Advances in Biochemical Engineering, Volume 13

216 Pages·1979·4.07 MB·English

by  Werner A. Hampel (auth.), Prof. Dr. T. K. Ghose, Prof. Dr. N. Blakebrough, Professor Dr. A. Fiechter (eds.)

ADVANCES IN
BIOCHEMICAL
ENGINEERING
Volume 13
Editors: T. K. Ghose, A. Fiechter,
N. Blakebrough
Managing' Editor: A. Fiechter
With 134 Figures
Springer-Verlag
Berlin Heidelberg New York 1979
ISBN 3-540-09468-7 Springer-Verlag Berlin Heidelberg New York
ISBN 0-387-09468-7 Springer-Verlag New York Heidelberg Berlin
This work is subject to copyright. All rights are reserved, whether the whole or part
of the material is concerned, specifically those of translation, reprinting, re-use of
illustrations, broadcasting, reproduction by photocopying machine or similar means,
and storage in data banks. Under § 54 of the German Copyright Law where copies
are made for other than private use, a fee is payable to the publisher, the amount
of the fee to be determined by agreement with the publisher.
© by Springer-Verlag Berlin - Heidelberg 1979
Library of Congress Catalog Card Number 72-152360
Printed in Germany
The use of registered names, trademarks, etc. in this publication does not imply, even
in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
Typesetting, printing,and bookbinding: Brtihlsche Universit/~ tsdruckerei Lahn-GieBen.
2152/3140-543210
Editors
Prof . Dr . T. K . G h o s e
H e a d , B i o c h e m i c a l E n g i n e e r i n g R e s e a r c h Cen t r e , I n d i a n In s t i t u t e o f T e c h n o l o g y
H a u z K h a s , N e w De lh i 110029 / Ind ia
Prof . Dr . A . F i e c h t e r
E idgen . T e chn . H o c h s c h u l e , H a n g g e r b e r g , C H - 8 0 9 3 Zt i r ich
Prof. Dr . N . B l a k e b r o u g h
T h e Un ive r s i ty of Read ing , N a t i o n a l Col lege of F o o d T e c h n o l o g y
W e y b f i d g e Sur rey K T 1 3 0 D E / E n g l a n d
Managing Editor
Pro f e s s o r Dr . A . F i e c h t e r
E idgen . Techn . H o c h s c h u l e , H 6 n g g e r b e r g , C H - 8 0 9 3 Zi i r ich
Editorial Board
Prof. Dr. S. Aiba Prof. Dr. R.M. Lafferty
Biochemical Engineering Laboratory, Institute of Applied Techn. Hochschule Graz, Institut fiir Biochem. Technol.,
Microbiology, The University of Tokyo, Bunkyo-Ku, Schl/Sgelgasse 9, A-8010 Graz
Tokyo, Japan
Prof. Dr. L.K.Nyiri
Prof. Dr. B. Atkinson Dept. of Chem. Engineering, Lehigh University, Whitaker
University of Manchester, Dept. Chemical Engineering, Lab., Bethlehem, PA 18015/USA
Manchester/England
Prof. Dr. H.J. Rehm
Dr. J. B6ing Westf. Wilhelms Universit~it, Institut fiir Mikrobiologie,
RiShm GmbH, Chem. Fabrik, Postf. 4166, D-6100 Darmstadt Tibusstral3e 7-15, D-4400 MiJnster
Prof. Dr. P. L. Rogers
Dr. E. Bylinkina
School of Biological Technology,. The University of New
Head of Technology Dept., National Institute of Antibiotika,
South Wales, PO Box 1, Kensington, New South Wales,
3a Nagatinska Str., Moscow M-105/USSR
Australia 2033
Prof. Dr. H.Dellweg Prof. Dr. H. Sahm
Tecbn. Universit~it Berlin, Lehrstuhl fiir Biotechnologie, Institut ftir Biotechnologie, KernforschungsanlageJ iilich,
Seestrage 13, D-1000 Berlin 65
D-5170 Jiilich
Dr. A. L. Demain Pro Dr. W.Schmidt-Lorenz
Massachusetts Institute of Technology, Dept. of Nutrition Eidgen. Techn. Hochschule, lnstitut flit Lebensmittelwissen-
& Food Sc., Room 56-125, Cambridge, Mass. 02139/USA schaft, TannenstrafSe 1. CH-8092 Ziirich
Prof. Dr. R.Finn Prof. Dr. K.Schiigerl
Institut ftir Technische Chemie, Technische Universitiit
School of Chemical Engineering,
Hannover, CallinstraBe 3, D-3000 Hannover
Olin Hall, Ithaca, NY 14853/USA
Prof. S. Fukui Prof. Dr. H.Sunmalainen
Dept. of Industrial Chemistry, Faculty of Engineering, Director, The Finnish State Alcohol Monopoly, Alko,
Sakyo-Ku, Kyoto 606, Japan P.O.B. 350, 00101 Helsinki 10/Finland
Prof. G.T.Tsao
Dr. K. Kieslich
Director, Lab. of Renewable Resources Eng., A.A.Potter
Schering AG, Werk Charlottenburg, Max-Dohrn-Strage,
Eng. Center, Room 216, Purdue University, West Lafayette,
D-1000 Berlin 10
IN 47907/U S A
Contents
Application of Microcomputers
in the Study of Microbial Processes
W. Hampel, Vienna (Austria)
Dissolved Oxygen Electrodes 35
Y. H. Lee, Philadelphia, Pennsylvania (USA)
G. T. Tsao, West Lafayette, Indiana (USA)
Power Consumption in Aerated Stirred 87
Tank Reactor Systems
H. Brauer, Berlin (Germany)
Loop Reactors 121
H. Blenke, Stuttgart (Germany)
Application of Microcomputers
in the Study of Microbial Processes
Werner A. Hampel
Ins t i tu te o f Biochemical Techno logy and Microbiology
Universi ty o f Techno logy Vienna
A-1060 Wien, Austr ia
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Application of Computers to Analyze, Optimize, and Control Microbial Processes . . . . . . 2
2.1 Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2 Data Reduction and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Process Control and Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Microelectronics and Microcomputers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.1 Computer Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 Minicomputers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3 Microcomputers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.4 Distributed Computer Network (Multiprocessor Systems) . . . . . . . . . . . . . . . . . 12
4 Microcomputers Coupled to Bioreactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.1 Hardware Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1.1 Microcomputer (Desk-top Calculator) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.1.2 Interface Units for Peripheral Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.1.3 Process Periphery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.1.3.1 Data Input Periphery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.1.3.2 Data Output Periphery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2 Efficacy and Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.3 Cost Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5 Broadening Applicability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
6 Concluding Remarks and Future Tendencies . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
7 Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
The applications of computers and microcomputers in particular interfaced to bench-top bioreactors
are briefly discussed. Compact, quick-access microcomputers (e.g., desk-top calculators) offer the
possibility of on-line data acquisition and analysis as well as process control (sequencing, interactive
control) with relatively high efficiency at reasonably low costs of installation in laboratory experi-
ments. The configuration of such a computerized system is described in detail along with the capa-
bilities and features of interfacing hardware components. The limits in applicability due to slow
operating speeds of fully developed microcomputer systems in particular are pointed out and a sur-
vey on investment costs for high-performance, compact desk-top calculators and some peripheral
devices is given. Examples of on-line acquisition of several directly accessible environmental process
parameters and computations of directly inaccessible state variables are presented in the form of
type-writer print-outs. The advantages of onqine experiments for establishing sophisticated control
algorithms and for studying the physiological behaviour of the microbial population are demon-
strated.
2 W. A. Hampel
1 Introduction
The advantages and various applications of a system consisting of a computer directly
connected to one or several bioreactors have been described in detail in several publica-
tions 16, 39, s6) during the last decade and have occasionally been discussed with great
vehemence. However, there was no widespread use of such systems which was obviously
due to the high installation costs for efficient and reliable computers. It has been pos-
sible to increase the utilization of computers in recent years by the development of small,
reliable computer systems (minicomputers) thus permitting computer use for pilot
plants or even smaller microbial systems at a reasonable price.
The great success achieved with such system configurations has, furthermore, resulted
in an increased interest in on-line systems. In the case of bench-top bioreactors the de-
cisive breakthrough was the development of microprocessors and microcomputers re-
spectively. Especially the rapid emergence of compact, quick-access microcomputing
systems, e.g., programmable desk-top calculators, capable and flexible enough to meet
data processing requirements, made it possible to adapt data acquisition, data reduction
and process control even for bench-top bioreactors at reasonable costs. It remains to be
investigated as to what extent the microcomputer might replace the minicomputer so
far predominantly used in biological growth experiments. In particular, the question
posed is which microcomputing system offers optimum possibilities and where the pos-
sible limits of the applicability are.
2 Application of Computers to Analyze, Optimize,
and Control Microbial Processes
The on-line, real-time utilization of the computer in highly instrumented microbial cul-
tivation systems offers the possibility of
1) collecting large amounts of data derived from various kinds or sensors as part of
a data acquisition system,
2) automatically and instantaneously reducing certain data in order to determine
the state of the system, as well as of
3) optimizing and controlling the process.
2.1 Da ta Acquis i t ion
Various methods for determining process parameters based on physical, chemical, and
enzymatic analyses have been extensively described 1' 6, 7, 29, 30, as, 46, 48, 49, 53, 67, 69,
71, 76, "78, 84, 85, 94). Some of these methods do not fulfill many principal criteria TM
such as reliability, accuracy, reproducibility and prevention of contamination, or they
are characterized by a low maximum frequency for handling successively measured
Application of Microcomputers in the Study of Microbial Processes
values. The latter holds true for several chemical and physicochemical methods (e.g.,
wet chemical analysis, chromatographic procedures) so that these methods are only ap-
plicable to specific, predominantly scientific, problems.
There are only very few adequate systems for continuously and directly (in situ)
measuring microbial process parameters, i.e., acquiring both physical and chemical fac-
tors, such as temperature, pressure, flow rates of gases and liquids, power consumption
of the mixing system, pH, rH, dissolved gas concentrations, exhaust air composition.
The application of computers in a data acquisition system does offer some advantages
of which the following are pointed out:
Improved accuracy andreliability. Due to statistical methods for obtaining parameter
values (e.g., computation of the mean and variance) false and noisy signals can be re-
jected (digital filtering proceduress ' 45, 47, 77)). Periodical recalibration automatically
performed with several sensor systems may correct drifting.
Increased number of sensing systems. Owing to the possibility of supervising com-
plex measuring devfces by the computer, the number of accessible parameters may be
considerably increased; e.g., by incorporating batch methods of analysis or by using
methods requiring extensive numerical computations for obtaining results.
Cost reduction. Low-cost and simple measuring devices without additional electronic
circuits necessary for signal corrections may be used. That is a linearization of sig-
nals 6, 30 ), compensation of disturbing influences 23' 83), correction of time lags 21' 69),
etc., are easily performed by the computer in computer-aided data acquisition systems.
Conservation of data. Frequently it is not possible to immediately analyze the infor-
mation received by the aid of a data acquisition system and thus storage of measured
values for reuse is desirable. The length of a single microbial experiment as well as the
high frequency of data sometimes necessary results in a large number of values. The
computer offers the possibility to store these values on a suitable data base (punched
tape, magnetic tape, or disk) after these have been sorted if necessary. These stored data
may be used later on, optionally fed to a large mainframe computer, to answer questions
concerning the microbial cultivation process (modeling, optimization 11-13, 64, 81, 95)).
2.2 Data R e d u c t i o n and Analysis
Several of the monitoring systems are "gateway" sensors, i.e., they open the way through
combination with other sensor systems and data for obtaining further information about
the microbial cultivation 1' 39, 56). Thus, process indicators or directly inaccessible state
variables can be obtained which are related to the actual physicochemical, physiological,
and biochemical conditions of the culture sT' 60, 61)
For example, the use of reliable oxygen- and carbon dioxide gas analyzers as well as
devices for measuring the gas flow rates and volume of the culture liquid, permit a de-
tailed, real-time study of oxygen uptake and carbon dioxide evolution of microbial
cultures. In this respect the bioreactor is comparable to a differential respiromet-
er 1, 23, 60, 61). Gas exchange rates are computed according to Eqs. (1) and (2):
W. A. Hampel
Wi • x o ))
Qo= • X = Fi Xi . f , (1)
V a ~ ( X o ; yo
Qc°: " x = F i ( w i ' y ° )
V" a--(Xo+ yo) -y i "f
(2)
273 Pi a
f
= 2 7 3 + T i ' T ' l + h '
where F i represents the molar inlet gas flow rate at experimental conditions and wi, xi,
Yi the mole fractions of inert gas, oxygen, and carbon dioxide, respectively, in inlet air,
while Wo, Xo, Yo are the corresponding mole fractions in outlet air. For conversion to
standard conditions the absolute pressure (Pi), the temperature (Ti), and the absolute
humidity (h) of inlet air must be known. These intermediate data may then be used to
compute the volumetric gas transfer rate (kL a) for which additional values of the dis-
solved gas concentration in the culture liquid are needed. Thus information is provided
on existing conditions of mass transfer in the bioreactor at the given operating condi-
tions of agitation and aeration
Qo2 X
kLa - . (3)
(c L - e L)
On the other hand the oxygen uptake rate and the carbon dioxide evolution rate by the
culture can be combined to yield the respiratory quotient (RQ) according to Eq. (4),
RQ = Qco=/Qo~ (4)
thus providing information on the physiological behaviour of the culture. It has been
tried several times to establish statistically proven correlations between particular physio-
logical conditions of the microorganism and on-line process intbrmation 23' s6, 6o). For
instance, the RQ represents a quantitative indicator of ethanol formation and utilization
in an aerobic culture of Saccharomyces cerevisiae. As it has been described 86), the stages
of ethanol formation (RQ > 1.0), oxidative growth (RQ 1.0-0.9), endogenous meta-
bolism (RQ 0.8-0.7) as well as that of ethanol utilization may be distinguished by the
corresponding values for the RQ. The coincidence of definite metabolic activities (e.g.,
nucleic acid synthesis, protein synthesis, ethanol formation) with characteristic changes
of the RQ was demonstrated during the transition from lag to the logarithmic phase of
growth in the case of a Candida utilis culture by means of wet chemical analysis 6°).
Heat and material balances not only give insights with respect to the gas exchange
of microbial populations 14' 19-21, 26, 40, 41, 86, 87), but by also using additional sensing
systems (carbohydrate feed rate, turbidity, amount of neutralizing agents, etc.), infor-
mation pertaining to the biochemical characteristics of the process can be gained. From
data on gas exchange and molasses addition during the cultivation of baker's yeast, on-
line computations of growth rates and different yield coefficients (Yx/s, Yo/s) were per-
Application of Microcomputers in the Study of Microbial Processes 5
formed using carbon balances 86). A comparison of the compu ted cell yields with organic
energy yields using different substrates might reflect the mechanism of the breakdown
of organic compounds. The correlation between these variables will aid in the elucida-
tion of the main metabolic pathways and reflect the efficiency of the cultivation pro-
cess86, 97)
Easily and directly accessible process parameters can be used to estimate direct-
ly inaccessible parameters if appropriate models exist (indirect measurement con-
cept36, 40, 41,45)). Figure 1 shows the estimation of biomass concentration and specific
growth rate from data on exhaust air composition and air flow rate. After computing
the amount of assimilated oxygen using a material balance oxygen uptake can be dis-
tributed with respect to maintenance and growth according to a very simple model 64).
The values of the constants c m and Yx/s necessary for biomass calculation can then be
determined by using the appropriate evaluation procedures from experiments in which
oxygen uptake and biomass concentration are independently measured 9' 66, 68, 97)
2.3 Process C o n t r o l and Optimization
Owing to very encouraging results and adequate economic justification, computer con-
trol is widely used in batch chemical processes 38). Hence, several control strategies and
systems have been also developed for large-scale processes 27' 28, 93) Triggered by the
introduction of highly instrumented bioreactors 22" 33, 34) for biochemical and micro-
bial research, control structures and algorithms as well as optimization procedures have
been tested several times 2' 40, 42, 59, 61, 73, 74, 82, 96)
In process control applications control tasks performed by the computer may first
include "On/Off-control". Thus, the timing of several events before, during and follow-
Fo
)
So
F i
Si
Ba lance : Q s " X • V = F I - S ~ - Fo • So
e.g. for oxygen
W i • X o
Q o 2 " X ' V = F i • ( x i 1 - ( x o + y o) )
1 dX
M o d e l : Q s " X = Cm • X + - - •
Yx,s d t
-Cm " Y~s • t / t "Yxls" t
Es t i m a t i o n : X ( t ) = e . ( X a + y x / s . j e c ~ . O s . X . d t )
a
P = Yx,~" ( Os cm )
Fig. 1. Indirect estimation of biomass and specific growth rates. After computing the amount of as-
similated substrate by balancing methods, it is distributed to maintenance and growth according to
a simple model proposed by Pirt 64)
6 W.A. Hampel
ing the microbial cultivation may be performed. Examples of this type of process con-
trol include:
heating for sterilization and cooling 28' s2, 93); sampling and sample preservation for
off-line determination of different process parameters by physical, chemical, and enzy-
matic methods 3s' 84, 8s); switching of sample streams from several reactor units to shared
analyzers43, 49); stopping the process and isolating culture liquid, biomass, and/or pro-
ducts28, 93).
For further control tasks two basic strategies are currently available: "Direct Setpoint
Control" (DSC) and "Direct Digital Control" (DDC 2a' 38, so)). Conventional analog
control loops make it possible to maintain environmental parameters at previously de-
fined values (setpoints). In monitoring control the computer provides the setpoint se-
quencing and timing for the analog controller, whereas in direct digital control, the
computer directly controls the position of the final control elements. The advantage of
DSC is that in the case of computer failure, control may be returned to the local analog
element with manual override; on the other hand, DDC permits greater flexibility and
more precise representation, since control responses can be modelled as algebraic func-
tions: i.e., costly and time consuming redesign of controller hardware is not required.
At first, computer-aided process control enables the construction of control loops
for directly accessible environmental parameters, thus implementing sophisticated con-
trol strategies by using selected combinations of proportional, integral and differential
(PID) control algorithms. Process indicators or state variables, obtained by data reduc-
tion, may likewise be used in control loops 4°' 42, sg, 61, 86). When using these variables
(interactive control), the fact will have to be taken into consideration that the change
of even one environmental parameter influences the values of several other variables,
so that their readjustment will be necessary. This adjustment must be carried out ac-
cording to the complex physiological needs of the cultivated cells, hence, a detailed
knowledge of microbial physiology is of utmost importance. These control operations
may act on several levels for instance connected in cascades. The target values of en-
vironmental variables are maintained by means of conventional control elements (DDC
or DSC), but the setpoint of each control loop is altered according to the state of the
culture by taking into consideration the interactive effects of the system elements.
The algorithms of interactive control can be developed most effectively from on-line
computer-aided experiments while observing the response of a culture to a sudden
change of one variable (perturbation testing 2s' 61, 63)). Results of modelling experiments
may likewise be used after process identification and parameter estimation 9' ~~ )
Computer-aided interactive control was successfully implemented in experiments
using the computed kL .a - value to alter agitation speed and air flow rate in order to
maintain the predetermined concentration of dissolved oxygen. A physiological process
parameter calculated on-line, in particular the RQ, was utilized to control the feed rates
of carbohydrate and nitrogen sources in a cultivation of Candida utilis 61), or to control
the molasses feed rate in the case of fed-batch cultures of baker's yeast, respective-
ly86, 89).
Optimization procedures are typically designed to increase the concentration of
some extracellular product, to enhance particular metabolic functions, or to increase