434 10.

MASS TRANSFER

PROBLEMS
10.1 Rate-controlling processes in fermentation
Serratia marcescens bacteria are used for the production of threonine. The maximum
specific oxygen uptake rate of S. marcescens in batch culture is 5 mmol O2 g21 h21. It is
planned to operate the fermenter to achieve a maximum cell density of 40 g l21. At the
fermentation temperature and pressure, the solubility of oxygen in the culture liquid is
8 3 1023 kg m23. At a particular stirrer speed, kLa is 0.15 s21. Under these conditions, will
the rate of cell metabolism be limited by mass transfer or depend solely on metabolic
kinetics?
10.2 Test for oxygen limitation
An 8-m3 stirred fermenter is used to culture Agrobacterium sp. ATCC 31750 for production
of curdlan. The liquid medium contains 80 g l21 sucrose. Under optimal conditions, 1.0 g
dry weight of cells is produced for every 4.2 g of sucrose consumed. The fermenter is
sparged with air at 1.5 atm pressure, and the specific oxygen demand is 7.5 mmol per g
dry weight per h. To achieve the maximum yield of curdlan, the fermentation temperature
is held constant at 32 C. The solubility of oxygen in the fermentation broth is 15% lower
than in water due to solute effects. If the maximum kLa that can be achieved is 0.10 s21,
does the fermenter’s mass transfer capacity support complete consumption of substrate?
10.3 kLa required to maintain critical oxygen concentration
A genetically engineered strain of yeast is cultured in a bioreactor at 30 C for production
of heterologous protein. The oxygen requirement is 80 mmol l21 h21; the critical oxygen
concentration is 0.004 mM. The solubility of oxygen in the fermentation broth is estimated
to be 10% lower than in water due to solute effects.
(a) What is the minimum mass transfer coefficient necessary to sustain this culture with
dissolved oxygen levels above critical if the reactor is sparged with air at
approximately 1 atm pressure?
(b) What mass transfer coefficient is required if pure oxygen is used instead of air?
10.4 Oxygen transfer with different impellers
A 10-m3 stirred fermenter with liquid height 2.3 m is used to culture Trichoderma reesei for
production of cellulase. The density of the culture fluid is 1000 kg m23. An equation for the
oxygen transfer coefficient as a function of operating variables has been developed for T.
reesei broth:

0:7
PT
kL a 5 2:5 3 1023 u0:3
G
VL

where kLa has units of s21, PT is the total power input in W, VL is the liquid volume in m3,
and uG is the superficial gas velocity in m s21. The fermenter is sparged using a gas flow
rate of 0.6 vvm (vvm means volume of gas per volume of liquid per minute). The vessel is
stirred with a single impeller but two alternative impeller designs, a Rushton turbine and
a curved-blade disc turbine, are available. Both impellers are sized and operated so that
their ungassed power draw is 9 kW.
(a) If the power loss with gassing is 50% for the Rushton and 5% for the curved-blade
turbine, compare the kLa values achieved using each impeller.

3. PHYSICAL PROCESSES
 PROBLEMS 435

(b) What is the percentage contribution to PT from gassing with the two different
impellers?
(c) If the cell concentration is limited to 15 g l21 using the Rushton turbine because of
mass transfer effects, estimate the maximum possible cell concentration with the
curved-blade disc turbine.
It is decided to install the Rushton turbine, but to compensate for the effect on kLa of its
loss of power with gassing by increasing the gas flow rate.
(d) Estimate the gas flow rate required to obtain the same maximum cell concentration
using the Rushton turbine as that achieved with the curved-blade disc turbine.
Express your answer in vvm. What are your assumptions? (Iterative solution may be
required.)
10.5 Foam control and oxygen transfer
Foaming is controlled routinely in fermenters using a foam sensor and pump for
automatic addition of antifoam agent. As shown in Figure 10P5.1, the foam sensor is
located at the top of the vessel above the liquid surface. When a head of foam builds up so
that foam contacts the lower tip of the sensor, an electrical signal is sent to the pump to
add antifoam. The antifoam agent destroys the foam, the foam height is therefore reduced,
contact with the foam sensor is broken, and the pump supplying the antifoam agent is
switched off. Further build-up of foam reactivates the control process.
If the position of the foam sensor is fixed, when the gap between the liquid surface
and sensor is reduced by raising the liquid height, a smaller foam build-up is tolerated
before antifoam agent is added. Therefore, antifoam addition will be triggered more often
if the working volume of the vessel is increased. Although a greater fermenter working
volume means that more cells and/or product are formed, addition of excessive antifoam

FIGURE 10P5.1 Stirred fermenter

Antifoam with automatic foam control system.

Foam
sensor
Foam
Liquid height layer

3. PHYSICAL PROCESSES
436 10. MASS TRANSFER

agent could reduce kLa significantly, thereby increasing the likelihood of mass transfer
limitations.
A stirred fermenter of diameter 1.5 m is used to culture Bacillus licheniformis for
production of serine alkaline protease. The fermenter is operated five times with automatic
antifoam addition using five different liquid heights. The position of the foam sensor is the
same in each fermentation. The volume of antifoam added and the kLa at the end of the
culture period are recorded.

Liquid height (m) Antifoam added (l) kLa (s21)

1.10 0.16 0.016

1.29 0.28 0.013
1.37 1.2 0.012
1.52 1.8 0.012
1.64 2.4 0.0094

Under ideal conditions, the maximum specific oxygen uptake rate for B. licheniformis is
2.6 mmol g21 h21. When glucose is used as the carbon source at an initial concentration of
20 g l21, a maximum of 0.32 g of cells are produced for each g of glucose consumed, and
0.055 g of protease is produced per g of biomass formed. The solubility of oxygen in the
broth is estimated as 7.8 g m23.
(a) Using the kLa values associated with each level of antifoam addition, estimate the
maximum cell concentrations supported by oxygen transfer as a function of liquid
height. Assume that antifoam exerts a much stronger influence on kLa than on other
properties of the system such as oxygen solubility and specific oxygen uptake rate.
(b) Calculate the maximum mass of cells and maximum mass of protease that can be
produced based on the oxygen transfer capacity of the fermenter as a function of
liquid height.
(c) Is protease production limited by oxygen transfer at any of the liquid heights tested?
(d) What operating liquid height would you recommend for this fermentation process?
Explain your answer.
10.6 Improving the rate of oxygen transfer
Rifamycin is produced in a 17-m3 stirred fermenter using a mycelial culture, Nocardia
mediterranei. The fermenter is sparged with air under slight pressure so the solubility of
oxygen in the broth is 10.7 g m23. Data obtained during operation of the fermenter are
shown in Figure 10P6.1. After about 147 hours of culture, vegetable oil is added to the
broth to disperse a thick build-up of foam. This has a severe effect on the oxygen transfer
coefficient and reduces the dissolved oxygen tension.
(a) Calculate the steady-state oxygen transfer rate before and after addition of the
vegetable oil.
(b) The relationship between kLa and the fermenter operating conditions is:

0:5
PT
kL a ~ uG 0:3
VL

3. PHYSICAL PROCESSES
 PROBLEMS 437
FIGURE 10P6.1

Dissolved oxygen tension (% air saturation)

350 Online time-
Vegetable oil course data from a stirred fermen-
300 addition ter used for rifamycin production.
250

kLa (h–1) 200

150

100

50

0
130 135 140 145 150 155 160
Time (hours)

Because increasing the gas flow rate would aggravate the problems with foaming, it is
decided to restore the value of kLa by increasing the power input by stirring. If the
power contribution from gas sparging is negligible, by how much does the stirrer
power need to be increased to overcome the effects of the vegetable oil on kLa?
(c) To save the cost of increasing the power input, instead of (b), it is decided to improve
oxygen transfer by sparging the fermenter with oxygen-enriched air. The total gas
flow rate and pressure are unchanged. To restore the rate of oxygen transfer after
vegetable oil addition to that before oil was added, what volume percentage of oxygen
is required in the sparge gas if the desired dissolved oxygen concentration in the broth
is 6.2 3 1023 kg m23?
10.7 Oxygen transfer for different cell types
The specific oxygen demands and critical oxygen concentrations for typical microbial,
plant, and animal cell cultures are listed below.

Cell culture qO Ccrit (mmol l21)

Escherichia coli 8.5 mmol (g dry weight)21 h21 0.0082

Vitis vinifera (grape) 0.60 mmol (g dry weight)21 h21 0.055
Chinese hamster ovary (CHO) 3.0 3 10210 mmol cell21 h21 0.020

(a) Estimate the kLa required to achieve cell concentrations of 25 g dry weight l21 for
E. coli and V. vinifera and 3.0 3 109 cells l21 for CHO cells, while maintaining the
dissolved oxygen concentration above critical. The oxygen solubility in the media used
for the cultures is 7.2 3 1023 kg m23.
(b) The relationship between kLa and the power input to a 1-m3 stirred bioreactor is:

0:5
PT
kL a ~
VL

3. PHYSICAL PROCESSES
438 10. MASS TRANSFER

Compare the bioreactor power requirements for culture of the three different cell types
under the conditions described in (a).
10.8 Single-point kLa determination using the oxygen balance method
A 200-litre stirred fermenter contains a batch culture of Bacillus subtilis bacteria at 28 C.
Air at 20 C is pumped into the vessel at a rate of 1 vvm (vvm means volume of gas per
volume of liquid per minute). The average pressure in the fermenter is 1 atm. The
volumetric flow rate of off-gas from the fermenter is measured as 189 l min21. The exit gas
stream is analysed for oxygen and is found to contain 20.1% O2. The dissolved oxygen
concentration in the broth is measured using an oxygen electrode as 52% air saturation.
The solubility of oxygen in the fermentation broth at 28 C and 1 atm air pressure is
7.8 3 1023 kg m23.
(a) Calculate the oxygen transfer rate.
(b) Determine the value of kLa for the system.
(c) The oxygen analyser used to measure the exit gas composition was incorrectly
calibrated. If the oxygen content has been overestimated by 10%, what error is
associated with the result for kLa?
10.9 Steady-state kLa measurement
Escherichia coli bacteria are cultured at 35 C and 1 atm pressure in a 500-litre fermenter
using the following medium:

Component Concentration (g l21)

glucose 20
sucrose 8.5
CaCO3 1.3
(NH4)2SO4 1.3
Na2HPO4 0.09
KH2PO4 0.12

Air at 25 C and 1 atm is sparged into the vessel at a rate of 0.4 m3 min21. The dissolved
oxygen tension measured using a polarographic electrode calibrated in situ in sterile
culture medium is 45% air saturation. The gas flow rate leaving the fermenter is measured
using a rotary gas meter as 6.3 l s21. The oxygen concentration in the off-gas is 19.7%.
(a) Estimate the solubility of oxygen in the fermentation broth. What are your
assumptions?
(b) What is the oxygen transfer rate?
(c) Determine the value of kLa.
(d) Estimate the maximum cell concentration that can be supported by oxygen transfer in
this fermenter if the specific oxygen demand of the E. coli strain is 5.4 mmol g21 h21.
(e) If the biomass yield from the combined sugar substrates is 0.5 g g21, is growth in the
culture limited by oxygen transfer or substrate availability?
10.10 Oxygen transfer in a pressure vessel
A fermenter of diameter 3.6 m and liquid height 6.1 m is used for production of ustilagic
acid by Ustilago zeae. The pressure at the top of the fermenter is 1.4 atma. The vessel is

3. PHYSICAL PROCESSES
 PROBLEMS 439

stirred using dual Rushton turbines and the fermentation temperature is 29 C. The
dissolved oxygen tension is measured using two electrodes: one electrode is located near
the top of the tank, the other is located near the bottom. Both electrodes are calibrated
in situ in sterile culture medium. The dissolved oxygen reading at the top of the fermenter
is 50% air saturation; the reading at the bottom is 65% air saturation. The fermenter is
sparged with air at 20 C at a flow rate of 30 m3 min21 measured at atmospheric pressure.
Off-gas leaving the vessel at a rate of 20.5 m3 min21 contains 17.2% oxygen. The solubility
of oxygen in the fermentation broth is not significantly different from that in water. The
density of the culture broth is 1000 kg m23.
(a) What is the oxygen transfer rate?
(b) Estimate the pressure at the bottom of the tank.
(c) The gas phase in large fermenters is often assumed to exhibit plug flow. Under these
conditions, no gas mixing occurs so that the gas composition at the bottom of the tank
is equal to that in the inlet gas stream, while the gas composition at the top of the tank
is equal to that in the outlet gas stream. For the gas phase in plug flow, estimate the
oxygen solubility at the top and bottom of the tank.
(d) What is the value of kLa?
(e) If the cell concentration is 16 g l21, what is the specific oxygen demand?
(f) Industrial fermentation vessels are rated for operation at elevated pressures so they
can withstand steam sterilisation. Accordingly, the fermenter used for ustilagic acid
production can be operated safely at a maximum pressure of 2.7 atma. Assuming that
respiration by U. zeae and the value of kLa are relatively insensitive to pressure, what
maximum cell concentration can be supported by oxygen transfer in the fermenter
after the pressure is raised?
10.11 Dynamic kLa measurement
The simple dynamic method is used to measure kLa in a fermenter operated at 30 C and
1 atm pressure. Data for the dissolved oxygen concentration as a function of time during
the reoxygenation step are as follows.

Time (s) CAL (% air saturation)

10 43.5
15 53.5
20 60.0
30 67.5
40 70.5
50 72.0
70 73.0
100 73.5
130 73.5

(a) Calculate the value of kLa.

(b) What additional experiments are required to check the reliability of this kLa result?

3. PHYSICAL PROCESSES
440 10. MASS TRANSFER

10.12 kLa measurement using the dynamic pressure method

The dynamic pressure method is applied for measurement of kLa in a 3000-l stirred
fermenter containing a suspension culture of Micrococcus glutamicus. The stirrer is operated
at 60 rpm and the gas flow rate is fixed at 800 l min21. The following dissolved oxygen
concentrations are measured using a polarographic dissolved oxygen electrode after a step
increase in fermenter pressure.

Time (s) CAL (% air saturation)

6 50.0
10 56.1
25 63.0
40 64.7

The steady-state dissolved oxygen tension at the end of the dynamic response is 66% air
saturation.
(a) Estimate the value of kLa.
(b) An error is made determining the steady-state oxygen level, which is taken as 70%
instead of 66% air saturation. What effect does this 6% error in CAL have on the result
for kLa?
(c) At the end of the kLa experiment, the electrode response time is measured by
observing the output after a step change in dissolved oxygen tension from 0 to 100%
air saturation. The stirrer speeds tested are 40, 50, and 60 rpm. Figure 10P12.1 at the
bottom of page shows a chart recording of the results at 60 rpm; the results at 50 rpm
are not significantly different. From this information, how much confidence do you
have in the kLa measurements? Explain your answer.
10.13 Surface versus bubble aeration
Hematopoietic cells used in cancer treatment are cultured at 37 C in an 8.5-cm diameter
bioreactor with working volume 500 ml. The culture fluid is mixed using a stirrer speed of
30 rpm. The reactor is operated at ambient pressure.

FIGURE 10P12.1 Chart recording of

100 the electrode response at 60 rpm to a
90
Dissolved oxygen reading

step change in dissolved oxygen tension.

80
(% air saturation)

70
60
50
40
30
20
10
0
0 1 2 3 4 5 6 7 8 9
Time (s)

3. PHYSICAL PROCESSES
 PROBLEMS 441

(a) The dissolved oxygen tension is controlled at 50% air saturation using surface aeration
only. A 50:20:30 mixture of air, oxygen, and nitrogen is passed at a fixed flow rate
through the headspace. The specific oxygen uptake rate for hematopoietic cells is
7.7 3 10212 g cell21 h21 and the cell concentration is 1.1 3 109 cells l21. Estimate the
value of kLa for surface aeration.
(b) Instead of surface aeration, the bioreactor is sparged gently with air. When the
dissolved oxygen tension is maintained at the critical level of 8% air saturation, the
cell concentration is 3.9 3 109 cells l21. What is the kLa for bubble aeration?
(c) Surface aeration is preferred for this shear-sensitive culture, but the surface kLa needs
improvement. For the gassing conditions applied in (a), estimate the vessel diameter
required for surface aeration to achieve the same kLa obtained with sparging. What are
your assumptions?
10.14 Shake-flask aeration
A mixed culture of heterotrophic microorganisms isolated from the Roman baths at Bath is
prepared for bioleaching of manganese ore. One hundred ml of molasses medium is used
in 300-ml flasks with 4-cm-long silicone sponge stoppers. The width of the flask opening is
3.2 cm. The cultures are incubated at 30 C on an orbital shaker operated at 80 rpm.
(a) With the flask closure removed, kLa for gas
liquid mass transfer is estimated using
the dynamic method. During the reoxygenation step, the dissolved oxygen tension
measured using a small, rapid-response electrode is 65% air saturation after 5 s and
75% after 30 s. The steady-state oxygen tension is 90% air saturation. What is the
resistance to oxygen transfer in the flask? What are your assumptions?
(b) An expression for the mass transfer coefficient Kc for the flask closure is:
De Ac
Kc 5
Lc VG

where De is the effective diffusion coefficient of oxygen in the closure material, Ac is

the cross-sectional area of the closure, Lc is the closure length, and VG is the volume of
gas in the flask. If De for silicone sponge is 20.8 cm2 s21, what resistance to oxygen
transfer is provided by the flask closure?
(c) What proportion of the total resistance to oxygen transfer does the flask closure
represent?
(d) It is decided to improve the rate of gas
liquid oxygen transfer so that the resistance
to oxygen transfer within the flask is approximately equal to that of the flask closure.
A study of the dependence of kLa on shake-flask operating parameters yields the
relationship:

0:85
VF
kL a ~ N 1:2
VL

where N is the shaker speed in rpm, VF is the flask size in ml, and VL is the liquid
volume in ml. If the shaker speed can be increased to a maximum of 150 rpm:
(i) What size flask is needed if the culture volume remains at 100 ml?
(ii) If 300-ml flasks are the only shake flasks available, what culture volume should
be used?

3. PHYSICAL PROCESSES
442 10. MASS TRANSFER

References
[1] R.E. Treybal, Mass-Transfer Operations, third ed., McGraw-Hill, 1980.
[2] Perry’s Chemical Engineers’ Handbook, eighth ed., McGraw-Hill, 2008.
[3] T.K. Sherwood, R.L. Pigford, C.R. Wilke, Mass Transfer, McGraw-Hill, 1975.
[4] J.M. Coulson, J.F. Richardson, J.R. Backhurst, J.H. Harker, Coulson and Richardson’s Chemical Engineering,
vol. 1, sixth ed., Butterworth-Heinemann, 1999 (Chapters 10 and 12).
[5] M.J. Johnson, Metabolism of penicillin-producing molds, Ann. N.Y. Acad. Sci. 48 (1946) 57
66.
[6] C.E. Clifton, A comparison of the metabolic activities of Aerobacter aerogenes, Eberthella typhi and Escherichia
coli, J. Bacteriol. 33 (1937) 145
162.
[7] G.C. Paul, M.A. Priede, C.R. Thomas, Relationship between morphology and citric acid production in sub-
merged Aspergillus niger fermentations, Biochem. Eng. J. 3 (1999) 121
129.
[8] O. Rahn, G.L. Richardson, Oxygen demand and oxygen supply, J. Bacteriol. 41 (1941) 225
249.
[9] P. Gikas, A.G. Livingston, Use of specific ATP concentration and specific oxygen uptake rate to determine
parameters of a structured model of biomass growth, Enzyme Microb. Technol. 22 (1998) 500
510.
[10] E.B. Chain, G. Gualandi, G. Morisi, Aeration studies. IV. Aeration conditions in 3000-liter submerged fermen-
tations with various microorganisms, Biotechnol. Bioeng. 8 (1966) 595
619.
[11] A.H.E. Bijkerk, R.J. Hall, A mechanistic model of the aerobic growth of Saccharomyces cerevisiae, Biotechnol.
Bioeng. 19 (1977) 267
296.
[12] A.L. Jensen, J.S. Schultz, P. Shu, Scale-up of antibiotic fermentations by control of oxygen utilization,
Biotechnol. Bioeng. 8 (1966) 525
537.
[13] K. Ozergin-Ulgen, F. Mavituna, Oxygen transfer and uptake in Streptomyces coelicolor A3(2) culture in a batch
bioreactor, J. Chem. Technol. Biotechnol. 73 (1998) 243
250.
[14] W.H. Bartholomew, E.O. Karow, M.R. Sfat, R.H. Wilhelm, Oxygen transfer and agitation in submerged fer-
mentations, Ind. Eng. Chem. 42 (1950) 1801
1809.
[15] A. Pinches, L.J. Pallent, Rate and yield relationships in the production of xanthan gum by batch fermenta-
tions using complex and chemically defined growth media, Biotechnol. Bioeng. 28 (1986) 1484
1496.
[16] P.A. Bond, M.W. Fowler, A.H. Scragg, Growth of Catharanthus roseus cell suspensions in bioreactors: on-line
analysis of oxygen and carbon dioxide levels in inlet and outlet gas streams, Biotechnol. Lett. 10 (1988)
713
718.
[17] A. Kato, S. Nagai, Energetics of tobacco cells, Nicotiana tabacum L., growing on sucrose medium, Eur. J. Appl.
Microbiol. Biotechnol. 7 (1979) 219
225.
[18] P. Ducommun, P.-A. Ruffieux, M.-P. Furter, I. Marison, U. von Stockar, A new method for on-line measure-
ment of the volumetric oxygen uptake rate in membrane aerated animal cell cultures, J. Biotechnol. 78 (2000)
139
147.
[19] S. Tatiraju, M. Soroush, R. Mutharasan, Multi-rate nonlinear state and parameter estimation in a bioreactor,
Biotechnol. Bioeng. 63 (1999) 22
32.
[20] S.P.S. Andrew, Gas
liquid mass transfer in microbiological reactors, Trans. IChemE 60 (1982) 3
13.
[21] J.J. Heijnen, K. van’t Riet, A.J. Wolthuis, Influence of very small bubbles on the dynamic kLA measurement in
viscous gas
liquid systems, Biotechnol. Bioeng. 22 (1980) 1945
1956.
[22] K. van’t Riet, Review of measuring methods and results in nonviscous gas
liquid mass transfer in stirred
vessels, Ind. Eng. Chem. Process Des. Dev. 18 (1979) 357
364.
[23] A. Ogut, R.T. Hatch, Oxygen transfer into Newtonian and non-Newtonian fluids in mechanically agitated
vessels, Can. J. Chem. Eng. 66 (1988) 79
85.
[24] Y. Kawase, M. Moo-Young, The effect of antifoam agents on mass transfer in bioreactors, Bioprocess Eng. 5
(1990) 169
173.
[25] A. Prins, K. van’t Riet, Proteins and surface effects in fermentation: foam, antifoam and mass transfer,
Trends Biotechnol. 5 (1987) 296
301.
[26] International Critical Tables (1926
1930) McGraw-Hill; first electronic ed., International Critical Tables of
Numerical Data, Physics, Chemistry and Technology (2003), Knovel.
[27] Y.H. Lee, G.T. Tsao, Dissolved oxygen electrodes, Adv. Biochem. Eng. 13 (1979) 35
86.
[28] V. Linek, J. Sinkule, V. Vacek, Dissolved oxygen probes, in: M. Moo-Young (Ed.), Comprehensive Biotechnology,
vol. 4, Pergamon Press, 1985, pp. 363
394.

3. PHYSICAL PROCESSES
Chapter 10
Mass Transfer

10.1 Rate-controlling processes in fermentation

Converting the units of the maximum specific oxygen uptake rate using the molecular weight of oxygen =
32.0 (Table C.1, Appendix C):
1 gmol 32.0 g 1h
qO = 5 mmol g −1 h −1 . . . = 4.44 × 10−5 g g −1 s −1
1000 mmol 1 gmol 3600 s

At a cell density of 40 g l–1, the maximum oxygen requirement is:

1000 l 1 kg
qO x = 4.44 × 10−5 g g −1 s −1 (40 g l−1 ) . . = 1.78 × 10−3 kg m −3 s −1
1 m3 1000 g

The rate of oxygen transfer is given by Eq. (10.39); NA takes a maximum value when CAL = 0:
N A = kL a CAL
*
= 0.15 s −1 (8 × 10−3 kg m −3 ) = 1.20 × 10−3 kg m −3 s −1
As the maximum oxygen demand of the culture (1.78 × 10–3 kg m–3 s–1) is greater than the maximum
oxygen transfer rate in the fermenter (1.20 × 10–3 kg m–3 s–1), the system will be limited by mass transfer.
Answer: Limited by mass transfer

10.2 Test for oxygen limitation

qO = 7.5 mmol g–1 h–1. Converting the units of qO to g g–1 s–1 using the molecular weight of oxygen = 32.0
(Table C.1, Appendix C):
1 gmol 32 g 1h
qO = 7.5 mmol g −1 h −1 . . . = 6.67 × 10−5 g g −1 s −1
1000 mmol 1 gmol 3600 s
*
CAL for oxygen in water at 1 atm air pressure and 32°C can be interpolated from the values for 30°C and
35°C listed in Table 10.2:
*
CAL (O 2 in water at 1 atm air pressure and 32 o C)

(32 − 30) o C
= 8.05 × 10−3 kg m −3 − (8.05 − 7.52) × 10−3 kg m −3
(35 − 30) o C
= 7.84 × 10−3 kg m −3
*
From Henry’s law Eq. (10.45), CAL is directly proportional to the total gas pressure. As the pressure in the
fermenter is 1.5 atm rather than 1 atm:
 1.5 atm 
*
CAL (O 2 in water at 1.5 atm air pressure and 32 o C) = 7.84 × 10−3 kg m −3  −2
 = 1.18 × 10 kg m
−3

 1 atm 
The solubility of oxygen in the fermentation broth is 15% lower than in water; therefore:
*
CAL (O 2 in fermentation broth at 1.5 atm air pressure and 32 o C) = 0.85 (1.18 × 10−2 kg m −3 )
= 9.99 × 10−3 kg m −3
Converting to units of g l–1:
*
CAL (O 2 in fermentation broth at 1.5 atm air pressure and 32 o C)

1
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

1000 g 1 m3
= 9.99 × 10−3 kg m −3 . .
1 kg 1000 l
= 9.99 × 10−3 g l−1
If 1.0 g of cells is produced for every 4.2 g of sucrose consumed, then complete consumption of 80 g l–1 of
sucrose produces (80 g l–1)/(4.2 g g–1) = 19.05 g l–1 of cells. The kLa required to achieve this cell
concentration under conditions allowing the maximum rate of mass transfer is estimated using Eq. (10.42):
xmax qO 19.05 g l−1 (6.67 × 10−5 g g −1 s −1 )
kL a = = = 0.13 s −1
*
CAL 9.99 × 10−3 g l−1
As this value of kLa is greater than the maximum that can be achieved in the fermenter (0.10 s–1), the
fermenter’s mass transfer capacity does not support complete consumption of substrate.
Answer: No

10.3 kLa required to maintain critical oxygen concentration

QO = 80 mmol l–1 h–1. Converting the units of QO to kg m–3 s–1 using the molecular weight of oxygen =
32.0 (Table C.1, Appendix C) and 1 m3 = 103 l (Table A.2, Appendix A):
1 gmol 32.0 g 1 kg 103 l 1h
QO = 80 mmol l−1 h −1 . . . . 3
. = 7.11 × 10−4 kg m −3 s −1
1000 mmol 1 gmol 1000 g 1 m 3600 s

Converting the units of the critical oxygen concentration to kg m–3:

1 gmol 32.0 g 1 kg 103 l
Ccrit = 0.004 mM = 0.004 mmol l−1 . . . . = 1.28 × 10−4 kg m −3
1000 mmol 1 gmol 1000 g 1 m3

(a)
From Table 10.2, the solubility of oxygen in water at 30°C under 1 atm air pressure is 8.05 × 10–3 kg m–3.
If the solubility in fermentation broth is 10% lower than this:
*
CAL = 0.9 (8.05 ×10−3 kg m −3 ) = 7.25 × 10−3 kg m −3
To maintain the oxygen concentration in the broth at the critical level, from Eqs (10.40) and (10.43):
QO 7.11 × 10−4 kg m −3 s −1
(kL a )crit = = = 0.10 s −1
*
(CAL − Ccrit ) (7.25 × 10−3 − 1.28 × 10−4 ) kg m −3
Answer: 0.10 s–1

(b)
From Section 10.6.5, if pure oxygen is used instead of air, the solubility of oxygen in the fermentation
broth is increased by a factor of 4.8:
*
CAL = 4.8 (7.25 × 10−3 kg m −3 ) = 3.48 × 10−2 kg m −3
Therefore, to maintain the critical oxygen concentration:
QO 7.11 × 10−4 kg m −3 s −1
(kL a )crit = = = 0.021s −1
*
(CAL − Ccrit ) (3.48 × 10−2 − 1.28 × 10−4 ) kg m −3
Answer: 0.021 s–1

2
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

10.4 Oxygen transfer with different impellers

VL = 10 m3; HL = 2.3 m; ρ = 1000 kg m–3; P0 = 9 kW = 9000 W. The gas flow rate is 0.6 m3 m–3 min–1;
therefore, for a tank with liquid volume 10 m3, the volumetric gas flow rate Fg is:
1 min
Fg = 0.6 m3 m −3 min −1 (10 m3 ) . = 0.1 m3 s −1
60 s

The superficial gas velocity uG is equal to Fg divided by the cross-sectional area of the fermenter (Section
10.9). The cross-sectional area of the tank A can be calculated from VL and HL, as VL for a cylindrical tank
is equal to AHL. Therefore:
VL 10 m3
A= = = 4.35 m 2
H L 2.3 m
and
Fg 0.1 m3 s −1
uG = = 2
= 0.023 m s −1
A 4.35 m
The power input by gassing Pv is evaluated using Eq. (8.13) with g = 9.8066 m s–2 (Section 2.3) and 1 W =
1 kg m2 s–3 (Table A.8, Appendix A):
1W
Pv = 0.1 m3 s −1 (1000 kg m −3 ) (9.8066 m s −2 ) (2.3 m) = 2256 kg m 2 s −3 . = 2256 W
1 kg m 2 s −3

(a)
For the Rushton turbine, the power input by the impeller is 0.5P0 = 4500 W. Adding this to the power
input by gassing Pv, PT = 4500 W + 2256 W = 6756 W. Substituting values for PT, VL and uG into the
equation for kLa using the units specified gives:
0.7
 6756 
kL a = 2.5 × 10−3  0.3
 (0.023) = 0.077 s
−1

 10 
For the curved-blade disc turbine, the power input by the impeller is 0.95P0 = 8550 W. Adding this to the
power input by gassing Pv, PT = 8550 W + 2256 W = 10,806 W. Therefore:
0.7
 10,806 
kL a = 2.5 × 10−3  0.3
 (0.023) = 0.107 s
−1

 10 
Answer: For the Rushton turbine, kLa is 0.077 s–1; for the curved-blade disc turbine, kLa is 39% higher at
0.107 s–1

(b)
For the Rushton turbine:
Pv 2256 W
= × 100% = 33%
PT 6756 W
For the curved-blade disc turbine:
Pv 2256 W
= × 100% = 21%
PT 10,806 W

Answer: 33% for the Rushton turbine, 21% for the curved-blade disc turbine

3
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

(c)
*
From Eq. (10.42), as CAL and qO do not vary with impeller design, xmax is directly proportional to kLa.
Therefore, using the results from (a), if xmax = 15 g l–1 with the Rushton turbine, for the curved-blade disc
turbine:
0.107 s −1
xmax = (15 g l−1 ) = 20.8 g l−1
0.077 s −1
Answer: 20.8 g l–1

(d)
kLa depends on the gas flow rate through both uG and PT. PT can be expressed in terms of uG using Eq.
(8.13) and the definition of superficial gas velocity (Section 10.9). If P is the power input by the impeller:
PT = P + Pv = P + Fg ρ gH L = P + uG Aρ gH L

Assuming that the fractional impeller power loss with gassing is independent of uG, we can substitute
values into this equation for the Rushton turbine:
1W
PT = 4500 W + uG (4.35 m 2 ) (1000 kg m −3 ) (9.8066 m s −2 ) (2.3 m) .
1 kg m 2 s −3
= (4500 + 9.812 × 104 uG ) W

where uG has units m s–1. From Eq. (10.42) and the results from (a), to achieve the same xmax using the
Rushton turbine as for the curved-blade disc turbine, kLa must be equal to 0.107 s–1. Using the above
expression for PT together with values for kLa and VL in the appropriate units, the expression for kLa
becomes:
0.7
 4500 + 9.812 × 104 uG 
0.107 = 2.5 × 10−3  0.3
 uG
 10 
This equation can be solved using iterative methods. Starting with an estimate for uG of 0.06 m s–1, the
value of the right side of the equation is:
0.7
 4500 + 9.812 × 104 (0.06) 
−3 0.3
2.5 × 10   0.06 = 0.139
 10 
As this value is greater than the left side of the equation (0.107), the estimated value of uG was too high.
Using uG = 0.04 m s–1, the value of the right side of the equation is:
0.7
 4500 + 9.812 × 104 (0.04) 
−3 0.3
2.5 × 10   0.04 = 0.106
 10 
This is close to but slightly less than 0.107, indicating that uG should be increased a little. Using uG =
0.042 m s–1, the value of the right side of the equation is:
0.7
 4500 + 9.812 × 104 (0.042) 
−3 0.3
2.5 × 10   0.042 = 0.110
 10 
This result indicates that 0.042 m s–1 is too high. Using uG = 0.0405 m s–1, the value of the right side of the
equation is:
0.7
 4500 + 9.812 × 104 (0.0405) 
2.5 × 10−3  0.3
 0.0405 = 0.107
 10 

4
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

Therefore, a kLa value of 0.107 s–1 is achieved using the Rushton turbine by operating the fermenter with
uG = 0.0405 m s–1. Converting this result to a volumetric gas flow rate using the definition of superficial
gas velocity (Section 10.9):
60 s
Fg = uG A = 0.0405 m s −1 (4.35 m 2 ) . = 10.57 m3 min −1
1 min

As VL = 10 m3, this gas flow rate can be expressed as (10.57 m3 min–1)/10 m3 = 1.06 vvm.
Answer: 1.06 vvm, assuming that the fractional power loss with gassing does not vary with gas flow rate

10.5 Foam control and oxygen transfer

DT = 1.5 m. Calculating the cross-sectional area of the fermenter:
π π
A= DT2 = (1.5 m)2 = 1.767 m 2
4 4
qO = 2.6 mmol g–1 h–1. Converting to units of g g–1 s–1 using the molecular weight of oxygen = 32.0 (Table
C.1, Appendix C):
1 gmol 32.0 g 1h
qO = 2.6 mmol g −1 h −1 . . . = 2.31 × 10−5 g g −1 s −1
1000 mmol 1 gmol 3600 s

s0 = 20 g l–1; YXS = 0.32 g g–1; YPX = 0.055 g g–1; CAL

*
= 7.8 g m–3. Converting the units of CAL
*
using 1 m3
3
= 10 l (Table A.2, Appendix A):
1 m3
*
CAL = 7.8 g m −3 . 3
= 7.8 × 10−3 g l−1
10 l

(a)
For the different liquid heights in the table, the maximum cell concentration is estimated using Eq.
(10.42). For example, at HL = 1.10 m, kLa = 0.016 s–1 and:
0.016 s −1 (7.8 × 10−3 g l−1 )
xmax = = 5.40 g l−1
2.31 × 10−5 g g −1 s −1
*
Assuming that antifoam addition does not affect qO or CAL , the results for all operating liquid heights are
listed below.
HL (m) xmax (g l–1)
1.10 5.40
1.29 4.39
1.37 4.05
1.52 4.05
1.64 3.17

(b)
The liquid volume in the fermenter VL is equal to AHL. Therefore, for HL = 1.10 m, VL = 1.94 m3.
Multiplying xmax by VL gives the maximum mass of cells produced, Xmax. Using the result in (a) for HL =
1.10 m:

5
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

103 l 1 kg
X max = 5.40 g l−1 (1.94 m3 ) . . = 10.5 kg
1 m3 1000 g

As 0.055 g of protease is produced per g of biomass formed, for HL = 1.10 m, the maximum mass of
protease produced Pmax is:
Pmax = YPX X max = 0.055 g g −1 (10.5 kg) = 0.58 kg
Repeating these calculations gives results for VL, Xmax and Pmax for all operating liquid heights.
HL (m) VL (m3) Xmax (kg) Pmax (kg)
1.10 1.94 10.5 0.58
1.29 2.28 10.0 0.55
1.37 2.42 9.81 0.54
1.52 2.69 10.9 0.60
1.64 2.90 9.19 0.51
The values of Xmax and Pmax vary depending on the balance of effects from the increase in liquid volume
and decrease in kLa as HL increases and more antifoam is required.

(c)
The values of Xmax and Pmax evaluated in (b) represent the maximum masses of cells and protease that can
be supported by oxygen transfer in the fermenter. These results can be compared with the mass of cells
′ and the mass of protease Pmax
X max ′ produced by complete conversion of substrate in the absence of mass
transfer limitations:
′ = s0 VL YXS
X max
and
′ = YPX X max
Pmax ′

As an example, for HL = 1.10 m:

103 l 1 kg
′ = 20 g l−1 (1.94 m3 ) (0.32 g g −1 ) .
X max 3
. = 12.4 kg
1 m 1000 g

and
′ = 0.055 g g −1 (12.4 kg) = 0.68 kg
′ = YPX X max
Pmax
Because VL increases as HL increases, X max′ and Pmax′ at the other liquid heights must be greater than at HL
= 1.10 m. As all the Pmax values in (b) are smaller than the corresponding Pmax ′ values that would be
achieved in the absence of mass transfer limitations, we can conclude that protease production is limited
by oxygen transfer at all the liquid heights tested.
Answer: Yes, at all the liquid heights tested

(d)
From the table in (b), Xmax and Pmax achieve maximum values at HL = 1.52 m. Therefore, operation at this
liquid height is recommended. In drawing this conclusion, we assume that the higher antifoam
consumption and increased antifoam cost for operation at this liquid height is off-set by the greater mass
of protease produced compared with operation at lower heights.
Answer: 1.52 m

6
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

10.6 Improving the rate of oxygen transfer

VL = 17 m3; CAL
*
= 10.7 g m–3. Converting the units of CAL
*
to kg m–3:
1 kg
*
CAL = 10.7 g m −3 . = 1.07 × 10−2 kg m −3
1000 g

(a)
From the time-course data, before addition of the vegetable oil, kLa = 250 h–1 and CAL = 77% air
saturation. The rate of oxygen transfer is determined using Eq. (10.39):
N A = 250 h −1 (1.07 × 10−2 − 0.77 × 1.07 × 10−2 ) kg m −3 = 0.615 kg m −3 h −1
After addition of the vegetable oil, the steady-state value of kLa = 100 h–1 and CAL = 55% air saturation.
The rate of oxygen transfer is:
N A = 100 h −1 (1.07 × 10−2 − 0.55 × 1.07 × 10−2 ) kg m −3 = 0.482 kg m −3 h −1
Answer: 0.62 kg m–3 h–1 before oil addition and 0.48 kg m–3 h–1 after oil addition

(b)
From the equation for kLa as a function of power input, if uG is unchanged:
0.5
(kL a )1  ( PT )1 
= 
(kL a ) 2  ( PT ) 2 

where subscript 1 refers to the conditions at kLa = 100 h–1 and subscript 2 refers to the conditions required
to achieve kLa = 250 h–1. Manipulating this equation gives:
2
 (kL a )1  ( PT )1
  =
 (kL a ) 2  ( PT )2
or
2
 (k a) 
( PT )2 = ( PT )1  L 2 
 (kL a )1 
As (kLa)2/(kLa)1 = 250/100 = 2.5:
( PT )2 = ( PT )1 (2.5)2 = 6.25 ( PT )1
Therefore, if the contribution to PT from gas sparging is negligible, the stirrer power must be increased by
a factor of 6.25.
Answer: By a factor of 6.25

(c)
*
The use of oxygen-enriched air increases the value of CAL but does not affect kLa. From (a), the rate of
–3 –1 *
oxygen transfer before oil addition is 0.615 kg m h . The value of CAL needed to produce this transfer
–1 –3 –3
rate when kLa = 100 h and CAL = 6.2 × 10 kg m is determined from Eq. (10.39):
NA 0.615 kg m −3 h −1
*
CAL = + CAL = + 6.2 × 10−3 kg m −3 = 1.24 × 10−2 kg m −3
kL a 100 h −1
*
This value of CAL can be compared with 1.07 × 10–2 kg m–3 calculated above for air containing 0.2099
mole fraction oxygen (Section 2.4.5). The mole fraction of oxygen yAG in the gas phase corresponding to

7
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________
*
CAL = 1.24 × 10–2 kg m–3 is evaluated using Henry’s law and Eq. (10.46). When the total pressure pT is
unchanging:
*
CAL2 1.24 × 10−2 kg m −3
yAG2 = y AG1 = (0.2099) = 0.243
*
CAL1 1.07 × 10−2 kg m −3
For gases at relatively low pressure, volume % = mole % (Section 2.4.5). Therefore, oxygen-enriched air
containing 24.3 volume % oxygen is required to restore the rate of oxygen transfer.
Answer: 24.3 vol%

10.7 Oxygen transfer for different cell types

(a)
*
CAL = 7.2 × 10–3 kg m–3. Converting the units of CAL
*
to mmol l–1 using the molecular weight of oxygen =
32.0 (Table C.1, Appendix C) and 1 m3 = 103 l (Table A.2, Appendix A):
1000 g 1 gmol 1000 mmol 1 m3
*
CAL = 7.2 × 10−3 kg m −3 . . . . 3 = 0.225 mmol l−1
1 kg 32 g 1 gmol 10 l

The kLa required to maintain the dissolved oxygen concentration at critical level is determined for each
cell type using Eq. (10.43). For E. coli:
8.5 mmol g −1 h −1 (25 g l−1 )
(kL a )crit = = 980 h −1
(0.225 − 0.0082) mmol l−1
For grape cells:
0.60 mmol g −1 h −1 (25 g l−1 )
(kL a )crit = −1
= 88 h −1
(0.225 − 0.055) mmol l
For CHO cells:
3.0 × 10−10 mmol cell−1 h −1 (3.0 × 109 cell l−1 )
(kL a )crit = = 4.4 h −1
(0.225 − 0.020) mmol l−1
Answer: For E. coli 980 h–1, for grape cells 88 h–1, for CHO cells 4.4 h–1. This problem highlights the very
high demands on oxygen transfer inherent in microbial cell culture relative to plant and animal cell
systems.

(b)
From the equation provided, when VL remains constant:
2 2
( PT )grape  (kL a )grape   88 h −1 
=  = −1 
= 400
( PT )CHO  (kL a )CHO   4.4 h 
Therefore, the power required to maintain critical oxygen tension in the grape cell culture is 400 times that
required for CHO cells. Similarly:
2 2
( PT ) E . coli  (kL a ) E . coli   980 h −1 
=  = −1 
= 4.96 × 104
( PT )CHO  (kL a )CHO   4.4 h 

The power required to maintain critical oxygen tension in the E. coli culture is 5 × 104 times that required
for CHO cells.

8
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

Answer: The power required for culture of grape cells is 400 times that required for culture of CHO cells;
the power required for culture of E. coli is 5.0 × 104 times greater than that required for CHO cells

10.8 Single-point kLa determination using the oxygen balance method

(a)
VL = 200 l. The oxygen transfer rate for kLa determination by the oxygen balance method is calculated
using Eq. (10.53). Air contains 21 mole % oxygen (Section 2.4.5); therefore, from Henry’s law Eq.
(10.45), the partial pressure pAG of oxygen in the inlet air at 1 atm is (0.21 × 1 atm) = 0.21 atm. Assuming
that the exit gas leaves the fermenter at the fermentation conditions (1 atm pressure and 28°C), the partial
pressure of oxygen in the exit gas is (0.201 × 1 atm) = 0.201 atm. The inlet air flow rate Fg is 1 vvm or
200 l min–1. Using R = 0.000082057 m3 atm K–1 gmol–1 (Table B.1, Appendix B) and converting the
temperatures to degrees Kelvin using Eq. (2.27), Eq. (10.53) becomes:
1
NA =
0.000082057 m atm K −1 gmol−1 (200 l)
3

 −1 1 min
  1 min 
 200 l min . . 0.21 atm   189 l min −1 . . 0.201 atm  
 60 s − 60 s 
 (20 + 273.15) K   (28 + 273.15) K 
    
    
−3 −1
= 0.0174 gmol m s
Answer: 0.0174 gmol m–3 s–1

(b)
Using Eq. (10.39) with the units of NA from (a) converted to mass using the molecular weight of oxygen =
32.0 (Table C.1, Appendix C):
32.0 g 1 kg
0.0174 gmol m −3 s −1 . .
N 1 gmol 1000 g
kL a = * A = = 0.15 s −1
(CAL − CAL ) (7.8 × 10 − 0.52 × 7.8 × 10−3 ) kg m −3
−3

Answer: 0.15 s–1

(c)
If the measured exit gas composition of 20.1% O2 is an overestimation, the actual value is 20.1% × 1/1.1 =
18.3% O2. Therefore, the partial pressure of oxygen in the exit gas is (0.183 × 1 atm) = 0.183 atm. From
Eq. (10.53):
1
NA =
0.000082057 m atm K −1 gmol−1 (200 l)
3

 −1 1 min
  1 min 
 200 l min . . 0.21 atm   189 l min −1 . . 0.183 atm  
 60 s − 60 s 
 (20 + 273.15) K   (28 + 273.15) K 
    
    
−3 −1
= 0.0289 gmol m s
Therefore, from Eq. (10.39):

9
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

32.0 g 1 kg
0.0289 gmol m −3 s −1 . .
N 1 gmol 1000 g
kL a = * A = = 0.25 s −1
(CAL − CAL ) (7.8 × 10 − 0.52 × 7.8 × 10−3 ) kg m −3
−3

The kLa value obtained in (b) using the incorrectly calibrated oxygen analyser is 60% of the actual kLa
value; the error is therefore 40%.
Answer: 40%. This calculation illustrates the sensitivity of the oxygen balance method to the accuracy of
the measured parameters used in Eq. (10.53). This sensitivity arises from the subtraction of two numbers
of similar magnitude for the moles of oxygen in and out of the system. When errors in both Fg terms are
taken into account, the error in the final kLa value can be very large.

10.9 Steady-state kLa measurement

(a)
The effect of medium components on the solubility of oxygen is evaluated using the methods of Section
10.8.3. From Tables C.8 and C.1 (Appendix C), the molecular formulae for glucose and sucrose and the
molecular weights of the medium components are: glucose (C6H12O6) = 180.2, sucrose (C12H22O11) =
342.3, CaCO3 = 100.1, (NH4)2SO4 = 132.1, Na2HPO4 = 142.0 and KH2PO4 = 136.1.
The parameter values for application in Eq. (10.49) are listed below. Values of Hi and Kj are taken
from Table 10.5.
Medium component Hi or Kj (m3 mol–1) zi CiL or CjL (mol m–3)
1 mol 1000 l
Glucose 0.119 × 10–3 – 20 g l−1 . . = 111
180.2 g 1 m 3
1 mol 1000 l
Sucrose 0.149 × 10–3 – 8.5 g l−1 . . = 24.8
342.3 g 1 m 3
1 mol 1000 l
Ca2+ –0.303 × 10–3 2 1.3 g l−1 . . = 13.0
100.1 g 1 m 3
1 mol 1000 l
CO32– 0.485 × 10–3 2 1.3 g l−1 . . = 13.0
100.1 g 1 m 3
1 mol 1000 l
NH4+ –0.720 × 10–3 1 2 × 1.3 g l−1 . . = 19.7
132.1 g 1 m 3
1 mol 1000 l
SO42– 0.453 × 10–3 2 1.3 g l−1 . . = 9.8
132.1 g 1 m3
1 mol 1000 l
Na+ –0.550 × 10–3 1 2 × 0.09 g l−1 . . = 1.3
142.0 g 1 m3
1 mol 1000 l
HPO42– 0.485 × 10–3 2 0.09 g l−1 . . = 0.63
142.0 g 1 m 3
1 mol 1000 l
K+ –0.596 × 10–3 1 0.12 g l−1 . . = 0.88
136.1 g 1 m 3
1 mol 1000 l
H2PO4– 1.037 × 10–3 1 0.12 g l−1 . . = 0.88
136.1 g 1 m 3

Substituting these values into Eq. (10.49) gives:

10
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

 (−0.303 × 10− 3 ) (2) 2 13.0 + (0.485 × 10−3 ) (2) 2 13.0 + ( − 0.720 × 10−3 ) (1) 2 19.7 + 
C *
AL0
  −3 2 −3 2 −3 2

log10  *  = 0.5  (0.453 × 10 ) (2) 9.8 + (−0.550 × 10 ) (1) 1.3 + (0.485 × 10 ) (2) 0.63 + 
C AL   −3 2 −3 2 
 (−0.596 × 10 ) (1) 0.88 + (l.037 × 10 ) (1) 0.88 
(
+ (0.119 × 10−3 ) 111 + (0.149 × 10−3 ) 24.8 )
= 2.39 × 10−2
Therefore:
*
CAL0 2.39×10−2
*
= 10 = 1.06
CAL
or
*
CAL = 0.95 CAL0
*

This result indicates that solutes in the medium reduce the oxygen solubility by about 5% compared with
the oxygen solubility at zero solute concentration. From Table 10.2, the solubility of oxygen in water at
35°C and 1 atm air pressure is 7.52 × 10–3 kg m–3. However, if the gas phase in the fermenter is well
mixed, the mole fraction of oxygen in the bubbles is equal to that in the off-gas, 0.197, which is less than
that in air (0.2099, Section 2.4.5). Therefore, applying Henry’s law Eq. (10.46):
0.197
*
CAL0 = (7.52 × 10−3 ) kg m −3 = 7.06 × 10−3 kg m −3
0.2099
*
so that CAL in the fermentation medium at 1 atm and 35°C is 0.95 × (7.06 × 10–3 kg m–3) = 6.70 × 10–3 kg
–3
m .
Answer: 6.70 × 10–3 kg m–3, assuming that the gas phase in the fermenter is well mixed, and that medium
solutes but not other broth components such as cells affect oxygen solubility

(b)
The oxygen transfer rate is determined using Eq. (10.53). The mole fraction of oxygen in the incoming air
is 0.2099 (Section 2.4.5); therefore, from Henry’s law Eq. (10.45), the partial pressure of oxygen in the
inlet air at 1 atm is 0.2099 × 1 atm = 0.2099 atm. The mole fraction of oxygen in the off-gas is 0.197;
therefore, from Henry’s law Eq. (10.45), at an off-gas pressure of 1 atm, the partial pressure of oxygen in
the off-gas is 0.197 × 1 atm = 0.197 atm. Using R = 0.082057 l atm K–1 gmol–1 (Table B.1, Appendix B)
and converting the temperatures to degrees Kelvin using Eq. (2.27), Eq. (10.53) becomes:
1
NA =
0.082057 l atm K −1 gmol−1 (500 l)
 3 −1 10 l
3
1 min  
 0.4 m min . . . 0.2099 atm  
 −  6.3 l s (0.197 atm)  
3 −1
 1m 60 s
 (25 + 273.15) K   (35 + 273.15) K  
  
  
= 1.623 × 10−5 gmol l−1 s −1
Answer: 1.62 × 10–5 gmol l–1 s–1

(c)
kLa is determined from NA using Eq. (10.39). CAL = 45% air saturation; this oxygen tension must be
converted to a dissolved oxygen concentration. From Table 10.2, the solubility of oxygen in water at 35°C
and 1 atm air pressure is 7.52 × 10–3 kg m–3. Using the result in (a) for the effect of medium solutes, air

11
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

saturation in fermentation medium corresponds to 0.95 × (7.52 × 10–3 kg m–3) = 7.14 × 10–3 kg m–3;
therefore, CAL = 0.45 × (7.14 × 10–3 kg m–3) = 3.21 × 10–3 kg m–3. Converting the units of NA in (b) to
mass using the molecular weight of oxygen = 32.0 (Table C.1, Appendix C) and applying the value of
*
CAL from (a), Eq. (10.39) becomes:
32.0 g 1 kg
1.623 × 10−5 gmol l−1 s −1 . .
NA 1 gmol 1000 g
kL a = = 3
= 0.15 s −1
(C − CAL )
*
1m
AL
(6.70 − 3.21) × 10−3 kg m −3 .
1000 l

Answer: 0.15 s–1

(d)
qO = 5.4 mmol g–1 h–1. The maximum cell concentration supported by oxygen transfer is evaluated using
*
Eq. (10.42). CAL in this equation depends on the medium composition and the gas-phase oxygen partial
pressure. We will assume that the oxygen partial pressure in the bubbles is 0.197, although this may vary
*
with cell concentration, so that the value of CAL is that determined in (a). Using this result together with
the value of kLa from (c) and converting qO to mass units using the molecular weight of oxygen = 32.0
(Table C.1, Appendix C):
0.15 s −1 (6.70 × 10−3 kg m −3 )
xmax = = 20.9 g l−1
1 gmol 32.0 g 1 h
5.4 mmol g −1 h −1 . . .
1000 mmol 1 gmol 3600 s
Answer: 20.9 g l–1

(e)
YXS = 0.5 g g–1. For complete conversion of 20 g l–1 glucose and 8.5 g l–1 sucrose:
x = 0.5 g g −1 (20 + 8.5) g l−1 = 14.3 g l−1
As this value of x is less than the maximum cell concentration supported by oxygen transfer evaluated in
(d), growth is limited by substrate availability, not oxygen transfer.
Answer: Substrate availability

10.10 Oxygen transfer in a pressure vessel

(a)
DT = 3.6 m; HL = 6.1 m. The volume of liquid VL in a cylindrical vessel of these dimensions is:
π π
VL = DT2 H L = (3.6 m)2 (6.1 m) = 62.09 m3
4 4
The oxygen transfer rate is determined using Eq. (10.53). The inlet air contains 20.99 mole % oxygen
(Section 2.4.5); therefore, from Henry’s law Eq. (10.45), the partial pressure of oxygen in the inlet air
measured at 1 atm is (0.2099 × 1 atm) = 0.2099 atm. The mole fraction of oxygen in the off-gas leaving
from the top of the vessel is 0.172 at a pressure of 1.4 atm; therefore the partial pressure of oxygen in the
off-gas is (0.172 × 1.4 atm) = 0.241 atm. Using R = 0.000082057 m3 atm K–1 gmol–1 (Table B.1, Appendix
B) and converting the temperatures to degrees Kelvin using Eq. (2.27), Eq. (10.53) becomes:

12
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

1
NA =
0.000082057 m atm K −1 gmol−1 (62.09 m3 )
3

 −1 1 min
  1 min 
3
 30 m min . . 0.2099 atm   20.5 m 3 min −1 . . 0.241 atm  
 60 s − 60 s 
 (20 + 273.15) K   (29 + 273.15) K 
    
    
= 0.0168 gmol m −3 s −1
Answer: 0.0168 gmol m–3 s–1

(b)
The hydrostatic pressure at the bottom of the tank due to the weight of liquid is calculated using Eq.
(10.58) with g = 9.8066 m s–2 (Section 2.3):
ps = 1000 kg m −3 (9.8066 m s −2 )(6.1 m) = 5.98 × 104 kg m −1 s −2
Converting units using 1 atm = 1.013 × 105 kg m–1 s–2 (Table A.5, Appendix A):
1 atm
ps = 5.98 × 104 kg m −1 s −2 . = 0.590 atm
1.013 × 105 kg m −1 s −2

The pressure at the bottom of the tank is equal to the pressure at the top + ps:
Pressure at the bottom of the tank = 1.4 atm + 0.590 atm = 1.99 atm
Answer: 1.99 atm

(c)
From Table 10.2, the solubility of oxygen in water at 29°C and 1 atm air pressure is 8.17 × 10–3 kg m–3. At
the bottom of the tank, the composition of the gas phase is the same as air but, from (b), the pressure is
*
1.99 atm rather than 1 atm. Using Henry’s law Eq. (10.46) to evaluate CAL at the bottom of the tank:
pT2 yAG2 pT2
*
CAL2 = CAL1
*
= CAL1
*
pT1 yAG1 pT1
where subscript 1 refers to air at 1 atm and subscript 2 refers to air at 1.99 atm. Substituting values gives:
 1.99 atm 
*
CAL2 = 8.17 × 10−3 kg m −3  −2
 = 1.63 × 10 kg m
−3

 1 atm 
At the top of the tank, the mole fraction of oxygen in the gas phase is 0.172 and the pressure is 1.4 atm.
*
Using Eq. (10.46) to evaluate CAL at the top of the tank:
pT2 yAG2
*
CAL2 = CAL1
*
pT1 yAG1
where subscript 1 refers to air at 1 atm and subscript 2 refers to off-gas at 1.4 atm. Substituting values
gives:
 1.4 atm × 0.172 
*
CAL2 = 8.17 × 10−3 kg m −3  −3
 = 9.37 × 10 kg m
−3

 1 atm × 0.2099 
Answer: 9.37 × 10–3 kg m–3 at the top of the tank and 1.63 × 10–2 kg m–3 at the bottom

13
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

(d)
kLa is determined using Eq. (10.39) modified for large vessels, where the logarithmic-mean concentration
*
difference given by Eq. (10.59) is a better representation than ( CAL – CAL) of the concentration-difference
driving force for oxygen transfer.
At the top of the fermenter, CAL = 50% air saturation. The measuring probe is calibrated in situ;
therefore, air saturation at this location corresponds to the solubility of oxygen in water at 29°C under air
at 1.4 atm pressure. From Table 10.2, the solubility of oxygen in water at 29°C under 1 atm air pressure is
8.17 × 10–3 kg m–3; therefore, using Eq. (10.46), air saturation at the top of the fermenter means an oxygen
concentration of (1.4 atm/1 atm) × 8.15 × 10–3 kg m–3 = 1.14 × 10–2 kg m–3. If the dissolved oxygen
tension is 50% air saturation, CAL at the gas outlet = 0.5 × (1.14 × 10–2 kg m–3) = 5.71 × 10–3 kg m–3.
Similarly, at the bottom of the fermenter, CAL = 65% air saturation. Air saturation at this location
corresponds to the solubility of oxygen in water at 29°C under air at 1.99 atm pressure. Therefore, air
saturation at the bottom of the fermenter means an oxygen concentration of (1.99 atm/1 atm) × 8.15 × 10–3
kg m–3 = 1.62 × 10–2 kg m–3. If the dissolved oxygen tension is 65% air saturation, CAL at the gas inlet =
0.65 × (1.62 × 10–2 kg m–3) = 1.05 × 10–2 kg m–3.
*
Substituting these values for CAL and the results for CAL from (c) into Eq. (10.59):
(9.37 × 10−3 − 5.71 × 10−3 ) kg m −3 − (1.63 × 10−2 − 1.05 × 10−2 ) kg m −3
*
(CAL − CAL )lm =
 (9.37 × 10−3 − 5.71 × 10−3 ) kg m −3 
ln  −2 −2 −3 
 (1.63 × 10 − 1.05 × 10 ) kg m 
= 4.65 × 10−3 kg m −3
Converting to mole units using the molecular weight of oxygen = 32.0 (Table C.1, Appendix C):
1 gmol 1000 g
*
(CAL − CAL )lm = 4.65 × 10−3 kg m −3 . . = 0.145 gmol m −3
32 g 1 kg

Calculating kLa using Eq. (10.39) with the logarithmic-mean concentration difference and NA from (a):
NA 0.0168 gmol m −3 s −1
kL a = = = 0.116 s −1
*
(CAL − CAL )lm 0.145 gmol m −3
Answer: 0.12 s–1

(e)
From Section (10.5.2), at steady state, the rate of oxygen transfer NA is equal to the rate of oxygen uptake
by the cells QO. Using the definition of QO in Eq. (10.40) and converting units using the molecular weight
of oxygen = 32.0 (Table C.1, Appendix C) and 1 m3 = 103 l (Table A.2, Appendix A):
QO N A 0.0168 gmol m −3 s −1
qO = = = 3
= 3.36 × 10−5 s −1
x x 10 l 1 gmol
16 g l−1 . .
1 m3 32.0 g

Answer: 3.4 × 10–5 s–1

(f)
Because of the effect of the hydrostatic pressure ps, the maximum pressure of 2.7 atm applies to the
bottom of the tank. Therefore, using the result for ps from (b), the pressure at the top of the tank is (2.7–
0.590) atm = 2.11 atm.
*
CAL depends on the gas-phase oxygen partial pressure. To determine a maximum value for x, we
will assume that the mole fraction of oxygen in the bubbles everywhere in the tank is the same as that in

14
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

air, 0.2099 (Section 2.4.5), although this may vary with cell concentration. Using the solubility of oxygen
in water at 29°C and 1 atm air pressure as a basis (Table 10.2), applying Henry’s law and Eq. (10.46)
gives:
 2.7 atm 
*
CAL at the bottom of the tank = 8.17 × 10−3 kg m −3  −2
 = 2.21 × 10 kg m
−3

 1 atm 

 2.11 atm 
*
CAL at the top of the tank = 8.17 × 10−3 kg m −3  −2
 = 1.72 × 10 kg m
−3

 1 atm 
As explained in Section 10.5.2, the maximum cell concentration occurs when CAL = 0. xmax can be
determined using Eq. (10.41) with CAL = 0; however, for large vessels, the logarithmic-mean
* *
concentration difference ( CAL – CAL)lm is more appropriate in Eq. (10.41) than ( CAL – CAL). Evaluating
*
( CAL – CAL)lm from Eq. (10.59) with CAL = 0:
*
(CAL )o − (CAL
*
)i (1.72 × 10−2 − 2.21 × 10−2 ) kg m −3
*
(CAL − CAL )lm = = = 1.95 × 10−2 kg m −3
*
(CAL )o  1.72 × 10−2 kg m −3 
ln * ln  −2 −3 
(CAL )i  2.21 × 10 kg m 
Using this result in Eq. (10.41) together with the values of kLa and qO from (d) and (e):
1 m3 1000 g
0.116 s −1 (1.95 × 10−2 kg m −3 ) . .
103 l 1 kg
xmax = = 67.3 g l−1
3.36 × 10−5 s −1
Answer: 67 g l–1

10.11 Dynamic kLa measurement

(a)
kLa is estimated using Eq. (10.57) and the methods described in Section 10.10.2. Let t1 = 10 s and CAL1 =
43.5% air saturation. From the measured data, the steady-state dissolved oxygen tension CAL = 73.5% air
saturation. Calculated values of
 C − CAL1 
ln  AL 
 CAL − CAL2 
and corresponding values of (t2 – t1) are listed and plotted below.
 C − CAL1 
ln  AL  (t2 – t1) (s)
 CAL − CAL2 
0 0
0.41 5
0.80 10
1.6 20
2.3 30
3.0 40
4.1 60

15
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

 CAL - CAL1  4

 CAL - CAL2 
3

2
ln 

0
0 10 20 30 40 50 60 70

(t2 - t1) s

From Eq. (10.57), kLa is equal to the slope of the straight line in the plot through the origin = 0.072 s–1.
Answer: 0.072 s–1

(b)
The simple dynamic method cannot be considered to give reliable results for kLa unless further
experimental checks are performed to show that the electrode, boundary layers and gas-phase dynamics do
not influence the measurement system. The effect of the electrode response time and boundary layers
should be determined at a range of stirrer speeds using the techniques described in Section 10.10.2
(Electrode Response Time and Liquid Boundary Layers subsection). The effect of gas-phase dynamics
should be tested using different methods of deoxygenation as described in Section 10.10.2 (Gas-Phase
Dynamics subsection).

10.12 kLa measurement using the dynamic pressure method

(a)
kLa is estimated using Eq. (10.57) and the methods described in Section 10.10.2. Let t1 = 6 s and CAL1 =
50% air saturation. The steady-state dissolved oxygen tension CAL = 66% air saturation. Calculated values
of
 C − CAL1 
ln  AL 
 CAL − CAL2 
and corresponding values of (t2 – t1) are listed and plotted below.
 C − CAL1 
ln  AL  (t2 – t1) (s)
 CAL − CAL2 
0 0
0.480 4
1.674 19
2.510 34

16
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

3.0

2.5

2.0
 CAL - CAL1 

 CAL - CAL2 

1.5
ln 

1.0

0.5

0.0
0 10 20 30 40

(t2 - t1) s

From Eq. (10.57), kLa is equal to the slope of the straight line in the plot through the origin = 0.078 s–1.
Although the dynamic pressure method reduces the influence of gas-phase dynamics on kLa measurement
(Section 10.10.2, Modified Dynamic Methods subsection), before this result can be considered reliable the
effects of the electrode response time and boundary layers should be checked using the techniques
described in Section 10.10.2 (Electrode Response Time and Liquid Boundary Layers subsection).
Answer: 0.078 s–1

(b)
Repeating the calculation with t1 = 6 s, CAL1 = 50% air saturation and CAL = 70% air saturation:
 C − CAL1 
ln  AL  (t2 – t1) (s)
 CAL − CAL2 
0 0
0.364 4
1.050 19
1.328 34

17
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

1.6

1.4

1.2
 CAL - CAL1 

 CAL - CAL2 
1.0

0.8

0.6
ln 

0.4

0.2

0.0
0 10 20 30 40

(t2 - t1) s

kLa = the slope of the straight line in the plot through the origin = 0.043 s–1. Therefore, a 6% error in CAL
results in an error in kLa of:
0.078 − 0.043
× 100% = 45%
0.078
Answer: The measured kLa value is 0.043 s–1, representing an error of 45%

(c)
The electrode response time is the same at 50 rpm and 60 rpm. This suggests that boundary layer effects
were eliminated at 50 rpm. As the kLa measurements were conducted at 60 rpm, we can conclude that
significant boundary layers were not present during the measurement procedure.
In the absence of boundary layers, the time taken for the electrode to record 63.2% of the step
change from 0 to 100% air saturation is about 2.2 s; therefore, the electrode response time τE = 2.2 s
(Figure 10.17). This result can be compared with 1/kLa and 0.2/kLa to determine whether the electrode
influenced the measured value of kLa (Section 10.10.2, Electrode Response Time and Liquid Boundary
Layers subsection). Using the result for kLa from (a), 1/kLa = 1/(0.078 s–1) = 12.8 s, and 0.2/kLa = 2.6 s. As
τE < 0.2/kLa, we can conclude that the error in kLa due to the electrode response is very small.
The test experiments show that there was a negligible effect on kLa due to the electrode response
time and boundary layers. The influence of gas-phase dynamics was not tested; however, use of the
dynamic pressure method for measuring kLa avoids many of the problems with transient gas-phase
composition and the loss of gas hold-up that occur using the simple dynamic method.
Answer: We can have reasonable confidence in the kLa measurements, as the effects of the electrode
response time, boundary layers and gas-phase dynamics are considered negligible

10.13 Surface versus bubble aeration

(a)
From Section 10.5.2, at steady state, the rate of oxygen transfer NA is equal to the rate of oxygen uptake by
the cells QO; therefore, kLa can be evaluated using Eq. (10.41). CAL = 50% air saturation; this oxygen
tension must be converted to a dissolved oxygen concentration. Assuming that the solubility of oxygen in

18
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________

culture medium is the same as that in water, the dissolved oxygen concentration corresponding to air
saturation at 37°C and 1 atm can be interpolated from the values for 35°C and 40°C in Table 10.2:
(37 − 35) o C
*
CAL (air at 1 atm and 37 o C) = 7.52 × 10−3 kg m −3 − o
(7.52 − 7.07) × 10−3 kg m −3
(40 − 35) C
= 7.34 × 10−3 kg m −3
Therefore, CAL = 0.5 × (7.34 × 10–3 kg m–3) = 3.67 × 10–3 kg m–3. CAL*
in the bioreactor must be evaluated
for operation with a 50:20:30 mixture of air:oxygen:nitrogen at 1 atm. The oxygen mole fraction yAG in the
gas mixture is determined using 0.2099 for the mole fraction of oxygen in air (Section 2.4.5):
20 + 0.2099 (50)
yAG = = 0.305
100
If gas with 0.305 mole fraction oxygen is used instead of air, the solubility of oxygen in the bioreactor is
*
greater than the value for CAL under air. From Henry’s law and Eq. (10.46):
0.305
*
CAL (gas mixture at 1 atm and 37 o C) = (7.34 × 10−3 kg m −3 ) = 1.07 × 10−2 kg m −3
0.2099
Applying Eq. (10.41) with qO = 7.7 × 10–12 g cell–1 h–1, x = 1.1 × 109 cells l–1 and the above values for CAL
*
and CAL :
1000 l 1 kg
7.7 × 10−12 g cell−1 h −1 (1.1 × 109 cell l−1 ).
3
.
qO x 1m 1000 g
kL a = = −2 −3 −3
(C − CAL )
*
AL (1.07 × 10 − 3.67 × 10 ) kg m
= 1.20 h −1
Answer: 1.2 h–1

(b)
x = 3.9 × 109 cells l–1 when CAL = 8% air saturation = 0.08 × (7.34 × 10–3 kg m–3) = 5.87 × 10–4 kg m–3.
Recalculating kLa using Eq. (10.41) for air sparging:

1000 l 1 kg
7.7 × 10−12 g cell−1 h −1 (3.9 × 109 cell l−1 ).
3
.
1m 1000 g
kL a = −3 −4 −3
(7.34 × 10 − 5.87 × 10 ) kg m
= 4.45 h −1
Answer: 4.5 h–1

(c)
The kLa for surface aeration must be increased by a factor of 4.45/1.20 = 3.71 to match the kLa for bubble
aeration. In surface aeration, the term a in kLa represents the surface area of liquid at the liquid–gas
interface. Let us assume that a for a cylindrical vessel is the area of a circle with diameter equal to the tank
diameter DT:
π
a= DT2
4
This assumption is valid when the liquid is stationary; however, if the liquid is agitated, the area available
for oxygen transfer may be higher due to rippling and movement of the gas–liquid interface. Assuming
that the value of kL is unchanged, i.e. increasing the tank diameter does not affect the velocity of the liquid
at the surface, a must be increased by a factor of 3.71 to achieve the required improvement in kLa. This

19
 P.M. Doran – Bioprocess Engineering Principles – Solutions Manual
________________________________________________________________________________________________________

can be achieved by increasing DT by a factor of 3.71 = 1.93. Therefore, DT must be increased from 8.5
cm to 1.93 × 8.5 cm = 16.4 cm while keeping the liquid volume constant at 500 ml.
Answer: 16.4 cm, assuming that the surface liquid velocity and the effect of liquid movement on the gas–
liquid interfacial area are unaffected by the increase in tank diameter. A vessel diameter of 16.4 cm for a
liquid volume of 500 ml corresponds to a liquid height of only 2.4 cm. This calculation shows that
increasing DT is not a practical approach for obtaining surface aeration kLa values similar to those
achieved using air sparging.

10.14 Shake-flask aeration

(a)
By analogy with Eq. (10.7), the resistance to oxygen transfer is equal to 1/kLa. kLa for the dynamic method
is estimated using Eq. (10.57) with t1 = 5 s, CAL1 = 65% air saturation, t2 = 30 s, CAL2 = 75% air saturation
and CAL = 90% air saturation:
 90 − 65 
ln  
kL a = 
90 − 75 
= 0.0204 s −1
(30 − 5) s
Therefore, 1/kLa = 48.9 s. The assumptions involved in the simple dynamic method apply as described in
Section 10.10.2. The calculated value for kLa is not reliable unless electrode response and liquid boundary
layer effects can be shown to be negligible. Because bubbles and gas hold-up are not involved in shake-
flask aeration, gas-phase dynamics is not a significant issue in this case.
Answer: 49 s, assuming that electrode and boundary layer effects are negligible

(b)
The closures have cylindrical geometry. The cross-sectional area of the closure Ac is equal to:
π
Ac = Dc2
4
where Dc is the diameter of the closure. As Dc is equal to the width of the flask opening:
π
Ac = (3.2 cm) 2 = 8.04 cm 2
4
The volume of gas in the flask VG can be estimated as the overall flask volume minus the liquid volume:
VG = (300 − 100) cm3 = 200 cm3
Substituting values into the equation for Kc:
20.8 cm 2 s −1 (8.04 cm 2 )
Kc = = 0.209 s −1
4 cm (200 cm3 )
The resistance due to the flask closure is equal to 1/ Kc = 1/(0.209 s–1) = 4.78 s.
Answer: 4.8 s

(c)
The total resistance to mass transfer is the sum of the resistance due to the flask closure and the resistance
due to the liquid boundary layer = (4.78 + 48.9) s = 53.68 s. Therefore, the proportion of the total
resistance due to the flask closure is (4.78 s)/(53.68 s) × 100% = 8.9%.
Answer: 8.9%

20
 Chapter 10 – Mass Transfer
________________________________________________________________________________________________________
(d)
kLa must be increased by a factor of 0.209/0.0204 = 10.25 for the resistance within the flask to be equal to
that of the flask closure. From the equation for kLa as a function of shake-flask operating parameters,
increasing the shaker speed from 80 rpm to 150 rpm increases kLa by a factor of:
1.2
 150 rpm 
  = 2.13
 80 rpm 
As this is less than the 10.25-fold increase required, additional changes must be made to the shake-flask
system to obtain a further 10.25/2.13 = 4.8-fold increase in kLa.

(i)
From the equation for kLa as a function of system parameters, for the required additional improvement in
kLa to be achieved:
0.85
 VF 
 
 VL 
must be increased by a factor of 4.8. Therefore, VF/VL must be increased by a factor of (4.8)1/0.85 = 6.33. If
the liquid volume VL remains constant, VF must be increased 6.33-fold, from 300 ml to 6.33 × 300 ml =
1899 ml = 1.9 l.
Answer: 1.9 l

(ii)
From the calculation in (i), if VF/VL must be increased by a factor of 6.33 while keeping VF constant, VL
must be reduced 6.33-fold, from 100 ml to (100 ml)/6.33 = 15.8 ml.
Answer: 16 ml

21
Chapter 11
Unit Operations

11.1 Overall product recovery

The overall product recovery is obtained by multiplying together the fractional recoveries for each
downstream processing step:
Overall product recovery = (0.89) (0.74) (0.71) (0.82) (0.68) = 0.26
Answer: 0.26

11.2 Product yield from transgenic goats’ milk

(a)
The overall product recovery is obtained by multiplying together the fractional recoveries for each
processing step:
Overall product recovery = (0.75) (0.60) (0.85) (0.85) = 0.325
Answer: 0.33

(b)
Before product recovery, the mass of anti-thrombin III produced per week is (20,000 l) × (10 g l–1) =
200,000 g = 200 kg. Applying the result for the overall fractional product recovery from (a), the mass of
purified anti-thrombin III recovered per week is 0.325 × 200 kg = 65 kg.
Answer: 65 kg

(c)
To produce 80 kg of purified anti-thrombin III per week, the overall fractional product recovery required
is (80 kg)/(200 kg) = 0.40. If the increase in overall recovery is achieved by improving the salt
precipitation and filtration step only, the product yield for this step must be:
0.40
Recovery for salt precipitation and filtration step = = 0.738
(0.75) (0.85) (0.85)

Therefore, the percentage increase in product yield required for the salt precipitation and filtration step is:
0.738 − 0.60
× 100% = 23%
0.60
Answer: 23%

11.3 Laboratory algal filtration

The filter area A is the area of a circle of diameter 6 cm = 6 × 10–2 m:
2 2
D  6 × 10−2 m  −3
A = π R2 = π   = π   = 2.83 × 10 m
2

2  2 
∆p = 5 psi; therefore, from Table A.5 (Appendix A):
6.895 × 103 kg m −1 s −2
∆p = 5 psi . = 3.45 × 104 kg m −1 s −2
1 psi

266