Biological Phosphorus Removal Potential Test

BPR Potential Test

Experimental observations by Wentzel et al. (1985) and Abu-ghararah and Randall (1991) indicated the magnitude of biological excess phosphorus uptake was strongly linked to the magnitude of phosphorus release in the anaerobic reactor. The ratio of phosphorus uptake to release has been found to be relatively constant. Using this ratio together with the phosphorus release obtained from a batch test will herein be called the BPR potential test.

Phosphorus Uptake/Release Ratio

A series of pilot-scale experiments was conducted by Wentzel et al. (1985) with the modified UCT process at sludge ages of 8, 10, 15, and 20 days with a raw municipal wastewater having an influent COD of approximately 500 mg COD/L. As shown in Figure 6, there was a close linear relationship between phosphorus release and uptake, with the slopes having nearly the same magnitude irrespective of sludge age (q_c). The ratio of phosphorus uptake to phosphorus release ranged from 1.15 to 1.2.

Figure 6. Relationship between phosphrous release and uptake under different sludge ages.

Abu-ghararah and Randall (1991) conducted pilot-scale tests using the UCT process. Seven different SCFAs (formic acid, acetic acid, propionic acid, butyric acid, isobutyric acid, valeric acid, and isovaleric acid) were separately added to the influent wastewater, and their effects on the performance of the system were observed. The actual anaerobic hydraulic retention time was 2.1 hours, and the mixed liquor volatile suspended solids (MLVSS) concentration in the anaerobic zone was 1,300 mg/L. The recycling ratio from the anoxic zone to the anaerobic zone was unity. The average contents of phosphorus in MLVSS were 4.3, 9.8, 8.2, 7.4, 7.0, 6.9, and 6.1% for the seven different SCFAs, respectively. A linear relationship with a correlation coefficient of 0.99 and a slope of 1.2 was found between phosphorus uptake and phosphorus release. The results indicated the ratio of phosphorus uptake to phosphorus release was a constant of 1.2.

Rubens (1994) and Karlovich (1994) operated A2O and modified UCT processes, respectively, to treat a municipal wastewater. Both studies showed a good linear relationship between phosphorus release and uptake with the correlation coefficient of 0.99. Table 2 summarizes the linear relationship between phosphorus release and uptake from the literature. The slightly lower values of the ratio in the studies of Rubens (1994) and Karlovich (1994) were thought to be more the result of a low influent phosphorus concentration rather than the actual phosphorus uptake potential. It is believed that the ratio of phosphorus uptake to release ranges from 1.15 to 1.2. As a conservative estimate of phosphorus uptake, 1.15 is thought to be a reasonable value.

Relationship Between Phosphorus Uptake and Release

By accepting the observed linear relationship and postulating that the phosphorus value of the intercept on the uptake axis is equal to the basic metabolic phosphorus requirements of the total organism mass, Wentzel et al. (1985) proposed the following relationship for phosphorus release and phosphorus uptake:

where

P(uptake) = total phosphorus uptake, mg TP/L;
a = a constant ranging from 1.15 to 1.2;
P(release) = released phosphorus during anaerobic conditions, mg P/L; and
P(metabolic) = phosphorus requirement for removing BOD, mg P/L.

Practical Application

The magnitude of released phosphorus is strongly dependent on the characteristics of the wastewater of interest. Using the measured amount of phosphorus release under the anaerobic condition, a conservative P uptake/P release ratio of 1.15, and a ratio for metabolic phosphorus requirement of BOD/P, which is a function of sludge age (Benefield and Randall 1980), the feasibility of BPR can be evaluated. Therefore, the soluble effluent phosphorus concentration can be predicted as follows:

For example, if the BOD and phosphorus values of the wastewater of interest are 230 and 6 mg/L, respectively, and the amount of phosphorus release measured with the BPR potential test is 20 mg/L, then the predicted excess uptake phosphorus will be 23 mg/L (20 ´ 1.15 = 23), and the metabolic phosphorus requirement will be 2 mg/L at the sludge age of 5 days. Finally, the removed phosphorus will be 5 mg/L (23 + 2 - 20 = 5), and consequently, the effluent phosphorus concentration will be 1 mg/L (6 - 5 = 1).

Proposed BPR Potential Test

The following procedure is used to measure the magnitude of released phosphorus, which is influenced by the characteristics of the wastewater of interest:

1. Prepare two batch reactors, each containing 2 L of the wastewater.
2. Add 25 mg COD/L of sodium acetate to one reactor only.
3. Add 2 L of PAO-containing activated sludge taken at the end of the aerobic phase to each reactor and have the MLVSS concentration at about 1,400 mg/L (or at the level that an actual WWTP will target) and pH at about 7.
4. Obtain filtered samples every 15 minutes during the 2 hours in the anaerobic stage.
5. Examine the phosphorus release rate.
6. Evaluate the phosphorus release rate obtained from the one with acetate addition for potential inhibition of PAOs by an unknown compound(s) in wastewater. (If there is no indication of industrial wastewater input, this step can be skipped; instead, a duplicate test can be conducted.)
7. Evaluate whether BPR is feasible by determining the potential phosphorus uptake and effluent phosphorus concentration.

Figure 7 shows the BPR potential test procedure. If recycle streams contain a high level of phosphorus, this should also be taken into account by increasing the initial phosphorus concentration.

Figure 7. BPR potential test procedure.

BPR Potential Test Results

Five wastewater samples from the Oakfield, Green Lake, Ashland, Campbellsport, and Green Bay WWTPs were used for this test. The Green Bay WWTP is an advanced secondary plant with a total design flow of about 35 MGD. The plant receives significant industrial wastewater containing high soluble BOD and low phosphorus. Treatment includes primary sedimentation, anoxic selector basins, nitrifying activated sludge, and final clarification. Primary sludge and secondary sludge are mixed, dewatered, and incinerated. The Campbellsport WWTP treats primarily domestic wastewater. The design flow is 0.47 MGD, and the plant currently treats about 0.25 MGD. The plant is an oxidation ditch with sand filtration and aerobic sludge digestion. The Ashland WWTP treats mostly domestic wastewater in an oxidation ditch with aerobic sludge digestion. The design flow is 1.92 MGD. The Green Lake WWTP has a design flow of about 0.282 MGD and currently treats about 0.2 MGD of primarily domestic wastewater. The plant is a complete-mix activated sludge plant with aerobic digestion. The grab samples were taken from raw wastewater except for the Green Bay sample, which was taken at the primary sedimentation tank effluent channel. The characteristics of each grab wastewater sample are summarized in Table 3. According to the commonly used criteria—the values of BOD/P and COD/P—all five wastewater samples are considered adequate to achieve an effluent phosphorus concentration < 1 mg TP/L when a BPR process is employed.

BPR potential test results are shown in Figures 8 to 12. Oakfield, Green Lake, and Ashland wastewater samples showed poor phosphorus release, while the acetate-added wastewater samples had relatively high phosphorus release. This indicates there was insufficient readily biodegradable soluble COD in these wastewater samples. The BOD/COD ratios of the Oakfield, Green Lake, and Ashland wastewater samples were 0.24, 0.4, and 0.42, and the readily biodegradable COD fractions were 0.26, 0.44, and 0.18, respectively. In a previous study of Ashland’s wastewater, four 24-hour composite samples were characterized over a 1-year period. The readily biodegradable soluble COD fraction ranged from 0.19 in the winter to 0.3 in the summer (Park et al. 1997). Although the grab sample used for the BPR potential test showed poor phosphorus release, it may still be possible to adapt BPR processes, considering the readily biodegradable soluble COD fractions measured using composite samples taken for the earlier study by Park et al. (1997). We highly recommend that the BPR potential test be conducted with several composite samples taken during various seasons to represent the wastewater characteristics over a period of a year and that consideration be given to recycle streams.

Figure 8. Phosphorus release profile for Oakfield Wastewater Treatment Plant.

Figure 9. Phosphorus release profile for Green Lake Wastewater Treatment Plant.

Figure 10. Phosphorus release profile for Ashland Wastewater Treatment Plant.

Figure 11. Phosphorus release profile for Campbellsport Wastewater Treatment Plant.

Figure 12. Phosphorus release profile for Green Bay Wastewater Treatment Plant.

The Campbellsport wastewater samples with and without acetate had similar phosphorus release for the first 15 minutes; however, after 15 minutes the phosphorus release for the sample without acetate was much lower than that with acetate. This indicates that there was insufficient readily biodegradable soluble COD despite the high readily biodegradable soluble COD fraction of 0.48. In the case of the Green Bay wastewater samples, the sample without acetate had phosphorus release similar to that with acetate for the first 30 minutes and then had slightly lower release than that with acetate. Due to the high readily biodegradable soluble COD fraction of 0.57, there appeared to be sufficient readily biodegradable soluble COD required for the growth of PAOs in the anaerobic zone.

The total phosphorus removal for each wastewater sample can be calculated using Equation 2. The results are summarized in Table 4.

^a The sludge age of 5 days was used; thus, BOD:P = 115:1
^b NR 217 of the Wisconsin Administrative Code requires that all existing wastewater treatment plants that discharge in excess of 150 pounds of total phosphorus per month to surface water must meet an effluent total phosphorus limit of 1 mg/L. If an enhanced biological phosphorus removal process is used, 90% of the total phosphorus required to meet 1 mg/L of total phosphorus in the effluent must be removed.

Randall et al. (1992) showed that if BOD/P is > 20~30 or COD/P is > 35, BPR is, in general, feasible. However, from Table 4, only the Green Bay wastewater would meet the phosphorus effluent permit. The Green Bay WWTP is currently removing phosphorus to below 0.5 mg TP/L by employing an anaerobic selector, originally installed for bulking control, ahead of the aeration basin. In the five wastewaters we studied, the BOD/P and the readily biodegradable COD fraction for Green Bay were the highest because the Green Bay WWTP receives wastewater from industries that contribute large amounts of soluble BOD with low phosphorus. The BPR test result of the Campbellsport wastewater showed that 5.2 mg/L of phosphorus could be removed. However, it may not meet the effluent phosphorus limit as the total phosphorus concentration in influent was 8.2 mg TP/L. The BPR potential test result for Ashland wastewater showed that 2.7 mg TP/L could be removed, but it could not meet the discharge limit because of the high influent total phosphorus concentration of 6.5 mg TP/L. A previous study found the phosphorus concentration in Ashland wastewater ranged from 2.9 to 5.7 mg TP/L (Park et al. 1997). The BPR potential test results of the Green Lake and Oakfield wastewater samples showed that almost no excess phosphorus could be removed due to low phosphorus release. This is assumed to be because fermentation of raw wastewater does not occur in a collection system with a short travel time.

More information on this topic: Gerry Novotny