even if they have heights of several meters. In contrast, the Zymotis bioreactor
allows for large substrate loadings [131]. Theoretical modeling work done for
the Zymotis bioreactor suggests that the best operational strategy is to have a
control system for the cooling water, which reduces the temperature of the cooling water in response to the temperature measured at the air outlet end at the
mid point between the heat transfer plates, which is where the maximum temperature is expected. Using such a strategy and using a low spacing between
plates of 5 cm means that near optimal performance can be obtained [143].
Water balances have been noted as a potential problem in packed bed bioreactors [144], but they have received little attention. Most modeling work has
ignored them, and experimental work has not been done to test how water contents vary with height. One strategy to minimize the importance of the water
balance is to use a substrate in which large decreases in water content can occur
before the water activity falls to undesirably low values [37]. Selection of such a
substrate requires knowledge of sorption isotherms of the substrate and the
influence of water activity on the growth of the process organism.
The other phenomena which are quite important are related to air flow
through the column. Most modeling work has assumed plug flow of the air with
a flat velocity profile across the column, but no experimental work has been
done to confirm this. Pressure drop through the column is a potentially important phenomenon. At high biomass concentrations the intrinsic permeability of the bed can be less than 5% of the initial intrinsic permeability
[108]. Pressure drops as high as 2.75 m of water per meter of bioreactor have
been noted, during the growth of Aspergillus niger on a substrate consisting of
a nutrient and sucrose solution adsorbed onto sugar cane bagasse and with a
low aeration rate of only 3.5 vvm (air volumes per bioreactor volume per
minute) [145]. Other systems with similar air flow rates (in vvm) have given
maximum pressure drops ranging from 0.1 m to 0.7 m of water per meter of
bioreactor [108, 145]. With a higher vvm of 11, Gumbira-Sa’id et al. [127] obtained a pressure drop of 1.45 m water per meter of bioreactor. Note that many
of these studies have been done with relatively small laboratory reactors, where
there might be significant effects of the walls and the bioreactor ends.
Pressure drops of 1 m of water per meter of bioreactor or higher could
quickly put a limit on bioreactor height. These pressure drops need to be
carefully considered in the design of the aeration equipment for packed bed
bioreactors. If pressure drop is a problem and infrequent mixing can be
tolerated, then this mixing can be used to limit the pressure drop [127].
A consequence of the decrease in particle size noted in Sect. 4.5 is that the
substrate bed within a packed bed can shrink and may pull away from the bioreactor walls [127]. This is undesirable since air passes preferentially between
the bed and the wall, rather than through the bed itself. Such shrinkage and its
consequences can be avoided by using inert materials on which substrates are
absorbed or by using a natural substrate in which the solid structure is provided by a polymer which is not attacked by the microorganism [37].
In conclusion regarding packed-bed bioreactors, the Zymotis design of
Roussos et al. [131] is the most appropriate bioreactor for those processes in
which the substrate bed must remain static throughout the fermentation, since
Biochemical Engineering Aspects of Solid State Bioprocessing
109
allows for large substrate loadings [131]. Theoretical modeling work done for
the Zymotis bioreactor suggests that the best operational strategy is to have a
control system for the cooling water, which reduces the temperature of the cooling water in response to the temperature measured at the air outlet end at the
mid point between the heat transfer plates, which is where the maximum temperature is expected. Using such a strategy and using a low spacing between
plates of 5 cm means that near optimal performance can be obtained [143].
Water balances have been noted as a potential problem in packed bed bioreactors [144], but they have received little attention. Most modeling work has
ignored them, and experimental work has not been done to test how water contents vary with height. One strategy to minimize the importance of the water
balance is to use a substrate in which large decreases in water content can occur
before the water activity falls to undesirably low values [37]. Selection of such a
substrate requires knowledge of sorption isotherms of the substrate and the
influence of water activity on the growth of the process organism.
The other phenomena which are quite important are related to air flow
through the column. Most modeling work has assumed plug flow of the air with
a flat velocity profile across the column, but no experimental work has been
done to confirm this. Pressure drop through the column is a potentially important phenomenon. At high biomass concentrations the intrinsic permeability of the bed can be less than 5% of the initial intrinsic permeability
[108]. Pressure drops as high as 2.75 m of water per meter of bioreactor have
been noted, during the growth of Aspergillus niger on a substrate consisting of
a nutrient and sucrose solution adsorbed onto sugar cane bagasse and with a
low aeration rate of only 3.5 vvm (air volumes per bioreactor volume per
minute) [145]. Other systems with similar air flow rates (in vvm) have given
maximum pressure drops ranging from 0.1 m to 0.7 m of water per meter of
bioreactor [108, 145]. With a higher vvm of 11, Gumbira-Sa’id et al. [127] obtained a pressure drop of 1.45 m water per meter of bioreactor. Note that many
of these studies have been done with relatively small laboratory reactors, where
there might be significant effects of the walls and the bioreactor ends.
Pressure drops of 1 m of water per meter of bioreactor or higher could
quickly put a limit on bioreactor height. These pressure drops need to be
carefully considered in the design of the aeration equipment for packed bed
bioreactors. If pressure drop is a problem and infrequent mixing can be
tolerated, then this mixing can be used to limit the pressure drop [127].
A consequence of the decrease in particle size noted in Sect. 4.5 is that the
substrate bed within a packed bed can shrink and may pull away from the bioreactor walls [127]. This is undesirable since air passes preferentially between
the bed and the wall, rather than through the bed itself. Such shrinkage and its
consequences can be avoided by using inert materials on which substrates are
absorbed or by using a natural substrate in which the solid structure is provided by a polymer which is not attacked by the microorganism [37].
In conclusion regarding packed-bed bioreactors, the Zymotis design of
Roussos et al. [131] is the most appropriate bioreactor for those processes in
which the substrate bed must remain static throughout the fermentation, since
Biochemical Engineering Aspects of Solid State Bioprocessing
109