Estimation of Kinetic Parameters

Aerobic Growth of Heterotrophs. Because of all of the terms in ASM No. 1 and the generation of soluble substrate from slowly biodegradable substrate, it is not possible to use the approach described in Section 8.2 to obtain |iH and Ks. Conse quently, an alternative approach must be used. The batch techniques described in Section 8.4 may be applied using the soluble fraction of the wastewater, although as discussed in that section, it is currently unclear whether intrinsic or extant parameter estimates are more suitable when using the model. Intrinsic kinetic parameter estimates can be obtained by using coagulated and filtered wastewater in batch experiments like those described in Section 8.4.2. In this case, however, the substrate will be the readily biodegradable organic matter rather than a single organic compound. As a result, the Ks value obtained will be larger than that associated with single compounds. If the value of Sso used in the batch test is larger than the resulting Ks value, then the parameters can be considered to be intrinsic. If this condition is not met, the resulting parameter values will be undefined and should not be used. Because the main function of (1„ and Ks in ASM No. 1 is to allow the maximum oxygen uptake rate of an operating bioreactor to be calculated, it may be better to use one of the extant parameter techniques described in Section 8.4.3 to obtain |iH and Ks, provided that the biomass is taken from a bioreactor with a short SRT. The biomass concentration used in the estimation of the specific rates should correspond to that of the active heterotrophic biomass, XBH. Sufficient information in the form of stoichiometric and kinetic parameters is available at this point to allow its estimation for the bioreactor from which the biomass in the test is obtained, thereby allowing its estimation for the batch test reactor. Substrate injections should be made with coagulated and filtered wastewater, thereby providing only readily biodegradable substrate during the test.

Aerobic Growth of Autotrophs. As far as the design and control of nitrifying bioreactor systems are concerned, the maximum specific growth rate coefficient for autotrophic biomass, |1A, is the most critical parameter value. There are two reasons for this. First, it determines the SRT at which nitrifying biomass will be lost from the system, thereby fixing the minimum acceptable SRT at which the system can be operated. Second, it can be affected strongly by chemicals in the wastewater; more strongly, in fact, than the half-saturation coefficient. Thus, it is important that an accurate assessment of |iA be obtained in an environment that represents the wastewater undergoing treatment. Several procedures have been proposed for measuring (1A,H lh 2" but the simplest involves a batch experiment started with a small amount of biomass. Effluent should be collected from one of the continuous CSTRs, preferably one with a short SRT so that little nitrification will have occurred, making the initial concentrations of nitrate-N and nitrite-N small. If necessary, ammonia-N should be added to the bioreactor to bring the concentration to approximately 40 mg/L. The bioreactor should then be seeded with biomass from a bioreactor with an active nitrifying population, but the initial nitrifying biomass concentration should be less than 1.0 mg/L. Because the production of nitrate-N and nitrite-N is proportional to the amount of nitrifying bacteria formed, the change in the concentration of oxidized nitrogen (NO,-N + NO:-N) can be used to estimate p-A.1'"1 The concentration of oxidized nitrogen in the bioreactor should be measured over time as it increases through growth of the autotrophs and the natural log of (NO, -N + NO; -N) should be plotted versus time. The plot should give a straight line, with a slope of |iA -b, A. The value of b, A is an assumed value, as indicated in Table 8.1, and thus the value of |iA can be calculated. Examples of this procedure may be found in the literature."4

Although the half-saturation coefficient for nitrifying bacteria, KNM, is not affected as strongly by organic contaminants as the maximum specific growth rate, it may be influenced and should be determined. This may be done with the extant kinetic parameter technique of Lamb et al." and Chudoba and colleagues4 s discussed in Section 8.4.3. Biomass should be removed from a CSTR that is fully nitrifying and placed into a respirometer without dilution. Small quantities of ammonia-N should be injected and the net respiration rate measured in response to the injections. This provides data on OUR as a function of injected ammonia-N concentration which can be analyzed as discussed above for heterotrophic biomass. However, in this case the concentration of nitrifying biomass is unknown so that SOUR values cannot be calculated. As a result, the analysis will yield only a value of the maximum OUR in addition to the value of KNII. However, the value of KNH is valid for use with the value of (1A determined from the batch test described in the preceding paragraph.

Decay of Heterotrophs. The decay coefficient, b, H, is very important to predictions of biomass production and oxygen requirements, so it must be determined for the particular wastewater under study. Therefore, biomass should be removed from one of the CSTRs and used in the batch procedure of Section 8.3.2 to determine the traditional decay coefficient, b„, which can be used to determine X,,, as discussed in Section 8.5.2. Biomass can be removed from any of the CSTRs, but correction for the effects of nitrification will be easier if the biomass is fully nitrifying as discussed in Section 8.3.1. Once the value of b„ has been obtained it can be used to calculate bI H with Eq. 3.69:

The value of f,'> should be assumed to be 0.08.

Correction Factors for Anoxic Conditions, t|,, and r|(1. Two important parameters in ASM No. 1 are t^ and % because they correct the rates of growth and hydrolysis reactions when they occur under anoxic conditions. As discussed in Section 6.1.2, this is required because only a portion of the biomass will be capable of functioning under anoxic conditions and the model needs some way to reflect that fact. Tests to measure t^ and t|s are performed at the same time by evaluating oxygen and nitrate consumption rates in two batch bioreactors which are equivalent in every respect except for the terminal electron acceptor.'" The biomass for the tests should come from the MLE bioreactor run as part of the parameter estimation study because it will contain biomass capable of functioning under both aerobic and anoxic conditions. The rationale for the test is as follows. Immediately after biomass is brought into contact with wastewater in a batch bioreactor, the activity in the bioreactor will be dominated by growth of the heterotrophs on the readily biodegradable substrate. However, as soon as the readily biodegradable substrate is exhausted, the activity will be predominantly due to the use of substrate arising from hydrolysis of the slowly biodegradable substrate. Therefore, by comparing the activity of a biomass sample in both of these regions under both aerobic and anoxic conditions it is possible to estimate rip and iqh.

Conceptually, the experiment is very simple. Biomass is removed from the MLE system and placed into two batch bioreactors, one of which is maintained under aerobic conditions with oxygen as the terminal electron acceptor and the other of which is kept under anoxic conditions with nitrate as the terminal electron acceptor. The latter bioreactor should be constructed to minimize oxygen transfer to the liquid. Wastewater is then added to both bioreactors and the OUR and nitrate utilization rate (NUR) are measured in the appropriate bioreactors as the substrate is depleted. The OUR is normally measured by placing a dissolved oxygen (DO) probe in the aerobic bioreactor. By providing a mechanical mixer to keep the biomass in suspension, it is possible to turn off the air supply periodically and measure the OUR directly by the rate of decrease in the DO concentration over a short time period. The NUR is normally measured by manually removing samples from the bioreactor over time, stopping the reaction, removing the biomass, and measuring the nitrateN concentration. The NUR in the two reaction regions is then determined from the slopes of the plot of nitrate-N over time as illustrated in Figure 8.8. Care should be exercised in the measurement of the NUR to ensure that nitrite is not accumulating in the bioreactor. If it is, then the results should be expressed in terms of the net amount of electron acceptor used expressed as the equivalent amount of oxygen used. Care should also be taken to ensure that an appropriate substrate to biomass (F/M) ratio is used in the tests. Figure 8.9 illustrates why this is important in terms of the OUR. If the F/M ratio is too low, the time required for readily biodegradable substrate removal will be too short to get a good measure of the rate. Conversely, if the ratio is too large, the difference in rate between the two zones will not be sufficiently large to clearly distinguish between them. The value of r^ can be calculated once data are available for OURtt and NUR,,:

Figure 8.8 Nitrate utilization in a batch reactor. The faster rate corresponds to use of readily biodegradable substrate (NURB) whereas the slower rate corresponds to use of slowly biodegradable substrate (NUR„). (From S. W. Givens, et al.. Biological process design and pilot testing for a carbon oxidation, nitrification and denitrification system. Environmental Progress 10:133-146, 1991. Copyright © American Institute of Chemical Engineers; reprinted with permission.)

Time, min

Figure 8.8 Nitrate utilization in a batch reactor. The faster rate corresponds to use of readily biodegradable substrate (NURB) whereas the slower rate corresponds to use of slowly biodegradable substrate (NUR„). (From S. W. Givens, et al.. Biological process design and pilot testing for a carbon oxidation, nitrification and denitrification system. Environmental Progress 10:133-146, 1991. Copyright © American Institute of Chemical Engineers; reprinted with permission.)

Figure 8.9 Effect of changing the substrate to biomass ratio (F/M) on the OUR in a batch reactor. The faster rate corresponds to use of readily biodegradable substrate (OUR,.) whereas the slower rale corresponds to use of slowly biodegradable substrate (OUR,,). (From G. A. Ekama, et al.. Procedures for determining influent COD fractions and the maximum specific growth rate of heterotrophs in activated sludge systems. Water Science and Technology 18(6): 91-114, 1986. Copyright © Elsevier Science Ltd.; reprinted with permission.)

Time, min

Figure 8.9 Effect of changing the substrate to biomass ratio (F/M) on the OUR in a batch reactor. The faster rate corresponds to use of readily biodegradable substrate (OUR,.) whereas the slower rale corresponds to use of slowly biodegradable substrate (OUR,,). (From G. A. Ekama, et al.. Procedures for determining influent COD fractions and the maximum specific growth rate of heterotrophs in activated sludge systems. Water Science and Technology 18(6): 91-114, 1986. Copyright © Elsevier Science Ltd.; reprinted with permission.)

Likewise, the value of t|h can be calculated from OURh and NURh: 2.86 X NURh

Hydrolysis and Ammonification. Three parameters remain to be evaluated. Two are the parameters characterizing hydrolysis of slowly biodegradable substrate, kh and Kx. The third is the parameter describing ammonification, k,. These parameters can be evaluated with data collected with the CSTR receiving feed conforming to a daily cyclic square wave pattern."' This is done by nonlinear curve fitting of ASM No. 1 to the OUR data collected during one feed cycle. All other parameter values are known or assumed. The OUR pattern for such a bioreactor was shown in Figure 8.7. The plateau in the OUR after feed cessation is due to degradation of organic substrate released by hydrolysis of slowly biodegradable substrate. The existence of a sustained plateau is evidence that the biomass is saturated and that hydrolysis is occurring at the maximum rate, thereby allowing evaluation of kh. Operation of the bioreactor at an SRT of two days ensures that this condition will exist. The pattern by which OUR declines with time is determined by Kx. The best way to estimate kh and Kx is by curve-fitting techniques to match the OUR response of the model to the OUR pattern in Figure 8.7. Because of the short SRT involved, nitrification will not be occurring. Furthermore, since the bioreactor is fully aerobic, denitrification need not be considered. Thus, a simplified form of the model may be used. Since all of the other parameters have been selected, the only unknowns for the curve-fit are kh and Kx. In addition, because nitrification is excluded from the bioreactor, ammonia will build up as ammonification occurs. Consequently, estimation of ka can be based on the release of ammonia during the nonfeed period.1"

8.5.4 Order of Determination

Evaluation of the parameters and the wastewater characteristics must proceed in a particular order because the values of some are needed before others can be obtained. Table 8.2 summarizes the order of their determination.

Table 8.2

Parameters and Characteristics Which Must Be Evaluated and Information

Needed

0 0

Post a comment