WATER POLLUTION CONTROL RESEARCH SERIES 17010DXD08/70 PHOSPHOROUS REMOVAL BY AN ACTIVATED SLUDGE PLANT U.S. ENVIRONMENTAL PROTECTION AGENCY ------- WATER POLLUTION CONTROL RESEARCH SERIES The Water Pollution Control Research Series describes the results and progress in the control and abatement of pollution in our Nation's waters. They provide a central source of information on the research, development and demonstration activities in the Environmental Protection Agency, through inhouse research and grants and contracts with Federal, State, and local agencies, research institutions, and industrial organizations. Inquiries pertaining to Water Pollution Control Research Reports should be directed to the Chief, Publications Branch (Water), Research Information Division, R&M, Environmental Protection Agency, Washington, B.C. 20460. ------- PHOSPHORUS REMOVAL BY AN ACTIVATED SLUDGE PLANT by Sewerage Commission of the City of Milwaukee Milwaukee, Wisconsin 53201 for the ENVIRONMENTAL PROTECTION AGENCY Program #17010 DXD Grant #WPD 188-01-67 188-02-68 188-03-69 August, 1970 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, B.C. 20402 - Price $1.00 ------- EPA Review Notice This report has been reviewed by the EPA, and approved for publication. Approval does not signify that the con- tents necessarily reflect the views and policies of the Environmental Protection Agency, nor does mention of trade names or comnercial products constitute endorsement or recommendation for use. ii ------- ABSTRACT The Milwaukee plants removed an average of Q0% of the influent total phosphorus (TP). Milwaukee offered an opportunity for plant scale demonstration and study of the activated sludge process parameters effects on TP removal. The high TP removals were due to sufficient solids production from the amounts of TP, BOD and suspended solids in Milwaukee sewage. Phosphorus balances demonstrated an average of 95.7$ of the phospho- rus removed was recovered in the waste sludge withdrawn as Milorganite. An equation was developed for predicting % TP removal from East Plant (EP) 1968 data. Applied to 1969 and 1970 data when TP removals exceeded 80#, the significant parameters were food to microorganisms (F/M), MLSS, and TP to microorganisms (TP/M). The significance of ML-DO, air application rate and detention time could not be shown. Plant scale studies revealed that F/M, MLSS and TP/M mainly affected soluble orthophosphate (SOP) removal. The removal of SOP was associated with MLSS biological activity by oxygen uptake rate measurements in tank studies. A ML-DO^ 2.0 mg/L past the tank turn- point was effective for SOP removal. Insolubilization of SOP by sewage soluble cations appeared to be insignificant in the process. It appears that brewery waste water aids soluble phosphorus removal at the Milwaukee plants. The cyclic removal of TP in an activated sludge plant was shown by the analyses of hourly influent and effluent samples for BOD, TP and SOP during a year's study. These cycles corresponded to the hourly changes in sewage composition and flows and clarifier sludge blanket build-up. At times TP removal efficiency was greatly reduced by the loss of solids from overloaded clarifiers at peak flows. Continuous addition of ferrous sulfate waste pickle liquor to EP-ML at a 15 mg/L-Fe rate provided hourly effluent soluble phosphorus residuals of 0.05 mg/L-P. X-ray diffraction studies of freeze dried sludges showed iron present as vivianite. This report was submitted in fulfillment of Grant No. WPD 188-01-67, 188-02-68, and 188-03-69, Program No. 17010 DXD, between the Federal Water Quality Administration (now Environmental Protection Agency) and the Sewerage Commission of the City of Milwaukee, Wisconsin. Key Words: Phosphorus removal, activated sludge process, process parameters, wastewater treatment, biological treatment. 111 ------- CONTENTS Section Abstract iii v Contents List of Figures vii List of Tables lx List of Appendices xl I Conclusions 3 II Recommendations III Introduction 5 IV Literature Survey V Procedures A. Milwaukee Waste Water Treatment Facilities 11 at Jones Island B. Sampling and Analytical Techniques 13 VI Research Phases, Results & Discussion A. Correlation of Activated Sludge Process 15 Parameters to Phosphorus Removal on a Plant Scale B. Demonstration of Cyclic Removal of Phosphorus 19 from Sewage by an Activated Sludge Plant C. Plant Study of the Effect of Clarifier Blanket 23 Depths on Clarifier Effluent Residual SOP D. Plant Loading Study (F/M and TP/M) 26 E. Effect of Brewery Waste Load 30 F. Plant Scale Phosphorus Balances 36 ------- CONTENTS Section G. Relationship Between the Removal of Total BOD 38 and Removal of TSP in an Activated Sludge Plant (BODR/TSP/R) H. Study of Soluble Phosphorus Uptake in East Plant 41 Aeration Tanks (General) I. Effect of MLSS-02 Uptake Rates and Air Appli- 42 cation Rates on Soluble Phosphorus Uptake in an Aeration Tank J. Evaluation of MLSS Activity by Glucose Dehydro- 52 genase Assay K. X-Ray Diffraction Studies of Sewage Suspended 56 Solids and Waste Sludge Solids L. Cationic Removal of Phosphorus from Sewage 57 M. Effect of Iron Addition to an Aeration Tank on 60 Soluble Phosphorus Removal N. Plant Scale Iron Addition Study 66 VII Acknowledgements 77 VIII References 79 IX Nomenclature and Glossary 83 X Appendices 85 vi ------- FIGURES PAGE 1. JONES ISLAND SEWAGE TREATMENT PLANT 12 2. SEWAGE AND EFFLUENT AVERAGE HOURLY COMPOSITION 20 13/3/67 to 6/15/68 3. SOP RELEASE AND DIFFUSION FROM SLUDGE BLANKET 24 VERSUS TIME U. EFFECT OF PLANT LOADINGS ON PLANT EFFLUENT 28 SOP LEVELS 5. EFFECT OF BREWERIES SHUTDOWN ON RS COMPOSITION 33 6. FREQUENCY DISTRIBUTION BODR/TSPR WEEKDAY DATA 39 7. FREQUENCY DISTRIBUTION BODR/TSPR WEEKEND DATA 40 8. AERATION TANKS DO PROFILES (9-30-1969, 8:00 A.M.) 43 9. AERATION TANKS DO PROFILES (10-6-1969, 12:30 P.M.) 45 10. EFFECT OF AIR APPLICATION RATE ON ML-DO LEVEL 47 11. EFFECT OF AIR APPLICATION RATE ON MLSS-02 UPTAKE RATE 48 12. EFFECT OF AIR APPLICATION RATE ON ML-TSP LEVEL 49 13. EFFECT OF AIR APPLICATION RATE ON ML-SOLUBLE BOD LEVEL 50 lU. EFFECT OF IRON ADDITION ON EAST PLANT RESIDUAL SOP 68 15. COMPARISON OF EAST AND WEST PLANT SDI'S 69 16. ACCUMULATION OF IRON IN EP-RS 71 17. RETURN SLUDGE ASH CONTENT (DRY BASIS) 72 18. RETURN SLUDGE NITROGEN CONTENT (DRY BASIS) 73 19. TECHNICON AUTOANALYZER SCHEMATIC 87 VII ------- TABUS Table pafte 1. Effect of Clarifier Sludge Blanket Depths on 22 East Plant Effluent Quality 2. Comparison of SOP Concentration (mg/L-P) of Aeration 25 Tank No. 10 Outlet ML, Clarifier No. 5 ML - Feed, Clarifier No. 5 Effluent in Relation to Clarifier Blanket Depth 3. Averages of Daily Data by Periods 27 U. Summary Period Averages 31 5. Comparison of Period Averages 32 6. Relation of Phosphorus Removed to Solids Synthesis 35 7. Calculated Weekly Phosphorus Recoveries 37 8. Average BODR/TSPR Ratios 38 9. Air Application Rates 42 10. Effect of Air Application Rates 44 11. Daily Process Parameter Data 46 12. MLSS-02 Uptake Rate After Six Hours Aeration 51 13. Dehydrogenase Assay Variability 53 lU. Dehydrogenase Assay Data 53 15. Variability of ML - Dehydrogenase Activity and MLVSS 54 at Different Sample Locations 16. Comparison of Coefficients of Variation (C.V.) 54 17. Average Cation Concentrations (mg/L) for July, 19&9 58 18. Effect of Iron Dosage on ML - TSP Level 61 ------- TABLES Table 19. Run 6 Observation Data 62 20. Reduction of Soluble Iron Concentration in an 63 Aeration Tank 21. Average Iron Recoveries 64- 22. Soluble Iron and TSP Release by MLSS 65 23. Effluent SOP (mg/L-P) 67 21*. Average Return Sludge Composition 75 25. Aerobic Digestion of Sludge 92 ------- APPENDICES Appendix Title A Phosphorus Determination with Technicon Autoanalyzer 85 B Procedure used for MLSS and RSSS Determination 88 C BOD Determination 89 D Discussion of Material Found Floating on the Surface 91 of the EP Aeration Tanks and the Aerobic Digestion of Waste Sludge E Average Daily Screened Sewage Characteristics and 94 Plant Operation Data F Total Phosphorus Removal at the Jones Island Plants 95 Based on Plant Flows xi ------- SECTION I CONCLUSIONS A three-year plant scale study on the phosphorus removal by an acti- vated sludge plant conducted at Jones Island Plant, Milwaukee, Wis- consin, resulted in the following observations: 1. The Milwaukee plants on the average removed 80# of the TP from sewage. The effluent TP concentrations coincided with the hourly changes in sewage composition and flow, and at times with the hourly build-up of clarifier sludge blankets. The effluent TP concentrations increased as the sewage flow, BOD and TP concentrations increased. Low effluent TP concentrations were usually observed with low sewage flow, BOD and TP concentrations. 2. The relationship of total BOD removal to total soluble phosphorus removal was found to vary according to the day of the week; on the average, the ratios were 88:1 for weekdays, 62:1 for Saturdays, and Vf:l for Sundays. 3. Weekly plant scale phosphorus balances during a 37 week study showed that an average of 95.1% of the sewage phosphorus removed by both plants could be accounted for in the Milorganite produced. This indicated that the removal of waste sludge (as Milorganite) was the vehicle for withdrawal of phosphorus from the waste treat- ment system. 1*. Overloaded clarifiers at peak flows resulted in lower TP removals due to spewing of solids over the weirs. However, the detention of solids for 2 to 3 hours at different sludge blanket depths had very little influence on the clarifier effluent SOP concentration. 5. A substantial release of SOP was observed when the RS was mixed with the screened sewage. The concentration of SOP in the resulting ML was 2 to 3 times greater than that expected from the calculated amount in a corresponding mixture. 6. The removal of TSP by MLSS appears to be related to MLSS metabolic activity as was shown by MLSS-02 uptake rate measurements along the aeration tank. 7. A loading study with the two plants showed that an average F/M and TP/M loadings of 0.291 and .011, respectively, resulted in slightly higher average SOP removals compared to the higher average loadings of 0.511* and 0.019» respectively. ------- 8. There was a marked drop in the soluble phosphorus removal during the shutdown of the Milwaukee breweries in 1969, compared with the high soluble phosphorus removals before and after the shutdown. The rapid recovery of soluble phosphorus removal by both plants indicates the brewery waste water aids soluble phosphorus removal. 9. Aeration tank studies in the East Plant indicated that an air application rate of 0.86 cu. ft. /gal. of influent or larger and a ML-DO of 2.0 ag/L beyond the tank turnpoint was adequate and effective in reducing soluble BOD to 10 ag/L and TSP to 0.5 mg/L-P at MLSS concentrations of 2530 to 2910 mg/L. A ML-DO of 1.0 ag/L beyond the tank turnpoint did not result in as much SOP removal as was observed with a ML-DO of 2.0 mg/L. 10. The quantity of insoluble phosphorus (58£ for 1969) in Milwaukee sewage is unusually high compared to normally reported values for municipal waste waters. No significant differences were found in influent and plant effluent soluble iron, aluminum, calcium and mag- nesium concentrations. This indicated that very little insolubli- zation of phosphorus by these cations occur in the activated sludge process. It appears that the cationic fixation of phosphorus probably occurs in the sewage prior to its reaching the plant. 11. Limited X-ray diffraction studies on freeze dried WS samples revealed that iome of the iron in the WS was in the form of a crystal- like iron orthophosphate, vivianite Pe3 12. Limited phosphorus removal studies with the addition of waste pickle liquor to the mixed liquor indicated that enhanced SOP removals can be achieved by iron addition. The waste pickle liquor appeared to have no effect on the ML biota. ------- SECTION II RECOMMENDATIONS The results of this three year study have shovn that the City of Milwaukee Severage Commission's East Plant can consistently average 80$ total phosphorus removal from sewage. A short term study of continuous iron addition, as ferrous sulfate waste pickle liquor, to the East Plant has demonstrated the feasibility of a long term plant study. The preliminary data indicated that it could be possible to consistently produce a plant effluent with a TSP residual of approximately 0.5 mg/L-P or less. It could also be possible to improve the plant total phosphorus removal efficiency from an average of &0% to an average of 90$. A one year study to determine the long term effects of con- tinuous iron addition on the 115 mgd East Plant and to compare its performance to the 85 mgd West Plant which will receive no additional iron is proposed. The effects of continuous iron addition upon mixed liquor flora, mixed liquor settleability, waste sludge conditioning requirements and plant physical facilities would be evaluated along with effluent phosphorus and iron concentrations. This would require the continuous feeding of ferrous iron (as waste pickle liquor) to the mixed liquor feed channel of the 115 »gd East Plant. Twenty four hour composite samples of East and West Plant effluent would be analyzed to determine the ability of ferrous iron to increase phosphorus removal. Microscopic examination of mixed liquor samples would be utilized to determine if iron affected the mixed liquor culture. Extension of the test period over one year would establish the feasibility of iron addition as a method for enhanced sewage phosphorus removal. If iron addition proves to be feasible it could provide an economic method of phosphorus removal for existing acti- vated sludge plants, and provide effluents with consistently low phosphorus residuals. The completed work has shown that excellent removal of phosphorus from sewage can be obtained in an activated sludge plant. However, due to the size of the operation (115 mgd) and limitations in the method of waste sludge removal (fertilizer production) it was extremely difficult to control and/or change the process parameters on a day to day basis. It is therefore suggested that consideration b« given to modifying the East Plant. The modifications would be the isolation of two aeration tanks and one clarifier and a separated mixed liquor feed. This would provide a 12 mgd plant operated on its own return sludge, in which all process parameters could be controlled and varied as required to establish the mechanisms of phosphorus fixation in the activated sludge process. ------- SECTION III INTRODUCTION It has long teen recognized that phosphates are one of the major factors contributing to the progressive fertilization of streams and lakes which result in frequent and highly objectionable algae blooms. Although over 90% of the biologically oxidizable organic matter and suspended solids are removed in the conventional activated sludge process, the reduction of dissolved mineral nutrients is gener- ally very low. Most municipal plants employing the activated sludge treatment process, report phosphorus removals from sewage of 20 to 30%. There are a few exceptions, such as the Milwaukee (l), San Antonio (2), and Baltimore (3) plants where phosphorus removals as high as &Q% to 96% have been reported. No significant conclusions have been reached as to the reason for such wide variation in the effectiveness of the convention- al activated sludge process to remove total phosphorus from sewage. One group of investigators, Menar and Jenkins (U), hypothesized that based on a normal BOD of about 200 mg/L and total phosphorus of 10 mg/L-P in sewage, the biological removal of phosphorus in the form of waste activated sludge should be 20 to 30% and any additional re- moval observed is the result of the insolublization of phosphate by the soluble cations (mainly calcium and iron) present either in the hard carriage waters or coining from steel industries or both. The second group (Levin and Shapiro (5), and Vacker, et. al. (6)) hold the theory that the activated sludge microorganisms under optimum operating conditions can store phosphorus far in excess of their metabolic requirements (100 BOD: IP) and have called this exces« uptake "luxury uptake". The Milwaukee Jones Island plants have been consistently showing good total phosphorus removals (usually over 80%). The two plants (East and West) are operated in parallel and each plant treats a part of the total flow. In 1967, the Federal Water Pollution Con- trol Administration initiated a three year research and demonstration project with the Sewerage Commission of the City of Milwaukee. The objective of the study was to demonstrate and optimize the effects of the activated sludge process parameters on a plant scale for the removal of total phosphorus from sewage. ------- SECTION IV LITERATURE SURVEY Sewage phosphorus removals "by the activated sludge process have been reported by a number of investigators, Owens (7) 2 to U6$, Stone (8), 6U to ?1#, for three Chicago waste water treatment plants Hurwitz (9) reports removals of 76.7$, 36.6$, and 53.9$.Some of the short term plant scale investigations made by the Federal Water Pollution Administration as summarized by Witherow (3) have indicated total phosphorus removals of 51 to 15% in the Northside plant of Amarillo and the Village Creek plant in Fort Worth. The Rilling plant in San Antonio and the Back River plant in Baltimore have been report- ed to consistently remove an average of 80$ and 90$ of the total phosphorus respectively. Phosphorus has also been successfully removed from waste waters by using chemicals such as lime, alum, ferric chloride, ferric sulfate, ferrous sulfate, and sodium aluminate. These chemicals have been used to precipitate phosphorus in tertiary treatment by Owens (7), Lea, et. al. (10), Rohlich (ll), Malhotra, et. al. (12) and Nesbitt (13). Chemicals have also been used to insolublize phosphorus in the biological treatment units by their addition to the aeration units, Tenney and Stumm (lU), Barth and Ettinger (15), Eberhardt and Nesbitt (16), or by their addition prior to the primary sedimentation units, Rudolf (17), Neil (l8), Schmid and McKinney (19). A review of the current literature suggests that two schools of thought are being used to explain the wide variation in the phospho- rus removals observed at different plants in the United States. Sawyer (20), Sekikawa, et. al. (21), Hall and Engelbrecht (22), Menar and Jenkins (U), support the theory that the biologically incorporated phosphorus in the activated sludge solids is between 2 to 3% phosphorus on the volatile mass basis and any additional removal is cationic. Menar and Jenkins (H) indicate that biological phosphorus removal is not affected by the sludge growth rate or by standard process operating parameters and that the phosphorus removal is directly proportional to the net sludge growth. This means that the phosphorus removed biologi- cally in a conventional activated sludge sewage treatment plant would be in the range of approximately 100 parts of BOD removed to 1 part of soluble phosphorus removed. Based on an average BOD removal of 200 mg/L and influent total phosphorus content of 10 mg/L-P, about 20 to 30$ of the influent phosphorus would be removed biologically. The second theory was presented in papers by Levin and Shapiro (5), Borchardt and Azad (23), and Connell and Vacker (2); they indicated that activated sludge solids under certain conditions were capable of removing more phosphorus than they require for cell growth. ------- The "biological uptake of excess phosphorus was called "luxury uptake". They reported that luxury uptake vas enhanced by high ML-DO concen- trations, and the phosphorus incorporated by this mechanism was re- leased from the solids by anaercaic or acidic conditions. They report- ed that luxury uptake leads to sludges containing as high as 1% phos- phorus on the VSS basis. Most investigations, Sawyer (20), Hall and Engelbrecht (22), Levin and Shapiro (5), were laboratory experiments to study the acti- vated sludge process parameters effect on the removal of SOP from sewage. Connell and Vacker (2), Witherow (3), have reported the process parameter effects based on full scale plant studies which showed good phosphorus removals. Sawyer (20) found that an addition of glucose to increase the sewage BOD by 1*00 mg/L reduced the sewage phosphorus from 2.68 mg/L-P to 0.00 mg/L-P. Sekikawa, et. al. (21), Hall and Engel- brecht (22), and Levin and Shapiro (5), also observed higher phosphorus removals when the initial BOD concentration of the substrate was in- creased. Witherow (3) reported that the phosphorus removal in the aeration tank was affected by the BOD load applied to the tank and he observed good phosphorus removals with the influent BOD in the range of 118 to 202 mg/L. Menar and Jenkins (U) found that the phosphorus content of the activated sludge volatile solids did not vary signifi- cantly over a wide range of substrate removal rates and the weighted average per cent phosphorus on the VSS basis was 2.62%. The effect of high initial phosphorus content of the sub- strate on phosphorus removal has been reported by Witherow (3). He observed good phosphorus removals when the influent phosphorus was in the range of U.2 to 10.5 mg/L-P. Higher phosphorus levels in the in- fluent resulted in decreased % phosphorus removals in the activated sludge process. The effect of detention time on phosphorus removal has been reported by Srinath (2U), Alarcon (25), Hall and Engelbrecht (22), Levin and Shapiro (5), Srinathfs (2U) data showed that the phosphorus and BOD removal rates appear to coincide, approximately 10% removal after one hour and then gradually increasing to 90$ after five addition- al hours of aeration. Levin and Shapiro (5) observed over 7055 soluble phosphorus removals in an aeration time of 3 hours. Witherow (3) recommends a modal detention of > 2.5 hours in the aeration tank for good phosphorus removal. Sekikawa et. al. (21) and Alarcon (25) found that after all the BOD had been consumed, further aeration of ML caused the release of phosphorus from the sludge solids due to cell oxidation (i.e., endogenous respiration). There are divided opinions on the role of MLSS concent- ration in the removal of phosphorus in the aeration tanks. Hall and Englebrecht (22) and Sekikawa, et. al. (21) found MLSS concentrations had very little effect on the rate or the extent of soluble uptake. ------- Srinath, et. al. (2U) reported that maximum phosphorus uptake was produced "by 20 to 30 percent return sludge. Feng (26) observed low MLSS (500 mg/L) to effect the best phosphorus removals when the aer- ation rates were high (18 cu.ft./gallon of ML). Connell and Vacker (2) obtained maximum phosphorus removal at an average daily BOD loading rate of about 50 Ibs. of BOD/100 Ibs. of aeration solids. Witherow (3) observed good phosphorus removals with MLSS >1200 mg/L and loading rates ranging from 0.26 to 0.35 Ibs. of BOD/lb. of MLSS/day. Levin and Shapiro (5) found that ML pH had a pronounced influence on phosphorus removal, a pH range of 7 to 8 was most effec- tive while pH values >8 and <6 resulted in phosphorus release from the sludge solids. Rudolf (17) in 19^7 reported a release of soluble phospho- rus during anaerobic sludge digestion. In 196l,Alarcon (25) demon- strated that the soluble phosphorus content of ML decreased during aeration and increased when aeration was stopped. Campbell (27) reported that return sludges kept under anaerobic conditions at room temperature for four hours released phosphorus from the sludge solids into the liquid. Levin and Shapiro (5), Connell and Vacker (2), Han and Englebrecht (22), and Witherow (3) indicate that a ML-DO level of approximately 2 mg/L was necessary for the optimum uptake of soluble phosphorus. Hall and Engelbrecht (22) and Connell and Vacker (2) found that the release of phosphorus from the solids back into the liquid (such as in the final clarifier) was not signi- ficant if a high dissolved oxygen level had been sustained during the second half of the aeration period. ------- SECTION V PROCEDURES MILWAUKEE WASTE WATER TREATMENT FACILITIES AT JONES ISLAND The wastewater treatment system of the Sewerage Commission of the City of Milwaukee on Jones Island consists of the East and West Plant. The capacity of the activated sludge plants is 115 mgd, and 85 mgd respectively. The physical layout of these plants is shown in Figure 1. The Milwaukee plants serve a total drainage area of UlO square miles having a connected population of approximately 1,000,000. Combined sewers serve approximately 6.6% of the total area and the remainder of the area has a separate sewer system. The volume of wastewater from this highly industrial area consists of approximately 225? industrial and 78$ domestic wastewater (28). The daily average characteristics of screened sewage and some daily average operation data for 1967» 1968 and 19^9 are given in appendix E. The wastewater flows through mechanically cleaned "bar screens (l inch openings) and then flows through a battery of eight grit chambers (8 x 8 x 90 feet long each) at a velocity of approxi- mately one foot per second. Fine screenings are removed by passage of the degritted sewage through eight rotary drum screens having 3/32 inch slots. Approximately 60 wet tons of screenings and grit are removed and incinerated daily in a 5 stage multiple hearth furnace. The screened sewage flow is then divided between the two activated sludge plants (East and West). After screening the sewage flows into aeration tanks where air is supplied through plate diffusers arranged in a ridge and furrow type pattern. Design detention time is 6 hours. Final sedimentation in peripheral feed clarifiers is accomplished during a 2 hour design time for detention. The waste sludge is disposed of by processing it into Milorganite fertilizer. Further details of the Jones Island plants can be found in reference 28. 11 ------- HIGH • LOW iFvfl. V tCWIGt IWHONt? If HARBOR ENTRANCE KINNICKINNIC RIVER EWERAGE COMMISSION CITY OF MILWAUKEE GENFRAL PLAN JONES ISLAND SEWAGE TREATMENT PLANT FIGURE I ------- SAMPLING AND ANALYTICAL TECHNIQUES Sewage sampling - Initially, screened sewage samples were single grab samples which were used for the hourly study. Aliquots proportional to flow from these hourly samples were used to make the 2^-hour composite sample. On 6-13-68, a Phipps - Bird sewage sampler was put into operation to provide hourly sewage composites (30 - 200 ml. portions/ hour). Aliquots proportional to flow from these hourly composites were used to make the 2k hour composite. This sampling procedure markedly reduced the fluctuations in the sewage phosphorus and BOD concentrations exhibited by the hourly grab samples. Effluent sampling - Plant effluent samples were hourly grab samples. The 2k hour composites were prepared the same way as for sewage. Hourly grab samples of the plant effluent showed very little fluctuation in BOD and phosphorus concentrations from hour to hour compared to screened sewage hourly grab samples. Others - All other samples used in short term studies were seven liter grab samples. Analytical methods - Listed below are brief descriptions of the analytical methods and instrumentations used in this study. Some of the analytical methods used are described in the appendix and the instruments manufacturer operation manuals are listed in the references. Angel Reeves glass fiber filter pads (2.U cm., 93^AH) were used to provide the filtrates for the analyses of "soluble components". Determination of phosphorus - Technicon Autoanalyzer as described in Appendix A was used. The Autoanalyzer was used only for SOP analyses because the Technicon acid digestion system, auto- matic sampler for liquids with colloidal suspended solids and the automatic filtration system were found to be inadequate. Dissolved oxygen measurements - were made by a variety of instruments. The ML-DO was continuously monitored at the outlets of six EP aeration tanks by means of "galvanic cell" probes. Spot ML-DO measurements were made by a YSI Model 5^18 probe and Model 51 meter. The YSI Model 5^20 probe and Model 5^ meter were used to determine the DO levels for the BOD determination. The MLSS-02 uptake rates were determined with a YSI Model 53 Biological Oxygen Monitor. The procedure used was as described in the suppliers manual (29) using 1 to 2 and 1 to 5 dilutions on each ML sample. 13 ------- East Plant effluent turbidity - vas continuously monitored by a Hach CR surface turbidimeter Model 1889. Spot turbidity measure- ments were made with a Hach Laboratory Turbidimeter Model 2100. The turbidimeters were operated and calibrated as described in manufactur- er's "Instruction Manuals" (30) and (31). The MLSS and RSSS concentrations - were ascertained by a procedure in use in the Milwaukee Sewerage Commission's Laboratory (Appendix B). The total and total soluble cations concentrations (Fe, Al, Ca and Mg) - were determined on "ternary acid digestates" of the sample by means of an Atomic Absorption instrument (instrumentation Laboratory, Inc. Model No. 153). The instrument operation and cali- bration were as described in the manufacturer's "Procedure Manual" (32). Biochemical oxygen demand (BOD) - the "Azide Modification of the lodometric method" for the BOD analyses as given in Standard Methods 12th edition (33) was compared with a modified procedure using a YSI Model 5^20 probe and Model 5*+ meter to determine DO levels. The modified procedure is described in appendix (C). A comparative study of these methods showed good agreement. A study to determine a factor for converting U day and 6 day BOD's to 5 day BOD's (l.lU and 0.93) showed good agreement with those reported by B. L. Goodman & J. W. Foster (31*) of 1.13 & 0.91 respectively for sewage. 14 ------- SECTION VI RESEARCH PHASES, RESULTS AMD DISCUSSIONS CORRELATION OF ACTIVATED SLUDGE PROCESS PARAMETERS TO PHOSPHORUS REMOVAL ON A PLANT SCALE. In this study, all pertinent data from the Milwaukee Sewerage Commission East Plant located on Jones Island for the 1968 year and for the 1969 year thru September 30 was transferred to IBM computer cards. The month of January, 1970 was used for the comparison study of 1968 loge equation. The data from these cards was then used to establish the variables included in this analysis. It was initially assumed that the percent of total phosphorus removed by the activated sludge process is a linear function of a set of process parameters (note : effluent soluble orthophosphate was considered as a variable not as a process parameter). The predictive equation of phosphate removal can then be expressed as : Y = ai + a2x2 a-^ = intercept of the regression line a^ = regression coefficient for ith parameters , i = 2, 3, ... 8 where Y = % removal of total phosphate Xp = food to microorganism ratio (F/M) F = Ibs. of BOD/ day M = Ibs. of volatile suspended solids (MLVSS) x~ = detention time (hours) in aeration tank x^ = cu. ft. of air/ gallon of sewage Xc - mixed liquor suspended solids (mg/liter) (MLSS) x/r = dissolved oxygen in the effluent of aeration tank (mg/liter) X.T = total phosphate in the influent in Ibs/day (TP) microorganisms (M) (MLVSS) xg = soluble orthophosphate in the effluent (mg/liter) 15 ------- Another set of statistical evaluations vas made on the basis of the assumption that the distribution of parameters influenc- ing percent removal of total phosphate (x2, x^, ... xg) is exponential, such that the logarithm of the dependent variable (percent removal of total phosphate) varies linearly with the logarithm of the independent variables. The regression equation for this assumption can be ex- pressed as Y = aixp X3 ' * * X8 The data was analyzed in the following groups with several trials in each group. 1968 Data - Natural Form (362 cards) 1969 Data - Natural Form (273 cards) 1/1/69 - 9/30/69 1968 + 1969 Data - Natural Form (601 cards) 1969 excluding brewery strike period - Natural Form (217 cards) 1968 Data - Logarithmic Form (3U2 cards) 1968 Data - Logarithmic Form (231* cards) 1969 E&ta - Logarithmic Form (259 cards) 1968 + 1969 Data - Logarithmic Form (601 cards) 1969 excluding brewery strike period - Logarithmic Form (217 cards) During 1969, a brewery strike from June 9 to July 15 caused significant differences to occur at the sewage treatment plant Thus, the 1969 data was analyzed including and excluding the strike period to determine the effect of large volumes of brewery wastes on the Milwaukee treatment plant. Method of analysis - A multiple regression analysis was performed with the data from 1968 and 1969 in both the natural form and in the logrithmic form (log to the base e) and is explained completely in a separate report ( 35 ) • Results - Several equations were developed for each set of data both in natural form and in logarithmic form. Generally the significant variables were: Xg F/M ratio Xg Detention Time xc Mixed Liquor Suspended Solids x.' TP/M ratio 16 ------- As an example, the best equation developed for the 1968 data was the following- Equation Number 1 Loge Y a 3.55118 - 0.02855 Loge x2 + 0.05819 Loge x - 0.10076 Loge x~ This equation used data from those days (23^) in 1968 when the effect of overloaded clarifiers was not evident. This was determined by using the following criteria: 1. Effluent BOD < 20 mg/1 2. Effluent SS < 20 mg/1 3. Clarifier Blanket < lU' k. (Total P - Soluble P) in the effluent < O.U mg/l-P If any 3 of the above U criteria were satisfied, the data for the respective date in 1968 was accepted and analyzed in the multiple regression analysis. The above equation was then tested against the 1969 data from January 1 to September 30, minus the June 9 to July 15 brewery- strike period, plus January 1, 1970. The same criteria for over- loaded clarifiers was used. A total of 256 days during the above period was compared. Comparison is best shown as follows by relating the measured % phosphorus removal to that removal calculated in the above equation (l). On 193 days, the calculated $ removal was within 10$ of the measured % removal. On 115 days, the calculated % removal was within 5# of the measured % removal. On 102 days, the calculated % removal was within U$ of the measured % removal. On 86 days, the calculated % removal was within 3>% of the measured % removal. On 65 days, the calculated % removal was within 2% of the measured % removal. On those days (196 out of 256) when the measured % removal was > 80$, the calculated % removal was within 10% except for 3 days. 17 ------- A conclusion can be drawn from this analysis. When the removal of phosphorus exceeds B0% in the Jones Island plant, the significant parameters are Food/Microorganism ratio, mixed liquor suspended solids and Total Phosphorus/Microorganism ratio as related in Equation (l). 18 ------- DEMONSTRATION OF CYCLIC REMOVAL OF PHOSPHORUS FROM SEWAGE BY AN ACTIVATED SLUDGE PLANT. In the literature prior to 1966, the majority of the studies reported on phosphorus removal from sewage by the activated sludge^ process were small scale laboratory experiments under ideal conditions. A few plant scale investigations were reported but were limited to the study of SOP levels in the effluent from the aeration tanks and not in the effluent from the secondary clarifier. Very little information from long term studies was available on an activated sludge plant's capability to remove total and soluble phosphorus from sewage on a hour- ly basis. At the outset of this investigation, the influent and the effluent of a 115 mgd activated sludge plant were monitored on an hour- ly basis, 7 days a week. This initial approach was taken to study the effects of plant operation practices, sewage flow and composition on the hourly removal of sewage phosphorus by an activated sludge plant. This study was carrried over a period from 12-3-67 to 12-19-68. Screened sewage and East Plant effluent grab samples were taken every hour and stored under refrigeration. These samples were analyzed the next day for BOD, TP and SOP. The data from the hourly grab samples of screened sewage exhibited extreme fluctuations, where- as the data from the effluent hourly grab samples showed very little fluctuation. Data for each hour for each day of the week was aver- aged because the data showed that sewage total phosphorus removal was influenced by the sewage composition and volume of flow, and by operation of the clarifiers which varied from day to day. For example, on Sunday and Saturday sewage volume of flow, BOD and phosphorus concentration were consistently lower than those for Monday thru Friday. The data for Monday thru Friday was considered separately because insufficient capacity in the dewatering facilities on those days controlled the MLSS level and the clarifier blanket depths. Because of this, it was usually not possible to maintain a given MLSS level in the aeration tanks from day to day during this study. The averaged hourly data for the screened sewage and for the East Plant effluent are presented in Figure 2. The East Plant effluent data was adjusted by 9 hours to compensate for the plant flow thru time. After this adjustment, the effluent data correspond- ed fairly close to the influent data. Figure 2 exhibits the cyclic nature of phosphorus re- moval by the East Plant. It also demonstrates that three cycles occur during each week in the East Plant. These cycles are related to three sets of circumstances which occur weekly and may vary from week to week. 19 ------- 4OO 360 320 280 -1240 j 200 160 O O 120 m so : o LEGEND SEWAGE EFFLUENT FLOW / X / x / X / X 140 H 120 110 O 100 O 90 80 £ O 70 J^ 60 51 I • X / A V / ~^—r~ N TUESDAYS WEDNESDAY N THURSDAY M M FRIDAYS SATURDAYS FIGURE 2 SEWAGE a EFFLUENT AVERAGE HOURLY COMPOSITION 12/3/67 TO 6/15/68 (EFFLUENT ADJUSTED FOR 9 HOUR DETENTION) ------- The first cycle was labeled the "Sunday Cycle". It started at approximately U PM (see Figure 2) on Saturday (actual effluent time would "be 1 to 2 AM Sunday) and continued until 6 AM Monday (actual effluent time, 3 to k PM Monday.). The reduced sewage flow and BOD concentrations occurring during this cycle resulted in increased detention time and air application rate which in turn in- creased the ML-DO. The most striking characteristic of the Sunday Cycle was the consistency of the low levels of TP, SOP and BOD in the East Plant effluent from hour to hour, which averaged below 1.0 mg/L-P, 0.5 mg/L-P and 10 mg/L respectively. This period appeared to have the most favorable conditions for the activated sludge process to remove phosphorus from sewage. The second cycle was called the "Daily Cycle". It differed from the first in that the effluent TP, SOP and BOD levels fluctuated over a 2k hour period in the same manner as' the sewage flow and nutrient (phosphates) concentrations. This cycle started with peak flow and BOD levels at 2 to 3 PM on any day and gradually decreasing to the lowest values at about 3 AM. The daily cycle occurred Monday thru Friday as shown in Figure 2. The third cycle was the "Weekly Cycle". The main charac- teristics of this cycle were the increase in the amount of insoluble phosphorus and BOD in the plant effluent starting from Monday which kept on increasing as the week progressed. This cycle was found to be dependent on the mixed liquor settleability and the capacity of the dewatering facilities for removing the waste sludge. Most of the time dewatering facilities were unable to remove the waste sludge as fast as it was produced. As a result, the unremoved waste sludge accumulated in the clarifiers causing higher sludge blankets during hours of peak flow on certain days. The amount of waste sludge accumulated in the system generally increased from Monday to Saturday, resulting in high sludge blankets during the peak flow hours. The final result was that the overloaded clarifiers sometimes discharged MLSS over the weir during peak flow hours and caused an increase in the effluent BOD and insoluble phosphorus concentration as shown in Figure 2. The effect of clarifier sludge blanket depths on the TP, SOP and SS concentrations of the East Plant effluent are summarized in Table 1 on the next page. 21 ------- TABLE 1 EFFECT OF CLARIFIER SLUDGE BLANKET DEPTHS ON EP EFFLUENT QUALITY Maximum Clarifier Daily Averages from Hourly Number Blanket Depth Effluent Sample Data of Days Reached in 24 Hours 12/3/67 to 9/15/68 Observed Feet 0 to U 5 to 8 9 to 13.5 TP mg/L-P 1.11 1.04 2.10 SOP mg/L-P 0.70 0.64 0.88 Insoluble -P* mg/L-P 0.41 0.40 1.22 • ss mg/L 16 14 45 107 75 102 « Insoluble P = TP - SOP The above data shovs that the effluent SOP levels on the average did not change greatly with increasing sludge blanket depths, 22 ------- PLANT STUDY OF THE EFFECT OF CLARIFIER BLANKET DEPTHS ON CLARIFIER EFFLUENT SOP RESIDUAL It is well known that SOP is released by sludge solids when they are detained in the final clarifiers for extended periods of time. Thus, the release of SOP from the clarifier sludge solids could in- crease the clarifier effluent SOP concentration. If this occurred,the effects of the process parameters in removing SOP by MLSS during aeration would be obscured. The effect of sludge detention time on the release of SOP from sludge solids was first studied in the laboratory. A typical laboratory experiment consisted of using a ML sample which had been taken from an aeration tank after six hours of aeration ( .and contained 0.3 mg/L - P of SOP). This ML sample was then divided into five portions which were placed into five one liter graduate cylinders and allowed to settle at room temperature (72°F). Then at hourly intervals (from one cylinder each hour) the supernatant was carefully removed in 200 ml aliquots (four in all). The supernatant fractions and the sludge fraction were then analyzed for SOP. The SOP release and dif- fusion data are presented in Figure 3. This experiment showed that the amount of SOP released from the sludge solids increased with increased sludge detention time as expected. However, as shown in figure 3, the majority of the released SOP remained in the quiescent sludge blanket fraction and only a small amount of released SOP diffused into the supernatant with time. This means that the clarifier effluent can contain some SOP released by the sludge solids detained in the clarifier by the process of diffusion. But, more important are clarifier operations during periods of high blanket and peak flows. The data indicates that it is possible that if the inflow ML causes turbulent conditions during periods of high blankets the clarifier effluent could contain SOP released by the solids detained in the clarifiers. Plant studies to determine if clarifier sludge SOP release had a significant influence on the clarifier effluent SOP residual were also made. During a ten day period the ML effluent from an aeration tank (EP Tank #10), the ML influent to clarifier #5 and the final effluent from a clarifier (EP Clarifier #5) were sampled 5 times a day. The SOP content was determined on each sample. The data in Table 2 showed that during this study period there was very little diffusion of SOP from the clarifier sludge blankets to the clarifier effluent. 23 ------- SOP MG/L - P V) f~ O o m CD m H m co co - m H X O O o c ^) m CM co o T) r m co m > o o d m en CO o 00° 5 -< m CD ro o SLUDGE BLANKET FRACTION 4 SUPERNATA FRACTION 3 r-o cnO r LIQUID- FRACTION 2 C/) FRACTION m Q ------- TABLE 2 COMPARISON OF SOP CONCENTRATION (MG/L-P) OF AERATION TANK-10 OUTLET ML, CLARIFIER #5 ML-FEED, CLARIFIER #5 EFFLUENT IN RELATION TO CLARIFIER BLANKET DEPTH SAMPLING TIME 12-23 1968 12-2U 12-26 12-27 12-30 12-31 1-3 1969 1-1* 1-6 1-7 Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth Outlet Feed Effluent Blanket Depth 7:30 AM O.U1* - 0.05 6 0.07 0.03 0.22 6 0.19 0.21 0.28 U 0.12 0.3U 0.37 6 0.67 0.77 0.76 1 O.U2 0.66 0.63 1 1.1* 1.1* 1.1* 1 1.0 0.92 1.5 1 3.1 3.2 2.9 1 3.7 1*.0 3.7 1 9:30 AM _ 0.03 6 0.19 0.06 0.13 6 0.20 0.19 0.27 U 0.06 0.19 0.2k 6 0.61 0.57 0.61* 1 0.13 O.lU 0.37 1 1.3 1.3 1.1* 1 0.27 0.2l* 0.70 1 3.0 3.0 3.0 1 2.9 2.7 2.7 2 11:30 AM 0.05 0.02 0.03 7 O.Ol* O.Ol* 0.07 6 0.26 0.22 0.29 1* 0.06 0.08 0.13 6 0.1*2 0.1*1 O.U6 1 0.15 0.13 0.19 1 1.1 1.1 1.0 1 0.12 0.06 0.16 1 3.0 2.9 2.9 1 2.0 1.8 1.7 2 1:30 PM 0.06 0.03 o.ou 8 O.Ol* 0.02 — 6 0.35 0.36 0.3** 1* 0.05 0.05 0.07 k 0.32 0.27 0.31 1 0.07 0.11 - 1 0.88 0.71* 1.0 1 o.oi* 0.06 0.09 1 2.7 2.7 2.6 1 1.6 1.2 1.3 3 3:00 PM 0.05 0.01 _ 8 _ _ _ 6 0.36 0.31 _ U O.Ol* 0.06 _ 1* 0.39 0.35 _ 1 _ _ — 1 0.83 0.71* _ 1 o.ou 0.07 _ 1 2.7 2.6 ^ 1 1.6 1.3 _ k Blanket nepth - Feet 25 ------- PLANT LOADING STUDY (F/M AND TP/M) The first year's data indicated that the plant loading parameters, food to microorganism ratio (F/M) and total phosphorus to microorganism ratio (TP/M), appeared to be related to the removal of phosphorus by the activated sludge process. The effects of the F/M and TP/M loadings were studied simultaneously in both the acti- vated sludge plants. The following loadings were chosen for this study: Plant Loading Ratios Remarks F/M TP/M East (EP) 0.300 0.012 Optimal Loading West (WP) 0.600 0.02U High Loading Under these loadings the EP was expected to yield good phosphorus removals as compared to the WP. It was further planned to reverse the above loadings between the two plants to see if the reverse results could be observed. The desired F/M and TP/M loadings were obtained by modifying the operations of both plants. The pertinent data summarized in Table 3 are averages of the daily data for the periods indicated. In general, the data indicated that the high loadings (with average values of 0.511* F/M and 0.019 TP/M in the WP for period II) reduced the WP-SOP removal efficiency. The effect of the plant loadings on the daily EP and WP effluent SOP concentrations are shown in Figure U. The initial plan was to then reverse the plant loadings to see if the EP - SOP removal efficiency would be reduced. A viscous floating material appeared on the surface of the EP aeration tanks in period II but none was observed in the WP. The amount of this viscous material continued to increase in the EP until it was felt that this material and the continuation of the loading study would jeopardize the EP treatment efficiency. Regular defoaming agents were ineffective in breaking this foam. Vacuum skimming of the EP aeration tanks and clarifier feed channels reduced the amount of this foam and aided in overcoming this problem. Thus, the plant loadings were not reversed and this study was suspended. The sewage flow distribution and other process parameters were then changed in both plants and were gradually restored to their normal values in period III. Further details on this foam problem are given in appendix (D). 26 ------- TABLE 3 AVERAGES OP DAILY DATA BY PERIODS. K> Period Plant % Sewage Distribution % Return Sludge I - Prior to Study Period 2/1 to 2/16- Dnr East 58 25 Air - Million cu.ft./day 133 Vest 1*2 25 107 II - East 50 35 Loading Study Period 2/17 to 3/10- Dry Flow West 50 25 III - East 25 After the Study Period 3/11 to 3/18- Dry flov West U6 25 IV - 3/19 East 51* 25 Period of Excessive Rainfall to 3/31-Rain West U6 25 133 107 133 107 118 107 SEWAGE SOP TSP TP BOD SS FLOW EFFLUENT SOP TSP TP BOD SS MG/L-P MG/L-P MG/L-P MG/L MG/L MGD MG/L-P MG/L-P MG/L-P MG/L MG/L 2.5 M 9.8 258 218 100.1 1.5 1.9 2.2 15.8 18 69.1 1.7 2.2 2.1* 15.8 19 2.7 I*. 2 9.7 260 253 88.1 1.1 1.1* 1.9 17.0 23 86.1 1.8 2.1 2.9 21.0 26 2.5 k.Q 9.9 306 258 93.0 0.59 0.91* 1.3 19.6 21 77.0 1.6 2.2 2.6 22.2 26 2.0 3.2 8.5 21*7 232 102.2 0.28 0.1*1* 0.82 18.8 18 86.1 0.77 0.97 1.5 20.5 28 PROCESS PARAMETERS F/M T?/M MLSS MG/L DO MG/L DETENTION HRS AIR CU.FT./GAL % REMOVAL SOP TSP TP O.U36 0.0167 2l*6l U.3 7.3 1.3H 0.372 0.0ll»2 2513 7.5 1.57 0.291 0.0109 2953 3.1 7.9 1.53 0.511* 0.0188 2273 6.2 1.26 0.366 0.0319 2989 3.9 7.7 l.UU 0.1*72 0.0152 2657 6.9 i.Uo 0.320 0.0109 3051* 3.1 7.1 1.18 0.335 0.0118 3291* 6.2 1.26 1*0.0 59.6 77.6 32.0 53.2 75.5 59.3 66.7 80.1* 33.3 50.0 70.1 76.1* 76.5 86.9 28.0 1*5.0 73.7 86.0 61.5 86.3 69.7 90.1* 82.U ------- I ' ,, STUDY PERIOD AFTER THE STUDY PERIOD 3/11/69 TO 3/18/69 PERIOD OF CONTINUOUS RAINFALL 3/19/69 TO 3/31/69 PRIOR TO STUDY PERIOD 2/17/69 TO 3/10/69 2/1/69 TO 2/16/69 LEGEND - EAST PLANT WEST PLANT SEWAGE So 3/23 So 3/30 FIGURE 4 EFFECT OF PLANT LOADINGS ON PLANT EFFLUENT SOP LEVEL ------- Reduced TP/M and F/M loadings observed in period IV vere caused "by surface runoff due to continuous rainfall in this period (1.02 inches in 13 days) which affected influent composition and flow volumes. The reduced loadings frequently happen because of infil- tration, since approximately 6.6% of the Commission's service area is served by combined sewers. Throughout this three year study, improved SOP removals in both plants were usually observed during such periods of continuous rainfall similar to the data shown in period IV. 29 ------- EFFECT OF BREWERY WASTE LOAD The effect of the industrial wastewater from the brewery in- dustries in Milwaukee on the activated sludge plants operations were studied in 1969 when they were shut down due to a strike lasting from June 9 to July 15. The loss of the brewery industrial wastewater and the changes in sewage treatment plant operations made to offset the effects of this loss offered an excellent opportunity for another plant scale loading study (F/M and TP/M) in the East and West plants. One of the significant effects of this industrial waste loss was a substantial reduction in the average sewage BOD when compared to the periods prior to and after the shutdown period as shown in Table U. The data indicated that the Milwaukee breweries average BOD contri- bution comprised approximately 22% of the normal total BOD load. The BOD removal efficiencies for both plants *rere not affected; however, they continued to exhibit greater than 90% BOD removals. It was also observed in a previous breweries shutdown in 1953 that the sewage BOD was substantially lower. During this previous period of low BOD the Milorganite nitrogen content decreased to an average of 5.31J& from a normal value of 6.0%. During the 1969 shut- down, the plant operations were modified to reduce nitrification by decreasing the MLSS level, decreasing the detention time and decreas- ing the air applied in both plants. These plant operational changes were successful during the 19^9 brewery shutdown, as evidenced by the small decrease in the average Milorganite nitrogen content, 5.75$ N vs normal of 6,0% N. During the 1969 strike period however, the TP removal effi- ciencies for both plants dropped from a monthly average of B5% to 63%. This decrease in TP removal efficiency was due to a significant reduction in the SOP removal efficiency from a monthly average of B0% to an average of 12.55? and 0.2% in the East and West plants respect- ively as shown in Table U. The decreased SOP removal was apparently influenced by an increase in F/M, an increase in TP/M and a decrease in the MLSS levels, these are shown in Tables U and 5. After the breweries resumed their normal production and the sewage treatment plant returned to their normal operating parameters, the SOP removal efficiency for both plants improved tremendously. During this long period of continuously poor removal of soluble phosphorus from sewage, one could expect to observe the RSSS phosphorus content to decrease. In Figure 5 are plotted the nitrogen and phosphorus contents of RS based on VSS (from the weekly composites for EP and WP return sludges). The data in Figure 5 shows that the RSVSS phosphorus did not decrease during this period of poor soluble phosphorus removal. Therefore, it may be assumed that a portion of 30 ------- TABLE U SUMMARY PERIOD AVERAGES Period April, 1969 May, 1969 Breweries Shutdown August, 1969 June 9th - July 15th Plant % Sewage % Return SEWAGE SOP TSP TP BOD SS FLOW EFFLUENT SOP TSP TP BOD SS Distribution Sludge MG/L-P MG/L-P MG/L-P MG/L MG/L MOD MG/L-P MG/L-P MG/L-P MG/L MG/L East 56 21* 2,1 3.3 7.8 230 210 109. 1* 0.31 0.51* 0.70 9.9 11 West 1*1* 26 85.0 0.32 0.5l* 0.96 13.2 18 East 60 23 2.0 3.1 7.7 238 238 112.1 0.27 0.1*8 0.79 17.5 19 West 1*0 27 7»». 9 0.26 0.1*7 0.90 15.2 31 East 58 22 2.1 3.6 7.1 182 196 110.9 1.9 2.1 2.1* 10.5 12.6 West 1*2 26 79 2. 2. 2. 12. 16. .9 1 3 7 0 5 East 59 2»* 2.3 2.8 7.3 221 218 115.6 0.50 0.60 0.93 9.U 15 West 1*1 26 81 0. 0. 1. 12. 28 .5 1*1 53 2 9 PROCESS PARAMETER F/M TP/M MLSS MG/L DO MG/L DETENTION HOURS AIR CU.FT./GAL % REMOVAL SOP TSP TP 0.352 0.012 2880 3.2 6.7 1.08 0.366 0.012 2970 6.3 1.21* 0.381* 0.012 2700 1.7 6.5 1.10 0.350 0.011 2770 6.9 1.32 0.1*76 0.019 1850 3.6 6.2 0.99 0.529 0.021 1870 5.3 1.10 0.1*35 O.Oll* 2390 U.9 6.1 1.09 0.389 0.013 2590 6.2 1.25 86 8U 91 8U 83 87 87 85 90 87 85 12.5 1*0 65 0.2 35 62 79 80 87 81 82 81* ------- TABLE 5 COMPARISON OF PERIOD AVERAGES; LOADING STUDY VS. BREWERIES SHUTDOWN Period 1969 Plant % Sewage Distribution % Return Sludge SEWAGE SOP MG/L-P TSP MG/L-P TP MG/L-P BOD MG/L SS MG/L FLOW MGD Loading Study Feb. 17th - Mar. 10th East 50 35 2.7 U.2 9.7 260 253 88.1 West 50 25 86.1 Breweries Shutdown June 9th - July 15th EFFLUENT SOP TSP TP BOD SS MG/L-P MG/L-P MG/L-P MG/L MG/L 1.1 l.U 1.9 17.0 23 PROCESS PARAMETER F/M 0.291 TP/M 0.011 MLSS MG/L 2950 DO MG/L 3.1 DETENTION HRS. 7.9 AIR CU.FT./GAL 1.53 % REMOVAL SOP 59 TSP 67 TP 80 1.8 2.1 2.9 21.0 26 0.511* 0.019 2270 6.2 1.26 33 50 70 East 58 22 2.1 3.6 7.1 182 196 110.9 1.9 2.1 2.1* 10.5 12.6 12.5 UO 65 West U2 26 79.9 2.1 2.3 2.7 12.0 16.5 O.U76 0.019 1850 3.6 6.2 0.99 0.529 0.021 1870 5.3 1.10 0.2 35 62 32 ------- • i i j SHUTDOWN IN OPERATION OPERATION % NITROGEN BASED ON RSVSS % PHOSPHORUS BASED ON RSVSS 3.0 2.0 4/13 4/20 4/27 5/45/M 5/18 5/25 6/1 6/8 6/15 6/22 6/29 7/6 7/13 7/20 7/27 8/3 8/10 8/17 FIGURE 5 EFFECT OF BREWERIES SHUTDOWN ON R S COMPOSITION (WEEKLY COMPOSITES) ------- the sewage insoluble phosphorus vas made available for assimilation by the microorganisms either in the initial aeration period vith MLSS or in the subsequent recycling vith the RS. In the Milwaukee vaste treatment plants the solids synthe- sized from the sewage BOD, suspended solids, and phosphorus are continuously removed from the system as Milorganite. During the breweries shutdown Milorganite production dropped significantly as shown in Table 6. Weekly balances are presented because it was difficult to calculate a reliable estimate for solids detention or withdrawal from the two plants on a day to day basis. The weekly pounds applied of sewage BOD and VSS also declined for the same period as shown in Table 6. The Milorganite phosphorus content did not appear to change significantly before, during, or after the shutdown as shown in Table 6. Phosphorus balance data showed that, on a weekly basis, the phosphorus removed from sewage was recovered in the waste solids withdrawn from the system as Milorganite. It appears that the reduction in phosphorus removal observed during the breweries shutdown was due to insufficient synthesis of solids for the quantity of phosphorus present in the sewage. In the Milwaukee plants the amount (and/or efficiency) of phosphorus removed from sewage appears to be influenced more by the amount of solids produced in the activated sludge process than by the activated sludge process parameters studied. In addition, it appears that the brewery waste water aids soluble phosphorus re- moval at the Milwaukee plants. 34 ------- TABLE 6 RELATION OF PHOSPHORUS REMOVED TO SOLIDS SYNTHESIS (1969 WEEKLY DATA) SOLIDS PRODUCED (DRY BASIS)* WEEK 3-30 U-6 U-13 1*-20 U-27 5-U 5-11 5-18 5-25 6-1 6-8 ' 6-15 6-22 6-29 7-6 7-n 7-20 7-27 8-3 8-10 8-17 8-2U OF to 11 ti it it ii it it it it —TT it 11 it ii it —IT II It II II ti l*-5 lt-12 1*-19 l*-26 5-3 5-10 5-17 5-21* 5-31 6-7 6-lU 6-21 6-28 7-5 7-12 7-19 7-26 8-2 8-9 8-16 8-23 8-30 TONS WEEK 1376 1682 1650 183U 1655 1537 13W 1656 161*0 lll*8»* 15U7*** 1309 1076 8Ul 1026 1171 lU7 11*1*6 1509 1678 151*5 151*9 % P TONS PHOSPHORUS 2. 2. 2. 2. 2. 2. 2. 2. 2. ?. 2. 2. 2. 2. 2. ?. 2. 2. 2. 2. 2. 2. 25 26 27 18 23 30 3»4 28 30 35 57 59 38 19 30 30 17 33 U3 5H 53 63 31. 38. 37. 1*0. 36. 35. 31. 37. 37. 27- 39. 33. 25. 18. 23. 26. 31. 33. 36. U2. 39. UO. 0 1 5 0 9 1* 5 8 7 0 8 9 6 )* 6 9 5* 7 7 6 1 7 TONS APPLIED/WEEK (BOTH PLANTS) PHOSPHORUS BOD M*. 8 1*7.0 1*1.3 1*1*. 8 1*2.8 1*2.5 1*1*. 5 1*2.1* UO.U 1*1.3 U3.7 1*1*. 0 37.6 30.5 37.9 39.3 - 35.1 Uo.i* 1*1*. 8 1*1*. 3 Ul.9 1292 ll*51 1061* ll*07 1358 1335 13U9 1218 1298 1189 1098 1139 1011* 829 957 1013 1251 1260 1167 1389 1190 ll*23 vss ll*ll ll*91 1293 131*6 1321* ll*56 1503 903 963 9^2 917 960 8U8 659 775 772 9W 937 551+ 1093 1053 1091 BREWERIES OPERATION IN OPERATION SHUTDOWN IN OPERATION NOTE *Milorganite production adjusted to account for either solids detention or withdrawal from the system. **Abnoraally low Milorganite production due to a holiday and the dewatering facilities shut down for two and a half days. ***High Milorganite production because the MLSS levels in the system were reduced from a weekly average of 2800 riR/L to the low level of 2000 mg/L (i 1*00) for the East Plant. ------- PLANT SCALE PHOSPHORUS BALANCES The phosphorus balances shown in Table 7 were made to check the accuracy of phosphorus and plant flov data obtained in this project. The influent and effluent TP contents were determined by a Ternary Acid Digestion procedure (as described in Appendix A). The Milorganite phosphorus content was determined by the gravimetric AOAC Quinoline Molybdate Method 2.023 and 2.021* (36). Phosphorus recovered in the Milorganite was compared with that removed from the influent by the two plants. The initial phosphorus balance calculations were made with the daily data but it was found too difficult to accurately account for phosphorus associated with the solids in the system. The phosphorus balances using weekly phosphorus removals, plant flows, and Milorganite production data were considered more real- istic. The estimated solids storage in the system on a weekly basis usually exhibited less fluctuation than on a daily basis. An average of 95.7/S of the phosphorus removed by both plants could be accounted for in the Milorganite produced over the 37 week period. As evidenced by the weekly phosphorus balances as shown in the Table 7, plant phosphorus and flow data were reasonably accurate. The phosphorus balances for both plants established the high phosphorus removals by the Milwaukee Sewage treatment plants and demonstrated that the phosphorus was removed from the process as Milorganite produced from waste sludge. ------- TABLE 7 CALCULATED WEEKLY PHOSPHORUS RECOVERIES #TP Removed/Week in 1969 Weekly Period Feb. 10-16 17-23 21+-2 Mar. 3-9 10-16 17-23 21+-30 31-6 Apr. 7-13 11+-20 21-27 28-1* May 5-11 12-18 19-25 26-1 June 2-8 9-15 16-22 23-29 30-6 July 7-13 11+-20 21-27 27-2 Aug. 3-9 10-16 17-23 2U-30 31-6 Sept. 7-13 lU-20 21-27 28-1+ Oct. 5-11 12-18 19-25 26-1 Nov. 2-8 9-15 16-22 23-29 Average % WP 29,81+0 38,559 37,355 32,113 32.91U 28,571+ 31+, 805 31*, 851 33,960 36,021+ 30,261 33,297 30,33** 29,696 31,257 29,851 22,135 36,266 2l+,8l9 18,120 17,325 16.1U2 21,1+71+ _ 21,762 29,779 3U.007 30,575 23,81*5 17,906 21+.1+21 22,690 26,993 9,1*79 10,551* 30,682 37,657 30,508 33,277 31*, 707 35,758 28,857 Recovery EP 1*0,026 32,206 1*9,009 37,828 1*1,058 1*5,337 U, 857 Ul+,903 J+l+,1+52 1*6,81*1 1*5,568 1*6,195 1*3,771 1*5,568 1*5,971* 1*6,1+22 38,708 149,028 1*1,953 30,653 22,268 21,U67 31,392 _ 3U ,216 1*0,796 U5.022 1*8,187 1*3,1*96 29,777 1+3,529 37,858 1+2,01+1 1*6,953 38,166 1*6,181 59,887 1+1+ , 977 1+9,807 1*9,251 1*8,1*12 1*1,398 Both Milorganite Plants 69,866 70,765 86,361* 69,91*1 73,972 73,911 79,662 79,751* 78,1*12 82,865 75,829 79,1*92 71+.105 75,261+ 77,231 76,273 60,81+3 85,291* 66,772 1*8,773 39,593 37,609 52,866 - 55,978 70,075 79,029 78,762 67,31*1 1*7,683 67,950 6o,5l+8 69,031* 56,1*32 1+8,720 76,863 97,5U1* 75,1*85 83,081+ 83,958 81*. 170 70,255 69,388 69,056 72,338 66,709 66,1+39 69,572 67,910 65,582 71,610 78,71+1* 7!*, 802 70,800 69 ,986 68,1*59 69,51*1* 77,982 1*8,1+1+5 86,939 72,105 52,837 1+0,795 36,681 1*6,336 - 63,760 69,29!* 7U.771 75,871* 71*. 090 51,729 6U, 1*97 57,753 6U, 691 58,232 56,726 61+, 1+00 70,896 67,691 78,371 76,175 71,562 61+, 377 Phosphorus Recovery .3 .6 99. 97. 83.8 95.1+ 89.8 9l*.l ,2 ,2 ,3 85 82 91 .0 .2 95.0 98.6 89.1 9U.U 91.0 90. 102. T9.6 101.9 108.0 108.3 103.0 97.5 87.6 113.9 95.7 (») ^Recovery (#TP removed in Milorganite-f-#TP removed in the plant) X 100 37 ------- RELATIONSHIP BETWEEN TOTAL BOD REMOVAL AND TSP REMOVAL IN AN ACTIVATED SLUDGE PLANT (BODR/TSPR) Sawyer (20) studied the removal of BOD in conjunction vith soluble phosphorus removal by the activated sludge process. He demonstrated in the laboratory, that by adding additional BOD (as glucose) he could obtain increased soluble phosphorus removal. From this work he concluded that 100 pounds of BOD were removed for each pound of soluble phosphorus in the activated sludge process. Hall and Engelbrecht (22) concluded from their studies that 80 to 110 pounds of COD were removed per pound of soluble phosphorus. It was decided to determine how applicable this concept was on a plant scale* The BOD to TSP removal ratios (i.e., BODj^/TSPp) for both plants were calculated from the daily 2U hour sewage and effluents composites data. The calculated removal ratios from 2/9/69 to 11/30/69 are presented as frequency diagrams in Figures number 6 and 7. The data used in these diagrams are for periods of normal operating conditions. The magnitude of the BODR/TSPR ratio appears to be dependent on the day of the week as shown in Table 8. TABLE 8 Average BODp/TSPR Ratios Plant Weekdays Weekend Days Monday thru Friday Saturday Sunday West Plant 86 60 kk East Plant 90 6U 50 The data appears to indicate that in the Milwaukee activated sludge plants the removal of soluble phosphorus from sewage is not dependent on the removal of total BOD as we always observed greater than 90% BOD removals. 38 ------- vo 24 20 16 12 UJ 8 O Z UJ 4 (K O 0 V 0 32 > 28 2 ^ 24 O UJ . u. 16 12 8 4 WEST PLANT DATA AVERAGE 86LBS BODR/LB. TSPR r-m EAST PLANT DATA 1 -AVERAGE 90LBS BODR/LB TSPR r -t- -I- 6 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 I9O 200 9 19 29 39 49 59 69 79 89 99 109 119 129 139 149 159 169 179 189 199 209 POUNDS BOD REMOVED/POUND TSP REMOVED FIGURE 6 FREQUENCY DISTRIBUTION BODR/TSPR WEEKDAY DATA ------- 8 • 6 • 4 • 2 • 6 • A IRRENCE 3 l\5 4 ^ w O o o Q . u. B o 6 O 5 4 O o- Ul 2 QC 8 6 4 2 SATURDAYS WEST PLANT EAST PLANT 1 I SUNDAYS WEST PLANT x If 1 EAST PLANT 0 9 10 20 30 40 19 29 39 49 AVER/ r~ N ^AVERAGE 60 I , , 1 1 !*•: . AVLRAGt 64 1 1 1 \GE 44 I 1 1 ""-•—- AVERAGE 50 1 1 50 60 70 80 90 100 110 120 130 59 69 79 89 99 109 119 129 139 POUNDS BOD REMOVED/POUND TSP REMOVED FREQUENCY DISTRIBUTION BODR/TSFk WEEKEND DATA FIGURE 7 40 ------- STUDY OF SOLUBLE PHOSPHORUS UPTAKE IN EAST PLANT AERATION TANKS (GENERAL) Eight separate studies were made during the course of tvo years on soluble phosphorus uptake in the aeration tanks. The ML de- tention time and air application rates were kept fairly constant at 6 hours and 3300 to 3500 cfm respectively. No attempt was made to control the MLSS levels and the F/M and TP/M loadings. Thus, they were the varying parameters which affected the ML-DO level and phosphorus re- moval efficiencies. ML Spot samples (over a six hour period) were taken at each hour along the aeration tank and "brought immediately to the labora- tory and filtered. At the location from which the ML samples were taken the ML-DO was also measured. The ML filtrates were analyzed for SOP and/or TSP, soluble BOD, and pH. The SS were also determined on the ML samples. In addition, the zero hour detention grab samples of screened sewage and RS samples were analyzed for SOP or TSP. The data from the majority of these studies were incon- clusive because the studies were conducted on influent arriving at 8 to 9 AM when the influent BOD, TP and SOP concentrations were near their lowest average concentration in the twenty four hour period (see Figure 2). As a result, practically no dramatic differences in effluent residual SOP were observed which could be related to the pro- cess parameters under investigation. It was consistently observed in these studies, that the SOP concentration of ML resulting from the mixing of screened sewage and RS in the mixing and ML feed channel as well as at zero hour aeration time in the aeration tanks was usually 2 to 3 times greater than the expected calculated SOP concentration in a corresponding mixture of sewage and RS. Also, at times the concentration of ML-SOP was greater than the sewage TP. It is unlikely that all the sewage insoluble phosphorus could be instantly converted to SOP when sewage is mixed with RS. Also, it is well known and easily demonstrated that the RSSS readily release SOP under anaerobic conditions. The RS was therefore considered as the logical source for the SOP in- crease in the ML. Sewage and effluent samples were also analyzed for SOP and TSP in several studies. The difference between TSP concentration and SOP concentration in the sewage was consistently greater than 1 mg/L-P. In the aeration tank effluent, however, as well as in the plant effluents, the difference between TSP concentration and SOP concentration was consistently only 0.2 - O.H mg/L-P. This indicated, that in the activated sludge process the MLSS rapidly degrade or remove and retain soluble poly-and organic-phosphorus compounds. 41 ------- EFFECT OF MLSS-Og UPTAKE RATES AND AIR APPLICATION RATES ON SOLUBLE PHOSPHORUS UPTAKE IN AN AERATION TANK. In the activated sludge process the aerobic microorgan- isms require C, P, N, etc., from sewage in order to maintain their metabolic and reproduction activity. The amount of dissolved oxygen in the ML is dependent on the air application rate, the oxygen trans- fer efficiency of the aeration devices employed, and the oxygen demand rate of the microorganisms. The aerobic microorganisms in the ML also require sufficient amounts of dissolved oxygen for these activities. Numerous reports in the literature have shown that the ML dissolved oxygen level is a significant activated sludge process parameter affecting soluble phosphorus removal (2,3,5,22). Accord- ingly, the effect of air application rate was evaluated using three tanks of the East Plant. This experiment was not performed on a plant scale because the MLSS level and plant loadings could not be controlled from day to day. Oxygen utilization rate measurements have been frequently used to assess the metabolic activity or proportion of viably active cells in MLSS. In the literature no information was discovered on the relation of MLSS-02 uptake rate to ML-soluble phosphorus levels. The establishment of such a relationship would provide a means to assess the MLSS capability to remove soluble phosphorus in the activated sludge process. Preliminary DO-profile studies of aeration tanks resulted in the selection of tanks 8, 9, and 11 of the East Plant for this study because these tanks had similar air distribution patterns (Figure 8). The ML flows to the three tanks were maintained at 7.5 mgd each with 3C# RS to provide 6 hours detention time in each tank. The experiments were started at 2 PM when the ML oxygen demand rate due to high sewage BOD was usually at its peak. The air application rates tested during this four day study were as follows in Table 9. TABLE 9 AIR APPLICATION RATES Tank No. CFM Cu. ft./gal sewage 8 2300 0.57 33% below normal 9 3500 0.86 Normal Rate 10 U700 1.16 33% above normal 42 ------- i 1 5.0 4.0 3.0 o 2.0 1.0 TANK 8 TANK 9 AIR CFM CFM TANK II -- AIR 3500 CFM 0 I 23456789 10 I 234 15 6789 INLET HEADER 19 8 7 6 5 ,4 3 2 I 10 9 8 7 6 5 4 3 NUMBER FIGURE 8 AERATION TANKS DO PROFILES ( 9-30-1969, 8:00 A.M. ) ------- The changes observed in the tank DO-profiles resulting from these air application rates are shown in Figure 9. The difference in the DO-profiles given in Figure 8 and Figure 9 is partly due to the different BOD loadings on these tanks when the DO-profile data was taken. At the start of each daily run, the 1 PM to 2 PM hourly composite of screened sewage was taken to the laboratory for BOD, SS, TP and TSP analyses. In addition a 7 liter grab sample of EPRS was taken at the same time and analyzed for SS and TSP. At 2 PM (zero detention time) ML-DO readings and a 7 liter grab sample of ML was taken at the inlet of each tank. The ML samples were immediately filtered and analyzed for TSP and soluble BOD. A portion of the un- filtered ML sample from each tank was analyzedfor SS^jHancl MLSS-02 uptake rate. This procedure was repeated every hour at sampling locations along the aeration tanks based on the detention time to follow the plug flow of ML through the aeration tank. The ML final pH levels did not change significantly due to the changes in air application rates as shown below in Table 10. TABLE 10 Effect of Air Application Rate on ML pH Air Rate Average ML pH eu.ft./gal Initial 0.57 7.^0 0.86 7.**0 1.16 7.^1 The process parameter data for these runs as presented in Table 11 are based on 6 hours detention, 115.5 mgd sewage flow, 30? RS, and a MLVSS content of 68.92. The data, listed in Table 11, showed that in an aeration tank at a given time the air application rate of 0.57 cu. ft./gal of sewage was inadequate for good phosphorus removal. However, an aeration rate of 0.86 cu. ft./gal of sewage or greater exhibited significantly improved phosphorus removals as indicated by the low TSP residuals in the aeration tank effluent. The hourly data from the four runs showed little vari- tion. As a result, the daily data was averaged and plotted in Figures 10, 11, 12, and 13. Figure 10 indicates, that to maintain a DO of at least 2.0 mg/L in the aeration tank beyond the turnpoint, an air application rate of at least 0.86 cu. ft./gal. was required. 44 ------- o o ci 8.0 6.0 4.0 2.0 TANK 8-- TANK 9 - TANK II — AIR 2300 CFM •AIR 3600 CFM •AIR 4800 CFM 01 23456789 10 I 23456789 20 I TP 10 19 INLET TURN POINT HEADER NUMBER 7654321 10 98765432 I OUTLET FIGURE 9 AERATION TANKS D.O. PROFILES (10-6-1969, 12:30RM. ) ------- Date TABLE 11 DAILY PROCESS PARAMETER DATA (Detention time controlled to 6 hours) At Tank Outlet Aeration Tank 10-6 Mon. 10-7 Tues 10-8 Wed. 10-9 Thurs. 8 9 11 8 9 11 8 9 11 8 9 11 F/M 0.517 0.635 TP/M 0.03^9 0.0196 0.0215 0.0159 MLSS mg/L 2530 2610 2770 2910 Air cu. ft. /gal. 0.57 0.86 1.16 0.57 0.86 1.16 0.57 0.86 1.16 0.57 0.86 1.16 ML-DO mg/L 1.0 6.0 7.8 « « * 0.5 1.2 3.5 0.5 1.0 2.5 ML-TSP mg/L-P 6.U 0.3H 0.28 lU.U 0.2U 0.18 11.5 0.20 0.16 9.7 0.2U 0.2U ML Soluble BOD mg/1 9.U U.U k.k 1U.U 5.U U.2 15. U 5.6 5.8 1^.0 6.2 5.U * D. 0. meter inoperative ------- A 2300 C.FM. AVE. 4 RUNS O 3500 <• " 4 " X 3600 " " 6 " • 4700 » » 4 O I _J v. O O O 1234 DETENTION HOURS FIGURE 10 EFFECT OF AIR APPLICATION RATE ON ML-DO LEVEL 47 ------- 60 cc CO CO 50 A 2300 CFM AVE. 4 RUNS O 3500 " » 4 ii X 3600 •• •• 6 • 4700 n i. 4 ., 2 3 DETENTION 4 HOURS FIGURE 11 EFFECT OF AIR APPLICATION RATE ON MLSS-02 UPTAKE RATE 48 ------- 20 A 2300 CFM AVE. 4 RUNS O 3500 " " 4 " X 3600 " " 6 • 4700 " " 4 " 234 DETENTION HOURS FIGURE 12 EFFECT OF AIR APPLICATION RATE ON ML-TSP LEVEL 49 ------- o Q O 03 LJ _l CO 2300 CFM AVE 4 RUNS 3500 " ii 4 'i 3600 " ii 6 '• 4700 » » 4 " 1234 DETENTION HOURS FIGURE 13 EFFECT OF AIR APPLICATION RATE ON ML-SOLUBLE BOD LEVEL 50 ------- Aa plotted in Figure 11, MLSS-Og uptake rates observed in this study at 0-hour detention time in the aeration tank vere about half of the values reported by Eckenfelder (37). This may be due to approximately 20 minutes aeration of the ML in the long feed channels prior to entrance of the ML into the aeration tanks. TABLE 12 MLSS-02 UPTAKE RATE AFTER 6 HOURS AERATION AIR APPLICATION RATES mg 02/gm VSS/hr. CFM cu. ft./GAL. 2300 0.5T 3500 0.86 UJOO 1.16 The above values in Table 12 compare favorably with those reported by Okum and Lynn (38) of 22 to 28 mg 02/gm VSS/hr. At no time vere the endogeneous oxygen uptake rates of 1.9 to 9.8 mg 02/hr/1000 mg sludge as reported by Eckenfelder (37) observed. An examination of Figures 12 and 13 indicates that an aeration rate of 0.57 cu.ft./gal. of sevage was insufficient for soluble phosphorus removal but did not affect the BOD removals. AVERAGE (H RUNS) 32. U 25.1 23.8 RANGE LOW 28.9 22.3 19.5 HIGH 36.0 27.1 26.1 51 ------- EVALUATION OF MLSS METABOLIC ACTIVITY BY GLUCOSE DEHYDROGENASE ASSAY Observations of the daily process parameters data indicated that they are sufficient to explain why poor or good soluble phosphorus removal occurs in the Milwaukee activated sludge plants. Ford, Young and Eckenfelder, (39) investigated the use of a glucose dehydrogenase assay for evaluating the biological activity of sludges. It was then thought that if the removal of soluble phospho- rus in the activated sludge process could be related to the biological activity of the MLSS, this additional information could probably explain the observed deviations. In an aeration tank, the oxygen uptake rate of the MLSS decreases with increasing detention and air application rate (Refer to Figure 11). Thus, the metabolic or oxidative activity as represented by the MLSS-glucose dehydrogenase activity could be ex- pected to decrease from the tank inlet to the tank outlet and perhaps coincide with the SOP being removed by the MLSS. The data of Ford, et. al. (39) showed that the MLSS- glucose dehydrogenase activity within a sampling location exhibited a very large coefficient of variation (C.V.). They performed one hundred triplicate analyses on samples from three sampling locations and obtained C.V.'s of 31.6#, 2U.5JJ and 23.85? respectively. For this study to be worthwhile it was necessary to determine whether the large variation in MLSS-glucose dehydrogenase activity reported by Ford was caused by the assay itself or was caused by the samples that were taken from a specific sampling site. In this study, seven liter ML samples taken from the inlet, turn- point and outlet of an aeration tank were each analyzed in triplicates for MLSS and MLVSS (Appendix B) and glucose dehydrogenase activity. The standard deviation of the dehydrogenase assay was calculated from the grand average range of the range of triplicate analyses per sample (19 samples from each of three sampling sites) according to the procedure given by Natrella (^0). The values used in this calculation were the range of triplicate determinations per sample as shown in Table 13. 52 ------- TABLE 13 DEHYDROGENASE ASSAY VARIABILITY ( jiM TPF/5 ml of ML) SAMPLING RANGE OF TRIPLICATES STANDARD SITE AVERAGE HIGH LOW DEVIATION INLET .032 .072 .013 .0189 TURNPOINT .OU9 .097 .000 .0289 OUTLET .050 .18U .000 .0295 The actual value from the assay used in calculating MLSS-glucose dehydrogenase activity for each sample was the average of the triplicate analyses for each sample. These values exhibited wide variations for the ML samples analyzed as shown in Table lU. TABLE lU DEHYDROGENASE ASSAY DATA TPF/5 ml of ML) Sampling Average Range Site High Low INLET 0.822 1.150 O.U15 TURNPOINT 0.762 1.0U3 0.285 OUTLET 0.729 1.051 0.212 The comparison of data in Tables 13 and lU indicated that the reproducibility of the assay was fairly good and that the vide variations in MLSS-glucose dehydrogenase activity was due to the samples which were taken within a sampling site for analyses. 53 ------- The variability of the MLSS-dehydrogenase activity according to sample sources (same 19 runs) vas also determined as shown in Table 15. The MLSS-dehydrogenase activity vas expressed as juM TPF/mg MLVSS (jiM Triphenylformazin produced/mg MLVSS). TABLE 15 VARIABILITY OF ML - DEHYDROGENASE ACTIVITY AMD MLVSS AT DIFFERENT SAMPLE LOCATIONS Aeration Tank ML-Dehydrogenase Activity MLSS MLVSS Sample Location Inlet Turnpoint Outlet Average Average uM TPF/mg MLVSS 0.072 0.068 0.063 Standard Deviation 0.01^9 0.0130 O.OlUl % CV 20. 19. 22. 20. 7 1 U 7 Ave. mg/1 3030 3160 3130 Ave. ?VSS 75.2 7^.3 73.9 Std. Dev. 0.89 0.89 0.79 i CV 1.18 1.20 1.07 1.15 Std. Dev. at Standard Deviation The above data continued to show that the large variation in the glucose dehydrogenase activity within a sampling site was due to the variation from sample to sample. The CV of the MLSS-glucose dehydrogenase activity at different sample locations conform to those reported by Ford as shown in Table 16. TABLE 16 COMPARISON OF COEFFICIENTS OF VARIATION (C.V.) Milwaukee Data Sample Location % C.V, Aeration Tank Inlet 20.7/5 Aeration Tank Turnpoint 19.1? Aeration Tank Outlet 22.U? Average 20.7? Ford's Data Sample Location ? C.V. Contract Tank 31.6? Stabilization Tank (Sampling Point A*) 2U.5? Stabilization Tank (Sampling Point B«) 23.8? •Sampling Location Not (39) Specified in Reference The usage of the glucose dehydrogenase assay to evaluate the metabolic or oxidative activity of the MLSS in an activated sludge plant was found to be of little practical value as a large number of samples would have to be analyzed to provide significant 54 ------- data. It should be noted that the data on the average did show that the MLSS-glucose dehydrogenase activity decreases with increasing detention time, (see Table 15). Jones and Prasad (Ul) made a detailed study on the use of tetrazolium salts as a measure of sludge activity. They concluded that, "The TTC dehydrogenase test in mixed cultures and complex substrates at best can be considered as a gross measurement of activity. The interpretation of the data should be approached with caution particularly if it is being used as a research tool." 55 ------- X-RAY DIFFRACTION STUDIES OF SEWAGE SUSPENDED SOLIDS AND WASTE SLUDGE SOLIDS In the literature, the removal of sewage phosphorus by the activated sludge process has been attributed to cationic precipi- tation or biological uptake, or a combination of both. Therefore, a knowledge of the mechanism or mechanisms prevailing vould not only enable us to obtain a better understanding of the phosphorus removal process but also to develop concepts for the improvement of phosphorus removal in the activated sludge system. Many cations (iron, aluminum, calcium and magnesium) have been successfully added to aeration units by many investigators to enhance the removal of soluble phosphorus. If a qualitative and quantitative procedure to determine the presence of these inorganic phosphate compounds in sewage suspended solids and vaste sludge solids were available, the mechanism of phosphorus re- moval by the activated sludge process could be better explained. Accordingly, staff members of Marquette University, Milwaukee were consulted on the use of X-Ray diffraction analysis to determine if these inorganic phosphate compounds were present in the sewage suspended solids and waste sludge solids. The preliminary X-Ray diffraction studies (U2, U3, UU) re- vealed that a crystal-like compound in the 103°C oven dried RS gave an X-Ray diffraction pattern very similar to that of Iron (ill) orthophosphate (FePOjj). This observation could not be confirmed and studied further because of the appearance of too many background lines in the X-Ray diffraction patterns obtained in the subsequent studies due to amorphous material in the sludge. The sludge samples were then heated at different temperatures to see if the background dark- ness could be decreased. Heat treatment of the samples (500°C to 600°C) decreased the background darkening considerably, but this high temperature heat treatment changed the nature of the compound formed and resulted in a homogeneous mass. Freeze drying of the sample was tried. This did not help in decreasing the background but did result in a heterogenous mass containing some black crystals. These black crystals were separated and found to give patterns similar to vivianite, Fe3 (POj^ • 8H20. Further research will be conducted in an attempt to purify and evaluate the concentration of these black crystals. Magnetic separation techniques are currently under study and show promise of success. These X-Ray studies are being continued under Project Grant Number 11010, FLQ, "Phosphorus Removal With Pickle Liquor In A 115 MOD Activated Sludge Plant". 56 ------- CATIONIC REMOVAL OF PHOSPHORUS FROM SEWAGE A number of investigators have studied the ability of certain cations (Fe+++, Fe++, Al***, Ca^, and Mg**) to enhance the soluble phosphorus removals in the activated sludge process. Very little information is available in the literature on the concentrations of these soluble cations in sewage and their effect on sewage phospho- rus removal. Studies were undertaken to assess the role of Fe, Al, Ca and Mg ions, normally present in Milwaukee sewage, in the removal of SOP by the two Milwaukee activated sludge plants. Currently, it was assumed that the insoluble cations (i.e., the cations present in the sewage suspended solids) are tied up possibly as Fe(OH)3 and are not available for reaction with SOP to form insoluble phosphate compounds during the activated sludge process. A comparison of the influent and of the effluent soluble cation concentrations would indicate if these cations were removed in the activated sludge process. It was hypothesized that if a decrease in the soluble cation concentration was observed, this decrease was due to the formation of an insoluble phosphate compound through its reaction with SOP. The cation analyses were performed on the unneutralized ternary acid digestate of 2H hour composite samples of screened sewage, EP and WP effluents by atomic absorption. The initial work was carried out using the 2H hour composite samples to determine if there was a significant change in the soluble cations concentrations on a daily basis during July of 1969. The daily data exhibited marked fluctuations because of the limited experience on the atomic absorption unit by laboratory personnel. Therefore, only monthly averages are presented in Table IT 57 ------- TABLE 17 AVERAGE CATION CONCENTRATIONS (mg/L) FOR JULY, 1969 TOTAL CATION CONCENTRATIONS TOTAL SOLUBLE CATION CONCENTRATIONS Screened Effluent Fe & Fe Al Ca 5.00* 0.71 0.59 2.00 0.53 0.41 52.4 48.8 47.9 Screened Sewage 0.55 0.44 48.6 Effluent WP 0.32 0.33 47.7 EP 0.29 0.27 47.2 Mg++ 31.8 30.7 30.2 30.4 30.5 30.3 ^Partly influenced by filtrate from vacuum filters The above data showed that both activated sludge plants removed very little of the soluble cations of Fe, Al, Ca and Mg. Based on this data, the removal of SOP in the Milwaukee activated sludge plants by the sewage soluble cations would be very insignifi- cant From the above data one could assume that the cationic fix- ation of soluble phosphorus probably takes place in the sewage prior to its reaching the plant. The consistently low amount of total soluble iron in Milwaukee sewage (usualljK 1.0 mg/l-Fe) may be due to the low solubility of the iron phosphate compound present in the sewage suspended solids. The initial X-Ray diffraction studies on the 103°C dried sludge samples indicated that one of the insoluble cation phosphate compounds was possibly FeP04 which in the wet sludge may be present as FePOA . 2H00 or FeP04 . 4H20. The theoretical solubility product of FePoZ - 2Hfo is 1 x 10'33 in distilled water according to Chang and Jackson (45). The solubility product of FeP04 • 2H20 in sewage was calculated from the observed data and was found to average 6.7 x 10 at pH 8.0. This higher value would be expected since sewage contains substances, such as detergents, polyphosphates, chelating agents, salts and other unknown compounds which could increase the solubility of FePOA . 2H90 in sewage. The above assumptions are reasonable because the calculated solubility product agreed with that reported by Galal- Grochev and Stumm (46) of 1 x lO"23 at a PH greater than 7.0 at 25°C for FeP04. 58 ------- Since the activated sludge process removes polyphosphates, surfactants, etc., from sewage the solubility product of ferric phosphate vould be expected to decrease. The calculated solubility product of FePOii 4 2H2° in ^ afterp£ nours aeration was found to decrease to an average of 1.3 x 10~ ' at pH 7«0. 59 ------- EFFECT OF IRON ADDITION TO AN AERATION TANK ON SOLUBLE PHOSPHORUS REMOVAL During June and July of 1969 the TSP and SOP removal efficiencies for both plants dropped from an average of 80$ to an average of Uo#. The period of reduced removals coincided with the breweries shutdown which was discussed in a previous section. This was an opportune time to begin a cationic addition experiment, since this period of extremely reduced SOP and TSP removals could provide more reliable and significant data on the cationic removal of soluble phosphorus. This phase of the study was a cooperative venture with the A. 0. Smith Corporation of Milwaukee, who supplied the cationic com- pound, ferrous sulfate as waste pickle liquor needed for the addition of iron to the ML. The industrial waste used contained from O.U to 1.0 pound of iron per gallon and usually had a pH of about 1.0. Neutralized and unneutralized pickle liquor were used in this study. The initial work was limited to the addition of iron to one of the aeration tanks of the EP. This approach was taken to determine the effective iron dose, the desired addition site, and the material handling problems associated with the pickle liquor addition. We were also interested in determining if this waste product would have any detrimental influence on the activated sludge process or on the air diffuser plates. The waste pickle liquor was first added to the aeration tank at the turning point, providing 3 hours detention, and then at the inlet, providing 6 hours detention. The ML flows to the aeration tanks were held at 7.0 to 7«5 mgd to provide about 6 hours detention, and the air application rate was maintained at approximately 3200 cfm. The iron was added 2k hours'/day from Monday to Friday and the pickle liquor addition rate was manually controlled. This study was conducted for two periods of three weeks each. An adjacent aeration tank with the same ML flow and aeration pattern was used as a control. The experimental and control tanks ML were sampled at the inlet, turn- point and outlet every 2 to 3 hours continuously. The ML samples were analyzed for total iron, total soluble iron, TP, TSP, MLSS, pH, and ML-SDI, and ML-DO at the tank outlets. The iron dosing rates maintained were approximately 7.5, 15, and 30 mg/L-Fe based on the hourly ML flow. The averages of the bi- hourly and trihourly data by week are presented in Table 38. The data such as given in Table 18 showed that the addition of ferrous iron in the form of waste pickle liquor was very effective in the reduction of ML-TSP. Actual iron dosing rates of 11.0 to 27.k mg/L-Fe (based on ML flow) resulted in average ML-TSP residuals between 0.23 to O.U9 mg/ L-P and reduced the inlet ML-TSP levels (3.5 to 8.5 mg/L) by 77.3 to 60 ------- TABLE 18 EFFECT OF IRON DOSAGE ON ML-TSP LEVEL Cr> Run No 1 2 3 U 5 6 Date Iron Add! 1969 Dosage mg/L-F 7/1 to 7/3 7/8 to 7/11 7/15 to 7/18 8/U to 8/8 8/11 8/18 to 8/15 to 8/22 11.0 6.8 27. U 30.0 15.0 13.3 15.5 17.0 tion Average ML-T (1) Site Inlet (2) 'e TP (U) TP TP (5) past TP (6) (5) past TP Inlet Inlet Inlet 'SP mg/L-P Outlet (3) Exptl. Control Tank Tank 3. k. U. U. 2. 5. 6. 8. 5 85 65 l 8 U 9 5 0. 1. 0. 0. 0. 0. 0. 0. U6 1 U6 31 23 HO U5 U9 1.9 2.8 2.5 2.0 0.76 0.88 1.2 1.1 % Reduction ML-TSP Exptl. Tank 86.9 77.3 90.1 92. U 91.8 92.6 93.5 9U.2 Control Tank H5.7 H2.3 U6.3 51.2 72.9 83.7 82.6 87.1 (l) Dosage based on Pickle Liquor iron analyses (2) Average of experimental and control tanks inlet ML-TSP (3) Outlet ML samples (6 hours detention) (k) TP * tank turnpoint and is 1/2 of tank length (5) In run 3 these dosages are estimates (6) Past TP (2 hours detention) ------- All three iron addition sites (2,3 and 6 hours detention) were found to be effective, provided there was sufficient ML-DO. The role of ML-DO level on the effectiveness of ferrous iron in TSP removal was observed in run 6 and a typical observation at an aeration rate of 3600 cfm. is given below in Table 19 TABLE 19 RUN 6 OBSERVATION DATA Turnpoint Outlet D.O. MG/L ML-TSP MG/L D.O. MG/L ML-TSP MG/L Control Tank 2.7 1.1 5.0 0.35 Experimental Tank 0.5 5.2 6.6 0.22 In Table 18 it can be seen that the average removal of TSP by the MLSS in the control tank was much better in runs U, 5 and 6 than in runs 1, 2 and 3, even though the inlet average ML-TSP levels were much higher in runs U, 5 and 6 compared to the early runs. This difference is attributed to the effect of the breweries waste water on the Milwaukee plants. First, the initial iron addition runs (7-1-69 to 7-18-69) were carried out during the breweries shutdown period (6-9-69 to 7-15-69) when the daily soluble phosphorus removals averaged less than H0#. In the second set of runs (8-U-69 to 8-22-69) the breweries were back in operation and the daily soluble phosphorus removals improved to an average of greater than 80%. The results in- dicated that the addition of ferrous iron as pickle liquor to ML was effective in maintaining low residuals of TSP in ML after 6 hours of aeration. Added soluble ferrous sulfate as pickle liquor was rapidly incorporated into the MLSS. This was found by the analysis of ML samples for total soluble iron taken at ten foot interval sampling points from the iron addition site at the tank turnpoint as shown in Table 20. 62 ------- TABLE 20 REDUCTION OF SOLUBLE IRON CONCENTRATION IN AN AERATION TANK (Iron * was added as waste pickle liquor at the aeration tank turnpoint) DISTANCE FROM TOTAL SOLUBLE IRON TANK TURNPOINT FOUND IN THE ML (FEET) (mg/L-Fe) 10 l.lU 20 l.lU 30 0.58 UO 0.50 50 0.56 60 0.58 70 O.M 80 O.UU 90 O.U3 * Iron dosing rate 7.5 mg/L - Fe based on ML flow. The average ML total soluble iron concentrations at the inlets and outlets were practically the same, 0.23 to 0.39 mg/L-Fe respectively during the 6 test runs in the experimental tank. The iron added, as obtained by calculation, from the pickle liquor data compared favorably with the average total iron found in the inlet and outlet MLSS as shown in Table 21 • The clarifier detention times are approximately 2 to 3.5 hours. During this time period it has been previously shown that soluble phosphorus is released from the solids. Ten iron and phosphorus tests were performed in the laboratory with the samples of ML from the control and experimental tanks outlet and then the release of phosphorus and iron after two hours detention was determined. The average results of the ten test runs as given in Table 22 , indicate that in a period of 2 hours the MLSS treated with iron release less TSP than the untreated MLSS from control tanks and that the iron treated MLSS released more soluble iron than the untreated MLSS. 63 ------- TABLE 21 AVERAGE IRON RECOVERIES Observed Average Run Ho. 1 2 3 U 5 6 Planned iron ca Addition Rates Ad gprn 1.0 0.5 2.0 2.0 1.0 1.0 0.5(b) 1.0 1.0 mg/L-Fe 15 7.5 30 30(a) 15(a) 15 15 15 15 icuiatea iron dition Rates mg/L-Fe 11.0 6.8 27. U 13.3 13.2 15.5 17.0 /iW UU«*J. J.X 1/4* i Wfc****-* sxi In the Outlet MLSS R< m«t/L-Fe Exptl Tank 70.6 55.0 80.5 8U.3 75.1 68.5 78.1 90. U Control Tank 59.2 U9.9 56.7 60.8 60.0 55.9 59.8 7U. U scovery of Added Iron Dose rag/L-Fe 11. U 5.1 23.8 23.5 15.1 12.6 19.0(c) 16.0 (a) Iron dosing rate estimated from pickle liquor specific gravity. (b) Concentrate (undiluted pickle liquor) (c) The large difference in run 5 may be partly due to inadequate manual control of pickle liquor flow. ------- TABLE 22 SOLUBLE IRON AND TSP RELEASE BY MLSS (average of 10 tests) Control Tank Detention Hours 0 hour 2 hours Total soluble Iron 0.21 0.30 mg/L-Fe TSP mg/L-P 1.61* 3.01 Experimental Tank 0 hour 0.27 0.29 2 hours 0.71 1.15 In general, the neutralized and the unneutralized waste pickle liquor were observed to have no detrimental effect on the activated sludge process. This was shown by the experimental and control tanks ML having similar pH's, MLSS-02 uptake rates, SDI values. Microscopic examinations showed no changes in the biota. In this limited study no apparent clogging of aeration tank diffuser plates was observed. 65 ------- PLANT SCALE IRON ADDITION STUDY This was a feasibility study of plant scale iron addition for improved soluble phosphorus removal during a two week period. The EP was used as the experimental plant and the WP as the control. The iron was added in the form of ferrous sulfate as unneutral- ized pickle liquor, a waste product which was supplied by the A. 0. Smith Corp. Milwaukee, in a cooperative venture. The A. 0. Smith Corp. also provided personnel to monitor the pickle liquor iron concentrations and to control the iron dosing rate 2U hours a day. The pickle liquor was added to the EP sewage feed channel 65 feet upstream from the EP RS feed pipe. The ferrous iron concent- ration in the pickle liquor was determined by titration with 0.10N potassium dichronate. The iron addition rate was then based on the pickle liquor iron content and the hourly ML flow rate. A dosing rate of about 15 mg/L-Pe based on the ML flow was used from 11/3 to 11/7 (2U hours/day for U-l/3 days) during which a total of 110,898 gallons of pickle liquor (7^,300 pounds of iron) was added to the ML. From 11/10 to 11/lU (2U hours/day for U-l/3 days), 52,3^5 gallons (37,200 pounds of iron) of pickle liquor was added to ML for a dosing rate of about 7.5 rag/L-Fe. The sampling pro- cedures and the analytical methods used were as described in pre- vious sections. The data from the 2U hour WP and EP composite effluent samples do not readily show the effectiveness of the added iron in reducing the effluent residual SOP as shown in Table 23. However, a comparison of the SOP content of the sewage and EP effluent bihourly composite samples (Figure lU) illustrates very dramatically that a 15 mg/L-Fe dosing rate consistently pro- duced an effluent SOP residual of 0.05 mg/L-P for U-l/3 days. The 7.5 mg/L-Fe dosing rate was not very effective as shown in Figure lU. In the previous section it was found that in an aeration tank the added soluble iron was rapidly removed, and did not appear to significantly increase the ML total soluble iron content. This was also observed during this study in the daily data from both plant effluents. The total soluble iron in the plant effluent averaged 0.3U mg/L-Fe for the WP and 0.31 mg/L-Fe for the EP. Plotted in Figure 15 are the WP and the EP SEE and MLSS shift averages on a daily basis. The plots show that during this study period the ML settleability fluctuated with a weekly cycle, improving on Sunday, Monday and Tuesday and deteriorating steadily from Wednesday to Saturday for both the plants. The plot also shows that an iron dosing rate of 15 mg/L-Fe was more effective than the 7.5 mg/L-Fe dosing rate in improving the settleability (SDI) 66 ------- No Iron Addition TABLE 23 EFFLUENT SOP mg/L-P Iron Addition No Iron Addition 10/27/69 to 11/2/69 Monday Tuesday Wednesday Thursday Friday Saturday Sunday Date 10-27 10-26 10-29 10-30 10-31 11-1 11-2 WP O.lU 0.16 O.U5 0.1*5 0.51 0.13 0.10 EP 0.15 0.21 0.1*6 0.32 0.25 O.ll* 0.11 15 mg/L-Fe 11/3/69 to 11/9/69 Date 11-3 11-1* 11-5 11-6 11-7 11-8 11-9 WP 0.17 0.10 0.19 0.53 0.82 0.31 0.10 EP 0.07 0.05 0.05 0.05 0.06 0.05 0.18 7.5 mg/L-Fe 11/10/69 to 11/16/69 Date 11-10 11-11 11-12 11-13 ll-ll* 11-15 11-16 WP O.U6 0.59 0.60 0.70 0.99 0.27 0.13 EP 0.1*1 0.12 0.22 0.27 0.1*1* 0.19 O.U7 11/17/69 to 11/23/69 Date 11-17 11-18 11-19 11-20 11-21 11-22 11-23 WP 0.20 0.31 0.21 0.30 0.53 0.1*5 0.23 EP 1.6 0.75 0.58 0.52 0.70 0.31 O.Ul Note: Iron added only to the East Plant Dosing Rate: 15 mg/L-Fe; 7 A.M. - 11/3/69 to 3:30 P.M. on 11/7/69 7.5 mg/L-Fe; 7 A.M. - 11/10/69 to 1:05 P.M. on ll/lU/69 ------- 10/27/69 TO 11/21/69 LEGEND - SEWAOE SOP EFFLUENT SOP 00 o 3 Q O C/1 NO IRON ADDITION IRON ADDITION I5MG/L - FE NO IRON ADDITION IRON ADDITION 75 MG/L - FE NO IRON ADDITION I M I M I M I M M M I M I M I M 1 I M M I I M M I ' I M M I I M M I M SU I M W | TH SA | SU TH SA | SU I M T I W | TH SA | SU W | TH SA FIGURE 14 EFFECT OF IRON ADDITION ON EAST PLANT RESIDUAL SOP ------- o-. 1.20 LEGEND O—OWEST PLANT EAST PLANT •60' T I I I I I I I I I I I ' I I ' A 3 5 7 9 II 13 15 17 19 21 23 25 27 26 28 30 | OCT. \ FI6UREI5 COMPARISON OF EAST a WEST PLANT SDl'S 1969 ------- of the EP MLSS vhen compared to that of the WP MLSS. This was probably due to the iron accumulation in the MLSS vhich increased the density of EP MLSS. The accumulation of added iron in the East Plant Return Sludge with time is plotted in Figure l6. The data was obtained from RS samples taken continuously at intervals of 2 hours. The first iron addition (15 mg/L-Pe) began at 7 A.M. on 11/3/69 and the accumulation of the iron in the RSSS (approximately 2.5JO was observed to begin with the U P.M. sample on 11/3/69. From this point, the iron continued to accumulate in the RS until 8 A.M. on Thursday, 11/6/69 (6U hours of iron addition) and then the RS iron content stayed at a fairly constant level (at an average of 6.01*) until midnight of 11/7/69 (next Uo hours). This indicated that an equilibrium condition was reached when the iron wasting rate (i.e., the iron in EP-WS) was equivalent to the iron addition rate. From k A.M. on Saturday 11/8/69 the EP-RS iron content slowly decreased due to the cessation of iron addition. The second iron addition (7.5 mg/L-Fe) was started at 7 A.M. on Monday 11/10/69* Figure 16 shows that at this iron addition rate the iron did not show any further buildup in the EP-RS (average 5.26JO. This indicated that the equilibrium iron concentration was reached for these iron dosing and sludge wasting rates. After iron addition was stopped at 1 P.M. on Friday ll/lU/69, the RS iron content was observed to continually decrease from 5.26j» at 10 P.M. on Friday until it approached the initial RS iron level of 2.5# on Thursday, 11/20/69. The total phosphorus contents were also determined on these bihourly RS samples. These values are also plotted in Figure 16. A comparison of the RS phosphorus and iron plots reveals that the phosphorus did not accumulate in the RS to the same extent that the iron accumulated. It was anticipated that the accumulation of iron in EP-RS would increase its ash content, and the effect of the increased ash would be to reduce nitrogen content of the RS. Each day, one liter sample of WP and EP-RS were centrifuged and the concentrated solids dried at 10H°C for 2k hours. Ash and Total Kjeldahl Nitrogen (AOAC-2.0U2 and 2.0U3, reference 36) analyses were performed on the dried material. In Figures 17 and 18 are plotted the results of these daily analyses. These figures show that as iron addition continued the EP-RS ash content progressively increased as the iron was added, and the EP-RS ash content was much higher than that of the WP-RS. When iron addition was stopped on ll/lU/69 the EP-RS ash content decreased as the accumulated iron was removed from the system as waste sludge. 70 ------- '• -IRON ADDITION - 18 MG/U -NO IRON- -IRON ADDITION- 75 MO/L -NO IRON NMNMNMNMNMNMNMNM N M15 MN5NMNBN~M 3 !j £ J 1 1 .| M/J MON | MM Tue | ,,/5 WED. | ,1/6 THUR | M/7 FR, | M/e SAT | ,,/=, SUN | M/,0 MON | „/,, TUE | M/B WED | ,,/,, THUR | M/,, FR, | ,,/B SAT | „/,. SUN | M/,7 MON | M/B TUES | M/,9 WED | M/20THUR | ,1/3, |FR, | FIGURE 16 ACCUMULATION OF IRON IN EP-RS 1969 ------- 30 29 to < CD cc 0 28 CO cc I (O 27 26 25 24 LEGEND —O WEST PLANT £t A EAST PLANT IRON ADDITION: I5MG/L; 7AM. 11/3 TO 3:30 PM. 11/7 75 MG/L; 7A.M. 11/10 TO 1:05PM 11/14 IRON 15 MG/L IRON 75 MG/L 16 i — r— i — r— i — i i i — i IO/27 i — i— i — i -II/I SU 11/2- 11/8 SU II/9-II/I5 SU 11/16-11/20 Su FIGURE 17 RETURN SLUDGE ASH CONTENT (DRY BASIS) 1969 72 ------- 6.7 6.6 GO CO < CD CC CO oL e> o cc 6.4 63 e, 6.1 LEGEND O OWEST PLANT A A EAST PLANT IRON ADDITION: 15 MG/L; 7A.M. 11/3 TO 3:30PM 11/9 75MG/LJ7AM II/IOTO 1:05PM. 11/14 IRON 75 MG/L 16 ' ,0/27- 111/1 SU 11/2-I 1/8 SU 11/9-11/15 SU 11/16-11/20 SU FIGURE 18 RETURN SLUDGE NITROGEN CONTENT (DRY BASIS) 1969 73 ------- The effect of the increased ash content vas to reduce the EP-RS nitrogen content as shown in Figure 18. The magnitude of the effect of ash content on the nitrogen content of this vaste sludge can be obtained by a comparison of the average values of ash and nitrogen contents of RS during the iron addition periods as shown in Table 2k. 74 ------- TABLE 2U AVERAGE RETURN SLUDOE COMPOSITION Iron Addition % Ash * Nitrogen EP Difference WP EP Difference 15 mg/L 11/3 to 11/7 25.9 29.1 3.2 6.U9 6.27 0.21 % N based on VSS (8.76) (8.85) (0.09) 7.5 11/10 to ll/U 25.2 28.8 3.6 6.59 6.32 0.27 % * based on VSS (8.82) (8.87) (0.05) 75 ------- SECTION VII ACKNOWLEDGEMENTS This report was prepared by Raymond D. Leary, Chief Engineer and General Manager; Lawrence A. Ernest, Director of Laboratory; Roland S. Powell, Assistant Director of Laboratory; and Lawrence H. Docta, Research Supervisor of the Sewerage Commission of the City of Milwaukee. We acknowledge the assistance given by Mr. H. W. Boston and Mr. C. Risley and the staff of the Chicago office of the Environmental Protection Agency. Our acknowledgements are due to Dr. R. N. Kinman and Dr. R. L. Bunch whose guidance from time to time as project officers led to the completion of this project. The Commission Staff members wish to express their gratitude to the Marquette University Staff who conducted the statistical and X-Ray diffraction studies under the guidance of the project consultants Dr. Raymond J. Kipp and Dr. Sudershan K. Malhotra. The Commission commends the A. 0. Smith Corp. (Milw., Wis.) for their cooperation and contribution of the waste pickle liquor for the iron addition studies as well as for the assistance and cooperation of Mr. Milton Johnson and other A. 0. Smith personnel in making that study successful. The assistance of Laboratory Technicians, Miss Gloria Aldenhoff, Mrs. Minne Ness, Miss Elizabeth Merscher, and Mr. Gerald Hertzfeldt in conducting the laboratory analyses is gratefully acknowledged. 77 ------- SECTION VIII REFERENCES 1. Unpublished data. Sewerage Commission of the City of Milwaukee Laboratory, Wisconsin. 2. Connell, C. H., and D. Vacker, "Parameters of Phosphate Removal by Activated Sludge," Proceedings of the 7th Industrial Water and Waste Water Conference, Univ. of Texas, Austin, (June 1 & 2, 1967). 3. Witherow, Jack, L., "Phosphate Removal by Activated Sludge," A report by the U.S. Depart, of Interior, FWPCA, Robert S. Kerr Water Research Center, Ada, Oklahoma (May 1969). U. Menar, A. B., and D. Jenkins, "The Mechanism of Enhanced Phosphate Removal in the Activated Sludge Process," Part II, SERL Report 68-6, Univ. of California, Berkeley, (Aug., 1968). 5. Levin, G. V., and J. Shapiro, "Metabolic Uptake of Phosphorus by Waste Water Organisms," J. Water Pollution Control Federation, 37_,(6),800 (1965). 6. Vacker, D., C. H. Connell, and W. N. Wells, "Phosphate Removal Through Municipal Waste Water Treatment at San Antonio, Texas," J. Water Pollution Control Federation, 32,(5),750 (1967). 7. Owen, R., "Removal of Phosphorus from Sewage Plant Effluents with Lime," Sew. and Ind. Wastes, 25., (5), 5**8 (1953). 8. Stone, T., "Iron and Phosphate Changes during Sewage Treatment," Sew. and Ind. Wastes, 31, (8), 98l (1959). 9* Hurwitz, E., R. Beaudoin, and W. Waters, "Phosphates, Their Fate in a Sewage Treatment Plant-Waterway System," Water and Sew. Works, 112. (3), 8U (1965). 10. Lea, W. L., G. A. Rohlich, and W. J. Katz, "Removal of Phosphates from Treated Sewage," Sew. and Ind. Wastes, £6, (3), 26l (195*0. 11. Rohlich, G. A., "Methods for the Removal of Phosphorus from Sewage Plant Effluents," Inter. J. of Air and Water Pollution, J_ U27 (1963). 79 ------- 12. Malhotra, S. K. , G. F. Lee, and G. A. Rohlich, "Nutrient Removal from Secondary Effluent "by Alum Flocculation and Lime Precipi- tation," Inter. J. of Mr and Water Pollution, £, 1*87 (196U). 13. Nesbitt, J. B., "Removal of Phosphorus from Municipal Sewage Plant Effluents," Eng. Research Bulletin B-93, Penna. State Univ., College of Eng., University-Park, Penna., (1966). lit. Tenney, M. W. , and W. Stumm, "Chemical Flocculation of Micro- organisms in Biological Waste Treatment," J. Water Pollution Control Federation, 37.. (10), 1371 (1965). 15. Barth, E. F. , and M. B. Ettinger, "Mineral Controlled Phosphorus Removal in the Activated Sludge Process," J. Water Pollution Control Federation, 3£, (8), 1362 (1967). 16. Eberhardt, W. A. and J. B. Nesbitt, "Chemical Precipitation of Phosphorus in a High Rate Activated Sludge System," J. Water Pollution Control Federation, jj£, (7), 1239 (1968). 17. Rudolf, W. , "Phosphates in Sewage and Sludge Treatment. II. Effect on Coagulation, Clarification and Sludge Volume*', Sewage Works J., 19., (2), 178 (19^7). 18. Neil, J. H., "Problems and Control of Unnatural Fertilization of Lake Waters," Proceedings of the 12th Ind. Wastes Conf . , Purdue Univ., (May 13-15, 1957). 19. Schmid, L. A. , and R. E. McKinney, "Phosphate Removal by Lime- Biological Treatment Scheme," J. Water Pollution Control Feder- ation, la, (7), 1259 (1969). 20. Sawyer, C. N. , "Biological Engineering in Sewage Treatment," Sewage Works J. , l£, (9), 925 21. Sekikawa, Y. , S. Nishikawa, M. Okazaki, and K. Kato, "Release of Soluble Phosphates in the Activated Sludge Process," 3rd International Conference on Water Pollution Research, Munich, Germany, (Sept., 1966). 22. Hall, M. W., and K. Engelbrecht, "Uptake of Soluble Phosphate by Activated Sludge; Parameters of Influence," Proceedings of the 7th Industrial Water and Waste Water Conference, Univ. of Texas, Austin, p. II, 8, (1967). 80 ------- 23. Borchardt, J. A., and H. S. Azad, "Biological Extraction of Nutrients," J. Water Pollution Control Federation, UP, (10), 1739 (1968). 2U. Srinath, E. G., C. A. Sastry, and S. C. Pillai, "Rapid Removal of Phosphorus from Sewage by Activated Sludge," Experientla, 15, 9 (1959). 25. Alarcon, G. 0., "Removal of Phosphorus from Sewage," Masters Essay, The John Hopkins Univ., Baltimore, Maryland, (l96l). 26. Feng, T. H., "Phosphorus and the Activated Sludge Process," Water and Sewage Works, 109* (ll), ^31 (1962). 27. Campbell, L. A., "The Role of Phosphate in the Activated Sludge Process," Proceedings of the 21st, Purdue Industrial Waste Conference, (May, 1966). 28. Leary, R. D., and L. A. Ernest, "Industrial and Domestic Waste- water Control in the Milwaukee Metropolitan District," J. Water Pollution Control Federation, 32. (7), 1223 (1967). 29. "Instructions For YSI Model 53 Biological Oxygen Monitor," Yellow Springs Instrument Co. Inc., Yellow Springs, Ohio. 30, "Instruction Manual Hach C R Surface Scatter Turbidimeters Model 1889," Hach Chemical Company, Ames, Iowa. 31. "Instruction Manual Hach Laboratory Turbidimeter Model 2100," Hach Chemical Company, Ames, Iowa. 32. "Procedure Manual For Atomic Absorption Spectrophotometry," Instrumentation Laboratory Inc., Lexington, Mass. 33. "Standard Methods for the Examination of Water and Wastewater." 12th Ed. Amer. Pub. Health Assn., New York (1965). 3U. Goodman, B. L., and J. W. Foster, "Notes On Activated Sludge ."2nd Edition, Smith and Loveless Division of Union Tank Car Company, Lenexa, Kansas. 35. Kipp, R. J., "Statistical Analysis of Phosphate Removal Data for 1968, 1969; Jones Island Sewage Treatment Plant, Milwaukee, Wisconsin," Marquette Univ. (May 1970). 81 ------- 36. "Official Methods of Analysis of the Association of Official Agricultural Chemists." 10th Ed., Aasn. of Official Agricultural Chemists, Washington, D.C."Tl965). 37. Eckenfelder, Jr., W. W., and D. J. Q'Connor."Biological Waste Treatment." Pergamon Press, New York, New York (1961). 38. Okun, D. A., and W. R. Lynn, "Preliminary Investigations^into the Effect of Oxygen Tension on Biological Sewage Treatment, in J. McCabe and W. W. Eckenfelder Jr., "Biological Treatment of Sewage and Industrial Wastes." Volume I, Rheinhold, New York, New York (1956). 39. Ford, D. L., J. T. Young, and W. W. Eckenfelder Jr., "Dehydro- genase Enzyme as a Parameter of Activated Sludge Activities, Proceedings of the 21st_ Ind. Waste Conf., Purdue Univ., Part I, (May 3 to 5, 1966). HO. Natrella, M. G., "Experimental Statistics".NBS Handbook No. 91, U.S. Government Printing Office, Wash., D.C. (1963). Ul. Jones, P.H., and D. Prasad, "The Use of Tetrazolium Salts as a Measure of Sludge Activity," J. Water Pollution Control Feder- ation, Ul, (11, Part 2), R M»l (1969). U2. Gopalakrishna, E. M., J. Winters, and L. Gartz "Report on the X-Ray Analysis of Sludge Material," Marquette, Univ. (June 1969). H3. Natarajan, Dr., M. Seitz. J. Winters, and R. J. Riedner, "Sludge X-Ray Analysis Report," Marquette Univ. (December 1969). UU. Seitz, M. A., R. Riedner, and J. Winters, "X-Ray Diffraction Studies of Sewage Sludge Residue," Marquette Univ. (March 1970). U5. Chang, S. C., and M. L. Jackson, "Solubility Product of Iron Phosphate," Soil Science of America Proceedings, 21, (3), 265 (1957). U6. Galal-Grochev, H., and W. Stumm, "The Reaction of Ferric Iron with Ortho-Phosphate," J. Inorg. Nucl, Chem., 2£, 576 U9&3). 82 ------- SECTION IX PHOSPHORUS NOMENCLATURE AND ABBREVIATION'S GLOSSARY PHOSPHORUS NOMENCLATURE 1. Total Phosphorus (TP). All the phosphorus present in sample (whether in the soluble or insoluble state and present as ortho, poly, organic, etc., phosphorus compounds) which is converted by ternary acid digestion to soluble ortho-phosphate. 2. Soluble Ortho - Phosphate (SOP). All phosphorus measured by direct colorimetric analysis of sample filtrate. (Angel Reeve Glass Fiber Pad No. 93UAB). 3. Total Soluble Phosphorus (TSP). All the phosphorus compounds in the sample filtrate converted by ternary acid digestion to ortho-phosphate. 1*. Suspended Solids Phosphorus (SS-P). Represents the phosphorus present in the sample suspended solids. (SS-P = TP - TSP). ABBREVIATIONS GLOSSARY 1. BOD - five day biochemical oxygen demand. 2. BOD/TSP - ratio of BOD removed/day to TSP removed/day. 3. CV - coefficient of variation (standard deviation divided by the average, multiplied by 100). U. CFM - cubic feet per minute. 5. COD - chemical oxygen demand. 6. DO - dissolved oxygen. T. EP - East Plant. 8. EPML - East Plant mixed liquor. 9. EPRS - East Plant return sludge. 10. EPWS - East Plant waste sludge. 11. F/M - (Ratio of food to microorganisms) /Day. 83 ------- iEBOD/DAY #MLVSS in the Aeration Capacity 12. MOD - million gallons/day. 13. ML - mixed liquor. lU. ML-DO - mixed liquor dissolved oxygen. 15. ML-SDI - mixed liquor sludge density index. 16. MLSS - mixed liquor suspended solids. 17. MLSS-0- - mixed liquor oxygen uptake rate. mg 02/mg VSS/hour). 18. MLVSS - mixed liquor volatile suspended solids. 19. N - nitrogen. 20. P - phosphorus. 21. RS - return sludge. 22. RSSS - return sludge suspended solids. 23. SOP - soluble ortho-phosphate. 2k. SS - suspended solids. 25. TP - total phosphorus. 26. TSP - total soluble phosphorus 27. TP/M - (ratio of total phosphorus to Microorganisms)/Day, jfTP/DAY #MLVSS in the Aeration Capacity 28. MM TPF/5 mis of ML - micro-moles of triphenylformazin/ 5mls of mixed liquor. 29. VSS - volatile suspended solids. 30. WP - West Plant. 31. WS - vaste sludge. 84 ------- SECTION X APPENDIX A PHOSPHORUS DETERMINATION WITH TECHNICON AUTOANALYZER Sample Preparation A« Total Phosphorus 1. Pipette unfiltered sample into a 100 ml. volumetric flask (20 mis effluent, 5 mis for sewage). 2. Add 5 mis of Ternary Acid Mixture and 3 glass "beads. 3. Heat on hot plate to dense fumes of perchloric acid, plus 5 minutes, cool. U. Add 20 mis of distilled water, bring to a boil, boil 5 minutes, cool. 5. Add 1 drop of phenolphthalein. neutralize vith 10 NaOH to a faint pink color. 6. Just discharge pink color with INHgSO^, dilute to 100 ml, mix. 7. Place solution from step 6 in the sampling cup of the aut oanaly zer . 8. Obtain the phosphorus concentration of the sample from the standard curve. B. Total Soluble Phosphorus 1. Same as total phosphorus, except the aliquot is filtered through an Angel Reeves glass fiber pad 93UAH. C. Soluble Ortho-Phosphate 1. Filter through Angel Reeves glass fiber pad 93UAH. 2. Dilute filtrate if needed. 3 Place in sampling cup of the autoanalyzer. 85 ------- Reagents A. Afflmonium Molybdate - Dissolve 200 gm of (MHj,)/- Mo7 0 mmonum oyae - jg ? . .g in 10 liters of distilled vater. Add 1680 ml of con57H2soU and dilute to 20 liters. B. ANSA Stock Solution - Dissolve 219 gm Na2S20 and 8 gm Na2 SO- in 700 ml of distilled water (temperature <50° C), add 1* gm of 1-amino- 2-naphthol -U- sulfonic acid. Dilute to 2 liters. For daily use, prepare a 1:10 dilution. C. Phosphorus Standard Curve use standards, both digested and undigested from 0.1 to 1.2 mg/L-P in increments of 0.1 mg/L-P from a 1000 mg/L-P stock (U.386 gm of oven dried at 103°C for U hours, in one liter). D. Ternary Acid Mixture - Add 100 mis of 96% HgSO^ to 500 mis of 1Q% HK03, mix. Add 200 mis 70? HCIO^, mix and cool. 86 ------- oo SAMPLER II RATE.40_pERHR. 1:2 CAM WATER RINSE EVERY 4tb SAMPLE TO WELL O TUBE SIZE (INCHES) 0.090 WATER QO73 ANSA FIGURE 19 TECHNICON AUTOANALYZER SCHEMATIC ------- APPENDIX B PROCEDURE USED FOR MLSS AND RSSS DETERMINATION: A. Determination of SS concentration by weight. 1. Mix sample veil, pour (100 mis for ML., 50 mis for RS) into a 100 ml graduate cylinder. 2. Add 5J& chlorhydrol solution (6 drops for ML and 12 drops for RS) mix 3 times by inversion, let sit for 5 minutes. 3. Filter under vacuum thru tared filter paper (9cm, S and S Sharkskin for ML, Whatman No. 3 for RS) placed on a buechner funnel. U. Dry the paper for 1 hour at 103° C in a Forced Draft Oven. 5. Cool it for 5 minutes in a dessicator and weigh back. 6. For MLSS subtract from this weight the tare weight of the paper. This gives solids in units of grams which in turn is equivalent to per cent by weight. (For RSSS since we used 50 ml sample, therefore multiply the above weight by a factor of two to get per cent RSSS by weight). B. Determination of MLVSS 1. Place the filter paper with MLSS (use low ash S and S filter paper) after step 5 in a previously ashed and tared silica dish, and then ash for 15 minutes at 600°C. 2. Cool for 30 minutes in a dessicator and then weigh back. 3. The MLSS less the MLSS ash divided by the MLSS when multiplied by 100 equals per cent MLVSS. 88 ------- APPENDIX C BOD DETERMINATION Dilution Water 1. Add 1 ml of the following solutions (33) to each liter of aged distilled water, 5 or more days. Phosphate Buffer Solution Magnesium Sulfate Solution Calcium Chloride Solution Ferric Chloride Solution 2. Aerate this mixture for 5 minutes. Meter Setting 1. Place probe in aerating distilled vater for 5 minutes. 2. Check zero and adjust if necessary. 3. Check red line and adjust if necessary. 1*. Determine barometric pressure. 5. Read temperature on meter. 6. Determine DO setting vith pressure-temperature chart. 7. Adjust calibration for DO of the day. Procedure 1. Discharge stale water in buffer line. 2. Siphon dilution water into BOD bottle for blank. 3. Place probe in BOD bottle and read DO when meter stabilizes. U. Stopper with rubber seal and incubate for 5 days, dilution vater blank. 5. Determine dilutions to be made (set of 2) for samples. 6. Siphon dilution water to cylinder, filling to required mark. 7. Add sample desired with pipette or siphon. 8. Mix well with plunger and siphon to BOD bottles. 9. Determine DO, stopper and seal, and incubate 5 days. 10. After 5 days remove from incubation and determine DO as before, 89 ------- Calculations 1. 2. Initial DO - 5 day Conversion factors 5 day BOD, dividing x dilution factor = 5 day BOD of U and 6 day sevage and effluent BO] by appropriate factor listed below. U day Sewage .875 Effluent .796 parts Sample 1 1 1 1 1 1 1 1 parts dilution H20 + 1 * 3 + 9 + 19 + 39 + 1*9 + 79 + 99 DILUTION TABLE ml dilution H20 150 ml 225 " 270 " 285 292.5 " 204 296.25 " 297 " 6 day 1.076 1.231 add ml of Sample 150 ml 75 30 15 7.5 " 6 3.75 " 3 90 ------- APPENDIX D DISCUSSION OF MATERIAL FOUND FLOATING ON THE SURFACE OF THE EP AERATION TANKS AND THE AEROBIC DIGESTION OF WASTE SLUDGE. The principal microorganism found in this material floating on the surface of the E.P. aeration tanks was identified as belonging to Actinomycetaceae and to the Genus Nocardia. A sample of this material was found to contain 85? volatile matter and 31% hexane solu- bie materials. Regular defoaming agents as mentioned before were ineffective in breaking this foam. Vacuum skimming of aeration tanks and clarifier feed channels reduced the amount of foam and aided in overcoming this problem. This floating material did not appear during all this time in the WP. The effect of extended aeration on the stability of the Actinomycetaceae foam was also investigated. The East Plant Reserve Return Sludge channel was filled with approximately 286,000 gallons of waste sludge. This waste sludge was vigorously aerated for 21 days (3-25-69 to lt-15-69). The air application rate could not be defined because the air supply to this RS channel was not metered. At no time during this lengthy aeration period was a decrease in the quantity or the density of the Acti nomy cetaceae foam observed. During this study it was also decided to observe the effect of extended aeration on the aerobic digestion of the waste sludge containing this floating matter. Seven liter composite samples were taken out of the aerated sludge for several days during this experiment and analyzed for TP,TSP, SOP, S3, VSS, pH and total Nitrogen. The results of these analyses are presented in Table 25. Some of the variation in the data was caused by sampling and analytical errors. Further, a part of the reduction of SS and TP in the aerated waste sludge was due to the dilution caused by some waste water entering the reserve RS channel. The following general observations were made regarding the aerobic digestion of sludge in the reserve RS channel. 1. There appeared to be an acclimatization period of approxi- mately 2 to 3 days before the aerobic digestion of the solids began. 2. After the acclimatization period, aerobic digestion of the sludge solids was significant as indicated by a sudden increase in the soluble phosphorus and a decrease in the $VSS and % nitrogen in the waste sludge solids. 91 ------- TABLE 25 AEROBIC DIGESTION OF SLUDGE Date 1969 3-25 3-27 3-28 3-31 U-2 U-3 H-7 l*-9 1*-11 U-lU U-15 Day Tu Th F M W Th M W F M Tu * Phosphorus 0 2 3 6 8 9 13 15 17 20 21 TP - 280 270 266 2U6 258 238 226 239 220 205 TSP - 2.1* 3.1* - 25.U 32.8 1*8.6 51.6 58.3 56.0 53.1 SOP - 2.1 2.8 1U.6 lU.7 32.5 1*6.9 1*9.1 53.1* 53.6 52.9 Sludge Solids PH in % Sludge MG/L VSS (Dry) - 10,000 9,1*30 8,31*0 7,890 7,180 5,910 6,300 5,560 5,010 U.350 - 61*. 5 63.8 56.6 59.3 56.U 55.3 56.2 55.7 53.5 - - 7.1 7.3 7.2 7.0 7.1 7.1 7.0 7.0 7.1 - - 6.61* - 6.70 - 5.95 5.86 - 6.05 - 5.90 in Sludge (Dry) - 2.78 2.83 3.13 2.80 3.11* 3.20 2.78 3.25 3.27 3.1*9 "Number of days aerated. 92 ------- 3. In a period of 6 to 8 days over k$% of the volatile matter in the waste sludge solids was reduced and after that the % reduction of volatile matter did not increase significantly. H. During aerobic digestion of sludge solids, the solids appeared to hold the remaining phosphorus very strongly. This was shown by the increase in the % phosphorus in the solids on a dry basis as the digestion progressed. 5. Aerobic digestion of sludge solids had no effect on the pH value of the sludge. It was observed that similar floating material started to reappear in November (1969) in small quantities and on limited occasions in both plants when no changes in loading or other process parameters were made. This foam then continued to build up to the extent where it started affecting the operation of both the plants. The cause or causes of the appearance of this floating material and the controlling of this material by process parameters changes is not yet fully known but we did seem to have limited success in overcoming this floating material by increasing F/M loadings. 93 ------- APPENDIX E 1967 1968 1969 SCREENED SEWAGE CHARACTERISTICS Flow BOD SS TP TSP SOP % of TP in SS PLANT OPERATION Plow BOD SS TP TSP SOP mgd mg/L mg/L ng/L mg/L mg/L mgd mg/L mg/L mg/L mg/L mg/L DETENTION TIME UNDER AIR HOURS Air Applied Cu. ft Food/Microorganism MLSS MG/L < RS ./gal. Sewage Ratio 183. 297 301* 8. ... ... — WEST 71*. 5 15.1 22 1.2 ... — 6.8 l.Uo O.U25 2800 28.0 i* u EAST 109.0 18.2 25 1.3 ... ... 6.6 1.22 0.1*53 2700 25. U 183.1* 306 311* 9.6 ... 2.8 ... WEST 76.3 26.8 1*7 2.2 — 0.87 6.8 1.37 O.U25 2900 28.2 EAST 107.2 19.2 30 1.7 ... 0.83 6.8 1.21 0.1*1*0 2800 25. U 181.6 239 227 81 .1* 3.5 2.3 57.2 WEST 76.1* 21.3 1*1 2.0 0.99 0.86 6.6 1.36 0.391 2600 27.5 EAST 105.2 1U.8 23 1.1* _ rt /* 0,86 0.73 6.8 1.20 0.388 2600 25.1* Lbs. BOD per 1000 cu. ft. of Aeration Tank Capacity 58.1 60.2 59. 60.5 1*7.8 1*7.5 ------- APPENDIX F TOTAL PHOSPHORUS REMOVAL AT THE JONES ISLAND PLANTS BASED ON PLANT FLOWS The values used in the calculations are yearly averages. Flow MOD TP mg/L-P TP pounds Lant Effluent Flow MOD TP mg/L-P TP pounds Dtal TP pounds Removal 183. U 8. It 12.8U8 W E 7U.5 109.0 1.2 1.3 7U6 1182 1928 85.0 183. U 9.6 1H.68H W E 76.0 107.2 2.2 1.7 ll*00 1520 2920 80.1 181.6 8.U 12,722 W E 76. U 105. 2.0 1. 127H 1228 2502 80.3 2 U W » West Plant E = East Plant 7350 81.7 ------- Accession Number w Subject Field & Group 05 SELECTED WATER RESOURCES ABSTRACTS INPUT TRANSACTION FORM Organization Sewerage Commission of the City of Milwaukee, Milwaukee, Wisconsin Title Phosphorus Removal By An Activated Sludge Plant 10 Authors) Leary, R. D. Ernest, L. A. Powell, R. S. Docta, L. H. Malhotra, S. K. Kipp, R. J. Project Designation pj.ogram # IJQIO DXD Grant # WPD 188-01-67, 188-02-68, 188-03-69 21 Note 22 Citation Water Resources Research Catalog. Vol. 3, Dec. 1967, p 585, Abstract 5.1362 23 Descriptors (Starred First) •Phosphorus removal, * Activated Sludge Process, Process parameters, Wastewater treatment, •Biological treatment. 25 Identifiers (Starred First) 27 Abstract jo« tb. actirtea sludge plants elsewhere in this country. A detailed plant- -rs sLss^sr^s^-at vithout «r apparent detri^ntal e«ect on the process or equipment. Abstractor J. Kil Institution Marquette University, Milwaukee, Wisconsin WR:102 (REV. JULY 1969) WRSIC "SEND, WITH COPY OF DOCUMENT, ' U.S. DEPARTMENT OF THE INTERIOR WASHINGTON. D. C. 2024O * GPO: 197U-389-930 ------- |