y>EPA United States Environmental Protection Agency Municipal Environmental Research Laboratory Cincinnati OH 45268 Research and Development EPA-600/S2-81-155 Sept. 1981 Project Summary Parallel Evaluation of Air- and Oxygen-Activated Sludge Scott Austin, Fred Yunt, and Donald Wuerdeman To provide data on the relative merits of air and of oxygen in the activated sludge process, two 1,900- mVday (0.5-mgd) activated sludge pilot plants, one air and one oxygen system, were operated side-by-side at the Joint Water Pollution Control Plant, Carson, California. Although both pilot plants met the applicable discharge limitations for everything but three trace metals, the oxygen system provided a more stable opera- tion. Primary differences in performance concerned ammonia nitrogen removals. Calculated differences in energy consumption indicate a savings might be expected with the oxygen system. Differences in sludge production were not significant. This Project Summary was devel- oped by EPA's Municipal Environ- mental Research Laboratory. Cincin- nati, OH, to announce key findings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction Since the introduction of high-purity, oxygen-activated sludge, a controversy has existed concerning the relative merits of air and of oxygen in the activated sludge process. Very little data, however, were available on side- by-side operation of relatively large- scale systems with comparable engi- neering. As part of the research effort involved with federally-mandated secondary treatment at the Joint Water Pollution Control Plant (JWPCP) in Carson, California, the County Sanitation Dis- tricts of Los Angeles County constructed two 1,900-m3/day (0.5-mgd) activated sludge demonstration plants. One incorporated the UNOX high-purity oxygen process, and one used an air- sparged mechanical aerator. The primary purpose of the study was to obtain data pertinent to the selection and design of an activated sludge system at the JWPCP, but the nature of the research facilities allowed a direct comparison of the two activated sludge processes. The pilot plants were operated on identical feed. Equal engineering care was taken in the design of the aeration system, and identical clarifiers were used. The research motivations in establishing the operating parameters for the two plants were different: the oxygen system was operated to refine specified design parameters, whereas the air system was operated to determine its capabili- ties and limitations. The JWPCP is a 15-mVsec (350- mgd) primary treatment plant treating a mixture of domestic and industrial wastes. This facility allowed a good' comparison of the two activated sludge alternatives for treating relatively concentrated municipal wastewater. Selection and Description of the Pilot Plants Air-Sparged Turbine System Locating the Districts' JWPCP in an urban area placed a definite land constraint on the proposed secondary ------- treatment system for that plant. When preliminary site layouts were made for a conventional activated sludge system with the standard 4.6-m (15-ft) deep aeratiorrtanks and an optimistic 6-hr aeration period, no excess land was available for waste activated sludge processing. Because of this land con- straint, the Sanitation Districts proceeded to evaluate activated sludge systems that could reduce the land area required for secondary treatment. One of those alternatives was the deep tank sub- merged turbine (DTST) system. The DTST system was selected not only because of the land savings from the deeper tank (7.6 m or 26 ft) but also because the submerged turbine is a more efficient oxygen transfer device than the conventional coarse bubble air diffusers. The land savings from the deeper tank and the possibility of reducing the aeration period made the DTST system a realistic candidate system for secondary treatment at the JWPCP. High-Purity Oxygen System One of the major advantages offered by the pure oxygen biological treatment process is the ability to reduce the period of time required for treatment of wastewater by increasing the rate at which oxygen can be dissolved into the mixed liquor within the biological reactor. The results of preliminary studies using Union Carbide's 0.6- L/sec (10-gpm) mobile pilot plant verified this claim since acceptable effluent quality was achieved at aeration periods as short as 1.5 hr (V/Q). As a result of competitive bidding. Union Carbide Corporation* constructed the pure oxygen biological reactor, which was to utilize the existing pilot plant influent pumping station and final clarifier system. The reactor was designed to incorporate a submerged turbine/gas recirculation compressor arrangement for oxygen dissolution in each reactor stage. Table 1 compares the design criteria for the air-sparged system and the high- purity oxygen system as well as the associated final clarifiers. Tables 2 and 3 summarize the operational parameters for the air and oxygen systems, respec- tively. Table 1. Design Criteria for Pilot Plants Item Air System Oxygen System Biological Reactors: Average flow, m3/day (mgd) 1900 (0.5) 1900 (0.5) Length, m (ft) 6.1 (20) 7.3 (24) Width, m (ft) 6.1 (20) 7.3 (24) A verage water depth, m (ft) 7.6 (25) 3.7 (12) No. of stages 1 4 Detention time (V/Q), hr 3.5 2.5 Oxygen Storage Tank: Number — 1 Volume, m3 (ft3) NTP — 3900 (350.000) Capacity, m3/hr (ft3/hr) — 740 (4940) Standard Large Final Clarifiers: Number 2 1 Length, m (ft) 22 (72) 34 (111) Width, m (ft) 3.0 (10) 3.0 (10) Average water depth, m (ft) 3.0(10) 3.0(10) Overflow rate. m3/m2/day (gpd/.f?) 28.5 (700) 18.3 (450) Detention time (Q x 1/3 return), hr 2.0 3.0 Weir loading rate. m3/m/day (gpd/ft) 62.1 (5000) 62.1 (5000) Flowthrough velocity (Q x 1/3 return), mm/sec (ft/min) 3.2 (0.6) 3.2 (0.6) 'Mention of trade names or commercial products does not constitute endorsement or recommenda- tion for use by the U.S. Environmental Protection Agency. Discussion of Results Effluent Quality Activated sludge systems consist of two component units—the reactor and the final clarifier. The quality of the final effluent is related to the interaction of the component parts, and poor effluent quality may be caused by an inadequacy of only one part. The effluent quality of the air and oxygen systems is described in Tables 4 and 5. Soluble BOD A primary indicator of the adequacy of the reactor in terms of oxygen transfer and treating the wastewater is the removal of soluble organics. In all phases, for both pilot plants, the soluble BOD5 concentrations were 6 mg/L or less. These BOD measurements are low enough that 'differences between the two systems are not considered signifi- cant. Suspended Solids Secondary effluent solids concentra- tions depend on the effectiveness of the final clarifier. High effluent suspended solids, however, may be an indication of poor clarifier design, poor aerator design, or poor plant operation. During startup, both 1,900-m /day <0.5-mgd) pilot plants experienced periods of high effluent suspended solids and turbidity, which were alleviated by reducing the power input to the final stages of the reactors. Sludge Production One of the most important claims made on behalf of pure oxygen is thai the net growth of solids in these systems will be less than a similar air system when operated at the same mean cell residence time (MCRT). Since a large portion of the cost of wastewater treatment is usually associated with solids processing and sludge handling, this claim would represent a significant savings in both capital and operating costs. The claim is based on a comparison between the two systems that shows the net sludge production of air systems to be greater for any given organic loading rate than a similarly operated oxygen system. From an analysis of the data collected both from this and an earlier, smaller- scale study, the Districts have concluded there is little difference between the an ------- Table. 2. Summary of Operational Parameters—Air-Sparged Turbine System Parameter . Phase DATES: Start End Duration, days Flow Pattern REACTOR: Influent Flow, m3/day (mgd) Recycle. % Hydraulic Detention Time: V/Q, hr V/(QxR). hr MLSS, mg/L Volatility. % Mean Cell Residence Time: Reactor So/ids, days Total System Solids, days Organic Loading Rate: BODR/MLVSS. kg/ kg/ day BODn/TPVSS. kg/kg/day CODR/MLVSS, kg/kg/day CODn/TPVSS, kg/kg/day flODA. kg/ m3/ day fib/ ft3 /day) Sludge Production: VSS/BOD*. kg/ kg VSS/CODn, kg/ kg CLARIFIER: Overflow Rate, m3/m2/day (gpd/ff) Dentention Time: V/Q, hr V/(Q+R), hr Solids Loading Rate, kg/ m3/ day (Ib/ft3/day) Return Sludge Concentration, % SVI. ml/g 1 2/9/75 3/1/75 21 Steady 1200 (0.32) 90 5.6 2.9 3100 72 5.1 6.8 0.34 0.26 0.80 0.60 0.75 (12.0) 0.51 0.22 18.3 1450) 4.0 2.1 107 (1714) 0.7 252 II 3/9/75 3/29/75 21 Steady 1700 (0.45) 65 4.0 2.4 3400 73 3.7 5.4 0.38 0.27 1.07 0.74 1.00 (16.0) 0.64 0.27 21.3 (523) 2.8 1.7 117 (1874) 0.9 183 III 4/6/75 5/3/75 29 Steady 1700 (0.45) 45 4.0 2.8 2600 74 2.2 3.3 0.49 0.33 1.30 0.87 1.03 (16.5) 0.79 0.34 16.9 (415) 4.3 3.0 63 (1009) 0.9 163 IV 5/11/75 6/21/75 42 Steady 1900 (0.50) 40 3.5 2.5 4000 73 3.7 5.5 0.30 0.23 0.90 0.60 1.15 (18.4) 0.73 0.30 18.3 J450) 4.1 2.9 103 (1650) 0.9 165 V 7/20/75 8/30/75 42 Steady 1700 (0.45) 44 4.0 2.8 2300 73 1.8 2.8 0.70 0.47 1.61 1.06 1.34 121.5) 0.70 0.35 16.1 (395) 4.5 3.1 54 (865) 0.9 227 VI 9/28/75 10/25/75 28 Steady 1500 (0.40) 29 4.5 3.5 3300 70 3.0 4.3 . 0.49 0.33 1.16 0.82 1.24 (19.9) 0.56 0.26 14.7 (361) 5.0 3.9 63 (1009) 1.2 200 VII 10/26/75 11/20/75 26 Steady 1500 (0.40) 38 4.5 3.3 3300 71 3.2 4.3 0.44 0.30 1.00 0.75 1.12 (17.9) 0.63 0.31 14.7 1361) 5.0 3.6 68 (1089) 1.1 160 VIII 11/27/75 12/25/75 29 Steady 1500 (0.40) 50 4.5 3.0 3600 70 3.4 4.5 0.44 0.30 1.00 0.75 1.20 (19.2) 0.63 0.30 14.7 (361) 5.0 3.3 83 (1329) 1.1 173 IX 3/4/76 3/25/76 22 Steady 1300 (0.34) 47 5.3 3.6 2900 70 3.6 5.9 0.45 0.29 1.10 0.68 0.97 (15.5) 0.60 0.27 19.4 (476) 3.6 2.5 83 (1329) 0.9 146 and oxygen systems in terms of sludge production. When an analysis of the system is made based on the mass of micrporganisms contained within the biological reactor (which is the method used by proponents of pure oxygen), the data do indeed indicate that the oxygen systems produces less sludge. The authors believe, however, that the mass of solids within the entire biological system must be considered to obtain a true indication of the level of sludge production. This means that the solids present in the final clarifiers must be included when the total system solids are calculated. When the data are reexamined in this way, the oxygen system will no longer demonstrate an advantage over air systems in terms of sludge production. This reversal is because a greater portion of the total system solids will be contained within le clarifiers of an oxygen system than is typically encountered in air-activated sludge systems. Improved sludge settling and oxygen transfer capability allows the oxygen system to be operated as a high-rate system. As a result, as much as 50% of the total system solids will be carried in the final clarifiers. If the comparison of air and oxygen systems is based on reactor solids only, then a significant portion of the oxygen solid will be eliminated from the analysis, thus falsely indicating a higher organic loading rate than that imposed on the air system. Sludge Settleability Two parameters are commonly used to indicate sludge settleability. The sludge volume index (SVI) is the inverse of the settled sludge concentration expressed in ml/g, and the initial settling rate (ISR) is the maximum rate at which the sludge interface drops during the test. ISR data are reported from one series of tests conducted during a period when the performance of both pilot plants was characterized as "good." In this series of tests, the oxygen sludge settled about three times as fast as the air sludge. Although these results are the product of limited testing, they are in qualitative agreement with the general operating experience of the JWPCP pilot plants. The oxygen sludge definitely settled and gravity thickened better than the air sludge during this project. At this time, however, it's impossible to determine the extent to which this is an innate property of oxygen-activated sludge or a function of the reactor design. One factor that affected the sludge settleability in both of these systems was power input. To produce an acceptable eff I uent during the startup of ------- Table 3. Summary of Operational Parameters—Oxygen System Parameter Phase I II III IV VI VII VIII IX XI DATES' Start End Duration, days Flow Pattern REACTOR: Influent Flow, rrf/day fmgd) Recycle. % Hydraulic Detention Time- V/Q, hr V/(Q+R), hr MLSS. mg/L Volatility. % Mean Cell Residence Time: Reactor Solids, days Total System Solids, days Organic Loading Rate: BODH/MLVSS. kg/kg/day BODn/TPVSS. kg/kg/day CODR/MLVSS. kg/kg/day CODR/TPVSS, kg/kg/day BODA, kg/m3/day (Ib/ft3/day) Oxygen Utilization. 02/BODK, kg/kg Oi/CODn, kg/kg Sludge Production: VSS/BODR, kg/kg VSS/COD*. kg/kg CLARIFIER: 9/22/75 10/27/75 12/1/75 2/1/76 2/18/76 3/31/76 6/21/76 9/30/76 10/28/76 11/9/75 12/10/71 9/25/75 11/10/75 12/30/75 2/17/76 2/29/76 5/20/76 9/14/76 10/13/76 11/7/76 11/24/76 12/23/71 4 15 30 17 12 51 85 14 11 16 14 Diurnal Steady Steady Steady Steady Steady Steady Steady Dirunal Dirunal Diurnal 1900 (0.511 40 2.5 1.8 3800 75 1.8 3.4 0.70 0.31 1.67 0.89 2.15 134.4) 1.36 0.71 0.97 0.48 1500 (0.40) 40 31 2.2 2800 73 2.5 5.9 074 0.31 1.52 0.64 1.73 127.7) 0.60 0.29 1400 fO 37) 44 3.4 2.3 4200 74 3.4 6.8 1700 (0.45) 44 2.8 1.9 4600 72 1.9 5.6 1900 (0.51) 42 2.5 1.6 3300 75 1.7 34 1900 (0.51) 40 2.5 1.8 3900 74 1.9 4.4 1800 (0.48) 38 2.6 1.9 4100 77 2.7 4.8 1900 (0.51) 40 2.5 1.8 4420 75 2.1 3.8 0.52 0.26 1.14 0.56 1.62 (25.9) 0.60 0.20 1.31 045 2.03 (32.5) 0.83 042 1.61 0.81 2.05 132.8) 0.69 0.29 1.54 0.64 2.00 (32.0) 0.57 ' 0.33 1.15 0.66 1.76 (282) 0.48 0.27 0.95 0.54 1.63 (26.1) 0.64 0.28 0.63 0.29 0.78 0.40 0.80 0.36 0.69 0.33 1.52 0.81 0.84 0.42 1900 (0.51) 39 2.5 1.8 3700 70 2.0 4.2 0.67 0.32 1.46 0.69 1.94 (31.0) 1.24 0.69 0.98 0.38 1600 (0.43) 47 3.1 2.1 3990 70 3.0 6.6 0.55 0.24 1.07 0.47 1.54 (24.6) 1.48 0.71 0.74 0.38 1600 (0.43) 39 3.0 2.2 3840 77 2.8 5.4 0.51 0.27 1.05 0.55 1.44 (23.0) 1.49 070 0.66 0.37 Overflow Rate, m3/ rrf/day (gpd/ff) Detention Time: V/Q, hr V/IQ+R). hr Weir Loading Rate, m3/m/day (ff/ft/day) Solids Loading Rate, kg/ m3/ day (Ib/ff/day) Return Sludge Concentration, % SVI, ml/g 18.7 (459) 3.7 2.8 79.1 (852) 98 (1568) 1.05 78 23.2 (570) 3.0 2.2 62.6 (674) 90 (1440) 1.06 153 21.2 (521) 3.3 2.3 52.2 (562) 127 (2032) 1.40 99 25.4 (625) . 2.8 1.9 68.9 (741) 168 (2688) 1 54 65 28.4 (698) , 2.4 1.7 77.0 (829) 134 (2144) 1.18 83 27.9 (686) 2.5 1.8 101.2 (1089) 152 (2432) 1.36 77 27.5 (676) 2.5 1.8 99.4 (1070) 141 (2256) 1.22 83 18.1 (445) 3.8 2.7 101.5 (1092) 113 (1808) 1.34 113 28.4 (698) 25 1.8 102.3 (1101) 147 (2352) 0.88 124 23.3 (573) 2.9 2.0 84.2 (906) 141 (2256) 0.99 114 23.2 (570) 2.9 2.1 85.8 (923) 126 (2016) 094 101 each pilot plant, the mixer power had to be reduced. Excessive power input shears the floe, which can cause poor settleability of the sludge and a turbid effluent. Power Consumption In the present economic climate, energy consumption is one of the most important factors involved in comparing the air and oxygen activated sludge processes. Since power intensity prob- lems in both pilot plants required the aeration equipment to be operated at speeds lower than design, a comparison based on the pilot plant data is inappro- priate. Additionally, because the effects of scale would be difficult to predict, estimates based on typical aerator efficiencies produce more applicable results. The results of power consumption estimates made usi ng the above ground rules indicate that the oxygen systems use substantially less energy. The surface aerator oxygen system, in fact, is estimated to require only 52% of the energy used by the air system, and the submerged turbine oxygen system, 62%. Because of land constraints at the JWPCP, aeration tank depths greater than 5 m (15 ft) would be required with an air system, so surface air aeration was not evaluated. Conclusions Both air- and oxygen-activated sludge systems can produce effluents meeting the JWPCP discharge limitations for everything but certain trace metals, which require source control. The oxygen system is somewhat more stable and flexible in its operation. The two systems obtained good removals of soluble organics, and factors affecting solids separation in the final clarifier are most significant ii terms of their effects on effluent quality The most notable detrimental factor encountered in the study were excessiv input of aerator power, which sheare the floes in both systems, and nitrifica tion-denitrification, which caused trv settled sludge from the air system t resuspend. The major difference between the tw systems in terms of pollutant removal concerns ammonia nitrogen. The oxyge system did not nitrify. At the JWPCF where the ammonia discharge limitatio is high enough to impose no constrain' this characteristic is an advantage i that it eliminates rising sludge resultin from nitrification-denitrification. Claims have been made that oxygen activated sludge processes produce les sludge than air-activated sludg processes. In this study, the total plar solids were compared and the different^ ------- Table 4. Summary of Effluent Quality— Air Stream Parameters Aeration Period (V/Qj. hr MCRT (Total System), days Flow Pattern Suspended Solids: Influent, mg/L Effluent, mg/L Removal, % Total BODS: Influent, mg/L Effluent. mg/L Removal. % Soluble BODS: Influent, mg/L Effluent. mg/L Removal, % Total COD: Influent, mg/L Effluent, mg/L Removal, % Soluble COD: Influent, mg/L Effluent, mg/L Removal. % Grease (By Hexane Extraction}: Influent, mg/L Effluent, mg/L Removal, % Ammonia Nitrogen: Influent, mg/L Effluent, mg/L Removal, % 1 5.6 6.8 Steady 167 89 47 178 75 92 118 2 98 458 118 74 262 49 81 51 8 84 35 14 6O II 4.0 5.4 Steady 179 80 55 167 17 90 102 3 97 447 152 66 247 56 77 41 6 85 32 20 38 III 4.0 3.3 Steady 167 67 60 172 15 91 98 3 97 453 130 71 234 59 75 37 5 86 35 28 20 IV 3.5 5.6 Steady 170 22 87 171 8 95 101 4 96 460 77 83 241 56 7*7 38 1 97 35 32 9 Phase V 4.0 2.8 Steady 204 110 46 224 16 93 126 5 96 513 191 63 265 72 73 — — — 31 28 10 VI 4.5 4.3 Steady 204 36 82 234 12 95 132 4 97 556 91 84 257 57 78 — — — 36 32 11 VII 4.5 4.3 Steady 165 37 78 212 12 94 129 2 98 483 92 81 270 55 80 — — — 33 28 15 VIII 4.5 4.5 Steady 216 54 75 226 13 94 109 2 98 515 111 78 256 48 81 — — — 34 32 6 IX 5.3 5.9 Steady 177 29 84 211 18 92 119 2 98 517 84 84 282 54 81 — — — 38 31 18 was found to be insignificant at the 90% confidence level. The trend, however, was for the oxygen system to produce more sludge. Because of modifications made to the pilot plant's aeration equipment to prevent floe shear, an energy consump- tion comparison was considered inap- propriate. A paper study indicates that substantial energy savings may be expected with the oxygen system. The full report was submitted in partial fulfillment of Contract No. 14- 12-150 by Los Angeles County Sanita- tion Districts under the sponsorship of the U.S. Environmental Protection Agency. ------- Table 5. Summary of Effluent Quality—Oxygen System Parameters Aeration Period IV/Q). hr MCRT (Total System), days Flow Pattern Suspended Solids: Influent, mg/L Effluent, mg/L Removal, % Total BOD: Influent, mg/L Effluent, mg/L Removal, % Soluble S005: Influent, mg/L Effluent, mg/L Removal, % Total COD: Influent, mg/L Effluent, mg/L Removal, % Soluble COD: Influent, mg/L Effluent, mg/L Removal, % Grease (By Hexane Extraction): Influent, mg/L Effluent, mg/L Removal, % Ammonia Nitrogen: Influent, mg/L Effluent, mg/L Removal, % 1 2.5 3.4 • Diurnal 189 17 91 219 11 95 131 4 97 467 81 83 249 62 75 43 1 98 32 26 19 II 3 1 5.9 Steady 165 18 89 221 7 97 132 3 98 523 87 83 213 68 68 38 1 97 34 31 9 III 3.4 6.8 Steady 242 28 88 231 12 95 105 3 97 554 94 83 258 58 78 47 3 94 33 31 6 IV 2.8 5.6 Steady 201 54 73 238 20 92 122 5 96 561 122 78 279 59 79 56 4 93 32 31 3 V 2.5 34 Steady 172 28 84 219 21 90 121 6 95 486 100 79 283 67 76 42 3 93 36 31 14 Phase VI 2.5 4.4 Steady 202 21 90 212 12 94 115 3 97 536 88 84 279 66 76 62 2 97 37 32 14 VII 26 4.8 Steady 142 17 88 187 8 96 93 2 98 438 82 81 255 64 75 64 2 97 32 30 6 VIII 2.5 38 Steady 140 14 90 176 5 97 90 1 99 400 71 82 260 58 78 46 1 98 34 29 15 IX 2.5 4.2 Diurnal 150 48 68 204 13 94 134 1 99 415 116 72 272 64 77 46 6 87 28 28 0 X 3.1 6.6 Diurnal 130 34 74 173 12 93 100 2 98 431 97 78 280 63 78 39 3 92 34 29 15 XI 3.0 5.4 Diurnal 120 20 83 185 6 97 124 2 98 446 83 81 305 65 79 41 2 95 37 34 8 Scott Austin and Fred Yunt are with, and Donald Wuerdeman was with, Los Angeles County Sanitation Districts, Whittier, CA 90607. Irwin J. Huge/man was the EPA Project Officer (see below). The complete report, entitled "Parallel Evaluation of Air- and Oxygen-Activated Sludge," fOrder No. PB 81-246 712; Cost: $8.00, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 Richard C. Brenner, the present contact, can be reached at: Municipal Environmental Research Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 ft US GOVERNMENT PRINTING OFFICE, 1981 — 757-012/7355 ------- United States Center for Environmental Research pees pajd Environmental Protection v Information Environmental Agency Cincinnati OH 45268 Protection Agency EPA 335 Official Business Penalty for Private Use $300 RETURN POSTAGE GUARANTEED Third-Class Bulk Rate LOU W TILLEY REGION V EPA LIBRARIAN 230 S DEARBORN ST CHICAGO IL 60604 ------- |