环境工程概论（一）：Chapter7

7 Suspended Growth Biological Treatment Processes 7-1 Introduction to the Activated-Sludge Process Historical Development The activated-sludge process is now used routinely for biological treatment of municipal and industrial wastewaters. The antecedents of the activated-sludge process date back to the early 1880s to the work of Dr. Angus Smith, who investigated the aeration of wastewater in tanks and the hastening of the oxidation of the organic matter. The aeration of wastewater was studied subsequently by a number of investigators, and in 1910 Black and Phelps reported that a considerable reduction in putrescibility could be secured by forcing air into wastewater in basins. In experiments conducted at the Lawrence Experiment Station during 1912 and 1913 by Clark and Gage with aerated wastewater, growths of organisms could be cultivated in bottles and in tanks partially filled with roofing slate spaced about 25 mm (1 in) apart and would greatly increase the degree of purification obtained (Clark and Adams, 1914). The results of the work at the Lawrence Experiment Station were so striking that knowledge of them led Dr. G. J. Fowler of the University of Manchester, England to suggest experiments along similar lines be conducted at the Manchester Sewage Works where Ardern and Lockett carried out valuable research on the subject. During the course of their experiments, Ardern and Lockett found that the sludge played an important part in the results obtained by aeration, as announced in their paper of May 3, 1914 (Ardern and Lockett, 1914). The process was named activated sludge by Ardern and Lockett because it involved the production of an activated mass of microorganisms capable of aerobic stabilization of organic material in wastewater (Metcalf & Eddy, 1930). Fig. 7-1 Description of Basic Process By definition, the basic activated-sludge treatment process, as illustrated on Fig. 7-la and b, consists of the following three basic components: (1) a reactor in which the microorganisms responsible for treatment are kept in suspension and aerated; (2) liquid-solids separation, usually in a sedimentation tank; and (3) a recycle system for returning solids removed from the liquid-solids separation unit back to the reactor. Numerous process configurations have evolved employing these components. An important feature of the activated-sludge process is the formation of flocculent settleable solids that can be removed by gravity settling in sedimentation tanks. In most cases, the activated-sludge process is employed in conjunction with physical and chemical processes that are used for the preliminary and primary treatment of wastewater, and posttreatment, including disinfection and possibly filtration. Historically, most activated-sludge plants have received wastewaters that were pretreated by primary sedimentation, as shown on Fig. 7-la and b. Primary sedimentation is most efficient at removing settleable solids, whereas the biological processes are essential for removing soluble, colloidal, and particulate (suspended) organic substances; for biological nitrification and denitrification; and for biological phosphorus removal. For applications such as treating wastewater from smaller-sized communities, primary treatment is often not used as more emphasis is placed on simpler and less operator-intensive treatment methods. Primary treatment is omitted frequently in areas of the world that have hot climates where odor problems from primary tanks and primary sludge can be significant. For these applications, various modifications of conventional activated-sludge processes are used, including sequencing batch reactors, oxidation ditch systems, aerated lagoons, or stabilization ponds. Evolution of the Activated-Sludge Process A number of activated-sludge processes and design configurations have evolved since its early conception as a result of (1) engineering innovation in response to the need for higher-quality effluents from wastewater treatment plants; (2) technological advances in equipment, electronics, and process control; (3) increased understanding of microbial processes and fundamentals; and (4) the continual need to reduce capital and operating costs for municipalities and industries. With greater frequency, activated-sludge processes used today may incorporate nitrification, biological nitrogen removal, and/or biological phosphorus removal. These designs employ reactors in series, operated under aerobic, anoxic, and anaerobic conditions, and may use internal recycle pumps and piping. Since the process came into common use in the early 1920s and up until the late 1970s, the type of activated-sludge process used most commonly was the one in which a plug-flow reactor with large length to width ratios (typically > 10:1) was used (see Fig. 7-la). In considering the evolution of the activated-sludge process, it is important to note that the discharge of industrial wastes to domestic wastewater collection systems increased in the late 1960s. The use of a plug-flow process became problematic when industrial wastes were introduced because of the toxic effects of some of the discharges. The complete-mix reactor was developed, in part, because the larger volume allowed for greater dilution and thus mitigated the effects of toxic discharges. The more common type of activated-sludge process in the 1970s and early 1980s tended to be single-stage, complete-mix activated-sludge (CMAS) processes (see Fig. 7-lb), as advanced by McKinney (1962). In Europe, the CMAS process has not been adopted generally as ammonia standards have become increasingly stringent. For some nitrification applications, two-stage systems (each stage consisting of an aeration tank and clarifier) were used with the first stage designed for BOD removal, followed by a second stage for nitrification. Other activated-sludge processes that have found application include the oxidation ditch (1950s), contact stabilization (1950s), Krause process (1960s), pure oxygen activated sludge (1970s), Orbal process (1970s), deep shaft aeration (1970s), and sequencing batch reactor process (1980). With the development of simple inexpensive program logic controllers (PLCs) and the availability of level sensors and automatically operated valves, the sequencing batch reactor (SBR) process (see Fig. 7-1c) became more widely used by the late 1970s, especially for smaller communities and industrial installations with intermittent flows. In recent years, however, SBRs are being used for large cities in some parts of the world. The SBR is a fill-and-draw type of reactor system involving a single complete-mix reactor in which all steps of the activated-sludge process occur. Mixed liquor remains in the reactor during all cycles, thereby eliminating the need for separate sedimentation tanks. In comparing the plug-flow (Fig. 7-la) and complete-mix activated-sludge (CMAS) (Fig. 7-1 b) processes, the mixing regimes and tank geometry are quite different. In the CMAS process, the mixing of the tank contents is sufficient so that ideally the concentrations of the mixed-liquor constituents, soluble substances (i.e., COD, BOD, NH4-N), and colloidal and suspended solids do not vary with location in the aeration basin. The plug-flow process involves relatively long, narrow aeration basins, so that the concentration of soluble substances and colloidal and suspended solids varies along the reactor length. Although process configurations employing long, narrow tanks are commonly referred to as plug-flow processes, in reality, true plug flow does not exist. Activated-sludge process designs before and until the late 1970s generally involved the configurations shown on Fig. 7-1a and b. However, with interest in biological nutrient removal, staged reactor designs consisting of complete-mix reactors in series have been developed (see Fig. 7-2). Some of the stages are not aerated (anaerobic or anoxic stages) and internal recycle flows may be used. For nitrification, a staged aerobic reactor design may also be used to provide more efficient use of the total reactor volume than a single-stage CMAS process. Recent Process Developments As noted above, numerous modifications of the activated-sludge process have evolved in the last 10 to 20 years, aimed principally at effective and efficient removal of nitrogen and phosphorus. Because of the development of improved membrane design, principally for water treatment applications, membrane technology has found increasing application for enhanced solids separation for water reuse, and more recently for use in suspended growth reactors for wastewater treatment. Membrane biological reactors (MBRs) may change the look of wastewater-treatment facilities in the future. Because the design and operation of the activated-sludge process is becoming more complex, computer modeling is an increasingly important tool to incorporate the large number of components and reactions necessary to evaluate activated-sludge performance. 7-2 Wastewater Characterization Activated-sludge process design requires determining (1) the aeration basin volume, (2) the amount of sludge production, (3) the amount of oxygen needed, and (4) the effluent concentration of important parameters. To design an activated-sludge treatment process properly, characterization of the wastewater is perhaps the most critical step in the process. For biological nutrient-removal processes, wastewater characterization is essential for predicting performance. Wastewater characterization is an important element in the evaluation of existing facilities for optimizing performance and available treatment capacity. Flow characterization is also important including diurnal, seasonal, and wet-weather flow variations. Without comprehensive wastewater characterization, facilities may either be under- or overdesigned, resulting in inadequate or inefficient treatment. Key Wastewater Constituents for Process Design Carbonaceous Constituents. Carbonaceous constituents measured by BOD or COD analyses are critical to the activated-sludge process design. Higher concentrations of degradable COD or BOD result in (1) a larger aeration basin volume, (2) more oxygen transfer needs, and (3) greater sludge production. While the BOD has been the common parameter to characterize carbonaceous material in wastewater, COD is becoming more common. By using a COD mass balance, the fate of carbonaceous material between the amount oxidized and the amount incorporated into cell mass is followed more easily. The various forms of the COD in wastewater are shown on Fig. 7-3. Unlike BOD, some portion of the COD is not biodegradable, so the COD is divided into biodegradable and nonbiodegradable concentrations. The next level of interest is how much of the COD in each of these categories is dissolved or soluble, and how much is particulate, comprised of colloidal and suspended solids. The nonbiodegradable soluble COD (nbsCOD) will be found in the activated-sludge effluent, and nonbiodegradable particulates will contribute to the total sludge production. Because the nonbiodegradable particulate COD (nbpCOD) is organic material, it will also contribute to the VSS concentration of the wastewater and mixed liquor in the activated-sludge process, and is referred to here as the nonbiodegradable volatile suspended solids (nbVSS). The influent wastewater will also contain nonvolatile influent suspended solids that add to the MLSS concentration in the activated-sludge process. These solids are influent inert TSS (iTSS) and can be quantified by the difference in influent wastewater TSS and VSS concentrations. For biodegradable COD, understanding the fractions that are measured as soluble, soluble readily biodegradable COD (rbCOD), and particulate is extremely important for activated-sludge process design. The rbCOD portion is quickly assimilated by the biomass, while the particulate and colloidal COD must first be dissolved by extracellular enzymes and are thus assimilated at much slower rates. The rbCOD fraction of the COD has a direct effect on the activated-sludge biological kinetics and process performance.Process applications where the rbCOD concentration affects the process design and performance are summarized in Table 7-1. The rbCOD consists of complex soluble COD that can be fermented to volatile fatty acids (VFAs) in the influent wastewater. Wastewaters that are more septic, for example, from collection systems in warm climates with minimal slopes, will contain higher concentrations of VFAs. Nitrogenous Constituents. The total Kjeldahl nitrogen (TKN) is a measure of the sum of the ammonia and organic nitrogen. About 60 to 70 percent of the influent TKN concentration will be as NH4-N, which is readily available for bacterial synthesis and nitrification. Organic nitrogen is present in both soluble and particulate forms, and some portion of each of these is nonbiodegradable. The particulate degradable organic nitrogen will be removed more slowly than the soluble degradable organic nitrogen because a hydrolysis reaction is necessary first. The nondegradable organic nitrogen is assumed to be about 6 percent of the nondegradable VSS as COD in the influent wastewater (Grady et al., 1999). The particulate nondegradable nitrogen will be captured in the activated-sludge floc and exit in the waste sludge, but the soluble nondegradable nitrogen will be found in the secondary clarifier effluent. The soluble nondegradable nitrogen contributes to the effluent total nitrogen concentration and is a small fraction of the influent wastewater TKN concentration (<3 percent). The soluble nondegradable organic nitrogen concentration in domestic wastewater typically ranges from 1 to 2 mg/L as N. Alkalinity. Alkalinity concentration is an important wastewater characteristic that affects the performance of biological nitrification processes. Adequate alkalinity is needed to achieve complete nitrification. Measurement Methods for Wastewater Characterization Readily Biodegradable COD. The rbCOD concentration is either determined from a biological response or estimated by a physical separation technique. In the biological response method the oxygen uptake rate (OUR) is followed and recorded with time after mixing the wastewater sample with an acclimated activated-sludge sample. The wastewater may be preaerated so that upon contact with the activated sludge a high DO concentration is present to allow an immediate measurement of the OUR. The wastewater sample and activated sludge are mixed in a batch reactor with separate aeration and mixing. An idealized example of the OUR response for a wastewater sample using an activated sludge containing nitrifying bacteria is shown on Fig. 7-4. The OUR versus time can be divided into four areas, which can be used to determine the oxygen consumed for the reaction indicated by the area. Area A is the oxygen used for rbCOD degradation, area B for zero-order nitrification, area C for particulate COD degradation, and area D for endogenous decay Nitrogen Compounds. For the nitrogen compounds, the soluble organic nitrogen concentration is of interest from the standpoint of its effect on the effluent total nitrogen concentration. A filtered sample from the plant effluent or from a bench-scale treatability reactor can be used to determine the total effluent soluble organic nitrogen concentration by the difference between the TKN concentration of the filtered sample and the effluent NH4-N concentration. Recycle Flows and Loadings The impact of recycle flows must also be quantified and included in defining the influent wastewater characteristics to the activated-sludge process. The possible sources of recycle flows include digester supernatant flows (if settling and decanting are practiced in the digestion operation), recycle of centrate or filtrate from solids dewatering equipment, backwash water from effluent filtration processes, and water from odor-control scrubbers. Depending on the source, a significant BOD, TSS, and NH4-N load may be added to the influent wastewater. Compared to untreated wastewater or primary clarifier effluent, the BOD/VSS ratio is often much lower for recycle streams. In addition, a significant NH4-N load can be returned to the influent wastewater from anaerobic digestion-related processes. Concentrations of NH4-N in the range of 1000 to 2000 mg/L are possible in centrate or filtrate from the dewatering of anaerobically digested solids. Thus, the ammonia load from a return flow of about one-half percent of the influent flow can increase the influent TKN load to the activated-sludge process by 10 to 20 percent. The return solids load from effluent polishing filters can be estimated by a mass balance on solids removed across the filtration process, and thus released in the backwash water flow. In all cases, a mass balance for flow and important constituents, such as BOD, TSS/VSS, nitrogen compounds, and phosphorus should be done to account for all contributing flows and loads to the activated-sludge process. 7-3 Fundamentals of Process Analysis and Control The purpose of this section is to introduce (1) the basic considerations involved in process design, (2) process control measures, (3) operating problems associated with the activated-sludge process, and (4) activated-sludge selector processes. Process Design Considerations In the design of the activated-sludge process, consideration must be given to (1) selection of the reactor type, (2) applicable kinetic relationships, (3) solids retention time and loading criteria to be used, (4) sludge production, (5) oxygen requirements and transfer, (6) nutrient requirements, (7) other chemical requirements, (8) settling characteristics of biosolids, (9) use of selectors, and (10) effluent characteristics. Selection of Reactor Type. Important factors that must be considered in the selection of reactor types for the activated-sludge process include (1) the effects of reaction kinetics, (2) oxygen transfer requirements, (3) nature of the wastewater, (4) local environmental conditions, (5) presence of toxic or inhibitory substances in the influent wastewater, (6) costs, and (7) expansion to meet future treatment needs. Selection of Solids Retention Time and Loading Criteria. Certain design and operating parameters distinguish one activated-sludge process from another. The common parameters used are the solids retention time (SRT), the food to biomass (F/M) ratio (also known as food to microorganism ratio), and the volumetric organic loading rate. While the SRT is the basic design and operating parameter, the F/M ratio and volumetric loading rate provide values that are useful for comparison to historical data and typical observed operating conditions. Solids Retention Time. The SRT, in effect, represents the average period of time during which the sludge has remained in the system. SRT is the most critical parameter for activated-sludge design as SRT affects the treatment process performance, aeration tank volume, sludge production, and oxygen requirements. For BOD removal, SRT values may range from 3 to 5 d, depending on the mixed-liquor temperature. At 18 to 25。C an SRT value close to 3 d is desired where only BOD removal is required. To limit nitrification, some activated-sludge plants have been operated at SRT values of 1 d or less. At 10。C, SRT values of 5 to 6 d are common for BOD removal only. Temperature and other factors that affect SRT in various treatment applications are summarized in Table 7-2. For nitrification design, a safety factor is used to increase the SRT above that calculated based on nitrification kinetics and the required effluent NH4-N concentration. A factor of safety is used for two reasons: (1) to allow flexibility for operational variations in controlling the SRT, and (2) to provide for additional nitrifying bacteria to handle peak TKN loadings. The influent TKN concentration and mass loading can vary throughout the day (a peak to average TKN loading of 1.3 to 1.5 is not unusual, depending on plant size) and can also be affected by return flows from digested and dewatered biosolids processing. By increasing the design SRT, the inventory of nitrifying bacteria is increased to meet the NH4-N concentration at the peak load so that the effluent NH4-N concentration requirement is achieved. Food to Microorganism Ratio. A process parameter commonly used to characterize process designs and operating conditions is the food to microorganism (biomass) ratio (F/M). Typical values for the BOD F/M ratio reported in the literature vary from 0.04 g substrate/g biomass.d for extended aeration processes to 1.0 g/g.d for high rate processes. The BOD F/M ratio is usually evaluated for systems that were designed based on SRT to provide a reference point to previous activated-sludge design and operating performance. Volumetric Organic Loading Rote. The volumetric organic loading rate is defined as the amount of BOD or COD applied to the aeration tank volume per day, Organic loadings, expressed in kg BOD or COD/m3.d, may vary from 0.3 to more than 3.0. Higher volumetric organic loadings generally result in higher required oxygen transfer rates per unit volume for the aeration system. Sludge Production. The design of the sludge-handling and disposal/reuse facility depends on the prediction of sludge production for the activated-sludge process. Sludge will accumulate in the activated-sludge process if it cannot be processed fast enough by an undersized sludge-handling facility. Eventually, the sludge inventory capacity of the activated-sludge system will be exceeded and excess solids will exit in the secondary clarifier effluent, potentially violating discharge limits. The sludge production relative to the amount of BOD removed also affects the aeration tank size. Two methods are used to determine sludge production. The first method is based on an estimate of an observed sludge production yield from published data from similar facilities, and the second is based on the actual activated-sludge process design in which wastewater characterization is done and the various sources of sludge production are considered and accounted for. For a given wastewater, the Yobs value will vary depending on whether the substrate is defined as BOD, bCOD, or COD. Observed volatile suspended solids yield values, based on BOD, are illustrated on Fig. 7-5. The observed yield decreases as the SRT is increased due to biomass loss by more endogenous respiration. The yield is lower with increasing temperature as a result of a higher endogenous respiration rate at higher temperature. The yield is higher when no primary treatment is used, as more nbVSS remains in the influent wastewater.The total mass of dry solids wasted/day includes TSS and not just VSS. The TSSI includes the VSS plus inorganic solids. Oxygen Requirements. The oxygen required for the biodegradation of carbonaceous material is determined from a mass balance using the bCOD concentration of the wastewater treated and the amount of biomass wasted from the system per day. If all of the bCOD were oxidized to CO2, H20, and NH3, the oxygen demand would equal the bCOD concentration. However, bacteria oxidize a portion of the bCOD to provide energy and use the remaining portion of the bCOD for cell growth. Oxygen is also consumed for endogenous respiration, and the amount will depend on the system SRT. For a given SRT, a mass balance on the system can be done where the bCOD removal equals the oxygen used plus the biomass VSS remaining in terms of an oxygen equivalent. The oxygen requirements for BOD removal without nitrification can be computed. As an approximation, for BOD removal only, the oxygen requirement will vary from 0.90 to 1.3 kg O2/kg BOD removed for SRTs of 5 to 20 d, respectively (WEF, 1998). NOx is the amount of TKN oxidized to nitrate. A nitrogen mass balance for the system that accounts for the influent TKN, nitrogen removed for biomass synthesis, and unoxidized effluent nitrogen is done to determine NOx. The nitrogen mass balance is based on the assumption of 0.12 g N/g biomass (C5H7NO2 for biomass). Nutrient Requirements. If a biological system is to function properly, nutrients must be available in adequate amounts. Using the formula C5H7NO2, for the composition of cell biomass, about 12.4 percent by weight of nitrogen will be required. The phosphorus requirement is usually assumed to be about one-fifth of the nitrogen value. These are typical values, not fixed quantities, because it has been shown that the percentage distribution of nitrogen and phosphorus in cell tissue varies with the system SRT and environmental conditions. The amount of nutrients required can be estimated based on the daily biomass production rate. It should be noted that nutrient limitations can occur when the concentrations of nitrogen and phosphorus are in the range of 0.1 to 0.3 mg/L. As a general role, for SRT values greater than 7 d, about 5 g nitrogen and 1 g phosphorus will be required per 100 g of BOD to provide an excess of nutrients. Other Chemical Requirements. In addition to the nutrient requirements, alkalinity is a major chemical requirement needed for nitrification. The amount of alkalinity required for nitrification, taking into account cell growth, is about 7.07 g CaCO3/g NH4-N. In addition to the alkalinity required for nitrification, additional alkalinity must be available to maintain the pH in the range from 6.8 to 7.4. Typically the amount of residual alkalinity required to maintain pH near a neutral point (i.e., pH ≈ 7) is between 70 and 80 mg/L as CaCO3. Mixed-Liquor Settling Characteristics. Clarifier design must provide adequate clarification of the effluent and solids thickening for the activated-sludge solids. In the design of installations where sludge characteristics are not known, data from other installations must be assumed or experience of the designer with similar suspended growth processes must be utilized. Two commonly used measures developed to quantify the settling characteristics of activated sludge are the sludge volume index (SVI) and the zone settling rate (WEF, 1998). The SVI is the volume of 1 g of sludge after 30 min of settling. The SVI is determined by placing a mixed-liquor sample in a 1- to 2-L cylinder and measuring the settled volume after 30 min and the corresponding sample MLSS concentration. For example, a mixed-liquor sample with a 3000 mg/L TSS concentration that settles to a volume of 300 mL in 30 min in a 1-L cylinder would have an SVI of 100 mL/g. A value of 100 mL/g is considered a good settling sludge (SVI values below 100 are desired). SVI values above 150 are typically associated with filamentous growth (Parker et al., 2001 ). Because the SVI test is empirical, it is subject to significant errors. For example, if sludge with a concentration of 10,000 mg/L did not settle at all after 30 min, the SVI value would be 100. To avoid erroneous results and to allow for a meaningful comparison of SVI results for different sludges, the diluted SVI (DSVI) test has been used (Jenkins et al., 1993). In Jenkins's analysis, the sludge sample is diluted with process effluent until the settled volume after 30 min is 250 mL/L or less. The standard SVI test is then followed with this sample. Many SVI tests at wastewater treatment plants are done in a 2-L settleometer that has a larger diameter than 1- or 2-L graduated cylinders (see Fig. 7-6). To eliminate well effects on solids settling in a small-diameter test apparatus, use of a slow-speed stirring device is encouraged (Wahlberg and Keinath, 1988). The test is called a stirred SVI when a stirring device is used (see Standard Methods, WEF, 1998). The stirred SVI test is used frequently in Europe. Secondary Clarification. Several approaches are used in the design of secondary clarification facilities. The approach used most commonly is to base the design on a consideration of the surface overflow rate and the solids loading rate. Because steady-state operations seldom occur due to fluctuations in wastewater flowrate, return activated-sludge flowrate, and MLSS concentrations, attention to the occurrence of peak events and use of safety factors are important design considerations. Overflow rates are based on wastewater flowrates instead of on the mixed-liquor flowrates because the overflow rate is equivalent to an upward flow velocity. The return sludge flow is drawn off the bottom of the tank and does not contribute to the upward flow velocity. Selection of a surface overflow rate is influenced by the target effluent requirements and the need to provide consistent process performance. The solids loading rate on an activated-sludge settling tank may be computed by dividing the total solids applied by the surface area of the tank. The commonly used units for SLR are kilograms per square meter per hour (kg/m2.h). If peak flowrates are of short duration, average 24-h values may govern; if peaks are of long duration, peak values should be assumed to govern to prevent the solids from overflowing the tank. In effect, the solids loading rate represents a characteristic value for the suspension under consideration. In a settling tank of fixed surface area, the effluent quality will deteriorate if the solids loading is increased beyond the characteristic value for the suspension. Higher rates should not be used for design without extensive experimental work covering all seasons and operating 'variables. While the surface overflow rate has been the historical clarifier design parameter, the solids loading rate is considered by some to be the limiting parameter that affects the effluent concentration. Parker et al. (2001) have shown that with a proper hydraulic design and management of solids in the sedimentation tank, the overflow rate has little or no effect on the effluent quality over a wide range of overflow rates, and the design can be based on the solids loading rates. Wahlberg (1995) supports Parker's position and, based on the evaluation of secondary clarifier performance for a number of facilities, found no effect of using surface overflow rates up to 3.4 m/h. Use of Selectors. Because solids separation is one of the most important aspects of biological wastewater treatment, a biological selector (a small contact tank) is often incorporated in the design to limit the growth of organisms that do not settle well. Selectors are naturally incorporated into the biological nitrogen- and phosphorus. removal processes described. For BOD removal only or BOD removal and nitrification processes, an appropriate selector design can be added before the activated-sludge aeration basin. Effluent Characteristics. The major parameters of interest that determine effluent quality from biological treatment processes consist of organic compounds, suspended solids, and nutrients as indicated by the following four constituents: 1. Soluble biodegradable organics a. Organics that escaped biological treatment b. Organics formed as intermediate products in the biological degradation of the waste c. Cellular components (result of cell death or lysis) 2. Suspended organic material a. Biomass produced during treatment that escaped separation in the final settling tank b. Colloidal organic solids in the plant influent that escaped treatment and separation 3. Nitrogen and phosphorus a. Contained in biomass in effluent suspended solids b. ,Soluble nitrogen as NH4-N, NO3-N, N2-O, and organic N c. Soluble orthophosphates 4. Nonbiodegradable organics a. Those originally present in the influent b. Byproducts of biological degradation In a well-operating activated-sludge process treating domestic wastes with an SRT -> 4 d, the soluble carbonaceous BOD of a filtered sample is usually less than 3.0 mg/L. With a proper secondary clarifier design and good settling sludge, the effluent suspended solids may be in the range of 5 to 15 mg/L. Process Control To maintain high levels of treatment performance with the activated-sludge process under a wide range of operating conditions, special attention must be given to process control. The principal approaches to process control are (1) maintaining dissolved oxygen levels in the aeration tanks, (2) regulating the amount of return activated sludge (RAS), and (3) controlling the waste-activated sludge (WAS). The parameter used most commonly for controlling the activated-sludge process is SRT. The mixed-liquor suspended solids (MLSS) concentration may also be used as a control parameter. Return activated sludge is important in maintaining the MLSS concentration and controlling the sludge blanket level in the secondary clarifier. The waste activated-sludge flow from the recycle line is selected usually to maintain the desired SRT. Oxygen uptake rates (OURs) are also measured as a means of monitoring and controlling the activated-sludge process. Routine microscopic observations are important for monitoring the microbial characteristics and for early detection of changes that might negatively impact sludge settling and process performance. Dissolved Oxygen Control. Theoretically, the amount of oxygen that must be transferred in the aeration tanks equals the amount of oxygen required by the microorganisms in the activated-sludge system to oxidize the organic material. In practice, the transfer efficiency of oxygen for gas to liquid is relatively low so that only a small amount of oxygen supplied is used by the microorganisms. When oxygen limits the growth of microorganisms, filamentous organisms may predominate and the settleability and quality of the activated sludge may be poor. In general, the dissolved oxygen concentration in the aeration tank should be maintained at about 1.5 to 2 mg/L in all areas of the aeration tank. Higher DO concentrations (>2.0 mg/L) may improve nitrification rates in reactors with high BOD loads. Values above 4 mg/L do not improve operations significantly, but increase the aeration costs considerably. Return Activated-Sludge Control. The purpose of the return of activated sludge is to maintain a sufficient concentration of activated sludge in the aeration tank so that the required degree of treatment can be obtained in the time interval desired. The return of activated sludge from the final clarifier to the inlet of the aeration tank is the essential feature of the process. Ample return sludge pump capacity should be provided and is important to prevent the loss of sludge solids in the effluent. The solids form a sludge blanket in the bottom of the clarifier, which can vary in depth with flow and solids loadings variations to the clarifier. At transient peak flows, less time for sludge thickening is available so that the sludge blanket depth increases. Sufficient return sludge pumping capacity is needed, along with sufficient clarifier depth (3.7 to 5.5 m), to maintain the blanket below the effluent weirs. Return sludge pumping rates of 50 to 75 percent of the average design wastewater flowrate are typical, and the design average capacity is typically of 100 to 150 percent of the average design flowrate. Return sludge concentrations from secondary clarifiers range typically from 4000 to 12,000 mg/L (WEF, 1998). Several techniques are used to calculate the desirable return sludge flowrate. Common control strategies for determining the return sludge flowrate are based on maintaining either a target MLSS level in the aeration tanks or a given sludge blanket depth in the final clarifiers. The most commonly used techniques to determine return sludge flowrate are (1) settleability, (2) sludge blanket level control, (3) secondary clarifier mass balance, and (4) aeration tank mass balance. Settleability. Using the settleability test, the return sludge pumping rate is set so that the flowrate is approximately equal to the percentage ratio of the volume occupied by the settleable solids from the aeration tank effluent to the volume of the clarified liquid (supernatant) after sealing for 30 min in a 1000-mL graduated cylinder. This ratio should not be less than 15 percent at any time. For example, if the settleable solids occupied a volume of 275 mL after 30 min of settling, the percentage volume would be equal to 38 percent [(275 mL / 725 mL) * 100]. If the plant flow were 2 m3/s, the return sludge rate should be 0.38 × 2 m3/s = 0.76 m3/s. Sludge Blanket Level. With the sludge blanket level control method, an optimum sludge blanket level is maintained in the clarifiers. The optimum level is determined by experience and is a balance between settling depth and sludge storage. The optimum depth of the sludge blanket usually ranges between 0.3 and 0.9 m. The sludge blanket method of control requires considerable operator attention because of the diurnal flow and sludge production variations and changes in the settling characteristics of the sludge. Several methods are available to detect the sludge blanket levels, including withdrawing samples using air-lift pumps, gravity-flow tubes, portable sampling pumps, and core samplers, or using sludge-supernatant interface detectors. Mass-Balance Analysis. The return sludge pumping rate may also be determined by making a mass-balance analysis around either the settling tank or the aeration tank. The appropriate limits for the two mass-balance analyses are illustrated on Fig. 7-7. Assuming the sludge-blanket level in the settling tank remains constant and that the solids in the effluent from the settling tank are negligible, the mass balance around the settling tank shown on Fig. 7-7a is as follows: The required RAS pumping rate can also be estimated by performing a mass balance around the aeration tank (see Fig. 7-7b). The solids entering the tank will equal the solids leaving the tank if new cell growth can be considered negligible. Under conditions such as high organic loadings, this assumption may be incorrect. Solids enter the aeration tank in the return sludge and in the influent to the secondary process. Sludge Wasting. To maintain a given SRT, the excess activated sludge produced each day must be wasted. The most common practice is to waste sludge from the return sludge line because RAS is more concentrated and requires smaller waste sludge pumps. The waste sludge can be discharged to the primary sedimentation tanks for co-thickening, to thickening tanks, or to other sludge-thickening facilities. An alternative method of wasting sometimes used is withdrawing mixed liquor directly from the aeration tank or the aeration tank effluent pipe where the concentration of solids is uniform. The waste mixed liquor can then be discharged to a sludge-thickening tank or to the primary sedimentation tanks where it mixes and settles with the untreated primary sludge. Oxygen Uptake Rates. Microorganisms in the activated-sludge process use oxygen as they consume the substrate. The rate at which they use oxygen, known as the oxygen uptake rate (OUR), is a measure of the biological activity and loading on the aeration tank. Values for the OUR are obtained by performing a series of DO measurements over a period of time, and the measured results are conventionally reposed as mg O2/L.min or mg O2/L.h. Oxygen uptake is most valuable for plant operations when combined with VSS data. The combination of OUR with MLVSS yields a value termed the specific oxygen uptake rate (SOUR) or respiration rate. The SOUR is a measure of the amount of oxygen used by microorganisms and is reported as mg O2/g MLVSS.h. It has been shown that the mixed liquor SOUR and the final effluent COD can be correlated, thereby allowing predictions of final effluent quality to be made during transient loading conditions. Changes in SOUR values may also be used to assess the presence of toxic or inhibitory substances in the influent wastewater. Microscopic Observations. Routine microscopic observations provide valuable monitoring information about the condition of the microbial population in the activated-sludge process. Specific information gathered includes changes in floc size and density the status of filamentous organism growth in the floc, the presence of Nocardia bacteria, and the type and abundance of higher life-forms such as protozoans and rotifers. Changes in these characteristics can provide an indication of the changes in the wastewater characteristics or of an operational problem. A decrease in the protozoan population may be indicative of DO limitations, operation at a lower SRT inhibitory substances in the wastewater. Early detection of filamentous or Nocardia growth will allow time for corrective action to be taken to minimize potential problems associated with excessive growth of these organisms. Procedures may be followed to identify the specific type of filamentous organism, which may help identify an opeating or design condition that encourages their growth (Jenkins et al., 1993). Operational Problems The most common problems encountered in the operation of an activated-sludge plant are bulking sludge, rising sludge, and Nocardia foam. Because few plants have escaped these problems, it is appropriate to discuss their nature and methods for their control. Bulking Sludge. In many cases MLSS with poor settling characteristics has developed into what is known as a bulking sludge condition, which defines a condition in the activated-sludge clarifier that can cause high effluent suspended solids and poor treatment performance. In a bulking sludge condition, the MLSS floc does not compact or settle well, and floc particles are discharged in the clarifier effluent. With good settling sludge, sludge levels may be as low as 10 to 30 cm at the bottom of the clarifier. In extreme bulking sludge conditions, the sludge blanket cannot be contained and large quantities of MLSS are carried into the system effluent, potentially resulting in violation of permit requirements, inadequate disinfection, and clogging of effluent filters. Two principal types of sludge bulking problems have been identified. One type, filamentous bulking, is caused by the growth of filamentous organisms or organisms that can grow in a filamentous form under adverse conditions, and is the predominant form of bulking that occurs. The other type of bulking, viscous bulking, is caused by an excessive amount of extracellular biopolymer, which produces a sludge with a slimy, jellylike consistency (Wanner, 1994). As the biopolymers are hydrophilic, the activated sludge is highly water-retentive, and this condition is referred to as hydrous bulking. The resultant sludge has a low density with low settling velocities and poor compaction. Viscous bulking is usually found with nutrient-limited systems or in a very high loading condition with wastewater having a high amount of rbCOD. Bulking sludge problems due to the growth of filamentous bacteria are more common. In filamentous growth, bacteria form filaments of single-cell organisms that attach end-to-end, and the filaments normally protrude out of the sludge floc. This structure, in contrast to the preferred dense floc with good settling properties, has an increased surface area to mass ratio, which results in poor settling. On Fig. 7-8, a good settling, dense nonfilamentous floc is contrasted to floc containing filamentous growth. Many types of filamentous bacteria exist, and means have been developed for the identification and classification of filamentous bacteria found commonly in activated-sludge systems (Eikelboom, 2000). The classification system is based on morphology (size and shape of cells, length and shape of filaments), staining responses, and cell inclusions. Common filamentous organisms are summarized in Table 7-3, along with the operating conditions that favor their growth. Sludge bulking can be caused by a variety of factors, including wastewater characteristics, design limitations, and operational issues. Individual items associated with each of these categories are identified in Table 7-4. Activated-sludge reactor operating conditions (low DO, low F/M, and complete-mix operation) clearly have an effect on the development of filamentous populations. One of the kinetic features of filamentous organisms that relates to these conditions is that they are very competitive at low substrate concentrations whether it be organic substrates, DO, or nutrients. Thus, lightly loaded complete-mix activated-sludge systems or low DO (<0.5 mg/L) operating conditions provide an environment more favorable to filamentous bacteria than to the desired floc-forming bacteria. Filamentous bacteria such as Beggiatoa and Thiothrix grow well on hydrogen sulfide and reduced substrates, respectively, that would be found in septic wastewaters (Wanner, 1994). When the influent wastewater contains fermentation products such as volatile fatty acids and reduced sulfur compounds (sulfides and thiosulfate), Thiothrix can proliferate. Prechlorination of the wastewaters has been done in some cases to prevent their growth. Besides causing bulking problems in activated-sludge systems, Beggiatoa and Thiothrix can create problems in fixed-film systems, including trickling filters and rotating biological contactors. In the control of bulking, where a number of variables are possible causes, a checklist of items to investigate is valuable. The following items are recommended: (1) wastewater characteristics, (2) dissolved oxygen content, (3) process loading, (4) return and waste sludge pumping rates, (5) internal plant overloading, and (6) clarifier operation. One of the first steps to be taken when sludge settling characteristics change is to view the mixed liquor under the microscope to determine what type of microbial growth changes or floc structure changes can be related to the development of bulking sludge. A reasonable quality phase-contrast microscope with magnification up to 1000 times (oil immersion) is necessary to view the filamentous bacteria structure and size. Wastewater Characteristics. The nature of the components found in wastewater or the absence of certain components, such as trace elements, can lead to the development of a bulked sludge. If it is known that industrial wastes are being introduced into the system either intermittently or continuously, the quantity of nitrogen and phosphorus in the wastewater should be checked first, because limitations of both or either are known to favor bulking. Nutrient deficiency is a classic problem in the treatment of industrial wastewaters containing high levels of carbonaceous BOD. Wide fluctuations in pH are also known to be detrimental in plants of conventional design. Variations in organic waste loads due to batch-type operations can also lead to bulking and should be checked. Dissolved Oxygen Concentration. Limited dissolved oxygen has been noted more frequently than any other cause of bulking. If the problem is due to limited oxygen, it can usually be confirmed by operating the aeration equipment at full capacity by decreasing the system SRT, if possible, to reduce the oxygen demand. The aeration-equipment should have adequate capacity to maintain at least 2 mg/L of dissolved oxygen in the aeration tank under normal loading conditions. If 2 mg/L of oxygen cannot be maintained, installation of improvements to the existing aeration system may be required. Process Loading/Reactor Configuration. The aeration SRT should be checked to make sure that it is within the range of generally accepted values. In many cases, complete-mix systems with long SRTs and subsequent low F/M ratios experience filamentous growths. In such systems, the filamentous organisms are more competitive for substrate. Laboratory research and full-scale investigations have led to activated-sludge design configurations that provide conditions favoring the dominance of floc-forming bacteria over filamentous organisms (Jenkins et al., 1993). Reactors in series with various types of environmental conditions, i.e., aerobic, anoxic, and anaerobic, are generally used to augment or replace a complete-mix reactor. The series configurations are called selector processes because they provide conditions that cause selection of floc-forming bacteria in lieu of filamentous organisms as the dominant population. Internal Plant Overloading. To avoid internal plant overloading, recycle loads should be controlled so they are not returned to the plant flow during times of peak hydraulic and organic loading. Examples of recycle loads are centrate or filtrate from sludge dewatering operations and supernatant from sludge digesters. Clarifier Operation. The operating characteristics of the clarifier may also affect sludge settling characteristics. Poor settling is often a problem in center-feed circular tanks where sludge is removed from the tank directly under the point where the mixed liquor enters. Sludge may actually be retained in the tank for many hours rather than the desired 30 min and cause localized septic conditions. If this is the case, then the design is at fault, and changes must be made in the inlet feed well and sludge withdrawal equipment. Temporary Control Measures. In an emergency situation or while the aforementioned factors are being investigated, chlorine and hydrogen peroxide may be used to provide temporary help. Chlorination of return sludge has been practiced quite extensively as a means of controlling bulking. A typical design for a low (5 to 10 h) τ system uses 0.002 to 0.008 kg of chlorine per kg MLSS.d (Jenkins et al., 1993). Although chlorination is effective in controlling bulking caused by filamentous growths, it is ineffective when bulking is due to light floc containing bound water. Chlorination normally results in the production of a turbid effluent until such time as the sludge is free of the filamentous forms. Chlorination of a nitrifying sludge will also produce a turbid effluent because of the death of the nitrifying organisms. The use of chlorine also raises issues about the formation of trihalomethanes and other compounds with potential health and environmental effects. Hydrogen peroxide has also been used in the control of filamentous organisms in bulking sludge. Dosage of hydrogen peroxide and treatment time depend on the extent of the filamentous development. Rising Sludge. Occasionally, sludge that has good settling characteristics will be observed to rise or float to the surface after a relatively short settling period. The most common cause of this phenomenon is denitrification, in which nitrites and nitrates in the wastewater are converted to nitrogen gas. As nitrogen gas is formed in the sludge layer, much of it is trapped in the sludge mass. If enough gas is formed, the sludge mass becomes buoyant and rises or floats to the surface. Rising sludge can be differentiated from bulking sludge by noting the presence of small gas bubbles attached to the floating solids and the presence of more floating sludge on the secondary clarifier surface. Rising sludge is common in short SRT systems, where the temperature encourages the initiation of nitrification, and the mixed liquor is very active due to the low sludge age. Rising sludge problems may be overcome by (1) increasing the return activated-sludge withdrawal rate from the clarifier to reduce the detention time of the sludge in the clarifier, (2) decreasing the rate of flow of aeration liquor into the offending clarifier if the sludge depth cannot be reduced by increasing the return activated-sludge withdrawal rate, (3) where possible, increasing the speed of the sludge-collecting mechanism in the settling tanks, and (4) decreasing the SRT to bring the activated sludge out of nitrification. For warm climates where it is very difficult to operate at a low enough SRT to limit nitrification, an anoxic/aerobic process is preferred to denitrification to prevent rising sludge and to improve sludge settling characteristics. Nocardia Foam. Two bacteria genera, Nocardia and Microthrix parvicella, are associated with extensive foaming in activated-sludge processes. These organisms have hydrophobic cell surfaces and attach to air bubbles, where they stabilize the bubbles to cause foam. The organisms can be found at high concentrations in the foam above the mixed liquor. Both types of bacteria can be identified under microscopic examination. Nocardia has a filamentous structure, and the filaments are very short and are contained within the floc particles. Microthrix parvicella has thin filaments extending from the floc particles. Foaming on an activated-sludge basin and a microscopic view of Nocardia are shown on Fig. 7-9. The foam is thick, has a brown color, and can build up in thickness of 0.5 to 1 m. The foam production can occur with both diffused and mechanical aeration but is more pronounced with diffused aeration and with higher air flowrates. Problems of Nocardia foaming in the activated sludge can also lead to foaming in anaerobic and aerobic digesters that receive the waste-activated sludge. Nocardia growth is common where surface scum is trapped in either the aeration basin or secondary clarifiers. Aeration basins that are baffled with flow from one cell to the next occurring under the baffles, instead of over the top, encourage Nocardia growth and foam collection. Methods that can be used to control Nocardia include (1) avoiding trapping foam in the secondary treatment process, (2) avoiding the recycle of skimmings into the secondary treatment process, and (3) using chlorine spray on the surface of the Nocardia foam. The use of a selector design may help to discourage Nocardia foaming, but significant foaming has been observed with anoxic/aerobic processes. The addition of a small concentration of cationic polymer has been used with some success for controlling Nocardia foaming (Shao et al., 1997). The presence of Nocardia has also been associated with the presence of Nocardia-Microthrix with fats and edible oils in wastewater. Reducing the oil and grease content from discharges to the collection system from restaurants, truck stops, and meatpacking facilities by effective degreasing processes can help control potential Nocardia problems. Activated-Sludge Selector Processes In the above discussion, problems caused by nuisance microorganisms in activated sludge were presented, including the effect of filamentous bacteria on sludge settling characteristics and the potential of sludge bulking when the filamentous bacteria are present in high numbers. Prior to the 1970s, filamentous bulking was considered an inevitable consequence of activated-sludge treatment, but work by Chudoba et al. (1973) with staged versus complete-mix activated-sludge reactors led to the concept that reactor configuration designs, now termed selectors, could be used to control filamentous bulking and improve sludge-settling characteristics. The concept of a selector is the use of a specific bioreactor design that favors the growth of floc-forming bacteria instead of filamentous bacteria to provide an activated sludge with better settling and thickening properties. The high substrate concentration in the selector favors the growth of nonfilamentous organisms. A selector is a small tank (20 to 60 min contact time) or a series of tanks in which the incoming wastewater is mixed with return sludge under aerobic, anoxic, and anaerobic conditions. Various types of selectors are shown on Fig. 7-10. The selector reactor precedes the activated-sludge aeration tank and may be designed as a separate reaction stage for a complete-mix reactor (see Fig. 7-10a) or as individual compartments in a plug-flow system (see Fig. 7-10b and c). Sequencing batch reactors may also be operated to employ the selector concept. The goal in the selector is to have most of the rbCOD consumed by the floc-forming bacteria. Because the particulate degradable COD decomposes at a much slower rate and will be present in the aeration tank, the rbCOD must be utilized for the benefit of the floc-forming bacteria. Selector designs are based on either kinetic or metabolic mechanisms. The kinetics-based selector designs are called high F/M selectors, and the metabolic-based selectors are either anoxic or anaerobic processes. Kinetics-Based Selector. Selector designs based on biokinetic mechanisms provide for reactor substrate concentrations that result in faster substrate uptake by the floc-forming bacteria. While filamentous bacteria are more efficient for substrate utilization at low substrate concentrations, the floc-forming bacteria have higher growth rates at high soluble substrate concentrations. A series of reactors at relatively low τ values (minutes) is used to provide high soluble substrate concentrations, in contrast to feeding influent wastewater to aeration tanks with τ values on the order of hours. For three reactors in series, the following COD F/M ratios, based on the: influent to flowrate and COD concentration, are recommended. . First reactor, 12 g COD/g MLSS.d . Second reactor, 6 g COD/g MLSS.d . Third reactor, 3 g COD/g MLSS.d The F/M ratio is calculated for the first reactor using the volume and MLSS concentration at that reactor and the influent wastewater flowrate and COD concentration. The F/M value shown for the second reactor includes the volume of the first and second reactor and the applied loading as the product of the influent flowrate and COD concentration. Albertson (1987) recommended a similar approach based on a BOD F/M loading of 3 to 5 g BOD/g MLSS in the first reactor, with the second and third reactors being equal to and twice the first reactor volume, respectively. Albertson further notes that if the loading to the first reactor is too high (F/M > 8 g BOD/g MLSS.d), a viscous, nonfilamentous-type bulking can develop. The kinetic concept of a high F/M selector suggests that it be aerobic, and high DO concentrations are needed to maintain an aerobic floc (>6 to 8 mg/L). In many cases, such high DO concentrations are not practical or provided, and the staged selector design (described above) is operated at a low to zero DO concentration so that a metabolic selector mechanism is involved. A sequencing batch reactor (SBR) can also act as a very effective high F/M selector, depending on the wastewater strength and feeding strategy. For high-strength wastewaters with a relatively large fraction of the SBR volume occupied by the influent wastewater, a high initial F/M ratio can occur. The subsequent reaction by the batch process is equal to that for a plug-flow reactor. Metabolic-Based Selector. With biological nutrient-removal processes, improved sludge-settling characteristics, and, in many cases, minimal filamentous bacteria growth has been observed. The anoxic or anaerobic metabolic conditions used in these processes favor growth of the floc-forming bacteria. The filamentous bacteria cannot use nitrate or nitrite for an electron acceptor, thus yielding a significant advantage to denitrifying floc-forming bacteria. Similarly, the filamentous bacteria do not store polyphosphates and thus cannot consume acetate in the anaerobic contact zone in biological phosphorus-removal designs, giving an advantage for substrate uptake and growth to the phosphorus-storing bacteria. In some wastewater-treatment facilities (Seattle and San Francisco, for example), an anaerobic selector has been used before the aeration tank in low SRT activated-sludge systems designed for BOD removal, even though phosphorus removal is not required. Where nitrification is used and phosphorus removal is not required, anoxic selectors (either the staged high F/M gradient or the single-stage designs) have been used. For the high F/M anoxic or anaerobic selectors, the resultant mixed-liquor SVI may be in the range of 65 to 90 mL/g, and for single-tank anoxic selectors, SVI values in the range of 100 to 120 mL/g are more commonly obtained. The use of selector designs in activated sludge is more common because of the many advantages derived from the minimal investment in a relatively small reactor volume. By improving sludge settling, the activated-sludge treatment capacity may be increased, as higher MLSS concentrations are usually possible. The hydraulic capacity of the secondary clarifiers is also increased. 7-4 Processes for BOD Removal and Nitrification Process Design Considerations For BOD removal and nitrification processes, the rbCOD concentration is important for evaluating the oxygen demand profiles for plug-flow, staged, and batch-fed processes. The effect of nbVSS concentration in the influent will be significant in process sludge production and aeration volume requirements. In the following paragraphs, three activated-sludge process design examples are provided to demonstrate application of these fundamental principles to BOD removal and nitrification processes. Complete-Mix Activated-Sludge Process In a typical complete-mix activated-sludge (CMAS) process, effluent from the primary sedimentation tank and recycled return activated sludge are introduced typically at several points in the reactor. Because the tank contents are thoroughly mixed, the organic load, oxygen demand, and substrate concentration are uniform throughout the entire aeration tank and the F/M ratio is low. Care should be taken to assure that the CMAS reactor is well mixed and that influent feed and effluent with-drawn points are selected to prevent short-circuiting of untreated or partially treated wastewater. The complete-mix reactor is usually configured in square, rectangular, or round shapes. Tank dimensions depend mainly on the size, type, and mixing pattern of the aeration equipment. Sequencing Batch Reactor Process The sequencing batch reactor (SBR) process utilizes a fill-and-draw reactor with complete mixing during the batch reaction step (after filling) and where the subsequent steps of aeration and clarification occur in the same tank. All SBR systems have five steps in common, which are carded out in sequence as follows: (1) fill, (2) react (aeration), (3) settle (sedimentation/clarification), (4) draw (decant), and (5) idle. Each of these steps is illustrated on Fig. 7-11 and described in Table 7-5. For continuous-flow applications, at least two SBR tanks must be provided so that one tank receives flow while the other completes its treatment cycle. Several process modifications have been made in the times associated with each step to achieve nitrogen and phosphorus removal. Sludge Wasting in SBRs. Sludge wasting is another important step in the SBR operation that greatly affects performance. Wasting is not included as one of the five basic process steps because there is no set time period within the cycle dedicated to wasting. The amount and frequency of sludge wasting is determined by performance requirements, as with a conventional continuous-flow system. In an SBR operation, sludge wasting usually occurs during the react phase so that a uniform discharge of solids (including fine material and large floc particles) occurs. A unique feature of the SBR system is that there is no need for a return activated-sludge (RAS) system. Tab. 7-5 Description of operational steps for the sequencing batch reactor Because both aeration and settling occur in the same chamber, no sludge is lost in the react step and none has to be returned to maintain the solids content in the aeration chamber. The SBR process can also be modified to operate in a continuous-flow mode as discussed later in this chapter. Because of the substrate concentration changes with time, the substrate utilization and oxygen demand rates change, progressing from high to low levels. The aeration system should be designed to reflect the changing requirements in oxygen demand. Process Design of SBRs. Because of the many design variables involved in an SBR design, an iterative approach is necessary in which key reactor design conditions are first assumed. A set of different design conditions can be evaluated by use of a spreadsheet analysis to determine the most optimal choice. The key design conditions selected are (1) the fraction of the tank contents removed during decanting and (2) the settle, decant, and aeration times. Because the fill volume equals the decant volume, the fraction of decant volume equals the fraction of the SBR tank volume used for the fill volume per cycle. The design procedure for the SBR system is presented in Table 7-6. Tab. 7-6 Computation approach foe the design of a SBR Staged Activated-Sludge Process In the conventional plug-flow activated-sludge system, the tank hydraulics and mixing regime may result in two to four effective stages from the standpoint of biological kinetics. Activated-sludge processes can be designed with baffle walls to intentionally create a number of complete-mix activated-sludge zones operating in series. For the same reactor volume, reactors in series can provide greater treatment efficiency than a single complete-mix reactor, or provide a greater treatment capacity. As a consequence, staged activated-sludge process configurations are used at several full-scale installations. Oxygen Demand in Staged Designs. The oxygen demand varies in staged complete-mix reactor designs and can be high enough in the first stage to challenge the volumetric oxygen transfer capability of aeration equipment. With high-density fine bubble aeration diffusers, such as membrane aeration panels oxygen transfer rates of 100 to 150 mg/L.h are possible, with some manufacturers claiming higher rates. The changes in oxygen uptake rates (OURs) in each stage of a four-stage activated-sludge process (defined as a function of oxygen needed for nitrification, rbCOD removal, particulate degradable COD, and endogenous respiration) are depicted on Fig. 7-12. Most of the rbCOD will be consumed in the first stage, and the OUR for pCOD degradation will decrease from stage to stage as a function of the degradation kinetics. Nitrification rates may be at a maximum zero-order kinetic rate for the first one to three stages due to higher NH4-N concentrations in the early stages. Oxygen demand for endogenous respiration will be relatively constant from stage to stage. The oxygen demand distribution may be estimated to determine the aeration design for staged processes. The percent of the total oxygen consumption may range from 40, 30, 20, and 10 percent, respectively, for a four-stage system. One design approach that can be used to obtain an estimate of the oxygen demand in a staged system is to calculate the total oxygen demand as would be done for a CMAS process, and then estimate the oxygen demand distribution with consideration to the various components described above. With proper selection of the type and placement of the diffusers and by providing an air supply system with DO control in each portion of the system, the air can be provided where needed. Generally, the approach outlined above is satisfactory because during the life of the process, the oxygen demand will vary across the tank as the load changes. Use of Simulation Models. The other approach involves the use of simulation models, in which the kinetics and changes in constituent concentrations in each stage are taken into consideration. This approach will typically result in a more optimal design and can be used to assess the real capacity of a given activated-sludge design. The simulation approach involves solving a set of equations in each stage for each constituent, which includes rbCOD, pCOD, NH4-N, endogenous respiration, and biomass concentration. Models also include phosphorus and the effect of biological phosphorus removal on design and performance. Alternative Processes for BOD Removal and Nitrification Over the last 30 years numerous activated-sludge processes have been developed for the removal of organic material (BOD) and for nitrification. Some of the processes modifications or variations of basic processes that have evolved to meet different performance objectives. Descriptions and flowsheets are presented in Table 7-7 for representative processes used for BOD removal and nitrification. The processes grouped according to the basic reactor configuration: plug-flow, complete-mix, and sequentially operated systems. The processes differ in terms of their aeration configuration, aeration equipment design, solids retention time, operating mode, and ability to remove nitrogen, and some are proprietary. The high-rate aeration, contact stabilization, and high-purity oxygen processes are used primarily for BOD removal only, are designed for relatively~ short SRTs, and require less space than other processes. Where nitrification is not needed to meet treatment discharge limits, the three processes cited above are particularly attractive for large municipalities where space is limited. The conventional plug-flow, step-feed, and complete-mix processes are used for both BOD removal and nitrification and are applied over a wide range of SRTs, depending on the wastewater temperature and treatment needs. The Kraus process is seldom used, but it is included to show how oxidized nitrogen can be used to help BOD degradation in the first pass of a plug-flow aeration tank. In contrast to the processes described above, conventional extended aeration, oxidation ditch, OrbalTM, and BiolacTM processes represent a different approach to biological wastewater treatment (for the latter three, see Fig. 7-13). The processes employ a much simpler system by generally eliminating primary treatment and anaerobic digestion from the overall treatment system. Larger aeration tanks with longer SRTs, usually exceeding 20 d, are used. The process approach is attractive for smaller communities where space is not an issue and less complex operation is preferred. The large aeration tank volume provides good equalization at high flow and loading occurrences, and a high-quality effluent is produced. With the exception of the conventional extended aeration process, the systems are operated usually to promote denitrification in addition to nitrification. The aeration and mixing of the channel-flow processes (oxidation ditch, OrbalTM, and CCASm) require much less energy for mixing than needed for aeration so that aeration equipment design is based on meeting oxygen requirements instead of tank mixing. Less energy is required in comparison to conventional extended aeration processes. In the past, the oxidation ditch and extended aeration processes were thought to need long SRTs to provide well-stabilized biosolids for reuse. However, with stricter regulations governing biosolids stabilization, separate aerobic digestion facilities are used to meet the requirements for reuse. Several sequentially operated activated-sludge processes that do not use separate tanks for liquid-solids separation are also described in Table 7-8. The processes include the sequencing batch reactor, batch decant reactor, and the cycle activated-sludge system. Operation is based usually on long r and SRT values. The processes are attractive to small communities because of the simplicity of operation and relatively low cost. Sequentially operated processes are also adaptable to nitrogen removal. Tab. 7-8 Typical design parameters for activated sludge processes Process Selection Considerations Selection of an activated-sludge process for BOD removal and nitrification is a function of many considerations including specific site constraints, compatibility with the existing process, compatibility with existing equipment, present and future treatment needs, level of capability of the operating staff, capital costs, and operating costs. Significant features and limitations of the various activated-sludge process alternatives that affect process selection in certain applications are summarized in Table 7-9. Tab. 7-9 Advantages and limitations of activated sludge processes for BOD removal and nitrification 7-5 Process for Biological Nitrogen Removal Nitrogen removal is often required before discharging treated wastewater to sensitive water bodies (to prevent eutrophication), or for groundwater recharge or other reuse applications. Nitrogen removal can be either an integral part of the biological treatment system or an add-on process to an existing treatment plant. The purpose of this section as in the previous section, is to illustrate in detail the design procedure for processes used to remove the nitrogen from wastewater biologically. However, before considering the design examples, it is appropriate to present an overview of the biological nitrogen-removal process and design issues for the anoxic/aerobic process. Following the discussion of design issues, design examples are provided for (1) the anoxic/aerobic process, (2) step-feed anoxic/aerobic process, (3) intermittent aeration, (4) a sequencing batch reactor, and (5) postanoxic denitrification with methanol addition. Descriptions and flow diagrams for several alternative processes for nitrogen removal, typical process design parameters, and process selection considerations are presented following the design examples. Overview of Biological Nitrogen-Removal Processes All of the biological nitrogen-removal processes include an aerobic zone in which biological nitrification occurs. Some anoxic volume or time must also be included to provide biological denitrification to complete the objective of total nitrogen removal by both NH4-N oxidation and NO3-N and NO2-N reduction to nitrogen gas. Nitrate reduction requires an electron donor, which can be supplied in the form of influent wastewater BCD, by endogenous respiration, or an external carbon source. The types of suspended growth biological nitrogen-removal processes can be categorized as (1) single-sludge or (2) two-sludge. The term "single-sludge" means only one solids separation device (normally a secondary clarifier) is used in the process(see Fig. 7-14a, c, and d). The activated-sludge tank may be divided into different zones of anoxic and aerobic conditions and mixed liquor may be pumped from one zone to another (internal recycle), but the liquid-solids separation occurs only once. In the two-sludge system, the most common system consists of an aerobic process (for nitrification) followed by an anoxic process (for denitrification), each with its own clarifier, thus producing two sludges (see Fig. 7-14e). For postanoxic denitrification, an organic substrate usually methanol, must be added to create a biological demand for the nitrate. Single-Sludge Biological Nitrogen-Removal Processes The single-sludge biological nitrogen-removal processes are grouped accordin to whether the anoxic zone is located before, after, or within the aerobic nitrification zone. These three possibilities, illustrated on Fig. 7-14a, c, and d, are termed (1) preanoxic where initial contact of the wastewater and return activated sludge is in an anoxic zone (2) postanoxic, where the anoxic zone follows the aerobic zone; or (3) simultaneous nitrification-denitrification (SNdN) processes where both nitrification and denitrification occur in the same tank. In the preanoxic configuration, nitrate produced in the aerobic zone is recycled to the preanoxic compartment. The late of denitrification is affected by the rbCOD concentration in the influent wastewater, the MLSS concentration, and temperature. Postanoxic designs (Fig. 7-14c) may be operated with or without an exogenous carbon source. Without an exogenous source, postanoxic processes depend on the endogenous respiration of the activated sludge to provide electron donor for nitrate consumption in lieu of oxygen. The denitrification rate is much slower, by a factor of 3 to 8, compared to preanoxic applications that use influent waste water BCD for the electron donor. A long detention time would be requited in this type of postanoxic tank to achieve high nitrate-removal efficiency. Single-tank designs have also been used in which nitrification and denitrification occur simultaneously. The simultaneous nitrification-denitrification (SNdN) applications require DO control or other types of control methods to assure that both nitrification and denitrification occur in a single tank. A significant amount of nitrogen removal has been observed in single-sludge activated-sludge systems without distinctive separate anoxic zones. Van Huyssteen et al. (1990) reported significant nitrogen loss in a nitrification aeration basin mixed and aerated by surface mechanical aerators. Significant nitrogen loss can occur in reactors, such as the oxidation ditch or similar process (see Fig. 7-14d) that have long hydraulic retention times. The combination of both nitrification and denitrification may be explained by two possible mechanisms. First, regions of low DO or zero DO concentration may be present within the basin as a function of the mixing regime. Van Huyssteen et al. (1990) opined that as the mixed liquor traveled away from the surface mechanical aerators, the DO was depleted, creating conditions more favorable for anoxic reactions. Second, activated-sludge floc can contain both aerobic and anoxic zones, as illustrated in a simplified view of a biological floc on Fig. 7-15. Dissolved oxygen and dissolved substrates outside of the floc diffuse into the aerobic zone, and, depending on the DO concentration and concentration of ammonia and bCOD, oxygen may be depleted at significant rates within the floc so that the DO cannot penetrate the entire floc depth. Nitrate produced by nitrification in the aerobic zone can diffuse into the inner anoxic zone along with substrate so that denitrification occurs within the floc depth. The existence of anoxic zones within the biological floc has been supported by Stenstrom and Song (1991), in which they showed that nitrification rates were related not only to bulk liquid DO concentration but also to the amount of BOD present. At higher soluble BOD concentrations, higher oxygen uptake rates occurred, and lower nitrification rates were observed for the same bulk liquid DO concentration, suggesting that the aerobic zone of the activated-sludge floc decreased. Nitrification and denitrification rates should both be at less than optimal levels for simultaneous nitrification/denitrification (SNdN) processes. Only a portion of the biomass is used for each of these reactions. In addition, the nitrification rate is lower due to the low DO concentration, and the denitrification rate is lower due to substrate consumption in the aerobic portion of the floc. However, systems with very long detention times, such as oxidation ditches, have sufficient volume to accommodate lower rates for both nitrification and denitrification. Of the three basic process configurations, the preanoxic nitrification/denitrification process (Fig. 7-14a) is used most often because of (1) the relative ease of retrofit to existing plants, (2) the benefits of the selector operation for control of bulking sludge, (3) the production of alkalinity before the nitrification step, and (4) the ability to convert an existing biological treatment system to nitrogen removal with relatively short to moderate basin detention times. Most of the nitrogen-removal processes can incorporate biological phosphorus removal, as discussed in Sec. 8-6. Design issues and design examples for commonly used biological nitrogen-removal processes are presented below. The fundamental design concepts that are exemplified can be of use in evaluating other types of suspended growth biological nitrogen-removal processes. Process Design Considerations Anoxic/Aerobic Reactor Design Considerations. In the anoxic/aerobic process (Fig. 7-14a) nitrate is fed to the anoxic reactor from nitrate in the return activated-sludge flow and by pumping mixed liquor from the aerobic zone. In step-feed anoxic/aerobic processes, nitrate will be fed to the anoxic zone by flow of mixed liquor from a previous nitrification step. The electron donor is provided by the influent wastewater fed to these preanoxic zones. Key design parameters that affect the amount of nitrogen removed are (1) anoxic zone detention time, (2) mixed-liquor volatile suspended solids (MLVSS) concentration, (3) internal recycle rate and return sludge flow, (4) influent BOD or biodegradable COD (bCOD) concentration, (5) the readily biodegradable COD (rbCOD) fraction, and (6) temperature. The influent rbCOD concentration has a significant effect on the denitrification rate in the anoxic zone. Wastewaters with the same influent bCOD, but with a higher fraction of rbCOD, will undergo higher denitrification rates in the anoxic zone. Anoxic zones have been designed as single- stage or a series of complete mix tanks with equal or different detention times. Typical power requirements for mechanical mixing in the anoxic zone range from 8 to 13 kW/103 m3). Two design approaches are used to design the anoxic zone volume and to determine the amount of nitrogen removal. One is a desktop design approach that employs mass balances for nitrogen and a commonly used design parameter, the specific denitrification rate (SDNR) in g NO3-N reduced/g MLVSS.d. More recently, comprehensive mechanistic simulation models have been developed that can relate denitrification rates to fundamental biokinetics, wastewater characteristics, activated-sludge tank volumes and configurations, and SRT. Most simulation models are based on the ASM1 model developed by a committee under the International Association of Water Pollution Research Control (IAWPRC). The ASM1 model is now referred to as the IWA (International Water Association) model to account for the organization name change. The model is developed around a basic activated-sludge model that can be used to describe biomass growth rates, and follows the fate of degradable COD and nitrogen (in both soluble and particulate forms) for systems that have both aerobic and anoxic treatment zones. An ASM2 version of the model incorporates biological phosphorus removal as well as nitrogen removal (Barker and Dold, 1997). In these models, nitrate is consumed by heterotrophic biomass under anoxic conditions during the consumption of either rbCOD or short-chain volatile fatty acids. The rbCOD substrate consumed in the anoxic reactor is from the influent wastewater plus that produced in the reactor by hydrolysis of influent pCOD and released biomass material due to cell lyses. Anoxic/Aerobic Process Design Procedure. The influent wastewater rbCOD fraction is a critical design parameter, and if unknown, a conservative value in the range of 15 to 25 percent of the total bCOD can be used. The aerobic volume is based on using an aerobic SRT for nitrification, and only the aerobic basin volume and mixed liquor are used to compute the sludge wasting for that SRT. The total process SRT will be longer when the mixed liquor and the volume of the anoxic reactor are included. The aeration oxygen requirement will be less than that for the nitrification-only design as nitrate will be used to consume some of the influent bCOD in the preanoxic zone. A mass balance on nitrogen must be done to determine (1) how much nitrate is produced in the aeration zone, and (2) what the internal recycle ratio must be to meet the desired effluent nitrate concentration. The mass balance accounts for the nitrate produced in the aerobic zone. The amount of nitrate produced in the aerobic zone is based on the influent flowrate and nitrogen concentration, the amount consumed for cell synthesis, and the effluent NH3-N and soluble organic nitrogen concentrations. As a conservative design approach, all of the influent TKN is assumed to be biodegradable and the effluent soluble organic nitrogen concentration is ignored. The nitrate produced is contained in the total flow leaving the aerobic zone, which includes internal recycle, RATS, and effluent flows. The effect of the IR ratio on the effluent NO3-N concentration for a given amount of nitrate produced (NOx) and for a RAS recycle ratio of 0.50 is illustrated on Fig. 7-15. A greater IR ratio is needed to produce the same effluent NO3-N concentration when more NOx is produced in the aerobic zone. To meet a standard of 10 mg TN/L or less a design effluent NO3-N concentration of 5 to 7 mg/L should be used. An internal recycle ratio in the range of 3 to 4 is typical, but ratios in the range of 2 to 3 are also applied for wastewaters with a lower influent wastewater TKN concentration. Recycle ratios above 4 are generally not warranted, as the incremental removal of NO3-N is low and more DO is recycled from the aeration zone into the anoxic zone. The amount of DO fed to the anoxic zone due to the internal recycle flow from the aerobic zone must be minimized because the oxygen will consume rbCOD, leaving less available for NO3 reduction. In some designs, sections of the aerobic zone are baffled with DO control so that the DO concentration in the recycle can be controlled and minimized. Care must also be taken to ensure that the influent wastewater is not overly aerated when passing through the plant en route to the anoxic tank. Single-Sludge Simultaneous Nitrification Denitrificatlon (SNdN) Processes. Both high levels of nitrification and denitrification have been reported for oxidation ditch systems operated with low DO concentrations (0.10 to 0.40 mg/L) and with relatively long τ and SRT values. The low DO concentration will result in lower nitrification rates, as the activated-sludge floc will be only partially aerobic. Thus, only a portion of the nitrifying bacteria contained in the floc will be active. In addition, nitrification rates are lower at low DO concentrations. Denitrification occurs in the anoxic zones established within the floc particles due to oxygen depletion, with the result that simultaneous nitrification and denitrification takes place. The nitrification and denitrification rates are a function of the reaction kinetics, floc size, floc density, floc structure, rbCOD loading, and bulk liquid DO concentration. Because of the complex physical factors, the nitrification and denitrification rates cannot be predicted accurately with present models. Basic modifications to the Monod growth model, however, can be used to estimate the effects of a low DO concentration on nitrification and denitrification rates and system performance. Anoxic/Aerobic Process Design Step-Feed Anoxic/Aerobic Process Design Step feed for nitrogen removal is similar to the step-feed process for BOD and nitrification. For nitrogen removal, wastewater is introduced at several feed points. In most eases, where a step-feed process is in place for BOD removal and nitrification, it will be relatively easy to upgrade it to a step-feed anoxic/aerobic biological nitrogen-removal process. For such applications the influent feed points and volumes of the individual channels in the reactor (passes) are already determined. The tank layout is generally symmetrical and the volume in each pass is equal. For a new tank design, it is possible to use a nonsymmetrical step-feed design where the feed split is somewhat equal, but the volume of each pass increases as the mixed-liquor concentration decreases from the first to last pass. The nonsymmetrical design approach may utilize the tank volumes more efficiently by using a similar F/M ratio for each pass. The variables involved in the design of a step-feed biological nitrogen-removal process for an existing basin are (1) the flow distribution between passes, (2) the relative split between anoxic and aerobic volumes, and (3) the final pass MLSS concentration. The selection of the final pass MLSS concentration is normally based on using an acceptable solids loading for the secondary clarifier. As will be shown in the following design example, the selection of the final pass MLSS concentration, the RAS ratio, the influent flow split, and wastewater characteristics determine the system SRT. With the known SRT value, the biomass and nitrifying bacteria concentration in the mixed liquor can be determined, which can then be used to determine the nitrification and denitrification capacity of the system. The process design procedure involves successive iterations with varying anoxic/aerobic volumes and flow splits evaluated to find the most satisfactory design. Intermittent Aeration Process Design Long SRT systems, such as oxidation ditch processes, may employ intermittent aeration to accomplish both nitrification and denitrification in a single tank. During the aeration off period, the aeration tank operates essentially as an anoxic reactor as nitrate is used in lieu of DO for BOD removal. During the anoxic period, the tank operation is similar to a preanoxic tank because influent BOD is added continuously to drive the denitrificaton reaction. Operation of an oxidation ditch using intermittent aeration is shown on Fig. 7-16. Intermittent aeration systems typically are operated with SRT values in the range of 18 to 40 d and hydraulic detention times in excess of 16 h. During the anoxic reaction period (Fig. 7-16b), aeration is stopped, a submerged mixer is turned on, and nitrate is used as the electron acceptor. The reactor is operated as a complete mix activated sludge anoxic process. During the anoxic period, DO and nitrate are depleted and the ammonia concentration increases (see Fig. 7-16c). The time for the anoxic and aerobic periods is important in determining the system's treatment performance. The anoxic/aerobic cycle times may be adjusted manually as part of the system operation to optimize the process performance. Alternatively, a patented process, NitroxTM, uses an oxidation-reduction potential (ORP) measurement to control the intermittent operation. The ORP response during an aeration-off period is shown on Fig. 7-16c. As the DO concentrations decline, the ORP value decreases. When the NO3-N is depleted, a dramatic decline in the ORP value occurs. The ORP decline is called the ORP knee and can be identified by calculating the ORP slope with time. In the NitroxTM process, the ORP values are logged onto a computer, which is programmed to turn on the aeration based on the changing slope of the OPP. The aeration off periods are selected to occur during different times of the day; the more ideal time is when the influent BOD concentration is high st) that nitrate reduction occurs at a faster rate. The process behaves much like an anoxic selector as improved SVIs have been reported. Reported plant performance data for intermittent aeration processes indicate effluent NO3-N concentrations range from 3.0 to 4.8 mg/k The processes are also predicted to produce affluent total nitrogen (TN) concentrations <8.0 mg/L (U.S. EPA, 1993). The relatively long v values used provide sufficient dilution to minimize the effluent NH4-N concentrations during the OFF period. A sufficiently long SRT is also needed to provide enough nitrification capacity to allow the aeration system to be operated intermittently. Because of the long SRT and τ the denitrification kinetics axe related to the over all degradation of bCOD, pbCOD, and endogenous decay and are not as strongly influenced by the rbCOD fraction as for the preanoxic denitrification application with the relatively short τ Daring the complete-mix anoxic period in an intermittent oxidation ditch operation or in the anoxic zone of a continuously aerated oxidation ditch, the specific denitrification rate is affected by both the endogenous respiration rate and the bCOD in the influent wastewater as continuous feeding occurs. The average specific denitrification rate,which includes these effects, can be estimated by the following equation. Postanoxic Endogenous Denitrification Postanoxic denitrification can be done in separate tanks within the same single-sludge system following nitrification, as first proposed by Wuhrmann (1964). The Bardenpho process is a good example of this application. After nitrification, the rbCOD is fully depleted and, depending upon the system SRT, most of the pbCOD is likely to be depleted. Thus, the electron donor that creates the demand for nitrate reduction is mainly from activated-sludge endogenous respiration. Observed SDNRs have ranged from 0.01 to 0.04 g NO3-N/g MLVSS under endogenous respiration (U.S. EPA, 1993; Stensel et al., 1995). The endogenous oxygen equivalent utilization rate under anoxic conditions has been found to be about 50 percent of that under aerobic conditions (Randall et al., 1992; Wuhrmann, 1964). As discussed previously, the SDNRb is based on the biomass concentratiomn. The fraction of biomass from the MLVSS reaction declines as the SRT increases, so the SDNR value based on the MLVSS concentration would decrease with increasing SRT. The endogenous respiration SDNR value at 20。C as a function of SRT has also been described by the following empirical relationship (Burdick et al., 1982), in which the SDNR is based on the total MLVSS concentration. Sequencing Batch Reactor Process Analysis In the SBR process and other batch decant processes nitrate removal can be accomplished by three methods: (1) nitrate reduction by using a mixed nonaerated fill period, (2) cycling aeration On/Off during the react period, and (3) operating at a low DO concentration to encourage SNdN. Denitrification during a mixed nonaerated fill period provides the most efficient means of nitrate removal and also provides a selector operation to prevent filamentous sludge bulking. Most of the nitrate produced during the previous aerobic cycle remains in the SBR tank because the decant volume is only 20 to 30 percent of the total tank volume. The mass of nitrate remaining after decant can be reduced during the fill period if sufficient BOD and time are available. The following example is used to illustrate how to estimate the amount of nitrate removed during a mixed fill period for an SBR reactor. Postanoxic Denitrification with an External Carbon Source The biological nitrogen-removal processes discussed thus far in this text are single-sludge suspended growth processes with the electron donor for denitrification provided by BOD in the wastewater (e.g., preanoxic condition) or from mainly endogenous decay of the activated sludge (postanoxic). Prior to the late 1970s or early 1980s, the principal approach for nitrate removal was to add a process (either suspended growth or attached growth) after nitrification. Nitrification/denitrification using suspended growth processes is also termed the two-sludge process. Because only a negligible amount of BOD remains in the nitrified effluent, an external carbon source has to be added to supply the energy for the nitrifying organisms. For postanoxic denitrification, nitrified influent is fed to a mixed anoxic tank along with an external carbon source, which is commonly methanol (see Fig. 7-14e). Sufficient detention time and SRT are needed to consume the methanol with NO3-N as the electron acceptor and to assure good floc settling and thickening characteristics. An SRT of at least 5 d is normally used. The anoxic tank is followed by a short aeration time of 10 to 20 rain to release nitrogen gas bubbles from the mixed liquor to promote maximum suspended solids removal in the final clarifier. As in any suspended growth process, wasting of excess sludge is necessary. The postanoxic suspended growth process can be designed in a similar manner to an activated-sludge process, with the electron donor being the growth substrate. The NO3-N that must be removed serves as the electron acceptor in lieu of dissolved oxygen. Nitrate limits only the process kinetics at very low concentrations (less than 0.30 mg/L); thus, the process design is based on methanol degradation kinetics. The nitrate consumption is treated in a manner similar to supplying oxygen for an aerobic CMAS process. Because the process and the organic substrate used are so well defined, in contrast to the preanoxic processes using wastewater BOD, the SDNR design approach is not appropriate. The design process procedure is outlined as follows: 1. Determine the amount of NO3-N to be reduced. 2. Select the anoxic tank SRT. 3. Calculate the anoxic tank residual methanol concentration based on design SRT and kinetic coefficients. 4. Determine the methanol dose based on the amount of nitrate to be removed and the amount of dissolved oxygen in the influent wastewater to be consumed. 5. Calculate the total solids production. 6. Determine the anoxic tank volume based on the solids production and SRT. 7. Select a detention time for the postaeration tank prior to clarification. Nitrogen Removal in Anaerobic Digestion Recycle Streams Recycle flows from dewatering of anaerobically digested solids contain high NH4-N concentrations (>1000 mg/L) that can increase wastewater influent load by 15 to 20 percent. The recycle steam is also characterized by relatively high temperature and pH. A sidestream treatment process termed the SharonTM (single-reactor high-activity ammonia removal over nitrite) process has been developed at Delft University of Technology in the Netherlands to remove nitrogen biologically from digester recycle in a relatively short detention time reactor. The process takes advantage of the effect of high temperature on nitrification kinetics that favor more rapid growth of ammonia-oxidizing bacteria over nitrite-oxidizing bacteria. A complete-mix reactor without solids recycle is operated with intermittent aeration for nitrification and denitrification. The BOD in the anaerobic recycle stream is low compared to the NH4-N concentration, so for nitrite reduction, methanol is added during the anoxic period to provide an electron donor. In a bench-scale study, 80 to 85 percent nitrogen removal has been reported for an operation having a temperature of 35。C, 1.5 d SRT, 80 min of aeration, and 40 min of anoxic contacting. The SharonTM process can also be operated without the anoxic step and methanol addition to produce a nitrite-rich recycle stream that can be fed to a preanoxic zone or to the plant headworks. Alternative Process Configurations for Biological Nitrogen Removal A variety of activated-sludge process configurations are used to accomplish biological nitrogen removal. Representative process schematics and descriptions for biological nitrogen-removal processes are given in Table 7-10. Tab. 7-10 Description of suspended growth processes for nitrogen removal Process Design Parameters Typical parameters used for the design and operation of various biological nitrogen removal processes are presented in Table 7-12. Process Selection Considerations The selection of a specific process for biological nitrogen removal will depend on site-specific conditions, existing processes and equipment, and treatment needs. The advantage and limitations of the processes commonly used and their treatment capability in terms of effluent total nitrogen concentrations are summarized in Table 7-12. Tab. 7-12 Advantages and limitations of nitrogen removal processes The MLE process is one of most common methods used for biological nitrogen removal, and can be adapted easily to existing activated-sludge facilities. The amount of nitrate removal is limited by the practical levels of internal recycle to the preanoxic zone, and the process is used more generally to achieve effluent total nitrogen concentrations between 5 and 10 mg/L. Dissolved oxygen control should be used in the zone from which the recycle stream is taken to limit the amount of DO fed to the anoxic zone. The step-feed process is also applicable for meeting effluent total nitrogen concentrations of less than 10 mg/L. However, it is theoretically possible to achieve lower effluent nitrogen total concentrations (ranging from less than 3 to 5 mg/L) with step-feed BNR using internal recycle, such as in the MLE process, for the last pass of the anoxic-aerobic step-feed process. As in the MLE process, the DO concentration from the aerobic zone must be controlled to minimize dissolved oxygen addition to the anoxic zone. In the case of the step-feed process, more DO control points are required. Influent flow splitting measurement and control are necessary to optimize the step-feed reactor volume for nitrogen removal. The sequencing batch reactor process provides a high degree of flexibility for nitrogen removal. Mixing during the fill period provides an opportunity for anoxic conditions for nitrate removal. During the aeration react period, the DO concentration may be cycled to provide anoxic operating periods. The batch decant reactor designs are slightly less flexible than the SBR processes because they depend on internal recycle like the MLE for a major portion of the nitrate removal. Bio-denitroTM, NitroxTM, and the oxidation ditch with DO control are all processes with large reactor volumes for nitrogen removal, and represent various methods for optimizing biological nitrogen removal in oxidation ditch systems. Very low effluent total nitrogen concentrations (less than 5 mg/L) have been reported for the Bio-denitroTM process. The NitroxTM process is generally limited to effluent total nitrogen concentrations of 5 to 8 mg/L. During the aeration "off period" in the NitroxTM process, ammonia accumulates in the oxidation ditch, resulting in higher effluent NH4-N concentrations from the process. The effluent NH4-N and total nitrogen concentration is dependent on the total reactor volume and influent nitrogen concentrations. Higher influent TKN concentrations can result in higher effluent ammonia concentrations. The Bardenpho process and postanoxic processes with methanol addition have demonstrated the ability to achieve less than 3 mg/L total nitrogen. The second anoxic zone of the Bardenpho process has a very low denitrification rate, resulting in less efficient reactor volume utilization. The addition of methanol to the second anoxic reduces the reactor volume requirements. The SNdN processes, such as the OrbalTM and Sym-BioTM NADH control process, require large reactor volumes and some operator attention and skill. However, depending on the operating conditions, these processes have been shown to be capable of producing very low effluent total nitrogen concentrations. Besides removing nitrogen, the preanoxic and SNdN processes have additional advantages over the postanoxic processes. By removing nitrate before or during nitrification step, the alkalinity produced by denitrification is made available to offset the alkalinity depleted by nitrification. Because 3.57 g of alkalinity (as CaCO3) are produced per g NO3-N oxidized, and 7.14 g alkalinity (as CaCO3) are used per g NH4-N oxidized, almost half of the alkalinity used for nitrification can be provided by Preanoxic or SNdN processes. The recovery of alkalinity is very important for wastewaters that have low alkalinity. In some applications, alkalinity may have to be added in the form of lime or sodium hydroxide, at significant cost, to maintain an acceptable pH fort the nitrification process. Other advantages of these processes include aeration energy savings and the ability to produce a sludge that settles well. By using nitrate to oxidize influent BOD, the preanoxic and SNdN processes require less oxygen for aeration compared to postanoxic processes. Greater energy savings can be achieved with the SNdN processes as compared to the preanoxic processes, because they are operated at a lower DO concentration. By operating at a lower DO concentration, the DO gradient is larger, which results in more efficient oxygen transfer during aeration. Internal recycle pumping is also eliminated, which accounts for further energy reduction. The postanoxic suspended growth process with methanol addition lacks the benefits of the preanoxic process, i.e., energy savings, alkalinity production, and filamentous bulking control, and has higher operating cost because of the purchase of methanol. The selection of the postanoxic suspended growth process is driven mainly by the site lay out, existing reactor configuration, and equipment considerations. 7-6 Processes for Biological Phosphorus Removal Over the past 20 years, several biological suspended growth process configurations have been used to accomplish biological phosphorus removal, and they all include the basic steps of an anaerobic zone followed by an aerobic zone. Barnard (1974) was the first to clarify the need for anaerobic contacting between activated sludge and influent wastewater before aerobic degradation to accomplish biological phosphorus removal. Other modifications of the basic process include (1) combining the anaerobic/aerobic sequence with various biological nitrogen-removal designs, (2) recycling mixed liquor to the anaerobic zone from a downstream anoxic zone instead of only from the secondary clarifier undertow, (3) adding volatile fatty acids to the anaerobic zone as either acetate or a liquid stream from a fermentation reactor processing primary clarifier sludge, and (4) using multiple-staged anaerobic and aerobic reactors. The alternating exposure to anaerobic conditions can be accomplished in the main biological treatment process, or "mainstremn," or in the return sludge stream, or "side-stream." Several mainstream biological phosphorus-removal processes and one side-stream process, PhostripTM, are described in this section. Also included in this section are process design considerations, process control, analysis of biological phosphorous-removal performance, design parameters, and process selection considerations. The first "mainstream" biological phosphorus-removal process that was included with biological nitrogen removal is the Bardenpho process at Palmetto, FL, shown on Fig. 7-2b. Biological Phosphorus-Removal Processes Three biological phosphorus-removal (BPR) configurations that are more commonly used are shown on Fig. 7-17. Barnard (1975) used the term Phoredox to represent any process with an anaerobic/aerobic sequence to promote BPR. Since then, other process names have evolved m designate specific process configurations, such as the A/OTM (anaerobic/aerobic only) or A2O TM (anaerobic/anoxic/aerobic) processes. The A/OTM process is similar to the Phoredox process and was patented and marketed by Air Products and Chemicals, Inc., as well as the A20TM process. The main difference between the Phoredox (A/O) process and the A2O processes shown on Fig. 7-17a and b is that nitrification does not occur in the Phoredox (A/O) process. Low operating SRTs are used to prevent the initiation of nitrification. Deskable SRT values range from 2 to 3 d at 20。C and 4 to 5 d at 10。C for biological phosphorus removal to occur without nitrification (Grady et al., 1999). For applications where nitrification is needed to meet discharge requirements, the process must also include biological denitrification to prevent excessive amounts of nitrate from entering the anaerobic reactor by way of the RAS recycle. Heterotrophic bacteria will use nitrate to consume rbCOD in the anaerobic zone, which then leaves less rbCOD available for phosphorus-storing bacteria, thus decreasing the biological phosphorus-removal treatment efficiency. The A2O and UCT (University of Cape Town) processes are two basic types of mainstream systems used for nitrate removal with BPR. In the A2O process, the return activated-sludge (RAS) recycle, which contains nitrate, is directed to the anaerobic zone (see Fig. 7-17b). In the UCT process (see Fig. 7-17c), the RAS recycle is directed instead to an anoxic zone, and the mixed-liquor recycle to the anaerobic zone is drawn following the anoxic zone where the nitrate concentration is minimal. The UCT and similar processes (see later discussion in this section) are used generally for relatively weak wastewaters where the addition of nitrate would have a significant effect on the BPR performance. Process Design Considerations The process design considerations for BPR processes include (1) wastewater characteristics, (2) anaerobic contact time, (3) SRT, (4) waste sludge processing method, and (5) chemical addition capability. Wastewater Characteristics. Wastewater characterization, including rbCOD measurements, is essential to evaluate fully the design and performance of BPR systems. Biological phosphorus removal is initiated in the anaerobic zone where acetate (and propionate) is taken up by phosphorus-storing bacteria and converted to carbon storage products that provide energy and growth in the subsequent anoxic and aerobic zones. The rbCOD is the primary source of volatile fatty acids (VFAs) for the phosphorus-storing bacteria. The conversion of rbCOD to VFAs-occurs quickly through fermentation in the anaerobic zone and 7 to 10 mg of acetate results in about 1.0 mg P removal by enhanced phosphorus removal (Wentzel et al., 1989; Wentzel et al., 1990). The more acetate, the more cell growth, and, thus, more phosphorus removal. Because of the need for organic material for nitrate removal, the amount of rbCOD relative to the amount of TKN in the influent is also an important wastewater parameter. The diurnal variation in wastewater strength is also an important process consideration. Because the performance of phosphorus-storing bacteria depends on the availability of fermentation substrates, it is important to know if periods of low influent wastewater strength may affect BPR performance. For domestic wastewaters, the influent total BOD and rbCOD concentrations will vary with time over a 24-h period, with lower concentrations in the late evening and early morning hours. For smaller-sized communities, the variations are usually more pronounced and very little rbCOD may be present at certain times. During wet-weather conditions, especially in the winter, BPR may be difficult to achieve due to cold, low strength wastewater that does not readily become anaerobic. Extended periods of reduced rbCOD concentration have been reported to decrease BPR performance for a number of hours after the occurrence of low substrate concentration (Stephens and Stensel, 1998). The impact of continuous acetate feeding at plants where sludge fermentation has been done to produce additional VFAs has shown the benefit of a steady supply of rbCOD for biological phosphorus removal. In parallel modified Bardenpho trains at Kelowna, Canada, one train was fed fermentation liquor and the other train was used as the control. With continuous VFA addition, the effluent soluble phosphorus concentration decreased from 2.5 to 0.3 mg/L (Oldham and Stevens, 1985), and the VFA/P ratio was 6.7 g/g, an amount lower than the estimated 7 to 10 g/g. Based on these results, it appears that continuous acetate addition may provide more efficient biological phosphorus removal. Anaerobic Contact Time. The role of the anaerobic contact zone has been described in Sec. 7-11. Detention times of 0.25 to 1.0 h are adequate for fermentation of rbCOD. To account for the effect of the MLVSS concentration in the anaerobic contact zone, a l-d SRT is recommended for the anaerobic contact zone design (Grady et al., 1999). Barnard (1984) cautioned against using too long an anaerobic contact time due to the potential for a secondary release of phosphorus, which is phosphorus release not associated with acetate uptake. When secondary release occurs, bacteria have not accumulated polyhydroxybutyrate (PHB) for subsequent oxidation in the aerobic zone. Polyhydroxybutyrate provides energy for phosphorus uptake and storage. From SBR bench-scale studies, Stephens and Stensel (1998) found that secondary phosphorus release occurred for anaerobic contact times in excess of 3.0 h. Solids Retention Time. Processes with longer SRT values had less biological phosphorus removal for a given amount of influent BOD. These data were obtained without measurements of the rbCOD concentration in the wastewater, but provide a general trend on the effect of SRT on BPR removal efficiency. Two adverse effects on phosphorus removal efficiency are associated with lightly loaded long SRT processes. First, because the final amount of phosphorus removed is proportional to the amount of biological phosphorus-storing bacteria wasted, the phosphorus-storing biomass production is lower so that less phosphorus can be removed. Second, at long SRTs the biological phosphorus bacteria are in a more extended endogenous phase, which will deplete more of their intracellular storage products. If the intracellular glycogen is depleted, less efficient acetate uptake and PHB storage will occur in the anaerobic contact zone, thus making the overall BPR process less efficient. Waste Sludge Processing. Because phosphorus is removed in the sludge wasted from BPR processes, consideration must be given to the waste sludge processing methods and the potential to recycle excessive amounts of phosphorus back to the BPR process.Review of the biological phosphorus-removal mechanism showed that phosphorus is released when the bacteria that contain stored phosphorus are subject to anaerobic conditions. Anaerobic conditions in thickening and/or digestion can thus result in the release of significant amounts of phosphorus. The recycle stream from these processes would, in essence, increase the influent phosphorus concentration that would then require a greater amount of influent rbCOD to produce low effluent phosphorus concentrations. Thickening of waste sludge by dissolved air flotation, gravity belt thickeners, or rotary-drum thickeners is preferred over gravity thickening of waste sludge to minimize phosphorus release. Phosphorus release would be expected from anaerobic and aerobic digestion processes as well. However, less phosphorus release and recycle have been observed than expected for anaerobic and aerobic digestion evaluations (Randall et al., 1992). Based on bench-scale aerobic digestion studies on sludge from a BPR process at the Little Patuxent wastewater treatment plant (WWTP) in Howard County, MD, only 20 percent P release was observed. At the York River, VA, WWTP, only 27 percent of the phosphorus removed was released in an anaerobic digester. The formation of phosphorus precipitates, such as struvite and brushnite, was credited with keeping phosphorus out of solution. Direct land application of liquid, digested sludge or dewatered raw sludge followed by stabilization such as composting minimizes recycled phosphorus loads. Chemical Addition Capability. Many BPR facility designs include provisions for phosphorus removal by chemical precipitation in addition to biological removal. Where there are insufficient amounts of rbCOD in the influent wastewater, chemical addition is necessary to provide enough phosphorus removal to meet the effluent discharge concentration needed. Alum or iron salts may be used (see Chap. 6) and may be applied at a number of locations in the liquid stream treatment process. Where effluent filtration is used and the additional amount of phosphorus to be removed is small (less than 2.0 mg/L), chemicals may be added and mixed with the flow before filtration. Chemical addition prior to the secondary clarifier is also possible. Where primary treatment is used, alum or iron salts may be added to remove phosphorus prior to the biological process. The biological process is then depended upon to remove the phosphorus to the low concentration required. Present experience indicates that the phosphorus can be removed biologically to dissolved concentrations as low as 0.20 to 0.30 mg/L, provided sufficient rbCOD is available. Iron salts may be preferred in some cases over alum salts for primary treatment applications, because they have the additional advantage of removing sulfide to help reduce odors. The choice of the chemical addition point can affect the chemical dosage. When added for polishing to achieve low effluent phosphorus concentrations, the metal salt is added at dosages well above the stoichiometric ratio, which is the theoretical amount needed to form a metal/phosphorus precipitate compound. By adding the metal salt before the biological process, stoichiometric ratios can be used to minimize the chemical dose needed. Process Control BPR performance is not just dependent on placing an anaerobic zone in front of the aerobic zone of an activated-sludge process and the amount of rbCOD in the influent wastewater. Process performance is 'affected by a number of operating conditions including (1) nitrate-removal efficiency in processes in which nitrification occurs, (2) process SRT, (3) control of dissolved oxygen entering the anaerobic zone, (4) phosphotos in recycle streams, and (5) the system effluent suspended solids concentration. Effect of Dissolved Oxygen and Nitrate in Recycle Flows. Recycle flows to the anaerobic contact zone must be evaluated in terms of their possible effects on BPR, and some recycle streams should be avoided where possible. Filter backwash recycle flows should be sent to the aerobic zone instead of the anaerobic or anoxic zones. Recycle streams with significant concentrations of DO and nitrate can have an adverse impact on process performance. The nitrate concentration in the RAS flow can have a significant effect on the amount of influent rbCOD that is available for BPR Assuming a synthesis yield of 0.4 g VSS/g rbCOD removed, the amount of rbCOD used by nitrate and oxygen fed to the anaerobic zone can be estimated as follows: Based on the above rbCOD/NO3-N ratio and rbCOD/DO ratio, the impact of DO and nitrate fed into the anaerobic contact zone on the BPR performance can be evaluated. The rbCOD in the influent wastewater added to the anaerobic zone will most likely be removed by bacteria using oxygen and nitrate before it is available for biological phosphotos removal. Effect of Recycle Streams with Released Phosphorus. As discussed under Process Design Considerations, recycle streams from sludge thickening or digestion processes may contain high phosphorus concentrations. Equalization and control of the recycle flow and phosphorus load with time may help to minimize the impact of the recycled phosphorus on effluent quality. By adding the recycle streams during times of the day when the influent wastewater strength is higher, a better possibility exists of removing recycled phosphorus in the waste sludge. Recycle streams may also be treated separately with chemical addition to minimize the phosphorus load to the liquid treatment process. Effluent Suspended Solids. The phosphorus content in the mixed-liquor solids is greater than that from the conventional activated-sludge process due to the biological phosphorus storage. The phosphorus content, on a dry solids basis, may be in the range of 3 to 6 percent (Randall et al., 1992). Thus, the total phosphorus concentration in the effluent can be affected significantly by the system effluent TSS concentration. At 3 to 6 percent phosphorus in the solids, the phosphorus contribution in an effluent having a TSS concentration of 10 mg/L would be 0.3 to 0.6 mg/L, values that are significant if the effluent standard is less than 1.0 mg P/L. Fortunately, most BPR processes exhibit good settling characteristics and have secondary clarifier effluent TSS concentrations of 10 mg/L or less. To provide very low effluent phosphorus concentrations, effluent filtration may be required. Solids Separation Facilities Operation of the solids separation facilities affects the process design and performance. If chemical addition is necessary for effluent polishing to achieve low phosphorus levels in the effluent, sufficient clarifier capacity is required to handle the additional chemical precipitate. Methods to Improve Phosphorus-Removal Efficiency in BPR Systems The performance of BPR systems is very site-specific and depends on the wastewater characteristics and the plant process design and operation. For wastewaters with relatively low influent rbCOD concentrations, effluent soluble phosphorus concentrations may exceed 1.0 to 2.0 mg/L, whereas effluent concentrations below 0.5 or 1.0 mg/L have been shown for higher-strength wastewaters. Methods to improve performance for overall phosphorus removal include the following: 1. Provide supplemental acetate by direct purchase or by primary sludge fermentation. 2. Reduce the process SRT. 3. Add alum or iron salts in primary treatment or for effluent polishing. 4. Reduce the amount of nitrate and/or oxygen entering the anaerobic zone. Two methods used to provide additional rbCOD for biological phosphorus removal are to import (purchase) an exogenous carbon source (i.e., acetate) or to produce VFAs from fermentation of primary clarifier sludge. As shown on Fig. 7-18a, a fermentation reactor provides residence time and mixing of the primary sludge, and VFAs are released through the primary clarifier for feed to the secondary treatment process anaerobic zone. A deeper depth primary clarifier design (Fig. 7-18b) has also been proposed to provide sufficient holding time for the settled primary sludge for hydrolysis and acid fermentation (Barnard, 1984). The undertow sludge is recycled to release VFAs to the liquid stream. Operating issues of odors, mixing, and accumulation of rags in the fermenters must also be considered. The primary tank sludge fermenters are not heated and SRT values ranging from 3 to 5 d, depending on temperature, are generally used to stay below the point where methanogenic activity can start (Rabinowitz and Oldham, 1985). At SRT values greater than 4 to 5 d, methanogenesis activity can be high enough to consume the VFAs. The VFA production ranges from 0.1 to 0.2 g VFA/g VSS applied to the fermenters. Fermenters can add an additional 10 to 20 mg/L VFA concentration to the influent wastewater, and lower, more consistent effluent soluble phosphorus concentrations have been observed when fermenters are used. A good example of process changes that can improve BPR performance has been shown for the Kelowna, British Columbia, plant (Oldham and Stevens, 1985), which was one of the first noted successes with primary sludge fermentation. A two-train modified BardenphoTM facility is used to provide BPR. The first change was to ferment primary sludge in existing tanks that were operated as sludge thickeners with long holding times. The thickener overflow was fed to the Bardenpho anaerobic zone. The fermenter effluent VFA concentration ranged from 110 to 140 mg/L and, when mixed with the primary effluent, resulted in a VFA concentration of 9 to 10 mg/L. The additional VFA supply to the BPR process decreased the average effluent phosphorus concentration from about 1.5 to 0.5 mg/L. Alum was later added before the secondary clarifiers at a dosage of about 8 mg/L (as alum) to further reduce the effluent phosphorus concentration. With both prefermentation and alum addition, the effluent phosphorus concentration averaged less than 0.20 mg/L for the period from September 1989 to August 1990. During 1993, alum addition was stopped and the last two anoxic-aerobic stages of the BardenphoTM system were taken out of service, when a higher effluent nitrate concentration was allowed. Removing the latter two stages resulted in a low SRT for the process, and the average effluent phosphorus concentration decreased to 0.10 mg/L. The other benefit claimed for removing the second anoxic zone was that had longer than necessary detention times for nitrate removal. The second anoxic zone was suspected of causing undesired secondary phosphorus release. Biological Phosphorus-Removal Process Performance The following example is designed to illustrate the methodology used in evaluating the performance of a BPR process. Because biological phosphorus- and nitrogen-removal processes are very complex with many dependent interactions, comprehensive simiulation models (Barker and Dold, 1997) are very useful for evaluating biological phosphorus- and nitrogen-removal designs. Alternative Processes for Biological Phosphorous Removal Various modifications to the basic Phoredox process are used for both biological phosphorus and nitrogen removal. Most deal with different designs that are intended to minimize the amount of nitrate fed to the anaerobic zone, some involve short SRTs, and some involve using multiple stages for the anaerobic, anoxic, and aerobic zones. The principal mainstream processes and the Phostrip sidestream process are described in Table 7-13. Tab. 7-13 Process Design Parameters Typical parameters used in the design of biological phosphorus-removal systems are presented in Table 7-14. Process Select/on Considerations The processes described for biological phosphorus removal all incorporate the necessary anaerobic contacting between influent wastewater and activated sludge, followed by an aerobic zone for bio-oxidation of stored PHB and phosphorus uptake by the polyphosphate-storing bacteria. The specific designs selected are governed more by other treatment needs such as BOD or nitrogen removal. The advantages and limitations of the biological phosphorus-removal processes are presented in Table 7-14. Tab. 7-15 Advantages and limitations of phosphorus removal processes Biological phosphorus-removal efficiency is affected by the overall activated-sludge process as well as the influent wastewater characteristics (Randall et al., 1992). Lower phosphorus-removal efficiency occurs for systems with longer SRTs, more nitrate and/or oxygen input to the anaerobic zone, and less readily biodegradable COD in the influent wastewater. Shorter SRT systems like the VIP or Phoredox process provide more biological phosphorus removal for the same amount of influent BOD than the modified Bardenpho process, for example. In some applications, other sources of rbCOD are obtained to enhance biological phosphorus removal. 7-7 Selection and Design of Physical Facilities for Activated Sludge Process The physical facilities used in the design of activated-sludge treatment systems are presented and discussed in this section. The subjects discussed include (1) the aeration system, (2) aeration tanks and appurtenances, (3) solids separation, and (4) solids separation facilities. Aeration System The aeration system design for the activated-sludge process must be adequate to (1) satisfy the bCOD of the waste, (2) satisfy the endogenous respiration by the biomass, (3) satisfy the oxygen demand for nitrification, (4) provide adequate mixing, and (5) maintain a minimum dissolved oxygen concentration throughout the aeration tank If the oxygen transfer efficiency of the aeration system is known or can be estimated, the actual air requirements for diffused air aeration or installed power of mechanical surface aerators may be determined. To meet sustained organic loadings at peak conditions discussed in Chap. 3, aeration equipment should be designed with a peaking factor of at least 1.5 to 2.0 times the average BOD load. Aeration equipment should also be sized based on a residual dissolved oxygen (DO) of 2 mg/L in the aeration tank at the average load and 1.0 mg/L at peak load. The Ten States Standards require the aeration system to be capable of providing oxygen to meet the diurnal peak oxygen demand or 200 percent of the design average, whichever is larger with a residual DO of 2.0 mg/L. The aeration equipment must be designed with enough flexibility to (1) meet minimum oxygen demands, (2) prevent excessive aeration and save energy, and (3) meet maximum oxygen demands. Aeration Tanks and Appurtenances After the activated-sludge process and the aeration system have been selected and a preliminary design has been prepared, the next step is to design the aeration tanks and support facilities. The following discussion covers (1) aeration tanks, (2) flow distribution and (3) froth control systems. Aeration Tanks. Aeration tanks usually are constructed of reinforced concerete and left open to the atmosphere. In a typical aeration tank using porous tube diffusers, the rectangular shape permits common-wall construction for multiple tanks. The total tank capacity required should be determined from the biological process design. For plants in a capacity range of 0.22 to 0.44 m3/s, at least two tanks should be provided (a minimum of two tanks is preferred for smaller plants as well, for redundancy). In the range of 0.44 to 2.2 m3/s, four tanks are often provided to allow operational flexibility and ease of maintenance. Large plants, over 2.2 m3/s in capacity, should contain six or more tanks. Some of the largest plants have from 30 to 40 tanks arranged in several groups or batteries. Although the air bubbles dispersed in the wastewater occupy perhaps 1 percent of the total volume, no allowance is made for this in tank sizing. If the wastewater is to be aerated with diffused air, the geometry of the tank may significantly affect the aeration efficiency and the amount of mixing obtained. The depth of wastewater in the tank should be between 4.5 and 7.5 m) to maximize the energy efficiency of the diffuser systems. Freeboard from 0.3 to 0.6 m above the waterline should be provided. The width of the tank in relation to its depth is important if spiral-flow mixing is used in the plug-flow configuration. The width-to-depth ratio for such tanks may vary from 1.0:1 to 2.2:1, with 1.5:1 being the most common. In large plants, the channels become quite long and sometimes exceed 150 m per tank. Tanks may consist of one to four channels with round-the-end flow in multiple-channel tanks. The length-to-width ratio of each channel should be at least 5:1. Where complete-mix diffused air systems are used, the length-to-width ratio may be reduced to save construction cost. For tanks with diffusers on both sides or in a grid or panel pattern, greater widths are permissible. The important point is to restrict the width of the tank so that "dead spots" or zones of inadequate mixing are avoided. The dimensions and proportions of each independent unit should be such as to maintain adequate velocities so that deposition of solids will not occur. In spiral-flow tanks, triangular baffles or fillets may be placed longitudinally in the comers of the channels to eliminate dead spots and to deflect the spiral flow. For mechanical aeration systems, the most efficient arrangement is one aerator per tank. Where multiple aerators are installed in the same tank for best efficiency, the length-to-width ratio of the tank should be in even multiples with the aerator centered in a square configuration to avoid interference at the hydraulic boundaries. Two-speed aerators are desirable to provide operating flexibility to cover a wide range of oxygen demand conditions. Freeboard of about 1 to 1.5 m should be provided for mechanical aeration systems. Individual tanks should have inlet and outlet gates or valves so that they may be removed from service for inspection and repair. The common walls of multiple tanks must therefore be able to withstand the full hydrostatic pressure from either side. Aeration tanks must have adequate foundations to prevent settlement, and, in saturated soil, they must be designed to prevent flotation when the tanks are dewatered. Methods of preventing flotation include thickening the floor slab, installing hold-down piles, or installing hydrostatic pressure relief valves. Drains or sumps for aeration tanks are desirable for dewatering. In large plants where tank dewatering might be more common, it may be desirable to install mud valves in the bottoms of all tanks. The mud valves should be connected to a central dewatering pump or to a plant drain discharging to the wet well of the plant pumping station. Dewatering systems are commonly designed to empty a tank in 12 to 24 h. Flow Distribution. For wastewater treatment plants containing multiple units of primary sedimentation basins and aeration tanks, consideration has to be given to equilizing the distribution of flow to the aeration tanks. In many designs, the wastewater from the primary sedimentation basins is collected in a common conduit or channel for transport to the aeration tanks. For efficient use of the aeration tanks, a method of splitting or controlling the fiowrate to each of the individual tanks should be used. Methods commonly used are splitter boxes equipped with weirs or control valves or aeration tank influent control gates. Hydraulic balancing of the flow by equalizing the headloss from the primary sedimentation basins to the individual aeration tanks is also practiced, Flow regimes using a form of step feed particularly need a positive means of flow control Where channels are used for aeration tank influent or effluent transport, they can be equipped with aeration devices to prevent deposition of solids. The air required ranges from 0.2 to 0.5 m3/lin m.min of channel. Froth Control Systems. Wastewater normally contains soap, detergents, and other surfactants that produce foam when the wastewater is aerated. If the concentration of mixed-liquor suspended solids is high, the foaming tendency is minimized. Large quantities of foam may be produced during startup of the process, when surfactants are present in the wastewater. The foaming action produces a froth that contains sludge solids, grease, and large numbers of wastewater bacteria (see Fig. 7-19). The wind may lift the froth off the tank surface and blow it about, contaminating whatever it touches. The froth, besides being unsightly, is a hazard to those working with it because it is very slippery, even after it collapses. In addition, once the froth has dried, it is difficult to remove. It is important, therefore, to consider some method for controlling froth formation, particularly in spiral-flow tanks where the froth collects along the side of the tank, aerated channels, and free-fall from weirs. A commonly used system for spiral-roll tanks consists of a series of spray nozzles mounted above the surface in areas where the froth collects. Screened effluent or clear water is sprayed through these nozzles and physically breaks down the froth as it forms. Another approach is to meter a small quantity of antifoaming chemical additive into the spray water. Nocardia Foam Control. Nocardia foam is a thick layer of brown, biological foam that forms on the top of aeration tanks and clarifiers. When the Nocardia organisms grow in sufficient numbers, they tend to trap air bubbles that subsequently float to the surface and accumulate as scum. When Nocardia foam occurs, it should be removed from the system, as it will also cause foaming problems in anaerobic and aerobic digesters. Nocardia foam has been controlled by spraying a chlorine solution directly into the foam layer. In some cases, spray nozzles have been installed within a hood located across the width of plug-flow aeration tanks. The addition of a cationic polymer to the activated-sludge process has also been used to control the production of Nocardia foam (Shao et al., 1997). Chlorinating the return activated sludge (RAS) for bulking control may not be effective, as it may cause floc breakup and inhibit BOD removal and nitrification. Solids Separation The separation of solids in the activated-sludge process is a very important function in order to provide well-clarified effluent and concentrated solids that are returned to the biological treatment system or wasted to the solids processing facilities. The design secondary clarifiers following suspended growth biological treatment considers two functions: (1) a sufficient time is needed to provide gravity settling of particles and a relatively clear liquid and (2) thickening of the settled solids to provide higher solids concentration in the return activated sludge. For solids-thickening considerations, the solids loading rate (kg/m2.d), which is the rate of solids feed relative to the clarifier cross-sectional area, is the primary process parameter. In this section, methods are presented to determine solids loading rates as a function of the sludge-settling properties and clarifier return sludge flowrate. The analysis to determine acceptable loadings requires knowledge of the sludge-thickening characteristics. Thus, the procedures presented in this section are more applicable for the evaluation and optimization of existing systems than the design for new systems. Two methods of clarifier analysis are described: solids flux and state point analyses: Both methods are based on the same principles. Solids Flux Analysis. The area required for thickening of the applied mixed liquor depends on the limiting solids flux that can be transported to the bottom of the sedimentation basin. Because the solids flux varies with the characteristics of the sludge, column settling tests should be conducted to determine the relationship between the sludge concentration and the settling rate. The required area can then be determined using the solids flux analysis procedure described below. The depth of the thickening portion of the sedimentation tank must be sufficient to (1) ensure maintenance of an adequate sludge blanket depth so that unthickened solids are not recycled, and (2) temporarily store excess solids that may be applied. A method used to determine the area required for hindered settling is based on an analysis of the solids (mass) flux. Data derived from settling tests must be available when applying this method, which is based on an analysis of the mass flux (movement across a boundary) of the solids in the settling basin. Design of Solids Separation Facilities Solids separation is the final step in the production of a well-clarified, stable effluent low in BOD and suspended solids and, as such, represents a critical link in the operation of an activated-sludge treatment process. Although much of the information presented for the design of primary sedimentation tanks is applicable, the presence of a large volume of flocculent solids in the mixed liquor requires that special consideration be given to the design of activated-sludge settling tanks. As mentioned previously, these solids tend to form a sludge blanket in the bottom of the tank that will vary in thickness. The blanket may fill the entire depth of the tank and overflow the weirs at peak flowrates if the return sludge pumping capacity or the size of the settling tank is inadequate. Further, the mixed liquor, on entering the tank, has a tendency to flow as a density current, interfering with the separation of the solids and the thickening of the sludge. To cope successfully with these characteristics, the following factors must be considered in the design of secondary sedimentation tanks: (1) tank types, (2) surface and solids loading rates, (3) sidewater depth, (4) flow distribution, (5) inlet design, (6) weir placement and loading rates, and (7) scum removal. Tank Types. The most commonly used types of activated-sludge settling tanks are either circular or rectangular. Square tanks are used on occasion but they are not as effective in retaining separated solids as circular or rectangular tanks. Solids accumulate in the comers of the square tanks and are frequently swept over the weirs by the agitation of the sludge collectors. Circular tanks have been constructed with diameters ranging from 3 to 60 m, although the more common range is from 10 to 40 m. The tank radius should preferably not exceed five times the sidewater depth. Two basic types of circular tanks are used for secondary sedimentation: center-feed and rim-feed. Both types use a revolving mechanism to transport and remove the sludge from the bottom of the clarifier. Mechanisms are of two types: those that scrape or plow the sludge to a center hopper similar to the types used in primary sedimentation tanks, and those that remove the sludge directly from the tank bottom through suction orifices that serve the entire bottom of the tank in each revolution. Of the latter, in one type the suction is maintained by reduced static head on the individual suction pipes (Fig. 7-20a). In another patented suction system, sludge is removed through a manifold either hydrostatically or by pumping. Spiral-type scrapers are also used to accelerate movement of settled solids from the tank periphery to the collection sump (Fig. 7-20b). Rectangular tanks must be proportioned to achieve proper distribution of incoming flow so that horizontal velocities are not excessive. The maximum length of rectangular tanks normally should not exceed 10 times the depth, but lengths up to 90 m have been used successfully in large plants. Where widths of rectangular tanks exceed 6 m, multiple sludge collection mechanisms may be used to permit tank widths up to 24 m. Regardless of tank shape, the sludge collector selected should be able to meet the following operational conditions: (1) the collector should have enough capacity so that when a high sludge-recirculation rate is desired, channeling of the over-lying liquid through the sludge will not result, and (2) the mechanism should be sufficiently rugged to transport and remove very dense sludges that could accumulate in the settling tank during periods of mechanical breakdown or power failure. Two types of sludge collectors are commonly used in rectangular tanks: (1) traveling flights, and (2) traveling bridges (see Fig. 7-21). Traveling flights are similar to those used for the removal of sludge in primary settling tanks. For very long tanks, it is desirable to use two sets of chains and flights in tandem with a central hopper to receive the sludge to minimize the sludge transport distance. Sludge may be collected at the influent or effluent end of the tank. The traveling bridge, which is similar to a traveling overhead crane, travels along the sides of the sedimentation tank or on a support structure if several bridges are used. The bridge serves as the support for the sludge-removal system, which usually consists of a scraper or a suction manifold from which the sludge is pumped. The sludge is discharged to a collection trough that runs the length of the tank. Other types of settling tanks that are used include stacked clarifiers, tube and plate settlers, and intrachannel clarifiers. Stacked clarifiers are used in installations where limited land area is available for clarifiers. Stacked clarifiers are used at the Deer Island Wastewater Treatment Plant in Boston, MA, for secondary sedimentation and were selected because of limited land area. The efficiency of conventional or shallow clarifiers may be improved by the installation of tubes or parallel plates to establish laminar flow. Constructed of bundles of tubes or plates set at selected angles (usually 60O) from the horizontal, tube and plate settlers have a very short settling distance and circulation is dampened because of the small size of the tubes. Solids that collect in the tubes or on the plates tend to slide out due to gravitational forces. The major drawback in wastewater treatment is a tendency of the tubes and plates to clog because of the accumulation of biological growths, grease, and small objects that pass through coarse screens. Intrachannel clarifiers are proprietary devices that have been developed to improve the performance of the oxidation ditch activated-sludge process and to save the cost of constructing external final clarifiers. The concept incorporates a clarifier, configured somewhat similar to a boat, installed in the oxidation ditch channel. These devices permit liquid and solids separation and sludge return to occur within the aeration channel Sludge wasting is done from either the aeration channel or the intrachannel clarifier. Disadvantages of intrachannel clarifiers include loss of independency of reactors and clarifiers, wasting dilute sludge, loss of RAS chlorination potential for controlling bulking sludge, and lack of metered RAS control (WEF, 1998). Surface and Solids Loading Rates. When it is necessary to design secondary settling facilities without the benefit of settling tests, published values for surface and solids loading rates are often used. Because of the large amount of solids that may b e lost in the effluent if design criteria are exceeded, effluent overflow rates based on peak flow conditions are used commonly. An alternative design basis is to use the average dry-weather flowrate (ADWF) and a corresponding surface loading rate and also check for peak flow and loading conditions. Either condition may govern the design parameters. Sidewater Depth. Liquid depth in a secondary clarifier is normally measured at the sidewall in circular tanks and at the effluent end wall for rectangular tanks. The liquid depth is a factor in the effectiveness of suspended solids removal and in the concentration of the return sludge. Other factors such as inlet design, type of sludge-removal equipment, sludge blanket depth, and weir type and location also affect clarifier performance. In recent years, the trend has been toward increasing liquid depths to improve overall performance. Typical sidewater depths are presented. Current practice favors a minimum sidewater depth of 3.5 m for large secondary clarifiers, and depths ranging up to 6 m have been used. The cost of tank construction has to be considered in selecting a sidewater depth, especially in areas of high groundwater levels. Tanks with depths less than 3.5 m often have difficulty containing the typically low-density activated sludge, and low-density sludge blankets are more easily disturbed by hydraulic fluctuations. Deeper tanks therefore provide greater flexibility of operation and a larger margin of safety when changes in the activated-sludge system occur. Flow Distribution. Flow imbalance between multiple process units can cause under- or overloading of the individual units and affect overall system performance. In plants where parallel tanks of the same size arc used, flow between the tanks should be equalized. In cases where the tanks are not of equal capacity, flows should be distributed in proportion to surface area. Methods of flow distribution to the secondary sedimentation tanks include weirs, flow distribution boxes, flow control valves, hydraulic distribution using hydraulic symmetry, and feed gate or inlet port control. Effluent weir control, although frequently used to effect flow splitting, is usually ineffective and should be used only where there are two tanks of equal size. Tank Inlet Design. Poor distribution or jetting of the tank influent can increase the formation of density currents and scouring of settled sludge, resulting in unsatisfactory tank performance. Tank inlets should dissipate influent energy, distribute the flow evenly in horizontal and vertical directions, mitigate density currents, minimize sludge blanket disturbance, and promote flocculation. In circular center-feed tanks, a common design is to use small, solid-skirted, cylindrical baffles to dissipate the influent energy and distribute flow. Studies indicate that a density current waterfall can be created using skirted baffles resulting in poor vertical flow distribution (Crosby and Bender, 1980). Methods to overcome these problems include the use of a large ceenter diffusion well or a flocculating-type clarifier. The large center diffusion well, with a minimum diameter of 25 percent of the tank diameter, provides a greater area for dissipation of the influent energy and distribution of the incoming mixed liquor. The bottom of the feed well should end well above the sludge blanket interface in order to minimize turbulence and resuspension of the solids. Flocculating center-feed clarifiers can incorporate an energy-dissipating inlet (EDI) and means to promote flocculation in the center-feed well (see Fig. 7-23a). Typical flocculation feed wells have diameters of 30 to 35 percent of the tank diameter. An alternative device for dissipating energy, developed by the City of Los Angeles, is shown on Fig. 7-23b. Operationally, the flow is discharged from a centerwell through a series of downward-facing discharge ports. By arranging the discharge ports so they discharge facing each other, the momenturn energy is dissipated as the discharge streams impact each other. In rectangular tanks, inlet ports or baffles should be provided to achieve flow distribution. Inlet port velocities are typically 75 to 150 mm/s (WPCF, 1985). Weir Placement and Loading. When density currents occur in a secondary clarifier, mixed liquor entering the tank flows along the tank bottom until it encounters a countercurrent pattern or an end wall. Unless density currents are considered in the design, solids may be discharged over the effluent weir. Experimental work performed by Anderson (1945) at Chicago on tanks approximately 38 m in diameter indicated that a circular weir trough placed at two-thirds to three-fourths of the radial distance from the center was in the optimum position to intercept well-clarified effluent. With low surface loadings and weir rates, the placement of the weirs in small tanks does not significantly affect the performance of the clarifier. Circular clarifiers are manufactured with overflow weirs located near both the center and the perimeter of the tank. If weirs are located at the tank perimeter or at end walls in rectangular tanks, a baffle should be provided to deflect the density currents toward the center of the tank and away from the effluent weir. Alternative baffle arrangements are shown on Fig. 7-24. Weir loading rates are used commonly in the design of clarifiers, although they are less critical in clarifier design than hydraulic overflow rates. Weir loading rates used in large tanks should preferably not exceed 375 m3/lin m.d of weir at maximum flow When located away from the upturn zone of the density current, or 250 m3/lin m,d when located within the upturn zone. In small tanks, the weir loading rate should not exceed 125 m3/lin m-d at average flow or 250 m3/lin m.d at maximum flow. The upflow velocity in the immediate vicinity of the weir should be limited to about 3.5 to 7 m/h. Scum Removal. In many well-operating secondary plants, very little scum is formed in the secondary clarifiers. However, occasions arise when some floating material is present, necessitating its removal. Where primary settling tanks are not used, skimming of the final tanks is essential. Most designs in recent years provide scum removal for both circular and rectangular secondary clarifiers. Typical scum-removal equipment includes beach and scraper type, rotating pipe-through skimmer, and slotted pipes. Scum should not be returned to the plant headworks because microorganisms responsible for foaming (typically Nocardia) will be recycled, causing foaming problems to persist because of continuous seeding of the unwanted microorganisms. In some plants, scum is discharged to sludge-thickening facilities or is added directly to digester feed streams, as appropriate. 7-8 Suspended Growth Aerated Lagoons Suspended growth lagoons are relatively shallow earthen basins varying in depth from 2 to 5 m, provided with mechanical aerators on floats or fixed platforms. The mechanical aerators are used to provide oxygen for the biological treatment of wastewater and to keep the biological solids in suspension. In a few cases, diffused aeration has also been used. Suspended growth aerated lagoons are operated on a flow-through basis or with solids recycle. The purpose of this section is to (1) describe the principal types for suspended growth lagoons and (2) present and illustrate the design consideration for the suspended growth flow-through type of lagoon. Lagoons with solids recycle are essentially the same as the activated-sludge process. The dual-powered flow-through lagoon system, a modification of the conventional flow-through lagoon, is also considered in this section. Tab. 7-16 Typical characteristics of different types types of aerated aerated suspended growth lagoons Types of Suspended Growth Aerated Lagoons The principal types of suspended growth lagoon processes, classified based on the manner in which the solids are handled (Arceivala, 1998), are: 1. Facultative partially mixed 2. Aerobic flow through with partial mixing 3. Aerobic with solids recycle and nominal complete mixing Differences in the manner in which the solids are handled will affect the treatment efficiency, power requirements, hydraulic and solids retention time, sludge disposal, and environmental considerations. The general characteristics of these lagoon systems are summarized in Table 7-16. Each of the lagoon processes is described below. Facultative Partially Mixed Lagoon. In the facultative partially mixed lagoon system, the energy input is only sufficient to transfer the amount of oxygen required for biological treatment, but is not sufficient to maintain the solids in suspension. Because the energy input will not maintain the solids in suspension, a portion of the incoming solids will settle along with a portion of the biological solids produced from the conversion of the soluble organic substrate. In time, the settled solids will undergo anaerobic decomposition. The term facultative is derived from the observation that the biological conversion, which occurs in the lagoon, is partially aerobic and partially anaerobic. Eventually facultative lagoons must be dewatered and the accumulated solids removed. Because there is essentially no way to manage the events in an aerated facultative lagoon (e.g., wind-driven circulation patterns), the use of facultative lagoons has diminished, especially where discharge limits must be met reliably. Because of the lack of a rational approach to the analysis of facultative lagoons, they are not considered further in this section. Aerobic Flow-Through Partially Mixed Lagoon. In aerobic flow-through lagoons, the energy input is sufficient to meet the oxygen requirements, but is insufficient to maintain all of the solids in suspension. Operationally, the hydraulic and solids retention times are the same (i.e., r = SRT). The solids in the effluent are removed prior to discharge in an external sedimentation facility. The analysis and design of aerobic flow-through lagoons is considered in detail in the balance of this section. Aerobic Lagoon with Solids Recycle. The aerobic lagoon with solids recycle is essentially the same as the extended aeration activated-sludge process, with the exception that an earthen (typically lined) basin is used in place of a reinforced-concrete reactor basin. As shown in Table7-16, the hydraulic detention time, at the high end (i.e., 2 d), is longer than that used in a conventional extended aeration process (1 d). It should be noted that the aeration requirement for an aerobic lagoon with recycle must be higher than the values for an aerobic flow-through lagoon to maintain the solids in suspension. The power levels given in Table 7-16 are designed to maintain most of the solids in suspension, hence the term nominal complete mix is used. As noted in the introductory paragraph, because the analysis for the aerobic lagoon with recycle is the same as the activated-sludge process, this process will not be considered further in the following discussion. Process Design Considerations for Flow-Through Lagoons Factors that must be considered in the process design of suspended growth flow-through lagoons include (l) BOD removal, (2) effluent characteristics, (3) temperature effects, (4) oxygen requirement, (5) energy requirement for mixing, and (6) solids separation. BOD Removal. Because a suspended growth aerobic flow-through lagoon can be considered to be a complete-mix reactor without recycle, the basis of design is SRT, which in this case is equal to the hydraulic retention time τ under ideal flow conditions. The SRT should be selected to assure that (1) the suspended microorganisms will bioflocculate for easy removal by sedimentation and (2) an adequate safety factor is provided. Typical design values of SRT for aerated lagoons used for treating domestic wastes vary from about 3 to 6 d. Once the value of SRT has been selected, the soluble substrate concentration of the effluent can be estimated, and the removal efficiency can then be computed using the equations given earlier in this chapter. An alternative approach is to assume that the observed BOD removal (either overall, including soluble and suspended solids contribution, or soluble only) can be described in terms of a first-order removal function. The BOD removal is measured between the influent and the lagoon outlet (not the outlet of the sedimentation facilities following the lagoon). Effluent Characteristics. The important characteristics of the effluent from an aerated lagoon include the BOD and the TSS concentration. The effluent BOD will be made up of those components previously discussed in connection with the activated-shdge process and occasionally may contain the contribution of small amounts of algae. The solids in the effluent are composed of a portion of the incoming suspended solids, The biological solids produced from waste conversion, and occasionally small amounts of algae. It should be noted that if the effluent from a suspended growth flow-through aerated lagoon is to meet the minimum standards for secondary treatment, settling facilities are needed. Where greater removal of effluent solids is required, slow sand or rock filters are used commonly (Rich, 1999). Temperature. Because suspended growth aerobic flow-through lagoons are often installed and operated in locations with widely varying climatic conditions, the effects of temperature change must be considered in their design. The two most important effects of temperature are (1) reduced biological activity and treatment efficiency and (2) the formation of ice. Alternatively, if climatological data are available, the average temperature of the lagoon may be determined from a heat budget analysis by assuming the lagoon is completely mixed. It should be noted that surface aerators tend to further cool lagoons in cold weather, but submerged diffused air systems add heat to some extent. Where icing may be a problem, the effect on the operation of lagoons may be minimized by increasing the depth of the lagoon or by altering the method of operation. As computed, reducing the area by one-half increases the temperature about 2.2。C, which corresponds roughly to about a 20 percent increase in the rate of biological activity. As the depth of the lagoon is increased, maintenance of a completely mixed flow regime becomes difficult. If the depth is increased much beyond 3.7 m, draft tube aerators or diffused aeration must be used. In multiple lagoon systems, cold-weather effects can be mitigated by seasonal changes in the method of operation. During the warmer months, the lagoons would be operated in parallel; in the winter, they would be operated in series. In the winter operating mode, the downstream aerators could be turned off and removed, and the lagoon surface is allowed to freeze. In spring when the ice melts, the parallel method of operation is again adopted. With this method of operation it is possible to achieve a 60 to 70 percent removal of BOD even during the coldest winter months. Still another method that can be used to improve performance during the winter months is to recycle a portion of the solids removed by settling. Oxygen Requirement, Based on operating results obtained from a number of industrial and domestic installations, the amount of oxygen required has been found to vary from 0.7 to 1.4 times the amount of BOD removed. Mixing Requirements. The power requirements for maintaining the solids in suspension are functions of several variables including (1) the type and design of the aeration system, (2) the concentration and nature of the suspended solids, (3) the temperature of the lagoon contents, and (4) the size and geometry (i.e., aspect ratio) of the lagoon. Other relationships have been proposed by McKinney and Benjes (1965), Fleckseder and Malina (1970), and Murphy and Wilson (1974). The threshold energy input value for the suspension of biosolids is about 1.5 to 1.75 kW/l03 m3. The energy requirement to maintain all of the solids in suspension is considerably greater than the threshold value. As important as the amount of energy required is the placement of the aerators, to achieve effective dispersion of the oxygen and mixing with sufficient overlap of the dispersion and mixing zones. Because so many different types of aerators are used in aerated lagoons (e.g., high and low speed and aspirating aerators, draft tube aerators, static tubes, and perforated tubing) manufacturers' literature should be consulted to determine appropriate aerator spacing. In a report prepared for the U.S. EPA, the recommended maximum spacing between surface aerators should not exceed 75 m (Crown Zellerbach, 1970). In general, a number of smaller aeration devices, spaced more closely, will be more effective than a few larger devices (Rich, 1999). For depths greater than 3.7m, aerators with draft tubes may be considered to prevent solids deposition. Solids Separation. The separation of solids from suspended growth flow through aerated lagoons is accomplished most commonly in a large shallow earthen sedimentation basin (lagoon) designed expressly for the purpose, or in more conventional settling facilities. Where large earthen basins are used, the following requirements must be considered carefully: (1) the detention time must be adequate to achieve the desired degree of suspended solids removal, (2) sufficient volume must be provided for sludge storage, (3) algal growth must be minimized, (4) odors that may develop as a result of the anaerobic decomposition of the accumulated sludge must be controlled, and (5) the need for a lining must be assessed. In some cases, because of local conditions, these requirements may be in conflict with each other. In most cases, a minimum detention time of 1 d is required to achieve solids separation (Adams and Eckenfelder, 1974). If a 1-d detention time is used, adequate provision must be made for sludge storage so that the accumulated solids will not reduce the actual liquid detention time. Further, if all the solids become deposited in localized patterns, it may be necessary to increase the detention time to counteract the effects of poor hydraulic distribution. Under anaerobic conditions, about 40 to 60 percent of the deposited volatile suspended solids will be degraded each year. Two problems that are often encountered with the use of settling basins are the growth of algae and the production of odors. Algal growths can usually be controlled by limiting the hydraulic detention time to 2 d or less. If longer detention times must be used, the algal content may be reduced by using either a rock filter or a microstrainer. Odors arising from anaerobic decomposition can generally be controlled by maintaining a minimum water depth of 1 m. In extremely warm areas, depths up to 2 m have been needed to eliminate odors, especially those of hydrogen sulfide. If space for large settling basins is unavailable, conventional settling facilities can be used. To reduce the construction costs associated with conventional concrete and steel settling tanks, lined earthen basins can be used. As noted previously, where greater removal of effluent solids is required, it has become common practice to use slow sand or rock filters (Rich, 1999). Rock filters, which may be effective in algae removal, consist of a submerged bed of rocks through which lagoon effluent passes vertically or horizontally. Although rock filters are simple to operate and maintain, effluent quality and dependability of long-term operation are not well documented (U.S. EPA, 1992). Dual-Powered Flow-Through Lagoon System The dual-powered flow-through lagoon system is a modification of the conventional flow-through lagoon pioneered by Professor Rich (1982, 1996, and 1999). The dual-powered flow-through lagoon system is comprised of a complete-mix lagoon followed by two or three facultative lagoons that serve as settling basins. The operating principle of the dual-powered flow-through lagoon system may be outlined as follows: From basic considerations, it can be reasoned that secondary treatment involves the following functions: (1) bioconversion of the influent substrate to biomass and end products, (2) flocculation of the biomass, (3) separation of the solids, (4) stabilization of the solids, and (5) storage of the solids until they are reused or disposed of. These functions are optimized in the dual-powered flow-through lagoon system. Bio-conversion and flocculation occur in the first complete-mix lagoon. The solids separation, stabilization, and storage are accomplished in the facultative lagoons following the complete-mix lagoon. To minimize the growth of algae, the detention time should be limited and the facultative lagoons should be divided into a series of cells (Rich, 1999). The hydraulic detention time for the complete-mix lagoon will typically vary from 1.5 to 3 d. The energy input in the complete-mix lagoon is on the order of 6 kW/103 m3. The total retention time for the facultative lagoons is on the order of 3d The energy input into the facultative lagoons is below the threshold value for the suspension of solids, but is sufficient to oxidize soluble organic' constituents released from the anaerobic degradation of the solids accumulated on the bottom of the facultative lagoons. The corresponding energy input of the facultative lagoons is on the order of 1.0 to 1.25 kW/103 m3). The overall detention time for the two lagoons is typically 4.5 to 6 d. The same procedure used for the design of the flow-through suspended growth lagoon, as detailed in Example 8-14, is used for the design of the complete-mix lagoon of the dual-powered lagoon system. 7-9 Biological Treatment with Membrane Separation Membrane biological reactors (MBRs), consisting of a biological reactor (bioreactor with suspended biomass and solids separation by microfiltration membranes with nominal pore sizes ranging from 0.1 to 0.4 gm, are finding many applications in wastewater treatment. Membrane biological reactor systems may be used with aerobic or anaerobic suspended growth bioreactors to separate treated wastewater from the active biomass. The membrane systems described in this section produce an effluent quality equal to the combination of secondary clarification and effluent microfiltration Membrane biological reactors have been used for treatment of both municipal and industrial wastewater (Brindle and Stephenson, 1996; Van Dijk and Roncken, 1997; Trussell et al., 2000) and for water-reuse applications (Cicek et al., 1998). Overview of Membrane Biological Reactors The concept of MBR systems consists of utilizing a bioreactor and microfiltration as one unit process for wastewater treatment thereby replacing, and in some cases supplementing, the solids separation function of secondary clarification and effluent filtration. The ability to eliminate secondary clarification and operate at higher MLSS concentration provides the following advantages: ( 1 ) higher volumetric loading rates and thus shorter reactor hydraulic retention times; (2) longer SRTs resulting in less sludge production (3) operation at low DO concentrations with potential for simultaneous nitrification denitrification in long SRT designs; (4) high-quality effluent in terms of low turbidity bacteria, TSS, and BOD; and (5) less space required for wastewater treatment. Disadvantages of MBRs include high capital costs, limited data on membrane life, potential high cost of periodic membrane replacement, higher energy costs, and the need to control membrane fouling. Process Description Membrane bioreactor systems have two basic configurations: (1) the integrated bioreactor that uses membranes immersed in the bioreactor and (2) the recirculated MBR in which the mixed liquor circulates through a membrane module situated outside the bioreactor. In the integrated MBR system, the key component is the microfiltration membrane that is immersed directly into the activated-sludge reactor. The membranes are mounted in modules (sometimes called cassettes) that can be lowered into the bioreactor. The modules are comprised of the membranes, support structure for the membranes, feed inlet and outlet connections, and an overall support structure. The membranes are subjected to a vacuum (less than 50 kPa) that draws water (permeate) through the membrane while retaining solids in the reactor. To maintain TSS within the bioreactor and to clean the exterior of the membranes, compressed air is introduced through a distribution manifold at the base of the membrane module. As the air bubbles rise to the surface, scouring of the membrane surface occurs; the air also provides oxygen to maintain aerobic conditions. A proprietary integrated MBR system is shown on Fig. 7-25, the Zenogem process manufactured by Zenon Environmental, Inc. The Kubota submerged membrane bioreactor marketed by Enviroquip is similar. The Zenogem microfiltration cassette shown on Fig. 7-25 is composed of hollow fiber membranes and has overall dimensions of 0.91 m wide by 2.13 m long, and approximately 2.44 m high. Support facilities required include permeate pumps, chemical storage tanks, chemical feed pumps and all process controls for the membranes, including the motor control center for the membrane process. The membrane support equipment also includes an air-scour system and a back-pulse water-flushing system. The air-scour system consists of coarse bubble diffusers located in the aeration basin and provides continuous agitation on the outside of the membranes to minimize solids deposition. The air supply for the air-scour system is typically provided in addition to the activated-sludge process air. The membrane unit manufactured by Kubota consists of a series of membrane cartridges composed of fine porous membranes mounted on both sides of a supporting plate. The cartridges, which have a thickness of 6 m, can be removed individually for cleaning and replacement. Each of the cassettes or modules has a series of tubes that extract effluent from the membranes and connect to a collection manifold. The permeate support facilities are similar to those for the Zenon system. As of 2001, in addition to Zenon and Kubota, submerged MBR systems in the United States are also marketed by Mitsubishi Rayon Corporation (Mitsubishi) from Japan (Merlo et al., 2000). Besides submerged membranes, an in-line, recirculated MBR design is being marketed by Lyonnaise-des-Eaux Degremont of France. For this device, activated sludge from the bioreactor is pumped to a pressure-driven tubular membrane where solids are retained inside the membrane and water passes through to the outside. The driving force is the pressure created by high cross velocity through the membrane. The solids are recycled to the activated-sludge basin. The membranes are backwashed systematically to remove solids and cleaned chemically to control pressure buildup. By replacing solids separation by gravity settling in secondary clarifiers, membranes avoid issues of filamentous sludge bulking and other floc settling and clarification problems, and the aeration tank MLSS concentration is no longer controlled by secondary clarifier solids loading limitations. The MBR systems can operate at much higher MLSS concentrations (15,000 to 25,000 mg/L) than conventional activated-sludge processes (Cote et al., 1998). Although high concentrations of MLSS as noted above have been reported, MLSS concentrations in the range of 8000 to 10,000 mg/L appear to be most cost-effective when all factors are considered. The membrane flux rate, defined as the mass or volume rate of transfer through the membrane surface [in terms of L/m2.h is an important design and operating parameter that affects the process economics. Lower flux rates are expected at higher MLSS concentrations. As more applications occur for MBR with activated-sludge treatment, more information will be available for long-term membrane flux rates and membrane life. Membrane Fouling Control In the activated-sludge reactor, biomass coats the outer layer of the membranes used in an integrated MBR during effluent withdrawal. Finer particles may penetrate the inner pores of the membrane, causing an increase in pressure loss. Continuous membrane fouling control methods are used during the operation of the MBR, with periodic more aggressive cleaning to maintain the filtration capacity of the membrane. A method developed by Zenon Environmental to control fouling on the outside surfaces of the membrane fibers involves a three-step process. First, coarse bubble aeration is provided at the bottom of the membrane tank directly below the membrane fibers. The air bubbles flow upward between the vertically oriented fibers, causing the fibers to agitate against one another to provide mechanical scouring. Second, filtration is interrupted about every 15 to 30 min and the membrane fibers are backwashed with permeate for 30 to 45 s (Giese and Larsen, 2000). The system remains on-line during backwashing, and the total time for backflushing is about 45 min per day. Typically, a low concentration of chlorine (<5 mg/L) is maintained in the backflush water to inactivate and remove microbes that colonize the outer membrane surface. Third, about three times per week a strong sodium hypochlorite solution (about 100 mg/L) or citric acid is used in the backflush mode for 45 min in a procedure called a "maintenance clean." After the 45-min in situ cleaning, the system is flushed with permeate for 15 min. An additional permeate flush-to-drain operation is performed for 10 to 15 min to purge the system of free chlorine once the vacuum is initiated. The total system downtime during a maintenance clean is about 75 min. The back-pulse system is similar to a typical flushing or cleaning system. A chlorine solution is pumped periodically through the membranes in the reverse direction, similar to backwashing a filter. The cassettes can be removed easily from the aeration basin by an overhead hoist system for a periodic chemical-bath cleaning. When removed for cleaning, the cassettes are submerged in a high-concentration chlorine solution bath in a separate small tank or basin located adjacent to the aeration basin. External cleaning occurs about every 3 to 6 months. The combination of air scour, backflushing, and maintenance cleaning is not completely effective in controlling membrane fouling, and the pressure drop across the membrane increases with time. The pressure drop across the membranes is monitored to indicate fouling problems and cleaning needs. At a maximum operating pressure drop of ≈60 kPa, the membranes are removed from the aeration basin for a recovery cleaning (Fernandez et al., 2000). During recovery cleaning, a membrane cassette is soaked in a tank containing a 1500 to 2000 mg/L sodium hypochlorite solution for about 24 h. Spare membranes are typically installed in the aeration tank during recovery cleaning so that there is no reduction in treatment capacity. A similar fouling control procedure has been reported for the Mitsubishi MBR membrane (Merlo et al., 2000). In the Kubota MBR process, the flat-plate membranes are not removed for cleaning, and an infrequent backflush with a 0.5 percent solution of hypochlorite has been shown to be effective (Stephenson et al., 2000). Process Capabilities The treatment capability of MBR is evaluated in terms of BOD, TSS, coliform, and nitrogen removal based on laboratory, pilot-plant, and full-scale plant studies. Because the activated-sludge effluent from MBRs is treated by filtration through a nominal 0.40μm membrane, very low concentrations of effluent suspended solids, turbidity, and BOD are produced that provide an effluent suitable for water reuse following disinfection. Low effluent BOD and turbidity concentrations are possible for MBR systems with MLSS concentrations in the range of 6000 to 16,000 mg/L. Full-scale and pilot-plant systems have been operated with the anoxic/aerobic MLE biological nitrogen-removal process with the result that effluent total nitrogen concentrations of <10 mg/L have been achieved (Mourato et al., 1999; ReVoir et al., 2000; and Giese et al., 2000). Influent recycle flowrate ratios of 4.0 to 6.0 have been used in those studies to feed nitrate to a separate preanoxic tank. Nitrogen removal in MBR systems has also been done in bench- and pilot-scale tests by using intermittent aeration or operation at low DO concentration to achieve SNdN. An intermittent aeration study was conducted with an MLSS concentration of 6000 mg/L and on/off aeration intervals of 15 min (Fernandez et al., 2000). An effluent total nitrogen concentration ranging from 7 to 10 mg/L was achieved. Full nitrification was not possible at high MLSS concentrations (up to 13,000 mg/L), probably due to DO limitations in the activated sludge. In a laboratory study, an MBR was operated successfully at low DO concentrations (< 1.0 mg/L) to accomplish SNdN (Choo and Stensel, 2000).