Chemical Engineering November 2013 - 44

Feature Report
Engineering Practice
Heat Transfer for HugeScale
Fermentation
The impact of ten-fold scaleup is explored here
Jim Gregory and Bob Green
Fluor
I
ndustrial microbiology continues
to hold the promise of producing an
increasingly large array of commercially
important chemicals because
an ever-wider number of useful microbes
is being developed and successfully
deployed at commercially viable
scales. Today, many of these bio-based
chemicals are needed in quantities so
large that existing fermentation technology
is inadequate, and this is pushing
the envelope in terms of the size of
today's largest-scale fermenters.
Currently, aerobic fermenters exist
up to about 100,000 gal. Fermenters
ten times bigger are needed. This
challenge raises a number of important
engineering questions: How big
can a fermenter get? Is there a limit,
in terms of economies of scale, above
which the use of additional units is
the preferred option for gaining additional
capacity? What does a " hugescale "
fermenter look like? How big
do demonstration-plant fermenters
need to be to justify building the fullscale
plant? What testing is required
at small scale to support the full-scale
design? This article on heat transfer
in huge fermenters is intended to offer
answers to these questions. Hugescale
projects also face issues related
to mass transfer, momentum transfer,
cleaning and sterilization, and piping
specifications, but these are beyond
the scope of this article.
The impact of scaleup
As reaction vessels get larger, their
liquid volume (and thus heat load)
grows by the cube, while the jacket
surface area grows by the square (Figure
1). Therefore, with scaleup, heattransfer
capability is reduced. As a
Jacket heat transfer
area versus scale
Working volume, gal
Jacket area, ft2
Ratio of jacket area to
working volume, ft2
1,000
140
/gal
0.14
10,000
650
0.06
100,000
3,000
0.03
1,000,000
15,000
0.015
FIGURE 1. The amount of jacket surface area required for a given reactor is a function
of both area and scale and grows by the square, while the liquid volume (and
thus heat load) grows by the cube. As a result, heat transfer capability falls with progressive
scaleup and steps must be taken to address this issue
result, larger fermenters need colder
coolants or additional modifications
to increase surface area, or both.
Figure 1 shows how fermenter attributes
scale up geometrically. This is intended
for illustrative purposes only, to
show how heat-transfer capacity drops
significantly even when fermenters
look similar. The authors understand
that 100,000-gal and larger fermenters
will not look anything like this. They
may not have cooling jackets, they can't
be so tall, or be agitated in this way;
they may not even be pressure vessels.
Since chilled water (~35°F) is the
coldest practical coolant for this application,
large fermenters typically
use a combination of jacket, coil and
baffle-type heat-transfer methods to
increase the amount of surface area
available for heat transfer. For a
highly aerobic fermentation system, a
jacket alone is not sufficient for tanks
above about 30,000 gallons.
Coil and baffle-type heat-transfer
surfaces can greatly increase the
available fermenter heat-transfer
area. However, the effectiveness of
these supplemental heat-transfer surfaces
decreases with increasing scale,
just as in the case for the jacket.
In addition, jacket, coil, and baffle44
ChemiCal engineering www.Che.Com november 2013
type heat exchangers require multiple,
parallel paths in order to meet the
design flow velocity and outlet temperature
parameters. A jacket made
of 4-in. half pipe, or an internal cooling
coil made of 3-in. tubing, both would be
designed for about 5 ft/s velocity, which
accommodates a flow of about 100 gal/
min. Thus, for a huge fermenter that
needs tens of thousands of gal/min of
coolant, hundreds of parallel coolant
paths would be required. This would
engender significant costs.
By comparison, an external-type
heat exchanger would allow for many
fewer parallel paths by utilizing much
larger pipe. Therefore, huge fermenters
are able to benefit from external
heat exchangers to reject heat, at least
for aerobic processes.
Similarly, external heat exchangers
have the advantage of unlimited heattransfer
area and do not require the
internal structure and piping that are
necessary to support the coils and baffles
(which displace product and make
the fermenter difficult to keep clean).
Additionally, external heat-transfer can
be proven at small scale in a way that
helps to reduce the risk during scaleup,
and thus potentially reduce the need
for demonstration-scale plants.

http://www.Che.Com

Chemical Engineering November 2013

Table of Contents for the Digital Edition of Chemical Engineering November 2013