510
Chapter 11
Another factor to keep in mind is that the requirement for culture stability increases. The actual
number of generations in the large-scale process depends on the inoculation density. However,
even if a high inoculation density is used, the number of generation from the stock culture to the
final harvesting increases rather much, and possible genetic instabilities of the production strain
will therefore be more pronounced.
A final, unpleasant, surprise experienced in many large-scale processes is foaming. Foaming is
caused by surface-active components that are excreted or released through cell breakage.
Foaming is normally manageable in lab-scale experiments, but in a large-scale reactor
unexpected foaming can become a major problem.
11.5. Scale-up in Practice
At this point, we may ask ourselves, what is a suitable practical approach to scale up? Suppose
that a very successful lab-scale or pilot scale process has been developed. Could we not just
maintain the same process conditions and scale up the process? From the previous sections in
this chapter it should be clear that this, unfortunately, is physically impossible. This is further
illustrated in Table 11.8, which is based on a classical table published by Oldshue (1966).
Four different parameters were maintained constant during the scale-up in Table 11.8, i.e. the
specific power input, the stirrer rate (which corresponds to maintaining the mixing time), the tip
speed (which gives approximately the same maximum shear rate) and the Reynolds number
(which is suggested by the dimensionless Navier-Stokes equation). It is clear that in fact most of
the other parameter values, except the one chosen constant, change during scale-up.
Four in principle different approaches to scale-up in practice can be distinguished (Kossen and
Oosterhuis, 1985):
1. Fundamental methods
2. Semifundamental methods
3. Dimensional analysis
4. Rules of thumb
Table 11.8. Effect of scale-up on charateristic properties when scaling up from a 100 L to 12.5 m3
Property
Pilot scale
(100 L)
Plant scale (12.5 m3)
p
1
125
3125
25
0.2
P/V
1
1
25
0.2
0.0016
N
1
0.34
1
0.2
0.04
ds
1
5
5
5
5
Vpump
1
42.5
125
25
5
tc
1
0.34
1
0.2
0.04
Nd,
1
1.7
5
1
0.2
Res
1
8.5
25
5
1
previous page 532 Bioreaction Engineering Principles, Second Edition  read online next page 534 Bioreaction Engineering Principles, Second Edition  read online Home Toggle text on/off