October 12, 2005

Phosphorus Release in Aerobic Sludge Digestion

By Ju, Lu-Kwang; Shah, Hemant K; Porteous, Jim


The objectives of this study are to examine the phosphorus release in aerobic sludge digestion and to better understand its governing mechanisms. In this study, phosphorus release was examined using the secondary sludge from both conventional and biological nutrient removal processes. The experiments were carried out at room temperature (22 2C), with or without automatic control of pH (4.5 to 7.8), and under three aeration schemes: fully aerobic (dissolved oxygen [DO] at 3 to 4 mg/L), low DO (0.2 to 0.8 mg/L), and cyclic (with alternate on/off aeration). The released phosphorus concentrations were 20 to 80 mg/L for the conventional sludge and 60 to 130 mg/L for the biophosphorus sludge. Higher phosphorus release also occurred at low pH (

KEYWORDS: biological nutrient removal, inorganic phosphorus, organic phosphorus, biophosphorus, polyphosphate, low-dissolved- oxygen digestion, cyclic aerobic-anoxic digestion.



Removal of phosphorus (P) from wastewater is important for preventing eutrophication of lakes, inland seas, and other natural waters. It has been established that, for controlling excessive algae growth in water bodies, the phosphorus concentration has to be kept lower than 0.03 mg/L (Gachter and Imboden, 1985). Chemical precipitation (CP) and biological nutrient removal (BNR) are the two common methods used for phosphorus removal from wastewater. These methods essentially transfer phosphorus from the liquid to the sludge phase, which subsequently needs to be treated and disposed of. The overall success of these methods depends on the fate of phosphorus during the sludge treatment.

Because of the growing environmental concerns, shifting economy, and operational benefits, many conventional wastewater treatment plants are being converted, modified, or retrofitted to BNR processes. In BNR processes, the majority of the nutrients, including nitrogen and phosphorus are removed in the treatment stage. In general, the BNR processes produce less secondary sludge and require lower running costs than the chemical processes for phosphorus removal (Toerien et al., 1990). On the other hand, while the chemically precipitated phosphorus remains with the treated solids (Seyfried et al., 1988), the biophosphorus accumulated in the BNR sludge may cause problems when the sludge is dewatered, thickened, and digested. For example, up to 20 mg/L of phosphorus was released in the overflows from a gravity-thickening process (Pitman et al., 1991). Moreover, 60 to 80% of the phosphorus removed in the BNR processes may be released during anaerobic sludge digestion (Murakami et al., 1987). Under the anaerobic conditions, a portion of this biophosphorus may precipitate in complex forms such as magnesium ammonium phosphate (struvite), depending on the pH and dissolved oxygen (DO) concentration (Jardin and Popel, 1996). The substantial release of phosphorus (and nitrogen) results in contaminated liquors or requires the addition of chemicals (typically lime and/or salts of iron, aluminum, or calcium) to reprecipitate the released phosphorus (Hwang and Choi, 1998). In addition, struvite precipitation during anaerobic digestion tends to cause blockage of the digester's outlet pipes. These problems significantly compromise the overall performance of BNR and other secondary treatment processes targeting for biophosphorus removal and make the anaerobic digestion questionable in its applicability to such sludge (Pitman, 1999).

While many have examined the phosphorus release during anaerobic sludge digestion, the work in aerobic digestion processes has been scarce. Tonkovic (1999) studied the aerobic digestion of sludge samples taken from seven different plants. Two of the plants used BNR for biological phosphorus and nitrogen removal. One plant was designed for biological phosphorus removal only. Another was to remove nitrogen biologically, by nitrification and denitrification, but remove phosphorus by chemical precipitation. The remaining three plants used conventional activated sludge processes without enhanced BNR. Only the results obtained at the end of seven days of digestion were reported. According to the results, under aerobic digestion, only 0.5 to 21% of the total phosphorus in the sludge from biophosphorus removal processes were released, which was similar to or less than that from the conventional sludge (9 to 28%). More importantly, the phosphorus release from biophosphorus sludge under aerobic digestion appeared to be far lower than the 60 to 80% release reported elsewhere for anaerobic digestion (Murakami et al., 1987).

However, the experiments by Tonkovic (1999) were made under no pH control and for only seven days. The longer-term profiles and the effects of pH, DO, and intermittent aeration (with on/off cycle of aeration) remained unknown. These have been investigated systematically in the current study with sludge samples from both conventional and BNR plants. The results obtained and the responsible mechanisms hypothesized are presented in this work.

Table 1-Experimental design chart.


Sludge Samples. In this study, phosphorus release was examined in batch digestion experiments using the secondary sludge from two municipal wastewater treatment plants: one at the nearby Akron, Ohio, and the other at Ozark, Missouri. The Akron sludge was collected fresh from the returned activated sludge (RAS) of the secondary clarifier and studied within one hour following the collection to minimize the change of sludge properties. With an average flow of approximately 3000 to 3500 L/s (70 to 80 mgd), the Akron plant used aeration ditches for the secondary biological treatment to remove ammonium and organic substances. The primary influent had the following average properties: five-day biological oxygen demand (BOD^sub 5^) = 100 to 110 mg/L, total suspended solids (TSS) = 160 mg/L, and total Kjeldahl nitrogen (TKN) = 20 mg/L. The BOD^sub 5^ of the influent to secondary treatment was 70 to 90 mg/ L. The solids concentration in the Akron plant's secondary sludge was typically approximately 1.5% (w/v). Lime was occasionally added for phosphorus removal and pH adjustment.

The much smaller Ozark plant had a design capacity of approximately 77 L/s (1.75 mgd). It used an anaerobic/oxic (A/O) Ditch process that consisted of a three-stage anaerobic selector (including a RAS denitrification stage and two subsequent anaerobic stages) and a single oxidation ditch. The typical influent values were as follows: BOD^sub 5^ = 300 mg/L, TSS = 350 mg/L, and TKN = 40 mg/L. The solids concentration in the secondary sludge was typically approximately 1.5%. Because the phosphorus release was likely extensive in the anaerobic conditions encountered during the shipping from the plant to our laboratory, the sludge samples received could not be used directly for the current study. Therefore, both samples of waste activated sludge (WAS) and the influent to secondary treatment were shipped overnight in ice packs separately. The sludge was then reacclimatized in the laboratory using a sequential batch operation to simulate the plant condition. The sludge was mixed with the influent water at a ratio of 1:2 (the average ratio used in the plant) and the resultant mixed liquor was subjected to 1.5 h of gentle mixing without aeration, 26 h of aeration, and 7 h of sludge settling. The settled sludge was then remixed with the influent sample from the plant and the cycle of acclimation repeated. The sludge was acclimatized for 3 to 5 cycles before being collected for the digestion experiment.

Digestion Experiments. Twenty experiments of batch digestion were conducted in six runs to investigate the effects on phosphorus release of the following factors: conventional versus BNR sludge, pH, DO, and cyclic aeration (with alternate on/off-aeration). The experimentaldesigns are summarized in Table 1. Three runs (denoted as RlA, R2A, and R3A) were made with the sludge samples from the conventional Akron plant. Another three runs were conducted using the Ozark's BNR sludge, two of which (R2M and R3M) had been acclimated with the plant's secondary influent, one (RlM) with the influent to the secondary clarifier (i.e., the mixed liquor from the oxidation ditch). Depending on the designed experimental conditions, the experiments were done with or without automatic control of pH (ranging 4.5 to 7.8) and under three aeration schemes: fully aerobic (high DO at 3 to 4 mg/L), low DO (0.2 to 0.8 mg/L), and cyclic aeration.

The batch digestion experiments were carried out at room temperature (22 2 C) in 2-L plastic containers. The plastic container had a lid with ports for aeration and sampling and for introducing DO (YSI 5739, Yellow Springs Instruments, Inc., Yellow Springs, Ohio) and pH (Mettler Toledo 465, Columbus, Ohio) probes. For the pH-controlled systems, the base (0.5 N sodium hydroxide [NaOH]) was added by a peristaltic pump, which was activated automatically by a pH controller. The aeration was provided by sparging humidified air through a diffuser placed at the bottom of the reactor. A magnetic stirrer was used for mixing. The stirring speed and aeration rate were adjusted to maintain DO in the range designed for the specific experiment. For fully aerobic digestion, DO was maintained at 3 to 4 mg/L. For the systems of low DO (0.2 to 0.8 mg/L), the aeration rate was reduced significantly, while the agitation was kept sufficient for adequate bulk mixing. The aeration was turned on and off for the systems studied under cyclic aerobic and anoxic operation.

Glutaraldehyde and Sodium Acetate Treatment. To examine the phosphorus release under no biological activities, some experiments were carried out with the Ozark sludge treated with glutaraldehyde. Glutaraldehyde is known to crosslink all amino groups in cellular components, especially the proteins (including enzymes), causing cell death while preserving the physical structure of the organism. To differentiate between the release from the polyphosphate (poly- P) and that from the non-poly-P, two sets of experiments were conducted. The sludge used in one set was prepared as follows: acclimated sludge (the same as that used in R3M, ~ 400 mL) was treated with 3% (w/v) glutaraldehyde for 2 h under constant stirring. The glutaraldehyde-fixed sludge was then centrifuged and the supernatant decanted. (Samples were taken before and after the glutaraldehyde treatment for later analysis, to check if and/or how the treatment affected the phosphorus concentration in the water.) The solids were resuspended in aqueous solutions of 1 g/L sodiuim chloride (NaCl), which was used to minimize the phosphorus release from solids because of the change of ionic strength (and, thus, osmotic pressure across the cell membrane). The procedure of centrifugation, decantation, and resuspension was repeated once to wash off the residual glutaraldehyde. The final suspension of the treated solids was separated into four magnetically stirred vials; two were open to air (aerobic) and the other two sealed after being purged with nitrogen (N^sub 2^) (anaerobic). For the two vials under either aerobic or anaerobic condition, one was maintained at pH 7 and the other at pH 5 by addition of base or acid. Subsequently, the profiles of phosphorus release, under no biological activities and different operating (aeration and pH) conditions, were followed with periodical sampling. The pH was adjusted every other day after sampling.

Figure 1-Time profiles observed in a batch digestion experiment with the BNR sludge from the Ozark plant. The sludge had an initial phosphorus content (P/TS^sub i^) of approximately 7%.

Figure 2-Supernatant phosphorus concentrations at the end of 16- d digestion.

The design for the second set of experiments was to induce the release of poly-P as much as possible before fixing the solids with glutaraldehyde. The subsequent phosphorus release profile would thus come primarily from the non-poly-P in the solids under no biological activities. The differences between the phosphorus release profiles observed in the two sets of experiments would correspond to the contribution from the poly-P in the sludge.

For the second set of experiments, the active poly-P release before the glutaraldehyde treatment was achieved by addition of acetate under anaerobic condition, which is well known to trigger the polyphosphate hydrolysis in phosphate accumulating organisms (PAOs) (Ju et al., 1995; Mino et al., 1998). Approximately 400 mL of the sludge sample (R3M) were added with 1 g/L of sodium acetate and kept anaerobic for 24 h. The sludge was then treated with glutaraldehyde, washed with mineral solution, and monitored for phosphorus release under different conditions. The procedures and conditions used were the same as those described above for the first set of experiments.

Analytical Methods. The analyses for total solids (TS), volatile solids (VS), phosphorus, ammonium, nitrate, and nitrite concentrations were performed in accordance with Standard Methods (APHA et al., 1989). The specific procedures followed for measuring TS and VS were 254OB and 254OE, respectively. Phosphorus was measured using Method 4500-P D.


Time Profiles. The time profiles of phosphorus release and solid digestion were closely followed in all of the 20 digestion experiments. As an example, the results observed in the experiment conducted with the acclimated Ozark sludge, under fully aerobic conditions and a pH of 7.0, are shown in Figure 1. Appreciable amounts of phosphorus were released into the water along with the sludge digestion. In the following section, the extents of phosphorus release are summarized and compared for the digestion experiments conducted under different conditions. The important factors and potential mechanisms are further analyzed and discussed in the Discussion section.

Released Phosphorus in Water. The total lengths of digestion varied from 384 to 510 h in different runs of experiments. To be more consistent, the phosphorus concentrations in water (collected as the centrifuged supernatant) from various experiments are compared at the end of 16 days of digestion (the shortest length used). The data are plotted in Figure 2 against both pH and the initial phosphorus content in solids. The phosphorus concentrations were found in the range 20 to 130 mg/L, with a clear dependency on the initial phosphorus content in the solids, i.e., (P/TS)^sub i^. As expected, (P/TS)^sub i^ differed significantly between Akron and Ozark sludge samples; 6 to 7% for the Ozark sludge (except RlM, which was acclimated with the leaner secondary effluent instead of primary effluent) and 2 to 3.5% for the Akron sludge, which was typical of conventional activated sludge (Bishop and Farmer, 1978). The supernatant phosphorus concentration was 20 to 80 mg/L for the Akron sludge and increased to 60 to 130 mg/L for the Ozark's biophosphorus sludge.

A low pH, especially when lower than 6.0, was also found to cause higher phosphorus release, for both Akron and Ozark sludge. Taking some low-DO systems as examples, for the Ozark sludge with (P/ TS)^sub i^ = 7.2%, the supernatant phosphorus concentration reached 138 mg/L at pH 5 while only 82 mg/L at pH 7; for the Akron sludge with (P/TS)^sub i^ = 2.1%, the supernatant phosphorus concentration was 45 mg/L at pH 4.7 while only 31 mg/L at pH 6.6. As for the effect of DO, the supernatant phosphorus concentrations were higher under high-DO (fully aerobic) digestion while those under low-DO and cyclic (alternating on/off-aeration) operations were comparable.

Of course, the supernatant phosphorus concentration should not be assumed independent of the solid digestion. The normally desirable conditions that give higher percentages of VS digestion might also lead to more release of phosphorus. While more discussion on the interrelated issues is given later in the Discussion section, the basic kinetics of the VS digestion observed in the experiments is described below.

Table 2-Kinetic parameters for the first-order VS digestion of Akron and Ozark sludge.*

Figure 3-Phosphorus release profiles in digestion experiments made with alternate on/off aeration.

The best-fit kinetic parameters, k^sub d^ and (VS^sub f^/VS^sub i^), are summarized in Table 2 for the various experiments conducted. The k^sub d^ values observed for the aerobic digestion of Ozark sludge were comparable to those reported by Tonkovic (1999) for aerobic digestion of BNR biosolids. The k^sub d^ values for the Akron's conventional sludge (0.006 to 0.024 h^sup -1^) were clearly higher than those for the Ozark sludge (0.002 to 0.008 h^sup -1^) when digested at similar conditions. The slower VS digestion of the biophosphorus sludge observed in this study was consistent with the report by Wentzel et al. (1989) that the specific endogenous decay rate of PAOs, mainlyAcinetobacter, was ~0.05 d^sup -1^ (i.e., - 0.002 h^sup -1^), significantly smaller than that of the common heterotrophs, i.e., ~0.24 d^sup -1^ (i.e., ~0.01 h^sup -1^). On the other hand, when studying an enhanced biophosphorus removal process, Jardin and Popel (1996) observed similar decay rates between the PAOs and the non-PAOs. A more extensive study with sludge samples from many more plants is required before a more statistically meaningful conclusion can be reached.

Phosphorus Release under Alternating On/Off Aeration. As shown in Figure 3, distinctly different behaviors were observed between the Akron and Ozark sludge when digested under intermittent aeration. The aeration was on 16 h and off 8 h for the Akron sludge, but less regular for the Ozark sludge (on/off indicated with arrows in the figure). While no apparent responses to the on/off switch were observed for the Akron sludge, a \cyclic profile was obtained for the Ozark sludge, showing significant phosphorus release when aeration was turned off and subsequent phosphorus uptake when aeration was turned back on. This profile was consistent with the commonly accepted metabolism of PAOs. The lack of such profile for the Akron sludge confirmed the negligible PAO population in this conventional sludge.

The cyclic profile of the supernatant phosphorus for the Ozark sludge was apparent even after 500 h of digestion, although the uptake/release amplitudes decreased with time. The fact that the, PAOs were active through such a long digestion further supported their lower kd values, as described earlier. For comparison, our previous work on the Akron sludge digestion had shown very low specific oxygen uptake rates (SOUR), 0.05 to 0.4 (mg O2/g TS-h) after 20 days of digestion, and the minimal residual biological activity left was attributed primarily to the surviving autotrophic nitrifiers.


Conceptually, the solid phosphorus present in the wastewater sludge exists in three forms: (1) inorganic phosphorus, as precipitated mineral complexes; (2) organic cellular phosphorus, including nucleic acids (RNA, DNA), nucleotides, adenosine phosphates (ATP, ADP, AMP), phospholipids in cell membrane, etc.; and (3) poly-P, accumulated in PAOs. During the sludge digestion process, the three forms may be released according to different mechanisms. These are discussed and explored below for enhancing our fundamental understanding.

Effects of pH on Inorganic Phosphorus Precipitation and Dissolution. Dissolution of the inorganic phosphorus precipitates was controlled by the equilibrium relationship governed quantitatively by the solubility product constant, K^sub sp^. Many have studied the chemical precipitation of phosphates (e.g., Green et al., 1974; Hais et al., 1973; Menar and Jenkins, 1972). With the commonly used chemicals, the inorganic precipitates may be aluminum phosphate (AlPO^sub 4^) with pK^sub sp^ = 18.24 (at 20C); calcium phosphate dibasic (CaHPO^sub 4^), and calcium phosphate [Ca^sub 3^(PO^sub 4^)^sub 2^] with pK^sub sp^ = 7.0 and 28.7, respectively; or ferric phosphate (FePO^sub 4^) with pK^sub sp^ = 21.89. These minerals may also be naturally present in the wastewater even if they are not added on purpose. The change of supernatant phosphorus concentration due to the dissolution (or formation) of these inorganic precipitates is expected to depend on physical and chemical conditions, especially temperature and pH/alkalinity. With increasing temperature, K^sub sp^ increases and, consequently, promotes phosphorus release into the water. The effects of pH/ alkalinity are more complicated because pH/alkalinity has opposite effects on the concentrations of the mineral and phosphate ions. On one hand, these mineral ions also form precipitates with the hydroxyl (OH^sup -^) or carbonate (CO^sup 2-^^sub 3^) ions, such as aluminum hydroxide [Al(OH)^sub 3^], CaCO^sub 3^, Ca(OH)^sub 2^, and Fe(OH)^sub 3^. With increasing pH or alkalinity, these competing precipitation reactions reduce the availability of the mineral ions for forming phosphorus precipitates, thus tending to cause the phosphorus release. On the other hand, phosphoric acid is triprotic with pKa^sub 1^ = 2.15, pKa^sub 2^ = 7.20, and pKa^sub 3^ = 12.38. With increasing pH, the precipitating ions (PO^sup 3-^^sub 4^ and HPO^sup 2-^^sub 4^) are present in larger fractions, thus promoting the phosphorus precipitation. Which mechanism dominates depends on which ions (minerals vs. precipitating phosphates) are stoichiometrically limiting in the sludge.

Figure 4-Effect of pH on supernatant phosphorus concentration, observed in the experiment with the Akron sludge being digested under varying DO (0.3 to 1.0 mg/L) without pH control.

Figure 5-Phosphorus release profiles from glutaraldehyde-treated sludge samples.

In this study, the latter mechanism was found to dominate in the commonly encountered conditions for sludge digestion. This phenomenon is clearly shown in Figure 4 for the low-DO experiment in Group R2A. In this experiment, the DO fluctuated to higher levels (~1 mg/L) than in other low DO experiments (typically 0.2 to 0.5 mg/ L), causing the corresponding pH cycles: decreasing with increasing DO due to nitrification, and increasing with decreasing DO due to slower nitrification and the occurrence of denitrification. The supernatant phosphorus profiles basically mirrored the profiles of pH change. Note that no biological mechanisms would cause the decrease of supernatant phosphorus concentration in the above profiles because no cell growth or poly-P accumulation was expected from the decaying Akron sludge with negligible PAO population. The only plausible explanation was the chemical effects caused by pH variation.

The pH effects on the dissolution/formation of inorganic phosphorus precipitates are also largely responsible for the higher supernatant phosphorus concentrations observed in the lower pH digestion experiments, as summarized earlier in Figure 2.

Phosphorus Release from Polyphosphate Under No Biological Activity. As described in the Materials and Methods section, the glutaraldehyde killed the cells while preserving their physical structures. The phosphorus release from glutaraldehyde-treated sludge was thus not associated with any biological activity. In addition, two sets of experiments were made with and without the sludge being induced to release the poly-P as much as possible before being treated with glutaraldehyde by acetate addition under anaerobic condition. The phosphorus release from the acetate- induced, glutaraklehyde-treated sludge would thus come predominantly from the inorganic phosphorus in the solids. (The organic phosphorus release would be minimal without the biological digestion of the organics to free the covalently bonded phosphorus. The glutaraldehyde treatment also minimized the lysis of cells for release of cellular components into the water.) On the other hand, the phosphorus release from the gJutaraldehyde-treated sludge without acetate induction would include the contribution from polyP. The differences between the phosphorus release profiles observed in the two sets of experiments would indicate the poly-P hydrolysis and release under no biological activity.

The profiles obtained at different pH (5 vs. 7) and with and without acetate addition (to induce poly-P release before glutaraldehyde fixation) are summarized in Figure 5. The release profiles were similar for the systems with and without acetate induction, indicating that there was negligible phosphorus release from poly-P hydrolysis in the absence of biological activity. The phosphorus release observed in Figure 5 came predominantly from the dissolution of inorganic phosphorus precipitates. This conclusion was also supported by the following observations: (1) the majority of the phosphorus release occurred initially, followed by a gradual ascending trend of supernatant phosphorus concentration, especially for the acetate-induced systems, and (2) the phosphorus release was higher in pH 5 systems than in pH 7 systems. As described earlier, after the glutaraldehyde treatment, the solids were centrifuged and resuspended in a NaCl solution (pH ~ 7.0) containing no phosphorus- precipitating mineral ions. The initial phosphorus release could thus be a result of reaching the new equilibrium by dissolution of the existent phosphorus precipitates. The higher phosphorus release at pH 5 was also consistent with the earlier discussion for the pH effects on inorganic phosphorus dissolution/ precipitation.

Phosphorus Release from Biological Volatile Solids Digestion. The phosphorus release from organic phosphorus is expected to correlate with the VS digestion, which lyses the cells and enzymatically frees the phosphorus covalently bonded with the organic matters. As discussed above, the phosphorus release from poly-P also appeared to be mainly regulated by biological mechanisms, although its dependency on VS digestion was unknown. On the other hand, the phosphorus release from inorganic phosphorus precipitates is controlled primarily by physical and chemical environments, such as temperature, pH, and the presence of precipitating mineral ions. Among these environmental factors, only pH was varied in this study by being adjusted to the intended control value from the very beginning. Therefore, the chemical effects on phosphorus release from inorganic phosphorus mostly appeared during the initial pH adjustment.

Figure 6-The average values of released phosphorus per unit VS digested, calculated for long periods of relatively constant pH. For the Akron sludge, the values corresponded only to the phosphorus release from organic phosphorus. For the Ozark sludge, the values included the release of organic phosphorus and some poly-P.

The results are summarized in Figure 6 where the average ΔP/ ΔVS for each experiment was plotted against pH and the initial phosphorus content in the solids, i.e., (P/TS)^sub i^. Those experiments with cyclic on/off aeration or more fluctuating pH were not included. The standard deviations of the values (not shown in the figure) were rather large, ranging (15 to 50)%, involving experimental errors from both measurements of phosphorus concentration and VS.

Taken as groups, the Ozark sludge had appreciably larger values of ΔP/ΔVS than the Akron sludge did. This observation indicated that poly-P was hydrolyzed and released accompanying the biological VS digestion, assuming that the higher (P/TS)^sub i^ in the Ozark sludge was predominantly due to the presence of poly-P.

The current theory strongly links the poly-P release with its role of providing energy (in terms of ATP) for cell growth, for transport and conversion of volatile fatty acids (VFAs) to storage substances, poly(hydroxy-alkanoates) (PHA), and maintenance (survival) \activities. The last is most relevant to sludge digestion. Poly-P hydrolysis is generally believed to provide the maintenance energy during the anaerobic stage of BNR processes. It has been estimated to require the phosphorus release at approximately 2.4 to 3.7 mg P/ (g PAO VSS)-h (Smolders et al., 1994; Wentzel et al, 1989). More important to this work is what happens under aerobic conditions. Under good secondary-treatment operation, PHA oxidation provides the energy for all cellular activities, and the sludge retention time in the aeration zone should not be too long to exhaust the PHA reserves. In a study on the effects of excessive aeration, it was shown that after the depletion of PHA, the sludge glycogen content began to decrease while no phosphorus release occurred (Brdjanovic et al., 1998). Glycogen oxidation, instead of poly-P degradation, appeared to provide the energy required for survival. Therefore, poly-P release may not occur during the initial stage of aerobic digestion. The above study did not go beyond one day of aeration and glycogen was not completely depleted. The results of the current work, however, suggested that the cells would degrade poly-P for maintenance energy after glycogen was exhausted in the long-term digestion.

Low DO operation had no certain effect for the conventional Akron sludge but a clearly beneficial effect in minimizing the phosphorus release (per unit VS digested) from the poly-P containing Ozark sludge. More study is required to confirm the. observation and better understand the responsible mechanism(s). The effect of pH for the Akron sludge was uncertain because of the relatively narrow pH range involved, i.e., 6.5 to 7.5. The value of ΔP/ΔVS was, however, smaller at pH = 5.0 than at pH = 7.0 for the Ozark sludge. Some previous studies have indicated that raising the mixed liquor pH resulted in increased phosphorus release during anaerobic uptake of VFAs by BNR sludge (Liu et al., 1996; Smolders et al., 1994). Bond et al. (1999) further showed that it might be the intracellular pH that controls the phosphorus release. By adding a small weak organic base, diethanolamine (DEA), which can permeate through cell membrane, they raised the intracellular pH of the sludge. The action blocked the poly(hydroxybutyrate) (PHB) production and glycogen degradation, typically observed under the induction of VFA (in this case, 5 mM acetate). Similar amounts of phosphorus, however, were released with and without the addition of DEA. Furthermore, the phosphorus release was inhibited by the subsequent addition of weak organic acid that lowers the intracellular pH. The lower ΔP/ ΔVS observed in this study for the Ozark sludge is consistent with the above literature reports.


The phosphorus release clearly depended on the type of sludge being digested. After 16 to 21 days of digestion under various conditions, the supernatant phosphorus concentrations were 20 to 80 mg/L for the conventional Akron sludge and increased to 60 to 130 mg/ L for the Ozark's biophosphorus sludge. The low pH (

The solid phosphorus present in the wastewater sludge was conceptually divided into three groups: inorganic phosphorus, organic cellular phosphorus, and poly-P. The release of the inorganic phosphorus was largely controlled by the physical and chemical conditions, with pH being the most important factor in this study. Lowering the pH to ~4 to 6 clearly promoted the release of inorganic phosphorus, while the pH effect was less apparent at the neutral range (6.5 to 7.5). Poly-P hydrolysis was found primarily regulated by biological mechanisms and was insignificant in the deactivated (glutaraldehyde-treated) Ozark sludge. As expected, the release of organic phosphorus appeared to accompany the . VS digestion.


American Public Health Association; American Water Works Association; Water Environment Federation (1989) Standard Methods for the Examination of Water and Wastewater, 17th ed., Lenore, L. S.; Arnold E. G.; Trussell, R. R. (Eds.); Washington, D.C.

Bishop, P. L.; Farmer, M. (1978) Fate of Nutrients During Aerobic Digestion. J. Environ. Eng., 104, 967-979.

Bond, P. L.; Keller, J.; Blackall, L. L. (1999) Anaerobic Phosphate Release from Activated Sludge with Enhanced Biological Phosphorus Removal. A Possible Mechanism of Intracellular pH Control. Biotechnol. Bioeng., 63, 507-515.

Brdjanovic, D.; Slamet, A.; van Loosdrecht, M. C. M.; Hooijmans, C. M.; Alaerts, G. J.; Heijnen, J. J. (1998) Impact of Excessive Aeration on Biological Phosphorus Removal from Wastewater. Water Res., 32, 200-208.

Gachler, R.; Imboden, D. M. (1985) Lake Restoration. In Chemical Processes in Lakes. John Wiley and Sons: New York, 365-388.

Green, O.; VanDam, D.; LaBeau, B.; Campbell, T. L.; Daniels, S. L. (1974) Ferric Chloride and Organic Polyelectrolytes for the Removal of Phosphorus, EPA-670/2-73-103; U.S. Environmental Protection Agency: Washington, D.C.

Hais, A. B.; Stamberg, J. B.; Bishop, D. F. (1973) Alum Addition to Activated Sludge with Tertiary Solids Removal, EPA-670/2-73-037; U.S. Environmental Protection Agency: Washington, D.C.

Hwang, H.-J.; Choi, E. (1998) Nutrient Control with Other Sludges in Anaerobic Digestion of BPR Sludge. Water Sd. Technol., 38, 295- 302.

Jardin, N.; Popel, H. J. (1996) Influence of the Enhanced Biological Phosphorus Removal on the Waste Activated Sludge Production. Water Sd. Technol, 34, 17-23.

Ju, L.-K.; Yang, X.; Lee, J. F.; Armiger, W. B. (1995) Monitoring of the Biological Nutrient Removal Process by an On-Line NAD(P)H Fluorometer. Biotechnol. Prog., 11, 545-551.

Liu, W. T.; Mino, T.; Nakamura, K.; Matsuo, T. (1996) Glycogen Accumulating Population and its Anaerobic Substrate Uptake in Anaerobic-Aerobic Activated Sludge Without Biological Phosphorus Removal. Water Res., 30, 75-82.

Menar, A. B.; Jenkins, D. (1972) Calcium Phosphate Precipitation in Wastewater Treatment, EPA-R2/72-064; U.S. Environmental Protection Agency: Washington, D.C.

Mino, T.; van Loosdrecht, C. M.; Heijnen, J. J. (1998) Microbiology and Biochemistry of the Enhanced Biological Phosphate Removal Process. Water Rex., 32, 3193-3207.

Murakami, T.; Koike, N.; Taniguchi, N.; Esumi, H. (1987) Influence of Return Flow Phosphorus Load on Performance of the Biological Phosphorus Removal Process. Adv. Water Pollut. Control, 237-245; IAWPRC Conference Series, Pergamon Press: London.

Pitman, A. R. (1999) Management of Biological Nutrient Removal Plant Sludges-Change the Paradigms? Water Res., 33, 1141-1146.

Pitman, A. R.; Deacon, S. L.; Alexander, W. V. (1991) The Thickening and Treatment of Sewage Sludges to Minimi/e Phosphorus Release. Water Res., 25, 1285-1294.

Seyfried, C. F.; Kruse, H. D.; Schmitt, F. (1988) Influence of Sludge from Chemical Biological Wastewater Treatment on Nitrification and Digestion. Pretreatment in Chemical Water and Wastewater Treatment. Springer- Verlag: New York, 307-317.

Smolders, G. J. F.; van der Meij, J.; van Loosdrecht, M. C. M.; Heijnen, J. J. (1994) Model of the Anaerobic Metabolism of the Biological Phosphorus Removal Process: Stoichiometry and pH Influence. Biotechnol. Bioeng., 42, 461-470.

Toerien, D. F.; Gerber, A.; Lotter, L. H.; Cloete, T. E. (1990) Enhanced Biological Phosphorous Removal in Activated Sludge Systems. Adv. Microb. Ecoi, 11, 13-230.

Tonkovic, Z. (1999) Aerobic Stabilization Criteria for BNR Biosolids. Water Sci. Technol., 39, 167-174.

Wentzel, M. C., Ekama, G. A., Loewenthal, R. E., DoId, P. L.; Marais, G. V. R. (1989) Enhanced Polyphosphate Organism Cultures in Activated Sludge. Part II: Experimental Behaviour. Water SA, 15 (2), 71-88.


Credits. This research work was supported by Enviroquip, Inc. (Austin, Texas).

Authors. At the time of this work, Lu-Kwang Ju was Professor and Hemant K. Shah was a graduate student in Department of Chemical Engineering at the University of Akron (Akron, Ohio), and Jim Porteous was President of Enviroquip, Inc. Correspondence should be addressed to Lu-Kwang Ju, 230 Whitby Hall, Department of Chemical Engineering, University of Akron, Akron, OH 44325-3906; e-mail: [email protected]

Submitted for publication February 25, 2003; revised manuscript submitted October 10, 2003; accepted for publication November 4, 2003.

The deadline to submit Discussions of this paper is January 15, 2006.

Copyright Water Environment Federation Sep/Oct 2005