Fung and Mitchell (1995) investigated the effect of the presence and absence of baffles on the performance of a 200-L rotating-drum bioreactor, in which Rhizopus oligosporus was grown on wheat bran. The drum had an internal diameter of 56 cm and an internal length of 85 cm. When baffles were used, four baffles of 17 cm width and 85 cm length were attached at right angles to the inner wall of the drum, with uniform spacing between them (i.e., in the manner indicated in Fig. 8.1). There was no external temperature control; the bioreactor operated within a room that varied from 17 to 26°C. Pre-humidified air was blown through the bioreactor at the optimum temperature for growth of the organism of 37°C.
The aeration at 37°C was not sufficient to maintain the bed temperature at a value suitable for initial growth. As a result, there was a long lag period, with bed temperatures below 30°C (Fig. 8.4(a)). This was especially true for the baffled drum, for which heat transfer between the bed and surroundings was more efficient. In the unbaffled drum the temperature was slightly higher during the lag phase and, as a result, the lag phase was slightly shorter.
The temperature then increased, over a period of 10 h, to values around 45°C. In the baffled drum the temperature then decreased quickly again. The bed was at temperatures of above 40°C for only 10 h. In the unbaffled bioreactor the temperature remained at values above 40°C for 30 h.
Peak O2 consumption rates were higher in the baffled drum, as shown by the greater slope of the cumulative O2 uptake profile in Fig. 8.4(a). However, due to the longer lag phase, the O2 uptake was not significantly better in the baffled drum. Note that the baffled drum would have outperformed the unbaffled drum over the first 30 h, if the bed temperature in both drums had been maintained at 37°C during the first 10 h. This could be achieved with a water jacket or by placing the bioreactor in a 37°C room. Alternatively, it might be sufficient simply to insulate the outer surfaces of the bioreactor during the early stages of the fermentation, such that heating of the bed by the inlet air would be more efficient. Obviously, such insulation would need to be removed once rapid growth began.
These results, obtained at pilot scale, demonstrate a major challenge to be overcome in rotating-drum bioreactors, namely the adequate removal of the waste metabolic heat from the substrate. For example, in the fermentations described above that were undertaken with R. oligosporus, it was highly desirable to avoid temperatures above 40°C, but this value was exceeded for long periods.
Stuart (1996) grew Aspergillus oryzae on wheat bran in the same 200-L bioreactor, without baffles. Performance in terms of O2 consumption was significantly better at 9 rpm than at 2 rpm (Fig. 8.4(b)). This is most likely due to the effect of the rotational speed on the effectiveness of the mixing within the bed. At 2 rpm the bed slumped within the bioreactor while at 9 rpm there was a tumbling flow regime (flow regimes in unbaffled rotating-drum bioreactors are discussed in more detail in Sect. 8.4.1). The maximum temperatures reached during the fermentations were about 43°C at 9 rpm and about 38°C at 2 rpm. The higher temperature occurred at the higher rotational rate due to better mixing, which allowed better O2 transfer from the headspace into the bed, which in turn allowed faster growth.
Was this article helpful?