Automated Measurement and Monitoring of Bioprocesses: Key Elements of the M3C Strategy - Measurement, Monitoring, Modelling, and Control of Bioprocesses - page 19

FIA was originally defined as ''… information gathering from a concentration
gradient formed from an injected, well-defined zone of a fluid, dispersed into a
continuous unsegmented stream of a carrier …'' [ 75 ]. Accordingly, basic com-
ponents of any FIA equipment are a transport system consisting of tubing, pumps,
valves, and a carrier stream into which a system (practically always an injection
valve attached to an injection loop of defined, constant volume) injects a sample.
Rehbock et al. [ 76 ] assign FIA a major key role in bioprocess monitoring. In
contrast, a sequential injection analysis (SIA) system does not work continuously,
since it must aspirate the necessary components sequentially and ejects them
afterwards, and the flow may be segmented; however, this setup can save reagents
and reduce waste. The principles are briefly sketched in Fig. 2 .
A central element in the FIA is the injection valve used to load an injection loop
(position as shown: sample is pumped through the loop and further to waste) and
then expel this defined volume with carrier fluid after switching of the valve. The
sample can be diluted (not shown) and mixed with reagent(s). The mixture is
allowed to react for a time given by the flow rate and the volume of the residence
time reactor, usually a long, thin tube (tubular microreactor), and finally directed to
a flowthrough detector.
The SIA works with a precise bidirectional pump, usually a syringe pump. It
first aspirates an appropriate amount of the sample and—in sequence, using the
selection valve—the necessary reagents (in appropriate amounts), which are then
expelled into a (micro)reactor to be mixed and react. Thereafter, the reaction
mixture is pumped to a detector.
A fascinating property of FIA is the elegant implementation of the dilution step
(Fig. 3 ). This is highly desirable for monitoring batch and probably also fed-batch
processes: some components change their concentrations more than a few decades
during a cultivation, but good resolution at low concentrations (e.g., while a
substrate is being limiting) is highly desirable.
Whenever injection loops are implemented, it is mandatory to avoid the
occurrence of gas bubbles in the sample stream, since they falsify the volumetric
dosage. This error would, of course, be propagated by the dilution factor.
Degassing vessels (i.e., sedimenters for the liquid phase) are simple to make but
add a mixing device upstream of the dilution and/or analysis, whereas degassing
membranes may not be as effective but they do not mix (they just disperse axially).
Downstream of the dilution section everything can be arranged to work in
continuous flow mode, e.g., adding reagents ahead of static mixers, ''inactive''
tubes to achieve a necessary hydraulic residence time, flowthrough cartridges
hosting immobilized (bio)catalysts, membrane reactors retaining cells but allowing
to wash them, etc. Finally, there is a detector, appropriately selected for the final
product to be quantified. Optical and electrochemical detectors qualify well, and
biosensors are equally useful in this position since they need not be sterilized there
[ 77 ]. However, any other instrument is welcome if appropriate: as described in
Sect. 3.2.5 , a flow cytometer is the final detector and certainly much more
expensive than the preceding FIA. Vojinovic et al. [ 78 ] have compared several