Info

20 40

20 40

Fig. 24.5. (a) Temperature as a function of time and position for a superficial velocity (VZ) of 0.1 m s-1, twice as high as the value used to produce Fig. 24.2(a). All other parameters are the same as in Table 24.1. For ease of comparison, the same Y-axis range has been used as was used in Fig. 24.2(a). (b) Growth as a function of position for a superficial velocity (VZ) of 0.1 m s-1

Effect of bioreactor height and fungal specific growth rate. Obviously, with all operating conditions held constant, the height of the bioreactor will affect the maximum temperature reached, due to the unavoidable presence of an axial temperature gradient. In turn, this will affect the performance of the bioreactor. Figure 24.6 shows how the bioreactor height affects the time for the average biomass concentration (i.e., averaged over the whole bed) to reach 90% of the maximum biomass concentration. This time is denoted as t90: the larger the value of t90, the poorer the performance of the fermentation. This criterion is used to compare bioreactors since, for logistic growth kinetics, over a wide range of microbial and system parameters, the productivity of the fermentation, in terms of g-biomass m-3-bioreactor h-1, reaches a maximum when the biomass reaches around 90% of its final value (Mitchell et al. 2002b). In fact, t90 is inversely proportional to the productivity. The simulations were done for different specific growth rates.

The value of t90 increases approximately linearly with bioreactor height. This occurs because the greater the height, the greater the average deviation of the temperature from the optimum for growth. The value for zero height is the time that it would take for the biomass to reach 0.9Xm if the whole of the bioreactor remained at the optimum temperature for growth throughout the entire period.

Summary of strategies for optimizing the operation of traditional packed beds. The previous sections involved only one-by-one changes of variables. Obviously it is possible for more than one variable to be changed simultaneously. Simulations will not be shown for simultaneous changes (readers can use the program supplied with this book to undertake their own explorations), but, in general terms, to improve the performance of a traditional packed-bed, it is necessary to

Fig. 24.6. The time taken for the average biomass concentration to reach 90% of Xm (t90), as a function of the bed height within the packed-bed bioreactor, shown for various values for the parameter ^iopt. Key: (•) /xop, = 0.1 h-1 (A) /xop, = 0.236 h-1 (■) /xop, = 0.5 h-1

Height (m)

Fig. 24.6. The time taken for the average biomass concentration to reach 90% of Xm (t90), as a function of the bed height within the packed-bed bioreactor, shown for various values for the parameter ^iopt. Key: (•) /xop, = 0.1 h-1 (A) /xop, = 0.236 h-1 (■) /xop, = 0.5 h-1

decrease the height of the bed within the bioreactor, increase the superficial velocity and use a control system to reduce the temperature of the inlet air in response to temperature increases in the outlet air. Note that, in order to minimize operating costs, it would be preferable not to have to refrigerate the inlet air.

The implications of changes in bed height might need to be considered either at the bioreactor design stage or in an attempt to optimize the performance of a bio-reactor that has already been built. At the design stage, decreasing the height while maintaining the superficial velocity constant means that the bioreactor will need to be wider to hold the same amount of substrate, occupying more floor space. Once a bioreactor is built, decreasing the height will mean that the unutilized volume within the bioreactor will increase and therefore the volumetric productivity of the bioreactor will fall, if the calculation is based on the total bioreactor volume and not the bed volume. Therefore, provided the aeration system has the capacity, it would be preferable to increase the superficial velocity than to decrease the bed height, although problems may occur with high pressure drops.

A model similar to the one used in the simulations above was used by Ashley et al. (1999) to investigate whether reversing the airflow direction would help to overcome the problem of overheating at the top of the bed. Figure 24.7 shows that the model predicts that indeed such a strategy will prevent the temperature at the ends of the beds from reaching deleteriously high values. Unfortunately, it is not a useful strategy since the cooling of the middle sections of the bed is very inefficient, allowing them to reach very high temperatures.

Insights into scale-up of traditional packed-beds gained from modeling work.

If you are considering using a traditional packed-bed bioreactor due to the inability of your microorganism to tolerate agitation, then, on the basis of the results in bed

Was this article helpful?

0 0

Post a comment