We see in Section 3.2.7 that the kinetics of growth of autotrophic nitrifying bacteria can be represented in the same manner as that of heterotrophic bacteria. Consequently, the general conclusions derived in Chapter 5 about biomass growth in CSTRs are equally applicable to them. We also see in Section 3.2.10 that the values of their kinetic coefficients are quite different from those for heterotrophs. This means that the specifics of their behavior in a given reactor environment will differ somewhat from that of the heterotrophs. In this section, we will investigate some of the characteristics of nitrifying bacteria that require special recognition and see how autotrophic and heterotrophic bacteria might influence one another when grown in the same bioreactor.
6.3.1 Special Characteristics of Nitrifying Bacteria
Comparison of the typical (1 values for heterotrophic and autotrophic bacteria in Table 6.3 shows that the value for autotrophs is almost an order of magnitude lower than that for heterotrophs, suggesting that the minimum SRT for nitrifying bacteria is almost an order of magnitude larger. As a consequence, they can be lost from bioreactors under conditions that allow heterotrophic bacteria to grow freely. This situation is aggravated by the fact that nitrifying bacteria are more sensitive to low temperatures and low dissolved oxygen concentrations, as will be discussed later in this section. Therefore, special consideration must be given to the choice of the SRT in systems containing autotrophic bacteria and it cannot be assumed that conditions suitable for the removal of soluble organic matter are suitable for the conversion of ammonia-N to nitrate-N.
Another characteristic of nitrifying bacteria is that their half-saturation coefficient is very low; the typical value given in Table 6.3 is 1.0 mg/L as N. It will be recalled that the half-saturation coefficient is the substrate concentration allowing bacteria to grow at half of their maximal rate. This means that CSTRs containing autotrophic bacteria will have low ammonia-N concentrations even when the bacteria are growing relatively fast and that very low concentrations will result whenever the SRT is large enough to ensure stable growth. It also means that the ammonia-N concentration will rise rapidly as the SRT is decreased to the point of washout. As a consequence, nitrification has gained the reputation of being an all-or-none phenomenon. In other words, since the nitrogen concentration entering municipal wastewater treatment systems is on the order of 30 to 40 mg/L, the percent nitrification approaches 100 whenever the SRT is long enough to give stable growth and rapidly falls to zero as washout occurs. This is illustrated in Figure 6.4,~ which includes data from the literature as well as the results of steady-state simulations with a model similar to the one employed herein.
There are some occasions in which the all-or-none phenomenon will not occur or will not be as drastic as that shown in Figure 6.4. One is when the bioreactor is subjected to diurnal loading, like that shown in Figure 6.2. It will be recalled from Section 6.2.2 that the flow-weighted average effluent substrate concentration from a CSTR receiving a diurnal input is not as low as the concentration from the same reactor receiving a constant input. The degree of difference between the two responses depends on the bioreactor SRT, as shown in Figure 6.5," which was prepared by simulation with a model similar to the one presented herein. It also shows that complete nitrification can still be achieved in a CSTR receiving a diurnal input, but that longer SRTs are required. Another situation that can lead to incomplete nitrification is when the influent contains such a high ammonia-N concentration that both substrate and product inhibition can occur.""" In that case the attainment of complete nitrification requires the use of multireactor systems in which denitrification can be used to reduce the product concentration, i.e., nitrite and nitrate. As pointed out earlier, substrate and product inhibition effects cannot yet be reliably modeled. Con-
Was this article helpful?