tion, pH, or water activity gradients but some experimental and modeling work
has addressed the diffusion of enzymes, oxygen, and soluble nutrients.
To date there has been little experimental effort to measure nutrient concentration gradients within particles. Mitchell et al. [109], by cutting slices off the
faces of cubes of an artificial gel-based solid substrate, showed that starch and
glucose concentrations were higher in the interior portion of the substrate
particle than in the outer portion. Protein concentrations, used as a measure of
biomass, were higher at the surface. Finer cuts would characterize the concentration profiles better, the fineness of the cut being limited by the sensitivity of
the detection method for the compound whose concentration profile is being
studied. Oxygen microprobe measurements demonstrate that during peak
growth periods oxygen concentrations fall from atmospheric levels to zero
within 100 mm depth for a fast growing organism such as Rhizopus oligosporus
(m = 0.3 h –1 ) and within 200 mm depth for a slow growing organism such as
Coniothyrium minitans (m = 1 day –1 ) [86]. The modeling studies described
below have given more insight into these experimental results.
In overculture studies with a gel substrate in which starch was embedded,
Mitchell et al. [84] showed that the diffusional limitation of the glucoamylase
released by the fungus at the surface meant that the enzyme tended to be concentrated near the surface most of the time, and as a result the starch at the surface was quickly utilized (Fig. 4a). The boundary between the cleared region
and the region containing starch receded from the surface. The glucoamylase in
the cleared region lacked substrate to act upon. The cleared region receded
from the surface faster than the glucoamylase diffused away from the surface,
the result being that the rate of glucose production fell quite drastically
(Fig. 4b). It could be as low as 20% of the activity that would be obtained if all
the enzyme were in contact with starch. For much of the fermentation, growth
can be limited by the rate at which this glucose is generated and diffuses to the
surface. In this case growth can be assumed to be limited by enzyme diffusion,
since the effect arises due to the low diffusivity of glucoamylase. At high diffusivities the glucoamylase would distribute evenly throughout the substrate
and starch hydrolysis would occur throughout the substrate rather than being
localized near the surface.
In this work of Mitchell et al. [84] the biomass had no structure and diffusion
of glucose within the biomass layer was not considered. Furthermore the role of
oxygen diffusion and consumption was not considered. The model was extended by Rajagopalan and Modak [85] for the growth of unicellular organisms
in a biofilm of constant density (shown on the right of Fig. 4a). They concluded
that oxygen is more likely to limit growth within the film than lack of glucose,
even when high oxygen concentrations can be maintained at the outer edge of
the film [85, 122]. After a short period in which neither glucose or oxygen
are lacking, the oxygen becomes limiting within the lower parts of the biofilm.
This situation is maintained until glucose becomes limiting later in the
fermentation when the starch in the substrate is almost completely utilized.
These results have been confirmed in a somewhat different SSF system in which
glucose solution is sprayed onto the surface of yeast pellets in a gas-solid
fluidized bed bioreactor [89, 90]. In the production of baker’s yeast in this
Biochemical Engineering Aspects of Solid State Bioprocessing
93
has addressed the diffusion of enzymes, oxygen, and soluble nutrients.
To date there has been little experimental effort to measure nutrient concentration gradients within particles. Mitchell et al. [109], by cutting slices off the
faces of cubes of an artificial gel-based solid substrate, showed that starch and
glucose concentrations were higher in the interior portion of the substrate
particle than in the outer portion. Protein concentrations, used as a measure of
biomass, were higher at the surface. Finer cuts would characterize the concentration profiles better, the fineness of the cut being limited by the sensitivity of
the detection method for the compound whose concentration profile is being
studied. Oxygen microprobe measurements demonstrate that during peak
growth periods oxygen concentrations fall from atmospheric levels to zero
within 100 mm depth for a fast growing organism such as Rhizopus oligosporus
(m = 0.3 h –1 ) and within 200 mm depth for a slow growing organism such as
Coniothyrium minitans (m = 1 day –1 ) [86]. The modeling studies described
below have given more insight into these experimental results.
In overculture studies with a gel substrate in which starch was embedded,
Mitchell et al. [84] showed that the diffusional limitation of the glucoamylase
released by the fungus at the surface meant that the enzyme tended to be concentrated near the surface most of the time, and as a result the starch at the surface was quickly utilized (Fig. 4a). The boundary between the cleared region
and the region containing starch receded from the surface. The glucoamylase in
the cleared region lacked substrate to act upon. The cleared region receded
from the surface faster than the glucoamylase diffused away from the surface,
the result being that the rate of glucose production fell quite drastically
(Fig. 4b). It could be as low as 20% of the activity that would be obtained if all
the enzyme were in contact with starch. For much of the fermentation, growth
can be limited by the rate at which this glucose is generated and diffuses to the
surface. In this case growth can be assumed to be limited by enzyme diffusion,
since the effect arises due to the low diffusivity of glucoamylase. At high diffusivities the glucoamylase would distribute evenly throughout the substrate
and starch hydrolysis would occur throughout the substrate rather than being
localized near the surface.
In this work of Mitchell et al. [84] the biomass had no structure and diffusion
of glucose within the biomass layer was not considered. Furthermore the role of
oxygen diffusion and consumption was not considered. The model was extended by Rajagopalan and Modak [85] for the growth of unicellular organisms
in a biofilm of constant density (shown on the right of Fig. 4a). They concluded
that oxygen is more likely to limit growth within the film than lack of glucose,
even when high oxygen concentrations can be maintained at the outer edge of
the film [85, 122]. After a short period in which neither glucose or oxygen
are lacking, the oxygen becomes limiting within the lower parts of the biofilm.
This situation is maintained until glucose becomes limiting later in the
fermentation when the starch in the substrate is almost completely utilized.
These results have been confirmed in a somewhat different SSF system in which
glucose solution is sprayed onto the surface of yeast pellets in a gas-solid
fluidized bed bioreactor [89, 90]. In the production of baker’s yeast in this
Biochemical Engineering Aspects of Solid State Bioprocessing
93