Automated Measurement and Monitoring of Bioprocesses: Key Elements of the M3C Strategy - Measurement, Monitoring, Modelling, and Control of Bioprocesses - page 4

systems and should, therefore, be monitored using multi-analyzer techniques [ 12 ,
13 ]. We can retrieve the parameters by appropriate evaluation of the variables,
which is also known as the inverse problem or system identification problem. The
entire chapter will use this definition in spite of the widespread misuse of the term
''process parameter'' in place of process variable.
2 State of Routine
The variables treated in this section are also quite important for nonbiological or
chemical processes; they are not typical of or particular to bioprocesses. However,
the respective commercially available sensors are robust, reliable, and quite mature.
2.1 Monitoring Physical Variables
Among these are sensors for agitator speed, aeration flux, weight of vessels, and
probably electrical power consumption. They are not critical with respect to con-
tamination risks because all these measurements take place outside the sterile barrier.
Gas flow rates can be measured as volumetric or mass flow rates. The latter are
recommended because gas mass flow rates occur in the respective mass balances
needed to evaluate process data, e.g., to determine a volumetric or specific oxygen
consumption rate. The instruments at hand are thermoanemometers provided by
various companies.
Some care must be taken when using balances or scales to determine the weights
of vessels and to conclude from these weights the respective filling degrees or
changes thereof; liquid feed or harvest mass flow rates can be accurately and pre-
cisely derived from the time trajectories. However, a proper force-neutral mechan-
ical setup is required and must not be changed during the entire operation period.
Omnipresent is the monitoring of temperature and, although usually not at the
very small scale, pressure. Respective sensors and amplifiers are available from
various suppliers. However, these sensors, as well as those discussed in the fol-
lowing, are to be mounted in situ. They must not be prone to drifts or gain changes
during sterilization, and their housing, installation method, and sterile sealing are
critical with respect to contamination.
2.2 Monitoring Chemical Variables
When moving from physical to chemical variables, the sensors are more typical for
biochemical processes, namely those that are specific or selective for protons,
dissolved gases, and dissolved redox couples. Several alternative brands are
available commercially.