Insights into Mixing and Transport Phenomena in Group IVa Bioreactors

Relatively little work is available that allows insights into mixing and transport phenomena in continuously-mixed, forcefully-aerated bioreactors. More work has been done on intermittently-mixed, forcefully-aerated bioreactors (see Chap. 10). This is not altogether surprising: Although some microorganisms can tolerate continuous mixing, the majority performs better when the mixing is intermittent.

Wall cooling can be effective in heat removal at small scale but it will not be sufficient to maintain the bed temperature at the desired value as scale is increased, if the bioreactor is scaled up on the basis of geometric similarity. This is demonstrated in Fig. 9.7, which is the result of a case study undertaken by Nagel et al. (2001a). If the length-to-diameter ratio is maintained constant, then the surface area of the wall per unit volume of bed decreases. Therefore, as scale is increased, eventually a "critical bioreactor volume" is reached, above which wall cooling alone cannot control the bed temperature. In Fig. 9.7 this critical volume corresponds to the bioreactor volume at which the curve intersects the horizontal line. Of course the critical bioreactor volume will depend on the maximum growth rate of the organism, the temperature difference between the bed and the cooling water and the length to diameter ratio of the bioreactor.

One strategy for heat removal at volumes above the critical bioreactor volume might be to include internal heat transfer surfaces. This could possibly be done by incorporating baffles, or by circulating cooling water through the mixing paddles. However, a situation will quickly be reached in which a further increase in internal heat transfer surfaces will interfere with the ability to mix the bed. The other strategy is to promote evaporation by using dry air. In this case, it will be essential to make periodic water additions in order to prevent growth from being limited by low water contents. Note that the sudden decrease in the O2 uptake rate in Fig. 9.4(c) was due to the decrease in the water activity of the bed.

Flow patterns within mixed solid beds in SSF bioreactors have received little attention. It is often assumed that the bed is well mixed in such bioreactors. However, this might not necessarily be the case. Rather, there might be defined circulation patterns and the effectiveness of mixing may be different in different regions of the bed. This is best illustrated by the study of mixing within a conical solids mixer that was undertaken by Schutyser et al. (2003b). Positron emission particle tracking was used to follow the circulation of individual particles within the bed. Figure 9.8 shows how such studies can give information about the circulation patterns of particles within the bed. However, note that such studies require access to quite sophisticated equipment. Schutyser et al. (2003b) compared the experimental results obtained with positron emission particle tracking with predictions made using a discrete-particle mixing model (see Fig. 8.10(a) for a simple explanation of the basis of discrete-particle modeling). Once such a model has been validated, it can then be used to predict flow patterns.

Was this article helpful?

0 0

Post a comment