This is a relatively high value, but necessary to meet the process goals. Normally, for economic final settler performance, the solids recycle ratio, a, typically lies between 0.5 and 1.0. This means that the MLR ratio, 3, should be at least 4.38. If a value of 4.5 is chosen, the MLR flow rate must be (4.5) (40,000 m'/day) or 180,000 mVday. A final decision can be made only after the requirement for summer operation is determined. This will require re-computation of the mass of MLSS in the system in the summer, thereby giving the MLSS concentration associated with the system volume. This fixes the process loading factor for the anoxic zone, which determines the specific denitrification rate. The nitrogen available to the nitrifiers must also be recalculated for 25°C, allowing computation of the mass production rate of nitrate-N. Finally, the information can be combined to determine the fraction of the nitrate-N that is denitrified, allowing a + (3 to be calculated. Determination of the MLR rate needed for summer operation is left as an exercise for the reader.

d. What is the net alkalinity consumption at 15°C?

As calculated in Part a, the concentration of nitrate-N leaving the system in the absence of nitrification would be 33.3 mg/L as N. Its formation would lead to the destruction of (7.23)(33.3) = 241 mg/L as CaCO, of alkalinity. Since the effluent nitrate-N concentration is 5.7 mg/L as N, the amount of nitrate-N denitrified is equivalent to 27.6 mg/L as N. This would return

(3.5)(27.6) = 96.6 mg/L as CaCO, of alkalinity to the system. Thus, the net alkalinity destruction will be 241 - 96.6 = 144.4 mg/L as CaCO,. Since 150 mg/L of the alkalinity in the influent could be destroyed while maintaining the needed residual of 50 mg/L, it would not be necessary to add chemicals in the winter. This is in contrast to the system with the anoxic selector, for which chemical addition was still required, although in lesser amount than the system without denitrification.

The remainder of the process design would proceed in exactly the same way as for the other systems. The oxygen requirement would have to be calculated for summer conditions and the savings due to denitrification determined in the same way as in Example 11.3.1.2. The heterotrophic oxygen requirement in the aerobic zone will be relatively low because the majority of the degradable organic matter will have been removed in the anoxic zone. This fact, coupled with the fact that the tank is large to accommodate the biomass needed to achieve a long SRT, means that the power requirements are associated primarily with solids suspension rather than with oxygen transfer. Investigation of these issues will be left as an exercise for the reader.

Examples 11.3.1.2 and 11.3.1.3 illustrate that initial anoxic zones of two quite different sizes may be used in systems of the same basic configuration, depending on process objectives. If an anoxic selector is used to remove readily biodegradable organic matter, then a relatively small anoxic zone results due to the rapid utilization of that substrate. Meaningful removal of nitrate-N will occur as a result of the presence of an anoxic selector, but the effluent nitrate-N concentration may still be significant. Relatively modest MLR rates may be adequate for these applications because the fraction of the nitrate-N that must be recirculated to the selector is fairly modest. Furthermore, a noticeable reduction in the oxygen requirement in the aerobic zone and in the net alkalinity consumption will occur. The reduction in power requirements associated with the decrease in the oxygen requirement will partially or completely off-set the power required to mix the selector and to recirculate mixed liquor. A larger initial anoxic zone may be required if the objective is to meet a specified effluent nitrate-N concentration through use of the MLE process. This occurs because both readily and slowly biodegradable organic matter may be required as electron donors to remove the larger mass of nitrate-N, and utilization of slowly biodegradable substrate results in slower denitrification rates. The greater degree of niiraie-N utilization in such applications results in increased MLR flow rates since a higher fraction of the nitrate-N formed must be recirculated to the initial anoxic zone. Greater reductions in the oxygen requirement and the associated power requirement in the aerobic zone occur, and net alkalinity consumption is reduced more. However, the further reduction in the power requirement for oxygen transfer may be small in comparison to the increase in power required to mix the larger anoxic zone and to pump the increased MLR flow.

Four-Stage Bardenpho Process—Addition of Second Anoxic and Aerobic Zones. A second anoxic zone, such as in the four-stage Bardenpho process (Figure 11.5), may be used to further reduce the nitrate-N concentration below that which can be economically achieved using an MLE system. However, because essentially all of the readily and slowly biodegradable organic matter will have been removed in the initial anoxic and aerobic zones, the primary source of electrons for the ad-

dilional denitrification is biomass decay. Since the decay of heterotrophs is a relatively slow process, particularly under anoxic conditions, and since not all of the heterotrophic bacteria are capable of denitrification, the resulting specific rate of denitrification will be low. This means that the size of a second anoxic zone can be significant. In addition, a small aerobic zone is required to prepare the MLSS for settling. Typically, an HRT of 30 min is used, which is sufficient to strip entrained gases from the denitrified mixed liquor exiting the second anoxic zone and to add dissolved oxygen to it before it enters the clarifier. The MLSS concentration is uniform throughout the four-stage Bardenpho system, making the SRT in each zone directly proportional to its volume.

As with the design of an MLE system, the best way to estimate the impact of adding second anoxic and aerobic zones is through the use of simulation with a model like ASM No. I. Care should be exercised in the selection of the anoxic hydrolysis factor, t)h, because it will have a large impact on the rate of nitrate-N utilization associated with decay in the anoxic zones.

In the absence of a kinetic characterization sufficient to allow simulation, or as a prelude to simulation, it is possible to roughly estimate the nitrate-N requirement in the second anoxic zone by partitioning the electron acceptor requirement due to decay in much the same way that the oxygen requirement due to decay was partitioned to the various reactors in a CAS system, as illustrated in Section 10.3.4. The major uncertainty associated with this is the effect that the anoxic conditions have on the rate of decay.

An alternative to estimating the denitrification rate due to decay is to take an empirical approach. As they did for the specific denitrification rate in the first anoxic zone. Burdick, et al.'" have reported an empirical relationship for the specific rate of denitrification due to decay in a second anoxic zone, qM,xu. In this case, since no substrate enters the second anoxic zone, they were able to correlate the specific denitrification rate with the system SRT:

This relationship indicates that the specific denitrification rate will decrease as the SRT is increased, as expected. It was developed at a temperature of 20°C, but it can be corrected to other temperatures using Eq. 3.95 with an appropriate 0 value; a value around 1.02 appears reasonable. This relationship has been widely reported and used;" "" and it is appropriate when other data are not available.

As with Eq. 11.11, use of Eq. 11.12 requires an iterative procedure. As with all previous nitrogen removal systems, selection of the system size should be done for winter operating conditions. After the first anoxic and aerobic zones have been sized using the procedures for the MLE system, an SRT is assumed for the second anoxic zone, thereby increasing the system SRT. Since the MLSS concentration is uniform throughout the system, its volume is calculated in proportion to the volume of the first anoxic and aerobic zones as suggested by Eqs. 11.1 and 11.2. Typically, the second aerobic zone is sized to give an HRT of around 30 min, as mentioned previously, thereby increasing the system volume. The SRT of this zone is also calculated by proportion, giving the new system SRT. The new system SRT allows computation of qM)XH with Eq. 11.12 and the mass of MLSS in the system with Eq. 9.11. The mass of MLSS in the second anoxic zone can then be calculated by proportioning with respect to its SRT as a fraction of the total SRT. Multiplication of that mass by qN,,Xn allows calculation of the mass denitrification rate due to biomass decay in the second anoxic zone, ANxlt. Division of that value by the flow rate gives the additional amount by which the effluent nitrate-N concentration is reduced. Subtraction of that value from the MLE effluent nitrate-N concentration will determine whether the required effluent nitrate-N concentration is attained. If it is not, then another SRT must be assumed for the second anoxic zone and the process repeated. Generally, the volume of the second aerobic zone is held constant at a value giving an HRT of about 30 min. The system size determined by this procedure will be approximate because each time the system SRT is increased the amount of nitrogen available to the nitrifiers and the rate of denitrification in the first anoxic zone will change. However, recalculation around the entire system is not justified because of the approximate nature of the empirical relationships used for denitrification in both anoxic zones. As stated earlier, the most accurate method of arriving at a final design is through simulation, provided the needed kinctic information is available.

The presence of a second anoxic zone will increase the amount of alkalinity recovery in the system, which will increase the alkalinity of the final effluent. However, the alkalinity produced in the second anoxic zone will have little impact on the alkalinity in the main aerobic zone because little of it will be recirculated through the system. Thus, it will not change the amount of chemical required to achieve stable nitrification from that required by an MLE system.

The inclusion of a second anoxic zone in a nitrogen removal process will increase the MLR requirements, which should be calculated for summer conditions. This occurs because the second anoxic zone will reduce the nitrate-N concentration entering the secondary clarifier, which reduces the nitrate-N concentration in the RAS flow to the initial anoxic zone, thereby decreasing the mass rate of nitrate-N return. If it is assumed that the nitrate-N concentration in the RAS is zero, Eq 11.8 can be modified to show the effect of the RAS and MLR ratios on the fraction of nitrateN formed in the aerobic zone that is recirculated to the initial anoxic zone:

Likewise, Eq. 11.9 can be modified to allow calculation of the MLR ratio required to allow denitrification of a specified fraction of the nitrate-N in the initial anoxic zone:

1 Isol)

It is important to note that the value of fN(l l) used in Eq. 11.14 is exactly the same as the value used in Eq. 11.9 because the fractional nitrate removal in the first anoxic zone is the same. As noted previously, the MLR flow rate should be calculated for summer conditions because it will be largest then.

The presence of the second anoxic zone will have little effect on the oxygen requirement in the first aerobic zone because the additional decay resulting from the increase in system SRT provides the electrons for denitrification in the second anoxic zone. The oxygen requirement in the second aerobic zone will be low, and thus the aeration rate in it will be governed primarily by the need to keep solids in suspension.

The impact of the additional SRT on the solids wastage rate can be calculated for winter conditions using the same procedures as used in all other designs.

The design of a second anoxic zone is illustrated in the following example.

A second anoxic ¿one is to be added to the MLE process sized in Example 11.3.1.3. It is to reduce the nitrate-N concentration from 6 mg/L as N to 2 mg/L as N at 15°C. The second aerobic zone will have an HRT of 30 min. Assume that the temperature coefficient for the specific denitrification rate in the second anoxic zone is 1.02.

a. What is the required size of the second anoxic zone if the sizes and SRTs of the first anoxic and aerobic zones remain the same as in the MLE system? As a first guess, assume an SRT of 4 days for the second anoxic zone, from tixample 1 1.3.1.3. the SRT of the MLE system is 16 days and its volume is 21,600 m'. Thus, the volume of the second anoxic ¿one, V\NS is given bv proportion:

The second aerobic zone has an HRT of 30 min. Since the influent flow rate is 40.000 m'/day, its volume is 833 m'. The total system volume is:

By proportion, the total system SRT is:

Therefore, the SRT of the second aerobic /one is 0.6 day

Substitution of the system SRT into Eq. 11.12 allows calculation of the specific denitrification rate in the second anoxic zone at 20°C:

qv,M..-.i = (0.12)(20.6) = 0.0142 g NO,-N/(g MLSS-day)

Since the design is being performed for 15"C\ the specific denitrification rate must be corrected to that temperature using Eq. 3.95:

= (0.0142)( 1.02)'" '" = 0.013 g NO,-N/(g MLSS-day)

This must now be multiplied by the mass of MLSS in the second anoxic zone to determine the mass removal rate of nitrate-N in the zone. The mass of MLSS in the system can be estimated with Eq. 9.1 1, as has been done several times, giving a value of 79,000,000 g. Since the SRT in the second anoxic ¿one is 4 days and the system SRI' is 20.6 days, the mass of ML.SS in the second anoxic zone is calculated by proportion to be 15,300,000 g. Therefore, the mass rate of denitrification in the second anoxic zone is:

The mass rate of NO.-N release from the MLE system was calculated in Part a of Example 1 1.3.1.3 to be 226,000 g NO<-N/day. Since the second anoxic zone can remove 199,000 g NO.-N/day, the discharge into the second aerobic zone is 27,000 g NOi-N/day. This corresponds to a concentration of 0.67

mg/L as N. Because the concentration is below the target value of 2.0 mg/ L as N, it is possible that a smaller second anoxic zone could be used. However, several uncertainties exist in the design that suggest that it would be prudent to maintain an SRT of 4 days in the second anoxic zone. First, some additional nitrification will occur in the second aerobic zone on the ammonia-N released by the decay reactions in the second anoxic zone. Second, because of the longer SRT, the mass of MLSS in the first anoxic zone will be somewhat smaller than the amount calculated in Example 11.3.1.3. This means that a little less denitrification will occur there, increasing the mass flow rate of nitrate-N into the second anoxic zone. Consequently, it would be prudent to retain an SRT of 4 days in the second anoxic zone. Of course, if possible, simulations with a model like ASM No. 1 should be done to refine the design.

What MLR flow rate to the first anoxic zone is required for this application at 15°C?

Equation 11.14 is used to perform this calculation. From Part c of Example 11.3.1.3, 83"/{ of the nitrate-N produced in the first aerobic zone must be recirculated to the first anoxic zone for denitrification. Typical solids recycle ratios for applications such as these range from 0.5 to 1.0. Calculate (3 for these two values of ot. For a = 0.5,

These are high values, which may not be practical. The alternative would be to reduce the size of the initial anoxic zone to reduce the fraction of the nitrate-N that would be reduced in that zone. This will necessitate an increase in the size of the second anoxic zone since more nitrate-N must be removed there. The entire computational procedure of this and (he preceding example would have to be repeated to arrive at an estimate of the performance of an alternative system. Doing this several times would provide the information required to choose the optimal system design. Consideration of the effort involved in doing this demonstrates clearly the benefits associated with being able to use simulation to investigate alternative designs.

Simultaneous Nitrification and Denitrification. As discussed in Section 11.1.3, simultaneous nitrification and denitrification can be a significant nitrogen removal mechanism in a nitrifying activated sludge system oxygenated with a point source aerator or in a system with a uniformly low DO concentration, even though the bioreactor does not have a distinct and separate anoxic zone. Denitrification of as much as 50% of the nitrate-N produced has been reported in some applications."4 "'' The occurrence of simultaneous nitrification and denitrification requires three factors:

(1) an oxygen transfer system that allows the development of zones of high and low DO concentration (either on a macroscopic or microscopic scale), (2) control of the oxygen input rate to the process, and (3) a sufficiently long SRT to allow full nitrification to occur even though parts of the bioreactor contain very low DO concentrations.

The impact of an oxygen transfer system that develops zones of high and low DO concentration was discussed in detail in Section 11.1.3. However, even if such a system is being used, regions of low DO concentration will not develop if the potential oxygen transfer rate to the system greatly exceeds the oxygen requirement. Rather, they will only develop when the potential oxygen transfer rate is less than the oxygen requirement. Furthermore, the mass rate of denitrification will be determined by the difference between the oxygen requirement for a totally aerobic system and the mass rate at which oxygen is actually being transferred to the liquid as an electron acceptor, TO:

Was this article helpful?

## Post a comment