Automated Measurement and Monitoring of Bioprocesses: Key Elements of the M3C Strategy - Measurement, Monitoring, Modelling, and Control of Bioprocesses - page 2

3.3 Components Dissolved in the Medium...................................................................... 18

4 Conclusions.......................................................................................................................... 27

References.................................................................................................................................. 27

1 Introduction

Successful process monitoring and control in industry depends heavily on proper
and functioning measurement and monitoring techniques. Especially in the bio-
process industry, the demands on measurements and bioprocess analysis are great
challenges for inventors, instrument suppliers, and industrial users for a number of
reasons. One reason is that biological systems are very complex from an analytical
perspective, with components of varying molecular size and concentration, all in a
quite complicated and dynamically changing matrix. Another reason is the fact
that most bioprocesses in research or industrial production are operated behind a
sterile barrier and must provide some bio-active product with critical quality
attributes. This makes the task of quantifying critical process variables directly in
or at the process and (almost) in real time more demanding than measuring
withdrawn samples in the laboratory. As a consequence, only a few sensors are
routinely used and only a few instrumental analytical methods and instruments are
adapted to and supported as process analyzers for bioprocess engineering [ 1 - 3 ].
However, the process analytical technology (PAT) initiative and the more recent
Guidance for Industry: Process Validation of the Food and Drug Administration
(FDA, USA) strongly support and favor these developments [ 4 , 5 ]. This forms a
central subset of the M 3 C strategy. The papers cited above should be used to find
references to older literature.
In the following, some technical terms will be used and may result in some
confusion. For instance, the term ''online'' has recently often been used when
results are made available in a short time after sampling and/or on a computer.
Here, ''online'' is used in the sense of ''fully automatic,'' meaning that no manual
interaction by an operator is required to provide results on a process computer. ''In
situ'' is used in the sense of ''mounted and exposed to a representative volume
element inside'' the bioreactor. ''In bypass'' is used to indicate that a representative
volume flux is withdrawn from the reactor and analyzed aside but close to the
reactor. This sample may be returned to the reactor if the analyzer works nonin-
vasively (i.e., does not affect the sample) and does not compromise the sterile
barrier; however, the sample must probably be treated in some harmful way before
analysis and therefore discarded thereafter. ''In real time'' is used in the sense that
the result is available ''quickly enough'' that it can be used to affect the process,
i.e., to close a control loop, meaning that delay or dead times are reasonably low
with respect to the dynamics of a bioprocess. This time horizon is of course shorter
for rapidly growing cells than for slowly growing ones; it should be as short as
possible in very dynamic situations, for instance, when the physiological status of