Scale-up of Bioprocesses
Note 11.2: Methods for characterizing mixing
Many different characterization methods have been applied to describe the flow pattern and mixing in
bioreactor systems. The simplest methods are based on tracer techniques in which a tracer is added to the
bioreactor and its concentration is measured as a function of time. A short description will be given of
these methods, which are easily implemented and therefore recommendable for initial studies of mixing
phenomena. For detailed studies of the mixing, characterization of the entire flow field is necessary as
discussed in section 11.3.5.
Many tracer techniques have been successfully applied to determine the
mixing time in a bioreactor:
based on electrolytes as tracer
based on acids or bases as tracer
in which dyes are used as tracer
Heat pulse methods,
in which a warm liquid is used as a tracer
based on radioactive isotopes as tracer
All these methods are easy to implement, but unfortunately each has one or more drawbacks in connection
with measurement in fermentation media. The choice of method should therefore be based on an evaluation
of the drawbacks (see below) when the method is applied in the actual system.
is inexpensive and easily implemented, but it has the disadvantage that most
fermentation media are themselves good conductors, and the sensitivity is therefore poor.
is very easily implemented since most bioreactors are equipped with pH electrodes. By
measuring the change in pH after addition of base (or acid), the mixing time can be determined. Pulses of
bases (or acids) can normally be added to a bioreactor without seriously affecting the fermentation process,
and the method can therefore be applied under real process conditions. However, microbial activity may
influence the results since many microorganisms produce acids as metabolic products, and it is therefore
important that the mixing time is much smaller than the characteristic time for acid production. The major
disadvantage of the pH method is that most fermentation media have a high buffer capacity, and large pulses
are therefore required in order to obtain good sensitivity.
is based on measurement of an inert dye, e.g. a fluorophore such as NADH, riboflavin
or coumarin. There are commercially available fluorescence probes that can be inserted through standard
ports in most bioreactors, and it is therefore possible to quantify the mixing time under real process
conditions. However, for many fermentation media the background fluorescence is high and the sensitivity
may therefore be poor. Instead of coloring the medium it is also possible to decolorize the medium. This can
be done by for instance lowering pH by adding acid in an alkaline phenolphtalein solution. This is very useful
for detecting stagnant zones in a reactor, since these zones will remain colored (Stein, 1992).
heat pulse method
is based on following the spread of warm water by high precision temperature sensors.
One problem with this method is to achieve a local distribution of heat
is based on addition of radioactive isotopes and measurement of the radioactivity using
scintilation counters. An advantage of this method is that the sensor can be placed outside the bioreactor, and
it can therefore easily be fitted. With the right choice of isotope a good sensitivity can be obtained for most
systems and there will be no traceable influence on the microbial activity from the radiation. A disadvantage
of the method is, however, the concern caused in production facilities when radioactive isotopes are added to
There have been relatively few studies concerning mixing in aerated systems compared to the
case of one-phase systems. At high aeration rates, mixing in the liquid phase is significantly
improved. However, the positive effects on mixing take place after flooding of the impeller. This