Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Methanol induced biological nutrient removal in a full-scale sequencing batch reactor Louzeiro, Nuno 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-0105.pdf [ 6.67MB ]
Metadata
JSON: 831-1.0064021.json
JSON-LD: 831-1.0064021-ld.json
RDF/XML (Pretty): 831-1.0064021-rdf.xml
RDF/JSON: 831-1.0064021-rdf.json
Turtle: 831-1.0064021-turtle.txt
N-Triples: 831-1.0064021-rdf-ntriples.txt
Original Record: 831-1.0064021-source.json
Full Text
831-1.0064021-fulltext.txt
Citation
831-1.0064021.ris

Full Text

METHANOL INDUCED BIOLOGICAL NUTRIENT REMOVAL IN A FULL-SCALE SEQUENCING BATCH REACTOR by NUNO LOUZEIRO B.A.Sc, The University of Waterloo, 1997 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Civil Engineering) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1999 © Nuno M . Louzeiro, 1999 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 of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that 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 reference and study. I f u r t h e r agree that permission f o r e x t e n s i v e copying of 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 granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n permission. The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada A B S T R A C T The primary goal of this research was to determine the potential for denitrification and phosphorus removal of a full-scale SBR, with and without the use of methanol as an external carbon source. This study was conducted at the Kent Wastewater Treatment Plant, in Agassiz, British Columbia, Canada, which has two SBRs working in parallel; one SBR was used as a control, and the other was used for the experiment. Methanol was fed into the experimental SBR during the anoxic cycle to achieve approximate concentrations of 4.1, 8.1, and 12.2 mg CH3OH/L. Each dosage was applied for a one-month period. The full-scale control SBR, without methanol addition, achieved negligible denitrification rates. However, two denitrification rates were observed in the full-scale experimental SBR, with methanol addition; an initial fast rate and a slower second rate. Methanol was utilized as the carbon source for denitrification during the first rate period. The denitrification rate (KDN, nig NOx/g MLVSS/day) during this period increased with increasing methanol concentration (M, mg CH3OH/L) according to the following relationship: K D N =-0.203-M2 + 3.93-M, until a maximum denitrification rate of approximately 19 mg NOx/g MLVSS/day was attained. Following the depletion of the methanol, denitrification reactions continued by using the available natural carbon in the influent, resulting in a slower second denitrification rate. Biological phosphate uptake and release was only observed to a significant extent in the experimental SBR with methanol addition. The phosphate release, which commenced once a very low nitrate concentration was achieved, had a rate of approximately 2.7 mg P/g MLVSS/day and was likely caused by a supply of natural short chain carbon in the influent wastewater. Methanol was not utilized to a significant extent as the carbon source for the enhanced biological phosphorus removal (EBPR) process. However, methanol addition was critical to the EBPR process, since it depleted the available nitrates, and, thus, allowed EBPR to take place. It was demonstrated during the full-scale study that the settleability was linearly related to the denitrification rate. The average total change in the ORP, during a four-hour SBR cycle, was also linearly related to the denitrification rate. ii T A B L E O F C O N T E N T S Abstract ii List of Tables vi List of Figures vii Acknowledgements x 1.0 Introduction 1 1.1 Experimental Significance 1 1.2 Research Objectives 2 2.0 Literature Review 4 2.1 Sequencing Batch Reactors 4 2.1.1 Historical Use of SBR Technology 4 2.1.2 Operation of a Sequencing Batch Reactor 5 2.1.3 Advantages and Disadvantages of SBRs 8 2.1.4 Programmable Logic Controller 10 2.2 Biological Nutrient Removal 11 2.2.1 Nitrogen Removal 12 2.2.2 Phosphorus Removal 15 2.2.3 Methanol Application in BNR 18 2.2.3.1 Properties ofMethanol 18 2.2.3.2 Methanol as an External Carbon Source 18 2.2.3.3 Environmental Implications 20 2.2.4 SBR Application for Biological Nutrient Removal 21 2.3 ORP Monitoring for Process Control 23 3.0 Experimental Methods and Analytical Techniques 26 3.1 Experimental Design and Setup 26 3.1.1 Full-Scale Experiment 26 3.1.2 Batch Experiments 31 iii 3.2 Sampling and Monitoring Program 32 3.2.1 Analytical and Sampling Methodology 36 3.2.1.1 Nitrogen and Phosphorus Analyses 36 3.2.1.2 Estimates of Carbon Content 37 3.2.1.3 Suspended Solids Analysis 38 3.2.1.4 ORP, pH, DO, and Temperature Measurements 39 3.2.2 Agassiz WWTP Sampling Program 39 3.2.3 Quality Assurance and Quality Control 40 4.0 Results and Discussion 42 4.1 Study of a Full-Scale SBR 42 4.1.1 Methanol Induced Biological Nitrogen Removal 43 4.1.1.1 Nitrification 48 4.1.1.2 Denitrification 53 4.1.2 Methanol Induced Biological Phosphorus Removal 61 4.1.3 Areal Analysis of SBR BNR Performance 69 4.1.4 Solids Production 71 4.1.5 Methanol-Induced BNR: Efficiency of Treatment Process 80 4.1.6 On-Line Monitoring Results 87 4.1.7 Seasonal Variations 92 4.2 Batch Study of Methanol Induced Denitrification 97 4.2.1 Continuously-Mixed Reactor 98 4.2.2 Settled Reactor 100 5.0 Conclusions and Recommendations 102 5.1 Summary and Conclusions 102 5.1.1 Biological Nitrogen Removal in the Full-Scale SBRs 102 5.1.2 Biological Phosphorus Removal in the Full-Scale SBRs 103 5.1.3 General Performance of the Full-Scale SBRs 103 5.1.4 Batch Experiments 104 5.2 Recommendations for Continued Research 105 6.0 References 106 iv Appendix A: MSDS for Methanol 112 Appendix B: Calculation of Required Methanol 117 Appendix C: Laboratory Instruments QA/QC Tests 118 Appendix D: Influent and Effluent Raw Data for Experimental Runs 121 Appendix E: SBR Raw Data for Experimental Runs 123 Appendix F: Historical Data - Jan. 1997 to June 1999 Monthly Averages 127 Appendix G: Historical Data - Jan. 1999 to June 1999 Weekly Averages 128 Appendix H : Determination of Nitrification and Denitrification Rates 129 Appendix I: Nitrate and Phosphate Plots (NTS) 130 Appendix J: Determination of Phosphate Release Rate 132 Appendix K : Solids Production Determination 133 Appendix L: ORP Monitoring - Average Four Hour Cycle Trend 134 Appendix M : ORP Monitoring - Average Daily Trend 137 Appendix N : Raw Data and Graphs from Batch Study 140 v L I S T O F T A B L E S Table 1: Description of Experimental Runs 31 Table 2: Agassiz WWTP Laboratory Analyses 39 Table 3: Summary of Nitrification and Denitrification Kinetic Rates 48 vi L I S T O F F I G U R E S Figure 1: Typical Operating Sequence for a SBR 6 Figure 2: Illustration of Reactor Conditions During Fill Phase 7 Figure 3: An Advanced Programmable Logic Controller System 11 Figure 4: Possible Microbial Nitrogen Conversions 15 Figure 5: Simplified Representation of Phosphorus Removal Process 17 Figure 6: Schematic Operation of the Control and Experimental SBRs 27 Figure 7: Schematic Diagram of Methanol Injection System 29 Figure 8: Nitrate in Experiment SBR Without Methanol Addition (June 30, 1998) 30 Figure 9: Batch Test Apparatus 32 Figure 10: Summary of Sampling and Monitoring Plan 33 Figure 11: Monitoring Probe Arm 34 Figure 12: SBR Sampling Schedule 35 Figure 13: Temporal SBR Operation 43 Figure 14: Nitrate Trend Without Methanol Addition (Mar. 10, 1999) 44 Figure 15: Nitrate Trend With 27 L/day Methanol Addition (Mar. 24, 1999) 44 Figure 16: Nitrate Trend With 27 L/day Methanol Addition (Apr. 7, 1999) 45 Figure 17: Nitrate Trend With 54 L/day Methanol Addition (Apr. 20, 1999) 45 Figure 18: Nitrate Trend With 54 L/day Methanol Addition (May 6, 1999) 46 Figure 19: Nitrate Trend With 81 L/day Methanol Addition (May 18, 1999) 46 Figure 20: Nitrate Trend With 81 L/day Methanol Addition (June 3, 1999) 47 Figure 21: Nitrification Rates and Temperature 48 Figure 22: Relationship Between Nitrification Rate and Temperature 50 Figure 23: Ammonia Trends (Dec. 1, 1998) 51 Figure 24: Ammonia Trend (June 3, 1999) 52 Figure 25: Summary of Denitrification Rates 53 Figure 26: Phosphate and Nitrate Concentration in Experimental SBR (Apr. 20, 1999). 56 Figure 27: Methanol Dosage and Denitrification Rate Linear Relationship 57 Figure 28: Phosphate and Nitrate Concentration in Experimental SBR (June 3, 1999)... 59 Figure 29: Methanol Dosage and Specific Denitrification Rate Relationship 61 vii Figure 30: Phosphate Trend in Experimental SBR without Methanol (June 30, 1998) ... 62 Figure 31: Phosphate Trend without Methanol (Mar. 10, 1999) 62 Figure 32: Phosphate Trend with 27 L/day Methanol Addition (Mar. 24, 1999) 63 Figure 33: Phosphate Trend with 27 L/day Methanol Addition (April 7, 1999) 63 Figure 34: Phosphate Trend with 54 L/day Methanol Addition (April 20, 1999) 64 Figure 35: Phosphate Trend with 54 L/day Methanol Addition (May 6, 1999) 64 Figure 36: Phosphate Trend with 81 L/day Methanol Addition (May 18, 1999) 65 Figure 37: Phosphate Trend with 81 L/day Methanol Addition (June 3, 1999) 65 Figure 38: Phosphate Release at Nitrate Depletion (May 6, 1999) 67 Figure 39: Determination of Phosphate Release Rate 68 Figure 40: Concentration of Nitrate in SBR 69 Figure 41: Concentration of Phosphate in SBR 70 Figure 42: Observed Solids Production from Methanol Addition 71 Figure 43: Derivation of Net Yield Coefficient 73 Figure 44: Theoretical and Kinetic Models for Solids Production 74 Figure 45: Microscopic Photograph (lOOx) of Sludge in Control SBR 75 Figure 46: Microscopic Photograph (lOOx) of Sludge in Experimental SBR 76 Figure 47: Sludge Volume Index Without Methanol Addition 77 Figure 48: Sludge Volume Index as a Function of Methanol Addition 78 Figure 49: Relationship Between Methanol and ASVI 78 Figure 50: SVI as a Function of the Denitrification Rate 79 Figure 51: Total Nitrogen Removal Efficiency and Methanol Dosage 80 Figure 52: Influent Total Nitrogen Concentration (June 30, 1998) 82 Figure 53: Relationship between TP Removal Efficiency and Methanol Dosage 83 Figure 54: COD Removal Efficiency as a Function of Methanol Dosage 85 Figure 55: Reduction in COD Removal Efficiency as a Function of Methanol Dosage.. 85 Figure 56: Total Nitrogen Influent Concentration 86 Figure 57: Solids Removal Efficiency as a Function of Methanol Dosage 87 Figure 58: Average Weekly Temperature 88 Figure 59: Average Weekly pH 89 Figure 60: Average DO Trend During SBR Cycle (week of May 18 to 25) 90 Figure 61: Daily Average DO Trend (week of Mar. 16 to 23) 90 V l l l Figure 62: Average Change in ORP per Cycle vs. Methanol Dosage 91 Figure 63: Average Change in ORP per Cycle vs. Denitrification Rate 92 Figure 64: Seasonal SBR Temperature Variations 93 Figure 65: Seasonal pH Variation 94 Figure 66: Seasonal Influent Flow Rate Variation 94 Figure 67: Seasonal Variation in Influent N H 3 Concentration 95 Figure 68: Seasonal Variation in Influent PO4 Concentration 95 Figure 69: Influent N H 3 Concentration as a Function of Flow 96 Figure 70: Influent PO4 Concentration as a Function of Flow 96 Figure 71: Influent BOD 5 and COD as a Function of Flow 97 Figure 72: Completely Mixed Batch Test with Acclimatized Sludge 98 Figure 73: Completely Mixed Batch Test with Non-Acclimatized Sludge 99 Figure 74: Settled Batch Test with Acclimatized Sludge 100 ix A C K N O W L E D G E M E N T S I wish to express my sincere thanks to the following people and organizations for their assistance and support of this research: Dr. D.S. Mavinic, Professor and Head of the Environmental Engineering Group, Department of Civil Engineering, UBC, Dr. W.K. Oldham, Professor Emeritus of Civil Engineering, UBC, and Dr. A. Meisen, Professor of Chemical Engineering, Department of Chemical and Bio-Resource Engineering, UBC, for their guidance, support and valuable advice throughout this research; Scott Jackson, Civil Engineering Electronics Shop technician, UBC, for his support, advice, and extensive assistance with the setup of my experimental apparatus; Ian Gardner, Steve Nuttal, and Keith Paisley, Agassiz WWTP, District of Kent, B.C., for allowing the experiment to take place at their facility, for their valuable advice and cooperation, and for their assistance with the setup of the experimental apparatus; Paula Parkinson and Susan Harper of the UBC Environmental Engineering Laboratory for their assistance in the sample analysis; Fred Koch, Jowitt Z.X. Li, and Jeff Koh for sharing their ideas and experience and for their assistance with the setup of the experimental apparatus; Harald Schrempp, Kurt Nielson, and Douglas Smith, Civil Engineering Machine Shop technicians, UBC, for their assistance in the fabrication of the experimental apparatus; The Natural Sciences and Engineering Research Council of Canada (NSERC), the District of Kent, B.C., the Methanex Corporation, Stantec Consulting Ltd., and the Department of Civil Engineering, UBC, for their financial support; and, Phil Barton, for sharing his insight, Chuck Stearns for his assistance with my seminar presentation, Kristina Laretei for letting me crash on her couch in Abbottsford, and the blue Caprice for never leaving Jeff and I stranded between Vancouver and Agassiz (it came close a few times). x 1.0 I N T R O D U C T I O N 1.1 Experimental Significance Eutrophication is a natural aging process in which the water becomes organically enriched, leading to increasing domination by aquatic weeds, transformation to marshland, and eventually to dry land [1]. Wastewater discharges containing nitrogen and phosphorus compounds may result in the accelerated eutrophication of receiving waters, thereby promoting the growth of unsightly aquatic vegetation and algal blooms that may interfere with the beneficial uses of water resources. Algal growths can cause the depletion of dissolved oxygen in the receiving water, and, hence, alter the growth of certain kinds of fish and other aquatic life. The presence of algae also affects the value of water for water supply because they often cause taste and odour problems. Furthermore, ammonia is toxic to fish and other aquatic organisms, and nitrates have been implicated as the cause of methanoglobanemia or "blue baby disease", in which nitrates interfere with the oxygen utilization of newborn babies [1]. Consequently, nutrient regulations for treated wastewater effluents are becoming increasingly restrictive. The Canadian Water Guidelines [2] specify a total ammonia limit of 1.13 to 1.81 mg N/L (depending on temperature and pH), and a nitrite limit of 0.02 mg N/L for discharges into freshwater bodies. Many wastewater treatment plants currently target a nitrate discharge concentration of approximately 5 mg N/L. In the near future, it is expected that nitrate discharge regulations will be reduced to as low as 1 or 2 mg N/L [3]. Regulations as low as 0.3 mg P/L for phosphorus discharges from treatment plants have been reported in literature [4]. As a result, the development of effective and inexpensive nutrient removal processes is required to meet these strict nutrient regulations. Various biological and physicochemical treatment processes have been employed to control the amount of nutrients discharged in the treated wastewater. Biological nutrient removal (BNR) treatment processes, with the use of carbon naturally developed in the wastewater, have gained widespread acceptance by designers and operators because the use of chemicals and external carbon sources have been eliminated. However, the use of naturally developed carbon sources often results in an inconsistent supply of carbon and slow reaction rates. Therefore, it is unlikely that wastewater treatment plants will be able to meet the expected lower nutrient limits without the use of exogenous carbon sources. 1 Biological nutrient removal processes require the use of sequential anaerobic, anoxic, and aerobic zones. The most commonly utilized mainstream BNR processes are the University of Cape Town (UCT) process and the Bardenpho processes (including 4 stage, 5 stage, 3 stage, and 2 stage) [5]. Although these processes achieve efficient and stable nutrient removal, their complexity makes them difficult for retrofitting and process optimization. The sequencing batch reactor (SBR) is a modification of the traditional fill-and-draw activated sludge treatment system. SBRs and conventional activated sludge systems perform the same unit processes. However, in conventional plants the processes are completed simultaneously in separate tanks. In SBR systems, flow equalization, biological reactions, and clarification processes are accomplished sequentially in the same tank. SBRs are extremely flexible, relatively inexpensive and very effective treatment systems that can produce the conditions required for BNR by simply modifying the application and duration of mixing and aeration [6]. Furthermore, the SBR's simplicity and flexibility allow it to be controlled and optimized more effectively under dynamic conditions, compared to the UCT and Bardenpho processes. Real-time process controls can be easily and effectively applied to SBRs by modifying the application of aeration and mixing according to measurements made by sensors such as oxidation reduction potential (ORP) probes [7]. Methanol (CH3OH) has gained acceptance as a relatively inexpensive and reliable external carbon source for denitrification [8]. Methanol is typically considered to have a negligible effect on biological phosphorus removal, as was concluded by Randall et al. [9]. However, Jones et al. concluded that methanol does support biological phosphorus removal [10]. Furthermore, limited research has been performed to assess the effectiveness of methanol in biological nutrient removal in a full-scale SBR. 1.2 Research Objectives The study described in this thesis was performed at a full-scale SBR system located at the Kent Wastewater Treatment Plant (WWTP) in Agassiz, British Columbia, Canada; this plant has two SBRs working in parallel. The primary goal of this research was to determine the potential of the SBR's existing treatment strategy for denitrification, and phosphorus removal, with and without the use of methanol as an external carbon source. One SBR was used as a control and the other was used 2 for the experiment. Methanol was fed into the experimental SBR during the anoxic cycle to achieve approximate concentrations of 4.1, 8.1, and 12.2 mg CH3OH/L. Each dosage was applied for a one-month period. The specific objectives of this study are summarized as follows: 1. Study of the nitrification, denitrification and phosphorus removal kinetics of the SBRs existing treatment strategy, with and without methanol addition; 2. Investigation of the effect of methanol addition on solids production and other general performance aspects of the SBR; and, 3. Determination of seasonal influences on the SBR performance. 3 2.0 L I T E R A T U R E R E V I E W This chapter presents a summary of research previously conducted that is relevant to the objectives of this thesis. The specific areas emphasized in this review include: 1) sequencing batch reactor technology; 2) biological nutrient removal processes and the role of methanol and SBRs in BNR processes; and 3) the use of oxidation-reduction potential as a process control tool. 2.1 Sequencing Batch Reactors The sequencing batch reactor is defined as a mixed-culture, suspended-growth or activated sludge system. Unlike conventional activated sludge systems, in which flow moves continuously along a series of tanks of preset volumes, the SBR is a time-oriented, periodic system, which can satisfy different treatment objectives by simply modifying the application and duration of mixing and aeration in a single tank. 2.1.1 Historical Use of SBR Technology Sequencing batch reactor technology preceded the use of continuous-flow activated sludge technology. In 1914, Ardern and Lockett developed an activated sludge reactor that operated according to the principles of SBR technology [11], However, widespread use of SBRs did not occur at that time because of equipment and operational limitations associated with SBRs and the need for larger facilities. Technological developments in the early 1960's revived interest in the fill-and-draw systems. In 1971, Irvine and Davis initiated systematic research on SBR technology [12]. The U.S. Environmental Protection Agency (EPA) attempted to revive interest in this technology in the early 1980s, and spent considerable sums of money evaluating the process on a full-scale basis. Results of this project led to the use of SBR technology at several municipal facilities [6]. 4 The present level of efficiency associated with SBRs is a result of technological improvements in aeration devices and computers, which have drastically improved the operation and control of SBR processes. All wastewaters commonly treated by conventional activated sludge plants can now be treated with SBRs [1]. The SBR technology has become more attractive because of its economical and operational benefits, and it can now successfully compete with conventional systems. Since 1970, the SBR technology has gained widespread popularity and it is currently being adopted rapidly on a worldwide scale for activated sludge treatment at small and medium-sized municipal treatment plants [13, 14]. 2.1.2 Operation of a Sequencing Batch Reactor The SBR system can be operated using a single tank or multiple tanks in parallel. There is no theoretical limit to the maximum wastewater flow rate that can be handled by a SBR [15]. The principle components of a SBR are the tank, inlet, outlet, mixing and aeration system, and controller. A typical operating sequence for a SBR is composed of the following five steps: fill, react (aeration), settle (sedimentation/clarification), draw (decant), and idle. Specific treatment objectives are achieved by modifying the duration of each step. This allows the SBR system to be extremely flexible in its ability to meet various objectives. Figure 1 illustrates the operating sequence of a typical SBR [6]. The volume in the tank and the cycle times are also shown in Figure 1. The amplitude and frequency of the system's periodicity can be chosen according to specific long term goals, such as sludge settleability, and size of the population of the nitrifiers, denitrifiers, or Bio-P bacteria. Average design hydraulic retention times range from 12 hours, where the objective is to meet organic and suspended solids reduction, to 24 hours, when flow rates are highly variable and nitrification, denitrification, and phosphorus removal is also required. Arora et al. [6] and Irvine et al. [15] discuss methods and considerations for the design of SBRs. 5 Percent of: w w * Purpose/Operation Max. Volume 25 to 100 REACT ioo 35 On/Cycle 100 20 SETTLE Air Off Clarify D R A W Effluent 100 to 35 15 Air Off 1 Renucire Effluent 35 to 25 I D L E Air On/Off "Waste Sludge Figure 1: Typical Operating Sequence for a S B R [6] Substrate, in the form of raw wastewater or primary effluent, is added to the reactor during the fill operation. The liquid level is allowed to rise from 25 percent of capacity (at the end of idle) to 100 percent. The fill process normally lasts approximately 25 percent of the total cycle time. The addition of influent may be carried out under static, mixed, or aerated conditions, depending on the treatment objectives. Static filling conditions result in minimum energy input and high substrate concentrations at the end of the fill phase. Mixed filling conditions will achieve denitrification, if nitrates are present, a subsequent reduction of oxygen demand and energy input, and the anoxic or anaerobic conditions required for the biological removal of phosphorus. Figure 2 illustrates the 6 behaviour of the substrate concentration, dissolved oxygen concentration, and nitrate concentration during a mixed filling phase of a typical SBR, with oxygen and nitrate initially present [16]. Dissolved oxygen and substrate are initially reduced by aerobic biological reactions occurring during the initial period of the mixed fill phase. Once the oxygen concentration is depleted, nitrate serves as the electron acceptor, and anoxic biological reactions will degrade the substrate. Fermentation will begin once oxygen and nitrate, the electron acceptors, have been depleted, and anaerobic conditions are achieved. Aerated filling conditions reduce the cycle time and maintain a low substrate concentration by commencing the aerobic biological reactions during the fill phase. The reactions that are initiated during the filling process are completed during the react step. The react step typically requires 35 percent of the full cycle time [6]. Mixed react conditions result in denitrification, if organics and nitrates are present, reduction of oxygen demand, and anoxic or anaerobic conditions required for biological phosphorus removal. Aerated conditions result in the completion of aerobic reactions and biosolids reduction, if aeration is extended. Solids separation is achieved during the settle operation, providing a clarified supernatant to be discharged as effluent. The clarified treated water is removed from the reactor during the draw operation. The duration of the draw period can range from 5 to 30 percent of the total cycle time [6]. The idle operation is an optional phase that is occasionally omitted. In a multiple tank system, this phase is used to provide time for one reactor to complete its fill cycle before switching to another unit. The idle phase will create conditions for endogenous respiration and it will also thicken the Nitrate: Cone. Time During Mixed Fill (%) Figure 2: Illustration of Reactor Conditions During Fill Phase [16] 7 sludge further. Sludge wasting greatly affects the operation of an SBR. The amount and frequency of sludge wasting is determined by performance. Sludge wasting usually occurs during the settle or idle phases, but wasting can occur in the other phases depending on the mode of operation. Return of activated sludge (RAS) is not necessary in a SBR operation. Since aeration and settling occur in the same tank, sludge is not lost during the react step, and none has to be returned from the clarifier to maintain the sludge content in the aeration chamber. The SBR acronym has become a synonym for various different batch process designs. However, many of these batch technologies differ significantly from the SBR process originally described by Irvine and Davis [12]. Some modifications of the SBR process also include continuous flow modes of operation with intermittent withdrawal. This type of system is called the intermittent cycle, extended aeration system (ICEAS) [6]. 2.1.3 Advantages and Disadvantages of SBRs The sequencing batch reactor has several advantages over conventional activated sludge processes. Some of the other major advantages include the following: • An increased flow rate can be treated for a given tank volume since only one tank is required to accomplish flow equalization, biological reactions, and biomass settling (clarification) [16]; • SBRs are physically very simple, have a very compact layout, and are easy to operate; • The SBR tank serves as an equalization basin during the fill phase, and, hence, it is able to handle unsteady conditions, such as peak flows and shock loads, without significant degradation in effluent quality and without the need for additional tanks [6]; • This system is extremely flexible and by simple adjustment of the time cycles of operation and the application and duration of mixing and aeration, it can meet various treatment objectives, including the following: - combined nitrogen and phosphorus removal; - effluent discharge is periodic and the effluent can be retained until it meets specific requirements; - filamentous growth can be easily controlled by varying the operating strategies during the fill phase [17]; and 8 - holding the mixed liquor solids in the tank as long as necessary can also prevent washout by hydraulic surges; • SBRs do not require sludge recycling (RAS) between tanks and this eliminates the need for additional pumping and piping; • The clarification process in an SBR is normally much more efficient than in continuous-flow systems because, in the settle mode, the reactor contents are under nearly ideal quiescent conditions [6]; • During the early design life, when flow is significantly lower than design capacity, liquid level sensors can be set at a lower level to utilize only a fraction of the SBR tank capacity, while maintaining the length of treatment cycles the same as design without wasting power unnecessarily by over-aeration [6]; • Dissolved oxygen concentrations are very low during the anoxic phase, which can provide a greater oxygen driving gradient during the react phase and result in a slightly higher overall oxygen transfer efficiency with the same aeration equipment [6]; and • The microorganisms in the SBR have a ribonucleic acid (RNA) content three to four times greater than conventional continuous flow system, and, thus, they are capable of processing a greater quantity of substrate at a greater rate (the growth rate of microorganisms is dependent on the RNA content of the cells) [18]. Many of the disadvantages associated with the SBR process are related to the lack of research and experience in full-scale SBR systems. There are limited design data available, and standards for the design of SBR systems are not widely accepted or known [6]. The composition of the microflora, the spatial arrangement of the microconsortia, and their interactions are still largely unknown. Therefore, it is still very difficult to predict the behaviour of a given system [19]. Equipment limitations are another major source of concern with SBR processes. The effluent quality depends upon a reliable decanting facility and many of the difficulties with existing facilities have been related to the decanting equipment [20]. Plugging of aeration equipment may also occur during settle, draw, and idle periods [21]. A significant disadvantage with the SBR process is that, as the system gets larger, there is an increasing sophistication associated with the timing units and level sensors required to control process parameters. As a result, the SBR is primarily suitable for smaller flow rates [21]. 9 2.1.4 Programmable Logic Controller The programmable logic controller (PLC) of a SBR may be designed very simply. Timers and liquid level switches are typically used to operate each of the phases. This type of control system is favored for small systems and systems where the influent flows and characteristics are reasonably predictable. In a two-tank system, level switches may be used to signal the fill phase of the second tank when the high water level is reached in the first tank. Intermediate liquid level switches are used to begin mixing or aeration sooner if the liquid level reaches a high level sooner than expected. This assures that adequate time is provided during peak flow periods. More complex control systems allow greater energy conservation, the use of smaller tanks, and more precise control. Computer assisted analysis of process performance and automatic control of plant units reduce expenses related to energy, chemicals, and man power [22]. Some systems have on-line monitoring meters and sensors, such as flow meters, dissolved oxygen, pH and temperature probes, sludge level indicators, and electricity consumption counters; these continuously feed data to a computer and allow for more efficient operation. The data is stored, organized, and further processed using advanced statistical and graphical methods to produce an information summary. These functions can be performed almost instantaneously and they can be displayed in selected formats according to preset options. Operators can analyze data quickly to locate failures and inefficient operations, and improve process control and operation. Alternatively, process performance can be modified automatically based on set points and set rules. These sophisticated PLC systems utilize on-line devices to monitor organic matter, oxygen uptake rate (OUR), oxidation-reduction potential (ORP), ammonia, and nitrate. These on-line devices can achieve long-term performance improvements and significant savings in energy and chemical consumption. They can also contribute to immediate repair of failure [22]. Figure 3 is a diagram of an advanced PLC system. Complex PLC systems are favoured for larger SBR processes where more skilled operators are available, and savings in energy consumption warrant the increased capital and operating costs of the control system. 1 0 ORGANIZATION & STOEAGE STATISTICS & GRAPHICS SJMUT.A~rON SliT POINTS PROCESS ANALYSIS & CONTROL H U M A N DECISION ::i::::iPR©CjESS;::;;i:: MANAGEMENT & OPERATION MOTORS I VALVES FEED5RS AUTOMATIC :CONTROL;: Figure 3: An Advanced Programmable Logic Controller System [22] 2.2 Biological Nutrient Removal Nutrient-rich wastewater discharges accelerate the eutrophication of inland and coastal waters and this has resulted in an increased demand for the removal of nitrogen and phosphorus from wastewater. The limiting nutrient controlling eutrophication is typically considered to be phosphorus in freshwater environments and nitrogen in marine waters [4]. The most common eutrophication control policy is simultaneous reduction of both nitrogen and phosphorus, since limiting nutrient dynamics are poorly understood for most bodies of water [23]. However, some scientists disagree about the necessity to decrease nitrogen discharges in certain environments, and even suggest it may result in increased eutrophication problems [4]. Various options have been considered to remove nutrients from wastewater. Biological nutrient removal (BNR) processes provide a reliable and effective mode for phosphorus and nitrogen removal from point sources, that is both environmentally and economically superior to other options [24]. 11 2.2.1 Nitrogen Removal Nitrogen is continuously cycled between its seven oxidation states according to the relationships defined by the global nitrogen cycle [25]. The nitrogen-containing compounds of most interest, from a water quality standpoint, are organic nitrogen; ammonia (NH3), nitrite (N02~), nitrate (NO3), urea [CO(NH2)2], and nitrogen gas (N2). The atmosphere serves as a nitrogen reservoir. Nitrogen transfer to and from the atmosphere is primarily as N 2 , with a smaller amount of transfer as nitrous oxides (N20 and NO), and as gaseous ammonia. The transfer of nitrogen between terrestrial and aquatic environments is primarily as organic nitrogen, ammonium ion (NH41") and nitrate ion. Nitrogen can be removed from the atmosphere biologically by nitrogen-fixing bacteria, chemically via lightning discharges, and a certain amount of nitrogen fixation occurs in the industrial production of nitrogen fertilizer [25]. The fixed nitrogen is converted by plants and bacteria to proteins (organic compounds containing nitrogen) [26]. Animals, including humans, are incapable of utilizing nitrogen from the atmosphere or from inorganic compounds, but must obtain nitrogen in an organic form. Subsequently, the organic nitrogen is broken down by bacterial activity into urea, the chief constituent of urine, and ammonia. Urea is also hydrolyzed rather rapidly by enzymes to ammonium carbonate [26], and, as a result, it is seldom found in wastewater that is not very fresh. Animal feces also contain substantial quantities of organic nitrogen that is readily converted to ammonia by the action of heterotrophic bacteria, under aerobic and anaerobic conditions. The ammonia is converted to nitrite and then nitrate through bacterial nitrification. Subsequently, bacterial denitrification converts nitrite and nitrate to nitrogen gas, which is released into the atmosphere. The nitrogen gas consists primarily of N 2 , but nitric oxide (NO) and nitrous oxide (N20) are also formed to a much smaller extent. Nitrogen in untreated domestic wastewater is typically found as ammonia and organic nitrogen, albeit in smaller quantities [1]. Concentrations of nitrite and nitrate in untreated domestic wastewater are negligible. Organic nitrogen is readily utilized as a substrate for biological reactions during aerobic and anaerobic conditions, and, subsequently, it is either utilized for biological synthesis or converted to ammonia and released into the extracellular environment. Ammonia nitrogen exists in aqueous solution as either the ammonium ion or ammonia, depending on the pH of the solution, in accordance with the following equilibrium reaction: 12 N H 3 + H 2 0 ^ N H 4 + + OH" . The ammonium ion is predominant at pH levels below 7, and at levels above pH 7 ammonia is prevalent. The principle mechanisms for the biological removal of nitrogen from wastewater are assimilation and nitrification-denitrification. Nitrogen is a nutrient and microorganisms present in the treatment process will assimilate the ammonia and incorporate it into the cell mass. A portion of this nitrogen will be returned to the wastewater on the death and lysis of the cells. Nitrification-denitrification accomplishes the removal of nitrogen in two conversion steps. The first stage is nitrification, whereby, under aerobic conditions, ammonia in the raw wastewater is converted to nitrite and then to nitrate by the autotrophic microorganisms of the genera Nitrosomonas and Nitrobacter, respectively [1]. Approximate biochemical reactions can be expressed as follows [1]: 55NH! +760, + 1 0 9 H C O : ^ - >C,H 7 0,N + 54NO: + 57H 2 0 + 104H 2CO 3 4 2 3 Nitrosomonas i l l I 2. 1 5 400NO" +NH; +4H 2 C0 3 + H C O 3 +19502 N i t r o b a c t e r >C 5 H 7 O 2 N + 3H 2O + 400NO3~ Therefore, the nitrification reactions cause the pH to decrease. Nitrifying bacteria are sensitive organisms and extremely susceptible to a wide variety of inhibitors, including pH. A narrow optimal range between pH 7.5 to 8.6 exists, and sufficient alkalinity in the system must exist to prevent the pH from lowering significantly below this range [1]. In the second stage, under anoxic conditions (i.e. absence of dissolved oxygen) and in the presence of an energy and organic carbon source, facultative heterotrophs reduce the nitrate to nitrogen gas [1, 8]. The latter process is called denitrification. Some of the genera of bacteria that accomplish denitrification include Achromobacter, Aerobacter, Alcaligenes, Bacillus, Brevibacterium, Flavobacterium, Lactobacillus, Micrococcus, Proteus, Pseudomonas, and Spirillum [1]. The denitrification reactions are driven by an organic carbon source, which is normally supplied from internal (endogenous) sources, including wastewater, storage-induced carbon, and endogenous respiration of activated sludge [27]. However, following nitrification, wastewater is usually deficient in organic carbon. Therefore, the use of endogenous carbon may result in low denitrification rates and limited overall nitrogen removal [28]. Therefore, exogenous carbon sources must be supplied in such cases to drive the denitrification reactions. The following substrates have been used to achieve denitrification: methanol, ethanol, glucose, acetate, propionate, various organic acids and salts, and even industrial wastewaters [29, 30, 13 31]. Methanol has been reported in literature as the most widely used exogenous carbon source [8]. The type and dose of organic carbon utilized affects the denitrification rate by influencing the type of bacteria that develop, organism growth rate, nitrate reduction rate, and the degree of accumulation of intermediate by-products [31, 32]. The reduction of nitrate involves assimilatory and dissimilatory enzyme systems [1]. The assimilatory nitrate reduction process occurs when nitrate is the only form of nitrogen available. During assimilatory nitrate reduction processes, nitrate is converted to ammonia nitrogen for use by the cells in biosynthesis. The dissimilatory nitrate reduction process results in the denitrification of the wastewater by forming nitrogen gas from nitrate. The reduction of nitrate towards nitrogen gas by the denitrification process, has several intermediates, such as HN0 2, NO, and N 2 0. Under conditions of limited electron donors these intermediates can be readily formed. There is some concern related to the impacts of releasing these intermediates from the treatment process into the environment [33, 34]. The overall denitrification reaction using methanol as the external carbon source, may be represented approximately by the following equation [1]: N 0 3 " + 1.08 C H 3 O H + H + 0.065 C 5 H 7 0 2 N + 0.47 N 2 + 0.76 C 0 2 + 2.44 H 2 0 A typical ammonia concentration for a medium strength untreated domestic wastewater is 25 mg N/L [1]. Based on the above equation, a wastewater treatment plant serving a small community of 5,000 people, which produces an average wastewater flow rate of 1,500 m3/day, would require approximately 90 kg of methanol per day. Denitrification utilizes hydrogen ions and restores some of the alkalinity consumed during the nitrification reactions. The optimal pH for denitrifiers lies between 7 and 8 [1]. Furthermore, the presence of dissolved oxygen will suppress the enzyme system needed for denitrification and it is a critical parameter [1]. Several investigations have indicated that a critical oxygen concentration of approximately 0.2 mg/L exists, above which denitrification is very poor. It is suspected that the denitrification activity is also adversely affected by lower oxygen concentrations [35]. Recently, a range of new biological nitrogen removal processes have been reported that may occur in wastewater treatment plants, including the following: aerobic denitrification and heterotrophic nitrification [36, 37], anaerobic ammonium oxidation [38], and denitrification by autotrophic nitrifying bacteria [39, 40]. Van Loosdrecht et al. [41] provides an excellent overview of these possible microbiological conversions that are illustrated in Figure 4. Biological nitrogen removal can also be achieved by simultaneous nitrification/denitrification during the aerobic period of 14 an intermittently operated system [42]. Simultaneous nitrification/denitrification (SBD) occurs under reaction conditions which permit nitrification and denitrification to take place at the same time in the same reaction basin without using any specific anoxic mixing sequences for nitrate reduction. These systems also show the possibility of nitrification and denitrification using nitrite instead of nitrate [42]. The underlying physical explanation is that a substantial anoxic mass fraction exists in the center of large biomass floes resulting from an oxygen diffusion limitation into the floes [5], In a continuous-flow plant, SND may reduce the size of a second anoxic basin or eliminate it altogether. For sequencing batch reactors, SND could reduce the time required to achieve complete nitrification and denitrification. Nitrification Assimilation Denitrification • • • • N-fixation • Anamrrox Don trificaticn by nitrifiers ISIlllllilllllilil!!! N 2H 4 N H 3 ^ ^ Q r g a n j c _ N Figure 4: Possible Microbial Nitrogen Conversions [41] 2.2.2 Phosphorus Removal Phosphorus typically appears in wastewater as orthophosphate (P0 4 3), polyphosphate (P2O7), and organically bound phosphorus [1]. Microorganisms utilize phosphorus during cell synthesis and energy transport. Orthophosphates, such as P04~3, HP0 4" 2, H2P04~, and F£3P04, are available for biological metabolism without further breakdown. Polyphosphates, which include molecules with 15 two or more phosphorus atoms, oxygen atoms, and, in some case, hydrogen atoms, undergo hydrolysis in aqueous solutions and slowly revert to the orthophosphate forms. The biological cycle does not destroy phosphorus, and organics will build up in the receiving waters and exert an oxygen demand in the presence of this limiting nutrient. Degradation of these organics makes phosphorus available for another cycle. Therefore, the phosphorus supply must be interrupted to stop the eutrophication of receiving waters. The two basic models currently accepted for enhanced biological phosphorus removal (EBPR) are the Comeau/Wentzel Model [43, 44, 45], and the Mino Model [46, 47]. The two models are in general agreement about the biochemical mechanism for phosphorus removal but differ with respect to the source of reducing power in the reactions. Alternating aerobic and anaerobic conditions result in competitive substrate utilization and selection of phosphorus accumulating organisms (PAOs). The anaerobic zone is required as a fermentation step for the production of volatile fatty acids (VFAs) by normally-occurring facultative organisms. Volatile fatty acids or carboxylic acids, especially acetic acid, are the best known fermentation products to support enhanced biological phosphorus removal [43]. Substrates other than VFAs can also be used for EBPR but other heterotrophs must first convert them to VFAs [48]. The PAOs use energy derived from the cleavage or hydrolysis of the stored polyphosphate to sequester these VFAs and store them as poly-hydroxyalkanoates (PHA). Inorganic phosphate is thus released into the liquid phase outside the cell. This is the primary release of phosphorus [43]. Secondary phosphorus release will occur in the absence of VFAs [24]. The Mino Model [46] suggests that energy from polyphosphate and reducing power supplied by the catabolism of glycogen through the Emden-Meyerhof-Parnas or Etner-Doudoroff pathway are used for the basal energy requirements of the cell and for the biosynthesis of PHAs from the VFAs. The ComeauAVentzel Model [43] differs only in that the anaerobic operation of the tricarboxylic acid (TCA) cycle rather than the utilization of glycogen produce the reducing power, in the form of NADH and FADH2. It has been implied that glycogen is the more significant pathway [20]. However, a recent study clearly indicates the importance of the TCA cycle for some PHA synthesis, suggesting that the two sources of reducing power are not mutually exclusive [45]. Under subsequent aerobic conditions, the PAOs utilize the stored PHAs in the presence of oxygen to generate energy for growth and uptake of the available orthophosphate into the cell from the surrounding bulk liquid and storing it mostly as polyphosphate. Glycogen is also replenished during this phase [47, 49]. A simplified representation of the phosphorus removal mechanism is shown in Figure 4 [24]. 16 V olatile F atty A cids Phosphorus Phosphorus Oxygen (Nitrate) \ BBS \ Carbon Dioxide Water Phosphorus Removing Bacteria Phosphorus Removing Bacteria Anaerobic Conditions Aerobic Conditions Figure 5: Simplified Representation of Phosphorus Removal Process [24] PAOs are able to adapt to various environmental conditions in the BNR plant. The success of the EBPR process is dependent on the outcome of competition with other heterotrophs, particularly denitrifiers, which can also consume fermentation products under the same anaerobic conditions. It is also well established that PAOs form dense floes that favour the selection process of sedimentation Secondary release of phosphorus is not associated with the storage of organic compounds, and, thus, it will not result in the removal of phosphorus by PAOs in the subsequent aerobic stage [24]. The efficiency of the biological phosphorus removal treatment is detrimentally affected by the secondary release of phosphorus. The secondary release of phosphorus may result from a long hydraulic retention time in the anaerobic zone, inadequate supply of organic compounds in the wastewater, and insufficient activity of the fermentation bacteria in the anaerobic zone [24]. Nitrate is considered to be an inhibitor to EBPR since it can be denitrified in the anaerobic zone, thereby reducing the supply of organic substrates available for uptake by PAOs. However, some PAOs can denitrify while taking up phosphorus and they will assist in the denitrification of the nitrates in the anoxic zone [24]. Investigations at the Westbank plant in British Columbia, Canada, clearly show the benefit of anoxic phosphorus uptake [24]. Not all PAOs have this capability, and even of those that have, not all will use this ability under normal conditions in an anoxic zone [24], [24]. 17 2.2.3 Methanol Application in BNR 2.2.3.1 Properties of Methanol Methanol (CH3OH) is usually derived from natural gas feedstock, which is then reformed to carbon oxides and hydrogen. The resulting "synthesis gas" is circulated under pressure through a conversion catalyst to form crude methanol, and subsequently distilled to form a high purity methanol. Methanol is a water-white, volatile, flammable liquid that has a faint odour and an approximate density of 0.79 kg/L at 20 °C. It is also hygroscopic and totally miscible with water. Furthermore, methanol is toxic and contact by skin, eyes, ingestion or inhalation may cause potentially serious health problems. Therefore, special attention should be given to shipping, handling, and storage of methanol. Given the proper precautions and handling methanol can be used safely for applications in wastewater treatment plants. The Material Safety Data Sheet (MSDS) for methanol is shown in Appendix A. 2.2.3.2 Methanol as an External Carbon Source The denitrification process has traditionally been performed following carbon oxidation and nitrification. As a result, the nitrified wastewaters are usually deficient in organic carbon. Furthermore, the use of endogenous carbon achieves low denitrification rates and limits the overall nitrogen removal. Therefore, exogenous carbon supplies are commonly required to drive the denitrification reactions. It is unlikely that wastewater treatment plants will be able to meet the expected lower nutrient limits without the use of exogenous carbon sources such as methanol. Methanol has been reported in literature as the most widely used exogenous carbon source [8], based on its effectiveness, cost, and availability on the market. During the last 45 years, considerable research has been performed into the denitrification of wastewater using methanol as a carbon source. Some of the recent studies to assess the effectiveness of methanol to the denitrification process are described in this section. Tarn et al. [50] investigated the effectiveness of exogenous carbon substrates (methanol, sodium acetate, and sodium propionate) on organic matter removal, nitrification, and denitrification in a modified bench scale SBR system with alternating anaerobic, anoxic, and aerobic stages. Both acetate and propionate were significantly more effective than methanol in enhancing denitrification. 18 However, the addition of methanol still produced drastically improved denitrification compared to the control reactor, which did not have an external carbon source. Carley and Mavinic [31] produced similar results from their research. However, Hallin et al. [51] observed that methanol resulted in faster denitrification rates, compared to acetate, after a long acclimation period. Biochemically, the glycolitic pathway and TCA cycle are the two major metabolic pathways for utilizing organic substrate as a source of energy and carbon in most microorganisms [25], Acetate and propionate enter the pathway directly via the easily formed acetyl-CoA and propionyl-CoA, respectively. However, one-carbon substrates, such as methanol, must be converted to acetyl-CoA or glyceraldehyde-3-phosphate before entering the two metabolic pathways [50]. This may explain the reduced denitrification efficiency of methanol compared to acetate and propionate, which has been observed in some studies. A significant disadvantage of methanol is that a long adaptation time is needed before satisfactory denitrification rates are achieved [51, 53]. It is speculated that the long adaptation period required by external carbon sources, such as methanol, is a result of the necessity for both the induction and synthesis of enzymes in the microflora, as well as genetic alterations in the microbial population [51]. However, Regan et al. observed effective denitrification with methanol without acclimation [54]. Despite its potentially reduced denitrification effectiveness compared to some other carbon sources, methanol has the strong advantages of good availability and relatively low cost on a chemical oxygen demand (COD) basis, compared to alternative carbon sources such as acetic acid [53, 54]. The supplementary costs of utilizing methanol to achieve higher denitrification rates may be considerably lower than the cost of upgrading with a new basin volume [53]. Furthermore, full-scale application of methanol as a carbon source for denitrification is stable and easy to operate, and high levels of nitrogen removal can be achieved [53]. It has been suggested that for substrates other than VFAs to be used for EBPR, they must first be converted to VFAs by heterotrophs [48]. This can be accomplished by providing an anaerobic zone or a separate fermenter for fermentation reactions to anaerobically break down the organic matter to form short chain volatile fatty acids. Primary sludge fermenters are routinely incorporated into BNR plants required to achieve effluent total phosphorus concentrations below 1.0 mg/L. Alternatively, some BNR plants incorporate an anaerobic bioreactor for fermentation of easily 19 degradable organic compounds in the wastewater to form VFAs. Most of the literature related to the optimization of VFA production describes attempts to optimize the acid phase of anaerobic digestion processes. Research was not found on the effectiveness of VFA production from methanol fermentation, for the purpose of enhancing the EBPR process. However, Carucci et. al [52] concluded that EBPR can also be obtained with organic substrates other than VFAs without their pre-conversion to VFAs. This would likely require a long acclimation period and would result in a reduction in reaction rates. Research on methanol as an external substrate for EBPR has resulted in contradicting conclusions. Randall et al. [9] experimented with the addition of different concentrations of methanol to observe its effect on biological phosphorus removal. The experiments were conducted on pilot scale SBRs, treating a synthetic wastewater influent. They concluded that external methanol addition had a negligible effect on phosphorus removal. Tam et. al [32] also found that methanol addition appeared not to enhance phosphorus release. However, Jones et al. [10] conducted an investigation with methanol on a continuous-flow reactor, treating variable domestic wastewater. They observed that methanol supported biological phosphorus removal. Studies have also been performed on the effects of methanol to combined nitrogen and phosphorus removal in an SBR. Tam et al. [32] investigated the effects of commonly used organic substrates, such as methanol, glucose, and sodium acetate, on a lab-scale SBR system with combined nitrogen and phosphorus removal. Sodium acetate was the most efficient and effective source in removing wastewater nitrogen, followed by methanol, and glucose was the least reliable substrate. However, the addition of methanol did not induce EBPR processes. Limited research has been performed on the effects of methanol addition to the biological removal of nitrogen and phosphorus in a full-scale SBR. 2.2.3.3 Environmental Implications The accumulation of greenhouse gases, such as carbon dioxide (C02), hydrofluorocarbons (HFCs), and methane, may act to increase Earth's greenhouse effect and ultimately lead to global warming. Such a buildup, primarily of C0 2, has occurred at an alarming rate over the last century. Since 1870, the atmospheric C0 2 concentration has increased by more than 20 percent at an exponential rate [55]. The largest source of additional C0 2 is the combustion of fossil fuels, which 20 has reduced Earth's largest stockpile of stored carbon. Carbon dioxide is being released into the atmosphere far faster than the Earth's ecosystem can absorb it. Methanol is mass-produced from natural gas, which is also a fossil fuel. The end products of biological reactions utilizing methanol include C0 2, which is released into the atmosphere. Therefore, the use of methanol as a carbon source for biological nutrient removal will also contribute to global warming. One of the strategies currently being considered for regulating the perceived causes of global climate changes is to coordinate carbon releases and sequestrations [55]. In the near future, a "carbon tax" may be applied to regulate the emissions of C0 2 derived from Earth's supply of stored carbon. As a result, this will increase the cost associated with the use of methanol as an external carbon source for BNR. Acetic acid is reported to be an effective carbon source for biological nitrogen and phosphorus removal. However, acetate is derived from methanol and, thus, has the same implications as methanol. 2.2.4 SBR Application for Biological Nutrient Removal The biological removal of nitrogen and phosphorus requires the use of alternating anaerobic, anoxic, and aerobic environments. The sequencing batch reactor process lends itself well to biological nutrient removal because it can attain these different zones very effectively and simply. The SBR system can be utilized to achieve any combination of carbon oxidation, nitrogen reduction, and phosphorus removal. These objectives can be realized with or without chemical addition by changing the application and duration of mixing and aeration in the SBR. Phosphorus removal can be achieved in a SBR biologically without coagulant addition. However, chemical treatment may be required to reduce the phosphorus concentration to low levels. This would result in large quantities of chemical sludge and an increase in operating costs. Another concern is the secondary release of phosphorus, which is detrimental to the performance efficiency of an SBR and should be avoided. Ketchum [16] provides an excellent review of common operating strategies used to meet various treatment objectives in a SBR. Successful performance of a SBR depends upon the specific local conditions. Therefore, pilot plant testing is recommended to develop operating data and design 21 criteria. The results of some of the research recently performed to assess the potential for biological nutrient removal in a SBR is summarized in this section. Demoulin et al. [42] experimented with co-current nitrification/denitrification and biological phosphorus removal in a cyclic activated sludge plant (SBR) with an ORP controlled cycle operation. Two full-scale SBR basins receiving continuous flow were used. The process demonstrated a very high degree of process stability and a considerably higher nutrient removal capacity than the conventional activated sludge process operated at the same loading. Biological phosphorus removal reduced an influent phosphorus concentration of 10 mg/L to 2 mg/L. A polishing dose of FeClS04 was also introduced to reduce the phosphorus concentration to less than 1 mg/L. Continuous monitoring of dissolved oxygen and ORP simplified the tuning of the cyclic process to meet effluent nitrogen and phosphorus concentrations. A lab-scale, single tank SBR was employed by Keller et al. [56] to evaluate the nutrient removal from an industrial wastewater (abattoir wastewater). This operation of the SBR clearly proved the potential of this simple process to achieve reliable and efficient nutrient reduction. Hamamoto et al. [57] experimented with simultaneous removal of nitrogen and phosphorus in a full-scale SBR system, with an on-line control system. The on-line control system selected optimal mixing and aeration times by analyzing measurements of DO, pH, ORP, and water level in the batch reactors. Average nitrogen and phosphorus removals in the full-scale plant were 96 and 93 percent, respectively. This study demonstrated that a SBR with an on-line control system is practical and effectively performs nitrogen and phosphorus removal. Some SBR research has been dedicated to developing strategies that promote the production of easily degradable organic substrate, such as VFAs, which are required for an efficient phosphorus removal process. Danesh et al. [58] experimented with a two-stage lab-scale SBR system, to remove phosphorus more effectively without the use of chemicals. The system was composed of two SBR units. The first unit was an anaerobic SBR, where volatile fatty acids were produced as a result of acid fermentation of the organic compounds in the raw wastewater. The fermented wastewater was conveyed to the second SBR where carbon and phosphorus were removed. Their results indicated an improvement in the wastewater quality for the purpose of nutrient removal, compared to a conventional SBR system. The prefermentation increased the available VFA content and provided improved phosphorus removal efficiency. 22 Marlkund [59] investigated the influence of low temperatures on small scale SBR biological phosphorus removal. It was concluded that phosphorus removal in the SBR could be maintained even at temperatures of 3 or 4 °C, although with lowered removal efficiency. Cycle times can be simply modified to compensate for changes in temperatures. Carrucci et al. [60] concluded, from their work on phosphorus removal using SBRs, that process performance, in terms of phosphorus removal, is highly sensitive to the competition for organic substrate between phosphorus accumulating and denitrifying bacteria. Gerber et al. [61] indicated a way to overcome this competition between denitrifiers and PAOs. They observed that PAOs could also release phosphorus under anoxic or even aerobic conditions, provided that acetate is available, and that PAOs can utilize nitrate as the final electron acceptor instead of oxygen. Vlekke et al. [48] successfully operated an anaerobic/aerobic SBR with nitrate addition in the anoxic phase, resulting in excellent phosphorus removal and combined denitrification; however, it was indicated that nitrogen, as a sole electron acceptor, may not be as efficient as oxygen for carbon oxidation and phosphate uptake. 2.3 ORP Monitoring for Process Control Wastewater treatment plants are dynamic systems which have been conventionally designed based on steady state principles and operation approaches. This has often resulted in poor system performance. Computerized, real-time, automatic monitoring and control systems have been developed to resolve such problems. The effectiveness of these systems is highly dependent on the reliability of on-line sensors, the suitability of monitored parameters, and the applicability of the control strategies. Recently, the use of oxidation-reduction potential (ORP) as a monitoring and control tool for biological nutrient removal has gained considerable attention. ORP is a highly sensitive and instantaneous on-line instrument that indirectly measures the amount of materials such as dissolved oxygen, organic substrate, activity of organisms and some toxic compounds in the reactor [62]. These parameters are indicative of operating conditions, such as overloading, under-loading, over-aerating, and under-aerating. Therefore, ORP measurements may be beneficial to a BNR process for making the distinction between anoxic, anaerobic, and aerobic periods. 23 ORP describes a solution's tendency to accept or donate electrons. The ORP (E) of a given oxidation-reduction equilibrium can be expressed as: ^ ^ RT, (Ox ^ E = E ° + In — , nF ^R D J where E° is the standard ORP of the given oxidation-reduction process at 25 °C, R is the gas constant, T is the absolute temperature, n is the number of electrons transferred in the reaction, F is the Faraday constant, O x is the activity of the oxidation agent, and R D is the activity of the reducing agent. In practice, the ORP is measured by determining the difference in potential between an inert electrode (generally a platinum electrode) immersed in the solution and a reference electrode. It is commonly accepted that ORP can be utilized as an indirect measure of dissolved oxygen (DO) at concentrations that can not be measured directly with oxygen probes. Monitoring of ORP is more flexible for process regulation under anoxic and anaerobic conditions, since measuring low DO concentrations is unreliable and the terms anoxic and anaerobic are only qualitative descriptions [63]. The ORP behaviour in the aerobic, anoxic and anaerobic zones was well described by Koch and Oldham [64]. They report that a "knee" occurs on the ORP curve at the point where anoxic conditions change into anaerobic conditions [64]. This variation can be used to identify bacterial activity and transition. The nitrate knee can be used to restart aeration at the point where the nitrate concentration approaches zero. Control strategies may also utilize the following relationships that have been observed: • the ORP value was determined to be linearly related to the logarithm of DO [65]; • a linear relationship between ORP and the nitrate concentration in the anoxic zone has also been discovered [65]; and • similarly, a linear relationship between phosphate and ORP also exists in the anaerobic and anoxic zones [64]. The magnitude of individual ORP readings is not necessarily precise, due to a number of inherent factors which affect its measurement [42]. It represents a spectrum or envelope of a process regime that has a measurable influence on the nutrient removal mechanisms. In comparison to the continuous-flow, activated sludge system, the SBR is a more versatile process that observes more dynamic changes. Continuous monitoring is therefore important for performance optimization. ORP can be used very effectively for fixed-time or real-time control of sequencing batch reactor operation, since quick and reliable modifications to the SBR process can be 24 performed by simply modifying cycle times [57, 66, 67]. The utilization of ORP can optimize the performance of an SBR process to achieve the most effective and economical results, by the control of variables such as air supply. ORP has traditionally been used at larger plants where the savings in operating costs justify the costs of implementing such a system. Yu et al. [62] tested the effectiveness of real-time on-line ORP and pH control of a pilot scale continuous flow SBR process. The results indicated that ORP and pH measurements could enhance the systems capability to remove nitrogen and phosphorus, while reducing the aeration energy requirements and cycle times. The on-line monitoring of ORP was also proven to be a practical technique for process control. Hamamoto et al. obtained a similar result using a full-scale SBR [57]. 25 3.0 EXPERIMENTAL METHODS AND ANALYTICAL TECHNIQUES The effectiveness of methanol as an external carbon source for biological nutrient removal in a full-scale SBR was studied in this research. This full-scale experiment consisted of four experimental runs. In addition, three batch tests were also conducted to determine the influence of acclimatization and mixing to the denitrification of wastewater with methanol. The materials used in the experiments, experimental procedures, and analytical methods are detailed in this chapter. 3.1 Experimental Design and Setup 3.1.1 Full-Scale Experiment This study was performed at the Kent Wastewater Treatment Plant, in Agassiz, British Columbia, Canada. The existing process sequence includes rotary screens, a degritter, two parallel SBRs, a chlorine contact tank with a S0 2 dechlorination facility, provision for final liquid discharge to the Fraser River, two parallel aerobic digesters, and a belt filter press plus polymer feed system. Two sludge lagoons are located on site for the storage of the processed sludge. The Kent plant is currently designed to handle a peak flow of about 5,400 m3/day. The average flow treated by the plant in 1997 was 2,400 m3/day. Local septage is also dumped into a separate holding tank and slowly incorporated into the plant flow by the plant operator. The plant is currently serving a community of approximately 4,000 people. Wastewater is constantly flowing into the SBRs, but the magnitude of the flow varies depending on the operation of the pumps at the headworks. Hence, the SBRs have a "semi-continuous" influent feed and an intermittent effluent withdrawal. A splitter box distributes the influent flow into two pre-mix chambers, which are separated from each SBR by baffle walls open at the bottom. The SBRs are operating on a four-hour cycle, and the operation of the two SBRs is offset by two hours. The Agassiz WWTP has two SBRs working in parallel; one was used as a control and the other as the experimental unit. Figure 6 illustrates the operation of the control and experimental reactors and the conditions that prevail in the reactors during the different phases. A similar strategy was employed by Ketchum et al. [68] to achieve biological nutrient removal and organic carbon reduction. The main differences between the two reactors are that methanol is mixed into the 26 experimental SBR following the completion of aeration to promote improved BNR, and the settling phase in the experimental SBR is fifteen minutes shorter than that in the control SBR. An explanation of the strategy shown in Figure 6 is provided below. ; Control Reactor (Existing: Conditions) :Fill:"~~i Fill Idle a r c Waste f Anoxic or Anaerobic):; Aeration Air On (Aercb c ) Fill 1 Settle Air Off 'Anox ic ) "ill :::::::::::::Decaht::::::::::::: : H : : ^ : : : n : A i r O f f : : : ^ : i : ! : : l : ; : (Anoxic or Anaerobic) : Time : (minutes) Experimental Reactor: f :ll! 1 Fil ;;;;;:;;ii:;i:i:;;:;::VVSste;; Idle and Waste Air Off (Anaerobic) i i iFer^ert^ ipniandj: : : : Phosphorus Release;; I Fill if-• Aeration Air On ::::!:::::(A :erbbic):::::::::: : Phosphorus Uptake:: and Nitt %cation : Mixed Methanol Addition Air OrVMixer On (Anoxic) Denitrificaticn Fill HI ;::;:;;;;;Settle;:;;;;;;;;; Air Off/Mixer Of t ::::::: (Anoxic):::: : : : : Denitrification Fill • ::::::::::::Deeant; :::::;:::::Decant:::::::::::: j j j [ j j j j i j j j «!! j j j=[ M M!. :: ;AnoxiC/Araercbic): Denitrificstior nr ; Fhosphorus Re.ease;; and Fcrmentaticin Timtt (Kirutos) Figure 6: Schematic Operation of the Control and Experimental SBRs Variations of mixing and aeration are used to achieve the conditions required for combined nitrogen and phosphorus removal with methanol addition in an SBR. Nitrification can be accomplished in an SBR process by providing a sufficiently long solids retention time (SRT of 5 to 10 days or more) to ensure the growth of nitrifying organisms [6], A sufficient aerated basin volume at adequate DO concentrations (2 mg/L) is also required [6]. These conditions will also achieve 27 stabilization of organic materials. Therefore, nitrification, with the strategy shown in Figure 6, will occur once adequate dissolved oxygen conditions prevail during the aeration phase. An anoxic period (presence of nitrates, but absence of DO) and a carbon source is necessary following the nitrification process to accomplish denitrification reactions. Once the aerators are turned off at the end of the aeration period, any available oxygen is quickly depleted by heterotrophs and anoxic conditions will then prevail. This occurs during the early period of the methanol addition phase, shown in Figure 6, when the reactor contents are well mixed. Alternating anaerobic (absence of DO and nitrates) and aerobic cycles are required for the induction of the enzyme system responsible for EBPR. In the configuration shown in Figure 6, raw substrate concentrations increase during the settle, decant, and idle phases as a result of minimum contact between the settled biomass and the influent substrate. The anaerobic conditions that may prevail during these phases favour fermenting organisms that use the influent raw wastewater to produce volatile fatty acids. Concurrently, the PAOs release stored polyphosphate into the bulk liquid to provide the energy needed to accumulate these VFAs. Once aerobic conditions prevail, during the aeration phase, the PAOs use stored VFAs to provide the energy needed to take up extracellular phosphorus and store it as intracellular polyphosphate. PAOs compete with denitrifying organisms for substrate until the oxidized nitrogen is eliminated. Subsequently, the PAOs are prepared for the next period of anaerobic conditions. Often a high percentage removal of both nitrogen and phosphorus is required and there is insufficient carbon available in the influent for performing both functions. Rusten et al. [13] observed that denitrification was limited by a lack of soluble carbon in several SBR experiments. The SBR at the Agassiz WWTP has a semi-continuous influent feed and this raw wastewater may provide an adequate supply of carbon to complete the denitrification reactions. However, improved BNR may be achieved by adding methanol to the anoxic zone to reduce the nitrates, leaving the available VFAs in the raw wastewater for phosphorus removal [24]. Consideration should be given to the fact that the addition of an external carbon source, such as methanol, has been shown to decrease the overall organic matter removal efficiency of a SBR system [50]. Methanol was added exclusively to the experimental SBR following the end of the aeration phase, as shown in Figure 6. The methanol injection system was composed of a pump, a mixer, and 28 an electrical timer. The pump was used to inject the methanol through a flexible tube that extends from the methanol tote tank, where the methanol was stored, to the propeller of a mixer. The mixer, which imparted a mixing energy of 2.74 W/m3 of tank volume, was specifically installed in the experimental reactor to allow the methanol to contact the mixed liquor throughout the entire volume of the SBR. A timer was used to turn the pump and mixer on at the end of the aeration phase (110th minute), and turn it off 15 minutes later (125th minute). Figure 7 shows a plan view of the methanol injection system. A slightly higher solids content was expected in the effluent from the experimental SBR, as a result of the shorter settling phase. Methanol Tote Tanks Pump Flexible Tube Inflow Pre-Mix Chamber t r i o Mixer Experimental J SBR • Outflow \ - - *• Pre-Mix Control Chamber SBR Outflow Figure 7: Schematic Diagram of Methanol Injection System The methanol dosage was varied during the experiment to determine the optimal dosage for biological nutrient removal. The required methanol dosage was calculated based on a preliminary analysis of the nitrate concentration in the SBR during the 4-hour cycle, as shown in Figure 8. Based on a maximum concentration of 3.3-mg N per L, it is required to inject 9.1 L of methanol per 4-hour cycle (refer to Appendix B). Therefore, a pumping rate of 0.6 L/min of methanol is required to achieve complete denitrification. Methanol was mixed into the experimental SBR over a 15-minute period at 0.3, 0.6, and 0.9 L/min. With six cycles daily, the methanol was consumed at rates of 27, 29 54, and 81 L/day, respectively. The average volume of the SBR is 875 m3; hence, the approximate methanol concentration in the experimental SBR for each dosage is 4.1, 8.1, and 12.2 mg CH3OH/L. Each dosage was applied for one month. Calibration of the methanol dosage was performed on a weekly basis. Air Off 0.5 0 -I ' ' ' ' 1 ' ' ' ' 1 ' ' ' ' 1 ' ' ' ' 1 ' ' 1 ' 1 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 8: Nitrate in Experiment SBR Without Methanol Addition (June 30, 1998) The full-scale experiment consisted of four experimental runs. The purpose of the first run was to establish an experimental baseline in both SBRs for various parameters. An areal analysis of the SBR was also performed during Run 1 to determine if there are areal variations in nutrient concentrations. The aerial analysis consisted of sampling a SBR during one typical 4-hour cycle at the front, middle, and back of the SBR, along the central catwalk. Methanol was not added to the experimental SBR during Run 1. During Runs 2 to 4, methanol dosages of 27, 54, and 81 L/day, respectively, were added to the experimental SBR, each for a period of four weeks, and the control SBR was operated continuously without methanol addition. Table 1 summarizes the different experimental runs related to the full-scale experiment. 30 Table 1: Description of Experimental Runs Experimental Run Description 1 -Methanol not added -Areal analysis of nutrient concentrations -Preliminary sampling to establish baseline for various parameters 2 -Methanol Dosage of 27 L/day 3 -Methanol Dosage of 54 L/day 4 -Methanol Dosage of 81 L/day 3.1.2 Batch Experiments Three batch experiments were conducted to determine the influence of mixing and acclimatization to denitrification reactions with methanol as a carbon source; one was conducted under unmixed conditions, and the other two were conducted under completely mixed conditions. One of the completely mixed batch experiments utilized a sludge sample from the full-scale control SBR that was not acclimatized to denitrification with methanol. All other experiments utilized sludge from the full-scale experimental SBR that had been acclimatized for a five week period to a concentration of 12.2 mg CH3OH/L. The unmixed batch experiment was designed to simulate the anoxic conditions in the full-scale experimental SBR. Each batch experiment consisted of four separate batch reactors that were operated simultaneously in sealed beakers. Three of the batch reactors were operated with different methanol dosages and the fourth reactor was used as a control, without methanol addition. Sludge samples were taken from the respective full-scale SBR at the end of the aeration cycle, when carbon content is low and nitrate concentration is high. The sludge sample was subsequently added to the beakers, sealed with a rubber stopper, and mixed for a short period with a multiple-beaker stirring device. Any residual oxygen and carbon would be depleted during this time. A balloon filled with nitrogen gas was subsequently added to the top of each reactor to fill the void resulting from the removal of samples from the system, thus ensuring that oxygen does not enter the reactor and disturb the anoxic conditions. An equal volume of nitrate and different methanol dosages were then spiked into each 31 reactor through a syringe. The reactors in each experiment were continuously mixed until after the methanol was added. At which time the stirring device was turned off in the unmixed batch experiment only, allowing the reactor contents to settle. Samples were withdrawn from the reactors every ten minutes using the syringe. The nitrate concentration in each sample was analyzed. Figure 9 shows the batch test apparatus that is described above. 3.2 Sampling and Monitoring Program The sampling program and analytical procedures used during this study are described in this section. Figure 10 describes the different sampling locations, type, frequency, and also the analyses that were performed on the samples. The sampling plan applies to experimental runs 2, 3, and 4. 32 Pre-Mix Chamber Influent Pre-Mix Chamber Experimental SBR c i Bl B2 C2 Control SBR Dl — •Effluent D2 Effluent Location Description Sample Frequency Analysis A Influent grab weekly TKN, TP, NO x, NH 3 , P0 4 , COD, TSS, VSS B l Middle of Exp. SBR grab weekly NO x, P0 4 , M L VSS B2 Middle of Control SBR grab weekly NO x, P0 4 , M L VSS CI Center of Exp. SBR probes 5 min. DO, ORP, pH, Temperature C2 Center of Control SBR probes 5 min. DO, ORP, pH, Temperature D l Exp. Effluent grab weekly TKN, TP, NO x, NH 3 , P0 4 , COD, TSS, VSS D2 Control Effluent grab weekly TKN, TP, NO x, NH 3 , P0 4 , COD, TSS, VSS Figure 10: Summary of Sampling and Monitoring Plan The following parameters were monitored on a continuous basis: pH, oxidation-reduction potential (ORP), temperature, and dissolved oxygen (DO). These parameters were monitored using probes submerged in the SBRs at a depth of approximately one meter. The probe arms were designed to float and adjust to changes in water level. It was placed at a central location in each of the SBRs. The locations of the probe arms are shown in Figure 10. An illustration of a probe arm submerged in one of the tanks is shown in Figure 11. 33 Figure 11: Monitoring Probe Arm Nitrification reactions consume alkalinity and thus cause a pH decrease. Nitrifying bacteria are sensitive organisms and extremely susceptible to a wide variety of inhibitors, including pH. A narrow optimal range between pH 7.5 to 8.6 exists, and sufficient alkalinity in the system must exist to prevent the pH from lowering significantly below this range [1]. Denitrification utilizes hydrogen ions and restores some of the alkalinity consumed during the nitrification reactions. The optimal pH for denitrifiers lies between 7 and 8 [1]. Furthermore, the presence of dissolved oxygen will suppress the enzyme system needed for denitrification and it is a critical parameter [ 1 ]. Temperature is also a critical parameter that controls the type of microorganisms that are able to live in an environment, and also the rate of biological reactions. ORP is a highly sensitive and instantaneous on-line instrument that indirectly measures the amount of materials, such as dissolved oxygen, organic substrate, activity of organisms and some toxic compounds in the reactor [62]. The monitoring of ORP is more flexible for process regulation under anoxic and anaerobic conditions than DO monitoring, since measuring low dissolved oxygen concentrations is unreliable and the terms anoxic and anaerobic are only qualitative descriptions [63]. Therefore, continuous monitoring of ORP may be beneficial to a BNR process in a SBR for making the distinction between anoxic, anaerobic, and aerobic periods, and for 34 performance optimization. As a result, these parameters were monitored continuously during the experiment. The main objective of this thesis was to determine the influence of methanol on the performance of the biological nutrient removal process in the SBR. As a result, it was necessary to analyze the different nitrogen and phosphorus constituents entering and leaving the reactor, to determine the efficiency of the BNR process. Grab samples were taken once per week from the influent to the reactors and the effluent from each reactor using a pole with a bottle secured to one end. The following parameters were analyzed from these samples: NO x (nitrite and nitrate), NH 3, total Kjeldahl nitrogen (TKN), total phosphorus (TP), soluble inorganic phosphorus (principally ortho-P04~3), chemical oxygen demand (COD), and suspended solids (SS). It was also necessary to analyze the different nitrogen and phosphorus components in the SBRs, to determine the extent of the biological reactions. Both SBRs were sampled on a weekly basis according to two different sampling plans. These samples were taken at an approximate 1-meter depth near the central catwalk separating the control and experimental SBRs, as shown in Figure 10. Samples were withdrawn at frequent time intervals throughout the SBR's 4-hour cycle during one sampling plan, and only at critical cycle times during the second sampling plan. The sampling times for both plans are shown in Figure 12. These sampling plans were alternated weekly, and each sample was analyzed for P0 4 and NOx content. Samples were also taken during the aeration phase of each SBR's cycle and analyzed for the MLVSS (mixed liquor volatile suspended solids) content. o u 3 •I I 1 0 10 30 Sampling Plan #1 Sampling Plan #2 60 90 110 125 140 155 170 Time (minutes) 190 210 240 Figure 12: SBR Sampling Schedule 35 3.2.1 Analytical and Sampling Methodology The analytical procedures used in this study followed the standard procedures outlined in Standard Methods [69]. All bottles and glassware used for storage or analytical purposes were carefully washed and dried. Plastic bottles were used to transport and store samples. The majority of the analyses were performed at the UBC Environmental Laboratory. The Agassiz plant operators also performed some analyses at the Agassiz WWTP laboratory. Samples that were to be analyzed for soluble contents (NOx, NH 3, TKN, TP, P0 4, and COD) were filtered through a Whatman #4 filter. One or two drops of sulfuric acid (H2S04) were added to preserve the samples at a pH less than two. The samples were stored at 4°C for no longer than 28 days before performing the analysis [70]. 3.2.1.1 Nitrogen and Phosphorus Analyses Total nitrogen is comprised of organic nitrogen, ammonia (NH3), nitrite (N02~), and nitrate (N03). The total Kjeldahl nitrogen (TKN) is the total of the organic and ammonia nitrogen. NOx is the combined nitrate and nitrite concentration. The growth rate of Nitrosomonas, which oxidizes ammonia to nitrite, is less than that of Nitrobacter, which oxidizes nitrite to nitrate. Thus, nitrite accumulation should not occur unless Nitrosomonas is inhibited. As a result, nitrite analyses will not be performed routinely and the NO x concentration will be assumed to be in the nitrate nitrogen form. An occasional nitrite analysis was performed to verify this assumption. Total phosphorus (TP) is a measurement of all the phosphorus forms present in a sample. The usual forms of phosphorus found in aqueous solutions include orthophosphate, polyphosphate, and organic phosphorus. The soluble inorganic phosphorus is predominantly orthophosphate, but soluble polyphosphate is also detected by this test. The organic phosphorus content is determined by subtracting the inorganic phosphorus from the TP. The procedure used for the analysis of NOx, P04, and NH 3 is as follows: • the samples were withdrawn into vials for analysis by the Lachat flow injection analyzer (operation of the Lachat instrument was performed by a qualified laboratory technician using QuikChem Method No. 10-107-04-1-E, 10-107-06-1-F, and 10-115-01-1-D [71]); and 36 • on a monthly basis, a N02" analysis was performed to verify that it is in limited quantities. The QuikChem digestion and extraction method (No.l0-107-06-2-D, and 10-115-0T-1-C), summarized below, was used for the determination of total phosphorus (TP) and total Kjeldahl nitrogen (TKN) [71]: • for each influent, SBR, effluent, the method blank, and the spiked samples 5 mL, 10 mL, 10 mL, 5 mL, and 10 mL, respectively, were withdrawn into digestion tubes; • standards were prepared by withdrawing 3 mL, 2 mL, and 1 mL of the 100 mg TKN/L and 50 mg TP/L solution, and 4 mL, 2 mL, and 1 mL of the 10 mg TKN/L and 10 mg TP/L solution into digestion tubes (this created TKN standards of 300, 200, 100, 40, 20, 10, and TP standards of 150, 100, 50, 40, 20, 10); • 10 mL of sulfuric acid digestion solution was added and mixed; • 2 to 3 boiling beads were add to prevent vigorous boiling during digestion; • the tubes were placed in the preheated block digester for 3.5 hours at 130°C to vaporize the H 20; • the digestion was continued for 2.5 hours at 380°C; • the samples were cooled and filled with ammonia-free water while shaking; • the tubes were inverted three times and then the contents were allowed to settle; • 5 mL was withdrawn for analysis of TP and TKN; and • a qualified technician measured TP and TKN using the Lachat Flow Injection Analyzer. 3.2.1.2 Estimates of Carbon Content The availability of an organic substrate is an important parameter, since it can limit the extent of biological reactions. The COD test is widely used as a means of measuring the organic strength of wastewater. It requires a very short time for evaluation and it is also an easier test to perform, compared to the BOD test. The procedure used for the analysis of COD is as follows [69]: • COD analysis vials were prepared with reagents for a maximum COD concentration of 200 ppm (if the concentration was higher than 200 ppm, than 1 mL of sample and 1 mL distilled water was used and a correction was made to the COD calculation); • 2 mL of each sample was put into tubes, which were capped tightly and well mixed; • standards were prepared (200, 100, 50, 20, 10, and 0 mg/L) with potassium hydrogen phthalate; 37 • the vials were placed in the block digester for 2 hours, allowed to cool to room temperature and settled; • analysis of the samples was performed using a spectrophotometer set at 600 nm to read the absorbance of light by the sample (the vials were cleaned and dried during the analysis); • the spectrophotometer was reset according to the 0 ppm standard and the absorbance of each sample was measured; and • a calibration curve (absorbance vs. concentration) was prepared and the COD in the samples was calculated. 3.2.1.3 Suspended Solids Analysis The total suspended solids (TSS) and volatile suspended solids (VSS) concentration was determined on a weekly basis for the influent and effluent from each SBR. In addition, the mixed liquor volatile suspended solids (MLVSS) concentration of the two SBRs was also measured. These samples were preserved at 4°C and they were analyzed within a 24-hour period after the samples were taken. The samples were brought to room temperature before conducting the analysis. The analysis of the total suspended solids and volatile suspended solids were performed as follows [69]: • the tare weight of an aluminum dish was measured and a glass fiber filter paper was seated onto a vacuum apparatus with distilled water; • the sample was mixed thoroughly and a sufficient quantity (1 L of final effluent and 10-25 mL of raw sewage or MLVSS) of well-mixed sample was filtered slowly through the glass fiber filter on the vacuum apparatus; • the cylinder was rinsed into the crucible, after the sample had gone through, with small amounts of distilled water; • the filter was placed on the aluminum dish; • the actual volumes dispensed, the sample source, and the dish identification was recorded, and the aluminum dish and filter paper were placed in the 103°C oven to dry for at least one hour; • the dish was transferred to a dessicator, allowed to cool, and weighed; the gain in weight represents the total suspended solids of the sample; and • the dish was fired in a muffle furnace at 550 °C for 15 minutes; after firing, the dish was allowed to cool somewhat before removing to a dessicator, where it was allowed to cool 38 completely before re-weighing it to determine the loss on ignition or volatile suspended solids. 3.2.1.4 ORP, pH, DO, and Temperature Measurements Two probe arms, each supporting four probes, were used to make continuous measurements of ORP, DO, pH, and temperature in the control and experimental SBRs. The data was collected using an 8-channel data logger and software fabricated by Lakewood Systems Ltd. Lakewood Systems also made the temperature probes used in this study. YSI dissolved oxygen meters, Model 54A, and Hanna Instruments ORP and pH probes and transmitters were also used for monitoring the SBRs during this study. All probes were cleaned on a weekly basis, using warm soapy water. The DO probes were calibrated on a weekly basis according to the manufacturer's instructions. Calibration of the ORP and pH probes was performed on a bi-weekly basis. The ORP was calibrated based on the Broadley James Corp. calibration procedure [72], and the pH was calibrated according to Standard Methods [69]. 3.2.2 Agassiz WWTP Sampling Program The Agassiz WWTP is equipped with trained laboratory technicians and equipment to perform various analyses. Table 2 lists the different parameters that are measured by their laboratory. Historical data is available for these parameters as of January 1997. Table 2: Agassiz WWTP Laboratory Analyses Analysis Units Location Frequency flow i m influent, waste sludge, digester decant, filter press water daily temperature °C air, influent, effluent, SBRs, digesters daily pH influent, effluent, SBRs, digesters daily DO mg/L influent, effluent, SBRs, digesters daily alkalinity influent, effluent monthly NH3 mg/L influent, effluent twice weekly TKN mg/L influent, effluent bi-weekly P04 mg/L influent, effluent twice weekly 39 Table 2: Agassiz WWTP Laboratory Analyses Analysis Units Location Frequency TSS mg/L influent, effluent, primary effluent, waste sludges, digesters twice weekly VSS mg/L influent, effluent, primary effluent, waste sludges, digesters twice weekly TSS mg/L SBRs daily VSS mg/L SBRs daily BOD mg/L influent, effluent weekly COD mg/L influent, effluent weekly fecal /lOOmL effluent bi-weekly C l 2 mg/L effluent daily S0 2 mg/L effluent daily colour FAU effluent daily turbidity pt effluent daily 3.2.3 Quality Assurance and Quality Control The majority of the laboratory analyses were performed at the UBC Environmental Laboratory. Qualified laboratory technicians who have many years of experience analyzing environmental parameters operate this laboratory. The technicians perform routine maintenance and calibrations of the analytical instruments using prepared standards. They also maintain historical data of test samples used to check the reliability of the different analyses. These test samples are used on a routine basis to identify and correct any instrument problems. The following quality assurance and quality control (QA/QC) measures were taken to control the quality of the sampling and analytical procedures: • Field Blanks were taken to detect any sampling, transport, and reagent, contamination; • a Standard Addition or spike of a known concentration of the analytes of interest (NOx, P0 4, NH 3, TKN, TP, and COD) was added to selected field samples to verify the absence of matrix effects; • duplicate samples were routinely taken to verify that the precision of the analytical procedure remains relatively stable; • the laboratory technicians performed frequent calibrations using single-point checking with a standard solution (Single-point checking requires the preparation of known concentrations of 40 a standard solution. The results are compared with the previous standard reading, and if the error is within 10 percent the sample analysis will proceed. If the results are outside this range, a secondary standard must be prepared and assessed similarly. If the second standard also fails, the calibration problem must be corrected before any further sample measurements are made Appendix C provides a historical summary of this QA/QC process.) [73]; and a Test Solution of a known concentration was also analyzed to test the calibration of the analytical instrument. 41 4.0 R E S U L T S A N D DISCUSSION The results obtained from the full-scale study and batch study are detailed and discussed in this section. This includes an evaluation of the nitrogen and phosphorus removal capabilities of the full-scale control SBR without methanol addition, and the full-scale experimental SBR with three different methanol dosages. A direct comparison between the control and experimental SBRs, in terms of nitrogen, phosphorus, and carbon removals, ORP, DO, temperature, pH, sludge settleability, and solids production, is also presented in this chapter. In addition, the seasonal influences on the performance of the two full-scale SBRs is discussed. The batch study discussion focuses on the influence of acclimatization and mixing to methanol induced denitrification. 4.1 Study of a Full-Scale SBR As described in Section 3.1.1, there were four experimental runs. These runs were distinguished by the applied methanol dosage. The two SBRs were operated without methanol addition during Run 1, and background data was collected during this experimental stage. Run 1 started on June 12, 1998 and finished on March 10, 1999; samples were taken at random during this period. During Runs 2 to 4, methanol dosages of 27, 54, and 81 L/day, respectively, were added to the experimental SBR, each for a period of four weeks, and the control SBR was operated continuously without methanol addition. The influent and effluent of the control and experimental SBRs were sampled on a weekly basis during Runs 2 to 4. The SBRs were sampled at critical times throughout the four-hour cycle on a weekly basis, and full samples of the entire four-hour cycle were taken on a biweekly basis. A summary of the raw data for the influent and effluent during the sampling periods described above is shown in Appendix D, and for the SBRs, it is summarised in Appendix E. The ORP, DO, pH and temperature in the SBRs were monitored throughout Runs 2 to 4. The plant staff also collected data for various parameters, from January 1997 to June 1999. Monthly average values for each of the parameters during this period are presented in Appendix F, and weekly average values for these parameters, from January 1999 to June 1999, are presented in Appendix G. 42 4.1.1 Methanol Induced Biological Nitrogen Removal The temporal operation of the control and experimental SBRs, during the time periods in which the SBRs were sampled, is shown in Figure 13. Note that the operation of the two SBRs is offset by two hours. The nitrite content was negligible in the SBR effluent (refer to Appendix D), and, as a result, the analysed NOx values were assumed to be in the nitrate form (refer to Figure 8 for the nitrate behaviour in the experimental SBR, without methanol addition, during a random 4-hour cycle). Figure 14 shows the nitrate profiles in both the control and experimental SBRs without methanol addition. The behaviour of the nitrate concentration at methanol dosages of 27, 54 and 81 L/day is shown in Figures 15 to 20. Two cycles were sampled for each experimental run. In each case, the SBRs were sampled throughout a 4-hour cycle, with the exception of the sampling period shown in Figure 20, in which the control SBR was sampled for a period of 6 hours. The raw data for these plots is shown in Appendix H. Each figure shows the best fit trend line for the nitrification and denitrification phases (determined with Microsoft Excel 97 regression software). The equations and R2 correlation coefficients for these trend lines are also shown. Zero-order reactions achieved a high correlation for both nitrification and denitrification reactions. Nitrification starts at the beginning of the Aeration phase, and denitrification starts after the completion of the Aeration phase and the depletion of any residual dissolved oxygen. Idle Aeration Methanol Injection Settle Decant (air off) (mixer and pump on, (air off) (air off) | | | air off) | | | 10:00 10:10 11:50 12:05 12:50 14:00 Experimental SBR Settle Decant Idle Aeration (air off) (air off) (air off) 9:50 10:50 12:00 12:10 13:50 Control SBR Figure 13: Temporal SBR Operation 43 Figure 14: Nitrate Trend Without Methanol Addition (Mar. 10, 1999) Figure 15: Nitrate Trend With 27 L/day Methanol Addition (Mar. 24, 1999) 44 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 16: Nitrate Trend With 27 L/day Methanol Addition (Apr. 7, 1999) 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 17: Nitrate Trend With 54 L/day Methanol Addition (Apr. 20, 1999) 45 5 y = 33.12x- 14.494 R 2 = 0.9812 O C t r l X E x p t 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 18: Nitrate Trend With 54 L/day Methanol Addition (May 6, 1999) 5 n ^ C t r l X E x p t 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 19: Nitrate Trend With 81 L/day Methanol Addition (May 18, 1999) 46 Figure 20: Nitrate Trend With 81 L/day Methanol Addition (June 3, 1999) The equations shown in Figure 8 and Figures 15 to 20 were used to calculate nitrification and denitrification rates. The mixed liquor volatile suspended solids (MLVSS) concentration (representing the concentration of microorganisms) is not steady state. The kinetic rates for nitrification and denitrification would typically be greater at a higher MLVSS. Therefore, the kinetic rates of nitrification and denitrification were calculated as a function of the MLVSS concentration (mg NOx-N/g MLVSS/day). A sample calculation and the data used to determine the kinetic rates are shown in Appendix H. Table 3 summarizes the zero-order kinetic rates of nitrification and denitrification for the two SBRs during the various methanol dosages. Carucci et al. [74] also concluded that zero-order kinetics provided a high correlation for denitrification reactions. 47 Table 3: Summary of Nitrification and Denitrification Kinetic Rates Date Methanol Dosage Nitrification Denitrification Experiment Control Experiment Control mg/L IVday mg N O x - N / mg N O x - N / mg N O x - N / mg N O x - N / g MLVSS/day g MLVSS/day g M L V S S / d a y g M L V S S / d a y 1st Rate 2nd Rate 06/30/98 0 0 7.20 0.06 03/10/99 0 0 9.31 0.61 -0.54 03/24/99 4.1 27 11.57 11.38 11.62 4.38 1.41 04/07/99 4.1 27 13.14 12.88 13.69 4.44 2.15 04/20/99 8.1 54 13.83 12.78 18.27 2.11 05/06/99 8.1 54 9.45 7.92 18.77 1.03 05/18/99 12.2 81 6.32 9.86 16.41 0.01 06/03/99 12.2 81 12.01 14.10 19.18 3.34 4.80 06/03/99 12.2 81 0.74 4.1.1.1 Nitrification The nitrification rates shown in Table 3 are plotted as a function of time in Figure 21. The average temperatures on the sampling dates are also shown in this plot. Methanol addition to the experimental SBR began on March 11 and was increased on a monthly basis until June 3. A similar nitrification rate trend is observed in both reactors throughout the time period shown. Nitrifying Figure 21: Nitrification Rates and Temperature 48 bacteria are extremely sensitive to various parameters including the presence of toxic substances, pH, dissolved oxygen concentration, mean cell retention time, and temperature. Any of these parameters may cause fluctuations in the nitrification rate, such as those observed in Figure 21. A variety of organic and inorganic agents can inhibit the growth and action of nitrifying bacteria. Trace metals, pH, and high concentration of ammonia and nitrous acid may be toxic to the nitrifiers. A narrow optimal range between pH 7.5 and 8.6 exists [1]. Systems acclimated to lower pH conditions have successfully nitrified and it is generally accepted that nitrification will work effectively between a range of pH 6 to 9. Nitrification reactions require dissolved oxygen concentrations greater than 1 mg/L [1]. Oxygen becomes the limiting nutrient if DO drops significantly below this level. Activated sludge systems designed for carbon oxidation and nitrification typically require DO concentrations greater than 2 mg/L. The pH and DO in both SBRs were continuously monitored. The results of the monitoring process are summarized and discussed in section 4.1.6. The pH in the experimental SBR fluctuated between pH 6.5 to 6.8, and had a standard deviation of 0.1. The control SBR varied from pH 6.5 to 6.7, with a standard deviation of 0.1. The average standard deviation between the pH in both reactors was less than 0.1. Therefore, the variation in pH in each SBR and the difference between the pH in both SBRs is negligible. The DO concentration in both SBRs was maintained at an average concentration of approximately 2 mg/L throughout the nitrification period. As a result, it is unlikely that pH and DO had a significant impact on the change in nitrification rate observed in Figure 21. The mean cell retention time (MCRT) is critical to the establishment of a viable population of nitrifying bacteria. If the MCRT is greater than the growth rate of the nitrifying organisms than the population of nitrifiers will be depleted or flushed from the SBR. Complete nitrification is typically assured in aerobic reactors if the MCRT is greater than 10 days. The MCRT in the control and experimental SBRs are significantly greater than this (refer to Appendix F). Thus, the MCRT does not have an influence on the behaviour observed in Figure 21. Temperature exerts a tremendous influence on the growth of nitrifying bacteria [1]. Temperature was monitored continuously in the SBRs. The average temperature for the day in which the samples were taken is shown in Figure 21. The nitrification rate appears to increase with an increase in temperature. The temperature started dropping after April 20 and the nitrification rate decreases drastically at this time. Slight variations in temperature are subsequently observed and the nitrification rate starts increasing slowly. The drastic decrease in nitrification rate is likely a shock 4 9 response to the decreasing temperature. Nitrifiers have an optimal temperature range of 25°C to 32°C. Therefore, the temperature in the SBRs is much colder than the temperature preferred by the organisms. Initially the nitrifiers were acclimated to the lower temperatures. The temperature slowly started increasing towards a higher and more optimal temperature for growth of nitrifying bacteria, and the nitrifiers lost their acclimation to the colder temperatures. As a result, the sudden decrease in temperature caused the observed drastic decrease in nitrification rate. The organisms slowly became acclimated to the lower temperatures and nitrification rates started increasing, even though the temperature remained relatively constant. This behaviour is demonstrated in Figure 22, which shows the relationship between nitrification rate and changes in SBR temperature. Therefore, the nitrifying bacteria in both SBRs were shown to be extremely sensitive to decreases in temperature. 15.0 -i 6.0 4 , 1 , 1 1 , , 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 Temperature (°C) Figure 22: Relationship Between Nitrification Rate and Temperature Methanol addition did not have an observable direct influence on the nitrification rate. Tam et al. also concluded that nitrification is not affected by carbon addition [50]. However, denitrification utilizes hydrogen ions and restores some of. the alkalinity consumed during the nitrification reactions. Therefore, improved denitrification as a result of methanol addition may have an indirect effect on the nitrifying bacteria by causing the pH to increase to a more optimal range. Although the pH was slightly higher in the experimental reactor, the difference is statistically negligible, and the methanol did not have a significant influence on the nitrification rate. 50 Figure 23 shows that both SBRs were achieving very effective ammonia removal. The maximum ammonia concentration will be reached at the end of the idle period, before aeration begins. In both the control and experimental SBRs, the ammonia concentration was reduced to approximately 0 mg N/L after 50 minutes of aeration. The ammonia concentration remained at 0 mg N/L until 100 minutes, after the start of the settling phase, for the control SBR and 70 minutes, after the start of the settling phase, for the experimental SBRs (methanol was not being injected into the experimental SBR during this sampling period and the settling phase started at 11.50, not 12.05, as shown in Figure 13). Subsequently, the influent began diffusing into the SBRs and, as a result, the ammonia concentration started increasing. I Expt - - Ctrl 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 Time Figure 23: Ammonia Trends (Dec. 1, 1998) The ammonia removal effectiveness of the two SBRs during the cycle shown in Figure 24 is significantly worse than that shown in Figure 23. The full 100-minute aeration cycle was required to reduce the ammonia concentration to approximately 0 and 0.3 mg N/L in the experimental and control SBRs respectively. This is twice as long as was required in Figure 22. The ammonia concentration in the control SBR started increasing at the start of the settling phase. In the experimental SBR, the ammonia concentration remained at less than 0.2 mg N/L, and then started increasing 45 minutes after the start of the methanol injection phase. Therefore, the ammonia concentration started increasing much earlier in Figure 24 than in Figure 23. The MLVSS 51 concentration in the control and experimental SBRs, respectively, was 3636 and 3337 mg/L on Dec. 1 1998, and 2448 and 2936 mg/L on June 3 1999. Therefore, the MLVSS was significantly less on June 3, and it is suspected that this resulted in a reduced overall nitrification rate compared to Dec. 1. The temperature in the SBRs was only 1 to 1.5°C higher on June 3, and, thus temperature likely had a negligible influence on the nitrification rate. ; Expt - - Ctrl 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 Time Figure 24: Ammonia Trend (June 3, 1999) Three important observations are made from the results of Figure 23 and Figure 24; firstly, the MLVSS concentration is critical to the overall nitrification rate; secondly, complete nitrification of the current influent ammonia loading may require the entire aeration cycle; and, thirdly, the time of ammonia reintroduction into the SBRs stresses the potential for short-circuiting to occur in a SBR system with a continuous influent feed (refer to section 3.1.1 for a more detailed definition of the influent feed operation). The third observation is discussed in more detailed below. The ammonia concentration in the experimental and control SBRs, as shown in Figure 24, starts increasing soon after the start of the settling phase and before decanting begins. This effect is not as pronounced in Figure 23, but the ammonia concentration still starts to increase during the 52 decant phase. Furthermore, the SBR uses level sensors to control the maximum attainable volume; the higher the level is in the SBR, the earlier the SBR begins decanting. This essentially maintains the total fluctuation in the volume of the SBR relatively stable at different flow rates. However, if the volume were kept constant, the concentration of the ammonia in the SBR would increase faster at higher ammonia loadings. Therefore, an increase in the influent ammonia concentration and influent flow would result in a decanted effluent with a higher ammonia concentration; essentially, the influent ammonia, during the decant phase, "short-circuits" to the decanter without nitrification. 4.1.1.2 Denitrification The denitrification rates shown in Table 3 are plotted as a function of time in Figure 25. The second denitrification rate observed in some runs, for the experimental SBR, are not shown in this plot. This plot also shows the respective methanol dosage being applied to the experimental SBR and the average temperature in each SBR on the sampling dates. The denitrification rates in the control SBR were typically slower than 2.2 mg NOx-N/g MLVSS/day. As a result, denitrification in the control SBR is considered to be very slow or insignificant. The fastest denitrification rate observed in the control SBR, 4.8 mg NOx-N/g MLVSS/day, was revealed on June 3. Two cycles were sampled on this day and the denitrification rates for the two cycles differ greatly (refer to Figure 20). This difference was potentially caused by a daily fluctuation in the concentration of natural short chain carbon (SCC) compounds in the influent to the control SBR. 1 4.0 12.5 / ^ _ _ ' \ / N N /l _ ~" - ^ ---T ? L / d a y ' / 5 4 L / d a y 81 L / d a y -- / / * --1 / / -- 1 X / >'/ -/ L 1 1 1 H 1 1 1 — - H ! 20 1 8 1 6 1 4 1 2 1 0 8 6 4 2 0 . o 3 O z - - X - - Expt T em p. — Ctrl T em p X — Expt k — C trl k Time Figure 25: Summary of Denitrification Rates 53 The decant from the digester and the filtrate from the belt-filter press is a potential source of the SCCs in the influent. Short chain carbons, or volatile fatty acids, are produced by fermentation processes, which require very low dissolved oxygen concentrations. The digesters are allowed to settle overnight and it is likely that the conditions required for fermentation are achieved. Subsequently, the thickened sludge from the digesters is pressed, and the filtrate from the belt-filter press and the decant from the digesters are discharged into the SBRs, along with any SCCs that may have been produced. The decant and filtrate were being discharged into the control SBR on June 3. It is possible that this may have caused the high denitrification rate observed in one of the control SBR's cycles. Methanol addition to the experimental SBR influenced denitrification in each of the experimental runs differently. Figure 15 and 16 shows the results for Rim 2, with a methanol dosage of 27 L/day. Two distinct denitrification rates are observed in the experimental SBR during Run 2. It is suspected that the initial fast rate is a result of the utilization of the available methanol for denitrification. However, once the methanol is depleted denitrification continues using the available natural carbon in the influent feed. This is represented in Figures 15 and 16 by the slower second rate of denitrification. Evidence that the observed initial fast rate was a result of the methanol addition is that the nitrate concentration was reduced by an equal amount of 1.9 mg/L, during the first rate period, in both Figures 15 and 16. The same methanol dosage of 27 L/day was added to the SBRs during the periods depicted by Figures 15 and 16, and hence it is expected that the nitrate concentration be reduced by an equal amount in both cases. Furthermore, according to the theoretical calculation for denitrification with methanol (refer to Section 2.2.1), for a methanol dosage of 27 L/day or 4.5 L per anoxic cycle, a total of 1.44 kg of NOx-N should be denitrified. Since the total change in nitrate concentration was 1.9 mg/L and the SBR volume is approximately 875 m3, the total denitrification observed in both Figure 15 and 16 is 1.66 kg of NOx-N, or 115 percent of the theoretical denitrification capacity. It is expected that the theoretical capacity be slightly lower than the actual denitrification capacity, since the denitrifying organisms will also utilize carbon in the influent wastewater not accounted for in the theoretical calculation. This provides further support for the hypothesis that the methanol was depleted at the end of the first rate. Also, the second denitrification rate was 4.4 mg NOx-N/g 54 MLVSS/day in both Figures 15 and 16, supporting the hypothesis that the second rate is a result of the utilisation of a background carbon source. Ekama et al. [75] also observed two linear denitrification rates in the anoxic zone. They attributed the initial fast denitrification rate, which persisted for a very short period, to the utilization of the readily biodegradable carbon in the influent, and the second slower denitrification rate, which lasted throughout the anoxic period, to the utilization of the slowly biodegradable carbon. The slowly biodegradable carbon fraction must be hydrolyzed and transformed into readily biodegradable substrate prior to becoming utilized by microorganisms [74]. Ffallin et al. [51] concluded that the bacteria in an anoxic reactor, with methanol as a carbon source consisted of one population denitrifying with methanol and another with compounds in the sewage. The results from the research performed by Ekama et al. [75]suggest that the slower second denitrification rate that is observed in Figures 15 and 16 is a result of the utilization of the slowly biodegradable carbon in the influent. Their results also suggest that three denitrification rates should be observed in Figures 15 and 16: the first fast rate is caused by the utilization of the readily biodegradable carbon in the influent and the methanol addition; the second rate is caused by the slowly biodegradable carbon in the influent and the methanol addition; and the third slow rate occurs after the depletion of the methanol, as a result of the utilization of the slowly biodegradable carbon in the influent. The first rate described above is likely not distinctively observed in Figures 15 to 20, since it did not persist for very long and samples were taken at 15 minute intervals. Two denitrification rates are also observed in the results for the experimental SBR shown in Figure 17, with a methanol addition of 54 L/day or 9 L/cycle. Once again, the first fast rate is a result of the utilization of methanol for denitrification. The theoretical denitrification capacity of methanol at a dosage of 9 L/cycle is 2.87 kg NOx-N, and, thus, the methanol should be depleted at the end of the first rate period. The total change in nitrate concentration observed in Figure 17 is approximately 2.5 mg N/L, corresponding to a total of 2.19 kg NOx-N removed or 76 percent of the theoretical denitrification capacity of the applied methanol dosage. The nitrate concentration at the end of the first rate was 1.6 mg N/L, suggesting that the limiting factor in the denitrification reaction was the availability of methanol. However, following the first rate period, the second denitrification rate is approximately 0 mg NOx-N/g MLVSS/day. A possible explanation for this observation is that the 55 influent biodegradable background carbon concentration was negligible, or that it was depleted in conjunction with the methanol during the first denitrification rate period. Figure 26 shows the phosphate and nitrate concentration in the SBR for April 20. Phosphate release appears to start at approximately 12:35. This time also coincides with the end of the denitrification period. Therefore, it is possible that a biodegradable carbon was available but the denitrifying organisms were not able to effectively compete for it with the phosphorus accumulating organisms or other heterotrophs. Another possible explanation is that the readily biodegradable carbon was indeed depleted, and the behaviour observed in Figure 26 is secondary phosphorus release. As a result, only endogenous denitrification, which is extremely slow, could occur. Figure 26: Phosphate and Nitrate Concentration in Experimental SBR (Apr. 20, 1999) The nitrate in the experimental SBR was completely denitrified in Figure 18 and 19, with methanol dosages of 54 L/day and 81 L/day respectively. Both of these plots show only one significant denitrification rate. However, two distinct reaction rates are observed in Figure 20, with a methanol dosage of 81 L/day or 13.5 L/cycle. The nitrate concentration was depleted by approximately 1.77 mg/L during the first rate phase. This corresponds to 1.55 kg NOx-N removed, which has a theoretically denitrification demand of 4.8 L CH3OH. Therefore, it is unlikely that the methanol was depleted following the end of the first rate period. The slower second rate may again 56 be the result of the denitrifying organisms' inability to effectively compete with other heterotrophs. However, unlike the situation described in Figure 17, in which the competition is for substrate in the influent, the competition in Figure 20 is for the remaining methanol. This hypothesis will be discussed further in this section. The first denitrification rate observed in the experimental SBR, as shown in Figures 15 to 20, is therefore associated with denitrification reactions utilizing methanol as a carbon source. Figure 27 shows a plot of the average denitrification rate as a function of the methanol dosage. The error bars indicate the 90 percent confidence level of the data about the mean. The relationship between the K D N =-0.0046-m2 + 0.5916 m denitrification rate and the methanol dosage can be expressed as: where: K D N = overall denitrification rate, mg NOx-N/g MLVSS/day; and m = methanol dosage, L CH3OH7day. This expression has a very high R2 correlation coefficient, as shown in Figure 27. The denitrification rate can also be expressed as a function of the methanol concentration (mg CH3OH/L) in the SBR by the following expression: K D N =-0.203-M2 +3.93-M where: M = methanol concentration, mg CH3OH/L. Figure 27: Methanol Dosage and Denitrification Rate Linear Relationship 57 Figure 27 suggests that a maximum denitrification rate is achieved at a methanol dosage of approximately 60 L/day. However, the denitrification rates at methanol dosages between 27 and 54 L/day are uncertain. It is possible that the maximum denitrification rate is reached with a smaller methanol dosage than that suggested by the relationship shown in Figure 27. A slight decrease in the denitrification rate is observed at a methanol dosage of 81 L/day. However, there is insufficient data to evaluate the significance of this small variation. The highest denitrification rate (Figure 20) was also observed at a dosage of 81 L/day, and it is possible that the results from May 18 (Figure 19), the second data set with a methanol dosage of 81 L/day, are suspect. The microbial growth rate will typically increase with increasing concentration of substrate until an optimal growth rate is achieved. At high substrate concentrations, the available enzymes required for biological reactions will eventually have all of their active sites occupied by the substrate or product molecules [25]. As a result, no further increases in methanol concentration will affect the denitrification reactions, as demonstrated by Figure 27. An excessive methanol concentration, as described above, was added to the SBR during the period shown in Figure 20. As previously discussed, two distinct denitrification rates are observed during the anoxic period. During the first rate, denitrification reactions are limited by the availability of unoccupied active sites on the enzymes. As a result, an excess of methanol remains in solution and is available for other heterotrophs to consume. Furthermore, with a sufficient acclimatization period and the appropriate dissolved oxygen conditions it is possible that organisms develop the ability to ferment methanol to formate [25]. An acclimatization period of one month to a methanol dosage of 81 L/day had elapsed at the time the samples shown in Figure 20 were taken. Also, the sludge in the experimental SBR had been acclimatized to methanol for a three month period. The availability of fermentation products should also promote E B P R processes. Figure 28 shows that significant phosphate release began at approximately 12:35, which corresponds to the end of the first denitrification period shown in Figure 20, and the onset of fermentation. This supports the hypothesis that the occurrence of a second slower denitrification rate, at a low nitrate concentration, was the result of competition for the available methanol between the denitrifying organisms and other heterotrophs. 58 Figure 28: Phosphate and Nitrate Concentration in Experimental SBR (June 3, 1999) The removal rate of nitrate and the microbial growth rate are affected by temperature, and, thus, denitrifying organisms are sensitive to changes in temperature [1]. Figure 25 compares the temperature in each reactor to the denitrification rate in each reactor. Initially, the denitrification rate in the experimental reactor increases with increasing temperature and increasing methanol dosage. The temperature started decreasing after April 20, but the denitrification rate remained stable at the same methanol dosage. Furthermore, the denitrification rate did not appear to be significantly affected by the lower temperature with a subsequent increase in methanol dosage. The control SBR exhibited an increase in the denitrification rate with an increase in temperature, and a subsequent decrease in the denitrification rate with decrease in temperature. However, the highest denitrification rate observed in the control SBR did not occur at the warmest temperature. It is likely that temperature does influences the denitrification rate to some extent. However, it is difficult to conclusively state how sensitive the denitrification rate is to changes in temperature from the information collected during this study. The control SBR is limited in carbon required for denitrification reactions. As a result, the denitrification rate would likely be more sensitive to changes in the available carbon in the SBR, than to changes in the SBR temperature. This explains 59 why the fastest denitrification rate observed in the control SBR occurred at a lower temperature. It is difficult to quantify the influence of temperature changes to the denitrification rate in the experimental SBR, since methanol dosage was changing during the duration of this experiment. However, the denitrification rate did not decrease at a methanol dosage of 54 L/day, after the temperature took its initial drastic decrease. This suggests that the denitrifying organisms are not extremely sensitive to slight decreases in temperature. The denitrification rate can be described by the following equation [1]: K D N = K D N 2 0 x 1 . 0 9 ( T - 2 0 ) x ( l - D O ) , where: K D N = overall denitrification rate, -mg N O X - N / g MLVSS/day; K D N 2 0 = denitrification rate at 20°C, -mg N O X - N / g MLVSS/day; T = wastewater temperature, °C; and, D O = dissolved oxygen in the wastewater, mg/L. This expression is a modification of the Arhenius temperature equation [26], and it assumes that the Streeter Phelps Temperature Sensitivity Coefficient is equal to 1.09. The specific denitrification rate at a standard of 20°C was determined from the above expression. A very low D O concentration was observed during the denitrification period (refer to Section 4.1.6), and, for the purpose of this calculation, the D O was assumed to be 0 mg/L. The denitrification rates shown in Figure 27 were measured at temperatures ranging from 12.4°C to 15.7°C (i.e., each denitrification rate is at a different temperature). Figure 29 shows the relationship between the calculated specific denitrification rate and the methanol dosage at a standard of 20°C. 60 T e m p e r a t u r e =20 °C 0 10 20 30 40 50 60 70 80 M e t h a n o l Dosage (L/day) Figure 29: Methanol Dosage and Specific Denitrification Rate Relationship (20°C) The experimental SBR is mixed for fifteen minutes to distribute the methanol throughout the entire SBR volume. Subsequently, the contents of the SBR were allowed to settle. The control SBR was not mixed. Mixing increases reaction rates and this was also a probable cause of the increased reaction rates observed in the experimental SBR, compared to the control SBR. 4.1.2 Methanol Induced Biological Phosphorus Removal Figures 30 and 31 show the phosphate behaviour during a random 4-hour cycle, without methanol addition. The behaviour of the phosphate with methanol addition is shown in Figures 32 to 37. Two cycles were sampled for each experimental run. In each case, the SBRs were sampled throughout a 4-hour cycle, with the exception of the sampling period shown in Figure 36, in which the control SBR was sampled for a period of 6 hours. The raw data for these plots is shown in Appendix H . Refer to Figure 13 for the temporal operation of the control and experimental SBRs during the time periods in which the SBRs were sampled. 61 Figure 30: Phosphate Trend in Experimental SBR without Methanol (June 30, 1998) Figure 31: Phosphate Trend without Methanol (Mar. 10, 1999) 62 3.4 3.2 A 2.2 A 2 -I ' ' ' 1 1 '—' '—' 1 '—' 1—1 1—' 1 ' ' 1 ' ' ' ' 1 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 32: Phosphate Trend with 27 L/day Methanol Addition (Mar. 24, 1999) 4.2 -, 3 -2.8 -I 1 1—1 1 1—1 1—1—1 1—1—1 1 1 1 1 1 — 1 1 1 1 1 1 1 1 9:00 10:00 11:00 12:00 13:00 14:00 Time Figure 33: Phosphate Trend with 27 L/day Methanol Addition (April 7, 1999) 63 3.8 3.6 eC 3.4 3.2 4 3 4 1 2.8 2.6 2.4 Air On 9:00 Air On ' 1 L 10:00 - t -11:00 12:00 Time 13:00 14:00 -C t r l X Expt Figure 34: Phosphate Trend with 54 L/day Methanol Addition (April 20, 1999) 3.4 3.2 a. « 2 : 2.6 "I 2.4 I 2.2 9:00 Air On 10:00 I 1 1 , , , 1 L 11:00 12:00 Time ' 1 L 13:00 14:00 • - - Ctrl — X - Expt Figure 35: Phosphate Trend with 54 L/day Methanol Addition (May 6, 1999) 64 Figure 36: Phosphate Trend with 81 L/day Methanol Addition (May 18, 1999) Figure 37: Phosphate Trend with 81 L/day Methanol Addition (June 3, 1999) 65 Biological phosphate release occurs under anaerobic conditions and in the presence of short chain carbons (SCC), or volatile fatty acid (VFA). The result of phosphate release is the biosynthesis of poly-hydroxyalkanoates (PHA), which are required for phosphate uptake to take place. Phosphate uptake will occur in the subsequent aerobic period. In the cases where methanol was not added to the SBR, phosphate uptake and release was not observed to a significant extent (refer to Figure 30, 31, and the profiles for the control SBR shown in Figures 32 to 37). Typically, the phosphate concentration remained constant throughout the settling phase. An increase in the phosphate concentration is observed in some cases at approximately the same time in which the aeration phase begins. This behaviour is especially pronounced in Figures 30, 35, and the first aerobic cycle shown in Figure 37. However, the anaerobic conditions required for phosphate release were not attained since the nitrate concentration in each case was greater than 2 mg N/L (refer to Figures 8, 18, and 20). As a result, the predominant cause of this behaviour is likely the result of mixing of a higher strength influent into the SBR, and not the result of biological phosphate release. Also, the phosphate concentration typically remained constant throughout the aeration phase in the SBRs that did not have methanol addition. However, Figures 30, 33, 35, and 37 show a slight decrease in the phosphate concentration, and some phosphate uptake may have occurred in these cases. Still, the phosphate uptake is negligible, except for the case observed in the experimental SBR without methanol addition, shown in Figure 30. A likely explanation for the possible phosphate uptake observed during this particular cycle is that phosphate release and the production of PHAs may have occurred during the previous cycle. This may have potentially happened if the previous cycle had a low nitrate concentration and a natural SCC source, allowing denitrification to take place. Figure 8 shows that the nitrate concentration at the end of the previous cycle was around 1 mg N/L, and, thus, the nitrates may have been sufficiently reduced for phosphate release to take place in the previous cycle. As a result, PHAs may have been produced and stored by the PAOs, to be utilized in the cycle shown in Figure 30 for biological phosphate uptake. In contrast, the experimental SBR appears to be achieving biological phosphorus uptake and release during the cases in which methanol was added. In Figures 32 to 37, the phosphate concentration in the experimental SBR decreased during the initial aeration period. It subsequently remained relatively constant during the methanol injection phase and the early part of the settling phase. Figure 38 shows the nitrate and phosphate profiles for a typical cycle in the experimental SBR 66 with a methanol dosage of 54 L/day. Appendix I shows the nitrate and phosphate profiles for each of the runs in the experimental SBR. In each case, the phosphate concentration begins increasing, once the nitrates have achieved a very low concentration. From the plots shown in Appendix I, phosphate release is evident with nitrate concentrations ranging from 0 to 1.5 mg N/L. This behaviour becomes more pronounced with increasing methanol dosages. 2.5 • Z e o U 1.5 Z 0.5 Time (Ins., not to scale) Figure 38: Phosphate Release at Nitrate Depletion (May 6, 1999) 3.2 3 a; be 2.8 2.6 2.4 u 2.2 -S o JS a. - - X - -— © -•NOx -P04 Once the nitrates are depleted, the PAOs, using energy derived from the cleavage or hydrolysis of the polyphosphate compounds in the cells, sequester the available VFAs. This results in the release of phosphate into the liquid phase outside the cell and the formation of PHAs. During the aeration phase, the PHAs are oxidized, releasing energy that is stored mostly as polyphosphate. As a result, an uptake of the available phosphate takes place. The influent sewage is suspected of having a natural supply of VFAs. It is also possible that heterotrophs utilize any remaining methanol not utilized during the denitrification process to produce VFAs. However, a very long acclimatization period is required to trigger such a biochemical reaction with methanol. It was suggested in Section 4.1.1.2 that this may have occurred on June 3, 1999 (refer to Figure 20). Figures 32 to 37 also indicate that an increased total phosphate release typically occurs with an increase in methanol dosage. However, phosphate release was still observed on April 20, 1999, although the methanol concentration was completely depleted following the denitrification reaction (refer to Figure 26 and 67 the discussion in Section 4.1.1.2). This suggests that phosphate release still occurred in the absence of methanol and that there must be a natural SCC source in the influent wastewater. The total phosphate release per MLVSS was determined for each of the experimental runs and plotted as a function of the duration of the phosphate release period. A linear relationship with a high correlation coefficient exists between these two parameters, as shown in Figure 39. The slope of this graph is the phosphate release rate, or 2.7 mg P/g MLVSS/day (refer to Appendix J for a summary of the data used in Figure 39). Figure 39: Determination of Phosphate Release Rate If a methanol residual existed following denitrification, it is expected that the quantity of residual methanol would vary depending on the nitrate concentration at the beginning of the anoxic cycle. From Figures 15 to 20, it is evident that the initial nitrate concentration fluctuated throughout the different experimental runs. Then, if this carbon source were being utilized for biological phosphate removal, it would be expected that fluctuations in the phosphate release rate would also be observed. However, the linear relationship shown in Figure 39 has a very high correlation coefficient and shows very little fluctuations. This suggests that a relatively constant supply of natural SCCs may have been present in the influent. Tarn et al. [32] and Randall et al. [9] also concluded, from their research, that methanol does not support EBPR. 68 Although methanol is likely not being used as the carbon source for biological phosphorus removal, it does allow for the biological phosphorus removal process to take place by depleting the nitrate concentration. In this respect, methanol is critical to the biological phosphorus removal process. Furthermore, it is still possible that the second slower denitrification rate observed in Figure 20 was caused by competition between the denitrifying organisms and other heterotrophs. Fermentation activity is relatively slow compared to E B P R reactions [58], and it is possible that the fermenters consume the methanol but do not process it quickly enough into a form that can be readily utilized by PAOs. 4.1.3 Areal Analysis of SBR BNR Performance The experimental SBR was sampled during one cycle at three different locations to determine if there are large variations in nutrient concentrations. Samples were withdrawn from the front, middle, and back of the SBR, along the central catwalk, at an approximate depth of one meter below the surface. Figures 40 and 41 show the results for nitrate and phosphate, respectively. Figure 40: Concentration of Nitrate in SBR 69 10:00 11:00 12:00 13:00 14:00 Time (hours) Figure 41: Concentration of Phosphate in SBR The nitrate concentration, shown in Figure 40, was typically larger at the front, smaller in the middle, and smallest in the back of the SBR. Therefore, samples withdrawn from the middle of the SBR are representative of approximate average conditions. The average standard deviation for the nitrate concentration observed at the three points was 0.26 mg N/L. Therefore, a negligible difference between the nitrate concentration at three sampling points exists. Similarly, Figure 41 also shows a negligible difference between the phosphate concentration at the three sampling points. The average standard deviation for the phosphate concentration observed at the three points was 0.05 mg P/L. As a result, the samples taken in the middle of the SBR provide a reliable approximation of the conditions throughout the entire SBR area, at a one meter depth below the surface. 70 4.1.4 Solids Production The addition of carbon, such as methanol, to a bioreactor will result in an increase in solids production. Solids production is a critical component of the wastewater treatment system, due to the large proportion of costs associated with the treatment and disposal of solids. The increase in solids production was determined by taking the difference between the mass of solids wasted from the experimental SBR with methanol addition, and the control SBR without methanol addition. Average values for the increase in solids production were determined for each of the experimental runs, at the various methanol dosages. This data, which is plotted in Figure 42, represents the observed solids production as a result of methanol addition during the full-scale experiment. The error bars indicate the range of error about the mean with a 90 percent confidence level. According to the linear model shown in Figure 42 (determined using Microsoft Excel 97 regression software), the observed solids production rate is 0.21 kg of sludge per L of methanol added. 5 0 4 0 « •a | | 3 0 y = 0 . 2 0 6 2 x + 4 . 7 6 6 R 2 = 0 . 2 9 7 4 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 - 1 0 Methanol Dosage (L/day) Figure 42: Observed Solids Production from Methanol Addition 71 However, a very low correlation exists between the data and the linear model of the actual solids production shown in Figure 42. Furthermore, the observed increase in sludge production varied significantly during each experimental run and there is a large error range. The solids wasting rate is controlled by the plant operator and it takes some time before the operator will notice an increase or decrease in the solids production and decides to modify the wasting rate accordingly. Therefore, the wasting rate does not accurately represent instantaneous changes in the solids production of the two SBRs. However, by using multiple wasting rates over a period of time, a reasonable estimate of the actual solids production at each of the methanol dosages can be made, by plotting a linear model through the average values (as shown in Figure 42). The longer the period of time, the more likely that the SBR solids concentration will reach a steady state and the more accurate the model will be. Each methanol dosage was applied for one month and wasting rates were determined weekly, based on biweekly measurements of sludge wasted from each SBR. The solids production in the experimental SBR, as a result of methanol addition, was also determined using a theoretical model and a kinetic model. The theoretical model employed the following expression [1]: N C V +108 C H 3 O H + H + -> 0.065 C 5 H 7 0 2 N + 0.47 N 2 +0.76 C 0 2 +2.44H 2 0 where C5H7O2N represents the formation of cell mass or sludge. The following kinetic model was used to determine the solids production [1]: dt f dL a — -b-S V dt • V where: V = volume, m3; S = solids concentration, mg/L; L = substrate concentration, mg/L; t = time, day; a = yield coefficient, mg VSS/mg COD; and, b = endogenous decay coefficient, day"1. This kinetic model can be manipulated so that the yield and endogenous decay are represented by a net yield coefficient, a', as follows: dt dL dt • V 72 The net yield coefficient was determined for the control and experimental SBRs by plotting the historical data for influent COD and the sludge mass wasted (kg/day). This plot is shown in Figure 43. The net yield coefficient for both SBRs was very close in magnitude. A net yield coefficient of 0.37 was used to determine the solids production based on the kinetic model. 180 ~i 160 A 140 •3 120 "S ioo 80 „ 60 ao •a 53 4 0 20 0 Experimental SBR y =0 .3658x R 2 = 0.4082 Control SBR y = 0.3506x R 2 = 0.2001 50 100 150 200 250 Influent COD (kg/day) oCtrl X Expt 300 350 Figure 43: Derivation of Net Yield Coefficient The solids production at each of the methanol dosages based on the theoretical and kinetic models is shown in Figure 44. The theoretical model reveals a solids production of approximately 0.17 kg of solids per L of methanol, whereas the results from the kinetic model provides a solids production of 0.44 kg of solids per L of methanol. The observed (Figure 42) and theoretical solids productions are very close in magnitude and this provides credibility to these two models. The kinetic model, however, utilizes a net yield coefficient for a generic carbon source, and thus this model is not specific to methanol. Appendix K summarizes the data and calculations used to determine the observed solids production, and the solids production based on the theoretical and kinetic models. 73 40 n Methanol Dosage (L/day) Figure 44: Theoretical and Kinetic Models for Solids Production The theoretical solids production for a methanol dosage of 54 L/day is 9.1 kg/day. The average solids production in the control SBR from January to June 1999 is approximately 61 kg/day. Therefore, methanol will constitute a theoretical increase of 15 per cent in the solids production, at a methanol dosage of 54 L/day. The settleability of the sludge is another important characteristic related to the solids in the SBR. The sludge volume index (SVI) is used as an indication of the settling characteristics of the sludge. The experimental SBR was observed to have poor settling properties, compared to the control SBR, and it was important to determine whether this was the result of the methanol addition or another parameter. Sludge bulking in the experimental SBR was initially suspected of causing the observed poor settling. A bulking sludge is one that exhibits poor settling typically caused by the growth of filamentous organisms such as Nocardia. A low dissolved oxygen concentration is the predominant 74 cause of bulking [4]. The DO was greater than 2 mg/L during the aeration phase; however, the DO was very low during the settling phase and Nocardia may have potentially formed in the experimental SBR at this time. However, scum formation typically forms on the surface of the water, as a result of this. A significant amount of scum formation was not observed on the surface of the experimental SBR. Furthermore, if inadequate DO caused bulking in the experimental SBR, it is expected that bulking would have also been observed in the control SBR, since the same DO concentration was maintained in both SBRs. Figure 45 and 46 show microscopic photographs of the sludge floe in the control and experimental SBRs respectively. Similar extents of filamentous organism growth are observed in both the experimental and control SBRs. Figure 45: Microscopic Photograph (lOOx) of Sludge in Control SBR (June 3, 1999) 75 Figure 46: Microscopic Photograph (lOOx) of Sludge in Experimental SBR (June 3, 1999) The average monthly SVI was determined in Appendix F, and the result is plotted in Figure 47 for 1997 and 1998. This plot indicates that very little difference exists between the SVI in the control and experimental SBRs, without methanol addition. Even when a drastic increase in SVI is observed, such as on March 1997, a similar increase was observed in both SBRs. 76 Figure 47: Sludge Volume Index Without Methanol Addition The average weekly SVI was calculated from January 1 to June 11, 1999. The calculation of this data is summarized in Appendix G and the results are plotted in Figure 48. Methanol was added to the experimental SBR from March 11 to June 11. During this period a distinction between the SVI in the control and experimental SBRs is observed in Figure 48, suggesting that methanol addition had an influence on the settling qualities of the sludge. The average difference between the SVI in the experimental and control SBRs (A SVI) is plotted as a function of the methanol dosage in Figure 49. The SVI was determined five times per week and the A SVI values shown in Figure 49 are based on an average during the period in which each methanol dosage was applied. The data shown in Figure 49 has a large error range, as indicated by the error bars that represent the 90 percent confidence level of the data about the mean. The relationship between the A SVI and the methanol dosage is similar to the relationship between the denitrification rate and the methanol dosage (refer to Figure 27). 77 Figure 48: Sludge Volume Index as a Function of Methanol Addition Figure 49: Relationship Between Methanol and AS VI 78 The average denitrification rate and average A SVI were computed for each experimental run. Figure 50 shows that a linear relationship exists between the average denitrification rate (KD N) and the average A SVI as follows: ASVI=1.73-K D N , where A SVI is the average difference between the SVI in the control and experimental SBRs (mL/g MLSS). As a result, the A SVI is not suspected of being directly influenced by the methanol dosage. Instead, the methanol dosage influences the denitrification rate, which, in turn, influences the A SVI. Therefore, the methanol dosage is indirectly related to the A SVI. A high correlation exists between the average A SVI and the average denitrification rate. The error bars shown in Figure 50 represent the 90 percent confidence level for the data about the average value. 40 i 35 -30 -Zero-Order Denitrification Reaction Rate Constant (mg NOj-N/g MLVSS/day) Figure 50: SVI as a Function of the Denitrification Rate As a result, the settleability deteriorates with an increase in the denitrification rate, for a constant concentration of MLVSS. A likely explanation for this phenomenon is that increases in the denitrification rate results in an increased amount of nitrogen gas (N2) production. Much of the nitrogen gas formed in the sludge layer is trapped in the sludge mass. Therefore, an increase in the production of nitrogen gas will result in an increase in the buoyancy of the sludge mass. If enough gas is formed, rising sludge, whereby the sludge floats to the surface, may eventually occur. 79 Furthermore, the influent nitrogen loading may be the limiting component for nitrogen gas production during periods in which there is a fast denitrification rate. Therefore, the large error range for the A SVI may be the result of fluctuations in the total nitrogen gas production during each experimental run. 4.1.5 Methanol-Induced BNR: Efficiency of Treatment Process The various nitrogen and phosphorus components in the influent and effluent of each SBR are shown in Appendix D for each experimental run. This data was used to determine the total nitrogen (TN) and total phosphorus (TP) removal efficiencies. The relationship between the average T N removal efficiency and the applied methanol dosage is shown in Figure 51 for the experimental SBR. The error bars indicate the 90 percent confidence level of the data. This plot indicates that the total nitrogen removal efficiency increases with an increase in methanol dosage, until a maximum removal efficiency is attained. This corresponds well with the relationship between the denitrification rate and the applied methanol dosage shown in Figure 27 and discussed in Section 4.1.1.2. The nitrogen removal efficiency, for a given hydraulic retention time and solids retention time, is therefore limited by the denitrification rate. 97 A 88 0 10 20 30 40 50 60 70 80 Methanol Dosage (L/day) Figure 51: Total Nitrogen Removal Efficiency and Methanol Dosage 80 The influent total nitrogen and short chain carbon (SCC) loading may vary daily and seasonally. For a given TN loading and methanol dosage, the TN removal efficiency should remain relatively constant. However, for the same methanol dosage, if there is an increase or decrease in the TN loading, then the TN removal efficiency will also change accordingly. Similarly, variations in the influent SCC loading will also influence the denitrification rate and, hence, the TN removal efficiency. This constitutes an explanation of the large range of error shown in Figure 51. Figure 51 shows a maximum increase in the TN removal efficiency of approximately 6 percent, as a result of the methanol addition. This result does not accurately represent the profound influence that methanol has on the biological nitrogen removal process. For example, in the control SBR, denitrification is not occurring to a significant extent. However, Figure 51 shows an 89 percent TN removal efficiency without methanol addition. Wastewater flow rates typically follow a diurnal pattern [4]. Based on the results from this study (refer to Section 4.1.7), the influent total nitrogen concentration at the Agassiz WWTP is inversely proportional to the flow rate. Minimum flows occur when water consumption is lowest during the early morning hours. The base flow during low flow periods consists of infiltration and small quantities of sanitary wastewater. The first peak flow generally occurs when wastewater from peak morning use reaches the treatment plant in the late morning. A second peak flow generally occurs in the early evening between 19:00 and 21:00 hours, but this varies with the size of the community and the length of the sewers. Metcalf and Eddy [1] provide a discussion and illustration of a typical diurnal flow pattern. The influent TN samples were taken at approximately 9:45 hours and the flow rate is likely increasing at this time. Thus, the influent TN concentration is likely decreasing at the time the sample was taken. Figure 52 indicates that this is the case and that the minimum influent TN concentration occurs after 14:00 hours. Furthermore, a 53 percent decrease in the influent TN concentration from 9:55 to 13:50 is observed in Figure 52. This is a significant reduction in the daily influent TN concentration. Thus, during the high flow periods, the SBR will be filled with an influent of low TN concentration, and during the low flow periods, the high TN concentration of the influent will subsequently be diluted in the SBR. Therefore, it is suspected that dilution is a predominant cause of the high TN removal efficiency, or, more appropriately, TN reduction efficiency, observed in the SBR without methanol addition. The collection of hourly grab samples of the influent and effluent throughout a 24-hour period would have clarified this issue. 81 Figure 52: Influent Total Nitrogen Concentration (June 30, 1998) A significant portion of the influent nitrogen is also utilized for biosynthesis. This nitrogen is wasted with the microorganisms to the digester for further processing. The experimental SBR during Run 2 had an average influent concentration of 64 mg TN/L and an average wastewater flow rate of 276 mVday in, or a TN loading of 18 kg/day. The average sludge production was 72 kg/day. Assuming the biomass is represented by C 5 H 7 N0 2 [1], then 9 kg/day of TN are theoretically required for biosynthesis. Therefore, biosynthesis reactions accounted for 50% of the TN removal during Run 2. The dilution of the influent TN loading in the SBR is not considered to be a treatment process. The total mass of excess nitrogen (i.e., nitrogen not utilized for biosynthesis) that is expelled as effluent will not be reduced, if dilution is the predominant nitrogen reduction mechanism. In order to reduce the mass of nitrogen in the SBR, denitrification needs to take place. As previously indicated, significant denitrification was not observed in the control SBR without methanol addition. However, fast denitrification rates were observed in the experimental SBR, as a result of the methanol addition. As a result, methanol was "critical" to the TN removal efficiency of this system. 82 The relationship between the average TP removal efficiency and the applied methanol dosage is shown in Figure 53 for the experimental SBR. The error bars indicate the error range with 90 percent confidence. This plot indicates that the total phosphorus removal efficiency increases with an increase in methanol dosage, according to the expression shown.' The denitrification rate increases with an increase in the methanol dosage until the maximum denitrification rate is attained. The nitrate will be depleted earlier at higher denitrification rates. However, the time in which the nitrate is depleted, during the denitrification period, is also dependent on the influent ammonia loading. Phosphate release is required for enhanced biological phosphorus removal and it will only take place once the nitrate is depleted. As a result, the initial increase in the TP removal efficiency, with an increase in the methanol dosage, is a result of the increasing denitrification rate. However, at a methanol dosage of 81 L/day, the TP removal efficiency still increases, although the denitrification rate remains constant. This is due to the significantly reduced influent nitrogen loading (refer to Appendix D), which allows the nitrate to be depleted earlier at a constant denitrification rate. 80 -, > < 20 - | , , , , , , , r -0 10 20 30 40 50 60 70 80 Methanol Dosage (L/day) Figure 53: Relationship between TP Removal Efficiency and Methanol Dosage 83 As with the TN removal efficiency, variations in the influent total phosphorus and short chain carbon (SCC) loading will influence the biological phosphorus removal rates and, hence, the TP removal efficiency. This explains the large range of error shown in Figure 53. Furthermore, based on the results from this study (refer to Section 4.1.7), the influent TP concentration at the Agassiz WWTP is also inversely proportional to the flow rate. The biological uptake and release of phosphorus was not observed in the control SBR without methanol. As a result, dilution is a predominant cause of the high TP reduction efficiency of 43 percent, observed in the SBR without methanol addition. A small portion of the influent phosphorus is also utilized for biosynthesis, and subsequently wasted to the digester along with the microorganisms. Therefore, the maximum increase of 26 percent to the TP removal efficiency, as shown in Figure 53, as a result of methanol addition, "underestimates" the importance of methanol to the biological phosphorus removal process. Furthermore, the TP removal efficiency is also limited by the operation of the existing programmable logic controller (PLC). The decanter starts lowering at the start of the decant phase (12:50), but it actually starts decanting much later. The actual start of the decanting process depends on the water level in the SBR. Therefore, biological phosphorus release is occurring during much of the decant phase (refer to Figures 32 to 37). Ideally, the decanting should start at the end of the denitrification period when the nitrate concentration and the phosphate concentration are both at a minimum. Therefore, the "potential" for increasing the TP removal efficiency in this SBR system, as a result of methanol addition, is actually much greater than that shown in Figure 53. This potential could be realized by implementing PLC modifications, improved process monitoring, and a real-time control system. The relationship between the average raw wastewater COD removal efficiency (does not include the contribution of methanol to the influent COD) and the methanol dosage in the experimental SBR is shown in Figure 54. The COD removal efficiency decreased with an increasing methanol dosage. A total reduction in the COD removal efficiency of 18 percent was observed in the experimental SBR at a methanol dosage of 81 L/day. This is a substantial reduction in the COD removal efficiency. However, the COD removal efficiency of the control SBR without methanol was also reduced. The COD removal potential of the SBR is dependent on several non-steady state parameters such as the influent COD concentration (e.g. a higher COD removal efficiency is expected 84 for a lower COD influent concentration), which may vary from day to day. By comparing the COD removal efficiency of the experimental SBR, with methanol addition, with that of the control SBR, without methanol addition, as shown in Figure 55, we reduce the influence of some of these parameters on the COD removal efficiency. Figure 55 shows an 8 percent reduction in the COD removal efficiency, as a result of a methanol dosage of 81 L/day. 90 -, 65 H 60 -I , 1 1 , , 1 1 , 1 0 10 20 30 40 50 60 70 80 90 Methanol Dosage (L/day) Figure 54: COD Removal Efficiency as a Function of Methanol Dosage 16 n 0 10 20 30 40 50 60 70 80 Methanol Dosage (L/day) Figure 55: Reduction in COD Removal Efficiency as a Function of Methanol Dosage 85 As previously mentioned, the influent TN loading will vary seasonally and these variations will influence the requirement for methanol. A higher methanol dosage will be required for a higher TN influent concentration. Conversely, if the TN influent concentration decreases, less methanol is required for complete denitrification to occur. The TN influent concentration based on morning grab samples, as shown in Figure 56, was initially increasing during the 27 L/day methanol addition. As a result, the methanol demand for denitrification would be increasing. Accordingly, a very small change in the COD removal efficiency is shown in Figure 55. Subsequently, during the 54 and 81 L/day methanol additions, the TN influent concentration started to decrease, and thus the methanol demand also decreased. Figure 55 shows an increasing rate in the reduction of the COD removal efficiency with an increase in methanol dosage. Since the methanol demand was decreasing and the methanol concentration was increasing, a residual methanol will result that will contribute to a reduced COD removal efficiency. Figure 56: Total Nitrogen Influent Concentration The average solids removal efficiency is plotted in Figure 57 as a function of methanol dosage. The pattern shown on this graph indicates that the solids removal efficiency is inversely 86 related to the denitrification rate (refer to Figure 27). As indicated in Figure 50, the denitrification rate is linearly related to the SVI. Therefore, at a higher rate of denitrification the settleability is reduced. This may potentially cause a reduction in the effluent solids quality and also the solids removal efficiency if the sludge blanket, after settling, is too high to accommodate decanting. However, the largest average reduction in the solids removal efficiency observed during the experiment was only 1.4 percent. Therefore, methanol addition had a negligible influence on the solids removal efficiency of the experimental SBR. 100.0 -i M a u | 98.0 -97.5 -I , , 1 , , 1 1 1 1 0 10 20 30 40 50 60 70 80 90 Methanol Dosage (L/day) Figure 57: Solids Removal Efficiency as a Function of Methanol Dosage 4.1.6 On-Line Monitoring Results The control and experimental SBR were monitored using submerged probes throughout experimental Runs 2 to 4. The SBR temperature did not demonstrate a significant diurnal variation or a variation during the SBR's 4-hour cycle. The average weekly temperature in the control and experimental SBRs is summarized in Figure 58 for experimental Runs 2 to 4. Initially, the 87 temperature in both SBRs increased gradually as a result of increasing atmospheric temperatures. The decrease in temperature, observed in late April, was likely caused by increased infiltration of cold groundwater due to a higher water table. Figure 58: Average Weekly Temperature The pH was also monitored with probes in each of the SBRs. However, consistent results were not obtained using the probes. The plant operators take one pH reading, in each of the SBRs, five times weekly. Figure 59 shows the average weekly pH for each of the SBRs during experimental Runs 2 to 4, using the data collected by the plant operators. A slight fluctuation in the pH is shown in Figure 59. The fluctuation pattern is similar in both SBRs. The standard deviation of the pH reading during runs 2 to 4 was approximately 0.1 in both SBRs. This small fluctuation in pH was likely the result of fluctuations in the composition of the influent waste stream. This would explain why a similar pattern was observed in both SBRs. 88 6.85 -i Figure 59: Average Weekly pH The experimental SBR typically had a slightly higher pH than the control SBR. Methanol addition caused an improvement in the denitrification rates of the experimental SBR. Denitrification reactions result in the replenishment of the alkalinity in the SBR, which in turn should increase the pH. However, the average standard deviation between the pH in the control and experimental SBRs was 0.06. This is a very insignificant difference in pH for two separate full-scale SBRs. Furthermore, the observed slight difference in pH between the control and experimental SBRs did not increase with increasing denitrification rates. The dissolved oxygen concentration pattern during each 4-hour SBR cycle was relatively consistent. Figure 60 shows the 30-period moving average trend for the DO in each SBR during the SBRs 4-hour cycle during a typical week. The DO concentration increased rapidly to approximately 2 mg/L at the start of the aeration cycle. It remained at this concentration throughout the aeration phase. At the end of the aeration phase, the residual DO was quickly scavenged and depleted to approximately 0 ppm. This concentration was maintained until the start of the following aeration cycle. The pattern described above and shown in Figure 60 was consistently repeated in the control and experimental SBRs throughout each four-hour cycle, during Runs 2 to 4. The average daily DO trend is shown in Figure 61, using a 30-period moving average for a typical week. A negligible difference between the DO in each of the SBRs was observed. 89 0:00 1:00 2:00 3:00 4:00 Cycle Time Figure 60: Average DO Trend During SBR Cycle (week of May 18 to 25) 3.5 . 3 Expt Ctrl 0:00 4:00 8:00 12:00 16:00 20:00 0:00 l i m e Figure 61: Daily Average DO Trend (week of Mar. 16 to 23) 90 The observed DO trend is ideal for nitrification, carbon oxidation, and denitrification. A DO concentration of 2 ppm is recommended for simultaneous nitrification and carbon oxidation [6], which is achieved during the aeration cycle. It is possible that methanol introduced at the end of the aeration cycle is not entirely consumed in the anoxic cycle during periods of low influent nitrogen loading and/or high methanol dosage. The residual methanol would contribute to an increased oxygen demand. However, a significant difference was not observed in the DO monitoring results for the control and experimental SBRs. The 4-hour cycle and daily 30-period moving average ORP trend for each week of experimental Runs 2 to 4 is shown in Appendix L and N respectively. The average ORP profile for a SBR cycle in the control SBR did not change significantly during Runs 2 to 4. However, a drastic change was observed in the experimental SBR. The results shown in Appendix M were used to determine the average weekly total change in ORP during a 4-hour cycle for each week. Subsequently, this information was used to determine the average change in ORP for each experimental run (refer to Appendix M). Figure 62 shows the relationship between the average total change in ORP and the methanol dosage. The relationship depicted in Figure 62 is very similar to the relationship between the denitrification rate and the methanol dosage. Figure 63 shows that the average denitrification rate is linearly related to the average total change in ORP. Lie et al. [35] and Fukagawa et al. [76] also concluded from their research that the denitrification rate is linearly related to the change in ORP. 300 i §2 o < s 50 0 0 10 20 30 40 50 60 70 80 90 Methanol Dosage (L/day) Figure 62: Average Change in ORP per Cycle vs. Methanol Dosage 91 300 250 -J 0 1 , , , , , , , , , , 0 2 4 6 8 10 12 14 16 18 20 Mean Zero-Order Denitrification Reaction Rate Constant (mg NO, -N/g MLVSS/day) Figure 63: Average Change in OPvP per Cycle vs. Denitrification Rate A nitrate breakpoint was rarely observed during Runs 2 to 4 in all of the monitored results for the experimental SBR, with the exception of a few cycles during the week of May 26; in this case, the nitrate breakpoint occurred during the cycle that commenced at 18:00. The influent total nitrogen concentration was very low during this week (refer to Figure 56 for the influent total nitrogen concentration determined from grab samples taken at mid-day). Furthermore, it is expected that there would be a high influent flow rate and, hence, an even lower influent ammonia concentration at 18:00 [1]. As a result, it is likely there was a low nitrate concentration following nitrification and the fast denitrification rate expected, from a methanol dosage of 81 L/day, achieved complete denitrification and the observed nitrate breakpoints. 4.1.7 Seasonal Variations Temperature is a critical parameter in the performance of a SBR. Section 4.1.1.1 suggests that the nitrification rate increases with an increase in temperature, and it decreases with decreases in temperature. It was also suggested that the nitrifiers are extremely sensitive to decreases in temperature. The effect of temperature on the denitrification rate was not conclusively determined from this research (refer to Section 4.1.1.2). However, literature indicates that denitrifying organisms are also very sensitive to temperature changes. The seasonal change in temperature is shown in 92 Figure 64 for the period from Jan. 1997 to June 1999. The SBR temperature appears to be typically at a maximum around August, and at a minimum in January. -Ctrl — Expt Tim Figure 64: Seasonal SBR Temperature Variations Figure 65 is a plot of the pH in each SBR from Jan. 1997 to June 1999. A seasonal variation in the pH is not apparent. However, the pH appears to be increasing slowly each year. This increase is likely the result of changes in the composition of the influent wastewater. The average monthly flow rate is plotted as a function of time in Figure 66 for Jan. 1997 to June 1999. The flow rates were significantly higher in 1997 than subsequent years. The highest flow rates are typically observed in Jan. and June. The lowest flow rates are typically observed in April and September. Figures 67 and 68 show the seasonal variation of influent NH 3 and P0 4 concentration respectively from January to June of 1999. The influent flow rate is also plotted on these figures for this time period. These figures indicate that the influent flow rate is inversely proportional to the influent NH 3 and P0 4 concentration. Figure 69 and 70 shows the relationship between the influent flow rate and the NH 3 and P0 4 concentrations, respectively. The average monthly values for flow rate, NH 3 and P0 4 concentration from the data collected by the plant operators Jan. 1997 to June 1999 were used in Figures 69 and 70. Therefore, from this relationship and Figure 66, which shows the yearly influent rates, we can predict the expected influent NH 3 and P0 4 concentration. Similarly, the influent BOD5 and COD concentrations are also inversely proportional to the influent flow rate. The relationship 93 between the influent flow rate and the influent BOD5 and COD concentration is shown in Figure 71. The plant's influent flow rate will vary seasonally. These seasonal variations are a result of seasonal variations in precipitation that will infiltrate into wastewater sewer systems and dilute the domestic wastewater. Therefore, at higher flow rates, lower concentrations will be observed, as shown in Figures 69 to 71 for NH 3, P04, BOD5 and COD. T i m e O X -Ctrl - Expt Figure 65: Seasonal pH Variation Jan Feb Mar Apr May Jun Jul Month Aug Sep Oct Nov Figure 66: Seasonal Influent Flow Rate Variation 94 0 0 CN V") ON NO o\ m NO o NO j£ ' 1 " H C N C N C N O to Jan. <u PH o Feb. oMa Mar. Mar. Apr. Apr. May May to Jan. +-» NO o o o o s j£ C N ts r^ o NO 0 0 3 Feb. C N •—»Feb. Feb. s Mar. i Apr. Ma; May T i m e Figure 67: Seasonal Variation in Influent NH3 Concentration C N ON NO m 0 NO C N n ' 1 C N , — * C N • — 1 C N Jan. toFel Feb. D Mai Mar. Mar. Apr. Apr. May May w - H 0 -»-» +J NO Q 0 0 +J 0 Q C N C N 0 NO 00 C N C N >. 9 Feb. Feb Mar. Mar. Api Apr. Ma; May en T O PM - X -- Flow -NH3 O — Flow - X P04 T i m e Figure 68: Seasonal Variation in Influent PO4 Concentration 95 25 -, Figure 70: Influent P 0 4 Concentration as a Function of Flow 96 800 -, 700 - \ x» x y =282764x-0-9614 600 - \ x R2 = 0.7522 O B O D 5 X C O D o 4 , , , , , , , , , 0 500 1000 1500 2000 2500 3000 3500 4000 4500 F l o w ( m 3 / d a y ) Figure 71: Influent BOD5 and COD as a Function of Flow 4.2 Batch Study of Methanol Induced Denitrification This section summarizes the results from three batch tests. In each case, multiple methanol dosages and a control were setup according to the procedure described in Section 3.1.2. Two of the batch tests were performed under completely mixed conditions and the third batch test was performed under unmixed conditions. This will indicate the significance of mixing to denitrification reaction kinetics. The two completely mixed batch tests were performed using a sludge acclimatized in the experimental SBR for a 5 week period to a methanol dosage of 12.2 mg CH3OH/L, and a sludge sample from the control SBR which was not acclimatized to methanol. Appendix N summarizes the data obtained from the batch studies. 97 4.2.1 Continuously-Mixed Reactor The denitrification rate at the different methanol dosages, using the acclimatized sludge, is plotted in Figure 72. This figure shows an increase in the denitrification rate with an increase in methanol dosage, according to the indicated relationship. The denitrification rates obtained in this batch test are significantly higher than the rates obtained under full-scale, unmixed conditions, with similar methanol concentrations. Furthermore, the control batch reactor, without methanol addition, achieved a very high denitrification rate. In fact, the denitrification rate obtained without any methanol and completely mixed conditions was approximately equal to the maximum achieved denitrification rate under unmixed, full-scale conditions with methanol addition. In addition, the denitrification rate did not plateau with increasing methanol dosage, as was observed during full-scale testing (refer to Figure 27). The denitrification rate is limited by the availability of enzymes. The absence of a plateau in Figure 72 is likely due to the mixing process, which would more effectively distribute the enzymes required for metabolic processes throughout the batch container and make them more available to the denitrifying organisms. Figure 72: Completely Mixed Batch Test with Acclimatized Sludge 98 Figure 73 shows the relationship between the denitrification rate and the methanol dosage under completely mixed conditions with a non-acclimatized sludge. The denitrification rates are significantly less with a non-acclimatized sludge. This suggests that a proper acclimatization period is required to attain the most efficient denitrification rates. This has been concluded in several other research studies that utilized methanol as an external carbon for denitrification [51, 53]. Furthermore, the control batch reactor had a negligible denitrification rate. Compare this result to the denitrification rate of 19 mg NO x /g MLVSS/day observed in the control reactor of the experiment, with an acclimatized sludge (refer to Figure 71). This further emphasizes the importance of acclimatization to the denitrification process. A likely explanation for this observation is that the control SBR sludge has a very small fraction of denitrifiers. Whereas, the experimental SBR sludge, after methanol introduction, gradually develops a large fraction of denitrifiers. Therefore, even without methanol addition, the denitrifiers that are present want to denitrify and they scavenge any available carbon in order to do so. Loosdrecht et al. [41] suggests that microorganisms generally respond to feast-famines regimes common to SBR systems by accumulating storage polymers and subsequently utilizing them for growth when external substrate is depleted. Microorganisms that have the capacity for substrate storage have a strong competitive advantage over other organisms in feast-famine regimes. Hence, the utilization of stored carbon may have resulted in the fast denitrification rates observed in the control reactor without methanol addition. 22 -, 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Methanol Concentration (mg C H , O H / L ) Figure 73: Completely Mixed Batch Test with Non-Acclimatized Sludge 99 It is inconclusive how much higher the denitrification rates shown in Figure 73 may be increased. However, it is evident that acclimatization plays a critical role in the denitrification process. The smallest observed increase in the denitrification rate between the results presented in Figures 72 and 73, as a result of acclimatization to methanol, was 55%. It is likely that much higher denitrification rates are possible after a sufficient acclimatization period to mixed conditions and higher methanol dosages. 4.2.2 Settled Reactor This batch test was designed to simulate the full-scale conditions during the anoxic period. A sludge sample from the experimental SBR, which had been acclimatized for a five-week period to a methanol concentration of 12.2 mg CH 3 OH/L , was utilized in this batch test. The relationship between the denitrification rate and the methanol concentration is shown in Figure 74. Once again, the control batch test, without methanol addition and an acclimatized sludge, showed a higher denitrification rate than did the control full-scale SBR, which was not acclimatized for denitrification. Also, the denitrification rates attained with an acclimatized completely mixed reactor (Figure 72) are 11 -1 0 -I , , , , , , , , , , • , r 0 1 2 3 - 4 5 6 7 8 9 10 11 12 13 Methanol Concentration (mg CH 3OH/L) Figure 74: Settled Batch Test with Acclimatized Sludge 100 significantly greater than those obtained without mixing and an acclimatized sludge. Mixing increases the contact between the microorganisms and the available substrate, and thus increases the denitrification rate. Comparison of Figures 72 and 74, both with methanol acclimatized sludges, revealed an increase in denitrification rates ranging from 660%, without methanol addition, to 200%, with a methanol dosage of 12.7 mg/L. Therefore, mixing appears critical to the denitrification process, to realize the best kinetic performance. The denitrification rates that are observed in Figure 74 are lower than the rates observed in Figure 27 under full-scale conditions. Based on the derived full-scale and batch-scale relationships between the denitrification rate and the methanol dosage, and at methanol dosages ranging from 2 to 10 mg/L, the denitrification rates were on average 130% higher under full-scale conditions than batch-scale conditions. The full-scale SBR was mixed for a 15 minute period at the onset of the anoxic phase, whereas the batch test was only mixed for 1 minute. Mixing increases the reaction rates significantly and is the likely cause of the difference in the denitrification rates observed between the full-scale experiment and the batch test. 101 5.0 C O N C L U S I O N S A N D R E C O M M E N D A T I O N S 5.1 Summary and Conclusions The primary goal of this research was to determine the potential of the existing treatment strategy of a full-scale SBR system, for denitrification, with and without the use of methanol as an external carbon source. The following conclusions were made, based on the results from three batch experiments and a full-scale study of two SBRs. 5.1.1 Biological Nitrogen Removal in the Full-Scale SBRs The nitrification and denitrification reactions followed zero-order reaction kinetics. The control SBR, without methanol addition, achieved negligible denitrification rates. Two denitrification rates were observed in the experimental SBR, with methanol addition; an initial fast rate and a slower second rate. Methanol was utilized as the carbon source for denitrification during the first rate period. Following the depletion of the methanol, denitrification reactions continued by using the available natural carbon in the influent, resulting in a slower second denitrification rate. The denitrification rate (KD N, mg NOx/g MLVSS/day) in the experimental SBR increased with increasing methanol concentration (M, mg CH3OH/L) according to the following relationship: K D N =-0.203-M 2+3.93-M, until a maximum denitrification rate of approximately 19 mg NOx/g MLVSS/day was attained. Further increases in the methanol concentration beyond 8.1 mg CH3OH/L had a negligible influence on the denitrification rate. As a result, the nitrogen removal efficiency also increased with increasing methanol dosage, until a maximum denitrification rate was achieved. The nitrifying organisms in both the control SBR, without methanol addition, and the experimental SBR, with methanol addition, were very sensitive to temperature fluctuations. The nitrification rates increased slowly with increases in temperature. Drastic decreases in the nitrification rates were observed with small decreases in temperature. After a period of stability at the lower temperature, the nitrification rates increases gradually. 102 5.1.2 Biological Phosphorus Removal in the Full-Scale SBRs Biological phosphate uptake and release was only observed to a significant extent in the experimental SBR with methanol addition. Phosphate release commenced once a very low nitrate concentration was achieved. The phosphate release rate was approximately 2.7 mg P/g MLVSS/day and was caused by a supply of natural short chain carbon in the influent wastewater. Methanol was not utilized to a significant extent as the carbon source for the enhanced biological phosphorus removal process. However, methanol addition is critical to the E B P R process, since it depletes the available nitrates, and thus allows E B P R to take place. As a result, the phosphorus removal efficiency increased with increasing methanol dosage. 5.1.3 General Performance of the Full-Scale SBRs The COD removal efficiency decreased with increasing methanol dosage, as a result of a concomitant decrease in total nitrogen concentration in the influent, and, hence, a decreasing demand for methanol to achieve complete denitrification. The observed solids production was 0.21 kg sludge per L of methanol, and the theoretical sludge production is 0.17 kg sludge per L of methanol. Therefore, at a methanol dosage of 54 L/day the solids production will be theoretically increased by fifteen percent. The sludge production, based on a kinetic model and a derived net yield coefficient of 0.37 mg COD/mg M L V S S , is 0.44 kg sludge per L of methanol. However, the net yield coefficient applies to a generic carbon source, and thus the kinetic model is not specific to methanol. The settling properties of the experimental SBR deteriorated with increasing methanol dosage, as a result of increased denitrification rates and, subsequently, an increased nitrogen gas production that gives the sludge mass buoyancy. The settleability, measured by the sludge volume index (SVI, mL/g M L V S S ) is linearly related to the denitrification rate by the following expression: SVI = 1.73 KDN. Furthermore, the solids removal efficiency also decreased with increasing denitrification rate as a result of the deteriorating settleability of the sludge. 103 A negligible difference in the pH of the control and experimental SBRs was observed and both reactors had a similar pH fluctuation pattern during the methanol addition experiment. Therefore, methanol had a negligible influence on the pH in the SBRs. The dissolved oxygen concentration increased rapidly to 2 mg/L during the aeration phase. It was diminished to approximately 0 mg/L DO shortly after the end of the aeration phase or the beginning of the settling phase, and remained constant until the subsequent aeration phase. The average total change in the ORP, during a four-hour SBR cycle, was determined from continuous ORP monitoring of each SBR. The following linear relationship between the total change in ORP magnitude (AORP, mV) and the denitrification rate was discovered: AORP = 8 K D N + 91. 5.1.4 Batch Experiments A minimum 55 percent increase in the denitrification rate was observed in a batch reactor with sludge acclimatized to methanol addition, compared to a batch reactor with a non-acclimatized sludge. The non-acclimatized batch reactor had a negligible denitrification rate without methanol addition. However, significant denitrification rates were observed in the acclimatized batch reactors, potentially caused by microbial storage or an increased population of denitrifiers that scavenge any available carbon. A completely mixed batch reactor, with sludge acclimatized to methanol addition during the anoxic cycle, had an increase in denitrification rates ranging from 660%, without methanol addition, to 200%, with a methanol dosage of 12.7 mg/L, compared to the unmixed batch reactor with an acclimatized sludge. Therefore, mixing appears critical to the denitrification process, to realize the best kinetic performance. The denitrification rates were on average 130% higher under full-scale conditions than batch-scale conditions at methanol dosages ranging from 2 to 10 mg/L, based on the derived full-scale and batch-scale relationships between the denitrification rate and the methanol dosage. The longer mixing period during the onset of anoxic conditions, in the full-scale experiment, the likely cause of 104 the difference in the denitrification rates observed between the full-scale experiment and the batch test. 5.2 Recommendations for Continued Research Further research work is recommended in the following areas: • Pilot-scale or full-scale experiments should be conducted to validate the results from the batch experiments from this study related to the affects of acclimatization and mixing to , denitrification rates; • A pilot-scale or full-scale study is required to determine the optimal P L C strategy to achieve efficient phosphorus and nitrogen removal performance. This study should also focus on determining the optimal mixing sequence during the methanol addition phase, to achieve the best kinetic performance; • Full-scale dye testing of a continuous influent feed sequencing batch reactor is required to determine the potential of this system to short-circuit during the decant phase; • Batch experiments should be conducted to determine the influence of temperature to the denitrifiers in the SBR, with methanol as a carbon source; • Controlled batch-scale experiments are required to validate the results related to the influence of temperature to nitrifiers, determined from a variable full-scale system during this study; • Batch experiments should be performed to verify the relationship between the change in ORP magnitude and the denitrification rate; • Diurnal sampling of the influent and effluent from both SBRs is required to verify the phosphorus and nitrogen removal performance of a SBR with methanol addition; and, • An analysis of the denitrification behaviour at different depths in the tank, during the anoxic cycle or settling phase, is required. 105 6.0 R E F E R E N C E S 1. Metcalf & Eddy, Inc., Wastewater Engineering: Treatment, Disposal, Reuse, Third Edition, M c G r a w - H i l l , Inc., 1991. 2. Canadian Water Quality Guidelines: Guidelines for Water Quality in Canada, Ministry of Supply and Services, Canada, 1995. 3. Oldham, W.K, and D.S. Mavinic, A personal communication, 1999. 4. Sweden's Nitrogen Debate, Water Quality International, September/October, 1998. 5. Zhao, H.W., Ph.D. Thesis: Oxidation-Reduction Potential and Organic Carbon Sources as Two Control Parameters for Simultaneous Nitrification and Denitrification in Biological Nutrient Removal Processes, The University of British Columbia, Canada, 1998. 6. Arora, M . L . , E.F. Barth, and M . B . Umphres, Technology Evaluation of Sequencing Batch Reactors, J. Water Pollution Control Federation, 57 (8), 867-875, 1985. 7. Yu, R.F., S.L. Liaw, C.N. Chang, H.J. Lu, and W.Y. Cheng, Monitoring and Control Using On-line ORP on the Continuous-Flow Activated Sludge Batch Reactor System, Water Science and Technology, 35 (1), 57-66, 1997. 8. Eckenfelder, W.W., and Y . Argaman, Kinetics of Nitrogen Removal for Municipal and Industrial Applications, In: Advances in Water and Wastewater Treatment - Biological Nutrient Removal (Edited by M.P. Wanielista and W W . Eckenfelder), Ann Arbor Sci. Publ. Inc., Ann Arbor, 23-41, 1979. 9. Randall, A . A . , L .D. Benefield, and W.E. Hi l l , Induction of Phosphorus Removal in an Enhanced Biological Phosphorus Removal Bacterial Population, Water Resources, 31 (11), 2869-2877, 1997. 10. Jones, P.H., A .D. Tadwacker, and C.L. Hsu, Enhanced Uptake of Phosphorus by Activated Sludge-Effect of Substrate Addition, Water Resources, 21, 301-308, 1987. 11. Ardern, E., and W.T. Lockett, Experiments on the Oxidation of Sewage Without the Aid of Filters, J. Soc. Chem. Ind., 33, 523-534, 1914. 12. Irvine, R.L. , and W.B. Davis, Use of Sequencing Batch Reactors for Waste Treatment - CPC International, Corpus Christi, Texas, Proc. Of the 26 t h Ann. Purdue Indust. Waste Conf, 26, pp. 450-462, 1971. 13. Rusten, B. , and H . Eliassen, Sequencing Batch Reactors for Nutrient Removal at Small Wastewater Treatment Plants, Water Science and Technology, 28 (10), 233-242, 1993. 106 14. Cybis, L.F. De A. , and N.J . He-ran, Protozoan and Metazoan Populations in Sequencing Batch Reactors Operated for Nitrification and/or Denitrification, Water Science and Technology, 35 (1), 81-86, 1997. 15. Irvine, R.L. , and L . H . Ketchum, Sequencing Batch Reactors for Biological Wastewater Treatment, CRC Critical Reviews in Environmental Control, 18 (4), 255-294, 1989. 16. Ketchum, L . H . , Design and Physical Features of Sequencing Batch Reactors, Water Science and Technology, 35 (1), 11-18, 1997. 17. Chiesa, S.C., and R.L. Irvine, Growth and Control of Filamentous Microbes in Activated Sludge: An Integrated Hypothesis, Water Resources, 19, pp.471, 1985 18. Irvine, R.L. , Structured Models for Biological Waste Treatment Systems, Research Grant, National Science Foundation, 1985. 19. Irvine, R.L. , P.A. Wilderer, and H.C. Flemming, Controlled Unsteady State Processes and Technologies - An Overview, Water Science and Technology, 35 (1), 1-10, 1997. 20. Irvine, R.L. , D.V.S. Murthy, M . L . Arora, J.L. Copeman, and J.A. Heidman, Analysis of a Full-scale SBR Operation at Grundy Center, Iowa, J. Water Pollution Control Federation, 59 (3), 132-138, 1987. 21. U.S. Environmental Protection Agency, Summary Report: Sequencing Batch Reactors, Report No. EPA-625-8-86-011, Office of Technology Transfer, Cincinnati, Ohio, 1986. 22. Brenner, A. , Use of Computers for Process Design Analysis and Control: Sequencing Batch Reactor Application, 35, (1), 95-104, 1997. 23. Randall, C.W., J.L. Barnard, and H.D. Stensel, Design and Retrofit of Wastewater Treatment Plants for Biological Nutrient Removal, Water Quality Management Library, Technomic Publishing Company, Inc., 5, 1992. 24. Barnard, J.L., and B. Rabinowitz, CIVL 559 Course Notes - Biological Phosphorus Removal, University of British Columbia, 1998. 25. Madigan, M.T., J .M. Martinko, and J. Parker, Brock Biology of Microorganisms, Eight Edition, Prentice-Hall, Inc., 1991. 26. Sawyer, C , P. McCarty, and G. Parkin, Chemistry for Environmental Engineering, McGraw-Hill, Inc., 1994. 27. Alleman, J.E, and R.L. Irvine, Storage-Induced Denitrification Using Sequencing Batch Reactor Operation, Water Research, 14, 1483-1489, 1980. 28. Leung, G.L.W., and N.F .Y . Tarn, Operation Strategy of a Sequencing Batch Reactor for Simultaneous Removal of Wastewater Organic Matter and Nutrients, Resources, Conservation and Recycling, 11, 209-223, 1994. 107 29. Manoharan, R., S. Liptak, P. Parkinson, D. Mavinic, and C.W. Randall, Denitrification of a High Ammonia Leachate Using an External Carbon Source, Environmental Technology Letters, 10, 701-716, 1989. 30. Winter, C.T., The role of Acetate in Denitrification and Biological Phosphate Removal in Modified Bardenpho Systems, Water Science and Technology, 21 (4-5), 375-385, 1989. 31. Carley, B .N . , and D.S. Mavinic, The effects of External Carbon Loading on Nitrification and Denitrification of a High-ammonia Landfill Leachate, Res. J. Water Pollution Control Federation, 63 (1), 51-59, 1991. 32. Tarn, N.F .Y. , Y.S . Wong, and G. Leung, Significance of External Carbon Sources on Simultaneous Removal of Nutrients from Wastewater, Water Science and Technology, 26 (5-6), 1047-1055, 1992. 33. Von Schultes R., D. Wild, and W. Gujer, Nitric and Nitrous Oxides from Denitrifying Activated Sludge at Low Oxygen Concentrations, Water Science Technology, 30, 123-132, 1994. 34. Czepiel, P., P. Crill, and R. Hariss, Nitrous Oxide Emissions from Municipal Wastewater Treatment, Environmental Science and Technology, 29, 2352-2356, 1995. 35. Lie, E. and T. Welander, Influence of Dissolved Oxygen and Oxidation-Reduction Potential on the Denitrification Rate of Activated Sludge, Water Science and Technology, 30 (6), 91-100, 1994. 36. Robertson, L .A . , and J.G. Kuene, Combined Heterotrophic Nitrification and Aerobic Denitrification in Thiosphaera pantotropha and Other Bacteria, Antonie van Leeuwenhoek, 57, 139-152, 1990. 37. Muller, E.B. , A . H . Stouthamer, and H.W. Verseveld, Simultaneous NH3 Oxidation and N2 Production at Reduced 02 Concentrations by Sewage Sludge Subculture with Chemolithotrophic Medium, Biodegradation, 6, 339-349, 1995. 38. Mulder, A. , A . A . Van der Graff, L .A . Robertson, and J.G. Kuenen, Anaerobic Ammonium Oxidation Discovered in a Fluidized Bed Reactor, FEMS Microbiol. Ecol., 16, 177-184, 1995. 39. Bock, E., R. Stueven, and D. Zart, Nitrogen Loss Caused by Denitrifying Nitrosomonas Cells Using Ammonium or Hydrogen as Electron Donors and Nitrite as Electron Acceptor, Arch. Microbiol., 163, 16-20, 1995. 40. Schmidt, I., and E. Bock, Anaerobic Ammonia Oxidation with Nitrogen Dioxide by Nitrosomonas eutropha, Archives of Microbiology, 167, 106-111, 1997. 41. Van Loosdrecht, M . , and M . Jetten, Microbiological Conversions in Nitrogen Removal, Water Science and Technology, 38 (1), 1-7, 1998. 108 42. Demoulin, G., M . C , Goronszy, K . Wutscher, and E. Forsthuber, Co-current Nitrification/Denitrification and Biological P-Removal in Cyclic Activated Sludge Plants by Redox Controlled Cycle Operation, Water Science and Technology, 35 (1), 215-224, 1997. 43. Comeau, Y . , K.J . Hall, R.E.W. Hancock, and W.K. Oldham, Biochemical Model for Enhanced Biological Phosphorus Removal, Water Research, 20 (12), 1511-1521, 1986. 44. Wentzel, M . C , L . H . Lotter, R.E. Loewenthal, and G.v.R. Marais, Metabolic Behaviour of Acinetobacter ssp. In Enhanced Biological Phosphorus Removal - A Biochemical Model, Water S.A., 12 (4), 209-224, 1986. 45. Louie, T .M. , T.J. Mah, W.K. Oldham, and W.D. Ramey, Use of Metabolic Inhibitors and Gas Chromatography/Mass Spectrometry to Study Poly-B-hydroxyalkanoates Metabolism Involving Cryptic Nutrients in Enhanced Biological Phosphorus Removal Systems, in progress. 46. Mino, T., V. Arun, Y . Tsuzuki, and T. Matsuo, Effect of Phosphorus Accumulation on Acetate Metabolism in the Biological Phosphorus Removal Process, In: Advances in Water Pollution Control no. 4: Biological Phosphate Removal from Wastewaters (Edited by R. Ramadori), 27-38, Pergamon Press, Oxford, 1987. 47. Satoh, H . , T. Mino, and T. Matsuo, Uptake of Organic Substrates and Accumulation of Polyhydroxyalkanoates Linked with the Glycolysis of Intracellular Carbohydrates Under Anaerobic Conditions in the Biological Excess Phosphate Removal Processes, Water Science and Technology, 26 (5-6), 933-942, 1992. 48. Vlekke, G.J.F.M., Y . Comeau, and W.K. Oldham, Biological Phosphate Removal from Wastewater with Oxygen and Nitrate in Sequencing Batch Reactors, Environmental Technology Letters, 9, 791-796, 1988. 49. Arun, V . , T. Mino, and T. Matsuo, Biological Mechanism of Acetate Uptake by Carbohydrate Consumption in Excess Phosphorus Removal Systems, Water Research, 22, 565-570, 1988. 50. Tam, N.F .Y. , G.L.W. Leung, and Y.S . Wong, The Effects of External Carbon Loading on Nitrogen Removal in Sequencing Batch Reactors, Water Science and Technology, 30 (6), 73-81, 1994. 51. Hallin, S., M . Rothman, and M . Pell, Adaptation of Denitrifying Bacteria to Acetate and Methanol in Activated Sludge, Water Resources, 30(6), 1445-1450, 1996. 52. Curucci, A. , K . Lindrea, M . Majone, and R. Ramadori, Different Mechanisms for the Anaerobic Storage of Organic Substrates and their Effect on Enhanced Biological Phosphate Removal (EBPR), Water Science and Technology, 39 (6), 21-28, 1999. 109 53. Nyberg, U . , H . Aspergen, B. Andersson, J. Jansen, and I.S. Villadsen, Full-Scale Application of Nitrogen Removal with Methanol as Carbon Source, Water Science and Technology, 26 (5-6), 1077-1086, 1992. 54. Regan, J., B. Koopman, S.A. Svoronos, and B. Lee, Full-Scale Test of Methanol Addition for the Enhanced Nitrogen Removal in a Ludzack-Ettinger Process, Water Environment Research, 70 (3), 376-381, 1998. 55. Yearbook of Science and the Future 2000, Encyclopedia Britannica, Inc., 1999. 56. Keller, J., K. Subramaniam, J. Gosswein, and P.F. Greenfield, Nutrient Removal from Industrial Wastewater Using Single Tank Sequencing Batch Reactors, Water Science and Technology, 35 (6), 137-144, 1997. 57. Hamamoto, Y . , S. Tabata, and Y . Okubo, Development of the Intermittent Cyclic Process for Simultaneous Nitrogen and Phosphorus Removal, Water Science and Technology, 35 (6), 145-152, 1997. 58. Danesh, S., and J.A. Oleszkiewicz, Use of a New Anaerobic-Aerobic Sequencing Batch Reactor System to Enhance Biological Phosphorus Removal, Water Science and Technology, 35 (6), 137-144, 1997. 59. Marklund, S., Cold Climate Sequencing Batch Reactor Biological Phosphorus Removal - Results 1991-92, Water Science and Technology, 35 (6), 275-282, 1993. 60. Carucci, A. , M . Majone, R. Ramadori, and S. Rossetti, Dynamics of Phosphorus and Organic Substrates in Anaerobic and Aerobic Phases of a Sequencing Batch Reactor, Water Science and Technology, 30 (6), 237-246, 1994. 61. Gerber, A. , R.H. Villiers, E.S. Mostert, and C.J.J, van Riet, The Phenomenon of Simultaneous Phosphateuptake and Release, and its Importance in Biological Nutrient Removal, Adv. Water Pollution, Proc. IAWPRC Conference, Rome, Italy, 123-134, 1987. 62. Yu , R.F., S.L. Liaw, C.N. Chang, H.J. Lu, and W Y . Cheng, Monitoring and Control Using On-line ORP on the Continuous Flow Activated Sludge Batch Reactor System, Water Science and Technology, 35 (1), 57-66, 1997. 63. Charpentier, J., H . Godart, G. Martin, and Y . Mogno, Oxidation-Reduction Potential (ORP) Regulation as a Way to Optimize Aeration and C, N and P Removal: Experimental Basis and Various Full-scale Examples, Water Science and Technology, 21, 1209-1223, 1989. 64. Koch, F.A., and W.K. Oldham, Oxidation Reduction Potential - A Tool for Monitoring, Control and Optimization of Biological Nutrient Removal Systems, Water Science and Technology, 17, 259-281, 1985. 110 65. Peddie, C . C , D.S. Mavinic, and J.J. Jenkins, Use of ORP for Monitoring and Control of Aerobic Sludge Digestion, J. of Environmental Engineering, A S C E , 116 (3), 461-471, 1990. 66. Wareham, D.G. , K.J . Hall, and D.S. Mavinic, An ORP Screening Protocol for Biological Phosphorus Removal in Sequencing Batch Reactors, Canadian J. of Civi l Engineering, 22, 260-269,1995. 67. Plisson-Saune, S., B. Capdeville, M . Mauret, A . Deguin, and P. Baptiste, Real-Time Control of Nitrogen Removal Using Three ORP Bending-Points: Signification, Control Strategy and Results, Water Science and Technology, 33 (1), 275-280, 1996. 68. Ketchum, L . H . , R.L. Irvine, R.E. Breyfogle, and J.F. Manning, A Comparison of Biological and Chemical Phosphorus Removals in Continuous and Sequencing Batch Reactors, J. Water Pollution Control Federation, 59, 13-18, 1987. 69. APHA, StandardMethods for the Examination ojWater andWastewater, 19th Edition, 1996. 70. Pamphlet: Collection and Preservation of Environmental Samples, Analytical Service Laboratories Ltd., British Columbia, Canada. 71. Methods Manual for the QuikChem® Automated Ion Analyzer, Lachat Instruments, Inc. 72. Pamphlet: Instructions - ORP (Redox) Combination Electrode: Testing Procedures, Broadley James Corp. 73. Ramamoorthy, S., Quality Assurance and Quality Control Related to Environmental Samples, No. 141, Northern River Basins Study, Edmonton, Alberta, 1997. 74. Carucci, A. , R. Ramadori, S. Rossetti, and M . C . Tomei, Kinetics of Denitrification in Single Sludge Systems, Water Research, 30 (1), 51-56, 1996. 75. Ekama, G.A., and G. Marais, Biological Nitrogen Removal, In: Theory Design and Operation of Nutrient Removal Activated Sludge Processes, pp. 6.1-6.26, South African Water Research Commission, Pretoria, 1984. 76. Fukagawa, M . , T. Harada, M . Takeuchi, H . Kawata, S. Murakami, and H . Nakanishi, A Study of Biological Treatment with Control of Redoxpotential, Ube Kogyo Koto Senmon Gakko Kenkyo Hokoki, 34, 51-57, 1988. I l l APPENDIX A : M S D S FOR METHANOL 1 - PRODUCT IDENTIFICATION PRODUCT NAME: FORMULA: FORMULA WT: CAS N O . : NIOSH/RTECS NO. COMMON SYNONYMS WOOD ALCOHOL; CARBINOL; METHYLOL; WOOD METHANOL CH30H 32.04 67-56-1 PC1400000 METHYL ALCOHOL; SPIRIT PRODUCT CODES: 9049, 9072, 9075, 907 6, 9071, 5217, 5370, 9074, P704, 9093, 5536, 9068 9073 ,9091 ,92 63 ,9069 ,9070 E F F E C T I V E : 0 9 / 2 6 / 8 6 REVISION #04 PRECAUTIONARY L A B E L L I N G BAKER S A F - T - D A T A ( T M ) SYSTEM HEALTH - 3 FLAMMABILITY - 3 REACTIVITY - 1 CONTACT - 1 HAZARD RATINGS ARE 0 TO 4 (0 = NO HAZARD; SEVERE (POISON) SEVERE (FLAMMABLE) SLIGHT SLIGHT 4 = EXTREME HAZARD) . LABORATORY PROTECTIVE EQUIPMENT GOGGLES & SHIELD; LAB COAT & APRON; VENT HOOD; PROPER GLOVES; CLASS B EXTINGUISHER PRECAUTIONARY LABEL STATEMENTS POISON DANGER FLAMMABLE HARMFUL I F INHALED CANNOT BE MADE NON-POISONOUS MAY BE FATAL OR CAUSE BLINDNESS I F SWALLOWED KEEP AWAY FROM HEAT, SPARKS, FLAME. DO NOT GET IN E Y E S , ON S K I N , ON CLOTHING. AVOID BREATHING VAPOR. KEEP IN TIGHTLY CLOSED CONTAINER. USE WITH ADEQUATE V E N T I L A T I O N . WASH THOROUGHLY AFTER HANDLING. IN CASE OF F I R E , USE ALCOHOL FOAM, DRY CHEMICAL, CARBON DIOXIDE - WATER MAY BE I N E F F E C T I V E . FLUSH S P I L L AREA WITH WATER SPRAY. S A F - T - D A T A ( T M ) STORAGE COLOR CODE: RED (FLAMMABLE) 112 2 - HAZARDOUS COMPONENTS COMPONENT METHANOL 90-100 CAS NO. 67 -56-1 3 - PHYSICAL DATA BOILING POINT: MELTING POINT: S P E C I F I C GRAVITY: (H20=l) S O L U B I L I T Y ( H 2 0 ) : 100 65 C ( 149 F) -98 C ( -144 F) 0 .79 VAPOR PRESSURE(MM H G ) : 96 VAPOR D E N S I T Y ( A I R = 1 ) : . 1 . 1 1 EVAPORATION R A T E : (BUTYL ACETATE=1) 4 . 6 COMPLETE (IN A L L PROPORTIONS) % V O L A T I L E S BY VOLUME: APPEARANCE & ODOR: CLEAR, COLORLESS LIQUID WITH CHARACTERISTIC PUNGENT ODOR. FIRE AND EXPLOSION HAZARD DATA FLASH POINT (CLOSED CUP 12 C ( 54 F) FLAMMABLE L I M I T S : UPPER - 36 .0 % LOWER NFPA 704M RATING: 1-3-0 6.0 % FIRE EXTINGUISHING MEDIA USE ALCOHOL FOAM, DRY CHEMICAL OR CARBON DIOXIDE. (WATER MAY BE I N E F F E C T I V E . ) SPECIAL F I R E - F I G H T I N G PROCEDURES FIREFIGHTERS SHOULD WEAR PROPER PROTECTIVE EQUIPMENT AND SELF-CONTAINED BREATHING APPARATUS WITH FULL F A C E P I E C E OPERATED IN P O S I T I V E PRESSURE MODE. MOVE CONTAINERS FROM FIRE AREA I F IT CAN BE DONE WITHOUT R I S K . USE WATER TO KEEP FIRE-EXPOSED CONTAINERS COOL. UNUSUAL FIRE & EXPLOSION HAZARDS VAPORS MAY FLOW ALONG SURFACES TO DISTANT IGNITION SOURCES AND FLASH BACK. CLOSED CONTAINERS EXPOSED TO HEAT MAY EXPLODE. CONTACT WITH STRONG OXIDIZERS MAY CAUSE F I R E . BURNS WITH A CLEAR, ALMOST I N V I S I B L E FLAME. TOXIC GASES PRODUCED CARBON MONOXIDE, CARBON DIOXIDE, FORMALDEHYDE 113 5 - HEALTH HAZARD DATA T L V LISTED DENOTES ( T L V - S K I N ) . THRESHOLD LIMIT VALUE ( T L V / T W A ) : SHORT-TERM EXPOSURE LIMIT (STEL) 260 MG/M3 ( 200 PPM) 310 MG/M3 ( 250 PPM) PERMISSIBLE EXPOSURE LIMIT ( P E L ) : 260 MG/M3 ( 200 PPM) T O X I C I T Y : LD50 (ORAL-RAT)(MG/KG) LD50 ( IPR-RAT)(MG/KG) LD50 (SCU-MOUSE)(MG/KG) LD50 (SKN-RABBIT) (G/KG) CARCINOGENICITY: NTP: NO IARC: NO - 5628 - 9540 - 9800 - 20 Z L I S T : NO OSHA REG: NO EFFECTS OF OVEREXPOSURE INHALATION AND INGESTION ARE HARMFUL AND MAY BE F A T A L . INHALATION MAY CAUSE HEADACHE, NAUSEA, VOMITING, D I Z Z I N E S S , NARCOSIS, SUFFOCATION, LOWER BLOOD PRESSURE, CENTRAL NERVOUS SYSTEM DEPRESSION. LIQUID MAY BE IRRITATING TO SKIN AND E Y E S . PROLONGED SKIN CONTACT MAY RESULT IN DERMATITIS. EYE CONTACT MAY RESULT IN TEMPORARY CORNEAL DAMAGE. INGESTION MAY CAUSE BLINDNESS. INGESTION MAY CAUSE NAUSEA, VOMITING, HEADACHES, D I Z Z I N E S S , GASTROINTESTINAL IRRITATION . CHRONIC EFFECTS OF OVEREXPOSURE MAY INCLUDE KIDNEY AND/OR L I V E R DAMAGE. TARGET ORGANS E Y E S , SKIN, CENTRAL NERVOUS SYSTEM MEDICAL CONDITIONS GENERALLY AGGRAVATED BY EXPOSURE NONE IDENTIFIED ROUTES OF ENTRY INHALATION, INGESTION, EYE CONTACT, SKIN CONTACT EMERGENCY AND FIRST AID PROCEDURES CALL A PHYSICIAN. I F SWALLOWED, I F CONSCIOUS, IMMEDIATELY INDUCE VOMITING. I F INHALED, REMOVE TO FRESH A I R . I F NOT BREATHING, GIVE A R T I F I C I A L RESPIRATION. I F BREATHING IS D I F F I C U L T , GIVE OXYGEN. IN CASE OF CONTACT, IMMEDIATELY FLUSH EYES OR SKIN WITH PLENTY OF WATER FOR AT LEAST 15 MINUTES WHILE REMOVING CONTAMINATED CLOTHING AND SHOES. WASH CLOTHING BEFORE R E - U S E . 114 6 - REACTIVITY DATA S T A B I L I T Y : STABLE HAZARDOUS POLYMERIZATION: WILL NOT OCCUR CONDITIONS TO AVOID: HEAT, FLAME, OTHER SOURCES OF IGNITION INCOMPATIBLES: STRONG OXIDIZING AGENTS, STRONG A C I D S , ALUMINUM DECOMPOSITION PRODUCTS: CARBON MONOXIDE, CARBON DIOXIDE, FORMALDEHYDE 7 - S P I L L AND DISPOSAL PROCEDURES STEPS TO BE TAKEN IN THE EVENT OF A S P I L L OR DISCHARGE WEAR SELF-CONTAINED BREATHING APPARATUS AND FULL PROTECTIVE CLOTHING. SHUT OFF IGNITION SOURCES; NO F L A R E S , SMOKING OR FLAMES IN A R E A . STOP LEAK I F YOU CAN DO SO WITHOUT RISK. USE WATER SPRAY TO REDUCE VAPORS. TAKE UP WITH SAND OR OTHER NON-COMBUSTIBLE ABSORBENT MATERIAL AND PLACE INTO CONTAINER FOR LATER DISPOSAL. FLUSH AREA WITH WATER. J . T . BAKER SOLUSORB(R) SOLVENT ADSORBENT IS RECOMMENDED FOR SPILLS OF THIS PRODUCT. DISPOSAL PROCEDURE DISPOSE IN ACCORDANCE WITH A L L A P P L I C A B L E FEDERAL, S T A T E , AND LOCAL ENVIRONMENTAL REGULATIONS. EPA HAZARDOUS WASTE NUMBER: U154 (TOXIC WASTE) 8 - PROTECTIVE EQUIPMENT V E N T I L A T I O N : USE GENERAL OR LOCAL EXHAUST V E N T I L A T I O N TO MEET T L V REQUIREMENTS. RESPIRATORY PROTECTION: RESPIRATORY PROTECTION REQUIRED I F AIRBORNE CONCENTRATION EXCEEDS T L V . AT CONCENTRATIONS ABOVE 2 00 PPM, A SELF-CONTAINED BREATHING APPARATUS IS ADVISED. E Y E / S K I N PROTECTION: SAFETY GOGGLES AND FACE S H I E L D , UNIFORM, PROTECTIVE S U I T , RUBBER GLOVES ARE RECOMMENDED. 9 - STORAGE AND HANDLING PRECAUTIONS S A F - T - D A T A ( T M ) STORAGE COLOR CODE: RED (FLAMMABLE) SPECIAL PRECAUTIONS BOND AND GROUND CONTAINERS WHEN TRANSFERRING L I Q U I D . KEEP CONTAINER TIGHTLY CLOSED. S T O R E ' I N A COOL, DRY, W E L L - V E N T I L A T E D , FLAMMABLE LIQUID STORAGE AREA. 115 10 - TRANSPORTATION DATA AND ADDITIONAL INFORMATION DOMESTIC ( D . O . T . ) PROPER SHIPPING NAME HAZARD CLASS UN/NA LABELS REPORTABLE QUANTITY METHANOL FLAMMABLE LIQUID UN1230 FLAMMABLE LIQUID 5000 L B S . INTERNATIONAL ( I . M . O . ) PROPER SHIPPING NAME HAZARD CLASS UN/NA LABELS METHANOL 3 . 2 , 6 .1 UN1230 FLAMMABLE LIQUID, POISON 116 APPENDIX B : CALCULATION OF REQUIRED METHANOL Given: - NO x concentration measured at the end of the aerobic cycle (refer to Figure 8) is approx. 3.3 mg N/L - ratio of 1 mole N0 3 to 1 mole methanol is 1.1.08 (refer to Section 2.2.1) - length x width of SBR = 24 m x 8.5 m = 204 m 2 (does not include pre-mix chamber which is 3.7 m long) Find: Mass of methanol per 4 hr. cycle required to denitrify remaining NO x. Solution: - Assume the average liquid depth of the SBR is approx. 4.3 m .-. Liquid Volume 877.2 m j 3.3 mg/LN 15 mg/L N0 3 For 1 L volume: - 15 mgN0 3 .-. required methanol (1:1.08) 0.24 moles N 0 3 0.26 moles CH 3 OH 8.3 mg CH 3 OH 8.3 mg/L CH 3 OH is required to denitrify, or for one cycle it is required to add 7.3 kg CH 3 OH m mass 9.1 L C H 3 O H in volume .-. 9.1 L C H 3 O H are required to denitrify the remaining N O x 117 APPENDIX C : LABORATORY INSTRUMENTS Q A / Q C TESTS i is Prepared Standard - 18.7 mg/L N H 3 Analysis Mean = 18.1 mg/L N H 3 10 l-Dec-96 ll-Mar-97 19-Jun-97 27-Sep-97 5-Jan-9S Date 15-Apr-98 24-Jul-98 l-Nov-98 9-Feb-99 1 11 Prepared Standard = 12.6 mg/L NO, Analysis Mean^ 11 3 mg/L NO* x x l-Dec-96 ll-Mar-97 19-]un-97 27-Sep-97 5-Jan-98 15-Apr-98 24-Jul-98 l-Nov-98 9-Feb-99 Date 118 •a &• 20 Analysis Mean = 18.9 mg/L TKN Prepared Standard = I8.7mg/L TKN 15 l-Dec-96 ll-Mar-97 19-Jun-97 27-Sep-97 5-Jan-9E Date 15-Apr-98 24-Jul-98 l-Nov-98 9-Feb-99 •a g 16 X X Analysis Mean = 16.1 mg/L P0 4 X X X X X X X X X X X X X X X X X X X X X X X X X . Prepared Standard = 15.6 mg/L P0 4 X X X X X 14.5 l-Dec-96 ll-Mar-97 19-Jun-97 27-Sep-97 5-Jan-98 Date 15-Apr-98 24-Jul-98 l-Nov-98 9-Feb-99 119 g 16.5 Analysis Mean = 16.1 mg/L TP Prepared Standard = 15.6 mg/L TP l-Dec-96 ll-Mar-97 19-Jun-97 27-Sep-97 5-Jan-9S Date 15-Apr-98 24-Jul-98 l-Nov-98 9-Feb-99 120 APPENDIX D : INFLUENT AND EFFLUENT RAW DATA FOR EXPERIMENTAL RUNS < Reduction in COD Removal Efficiency % Conf. - C S ro Reduction in COD Removal Efficiency % St. Dev. - CS Tf Reduction in COD Removal Efficiency % Mean cn 0 0 Reduction in COD Removal Efficiency % < Tf C N NO V » Tt ^ V O ° ^ ) COD Removal Efficiency % Expt. Conf. C N VI 00 vt COD Removal Efficiency % Expt. St. Dev. - in o COD Removal Efficiency % Expt. Mean NO OO «n oo r-r- oo NO COD Removal Efficiency % Expt. Effluent OO 0 0 ro ON oo o OO t- OO ON ro NO m ro NO r- oo oo —i ON O O t- >f': t -COD Removal Efficiency % Ctrl. Effluent r- .-H r-r- C N co ON v> ro vi t- oo ON ON Tf C N Tf oe oo ON oo « N O  t-~- r - oc r COD mgN/L Expt. Effluent ~- oo C N T-I O Vt fO oo rs C N C N rs CO O ON NO v» Tt »—* co NO r-H NO COD mgN/L Ctrl. Effluent tn VN vi Tt -H .-H m oo ro vi C S i-H ,-1 0 0 O ^ Tt CO to C S N O ON ^ ^ rs ~H (N — COD mgN/L Influent Tt O OO OO SO r-l oo oo in Ti OO 0 0 H H H fS to oo Tf «n Tt NO C N O H H H N i2 <- « « Methanol Dosage IVday o o o r- r- t--rs C N C N C N Tf Tf Tf Tf *n v> m v» 0 0 OO C C 0 0 Date 30-Jun-98 l-Dec-98 10-Mar-99 16-Mar-99 24-Mar-99 29-Mar-99 07-Apr-99 13-Apr-99 20-Apr-99 26-Apr-99 6-May-99 lO-May-99 18-May-99 26-May-99 3-Jun-99 Solids Removal Efficiency Expt. Conf. >TN © © © Solids Removal Efficiency Expt. St. Dev. NO © © © Solids Removal Efficiency Expt. Mean 99.7 98.9 98.4 99.3 Solids Removal Efficiency Expt. Effluent 99.7 99.1 99.2 99.3 98.1 98.0 97.8 99.0 98.5 99.2 98.8 99.3 99.8 Solids Removal Efficiency Ctrl. Effluent 99.2 99.6 99.6 99.3 99.3 99.4 99.5 99.0 99.3 98.6 99.4 99.8 Volatile Suspended Solids mg/L Expt. Effluent OO Tt C N ' T-i T-i CO ro ( N C N V « V V « Volatile Suspended Solids mg/L Ctrl. Effluent • C N C N ' ^ V ^ Tt _ _ ^ - 1 V V V 2 V V Volatile Suspended Solids mg/L Influent CO CO Tf CO NO NO t"* Tt Tf Tf O H N fS (N v> C N oo _ 2 2 2 °° NO O O V I O Tt" Tl" Total Suspended Solids mg/L Expt. Effluent m rs _ m H H ^ (N H H Tf N V I rj N Total Suspended Solids mg/L Ctrl. Effluent • m N N b ^ ^ V V —< oo ^ ,_, V V Total Suspended Solids mg/L Influent oo Vt (N O O ON NO ON ON ro H N (S fN1 rs 2 rs ^ OO TJ- -^•n rs r- r~-.-H .—1 Methanol Dosage L/day o r-~- r-~- r~- r-<N <N <N CN Tf Tf Tf Tf VI VI VI VI 0 0 0 0 0 0 0 0 Date lO-Mar-99 16-Mar-99 24-Mar-99 29-Mar-99 07-Apr-99 13-Apr-99 20-Apr-99 26-Apr-99 6-May-99 lO-May-99 18-May-99 26-May-99 Jun-99 121 TN Removal Efficiency Expt. Conf. m d 00 o-oi TN Removal Efficiency Expt. St. Dev. d 00 m OI cn cn TN Removal Efficiency Expt. Mean 88.9 92.0 94.6 94.4 TN Removal Efficiency Expt. Effluent 88.7 89.1 92.8 86.4 93.7 95.1 95.8 93.5 92.1 97.1 98.0 92.1 96.5 91.1 TN Removal Efficiency Ctrl. Effluent 83.6 94.6 80.6 95.6 96.5 94.6 94.3 96.6 98.7 87.1 95.7 96.1 89.9 86.0 89.0 K Expt. Effluent 2.805 1.470 4.097 2.559 6.736 3.640 3.598 3.107 3.669 5.441 1.637 1.012 2.935 1.577 3.172 Ctrl. Effluent 4.416 1.340 2.340 7.278 1.541 1.744 3.146 4.142 2.498 0.740 8.895 2.387 1.938 3.772 6.350 3.928 Influent 26.891 24.863 37.590 35.425 49.557 58.050 72.805 73.855 56.073 68.938 56.033 49.614 37.294 45.238 35.701 * g Expt. Effluent 0.195 0.340 0.280 0.130 0.160 0.110 0.180 0.100 0.090 0.320 0.180 0.060 0.710 0.100 0.750 Ctrl. Effluent 0.000 0.780 0.310 0.070 0.090 0.300 0.000 0.210 0.230 0.200 0.150 0.100 0.200 0.790 0.120 0.720 Influent 19.173 16.430 18.120 24.690 26.330. 31.230 41.320 35.975 33.800 29.895 38.840 32.850 28.030 23.330 24.720 15.770 TKN me, P/L Expt. Effluent oo „ o o oo d 1.726 4.821 2.655 1.522 2.044 1.222 4.140 1.285 0.802 2.805 1.208 3.022 TKN me, P/L Ctrl. Effluent 1.341 0 1.745 0.322 0 2.305 3.109 1.482 0 7.193 0.741 0.538 1.743 4.990 2.406 TKN me, P/L Influent 26.792 24.663 35.378 33.622 49.206 57.920 72.788 73.628 55.548 68.770 55.777 49.424 36.815 44.825 34.774 NO, mgN/L Expt. Effluent 2.805 1.470 3.289 0.833 1.915 0.985 2.076 1.063 2.447 1.301 0.352 0.210 0.130 0.369 0.150 NO, mgN/L Ctrl. Effluent 3.076 1.340 2.340 5.533 1.219 1.744 0.841 1.033 1.016 0.740 1.702 1.646 1.400 2.029 1.360 1.522 NO, mgN/L Influent 0.099 0.200 0.780 2.212 1.803 0.351 0.130 0.017 0.227 0.525 0.168 0.256 0.190 0.479 0.413 0.927 Methanol Dosage L/day o o o o r-. r. r- t--N N M N *r T T >/l to 00 00 00 00 Date 30-Jun-?8 l-Dec-98 9- Dec-98 10- Mar-99 O i O i O i Q \ C \ 0\ 0\ o\ i3 M M & * 4 ^ .—< (N O 13-Apr-99 20-Apr-99 26-Apr-99 6-May-99 lO-May-99 18-May-99 26-May-99 Jun-99 TP Removal Efficiency Expt. Conf. 15.0 VO oS 10.4 vd TP Removal Efficiency Expt. St. Dev. 12.9 11.6 12.6 O l 0 0 TP Removal Efficiency Expt. Mean 43.2 52.3 54.6 69.0 TP Removal Efficiency Expt. Effluent 34.1 52.3 41.8 57.8 66.1 43.7 41.2 47.0 68.4 61.6 80.7 62.6 68.3 64.4 TP Removal Efficiency Ctrl. Effluent 48.8 51.0 60.3 44.8 21.8 61.0 57.9 39.5 36.2 55.4 50.7 56.9 57.2 50.9 63.0 P g Expt. Effluent 1.799 1.561 1.839 1.746 2.828 4.190 3.818 2.869 2.163 3.215 1.768 3.618 1.641 1.872 Ctrl. Effluent 1.618 1.337 1.299 1.744 3.236 3.255 3.135 3.929 3.456 3.056 4.126 3.956 4.138 2.544 1.944 Influent 3.161 2.730 3.274 3.158 4.140 8.346 7.438 6.497 5.412 6.849 8.365 9.177 9.665 5.181 5.260 Phosphate mgP/L Expt. Effluent 1.590 2.030 2.445 2.433 2.564 2.868 3.712 2.992 2.836 2.111 2.523 1.310 2.825 1.188 1.389 Phosphate mgP/L Ctrl. Effluent 1.743 1.535 2.230 2.171 2.618 2.818 3.159 3.452 3.662 3.348 4.223 3.023 2.780 3.136 2.430 1.756 Phosphate mgP/L Influent 2.847 2.340 4.740 5.115 4.018 4.457 5.900 5.633 5.142 4.546 5.099 4.522 4.100 6.418 3.485 4.880 Methanol Dosage L/day o o o o r» c - r -CN CN CN (N •*r T -3 -m m iy-i OO 00 00 00 Date 30-Jun-98 l-Dec-99 9- Dec-99 10- Mar-99 Os 0\ 0\ Os ON o\ o\ o\ S ^ ^ n *i o\ rt N M O 13-Apr-99 20-Apr-99 26-Apr-99 6-May-99 lO-May-99 18-May-99 26-May-99 Jun-99 122 APPENDIX E : S B R RAW DATA FOR EXPERIMENTAL RUNS Run 1A Date 30-Jun-98 Experiment Time Nitrate Phosphate (mgN/L) (mgP/L) 9:55 1.5862 1.4926 10:10 1.6597 1.5581 10:20 1.0093 2.3159 10:32 1.1543 2.0908 10:50 2.035 2.0632 11:10 2.4083 1.914 11:30 2.8453 1.9002 11:50 3.1749 1.8291 12:10 3.0411 1.7984 12:30 3.1395 1.8357 12:50 3.0965 1.754 13:10 3.0156 1.7542 13:30 3.1197 1.8184 13:50 3.1056 1.7741 14:00 3.1174 1.7789 Run IB Date lO-Mar-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 10:08 1.843 2.222 10:08 1.303 2.315 10:20 1.89 2.211 10:30 1.685 2.604 10:35 1.715 2.169 11:00 2.446 2.478 10:50 1.981 2.3 11:30 3.016 2.885 11:10 1.865 2.227 11:50 3.337 2.379 11:30 1.94 2.194 12:05 3.334 2.344 12:00 5.423 2.366 12:20 3.309 2.387 12:30 5.589 2.468 12:35 3.363 2.405 13:00 2.458 12:50 3.328 2.387 13:30 3.383 2.475 13:10 3.201 2.389 13:50 3.695 2.453 13:30 3.222 2.398 14:00 2.466 123 R u n 2 A Date 19-Mar-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 1.361 2.508 11:50 1.72 2.383 10:50 1.394 2.454 12:50 1.215 2.379 12:00 1.325 2.607 14:00 2.089 2.439 Run 2Ii Date 24-Mar-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 1.592 2.73 10:00 0.639 2.764 10:05 1.808 2.684 10:30 0.528 2.869 10:20 1.521 2.776 11:00 1.447 2.661 10:35 1.663 2.733 11:30 2.302 2.484 10:50 1.556 2.946 11:50 2.901 2.446 11:10 1.603 3.072 12:05 2.1 2.3 11:30 1.317 2.699 12:20 1.809 2.373 12:00 1.29 2.745 12:35 1.446 12:30 1.969 2.524 12:50 1.003 2.482 13:00 2.314 2.968 13:10 1.067 2.426 13:30 3.477 2.876 13:30 0.561 2.485 13:50 4.007 2.808 14:00 0.435 2.746 R u n 2 C Date 29-Mar-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 0.838 3.061 11:50 2.122 2.632 10:50 0.814 3.108 12:50 0.941 2.733 12:00 0.608 3.168 14:00 0.441 2.824 R u n 2D Date 7-Apr-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 10:00 0.036 3.572 10:05 1.503 3.162 10:30 0.615 3.912 10:20 1.501 3.23 11:00 1.819 3.718 10:35 1.297 3.223 11:30 2.731 3.529 10:50 1.473 3.184 11:50 3.51 3.541 11:10 1.373 3.258 12:05 3.204 3.508 11:30 1.08 3.283 12:20 2.434 3.494 12:00 0.902 3.281 12:35 1.936 3.548 12:30 1.073 3.61 12:50 1.623 3.572 13:00 1.784 3.44 13:10 1.723 3.562 13:30 2.788 3.43 13:30 0.748 3.668 13:50 3.525 3.41 14:00 1.075 4.041 1 2 4 Run3A Date 13-Apr-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 1.588 3.709 11:50 2.268 3.038 10:50 1.098 3.565 12:50 0.845 3.102 12:00 0.889 3.596 14:00 0.15 3.269 Run 31$ Date 20-Apr-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 1.45 3.165 10:00 0.361 3.315 10:05 1.964 3.341 10:30 1.175 3.142 10:20 1.613 3.333 11:00 2.648 2.855 10:35 1.375 3.382 11:30 3.541 2.77 10:50 1.293 3.294 11:50 3.943 2.645 11:10 1.241 3.264 12:05 4.078 2.722 11:30 1.325 3.312 12:20 2.963 2.722 12:00 1.11 3.261 12:35 2.042 2.702 12:30 1.559 3.627 12:50 1.569 2.842 13:00 3.111 3.728 13:10 1.643 2.915 13:30 3.555 3.399 13:30 1.752 2.757 13:50 3.853 3.493 14:00 2.163 3.33 Run3C Date 26-Apr-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mg P/L) (mgN/L) (mgP/L) 9:50 2.348 4.141 11:50 2.209 1.998 10:50 1.613 4.064 12:50 0.286 2.11 12:00 1.928 4.633 14:00 0.396 2.858 Run 3D Date 6-May-99 Control Experiment Time Nitrate Phosphate Time Nitrate Phosphate (mgN/L) (mgP/L) (mgN/L) (mgP/L) 9:50 2.429 2.963 10:00 0.069 2.92 10:05 2.281 2.916 10:30 0.701 2.846 10:20 2.483 2.958 11:00 1.59 2.386 10:35 2.294 3.007 11:30 2.435 2.273 10:50 2.267 2.915 11:50 2.426 2.132 11:10 2.02 2.89 12:05 1.807 2.141 11:30 2.101 3.205 12:20 0.874 2.162 12:00 2.101 3.37 12:35 0.094 2.161 12:30 2.586 3.331 12:50 0.317 2.306 13:00 3.6 3.249 13:10 0.278 2.509 13:30 4.241 3.374 13:30 0.105 2.653 13:50 4.473 3.344 14:00 0.097 2.882 125 Run 4A Date lO-May-99 Control Experiment Time Nitrate (mgN/L) Phosphate (nig P/L) Time Nitrate (mgN/L) Phosphate (mgP/L) 9:50 1.6 2.88 11:50 2.07 1.17 10:50 1.36 2.83 12:50 0.28 1.29 12:00 0.93 3.02 14:00 0.11 1.84 Run 4B Date 18-May-99 Control Experiment Time Nitrate (mgN/L) Phosphate (mgP/L) Time Nitrate (mgN/L) Phosphate (mgP/L) 9:50 1.902 2.667 10:00 0.116 2.975 10:05 2.023 2.782 10:30 0.418 2.691 10:20 1.986 2.667 11:00 0.952 2.652 10:35 2 2.625 11:30 1.635 2.553 10:50 2.65 2.99 11:50 1.556 2.331 11:10 1.949 2.874 12:05 0.788 2.454 11:30 1.98 3.195 12:20 0.234 2.448 12:00 1.932 3.205 12:35 0.028 2.547 12:30 2.41 3.233 12:50 0.072 2.662 13:00 3.459 3.378 13:10 0.048 2.818 13:30 3.973 3.434 13:30 0.067 2.964 13:50 4.762 3.686 14:00 0.202 3.275 Run 4C Date 26-May-99 Control Experiment Time Nitrate (mgN/L) Phosphate (mgP/L) Time Nitrate (mgN/L) Phosphate (mgP/L) 9:50 1.847 2.378 11:50 2.478 1.067 10:50 1.767 2.442 12:50 0.237 1.18 12:00 1.06 2.663 14:00 0.174 1.688 Run 4D Date 3-Jun-99 Control Experiment Time Nitrate Phosphate Ammonia Time Nitrate Phosphate Ammonia (mgN/L) (mgP/L) (mgN/L) (mgN/L) (mgP/L) (mgN/L) 9:50 1.935 1.575 0 10:00 0.172 1.55 2.07 10:05 1.84 1.572 0 10:30 0.576 1.481 0.63 10:20 1.89 1.665 0.29 11:00 1.282 1.311 0.38 10:35 1.927 1.698 0.47 11:30 2.642 1.117 0.09 10:50 1.92 1.65 0.44 11:50 2.537 1.118 0.02 11:10 1.875 1.658 0.32 12:05 2.223 1.062 0.17 11:30 2.106 1.701 0.13 12:20 1.355 1.127 0.16 12:00 1.543 1.682 0.63 12:35 0.871 1.119 0.16 12:30 1.201 2.133 1.75 12:50 0.705 1.167 0.58 13:00 2.487 2.145 1.13 13:10 0.499 1.247 0.68 13:30 3.453 2.049 0.7 13:30 0.473 1.384 1.26 13:50 3.78 2.044 0.31 14:00 0.368 1.542 1.73 14:05 4.225 2.035 0.45 14:20 3.685 2.029 0.48 14:35 3.582 2.083 0.43 14:50 3.321 2.119 0.57 15:10 3.498 2.115 0.66 15:30 3.111 2.179 0.97 16:00 2.937 2.205 1.26 126 APPENDIX F : HISTORICAL DATA - JAN. 1997 TO JUNE 1999 MONTHLY AVERAGES M-tHN "jM - - S aoo \% £ aoa|> S 'oa *|> « aoa %\ § S ^ J S S H S ^ S S S K R R N S S S S N - " t S S S S -1 •od ^ - - - " « - • aoo *| 2 aoa W% oa ]j> N M O - L T E £ fa I A S I o 1 f- ? P P P r- P = 2 2 - = S H ssx "|» 8 5 ! § : s 5 ! f ! 5 3 3 : : s ? r ^ m f f l r ^ r ^ S P P r ^ P ( ^ P ™ S S c o P P P P f ^ r ^ ? P ^ ™ f ^ l ^ P s c ? c s s s t c c ? s • ; ; s : s . s ; s ? ? s ? s s | s 127 APPENDIX G : HISTORICAL DATA - JAN. 1999 TO JUNE 1999 W E E K L Y AVERAGES s i s ss a a a R a a s s s s i i s m * 4 2 2 3 i g g S 8 S S i § S S I § s B a S l ! = 4 woils o s - a a s a H a » a = s a s »« 2 -ssxf -« • § s § s ss 2 5 2 2 3 2 § s 4 _ x s a a g s g a s s s s E S S 3 S £ S ••if 8 s S S S i S S S S S S S S s S i S i S S B a s a l d'odf 2 » - : s : 5 a s ! 3 s s s s s : 5 2 » -I S s l g g g S s a § a 2 S o a f ~ <>^§ § § g S g g g g E 3 g 3 3 g g 3 g 3 B g 3 § - » 5 5 - " 2 i s ? s s s s S 5 s s ; 2 S s J iii l ^ g g g l l j j l l l l l l l l l l o a t -s a s s s s a ? a n a S S B ? , 0 0 0 ° " 1 o - n r ) 0 0 o o o ^ o « o [ ^ < t i n §1 l l l i i i § l l i § l i ! o a f -IAS I B S S K R S K 8 S S R S 1 1 1 * 1 1 1 1 111 E E g S = E 5 s » S s = E E E H = I S ; 5 5 •»f 1111 3 3 I § 11 a § § 11 § 1 1 § § 3 § I oa I ^ IAS u S 1 _ 128 APPENDIX H : DETERMINATION OF NITRIFICATION AND DENITRIFICATION RATES Nitrification I Ctrl mg NOx-N/ g MLVSS/day 11.38 12.88 12.78 7.92 9.86 14.10 Nitrification I Expt mg NOx-N/ g MLVSS/day 7.20 9.31 11.57 13.14 13.83 9.45 6.32 12.01 Nitrification I Ctrl slope 35.51 44.31 38.80 33.12 36.94 34.52 Nitrification I Expt slope 24.47 29.56 42.53 46.46 49.31 38.34 24.44 35.26 Methanol Dosage | L/day o o OJ tN 00 00 00 Methanol Dosage | mg/L o o od od 12.2 12.2 12.2 MLVSS | Ctrl MLVSS mg/L 3164 3120 3440 3036 4184 3747 2448 2448 MLVSS | Expt MLVSS mg/L 3399 3176 3676 3536 3564 4056 3868 2936 2936 Date 06/30/98 03/10/99 03/24/99 04/07/99 04/20/99 05/06/99 05/18/99 06/03/99 06/03/99 Denitrification | Expt | Conf. 0.44 1.70 0.41 2.28 Denitrification | Expt | St. Dev. 0.38 1.46 0.35 1.96 Denitrification | Expt | Mean mg NOx-N/ g MLVSS/day 0.34 12.65 18.52 17.79 Denitrification | Ctrl mg NOx-N/ g MLVSS/day -0.54 1.41 2.15 2.11 1.03 0.01 4.80 0.74 Denitrification | Expt mg NOx-N/ g MLVSS/day 2nd Rate | 4.38 4.44 3.34 Denitrification | Expt mg NOx-N/ g MLVSS/day 1st Rate | 0.06 0.61 11.62 13.69 18.27 18.77 16.41 19.18 Denitrification | Ctrl slope 1.70 -4.40 -7.40 -6.40 -4.32 -0.03 -11.74 -1.82 Denitrification | Expt slope 2nd Rate | -13.68 -15.27 -8.19 Denitrification | Expt slope 1st Rate | -0.22 -1.92 -42.72 -48.40 -65.13 -76.12 -63.46 -56.31 MLVSS | Ctrl MLVSS mg/L 3164 3120 3440 3036 4184 3747 2448 2448 MLVSS | Expt MLVSS mg/L 3399 3176 3676 3536 3564 4056 3868 2936 2936 Date 06/30/98 03/10/99 03/24/99 04/07/99 04/20/99 05/06/99 05/18/99 06/03/99 06/03/99 129 APPENDIX I: NITRATE AND PHOSPHATE PLOTS (NTS) 130 Experimental SBR Jun 30, 1 3 Sl E 2.8 r c o o o o o m o u ~ > o o o o O ro O M iS O ^ ^ U"> ^ IO O g o o o o g o i n o o o g O O O O O g O j J I O O O g 1 X NOT] I— o- PCK| O O O O g » / ) O L / J O O O O Mays OOC^OJ>'/>C>1J'>0000 I M NOx I |. • • - -P04| 131 APPENDIX J : DETERMINATION OF PHOSPHATE RELEASE R A T E MLVSS (mg/L) P-Release AP (mg/L) AP/MLVSS (mgP/g MLVSS) Start (hr:min) Duration (min) (day) 3676 13:10 50 0.035 0.45 0.12 3536 13:10 50 0.035 0.55 0.16 3564 12:35 85 0.059 0.69 0.19 4056 12:35 85 0.059 0.75 0.18 3868 12:20 100 0.069 0.94 0.24 2936 12:50 70 0.049 0.48 0.16 132 APPENDIX K : SOLIDS PRODUCTION DETERMINATION a 0.37 rag VSS/mg COD V 877.2 m 3 Methanol 1185320 mg/L COD Control SBR] Week Q 1 Q2 (CH30H) Q w w L, Lr S, m3/day m3/day m3/day m3/day kg/day COD (mg/L) COD (kg/day) COD (kg/day) VSS Jan. I to Jan. 8 545 0 545 9 56 307 0 167 6068 Jan. 8 to Jan. 15 552 0 552 9 57 296 0 163 6154 Jan. 15 to Jan. 22 672 0 672 8 50 258 0 173 6241 Jan. 22 to Jan. 31 465 0 465 6 41 225 0 105 6956 Feb. 1 to Feb. 5 481 0 481 8 50 325 0 156 6047 Feb. 5 to Feb. 12 488 0 488 9 52 406 0 198 6005 Feb. 12 to Feb. 19 381 0 381 9 52 486 0 185 5935 Feb. 19 to Feb. 26 364 0 364 8 50 370 0 135 5976 Feb. 26 to Mar. 4 483 0 483 8 50 415 0 200 6086 Mar. 4 to Mar. 11 386 0 386 9 55 492 0 190 6480 Mar. 11 to Mar. 16 317 0 317 10 68 568 0 180 7069 Mar. 16 to Mar. 24 292 0 292 10 60 589 0 172 6221 Mar. 24 to Mar. 29 261 0 261 9 47 794 0 207 5299 Mar. 29 to Apr. 7 255 0 255 9 60 0 6961 Apr. 7 to Apr. 13 246 0 246 9 45 0 5236 Apr. 13 to Apr. 20 244 0 244 9 54 787 0 192 5950 Apr. 20 to Apr. 26 262 0 262 10 34 549 0 144 3424 Apr. 26 to May 6 352 0 352 10 54 447 0 157 5145 May 6 to May 10 413 0 413 11 88 0 7677 May 10 to May 18 436 0 436 20 112 444 0 194 5586 May 18 to May 26 479 0 479 23 123 447 0 214 5310 May 26 to Jun. 3 574 0 574 18 89 287 0 164 4895 Jun 3 to Jun 11 693 0 693 11 58 240 0 166 5287 Experiment SBR2 Week Q 1 Q2 (CH30H) Q w w LCH30H S, Increase in Solids Wasted (kg/day) m3/day m3/day m3/day m3/day kg/day COD (mg/L) COD (kg/day) COD (kg/day) VSS A . Mean St. Dev. Conf. Jan. 1 to Jan. 8 545 0 545 11 80 307 0 167 7541 24 Jan. 8 to Jan. 15 552 0 552 11 73 296 0 163 6869 16 Jan. 15 to Jan. 22 672 0 672 10 64 258 0 173 6186 13 Jan. 22 to Jan. 31 465 0 465 8 70 225 0 105 8694 29 Feb. 1 to Feb. 5 481 0 481 11 64 325 0 156 6065 15 Feb. 5 to Feb. 12 488 0 488 11 61 406 0 198 5715 8 Feb. 12 to Feb. 19 381 0 381 11 60 486 0 185 5696 9 Feb. 19 to Feb. 26 364 0 364 9 54 370 0 135 5932 4 Feb. 26 to Mar. 4 483 0 483 8 51 415 0 200 6190 1 Mar. 4 to Mar. 11 386 0 386 8 52 492 0 190 6554 -4 12 10 5 Mar. 11 to Mar. 16 317 0.027 317 7 53 568 32 212 7443 -15 Mar. 16 to Mar. 24 292 0.027 292 8 48 589 32 204 5894 -11 Mar. 24 to Mar. 29 261 0.027 261 9 53 794 32 239 5997 6 Mar. 29 to Apr. 7 255 0.027 255 10 61 6144 1 -5 10 8 Apr. 7 to Apr. 13 246 0.054 246 11 63 5598 18 Apr. 13 to Apr. 20 244 0.054 244 11 71 787 64 256 6324 17 Apr. 20 to Apr. 26 262 0.054 262 13 67 549 64 208 5119 32 Apr. 26 to May 6 352 0.054 352 14 88 447 64 221 6271 34 26 9 7 May 6 to May 10 413 0.081 413 16 73 4671 -15 May 10 to May 18 436 0.081 436 27 131 444 96 290 4945 19 May 18 to May 26 479 0.081 479 32 169 447 96 310 5251 46 May 26 to Jun. 3 574 0.081 574 28 129 287 96 261 4620 39 Jun 3 to Jun 11 693 0.081 693 15 69 240 96 262 4545 11 20 24 18 Theoretical and Kinetic Calculation Theoretical Kinetic Model Q (CH30H) CHjOH CH,OH Sludge L , L , dlVdt a(dL/dt)V m3/day LMay/day kg/day kg/day COD (mg/L) COD (mg/L) mg/L/day kg/day 0.000 0 0.0 0.0 0 0 0 0 0.027 27 21.3 4.5 1185320 0 36 12 0.054 54 42.7 9.1 1185320 0 73 24 0.081 81 64.0 13.6 1185320 0 109 36 = Net YIELD Note: Substrate concentration in terms of COD =volumeofSBR 133 APPENDIX L : O R P MONITORING - A V E R A G E FOUR HOUR C Y C L E TREND 134 •31 o o o o >^ o o o o o o o o (?a o o o o m O VI O VI O £j iv"i o i*"> i / I O ' O O ' O (A">) JHO aSBJSAV (A<n) JHO »8EJ3AV =1 •51 O O O O °3 o o o o o o o o o o o o o o wi o v"i i/^ o vi o >o v " » o v n w - i o w - ) 0 < o o 135 136 APPENDIX M : O R P MONITORING - A V E R A G E DAILY TREND o o o o o o o o o o o o o o o o o o O V) O V) Vi O Vi O «~i O "~> V) O V) O Vi C\«0 dMO 93BJ3AV (A™) dHO 93EJ3AV SI o o o o o o o o o o o o o o o o o o vi o vi o vi vi o vi v> o vi vi o vi o vi (A«i) dHO aScjaAV (A«0 dMO S8EJ3AV 137 fl o (A™) dHO aScJSAy (A™) dHO aSejaAy [3 U (A«") dHO (A™) JHO aSEJJAV 138 (A«l) JMO OSRBAV (A»") dHO aScJaAy (A™) dHO aSiusAy 139 APPENDIX N : RAW DATA AND GRAPHS FROM B A T C H STUDY Mixed Batch Reactor with Acclimatized Sludge Reactor A Reactor B Reactor C Reactor D Methanol 0 4.23 8.47 12.7 Cone. mg CH3OH/L mg CH3OH/L mg CH3OH/L mg CH3OH/L Time Nitrate Nitrate Nitrate Nitrate hrs. mg/L mg/L mg/L mg/L 0 7.31 7.2 7.21 6.76 0.17 6.63 5.74 6.69 6.71 0.33 6.31 5.94 5.92 5.94 0.50 5.43 5.14 5.03 5.43 0.67 5.5 4.52 4.49 4.76 0.83 5.24 4.13 3.8 4.07 1.00 5.02 3.7 3.14 3.13 1.17 4.94 3.4 2.62 2.56 1.33 4.17 2.99 2.21 1.78 1.50 3.83 2.55 1.66 1.12 1.67 3.47 2.19 1.42 0.56 2.00 2.74 1.38 0.64 0 2.33 1.96 0.66 0.04 0.1 2.67 1.17 0.04 0.07 0.05 3.00 0.44 0.09 0.07 0.03 Mixed Reactor with Non-Acclimatized Sludge Reactor A Reactor B Reactor C Reactor D Methanol 0 4.66 8.47 12.7 Cone. mg CH3OH/L mg CH3OH/L mg CH3OH/L mg CH3OH/L Time Nitrate Nitrate Nitrate Nitrate hrs. mg/L mg/L mg/L mg/L 0 19.08 36.79 37.85 37.1 0.17 20.84 36.33 36.06 35.48 0.33 20.06 36.32 35.37 34.74 0.50 20.17 36.06 35.35 34.5 0.67 21.25 35.89 34.77 34.19 0.83 19.82 35.8 34.34 34.07 1.00 20.07 35.95 34.27 33.59 1.17 20.04 35.86 33.94 33.28 1.33 35.62 34.1 32.75 1.67 20.83 35.49 33.51 32 Non-Mixed Batch Reactor with Acclimatized Sludge Reactor A Reactor B Reactor C Reactor D Methanol 0 4.23 4.23 12.7 Cone. mg CH3OH/L mg CH3OH/L mg CH3OH/L mg CH3OH/L Time Nitrate Nitrate Nitrate Nitrate hrs. mg/L mg/L mg/L mg/L 0 8.75 9.04 0.17 8.8 8.89 8.8 8.51 0.33 8.58 8.48 8.37 0.50 9.32 8.35 8.4 7.87 0.67 8.75 8.65 8.59 8.42 0.83 8.57 8.45 8.11 8.12 1.00 8.39 8.48 8.16 8.31 1.17 8.57 8.27 8.09 7.17 1.33 8.68 8.05 8.03 7.85 1.50 8.63 8.31 8.04 7.65 1.67 8.46 7.97 7.79 6.89 2.00 8.41 7.96 7.78 6.89 2.33 8.49 7.77 8 7.06 2.67 8.25 7.51 7.34 6.49 3.00 8.62 7.61 7.47 6.37 140 Mixed Batch Test with Non-Acclimatized Sludge Elapsed Time (hrs.) Non-Mixed Batch Test with Acclimatized Sludge 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Elapsed Time (hrs.) 141 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0064021/manifest

Comment

Related Items