Conclusions on the Design and Operation of Rotating Drum Bioreactors

When operated in batch mode, rotating- and stirred-drum bioreactors should be designed to promote axial homogeneity. Axial mixing can be promoted by having an inclined axis and angled lifters that push the substrate along the drum (Fig. 8.11). However, it can also be achieved to some degree by ensuring that the head-space gases are well mixed, since convection and evaporation to the headspace gases will be major pathways for heat removal at large scale, and a uniform temperature within the headspace will tend to promote uniform rates of heat removal and therefore a uniform temperature along the substrate bed. However, it is not practical to insert a fan within the headspace. The other option is to introduce and remove air along the whole axial length (Fig. 22.10).

Due to the importance of evaporative cooling at large scale, it will be necessary to add water to the substrate during the fermentation. This will require internal piping with spray nozzles (Fig. 22.10).

piping to remain stationary

Fig. 23.10. Design features for large-scale rotating-drum bioreactors, including (1) multiple air inlets and outlets for promoting homogeneity within the headspace and (2) nozzles to allow the addition of water, which will be necessary at large scale due to the dependence on evaporative cooling. If the bioreactor is baffled, then it will be necessary to leave an appropriate clearance between the baffles and the air and water lines piping to remain stationary

Fig. 23.10. Design features for large-scale rotating-drum bioreactors, including (1) multiple air inlets and outlets for promoting homogeneity within the headspace and (2) nozzles to allow the addition of water, which will be necessary at large scale due to the dependence on evaporative cooling. If the bioreactor is baffled, then it will be necessary to leave an appropriate clearance between the baffles and the air and water lines

The model predictions suggest that the substrate-to-headspace heat and mass transfer coefficients are important. This is an area that has received only relatively little attention. Based on the predictions of the model, the "major contributor to heat removal" is likely to change quite significantly as scale is increased. Conduction through the bioreactor wall, which makes a large contribution to heat removal at small scales, is insufficient to remove the heat at large scales. This of course is due to the fact that the ratio of the surface area of the drum to the substrate bed volume decreases with scale (if geometric similarity is maintained). To try to maximize the heat transfer through the drum wall at large scales, you might consider either (1) including a water jacket, although this greatly complicates the design and increases power requirements for rotation; or (2) increasing the L to D ratio in order to minimize the reduction of the ratio of the bed-wall contact area to bed volume that occurs with increase in scale when geometric similarity is maintained. Of course there will be practical limits as to how long and thin the reactor can be.

Was this article helpful?

0 0

Post a comment