Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Laboratory study on aerated stabilization basin operation at 3°C 1973

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1973_A7 A88_3.pdf [ 4.85MB ]
UBC_1973_A7 A88_3.pdf
UBC_1973_A7 A88_3.pdf
Metadata
JSON: 1.0050532.json
JSON-LD: 1.0050532+ld.json
RDF/XML (Pretty): 1.0050532.xml
RDF/JSON: 1.0050532+rdf.json
Turtle: 1.0050532+rdf-turtle.txt
N-Triples: 1.0050532+rdf-ntriples.txt
Citation
1.0050532.ris

Full Text

A LABORATORY STUDY ON AERATED STABILIZATION BASIN OPERATION AT 3°C by JAMES WESLEY ATWATER B.A.Sc, University of Bri t i s h Columbia, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in the Department of C i v i l Engineering We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1973 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d " t h a t c o p y i n g or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a llowed w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f C i v i l Engineering The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Date October 9, 1973 A B S T R A C T Aerated stabilization basins (ASB), l i k e many other biological treat- ment systems, demonstrate a temperature dependency. A decrease in treatment efficiency usually results from a decreasing basin temperature and has often been related to a decrease in the reaction rate coefficient, K. This relation- ship to the reaction rate may well apply for other treatment systems, but i t has not been clearly demonstrated for aerated stabilization basins. This study develops data on steady-state performance at 3 C in order to present a coherent reference point for future ASB temperature studies and to define performance characteristics at 3°C. The following performance c r i - teria were documented in the study: 1. Substrate removal in terms of fi l t e r e d sub- strate removal. (61 - 80 per cent COD removal and 76 - 98 per cent BOD̂  removal for retention times of 1 - 16 days). 2. System treatment efficiency defined in terms of gross effluent COD and B0D5. (23 - 50 per cent COD removal and 18 - 80 per cent BOD̂  removal for retention times of 1 - 16 days). 3. Net biological solids production (0.25 lbs/lb BOD or COD used). 4. Oxygen u t i l i z a t i o n requirements (0.123 lbs O2/ lb COD removed and 0.143 lbs 02/B0D^ removed for retention times of 2 - 16 days. Endogenous respiration - 0.75 mg/hr/ gm MLSS). 5. Nitrogen transformation. (A transformation of Kjeldahl nitrogen in the biological solids to NH^ nitro- gen in the f i l t r a t e was found apparently as a function of retention time). i i i i i 6.. Post Settling. (One day's aeration with one day settling was found to give equivalent treatment as eight days aeration and one day settling). Data was obtained at two loadings to provide information on the influence of influent concentration on overall performance. Established i n the experiment was that any of the common mathema- t i c a l models used to describe ASB operation, McKinney's, Eckenfelder's, or first-order exponential, could predict system treatment efficiency at 3°C for retention times beyond two to four days. It was further shown that only the Chemostat model would describe the substrate removal measured in the study. TABLE OF CONTENTS Page LIST OF TABLES v i i LIST OF FIGURES v i i i LIST OF TERMS x CHAPTER I. INTRODUCTION 1 1.1 GENERAL. 1 1.2 RESEARCH OBJECTIVES 2 II. LITERATURE SURVEY 3 II. 1 GENERAL 3 11.2 DESIGN FORMULATIONS FOR SUBSTRATE REMOVAL IN ASBs . . . 3 11.3 TEMPERATURE COMPENSATION IN DESIGN FORMULATION 8 11.4 TREATMENT EFFICIENCY. . . 10 11.5 SOLIDS PRODUCTION AND SETTLING 11 11.6 NUTRIENT REQUIREMENTS . . . . . 12 III. RESEARCH METHODOLOGY 14 111.1 RATIONALE 14 111.2 GENERAL PROCEDURE 17 111.3 EQUIPMENT AND FLOW . . . 18 111.4 SUBSTRATE. . . 20 111.5 ANALYTICAL PROCEDURES 23 iv V CHAPTER .Page IV. RESULTS AND DISCUSSION 26 IV. 1 GENERAL 26 IV.2 CRITERIA FOR STEADY STATE OPERATION 26 IV.2.1 Low Loading Study 26 IV.2.2 High Loading Study 27 IV.3 PER CENT SUBSTRATE REMOVAL AND SYSTEM TREATMENT EFFICIENCY. 32 IV.3.1 Per Cent Substrate Removal (Substrate Uti l i z a t i o n ) . . . . 32 IV.3.2 System Treatment Efficiency 34 IV.4 EVALUATION OF MATHEMATICAL MODELS USED IN ASB DESIGN. . 36 IV.4.1 General 36 IV.4.2 Evaluation of O'Connor and Eckenfelder's Models . . . 36 IV.4.3 McKinney's Model. 41 IV.4.4 Chemostat 44 IV,A.5 First-Order Exponential 46 IV.5 ASB SOLIDS 50 IV.5.1 Solids Production 50 IV.5.2 COD - B0D5 of ASB Solids 53 IV.5.3 Settling at 3°C 56 IV.6 NITROGEN STUDIES 61 IV.6.1 General 61 IV.6.2 Nitrate Nitrogen 61 IV.6.3 Nitrogen Balance 62 IV. 7 pH 65 IV.8 OXYGEN UTILIZATION 65 V. SUMMARY 70 V.l STEADY STATE ; . . . . 70 V.2 PER CENT SUBSTRATE REMOVAL AND SYSTEM TREATMENT EFFICIENCY 70 V.3 MODEL EVALUATION 71 V.4 SOLIDS PRODUCTION 71 V.5 NITROGEN USAGE 72 V.6 pH -; 72 V.7 OXYGEN UPTAKE 72 VI. CONCLUSIONS 74 VII. RECOMMENDAT IONS 77 v i Page BIBLIOGRAPHY .79 APPENDIX A EXPERIMENTAL DATA 84 APPENDIX B TEST DATA PERTAINING TO THE DETERMINATION OF .STEADY STATE OPERATION - LOW LOADING 98 APPENDIX C CALCULATION OF CONSTANTS AND BOD5 CONCENTRATIONS - McKINNEY' S MODEL .100 LIST OF TABLES Table Page 1. TEMPERATURE COEFFICIENTS FOR BIOLOGICAL TREATMENT SYSTEMS 9 2. REACTION HYDRAULIC RETENTION TIME . 20 3. ANALYSIS OF POWDERED MILK WASTE 21 4. GREATER VANCOUVER WATER DISTRICT PHYSICAL AND CHEMICAL ANALYSIS OF WATER SUPPLIES 22 5. MEASURED AND CALCULATED SYSTEM TREATMENT EFFICIENCIES, O'CONNOR AND ECKENFELDER'S MODEL 40 6 . MEASURED AND CALCULATED SUBSTRATE CONCENTRATIONS, McKINNEY'S MODEL 42 7. MEASURED AND CALCULATED GROSS EFFLUENT BOD5 CONCENTRATIONS,' McKINNEY'S MODEL • 43 8. MEASURED AND CALCULATED SYSTEM TREATMENT EFFICIENCY, FIRST-ORDER EXPONENTIAL '49 9. NET SOLIDS PRODUCTION PER POUND SUBSTRATE REMOVED 52 10. EFFLUENT CHARACTERISTICS OF ASBs AT 3°C 54 11. GROSS KJELDAHL NITROGEN CONCENTRATIONS 63 12. AVERAGE NITROGEN CONCENTRATIONS — LOW LOADING 63 v i i LIST OF FIGURES Figure Page 1. SCHEMATIC OF MODEL ASB .18 2. STEADY STATE - 16 DAY REACTOR - LOW LOADING 28 3. CYCLIC FLUCTUATIONS OF FILTERED COD CONCENTRATIONS 29 4. HYDRAULIC EQUILIBRIUM - HIGH LOADING 30 5. STEADY STATE - SYSTEM TREATMENT EFFICIENCY COD - HIGH LOADING.31 6. PER CENT SUBSTRATE REMOVAL AS A FUNCTION OF MEAN HYDRAULIC RETENTION TIME AT 3°C 33 7. SYSTEM TREATMENT EFFICIENCY AS A FUNCTION OF MEAN HYDRAULIC RETENTION TIME 35 8. EVALUATION OF REACTION RATE CONSTANT (K) FOR SUBSTRATE REMOVAL (Se) (O'CONNOR AND ECKENFELDER) 37 9. EVALUATION OF REACTION RATE CONSTANT (K) FOR SYSTEM • TREATMENT EFFICIENCY AT 3°C (O'CONNOR AND ECKENFELDER). . . . 39 10. CHEMOSTAT FIT - COD SUBSTRATE CONCENTRATION 45 "11. EVALUATION OF REACTION RATE CONSTANT K FOR SUBSTRATE REMOVAL AT 3°C 47 12. EVALUATION OF REACTION RATE CONSTANT K FOR SYSTEM TREATMENT EFFICIENCY 48 13. MIXED LIQUOR SUSPENDED SOLIDS AS A FUNCTION OF MEAN HYDRAULIC RETENTION TIME AT 3°C 51 14. PER CENT COD REMOVAL WITH SETTLING TIME - 3°C - LOW LOADING . 57 15. PER CENT COD REMOVAL WITH SETTLING TIME - HIGH LOADING. . . . 58 16. SUPERNATANT MLSS vs. SETTLING TIME - 3°C - HIGH LOADING . . . 60 17. CONCENTRATIONS OF NITROGEN COMPOUNDS IN THE REACTOR SOLIDS AND FILTRATE AS A FUNCTION OF MEAN HYDRAULIC RETENTION TIME HIGH LOADING 64 v i i i ix / • Figure Page 18. OXYGEN UPTAKE AGAINST RETENTION TIME 66 19. OXYGEN CONSUMPTION PER DAY AS A FUNCTION OF SUBSTRATE REMOVED PER DAY - HIGH LOADING 68 20. OXYGEN UPTAKE RATE AS A FUNCTION OF REACTOR SUBSTRATE CONCENTRATION 69 LIST OF TERMS ASB - Aerated S t a b i l i z a t i o n Basin BOD, - 5 day Biochemical Oxygen Demand BOD - Ultimate Biochemical Oxygen Demand COD - Chemical Oxygen Demand MEAN HYDRAULIC RETENTION TIME Theoretical retention time of any given particle; equal to the re- actor volume divided by the mean flow rate. MLSS - Mixed Liquor Suspended Solids NUTRIENTS Elements in addition to carbon necessary for b i o l o g i c a l growth; usually refers, but not restricted, to nitrogen arid phosphorus. PER CENT SUBSTRATE REMOVAL Percentage decrease in applied substrate measured between the influent and the effluent. STEADY STATE Condition existing in the reactors when there is hydraulic and b i o l o g i c a l equilibrium. SYSTEM TREATMENT EFFICIENCY Per cent reduction in COD or BOD5 measured be- tween the influent and effluent; r e f l e c t s the COD or BODr of generated solids in the effluent. vss - Volatile Suspended Solids x A C K N O W L E D G M E N T The author wishes to express his thanks and appreciation to his supervisor, Dr. A. H. Benedict, for his guidance, enthusiasm and patience during the preparation and completion of this study. The author also wishes to thank his wife, Cheryl, and Dr. R. D. Cameron, Liza McDonald, Murray Hendren and Dale Wetter for their help and assistance. This study was financed through National Research Council Grant, NRC 67-8253. x i CHAPTER I INTRODUCTION 1.1 GENERAL The aerated stabilization basin (ASB) as a means of waste treat- ment was i n i t i a l l y developed from the upgrading of waste water holding ponds. Today, however, the ASB treatment system is recognized as having a biological basis. Unlike other biological treatment systems, ASBs do not have the com- plexities of sludge recycle, which makes them ideal systems for rural indus- tries and communities where land is cheap and operational supervision and available capital are minimal. Included among the many users of ASBs are numerous communities and industries located in northern areas where severe winter conditions are en- countered. ASBs operated in these northern areas have been successful, but they have invariably demonstrated a change in treatment efficiency with a change in basin temperature. It is this change which requires investigation. Descriptions of lagoons and basins, aerated lagoons, aerobic lagoons, faculative lagoons, oxidation ditches, photosynthetic ponds, and aerated stab- i l i z a t i o n basins can be found throughout the current literature. Unfortunately, one man's photosynthetic pond has often turned out to be another man's aerated lagoon. Therefore, in an attempt to avoid any semantic d i f f i c u l t i e s , the treatment system described in this paper, modelled in the laboratory and called an aerated stabilization basin has the following characteristics : 2 1. The basin i s hydraulically completely mixed, with the mean hydraulic retention time equal to the mean c e l l residence time (sludge age). 2. There i s no sludge recycle incorporated into the system. 3. The chemical and biological oxygen requirements of the treatment process are satisfied by mechanical means — generally, surface aerators, or diffuser systems. 4. Theoretically, there i s sufficient energy within the basin to maintain a l l the solids in suspension. Solids loss occurs only through oxidation or effluent carry-over. 1.2 RESEARCH OBJECTIVES The principle objective of this study was to obtain data on the operation of laboratory-scale ASBs at a low operation temperature (3°C), and to analyze the performance of these systems in terms of existing mathe- matical models used to describe ASB operation. To do this, information on five operating parameters was collected: substrate u t i l i z a t i o n , system treat- ment efficiency, solids production, nitrogen transformations and oxygen uptake. Of these-, the f i r s t two were used to evaluate the performance of the laboratory ASBs in terms of the existing mathematical models, while the other data were used to define performance characteristics. In addition information on the settling characteristics of ASB effluent at 3°C was collected. CHAPTER II LITERATURE SURVEY 11.1 GENERAL The aerated stabilization basin i s a recent innovation in the treatment of waste water, having only come into prominence since the early 1960s. It was not u n t i l 1959 that turbine aerators were used (21). Since their inception ASBs have been used with some success in northern climates, even though treatment efficiency reportedly decreases during the winter months (34) (42)(43). Today, the ASB is used for treating wastes throughout the industrial segment, from food processing to petrochemical wastes (14) (18). Besselievre (4) presents an extensive l i s t of references for industrial waste applications of ASB systems. 11.2 DESIGN FORMULATIONS FOR SUBSTRATE REMOVAL IN ASBs Many of the ASBs now in operation developed from overloaded photo- synthetic or facultative lagoons which were modified through the installation of aeration equipment (23)(36). Other basins have been designed empirically 2 using loading guidelines such as 1.7-2.3 lbs BOD/day/1000 f t . , 700 people/ day/acre, or 2000 lbs BOD5/acre/day for 10 foot depth (16)(8)(28). Today most ASB design manuals (19) (11) follow one or two design models. The f i r s t design model was developed by O'Connor and Eckenfelder (30). For this model substrate removal in a completely mixed basin is described by the following equation: where and equation where and S E = effluent substrate concentration, mg/l; S q = influent substrate concentration, mg/l; K = reaction rate coefficient, day ̂ ; t = mean hydraulic retention time, days. Equilibrium v o l a t i l e solids in the ASB are described by the S + aS X v = v o l a t i l e solids, mg/l; S Q = v o l a t i l e solids in influent waste, mg/l; a = yield factor, mg volatile solids produced/mg substrate used; S r = substrate ut i l i z e d , mg/l; b = endogenous coefficient, % loss/mg volatile solids; t = mean hydraulic retention time, days. In a completely mixed ASB mean hydraulic retention time (basin volume/flow), t, i s equal to the treatment time and the concentration of reactants in the effluent i s the same as the concentration in the basin. For a given waste and required soluble effluent concentration the mean hydraulic retention time, t, and therefore the basin size, can be calculated. 5 A second design model has been developed by McKinney (19). Un- like O'Connor and Eckehfelder, who assume a pseudo first-order substrate removal, McKinney assumes that a l l available BOD i s metabolized in the f i r s t twenty-four hours. He further assumes that the remaining treatment time is used for oxidizing the biological solids produced in u t i l i z i n g the substrate. The following three equations form the basis of McKinney's model: F = k 5 t + 1 (3) k,F M = 2 a 1/t + K- (4) and F = F + k M e 10 a (5) where F = unmetabolized waste B0D^,mg/l; F ± - influent waste B0D5, mg/l; k 5 = metabolism constant, 120 day - 1 at 5°C to 720 day - 1 at 30°C; t = mean hydraulic retention time, days; M = active microbial mass, mg/l; c l k 6 = synthesis constant, 83 days - 1 at 5°C to 500 days" 1 at 30°C; k 7 = endogenous metabolism coefficient, 0.16 days 1 at 5°C to 0.48 days - 1 at 20°C for t < 5 days and 0.04 day - 1 at 5°C to 0.12 at 20°C for t >_ 20 days; F e = effluent B0D5> mg/l; and k^Q = BOD5 proportionality constant, M).6 dimensionless. 6 For a given input waste loading and effluent requirement the retention time, t, i s calculated on a t r i a l and error basis. In addition to the two models described above, three other models have been developed to describe ASB operation; these are the Chemostat Model, the first-order exponential, and a specialized model for pulp and paper wastes. The Chemostat is a name coined by Novick and Szilard (30) for a single, homogeneous, completely stirred, constant volume, flow through reactor; an ASB. The equation describing the substrate remaining i n the Chemostat is the steady-state solution of two equations: the Monod formu- lation, describing substrate oxidation kinetics, and a dif f e r e n t i a l equation describing reactor hydraulics. The Chemostat model i s given by the equation K (D) m where and S = substrate concentration in the reactor, mg/l; K = a saturation constant, mg/l (numerically equal to sub- strate concentration when u = %umax.), (u = growth rate); u m = maximum growth rate constant, days ^; D = dilution rate, days 1 (reciprocal of hydraulic retention time). The Chemostat Model has not been widely applied to waste treatment systems, although some applications are available. The fourth model, the first-order exponential, is described by the equation S/S = e " k t (7) e o 7 where S G = effluent substrate concentration, mg/l; S q = influent substrate concentration, mg/l; k = reaction rate coefficient, days ̂ ; and t = mean hydraulic retention time, days. The first-order exponential model i s often presented by Eckenfelder (10), but the history or reasoning behind this equation is unknown. It appears to be s t r i c t l y based on empirical c r i t e r i a . The f i f t h mathematical model i s described by the following equations (13): L/L Q = (1 + 0.55t)" 0 , 7 8 (8) for no nutrient addition; and > L/L Q = (1 + 0.95t)~ 1 , 0 5 (9) for nutrient addition, where L = effluent B0D5> mg/l; L Q = influent B0D5, mg/l; and t = hydraulic retention time, days. This model was developed to describe the treatment of mixed pulp and paper wastes in ASBs, and i t s use i s therefore restricted to the pulp and paper industry. The coefficients were empirically derived. 8 II.3 TEMPERATURE COMPENSATION IN DESIGN FORMULATIONS The temperature dependency of biological systems has been reported by numerous authors (7) (9) (25) (32). O'Connor and Eckenfelder (31) decrease the reaction rate constant in their design model in order to compen- sate for a temperature drop. This decrease in the reaction rate constant, K, is related to the drop in temperature by the modified Van't Hoff-Arrhenius equation: K_ = K_ • 0T1'TR (10) 1  lR where and YL, = reaction rate constant, days -^ (at temperature T.°C); ll l K_ = reaction rate constant, days 1 (at a reference temperature, T R normally 20°C); 0 = temperature coefficient, theta, dimensionless; T, = ASB temperature, °C; T p = reference temperature, °C. R The temperature coefficient, 0, is a measure of the sensitivity of a system to temperature change. Reported values of 0 vary from 1.0 to 1.13, depending on the system in question. The commonly accepted 0 value for ASBs is 1.035 (7)(9)(37)(42). Table 1 shows a number of the reported 0 values for ASBs. Commonly assumed values for several other biological treatment systems are also shown (7)(9)(37). 9 TABLE 1 TEMPERATURE COEFFICIENTS FOR BIOLOGICAL TREATMENT SYSTEMS P R 0 C E S S 0 TEMPERATURE RANGE WASTE ASB 1.035 10-30°C Cotton textile ASB 1.046 13-20°C Domestic sewage ASB 1.026 2-10°C Pulp and paper ASB 1.058 10-30°C Pulp and paper ASB 1.16 4-20°C Fruit processing Stabilization ponds 1.072-1.085 3-35°C Activated sludge 1.0 -1.041 4-45°C Trickling f i l t e r 1.035 10-35°C Aerobic-facultative lagoon 1.06-1.18 4-30°C Extended aeration 1.037 10-30°C McKinney compensates for temperature changes by varying three of the four constants, K̂ , K^ and K^, used in his design equations. He assumes that the fourth constant, K̂Q (the ratio of BOD̂  to unit weight of active solids gen- erated), remains constant with temperature at ^0.6. No information was found on temperature compensation for the Chemo- stat or first-order exponential models, although for the latter i t is assumed that the ©-concept can be applied to the reaction rate coefficient, k. 10 Eckenfelder (10) developed the following equation for predicting the temperature of ASB contents in terms of ambient air temperature, i n - fluent temperature, flow and expected heat loss: (T - T )fA T i - T w = W Q 3 < n> where T^ = influent temperature, °F; T w = mean basin temperature, °F; T = mean ai r temperature, °F; A = basin area, square feet, Q = waste flow, U.S. mgd; and f = proportionality factor accounting for heat transfer, surface turbulence, wind and humidity effects (for central United States, f = 12 x 10~ ) (mgd/ft 2). This equation i s widely accepted in industry and has been reported to give excellent results (3)(19)(42). II.4 TREATMENT EFFICIENCY Treatment efficiencies have been reported for a number of f i e l d and laboratory ASBs operating at cold temperatures and over a range of temperatures. Carpenter, et al. (7) studying five different pulp and paper wastes at retention times of 2.5, 5 and 10 days and temperatures of 2°C, 10°C, 20°C and 30°C, found that treatment efficiency in the 2.5 day reactor increased from 56% at 2°C to 79% at 30°C, while in the ten day reactor treatment efficiency increased from 79% at 2°C to 88% at 30°C. Thus, the overall effect of temperature on treatment efficiency was shown to decrease with an 11 increase in retention time. Ling (24), studying the treatment of chemi- cal wastes in aerated lagoons, also reported a significant effect of temp- erature on treatment efficiency which was lessened with increasing reten- tion time. Timpany, et al. (42), studying three full-scale, five-day aerated lagoons treating pulp and paper wastes in northern British Columbia and Alberta, found that treatment efficiency increased 20% for an increase of 10°C i n the 14°C to 30°C range. Bartsch and Randall (3), reporting on the state of the art, showed that for a five-day aerated lagoon system, where some settling was occurring, 14.4°C was a c r i t i c a l temperature as treatment efficiency de- creased markedly from 90% at 14.4°C to 55% at 5°C. Esvelt, et al. (14) reported treatment efficiencies for apple processing waste of 84 to 88% at 4-7 C with a retention time of ten to eleven days. Reid (34), studying a basin treating domestic sewage in Alaska, reported that treatment efficiency remained above 80% even though temperatures were near-freezing. Goodrow (20) reported similar results at Regina, Saskatchewan when the treatment basin had an ice-cover. II.5 SOLIDS PRODUCTION AND SETTLING Literature directly pertaining to net solids production in ASBs at cold temperatures is almost non-existent. Goodman (19) shows the endo- genous oxidation rate to be decreased at low temperature, as do Esvelt, et al. (14); however, most of the information available is general or refers to elevated temperatures. Eckenfelder (10) maintains that the equilibrium solids concentration in an ASB from a soluble feed w i l l be 50% of the influent 12 BOD concentration. Gellman (17), citing a number of pulp and paper, pilot studies at 25-35°C, found that sludge accumulated at a rate of 0.15 to 0.30 pounds per pound of BOD removed, i t is not clear from Gellman's paper, or from the p i l o t studies cited, whether or not these figures refer s p e c i f i - cally to net solids production. Reported values of BOD associated with the bio-solids in ASBs are also of a general nature. Eckenfelder (10) graphs the BOD of the volatile solids vs. sludge age and shows mg of BOD per mg of volatile solids of 0.75 to 0.3 for sludge ages of 0 to 7 days. Goodman (19) shows the BOD per unit of active solids to be 0.6, regardless of sludge age or temperature. Gellman (17), again citing a number of pulp and paper p i l o t studies at 25-35°C, re- ports that the solids were well stabilized and the BOD per unit of solids was 0.1 to 0.2 pounds per pound for retention times of four to twenty days. How- ever, i t was not clear whether or not influent solids were included in the measurements. Secondary settling or polishing ponds are commonly used to improve the solids quality of ASB effluents (17)(23)(36), and this in turn provides some improvement in the overall BOD removal. For pulp and paper wastes re- ported improvements in BOD removal vary with settling from 2-15% for aeration times of two to ten days (1)(7)(35)(46). II.6 NUTRIENT REQUIREMENTS McKinney (27), discussing bacterial synthesis, shows the nitrogen and phosphorus requirements to be 11% and 2.5% of the dry weight of the bacter- i a l c e l l or, expressed as a carbon:nitrogen:phosphorus ratio, 20:4.4:1. Eckenfelder (10) describes nutrient requirements in biological treatment 13 systems in terms of the vo l a t i l e solids concentration by the following two empirical equations: Nitrogen (N) = 0.12 AXV + 1.0 mg/l (12) Phosphorus (P) = 0.02 AXv + 0.5 mg/l * (13) where AXv = change in volatile solids concentration, mg/l. Esvelt et al. (14) found that nutrient requirements for f r u i t pro- cessing wastes were a function of BOD concentration and the removal rate con- stant, k. This relationship was expressed by the two empirical equations: N/BOD ' = 0.087 BOD - 0.80 x 0.087k (14) removed cone P/BOD . = 0.016 BOD - 0.80 x 0.016k (15) removed cone where N/B0D r e m Q v e (j = lbs nitrogen required/lb BOD removed; P/B0D r e m o v e <j = lbs phosphorus required/lb BOD removed; k = removal rate constant (0.115 at 20°C). These authors also indicate that temperature had an effect on the nutrient requirements in ASBs. In the pulp and paper industry, the usual practice has been to describe nutrient requirements in terms of the BOD applied or removed. Ecken- felder (13) reports that for pulp and paper wastes optimum treatment should result from nutrient a v a i l a b i l i t y of 4.0 lbs of nitrogen and 0.6 lbs of phos- phorus per 100 lbs of BOD removed. Carpenter et al. (7) supplied 5 lbs of nitrogen and 1.0 lb of phosphorus per 100 lbs of BOD applied in their study. Blosser (5), studying de-inking and white water waste, also reported nitrogen and phosphorus addition of 5.0 lbs and 1.0 lb per 100 lbs of BOD applied. Amberg (1), reporting on a full-scale m i l l system, found that a B0D:N:P ratio of 300:7.5:1 was sufficient to support synthesis. CHAPTER III RESEARCH METHODOLOGY II I . l RATIONALE The rationale for undertaking this study and therefore the basis of the objectives, stems more from a lack of c l a r i f i c a t i o n as to what people have done in the past, rather than a need for investigation of a new system under new conditions. ASBs operating at cold temperatures, and the effect of temperature changes on ASB operation, have been previously studied under both f i e l d and laboratory conditions (7) (42) (34). The majority of these studies have been deficient in two areas: 1. The operating conditions under which the studies were conducted have not been specified, i.e., there is no indication as to whether or not steady- state conditions were achieved. 2. There has been no definition of treat- ment efficiency; i.e., whether treatment efficiency i s a measure of substrate u t i l i z a t i o n or a measure of the decrease in oxygen demand from the influent to the effluent. (The latter term is defined in this paper as system treatment efficiency). The detailing of steady-state conditions is necessary because a l l mathematical models used to describe ASB operation are based on the assumption that steady-state conditions exist. Therefore, any study undertaken to evaluate these models must be conducted under steady-state conditions. The need for a clear differentiation between substrate removal and the decrease in oxygen demand from the influent to effluent in ASBs i s per- haps less obvious, but equally important. In treatment systems such as a c t i - 14 15 vated sludge or t r i c k l i n g f i l t e r s , where there is a removal of biological solids by sedimentation, the difference between per cent substrate removal and the system treatment efficiency may be slight. However, for ASBs where there is a carry-over of biological solids, the numerical difference in the efficiency measurement can be significant. In terms of evaluating design models and predicting ASB operation, i t becomes necessary to clearly define what constitutes treatment efficiency. The models used to define ASB operation (O'Connor and Eckenfelder's, McKinney's, Chemostat, and first-order exponential) are reported to relate substrate u t i l i z e d or remaining, as a function of a biological reaction rate constant (K) and treatment time (31)(19(10); yet there has rarely been a clear distinction as to just what has been measured in previous ASB studies. In addition to the lack of c l a r i f i c a t i o n in previous studies, several of the assumptions behind temperature compensation in current ASB design prac- tice are questionable. In order to compensate for an expected decrease in treatment efficiency due to decreasing ASB temperatures, the current practice has been to increase treatment time by increasing the basin volume. The basis of this design practice i s that current design models relate substrate removal or substrate remaining to treatment time and a reaction rate constant (K). Theoretically, any increase in substrate remaining due to a decrease in the reaction rate constant can be compensated for by increasing the treatment time. The reaction rate constant at the lower temperature is simply calculated using the modified version of the Van't Hoff-Arrhenius equation, presented earlier: K = K • 0 T r TR (10) T, T 16 In current design practice, where temperature compensation for ASBs is considered, the following assumptions can be questioned: 1. The design models are applicable over the range of the temperature drop; 2. The biological population described by 0 does not change, either in population size or in species make-up; and 3. Only the residual substrate portion of the effluent is a function of temperature, i.e., the oxygen demand associated with effluent solids i s not considered. In summary, the rationale behind the objectives of this study were: 1. A need to clearly detail the operation con- ditions prior to the collection of ASB data, i.e., to estab- l i s h steady-state operation; 2. A need for a definite differentiation between substrate removal and system treatment efficiency in ASB studies; 3. To question a fundamental assumption of cold temperature operation prediction, i.e., current design models described the operation of ASBs over a broad temper- ature range; and 4. To question a further assumption that only the residual substrate portion of ASB effluent changed with temp- erature. On the basis of this rationale, the study was conducted so that 1. The laboratory ASB systems could function under steady- state conditions; 2. A clear differentiation between substrate u t i l i z a t i o n and system treatment efficiency existed (this was accomplished by measuring these parameters separately); 3. Sufficient data was collected to evaluate the mathe- matical models at a cold temperature (3°C); and 4. The proportions of the residual substrate and the biological solids could be determined and compared to similar data collected in the 15 - 25°C range. 17 It should be emphasized that this study was not undertaken to formulate a new model to describe ASB operation at cold temperatures, but simply to evaluate the existing models using data collected at 3°C under controlled steady-state conditions. The nitrogen, oxygen uptake, and settling data were collected and are presented as general information for ASB operation at a cold temp- erature (3°C). III.2 GENERAL PROCEDURE In order to develop experimental data on ASBs operating at a cold temperature, laboratory-scale continuous flow ASBs (operating at hydraulic retention times between one and sixteen days) were maintained at a controlled temperature of 3°C over a four month period. The reactor contents were fed a synthetic waste of powdered skim milk and tap water at two concentration levels: 630 mg/l COD and 1,240 mg/l COD. (Equivalent to 290 mg/l and 800 mg/l of BOD̂ , respectively). The reactors were monitored for COD, MLSS and Kjeldahl nitrogen following start-up to determine when steady-state operation was achieved. Considerable effort was expended on maintaining a constant hydraulic and applied load to the reactors so that ideal steady-state conditions were approached. Attainment of steady-state was verified through the measured stabilization of the substrate concentration, effluent quality, total nitrogen concentration of solids concentration. Once steady-state was attained, a f u l l testing program consisting of COD, B0D,j, Kjeldahl nitrogen, organic nitrogen, ammonia nitrogen, nitrate nitro- gen, and MLSS analysis was started. The ASBs were operated at each of the two 18 loadings u n t i l sufficient data on substrate u t i l i z a t i o n , system treatment efficiency, solids production, nitrogen usage and oxygen uptake were collec- ted. Batch settling tests were conducted following the completion of the continuous flow studies to evaluate the settling characteristics of the ASB effluent at 3°C. III.3 EQUIPMENT A schematic of the laboratory's ASB system is shown in Figure 1. F I G U R E I S C H E M A T I C O F M O D E L A S B 19 As shown in Figure 1, raw milk waste was fed to the ASB reactor by a precision volume pump which was controlled by a pulse timer. The reactor contents were kept completely mixed by an electric mixer and were aerated by diffusers. The reactors were plexiglass cylinders, capped at one end, and tapped along the sides so that the effluent would overflow into calibrated containers. The one-day reactor had a volume of 8.7 l i t e r s ; the other four were nominally 20 l i t e r s . The experimental apparatus was contained within a walk-in temperature room set at 3°C ± 0.5°C. The feed for the two, four, eight and sixteen day reactors was pumped from a single 20 l i t e r carboy, the contents of which were prepared and changed daily, Monday to Friday. The one day reactor was fed from another carboy which was r e f i l l e d every other day. Feed was maintained over the week- ends by syphoning from additional carboys. The carboys were stoppered and the air vents were plugged with cotton. Consequently, the build-up of bacterial solids in the feed bottles and feed lines was not a problem. In order to achieve steady-state conditions in the reactors, the hydraulic and applied load to each unit had to be maintained at a steady value. As the concentration of the synthetic waste feed was easily controlled in preparation, both the hydraulic and applied load could be maintained at a fixed level by controlling the pumping rate. The flow through each reactor was collected in calibrated containers and was checked daily. The pump flows were checked periodically using a grad- uated cylinder. Evaporation in the temperature room was found to be negligible. Over the period of the study, the flows through the reactors were reasonably constant, as shown in Table 2. Listed in Table 2 are the 20 nominal hydraulic retention times, the mean hydraulic retention times (reactor volume/mean flow) and the standard deviation about the mean hydraulic retention time for the five reactors. TABLE 2 REACTOR HYDRAULIC RETENTION TIME NOMINAL MEAN STANDARD DEVIATION H.R.T.,DAYS H.R.T.,DAYS . H.R.T.,DAYS 1 1.0 ±0.11 2 1.97 ±0.13 4 3.96 ±0.24 8 8.6 ±0.26 16 16.7 ±1.05 The nominal hydraulic retention times are used in the discussion of reac- tor performance; however, the mean hydraulic retention times, as l i s t e d in Table 2, were used in a l l calculations and graphs. III.4 SUBSTRATE The substrate used in the experiment was a synthesized mixture of powdered skim milk and aged tap water. An analysis of the raw milk waste is given in Table 3. The concentration of phosphorus and nitrogen in the milk, in the ratio of 100:7.5:1 (B0D^:N:P), was more than sufficient to supply any bio- chemical needs, assuming that both nutrients were in a readily useable form for the organisms. 21 TABLE 3 ANALYSIS OF POWDERED MILK WASTE 1000 mg/l Mixture ( I n i t i a l Analysis) Phosphorus 7 mg/l Inorganic carbon 3 mg/l Organic carbon 43.2 mg/l Organic nitrogen 52 mg/l Inorganic nitrogen 0.0 mg/l COD 1048 mg/l BOD o690 mg/l BOD 850 mg/l Suspended solids (Gooch) 0.0 mg/l Filtrable solids (Whatman #4) 0.0 mg/l Filtrable solids (Millipore 0.45y) 0.0 mg/l B0D5: Nitrogen rPhosphorus 100:7.5:1 600 mg/l Mixture (First Loading) Organic nitrogen 30 mg/l Inorganic nitrogen 0.3 mg/l COD 630 mg/l BOD,. 290 mg/l 1200 mg/l Mixture (Second Loading) Organic nitrogen Inorganic nitrogen COD BOD' Also from Table 3, i t can be seen that the COD/BODu ratio is 1.25, which means that 80 per cent of the measured COD i s biodegradable. The COD: B0D5 ratio (high loading) of 1.53 f a l l s within the reported C0D/B0D5 range for actual dairy wastes (38). 65.0 mg/l 1.0 mg/l 1240 mg/l 800 mg/l 22 Typical physical and chemical characteristics of the Vancouver tap water, as determined by the Greater Vancouver Water Di s t r i c t , are shown in Table 4 (47). TABLE 4 GREATER VANCOUVER WATER DISTRICT PHYSICAL & CHEMICAL ANALYSIS OF WATER SUPPLIES CAPILANO INTAKE Appearance Odour Turbidity pH Total residue Total fixed residue Total v o l a t i l e residue Total alkalinity as CaCO, Total hardness as CaCO^ Chloride as Cl Sulphate as SO^ Fluoride as F Si l i c a as S102 Ammonia as N Nitrate as N Nit r i t e as N Copper as Cu Total Iron as Fe Dissolved Oxygen Specific Conductance in micromhos/cm at 25°C Clear N i l 0.4 6.4 17.5 ppm 9.6 ppm 7.9 ppm 2.7 ppm 4.6 ppm 0.3 ppm 1.7 ppm Less than 0.05 ppm 3.2 ppm Less than 0.01 ppm Less than 0.1 ppm Less than 0.002 ppm Less than 0.02 ppm O.OS 1 ppm 11.7 ppm 13.7 Powdered milk was used in synthesizing the raw waste for the follow- ing reasons. 1. The waste would contain only a soluble and colloidal portion. 2. The feed mixture would pass through a 0.45 f i l t e r with- out a loss of colloidal solids, yet any biosolids in the effluent could be removed by this f i l t r a t i o n . The residual substrate 23 concentration could then be determined using normal analyti- cal techniques for COD or BOD̂ ; a simple subtraction from the COD or BOD̂  feed concentration would give substrate u t i l i z a t i o n . 3. The milk solids would pass through a gooch crucible and glass f i l t e r ; therefore the mixed liquor sus- pended solids determination would not be affected. 4. The mixture is representative of an industrial dairy waste. III.5 ANALYTICAL PROCEDURES A l l of the analytical procedures used in this study, with the excep- tion of the oxygen uptake rates, were as outlined in Standard Methods3 Thir- teenth Edition (40). Analyses were made on two types of sample: f i l t e r e d effluent and gross effluent. The f i l t e r e d samples were free of bacterial solids and contained only residual soluble and colloidal substrate. These samples were prepared in the following manner: 1. An aliquot of the reactor contents was centrifuged for 20 minutes at 2000 rpm to remove coarse solids. 2. The centrate was f i l t e r e d through a gooch crucible and glass f i l t e r , yielding a rough f i l t r a t e . 3. The rough f i l t r a t e was passed through a 0.45 m i l l i - pore f i l t e r , to remove bacterial cells and provide a sample having only substrates. The gross effluent samples were unaltered samples of the reactor contents. The following analyses were carried out on a continuing basis through- out the study. 1. Chemical Oxygen Demand (COD) on: (a) Feed: for the determination of influent oxygen demand. (b) Reactor contents, gross: for the determination of system treatment efficiency. 24 (c) Reactor contents, f i l t e r e d : for the determination of substrate u t i l i z a t i o n . (d) Settled effluent: for the determination of system treatment e f f i - ciency with post settling. 2. Biochemical Oxygen Demand — 5 Day (BOD^) on: (a) Feed: for the determination of Influent oxygen demand. (b) Reactor contents, gross: for the determination of system treatment efficiency. (c) Reactor contents, f i l t e r e d : for the determination of substrate u t i l i z a t i o n . 3. Mixed Liquor Suspended Solids (MLSS) on: (a) Reactor contents, gross: for the determination of reactor solids concentration. (b) Settled effluent: for the determination of suspended solids level In the settled effluent. 4. Kjeldahl Nitrogen on: (a) Feed: for the determination of the total nitrogen concentration in the feed. (b) Reactor contents, gross: for the determination of the total nitrogen concentration in reactors. 5. Organic and Ammonia Nitrogen on: (a) Feed: for the determination of organic and ammonia nitrogen level in feed for comparison with total Kjeldahl. (b) Reactor contents, f i l t e r e d ; for the determination of organic and ammonia nitrogen level in f i l t r a t e , in order to calculate the nitro- gen content in the biological solids. 6. Nitrate Nitrogen on: (a) Feed: for the determination of the background nitrate level. (b) Reactor contents, f i l t e r e d : for the determination of n i t r i f i c a t i o n , 7. £H: The pH of the reactor contents was determined periodically. 25 8. Settling: At the completion of each loading run, batch settling tests were conducted at 3°C using Imhoff cones. 9. Oxygen Uptake: The oxygen uptake rates were determined using a YSI Model 51 dissolved oxygen probe. The probe was calibrated periodically against a Winkler determination. CHAPTER IV RESULTS AND DISCUSSION IV.1 GENERAL The performance characteristics of the laboratory ASB systems oper- ating at 3°C are presented and discussed in this section in terms of present- day knowledge of ASB operation. The data points presented are mean values collected over the several weeks of steady-state operation. Variations about the mean of the suspended solids values are presented due to large fluctuations. The c r i t e r i a used in determining steady-state operation, and the re- sults from the batch settling tests are also presented. The raw steady-state data are presented in Appendix A. TV.2 CRITERIA FOR STEADY-STATE OPERATION Steady-state operation implies that both hydraulic and treatment equilibria have been reached. In a completely mixed, flow through system, such as that used i n this study, one criterion for steady-state operation i s that the solids,level in the basin reaches a stable concentration. At that point, bio- logical solids wash-out equals the net solids production from substrate u t i l i - zation. Steady-state operation can also be documented by the stabilization of either substrate removal or system treatment efficiency at hydraulic equilibrium. IV.2.1 Low Loading Study Steady-state conditions were achieved at the low loading (BOD^ = 290 mg/l) twenty-four to twenty-seven days after continuous flow operation was 26 27 initiated, as shown in Figure 2 for the 16-day reactor. At this point the Kjeldahl nitrogen concentration in the reactor had reached the feed level, signifying hydraulic equilibrium, and system treatment efficiency had reached a constant level. Similar data for the other reactors is presented in Appen- dix B. IV.2.2 High Loading Study Steady-state conditions were achieved in the high loading run (BOD^ = 800 mg/l) twenty-five to twenty-nine days after start-up. Two to three weeks after this run was started, an unusual but interesting phenomenon was noted in the reactors. The MLSS and fi l t e r e d COD levels started to fluc- tuate in a cyclic manner, while the fi l t e r e d BOD̂  and gross BOD̂  and COD levels were unaffected. This cyclic phenomenon i s shown in Figure 3 for the two and sixteen-day reactors. Cycling i n a bacterial system has been described by Gaudy, et al. (15) in their study of total oxidation of activated sludge and by Thirmurthi (41) studying photosynthetic ponds, who found wide and unexplained variations in fi l t e r e d COD, but not BOD, values. The fluctuations in the MLSS and f i l t e r e d COD concentrations raise the question of non steady-state operation. However, an examination of Figures 4 and 5 shows that steady-state conditions were reached in a l l reactors after twenty-nine days. As can be seen from Figure 4, hydraulic equilibrium was achieved in a l l the reactors by twenty-nine days. By the same time, as shown in Figure 5, system treatment efficiency had stabilized. S Y S T E M T R E A T M E N T E F F I C I E N C Y , % ( C O D ) O C m ro co H m > o -< CO H > H m CD o > -< 3J m > o H O 3J O r o > g z ro Ui v. ro ro- TJ m 3 o o 10 OV ro GO \ ro cr>- o P0 o OJ o o _4_ o _1_ f > < o -n m rn o o o z o OJ 9 OJ 3 «3 m r~ o x o a m z CO -< CO H m 3J m > m z m o m z o -< 1 CO H m > o -< CO H > H rn -r- o T r-o o I UJ o K J E L D A H L NITROGEN C O N C E N T R A T I O N , rng /-6 83 F I G U R E 3. C Y C L I C F L U C T U A T I O N S O F F I L T E R E D C O D C O N C E N T R A T I O N S 70 60 -E z UJ CD O CH X < -i 5 0 ^ UJ a : o t- o < UJ tr 2 day I day 40 6 day r STEADY S T A T E (hydraulic equilibrium) 1 • e eed concentration 66.0 m g / l 3 / 8 / 7 2 3 4 T E S T PERIOD, Weeks 7 14/9/72 F I G U R E 4 . H Y D R A U L I C EQUILIBRIUM - HIGH LOADING o +>. 6 0 . > O 50 • z LL! O Ix. 4 0 - U_ U J h * Z 3 0 - L U < 2 0 -t i l DC H Lti 10- I— CO > CO 0 • S T E A D Y - S T A T E o O O "O <? A A A • • © o A l 4 A I DAY . 2 DAY A 4 4 DAY i o o 8 DAY + + 16 DAY "TP 8 2 1 0 / 8 / 7 2 F I G U R E 5, - T ~ 6 7 1 4 / 9 / 7 2 3 4 5 T E S T P E R I O D , W E E K S S T E A D Y - S T A T E - ( S Y S T E M T R E A T M E N T E F F I C I E N C Y COD HIGH L O A D I N G Co 32 IV.3 PER CENT SUBSTRATE REMOVAL AND SYSTEM TREATMENT EFFICIENCY The principle operating parameter for an ASB is treatment e f f i - ciency. Presented in this section are the results for two efficiencies: per cent substrate removal and the system treatment efficiency. Per cent substrate removal is the per cent decrease in the applied substrate and is equal to 100 per cent minus the per cent of substrate remaining in the reactor. System treat- ment efficiency is the per cent decrease in the oxygen demand (BOD^ or COD) measured between the influent and the effluent, and as such, takes into con- sideration the oxygen demand associated with generated solids. . i v .3.1 Per Cent Substrate Removal (Substrate Ut i l i z a t i o n ) . Per cent substrate removal as a function of mean hydraulic retention time i s plotted in Figure 6. Curves A and B relate per cent substrate removal on a COD basis for the two loadings studied; curves C and D are on a BOD̂  basis. As can be seen from Curves C and D in Figure 6, there is v i r t u a l l y complete u t i l i z a t i o n (94 - 98 per cent) of the substrate in two to three days, as mea- sured by BODy On a COD basis, Curves A and B, substrate removal continues u n t i l about eight days, after which per cent removal i s constant at 77 - 80 per cent. From the milk waste analysis, Table 3, i t can be seen that the B0Du/C0D ratio is 1:1.25, i.e., 80 per cent of the influent COD i s biodegradable, i n - dicating that for retention times beyond eight days, there is virtua l l y complete u t i l i z a t i o n of the biodegradable portion of the substrate, measured by either COD or BOD̂ . Applying this same reasoning, that i s , only 80 per cent of the COD is biodegradable, there is 75 - 85 per cent u t i l i z a t i o n of the biodegradable substrate in one to two days. A possible explanation for the difference in time needed to achieve the same per cent substrate removal when measured on a COD or BODc basis may be P E R C E N T S U B T R A T E R E M O V A L 4̂  Ol CT> 00 CD O o o o o o o -o 3 3 3 3 «Q « Q 0 3 «rt ^ >s \ ^ 34 the formation of intermediate compounds in the reactors. It can be suggested, because of the limitations of the BOD̂  test, that the COD curves are a better representation of the effect of mean hydraulic retention time on substrate re- moval at 3°C. Comparative ASB substrate removal data could not be found in the literature. However, Hoover et al. (21) documented very rapid and complete oxidation of dairy wastes in batch studies at 30°C. They found that 500 ppm of sludge solids would oxidize 1000 ppm of milk solids in six hours. Comparing these data with data from this study, i t would appear that milk wastes can be oxidized at least three times as rapidly at 30°C as they can at 3°C. IV.3.2 System Treatment Efficiency. The system treatment e f f i c i e n - cies measured in the laboratory ASB systems are plotted in Figure 7 as a func- tion of mean hydraulic retention time. Curves A and B are on a COD basis and Curves C and D are on a BOD̂  basis for the two loadings. As shown in Figure 7, system treatment efficiency continues to increase with mean hydraulic retention time over the range of the study, reaching 80 per cent (BODtj) or 51 per cent (COD) at sixteen days. The system treatment efficiencies (BOD^) measured in this study at 3°C are generally equal to, or lower than, reported cold temperature ASB treat- ment efficiencies. Carpenter et al. (7), treating pulp and paper wastes in the laboratory at 2°C, found efficiencies of 56 per cent to 79 per cent for retention times of 2.5 to 10 days. Esvelt et al. (14), treating f r u i t processing waste, re- ported efficiencies of 85 to 88 per cent for retention times of 10.5 - 11.5 days in the 4° - 7°C range. Reid (34) and Goodrow (20) report treatment efficiencies in excess of 80 per cent for domestic sewage treated in twenty day lagoons at near zero temperatures in Alaska and at Regina which are comparable to those found in this study. 35 FIGURE 7 . S Y S T E M T R E A T M E N T E F F I C I E N C Y A S A F U N C T I O N O F M E A N H Y D R A U L I C R E T E N T I O N T I M E A T 3° C . 36 IV.4 EVALUATION OF MATHEMATICAL MODELS USED IN ASB DESIGN IV.4.1 General. A fundamental assumption in predicting ASB treatment efficiency at cold temperatures is that the design model used, given the right set of constants, w i l l be applicable at a predicted temperature. Four of the five models presented in the literature review were eval- uated as to their applicability at 3°C on the basis of the per cent substrate removal and system treatment efficiency data collected in this study. The models evaluated were O'Connor and Eckenfelder's, McKinney's, the Chemostat, and the first-order exponential. The f i f t h model, the retardent form of O'Con- nor and Eckenfelder's equation, was not evaluated as i t s use is restricted to the pulp and paper industry. IV.4.2 Evaluation of O'Connor and Eckenfelder's Model. By far the most commonly used and frequently reported model is O'Connor and Eckenfelder given by Equation 1: = 1 (1) S 1 + Kt V ' o To use O'Connor and Eckenfelder's model, the reaction rate constant, K, must be known or calculated. To calculate the constant, Equation (1) is manipulated int the linear form S /S = Kt+1; S/S (influent substrate concentration/effluent o e o e substrate concentration) i s then plotted against t. The slope of the straight line drawn through the data points is then K. Figure 8 is a plot of S o/S g against t on a COD basis for the two loadings. As can be seen from this figure, this relationship of S Q/S e to t is non-linear and therefore O'Connor and Eckenfelder's model is not applicable. A similar plot of SQ/S against t on a BOD- basis would yield a straight line, UJ r- < cr t- co m ID CO 5.0- a ^ 4.0 UJ o z u j 3.0- UJ UJ < 2 or — h- . ir> o oo z O o co o a> 2.0- 1-0- o CO S 0 / S e )\ K t + I where K is constant •A C O D - 630 mg/-e A COD - 1240 mg/ -£ 2 "T" 4 6 • T 10 12 MEAN HYDRAULIC RETENTION T I M E , days -f—  ~i— 14 -T— 16 -I— 18 F I G U R E 8. E V A L U A T I O N O F R E A C T I O N R A T E C O N S T A N T I K ) F O R S U B S T R A T E R E M O V A L ( S e ) - ( O ' C O N M E R a n d E C K E N F E L D E R ) . CO 38 but with a zero slope (S Q/S e is v i r t u a l l y constant with mean hydraulic reten- tion time after one day), which is a non-solution. Thus, in terms of describ- ing per cent substrate removal measured in the laboratory PSB at 3°C, O'Connor and Eckenfelder's model is not applicable. Evaluating the same model in terms of system treatment efficiency, S Q / S E (influent substrate concentration/effluent gross concentration) plots as a straight line function of the mean hydraulic retention time for t beyond two days, as shown in Figure 9. Reaction rate constants can be calculated from the study data for retention times beyond two days. The intercepts of these straight lines do not pass through 1.0, but vary from 1.3 to 2.1. The BODJJ data is f i t t e d using only one l i n e . The measured system treatment efficiencies and the system treat- ment efficiencies calculated using O'Connor and Eckenfelder's model with the constants calculated from Figure 9, are l i s t e d in Table 5 for comparison. As can be seen from Table 5, O'Connor and Eckenfelder's pseudo f i r s t order model w i l l describe only the system treatment efficiency measured in the laboratory ASBs at 3°C for retention times beyond two days. S Q / S * = O . I 5 0 t + 2.10 3 . 0 - 2 . 0 - 1.0 - • BO Dc = o BOD = 2 9 0 m g / | 800 m g / f - 0 . 030 t + 1.65 o • A . 4 S 0 / S ' e = 0 . 0 3 0 1 + 1.40 * COD = 6 3 0 m g / l A COD = 1240 m g / l 4 6 8 10 12 14 16 t = M E A N H Y D R A U L I C R E T E N T I O N T I M E , D a y s 18 FIGURE 9. E V A L U A T I O N OF REACTION RATE CONSTANTS K FOR S Y S T E M T R E A T M E N T E F F I C I E N C Y AT 3 ° C . ( O ' C O N N E R A N D E C K E N F E L D E R ) CO VO TABLE 5 MEASURED AND CALCULATED SYSTEM TREATMENT EFFICIENCIES O'CONNOR AND ECKENFELDER'S MODEL E = ^ X 1 0 ° % COD EFFLUENT BOD,. EFFLUENT RETENTION ABSOLUTE ABSOLUTE TIME LOADING MEASURED CALCULATED DIFFERENCE MEASURED CALCULATED DIFFERENCE (Days) 1 23.5% 36.0% 12.5% 35.6% 56.7% 21. 1% 2 36.0% 36.5% 0.5% 62.5% 59.0% 3.5% 4 •a 40.4% 39.4% 1.0% 61.0% 63.3% 2.3% 8 i-i w 46.0% 45.3% 0.7% 73.2% 70.3% 2.9% 16 50.5% 53.0% 2.5% 76.6% 77.8% 1.2% 2 31.2% 31.0% 0.2% 18.0% 59.0% 41.0% 4 s 36.9% 33.2% 3.7% 63.2% 63.3% ,1.1% 8 o 43.3% 39.0% 4.-3% 66.0% 70.3% 4.3% 16 44.9% 45.0% 0.1% 80.0% 77.8% 2.2% 41 IV.4.3 McKinneyfs Model. McKinney's model, in very general terms, is based on the premise that a l l of the influent substrate is u t i l i z e d within the f i r s t day and the remainder of the treatment or retention time is used for the oxidation of generated solids. Substrate remaining, active solids production, and the effluent oxygen demand are described by three inter-related equations: F i F = kJt+T <3> k F Ma = V (4) a 1/t + k ? F e = F + k 1 0 M a < 5 ) Equation (3), which describes substrate remaining in the basin, is the same form as O'Connor and Eckenfelder's equation, and like their equation, i t does not describe the substrate removal measured in this study. Listed in Table 6 are the substrate concentrations measured in the laboratory ASBs and calculated using McKinney's equation, F = F^/k^t +1. A similar comparison on a COD basis could not be made as COD constants are not available. The constant K̂  for 3°C (108 day "*") was extrapolated from McKinney's values for 5°C to 20°C presented in Goodman (19). As can be seen from Table 6, McKinney's model predicts a more rapid and complete substrate u t i l i z a t i o n than was measured in this study. This d i f - ference could suggest that McKinney's model cannot be extrapolated beyond 42 TABLE 6 MEASURED AND CALCULATED BOD5 SUBSTRATE CONCENTRATION McKINNEY'S MODEL RETENTION LOW LOADING HIGH LOADING TIME MEASURED CALCULATED MEASURED j CALCULATED 1 — — 187 mg/l 7.3 mg/l 2 55 mg/l 1.33 mg/l 22 mg/l 3.7 mg/l 4 12 mg/l 0.67 mg/l 23 mg/l 1.85 mg/l 8 17 mg/l 0.33 mg/l 16 mg/l 0.87 mg/l 16 14 mg/l 0.15 mg/l 17 mg/l 0.43 mg/l Gross effluent concentrations were calculated using McKinney's three equations; I.e., relating Fe, the effluent BOD̂ , to F i , the raw waste B0D,j. Presented in Table 7 are the measured and calculated effluent BOD,.. In these calculations the MLSS levels were assumed to be equivalent to the active mass when used in conjunction with an evaluated k^^ constant. As can be seen from Table 7, with the exception of the i n i t i a l points in each loading, there i s l i t t l e difference between the measured and calculated concentrations. The constants used in evaluating McKinney's model were extrapolated to 3°C from the values tabled in Goodman's Design Manual (19). Goodman tables a range of values for k^ from 0.04 days^to 0.16 days ^ for sludge of five to twenty days at 5°C. To find the appropriate value of k^ within McKinney's range tabled in Goodman (19), the test data was used to calculate an average k 7 and 43 values of k-̂ Q. The k^ values calculated were 0.086days for low loading and 0.076 days ^ for high loading. The k^^ values calculated were: 0.79 days ^ for the low loading and 0.70 days ^ for high loading. Details of the calculations used in defining constants k^ and k^^ can be found in Appendix C. TABLE 7 MEASURED AND CALCULATED GROSS EFFLUENT B0D5 CONCENTRATIONS McKINNEY'S MODEL 290 mg/l B0D5 RAW WASTE RETENTION CALCULATED MEASURED TIME B0D5 B0D5 DIFFERENCE 2 days 4 days 8 days 16 days 131 mg/l 114 mg/l 88 mg/l 63 mg/l 238 mg/l 109 mg/l 99 mg/l 59 mg/l 107 mg/l 5 mg/l 11 mg/l 4 mg/l 800 mg/l B0D5 RAW WASTE RETENTION CALCULATED MEASURED DIFFERENCE TIME BOD̂  BOD̂ 1 day 2 days 4 days 8 days 16 days 344 mg/l 323 mg/l 286 mg/l 224 mg/l 165 mg/l 513 mg/l 300 mg/l 306 mg/l 215 mg/l 177 mg/l 169 mg/l 23 mg/l 20 mg/l 9 mg/l 12 mg/l 44 IV.4.4 Chemostat. The Chemostat model i s described by the following steady-state equation: S = =• (5) V - D m As noted in the literature review the Chemostat i s based on the Monod equa- tion and an equation describing ASB hydraulics. As can be seen from Figure 10, a plot of biodegradable COD substrate concentration against dilution rate, the Chemostat model gives a reasonable estimate of the residual substrate concentration measured in the laboratory ASBs at a l l hydraulic retention times and at both loadings. As the Chemostat model describes available or useable substrate remaining in the reactor, the non-biodegradable portion of the feed COD was substracted from that concentra- tion measured in the reactors before i t was plotted in Figure 10 (126 mg/l' at the low loading and 250 mg/l at the high loading). The Chemostat curve used in approximating the f i l t e r e d substrate data at 3°C was fitted using the constants Ks = 323 mg/l and u = 2.2 days Calculating, using the Chemostat equation with the above constants, the washout retention times at the two substrate loadings are 14.8 hours and 17.9 hours, respectively, for the low and high loadings. The maximum growth rate under un- limited substrate conditions i s 2.2 days \ which implies a generation time of 10.9 hours. This generation time i s of the same order as the reported gener- ation times of a psychrophylic strain of Pseudonomonads at low temperatures (32). The BOD,, substrate concentration measured in the reactors did not change with mean hydraulic retention time after one day, and therefore could not be described by the Chemostat model. A ' I 1 1 1 1 • 1 1 1 0 £ 4 6 8 10 |2 14 16 18 t , M E A N H Y D R A U L I C R E T E N T I O N T I M E - D A Y S •I* 1.0 0.25 0.125 0.0625 D, DILUTION R A T E - l / t , D A Y S " 1 FIGURE 10. C H E M O S T A T F I T - C O D S U B S T R A T E CONCENTRATION 46 IV.4.5 First-Order Exponential. The first-order exponential model is described by the following equation: S e/S o = e- k t (7) Like O'Connor and Eckenfelder's model, the first-order exponential model can only be used when the reaction rate constant, k, i s known. The reaction rate constant, k, can be calculated from a semi-log plot of Se/S against the mean hydraulic retention time, t. As can be seen from Figure 11, semi-log plots of S e/S Q (COD and BOD,. for both loadings) against t, the first-order exponential equation does not describe the substrate removal measured in the laboratory ASBs at 3°C. On a BOD̂  basis, the reaction rate constant i s 0.0 or a non-solution. On a COD basis, the points cannot be approximated by a straight l i n e . The first-order exponential model can, however, be used to describe the gross effluent concentration (or on a per cent basis, system treatment efficiency) in the laboratory study. The reaction rate constants, on a COD and BOD̂  basis for the two loadings, are calculated from the semi-log plots shown in Figure 12. As can be seen from Figure 12, the gross effluent data can be approximated only for mean hydraulic retention times beyond two days. The first-order exponential equation used is in the form of S e/S Q = C £ where C varies from 0.44 to 0.70. Listed in Table 8 are the system treatment efficiencies measured in the laboratory and those calculated using the first-order exponential equation. It can be seen that the absolute per cent difference between measured and c a l - culated system treatment efficiency i s less than four per cent for mean hydraulic retention times greater than two days. 0f40 47 O o o Ul r - a: to EQ 3 - CO Q IU UJ a z o o UJ < a : H CO CD CO h - z UJ 3 u. UJ II <u CO co° 0.20 , - 0.10 0.08 0 .06 •{ 0.0 4 0.0 3 0.02 0.015 o © B 0 D 5 = 290 m g / l o B0D 5= 800 m g / l K = 0.0 o N 'x K * CONSTANT V - - - - 0 T r T 0 4 8 12 16 t - M E A N H Y D R A U L I C R E T E N T I O N T I M E , D a y s 0.80 0.60 0.40 0.30 -\ 0.20 • A COD = 630 m g / l 4 A COD = 1240 m g / l v K # CONSTANT ..A -A 0.15 1 i i i ,i i i i i » 0 4 8 12 16 t - M E A N H Y D R A U L I C R E T E N T I O N T I M E , D a y s . FIGURE II. E V A L U A T I O N OF REACTION RATE CONSTANT K FOR S U B S T R A T E R E M O V A L AT 3 ° C . 0 . 9 0 - 0 . 7 0 - 0 . 5 0 - 0 . 4 0 0 . 3 0 4. 0 . 2 0 0 .15 0 . 9 0 - 0 ,70 - 0 . 5 0 - 0 . 4 0 0 .30 - 0 . 2 0 - 0.15 0 . 9 0 - 0 .70- 0 . 5 0 - 0 .40 - 0 . 3 0 - 0 . 2 0 *. 48 • B O D g = 2 9 0 m g / l S'e/S0 = 0.49e - 0.046t i i 4 - r 8 I F — n - 16 1 n 2 0 t , d o y s o B 0 D 5 = 8 0 0 m g / l - 0 . 0 4 5 t s e / s 0 = 0 . 4 4 e -I 1 1 1 r— 4 8 12 t , d a y s l i l t 16 2 0 s ; / s o s 0 . 7 o e - ° - 0 , 5 t 4 COD = 6 3 0 m g / l - i 1 1 1 1 1 r- 4 8 12 16 t , d a y s — i 2 0 1.0 i 0 . 8 0 . 6 0 . 5 J 0 . 4 0 . 3 4 A 4 A C O D = 1240 m g / l , -,O.OI6t S / S r t = 0.62 e e o i i i • 8 12 t , d a y s 16 2 0 F I G U R E 12. E V A L U A T I O N O F R E A C T I O N R A T E C O N S T A N T K F O R S Y S T E M T R E A T M E N T E F F I C I E N C Y . TABLE 8 MEASURED AND CALCULATED SYSTEM TREATMENT EFFICIENCY , FIRST-ORDER EXPONENTIAL • E - (1 - Ce" k t) x 100% RETENTION TIME (Days) LOADING COD EFFLUENT BOD5 EFFLUENT MEASURED CALCULATED ABSOLUTE DIFFERENCE MEASURED CALCULATED ABSOLUTE DIFFERENCE 1 23.5% 39.0% 14.5% 35.6% 59.0% 23.4% 2 Xi 36.0% 40.0% 4.0% 60.0% 60.5% 0.5% 4 60 •H 40.4% 42.0% 1.6% 61.0% 63.5% 2.5% 8 48.0% 46.0% 2.0% 73.2% 71.0% 2.2% 16 50.5% 51.0% 0.5% 76.6% 78.0% 1.4% 2 31.2% 32.0% 0.8% 18.0% 56.0% 38.0% 4 36.9% 35.0% 1.9% 63.2% 60.0% 3.2% 8 & o 43.3% 41.0% 2.3% 66.0% 68.0% 2.0% 16 44.9% 46.0% 1.1% 80.0% 79.0% 1.0% 50 IV.5 ASB SOLIDS IV.5.1 Solids Production Solids production in ASBs, and the resulting characteristics and concentration of these solids in the effluent stream, are a major factor in the effectiveness of the ASB in producing high quality effluents. The carry-over of solids produced in the basin can significantly deteriorate the quality of the effluent. In this study, there was an increase of as much as 300 mg/l COD or BOD̂  when the solids were included in the effluent measurements. Shown in Figure 13 are the average mixed liquor suspended solids concentrations measured in the model reactors at the two loadings. The ranges of measured concentrations are also shown, as there was considerable variance due to the cycling previously discussed. The similarity between the curves in Figure 13 for hydraulic retention times of two to sixteen days, would suggest that the average values are representative. The trend in the solids concentra- tion with hydraulic retention time i s one of a slightly increasing concentra- tion from two to eight days, followed by a slight decrease to sixteen days. This trend goes against the usual decrease in solids concentration expected with i n - creasing hydraulic retention time (10). Solids production i n a closed biological treatment system i s generally expressed by the equation: Y = aS r - bMLSS (16) where Y = yield or net solids production in lbs/day; a = yield factor, lbs solids/lbs substrate removed; S r = substrate removed, lbs/day; b = endogenous coefficient in %/day; MLSS = mixed liquor suspended solids, lbs. 51 © HIGH L 0 A D I N G - I 2 4 0 mg/i C O D A LOW L O A D I N G - 6 3 0 m g / - f C O D T V A R I A T I O N A B O U T M E A N 1 T 1 r - A - 1. T •A -I 1 1 1 " r - 4 6 8 10 12 14 t - M E A N H Y D R A U L I C R E T E N T I O N T I M E , D o y s 16 FIGURE 13. MIXED LIQUOR SUSPENDED SOLIDS AS A FUNCTION OF MEAN HYDRAULIC RETENTION TIME A T 3 ° C. 52 In a flow-through system without solids recycle, such as the model ASBs used in the laboratory, the hydraulic washout of solids under steady- state conditions is just offset by the biological yield or net solids produc- tion from substrate u t i l i z a t i o n . That i s , net solids production equals the weight of solids washed out. Listed below in Table 9 are the values of net solids production per pound of substrate removal, calculated for the two loadings on a BODu and COD basis. TABLE 9. NET SOLIDS PRODUCTION PER POUND SUBSTRATE REMOVED Low Loading (630 mg/l COD) 0.27 lbs/lb COD 0.25 lbs/lb BOD u High Loading (1,240 mg/l COD) 0.25 lbs/lb COD 0.24 lbs/lb BOD u The values tabled are the calculated average of the data collected from the two to sixteen day reactors. The net solids production per pound of substrate removed, measured in the one-day reactor at the high loading, is 0.48 lbs/lb COD or 0.53 lbs/lb BOD . u Eckenfelder (10) maintains that ASB solids production from a soluble feed w i l l be about 50 per cent of the Influent feed concentration, i.e., 0.5 lbs/ lb BOD,.. However, Gellman 0-7), citing several ASB pilot studies, reports net solids production of 0.10 to 0.25 lbs/lb BOD removed. 53 The solids production measured In the laboratory ASBs is the same for both the high and low loadings and is in line with the values reported by Gellman. The reason for the difference in the net solids pro- duction measured in the one-day reactor, and that measured in the other four reactors, i s not readily apparent. It may possibly be due to a physical agglomeration of the milk solids in the bacteria in the one-day reactor, a l - though there is no evidence to support this. Another poss i b i l i t y may be the difference i n growth conditions in the reactors. The food to micro-organism ratio in the one-day reactor is considerably higher than in the other four. Busch (6) maintains that the yield coefficient, as well as the reaction rate of bacteria, i s dependent on the substrate concentration, and, therefore, at a higher food to micro—organism ratio, the yield or net solids production would be higher. IV.5. 2 COD - BOD„ of ASB'Solids-5 The carry-over of solids in an ASB effluent can contribute s i g n i f i - cantly to the COD or BOD^ of the effluent. This was particularly true for this study. A comparison of the respective curves in Figures 6 and 7 shows a marked difference between the per cent substrate removal and system treatment eff i c i e n - cy, the latter taking into consideration the BOD,, or COD of the effluent solids. The difference between the respective COD or BOD,, curves is a measure of the COD or BOD,, tied up with the biological solids. The respective differences between the measured substrate remaining and the gross effluent concentration, the reactor MLSS,.the calculated COD or BOD5 per unit of MLSS, and the percentage of the e f f l u - ent COD or BOD,, contributed by the residual substrate and by the solids, are summarized in Table 10. TABLE 10 EFFLUENT CHARACTERISTICS OF ASBs AT 3°C 1 i - NOMINAL " EFFL. COD EFFLUENT EFFL. EFFL. BOD EFFLUENT EFFL. RETENTION -EFFL. SUBSTRATE^ SOLIDS/ -EFFL. 5 BOD,./ SUBSTRATE/ SOLIDS/ TIME SUBSTRATE COD/ '•EFFLUENT EFFL. SUBSTRATE EFFLUENT EFFL. (Days) LOADING COD MLSS MLSS COD % COD % BOD5 MLSS MLSS BOD % 5 BOD,. : % 2 201 mg/l 104 mg/l 1.93 54% 46% 160 mg/l 104 mg/l 1.54 32% 68% 4 194 mg/l 108 mg/l 1.80 52% 48% 118 mg/l 108 mg/l 1.07 11% 89% 8 230 mg/l 116 mg/l 1.97 37% 63% 95 mg/l 116 mg/l 0.82 13% 87% 16 o •J 235 mg/l 112 mg/l 2.10 35% 65% 60 mg/l 112 mg/l 0.54 19% 81% 1 522 mg/l 385 mg/l 1.35 45% 55% 300 mg/l 385 mg/l 0.78 40% 60% 2 459 mg/l 226 mg/l 2.03 42% 58% 300 mg/l 226 mg/l 1.33 6% 94% 4 X 435 mg/l 232 mg/l 1.88 41% 59% 270 mg/l 232 mg/l 1.16 7% 93% 8 60 •H '368 mg/l 244 mg/l 1.51 43% 57% 195 mg/l 244 mg/l 0.80 9% 91% 16 W 320 mg/l 224 mg/l 1.43 47% 53% 140 mg/l 224 mg/l 0.65 11% 89% U l 55 The data in Table 10 shows that the major portion of the effluent BOD̂  i s due to biological solids concentrations. Both Eckenfelder (11) and Goodman (19) show that the major portion of the effluent BOD,, w i l l be con- tributed by the solids in the effluent. On the other hand, a study (28) of a number of ASBs treating pulp and paper wastes at various retention times found that only 30 per cent of the effluent BOD̂  was contributed by suspended solids. The B0D5 per unit of MLSS (0.54 - 1,54 lbs B0D5/lb MLSS) liste d in Table 10 are considerably higher than expected. Eckenfelder (10) shows a range of values from 0.75 to 0.30 lbs BOD^/lb MLSS for sludge ages of 0 to seven days. However, in his design models he has used a factor as low as 0.25 lbs B0D5/lb MLSS (11). Goodman (19) uses a value of 0.60 lbs BOD^/lb active solids for a l l temperatures and sludge ages. Gellman (17) quotes a number of investigators who found ratios of 0.10 - 0.26 lbs B0D5/lb solids for wastes treated at 20° to 35°C, as compared to 0.54 - 1.54 lbs B0D5/lb MLSS found in this study at 3°C. A direct comparison of values may, however, be misleading, as i t is possible that influent solids are incorporated in the values reported by Gellman. With the exception of the data for the one day reactor (high loading), the BOD̂  per unit of MLSS decreases with an increasing retention time. This would be expected i f endogenous oxidation was occurring. A similar trend exists on a COD basis at the high loading, whereas at the low loading the ratio i s nearly constant. It would appear that the pounds BOD̂  per pound MLSS ratios measured in this study at 3°C are significantly higher than ratios measured at higher temperatures. It is also evident from the data presented in Table 10 56 that the major portion of the effluent BOD̂ -COD is due to the presence of biological solids. Therefore, i t could be argued that the decrease in treat- ment efficiency with decreasing temperature, as reported in other studies, is due largely to an increase in the COD or BOD,, of the biological solids, possib- ly through a decrease in the endogenous oxidation rate. IV.5.3 Settling at 3°C . In order to reduce the solids concentration in ASB effluents, a common practice has been to follow an ASB with a settling basin. In conjunction with the settling of solids, system treatment e f f i c i - encies have been reported to increase by two to 10 per cent (7) (37) (46). The results of the batch settling tests conducted in this study would seem to be somewhat unusual, in that the improvement in system treatment efficiency with settling (COD basis) was much greater than indi - cated by the literature. System treatment efficiency improved by as much as 62 per cent (from 23 per cent to 85 per cent) in the one-day ASB to a low of 14 per cent (from 51 per cent to 65 per cent) in the sixteen day reactor with fiv e days' settling time. This can be seen by comparing the gross COD in the supernatant, as system treatment efficiency plotted against settling time, with the superimposed curve of COD removal with aeration time (Figures 14 and 15 respectively) for the low and high loading. F I G U R E 14. P E R C E N T C O D R E M O V A L W I T H S E T T L I N G T I M E - 3 ° C - L O W L O A D I N G 9 0 i co 20 J 1 1 — 1 1 1 : 1 1 r- — i r-^V i — i 2 3 4 5 6 7 8 9 10 16 t - MEAN HYDRAULIC RETENTION OR SETTLING TIME, DAYS FIGURE 15. PERCENT COD REMOVAL WITH SETTLING TIME - HIGH LOADING" 59 The effect of aeration time on the settling rate does not follow any trend. However, the effluent from the reactors with the highest applied load for the two loadings responded most rapidly, and had the greatest overall improvement in system treatment efficiency (62 per cent for the one-day reactor at high loading and 47 per cent for the two-day reactor at low loading) after five days settling. A comparison of Figure 15 with Figure 16, which i s a plot of supernatant MLSS against settling time at the high loading, shows a general trend of decreasing COD with decreasing suspended solids. A numerical evaluation of COD removal i n terms of MLSS settled, was not performed, as the MLSS numbers must be considered suspect. The very low MLSS levels, and the necessity of taking small samples (40 mis) so as not to unduly affect the settling test, resulted in very small weighing d i f - ferences after one day (0.0 to 1.3 mg). The substantial increase in system treatment efficiency with settling, relative to the two to 15 per cent reported (7) (35)(46), can be explained in part by the higher COD or BOD,, per unit of MLSS found in this study. The increase may also be due in part to a possible change in the settling characteristics of ASB biological solids at low temperature. This, however, would have to be substantiated by future studies. Referring to Figures 15 and 16, and considering both MLSS and gross substrate removal, i t can be seen that a number of combinations are available to give a desired removal efficiency and suspended solids level. For example, an effluent quality for 1200 4 0 0 • 60 CO CO < < z or U J a. 3 / CO 3 0 0 - ,A I DAY • 2 DAY - A 4 DAY 200- o 8 DAY + 16 DAY 100- F I G U R E 16. I 2 S E T T L I N G TIME , DAYS S U P E R N A T A N T M L S S VS S E T T L I N G T I M E 3 ° C - HIGH L O A D I N G 61 mg/l feed, of 60 mg/l mixed liquor suspended solids and a system treatment efficiency of 70 per cent can be achieved by: (a) four days'aeration with one-half days' settling; (b) one day's aeration with 2% days' settling; (c) eight days' aeration with 2% days' settling. In contrast, the desired effluent quality could not be achieved by using any combinations of the two and sixteen day aerations. Therefore, in terms of both percentage removal and suspended solids l e v e l , the effluent quality does not necessarily reflect the length of retention t i m e — aeration or settling, but is a function of both. It i s not the intent of this discussion to suggest that the same results would be found for other wastes, or even for this waste, at a temper- ature other than 3°C. Rather, i t is suggested that when there are effluent suspended solids restrictions and settling f a c i l i t i e s are required, the design of the system should take into consideration the relationship between settling and gross substrate removal and, In turn, settling time with aeration time. IV.6 NITROGEN STUDIES IV.6.1 General. Analysis for the various nitrogen compounds was under- taken to find: (a) i f n i t r i f i c a t i o n was occurring at 3°C; (b) the amount of nitrogen in the solids; and (c) i f the retention time had any effect on the balance of nitrogen compounds. Sufficient nitrogen was available in the feed to ensure that i t would not be growth limiting. The presence of NĤ  nitrogen in the reactors substantiates this. IV.6.2 Nitrate Nitrogen. The concentration of nitrate nitrogen in the reactors varied from 0.2 mg/l to 0.6 mg/l, with the majority of the samples f a l l i n g between 0.3 mg/l and 0.5 mg/l. No apparent relation exists between 62 the nitrate concentration and retention time, loading, or test duration. The feed water contained about 0.4 mg/l nitrate nitrogen and i t is l i k e l y that this is the major source of nitrate nitrogen in the reactors. Due to the high back- ground level i t was impossible to determine whether or not n i t r i f i c a t i o n was taking place. Wild et al. (44), studying n i t r i f i c a t i o n kinetics, documented o very reduced n i t r i f i c a t i o n at 5 C with a decrease in activity with decreasing temperature. IV.6.3 Nitrogen Balance. The gross kjeldahl nitrogen or total nitrogen in the system, ignoring the nitrate concentration, averaged 29.6 mg/l at the low loading and 66.4 mg/l at the high loading and in each case was within ex- perimental limits of the feed. Gross nitrogen concentrations measured in the reactors at the two loadings are lis t e d in Table 11. Plotted in Figure 17 against retention time are the concentrations of the f i l t e r e d organic and NĤ  nitrogen, and the concentrations of the nitro- gen tied up with the solids for the high loading. As can be seen from Figure 17, the concentration of nitrogen in the solids decreases with retention time and i s reflected by an increase in the f i l t e r e d NĤ  concentration. The organic nitrogen concentration remains constant with retention time, which suggests that bacteria capable of converting organic nitrogen to NĤ  are well established in each reactor. A rigorous explanation as to why the proportion of nitrogen tied up with the solids decreases with retention time is not available. How- ever, the increase in NĤ  nitrogen concentration in the reactors, without a change in the organic nitrogen concentration, would suggest that the decrease in nitrogen in the bacterial solids i s due to a release of NĤ  through endogen- ous oxidation of the solids. 63 TABLE 11 GROSS KJELDAHL NITROGEN CONCENTRATIONS Low Loading - Feed Concentration 30.3 mg/l: RETENTION TIME GROSS CONCENTRATION (Days) 2 29.4 mg/l 4 29.2 mg/l 8 30.4 mg/l 16 29.4 mg/l High Loading - Feed Concentration 66.0 mg/l: RETENTION TIME GROSS CONCENTRATION (Days) / 1 66.3 mg/l 2 65.5 mg/l 4 66.6 mg/l 8 67.1 mg/l 16 66.6 mg/l The nitrogen data for the low loading, as can be seen in Table 12, does not show any relationship to retention time. TABLE 12 AVERAGE NITROGEN CONCENTRATIONS - LOW LOADING RETENTION GROSS KJELDAHL NH3 INORGANIC .TIME. NITROGEN NITROGEN NITROGEN (Days) 2 29.0 mg/l 4.1 mg/l 4.8 mg/l 4 29.6 mg/l 4.9 mg/l 5.8 mg/l 8 30.4 mg/l 3.4 mg/l 5.3 mg/l 16 29.4 mg/l 3.3 mg/l 9.1 mg/l .TOTAL NITROGEN CONCENTRATION IN SOLIDS , N H 3 CONCENTRATION IN F ILTRATE ORGANIC NITROGEN 'IN FILTRATE 1 i l i i I 1 I i r 2 4 6 8 10 12 14 16 18 MEAN HYDRAULIC R E T E N T I O N T I M E , days F I G U R E 17 C O N C E N T R A T I O N O F N I T R O G E N C O M P O U N D S I N T H E R E A C T O R S O L I D S A N D F I L T R A T E A S A F U N C T I O N O F M E A N H Y D R A U L I C R E T E N T I O N T I M E ( H I G H L O A D I N G ) . 65 The nitrogen used per hundred pounds of BOD̂  applied measured in this study varied from 8.4 to 6.7 pounds at the low loading and 8.0 to 5.4 pounds at the high loading for hydraulic retention times of one to sixteen days. It should be noted that Ludzack et at. (25) found 5.6 to eight per cent nitrogen in activated sludge vo l a t i l e suspended solids at 5°C. These values are approximately twice the nitrogen addition per hundred pounds of BOD̂ removed reported for ASBs operating at temperatures of 20°C to 35°C (13)(7)(5) (1). This higher nitrogen requirement may be related to the higher COD and BOD of the biological solids measured in this study. However, i t is quite possibly just an example of excess nutrient uptake by bacterial solids when there i s a surplus of nutrients. IV.7 pH The pH of the reactor contents were nearly constant and remained slightly basic at 7.0 to 7.2. IV.8 OXYGEN UTILIZATION Oxygen uptake rates of 0.5 mg/1 per hour in the sixteen day reactor to 5.3 mg/l per hour in the one day reactor were recorded at the high loading (see Figure 18). Oxygen u t i l i z a t i o n in biological systems i s often expressed by the following equation: 0 2lbs/day = a'S rlbs/day - b'MLSS lbs/day (17) where a' = oxygen u t i l i z a t i o n coefficient, lbs 0 2 used/lb substrate removed; S = substrate removed, lbs/day; b^ = endogenous respiration coefficient, %/day; MLSS = mixed liquor suspended solids, lbs. F I G U R E 1 8 . O X Y G E N U P T A K E A G A I N S T R E T E N T I O N T I M E 67 Figure 19, a plot of lbs oxygen used/lb MLSS against lbs substrate used/lb MLSS, yields a' as the slope and b' as the intercept. On a BOD,, basis, a' is 0.143 lbs 02/lb substrate used in the two to sixteen day reac- tors. On a COD basis a' i s 0.123 lbs 0 2/lb substrate used in the two to sixteen day reactors and 0.156 lbs 02/lb substrate used in the one day reactor. The endogenous respiration rate, b', i s 0.18 lbs 02/lb MLSS per day, or 0.75 mg/hr per gram of MLSS in both cases — as would be expected. The endogenous respiration rate of 0.75 mg/hr per gram MLSS, or approximately 0.95 mg/hr per gram VSS (assuming VSS = 0.8 MLSS) is consider- ably lower than many reported values. Symons (10) reported a mean rate of 15 mg/hr per gram VSS for mixed sludges at room temperature. Porges (10) reported a rate of 12 mg/hr per gram VSS for dairy wastes. However, the rate is comparable to the 0.80 mg/hr per gram VSS respiration rate found by Esvelt et al. (14) for f r u i t processing wastes treated in an ASB at 6°C. The depressed rate of endogenous respiration would indicate that endogenous oxidation i s much slower at 3°C than at higher temperatures. Shown in Figure 20 i s a linear relationship between the f i l t e r e d COD concentration remaining and the oxygen uptake rates. A comparison of Figures 19 and 20 would suggest that relationship between the oxygen up- take rate and the substrate remaining i s lik e l y better for the short retention times than the relationship between oxygen consumed and substrate removed. 68 0.36-8 * 0.30- CO s 0.20- < a cc UJ CL O Ul 2 3 tO z o o z u i CIO o >-X o A I day R.T. a =0143 lb.0 2 /C0Drem 0 A 2 days RT. av= 0.123 lb 0 2 / l b CODrem A 4 days R.T. o A 8 days R.T. o A 16 doys R.T. O C 0 D - I 2 4 0 m g / ^ A B 0 D 5 - 8 0 0 m g / - E b 1 = 0.018 lbs 0 2 / M L S S - d a y T " 1 I r 0.0 1.0 2.0 lbs. SUBSTRATE REMOVED PER DAY / lb. MLSS FIGURE 19. OXYGEN CONSUMPTION PER DAY AS A FUNCTION OF SUBSTRATE REMOVED PER DAY - HIGH LOADING. 69 6.04 5-0-1 E 4,0 • LU < tr < 3.0 o_ z> z UJ o > o 2,0 1.0 i————r — i i i 0 100 200 300 400 500 R E A C T O R C O D S U B S T R A T E C O N C E N T R A T I O N , mg / 4 FIGURE 20. OXYGEN U P T A K E RATE AS A FUNCTION OF REACTOR S U B S T R A T E CONCENTRATION CHAPTER V SUMMARY The primary objectives of this study, the detailing of ASB operation at a cold temperature and the evaluation of that operation in terms of current ASB models, were f u l f i l l e d . Summarized below are the results of the study. V.l STEADY STATE Steady state conditions were achieved at both loadings twenty-seven to twenty-nine days after the runs were initiated. Cyclic fluctuations of the f i l t e r e d COD concentrations and, to a lesser extent, MLSS levels were noted at the higher loading. V.2 PER CENT SUBSTRATE REMOVAL AND SYSTEM TREATMENT EFFICIENCY 1̂  Seventy-five to 85 per cent of the biodegradable portion of the raw milk waste was u t i l i z e d in the ASBs within one to two days, and there was complete u t i l i z a t i o n of the biodegradable portion of the waste by eight days. The fi l t e r e d COD data indicated the possible production of intermediate compounds which were ut i l i z e d by about eight days at both loadings. 2* The system treatment efficiency varied with retention time from 18 per cent at two days to 80 per cent at 16 days on a B0D5 basis and from 23 per cent at one day to 51 per cent at 16 days on a COD basis. 3c The per cent substrate removal measured was higher than reported treatment efficiencies and i s l i k e l y a different measurement. On the other hand, 70 71 the measured system treatment efficiencies were comparable to reported treat- ment efficiencies in ASBs operating at low temperature. V.3 MODEL EVALUATION 1. Only the Chemostat model would describe the substrate removal data measured in the laboratory ASBs at 3°C. 2. O'Connor and Eckenfelder's, McKinney's, and the first-order exponen- t i a l models could not describe the substrate removal measured at 3°C. This i n - a b i l i t y to describe the substrate removal at 3°C could conceivably lessen the usefulness of these models. 3. O'Connor and Eckenfelder's, McKinney's, and the first-order exponential models would describe system treatment efficiency for mean hydraulic retention times greater than two days. These models appear to be applicable where endo- genous oxidation i s the main mechanism of BOD,- or COD decay, but, on the basis of this study, are not applicable under growth conditions. V.4 SOLIDS PRODUCTION 1. Net solids production in the two to sixteen day ASBs was 0.25 - 0.27 lbs per pound of COD or BOD ultimate removed. In the one-day reactor, the net yield was higher, at 0.48 lbs per pound COD and 0.53 pounds per pound BOD u l t i - mate. The net solids production was not significantly different from that re- ported in other ASB studies. 2„ The COD and BOD̂  of the generated solids were significantly higher than reported values at higher temperatures; 1.35 - 2.10 mg COD per mg MLSS and 0.54 - 1.54 mg BOD̂  per mg MLSS. It is conceivable, although not proven, that the change in the characteristics of the generated solids may account in 72 part for the change in treatment efficiency reported with changes in. the temperature of ASBs. 3. Post settling at 3°C of the ASB effluent resulted in signi- ficant increases (14% - 62%) in system treatment efficiency in addition to the removal of suspended solids. These increases in system treatment e f f i - ciency may be due to changes in settling characteristics as well as to the higher COD and BOD5 of the MLSS measured at 3°C. V.5 NITROGEN USAGE 1. N i t r i f i c a t i o n did not appear to be significant at 3°C. 2. The concentration of nitrogen compounds in the reactor solids decreased with retention time at the high loading. This decrease was re- flected by an increase in the NĤ  nitrogen concentration in the effluent with increased retention time. The organic nitrogen concentration remains constant, suggesting endogenous oxidation as a possible mechanism for the release of NĤ  nitrogen. V.6 pH The pH in the reactors was slightly basic at 7.0 to 7.2. V.7 OXYGEN UPTAKE 1. The oxygen uptake rates at the high loading varied from 5.25 mg/l/hr to 0.50 mg/l/hr for retention times of one to sixteen days. 2. The endogenous respiration rate was found to be 0.18 mg/l/hr or 0.75 mg/hr/gram MLSS. 73 3. Oxygen u t i l i z a t i o n was related linearly to substrate removed for retention times of two to sixteen days — 0.143 lbs O2 used/lb BOD,, used and 0.123 lbs 0 2 used/lb COD used. 4. The oxygen uptake rate was related linearly to the COD substrate concentration remaining in the reactors for retention times of one to sixteen days. CHAPTER VI CONCLUSIONS 1. At 3°C twenty-seven to twenty-nine days were required at both loadings to achieve steady-state conditions in the reactors. 2. Seventy-five to 85 per cent of the biodegradable portion of raw milk was uti l i z e d within one to two days and there was virtua l l y complete u t i l i z a t i o n within eight days. 3. System treatment efficiency varied from 18 to 80 per cent on a BOD,, basis and 23 to 51 per cent on a COD basis. 4. The per cent substrate removal measured was higher than reported treatment efficiencies and i s l i k e l y a different measurement. System treat- ment efficiency i s comparable to and i s assumed to be the same as reported treatment efficiencies. 5. The Chemostat method described the substrate removal data mea- sured in the laboratory ASBs at 3°C. O'Connor and Eckenfelder's, McKinney's and the first-order exponential models could not describe the substrate removal measured at 3°C. This i n a b i l i t y to describe the substrate removal could con- ceivably lessen the usefulness of these models. 74 75 6. O'Connor and Eckenfelder's, McKinney's and the first-order ex- ponential models would describe system treatment efficiency for mean hydraulic retention times greater than two days. These models appear to be applicable where endogenous oxidation i s the main mechanism of BOD̂  or COD decay, but on the basis of this study are not applicable under growth conditions. 7. Net solids production for retention times of two days or greater is 25 to 27 per cent of the removed COD or BOD . For a retention time of one u day net solids production i s about 50 per cent of the removed COD or BOD̂ . 8. The COD and BOD,, of the ASB solids were significantly higher than reported values at higher temperatures. It is conceivable although not proven, that the change in the characteristics of the generated solids may account in part for the changes in treatment efficiency reported with changes in the temp- erature of ASBs. The higher COD and B0D5 of the solids may be due to a decrease in the rate of endogenous oxidation. 9. Post settling of the ASB effluent at 3°C resulted in significant increases (14%-62%) i n system treatment efficiencies. The most significant i n - creases occurred at the shortest retention times and would suggest that dollars spent on increasing aeration time in response to expected cold temperatures, would be better spent on settling lagoons with a return of the solids for diges- tion during the warmer months. This point requires further investigation. 10. N i t r i f i c a t i o n did not appear to be significant at 3°C. The concen- tration of Kjeldahl nitrogen i n the ASB solids decreased with retention time at 76 the high loading. This decrease was reflected by an increase in the NĤ  nitro- gen concentration in the reactor f i l t r a t e . The organic nitrogen concentration in the f i l t r a t e remained constant suggesting endogenous oxidation as a possible mechanism for the release of NĤ  nitrogen. 11. The endogenous respiration rate was found to be 0.75 mg/hr/gram MLSS which supports the conclusion of the reduced rate of endogenous oxidation of solids at 3°C. 12. Oxygen u t i l i z a t i o n was related linearly to substrate removed for retention times of two to sixteen days - 0.143 lbs 02/lb BOD,, removed or 0.123 lbs O2 used/lb COD used. The oxygen uptake rate was also related linearly to the COD substrate concentrate remaining in the reactors for retention times of one to sixteen days. In addition the curve showing oxygen uptake rate as a function of hydraulic retention time takes the same form as the Chemostat model.which follows from the relationship to substrate concentration. CHAPTER VII RECOMMENDATIONS 1. The high levels of COD and BOD̂  per unit of suspended solids and the low endogenous respiration rate found in this study suggest that the decrease in system treatment efficiency at low temperatures may be related in part to the make-up of the solids produced and to endogenous oxidation. To prove this supposition, a continuation of this study at at least two other temp- eratures, preferably 8°C and 15°C i s recommended. In addition, another two reactors with retention times of % and 3/4 days could be added to verify the Chemostat model. Particular attention should be paid to improving MLSS deter- mination, possibly using a power f i l t e r . In addition to the tests conducted in this study, long-term oxidation studies are recommended in order to establish the temperature effect on endogenous oxidation. 2. Due to the i n a b i l i t y of the current ASB models, with the exception of the Chemostat, to describe the substrate removal measured i n this study, a complete evaluation of these models should be undertaken. The evaluation should take into consideration both substrate removal and system treatment efficiency over a range of temperatures. 3. In view of the practical implications of the settling tests, fur- ther and more comprehensive studies are recommended. These studies should i n - clude settling tests on the effluent from f i e l d installations, as well as from laboratory scale basins, over a range of retention times and temperatures. 77 78 4. A repetition of the tests using several industrial wastes i s also recommended, in order to reinforce any conclusions drawn from the milk waste study. B I B L I O G R A P H Y Amberg, H. R., J. H. Pritchard and D. W. Wise, "Supplementary > Aeration of Oxidation Lagoons With Surface Aerators," TAPPI3 47 (10) 27A (1964). Barnhart, E. L. "Treatment of Chemical Wastes in Aerated Lagoons," Chem. Eng. Prog. Symp. Ser. 643 903 111 (1968). Bartsch, E. H., and C. W. Randall. "Aerated Lagoons — A Report on the State of the Art," Jour. Water Poll. Control Federation, 433 699 (1971). Besselievre, P. E. The Treatment of I n d u s t r i a l Wastes. New York: McGraw-Hill Book Co.,(1969) Blosser, R. 0. "Oxidation Pond Study for Treatment of De-Inking Wastes," Proc. 16th Ind. Waste Conf.3 Purdue University (1961). Busch, A. W. Aerobic B i o l o g i c a l Treatment of Waste Water. Houston Texas: Oligodynamics Press (1971). Carpenter, W. L., James G. Vanuakias and I. Gellman. "Temperature Relationships in Aerobic Treatment and Disposal of Pulp and Paper Wastes," Jour. Water Poll. Control Federation3 403 733 (1968). Dawson, R. N., and J. W. Grange. "Proposed Design Criteria for Waste Water Lagoons in Arctic and Sub-Arctic Regions," Jour. Water Poll. Control Federation3 413 237 (1969). Eckenfelder, W. W., and A. J. Englande. "Temperature E f f e c t s on B i o l o g i c a l Waste Treatment Process," Inter. Symp. on Water Poll. Control in Cold Climates3 pp. 180-190 (1971). Eckenfelder, W. W. Water Quality Engineering. New York: Barnes and Noble Inc. (1970). Eckenfelder, W. W., and Davis L. Ford. Laboratory and Design Pro- cedures for Waste Water Treatment Process, University of Texas Pres Tech. Report, EHE-10-6802, Austin, Texas (1968). 80 Eckenfelder, W. W. I n d u s t r i a l Water P o l l u t i o n Control. New York: McGraw-Hill Inc. (1966). Eckenfelder, W. W. "Design and Performance of Aerated Lagoons for Pulp and Paper Waste Treatment," Proc. 16th Ind. Waste Conf., Purdue University, 115 (1961). Esvelt, L. A. and H. H. Hart. "Treatment of Fruit Processing Wastes by Aeration," Jour. Water Poll. Control Federation, 42, 1305 (1970). Gaudy, A. F., Jr., P. Y. Yang, and A. W. Obayoshi. "Studies in the Total Oxidation of Activated Sludge, With and Without Hydraulic Pre- treatment," Jour. Water Poll. Control Federation, 43, 40 (1971). Gehm, H. W. "The Application of Stabilization Ponds in the P u r i f i - cation of Pulp and Paper M i l l Wastes," Jour. Water Poll. Control Federation, 35, 1174 (1963). Gellman, S. B. "Aerated Stabilization Basin Treatment of M i l l Effluent," TAPPI, 48, 6, 106A (1965). Gloyna, E. F., S. 0. Brady, and H. Lyles. "Use of Aerated Lagoons and Ponds in Refining and Chemical Waste Treatment," Jour. Water Poll. Control Federation, 41, 429 (1969). Goodman, B. L. Design Handbook of Wastewater Systems: Domestic, In- dustrial, Commercial. Westport, Conn.: Technomic Publishing Co. (1971).. Goodrow, W. E. "Regina Aerates i t s Sewage Lagoon," Water and Pollu- tion Control, 104, 12, 17 (1966). Hoover, S. R., L. Josewicz and W. Porges. "Biochemical Oxidation of Dairy Wastes," Proc. 9th Ind. Waste Conf., Purdue University (1954). Lawrence, A. W., and P. L. McCarty. "Unified Basis for Biological Treatment Design and Operation," Jour. Sanitary Engineering Division, ASCE, 96, No. SA3, 757 (1970). 81 Lee, W. C. "Oxidation Ponds and Aerated Lagoons — Some Practical Aspects," Can. Jour. Pub. Health, 60, 435 (1969). Ling, J. T., "Pilot Study of Treating Chemical Wastes With An Aerated Lagoon," Jour. Water Poll, Control Federation, 35, 963 (1963). Ludzack, F. J., R. B. Schaffer and M. B. Ellinger. "Temperature and Feed as Variables in Activated Sludge Performance," Jour. Water Poll. Control Federation, 33, 141 (1961). Mancini, J. L., and F. L. Barnhart. "Industrial Waste Treatment in Aerated Lagoons," Advances in Water Quality Improvement, Vol. I, University of Texas Press, Austin, Texas, pp. 313-324 (1968). McKinney, R. E. Microbiology for Sanitary Engineers. New York: McGraw-Hill Inc. (1962). NCASI, "A Study of Mixing Characteristics of Aerated Stabilization," Stream Improvement Technical B u l l e i n §245, NCASI Inc., New York (1971). Nemerow, N. L. Liquid Wastes of Industry: Theories, Practices and Treatment. Reading, Mass.: Addison Wesley Publishing Co. (1971). Novick, A., and L. Sziland. "Experiments with the Chemostat on Spon- taneous Mutations of Bacteria," Proc. Nat. Acad. Sci., Wash., 36, 708 (1950). O'Conner, D. J., and W. W. Eckenfelder, Jr. "Treatment of Organic Wastes in Aerated Lagoons," Jour. Water Poll. Control Federation, 32, 365 (1960). Olsen, R. H., and J. Jezeski. "Some Effects of Carbon Source, Aeration, and Temperature on Growth of A Psychrophylic Strain of Pseudomona: Flourescens," Jour. Bacteriol, 86, 429 (1963). Quirk, T. "Aerated Stabilization Pond Treatment of White Water," Water and Wastewater Eng., 6, p. 1 (1969). Reid, L. C , Jr. The Aerated Sewage Lagoon in A r c t i c Alaska, Arctic Health Research Center, Env. Eng. Sec. Anchorage, Alaska. 82 Rice, W. D., and R. F. Weston. "Bio-Treatment Design for Pulp- Paper Wastes," Proc. 16th Ind. Waste Conf., Purdue University, 461 (1961). Robinson, W. E. "Wastewater Treatment in Kraft M i l l s , " Chem. Eng. Progr., 65, 6, 78 (1968). Sawyer, C. N. "New Concepts in Aerated Lagoon Design and Operation," Advances in Water Quality Improvement, Vol. I, University of Texas Press, Austin, Texas, pp. 325-335 (1968). Scheltinga, H. M. J., "Discussion Paper," XVII Int. Dairy Congress, Munich, 1966, pp. 483-487. Shih, Chia Shun, and V. T. Stack, Jr. "Temperature Effects on Energy Oxygen Requirements in Biological Oxidation," Jour. Water Poll. Con- trol Federation, 41, R 463 (1969). Standard Methods for the Examination of Water and Wastewater,• 13th Ed. New York: American Public Health Assoc. Inc. (1972). Thirmurthi, D. "Design Principles of Waste Stabilization Ponds," Jour. Sanitary Engineering Division, ASCE, 95, No. SA2, 311 (1969). Timpany, P. L., L. E. Harris and K. L. Murphy. Cold Weather Operation in Aerated Lagoons Treating Pulp and Paper Wastes, T. W. Beak Con- sultants Ltd. Townshead, A. R., Sahi Unsal, Boris I. Boyko, "Aerated Lagoon Design Methods: An Evaluation Based on Ontario Field Data," Proc. 24th Ind. Waste Conf., Purdue University, 327 (1969). Wildt, H. F., C. N. Sawyer and T. C. McMohen. "Factors Affecting N i t r i f i c a t i o n Kinetics," Jour. Water Poll. Control Federation, 42, 1845 (1971). Williams, S. W. Jr., and G. A. Hotto, Jr. "Treatment of Textile M i l l Wastes in Aerated Lagoons," Proc. 16th Ind. Waste Conf., Purdue Univer- sity, 518 (1961). 83 (46) White, M. T. "Surface Aeration As A Secondary Treatment System," TAPPI, 48, 10 128A (1965). APPENDIX A EXPERIMENTAL DATA 85 TABLE A-l FILTERED COD RESULTS (630 mg/l feed) DATE 2 DAY 4 DAY 8 DAY 16 DAY COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL 10/7/72 226 62.2 108 . 82.0 97 84.0 12/7/72 241 60.0 241 60.0 133 80.0 105 82.7 14/7/72 254 59.7 206 67.3 128 79.7 190 70.0 18/7/72 224 65.5 201 69.0 147 77.4 105 83.7 24/7/72* 340 42.0 250 57.5 222 62.2 257 56.5 AVERAGES 236 61.2 216 65.4 129 79.8 125 80.0 * Temperature room failed. 86 TABLE A-2 GROSS COD RESULTS (630 mg/l feed) DATE 2 DAY 4 DAY 8 DAY 16 DAY COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL 10/7/72 415 43.2 383 39.2 353 44.0 337 47.5 12/7/72 440 26.5 381 36.5 370 40.0 381 36.5 14/7/72 428 30.0 393 35.. 5 356 45.0 356 45.0 18/7/72 464 29.0 430 34.0 364 44.0 351 46.0 21/7/72 410 33.0 388 37.0 350 43.0 327 47.0 24/7/72 386 34.5 355 39.5 333 43.5 310 47.3 AVERAGES 31.2 36.9 43.3 44.9 87 T A B L E A-3 F I L T E R E D BOD RESULTS (290 m g / l f e e d ) DATE 2 DAY 4 DAY 8 DAY 16 DAY BOD % B O D 5 % B O D 5 % BOD5 % mg/I REMOVAL m g / l REMOVAL m g / l REMOVAL m g / l REMOVAL 6/7/82 — — 9 97.0 18 94.0 12 96.0 13/7/72* 140 52.0 61 79.0 29 90.4 23 92.0 21/7/72 55 81.0 15 94.8 15 94.8 15 94.8 AVERAGES 81.0 96.0 94.4 95.4 * C o n s i d e r e d s u s p e c t ; n o t i n . a v e r a g e T A B L E A-4 GROSS B O D 5 RESULTS (290 m g / l f e e d ) DATE 2 DAY 4 DAY 8 DAY 16 DAY BOD5 m g / l % REMOVAL BOD5 m g / l % REMOVAL BOD5 m g / l % REMOVAL BOD5 m g / l % REMOVAL 7/7/72* 120 58.5 118 59.5 119 59.0 93 68.0 13/7/72 242 16.5 102 65.5 120 58.5 35 88.0 21/7/72 233 19.5 108 62.5 58 80.0 49 83.0 AVERAGES 238 18.0 109 63.2 99 66.0 42 80.0 2 Day p o i n t q u e s t i o n a b l e 88 TABLE A-5 MLSS DATA mg/l (600 mg/l milk feed) D A T E 2 DAY 4 DAY . 8 DAY 16 DAY 3/7/72 104 111 108 105 10/7/72 112 140 123 120 17/7/72 102 128 127 122 21/7/72 111 90 - 96 24/7/72* 152 137 101 "103 Failure of temperature room TABLE A-6 NITROGEN DATA mg/l (600 mg/l milk feed) DATE 2 DAY 4 DAY 8 DAY 16 DAY 10/7/72 30.0 30.8 29.2 GROSS 14/7/72 29.7 29.7 30.0 30.3 KJELDAHL 18/7/72 28.3 29.0 30.5 28.6 NITROGEN AVERAGE 29.0 29.6 30.4 29.4 10/7/72 1.9 6.2 3.9 2.2 NH 14/7/72 5.9 3.3 3.4 3.1 NITROGEN 18/7/72 4.5 3.1 2.8 4.5 AVERAGE 4.1 4.9 3.4 3.3 10/7/72 3.4 2.8 5.3 5.6 INORGANIC 14/7/72 7.3 5.9 - 10.6 NITROGEN 18/7/72 3.9 8.6 5.3 11.2 AVERAGE 4.8 5.8 5.3 9.1 89 TABLE A-7 SETTLING DATA (630 mg/l COD feed) SETTLING TIME 2 DAY EFFLUENT 4 DAY EFFLUENT 8 DAY EFFLUENT 16 DAY EFFLUENT COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL 2 DAY 197 68.8 213 66.0 330 47.7 242 61.7 3 DAY 166 73.6 242 61.7 159 75.0 229 63.7 6 DAY 146 77.0 191 70.0 159 75.0 185 70.5 10 DAY 140 78.0 166 73.8 121 80.8 159 75.0 20 DAY 85 86.5 130 79.2 68 89.0 123 80.5 TABLE A-8 FILTERED COD DATA (1240 mg/l COD feed) DATE 1 DAY 2 DAY 4 DAY 8 DAY 16 DAY COD % COD % COD % COD % COD % mg/l REMOVAL mg/l REMOVAL mg/l REMOVAL mg/l REMOVAL mg/l REMOVAL 25/8/72 478 61.7 237 74.3 403 68.2 221 82.7 299 76.6 28/8/72 486 61.2 320 68.9 217 82.6 — — 312 75.0 30/8/72 474 62.1 388 64.5 306 74.6 140 88.8 342 72.6 1/9/72 457 63.7 442 72.8 466 62.5 202 83.8 213 82.7 3/9/72 455 64.0 332 78.7 248 79.6 316 74.0 301 75.2 8/9/72 348 71.2 256 78.9 248 79.7 310 74.2 233 80.4 11/9/72 402 68.3 258 78.9 205 83.2 455 62.7 311 74.5 14/9/72 374 69.4 253 79.2 298 " 75.4 351 71.0 298 75.4 AVERAGE 438 65.2 311 74.5 300 75.7 285 76.8 288 76.5 TABLE A-9 GROSS COD DATA (1240 mg/l COD feed); DATE 1 DAY 2 DAY 4 DAY 8 DAY 16 DAY COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL COD mg/l % REMOVAL 25/8/72 990 20.5 764 40.4 695 45.5 602 51.7 595 53.5 28/8/72 1045 16.5 786 37.0 740 40.5 635 49.0 635 49.0 30/8/72 980 21.5 783 37.4 769 38.5 644 48.5 590 52.8 1/9/72 922 26.6 798 35.8 737 40.7 660 46.8 621 50.0 3/9/72 920 27.1 780 35.8 750 38.3 634 47.9 618 49.1 8/9/72 1135 14.0 805 33.5 736 39.0 645 46.5 619 48.6 11/9/72 880 30.6 796 34.8 713 41.7 637 47.8 607 50.2 14/5/72 857 29.2 789 35.0 728 40.0 637 47.5 615 49.4 18/9/72 915 25.5 788 34.7 750 39.6 659 46.0 609 50.0 . AVERAGES 23.5 36.0 46.4 48.0 50.0 TABLE A-10 FILTERED BOD5 DATA (800 mg/l BOD ) DATE 1 DAY 2 DAY 4 DAY 8 DAY 16 DAY BOD5 mg/l % REMOVAL BOD5 mg/l % REMOVAL B0D5 mg/l % REMOVAL B0D5 mg/l % REMOVAL BOD5 mg/l % REMOVAL 23/8/72 160 80.0 17 97.9 43 94.7 10 98.7 11 98.6 25/8/72 230 71.2 28 96.5 18 97.7 16 98.0 20 97.5 1/9/72 212 73.5 25 96.9 ' 27 96.6 — — — — 8/9/72 140 82.5 19 97.6 14 98.2 20 97.5 13 98.4 14/9/72 — 20 97.5 16 98.0 17 97.9 23 97.0 AVERAGES 187 76.6 22 97.0 23 97.1 16 98.0 17 97.9 TABLE A - l l . GROSS BOD,. DATA (800 mg/l BOD feed) DATE 1 DAY 2 DAY 4 DAY 8 DAY 16 DAY BOD5 mg/l % REMOVAL BOD5 mg/l % REMOVAL BOD5 mg/l % REMOVAL BOD5 mg/l % REMOVAL BOD5 mg/l % REMOVAL 16/8/72 510 36.2 303 62.1 296 63.0 210 73.7 * 129 84.0 23/8/72 500 37.5 280 65.0 315 60.6 215 73.1 180 77.5 25/8/72 528 33.0 288 64.0 300 61.7 : 200 75.0 178 77.7 13/9/72 360 ** 55.0 325 59.4 315 60.6 233 70.9 220 72.5 AVERAGES 515 35.6 300 62.5 306 61.0 215 73.2 177 76.6 * Not at steady state ** Not average TABLE A-12 MLSS DATA (1200 mg/l milk feed) DATE 1 DAY 2 DAY 4 DAY 8 DAY 16 DAY 25/8/72 360 233 218 263 — 28/8/72 333 225 230 248 — 30/8/72 357 210 253 230 - 290 1/9/72 340 200 233 223 255 5/9/72 280 153 198 242 250 8/9/72 470 210 217 220 185 11/9/72 390 243 235 228 195 13/9/72 435 285 258 273 215 18/9/72 505 258 255 268 173 95 TABLE A-13 NITROGEN DATA mg/l (1200 mg/l milk feed) DATE 1 DAY 2 DAY. 4 DAY 8 DAY 16 DAY 25/8/72 67.3 67.0 66.3 67.2 GROSS KJELDAHL NITROGEN 1/9/72 8/9/72 13/9/72 65.0 63.0 65.0 66.5 65.0 66.9 66.5 66.5 66.1 66.5 67.5 66.6 67.6 69.5 28/8/72 18.0 9.8 21.8 24.0 14.6 NH NITROGEN 5/9/72 15.0 10.9 12.9 7.6 19.6 18/9/72 25/9/72 5.6 3.36 10.7 16.2 ' 14.0 14.0 24.2 12.9 22.8 24.2 28/8/72 6.2 5.6 4.5 4.2 4.8 ORGANIC NITROGEN 5/9/72 18/9/72 4.2 4.5 4.8 5.5 2.5 3.4 4.9 5.0 4.5 25/9/72 5.6 4.3 5.0 6.2 5.0 TABLE A-14 BATCH SETTLING DATA (1230 mg/l COD feed) SETTLING TIME (Hours) . 1 DAY EFFLUENT 2 DAY EFFLUENT 4 DAY EFFLUENT 8 DAY EFFLUENT 16 DAY EFFLUENT COD mg/l % REMOVAL MLSS mg/l COD mg/l % REMOVAL MLSS mg/l COD mg/l % REMOVAL MLSS mg/l COD mg/l % REMOVAL MLSS mg/l COD mg/l % REMOVAL MLSS mg/l 2 475 61.7 — 606 51.1 — 468 62.2 — 521 58.0 — 575 53.5 — 4 375 69.7 530 57.2 — 413 66.6 — 452 63.6 — 552 58.0 — 6 . 375 69.7 150 521 58.0 120 390 68.5 55 413 66.6 205 552 58.0 185 8 343 72.3 — 502 59.5 — 399 67.8 — 399 67.8 — 516 58.3 — 24 343 72.3 115 478 61.4 55 359 71.0 45 351 71.6 210 454 63.4 110 48 276 77.7 65 470 62.1 0 347 72.0 0 213 82.8 30 466 62.4 20 120 171 86.0 40 490 60.4 10 314 74.7 10 250 79.8 40 432 65.2 60 VO 97 TABLE A-15 NITRATE NITROGEN DATA LOADING 1 DAY 2 DAY 4 DAY 8 DAY 16 DAY 12/7/72 Low 0.48 mg/l 0.35 mg/l 0.25 mg/l 0.20 mg/l 18/7/72 Low 0.43 mg/l 0.36 mg/l 0.56 mg/l 0.23 mg/l 24/7/72 Low 0.56 mg/l 0.41 mg/l 0.31 mg/l 0.46 mg/l 12/9/72 High 0.46 mg/l 0.36 mg/l 0.41 mg/l 0.34 mg/l 0.44 mg/l 28/9/72 High 0.49 mg/l 0.46 mg/l 0.45 mg/l - negl ~ 0.51 mg/l APPENDIX B TEST DATA PERTAINING TO THE DETERMINATION OF STEADY STATE OPERATION LOW LOADING TABLE B-l SYSTEM TREATMENT EFFICIENCY LOW LOADING — COD START UP TO STEADY STATE DATE TIME ELAPSED 2 DAYS R.T. 4 DAYS R.T. 8 DAYS R.T. 27/6/72 11 Days 33% 42% 40% 29/6/72 13 Days 45% 68% 47% _ 4/7/72 18 Days 39% 40% 41% 7/7/72 21 days 42% 42% 42% :—' : STEADY STATE 14/7/72 28 days 30% 35% 45% 18/7/72 32 days 29% 34% 44% 21/7/72 35 days 33% 37% 43% 24/7/72 39 days 34% 39% 44% APPENDIX C CALCULATIONS OF THE CONSTANTS K7 AND K10 FOR USE IN McKINNEY'S DESIGN EQUATIONS 101 Average values for McKinney's constants K̂  and K^Q were found by f i t t i n g the test data to a linear equation incorporating McKinney's three equations. F = F i (3) K 5t + 1 K F M a - 5 7 F T K : <4> F e - F + Va <5> where and K 5 = 108 day - 1 (extrapolated to 3°C) K, = 72 day - 1 (extrapolated to 3°C) Assume for purposes of this calculation F - K,„M since FL nM » F e 10 a 10 a Substituting Equation (3) into Equation (4) gives K, • F. M ^ i a (1/t + K y)(K 5 • t + 1) Substituting for M in Equation (5) gives F = ^ 0 * K6 ' F i 6 (1/t + KyKKgt + 1) which is reworked to 102 in a where and y = mx + b form F (K, • t + 1) _ e J ^ 1_ y ~ F. • K, t X = F i ' K6 m =-K? B = K 1 0 Listed in Table C-l are the calculated x's and y's for the two loadings. Figure C-l i s the plot of x against y for the two loadings. From this plot the average values are 0.086 day * for the low loading and 0.076 day ^ for the high loading. The K^Q values are 0.79 mg/mg for the low loading and 0.70 mg/mg for the high loading. TABLE C-l RETENTION FEED F e • (K5t+1) x K ? ( F e ) • (K5t+1) mx TIME (mg/l) F i ' K 6 fc y F • K 1 6 1 800 515(108-1+1) 1 800 • 72 1 0.955 " K7 515(108-1+1) 800 • 72 -0.955K? 2 290 238(108-1.97+1) 290 • 72 1 1.97 1.22 -v 238(108-1.97+1) 290 • 72 -2.44K? 2 800 300(108-1.97+1) 800 • 72 1 1.97 0.57 " K7 300(108-1.97+1) 800 • 72 -1.15K? 4 290 100(108-3.96+1) • 290 • 72 1 3.96 ' 0.57 "*7 137(108-3.96+1) 290 • 72 -2.251^ 4 800 306(108-3.96+1) 800 • 72 1 3.96 0.57 ~h 306(108-3.96+1) 800 • 72 -2.25K? 8 290 99(108-8.6+1) 290 • 72 1 8.6 0.52 ~ K7 99(108r8.6+l) 790 • 72 -4.46K? 8 800 215(108-8.6+1) 800 • 72 1 8.6 0.40 215(108-8.6+1) 800 • 72 -3.44*^ 16 290 46(108-16.7+1) 290 • 72 1 16.7 . 0.31 215(108-16.7+1) 290 • 72 -5.121^ 16 800 177(108-16.7+1) 800 • 72 1 16.7 0.35 ^ 7 177(108-16.7+1) 800 • 72 -5.8 K ? 

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 23 1
United States 4 0
France 1 0
Canada 1 0
City Views Downloads
Beijing 23 0
Unknown 1 0
Thunder Bay 1 0
Montpelier 1 0
Mountain View 1 0
Redmond 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items