + CS/KI )

μm CS
=
KS + CS + CS2/KI + KS CS/KI

μm CS
If KI >> KS then μ=
KS + CS + CS2/KI

where KI is the substrate inhibition constant.

Monod model modified for cell growth with
noncompetitive substrate inhibition:
μm
μ=
(1 + KS/CS)(1 + CS/KI )

μm CS
=
KS + CS + CS2/KI + KS CS/KI

μm CS
If KI >> KS then μ=
KS + CS + CS2/KI

where KI is the substrate inhibition constant.

Monod model modified for cell growth with
competitive substrate inhibition:
μm CS
μ=
KS(1 + CS/KI) + CS

where KI is the substrate inhibition constant.

Monod model for two limiting substrates

CS1 CS2
μ= μm
KS1 + CS1 KS2 + CS2
Monod model modified for cell growth
with product inhibition:
Monod model does not model product inhibition (where increasing product concentration
beyond certain value reduces the cell growth rate)

For competitive product inhibition:

μm CS
μ=
KS(1 + Cp/Kp) + CS

For non-competitive product inhibition:

μm
μ=
(1 + KS/CS)(1 + Cp/Kp )

where Cp is the product concentration and Kp is a product inhibition constant.

Monod model modified for cell growth with
product inhibition:
Ethanol fermentation from glucose by yeasts is an example of non-competitive product
inhibition. Ethanol is an inhibitor at concentrations above nearly 5% (v/v). Rate expressions
specifically for ethanol inhibition are the following:

μm CS
μ= (1 + Cp/Cpm)
(KS + CS)

μm CS
μ= exp(-Cp/Kp)
(KS + CS)

where Cpm is the product concentration at which growth stops.

Monod model modified for cell growth with
toxic compound inhibition:
For competitive toxic compound inhibition:

μm CS
μ=
KS(1 + CI/KI) + CS

For non-competitive toxic compound inhibition:

μm
μ=
(1 + KS/CS)(1 + CI/KI )

where CI is the product concentration and KI is a constant to be determined.

Monod model extended to include cell death
kinetics:

μm CS
μ= - kd
KS + C S

where kd is the specific death rate (per time).

 Unstructured Batch Growth
•
Models
A brief discussion on modelling of bacterial growth has been
discussed in the beginning of the course using the simplest
approach of the form,
dx
 f (x )
dt
• Where from malthus law equation f(x)=µx
• µ=constant , describing the microbial growth rate
• Verlhurst in 1844 and pearl and reed 1920 put forwarded the
theory which included inhibiting factor to population growth
assuming that inhibition is proportional to square of cellular
biomass
dx
dt
 kx (1   x ) x (o )  x o
The above given Riccati equation was integrated to obtaine logistic Curve

x0
kt
e
x 
1  x0 (1  e kt
)
Were stationary population of size 1
x o


dc t

dx
dt dt
Were production rate of toxin a toxin is proportional to the population growth
So,
c t
( 0)  0

c t
  (x  x 0
)

Where x0 is negligible relative to x, substituting in above

equation
Another class of stationary approach a stationary population level can be
formulated based on limiting nutrient exhaustion. Assuming a constant
overall yield factor yx s
and
   (s )
Such as monod kinetics, nutrient and cell material balance can be combined
into a single equation of the form . A drawback of the logistic equation is its
failure to predict a phase o decline after the stationary population has
exhausted all available resources.
This feature is found in one model developed by Volterra
t

 k (t , r ) x(r )dr
o

Physically we may interpret such a term as a crude recognition that the

history k(t,r) of a population influence its growth rate.

dc
 k (t ) x (t )
dt C(0)=0
Specifically, let us suppose that k is a constant equal to k0 and that the
history.
Then the population size is described by t
dx
 kx(1   x )  k 0  x(t )dr
dt 0

x (0)  x 0

K0 the sign is taken as negative for an inhibitor and positive for a compound
which promotes growth.
K0 is negative the population size decline after passing through the maximum.
 Product Formation Kinetics
Unstructured model

Simplest type of product formation kinetics arise when there is a simple

stoichiometric connection between product formation and substrate utilization
or cell growth. The rate of product formation can be written as
r f p  YP S r f a
It can also be expressed as
r f p  YP X rf x

For anaerobic fermentation such as Lactobacillus delbrukii, the kinetics of

product formation is famously expressed by the Leudeking-Piret kinetics given by

rf p   fi  x

Such a form is normally used in establishing product formation data. The first
term in the above equation refers to the energy used for growth and the
second term refers to the energy used for maintenance.
Consider a batch fermentation model given by a logistic model below
dx
 kx (1  x)
dt
and product formation is given by Leudeking-Piret kinetics of the form
dp dx
  x
dt dt
And substrate utilization is given by
ds 1 dx 1 dp
   kx
dt Yx s dt Y p s dt

For substrate consumption,

Substitute dp/dt into ds/dt gives

ds 1  dx 
 (  ) (  k)x
dt YX S YP S dt Y p s

This equation reduce into

ds dx
  x
dt dt
 Were as 1  
  (  ) and   (  k)
YX S YP S Y p s

Integrating dx/dt gives a Riccati type equation of the form

x0
kt
e
x
1  x0 (1  e kt
)
By rearrange this equation
 x (t ) 
 
ln 
xs   kt  ln 

xs
 1


 x (t )  x 
1   0 
 xs 
 

Where β is substituted for stationary cell, xs

Plot a graph using above equation to find K as slope , and –ln(Xs/X0-1)

as a interscept in y- axis
 Structured kinetics Model
Unstructured model of the above type describe only the quantity of the biological phase.
Such models do not recognize nor represent the composition or what we might call the
quality of the biophase. In situation in which the cell population composition changes
significantly and in which these composition changes influence kinetics, structured
model should be used. Since writing the material balance equation for every component
of cell is not possible

The biophase variable employed in structured model are typically the mass x j, molar
concentration cj per unit volume of biophase, for a well mixed reactor the material
balance on component j gives
d 1  1 1
 v xc j  v xr  xc j
dt 
 c
r
  c
R fj
 c

Where ρc= mass density of cell (cell mass/ unit volume cells)
rfi = molar rate of formation of component J
Φx= mas of the cell added to the recator by flow. Time -1
VR= culture volume
X= cell mass concentration (cell mass/unit volume culture)
Cj= moles J/Unit volume cells
It is assumed that any cells added or removed from the reactor have the
Same state as the cell population in the reactor. If this condition is not met,
The balance equation must be modified.

Assuming the density of the cells ρc and volume of culture VR are time invariant may
be rewritten in the following form after carrying out the indicated differentiation

dc j  1 dx  x
 rf j  c j  
 j c
dt  x dt  xVR

For a batch reactor, Φx is zero and the quantity in parenthesis on the

right hand is simply the specific growth rate µ, the equ

dc j
 rf j  c j
dt

The Physical interpretation of r f j , the –µc j term represent reduction in

concentration caused by dilution from growth of the cell population.
 For product formation , using the following eq
dp dx
  x
dt dt

With riccate equation and upon integration gives

xs x
p (t )  p0   ( )(1  0 (1  e kt ))   ( x (t )  x0 )
k xs

The constant β is found at the stationary phase of batch culture

dp
stationary
  dt
xs

And α is determined by a plot of,

x x
p(t )  p0   ( s )(1  0 (1  e kt )aga int( x(t )  x0 )
k xs
The aerobic bioprocess modeling is an useful tool to accomplish several important tasks

(a)it can be the basis for adequate optimization and control

technique applications;
(b)it can provide the necessary information about the features
of the chosen bioprocessing system;
(c)it synthesizes the characteristics of the specified living
cells’ evolution and hence, it is the best technique to
predict the process efficiency.
• The unstructured global models are in use nowadays as the main tool for
both the bioprocess modeling, but also for being applied in overall computer
control.
• Their limit is they are a simplified representation of the bioprocess behavior:
conforming to this concept the bioprocess evolution depends directly and only
on the macroscopic variables representing the working conditions in the
bioreactor.
• Therefore the unstructured models are essentially kinetic equations that
describe the variation of substrate or product concentrations and of a unique
biological state variable-the cell concentration, and can also express the
influences of some important process variables (pH, pO2, temperature, and
others), and only sometimes they are balance equations.
• the structured models - the biotic phase is not any more
viewed as a homogenous component, but they provide
information about the physiological state of the cells, their
composition and regulatory adaptation to the environment.
• Conforming to this concept the cell mass is structured in
several intracellular compounds and functional groups,
which are connected to each other and to the environment
by fluxes of material and information.
• The structured models can be:
• multi compartment models,
• genetically structured models, and
• biochemical structured models.
 Structured product formation kinetic modeling
• Only little development for the structured kinetics of product formation

• Models are usually developed based on metabolite production

• Most are based on antibiotic fermentation such as Claviceps purpurea
• The model is divided into four steps, the cellular growth, the extracellular
and intracellular phosphate levels and the product formation.

For cellular growth dx

dt
 
 k1 1  e  k1 pi x  k 2 x 2

For extra cellular Phosphates

x  y p  pi k 2 x 2
dp p
 k3 x
dt p  k2
For extra cellular phosphate

dpi p
 k3  k1 (YP X  pi )(1  e  k1 pi )
dt p  k2
For Product formation

da k3 x
 k4
dt k3  pi2
 Compartment model
 It is a simplest structured model
 compartmentalization of components/section into small sizes
 synthetic components such as RNA and precursors
structural components such as DNA and proteins

This model could also be defined as the assimilatory component and a

synthetic component.

class of transport and reaction occurring in a cell population include

molecular collision
chemical reactions
diffusion
RNA turnover
protein synthesis
increase in cell number
completion of batch process
spontaneous mutation
The Process as the relaxation time after a perturbation (it’s a time
to reach steady state)
• This relaxtion time range from fraction of a second to hour.
what is important in modeling, from a dynamic veiw point is
the relationship b/w the time scale of change in cells
environment ƮE and the spectrum of relaxation time of cellular
processes.
• Those ceullar process which respond to fast environmental
changes is referred as quasi study state (small relaxation time/
ƮE ratio)
• for large relaxation time compared to ƮE, cellular system may
be assumed to be frozen at the initial state (example:
significant genetic changes are not usually expected during a
single batch cultivation)
• Willams proposed a simple two compartment model with assumption such
as

s
Component Rate constants Component
1 2

 The synthetics portion is produced by uptake of external nutrient S with yield

coefficient Y . The rate of cell synthetic component formation is first order in total
cell density x and nutrient mass concentration.
 The structural and genetics cell component is produced from component 1 at a
rate proportional to ρ1 and ρ2
doubling of component 2 is necessary and sufficient for cell division and the cell
density is proportional to the density of component 2 in the culture
Biomass comprised entirely of components 1 and 2
 ds 1
  k1sx
dt y
dx
 k1sx
dt
d1
 k1s ( 1   2 )  k 2 1  2  1
dt
d 2
 k 2 1  2   2 Eq 1
dt
Where   k1s

And 1   2   c  const
 eq 1 can be rewritten has

df 2
  k1sf 2  ( k 2  c ) f c (1  f 2 )
dt
2
Where f2 is c This fraction of cell mass comprised of component 2

Such a model has been used to simulate batch growth using a stationary, nutrient-
exhausted inoculum that corresponds to ρ1= 0.
Simulation results in several frequently observed features of batch microbial
cultures, include

•Existence of lag phase ,cell size increases.

•Exponential growth phase, cell size is maximum.

•Change in composition of cells during growth cycle, since these changes

are evident during exponential phase.

•Stationary phase with relatively small cells.

Other unstructured, nonsegregated models
(assuming one limiting substrate):

Luedeking-Piret model:

r P =  r X + β CX

Used for lactic acid formation by Lactobacillus debruickii where production of lactic
acid was found to occur semi-independently of cell growth.
 Student Learning Objective
(SLO)-7
 Apply in monod Inibition kinetics on
Microbial Batch system
 Inhibition Models
(Very similar to enzyme models)

• Substrate Inhibition
– High substrate concentration inhibits growth
– If a single-substrate enzyme catalyzed reaction is the
rate-limiting step then inhibition of enzyme activity
results in inhibition of microbial growth.
Noncompetitive Substrate Competitive Substrate
m mS
 
 K S  S   S 
1  1   K S 1    S
 S  K I   KI 
 Inhibition Models - cont.
• Product Inhibition
– High concentrations of product can be inhibitory
– Underlying mechanism of product inhibition is unknown
– Approximated as exponential or linear decay functions
Noncompetitive Product Competitive Product
m mS
 
 KS  P   P 
1  1   K S 1    S
 S  K P 
 KP 
 Inhibition Models - cont.
• Inhibition by Toxic Compounds
– inhibition of growth is analogous to enzyme inhibition
Noncompetitive Competitive
m mS
 
 K S  I   I 
1  1   K S 1    S
 S  K I   KI 
Uncompetitive Cell Death

mS mS
 
  k d'
KS I  KS  S
  S 1  
 (1  I / K I )  K I 
 Batch Reactors
• Cell Growth
dX  S
rX   X  m X
dt KS  S

• Substrate Utilization
dS X m S X
rS   
dt YX / S K S  S YX / S

• Product (cometabolic contaminants use negative sign)

1 dP
qp   YP / X  g   g  
X dt
1 dP m S
qp   
X dt KS  S
 Logistic Equation
0.7

• Batch Growth Equation

Cell Concentration (g/L)

0.6

0.5
• Combines Batch growth,
0.4
Monod and Yield
0.3 Coefficients
0.2
• No maintenance
0.1

0
0 5 10 15 20
Time (hr)
 Logistic Equation
Rate Expression for Growth
(1)
dX mS
rX   X  X
dt KS  S
Yield Expression
(2)
Substituting into Eq. 1 for S from Eq 2:
X  X 0  YX / S ( S 0  S ) Describes
Sigmoidal shape
batch growth curve
Integrating dX  m (YX / S S 0  X 0  X )
 X
dt ( K S YX / S  YX / S S 0  X 0  X )

( K S YX / S  YX / S S 0  X 0 )  X  K S YX / S

ln 
  ln(YX / S S 0  X 0  X ) / YX / S S 0    mt
(YX / S S 0  X 0 )  X 0  (YX / S S 0  X 0 )
Unstructured, Nonsegregated Models
• Disadvantage of Unstructured, Nonsegregated Models
– No attempt to utilize or recognize knowledge about cellular
metabolism and regulation
– no description on lag phase
– give no insight to the variables that influence growth
– assume a black box
– assume dynamic response of a cell is dominated by an internal
process with a time delay on the order of the response time
– most processes are assumed to be too fast (psuedo r/n) or too
slow to influence the observed response.
 Filamentous Organisms
• Types of Organisms
– mold
– bacteria or yeast entrapped in a spherical gel particle
– formation of microbial pellets in suspension
• Model - no mass transfer limitations dR
 k d  const
dt
– R - radius of the cell floc or pellet or mold colony
Then the growth of the biomass (M)can be written as 4 3
M   R
3
dM
  4R 2
dt
dR
 k p 4R 2 
dt
 Filamentous Organisms - cont.
dM 2
 M 3
dt

• Where as:   k p (36 )1/ 3

• Integrating the equation:

3 3
 1/ 3 t   t 
M  M0     
 3 3
• M0 is usually very small yield then M  t3
• M will be cubic dependence Model is supported by experimental
data 1.
 REFERENCE

• Biochemical Engineering and fundamentals – James

E. Bailey and David F. Ollis
• Bioprocess Engineering and Basic Concept- Micheal
L. Shuler and Fikret Kargi, page:176-183.
 Models of growth associated and non growth associated product
formation
Growth growth associated product formation
•End products of energy and carbon metabolism are primary metabolism
•Examples; ethanol production by anaerobic fermentation glucose by yeast
•Production of gluconic acid from glucose by gluconobactor
•These products are referred to as growth associated products as their rate of
production parallels the growth of the cell population

Non growth associated product formation

•Products such as vitamins and antibiotics are produced in a batch culture at the end of
exponential phase
•Such secondary metabolites are termed as Non growth associated products
•Their kinetics does not depend on the rate of growth of the culture
Partially growth associated product formation
•Product formation lies between the growth associated product formation and Non growth
associated product formation
•Examples: Amino acids, lactic acid, intermediates from citric acid cycle (citric acid),
extracellular polysaccharides (Xanthan and pullulan ) and solvents such as acetone and
butanol
Luedeking piret Model
Growth associated kinetics
 Non – Growth Associated Kinetics :The penicillin fermentation
• Penicillin production can be considered to be non growth associated as first
approximation

• The production of penicillin occurs after the phase of rapid growth of the fungus
Penicillium chrysogenum. This second phase is referred to as iodophase

• At the end of growth phase, two key enzymes are found to increase in activity
1. phenyl acetic acid- which form a side chain of penicillin
2.penicillin acyl-transferase which attaches activated phenylacetate to 6
aminopenicilianic acid, which is the penicillin nucleus

• Penicillin production is subjected to catabolite repression

• A slow supply of carbon source is necessary to ensure that intermediate metabolites

do not accumulate and reduce production by end product inhibition
• After the period of exponential growth on lactose (0- 24 hours)glucose is slowly fed to
the culture and penicillin production increases from 24 to 140 hours

• The non growth rate associated behavior at specific growth rates greater than 0.015hour

• Below this growth rate, growth – associated product formation is seen

Partial Growth and non growth associated kinetics. Lactic acid Production
• The kinetics of lactic acid production by Lactobacillus provide an example of product
kinetics that are intermediate between growth and non growth associated.

• Lactic acid- anaerobic fermentation of sugars such as glucose and lactose

• It is mainly used as stearyl-2-lactolate

• Although the anaerobic conversion of glucose to lactic acid is a direct route for energy
production, the production kinetics are not soley growth associated