BEN 520

FUNDAMENTALS OF
BIOENGINEERING I
ASST PROF. BETÜL GÜRÜNLÜ
USKUDAR UNIVERSITY
BIOENGINEERING DEPT.

FALL, 2022
COURSE DUTIES
• %30 MIDTERM
• %20 PROJECT
• %50 FINAL
COURSE BOOK

John Villadsen,
Fundamental Bioengineering,
Volume 1
WEEKLY SCHEDULE
• 1ST WEEK: Experimentally Determined Rates of Bio-Reactions
Redox Balances and Consistency Check of Experiments
• 2ND WEEK: Primary Metabolic Pathways and Metabolic Flux Analysis
• 3RD WEEK: A Primer to 13C Metabolic Flux Analysis
• 4TH WEEK: Genome-Scale Models
• 5TH WEEK: Kinetics of Bio-Reactions
• 6TH WEEK: Application of Dynamic Models for Optimal Redesign of
Cell Factories
• 7TH WEEK: Chemical Thermodynamics Applied in Bioengineering
• 8TH WEEK: MIDTERM
WEEKLY SCHEDULE
• 9TH WEEK: Design of Ideal Bioreactors, Mixing and Mass Transfer in
Industrial Bioreactors
• 10TH WEEK: Product Recovery from the Cultures
• 11TH WEEK: Purification of Bio-Products
• 12TH WEEK: Real-Time Measurement and Monitoring of
Bioprocesses
• 13RD WEEK: Control of Bioprocesses
• 14TH WEEK: Scale-Up and Scale-Down
• 15TH WEEK: Commercial Development of Fermentation Processes
DEFINITION OF BIOENGINEERING
• Bioengineering is a relatively
new addition to a long list of
terms starting with “bio.”

• Bioengineering can include

elements of chemical,
electrical and mechanical
engineering, computer
science, materials,
chemistry and biology.
Experimentally Determined Rates of Bio-
Reactions
• Rates of bioreactions are introduced as measured terms in steady-state mass
balances for a continuous stirred tank reactor (CSTR).
• Both mass balances and the reaction rates have the same form for
enzymatically catalyzed reactions and for reactions with living cells. In cell
reactions, the rate of biomass formation is included through a separate mass
balance.
• Reactants absorbed in the liquid phase from a gas phase are treated
separately, and it is shown how experimental errors can lead to errors in the
calculated rates.
• The black-box model for a cell-reaction stoichiometry is introduced and the
yield coefficients are defined. Finally, different methods of controlling the
CSTR at steady state are discussed.
 BIOREACTORS
• The rate of an enzymatically
catalyzed bioreaction, or of a
reaction that involves living cells
(microbial, animal, or plant
cells), can be determined
experimentally in a bioreactor.

• The bioreactors used in

academic research or in an
industrial R&D department to
obtain reaction rates are
normally glass vessels of 0.5–5 l
working volume V.
 BIOREACTORS
• In all cases, the mixing of liquid feed into the medium volume V is
supposed to be good enough to ensure that there is no spatial variation of
substrates or products in the reactor.
• Batch operation or continuous operation of the reactor is typically used,
and the assumption of perfect mixing in the medium volume V will ensure
that the concentrations Si of substrates and Pi of products are the same at
any point in the reactor.
• If the continuous stirred tank reactor (CSTR) is operated in steady state,
there is no accumulation of either products or substrates.
• The liquid flow vf and the feed concentrations of substrates Sf,i are kept
constant in time. Now the medium volume V and the concentrations of
substrates and products both in the reactor [Si, Pj] and in the effluent
concentrations [Se,i, Pe,i] are constant in time.
 BATCH REACTOR
• In the batch reactor, one starts with
a high concentration of substrates
s0i (+ a small amount of biomass for
a fermentation), and Si are
converted to Pj over time. The
volume of medium V in the reactor
is constant in time.
• A gas-phase substrate is introduced
to the liquid through a sparger. It is
absorbed in the liquid and is
consumed by the reaction. • Gaseous products are transferred to
the gas phase by desorption from
the liquid.
 CSTR
• Then, mass balances for substrates
and products are set up. These mass
balances define the reaction rates,
and solution of the mass balances
allows the rates to be calculated
based on measured concentrations
[Si, Pi].
• A steady state CSTR, and the
equipment is shown in Figure.

Figure 2 A commercial laboratory reactor (Biostat A plus from Sartorious) for medium volume between 0.4 and 5 l. The
reactor is supported by packages for either microbial cell or animal/plant cell cultures. The microbial package includes two
6-blade Rushton turbines, two gas-inlets, ports for inoculation, automatic and manual samplers, and temperature control
via a heating blanket and cooling finger. Control of pH, T, dissolved oxygen (DO), stirrer speed, air flow rate, and foam
control.
Mass Balances for a CSTR Operating at Steady
State
• The steady-state mass balances for a continuous reactor with a
working volume V, for example, in m3 liquid medium, in which an
enzymatically catalyzed reaction occurs, is given by Eq. (2.1).
• For each substrate, the volumetric rate of production qs,i of Si, for
example, in units of g Si or C-mol Si l-1 h-1, is multiplied by the reactor
volume V to give the production rate of Si (e.g., in g Si h-1).

• To the production term is added vsf,I, the amount of Si introduced

through the feed and subtracted vse,i, the substrate that leaves the
reactor.

• The sum of the three terms of the steady-state mass balance is zero.
The mass balance for product Pj contains the same terms.

• qs,i (i=1, 2, . . . , N) and qp,j ( j=1, 2, . . . ,M) are always defined as

(volumetric) rates of production. Hence qs,i are always negative and
– qs,i is, therefore, the volumetric rate of consumption of Si.
Catalyzed Reactions
• As is the case for any catalyzed reaction, the rate can be defined either
per volume reactor (q) or per unit of catalyst (r), for example, per unit
mass of catalyst added to each l of reactor.

• This second definition defines the specific reaction rates rs,i and rp,j,
which are obtained from qs,i and qp,j by division with e, the
concentration of enzyme E, for example, in units of g E l-1.

• The specific rates define the activity of the enzyme E to convert Si to Pj.
Catalyzed Reactions
• In reactions with living cells, the cell mass catalyzes the conversion of
Si to Pj, but the substrate is also used to form more biomass X – the
reaction is autocatalytic. Hence biomass is also a product, and similar
to Eq. (2.1) one obtains:
Catalyzed Reactions
• The unit of ri could be g Si produced/g biomass/h (i.e., rsi is negative),
g Pi (g X h)-1, g X (g X h)-1.
• rx is defined as the specific growth rate of the culture, and in most
biotechnology literature rx is called μ. We shall only use this latter
symbol when its meaning is obvious.
• It is seen that μ is the ability (or activity) of the biomass in the reactor
to make more biomass. An active culture can make much biomass per
g biomass present per hour – a resting culture has a low value of
rx=μ.
• Since some substrates (e.g., O2) are captured from the gas phase and
some products (e.g., CO2) are released to the gas phase, one needs to
add an extra term in Eqs. (2.1) and (2.2) for these reaction species.
This term is qTsk or qTpm (e.g., in moles of O2 or CO2 transferred (l h)-1).
• In Eq. (2.4), the term qTsk is positive (there is an influx of O2 to the
liquid), while qTpm is negative.
• The amount of reaction species k and m transferred between the gas
and the liquid phase can be determined from a mass balance on the
gas phase.
Redox Balances and Consistency Check of
Experiments
• A black-box stoichiometry for a single bioreaction is introduced using
element balances to calculate some of the yield coefficients.
• Using a redox balance on the stoichiometry is shown to reduce the
computational work. In a number of examples, redox balances are used to
check experimentally determined yield coefficients.
• The concept of a theoretical maximum yield of a bioreaction is defined and
used on several industrially important processes.
• A given experimental stoichiometry is split into several elementary
stoichiometries with the purpose of finding potential process
improvements.
• Finally, redox balancing is used to analyze typical bio-remediation
processes.
Black-Box Stoichiometry Obtained in a CSTR
Operated at Steady State
• The stoichiometry (1) of Example 2.1 and Eq. (3.1) quantitatively
describes how much biomass, ethanol, CO2, and glycerol is produced
by conversion of one Cmol glucose at a given steady-state operation
condition, that is, at given values of temperature, pH, and dilution
rate D=v/V.
In Eq. (3.1), there is one carbon source (glucose), and the stoichiometry is
normalized to describe the conversion of one C-mol glucose to four
carbon containing compounds, each normalized to one C-mol. Water is
formed as a by-product. There are four elemental balances for C, H, O,
and N.
With the normalization of the glucose consumption rate to one C-mole
per unit time, and with the rates of formation or consumption of the
other carbon-containing compounds also based on one C atom, the four
elemental balances can be written as follows:
• There are six yield coefficients to be determined, and there are four
constraints, the four linear equations in Eq. (3.2). When two rates are
measured, the remaining four rates can be calculated by solutions of the
linear equations.

• The pair [Ysx, Ysn] cannot be chosen since the two yield coefficients are
linearly dependent when all N consumed ends up in the biomass.

• The measurement of Ysn can be used only to check if the experimental

value obtained for Ysx is correct, or one may find Ysx on the basis of
measurement of the ammonia consumption.

• Another independent way to determine Ysx in Eq. (3.1) where no

protonated compounds is formed is to measure the amount of base used
to titrate the medium to a constant pH.
• If the relation Ysn=0.12 Ysx is inserted in the second equation of Eq.
(3.2) and two out of the six combinations [(Ysx, Yse), (Ysx, Ysg), (Ysx,
Ysc), (Yse, Ysg), (Yse, Ysc), and (Ysg, Ysc)] are chosen as measured
quantities, the three remaining yield coefficients can be determined
by solution of the three equations. For example, when (Ysx, Yse) are
the measured quantities:

The solution is [Ysg, Ysc, Ysw]=[0.0700, 0.2759, 0.0373], and

Ysn=0.01657.
• In a more complex stoichiometry with more substrates and products
such as Eq. (3.4), the calculation of nonmeasured quantities increases
dramatically and larger errors are to be expected.

• It is advantageous to normalize the stoichiometry to one C-mol of S1,

one of the N carbon-containing substrates, and to use a basis of 1 C-
mol also for all other substrates and carbon-containing products.
• In Eq. (3.4) where other N containing compounds besides X are
produced, there is, of course, no proportionality between the
biomass yield coefficient and Ys1,n.
Calculation of Stoichiometric Coefficients by
Means of a Redox Balance

• Another principle of preservation is that of redox in a stand-alone black-

box stoichiometric equation, an equation that relates substrates fed to
the reactor and products retrieved from the reactor.

• Balancing of redox will be used as a simpler method to calculate

stoichiometric coefficients.
1) The unit of redox is defined as one atom of hydrogen. Hence, H2 contains
two redox units.

2) For each element, a redox-neutral compound is defined. We choose H2O

for oxygen, CO2 for carbon, NH3 for nitrogen, and the acids H2SO4 and H3PO4
for sulfur and phosphorous, respectively.

3) With these definitions of redox-neutral compounds and with the defined

unit for redox, one obtains that the elements [O, C, N, S, P] have redox levels
[-2, 4, 3, -6, 5]. Any other set of redox-neutral compounds could have been
chosen. This would have no influence on redox balancing of the
stoichiometry since all elements are preserved in the reaction. The choice of
[H2O, CO2, NH3] for the elements [O, C, N] is convenient since usually these
compounds occur in the stoichiometry for bioreactions.
4) After obtaining the redox level of the commonly occurring elements,
one can now calculate the redox level of any compound in a
bioreaction. Thus, glucose C6H12O6 contains 24 redox units or 4 units
per C atom. Lactic acid (CH3CHOHCOOH) contains 12 redox units or 4
per C-atom. Lactic acid is one possible end product of glycolysis in
which glucose is converted to two molecules of lactic acid.
Ethanol (C2H5OH=(CH3O0.5)2 contains six redox units per Cmol. The
degree of reduction of a compound is denoted by κi, for example, κx for
the biomass. If the stoichiometry contains charged compounds, a
positive charge reduces the degree of reduction of the compound by
one, a negative charge increases κi by one unit. Hence, κNH4+ = κNH3- = 0
and κH2S=κHS-=κS-2=8. Both H+ and OH- are redox neutral.
5) In every stand-alone stoichiometry, the redox content of the products
must equal that of the substrates. Thus for Eq. (3.1),

6) In a pathway that does not stand alone but where redox has to be
delivered from another pathway, one needs to balance the pathway by
adding redox equivalents on one side of the stoichiometry. Thus, in the
reduction of glucose to glycerol and in the oxidation of glucose to CO2:

A stand-alone reaction can be constructed from Eqs. (3.5) and (3.6) by

balancing the redox units:
7) Returning to the stoichiometry Eq. (3.1), we find that the
computational work needed when the yield coefficients are
determined by a combination of the carbon balance and the redox
balance is significantly smaller than when the four elemental balances
were used to determine nonmeasured yield coefficients.
• We need two measured yield coefficients to find the remaining two
yield coefficients of carbon-containing compounds. But the problem
has reduced to a 2 × 2 linear algebraic problem from the 3 × 3 linear
algebraic problem of Eq. (3.3). With Ysx and Yse as the measured
quantities one obtains

• Inserting the values of [Ysx, Yse] from Eq. (3.1) one obtains [Ysc, Ysg]=
[0.2759, 0.0700], and to obtain this (correct) result the amount of
algebra involved is trivial compared to the solution of Eq. (3.3).
Applications of the Redox Balance
Consider the experimental data for aerobic yeast cultivation in Figure
2.3. The experiments were performed at pH 5.5 in a mechanically
stirred bioreactor with working volume 2.7 l, sparged with air at a flow
rate of 2.5 vvm (=2.5 2.7(l) min-1=405 l h-1). For D ≤0.25, the biomass
concentration in the effluent is approximately constant, 14 g l-1. At D≤
0.25 h-1, there is virtually no glucose in the effluent, and since so =28 g l-
1, Ysx =14/28 =0.5g g-1=(14/25.17) (28/30)=0.5959 C-mol X/C-mol

glucose for X=CH1.83O0.56N0.17. Since no other products but biomass and

CO2 are produced for D<0.25 h-1, the stoichiometry is
Composition of the Biomass X
Composition of the Biomass X
Composition of the Biomass X
Composition of the Biomass X
Combination of Black-Box Models
• In Example 3.2, a (semiquantitative) model for the influence of a side
reaction on the yield of citric acid was derived by a combination of
two black-box stoichiometries.
• Basically, all that is needed is to construct and combine
stoichiometries that satisfy the requirements of closing the carbon
and redox balances.
• In Example 3.2, one also needed insight into the metabolic pathways
of A. niger function, and the possible activation of oxaloacetate
hydrolase should be known, but often quite obvious combinations of
stoichiometries will suffice.
Combination of Black-Box Models
• Production of a desired metabolite is just one possible fate of the
substrate. If some of the enzymes in the pathway to the desired
metabolite have insufficient capacity, the carbon may overflow into
other pathways to produce undesirable end products. If the culture is
stressed, for example, at a pH lower than the optimal, substantial
carbon is used to keep the culture alive. This is maintenance substrate
consumption. A general black-box model for substrate consumption is
as follows:
Application of Carbon and Redox Balances in
Bio-Remediation Processes
• In bio-remediation, the substrates are broken down to small molecules of
CO2 and H2O, and (in anaerobic processes) some of the capital costs of the
installation may be recovered, for example, biogas and CH4.
• Environmental engineering challenges: The substrate composition and feed
rate fluctuates with time, the biodegradation of pollutants is mediated by
consortia of microorganisms, most of which are not sequenced, and the
composition of the microbial consortium changes with the composition of
the feed.
• The use of a redox balance in the analysis of bioremediation processes is an
excellent tool that gives insight into the nature of the process and supports
quantitative studies.
• Thus, anaerobic degradation of cellulose to the end products CH4 and
CO2 is accomplished in a series of redox-neutral reactions in which the
redox present in the substrate is distributed in the products:
• In Eqs. (3.14) and (3.15), fatty acids, for example, acetic acid and
butyric acid (HBu), are formed together with CO2 and H2. Acetic acid
is decomposed to CH4 and CO2 as shown in Eq. (3.16), and it would
be very desirable if butyric acid could also be decomposed by the
second reaction in Eq. (3.16). Example 9.4 gives a more detailed
description of the processes that actually occur.
• In bio-remediation biological oxygen demand (BOD) and chemical oxygen
demand (COD) are used as units of redox content of a substrate. BOD is the
most important. It is defined as the amount of O2 needed to oxidize all the
reduced matter (solids or solutes) in, for example, a sample of wastewater
to redox neutral compounds with the help of a consortium of
microorganisms, either naturally occurring in the sample or added to it.
• BOD is measured in units of, for example, mg O2/l wastewater or kg O2/ton
of wastewater.
• Thus, a solution containing 0.2 g/l cellulose (mostly as colloids) has a BOD
of 32 000 mg O2 (0.2/30) =213 mg l1, since the cellulose is degraded to
redox neutral products by

• Typical raw sewage has a BOD value of about 300 mg l-1.

Example 3.6 Designing an anaerobic wastewater
treatment unit for slaughterhouse waste
• The wastewater from an animal processing plant is sent to an anaerobic
digester. The major pollutant is a mixture of animal fats that will be
represented as glycerol-tripalmitate (tripalmitin), C51H98O6. The digester is
a stirred tank reactor that operates in steady state. Laboratory
measurements show that the feed stream has a BOD of 3180 kg per day of
operation. There are three product streams: (1) a gaseous mixture of CH4
and CO2. (2) A liquid effluent that contains no fat. This effluent stream is
measured to have a BOD of 330 kg/day, and the main component is
characterized as acetic acid, HAc. (3) A solid effluent of biomass, 160
kg/day with an ash content of 7 wt% (X=CH1.8O0.5N0.2). As the first step in
the design of the digester, the stoichiometry of the process is to be
determined.
A) First the feed rate of the tripalmitin solution is determined from the
BOD of the feed. The degree of reduction of palmitin is κs = 1/51.
(4.51+1.98-2.6)=5.686 (C-mol)-1. Since all the fat is digested, this
amount of redox has to be distributed between the products, CH4,
biomass, HAc and CO2 (which, of course, is redox neutral). Since the
total redox content of the feed corresponds to 3180/32=99.375 kmol
O2 per day needed to oxidize all the redox in the feed, the feed stream
is calculated from M*5.686=99.375*4 or M=69.91 kC-mol
tripalmitin/day.
B) The liquid effluent (HAc) is calculated from MHAc *4=(330/32)*4
or MHAc=10.31 kC-mol HAc/day. The solids stream is calculated
from MX =160*0.93/24.6 =6.05 kC-mol biomass/day. Thus, the
total production of CO2 and CH4 = MCO2 + MCH4= 69.91-10.31 -6.05
=53.55 kC-mol/day.
C) A total redox balance states how the redox content of the feed is
distributed in the products: 69.91*5.686 = 10.31*4 + 6.05*4.20 +
MCH4 *8 + MCO2 *0  MCH4*41.36 kC mol CH4/day. Hence
MCO2=12.19 kC-mol CO2/day.
 The stoichiometry of the black-box model for the digestion of
the fat:

• Oxygen is not the only redox sink (or electron acceptor), as a large
number of anions or cations can have a similar role. SO4-2, NO3-1, S2O3-2,
and Fe+3 are typical examples. The sulfate in stagnant oxygen-depleted
water is reduced to sulfide with organic waste material as carbon and
energy source:
• Eq. (3.18) is a typical dissimilatory redox process. Such processes are
utilized in many wastewater cleaning processes (which in contrast to
Eq. (3.18) produce environment-friendly end products). Thus, both
nitrate and ammonia can be removed by the following process:
A redox balance yields:

For Ysx <0.952 (when Ysx =0.952, all the redox from the
glucose is used to produce biomass), some nitrate can be
reduced to N2. The highest possible value α=0.4 for nitrate
removal is obtained when Ysx=0
• The stoichiometry Eq. (3.19) is, of course, not correct at the
approximately neutral pH of the wastewater treatment process. In
actual fact, the stoichiometry should be:

• the complicated formulation Eq. (3.22,) one should first make the
carbon and redox balances close. Next, the charge balance is closed
and the coefficient to H+ is obtained. Finally, Ysw is calculated, either
from the oxygen or from the hydrogen balance.