Theoretical Performance Of Fluidized Bed Biological Reactors

We saw in Section 18.2 that for a given superficial velocity the expansion of a fluidized bed depends on the size and density of the carrier particles, as well as on the thickness of the biofilm. In addition, for a given degree of expansion (porosity), the number of particles per unit bed volume also depends on those factors. Consequently, the biomass concentration in the bed is influenced by them as well. The effect of biofilm thickness on the biomass concentration was illustrated in Figure 18.10, and similar figures could be generated illustrating that values that maximize the biomass concentrations exist for the carrier particle size and density a)so.:x':'' Thus, from consideration of the effects of fluidization alone, it can be seen that complex interactions exist among the factors that influence FBBR performance. Fluidization effects do not tell the whole story, however. Because of the need for transport of reactants into the biofilm, not all of the biomass has the same activity. That is why the effectiveness factor is less than 1.0. Furthermore, the effectiveness factor depends on the size of the bioparticle and the thickness of the biofilm, as reflected in the modified Thiele moduli as used in Figures 18.11 and 18.12. This suggests that the combination of bioparticle characteristics that maximizes the ability of the FBBR to remove substrate is different from that which maximizes biomass concentration.Because each situation is unique and complex, mathematical models are required for their analysis and several have been developed that integrate submodels of the type discussed in the preceding section.1,2A

The theoretical performance of FBBRs can be examined with those models. The result from one such exercise is shown in Figure 18.13. "' It shows the effect of particle diameter and biofilm thickness on the time required for 90% removal of substrate by biomass with an intrinsic zero-order reaction in an FBBR containing a fixed mass of carrier particles operated with a fixed superficial velocity. The flow regime in the FBBR was characterized as plug-flow. Thus, the required reaction time corresponds to the fractional height in the bed at which 90% of the substrate is removed. In other words, it corresponds to a required bed height and media mass. Examination of the figure reveals that the optimal (smallest) reaction time (and therefore bed size) is associated with moderately thin biofilms growing on small carrier particles. In fact, others have shown that for a given carrier particle size, the bioparticle effectiveness factor is maximized when the biofilm thickness is slightly less than the thickness at which all substrate would be exhausted."4 Consequently, the optimum biofilm thickness depends on the diffusivity of the substrate in the biofilm and the biodégradation kinetics. The benefit of small carrier particles derives directly from the fact that for a given mass of carrier particles, the surface area for biofilm growth increases as the carrier particle diameter decreases. Nevertheless, the curvature associated with the optimal region in Figure 18.13 is relatively shallow in both dimensions, suggesting that the designer has some latitude in selecting a carrier particle size and the desired biofilm thickness, i.e., fluidization conditions. Similar conclusions regarding the relative effects of carrier particle size and biofilm thickness have also been reached with a model assuming first order intrinsic kinetics. 1 Thus, they can be considered to be general.

The above information considers selection of the optimal carrier particle size and biofilm thickness. Once they have been fixed and the mass of carrier particles

Figure 18.13 Combined effects of carrier particle diameter, d,„ and biofilm thickness, L,, on the time required to remove 90% of the influent nitrate in a denitrifying FBBR. (From W. K. Shieh, L. T. Mulcahy, and E. J. LaMotta, Mathematical model of the fluidized bed biofilm reactor. Enzyme and Microbial Technology 4:269-275, 1982. Copyright © Elsevier Science Ltd.; reprinted with permission.)

Figure 18.13 Combined effects of carrier particle diameter, d,„ and biofilm thickness, L,, on the time required to remove 90% of the influent nitrate in a denitrifying FBBR. (From W. K. Shieh, L. T. Mulcahy, and E. J. LaMotta, Mathematical model of the fluidized bed biofilm reactor. Enzyme and Microbial Technology 4:269-275, 1982. Copyright © Elsevier Science Ltd.; reprinted with permission.)

and porosity have been selected, then the required superficial velocity (and associated bed diameter) and bed height become fixed. A question then arises about the performance of the FBBR if the influent flow rate or substrate concentration changes. Some understanding of the response can be obtained by considering what would happen to the quantity of biomass in the system, which depends on how the FBBR is operated. Consider the case in which biomass wastage is practiced to maintain a fixed bed height and the recirculation rate is adjusted to maintain a constant superficial velocity through the bed. If the influent flow rate or concentration was increased while maintaining the same mass input rate of substrate and the same superficial velocity, then the impact on system performance would be minimal because the mass input of substrate per unit of biomass would stay about the same. On the other hand, if the mass input rate of substrate increased, the output substrate concentration would increase by a proportionally greater amount. Because of the increased input, the biomass would be exposed to higher substrate concentrations, which would cause it to grow faster, leading to thicker biofilms. This would cause the degree of expansion in the bed to increase, which would reduce the number of carrier particles associated with the fixed bed height. Even though each carrier particle left in the system would have a thicker biofilm, the mass of biomass in the system would decrease because the increased biomass on each carrier particle would not compensate for the loss of carrier particles. Furthermore, the effectiveness of each carrier particle would be decreased because of the increased film thickness. On the other hand, if the recirculation rate was decreased to maintain the same mass of carrier particles within the prescribed bed height, the increase in the output substrate concentration would not be as great because more biomass could be maintained in the system. Likewise, if the bed height was allowed to expand to accommodate the increased mass input rate of substrate, there would be little impact on performance.

The above suggests that an FBBR can be thought of somewhat like a suspended growth system. If a suspended growth CSTR is operated at a fixed solids retention time (SRT), the mass of biomass in the system increases in proportion to an increase in the influent mass flow rate of substrate, and the effluent concentration remains the same. On the other hand, if it is operated at a fixed mixed liquor suspended solids (MLSS) concentration, the process loading factor (U) increases (the SRT decreases), and the effluent concentration increases. The FBBR acts similarly. If it is operated in a manner that allows the mass of biomass to increase, the impact of an increase in the mass input rate is minimal. Conversely, if the operational practice results in the same or less biomass, performance suffers.

The concept of SRT in a fluidized bed is a helpful one, but one must recognize that the situation is more complex than in a suspended growth bioreactor because of the mass transfer limitations in biofilms.'' Long SRTs can lead to thick biofilms, which have a lower effectiveness factor. Thus, FBBRs with long SRTs can have lower volumetric removal rates. Conversely, at short SRTs, even though the biofilms are thin, the amount of biomass may be insufficient to get good removal. In other words, the fact that there is an optimal biofilm thickness associated with a given particle size means that there is also an optimum SRT for a given situation.

0 0

Post a comment