Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Oxidation-reduction potential and organic carbon sources as two control parameters for simultaneous nitrification… Zhao, Hong Wang 1998

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

Item Metadata

Download

Media
831-ubc_1998-272745.pdf [ 12.57MB ]
Metadata
JSON: 831-1.0050308.json
JSON-LD: 831-1.0050308-ld.json
RDF/XML (Pretty): 831-1.0050308-rdf.xml
RDF/JSON: 831-1.0050308-rdf.json
Turtle: 831-1.0050308-turtle.txt
N-Triples: 831-1.0050308-rdf-ntriples.txt
Original Record: 831-1.0050308-source.json
Full Text
831-1.0050308-fulltext.txt
Citation
831-1.0050308.ris

Full Text

OXIDATION-REDUCTION POTENTIAL A N D ORGANIC CARBON SOURCES As Two CONTROL PARAMETERS FOR SIMULTANEOUS NITRIFICATION A N D D E N I T R I F I C A T I O N I N B I O L O G I C A L N U T R I E N T R E M O V A L P R O C E S S E S by H O N G W A N G Z H A O B . A . S c , Tsinghua University, China, 1987 M . Sc., The University of Regina, 1993 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Department of Civil Engineering) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January 1998 ®HongW. Zhao, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of C' y, / Cru&rinfl The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT The main objective of this study was to demonstrate the feasibility of achieving carbon (C), nitrogen (N) and phosphorus (P) removal from domestic sewage, in a two-stage intermittent aeration (IA) process, under conditions favorable to simultaneous nitrification and denitrification (SND). It was demonstrated in this study that, under the above operating conditions, the two-stage process achieved levels of N, P and C removals, similar to the 3-Bardenpho process operated at a dissolved oxygen (DO) concentration of 3 mg/L. It was possible to simultaneously reduce the influent total nitrogen concentration from 24-32 mgN/L to 6-12 mgN/L and the influent total phosphorus concentration from 3.0-6.0 mgP/L to less than 1 mgP/L, in the two-stage process. Compared to the 3-Bardenpho process, the two-stage process removed the same amount of total COD (75-90%) from the influent, but it produced more solids (average 480 mg/L higher) containing a 1-3% higher volatile content. Also compared to the 3-Bardenpho process, the two-stage process produced sludge with higher SVIs; however, this did not lead to wash-out of solids. It was calculated through nitrogen balances that the unaccounted for nitrogen loss in the aeration tank (i.e., the amount of SND) accounted for up to 50% of the influent T K N for the two-stage process under low DO conditions and an average of 15% for the 3-Bardenpho process at a DO concentration of 3 mg/L. The experimental results suggested that aerobic denitrification and heterotrophic nitrification were the main causes of the loss in the current systems. However, anoxic microzone denitrification cannot be precluded. According to the pared 7-tests, significant differences in process performance (e.g., the percentages of nitrification and denitrification in the I A C M tank) were observed for the two-stage process, when different ORP ranges were used to control the intermittent aeration; this proved that ORP range can be used as a control parameter for SND (i.e., nitrogen removal) in the I A C M tank. As confirmed by the independent f-tests, acetate and methanol additions improved both N and P removals in all three processes at acetate dosages less than 50 mgCOD/L and methanol dosages less than 30 mgCOD/L, but not at high dosages (e.g., 100 mgCOD/L for acetate and 60 mgCOD/L for methanol). It was suggested that the key factor in optimizing N and P removal in the two-stage process is to maximize carbon storage in the anaerobic zone by using ORP to control the degree of nitrification in the I A C M tank. Process dynamic behavior, in response to instantaneous ammonium and nitrate shock loads, was also investigated for the three experimental systems. The dynamic responses more clearly showed that the nitrification and denitrification in the I A C M tank occurred simultaneously. Based on the dynamic responses, a technique was developed to determine maximum specific nitrification rate (SNR) and maximum specific denitrification rate (SDNR). It was found that the maximum SNRs in the IACM tank (0.39-1.69 mgNH+ 4-N/gMLVSSxh" J) were considerably lower than those in the 3-Bardenpho aerobic zone (3.4-8.1 mgNFfV N/gMLVSSxh"1); this indicated that low DO conditions inhibits nitrification. Further, the maximum SDNRs in the IACM tank were in a range of 0.16-1.26 mgNOx-N/gMLVSSxh"1, which were also considerably lower than that in the 3-Bardenpho anoxic zone (2.5 mgNOx-N/gMLVSSxh"1); this indicated that low DO conditions, compared to anoxic conditions, inhibits denitrification. i i T A B L E OF C O N T E N T S A B S T R A C T i i T A B L E O F C O N T E N T S ii i L I S T O F T A B L E S viii L I S T O F F I G U R E S ix A C K N O W L E D G M E N T S xiv 1. I N T R O D U C T I O N 1 1.1 Need for Biological Nutrient Removal 1 1.2 Need for Developing N e w Processes 2 1.3 Research Objectives 6 2. L I T E R A T U R E R E V I E W 8 2.1 Biological Nitrification 8 2.2 Biological Denitrification 12 2.3 Simultaneous Nitrification and Denitrification (SND) 17 2.4 Kinetics of Nitrification and Denitrification 21 2.5 Biological Excess Phosphorus Removal 23 2.6 Single-Sludge B N R Process 26 2.6.1 Mainstream B N R Process 26 2.6.2 Intermittent Aeration (IA) Process 27 2.6.3 B N R Process Using S N D 28 2.7 O R P Control under L o w D O Conditions 33 iii 2.8 Role of Organic Substrates in BNR processes 37 2.8.1 Role of Organic Substrate in Biological Phosphorus Removal 37 2.8.2 Role of Organic Substrate in Biological Nitrogen Removal 37 2.8.3 Role of Organic Substrate in SND Process 39 3. MATERIALS AND METHODS 42 3.1 Seeding Sludge and Wastewater Source 42 3.2 Reactor Systems 43 3.3 Data Acquisition and Control System 45 3.4 Process Evaluation under Steady-State Conditions 46 3.4.1 Experimental Design 46 3.4.2 Operating Conditions 50 3.4.3 Sampling 51 3.4.4 Methods of Calculation 52 3.4.5 Dynamic Study Under Steady-State Conditions 53 3.5 Dynamic Study Under Transient-State Conditions 54 3.5.1 Transient-State Study Under Continuous-Flow Conditions 55 3.5.2 Transient-State Study Under Batch Conditions 56 3.6 Analytical Methods 56 3.7 Quality Control and Assurance for Analytical Data 58 4. RESULTS AND DISCUSSION 61 4.1 Steady-State Performance Evaluation 61 4.1.1 Observations on ORP, DO and pH Recordings 63 4.1.1.1 General Observations 63 iv 4.1.1.2 Relationship between ORP and DO Concentration 69 4.1.2 Feed Characteristics and Variations 76 4.1.2.1 Diurnal Variations in the Feed Strength and Composition 77 4.1.2.2 Variation in the Influent TCOD/TKN Ratio 77 4.1.2.3 Variation in Methanol Dosage 80 4.1.3 Nitrogen 81 4.1.3.1 Overall Nitrogen Removal 81 4.1.3.2 Ammonia Loss and Denitrification in the Anaerobic Zone 92 4.1.3.3 Ammonia Loss and Denitrification in the Anoxic Zone 97 4.1.3.4 Nitrogen Removal due to Sludge Wastage and Nitrogen Content in Solids ; 101 4.1.3.5 Percentage of Nitrification and Overall TKN Removal 103 4.1.3.6 Denitrification in the Aeration Tank Ill 4.1.3.7 Summary 120 4.1.4 Phosphorus 124 4.1.4.1 Overall Phosphorus Removal 124 4.1.4.2 Phosphorus Release and Uptake 130 4.1.4.3 Phosphorus Content in the Sludge 139 4.1.5 Carbon 143 4.1.5.1 Overall COD Removal 143 4.1.5.2 Carbon Loss in the Anaerobic Zone 144 4.1.6 MLSS and Volatile Content 148 4.1.7 Sludge Volume Index (SVI) and Effluent Suspended Solids 151 V 4.1.8 Process Dynamics under the Steady-State Conditions 154 ^ 4.1.8.1 NOx and Ammonia 155 4.1.8.2 Ortho-P 157 4.1.8.3 Chemical Oxygen Demand (COD) 159 4.2 Process Dynamics under the Transient-State Conditions 160 4.2.1 Process Dynamics under the Batch Conditions 161 4.2.1.1 Dynamic Responses to the Ammonium Shock Loading 161 4.2.1.2 Dynamic Responses to the Nitrate Shock Loading 164 4.2.2 Process Dynamics under the Continuous-Flow Conditions 167 4.2.2.1 Dynamic Responses to the Ammonium Shock Loading 167 4.2.2.2 Dynamic Responses to the Nitrate Shock Loading 178 4.2.3 Kinetic Study 182 4.2.3.1 Analysis of Reaction Rate 184 4.2.3.2 Kinetic Rates 192 4.3 Mechanisms of Simultaneous Nitrification and Denitrification 200 4.3.1 Ammonia Oxidation under Anoxic Conditions 200 4.3.2 Ammonia Stripping in the Aeration Tank 202 4.3.3 Nitrogen Loss during Nitrification Process 203 4.3.4 Anoxic Microzone Inside of the Floes 204 4.3.5 Aerobic Denitrification and Heterotrophic Nitrification 206 4.4 Roles of ORP Control Range and External Substrates in Selection of Bacteria Population 209 4.4.1 Role of ORP Control Range 211 vi 4.4.2 Role of Organic Substrate 212 4.4.3 Practical Implications of SND 215 4.4.3.1 Implications for Mainstream BNR Processes 216 4.4.3.2 Implications for a Single-Sludge IA Process 217 4.4.3.3 Implications for Two-Stage IA Processes 218 4.4.3.4 Implications for a Novel BNR Process 219 5. CONCLUSIONS AND RECOMMENDATIONS 223 5.1 Summary and Conclusions 223 5.1.1 Conclusions Related to the Overall Process Performance 223 5.1.2 Conclusions Related to the Process Control 225 5.1.3 Conclusions Related to the Role of External Substrate Dosage 226 5.1.4 Conclusions Related to the SND Mechanisms 226 5.1.5 Conclusions Related to the Transient-State Experiments 226 5.2 Recommendations 227 REFERENCES 229 APPENDIX A - CALCULATION METHODS FOR MASS BALANCES 245 APPENDIX B - CALCULATION METHODS FOR THE AMOUNT OF NITRIFICATION AND THE UNACCOUNTED FOR NITROGEN LOSS 247 APPENDIX C ~ THE RESULTS OF T-TESTS 249 APPENDIX D ~ RAW DATA 251 vii LIST OF TABLES Number Page 3.1 The variable experimental conditions 49 3.2 The fixed experimental conditions 50 3.3 Summary of sampling locations and analytical parameters 52 3.4 Summary of mass balances 53 3.5 Experimental conditions for the transient-state experiments under the continuous-flow conditions 55 3.6 Sample handing 56 3.7 The Estimated precision for the analytical techniques 59 4.1 Typical ranges of ORP and DO concentration and typical cycle time for each run 64 4.2 Diurnal variations in the influent feed strength and composition 78 4.3 Mean influent TCOD/TKN and TCOD/TP ratios for each run 79 4.4 The results of the paired t-test on the effluent TN concentration for the acetate addition runs 83 4.5 The results of the paired t-tests on the effluent TN concentration for the methanol addition runs 84 4.6 The results of the independent Mests on overall TN removal efficiency between a pair of replicates 87 4.7 Comparison with the results from other similar studies 91 4.8 Summary of the maximum and real SNRs and the calculation procedure 193 4.9 Summary of the maximum and real SDNRs and the calculation procedure 196 4.10 The PHA content in the sludge obtained from run 11 219 viii LIST OF FIGURES Number Page 2.1 The different sensitivities of the denitrification systems of various bacteria to the DO concentration (from Kuenen and Robertson, 1987) 13 2.2 Scheme showing the various possible options for NAD(P)H utilization available to Thiosphaerapantotropha (from Robertson et al., 1988) 18 3.1 Schematic diagram of the three bench-scale reactor systems 44 3.2 The precision of ORP measurement 60 4.1 Solids levels in the three systems gradually reaching steady-state in run 1 (designed SRT= 15 day) 62 4.2 Typical pH profiles in the two IACM tanks (recorded during run 9) 66 4.3 The typical measured ORP and DO profiles corresponding to the nominal high and low ORP control ranges (recorded during run 5): a) ORP profiles and b) DO profiles 67 4.4 The typical measured ORP and DO profiles corresponding to the nominal wide and narrow ORP control ranges (recorded during run 3): a) ORP profiles and b) DO profiles 68 4.5 Typical relationship between ORP and DO concentration (data from run 3) 70 4.6 The typical linear relationship between the ORP and adjusted DO concentration in the high ORP IACM tank (data from run 5) 74 4.7 The typical linear relationship between the ORP and adjusted DO concentration in the low ORP IACM tank (data from run 5) 75 4.8 The overall TN removal efficiencies obtained from the acetate addition runs .... 83 4.9 The overall TN removal efficiencies obtained from the methanol addition runs . 84 4.10 The overall TN removal efficiencies obtained from the runs under the wide/narrow ORP control regime 85 4.11 The amount of ammonia loss in the anaerobic zone obtained from the three systems in the acetate and methanol addition runs with the same feed 92 ix 4.12 The percentages of N0X removal in the anaerobic zone obtained from the runs under the high/low ORP control regimes 94 4.13 The percentage of denitrification in the anaerobic zone obtained from the acetate addition runs 96 4.14 The percentage of denitrification in the anaerobic zone obtained from the methanol addition runs 96 4.15 The percentages of NOx removal in the anoxic zone obtained from the acetate and methanol addition runs 99 4.16 The percentages of denitrification in the anoxic zone obtained from the acetate and methanol addition runs 100 4.17 The organic nitrogen concentrations in the aerobic mixed liquors vs. the MLSS concentrations 102 4.18 The percentages of nitrification obtained from the acetate addition runs 105 4.19 The percentages of nitrification obtained from the methanol addition runs 105 4.20 The percentages of nitrification obtained from the runs under the wide/narrow ORP control regime 107 4.21 The percentage of NOx removal in the aeration tank obtained from the acetate addition runs 113 4.22 The percentage of NOx removal in the aeration tank obtained from the methanol addition runs 113 4.23 Percentage of denitrification in the aeration tank obtained from the acetate addition runs 115 4.24 Percentage of denitrification in the aeration tank obtained from the methanol addition runs 115 4.25 Percentage of NOx removal in the aeration tank obtained the runs under the wide/narrow ORP control regime 117 4.26 The overall TP removal efficiencies obtained from the acetate addition runs 124 4.27 The overall TP removal efficiencies obtained from the methanol addition runs .. 125 4.28 The individual data of the effluent ortho-P concentrations from the three systems during run 11 126 4.29 The overall TP removal efficiencies obtained from the runs under the wide/narrow ORP control regime 128 4.30 The overall TP removal efficiencies vs. the corresponding NOx levels in the return sludge 129 4.31 The ortho-P release in the anaerobic zone obtained from the acetate addition runs 132 4.32 The individual data of the Ortho-P release in the anaerobic zone during runs 5 and 6 135 4.33 The ortho-P release in the anaerobic zone obtained from the methanol addition runs 135 4.34 The amounts of phosphorus uptake in the anoxic zone obtained from the acetate and methanol runs vs. the external substrate dosages 136 4.35 The phosphorus contents in the sludge for the acetate addition runs 140 4.36 The phosphorus contents in the sludge for the methanol addition runs 141 4.37 Comparison of the effluent filtered COD concentration among the three processes 144 4.38 The amount of the filtered COD loss in the anaerobic zone obtained from all methanol addition runs 147 4.39 The MLSS levels vs. the amount of total COD removed in the total process .... 149 4.40 Comparison on the volatile content of the aerobic sludge among the three processes ; 150 4.41 Comparison of the aerobic mixed liquor SVI among the three processes 151 4.42 Comparison of the effluent total SS concentration among the three processes .. 154 4.43 Typical ammonia variations in long and short aeration cycles 155 4.44 Typical NOx variations in long and short aeration cycles 156 4.45 Typical ortho-P variations in long and short aeration cycles 157 xi 4.46 Typical filtered COD variations in long and short aeration cycles 160 4.47 The ammonium concentration profiles in the two IACM tanks in response to a 30 mg/L ammonium shock loading at time 0 162 4.48 The NOx concentration profiles in the two IACM tanks in response to a 30 mg/L ammonium shock loading at time 0 162 4.49 The ortho-P concentration profiles in the two IACM tanks in response to a 30 mg/L ammonium shock loading at time 0 163 4.50 The NOx concentration profiles in the two IACM tanks in response to a 30 mg/L nitrate shock loading at time 0 165 4.51 The ammonium concentration profiles in the two IACM tanks in response to a 30 mg/L nitrate shock loading at time 0 166 4.52 The ortho-P concentration profiles in the two IACM tanks in response to a 30 mg/L nitrate shock loading at time 0 166 4.53 The measured ammonia concentration profiles in various zones after the ammonium shock loading was imposed 171 4.54 Typical measured and theoretical ammonia profiles in an IACM tank 173 4.55 The measured NOx concentration profiles in various zones after the ammonium shock loading was imposed 174 4.56 The measured ortho-P concentrations in various zones in one of the methanol runs after the ammonium shock loading was imposed 175 4.57 The measured ortho-P concentrations in the two 2-stage process anaerobic zones and the 3-Bardenpho anoxic zone in one of the acetate runs after the ammonium shock loading was imposed 177 4.58 The measured ortho-P concentrations in the two IACM tanks and the 3-Bardenpho aerobic zone in one of the acetate runs after the ammonium shock loading was imposed 177 4.59 The measured NOx concentration profiles in various zones after the nitrate shock loading was imposed 178 4.60 Typical measured and theoretical NOx profiles in an IACM tank 179 xii 4.61 The measured ammonia concentration profiles in various zones after the nitrate shock loading was imposed 180 4.62 The measured ortho-P concentration profiles in various zones after the nitrate shock loading was imposed 181 4.63 Linear plots developed from measured ammonium profiles according to Equation 4.20 (data obtained from the third series of transient-state test) 191 4.64 Linear plots developed from measured NOx profiles according to Equation 4.20 (data obtained from the third series of transient-state test) 191 4.65 Comparisons of the maximum SNR with the corresponding real SNR 194 4.66 Comparisons of the maximum SDNR with the corresponding real SDNR 197 4.67 The configuration of the proposed BNR process using methanol 220 xiii ACKNOWLEDGMENTS I wish to express my sincere thanks to many people who assisted in the completion of this research: Dr. D. S. Mavinic, Professor and Head of Environmental Engineering Group, Civil Engineering Department, UBC and Dr. W. K. Oldham, Professor Emeritus of Civil Engineering, UBC for their supervision, guidance and encouragement throughout this research. Dr. J. L. Barnard, specialist consultant at Reid Crowther, and Dr. S. Duff, Associate Professor, Chemical Engineering, UBC, for serving on my committee and their constructive criticisms and suggestions in the preparation of the final draft. Fred Koch, for freely sharing his ideas and experience, enlightening discussion, and encouragement during the hard time, and Dr. W. D. Ramey for his guidance on the microbiology aspects of this research. Jufang Zhou, Paula Parkinson, and Susan Harper of the UBC Environmental Engineering Laboratory for their assistance in the sampling analysis. Guy Kirsch and John Wong, Civil Engineering Workshop technicians at UBC, for their assistance in setting up the experimental apparatus. My wife for her endless patience and support throughout my graduate studies at UBC. The Natural Sciences and Engineering Research Council of Canada (NSERC) and the Department of Civil Engineering, UBC, for financial support. xiv Chapter One I N T R O D U C T I O N 1.1 Need for Biological Nutrient Removal The discharge of wastewater containing nitrogen and phosphorus into aquatic environments can lead to eutrophication of surface water bodies, which can stimulate algae blooms and promote the growth of nuisance aquatic vegetation. In addition to being aesthetically unsightly, the presence of algae and aquatic plants may interfere with the beneficial uses of water resources. The other adverse effects, caused by high nitrogen level in treated effluent, may include: 1) depletion of dissolved oxygen in receiving waters, 2) toxicity toward aquatic life, and 3) health hazards to humans, such as methaemoglobinemia (blue baby syndrome) in infants under six months of age. Either nitrogen or phosphorus will be the limiting nutrient controlling eutrophication because of the relatively large quantities required for biomass growth, compared to other micro nutrients such as sulfur, potassium, calcium and magnesium. Usually, the limiting nutrient is considered to be phosphorus in freshwater environments and nitrogen in marine waters (Randall et al., 1992). However, phosphorus removal has been recommended even in cases where limnological studies have shown nitrogen to be the primary and phosphorus the secondary growth limiting nutrient in a receiving body of water (Porter, 1975). Also because the limiting nutrient dynamics is generally poorly understood for most bodies of 1 water, the best eutrophication control policy is simultaneous reduction of both nitrogen and phosphorus inputs (Randall et al., 1992). 1.2 Need for Developing New Processes Various treatment processes have been used, employing physiochemical and biological methods, to control the amount of nutrients discharged. The processes most used initially were biological nitrification for ammonia oxidation, separate-stage biological denitrification using methanol for nitrogen removal (also known as post-denitrification), and chemical precipitation for phosphorus removal. In recent years, a number of biological processes have been developed to remove both nitrogen, and phosphorus, using carbon naturally developed in the wastewater. These processes have considerable appeal to designers and operators because the use of chemicals and external carbon sources have been eliminated. Mainstream, single-sludge, biological nutrient removal (BNR) processes employ combinations of anaerobic, anoxic, and aerobic zones, with a single sedimentation step for recycle of return activated sludge. Single-sludge means each process train contains only one secondary clarifier and settled biomass must recycle through all zones. The most commonly used mainstream, single-sludge, BNR processes include: 1) the Bardenpho processes (including 4 stage, 5 stage, 3 stage and 2 stage) and 2) the University of Cape Town (UCT) process. The original Bardenpho process consists of four stages. It can achieve complete nitrogen removal and incidental phosphorus removal. Subsequent developments were 2 aimed at optimizing biological phosphorus removal (BPR), when it was recognized that anaerobic conditions were critical to the process. The three-stage and two-stage processes allow for less degrees of nitrogen removal in conjunction with phosphorus removal. The UCT process, developed at the University of Cape Town (Siebritz et al., 1983), was a modification of the Bardenpho process with additional internal recycle to ensure minimal nitrate concentrations in the anaerobic zone. In general, the above processes are among the first choice by the design engineers, because these processes are efficient and stable in nutrient removal, and also because a sufficient amount of research and full-scale data have been made available. Nevertheless, these processes are usually very complicated; therefore, they are unfavorable or inadequate in either retrofitting or process optimization. As far as process control and optimization are concerned, the mainstream, single-sludge, BNR processes would only attain the best performance under steady-state conditions, because fixed volumes of the reactors are dedicated to certain functions. Unless a high degree of operational flexibility is provided in the design, process optimization would be difficult to achieve under dynamic situations. Hao and Huang (1996) suggested that under less stringent nitrogen requirements, a single-stage, intermittent-aeration (IA), activated sludge process could be a viable alternative to the mainstream processes. It was noted by both Batchelor (1982) and Hao and Huang (1993) that intermittent aeration enables the process to be easily adjusted to dynamic conditions, such as influent variation or temperature change. Incorporating an anaerobic selector (for the selection of both phosphorus accumulating organisms (PAOs) 3 and non-filamentous organisms) into a single-stage IA process forms a two-stage IA process, which has a configuration similar to the two-stage Bardenpho (i.e., A/O) process. It was demonstrated (de la Menardiere et al., 1991; Lo et al., 1994) that the two-stage IA process can achieve both nitrogen and phosphorus removals. Compared to the mainstream BNR processes, the IA processes (including single and two stages) appear to have the following advantages. Firstly, the IA processes may be more readily incorporated into any given existing plants, at the lowest cost, because of their simple configurations. Secondly, they may be superior in process control and optimization under dynamic situations. With a data acquisition and control system (DAC), the intermittent aeration schedule in the IA processes may simply rely on preset sensor measurements, such as oxidation-reduction potential (ORP), dissolved oxygen (DO), pH, ammonium (NFT4) or nitrates (NOx). In response to dynamic situations, the aeration schedule (controlling blower on/off) could change according to preset sensor values. Therefore, process optimization can easily be realized in the IA processes. The IA processes studied so far are usually controlled either by a timer, DO concentration, or ORP. The mostly used aeration cycle time is in a range of 0.5 to 3 hours. During the air-on period, DO concentration is usually higher than 2 mg/L. During the air-off period, the DO concentration is zero. In this type of IA process, ammonia is converted to nitrate during the air-on period and the nitrate produced is used to remove organic substrate during the subsequent air-off period. Simultaneous nitrification and denitrification (SND) implies that nitrification and denitrification occur concurrently in an aeration tank, without forming distinct anoxic 4 zones. Unaccounted for nitrogen loss of up to 30% (due to SND) was very often observed in aerobic nitrifying activated sludge systems (US EPA, 1987; van Huyssteen et al., 1990; Randall et al., 1992; Rabinowitz and Barnard 1995). While the loss has been recognized for decades, there is a large body of designers that does not take this into account in design. The two mostly recognized explanations for the loss are: 1) denitrification in the anoxic microzones inside of the activated sludge floe in combination with autotrophic nitrification (Krul, 1976; Rittmann and Langeland, 1985) and 2) heterotrophic nitrification and aerobic denitrification (Robertson et al., 1984a; 1984b). Favorable conditions for the occurrence of microzone denitrification were identified as: 1) a low DO concentration, 2) a low turbulence level, 3) a long solids retention time (SRT) or low organic loading, and 4) a short aeration cycle (Rittmann and Langeland, 1985; Bakti and Dick, 1992). Favorable conditions to the growth of aerobic denitrifiers and heterotrophic nitrifiers are: 1) a low DO concentration, 2) a frequently alternating aerobic and anoxic environment, 3) high substrate loading/a short SRT (vanNiel, 1991). Therefore, for achieving SND, a low DO concentration and a short intermittent aeration cycle are required, no matter what mechanism SND is based on. 1.3 Research Objectives According to a literature search, a few studies on two-stage IA processes have been done, but none on a two-stage process using SND. Therefore, a comprehensive study on a two-stage IA process, using SND, was considered. 5 The study described in this thesis was designed to investigate the performance of a two-stage, IA process employing SND, in comparison with that of a three-stage Bardenpho process. The two-stage process used consisted of an anaerobic selector followed by an intermittently-aerated, completely-mixed (IACM) reactor, in which the intermittent aeration was controlled by ORP range. Recognized, favorable conditions to SND were provided to the process. For example, by lowering and narrowing down the preset ORP control range, a low DO concentration range and a short aeration cycle were maintained in the IACM tank. Two external substrates, acetate and methanol, were added into the process to enhance SND. In the process studied, phosphorus release and selection of non-bulking sludge would occur in the anaerobic selector, whereas SND and phosphorus uptake (as well as organic carbon oxidation) would occur in the IACM tank. A bench-scale, steady-state, evaluation was conducted, under various ORP control ranges and external substrate dosages. Process dynamics, in response to ammonium and nitrate shock loads, was also studied under various conditions. The main objective of this study was to demonstrate the feasibility of achieving carbon (C), nitrogen (N), and phosphorus (P) removals from domestic sewage, in a two-stage IA process under conditions favorable to SND. The specific objectives of this research may be summarized as follows: 1. Steady-state process performance evaluation; 2. Evaluation of the possible SND hypotheses; 3. Investigation of the feasibility of using ORP as a process control parameter; 6 4. Investigation of the effect of external substrates (acetate and methanol) on process performance; 5. Study of the process dynamics under transient-state conditions; and 6. Study of the process kinetics under both steady-state and transient-state conditions. 7 Chapter Two LITERATURE REVIEW This chapter presents a review of previous work that has been used in developing the objectives and methods used in this research. The specific areas emphasized in this review include: 1) heterotrophic nitrification and aerobic denitrification, 2) biological nutrient removal (BNR) processes using simultaneous nitrification and denitrification (SND), 3) oxidation-reduction potential (ORP) control under low dissolved oxygen (DO) conditions, and 4) the role of external carbon substrates in BNR processes. 2.1 Biological Nitrification Nitrification was originally defined as the autotrophic conversion of ammonia nitrogen to nitrate nitrogen. In recent years, heterotrophic nitrification has become more accepted and better understood in the laboratory. Many heterotrophic organisms have been found to be able to nitrify organic and inorganic nitrogen compounds (Painter, 1970; Verstraete, 1975; Castignetti and Hollocher, 1984). A broader definition for nitrification is the biological conversion of inorganic and organic nitrogen compounds from one reduced state to a more oxidized state (Alexander et al., 1960; Robertson and Kuenen 1992). Reviews on heterotrophic nitrification, a relatively new concept to engineers, have been given by Verstraete (1975), Killham (1986), and van Niel (1991); the reader is referred to them for the additional information. The review presented here will be limited 8 to the aspects related to the physiological differences and ecological conditions between autotrophic and heterotrophic nitrifiers. The major recognized autotrophic nitrifiers are the genera Nitrosomonas and Nitrobacter. Nitrosomonas oxidizes ammonia to nitrite through hydroxylamine (NH 2OH); Nitrobacter oxidizes nitrite to nitrate in a single step. The free energies generated from ammonium oxidation and nitrite oxidation are -273 kJ/mole and -84 kJ/mole, respectively (US EPA, 1993). Carbon dioxide is the primary carbon source for the growth of these bacteria. The fixation of carbon dioxide costs the autotrophs about 80% of the energy generated (Kelly, 1978). Yield values observed in experimentation are 0.04-0.13 gVSS/g NH+ -N for Nitrosomonas and 0.02-0.07 gVSS/gNH^ -TV for Nitrobacter (Painter, 1970). The conditions required for the growth of autotrophic nitrifiers are: 1) aerobic conditions with available ammonia, 2) neutral and slightly alkaline pH, and 3) temperatures ranging from 5 to 40 °C (Focht and Verstraete, 1977). Verstraet (1975) and Killham (1986), in their reviews, recognized not only the existence of heterotrophic nitrification but also the surprisingly heterogeneous group of bacteria involved. Some of them oxidize the amino- and nitro- groups of organic nitrogen compounds, releasing nitrate or nitrite, whereas another group (overlapping) oxidizes ammonia to nitrite or nitrate. Robertson and Kuenen (1988) reported that Thiosphaera pantotropha, a heterotrophic nitrifier, is able to oxidize ammonia to nitrite via hydroxylamine, which resembles the pathway of autotrophic nitrification. 9 The first important distinction between autotrophic and heterotrophic nitrifiers is in their manner of energy generation. The energy for the growth of heterotrophic nitrifiers does not appear to be generated from the oxidation of ammonia by heterotrophs (e.g., Robertson et al., 1988; Castignetti et al., 1990). Verstraete (1975) found that heterotrophic nitrifiers were only able to oxidize reduced nitrogen compounds when an external organic energy source was present; this suggests that heterotrophic nitrifiers derive their energy for growth from the oxidation of organic substrate. There is no universal explanation for heterotrophic nitrification. Verstraete (1975) suggested that heterotrophic nitrification is mainly involved in the conversion of organic nitrogen compounds into compounds with a specific function, such as biocidal products and certain nitrogen compounds necessary for growth. However, relative high rates and large production of nitrite by heterotrophic nitrifiers can not be explained by the 'specific function' mechanism. Therefore, Robertson et al. (1988) suggested that in some heterotrophic nitrifying bacteria, the reaction serves as a dump for excess reducing power. The second major difference between these two types of nitrifiers is in their rates of nitrification. For a long time, the apparent nitrification rates by heterotrophic nitrifiers were considered to be very slow, compared with those of the autotrophs (Focht and Chang, 1975; Focht & Verstraete, 1977). However, this conclusion may result from measuring the heterotrophic nitrification activity, based on the nitrite or nitrate accumulation. This measuring method would underestimate the activity of heterotrophic nitrifiers, since many heterotrophic nitrifying bacteria can also denitrify simultaneously (Robertson et al., 1989a). van Niel (1991), using chemostaf cultures, equivalent growth 10 rates and similar media, found that the nitrification rates for Nitrosomonas europaea (autotrophic nitrifier) were 10 times those observed with Thiosphaera pantotropha (heterotrophic nitrifiers). However, heterotrophic nitrifiers usually grow much more rapidly than autotrophic nitrifiers (e.g., u m a x = 0.08 and 0.6 h"1 for Nitrosomonas europaea and Thiosphaera pantotropha, respectively) under relevant conditions, provided that organic substrates are available. Given the dominance of heterotrophs over autotrophs in most ecosystems, the rates reported by van Niel (1991) imply that heterotrophic nitrifiers could significantly contribute to nitrification in nature (Robertson and Kuenen, 1992). Between these two types of nitrifiers, heterotrophs generally tend to grow more rapidly with higher yield, require a lower DO concentration, and tolerate a more acidic environment than autotrophs. For example, in many cases, lower numbers of autotrophic nitrifiers were found in acidic soils than could account for the quantity of nitrification products accumulated. Therefore, additional nitrifying activity must originate from heterotrophs (Focht and Verstraete, 1977). van Niel (1991) used substrate-limited (ammonia) chemostat cultures of autotrophic and heterotrophic nitrifiers to show the competition between these two types of nitrifiers. The experimental results clearly showed that heterotrophic nitrification became more important as the ratio of carbon to nitrogen (C/N) increased. Similar chemostat experiments also revealed that once DO concentration had been allowed to fall to less than 40 uM, Nitrosomonas europaea was washed out, whereas Thiosphaera pantotropha remained unaffected. In summary, the favorable environmental conditions for heterotrophic nitrification include: 1) acidic environment (Focht and Verstraete, 1977), 2) low oxygen concentrations (Focht and Verstraete, 1977; 11 vanNiel, 1991), 3) high C/N ratios (Killham, 1986; vanNiel, 1991), and 4) short retention time (van Niel, 1991). 2.2 Biological Denitrification Biological denitrification involves the microbial reduction of nitrate to nitrite, and ultimately nitrite to gaseous forms of nitrogen (e.g., N 2 and N 2 0). A relatively broad range of bacteria can accomplish denitrification (Payne 1981). In general, denitrifiers are facultative and utilize the same basic biochemical pathways during both aerobic (O2) and anaerobic (NOx) respiration. However, by using nitrate or nitrite in place of oxygen in the electron transport chain, less energy is generated (US EPA, 1993). Denitrification is commonly thought to occur only in the absence of molecular oxygen (US EPA, 1993). The critical DO concentrations, above which denitrification ceases in dispersed cells, have been reported in a number of pure culture studies: 0.2 mg/L (Focht and Chang, 1975), 0.13 mg/L (Nelson and Knowles, 1978), and less than 0.1 mg/L (Krul, 1976). However, many bacteria, including Thiosphaera pantotropha, were found to be capable of simultaneously utilizing nitrates and oxygen as terminal electron acceptors in respiration; this phenomenon is termed as aerobic denitrification (e.g., Robertson and Kuenen, 1984a). Robertson and Kuenen (1984b) reviewed the controversy, surrounding aerobic denitrification. They found that the conclusion that denitrification only takes place under anaerobic conditions was based on studies of a limited number of species. Recent studies (Meiberg et al., 1980; Mechsner and Hamer, 1983; Robertson and Kuenen, 1984a; 1984b; Lloyd et al., 1987; van Niel, 1991; Richardson and Ferguson, 1992; Carter et al., 12 1995) have shown that a number of species from different genera are capable of aerobic denitrification. Lloyd et al. (1987) suggested that "aerobic denitrification is the rule rather than the exception". DO concentration has been considered a regulating factor for aerobic denitrification by most researchers. Many aerobic denitrifiers have different DO 'thresholds', above which aerobic denitrification begin to be inhibited (Robertson and Kuenen, 1990). As shown in Figure 2.1, the thresholds could be anywhere between 0 to 100% of air saturation, depending on the species of bacteria. However, aerobic denitrification tends to be slower than anaerobic denitrification, and all of the aerobic denitrifiers thus far studied increase their denitrification rate, as the DO concentration in the culture falls (Robertson and Kuenen, 1992). % air saturation 0 50 100 • ' ' • ' J l _ I — J I I t T Hypomicrobmm X Paracoccus denithficans fhiobacillus versutus Alcahgenes sp. (USA) Figure 2.1 The different sensitivities of the denitrification systems of various bacteria to the DO concentration (from Kuenen and Robertson, 1987). Thiosphaera pantotropha Alcahgenes sp. (Wageningen) "Pseudomonas denithficans" Alcahgenes faecalis S6 13 The denitrifying enzyme system in aerobic denitrifiers is generally considered to be constitutive. For example, the denitrifying enzymes are presented in Thiosphaera pantotropha growing aerobically even without nitrate (Robertson et al., 1984). Therefore, aerobic denitrifiers appear to have an ecological advantage in niches with fluctuating aerobic/anaerobic periods (van Niel. 1991; Richardson and Ferguson, 1992; Neef et al., 1996). The change from one electron acceptor to the other can be instantly made in aerobic denitrifiers because of their constitutive denitrifying enzyme systems. In contrast, anaerobic denitrifiers need at least two to three hours to switch over to another electron acceptor, because of the need to synthesize the enzyme involved (Knowles, 1982). Therefore, anaerobic denitrifiers are expected to lose their denitrification potential in an alternating aerobic and anoxic environment. While a reasonable amount of evidence has been accumulated for the aerobic denitrification in the last 10 years, its activity appears to be unstable (Kuenen and Robertson, 1994). Robertson et al. (1988) observed that Thiosphaera pantotropha synthesized poly-P-hydroxybutyrate (PHB) as a storage product under a number of growth conditions, even when the cultures appeared to be carbon limited in certain continuous cultures. The enzymes involved with PHB storage appear to be constitutive, since a rapid PHB synthesis by Thiosphaera pantotropha in the presence of excess acetate was observed (van Niel, 1991). Therefore, it was suggested (van Niel, 1991) that the PHB formation could result in the unstableness of aerobic denitrification by Thiosphaera pantotropha. 14 The mechanism, by which oxygen controls denitrification in many bacteria, is not yet fully understood and may vary among species of denitrifiers (Simpkin and Boyle, 1988; Robertson and Kuenen, 1992). The influence of oxygen on denitrification can be either repression of synthesis of the denitrifying enzyme system or inhibition of its activity. The repression mechanism represents the concept of anaerobic denitrification. It is generally held that the denitrifying enzymes in many anaerobic denitrifiers are inducible and oxygen is apparently a repressor of synthesis of these enzymes (Payne, 1981). The synthesis may either require only the absence of oxygen, or the presence of nitrate or nitrite (Payne, 1981; Knowles, 1982). Furthermore, Wimpenny (1969) suggested that the synthesis of denitrifying enzymes is regulated not by oxygen itself, but by the redox potential of the culture. Pure cultures of denitrifiers grown under aerobic conditions typically require 2-3 hours for the synthesis of the denitrifying enzymes to be complete, after being shifted to the anoxic conditions (Williams, 1978; Knowles, 1982). Once produced, the enzymes only gradually disappear, regardless of environmental conditions, and continue to function for "some time", after aerobic conditions have been established (Payne, 1981; Knowles, 1982). In other bacteria, the denitrifying enzymes appear to be constitutive (Robertson and Kuenen, 1984; Simpkin and Boyle, 1988); their activities can be inhibited by oxygen (Focht and Chang 1975; Stouthamer et al., 1980; Alefounder and Ferguson 1981). However, enzyme inhibition is not an on-off mechanism, and a reduced enzyme activity is possible in the presence of the inhibitor (Wild et al., 1992; von Schulthess, 1992). Two inhibiting mechanisms of oxygen are currently available. 15 The first one is based on competition between oxygen and nitrate for the electrons in the electron transport chain. The redox levels of the cytochromes appear to control the direction of electron flow (Kucera and Dadak, 1983). Generally, electron transport to oxygen is so efficient that the cytochromes do not get reduced enough to pass the electrons to the denitrification enzymes. However, as is sometimes the case, there is a bottleneck in the electron transport chain between cytochrome c and oxygen, resulting in an accumulation of NAD(P)H (Robertson et al., 1988). The redox potential of the electron transport chain would therefore become more reduced, permitting the flow of electrons to the constitutive denitrification enzyme system (Robertson et al. 1988). The second inhibiting mechanism is based on the existence of a permeability barrier between the denitrifying enzyme and nitrates. Since the nitrate reductase is generally located on the inside of the cell membrane (Stouthamer et al., 1980), a permeability barrier between the enzyme and its substrate would be a very effective controlling factor. It has been found that oxygen actually altered the membrane, preventing nitrate uptake by cells (Alefounder and Ferguson, 1980; Hernandez and Rowe, 1988). Many organisms, such as Thiosphaera pantotropha, can express two distinct respiratory nitrate reductases (Bell et al., 1990). One, expressed only under anaerobic conditions, is membrane-bound and inactive in intact cells in the presence of oxygen. The second, nitrate reductase, is isolated in the periplasmic compartment and can be expressed under both anaerobic and aerobic conditions. It is this enzyme that provides Thiosphaera pantotropha with the biochemical mechanisms required for aerobic denitrification (Bell et al., 1990). 16 The bacteria possessing the periplasmic nitrate reductase were identified in a recent study (Carter et al., 1995). Twenty-nine strains, isolated from three soils and a freshwater sediment, were shown to comprise members of three genera (Pseudomonas, Aeromonas, and Moraxella). All of these strains expressed a nitrate reductase with an active site located in the periplasmic compartment. Twenty-two of the strains showed significant rates of nitrate respiration in the presence of oxygen, when assayed with physiological electron donors. Whether oxygen inhibits or represses denitrifying enzymes is of importance in the design of BNR processes. Simpkin and Boyle (1988) gave a general view for the behavior of the denitrifying enzymes under alternating anoxic/aerobic conditions in BNR processes. The synthesis of the denitrifying enzymes occurs even during the aerobic cycles of activated sludge and the decay of the enzymes is offset by the synthesis. Therefore, the denitrifying enzymes do not go through significant cycles of synthesis and decay in BNR processes. Further, inhibition of enzyme activity, not repression of enzyme synthesis, must be the most important effect of oxygen on denitrification. Therefore, it is not necessary to allow time for the enzyme synthesis in designing anoxic zones, and it is possible to use processes that have rapid changes in DO concentration. 2.3 Simultaneous Nitrification and Denitrification (SND) As mentioned in Sections 2.1, Thiosphaera pantotropha is also a heterotrophic nitrifier (Robertson and Kuenen, 1988). van Niel (1991), in his literature review, 17 suggested that all the aerobic denitrifiers studied are able to perfrom heterotrophic nitrification; therefore, they.can convert ammonia directly into gaseous products. Robertson and Kuenen (1989; 1990) postulated a working hypothesis to explain the aerobic denitrification and heterotrophic nitrification, as well as PHB formation by Thiosphaera Pantotropha. It is based on the existence of a potential limitation in the respiratory chain. P H B •< Denitrification i 0 2 Figure 2.2 Scheme showing the various possible options for NAD(P)H utilization available to Thiosphaera pantotropha (Robertson et al., 1988). The details shown in Figure 2.2 are: 1) the electron transfer between cytochrome aa3 and oxygen is rate-limiting, as a consequence of which, the reduced electron flow (from organic substrates) would cause a lower redox potential of the cytochromes; 2) if nitrate or nitrite is available, electrons would be able to flow to the denitrification pathway and these compounds could serve as electron donors immediately; and 3) if nitrate and nitrite are both absent, the reducing power would be accumulated. As NAD(H) is required 18 for heterotrophic nitrification, it could be re-oxidized either by heterotrophic nitrification, or in the production of PHB if sufficient organic substrate is available (Robertson and Kuenen, 1989). Both heterotrophic nitrification and PHB synthesis serve as a dump for excess reducing power. Based on this working model, accumulations of nitrite and nitrate by heterotrophic nitrification are impossible. It was indeed observed (Robertson 1988) that ammonia oxidation and nitrite reduction in aerobic cultures of Thiosphaera pantotropha were usually so well balanced that nitrite was not accumulated. Several strains of autotrophic nitrifiers, such as Nitrosomonas and Nitrobacter, are capable of denitrification (Steinmueller and Bock, 1977; Sundermeyer-Klinger et al., 1984; Poth and Focht, 1985; Poth, 1986; Bock et al., 1988). However, van Niel (1991) noted that denitrification by these bacteria, at high DO concentrations, has not yet been described in the literature. Nitrobacter is capable of anoxic growth, reducing nitrate to nitrite with organic substrates as electron donors (Steinmueller and Bock, 1977). There is strong evidence (Sundermeyer-Klinger et al., 1984) that nitrate reduction is using the same enzyme system that oxidizes nitrite to nitrate. Bock et al. (1988) found that the observed cell yield of Nitrobacter was surprisingly high under anoxic conditions, provided that the nitrite concentration remained limited, and suggested that Nitrobacter belongs to the group of organisms favored by alternating aerobic/anoxic conditions. Nitrosomonas europaea is able to produce nitrous oxide and nitrogen gas at low oxygen concentrations (Poth and Focht, 1985; Poth, 1986; Downes, 1988). Recently, Bock et al. (1995), in pure culture of Nitrosomonas species, showed that cells of 19 Nitrosomonas europaea and Nitrosomonas eutropha were able to nitrify and denitrify at the same time when grown under oxygen limiting conditions (0.2-0.4 mg/L). In mixed cultures of Nitrosomonas europaea with various heterotrophs (including both anaerobic and aerobic denitrifiers), growing under oxygen limiting conditions, the amount of nitrogen loss was strongly affected by the partners. For example, Nitrosomonas europaea, mixed with Thiosphaera pantotropha, could convert almost 100% of ammonia present to gaseous forms of nitrogen. Broda (1977) (cited in Kuenen and Robertson (1994)) noted that the energetics of ammonium oxidation with N O x rather than oxygen is theoretically favorable. A mixed culture converting ammonia to nitrogen gas has been obtained in an anaerobic wastewater treatment system in recent studies (van der Graaf et al., 1990; 1991). Nitrogen balances revealed that ammonia was disappearing from the system; the ammonia disappearing rates was proven to be,proportional to the amount of biomass present, and to be dependent on the presence of nitrate or nitrite. If nitrate became depleted, both the ammonia disappearance and gas production stopped, and only resumed when more nitrate or nitrite was added. Therefore, it was concluded that ammonia oxidation under anoxic conditions is indeed a form of biological, ammonia-dependent denitrification. Further, it was reported (Kuenen and Robertson, 1994) that enrichment cultures obtained from this system gave rates of ammonia oxidation to nitrogen gas comparable to the nitrification rates obtained with conventional nitrification systems producing nitrate. 20 2.4 Kinetics of Nitrification and Denitrification Both nitrification and denitrification are generally considered to be a two-step reaction; nitrite is the product of the intermediate step of both reactions. The intermediate steps have usually been found not to be rate-limiting (Betlach and Tiedje 1981; Timmermann and Van Haute, 1983; Turk, 1986); therefore, the overall kinetics of the reaction may be considered as a single step for both reactions. Both reactions essentially follow the zero-order kinetics until a low substrate concentration condition is encountered, such as ammonia concentration between 0-5 mgN/L for nitrification and nitrate concentration between 0-2 mgN/L for denitrification (Huang and Hopson, 1974; US EPA, 1975; Beccari et al., 1983). Within the low substrate concentration ranges, Monod kinetics has been applied to describe both nitrification and denitrification kinetics (Downing et al., 1964; Stensel et al., 1973). A number of environmental factors, such as temperature, DO concentration, pH, organic carbon loading and inhibitors have been found to affect the nitrification rate (US EPA, 1993). When accounting for the effects of above-mentioned environmental factors, a number types of kinetics have been employed in the nitrification modeling. The most widely used is the double-substrate limiting kinetics, which incorporates both ammonia and DO effects, as shown in the following (Prosser, 1990): 21 where juA and /Jm are specific and maximum specific growth rate of nitrifiers; SNH4 and So are ammonia and DO concentrations; KM4 and K0 are half saturation constants for ammonia and DO, respectively; bA is the decay constant for nitrifiers. In Equation 2.1, the recommended K0 values are in a range of 0.45 to 1.3 mg/L (US EPA, 1975; IAWPRC, 1986; Stenstrom and Song, 1991). Denitrification kinetics in the anoxic reactors of single-sludge BNR processes depends on the nature and amount of organic substrate in the sewage. For example, Barnard (1973) found three denitrification rates by spiking the mixed liquor from the first and second anoxic zones of the Bardenpho pilot plant, with nitrates. He attributed the first fast rate to the "byproducts of fermentation", the second to adsorbed COD and the third slower rate to endogenous respiration. Similarly three rates were also observed by Ekama et al. (1984); Zero-order, first order, and Monod kinetics have all been suggested to describe the denitrification rate (US EPA, 1993). Monod-type kinetics has been adopted by most of the investigators. Since DO may also inhibit denitrification, a triple-substrate limiting kinetic expression was used by the IAWPRC task group (1986), as shown in Equation 2.2, to account for the effects of organic carbon, nitrate, and DO on denitrification. /C< = / C (-^rirX K S n \ X T ^ T - ) ? , - K (2.2) 22 where SNOx is nitrate concentration, and % is the denitrifying heterotrophs correction factor, K0 and KNOx are half-saturation constants for 02 and nitrates, respectively, and bH is the decay constant for heterotrophs. In Equation 2.2, organic substrate Ss is referred only to the readily biodegradable fraction of available total COD (RBCOD). The slowly biodegradable COD (SBCOD), mainly in particulate form, must be hydrolyzed and transformed into RBCOD, prior to be used by the organisms. 2.5 Biological Excess Phosphorus Removal Comprehensive literature reviews on biological phosphorus removal (BPR) have been presented in US EPA (1987) and by Comeau (1989) and Toerien (1990). The generally accepted theory for BPR is that anaerobic-aerobic contacting results in a competitive substrate utilization and selection of phosphorus accumulating organisms (PAOs). Under anaerobic conditions, acetate and other fermentation end-products are produced from fermentation reactions by normally-occurring facultative organisms in the anaerobic zone or in a separate fermentor. The PAOs take up these fermentation products and store them as poly-|3-hydroxyalkanotes (PHA), using stored polyphosphate as an energy source. However, the success of this activity is dependent on the outcome of competition with a host of organisms, particularly denitrifiers, which can consume fermentation products also under the same anaerobic conditions. Under subsequent aerobic conditions, the PAOs can use the stored PHA in the presence of oxygen to generate energy for growth and uptake of polyphosphate. The growth of PAOs is not 23 dependent on external organic carbon, which is usually limiting under aerobic conditions. In this way, the PAOs compete effectively with other aerobes in the aerobic stages of BPR processes. Organisms of the genus Acinetobacter have often been found in significant concentrations in BPR processes, and have been shown to be capable of accumulating excess phosphorus (Lotter, 1985; Wentzel et al., 1991). Other bacterial groups, such as Pseudomonas (Suresh et al., 1985), Aeromonas (Brodisch and Joyner, 1983), Klebsiella (Gersberg and Allen, 1985), Micrococcus (Ye et al., 1988) Arthrobacter (Shoda et al., 1980), Moraxella and Enterobacter (Lotter and Murphy, 1985) are also capable of enhancing phosphorus removal. Brodisch (1985) suggested that excess phosphorus removal is dependent on the correct balance between different bacteria types. In many BNR processes, a net phosphorus uptake has been reported under anoxic conditions in both batch and continuous tests (e.g. Malnou et al., 1984; Vlekke at al., 1988); this indicates that nitrate can serve as an electron acceptor for the oxidation of stored carbon (e.g., PHA) and the uptake of phosphorus. However, due to lower efficiency of energy production from anoxic oxidation of PHA in comparison to aerobic respiration, the phosphate uptake rates under anoxic conditions are usually lower than those under aerobic conditions (Comeau et al., 1987; Wentzel et al., 1992). This was also explained by the presence of two groups of bacteria, one that could use both nitrate and oxygen as oxidants and another that could only use oxygen (Comeau et al., 1987; Gerber et al., 1987; Kerrn-Jespersen and Henze, 1993). 24 The capacity of denitrifying bacteria to store phosphorus under anoxic conditions has further been.studied in several recent studies (Kuba et al., 1993; Kuba et al., 1996; Sorm et al., 1996; Kuba et al., 1997). It has been demonstrated in these studies that phosphorus removal by the denitrifying bacteria has capacities and characteristics similar to phosphorus removal in anaerobic-aerobic processes. The main advantages of incorporating phosphorus uptake under anoxic conditions into BNR processes include the saving of energy and COD, and a lower sludge production. Phosphates release in BPR systems under anoxic conditions, in response to the additions of preferred substrates (e.g., acetate and propionate), has been documented (Wentzell et al., 1984; Comeau et al., 1987; Mostert et al., 1988). Mostert et al. (1988) speculated that simultaneous release and uptake of phosphorus occurs in the presence of acetate in the anoxic phase. Comeau et al. (1987) suggested that the PAOs, that are capable of denitrification, take up phosphates in the presence of N O x instead of DO, whereas other PAOs are expected to release phosphates, provided that preferred substrates are available. The observed phosphorus concentration in the anoxic zone, therefore, is the net result of the opposing phosphorus release and uptake reactions. A study done by Wentzell et al. (1984) further showed that the phosphorus release reaction occurred even in the aerobic zone, when a preferred substrate was present in significant concentration, and continued until the preferred substrate was consumed, where upon the phosphorus uptake commenced. Barnard (1984) noticed that a consistent release of phosphates took place in the second anoxic zone of a five-stage Bardenpho process, to which no nitrates were 25 discharged, and this release disappeared when the second anoxic zone was aerated. He called this 'secondary release', the phosphorus release not associated with acetate uptake. 2.6 Single-Sludge B N R Process 2.6.1 Mainstream BNR Process Since the early 1960's, significant developments have taken place in the activated sludge process for treating wastewater. The function of activated sludge processes has been expanded from carbonaceous energy removal to include progressively nitrification, denitrification by Ludzack and Ettinger (1962), and phosphorus removal by Barnard (1974). Single-sludge processes, using internal carbon sources, have become dominant in BNR processes since that time. Single-sludge systems, utilizing two anoxic zones, such as the five-stage Bardenpho process, are capable of achieving effluent total nitrogen (TN) concentration less than 3 mgN/L and total phosphorus (TP) concentration less than 1 mgP/L; single-sludge systems utilizing one anoxic zone are capable of accomplishing T N residuals of 6 to 12 mgN/L and TP concentrations of 0.5 to 4.6 mgP/L (US EPA, 1993). A more recent assessment conducted by Mines (1996) for eight full-scale plants indicated that the five-stage Bardenpho process with sand filters and the A 2 /0 with deep-bed sand filters were equally capable of meeting 5 mg/L CBOD 5 , 5 mg/L TSS, 3 mgN/L TN, and 1 mgP/L TP effluent limitations. 26 2.6.2 Intermittent Aeration (IA) BNR Process The IA process can be an alternative to the mainstream, single-sludge BNR processes. Many researchers (e.g., Lin and Tsang, 1989; Heduit, 1990; Nakanishi et al., 1990; Huang, 1993; Lefevre et al., 1993; US EPA, 1993) reported that the IA process was effective in T N reduction and achieved a range of 70 to 90% removal efficiency. However, deterioration of sludge settleability was also reported and was attributed to excessive growth of filamentous organisms, caused by low food/microorganisms (F/M) (e.g., Landine, 1989; Lo et al., 1994; Huang, 1993). It has also been suggested that incorporating an anaerobic selector ahead of the intermittent aeration tank can improve sludge settleability (e.g., Landine, 1989; Lo et al., 1994) and enhance phosphorus removal in the process (e.g., Charpentier et al., 1987; Charpentier et al, 1989; de la Menardiere et al., 1991; Lo etal., 1994). The IA processes studied so far were controlled either by a timer, DO concentration, or ORP. The cycle times used usually ranged from 0.5 to 3 hours. However, the recent studies (Huang, 1993; Waki et al., 1980a) called for the use of a low DO concentration during the aerobic cycle and the use of a short on/off cycle. The reasons for using a low DO in the aerobic cycle are to reduce the adverse effect of high DO on the denitrification in the subsequent anoxic cycle (Landine, 1989; Sen et al., 1990) and to reduce the cost for aeration (Huang, 1993). Generally, the use of a short aeration cycle can avoid anaerobic conditions. Besides that, Waki et al. (1980a; 1980b) suggested that a short cycle could increase the denitrification rate in the anoxic cycle and the best cycle time appeared to be in the range of 5 to 20 minutes. 27. 2.6.3 BNR Process Using SND In studying nitrogen removal in a bench-scale, unaerated-aerated system, Ludzack and Ettinger (1962) proposed to use an under-aerated first zone, instead of unaerated, without mixed liquor recycle from the second aerated zone, in order to promote SND. They emphasized the needs of high organic loading and low DO conditions in the under-aerated first zone. For a decade after this publication, SND in low DO systems has been considered as a means of nitrogen removal. However, most early SND processes were a compromise between nitrification and denitrification. The processes might not be truly simultaneous but could remove a certain amount of nitrogen in a single reactor (Matsche, 1972; Drews and Greeff, 1973; Matsche and Spatzierer, 1975; van der Geest and Witvoet, 1977). The keys to achieve nitrogen removal in these processes were: 1) to provide sufficient tank volume for both nitrification and denitrification, 2) to supply sufficient oxygen for nitrification, and 3) to create anoxic zones within the tank or anoxic periods during the aeration cycle for denitrification. By this definition, channel systems and IA systems are all considered as a SND process. A stricter definition of SND would be that nitrification and denitrification occur concurrently in a single reactor, without forming any distinct anoxic zones (e.g., in a channel system) or anoxic periods (e.g., in an IA process). Nitrification has always been believed to be a strictly aerobic process, even for heterotrophic nitrification, because of the requirement of ammonia monooxygenase for molecular oxygen (Hollocher et al., 1981). Therefore, the occurrence of denitrification in the presence of oxygen is crucial to achieve SND in a nitrogen removal process. 28 A net nitrogen loss has been reported frequently in the aeration tanks of many full-scale BNR processes (US EPA, 1987; van Huyssteen et al., 1990; Moriyama et al., 1993; Rabinowitz and Barnard, 1995). This loss could vary from 20% of the influent total nitrogen to all the nitrogen not captured in the sludge (Randall et al., 1992). Denitrification in the anoxic microzones inside of the floes and biofilms due to diffusion limitation has often been suggested to explain the observed net nitrogen losses in activated sludge processes (Rittmann and Langeland, 1985; Moriyama et al., 1993; Munch et al., 1996) and in trickling filters (Strand et al., 1985; Watanabe et al., 1992; Halling-S(|>rensen and J<|)rgensen, 1993). Evidence for anoxic microzones was the profiles of DO concentration using microelectrodes in microbial aggregates (Huang and Bungay, 1973; Chen and Bungay, 1981). This explanation is based on the traditional concepts that nitrification and denitrification are mutually exclusive. The favorable conditions for achieving the microzone denitrification were identified as a low DO concentration, a long SRT/low organic loading, and a short aeration cycle time (Rittmann and Langeland, 1985; Sen et al., 1990; Munch et al., 1996), as well as a low turbulence level (Bakti and Dick, 1992; Beccari et al., 1992). For example, under long SRTs (44 and 164 days) and low organic loading conditions, denitrification was accomplished in oxidation ditches without forming distinct anoxic zones (Rittmann and Langeland 1985). In this system, DO concentrations were maintained between 0.1-0.5 mg/L and aeration cycle times between 10-30 minutes. Nitrogen removal between 75% to 97% was recorded. In an IA process, Nakanishi et al. (1990) achieved high nitrogen removal (80-90%) by using a DO concentration range of 0-0.3 mg/L and short aeration 29 cycle times between 5-20 min. The SRTs in their system were in a range of 40 to 50 days. It was suggested that SND must have occurred under the above conditions. In the above two studies, the low DO conditions were believed to be necessary for forming anoxic microzones inside of the floes. The short aeration cycle was necessary for controlling DO concentration in the low ranges. The long SRT and the low organic loading made nitrification possible under such low DO conditions. However, both the nitrification and denitrification rates were likely depressed in the above systems, due to the low DO conditions. In contrast to the traditional concepts of nitrification as an aerobic autotrophic process and denitrification as purely anoxic, microbiologists have reported the existence of aerobic denitrifiers as well as heterotrophic nitrifiers and postulated the concepts of aerobic denitrification and heterotrophic nitrification (Sections 2.1, 2.2 and 2.3). While aerobic denitrification and heterotrophic nitrification have been studied extensively in pure culture by microbiologists, only a few engineering applications in wastewater treatment can be found in the literature. Kshirsagar et al. (1995) mixed Thiosphaera pantotropha (aerobic denitrifiers) with activated sludge to treat synthetic wastewater containing high nitrate. Significant denitrification (75-85%) occurred at bulk DO concentrations above 2.5 mg/L. It was suggested that the major mechanism for nitrate removal in their system was aerobic conversion. Gupta et al. (1994) studied a rotating biological contactor (RBC) system, inoculated with Thiosphaera pantotropha. The results from their study proved the 30 potential of Thiosphaera pantotropha to aerobically denitrify and the ability to coexist with autotrophic nitrifiers. A study by Bang et al. (1995) presented an example of the natural selection for aerobic denitrifiers in an RBC system, treating polyvinyl alcohol (PVA) wastewater. A significant nitrogen loss was observed at high DO concentrations above 5 mg/L. The PVA-decomposing bacteria, nitrifiers, and denitrifiers were found to co-exist in the biofilm. Further, the population of denitrifiers in the surface layer was 1 to 2 orders of magnitude higher than that in the middle and bottom layers; this indicated that the surface layer had a higher denitrifying activity. In their chemostat experiments, the biofilm was homogenized and suspended in a reactor with very strong agitation (stirrer speed > 600 rpm). The DO concentration of the reactor was maintained between 3 to 6 mg/L. The overall nitrogen removal efficiency in this system was 45% at a C/N ratio of 1.8 (C as TOC), and 84% at a C/N ratio of 3.6. Therefore, it was concluded that the nitrogen loss in the biofilm was due to aerobic denitrification, as proposed by Robertson and Kuenen (1984) and the intermediates produced through P V A degradation, such as ketone, carboxy, and methyl radical, may be the preferred substrates for aerobic denitrification. Kugelman et al. (1988; 1991) also claimed that aerobic denitrification may be a significant mechanism to account for the aeration tank nitrogen losses they observed in their laboratory and full-scale anaerobic-aerobic (A/O) systems. In the laboratory tests with the sludge taken from an A/O process, they showed inhibition of oxygen respiration by nitrite addition and nitrogen removal through nitrite to gaseous forms of nitrogen, at a DO concentration of 4 mg/L or above. On the other hand, they showed no nitrite 31 inhibition and no nitrogen loss, in the tests with the sludge obtained from strictly aerobic systems! Based on these results, inhibition of oxygen respiration by nitrite was suggested as a mechanism for aerobic denitrification. The anaerobic zone of the A/O system appeared to serve as a selector, in which denitrifying enzymes were synthesized and oxidase inhibitors were produced; upon entering the aerobic zone, the bacteria, having their oxygen respiration blocked, were forced to use N O x as an electron acceptor, even under highly aerobic conditions. Nitrogen losses in an aeration tank have been attributed to the production of nitrogenous gases, such as N 2 and N 2 0 , during the nitrification process (Wood et al., 1981; Poth, 1986; Bock et al. 1995). In general, nitrogen losses have often been observed to a much greater extent when nitrification occurs than when it does not, even when no opportunity exists for anaerobic denitrification (Painter, 1970). Wood et al. (1981), in a study on inhibition of nitrification, ran a series batch tests with nitrifying sludge under fully aerobic conditions (70% air saturation). They observed up to 43% inorganic nitrogen loss and significant nitrite accumulation. Moreover, the inorganic nitrogen loss appeared to be proportional to the extent of nitrification. Therefore, the nitrogen loss was proposed to be due to the reduction of nitrite to nitrous oxide ( N 2 O ) by a hydroxylamine-nitrite reductase in Nitrosomonas when sufficient nitrite was present. The reductase contains copper and would be inhibited by compounds such as thiourea which inhibit nitrification, this explains the finding that the losses were far less when an inhibitor was present. However, Zheng et al. (1994), in studying nitrous oxide gas ( N 2 O ) production during the nitrification process in activated sludge, found that high amounts of nitrous 32 oxide production were accompanied by incomplete nitrification. Low DO concentration and short SRT, the two conditions causing incomplete nitrification, were the two factors controlling the nitrous oxide gas production. The maximum N 2 0 production under the worst conditions (0.2 mg/L DO and 3 day SRT) only accounted for 16% of nitrified nitrogen. Compared to the findings from Wood et al. (1981), the results from Zheng et al. (1994) appeared to be more consistent with those obtained from pure culture studies (Section 2.3). , 2.7 ORP Control Under Low DO Conditions Redox potential is very easily defined for a simple oxidation-reduction reaction consisting of a pair of components, in which one (Ox) can be reduced reversibly to the other (RD) by means of moving n electrons (e): Ox + ne'oRd (2.3) and its value Eh for this equilibrium can be calculated using the Nernst equation: = — l n ^ (2.4) * nF (Rd) • • ( where Eh is the potential referred to the normal hydrogen electrode, E? is the standard potential of the system at 25 °C when the activities of all reactants are unity, R is the gas constant, T is the absolute temperature, F is the Faraday constant and n is the number of electrons transferred in the reaction. 33 The ORP of a solution depends on the oxidation and reduction agents present in it. 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 reference electrode. The accuracy of the measured value is influenced by the formation of surface oxide and sulfide coatings on the platinum electrode (Heduit and Thevenot, 1992). The complex and unspecified nature of ORP measurements has been one of the main obstacles to the wider use of ORP as a control parameter in biological wastewater treatment. However, the ORP generally shows a strong response to DO, even at low concentrations. It is commonly accepted that ORP is valuable as an indirect measurement of DO at concentrations that can not be measured directly with DO probes. Kjaergaard (1977), in reviewing several investigations into the use of ORP in fermentation processes, noted that the dependence of the redox potential on the DO concentration was significant, although the dependence could not be simply formulated using the Nernst equation. A linear relationship between DO and ORP was reported in activated sludge systems (Heduit and Thevenot, 1989; Heduit and Thevenot, 1992; Lie and Welander, 1994). The constants in the Nernst equation depended upon the sludge loading, the aeration conditions, the sludge level and other redox species (Heduit and Thevenot, 1989), as well as the make-up of the microbial population, floe characteristics, and carbon sources (Lie and Welander, 1994). From the Nernst equation, the advantages of using ORP probe instead of DO probe for monitoring and controlling under a low DO environment include. 1) a significant changes in redox potential represent very small changes in DO concentration; 2) due to the 34 availability of negative redox potentials, the useful operating range of the redox probe is much larger than the DO probe. ORP behavior in the anoxic and anaerobic states was well explained by Koch and Oldham (1985), Charpentier et al. (1989), and Wareham (1992). The results from these studies showed that a rapid drop in ORP occurred when nitrate was depleted. Furthermore, the results from Lie and Welander (1994) suggested that the denitrification activity of activated sludge seemed to be linearly related to the ORP level. Absolute ORP measurements have been used to control the aeration in wastewater treatment processes when operation at low DO concentrations has been required (Charpentier et al., 1987; Charpentier et al., 1989; Nakanishi et al., 1990). Charpentier et al. (1987) discussed both laboratory and full scale applications of ORP control in France. In a low loaded activated sludge plant, various N H / and M V effluent concentrations were recorded along with the attendant variations in ORP. Subsequently, ORP values of 80 to 120 mV were targeted and air to the aeration basin was cycled on and off at a rate just sufficient to keep the ORP between these limits. In this way, consistent effluent nitrogen levels were maintained. Charpentier et al. (1989) furthered this work by investigating relationships between effluent T N concentration and ORP. They found that the effluent quality and electrical power costs were simultaneously optimized by targeting upper and lower ORP values in the aeration cycle. Nakanishi et al. (1990) compared ORP control to DO control in an intermittent aerated, complete mixing-type reactor for nitrogen removal. They concluded that ORP was a better control parameter than DO in optimizing nitrogen removal because ORP 35 responded to the feed characteristics and the bacterial population as well as DO. Moriyama et al. (1993) also studied on-off ORP control in an "oxic-anoxic-oxic" sequence full-scale plant. An ORP range of 118-150 mV was used to control the aeration in the first oxic tank. It was observed that the denitrification reaction mainly (70% of overall nitrogen removal) took place in the first oxic zone. Similar on-off ORP control for aeration was also applied to the two-stage IA process (an anaerobic selector followed by an extended aeration treatment system), designed for both nitrogen and phosphorus removals (de la Menardiere et al., 1991; Lo et al., 1994). It was observed that the removal levels for nitrogen and phosphorus depended on the ORP control range. For example, Lo et al. (1994) observed that T N removal of 93% was attained at an optimum ORP set points of 110 mV and excellent phosphorus removal was achieved over a range of 70-180 mV, at a SRT of 20 days. This indicates that optimization for both nitrogen and phosphorus removal is possible. In addition to aeration control, the ORP measurements were used in the control of external substrate addition for denitrification in wastewater treatment. Watanabe et al. (1985), in a series of lab experiments, used an ORP set point of approximately -150 mV as an indicator of zero nitrate condition, to control the addition of an external carbon source (methanol) in order to ensure complete denitrification. As the biomass exhausted the carbon supply, the ORP rose above the set point and initiated methanol addition. In this way, ORP became a control index for methanol regulation and allowed consistent effluent N O x level of less than 1 mg/L. 36 2.8 The Role of Organic Substrates in BNR Processes Both denitrifiers and PAOs are heterotrophs and require organic matter for their metabolism. It was estimated that the COD consumption in single-sludge BNR processes is 8.6 gCOD/gN and 50 to 59 gCOD/gP removed from municipal wastewater (Ekama and Marais, 1984). 2.8.1 The Role of Organic Substrate in Biological Phosphorus Removal The concentration and nature of soluble COD provided to the anaerobic zone has been shown to strongly affect the degree of BPR (Nicholls and Osborn; 1979; Manoharan, 1988). Increasing the concentration of fermentation products such as VFAs in the feed to the anaerobic zone generally increases the degree of BPR (Oldham and Stevens, 1985; Rabinowitz, 1985; Lotter and Pitman, 1992). For substrates other than VFAs, little work has been done, and the results appear to be controversial. For example, Tarn et al. (1992) found that methanol addition appeared not to enhance phosphorus release. However, Jones et al. (1987), working with continuous laboratory-scale anaerobic-aerobic sequence systems with acclimated biomass, observed that alcohol, such as methanol and ethanol, added into the anaerobic reactor, enhanced phosphorus release and subsequent phosphorus uptake. 2.8.2 The Role of Organic Substrate in Biological Nitrogen Removal When wastewater lacking a suitable source of organic carbon and/or with a low COD/N ratio is treated for biological nitrogen removal, carbon from an external source must be added to achieve an acceptable removal level. Randall et al. (1992) suggested the 37 use of an additional carbon source for improving nitrogen removal when COD/TKN ratio was less than 9. The most widely used carbon sources are acetate and methanol. Generally both acetate and methanol are effective carbon sources for denitrification, when they are present in the anoxic zone. However, acetate was reported to give higher denitrification rates than methanol (Gerber et al., 1986; Carley and Mavinic, 1991; Tam et al., 1992). An immediate increase in denitrification rate was observed as an immediate response to acetate addition (Tam et al., 1992; Hallin et al., 1996; Isaacs et al., 1994). However, a period of adaptation to acetate was still needed for denitrifying bacteria (Hallin et al., 1996). Most reports concerning methanol addition have dealt with post-denitrification processes, in which methanol is used as sole carbon and energy source. Both lab and field experiments demonstrated a lag period before an increase in both nitrogen removal and denitrification rates (McCarty et al. 1969; Nyberg et al., 1992; Hallin et al. 1996). Very little information is available concerning methanol addition into pre-denitrification processes, in which sewage is used as the main carbon and energy source. Hallin et al. (1996), in studying the adaptation of activated sludge to methanol, suggested that changes in species composition or induction of enzymes were necessary for denitrifying bacteria to use methanol. They further suggested that activated sludge, after adaptation to methanol, appeared to consist of two groups of denitrifying bacteria, one denitrifying with methanol and the other with the substrate present in sewage. 38 Generally, in the presence of a high concentration of organic carbon, the heterotrophs may slow down the activities of nitrifiers in activated sludge systems by competing for oxygen and nutrients (Prakasam and Loehr, 1972; US EPA, 1975). Hanaki et al. (1990a; 1990b) demonstrated that the presence of biodegradable organic substrates inhibited ammonia oxidation, but did not inhibit nitrite oxidation. This inhibitory effect on ammonia oxidizers was attributed to competition from heterotrophs for ammonia. The effect of specific organic compounds on nitrification is not clear. Cooper and Catchpole (1973) found that the addition of pyruvic acid and glucose shortened the minimum retention time required for coke-oven liquor to achieve nitrification in activated sludge systems. Kiff (1972), using activated sludge in batch tests, observed an inhibiting effect of added acetate at a low DO concentration. 2.8.3 The Role of Organic Substrate in SND Process Ottow and Fabig (1983) suggested that the amount of available organic matter rather than the oxygen diffusion rate should be considered as the triggering factor for denitrification in natural systems. The higher the demand for electron acceptors, the greater are the chances for denitrification to occur even under aerobic conditions. Richardson and Ferguson (1992) studied the influence of carbon substrate on the activity of the periplasmic nitrate reductase in aerobically grown Thiosphaera pantotropha. It was found that the activity of periplasmic enzyme increased with the more reduced carbon substrate. This suggests that the nature of the organic substrates used, rather than the presence of nitrate or oxygen, is more important in regulating aerobic denitrification. 39 As mentioned in Section 2.2, Bang et al. (1995) found that the selection of aerobic denitrifiers appeared to be associated with some special organic compounds, which were produced through the decomposition of polyvinyl alcohol. Acetate is considered to be effective in promoting both aerobic denitrification and heterotrophic nitrification (Robertson and Kuenen, 1988; 1990; van Niel, 1991). A few of facultative methylotrophs, such as Hypomicrobium, Pseudomonas, and Paracoccus species, have been isolated as a predominant part of the denitrifying flora in post-denitrification systems, using methanol as sole carbon and energy source (Nurse, 1980; Payne, 1981; Timmermans and van Haute, 1983; Claus and Kutzner, 1985; Neef et al, 1996). It has been reported that Hypomicrobium (Wilkinson and Hamer, 1972; Wilkinson and Harrison, 1973; Mechsner and Hamer, 1983), Thiosphaera pantotropha (= Paracoccus denitrificans) and Pseudomonas denitrificans (Robertson and Kuenen, 1987; 1990; van Niel, 1991) were capable of aerobic denitrification. The selection of the above aerobic denitrifiers in post-denitrification systems suggests that methanol is a preferred substrate for aerobic denitrifiers. Many bacteria are known to synthesize PHB as a carbon and energy reserve material under certain environmental conditions, such as the limitation of nitrogen, oxygen, or phosphate (Dawes and Senior 1973). The synthesis of PHB involves the condensation of two molecules of acetyl CoA and a reduction with NADH. This synthesis is unique among energy storage compounds because the direct participation of ATP is not required. However, reducing power in the form of N A D H is essential and PHB formation may be regarded as a quasi-fermentation process permitting the reoxidation of N A D H into NAD + . 40 Such a process is particularly useful under conditions of oxygen limitation, which prevent the re-oxidation of N A D H by the electron transport chain, or under nitrogen limiting conditions, which result in intracellular accumulation of NADH, since ATP is not utilized for protein synthesis. Therefore, PHB reserves will accumulate when cells are limited in oxygen or in nitrogen, but still have a carbon source available (Dawes and Senior, 1973). The degradation of PHB results in the production of two molecules of acetyl CoA and one NADH. The degradation of PHB occurs when the internal concentration of N A D + and free CoA increases, while the concentration of acetyl CoA is low. For example, PHB is degraded in the presence of oxygen when the external carbon sources are limited. However, if both oxygen and an external carbon source are present, PHB should not be degraded (Dawes and Senior, 1973). Several researchers (e.g., Suzuki et al, 1986; Ueda et al, 1992; Kang et al, 1993) recently reported that aerobically-grown facultative methylotrophic bacteria, such as Paracoccus denitrificans and Methylobacterium, produced PHB from inorganic salt medium containing methanol (sole carbon and energy source) under nitrogen-limited conditions. Both PHB content and specific PHB production rate depended on the ratio of methanol to ammonia in the feed solution. 41 Chapter Three MATERIALS AND METHODS The performance of the two-stage biological nutrient removal (BNR) process under conditions favorable to simultaneous nitrification and denitrification (SND) was studied in this research. A total of 13 experimental runs were conducted in the three continuous bench-scale activated sludge systems under steady-state conditions. In addition, five series of transient-state experiments were conducted. Materials used in the experiments, experimental procedures, statistical methods, and analytical methods are detailed in this chapter. 3.1 Seeding Sludge and Wastewater Source The seeding sludge for the laboratory experimental systems was taken from the pilot-scale, activated-sludge plant operated in a biological phosphorus removal (BPR) mode at the University of British Columbia in Vancouver, BC. Raw sewage from the same facility was used as the influent feed and was totally of domestic origin. Every morning, the raw sewage was pumped into the two completely-stirred storage tanks, starting at 9:30 A M until the tanks were full (around 11:30 AM). From these tanks, the sewage was taken at the same time (around 12:00 AM) on every Friday during the entire study to reduce the variation in the sewage strength and composition. The sewage was then stored in the laboratory cold room at 4 °C, for the use in the bench-scale experiments. The sewage, screened through a mesh with a pore size of 1.5 42 mm, was added into a continuously-stirred feed bucket. The sewage in the feed bucket was fed into the experimental systems at a rate of 57.6 liters per day. Due to the low alkalinity of the raw sewage, approximately 100 mg/L additional alkalinity, using sodium bicarbonate, was added to the feed bucket. The feed bucket was replenished daily. Details of the feed characteristics and variations will be presented in Section 4.1.2. 3.2 Reactor Systems Continuous-flow studies were conducted in three bench-scale reactor systems, of which one, a three-stage Bardenpho process (3-Bardenpho process), served as a control system; the other two were identical two-stage BNR processes (two-stage process). The 3-stage Bardenpho process consisted of three distinct zones: anaerobic, anoxic, and aerobic. Air was continuously provided to the aerobic zone by fine bubble aerators. The two-stage process consisted of an anaerobic selector and an intermittently-aerated, completely-mixed (IACM) reactor, in which the intermittent aeration was controlled by using absolute ORP range. A schematic diagram of the three systems is shown in Figure 3.1. All reactors were made of Plexiglas. Mixing in all bioreactors was provided mainly by mechanical mixers continuously, but additional mixing was also provided to the three aeration tanks, by fine bubble aerators, during air-on periods. Floating covers were provided to the three anaerobic zones and the 3-Bardenpho anoxic zone for minimizing the air entrainment. The three separate cylindrical clarifiers were equipped with a 1 rpm scraper to prevent the sludge from adhering on the side walls. 43 co 0) -*-> CO CO o i a '•3 O a co < C O 4) 44 External substrates were pumped into the three systems through a separate container. All pumping work, such as feeding, adding external substrates, sludge return, and internal recycle, was performed by peristaltic pumps (Cole-Parmer Masterflex 7553-10). 3.3 Data Acquisition and Control System The data acquisition and control (DAC) system used in this study consisted of a computer, a connection board, an interface card (PCL812), and a user interface software (Lab-Tech Notebook). Sensors, such as DO (OxyGuard), ORP (Broadley James), and pH through a meter (Chemcadet/Cole-Parmer), were connected to the connection board, sending signals to the interface card. The signals were digitized by the interface card and stored in the computer. The signals were collected at a frequency of two per minute through ten channels at the same time. The D A C system was equipped with two mechanical relays and two solenoid valves for on-off control. The control signals (0, 1), which were generated from the computer by comparing the digitized ORP signals with the preset upper and lower ORP limits, were used to switch the solenoid valves (open or close), through the relays (on or off). To ensure reliable measurements, three ORP probes were immersed in one tank and the median out of three measurements was used in the aeration control. 45 3.4 Process Evaluation under Steady-State Conditions 3.4.1 Experimental design The operation of the three experimental systems, treating a common waste stream, allowed parallel comparisons between different process designs (the 3-Bardenpho and the two-stage process) and between operation modes (different ORP control ranges). The t-test for paired comparisons (Microsoft Excel) was used to process the data when the comparisons on process performance, among the three systems within each individual run, were made. The paired f-test is designed to improve the precision of comparison between the means of two data sets, by making comparisons within matched pairs of experimental data, where the experimental material may be subject to uncontrolled variations between different matched pairs, but each matched pair is tested on common sample of experimental material. Inferences are made on the difference between two means, by making inferences about the mean of the differences between the matched pairs (Montgomery, 1984). As described above, the paired comparison is designed to factor out variations in the experimental material. In this study, the experimental material was the influent feed, which was very likely to vary daily and seasonally. Since the three systems were operated in parallel, treating a common feed sewage, all three systems were subject to the same daily and seasonal variations. Therefore, the comparisons on process performance among the three processes were not subject to the influence of the variations in the feed sewage. The 46 matched pairs were the samples taken from the three systems, at the same location in the process train, at the same time. The /-test for comparisons between two small independent data sets with unequal variances (independent /-test) was used to process the data when the comparisons on process performance, among different runs (i.e., different external substrate dosages) within each experimental system, were made. Generally, the independent /-test requires two data sets generated from an experimental system under the same environmental conditions, except controlled (or testing) conditions. Thus, unlike the paired /-test, it usually is not appropriate for an experimental system, in which the experimental material may be subject to uncontrolled variations. In this study, the changes in external substrate dosage (controlled conditions) were relatively large and were believed to have a major effect on the process performance, compared to the feed variations with time (uncontrolled variations). Under these conditions, the independent /-test was considered to be valid. Furthermore, two replicate experimental runs were designed to assess the effect of the feed variations on the process performance. Eleven experimental runs, plus two replicate runs, were divided into two groups according to the two ORP control regimes: the high/low and wide/narrow groups. Under the high/low ORP control regime, the intermittent aeration in the two 2-stage processes, specifically in the two IACM tanks, was controlled by either a high ORP range (0-50 mV) or a low ORP range (-50-0 mV); the two processes are denoted as the high ORP process and the low ORP process, respectively. In the same way, under the wide/narrow ORP 47 control regime, the two-stage processes were assigned with either a wide ORP range (-100-100 mV) or a narrow ORP range (-25-25 mV) to control the intermittent aeration in their IACM tanks and are defined as the wide ORP process and the narrow ORP process, respectively. The external substrates supplemented were sodium acetate and methanol. Sodium acetate was added into the three anaerobic zones continuously. Methanol was added into the two IACM tanks during air-off periods and into the 3-Bardenpho anoxic zone continuously. The experimental runs in the high/low ORP group were further divided into two subgroups according to the external substrate used: acetate group and methanol group. In each subgroup, a 2x4 factorial experiment was conducted. The two independent variables were the ORP control range with 2 levels and the substrate dosage with 4 levels. The objective of these two subgroups of runs was to investigate the effects of the ORP control range and the external substrate dosage (including both acetate and methanol) on the process performance, as well as the interaction between them. In the wide/narrow ORP group, three runs were also divided into two subgroups. In each subgroup, a 2x2 experiment was conducted. The two independent variables were the breadth of ORP range (2 levels) and the substrate dosage (2 levels). The objective of this group of runs was focused on studying the effect of the breadth of ORP control range (wide and narrow). The details of the ORP range and dosage settings for all experimental runs are summarized in Table 3.1. 48 CD S3 O •o S s-<u •a a ss <u OX) e « :-"o s-o u .22 z .2 03 > o o u E C v a Ji 2 .2 'E cs > eS H Pi ~o ts c o o I is o col CM o o o t-l PL, O JS oo I 1) •a a CQ I x> 3 CO "3 E Si w op § SI Pi O (3 I gp CO P i O C S o S3 o c o O O Q o Q Q O cj, oo $ CJ CO S 2 O < t oo C3 1) CO 00 >n o "5 s t oo $ D CO P i I o u o o PH O 1 pa I ! PH o o 2 s I f I s s o O Q o Q 1 Pi 00 >/"} > 0 < N i < N i < N i i i i < N < N o o o o o o © o CO 5 o < t a" The 3-Bardenpho process, as the control system, was operated in parallel to the two test systems, in terms of the feed and the external substrate dosage; the DO concentration in the 3-Bardenpho aerobic zone was maintained manually at approximately 3 mg/L throughout this study. 3.4.2 Operating Conditions During the first two runs, the sewage was diluted to provide the three systems with a feed with a constant total COD of250 mg/L. However, the sewage without dilution was used for the rest of the runs. The reason for such change was to preserve the organic carbon present in the sewage. The experiments were conducted at room temperature of 20 °C throughout this research. The same solids retention time (SRT) of 15 days was maintained for all three systems by wasting mixed liquor daily from the aeration tanks. The other experimental conditions, such as reactor volume, hydraulic retention time (HRT) and flow rate, are presented in Table 3.2. Table 3.2 The fixed experimental conditions. Process 3-Bardenpho process 2-stage process A 2-stage process B Volume 0^ ) HRT (h) Volume (L) HRT (h) Volume (L) HRT (h) Anaerobic 0.6 0.75 0.6 0.75 0.6 0.75 Anoxic 3.0 .3.75 Aerobic 4.2 5.25 7.2 9 7.2 9 Total 7.8 9.75 7.8 9.75 7.8 9.75 Clarifier 4.0 5 4.0 5 4.0 5 Flow rate (L/d] Ratio Flow rate (L/d) Ratio Flow rate (Ltd) Ratio Influent 19.2 1.0 19.2 1.0 19.2 1.0 Internal recycle 57.6 3.0 Sludge return 9.6 0.5 9.6 0.5 9.6 0.5 50 3.4.3 Sampling At the beginning of this research, the three systems were operated under the conditions set for run 1, for approximately two months to acclimatize the sludge to the intermittent aeration conditions. Then, the three systems were operated for at least one SRT to acclimatize the sludge under the respective conditions set for each run. After the acclimatizing period, sampling was commenced and conducted every other day for the next two or three weeks, until 8 sets of samples were collected. In order to reduce the influence of routine operation, all daily routine work, such as wasting sludge and replenishing the feed bucket, was done at the same time of day. All sampling work was done at the time after replenishing the feed bucket for approximately 20 hours. Grab samples were taken for the sewage from the cold room and from the feed bucket to represent the influent feed at time 0 and 20 hour, respectively. Grab samples were taken from all bioreactors. Half-hour composite samples were collected from the clarifier overflows to represent the effluents. The unfiltered samples from the influent, the three effluents and the three aeration tanks were analyzed for total chemical oxygen demand (total COD), total suspended solids (TSS), volatile suspended solids (VSS), total phosphorus (TP), and total Kjeldahl nitrogen (TKN). The filtered samples from all bioreactors, as well as the influent and the effluents, were analyzed for soluble COD, orthophosphate (ortho-P), ammonia, nitrate + nitrite (NOx), and nitrite: The filtered samples from the influent and the three anaerobic zones were also analyzed for volatile fatty acids (VFAs). Sludge volume index (SVI) was 51 determined for the mixed liquor from the three aeration tanks. A summary of testing parameters, sample locations, and number of samples is given in Table 3.3. Table 3.3 Summary of the sampling location and the analytical parameters. Analytical parameters Sample Location Total samples (all three systems) (each run) Total Suspended Solids (TSS) Raw Sewage* >8 Volatile Suspended Solids (VSS) Actual Feed* >8 Anaerobic Mixed Liquor >8x3 Aerobic Mixed Liquor >8x3 Clarifier Overflow (effluent) >8x3 Total Chemical Oxygen Raw Sewage 8 Demand (Total COD) Actual Feed 8 Soluble COD Raw Sewage 8 Orthophosphate (ortho-P) Actual Feed 8 Nitrate+Nitrite (NOx) Anaerobic Zone 8x3 Nitrite Anoxic Zone 8 Ammonia Aeration Tank 8x3 Clarifier Overflow (effluent) 8x3 Total Phosphorus (TP) Actual Feed 8 Total Kreldahl Nitrogen (TKN) Anaerobic Mixed Liquor 8x3 Aerobic Mixed Liquor 8x3 Clarifier Overflow (effluent) 8x3 Volatile Fatty Acids (VFAs) Raw Sewage 8 Actual Feed 8 Anaerobic Zone 8x3 Anoxic Zone 8x3 *Raw sewage represents the sewage from cold room. "Actual feed represents the sewage after a 20-hour storage in the feed bucket. "Sampling was conducted on that parameter only during some specific runs. 3.4.4 Methods of Calculation Mass balances were performed to calculate the parameters of interest. In addition to the balances on the total process train, relevant mass balances on individual reactors were performed. A summary of the mass balances performed in this study is given in Table 3.4. The calculation formulas for mass balances are given in Appendix A. The methods of calculating the amounts of nitrification and unaccounted for nitrogen loss are presented in Appendix B. 52 For each parameter, usually eight data were available. The median of eight data was used to report the results, since the median is a more efficient estimator of a population's mean than is the arithmetic average for a small data set (Wadsworth and Stephens, 1986). Table 3.4 Summary of mass balances. Process Reactor Mass Balance Parameters Feed Bucket Soluble COD Reduction VFAs Reduction Ammonia Loss Anaerobic Zone Phosphate Release NOx Reduction Ammonia Loss VFAs Reduction Soluble COD Reduction Anoxic Zone Phosphate Uptake NOx Reduction Ammonia Loss Soluble COD Reduction Aeration Tank (including the 3- Phosphate Uptake Bardenpho aerobic zone) Ammonia Loss NOx Production Soluble COD Reduction Sludge Wastage TKN Loss TP Loss NOx Loss Clarifier Phosphate Change Ammonia Change NOx Change Soluble COD Change Total Process Total Nitrogen Removal Total Phosphorus Removal Total TKN Removal Total COD Removal Total Process + Sludge Wastage Amount of Nitrification Total Phosphorus Balance Total Process + All Individuals Amount of Unaccounted Nitrogen Loss in Total Process 3.4.5 Dynamic Analysis Under Steady-State Conditions Grab sampling was conducted to investigate the process dynamic behavior in response to the intermittent aeration during the last five steady-state runs. This was done 53 by tracking the ammonia, NO x , ortho-P, soluble COD concentrations in the two IACM tanks during one aeration cycle. 3.5 Dynamic Study Under Transient-State Conditions The transient-state conditions in the three systems were created by imposing either ammonium or nitrate shock load. The objective of this dynamic study was to determine the nitrification and denitrification reaction rates in the IACM tank under the intermittent aeration conditions, which were normally maintained during the steady-state phase. Originally, the two IACM tanks were used as two batch reactors, by interrupting the anaerobic flow (the flow from the anaerobic zone to the IACM tank). However, it was found from the first two transient-state tests, conducted during steady-state run 9, that the interruption of the anaerobic flow resulted in a failure of the intermittent aeration in the IACM tank, due to a lack of incoming organic carbon. To prevent such failure, continuous-flow conditions were maintained in the three systems for the rest of the transient-state experiments. 3.5.1 Transient-State Study Under Continuous-Flow Conditions After the completion of a steady-state phase, an instantaneous ammonium shock load, followed two days later by a nitrate shock load, was imposed on the three systems under continuous-flow conditions. The use of a two-day lag period between the two shock loads would allow process to return to the steady state from the disturbance, resulting from the first loading. For providing the ammonium shock load, ammonium chloride was added into the two IACM tanks and the 3-Bardenpho aerobic zone instantaneously. The 54 amount added was equivalent to additional 30 mgN/L of ammonium in the reactors, where the shock loading was imposed. After the systems returned to the steady-state from the disturbance, a nitrate shock loading was provided to the three systems by adding sodium nitrate into the two IACM tanks and the 3-Bardenpho anoxic zone, which gave additional 30 mgN/L of nitrate in the above reactors. To prevent the added chemical from being washed away, the outlets of the reactors were shut off just before pouring the chemical solution and were reopened after a two-minute mixing of the added chemical with the mixed liquor. The two-minute mixing time would allow the shock loading to be evenly distributed in the entire reactor. An intensive sampling program started immediately and continued up to 5 hours after shock loading was imposed. A total of five series of transient-state experiments were conducted under the continuous-flow conditions. Each series included two experiments: one for the ammonia shock loading and the other for the nitrate shock loading. Except for the shock loading, the rest of the experimental conditions for each transient-state experiment were kept the same as those of the corresponding steady-state run. Details of the experimental conditions are summarized in Table 3.5. Table 3.5 Experimental conditions for the transient-state experiments under the continuous-flow conditions. A B C 3-Bardenpho Process Two-stage IA Process Two-stage IA Process External Aerobic External ORP External ORP Series No. Substrate DO Substrate Range Substrate Range (mgCOD/L) (mg/L) (mgCOD/L) (mV) (mgCOD/L) (mV) 1 Acetate/30 3.0 Acetate/30 -25-25 Acetate/30 -100-100 2 Acetate/50 1.0 Acetate/50 -25-0 Acetate/30 -25-0 3 Methanol/15 3.0 Methanol/15 -50-0 Methanol/15 0-50 4 Methanol/15 3.0 Methanol/15 -100-100 Methanol/15 -25-25 5 Acetate/100 3.0 0 -50--0 Acetate/50 0-25 55 3.5.2 Transient-State Study Under Batch Conditions As mentioned earlier, similar transient-state experiments were also conducted in the two IACM tanks under batch conditions; that is, the two IACM tanks were used as two batch reactors after the interruption of the anaerobic flows. Other than the flow conditions, the transient-state experiments under batch conditions followed the same procedures for those under the continuous-flow conditions. Only two such experiments were conducted during steady-state run 9 . 3.6 Analytical Methods The procedures for sample preparation and preservation techniques are summarized in Table 3 . 6 . Filtration was done, immediately after collection, using a Whatman No. 4 filter, except SS samples using a No. 9 3 4 - A H filter. Only COD, SS and SVI samples were analyzed immediately after collection. The analytical methods are briefly summarized in this section. Table 3.6 Sample handing. Test Filtration Storage Preservative NOx/ortho-P yes up to 1 week at 4 °C phenyl mercuric acetate Nitrite yes up to 1 week at 4 °C phenyl mercuric acetate Ammonia yes up to 1 week at 4 °C 10% sulfuric acid, pH=3 VFAs yes up to 1 month, freeze 2% phosphoric acid, 0.1 mL Soluble COD yes no none Total COD no no none TKN/TP no up to 1 month, freeze none MLSS/MLVSS yes no none Inf. and Eff. SS yes no none SVI no no no 56 Chemical Oxygen Demand (COD). COD determination was carried out immediately after collection according to the colorimetric method outlined in Standard Methods (APHA, 1992). Suspended Solids (SS) and Volatile Suspended Solids (VSS). The TSS and VSS determinations were carried out immediately after collection according to the procedures outlined in Standard Methods (APHA, 1992). ORP, Dissolved Oxygen (DO) and pH. ORP and DO concentration were collected directly by the D A C system from the ten channels simultaneously. The pH data were collected by the D A C system through a pH meter. The resolution analogue/digital converter was 0.5 mV. ORP probes are made by Broadley James Corporation that combined indicating electrode (platinum) and reference electrode (Ag/AgCl). DO probes (OxyGuard) were made by the Point Four Systems Inc. All probes were cleaned every two days after sampling. ORP and DO probes were calibrated every two weeks according to the manuals provided by the manufacturers, and pH probe according to the Standard Methods (APHA, 1992). Nitrate and Nitrite (NOx), Nitrite, Ammonia and Ortho-P. Filtered samples for ammonia were preserved by adding one drop of 10% sulfuric acid and analyzed on a Lachat QuikChem Automated Ion Analyzer according to QuikChem method No. 10-107-06-1-D. Filtered samples for NO x , nitrite, and phosphate were preserved by adding 1 drop of phenyl mercuric acetate solution (0.1 g phenyl mercuric acetate in 20 mL acetone and 80 mL distilled water) and were analyzed on a Lachat QuikChem Automated Ion 57 Analyzer, according to QuikChem methods No. 10-107-04-1-E for nitrite/nitrate and No. 10-115-01 -1 -D for phosphate. Total Kjeldahl Nitrogen (TKN) and Total Phosphorus (TP). Unfiltered samples for T K N and TP were frozen immediately after collection for up to one month. Analyses were conducted on a Lachat QuikChem Automated Ion Analyzer, according to QuikChem Method No. 10-107-06-2-E for T K N and Method No. 10-115-01-1-C for TP. Volatile Fatty Acids (VFAs). Filtered samples for VFAs were preserved with 0.1 mL of 2% phosphoric acid solution and frozen immediately for up to two months. Analyses were conducted according to Supelco GC Bulletin 75IG, using a computer-controlled Hewlett-Packard 5880A gas Chromatograph, equipped with a flame ionization detector (FID) and using helium as the carrier gas. Sludge Volume Index (SVI). SVI was determined immediately after sample collection. The standard method (APHA, 1992) requires 1000 mL mixed liquor to be settled for 30 minutes in a 1 L graduated cylinder. However, because of the limitation of mixed liquor (due to the small scale of the experimental setup), the following changes were made upon the standard method: 1) 400-500 mL mixed liquor, instead of 1 L, was used in the determination; 2) subsequently, the container used was a 500 mL plastic graduated cylinder. 3.7 Quality Control and Assurance for Analytical Data Quality control programs are essential to ensure that analytical data are valid. Two key elements in a quality control program are the assessment of accuracy and precision of 58 analyses. In this research, the quality control program was carried out for all forms of nitrogen and phosphorus. The accuracy of these analytical parameters was controlled by a series of standards and test solutions that were prepared under the same conditions as samples. The precision of each parameter was estimated by analyzing certain numbers of duplicate or triplet samples at different time during this study. The results are presented in Table 3.7. Table 3.7 The estimated precision for the analytical techniques. Analytical Sample Standard Mean Coefficient of Remark Parameter Number Deviation (s) (x) Variance (s/x) (n) (mg/L) (mg/L) (%) NOx 92 0.127 4.9 2.6% Filtered N0 2 48 0.0012 0.11 1.1% Filtered NH4 44 0.157 2.6 6.0% Filtered Ortho-P 92 0.031 3.5 0.9% Filtered TKN 30 2.48 40.9 6.3% high SS* TKN 32 0.73 9.0 8.2% low SS* TP 30 1.0 16.3 6.1% high SS TP 32 0.17 3.0 5.7% lowSS *high SS represents the mixed liquor samples and low SS represents the influent and the effluent samples Generally, a relative error (coefficient of variance) less than or around 6% was found for all above parameters, except for the low suspended solids T K N sample (8.2%). It was noticed that the precision for filtered samples was generally higher than that for unfiltered samples. The detection limits for orthophosphate, N O x and ammonia were around 0.05 mg/L. The detection limits for T K N and TP would increase to 0.4 mg/L, considering the dilution ratio of 7.5, due to the preparations of T K N and TP samples. The precision of ORP measurement under in-situ conditions is shown in Figure 3.2. The three lines represent the measurements from the three ORP probes immersed in one aeration tank. The two measurements agree very well and deviate from the remaining one 59 up to 30 mV. Other than the ORP probe itself, these differences might result from uneven aeration in the tank. E - - 3 0 CC o -50 —o— Probe 1 - A — Probe 2 —x— Probe 3 1 1 1 1 1 1 1 1 0 1 2 3 4 5 6 7 8 9 Time (min) Figure 3.2 The precision of ORP measurements from the three probes immersed in one aeration tank. 60 Chapter Four RESULTS AND DISCUSSION The results obtained from the steady-state evaluations are detailed and discussed in this chapter. These include the observations from DO and ORP recordings and the direct comparisons among the three bench-scale activated sludge systems, in terms of nitrogen, phosphorus and carbon removals, and sludge settleability. The results obtained from the transient-state experiments are also presented in this chapter. The discussion under the broader aspects of simultaneous nitrification and denitrification (SND) is presented in the final two sections, by summarizing the results obtained from the steady-state and transient-state studies. 4.1 Steady-State Performance Evaluation As described in Section 3.4.1, the 13 experimental runs were divided into four groups according to the ORP control range and the external substrate addition. The results are also presented mainly in groups. However, the results obtained from the two sub-groups, under the wide/narrow ORP control regime, are presented together for conciseness. As noted earlier, the independent Mest comparison is subject to the influence of feed variations. Further, the assumption of independent samples for this 7-test is not satisfied by 61 the fixed-time interval sampling program, as conducted in this study. The procedures to rectify these two violations are. 1) the independent /-tests are only to be performed among the runs without significant differences in the feed C/N ratio; 2) only the data generated under the steady-state conditions are to be used. Quasi steady state was considered to be reached in one SRT. This assumption was judged by no significant changes (or trends) in measured parameters, such as MLSS, NO x , ammonia and ortho-P concentrations in the aeration tanks after one SRT period of cultivation. As an example, Figure 4.1 shows the MLSS levels, during run 1, approaching a quasi steady state, in response to a change in the feed total COD level. It took approximately one SRT for all three processes to reach a new steady state, after the feed total COD level was reduced from an average 300 to 220 mg/L. E, c o c 0 o c o o CO CO 4000 3500 3000 4 252000 1500 1000 — 3 -Bardenpho process • High ORP process —X—Low ORP process o> CO c 3 CD CM 3 CM 00 CM I 1 h -i- CM CO 3 3 ~3 3 "5 "5 Date 3 Figure 4.1 Solids levels in the three systems gradually reaching steady state in run 1 (designed SRT = 15 d). 62 It must be stated that a true steady state was never reached in the three systems, due to the intermittent aeration and the diurnal variations in the feed characteristics (see Section 4.1.2.1). 4.1.1 Observations on ORP, DO and pH Recordings 4.1.1.1 General Observations Based on the ORP and DO data recorded by the D A C system, typical measured ORP and DO ranges, as well as the time of air-on and air-off periods, are summarized in Table 4.1 for each run. The following results are generalized by examining Table 4.1. 1) The measured ORP range overrode the corresponding nominal ORP range by 5-10 mV when the aeration in the IACM tank was controlled by the three narrow ORP ranges (-50-0 mV, 0-50 mV, and -25-25 mV). However, no override was observed for the wide ORP range (-100-100 mV). The override resulted in a 5-10 mV overlap between the measured high and low ORP ranges. Compared to the breadth of the two ranges (50 mV), the overlap was relatively small; therefore, the two ORP control ranges were distinct from each other. 2) The measured DO concentrations in the IACM tank under control by the high and low ORP ranges were generally in a range of 0.2-1.0 mg/L. Within each individual run, while the lower limits of the two DO ranges were approximately the same, the upper limit of the DO range, corresponding to the high ORP range, was consistently 0.2-0.4 mg/L higher than that corresponding to the low ORP range. 63 TS © •c u Q. i 5 -•a o I h "S £ CD C 73 su e ,© "S su u e o u O Q §5 •a 3 CU E "73 '5. >. H — 73 H u 00 c C3 o 1.3 o 00 S ^ 2 > o Q 1 00 00 c o OB o c aj CO o U co i n o m M m O p p CN CN CN 00 00 m d o o i n CN co o d d o 0 ^ S 1 1 i n o CN m i n i n — CN m o CN ro —< 0 O O CO VO i n o i n CN m CN i~- 00 =tt % 3 o 00 I O m m CN 1—' '—• m in i n m i n 0 0 d 1 a o 0 - o o o <n i n o o m m o o CN CN P P CN CN i n r — d d ro co d d o o VO vo O O O i n m m O O O _ vo m » * % m o o — CN CN i n m i n 00 00 d d i n i n ro o d d o ^ ^ i n i n CN CN O i n m CN < < 2 2 < < < < < < CN £ z i 3 o < 1 t o > O CN o I • tl 3 00 o erf 1 S3 O U a-8 o 8 © o o o o d o o l<8 a o a. S3 u 00 , a >> M3 o a 3) The total aeration cycle time was generally in a range of 3-5 minutes, when the aeration in the IACM tank was controlled by the three narrow ORP ranges; the time was about four times longer (15-20 minutes) when the wide ORP range was used. This indicates that the breadth of ORP range governed the total cycle time. Also, the air-on period was generally equal to the air-off period, when the three narrow ranges were used; this was partially due to the slight adjustments of the air flow rates: That is, at the beginning of each run, the air flows into the two IACM tanks were slightly adjusted to give equal time for the air on/off periods. This action was based on the conclusion that the optimal aeration fraction (AF: the time ratio of air-on period to total cycle) for nitrogen removal is 50% (Batchelor, 1982; Bakti and Dick, 1992). By this action, the actual AFs, corresponding to the three narrow ORP ranges, were controlled in a range of 43-57%. However, relatively low AFs were observed in the systems under the wide ORP control, such as 33% and 37% for runs 9 and 12, respectively. Figure 4.2 presents the typical on-line pH profiles in the two IACM tanks recorded by the D A C system. A less than 0.1 pH unit variation was observed in both pH profiles; the variation appeared to be random, in other words, it did not respond to air on and air off. During the entire experiment, the pHs in the two IACM tanks were generally in a narrow range of 7.0-7.4. Therefore, it can be concluded that the pHs in these tanks were constant. The constant pH was probably due to the supplement of extra 100 mgCaCOs/L alkalinity to the feed. Discussion on the constant pH in the current systems will be presented later in this section. 65 7.40 i 7.00 J 1 1 1 1 1 1 L- 1 1 1 1 1 1 1 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min) Figure 4.2 The typical pH profiles in the two IACM tanks recorded during run 9. The typical ORP and DO profiles in the IACM tank are presented in Figure 4.3 for the high/low ORP control regime and in Figure 4.4 for the wide/narrow ORP control regime. In addition to confirming the findings summarized from Table 4.1, Figures 4.3 and 4.4 show the details of the aeration cycle patterns. In an aeration cycle, two phases can be identified according to the ORP profiles: the ascending portion represents the air-on period and the descending portion represents the air-off period. According to the DO profiles, the air-off period can further be broken down to two sub-periods: zero and non-zero DO periods. For example, in the DO profile corresponding to the wide ORP control (Figure 4.4b), the descending portion represents a non-zero DO period, whereas the flat portion represents a zero-DO period. Although the indicated DO concentration in the flat portion did not reach zero, it should be considered as a zero DO concentration, because it is recommended by the manufacturer of the DO probes (OxyGuard) that the baseline (or the flat portion) be denoted as a zero DO 66 concentration and be used as an offset for an adjustment of other non-zero DO measurements. a) ORP profiles • Low ORP IACM tank •High ORP IACM tank b) DO profiles • Low ORP IACM tank •High ORP IACM tank 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min) Figure 4.3 The typical measured ORP and DO profiles corresponding to the nominal high and low ORP control ranges (recorded during run 5): a) ORP profiles and b) DO profiles. 67 a) ORP profiles -110 -1 ' ' ' 1 1 1 1 1 1 —1 1 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (min) b) DO profiles 1.6 -i O.o -1 ' 1 ' 1 1 L - 1 1 1 1 1 L -0 2 4 6 8 10 12 14 16 18 20 22 24 Time (min) Figure 4.4 The typical measured ORP and DO profdes corresponding to the nominal wide and narrow ORP control ranges (recorded during run 3): a) ORP profdes and b) DO profiles. 68 As indicated by the length of the flat portion (Figure 4.4b), the time of the zero DO period was approximately 2 minutes, which was one tenth of the total cycle time for this particular run (run 3). It was observed that the time of the zero DO period in the wide ORP IACM tank during runs 9 and 12 was up to one third of the total cycle time. Finally, as shown in Figures 4.3b and 4.4b, the zero DO period was negligible, when the aeration in the IACM tank was under control by the three narrow ORP ranges. 4.1.1.2 Relationship between ORP and DO Concentration The relationship between the ORP and the DO concentration in the IACM tank were more complicated than the linear relationship observed by Heduit and Thevenot (1989) and Lie and Welander (1994). The relationship was explored by plotting sets of ORP and DO data in the semi-log scale. The data sets were taken from the data files recorded by the D A C system. Because the pH in the IACM tank was constant, it was not considered as a parameter that influenced the ORP in the tank. A typical relationship, as shown in Figures 4.5, could be described as an ellipse band with five distinct phases. The first phase, which represents the period between the set lower ORP limit to the lowest ORP point, is characterized by an increase in DO and a decrease in ORP. As the ORP in the IACM tank reached the lower ORP limit, the aeration was switched on. The DO probe responded to the aeration immediately and the DO concentration began to rise. However, the ORP continued to decrease until it reached its lowest point. 69 ORP (mv) Figure 4.5 Typical relationship between ORP and DO concentration (data from run 3). The second phase, which represents the period from the lowest ORP point to the set upper ORP limit, is characterized by increases in both DO and ORP. As the aeration continued, the DO concentration continued to increase. The ORP, after reaching the lowest point, began responding to the aeration and increased until it reached the upper limit. The third phase, which starts from the set ORP upper limit to the highest ORP points, is characterized by a decrease in DO concentration and an increase in ORP. After the ORP reached the upper limit, the aeration stopped. The DO probe again responded to the cessation of aeration immediately; the DO concentration began to decrease. However, the ORP continued to rise to its highest point. 70 The fourth phase is characterized by decreases in both ORP and DO. As the cessation of aeration was prolonged, the ORP, after reaching the highest point, began to decline. The DO concentration continued to decrease; however, the decrease stopped at the lowest DO point. The fifth phase is characterized by no changes in DO and a decrease in ORP. Although the ORP continued to decrease during the air-off period, the DO concentration stayed constant in this period. As noted earlier, the constant DO concentration in this period was considered as zero. Among the five phases, the second phase, during which both ORP and DO concentration increased simultaneously, and the fourth phase, during which both ORP and DO concentration decreased simultaneously, were the two largest portions of the ellipse-band and were observed in all ORP-DO concentration plots. The first and third phases were reduced significantly, as the two sides of the ellipse-band became closer, and completely disappeared as the ellipse band became a straight line. The fifth phase, representing a zero DO period, was found to be significant, when the aeration was controlled by the wide ORP range. According to the Nernst Equation, an ORP-DO concentration plot in the semi-log scale should be a straight line, if the concentrations of other redox species, such as soluble organic substrate, ammonium and dissolved CO2, are not changing. The fact that the second and fourth phases were the two largest portions of the ellipse-band suggested that the ORP change in the IACM tank was dependent mainly on the DO concentration. In fact, it was reported (Harrison, 1972) that in the majority of cases, the increase in ORP 71 during the air-on period was caused by the increase of DO. However, the ellipse-band shape indicated that the changes of other variables, especially the change of C 0 2 concentration, could also influence the ORP in the IACM tank. For a continuous aeration process, two forces with opposite direction simultaneously act upon the carbonate system in the mixed liquor. First a great amount of C 0 2 is continuously produced by microbial respiration in the process. Concurrently, aeration continues to strip C O 2 out of the process. Therefore, in a continuous aeration process, a quasi equilibrium in the carbonate system could be reached. However, in an intermittent aeration process, such as used in this study, the C 0 2 concentration in the system may change dramatically in the transition period between air on and off. When the air is off, due to a discontinuation of the stripping process, the dissolved C 0 2 concentration in the process increases. In a study on ORP and pH as control parameters for intermittent aeration, Huang (1993) suggested: 1) the rate of increase in dissolved C O 2 could be higher than the DO consumption rate and 2) C 0 2 and O2 have the same oxidation state. According to his suggestions, the net result would be an increase of ORP during the initial air-off period. The third phase of the ellipse band is an example for this scenario (Figure 4.5). On the other hand, during the initial air-on period, stripping a large amount of accumulated dissolved C O 2 and continuing addition of organic substrate could cause the ORP to continue to decrease, noted as the first phase of the ellipse-band shape. Since the dissolved C O 2 concentration in the IACM tank was not measured, the above two scenarios only serve as a possible explanation for the ORP-DO relationship during a transition period between air on and off. 72 In a recent study, Hao and Huang (1996) observed the rapid pH changes, at the beginning of an air on and air off period. The near constant pHs in the current two IACM tanks were due to the presence of sufficient alkalinity. As suggested in the preceding paragraph, intermittent aeration could cause significant changes in dissolved CO2 concentration. However, due to the presence of enough alkalinity (bicarbonate), the changes in dissolved C 0 2 concentration would not cause significant changes in pH. Hao and Huang (1996) also reported several well-defined pH control points, corresponding the completion of nitrification and denitrification, during the air-on and air-off periods, respectively. The total aeration cycle time used in their study was in a range 2-6 hour, which was considerably longer than the cycle time used in the current study (3-5 minutes). The extremely short aeration cycle used in this study would not allow time for these pH points to occur. Furthermore, it was also possible that the pH control points did not even exist because nitrification and denitrification occurred simultaneously in the current IACM tanks. It was noticed in most of the runs that the ellipse bands were elongated and could be approximated as a straight line. Shown in Figures 4.6 and 4.7 are plots of the ORP and the adjusted DO concentration in the IACM tank. As noted in these plots, the linear relationship between ORP and DO concentration could prevail down to an adjusted DO concentration of 0.03 mg/L. However, according to the results obtained by Schuldiner et al. (1966), the linear relation holds down to a DO concentration at least 5xlO"5 mg02/L. Based on the linear phases of the plots observed in the current study, Equations 4.1 and 4.2 are obtained: 73 (4.2) For nominal high ORP range: Eh = 89.2 + 52.5 x log(Oz) (4.1) (R2=0.90) For nominal low ORP range: Eh = 63.4 + 62.1 xlog(Oz) (^=0.86) where Eh is the oxidation reduction potential referred to normal hydrogen potential (mV/NHE); 02 is the adjusted DO concentration (mg/L). ration (mg/L) ration (mg/L) ration (mg/L) ration (mg/L) ration (mg/L) ration (mg/L) ration (mg/L) ration (mg/L) ration (mg/L) x o DO concent o__ DO concent -o — DO concent tarn DO concent DO concent o 3)0 O DO concent COQO Txn o DO concent win "'•111 CCD 1 -1 0 0 1 0 2 0 3 ORP 0 4 (mV) 0 5 0 6 0 70 Figure 4.6 The typical linear relationship between the ORP and adjusted DO concentration in the high ORP IACM tank (data from run 5). 74 0.1 c cs u *•* e <U u e o • y O Q o.oi -60 (X3EECBDCD1 mrnrnnr crrarax>oq or o rnrrm T i m o p L D O < T H T Tf> C M m »o o o o< •ft H t XI)_ i o cxxrrTTQ x>o o -50 -40 -30 -20 ORP (mV) -10 10 Figure 4.7 The typical linear relationship between the ORP and adjusted DO concentration in the low ORP IACM tank (data from run 5). Theoretically, the ORP of an oxygen-water system at pH 7 can be expressed as (Heduit and Thevenot, 1989): Eh = 846+ 15 xlog(Oz) (4.3) However, in experimental investigations (Heduit and Thevenot, 1989; Lie and Welander, 1994), the slope was reported to be around 60 mV per log(02) rather than 15 mV per log(02); the intercepts were in a range of 180-410 mV, which depended mainly on the aeration conditions and sludge loading. For example, +410 mV was obtained for sludge aerated for several hours without feeding, +265 mV for over-aerated/low-loaded sludge, and 180 mV for high-loaded activated sludge from plug-flow system. 75 In terms of the slope value, the current study further confirmed the results in the literature. However, the intercept values of 60-90 mV were significantly lower than those obtained by Heduit and Thevenot (1989). In combination with the corresponding low DO levels, the lower intercepts values suggest that the sludge in the two IACM tanks was under-aerated. Therefore, in line with the observations by Heduit and Thevenot (1989), it appears that the intercept of the Nernst Equation generally represents the average aeration state of the activated sludge and the slope reflects the change of ORP with DO. In summary, the results from this study clearly showed that the changes in ORP were dependent mainly on the changes in DO concentration. Further, the ORP differences were large and regular, whereas DO differences were small and not consistent; the ORP control allowed the zero-DO period to be monitored. As described in Section 4.1.3, using different ORP control ranges resulted in significant differences in the process performance (e.g., the nitrification and N O x removal in the IACM tank) of the two-stage IA process; this confirms the conclusion that absolute ORP measurements can be used as a control parameter for intermittent aeration (Watanabe et al., 1985; Charpentier et al., 1987; de la Menardiere et al., 1991). 4.1.2 Feed Characteristics and Variations The total COD concentrations in the sewage from the UBC pilot plant varied in a range of 300 to 370 mg/L with an average TCOD/TKN ratio of 12.0 and an average TCOD/TP ratio of 75.0. Nitrate and nitrite were negligible; ammonia accounted for approximately 40-60% of TKN. 76 4.1.2.1 Diurnal Variations in the Influent Strength and Composition Paired f-tests were conducted on the mean concentrations of filtered COD, ammonia and VFAs between the raw sewage in the cold room and the actual feed. The actual feed represented the sewage after the 20-hour storage in the feed bucket. As shown in Table 4.2, the Mests detected significant decreases, in the above three analytical parameters, between the raw sewage and the actual feed, throughout runs 4 to 13, except for runs 1, 2 and 3. The exceptions were due to either no data (runs 1 and 2) or not enough data points (run 3). As described in Section 3.1, a mechanical mixer was used in the feed bucket. The mechanical mixer could entrain significant amount of air into the stored sewage. Under such conditions, aerobic growth could decompose about 20-40% of filtered COD (average 57 mg/L) and 100% of VFAs (average of 20 mg/L), present in the stored sewage, to water and C 0 2 , and assimilate 2-3 mgN/L ammonia for growth, over the 20-hour storage period. However, details on this diurnal variation were unknown, because a tracking study to monitor 24-hour changes in the feed bucket was not conducted. The complete consumption of the VFAs present in the stored sewage suggests that the sewage, after the 20-hour storage, contained no readily biodegradable COD (RBCOD). 4.1.2.2 Variation in the Influent TCOD7TKN Ratio The ratio of total COD to T K N (TCOD/TKN) is considered the most important parameter to represent the feed characteristics. The mean ratio in the actual feed is presented in Table 4.3 for each run. 77 •a tin o <a ' | | to CO 3 o (2 Z >1 1>. 5H 00 As noted in Table 4.3, the ratios varied from 9.0 to 12.0. Since the external substrates were added into the three systems directly from a separate container, the addition certainly increased the feed TCOD/TKN ratios to the systems. The ratios, after counting the external substrate dosage, are also given in Table 4.3. The increase in the feed TCOD/TKN ratio, caused by the external substrate addition, ranged from 0.5-3.2. The total COD to TP ratios (TCOD/TP) are also presented in Table 4.3 for references. Table 4.3 Mean influent TCOD/TKN and TCOD/TP ratios for each run. TCOD/TKN Ratio TCOD/TP Ratio Actual Feed After Addition of Actual Feed After Addition of Run No. External Substrate External Substrate 1 10.4 10.4 59.3 59.3 2 9.2 9.2 45.0 45.0 7 11.2 12.4 85.0 93.5 4 9.6 11.7 72.9 89.0 8 10.0 13.2 63.0 81.5 10 (A*) 10.1 12.0 66.8 79.3 10 03*) 10.1 12.0 66.8 79.3 10 (C*) 10.1 11.2 66.8 74.3 13(A) 10.8 14.3 72.0 96.4 13(B) 10.8 10.8 72.0 72.0 13(C) 10.8 12.7 72.0 84.2 11 10.2 10.7 52.0 54.5 6 11.4 12.7 64.2 71.2 5 9.0 11.3 60.8 76.2 3 9.9 9.9 73.5 73.5 9 12.1 13.3 57.0 62.7 12 12.1 12.7 68.0 71.3 *A, B and C represents each individual system. In order to conduct the independent Wests on the process performance, the independent 7-tests on the actual feed TCOD/TKN ratio is required to verify the assumption that there were no significant variations in the feed among different runs. The tests were only conducted among the runs within each group, since the comparisons on the process performance were also to be conducted in the same way. The results of the t-tests are summarized below. Firstly, since the raw sewage was diluted to provide a constant total COD of 230-250 mg/L in the feed during the first two runs; the feed total COD concentrations for these two runs were significantly lower than those for the rest of the runs (300-370 mg/L). Therefore, the first two runs were considered significantly different from the others, in terms of the feed characteristics, despite the fact that the differences in the feed TCOD/TKN ratio between the first two runs and the rest of the runs were not significant. Secondly, among the five acetate addition runs (including the two replicate runs), there were no significant differences in the feed TCOD/TKN ratio. Thirdly, among the three methanol addition runs, the ratio for run 5 was found significantly lower than those for runs 6 and 11. Lastly, among the three runs under the wide/narrow ORP control regime, the ratio for run 3 was found significantly lower than those for runs 9 and 12. 4.1.2.3 Variation in Methanol Dosage During the methanol addition runs, the methanol solution with a known concentration was added into the two IACM tanks from a separate container only during air-off periods. Such intermittent addition was for investigatng the feasibility of using ORP to control the addition of external carbon. The flow rate of the addition was set to give the nominal dosage based on the influent flow, the concentration of methanol solution, and an aeration fraction (AF) of 0.5. Due to the use of ORP in aeration control, the aeration cycle was not fixed. Therefore, the actual methanol dosage added was varied, since it depended 80 on AF. Since the AFs did not deviate far from 0.5 for most of the runs, the variation in dosage was generally not considered to be significant. However, in run 12, the actual methanol dosage added into the wide ORP IACM tank, calculated based on the measured AF of 0.38, could be 25% higher than the nominal dosage (15 mgCOD/L). 4.1.3 Nitrogen Experimental results related to nitrogen removal, nitrification, and denitrification are reported in this section. The absolute amounts of TN, TKN, ammonia, and N O x losses were calculated by performing mass balances. Based on the absolute amounts, the relative terms, such as removal efficiency, were calculated and used in presenting the results. The relative terms used in this study were: overall T N removal efficiency, overall T K N removal efficiency, percentage of N O x removal, percentage of nitrification, and percentage of denitrification. The first three calculations were based on the influent T K N concentration. The percentage of nitrification was the ratio of the amount of ammonia nitrified to the amount of available T K N (see Section 4.1.3.5 for the definition). The percentage of denitrification was calculated based on the amount of N O x denitrified and the amount of available NO x , i.e., the amount of ammonia nitrified (see Section 4.1.3.6 for details). 4.1.3.1 Overall Nitrogen Removal The overall T N removal efficiencies obtained from the acetate and methanol addition runs are presented in Figures 4.8 and 4.9, respectively. The acetate and methanol addition runs in this thesis denote the runs with acetate and methanol additions, respectively, under 81 the high/low ORP control regime only. The results obtained from the runs under the wide/narrow ORP control regime are given in Figure 4.10. Comparisons among the Three Processes Acetate addition runs. As shown in Figures 4.8, the overall T N removal efficiencies in the 3-Bardenpho process were consistently about 10-15% higher than those in the high and low ORP processes. Between the two 2-stage processes, there were no significant differences in overall TN removal efficiency over the entire acetate dosage range, except at an acetate dosage of 50 mgCOD/L. At this dosage, the low ORP process achieved 7% higher T N removal than the high ORP process. Table 4.4 presents the results of the paired Ntests on the effluent T N concentrations among the three processes. As noted, the 3-Bardenpho process, compared to the two IA processes, achieved 2.2-7.5 mgN/L lower effluent T N concentrations over the entire acetate dosage range; this confirms the results shown in Figure 4.8. However, the results of the comparisons between the high and low ORP processes were not consistent. For example, the effluent T N concentration of the low ORP process was the same as that of the high ORP process at acetate dosages of 0 and 100 mgCOD/L (zero and high dosages), 5.5 mgN/L higher at 30 mgCOD/L (low dosage), and 2.1 mgN/L lower at 50 mgCOD/L (medium dosage). As shown in Figure 4.8, when the results from the replicate runs were taken into account, the difference between these two processes at the low acetate dosage became not significant. Therefore, it is concluded that the low ORP process achieved better T N removal than the high ORP process only at the medium acetate dosage. 82 Acetate dosage (mg COD/L) Figure 4.8 The overall TN removal efficiencies obtained from the acetate addition runs. Table 4.4 The results of the paired f-tests on the effluent TN concentration obtained from the acetate addition runs. Median effluent TN concentration (mg/L) Calculated r*/Significant? Dosage 3-B* Low* High* 3-B/Low Low/High 3-B/High 0 12.6 14.8 -2.0/yes 0 10.9 14.4 -4.2/yes 30 6.9 14.4 8.9 -9.4/yes 2.4/yes -3.6/yes 50 4.1 7.8 9.9 -2.5/yes -3.2/yes -7.3/yes 100 7.1 10.3 10.4 -13.0/yes -1.4/no -3.5/yes •represent the 3-Bardenpho process, the low ORP process and the high ORP process, respectively. "Critical t at a=0.05 is between 1.8-1.9 depending on the number of data pairs. Methanol addition runs. As shown in Figure 4.9, the 3-Bardenpho process generally achieved 5-10% higher overall T N removal than the low ORP process, except at a methanol dosage of 30 mgCOD/L (medium dosage). At this dosage, the difference between the two processes was not significant. The 3-Bardenpho process also achieved 5-20% higher T N removal than the high ORP process, at various methanol dosages. 83 Compared to the high ORP process, the low ORP process achieved the same T N removal at methanol dosages less than 15 mgCOD/L (zero and low dosages) and at least 10% higher T N removal at methanol dosages of 30 and 60 mgCOD/L (medium and high dosages). The results of the paired /-tests on the effluent T N concentration (Table 4.5) confirmed the above findings. Methanol dosage (mg COD/L) Figure 4.9 The overall TN removal efficiencies obtained from the methanol addition runs. Table 4.5 The results of the paired /-tests on the effluent TN concentration obtained from the methanol addition runs. Median effluent TN concentration (mg/L) Calculated /^ Significant? Dosage 3-B* Low* High* 3-B/Low Low/High 3-B/High 15 5.0 10.6 10.4 -4.0/yes 0.2/no -4.3/yes 30 5.8 5.5 7.3 . 0.5/no -2.6/yes -2.0/yes 60 4.5 6.1 11.1 -2.0/yes -5.3/yes -9.1/yes •represent the 3-Bardenpho process, the low ORP process and the high ORP process, respectively. "Critical t at ct=0.05 is in a range of 1.8-1.9 depending on the number of data pairs. The runs under the wide/narrow ORP control regime. Figure 4.10 shows that compared to the 3-Bardenpho process, the wide ORP process achieved 20% less T N removal at the 84 zero external substrate dosage, about 10% higher TN removal both at a acetate dosage of 30 mgCOD/L and at a methanol dosage of 15 mgCOD/L. The results of the paired t-tests on the effluent TN concentration indicate that the above observations are all statistically significant (Table C. 1 in Appendix C). Raw sewage Acetate run Methanol run Figure 4.10 The overall TN removal efficiencies obtained from the runs under the wide/narrow ORP control regime. Also shown in Figure 4.10, compared to the narrow ORP process, the wide ORP process achieved the same level, 30% higher, and 15% higher nitrogen removal at external substrate dosages of zero, 30 mgCOD/L acetate, and 15 mgCOD/L methanol, respectively. The above findings were also confirmed by the paired t-test comparisons (Table C l in Appendix C). However, as described in Section 4.1.2.3, in the methanol addition run of this group, the actual dosage added into the wide ORP process was approximately 25% higher than those added into the other two processes, due to the variable aeration cycles. There is no way to determine to what extent the 25% higher 85 methanol dosage could affect the above comparisons. The results of the f-tests (Table C l ) also shows that compared to the 3-Bardenpho process, the narrow ORP process had 2.3-7.1 mgN/L higher effluent T N concentrations under various conditions. It should be emphasized here that the 3-Bardenopho process was operated at an aerobic DO concentration of 3 mg/L with the internal recycle at a rate of 3 times of the influent. Three-stage Bardenpho processes could have been operated with a DO of 0.5 mg/L and a higher recycle rate, and achieved 85 to 90% T N removal as has been found in many full-scale plants (Barnard, 1997). Therefore, the above comparisons are strictly limited to the experimental conditions set in this study. Influence of Uncontrolled Conditions Before conducting the comparisons among different external substrate dosages, the influence of the uncontrolled conditions (mainly the influent variation and the variation in aeration cycle) was investigated by conducting the independent Ntest comparison, on T N removal efficiency, between each pair of replicates. Two experiments were considered as a pair of replicates, if they were conducted in the same process (i.e., under control by the same ORP range) at the same external substrate dosage, but conducted during a different time period. A total of six pairs were available in this study. The results of the Mests, as well as the corresponding mean feed TCOD/TKN ratios, are presented in Table 4.6. ' As noted, a significant difference (6.1-17.0%) in overall T N removal efficiency was observed in most pairs (4 out of 6). The difference was generally less than 8% (5 out of 6). However, a large difference of 17% was also observed in the pair that was conducted 86 in the low ORP process at an acetate dosage of 30 mgCOD/L. Considering the large difference as an unusual case, the difference in T N removal efficiency, resulted from the variations in the uncontrolled environmental concentrations, was normally up to 8%. Table 4.6 The results of the independent f-tests on TN removal efficiency between a pair of replicates. , Feed TCOD/TKN TN Removal Efficiency (%) Dosage in two replicates in two replicates Independent r-test Process (mgCOD/L) 1 2 1 2 Difference Significant? 30* Acetate 12.1 11.2 71.9 71.2 / 0.7% No 3-Bardenpho 100* Acetate 10.8 10.0 83.8 77.7 6.1% Yes 15* Methanol 10.2 12.1 78.0 70.3 7.7% Yes Low ORP 30* Acetate 10.1 11.2 68.3 51.4 17.0% Yes 50* Acetate 10.1 9.6 74.1 69.7 4.4% No High ORP 50* Acetate 10.8 9.6 65.6 58.8 6.8% Yes Represents the external substrate dosage. The differences between a pair of replicates could not be attributed to the feed variation only, because a higher efficiency was found to be associated with a lower feed TCOD/TKN ratio in two pairs. As described in Section 4.1.1, there were variations in the aeration cycle and the DO concentration in the IACM tank, when the aeration was under control by the same ORP range; these variations must have contributed a major part to the significant differences observed. Effect of the Dosage of External Substrates Figures 4.8 and 4.9 show that the overall T N removal efficiencies in all three processes generally increased with increasing acetate and methanol dosages. The independent f-tests were conducted to confirm the increases or decreases shown in Figures 4.8 and 4.9 (see Table C.2). It is noted that, where a pair of replicates were available, two data sets were combined into one; the combined data set was used in the Mests. 87 Zero and non-zero external substrate dosages. As shown in Figures 4.8 and 4.9, significant increases were found in all three processes between the zero and low acetate dosages (more than 15%) and between the zero and low methanol dosages (more than 20%). However, as noted in Section 3.4.2, the runs without an addition of external substrate were fed with the diluted sewage, whereas the runs with external substrate additions were fed with the undiluted sewage. As noted in Section 4.1.2.2, compared to the diluted sewage, the undiluted sewage contained an average of 100 mg/L higher total COD, although the TCOD/TKN ratios between these two primary feeds were not significantly different. Therefore, the significant increases could not be attributed to the increases in acetate and methanol dosages only, since the lower T N removal efficiencies at the zero substrate dosage were obtained with the diluted sewage, and the higher removal efficiencies at the low acetate and methanol dosages were obtained with the undiluted sewage. The average 100 mg/L higher total COD in the undiluted sewage must have also contributed, in part, to the significant increases. , Acetate dosage. A significant increase in T N removal efficiency was detected in the 3-Bardenpho process (8.1%) and in the low ORP process (11.1%) between the low and medium acetate dosages (30 and 50 mgCOD/L), but not detected between the medium and high acetate dosages (50 and 100 mgCOD/L). This suggests that the acetate addition improved the overall T N removal in the above two process at the low and medium dosages, but not at the high dosage. The effect of high acetate dosage on process performance will be discussed in Section 4.1.3.6. In the high ORP process, a significant increase in T N removal efficiency was detected only between the low and high acetate 88 dosages, but neither between the low and medium nor between the medium and high. This suggests that the acetate addition did increase the overall T N removal in this process, but a high dosage was required. The details of the /-tests are given in Table C.2. Methanol dosage. Between the low and medium methanol dosages (15 and 30 mg COD/L), a significant increase in T N removal efficiency was found in both 2-stage processes (6.7% and 13.8%, respectively), but not in the 3-Bardenpho process. Between the medium and high methanol dosages (30 and 60 mg COD/L), a 7.6% increase was observed in the 3-Bardenpho process, whereas a 11.6% decrease was observed in the high ORP process and no significant changes in the low ORP process. As discussed in Section 4.1.3.3, the high T N removal efficiency, obtained in the 3-Bardenpho process at the low methanol dosage, was considered an abnormal case. Excluding this case, it appeared that methanol addition improved the overall T N removal in the 3-Bardenpho process over the entire dosage range and in the two 2-stage processes at the low and medium dosages, but not at the high dosage. The high methanol dosage even caused a decrease in the T N removal in the high ORP process. The causes for this decrease will be discussed in Section 4.1.3.6. Comparisons with Other Studies The sewage used in this study was totally of domestic origin. In comparison to the C/N ratio of 4.-10 frequently reported as required for complete denitrification (Stensel et al., 1973; Christensen and Harremoes, 1977; Nakis et al., 1979; Ekama et al., 1984; Rittmann and Langeland, 1985), the feed TCOD/TKN ratios of 10-13 in this study were 89 apparently higher than required. However, as described in Section 4.1.2.1, the influent feed into the three systems, at the sampling time, probably contained no RBCOD, as indicated by the zero V F A concentration and the consumption of 57 mg/L filtered COD in the storage bucket. The absence of RBCOD in the feed must have imposed a limitation on the performance of nitrogen removal in the current three systems and explains the significantly lower overall T N removal efficiencies during the first three runs (no external substrate addition). The results from several other similar studies are presented in Table 4.7. As noted, T N reduction of 60-90% was achievable by the single or two-stage IA processes. However, the high removal efficiencies (>90%) obtained by Lo et al. (1994) were probably due to the use of synthetic sewage, which contained only soluble COD (mainly glucose) with a T C O D / T K N ratio of 10. The high removal efficiencies (>90%) obtained by de la Menardiere et al (1991) were probably due to the extremely long SRT and HRT (35 days and 100 hours, respectively). Also noted in Table 4.7 and in the literature review (Section 2.6.3), all systems that achieved nitrogen removal in a single reactor were operated at long SRTs up to 100 days. The results from this study, in line with a few recent studies (Lo et al., 1994; Huang, 1993) suggested that with better controls, the long SRT could be shortened. However, it was noted by Barnard (1997) that the long SRT could not be shortened to the level that is used in the systems with higher DO levels. 90 Table 4.7 Comparison with the results from other similar studies. References Process Feed Characteristics Operating Condition TN and TP removal 1. Loetal. (1994) Lab-scale, an IACM tank with an anaerobic selector Synthetic sewage: COD = 370- 420 mg/L TKN = 30- 43 mg/L SRT = 20 days Total HRT = 20 h ORP control ORP = 40-180 mV ON/Off= 15/15 min. DO = 0.0-0.5 mg/L TN: 80-90% (1.3-11.1 mg/L) Dissolved P: 71-100% 2. de la Menardiere etal. (1991) Full-scale, an IACM tank with an anaerobic selector Sewage: 15% domestic and 85% industrial origin): COD = 850 mg/L BOD5= 470 mg/L TKN = 80 mg/L SRT = 35 days IACM HRT = 96 h Anaerobic HRT = 14 h Total cycle time = varies (1-5.5 h) ORP control ORP = 150-550 mV Temp. =6-21 °C TN: 70-90% (7.0-22 mg/L) Dissolved P: >70% 3. Nakanishi et al. (1990) Lab-scale and pilot-scale IACM tank Artificial sewage (lab): BOD = 220-250 mg/L TKN = 50-60 mg/L Domestic sewage (pilot) BOD = 30-80 mg/L TKN = 15-30 mg/L SRT = 40-50 days Temp. = 17-29 °C ON/OFF = 5-20 min. DO and ORP control DO = 0-0.3 mg/L ORP = 250-350 mV TN: 80-85% 4. Huang (1993) Lab-scale, IACM tank Sewage: COD = 140-220 mg/L TKN = 36-44 mg/L SRT = 7-20 days IACM HRT = 10-12 h Timer control ON/OFF = varies (1/1, 1.5/1.5, 3/3 h) DO = 2.5-7.1 mg/L (air-on) Temp. = 17-23 °C TN: 72-83% (7-10 mg/L) 5. This study Lab-scale, an IACM tank with an anaerobic selector Raw sewage: COD = 230-320 mg/L TKN = 22-33 mg/L RBCOD = 0 Supplemented with Acetate and methanol SRT = 15 days IACM HRT = 9.0 h Anaerobic HRT = 0.75 h ORP control ORP range = various ON/Off = varies DO = varies TN: 40-80% (6-15 mg/L) Dissolved P: 75-95% (2.0-1.0 mg/L) In terms of aeration control, the ORP control ranges used by Nakanishi et al. (1990), de la Menardiere et al (1991), and Lo et al. (1994) were considerably higher than the ranges used in this study; the DO concentration ranges observed in the study by Lo et al. (1994) were close to those observed in this study. The aeration cycle times (5-20 min.) used by Nakanishi et al. (1990) were slightly longer than those used in this study. Overall, 91 the results in Table 4.7 suggests that the effluent T N concentrations in a range of 6-12 mgN/L are achievable in a two-stage, IA process without a supplement of external substrate, if a reasonable amount of RBCOD is present in sewage. 4.1.3.2 Ammonia Loss and Denitrification in the Anaerobic Zone Ammonia Loss in the Anaerobic Zone Ammonia loss in activated sludge systems is the net result of two opposite reactions: ammonification of organic nitrogen and ammonia assimilation. The amounts of ammonia loss in the three anaerobic zones, obtained from the acetate and methanol addition runs, are presented in Figure 4.11. 1,2 —•—3-Bardenpho —o— Low ORP process —x— High ORP process 6 7 Run No. 8 10 11 Figure 4.11 The amounts of the ammonia loss in the anaerobic zone obtained from the three systems in the acetate and methanol addition runs with the same feed. 92 As noted, the amount of the loss ranged from -1 to 2 mgN per liter of feed. In each individual run, no significant differences were observed among the three anaerobic zones. This suggests that the loss was not related to the process configurations and operations. Among different runs, the variation pattern was the same for each process; this suggests that the loss depended mainly on the feed composition. Denitrification in the Anaerobic Zone The NOx concentration in the feed sewage was negligible. The N O x in the return sludge was recycled to the anaerobic zone and denitrified there. The amount of denitrification in the anaerobic zone was calculated by performing an N O x balance around that zone. In this study, in performing the N O x balance around the anaerobic zone, the N O x concentration in the clarifier overflow was used to approximate the N O x level in the return sludge. Because no, sludge blankets were observed in all three clarifiers during the entire study, except during run 4 in one system (the low ORP process), the amount of denitrification in the sludge blanket was considered to be negligible. Furthermore, since the retention time of the return sludge in the tubing system (size i8 and 1.0 meter long) was estimated to be less than 5 minutes, the amount of denitrification in the tubing system was also assumed to be negligible. Therefore, it is reasonable to use the N O x concentration in the clarifier overflow to represent the actual N O x in the return sludge. However, it should be remembered that using the N O x level in the clarifier overflow to approximate the N O x level in the return sludge could over-estimate the amount of denitrification in the anaerobic zone. 93 Percentage of N0X removal. The percentage of N O x removal in the anaerobic zone, which was calculated by dividing the amount of N O x removal in that zone by the influent T K N concentration, represents the relative amount of N O x denitrified in the anaerobic zone. Figure 4.12 presents this percentage obtained from the acetate and methanol addition runs. As noted, the differences in the percentage among three processes are significant and consistent in all runs, except in run 8. The highest percentage (11-16%) was found in the high ORP process, second (4-14%) in the 3-Bardenpho process, and last (0-4%) in the low ORP process. No relationship between this percentage and the external substrate dosage could be found because the denitrification in the anaerobic zone was limited by the N O x itself. 24 —•—3-Bardenpho process —o— Low ORP process —x— High ORP process Figure 4.12 The percentages of NO x removal in the anaerobic zone obtained from the runs under the high/low ORP control regime. 94 Percentage of denitrification. The percentage of denitrification in an individual reactor, which represents the degree of denitrification, is defined as the amount of N O x denitrified in that reactor divided by the N O x entering that reactor. Figures 4.13 and 4.14 present the percentages in the anaerobic zone obtained from the acetate and methanol runs, respectively. As indicated by the N O x data, the residual N O x concentrations in the anaerobic zone were generally in a range of 0.2-0.6 mgN/L, regardless of acetate and methanol dosages. This suggests that the residual N O x at levels less than 0.6 mgN/L may result from sample analyses rather than carbon limited conditions on denitrification. As shown in Figures 4.13 and 4.14, because of the residual NO x , 100% of denitrification was not observed in the anaerobic zone even at the high acetate and methanol dosages. The lower percentages of denitrification (less than 80%), in combination with the residual N O x concentrations higher than 0.6 mgN/L (1.3, 1.0, and 1.2 mgN/L), observed in the anaerobic zones of the 3-Bardenpho and high ORP processes, indicate that the anaerobic denitrification in these two processes was limited by organic carbon rather than N O x itself, at the zero external substrate dosage. Further, the residual N O x concentrations of up to 0.6 mgN/L suggest that 80% denitrification or above generally indicated the complete denitrification in the anaerobic zone. Based on that, complete denitrification was achieved in the anaerobic zone of the above two processes, when external substrate was supplemented. As also shown in Figures 4.13 and 4.14, this percentage in the low ORP process decreases with the increasing acetate and methanol dosages, except with the high dosages. The decreasing trend does not mean that the degree of denitrification decreased with increasing dosages, because it was actually caused 95 by the decrease in N 0 X concentration in the return sludge with the increasing acetate and methanol dosages. Figure 4.13 The percentage of denitrification in the anaerobic zone obtained from the acetate addition runs. Figure 4.14 The percentage of denitrification in the anaerobic zone obtained from the methanol addition runs. 96 4.1.3.3 Ammonia Loss and Denitrification in the Anoxic Zone Ammonia Loss in the Anoxic Zone Only the 3-Bardenpho process had a distinct anoxic zone. The amounts of ammonia loss in the anoxic zone normally varied in a wide range from 0 to 6 mgN/L in all runs, except in runs 11 and 12. In these two runs, the unreasonable high losses of 9.3 and 7.8 mgN/L were obtained. These two losses were even higher than the corresponding amounts of T K N removal due to sludge wastage; this suggests that the oxidation of ammonia (nitrification) had to occur in the anoxic zone during these two runs. However, from the literature review (Section 2.1), it is generally believed that nitrification is an aerobic process. Further, the transient-state experiment under batch conditions clearly showed that the oxidation of ammonium under anoxic conditions was impossible in the current systems (Section 4.2.1.1). Therefore, the assumed oxidation in the anoxic zone was believed to occur under aerobic conditions, which was most likely to be the results of air entrainment in the anoxic zone. The air entrainment was caused by the over-aeration in the aerobic zone (DO concentration of 3 mg/L) and the high ratio of mixed liquor recycle (300%). Since both nitrification and denitrification occurred at the same time in the anoxic zone, a significant, unaccounted for N O x loss was expected to occur in this zone during runs 11 and 12. If this loss were not included in the calculation, underestimation of the N O x removal in the anoxic zone would be expected. On the other hand, by using the calculation methods outlined in Appendices A and B, this unaccounted for N O x loss would 97 be attributed to the N O x loss in the aerobic zone, although it actually occurred in the anoxic zone. This would cause overestimation in the amount of N O x removal in the 3-Bardenpho aerobic zone. The average of the ammonia losses in the anoxic zone in the rest of the runs was 2.98 mgN per liter feed. Assuming that this average loss was due to ammonification and assimilation, any ammonia losses higher than this average would be caused by the nitrification in the anoxic zone. In this way, the amount of nitrification in the anoxic zone is quantified. Subsequently, both the N O x loss in the anoxic zone and the N O x loss in the aerobic zone are adjusted. It is noted here that the adjusted values will be used in the 3-Bardenpho process for runs 11 and 12 in this thesis. Denitrification in the Anoxic Zone The N O x denitrified in the anoxic zone came from the aerobic zone through the internal recycle. The N O x balance around the anoxic zone was performed to calculate the amount of N O x removal in this zone. Percentage of NOx removal. The percentage of N O x removal in the anoxic zone (based on the influent TKN), with respect to the acetate and methanol dosages, is presented in Figures 4.15. As expected, this percentage increased from 5 to 35% with an increasing acetate dosage even though a carryover of VFAs into the anoxic zone was not observed, as indicated by the zero V F A concentration in the anaerobic zone. Based on the biological phosphorus removal (BPR) mechanism, except for the acetate consumed for anaerobic denitrification, the rest would be stored by phosphorus 98 accumulating organisms (PAOs) in the form of poly-P-hydroxyalkanoates (PHA) in the anaerobic zone. Therefore, it was the carbon reserves stored in the biomass, not acetate, that was actually available to the 3-Bardenpho anoxic zone. Although the other carbon sources, such as slowly biodegradable COD (SBCOD) in the raw sewage, were still available to non-PAOs denitrifiers for denitrification, the improvement in the N O x removal in the anoxic zone by the acetate addition had to be due to the denitrification mediated by PAOs using the stored carbon reserves. This is because other carbon sources present in the anoxic zone were relatively constant among different acetate runs and would not give the trends observed in Figure 4.15. The improvement in the denitrification in the anoxic zone by adding VTA to the anaerobic zone without a carryover of VFAs was also reported by Alleman and Irvine (1980) and Van Huyssteen et al. (1990). 0 -I -I 1 1 i 1 0 20 40 60 80 100 Acetate and methanol dosages (mg COD/L) Figure 4.15 The percentage of NO* removal in the anoxic zone obtained from the acetate and methanol addition runs for the 3-Bardenpho process. 99 As also shown in Figure 4.15, the percentage of N O x removal in the anoxic zone increased from 5 to 45% with an increasing methanol dosage. As described in the literature review, methanol is an effective substrate for denitrification. Therefore, the increase observed was expected. The smaller increase between the medium and high methanol dosages was due to the N O x limitation, as confirmed by the corresponding low residual N O x concentrations in the anoxic zone (0.4 and 0.32 mgN/L). Percentage of denitrification. The percentage of denitrification in the anoxic zone (based on the flux of NO x ) is presented in Figures 4.16. As noted, this percentage increased from 5 to 90% with an increasing acetate dosage. The increase suggests that the denitrification in the anoxic zone was limited by organic carbon at various dosages, except at the high dosage. At the high dosage, near 90% of denitrification and small residual N O x concentrations (0.7 and 0.3 mgN/L) indicate that the denitrification was complete. 100 T : -I Acetate and methanol dosages (mg COD/L) Figure 4.16 The percentage of denitrification in the anoxic zone obtained from the acetate and methanol addition runs. 100 As noted in Figure 4.16, near 90% of denitrification at the medium and high methanol dosages and the corresponding residual N O x concentrations (0.4 and 0.32 mgN/L) suggest that complete denitrification was achieved in the anoxic zone in the above two dosages. Therefore, increasing methanol dosage from the medium to the high did not improve much the N O x removal in the anoxic zone. Finally, it is not appropriate to use the data shown in Figures 4.15 and 4.16 to compare the effectiveness of these two substrates in improving denitrification, since acetate was not directly available to the anoxic zone, as was methanol. 4.1.3.4 Nitrogen Removal due to Sludge Wastage and Nitrogen Content in Solids The amount of nitrogen loss due to sludge wastage was calculated based on the measured T K N (unfiltered sample) and N O x (filtered sample) concentrations in the aerobic mixed liquor and the daily volume of the wastage. The percentages of nitrogen removal due to sludge wastage, the ratio of the amount of nitrogen loss due to sludge wastage to the influent TKN, were in a wide range of 13-30%. It was noticed that the percentages in the two 2-stage processes averaged 3.3% higher than those in the 3-Bardenpho process. The difference between the measured T K N and ammonia (filtered sample) concentrations in the aerobic mixed liquor closely represents the nitrogen content in the solids because the major part of organic nitrogen in the mixed liquor is from the solids. Through wasting sludge, the organic nitrogen in the solids is removed. Under steady-state conditions, this organic nitrogen loss is equivalent to the amount of ammonia assimilated for growth. 101 All data points of the organic nitrogen concentrations in the mixed liquor are plotted against the aerobic MLSS concentrations in Figure 4.17. As noted, the organic nitrogen concentration increased linearly with an increase in the MLSS concentration, because the organic nitrogen in mixed liquor was mainly from the solids, as already noted. All data points follow the same trend line; this indicates that the nitrogen contents in solids were the same for all three processes. Based on the slope of the trend line, the nitrogen content in solids was estimated to be 8.4% (gN/gMLSS) for all three processes. This number was slightly higher than that observed in the UBC BPR pilot plant (6-7% gN/gMLSS) (Koch, 1997), but in the range of 9.3 to 12% (based on MLVSS) observed by Turk (1986) in systems treating a high-strength ammonia waste. 1800 2300 2800 3300 3800 MLSS level in the aerobic mixed liquor (mg/L) Figure 4.17 The organic nitrogen concentrations in the aerobic mixed liquors vs. the MLSS concentrations. The results presented in Section 4.1.6 indicated that an average 480 mg/L higher MLSS level was maintained in the two 2-stage processes than the 3-Bardenpho process by 102 the same amount of COD removed in the total process. Based on the constant nitrogen content (8.4% gN/gMLSS) and an average influent T K N concentration of 28 mgN/L, the i 480 mg/L higher MLSS would result in 3.9% higher nitrogen removal due to sludge wastage, in the two 2-stage processes than in the 3-Bardenpho process. This explains why the two-stage process achieved an average 3.3% higher nitrogen removal due to sludge wastage, compared to the 3-Bardenpho process. It should be noted that the N O x concentration in the aerobic mixed liquor also influences the percentage of nitrogen removal due to sludge wastage. 4.1.3.5 Percentage of Nitrification and Overall TKN Removal Percentage of Nitrification As summarized from the literature review, the following assumptions are considered necessary for calculating the amount of nitrification and the amount of unaccounted for nitrogen loss in the total process. 1) It is usually considered that ammonia is the primary source for assimilation, and the ammonia uptake for assimilation has priority over ammonia oxidation (e.g., nitrification); 2) assimilation of nitrate or dissimilation of nitrate to ammonia is not considered in sewage containing plentiful ammonia; 3) ammonia and organic nitrogen (TKN) must be first oxidized to nitrite or nitrate and subsequently be removed from the liquid through denitrification of N O x to gaseous forms of nitrogen; 4) organic carbon, not ammonia, is considered as an electron donor for denitrification; and 5) nitrification only occurs under aerobic conditions. The details of calculation methods are given in Appendix B. 103 As described in the literature review, both organic and inorganic nitrogen compounds can serve as substrates for heterotrophic nitrification. Therefore, in this study, T K N (including ammonia and organic nitrogen) was considered as a substrate for nitrification. However, two portions of the T K N were deemed not available to nitrification: ammonia assimilated for bacterial growth and the residual organic nitrogen present in the effluent (= effluent T K N - effluent NH + 4 ) . Part of the residual organic nitrogen was from the residual biomass in the effluent, which had already been accounted for in the amount of ammonia assimilation. The other part was from the liquid phase. However, the second part usually existed in non-biodegradable organic compounds, thus was not considered to be available to nitrification either. The percentage of nitrification, which represents both the amount and degree of nitrification, was calculated by dividing the amount of nitrification by the amount of available T K N to nitrification. Acetate and methanol addition runs. The percentages of nitrification, obtained from the acetate and methanol addition runs are presented in Figures 4.18 and 4.19, respectively. As noted, this percentage in the 3-Bardenpho aerobic zone and in the high ORP IACM tank was near 100% at various acetate and methanol dosages; this indicates that the nitrification in these two aeration tanks was complete and was not affected by the external substrate dosages. As also noted in Figure 4.18 and 4.19, the percentage of nitrification in the low ORP IACM tank generally increased with increasing dosages of both acetate and methanol. During the acetate addition runs, the percentages were about 65% at acetate dosages less 104 than 30 mgCOD/L (zero and low dosages), and 70-85% at 50 mgCOD/L (medium dosage), and near 100% at 100 mgCOD/L (high dosage). 60 -I : 1 1— 1 1 1 0 20 40 60 80 100 Acetate dosage (mg COD/L) Figure 4.18 The percentages of nitrification obtained from the acetate addition runs. Methanol dosage (mg COD/L) Figure 4.19 The percentages of nitrification obtained from the methanol addition runs. 105 During the methanol addition runs, the percentages were 60-65% at methanol dosages less than 15 mgCOD/L (zero and low dosage) and near 90% at 30 and 60 mgCOD/L (medium and high). The above results clearly indicate that the nitrification in the low ORP IACM tank was partially inhibited and the inhibitory effect was relieved by the acetate or methanol addition at the medium and high dosages. Explanations for this improvement will be given later in this section. Runs under the wide/narrow ORP control regime. The percentages of nitrification obtained from the runs under the wide/narrow ORP control regime are given in Figure 4.20. At the zero dosage and the 15 mgCOD/L methanol dosage, nearly 100% available T K N was nitrified in both the narrow and wide ORP IACM tanks. However, at the 30 mgCOD/L acetate dosage, the percentages were 58% and 88% in the narrow and wide ORP IACM tanks, respectively; this indicates that nitrification was inhibited at an average ORP level of 0 mV, and the narrow ORP range inhibited nitrification 30% more than the wide ORP range. The average DO concentrations in the two IACM tanks, over one hour period, were calculated based on the typical adjusted DO profiles, recorded during this acetate addition run; they were 0.69 and 0.33 mg/L in the wide and narrow ORP tanks, respectively. The higher percent of nitrification in the wide ORP IACM tank, compared to the narrow ORP IACM tank, is believed to result from the higher average DO level in the wide ORP IACM tank. Discussion. As described in the literature review, nitrification in activated sludge systems is usually considered as the autotrophic conversion of ammonia to nitrite or 106 nitrate. Numerous studies have suggested that autotrophic nitrification seems not to be inhibited in a DO concentration range of 0.5-1.5 mg/L, under normal operating conditions (Stenstrom and Poduska, 1980; Jayamohan et al., 1988; Stenstrom and Song, 1991; Baku' and Dick, 1992). It has also been experienced that high organic loading usually results in low performance of nitrification (Downing et al., 1964; Balakrishnan and Eckenfelder, 1970). Kiff (1972) reported inhibition of nitrification in batch tests by the presence of added acetate at low DO conditions. Hanaki et al. (1990a; 1990b) revealed that the inhibitory effect of high organic loading on ammonia oxidation was enhanced by low DO (0.3-0.8 mg/L). Figure 4.20 The percentage of nitrification obtained from the runs under the wide/narrow ORP control regime. As noted in Chapter 3, the DO concentration in the 3-Bardenpho aerobic zone was maintained at 3 mg/L during the entire study. The complete nitrification under such high 107 DO concentration, regard less external substrate dosages, was expected to be carried out by autotrophic nitrifiers. Recent studies (Rittmann and Langeland, 1985; Nakanishi et al., 1990; Lo et al., 1994) suggested that, under low COD loading and long SRT conditions, complete nitrification could be achieved in low DO concentration ranges. The DO concentration ranges reported were 0.0-0.5 mg/L (Rittmann and Langeland, 1985), 0.1-0.5 mg/L (Lo et al. 1994), and 0-0.3 mg/L (Nakanishi et al., 1990). The results from the current study, whereby nitrification was completed in the high ORP IACM tank with a typical low DO concentration range of 0.0-0.6 mg/L, regardless of various substrate dosages, further confirmed the findings from the above three studies. As noted in Section 4.1.1, within each individual run, the DO concentration range in the low ORP IACM tank was lower than that in the high ORP tank. Compared to the complete nitrification in the high ORP IACM tank, the partial nitrification in the low ORP tank had to result from the lower DO concentration range. There are two possible explanations for the improvements in the nitrification in the low ORP tank by both the acetate and methanol additions at relatively high dosages. As described in the literature review, the results from pure culture studies indicated that heterotrophic nitrifiers can successfully compete with autotrophic nitrifiers for ammonia and that both readily assimilated organic compounds and low DO conditions are two selecting conditions for heterotrophic nitrifiers. Considering the experimental conditions provided in the low ORP IACM tank, the increasing trends of nitrification with increasing substrate dosages suggest that heterotrophic nitrification must have dominated 108 the ammonia oxidation process in the low ORP IACM tank, at relatively high dosages of both acetate and methanol. At the zero and low substrate dosages, heterotrophic nitrifiers would not grow in a large number on such feed even under the low DO environment. Therefore, the inhibitory effect of low DO on autotrophic nitrifiers would result in incomplete nitrification in the low ORP IACM tank. With both acetate and methanol addition at the medium and high dosages and still under low DO conditions, heterotrophic nitrifiers would significantly increase and so would their activities, which led to the observed improvements in the nitrification in the low ORP IACM tank. As suggested by the BPR mechanism, in combination with the zero V F A concentration in the anaerobic zone, it was the stored carbon (e.g., PHA), not acetate, that was actually available to the low ORP IACM tank at acetate dosages less than 100 mgCOD/L. Therefore, it was the increase in the amount of stored carbon that actually caused the increase in the nitrification in the low ORP IACM tank. When acetate at a rate of 10 mg/L was bled into the low ORP IACM tank, the nitrification in this tank was even complete; this indicates that acetate could also improve heterotrophic nitrification. In summary, the above results suggested that acetate and methanol, as well as stored carbon reserves, all could be preferred substrates for heterotrophic nitrification. The consistently complete nitrification in the high ORP IACM tank at various substrate dosages could not provide any information for analysis of the nitrifier population in that tank. However, the significant differences in nitrification between the high and low ORP processes indicate that, in addition to the organic substrate, the ORP control range was another crucial factor for selecting the heterotrophic nitrifiers. That is, the low ORP 109 control, through maintaining low DO environment, favors the growth of heterotrophic nitrifiers. More complete discussion on the roles of ORP control and external substrates in the selection for bacterial populations will be given in Section 4.4. However, considering the features of the ORP control used in the current two-stage processes, there is another possible option for the improvements in the nitrification in the low ORP IACM tank. As the external substrate dosages increased, the air supply into the low ORP IACM tank should also be increased to maintain the same redox level. The increased air supply could improve the autotrophic nitrification in the IACM tank, thus explaining the increasing trends observed in Figures 4.18 and 4.19. Overall TKN Removal T K N removal can be attributed mainly to two reactions: ammonia assimilation and nitrification. The 3-Bardenpho process and the high ORP process both achieved about 90-95% T K N removal at various acetate and methanol dosages, since the nitrification in these two processes was always complete. The low ORP process achieved 52.6%, 63.1%, and 75% T K N removal at acetate dosages of 0, 30 and 50 mgCOD/L, respectively, because the nitrification was incomplete at the above dosages. At the high acetate dosage, the low ORP process achieved 94.4% T K N removal due to the complete nitrification. Similar percentages were also observed in the low ORP process at various methanol dosages: 65.2%, 85.9% and 87.5% at methanol dosages of 15, 30 and 60 mgCOD/L, respectively. The T K N removal efficiencies were also calculated for the runs under the wide/narrow ORP control regime. Both the narrow and wide ORP processes achieved 110 similar high percentage of T K N removal in the raw sewage run (92-93%) and in the methanol addition run (96-97%). However, at an acetate dosage of 30 mgCOD/L, both processes achieved partial T K N removal: 83.6% and 48.2% in the wide and the narrow ORP processes, respectively. 4.1.3.6 Denitrification in the Aeration Tank Based on the assumptions summarized in Section 4.1.3.5, nitrogen removal in this study was considered due mainly to denitrification in various zones and due to sludge wastage. Ammonia stripping was also generally possible in the three aeration tanks. However, the amount of stripping was considered negligible since pH was controlled in a range of 7.0-7.4 in all three aeration tanks by supplementing the extra 100 mgCaC0 3/L alkalinity into the feed. Although the amounts of N O x removal in the anaerobic zone and the 3-Bardenpho anoxic zone, as well as the amount of nitrogen removal due to sludge wastage were all carefully accounted for, a nitrogen balance on the entire process train still could not be achieved. For the two-stage process, the unaccounted for nitrogen loss in the total process train accounted for up to 50% of the influent T K N and could occur either in the anaerobic zone or in the IACM tank, or in both reactors. However, at this point, the loss is assumed to be due to denitrification that occurred in the aeration tank. This assumption will be verified in Section 4.3. For the 3-Bardenpho process, the unaccounted for nitrogen loss in the anoxic zone could be significant due to air entrainment in the anoxic zone (Section 4.1.3.4). However, the loss in the anoxic zone had been quantified; therefore, it would not affect the calculation for the loss in the aeration tank. ill As described in Section 4.1.1, the zero DO period did not exist in the IACM tank when the aeration was controlled by the three narrow ORP ranges. Therefore, the denitrification that occurred under such conditions could only be attributed to denitrification in the presence of DO. However, the denitrification in the wide ORP IACM tank could be attributed to denitrification in the presence or absence of DO, because both zero DO period and non-zero DO period existed in the aeration cycle. The amount of N O x removal in the aeration tank was calculated based on a total of 8 measurements (4 N O x concentrations, 3 T K N concentrations and the daily volume of sludge wastage). As expected, a large variance in the data sets was observed, because a small error in each measurement was accumulated through calculation. To increase the reliability of comparison, where a pair of replicate data sets were available, they were combined and the combined data set was used in comparison. The results of the t-tests on percentage of N O x removal in the aeration tank are presented in Table C.3. The percentages of N O x removal (based on the influent TKN) in the aeration tank, obtained from the acetate and methanol addition runs, are presented in Figures 4.21 and 4.22, respectively. As noted, this percentage accounted for up to 50% of the influent T K N in the IACM tank at low DO concentrations and normally 10-20% in the Bardenpho aerobic zone at a DO concentration of 3 mg/L. The percentage of denitrification in the aeration tank was the ratio of the amount of the unaccounted for nitrogen loss in that tank to the amount of N O x produced in that tank (i.e., the amount of nitrification). This percentage is plotted against the acetate and methanol dosages in Figures 4.23 and 4.24, respectively. 112 Figure 4.21 The percentage of NO, removal in the aeration tank obtained from the acetate addition runs. Methanol dosage (mg COD/L) Figure 4.22 The percentage of NO* removal in the aeration tank obtained from the methanol addition runs. 113 Comparison among the Two IACM Tanks and the 3-Bardenpho Aerobic Zone Acetate and methanol addition runs. As noted in Figures 4.21 and 4.22, the 3-Bardenpho aerobic zone and the low ORP IACM tank achieved 8.2% and 5.4%, respectively, higher N O x removal than the high ORP IACM tank at the zero external substrate dosage. Over the acetate dosage range 30 to 50 mgCOD/L and the methanol dosage range of 15-60 mgCOD/L, the percentage of N O x removal in the low ORP IACM tank was the highest (29-50%), second in the high ORP IACM tank (27.5-36%), and last in the 3-Bardenpho aerobic zone (10-21%). However, at the high acetate dosage, the highest percentage was found in the high ORP IACM tank (36%), second in the low ORP IACM tank (30%). The differences shown in the Figures 4.21 and 4.22 were statistically significant, except for the difference between the 3-Bardenpho aerobic zone and the high ORP IACM tank at an acetate dosage of 30 mgCOD/L (Table C.3). As noted in Figures 4.23 and 4.24, the percentage of denitrification was 35-50% higher in the low ORP IACM tank than in the high ORP tank over the entire acetate and methanol dosage ranges, except for the high acetate dosage. This indicates that the low ORP control favors denitrification, compared to the high ORP control. The percentages of about 80%, in combination with the low residual N O x concentrations, suggests that the denitrification in the low ORP IACM tank was completed at the low and medium dosages of both acetate and methanol. The low percentages of less than 50%, observed in both the high ORP IACM tank and the 3-Bardenpho aerobic zone, indicates that the denitrification in these two tanks was not completed over the entire acetate and methanol dosage ranges. 114 100 • 3-Bardenpho aerobic zone • Low ORP IACM tank x High ORP IACM tank 20 40 60 Acetate dosage (mg COD/L) 80 100 Figure 4.23 Percentage of denitrification in the aeration tank obtained from the acetate addition runs. 100 1 0 -1 1 1 1 1 1 1 0 10 20 30 40 50 60 Methanol dosage (mg COD/L) Figure 4.24 Percentage of denitrification in the aeration tank obtained from the methanol addition runs. 115 The runs under the wide/narrow ORP control regime. Figure 4.25 clearly shows that the percentage of N O x removal was about 30% higher in the wide ORP IACM tank than in the other two aeration tanks, in both the acetate (30 mgCOD/L) and methanol (15 mgCOD/L) addition runs. As noted in Section 4.1.2.3, in the methanol run of this group, 25% more methanol was added into the wide ORP IACM tank than into the other two aeration tanks. Therefore, the higher percentage obtained in the wide ORP IACM tank may be partially due to the extra methanol addition. In the raw sewage run, the wide ORP IACM tank achieved a similar percentage of N O x removal, compared to the other two aeration tanks. This was probably because the carbon limiting conditions were experienced in all three aeration tanks in the raw sewage run. Effect of External Substrate Dosage As shown in Figures 4.21 and 4.22, the difference between a pair of replicate runs was generally less than 5%. However, a large difference of 13% was observed in the pair conducted in the low ORP process, at an acetate dosage of 30 mgCOD/L. It was the same pair, in which a large difference (17%) was observed in overall T N removal efficiency. Excluding this unusual case, it can be concluded that the difference in percentage of N O x removal in the aeration tank, caused by the variations of uncontrolled environmental conditions, was normally less than 5%. 116 Sewage run Acetate run Methanol run Figure 4.25 Percentage of NO* removal in the aeration tank obtained from the runs under the wide/narrow ORP control regime. Acetate and methanol addition runs. As shown in Figures 4.21 and 4.22, the percentage of N O x removal in the 3-Bardenpho aerobic zone was a constant (15 to 20%) at various acetate and methanol dosages. This percentage in the two IACM tanks generally increased with increasing acetate and methanol dosages, except for the high acetate (100 mgCOD/L) and methanol (60 mgCOD/L) dosages. For example, a 15.5% decrease was observed in the low ORP IACM tank between the medium and high acetate dosages; a 15.3% decrease was observed in the high ORP IACM tank between the medium and high methanol dosages. Details concerning the significance of the above changing patterns can be found in Table C.4. Between the zero and low acetate dosages and between the zero and low methanol dosages, large increases were found in the low ORP IACM tank (16.3-19.4%) and in the 117 high ORP IACM tank (19-21.7%) (Table C.4). However, these large increases cannot be attributed to the increase in the external substrate dosage only, since, compared to the diluted sewage, the undiluted sewage contained 100 mg/L higher total COD; this may have contributed to the large increases observed, in part. As shown in Figures 4.23 and 4.24, for each aeration tank, the changing pattern of percentage of denitrification was the same as that of percentage of N O x removal. However, there is a difference between the two patterns. In the low ORP aeration tank, the methanol addition at the medium dosage began to cause a decrease in the percentage of denitrification, unlike the decrease in the percentage of N O x removal, which was observed only at the high dosage. Nitrogen loss in an aeration tank depends on the two sides of SND: nitrification side and denitrification side. Percentage of nitrification represents the nitrification side; percentage of denitrification represents the denitrification side. Further, percentage of N O x removal represents the combination of the two sides. The same changing patterns between the percentage of N O x removal and the percentage of denitrification, observed in this study, suggest that the N O x removal in the current aeration tanks was generally governed by the denitrification side. This was true in the 3-Bardenpho aerobic zone and the high ORP IACM tank, because the nitrification in these two tanks was always complete. However, that was not true in the low ORP IACM tank, because both the nitrification and denitrification in this tank generally were incomplete and any one of them could be the limiting factor for nitrogen removal. 118 Since carbon reserves in solids such as PHA were not measured in this study, except for one run, the following explanations for experimental results has not been proven. As described in Section 4.1.3.3, increasing the acetate dosage into the anaerobic zone would increase the amount of carbon reserves (e.g., PHA) in PAOs. The increase in the amount of stored carbon was believed to cause an increase in the N O x removal in the high ORP IACM tank, in the same manner of causing the increase in the N O x removal in the 3-Bardenpho anoxic zone. That is, carbon reserves improved the denitrification side. However, the improvement in the N O x removal in the low ORP IACM tank was believed to be caused by the increase in the amount of stored carbon, through improving both the nitrification and denitrification sides. The bleeding of acetate into the low ORP IACM tank could either be stored as PHA or inhibit the degradation of the carbon reserves already stored in the previous anaerobic zone. Both cases would limit the use of carbon for denitrification, thus causing the significant decrease in the percentage of N O x removal in this tank. Methanol is generally believed to be an effective substrate for denitrification. It was expected that methanol addition would improve the denitrification, and subsequently the N O x removal in the two IACM tanks, since methanol was also directly available to these tanks. However, significant decreases in percentage of denitrification was observed in the low ORP IACM tank at both the medium and high methanol dosages and in the high ORP IACM tank at the high dosage. These decreases were believed to be caused by the carbon storage, induced by the presence of methanol at relatively high concentrations. While a decrease in percentage of denitrification was observed in the low ORP IACM tank at the 119 medium methanol dosage, the percentage of N O x removal still increased. This resulted from the large increase in the nitrification side at the medium methanol dosage (see Section 4.1.3.5). More discussion related to the role of carbon storage in aerobic denitrification will be presented in Section 4.4.2. Both the constant percentage of denitrification and the constant percentage of N O x removal, in the 3-Bardenpho aerobic zone, regardless of the acetate and methanol dosages, were believed to result from carbon limiting conditions to this aerobic zone, since the external substrate would be utilized in the zones, into which they were added. Runs under the wide/narrow ORP control regime. Figure 4.25 shows that a higher than 35% increase in percentage of N O x removal was found in the wide ORP IACM tank between acetate dosages of 0 and 30 mgCOD/L and between methanol dosages of 0 and 15 mgCOD/L. However, no significant increases were found in the narrow ORP IACM tank in the above two dosage ranges. The above results indicate that the wide ORP IACM tank was able to use external substrates more efficiently for denitrification than the narrow ORP IACM tank. The explanations will be given in Section 4.4.1. 4.1.3.7 Summary This section presents a summary for the results regarding nitrogen removal, nitrification, and denitrification in the three processes. The results from Section 4.1.3.1 indicated that the two-stage process achieved a T N reduction of 40-80%, with an effluent T N concentration of 5.5-15 mgN/L, under various conditions. 120 Comparison between the 3-Bardenpho Process and the Low ORP Process Compared to the 3-Bardenpho process operated at an aerobic DO concentration of 3 mg/L, the low ORP process achieved 5-10% less overall T N removal, at various acetate and methanol dosages (Section 4.1.3.1). Further comparisons between these two processes indicated that the low ORP process achieved average 3.3% higher nitrogen removal due to sludge wastage (Section 4.1.3.5) and average 5.4% less N O x removal in the anaerobic zone(Section4.1.3.3). The above results further indicated: 1) the total N O x removal in the low ORP process would be 8.3-13.3% less than that in the 3-Bardenpho process; and 2) the N O x removal in the low ORP IACM tank would be 3-8% less than the sum of the N O x removal in the anoxic and aerobic zones. Effect of A verage ORP Level The comparisons between the two 2-stage processes revealed the effect of ORP control on nitrogen removal performance. Compared to the high ORP control, the low ORP control only slightly improved the overall T N removal efficiency (Section 4.1.3.1). The detailed comparisons showed that most of the N O x removal, achieved in the low ORP process, occurred in the IACM tank, while the high ORP process achieved a significant amount of N O x removal in the anaerobic zone (11-16% of the influent TKN). Furthermore, nitrification in the high ORP process was always complete and nitrogen removal was limited by incomplete denitrification. In contrast, both nitrification and 121 denitrification in the low ORP process were usually not complete and were limiting factors for nitrogen removal. Significant differences in the percentage of nitrification and the percentage of N O x removal in the IACM tank between the high and low ORP control range indicated that ORP can serve as a control parameter for the aeration in the two-stage process. Effect of External Substrate Increasing acetate and methanol dosages generally led to an increase in overall T N removal in all three processes and an increase in N O x removal in the anoxic zone and in the two IACM tanks, except for the high acetate and methanol dosages. The increasing dosage also led to an increase in the amount of nitrification in the low IACM tank and a small increase in the nitrogen removal due to sludge wastage for all three processes. However, the increasing dosages did not affect the N O x removal in the three anaerobic zones, since the denitrification in the anaerobic zone was limited by N O x itself. The increasing dosages did not affect the N O x removal in the 3-Bardenpho aerobic zone, since the external substrates added were not available to this zone and denitrification in this zone mainly relied on the RBCOD portion present in the sewage and the endogenous COD. In both the 3-Bardenpho anoxic zone and the high ORP IACM tank, the substrate added was used to enhance the denitrification. However, in the low ORP IACM tank, the added substrate was used to improve both nitrification and denitrification. The fact that 122 the high acetate and methanol dosages did not improve the denitrification in the 3-Bardenpho anoxic zone much was due to the fact that N O x in the anoxic zone was limiting at the high dosages. The poorer N O x removal in the two IACM tanks at the high acetate and methanol dosages appeared to be related to the carbon storage in the IACM tank (see Section 4.4.2). Comparisons among the 3-Bardenpho, the Wide and Narrow ORP Processes With a supplement of small amount of external substrates either in the anaerobic zone or in the IACM tank, the two-stage process was capable of achieving better nitrogen removal than the 3-Bardenpho process operated under the conditions set in this study. Compared to the 3-Bardenpho process, the wide ORP process achieved 10% higher T N removal at an acetate dosage of 30 mgCOD/L and at a methanol dosages of and 15 mgCOD/L. The detailed comparisons indicated the N O x removal in the wide ORP tank was about 10% higher than the sum of the N O x removal in the 3-Bardenpho anoxic and aerobic zones at the above two external substrate dosages. Compared to the narrow ORP range, the wide ORP control range improved the TN removal in total process by 15-30% and the N O x removal in the IACM tank by 30%; this indicates that, in addition to the average level, the breadth of QRP control range is also a very important parameter for aeration control. The 30% higher N O x removal achieved in the wide ORP IACM tank was due to the longer aeration cycle, as well as the presence of a zero-DO period in the aeration cycle. 123 4.1.4 Phosphorus The experimental results, related to phosphorus removal and ortho-P release and uptake, obtained from the steady-state study, are presented in this section. 4.1.4.1 Overall Phosphorus Removal The difference between the total phosphorus (TP) concentrations of the feed and the effluent is considered as the phosphorus removal of the system and should be balanced by phosphorus loss due to sludge wastage. To account for the phosphorus loss due to sludge wastage, TP concentration was also determined for the mixed liquor in the aeration tank, from which sludge was wasted. The overall TP removal efficiencies, obtained from the acetate and methanol addition runs, are presented in Figures 4.26 and 4.27, respectively. 100 T 90 ->. O c Ci 80 'o £ 70 -"5 > < > o E 60 --O k_ Q. H 50 o "5 > O 40 -30 \ • 3-Bardenpho process • Low ORP process x High ORP process -+-20 40 60 80 Acetate dosage (mg COD/L) 100 Figure 4.26 The overall TP removal efficiencies obtained from the acetate addition runs. 124 Comparison among the Three Processes Acetate addition runs. As noted in Figure 4.26, the overall TP removal efficiencies in the 3-Bardenpho process appeared to be 5-10% higher than those in the low ORP process at various acetate dosages. However, considering that the difference between a pair of replicates was up to 10%, the differences between these two processes were not significant, especially at the medium and high acetate dosages (50 and 100 mgCOD/L). Compared to the higher ORP process, both the 3-Bardenpho process and the low ORP process achieved approximately 10-15% higher TP removal, at the acetate dosages less than 30 mgCOD/L, and the same TP removal at dosages higher than 50 mgCOD/L. 30 -I 1 1 1 1 H : 1 0 10 20 30 40 50 60 Methanol dosage (mg COD/L) Figure 4.27 The overall TP removal efficiencies obtained from the methanol addition runs. Methanol addition run. As shown in Figure 4.27, in general, the highest TP removal efficiencies were found in the low ORP process, second in the 3-Bardenpho process, and 125 last in the high ORP process, over a methanol dosage range of 15-60 mgCOD/L, except at 15 mgCOD/L. At this dosage, the TP removal efficiency in the high ORP process was abnormally high. Figure 4.28 presents the individual data of the effluent ortho-P concentrations obtained from the three processes at that dosage (run 11). As noted, a large variation in the effluent ortho-P concentration was observed only for the high ORP process; this suggests that the process did not reach the steady state during this trial. It is noted that this is one of a few cases, in which steady state was not reached during the testing period. It also appeared that the 3-Bardenpho process and the low ORP process did reach the steady state during the testing period, as indicated by the constant effluent ortho-P concentrations during the final part of the testing period, at least. Figure 4.28 The individual data of the effluent ortho-P concentration from the three systems during run 11. —•—3-Bardenpho process —x— High ORP process • B — Low ORP process Date 126 Therefore, the high removal efficiency in the high ORP process in this run probably resulted from non-steady-state conditions. Because the preceding trial was done at an acetate dosage of 50 mgCOD/L, it was expected to take time for the microbial population to adapt to methanol. Effect of External Substrate Dosage on Overall TP Removal Figure 4.26 shows that the TP removal efficiencies in all three processes generally increased with an increasing acetate dosage. However, no increase in TP removal efficiency was observed in all three processes between the medium and high acetate dosages. As indicated by the low ortho-P concentrations in the three aeration tanks (0.1-0.2 mg P/L), the phosphorus limiting condition was experienced in all three processes at the high dosage. As shown in Figure 4.27, there was large increases (20-40%) in overall TP removal efficiency, between the 0 and 15 mgCOD/L methanol dosages, and small increases over the rest of the dosage range. The large increase could resulted from the difference in organic substrate (average 100 mg/L total COD) between the diluted and undiluted sewage (Section 4.1.2.2). Further, the f-tests (results not shown) confirmed that the small increases were not significant. Therefore, the above results suggest that methanol addition into the 3-Bardenpho anoxic zone and the two IACM tanks appeared not to affect the phosphorus removal directly. 127 Runs Under the Wide/Narrow ORP Control Regime The overall TP removal efficiencies, obtained from the runs under the wide/narrow ORP control regime, are presented in Figure 4.29. As noted, the removal efficiencies were mainly related to the external substrates added. About 20% higher removal efficiency was obtained for all three processes in the acetate addition run than in the other two runs. Within each process, no significant differences between the raw sewage run and the methanol addition run were found. Furthermore, among the three process, approximately 10% difference in TP removal was observed only in the sewage run. Sewage run Acetate run Methanol run Figure 4.29 Overall TP removal obtained from the runs under the wide/narrow ORP control regime. Effect of the NOx Level in the Return Sludge on the Overall TP Removal The overall TP removal efficiencies obtained from all runs are grouped into two: acetate and non-acetate groups; the data are plotted against the N O x levels in the return 128 sludge in group in Figure 4.30. The data points obtained from all acetate addition runs (under both ORP control regimes) show that the TP removal efficiencies were constant, regardless of the N O x level in the return sludge. However, the data points obtained from all sewage runs and all methanol addition runs (non-acetate) showed that the efficiencies decreased as the N O x level in the return sludge increased. The results suggest that the adverse effect of the N O x in the return sludge on TP removal appeared to be significant only in the runs without acetate addition. 100 £ 80 c 0) o o 60 TO > O o 40 Q. H TO V > O 20 r_i I_I y = 0.76x + 84.4 ———^"•^  u Rz = 0.116 • • ^ - — • • • — ^ A y = -4.6x + 94*2 • R 2 = 0.68 • • « Raw sewage and methanol addition runs • Acetate addition runs —i i i 1 1 i 1 1 2 4 6 8 10 NOx level in the return sludge (mg/L) 12 Figure 4.30 The overall TP removal efficiency vs. the corresponding NO, level in the return sludge. As described in the literature review, the degree of phosphorus removal is considered to be strongly dependent on the availability of specific substrates, such as VFAs or, more generally, RBCOD, to the anaerobic zone. The results shown in Figure 4.30 strongly suggest that the methanol addition into the IACM tanks had an indirect 129 effect on phosphorus removal. That is, methanol addition directly reduced the N O x level in the return sludge (i.e., the effluent N O x level), subsequently, improving the TP removal. In contrast, the results shown Figures 4.8 and 4.26 suggested that acetate addition directly improved both nitrogen and phosphorus removal. The amounts of TP removed in this study were in a range of 1.6-2.7 mgP/L for the raw sewage runs and in a range of 3.2-4.3 mgP/L for the runs supplemented with various acetate and methanol dosages. Any overall phosphorus removal in excess of 1.0-1.5 mgP/L for every 200 mg/L of COD removed, can be classified as BPR, since it is generally accepted to be the normal metabolic requirement for the cell growth in such activated sludge systems (US EPA, 1975). On this basis, excess phosphorus removal appeared to be taking place during the entire study, even in the first two runs, in which the diluted sewage was used. 4.1.4.2 Phosphorus Release and Uptake In this study, phosphorus concentrations in the raw sewage and the effluent were measured both in terms of TP and ortho-P. However, only ortho-P was measured in the various bioreactors, since no simple technique was available for measuring unfiltered TP without including the phosphorus present in the solids. For this reason, only an ortho-P balance can be performed on each individual reactor. It should be noted that using only the ortho-P balance to calculate the amounts of ortho-P release and uptake by biomass is subject to an error, because the complex phosphorus present in the wastewater can also be transformed into ortho-P in various zones. However, the extreme values of phosphorus 130 release and uptake could be calculated, if an assumption were made on which reactor caused the complex phosphorus to be completely transformed to simpler ortho-P. Manoharan (1988) calculated various extreme values of phosphorus release and uptake and found that the difference among extreme values was not large. Therefore, in this study, it is assumed that the complex phosphorus was completely transformed into ortho-P in the anaerobic zones. That is, the difference between the total phosphorus and ortho-P of the influent was added to the ortho-P balance on the anaerobic zone. It should be noted that, with this method, the amount of ortho-P released by biomass was underestimated. Although the daily wastage of ortho-P could be included in the ortho-P balance, it was not included in this study, because its contribution was insignificant (due to the volume of wastage being very small compared to the daily influent flow). Phosphorus Release in the Anaerobic Zone The anaerobic zone is considered to be one in which neither DO nor N O x is present. However, as described in Section 4.1.3.2, N O x was introduced into the three anaerobic zones through the sludge return flow during most of the runs in this study. Acetate addition runs. The amount of ortho-P release in the anaerobic zone, with respect to the acetate dosage, is shown in Figure 4.31. In general, the highest amount of release was observed in the low ORP process, second in the 3-Bardenpho process, and last in the high ORP process. Compared to the low ORP process, the smaller amount of ortho-P release observed in the 3-Bardenpho process and the high ORP process at each dosages was believed to be due to the competition from denitrification for acetate. As described in 131 Section 4.1.3.2, significant higher amount of N O x was removed in the anaerobic zones of the 3-Bardenpho and the high ORP processes than in the anaerobic zone of the low ORP process. Acetate dosage (mg COD/L) Figure 4.31 The ortho-P release in the anaerobic zone obtained from the acetate addition runs. Also shown in Figure 4.31, ortho-P release increased almost linearly with the acetate dosages; this indicates that ortho-P release depended mainly on the presence of acetate. However, a large deviation from the trend line was found in the low O R P process at the high acetate dosage. As described in the literature review, the phosphorus release reaction can occur even under aerobic conditions, when preferred substrates, such as acetate, are present in significant concentration. The observed deviation was believed to be due to the phosphorus release in the IACM tank, caused by the carryover of VFAs (10 mg/L) into 132 that tank. However, it was impossible to measure the amount of phosphorus release in the IACM tank because phosphorus uptake dominated the ortho-P change in the IACM tank. It was also observed that the phosphorus content in sludge from the low ORP process was lower at this high acetate dosage than at the medium dosage, as shown in Figure 4.34. Therefore, it is suggested that the release in the IACM tank resulted in the lower phosphorus content, subsequently, led to the less phosphorus release in the anaerobic zone. The adverse effect of high acetate concentrations on biological phosphorus removal in a three-stage Bardenpho process was also reported by Randall and Chapin (1997). It was found that high concentrations of sodium acetate can cause failure of the process, with the progression of failure being typical of bacterial washout, due to a carryover of high acetate concentration, through anoxic zone, into the aerobic zone. The importance of limiting the nitrates in the feed or in the return activated sludge for BPR systems was addressed in the literature review (Barnard, 1974; Simpkins and McClaren, 1978; Rabinowitz, 1985). It is believed that the denitrification of N O x in the anaerobic zone utilizes the substrate that would otherwise be available to the PAOs. The results shown in Figure 4.31, whereby the amount of phosphorus released in the anaerobic zone linearly increased with the acetate dosage and was inversely proportional to the amount of denitrification in that zone, confirms the conclusion from the literature. However, it was noticed that the denitrification in the anaerobic zone had less adverse effect on overall phosphorus removal, as indicated by the results shown in Figure 4.26; that is, the three processes achieved similar phosphorus removal efficiencies at 133 relatively high acetate dosages (50 and 100 mgCOD/L). Therefore, it can be concluded that phosphorus removal could still occur at high removal efficiency, regardless of the nitrate level in the return sludge, if sufficient amount of VFAs is presented in the anaerobic zone. Methanol addition runs. Firstly, it should be noted that methanol was added into the 3-Bardenpho anoxic zone and into the two IACM tanks directly. Figure 4.32 presents the individual data points of the ortho-P release in the anaerobic zone, obtained from runs 5 and 6, during which the three processes were supplemented with methanol at dosages of 60 and 30 mg COD/L, respectively, for 50 days in total. As noted, the levels of the release in the three processes decreased from the high levels in the preceding acetate trial (dosage at 50 mg COD/L) to the new low levels in about 15 days. Furthermore, for an additional 35 days of acclimation with methanol, the releases remained at the low levels, except for a few unusual points. The amounts of ortho-P release are also plotted against methanol dosage in Figure 4.33. As noted, except for one unusual point at a methanol dosage of 15 mgCOD/L in the low ORP process, the amounts of release were generally less than 3 mg per liter of the feed. Unlike the results from the acetate addition runs, both the differences among the three processes and the increasing trend with respect to methanol dosage were not observed. This suggests that methanol addition into the IACM tanks or the anoxic zone could not improve phosphorus release in the anaerobic zone. 134 Date Figure 4.32 The individual data of the ortho-P released in the anaerobic zone during runs 5 and 6. -1 -I 1 1 1 1 1 1 0 10 20 30 40 50 60 Methanol dosage (mg COD/L) Figure 4.33 The ortho-P release in the anaerobic zone obtained from the methanol addition runs. 135 Phosphorus Uptake in the Anoxic Zone Net uptake occurred in the anoxic zone during the entire experiment; the amount could account for up to 90% of the total uptake. The amounts of uptake, with respect to the acetate and methanol dosages, are presented in Figure 4.34. As noted, the amounts increased with an increasing acetate dosage, but was independent of methanol dosage. It should be noted that the observation, showing that the amount of uptake in the anoxic zone was not associated with methanol dosage, was based on an acclimation period of 4 SRTs (total time for two consecutive methanol runs, such as runs 5 and 6, and runs 11 and 12). Acetate and methanol dosages (mg COD/L) Figure 4.34 The amounts of phosphorus uptake in the 3-Bardenpho anoxic zone vs. the external substrate dosages. The anoxic zone is one in which the electron acceptor is available in the form of N O x and not as DO. As described in the literature review, a subpopulation of PAOs is capable 136 of denitrification, that is, they use nitrate as an electron acceptor, instead of DO, for oxidation of carbon reserves and synthesis of polyphosphate. Other PAOs that can not use N O x as electron acceptor, are expected to release phosphorus, provided that sufficient readily biodegradable substrates are available. In the acetate addition runs, there were no carryover of VFAs from the anaerobic zone to the anoxic zone, even at the high acetate dosage. Therefore, large net anoxic uptake was expected in the 3-Bardenpho process. The near linear increase of the uptake with an increasing acetate dosage indicated that the amount of uptake was associated with the oxidation of PHA reserves, since the amount of carbon reserves was expected to increase, as the acetate dosage into the anaerobic zone increased. Generally, it was the PHA, that limited the ortho-P uptake in the anoxic zone, since the N O x level and ortho-P level in the anoxic zone was usually greater than 0.5 mg/L and 1.0 mg/L, respectively. As discussed in the literature review, the additions of preferred substrates, such as acetate and propionate, under anoxic conditions can cause an ortho-P release. No ortho-P release in the current 3-Bardenpho anoxic zone, with the presence of methanol for up to a 4 SRT (60 days) suggests that methanol is not a preferred substrate for PAOs. Phosphorus Uptake in the Aeration Tank Under aerobic conditions, the acclimated biomass takes up ortho-P from solution. Barnard (1984) and Nicholls et al. (1987) have indicated that long anaerobic HRTs will lead to excessive secondary ortho-P release without uptake of VFAs and that this secondary release will lead to no direct relationship between the phosphorus uptake in the 137 aerobic phase and the phosphorus release in the preceding anaerobic phase. However, for systems that use short anaerobic HRTs, the secondary release in the anaerobic zone can be minimized, and direct relationships exist in these systems (Manoharan, 1988; Gibb, 1995). In the current three systems with a anaerobic HRT of 0.75 hour, for all data points (n = 39) obtained for the phosphorus uptake and the phosphorus release, the linear regression analysis gave the following relationship between them. It is noted that for the 3-Bardenpho process, the sum amount of uptake in the anoxic zone and in the aerobic zone was used in developing the Equation 4.4. P uptake (mg/L) = 1.05 x P release + 2.92 (mg/L) (4.4) (where R2 = 0.988) However, in terms of the two coefficients, the relationships obtained from different studies are different, depending mainly on the following factors: 1) the MLSS level, 2) the significance of secondary release, and 3) the ratio of the preferred substrate to the feed phosphorus. As shown in Figure 4.45 in Section 4.1.8.2, a small amount of ortho-P release was observed in the IACM tank during the air-off period. Further, as noted earlier in this section, using only an ortho-P balance to calculate the amounts of release and uptake is subject to an error. Therefore, unless the amount of secondary release is determined and a TP balance is performed, the linear relationship shown by Equation 4.4 can only be used in quality rather than in quantity. It should also be noted that the biomass specific amounts of uptake and release could not be calculated in this study, because the MLSS level in the 138 anaerobic zone was not determined for most of the runs and was not always the same as that in the aeration tank. 4.1.4.3 Phosphorus Content in Sludge The ortho-P taken up from the solution during the aerobic or anoxic phase is stored in the sludge and physically removed from the system through wasting. Therefore, the phosphorus content in the aerobic sludge is a good indication of the extent of BPR: Since the soluble TP (<3 mg/L), was a small part of TP (about 100 mg/L) in the aerobic mixed liquor, the phosphorus content in aerobic sludge could be calculated by dividing the aerobic mixed liquor TP concentration by the aerobic MLSS concentration. The phosphorus contents in the aerobic sludge obtained in this study were in a range of 1.8-4.0%. The phosphorus contents obtained from the acetate addition runs are presented in Figure 4.35. As noted, the phosphorus content in all three processes generally increased with an increasing acetate dosage. However, a 0.3% decrease was observed in the low ORP process between the medium and high dosages. The decrease was believed to be due to an ortho-P release in the low ORP IACM tank caused by the bleeding of acetate into the tank at the high acetate dosage. Compared to the sludge in the two 2-stage processes, the sludge in the 3-Bardenpho process appeared to contain a higher phosphorus content over the entire acetate dosage range. However, the low ORP process could surpass the 3-Bardenpho process at the low and medium acetate dosages. Compared to the sludge in the high ORP process, the sludge 139 in the low ORP process contained 0.7% and 0.2% higher phosphorus at the low and medium acetate dosages, respectively, and the same content at the zero and high dosages. 2 . 0 -I 1 1 : 1 1 1 0 2 0 4 0 6 0 8 0 1 0 0 Acetate dosage (mg COD/L) Figure 4.35 The phosphorus contents in the sludge for the acetate addition runs. The phosphorus contents obtained from the methanol addition runs are presented in Figure 4.36. As noted, a significant increase in a range of 0.4-0.9% was observed in all three processes between the zero and low methanol dosages. However, further increasing methanol dosage did not lead to significant increases in phosphorus contents in all processes. If the phosphorus content obtained from the low ORP process at the low methanol dosage were considered abnormal, both two-stage processes would have a consistently lower (about average 0.5%) phosphorus content over the entire methanol dosage range, compared to the 3-Bardenpho process. The consistently lower content had to be due to the less effective ortho-P uptake in the IACM tank, under the low DO and 140 intermittent aeration conditions. More discussion regarding to the ortho-P uptake in the IACM tank is given in Section 4.1.8.2. 3.4 2.2 -I 1 : 1 1 1 1 1 0 10 20 30 40 50 60 Methanol dosage (mg COD/L) Figure 4.36 The phosphorus contents in the sludge for the methanol addition runs It was noted (Hoffmann and Marais, T977) that a phosphorus content in sludge above 1.5% is a direct indication of the degree of BPR. The results from this study that the phosphorus contents in the aerobic sludge in all three processes were above 1.5% suggested that BPR occurred during the entire study, including the raw sewage and methanol addition runs. It was suggested by Siebritz et al. (1983) and Manoharan (1988) that at least about 25 mg/L of RBCOD in the anaerobic zone is required for any BPR to occur. The results from newer work indicated that a feed VFA/P ratio of 3 or 4 to 1 is required for total P removal, and 10 mg/L of V F A should still remove 2.5 to 3.5 mg/L P (Oldham, 1997). 141 Manoharan (1988), in his BPR study, conducted the experimental runs with external substrates, starting with the highest dosage and subsequently reduced in steps. He observed that the phosphorus removal continued to occur for a short period of time (1-5 days) at the level corresponding to the higher dosage, before decreasing to a lower level, corresponding to the new reduced dosage. Further, Manoharan observed: 1) until the aerobic PHB concentrations were reduced to the new lower steady-state values, the effluent phosphorus concentration continued to maintain the same low level; 2) once the aerobic PHB concentrations reached the new steady-state values, the effluent phosphorus concentration started to increase towards the new higher steady-state levels. Therefore, he suggested that it was the internal carbon reserves, already stored in the aerobic sludge, that helped to maintain the same phosphorus uptake in the aerobic zone. On the contrary, Manoharan also observed that the excess phosphorus removal increased immediately without the presence of any significant lag period, when the dosage of the preferred substrates was increased. The same phenomenon was also observed in the full-scale BPR treatment plant in Kelowna, B.C. (Oldham, 1985a). Based on the findings from Manoharan (1988), the high phosphorus contents in the sludge obtained from the raw sewage and methanol addition runs can still be explained by the BPR mechanism, provided that the diurnal variation of the VFAs in the feed bucket was taken into consideration. The fresh sewage used in this study contained an average of 20 mg/L of VFAs. Therefore, it was believed that BPR would occur immediately in all three processes, when they were fed with fresh sewage. As described in Section 3.4.3, 142 sampling was always conducted at 20 hours after the replenishment of the feed bucket. At this time, all the VFAs in the feed bucket had been completely consumed. Therefore, VFAs were not available to the three anaerobic zones at this time. Significant phosphorus release was not observed in the three anaerobic zones at this time either. However, this did not mean that BPR ceased. Based on the findings from Manoharan (1988), the phosphorus uptake in the three aeration tanks still continued through the oxidation of the existed internal carbon reserves, which were synthesized during the period while VFAs were available to the anaerobic zone. 4.1.5 Carbon The experimental results, related to COD removal and carbon release in the anaerobic zone, are presented in this section. 4.1.5.1 Overall COD Removal The total COD removal efficiencies obtained from all runs were in a range of 75-95%. The effluent filtered COD concentrations are presented in Figure 4.37 for each run. As noted, there were no significant differences, in effluent filtered COD concentration, among the three processes in all runs; this indicates that the three processes actually achieved the same level of organic carbon removal. However, a time variation of approximate 35 mg/L was found. Because the variation patterns were the same for all three processes, the variation had to be related to some common, but unknown, factors to the three processes. 143 0> c V 3 20 10 3= UJ —•—3-Bardenpho process — B — Two-stage process A —x—Two-stage process B H h + 6 7 8 Run No. 10 11 12 13 Figure 4.37 Comparison of the effluent filtered COD concentration among the three processes. 4.1.5.2 Carbon Loss in the Anaerobic Zone Acetate Addition Runs As described in the literature, in the anaerobic zone of BPR processes, there are three basic reactions governing the V F A loss: fermentation, denitrification, and V F A uptake by PAOs. It is believed that the denitrification of N O x in the anaerobic zone utilizes the VFAs that would otherwise be available to the PAOs. In the current three anaerobic zones, to relate the VFAs loss to the denitrification and ortho-P release, the following assumptions were made. 1) Due to the use of a short (0.75 hours) anaerobic HRT in the three processes, the V F A production by fermentation was considered to be negligible. 144 2) In the presence of enough VFAs, denitrification would consume VFAs only; the amount of denitrification using slowly biodegradable COD was considered to be negligible. 3) The amount of V F A uptake by PAOs was directly proportional to the amount of ortho-P release, since the secondary release in the anaerobic zone with a 0.75 hour HRT was negligible, compared to the ortho-P release with concomitant of V F A uptake. For all the data points (n = 16) obtained for the ortho-P release, denitrification, and the V F A loss, in the anaerobic zone, the two-variables linear regression analysis gave the following relationship among them. VFAs Loss =7.5 xNOx Denitrified + 1.8 xP Release (4.5) (where R2 = 0.92) Equation 4.5 has two meaningful coefficients. The first coefficient, associated with the N O x denitrified, indicates that one mg of N O x denitrified would oxidize 7.5 mg VFAs as COD in the anaerobic zone; this is slightly lower than the number of 8.6 mg/L RBCOD obtained by Ekama and Marais (1984) under similar conditions, but significantly higher than the value of 3.6 mgCOD/L VFAs obtained by Rabinowitz (1985) in batch tests. The second coefficient, associated with phosphorus release, indicates that one mg of phosphorus released would store 1.8 mg VFAs as COD. This represents a molar ratio of 1.5 moles of phosphorus released per mole of acetate sodium stored. It is much lower than the values of 2.6 found by Manoharan (1988) in a bench-scale continuous-flow BPR system, but close to 1.76 and 1.4 found by Rabinowitz (1985) and Comeau (1987) in 145 batch tests. As pointed out by Manoharan himself, the higher ratio that he obtained might be due to the presence of higher than the added concentration of acetate, resulting from the fermentation of the raw sewage in the anaerobic zone. It should be noted that the HRT of anaerobic zone used in Manoharan's study was 1.5 hour, which doubled the anaerobic HRT used in the current systems. Methanol Addition Runs The amounts of filtered COD loss in the anaerobic zone, with respect to the methanol dosage, are presented in Figure 4.38. As noted, 1) both two-stage processes produced a higher amount of filtered COD (negative values representing production) in their anaerobic zones at various methanol dosages, compared to the 3-Bardenpho process; 2) the amount of filtered COD production in the anaerobic zone appeared to increase with an increasing methanol dosage added into the IA aeration tank or the Bardenpho anoxic zone. T-tests were conducted to verify the results shown in Figure 4.38. The results of the tests confirmed the first observation listed above. However, the results indicated that the second observation was significant only between methanol dosages of 30 and 60 mgCOD/L and only in the two-stage process. Therefore, it is concluded that methanol addition, only at the high dosage (60 mgCOD/L), caused a significant filtered COD production in the anaerobic zone of the two-stage process. 146 £ a> 20 -, 10 .E *-o 2 o 0 Q O * O a £ -10 -20 «J CD •= c *- o « N 2 » C .Q 3 O ° a> E <c < s *r -30 -40 -50 -60 -70 • y = -0.44x+ 15.8 • —• R 2 = 0.91 w • ~ ~i> y = -0.83x - 0 . 8 8 ^ - ^ • 3-Bardenpho process • Two-stage process R2 = 0.64 1 1— , 1 T 10 20 30 40 50 Methanol dosage (mg COD/L) 60 Figure 4.38 The amounts of filtered COD loss in the anaerobic zone obtained from all methanol addition runs. As described in the literature, PHB reserves will accumulate when cells are limited in oxygen or in nitrogen but still have a carbon source available (Dawes and Senior, 1973). Several researchers (e.g., Suzuki et al., 1986a; 1986b; Ueda et al., 1992; Kang et al., 1993) recently reported that aerobic-growth, facultative methylotrophic bacteria (also denitrifiers) produced PHB from an inorganic mixture containing methanol under nitrogen-limiting conditions and that the C/N ratio was considered as a governing factor for the PHB production. In the current continuous-flow systems, the ammonia concentration in the anaerobic flow (the flow from the anaerobic zone to the IACM tank =1.5 times of the influent flow) were approximately 10 mgN/L. Direct methanol addition into the IACM tank only during 147 the air-off period, at a dosage of 60 mgCOD/L relevant to the influent flow, would give the mass ratio of methanol to ammonia to 8 during the air-off period. Further, both IACM tanks were operated under low DO conditions. Under such environmental conditions, methanol was very likely to be stored in the biomass as PHB in the IACM tanks. Since the biomass that contained stored carbon (PHB) was recycled into the anaerobic zone, the stored carbon was released into the solution again. The release in the anaerobic zone could result from lysing since there would be a high oxygen demand (Barnard, 1997). Since electron acceptors (molecular oxygen and nitrate-bound oxygen) were limiting in the anaerobic zone, oxidation of the released carbon was not possible. Therefore, the net result would be a significant filtered COD production in the anaerobic zone. 4.1.6 MLSS and Volatile Content The volume of sludge wasted was adjusted based on the effluent TSS level, for maintaining a system SRT of 15 days in all three systems. The MLSS levels in the three aeration tanks during the entire study were in a range of 2000-4000 mg/L. Significant decreases in the MLSS concentration in all three processes were found during run 1, after the feed was switched from the raw sewage to the diluted one (Figure 4T); significant increases were observed during run 3, after switching the feed back to the raw sewage. In response to the acetate dosage increase from 30 to 100 mgCOD/L (between runs 7 and 8), a significant increase in MLSS concentration was observed in all three processes during the transition period. Other than the above three significant changes, the MLSS levels in each run were fairly constant. 148 Figure 4.39 shows that the MLSS levels increased as the amount of total COD removed in the total process increased. Two trend lines are shown in Figure 4.39: one for the 3-Bardenpho process and the other for the two 2-stage processes. The difference between the two trend lines indicates that the MLSS level was maintained at an average 480 mg/L higher, in the two-stage process, than in the Bardenpho process, by the same amount of total COD consumed. 4000 1600 ~\ 1 1 1 1 1 1 150 200 250 300 350 400 450 Total COD removed in the process (mg/L) Figure 4.39 The MLSS levels in the aeration tank vs. the amount of total COD removed in the total process. The volatile content of the aerobic sludge (MLVSS/MLSS%), obtained from all runs except for the two replicate runs, is presented in Figure 4.40. The reason for excluding the data from the two replicate runs was that the three systems during the two replicate runs were not operated under the same feed conditions. As noted in Figure 4.40, 149 the volatile content generally ranged from 71% to 80%; consistently 1-3% higher volatile content was observed in the two 2-stage processes than in the 3-Bardenpho. The time variation in volatile content was the same for all three processes, indicating the variation had to be caused by some common factors. 70 -\ 1 1 1 —I 1 1 1 1 1 ' 1 2 3 4 5 6 7 8 9 11 12 Run No. Figure 4.40 Comparison of the volatile content of the aerobic sludge among the three processes. The higher MLSS level and its higher volatile content, maintained in the two IACM tanks than in the 3-Bardenpho aerobic zone, under the same feed conditions, clearly indicate that less endogenous oxidation of bacteria cells occurred in the two-stage process. The less endogenous oxidation was probably caused, in part, by the low DO and intermittent aeration conditions in the two IACM tanks. 4.1.7 Sludge Volume Index (SVI) and Effluent Suspended Solids SVI Because of the limitation of the mixed liquor due to the small scale of the experimental setup, 400-500 mL, instead of 1000 mL, mixed liquor was used in the determination of SVI. The SVI in this study was the volume occupied by one gram of the aerobic mixed liquor after settling for 30 minutes in a 500 mL graduated cylinder. The SVIs for each experimental run are presented in Figure 4.41. 30 -I 1 1 1 1 1 1 1 1 1 1 1 1 1 2 3 4 5 6 7 8 9 10 11 12 13 Run No. Figure 4.41 Comparison of the aerobic mixed liquor SVI among the three processes. As noted, the SVI levels from the 3-Bardenpho process were usually less than 120 mL/g, except for the last three runs. The SVI levels from the two 2-stage processes were in a wide range of 60-300 mL/g. Furthermore, within each individual run, the sludge from the two IACM tanks consistently showed a substantially higher SVI level than that from 151 the 3-Bardenpho aerobic zone; this generally indicates that the two-stage process produced sludge with worse settling characteristics. The high SVIs observed in the 3-Bardenpho process during the last three runs were probably due to the presence of some substance, which could cause growth of the filamentous organisms, in the sewage, since, during this time period, sludge bulking was also observed in the UBC BNR pilot plant, which was fed with the same sewage. The poor SVIs in many single-stage IA processes appear to be related to the growth of filamentous bacteria caused by the low food to biomass (F/M) ratios (Landine, 1991; Lo et al., 1994; Gabb et al., 1996a; 1996b) and low DO tension in the aerobic zone (van Huyssteen, 1990). Landine (1991) observed poor sludge settleability in his single-stage IA systems and found that the addition of an anoxic selector to the process was an effective means to improve sludge settleability; this indicates that the sludge bulking that occurred in his systems was a low F / M type. Gabb et al. (1996a; 1996b) studied filamentous organism bulking in single continuously-fed, completely-mixed, intermittently-aerated reactor systems. They found: 1) intermittent aeration conditions sustained the growth of the low F / M ratio (long SRT) filaments; 2) changing from intermittent to continuous aeration ameliorated the low F/M filament bulking without a selector; and 3) incorporation of correctly sized aerobic selector did not control the proliferation of the low F / M filaments. Finally they suggested that the influent RBCOD does not play an important role in the proliferation of low F / M filaments. In this study, the use of a small anaerobic selector (0.75 hour) was found not to be effective in preventing sludge bulking. 152 Another explanation for sludge bulking is associated with the growth of filamentous bacteria, caused by low DO or intermittent aeration conditions. As indicated by the typical DO profiles (Figures 4.3 and 4.4), the DO concentration in the IACM tank was always in a low range. Under such conditions, the filamentous organisms would have an advantage, since they protrude out of the floe to make use of the low oxygen concentration outside the floe (Hao et al., 1983; Lau et al., 1984; van Huyssteen, 1990). The low SVI observed in the 3-Bardenpho aerobic zone, in which the DO concentration was maintained at a high level, also supports this explanation. Effluent SS Level A high SVI does not necessarily lead to a high effluent SS concentration (van Huyssteen, 1990). The effluent SS concentrations during the entire study are presented in Figure 4.42. As noted, an effluent SS concentration less than 12 mg/L was achieved in all three processes, during most of the runs, despite the high SVIs in the two-stage process. In addition, the effluent SS levels (less than 8 mg/L) from both two-stage processes during the high SVI period (runs 6-11) were even lower than those from the 3-Bardenpho process during the same period, during which lower SVIs were found. It is generally accepted (Sezgin et al., 1978; Palm et al., 1980; Lee at al., 1983) that the settling characteristics of the floe are influenced by the ratio of filaments to floc-formers. Too low a level of filaments results in small dispersed floes. Elevated levels of filaments, protruding from the floe, will produce light, but cohesive, sludge. This sludge has the tendency to enhance the attachment and entrap smaller floes. 153 As indicated by the SVI data, the low DO and intermittent aeration conditions in the IACM tank must have resulted in the growth of a larger number of filaments. This sludge did not lead to washout of solids, because the large clarifier (nominal HRT = 5 hour) used in this study provided enough time for it to settle. During the slower settling process, this sludge entrapped small floes, thus resulting in the lower effluent TSS level, compared to the sludge from the 3-Bardenpho process operated at an aerobic DO concentration of 3 mg/L. Therefore, it is concluded that clear effluents can be produced at high SVI values, but at a cost of a much large tank and/or long HRT. 30 n 1 2 3 4 5 6 7 8 9 10 11 12 13 Run No. Figure 4.42 Comparison of the effluent total SS concentrations among the three processes. 4.1.8 Process Dynamics under the Steady-State Conditions Unlike the 3-Bardenpho process, the two-stage processes, due to a continuous alteration between air-on and air-off conditions, may never reach a true steady state with 154 respect to process parameters, such as ammonia, N O x and ortho-P. Therefore, it is of importance to understand the intermittent aeration process dynamics, if process optimization is to be achieved. The process dynamics in the two IACM tanks was studied by tracking the concentration changes during one aeration cycle for the following parameters: ammonia, NO x , ortho-P, and filtered COD. The results from the tracking study conducted during steady-state run 9 are presented in this section as an example. 4.1.8.1 Nitrate and Ammonia Figures 4.43 and 4.44 present typical ammonia and N O x concentration profiles during the short and long aeration cycles, under control by the narrow and wide ORP ranges, respectively. Time (min) Figure 4.43 Typical ammonia variations during long and short aeration cycles. 155 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (min) Figure 4.44 Typical NOx variations in long and short aeration cycles. As noted, the ammonia and nitrate concentrations varied randomly in a narrow range throughout the entire short cycle; no significant changes, corresponding to the air on and off, could be detected. This indicates that the narrow ORP IACM tank was in a steady state, without an occurrence of sequential nitrification and denitrification; in other words, nitrification and denitrification occurred simultaneously. In contrast, ammonia concentration was found to decrease during the air-on period and to increase during the air-off period, in the long aeration cycle; opposite changes were observed for the N O x concentration, respectively. In a modeling study for IA processes, Batchelor (1982) suggested that the IACM reactor would behave just like a continuous-aeration process at a true steady state, as the cycle time ratio (CTR = aeration cycle time/HRT) approaches zero. The CTRs used in this 156 study were approximately 0.013 and 0.055 for typical short and long cycles, respectively. As indicated by the ammonia and N O x concentration profiles, the steady state was reached at a CTR of 0.013, but not at a CTR of 0.055. 4.1.8.2 Ortho-P The track concentrations of ortho-P in both short and long aeration cycles are presented in Figure 4.45. As noted, the cycle of ortho-P uptake and release, corresponding air-on and air-off, existed in both aeration cycles. However, the magnitude of uptake and release was very small (less than 0.13 mgP/L). As indicated by the corresponding nitrate traces (Figure 4.44), the uptake and release cycle occurred in the presence of N O x (1.4, 2.0 mg/L). ra c o c fl> o c o o (0 £ Q. 0) O .C a o .c t O 0.34 0.30 0.26 0.22 air on / « Short aeration cycle • Long aeration cycle 0.18 ~~ air off n j LJy air on \ n X • \ LJ J X . • c\ • 1 1 1— —i 1 h • 1— LJ — I — — I — — i 1 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (min) Figure 4.45 Typical ortho-P variations in long and short aeration cycles. 157 The significant ortho-P uptake and release cycle (up to 1 mgP/L) was observed in a single-stage IA process by Huang (1993). It was noted by the author that the release was significantly enhanced when nitrate concentration fell below 0.5 mgN/L, where anaerobic conditions actually started. Ip et al. (1987) also observed higher effluent phosphorus concentrations from the IACM system than the fully aerobic control system, possibly caused by the phosphorus release during the air-off period. It should be noted that the time of air-on/off periods used in the above two studies were 1.5/1.5 hours and 2.25/2.75 hours. It was suggested that these long air-off periods had resulted in anaerobic conditions in the IACM tank, subsequently causing significant phosphorus release. In contrast, by controlling air on/off time with a 15/15 minute cycle, Lo et al. (1994) observed an effluent ortho-P concentration less than 0.1 mg/L in their ORP controlled, two-stage IA process. Therefore, for two-stage IA processes that incorporates an anaerobic selector for phosphorus removal, a long air-off period should be avoided in the IACM tank. Schon et al. (1993) observed both phosphate release and uptake in activated sludge under DO limiting conditions. Based on his observations, he hypothesized that, under the low DO conditions, phosphate uptake generally occurs on the surface of the sludge floe, while in the interior, phosphate is released. Further, he suggested that both reactions depend on organic substrates present in the sludge; the organic substrates are believed to effect the DO concentration gradient within the sludge floes. For example, in activated sludge, to which no substrate was added, phosphate uptake began at minimum DO concentrations (between 0.0 and 0.1 O2/L). Under limiting DO concentrations, nitrate 158 reduced the phosphate release in the sludge from lowly loaded plants, but not in the sludge from highly loaded plants (without nitrification). An anaerobic microzone was likely to form inside of the floes in the current two IACM tanks, especially during air-off periods. The small release shown in Figure 4.45 occurred in the presence of N O x and with absence of VFAs. Under such conditions, ortho-P release msut have occurred in the anaerobic microzone inside of the floes. Based on the ortho-P profiles, the rate of ortho-P release was approximately 0.34 mgP/gMLVSSxh"1 in the presence of NO x . As noted in Sections 4.1.3.1 and 4.1.3.6, a relatively long aeration cycle, resulted from the wide ORP control, appeared to be favorable to the overall T N removal in the two-stage process. The results from this section indicate that the use of a short aeration cycle is an effective means to prevent the phosphorus release in the IACM tank during the air-off period. Therefore, the optimal aeration cycle time for a two-stage IA process, designed for both nitrogen and phosphorus removals, will probably be determined by the requirements for phosphorus removal rather than those for nitrogen removal. 4.1.8.3 Chemical Oxygen Demand (COD) Dynamic behavior of COD in the IACM tanks during both the short and long aeration cycles is presented in Figure 4.46. During this study, the precision of COD measurement was estimated by analyzing about a total of 100 duplicate samples. It was found that the absolute error in COD measurement was up to 30 mg/L. Based on that, the random oscillation in COD concentration with a range of 30 mg/L was considered due to 159 the analytical error. Therefore, it can be concluded that the filtered COD concentration in the IACM tank was not affected by the intermittent aeration. This suggests that COD was removed approximately at the same rate in both air-on and air-off periods, even for a long aeration cycle. In addition to the normal COD variation, Huang (1993) also observed a rapid increase in COD concentration during an air-off period of 3 hours in a single IACM reactor system, when all nitrate was exhausted. This further suggested that as long as nitrate is present, COD removal will not be affected by the long air-off period. 120 110 O) E, c o e 100 c 0 u c o o Q O O T3 V 60 -•— Short aeration cycle - B — Long aeration cycle H 1 1 h H H 1 h 8 10 12 14 16 18 20 22 24 Time (min) Figure 4.46 Typical filtered COD variations in long and short aeration cycles. 4.2 Process Dynamics under the Transient-State Conditions As described in Section 3.5, transient state conditions were created by imposing ammonium and nitrate shock loads to the three systems, in which a quasi steady state had been reached under the respective set conditions. 160 4.2.1 Process Dynamics under the Batch Conditions By interrupting the anaerobic flows, the two IACM tanks were used as batch reactors. Two batch experiments were conducted in these two IACM tanks during run 9, after the completion of the steady-state phase: one with ammonium shock loading and the other with nitrate shock loading. This means the sludge used in these two batch experiments was acclimatized under the conditions set for run 9. It was found that the interruption of the anaerobic flows resulted in a failure of intermittent aeration, since the bulk liquid ORP levels in the two IACM tanks were not able to reach the lower control limit, set for switching on the aeration, due to a lack of organic substrate. Therefore, these two batch tests were conducted under no air supply conditions. 4.2.1.1 Dynamic Responses to the Ammonium Shock Loading Ammonium responses. The ammonium concentration profiles in the two IACM tanks, in response to the ammonium shock loading, are shown in Figure 4.47. Before the shock loading was imposed, the background concentrations in the two IACM tanks were 1.9 and 7.2 mg/L; this indicates that the nitrification in both systems was not complete under the steady-state conditions. After the shock loading was imposed, the ammonium concentrations in the two IACM tanks were elevated to higher levels and kept at the higher levels for the 2-hour air-off period. As shown in Figure 4.48, the corresponding NQ X concentrations were less than 0.5 mgN/L, indicating anaerobic conditions prevailed in the two IACM tanks. The constant ammonium concentrations shown in Figure 4.47 suggest that nitrification cannot proceed under anaerobic conditions. 161 CD E, c o c o c o o .2 'E o E E < 20 19 18 17 16 15 14 - • - W i d e ORP IACM tank -o—Narrow ORP IACM tank -+-0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Time (hour) Figure 4.47 The ammonium concentration profiles in the two IACM tanks in response to a 30 mg/L ammonium shock loading at time 0. NOx responses. The N O x concentration profiles in the two IACM tanks, in response to the ammonium shock loading, are presented in Figure 4.48. E c o c o c o o X o Z •Wide ORP IACM tank • Narrow ORP IACM tank + 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Time (hour) Figure 4.48 The NO, concentration profiles in the two IACM tanks in response to a 30 mg/L ammonium shock loading at time 0. 162 The background N O x levels were 0.33 and 0.39 mgN/L, indicating that the denitrification in those two aeration tanks was complete under the steady-state conditions. As noted in Figure 4.48, the N O x concentrations during the 2-hour air-off period were in a low range of 0.2-0.45 mgN/L. Therefore, it can be concluded that anaerobic conditions prevailed in the two IACM tanks, during this 2-hour air-off period, because the low residual N O x concentrations resulted from the interference in the N O x sample analysis (see Section 4.1.3.2). Ortho-P responses. The ortho-P concentration profiles in the two IACM tanks were shown in Figure 4.49. Time (hour) Figure 4.49 The ortho-P concentration profiles in the two IACM tanks in response to a 30 mg/L ammonium shock loading at time 0. Before imposing the ammonium shock loading, the ortho-P concentrations in the two IACM tanks were 0.3 and 0.2 mg/L. As noted in Figure 4.49, the shock loading did 163 not cause any instantaneous changes in the ortho-P concentrations. After the aeration was turned off by the ORP control system due to a lack of reducing agents, the ortho-P concentrations began to increase immediately and reached up to 4.5 mgP/L in the 2-hour anaerobic period. As described in the literature review, the increase must have been due to a secondary release. The results from this study further show that this release can start immediately after the onset of anaerobic conditions. In a single IA reactor system, ortho-P was released up to 1 mgP/L in a 1.5-hour air-off period (Huang, 1993). The lower release in his process was probably due to a lower phosphorus content, because his process, employing only a single IACM tank, was not a BPR process, A near linear increase in the ortho-P concentration indicates that the release followed zero-order kinetics. The rates of release, based on the linear phase of the two ortho-P profiles, were 0.81 and 0.74 (mgP/gMLVSSxh"1) in the wide and narrow ORP IACM tanks, respectively. 4.2.1.2 Dynamic Responses to the Nitrate Shock Loading NOx responses. The N O x concentration profiles in the two IACM tanks, in responses to the nitrate shock loading, are shown in Figure 4.50. As noted, the NOx concentrations in both IACM tanks declined linearly in the next 2-hour air-off period; this indicates that denitrification was occurring, despite the lack of organic substrate supply. Further, as noted in Figure 4.52, during the initial period of 0.4 hour, ortho-P uptake was also occurring. The results from these two figures suggest that the denitrification during the initial period was using the stored carbon in PAOs. The linear decrease in N O x 164 concentration suggests that denitrification followed zero-order kinetics. The rates were about the same for both IACM tanks (1.09 and 1.04 mgNFf 4-N/gMLVSSxh"1). Time (hour) Figure 4.50 The NO* concentration profiles in the two IACM tanks in response to a 30 mg/L nitrate shock loading at time 0. Ammonia responses. The ammonium concentration profiles in the two IACM tanks, in response to the nitrate shock loading, are presented in Figure 4.51. As expected, the ammonium concentrations were constant in the next 2-hour air-off period; this indicates that ammonium oxidation completely ceased even at high nitrate level. Ortho-P responses. The ortho-P concentration profiles in the two IACM tanks were shown in the Figure 4.52. After the nitrate shock loading was imposed, the ortho-P concentrations in both IACM tanks began to decrease immediately; the decreasing lasted for about 0.4 hours, indicating that the nitrate pulse could enhance phosphorus uptake 165 under anoxic conditions, by using N O x as an electron acceptor for oxidizing the stored carbon reserves (e.g., PHA). For the next 1.6 hour period, ortho-P concentrations in two IACM tanks remained in a low range, indicating that secondary ortho-P release can be effectively prevented by the high N O x level. c o v u c o o E 3 'E o E E < E •Wide ORP IACM tank - Narrow ORP IACM tank 1 1 1 1 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Time (hour) Figure 4.51 The ammonia concentration profiles in the two IACM tanks in response to a 30 mg/L nitrate shock loading at time 0. •Wide ORP IACM tank •Narrow ORP IACM tank! 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Time (hour) Figure 4.52 The ortho-P concentration profiles in the two IACM tanks in response to a 30 mg/L nitrate shock loading at time 0. 166 4.2.2 Process Dynamics under the Continuous-Flow Conditions Before presenting the results, it must be emphasized that the operating conditions for the transient-state experiments were the same as those for the corresponding steady-state phase, except that the ammonium or nitrate shock loading was imposed upon the three systems. In addition to determining reaction rates, the other objective of the transient-state study was to investigate dynamic responses to the shock loading. The results obtained from the fourth series are presented as examples, because the responses to the ammonium and nitrate shock loads, obtained from different transient-state experiments, were similar. However, data sets from the other series are used, when necessary. 4.2.2.1 Dynamic Responses to the Ammonia Shock Loading Instantaneous Drop in Ammonium Concentration An instantaneous drop in ammonium concentration, after the ammonium shock loading was imposed, was observed in all three aeration tank during most of the transient-state experiments. The observations from the first batch experiment are described here as an example. Before the shock loading was imposed, the background .ammonium concentrations in the two IACM tanks were 1.9 and 7.2 mgN/L. After the instant addition of 855 mg ammonium chloride, which was equivalent to an additional 30 mgN/L ammonium in the two IACM tanks, the concentrations in the two IACM tanks should be 32 and 37 mgN/L, respectively. However, the measured concentrations, after two minutes mixing, were only 16.7 and 18.2 mgN/L, respectively, approximately half the 167 concentrations that supposed to be. This phenomenon is termed as instantaneous ammonium drop. In two other transient-state experiments both conducted during run 10, this phenomenon was carefully studied with two ammonium shock loads: the first one imposed at the beginning of the transient experiment and the second one imposed at the time of 6 hours after the first shock loading. In the first of these two experiments, instantaneous drops, ranging from 6-11 mgN/L, were observed in the three systems, for both first and second shock loads. However, in the second experiment, conducted under the same conditions as the first experiment, the instantaneous drops were not significant (0-2 mgN/L) for both shock loads, in all three systems. The other three experiments with ammonium shock loading all exhibited significant, but variable amounts of instantaneous drop (3-11 mgN/L) for both shock loads. It was also noticed that the simultaneous drop in ortho-P concentration was not observed; this indicates that ammonium instant drop was not associated with ortho-P. There are many possible explanations for the instantaneous ammonium drop: 1) chemical precipitation, 2) biological reactions, such as nitrification and assimilation for bacterial growth, 3) stripping, 4) adsorption/exchange of ammonium ion on the surface of the activated sludge floes. Ammonium can form a few number of insoluble compounds. However, of these, only the compound MgNILPO^aq, a salt well known in analytical chemistry, is applicable to sewage (Schulze-Rettmer, 1991). The experimental results, showing that the ammonium drop was not accompanied by an ortho-P drop, suggest that ammonium 168 precipitation was not the case in this study. Biological reactions are excluded as options, since they could not take place at such a high rate (up to 10 mg/L drop in two minutes). Ammonia stripping is also excluded, because the environmental conditions (pH 7.0-7.4 and 20 °C) were not favorable to stripping. From published literature (e.g., Mercer et al., 1970; Koon and Kaufman, 1975; Booker at al., 1996), it is revealed that ammonium can be adsorbed by clinoptilolite, a natural zeolite with a high affinity for ammonium. The mechanism involved is ion exchange between NFT 4 and Na + or Ca 2 + . A literature review on bioadsorption of activated sludge (e.g., Nelson, 1976; Xue et al., 1988; Artola et al., 1997) indicates that activated sludge can adsorb many heavy metals and suggests that the mechanism for bioadsorption is still ion exchange, but involving a competition between FT* and heavy metal ions for the cell surface binding sites. All of these studies suggest that ammonium, which has a positive charge, could be adsorbed on the cell surface of activated sludge in exchange of FT. Based on the chemical properties of ammonium, the binding is weak, compared to metal ions; this suggests that the significant ammonium adsorption can occur only when the mixed liquor is lacking in metal ions. As suggested by the BPR mechanism, metal ions, such as potassium, magnesium, and calcium, are co-transported with ortho-P for both its release and uptake; that is, in an anaerobic-aerobic sequence system, metal ion concentrations are low in the aerobic zone, but high in the anaerobic zone. In the current three aeration tanks, significant amount of ammonium adsorption by the floes could occur, because the metal ion concentrations were low; this would result in 169 the observed instantaneous ammonium drop. The experimental results, showing that the instantaneous drop was not stable, would result from the competitions from metal ions present in the mixed liquor for the ammonium binding sites. Further, when the sludge with the adsorbed ammonium was recycled to the anaerobic zone, through the sludge return flow, where the metal ion concentrations were high, ammonium would be released back to the liquid due to the competitions from metal ions for the binding sites. It was indeed observed that the ammonium concentration in the anaerobic zone increased, in response to the ammonium shock loading (Figure 4.53). After the anaerobic sludge, which bound with metal ions, re-entered into the aeration tank, the metal ions would be utilized by PAOs with concomitant ortho-P uptake, and the binding sites would again be available to ammonium. The actual HRT of the IACM tank was six hour; that is, the mixed liquor in the IACM tank would be replaced once by the anaerobic mixed liquor in six hour. Therefore, the instant drop occurred again, at the same amount, when the second shock loading was provided at six hour later after the first shock loading was imposed. In summary, the theory of adsorption/exchange of ammonium ion on the surface of the activated sludge floes appears to be applicable in explaining the experimental results. However, it must be noted that the phenomenon of ammonium adsorption by activated sludge has not been reported in literature and that the competitions between metal ions and ammonium ions for binding sites on the surface of the activated sludge floes is only a speculation. 170 Ammonium Responses in Various Zones The measured ammonium concentration profiles, in the two IACM tanks and their preceding anaerobic zones, as well as in the 3-Bardenpho anoxic and aerobic zones, are presented in Figure 4.53. After an instantaneous ammonium drop, the ammonium levels in the three aeration tanks steadily decreased during the testing period. Since the transient-state experiments were conducted under the continuous-flow conditions, the shock loading caused a linear increase in the ammonium concentration in the anaerobic zones of both two-stage processes, through the sludge return flow, and an ammonium concentration variation in the 3-Bardenpho anoxic zone, through the internal recycle flow. IACM anoxic 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (hour) Figure 4.53 The measured ammonium concentration profiles in various zones after the ammonium shock loading was imposed. 171 Due to the continuous-flow conditions, the steady decrease in the measured ammonium concentration alone could not show the magnitude of nitrification reaction. Therefore, a theoretical ammonium time-concentration profile, showing the changes of ammonium concentration, caused by the continuous-flow conditions, is required. Equation 4.6 is derived for the theoretical profiles in the IACM tank. It takes the concentration variation in the influent from the preceding anaerobic zone and the flow conditions (e.g., flow rate) into account, but excludes biological reactions, such as nitrification and denitrification. The derivation of Equation 4.6 will be presented in detail in Section 4.2.3. CA(t) = at + (C°A + Aox- a/m) + (CA(0) - (C°A + Aox- a/m))emt (4.6) where CA(0) and CA(t) are the ammonium concentrations at time 0 and time t, respectively, during a transient-state phase; m = Q/V; V and Q are the volume of the IACM tank and the actual flow through the IACM tank, respectively; C°A and Aox are the average ammonium concentration and the average amount of ammonium oxidized, respectively, in the IACM tank, during the corresponding steady-state phase; a is a regression constant in Equation 4.12 (Section 4.2.3) that is used to represent the linear increase of the ammonia concentration in the preceding anaerobic zone. Figure 4.54 presents both the measured and theoretical concentration profiles, as well as the difference between these two profiles, for one of the two IACM tanks. As noted, the measured profile lays well below the theoretical profile; this indicates significant nitrification occurred. The difference between the two profiles increases linearly with time; This suggests that the nitrification rate followed zero-order kinetics. 172 x Theoretical profile in the IACM tank « Measured profile in the IACM tank • Difference between the theoretical and measured profiles 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (hour) Figure 4.54 Typical measured and theoretical ammonium profiles in an IACM tank. NOx Responses in Various Zones The N O x concentration profiles in the three aeration tanks are shown in Figure 4.55. As noted, a gradual increase in N O x concentration, in response to the ammonium shock loading, was observed. As indicated by the results from the corresponding steady-state phase (run 12), the nitrification in the three aeration tanks was limited by the ammonia concentration. Therefore, ammonia shock loading, by increasing the ammonium level, should result in an increase in the nitrification rate, while the denitrification rate in the same tank should still remain the same, since the denitrification in the three aeration tanks was limited by organic carbon, but not N O x concentration. Therefore, the increase in the N O x concentration should be observed in the three aeration tanks after ammonium shock loading was imposed. 173 3-Bardenpho aerobic zone +-Wide ORP IACM tank —x— Narrow ORP IACM tank 0.0 1.0 2.0 3.0 4.0 5.0 Time (hour) Figure 4.55 The measured NO x concentration profiles in various zones after the ammonium shock loading was imposed. Ortho-P Responses in Various Zones The ortho-P concentration profiles in various zones, after the ammonium shock loading was imposed, are presented in Figure 4.56. It should be noted that the results shown in Figure 4.56 were obtained from run 12, one of the methanol addition runs. In response to the shock loading, the ortho-P concentrations in the anaerobic zones of both two-stage processes remained at the background levels for a half hour period, then increased rapidly from 2 to 10 mgP/L in a 1.5-hour period and remained at this higher level for the rest of the test period. The ammonium shock loading was added into the two IACM tanks, therefore, it would not exert any effect on the ortho-P release in the anaerobic zones immediately. After 0.5 hour, some 65% of the anaerobic mixed liquor 174 would be replaced by the aerobic mixed liquor, through the sludge return flow. The rapid increase occurring at this time indicates that the increase was related to the aerobic mixed liquor contained high ammonium concentration. —o—3-Bardenpho aerobic zone - o - W i d e ORP IACM tank —A— Narrow ORP IACM tank —o— 3-Bardenpho anoxic zone —I—Wide ORP anaerobic zone —x— Narrow ORP anaerobic zone 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 Time (hour) Figure 4.56 The measured ortho-P concentrations in various zones in one of the methanol runs after the ammonium shock loading was imposed. Also shown in Figure 4.56, the ortho-P concentrations in the 3-Bardenpho anoxic and aerobic zones, as well as in the two IACM tanks, all remained at the background level for about 0.5-1.0 hour, and then gradually increased until the end of the test; this indicates that the ammonium shock loading did not cause any immediate and large changes in the ortho-P concentration in these four reactors. Furthermore, the gradual increases appeared to lag behind the rapid increases, observed in the preceding anaerobic zones, for about 0.5 hour. Therefore, it is suggested that the gradual increases in the two IACM tanks were the 175 results of the rapid increase in the preceding anaerobic zone, due to the continuous-flow conditions. Based on the BPR mechanism, ortho-P release in the anaerobic zone, if associated with organic substrate storage, will result in an ortho-P uptake in the subsequent aerobic zone. Therefore, the gradual increases observed in the two IACM tanks, as well as the zero influent V F A concentration, suggest that the rapid ortho-P releases in the two anaerobic zones were not associated with an organic substrate uptake. Considering that the only difference between the steady-state and transient-state phases was the ammonium shock loading, the rapid increase, which was observed only during the transient-state phase, had to be related to the ammonium shock loading. However, the mechanism of causing phosphorus release, by the presence of ammonium at high concentration in the anaerobic zone, is unknown. The ortho-P responses observed in one of the three experiments with an acetate addition are presented in Figures 4.57 and 4.58. As noted, neither the rapid increase in the anaerobic zone nor the gradual increase were observed at an acetate dosage of 50 mgCOD/L. However, at an acetate dosage of 30 mgCOD/L, such two increases were still significant, but considerably less than the corresponding increases observed in the methanol run. The above observations suggest that the presence of ammonium at a high concentration can cause ortho-P release even in the presence of acetate at a relatively low dosage in the anaerobic zone. However, as acetate dosage increases, biological excess phosphorus removal would surpass the disturbance caused by the ammonium shock loading, thus dominating the changes of the ortho-P concentrations in various zones. 176 24 0 J 1 1 1 1 L. 1 1 1 1 0 1 2 3 4 5 6 7 8 9 Time (hour) Figure 4.57 The measured ortho-P concentrations in the two 2-stage process zones and the 3-Bardenpho anoxic zone in one of the acetate runs after the ammonium shock loading was imposed. 2.0 0 1 acetate dosage 30 mg COD/L acetate dosage 50 mg COD/L j i_ 3 4 5 6 Time (hour) • 3-Bardenpho aerobic zone •Low ORP IACM tank •Low ORP IACM tank Figure 4.58 The measured ortho-P concentrations in the three aeration tanks in one of the acetate runs after the ammonium shock loading was imposed. 177 4.2.2.2 Dynamic Responses to the Nitrate Shock Loading NOx Responses in Various Zones The N O x profiles obtained in various zones, after the nitrate shock loading was imposed, are presented in Figure 4.59. As noted, a steady decrease in N O x concentration was observed in the 3-Bardenpho anoxic zone and the two IACM tanks. A slight increase was observed in the three anaerobic zones only at the end of the testing period; this indicates that organic substrate limiting conditions were experienced in the three anaerobic zones, after the extra amount of nitrate was brought into these zones, through the sludge return flow. Figure 4.59 The measured NO x concentration profiles in various zones after the nitrate shock loading was imposed. 178 To show the magnitude of denitrification in the IACM tank, Equation 4.7 was developed to obtain theoretical N O x concentration profiles. The derivation of Equation 4.7 will be presented in detail in Section 4.2.3. CN(t) = (C°N + AW + (CN(0) - (C°N + Ndr))em (4.7) where CN(0) and Cu(t) are the N O x concentrations at time 0 and time /, respectively, during a transient-state phase; C°N and Nd„ are an average N O x concentration in the IACM tank and an average amount of N O x denitrified in the same tank, respectively, during the corresponding steady-state phase; and the other parameters, such as Q, V, and m, have all been defined earlier in Equation 4.6. Figure 4.60 presents the measured N O x profile in one of the two IACM tanks and the corresponding theoretical profile. • Measured NOx profile in the IACM tank x Theoretical NOx profile in the IACM tank • Difference between the two profiles 0.0 1.0 . 2.0 3.0 4.0 5.0 Time (hour) Figure 4.60 Typical measured and theoretical NO r profiles in an IACM tank. 179 As noted, the measured profile lays well below the theoretical one; this indicates that significant denitrification occurred in the IACM tank. Furthermore, the difference between the two profiles increases linearly with time; this suggests that the denitrification rate followed zero-order kinetics. The results shown in Figures 4 54 and 4.60 together clearly show that nitrification and denitrification occurred simultaneously in the IACM tank. Ammonium Responses in Various Zones The ammonium concentration profiles in various zones are presented in Figure 4.61. As noted, relatively constant ammonium concentrations in all zones suggest that the nitrate shock loading did not affect the nitrification in the aeration tank. —©— 3-Bardenpho anoxic zone -a-Wide O R P IACM tank —A— Narrow O R P IACM tank —o— 3-Bardenpho anaerobic zone —4—Wide O R P anaerobic zone -x-Narrow O R P anaerobic zone 0.0 1.0 2.0 3.0 4.0 5.0 Time (hour) Figure 4.61 The measured ammonia concentration profiles in various zones after the nitrate shock loading was imposed. 180 Ortho-P Responses in Various Zones Figure 4.62 presents the ortho-P concentration profiles in various zones. Due to the nitrate shock loading, the N O x level in the return sludge would increase significantly for all three processes. Therefore, the nitrate shock loading, through increasing the N O x level in the return sludge, resulted in a rapid decrease in the ortho-P concentration in the three anaerobic zones. •3-Bardenpho anoxic zone •Wide ORP IACM tank • Narrow ORP IACM tank - 3-Bardenpho anaerobic zone —I—Wide ORP anaerobic zone —x— Narrow ORP anaerobic zone 0.0 1.0 2.0 3.0 Time(hour) 4.0 5.0 Figure 4.62 The measured ortho-P concentration profiles in various zones after the nitrate shock loading was imposed. Although the nitrate shock loading was directly imposed on the two IACM tanks and the 3-Bardenpho anoxic zone, it did not cause any significant changes in the ortho-P concentrations in these zones during the initial one hour period; this indicates that the presence of nitrate at a high concentration does not enhance the ortho-P uptake under low 181 DO conditions. During the rest of the testing period, a gradual decrease in ortho-P concentration was observed in the three aeration tanks. The changing patterns shown in Figure 4.62 suggest that the gradual decrease resulted from the rapid decrease in the three preceding anaerobic zones, due to the continuous flow. 4.2.3 Kinetic Study In IA processes with a relatively long aeration cycle (>3 hours), ammonium is converted to nitrate during the air-on period, and the nitrate produced is used to remove organic substrate during the subsequent air-off period. Therefore, the nitrification and denitrification rates in this type IA processes can be determined in-situ, based on the nitrate production rate during the air-on period and the nitrate reduction rate during the air-off period (Huang, 1993). However, Batchelor (1982) predicted that an IA reactor would behave just like a continuous-aeration reactor, as the cycle time ratio (CTR, the ratio of aeration cycle time to reactor HRT) is less than 0.1; this suggested that the nitrate concentration in the IA reactor could become constant, when the aeration cycle used is short enough. As such, the in-situ determination of reaction rates would become difficult, if not impossible. Furthermore, due to SND, the in-situ method could significantly underestimate both nitrification and denitrification rates. For estimating reaction rates, the simplest and most frequently used method is a batch reactor, operated at constant temperature and volume. However, a simple batch reactor cannot simulate a continuous-flow process. When a continuous-flow reactor is used to obtain the rate data, a number of steady-state runs are required to obtain data 182 relating reaction rate to concentration. In other words, one steady-state run is unable to assess the reaction order. Furthermore, as a substrate becomes limiting, such as close to zero, the rates obtained are usually less than the potential rate (or maximum rate). This study introduces a technique that allows process kinetics to be studied in a continuous-flow process. For each studied reaction, the technique requires a steady-state run with an additional transient-state test, created by a shock loading, using a substrate of the studied reaction, after steady state has been reached. Such a technique can eliminate the influence of side reactions on the substrate of the studied reaction. For example, nitrate is a substrate of denitrification (the studied reaction), as well as a product of nitrification (a side reaction). Therefore, the determination of denitrification rate, based on nitrate reduction rate, in a reactor involving SND, is subject to the influence of nitrification. However, such influence can be eliminated, if the response to the nitrate shock loading, observed during the additional transient-state test, is used to determine the denitrification rate. This technique can overcome the drawbacks of using a batch reactor and continuous-flow reactor to determine reaction rates. Firstly, since the transient-state tests are conducted under the same conditions as the steady-state phase, except for a higher substrate concentration, resulting from the shock loading; therefore, the rates estimated are representative of continuous process. Secondly, substrate shock loading, by eliminating substrate limiting conditions, allows the maximum reaction rate to be determined. Thirdly, this technique is able to obtain data relating reaction rate to concentration in one run, thus allowing reaction order to be studied. Finally, the steady-183 state rate (also termed as real rate) can also be determined, based on mass balances; the comparison between the real rate and the maximum rate can be made and allows process performance to be assessed. However, the shock loading may cause unwanted changes in the process for a short period. In this study, the proposed technique was applied to determine the simultaneous nitrification and denitrification rates in the IACM tank, as well as the nitrification and denitrification rates in the three-stage Bardenpho aerobic and anoxic zones, respectively. The two particular objectives for this kinetic study were: 1) comparison of the SND rates in the IACM tank with the corresponding rates in the 3-Bardenpho aerobic and anoxic zones, and 2) investigation into the effects of the ORP control range and the external substrate dosage on the SND rates. 4.2.3.1 Analysis of Reaction Rate Steady State The specific nitrification and denitrification rates in a reactor under steady-state conditions are defined as (Randall et al., 1992): SNR(or SDNR) = A ^ o r N ^ (4.8) I x HRT where SNR and SDNR are the specific nitrification and denitrification rates, respectively; Aox and Nj„ are the amounts of nitrification and denitrification in the reactor, respectively; Xis the mixed liquor volatile suspended solids (MLVSS) level in the reactor; and HRT is the nominal hydraulic retention time of the reactor. 184 Equation 4.8 was applied to calculate the real nitrification and denitrification rates in the IACM tank and in the 3-Bardenpho aerobic and anoxic zones. Transient State For simplicity, the equations below refer generally to the two-stage , IA process. A general mass balance on a reactive constituent C around a continuous-flow, complete-mix reactor is written as follows: V^ = QCm+QC-rcV (4.9) Where V is the volume of reactor; Q is the actually flow rate; C and C,„ are the reactive constituent concentration in the reactor and in the influent, respectively; and rc is the sum rate of all reactions involving C. Generally, the change of ammonium concentration in an activated sludge system involves three reactions: nitrification (rN), ammonium assimilation for bacterial growth (rG), and ammonification of organic nitrogen (r0). Therefore, ammonium balances on the IACM tank during steady-state and transient-state phases can be written as follows: Steady-state phase: 0 = QC°A.m-QC0A-rN(C°A)V-(rc + ro)V (4.10) Where C°A.in and C°A are the average ammonium concentrations in the anaerobic zone and in the IACM tanks during a steady-state phase, respectively; r^fCf^V is the amount of ammonium oxidized (nitrification) and is also termed as Aox. 185 Transient-state phase: V(dCA/dt) = QCA-tn - QCA - rN(CA)V-(rG + rQ)V (4.11) Where CA.in and CA are the ammonium concentrations in the anaerobic zone and in the IACM tanks, respectively, during a transient-state phase; Vn(CA) indicates that the nitrification rate is a function of ammonium concentration. The substrate of ammonification is organic nitrogen, hence the rate of ammonification is dependent on the influent organic composition and bacteria population. Therefore, the change in ammonium concentration would not affect the rate of ammonification. Ammonium assimilation for bacterial growth has affinity constants as low as 10 p-mole (Kleiner, 1985); this means that the rate of assimilation can reach the maximum at a very low ammonium concentration. Therefore, in the three systems, which treated a sewage containing plentiful ammonium, it was reasonable to assume that the ammonium concentration change (e.g., by shock loading) does not affect the assimilation rates, because the maximum rate has already been reached at a very low ammonium level. As described earlier, the transient-state was created upon the steady-state by shock loading. In the tank where the ammonium shock loading was imposed, the shock loading only raised the ammonium concentration to a higher level, compared to the steady-state level. As discussed in the preceding paragraph, the change in ammonium concentration would not exert any effects on the ammonification and ammonium assimilation. Therefore, the rates of ammonium assimilation and ammonification, after the ammonium shock 186 loading was imposed, would remain unchanged; in other words, the reaction terms in Equation 4.11, rG and r0, are the same as those in Equation 4.10. Furthermore, the increase in the ammonium concentration in the preceding zone (Figure 4.53), caused by the ammonium shock loading, can be described by a linear equation for the anaerobic zone of the two-stage, IA process or by a quadratic equation for the 3-stage Bardenpho anoxic zone. Subtracting Equation 4.10 from Equation 4.11 and taking Equation 4.12 into account gives the following equation: Equation 4.13 clearly shows that the transient ammonium response (dCA/df) to the ammonium shock loading can be used to study nitrification kinetics without the influences of ammonification and ammonium assimilation. The change of nitrate concentration usually involves nitrification (rN) and denitrification (rD). The assimilation of nitrate for bacterial growth can be neglected, when the influent feed contains plentiful ammonium. Therefore, N O x balances on the IACM tank during the steady-state and transient-state phases can be written as follows: Steady-state phase: Cin = C°A-in + at + fit2 (4.12) V(dCA/dt) = Qat + Qpt2 - QCA - rM(CA)V+ (QC°A + AJ (4.13) 0 = QC°N.in - QC°N- rD(C°N)V- rNV (4.14) 187 where CV,„ and C°N are the average N 0 X concentrations in the anaerobic zone and in the IACM tank, respectively, during a steady-state phase; rD(C°N)V is the amount of denitrification in the IACM tank and is also termed as Ndn. Transient-state phase: V(dCN/dt) = QCN.in - QCN - rD(CN) V-rNV (4.15) where CN-W and CN are the N O x concentrations in the anaerobic zone and in the IACM tanks, respectively, during a transient-state phase; rD(Cu) indicates that the denitrification rate is a function of N O x concentration. Since the nitrate shock loading would not affect the nitrification rates (Figure 4.61), the nitrification rate during the transient-state phase would remain the same as that during the steady-state phase. Furthermore, as shown in Figure 5.59, the nitrate shock loading did not increase the N O x concentration in the preceding anaerobic zone. Therefore, the term of QCN-m in Equation 4.15 is equal to the term of QC°mn in Equation 4.14. Subtracting Equation 4.14 from Equation 4.15 gives the follow equation: V(dCN/dt)=-QCN-rD(CN)V+(QC°N + Ndn} (4.16) Equation 4.16 indicates that the transient N O x response to the nitrate shock loading can be used to study denitrification kinetics without the influence of nitrification. A review of the literature concerning the nitrification process shows diverse opinions regarding the reaction order (Ffalling-S(|)rensen and J<|)rgensen, 1993). However, it is generally accepted that nitrification follows zero-order kinetics to ammonium 188 concentration as low as 5 mgN/L. During the transient-state phase, the ammonium concentrations in the IACM tank, as well as in the 3-stage Bardenpho aerobic zone, were elevated to a level of 15-40 mg/L by the ammonium shock loading. Therefore, zero-order kinetics is assumed for the nitrification reaction during the transient-state phase: rN = kNX (4.17) where kN is the specific nitrification rate. Several investigators (Moore and Schroeder, 1970; Murphy and Dawson, 1972; Stensel et al, 1973) have all reported zero-order kinetics for the denitrification process, when the nitrate concentration is above 1-2 mg/L. During the transient-state phase, the ammonium concentration in the IACM tank, as well as in the 3-stage Bardenpho anoxic zone, was elevated to a level above 30 mg/L by the nitrate shock loading. Therefore, zero-order kinetics is also assumed for the denitrification reaction during the transient-state phase. rD = kDX (4.18) where kD is the specific denitrification rate. During the steady-state phase, a SRT of 15 day was maintained in the three systems by wasting the mixed liquor from the aeration tank daily. Therefore, the daily variation of MLSS level in the three systems, caused by the sludge wastage, was less than 7%. Considering the duration of the transient-state phase was less than 6 hour, the variation of the MLSS level during this phase would be less than 2%. Therefore, the solids level X can be considered as a constant during the transient-state phase. 189 Based on the assumptions of zero-order kinetics for both nitrification and denitrification, Equations 4.13 and 4.16 can be further simplified to Equation 4.19. C '(t) + mC(t) =at2 + bt + c (4.19) where in describing an ammonium profile, C(t) represents ammonium concentration at time t; m = Q/V, a= mf3, b = ma, c = (mC°A + mAox/Q - kNX). where in describing a nitrate profile, C(t) represents N O x concentration at time t; m = Q/V, a=0,b = 0,c = (mC°N + mNdn/Q - koX). Integration of Equation 4.19 gives: C(t) - P? + (2/3/m - a)t = (2fi/m - a + c)/m + Coemt (4.20) where Co is an integration constant. Equation 4.20 is a two-variable linear equation, with the following forms: x = emt (4.21) y = C(t)-pf + (2/3/m - a)t (4.22) Based on a measured time-concentration profile, a x-y plot, with regression constants, can be developed according to Equation 4.20 as shown in Figures 4.63 and 4.64. The intercepts from the linear plots are used to estimate specific reaction rates (kn and kD). kN=[m(C°A+Aox/Q-Intercept)-a+2pym]/X (4.23) kD = m(C°N + Ndn/Q - Intercept)/X (4.24) 190 * 1 5 5 • 3-Bardenpho aerobic zone • Wide ORP IACM tank A Narrow ORP IACM tank y = 57.9x-15.1 S<<> R2 = 0.996 ' y = 38.9x-11.6 ^ JS^"^ R* = 0.995 . y = 38.8x-12.4 • / ^ ^ ^ R 2 = 0.996 — l 1 h 1 1 1 1— : 1 1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 x Figure 4.63 Linear plots developed from measured ammonium profiles according to Equation 4.20 (data obtained from series 4 of the transient-state experiments). 191 As shown in Figures 4.63 and 4.64, the high correlation coefficients for the linear plots suggest zero order kinetics for both nitrification and denitrification reactions. However, a non-linear N O x plot was also observed in the 3-Bardenpho anoxic zone, in the experimental series with a methanol addition (series 3 and 4). The non-linear plot might result from using different carbon sources for denitrification and could be broken into two linear phases. Further explanations cannot be provided based on the existing experimental results. Assuming that the reaction terms are zero, (i.e., rN = 0 and rK = 0), Equations 4.13 and 4.16 can also be simplified to Equation 4.19 with c = m(C°A + Aox) for ammonium and c = m(C°N + Ndn) for NO x , respectively. Using the same integral method, Equations 4.6 and 4.7 (shown in Sections 4.2.2.1 and 4.2.2.2, respectively), that were used to calculate theoretical profiles, are developed. 4.2.3.2 Kinetic Rates Specific Nitrification Rates (SNR) The maximum and real SNRs, as well as the calculation procedure, are summarized in Table 4.8. Figure 4.65 presents a comparison between the two types of SNR. As noted, the maximum rate agrees with the real rate in a low value range (less than 0.8 mg N H + 4 -N/gMLVSSxh"1) and is significantly higher than the real rate in a high value range (greater than 0.8 mg NHVN/gMLVSSxh" 1). The discrepancy in the high value range is believed to result from the ammonia shock loading. 192 3 I X 1/3 > 60 s 00 CO > J5 x 0 1 * s s i §- 3 « a. 60 •a , 4 J Q5 «5 «5 <L> O o a. "2 o •& 1) •a *-< 03 CQ <u 60 ee a. B CJ •a a CQ < * as It was very likely that amrnonium limiting conditions were experienced during the steady-state phase, at high nitrification rates. Therefore, the ammonia shock loading eliminated the ammonium limiting conditions, thus, resulting in a higher maximum SNR. The gradual increase in the N O x concentration in the three aeration tanks, under the ammonium shock loading conditions, indirectly proved the above explanation (Figure 4.55). Maximum SNR (mgNH4-N)/(gMLVSS*h) Figure 4.65 Comparison of the maximum SNR with the corresponding real SNR As shown in Table 4.8, the real SNRs in the 3-stage Bardenpho aerobic zone were in a typical range of 1.4-2.0 (mgNFLrN/gMLVSSxh"1). These values are in good agreement with the literature data under the same conditions: Randall et al. (1992) reported that the SNRs were 1.78, 1.92 and 2.02 (mgNFLt-N/gMLVSSxh"1), at temperatures of 20, 15 and 194 10 °C in a single-sludge BNR process, treating municipal wastewaters, with a total SRT of 15 days and an aerobic SRT of 8 days. The maximum SNRs, in the two IACM tanks, were generally in a range of 0.4-1.7 (mgNHrN/gMLVSSxh"1); as expected, these rates were significantly lower than those in the 3-Bardenpho aerobic zone (3.4-8.1 mgNFLt-N/gMLVSSxh"1), at a DO concentration of 3 mg/L. As suggested by the steady-state evaluation, both autotrophic nitrification and heterotrophic nitrification could occur in the IACM tank. The lower rates in the IACM tank could result from an inhibition of autotrophic nitrification by low DO conditions or by the lower intrinsic heterotrophic nitrification rate (compared to the autotrophic nitrification rate). With reference to Table 3.5, the results shown in Table 4.8 also reveal the effect of ORP control on the SNR in the IACM tank. Within each series, the maximum SNRs under the high ORP control were 0.57 and 1.9 times higher than those under the low ORP control for series 5 and 3, respectively. In series 1, the wide ORP control gave a 46% higher SNR, compared to the narrow ORP control; however, in series 4, no difference in the maximum SNRs was observed between the wide and narrow ORP control. Specific Denitrification Rates (SDNR) The maximum and real SDNRs, as well as the calculation procedure, are summarized in Table 4.9. The comparisons between the two type SDNRs are presented in Figure 4.66. 195 as s 1 u 03 o I* 60 g 60 a GO 00 > 60 S 60 X OO oo > a X •2 I i 1) o. 60 13 ,1> s <D Z OO Q3 vo o a. 2 o <: o a y cs £ a o N «j '53 -o « 0 * ! a T3 —' m § 2P « 3 -a 1 5 r<-> 03 2-a C JJ <L> J3 w *^  <u ^ fr " *-< 0) o s, •a a d a g vo Ov It should be noted that the maximum rates, obtained from the cases with the two linear phases, were not included in the comparison, since the maximum rates of any phases do not represent the average, as do the real rates. As shown in Figure 4.66, similar results to the SNRs were also observed for the SDNRs. That is, the maximum SDNRs appear to agree with the real SDNRs in the low value range (less than 0.5 mgNOx-N/gMLVSSxh"1) and is significantly higher than the real rate in the high value range (0.5 mgNOx-N/gMLVSSxh"1). The nitrate shock loading (by eliminating the N O x limiting conditions) is believed to be the cause of the discrepancy between the two rates in the high value range. Maximum SDNR (mgNOx-N)/(gMLVSS*h) Figure 4.66 Comparison of the maximum SDNR with the corresponding real SDNR. As described in Section 4.1.3.3 and 4.1.3.6, during the runs with a supplement of acetate, the carbon reserves (e.g., PHA), stored by PAOs in the anaerobic zone, would be available to the two IACM tanks and the 3-Bardenpho anoxic zone. The SDNR at the 197 zero acetate dosage would be representative of the conditions in which the denitrification occurs with only the SBCOD fraction present in the sewage. The SDNRs at acetate dosages of 30, 50, and 100 mgCOD/L would be the yield of both SBCOD and stored carbon. For the runs with methanol addition, since methanol was available to the above three reactors, the SDNR at a methanol dosage of 15 mgCOD/L would be representative of the conditions in which denitrification occurs with the SBCOD and a small amount of RBCOD. As shown in Table 4.9, the maximum SDNR in the 3-Bardenpho anoxic zone, was 2.57 mgNOx-N/gMLVSSxh"1 at an external acetate dosage of 100 mgCOD/L (series 5). It is considerably lower than the corresponding rates obtained by Ekama and Marais (1984) in batch tests and Carucci et al. (1996) in a lab-scale, semi-batch system. However, the discrepancy is expected, since the supplemented acetate was not directly available to the denitrification in the anoxic zone, but was available to the batch or semi-batch systems. Furthermore, the value of 2.57 mgNOx-N/gMLVSSxh"1 falls into ranges of 1.25 to 6.25 mgNOx-N/gMLVSSxh"1 (U.S. EPA, 1975) and 1.6-3.3 mgNOx-N/gMLVSSxh"1 (Stensel, 1981; Burdick et al., 1982), reported for both pilot-scale and full-scale BNR plants. As noted in Table 4.9, the methanol addition at a dosage of 15 mgCOD/L gave significantly higher SDNRs (8.3 and 10.7 mgNOx-N/gMLVSS/h), during the first 20 minutes of the transient-state phase, in the 3-Bardenpho anoxic zone; these two rates fall in the wide range of 4.2-50 mgNOx-N/gTSSxh"1, reported for methanol addition systems (U.S. EPA, 1975). However, the methanol addition did not increase the maximum SDNRs 198 in the IACM tanks, probably due to PHB storage in these tanks, induced by methanol; this might limit methanol utilization for denitrification (Section 4.1.5.2). With reference to Table 3.5, the results shown in Table 4.9 reveals the effects of ORP control and external substrate dosage on the maximum SDNR in the IACM tank. The details are: 1) the low ORP control generally gave a 3.2 times higher SDNR, compared to the high ORP control (series 3); 2) the wide ORP control gave a 0.52-2 times higher SDNR, compared to the narrow ORP control (series 1 and 4); and 3) under the same low ORP control (series 2), acetate addition at an dosage of 50 mgCOD/L double the rate at an acetate dosage of 30 mgCOD/L. The above findings are all in agreement with results obtained from the steady-state evaluation; Summary As demonstrated in Section 4.2.3.1, the proposed technique can determine nitrification and denitrification rates without the influences of ammonification and ammonium assimilation and without the influence of nitrification, respectively. The technique is extremely flexible and can obtain data relating reaction rate to substrate concentration in one run. In fact, the experimental technique, one steady-state run with an additional transient-state test, could be used to study many other reactions in continuous-flow, complete-mix reactors. However, the mathematical method (integral method) used in this study is limited to reactions with zero-order kinetics. This does not mean that the experimental technique has limitations, because a differential, mathematical method can be used for reactions with non-zero-order kinetics. 199 4.3 Mechanisms of Simultaneous Nitrification and Denitrification As described in the literature review, unaccounted for nitrogen loss has been reported in the aeration basins of many, full-scale, BNR plants (Rabinowitz and Barnard 1995; US EPA, 1987; Rittmann and Langeland, 1985; Moriyama et al., 1993) as well as in many lab-scale wastewater treatment systems under aerobic conditions (Kugelman et al., 1991; Gupta et al., 1994; Bang et al., 1995; Kshirsagar et al., 1995; Munch et al., 1996). However, unaccounted for nitrogen loss can also occur under anoxic conditions; that is, ammonia oxidation using nitrate as electron acceptor (van der Graaf et al., 1990; 1991). Because of that, the unaccounted for nitrogen losses observed in this study, based on the nitrogen balance on the entire process train, cannot be simply attributed to the occurrence of SND in aeration tank, because the current experimental systems include an anaerobic zone (with presence of NO x) as well. Therefore, the assumption that the unaccounted for nitrogen loss was due to the SND in the aeration tank must be verified. Under aerobic conditions, a few mechanisms have seen suggested to explain the unaccounted for nitrogen loss: 1) ammonia stripping, 2) nitrogen loss during nitrification process, 3) denitrification in the anoxic microzone inside of the floes, 4) denitrification mediated by aerobic denitrifiers. These four possible explanations all will be examined in this section, based on the experimental results obtained in this study. 4.3.1 Ammonia Oxidation under Anoxic Conditions As described in the literature review, biological oxidation of ammonia under anoxic conditions was proposed by Broda (cited in Kuenen and Robertson, 1994). The bacteria 200 mediating this reaction are unknown. However, the phenomenon was observed by van der Graaf et al. (1990; 1991) in a sulfide-oxidizing, denitrifying column being fed from a methanogenic reactor (containing VFAs). The results obtained from the transient-state experiment under batch conditions (Section 4.2.1.2) clearly showed that ammonia oxidation by nitrate was impossible in the current systems. As indicated by the ammonia and N O x concentration profiles (Figure 4.51), denitrification preceded steadily in the two IACM tanks under the anoxic conditions; however, at the same time, ammonia oxidation was not observed at the elevated nitrate concentrations. Furthermore, the results from the steady-state evaluations indicated that a higher amount of N O x was denitrified in the anaerobic zone of the high ORP process than in that of the low ORP process, during all the runs under the high/low ORP control regime (Section 4.1.3.2). If ammonia oxidation by nitrate were significant in the anaerobic zone, it would result in 1) higher amounts of ammonia loss in the anaerobic zone of the high ORP process than that of the low ORP process, and subsequently 2) higher amounts of the total unaccounted for nitrogen loss in the high ORP process than in the low ORP process. However, the experimental results indicated that neither of them were true (Sections 4.1.3.1 and 4.1.3.2). However, in runs 11 and 12, unusually high amounts of ammonia loss were observed in the 3-Bardenpho anoxic zone (7.8 and 9.2 mgN/L). The results suggested that a significant amount of ammonia was oxidized in the anoxic zone. However, as described in Section 4.1.3.3, the oxidation was considered to be the aerobic nitrification process, due 201 to the air entrainment caused by over aeration in the aeration tank (DO 3 mg/L), through the large internal recycle flow (300% of the influent flow). 4.3.2 Ammonia Stripping in the Aeration Tank Ammonia stripping is generally possible; however, it was not the case in this study, where pH was controlled in a range of 7.0 to 7.4 in all three aeration tanks, throughout the entire experimental period. An ammonia stripping study conducted by Turk (1986), in a bench-scale activated sludge system, showed that at a pH of 7.5 and an ammonium concentration of 90 mgN/L, the average rate of stripping, over an air flow rate range of 400-1000 mL/min, was only 1 mgNHs-N/Lxd"1. In addition to pH, the amount of ammonia stripping should depend on aeration intensity. Since the aeration intensity used in the current systems was relatively constant among different runs and was almost the same between the two IACM tanks, the ammonia stripping in the aeration tank would not cause the increasing trend in the unaccounted for nitrogen loss with increasing acetate and methanol dosages (Section 4.1.3.6); similarly, it could not explain the significant difference in the unaccounted for nitrogen loss between the two, 2-stage processes. Ammonia stripping was also denied as one of causes for the unaccounted for nitrogen loss in several other studies (e.g., Wood et al., 1981; Kugelman and Spector, 1988), since a constant neutral pH was maintained in their experimental systems and all nitrogen was recovered with non-nitrifying sludge. 202 4.3.3 Nitrogen Loss During the Nitrification Process As described in the literature review, Nitrosomonas europaea, well recognized autotrophic nitrifier, is able to produce nitrous oxide and nitrogen gas at low oxygen concentrations (Poth and Focht, 1985; Poth, 1986; Downes, 1988). Bock et al. (1995) further showed that cells of autotrophic nitrifiers were able to nitrify and denitrify at the same time, when grown under oxygen limiting conditions (0.2-0.4 mg/L). However, in pure cultures, the maximum gas production only accounted for 19% of ammonia provided (Bock et al., 1995). In the activated sludge system, Zheng et al. (1994) found that high amount of nitrous oxide production was accompanied by incomplete nitrification. Low DO concentration (0.2 mg/L) and a short SRT (3 days), the two conditions causing incomplete nitrification, resulted in nitrous oxide gas production. The highest production only accounted for 16% of nitrified nitrogen. •s -In this study, a significant amount of the unaccounted for nitrogen loss (up to 36% of the influent TKN) was obtained with complete nitrification, such as in the high ORP process. Further, the amount of the unaccounted for nitrogen loss in the low ORP process, which was accompanied by incomplete nitrification, accounted for up to 50% of the influent T K N and 90% of the nitrified nitrogen. The gas production by Nitrosomonas is unlikely to account for such a high amount of nitrogen loss and is unable to explain the increasing trend of nitrification with increasing acetate and methanol dosages. Therefore, the nitrogen gas production by Nitrosomonas is precluded as a major cause for the unaccounted for nitrogen loss in this study. 203 4.3.4 Denitrification in the Anoxic Microzone Inside the Floes As described in the literature review, nitrogen compounds, such as ammonia and nitrate, are normally removed microbiologically in two steps, exploiting two different types of microorganisms. The first step is aerobic, and involves the oxidation of ammonia to nitrate by autotrophic bacteria, which derive energy from the reaction (nitrification). In the second step, the nitrate is reduced to nitrogen gas under anaerobic conditions, generally by heterotrophs or methylotrophs (also denitrifiers, but using methanol). A DO concentration of 0.1 mg/L or above has been reported to inhibit denitrification in dispersed cells (Focht and Chang, 1975; Nelson and Knowles, 1978; Krul, 1976). Based on the concepts of exclusive nitrification and denitrification, an unaccounted for nitrogen loss can be attributed to denitrification in the anaerobic interior of the floe, even though the bulk liquid is aerobic. As described in the literature review, it has been postulated that a DO gradient exists within the activated sludge floes and hence nitrification may occur on the surface of the floes, while denitrification may occur in the inner layers (Kurl, 1976; Rittmann and Langeland, 1985; Bakti and Dick, 1992; Moriyama et al., 1993; Munch et al., 1996). The simulation results obtained from modeling studies (La Motta and Shieh, 1979; Benefield and Molz, 1984; Bakti and Dick, 1992) showed that the DO gradient within the activated sludge floes is controlled by a number of environmental factors, such as bulk DO concentration, mixing intensity (or floe size), organic substrate loading (or SRT) and aeration cycle. Significant nitrate reduction due to denitrification in the anoxic microzone has been reported in activated sludge systems 204 under low DO conditions (e.g., Krul, 1976; Christensen and Harremoes, 1977; Rittmann and Langeland, 1985; Moriyama etal., 1993). The predictions based on the anoxic microzone theory would be: 1) both nitrification and denitrification would occur simultaneously at a reduced rate, since environmental conditions are not optimal for either of them; 2) denitrification under highly aerobic conditions would be impossible because full penetration of the activated sludge floes by DO is expected; 3) as the other three factors (e.g., DO concentration, mixing intensity, and aeration cycle) hold constant, increasing organic substrate concentration, especially small molecule carbon concentration, would enhance the denitrification side of SND; at the same time, there would be inhibition on the nitrification side, since organic substrate would penetrate further into/the floes, and subsequently increase the relative volume of the anoxic zone inside of the floes and reduce the volume of the aerobic zone. The anoxic microzone theory fits some observations in this study, but conflicts with some others. The pro observations are: 1) compared to the high ORP process, the low ORP process, in which DO concentration was relatively lower within each individual run, achieved a larger amount of the unaccounted for nitrogen loss; 2) the amount of the unaccounted for nitrogen loss in both two-stage processes increased with increasing acetate and methanol dosages, except at the high dosage; 3) incomplete nitrification was observed in the low ORP process, whereas complete nitrification in the high ORP process; and 4) the specific nitrification and denitrification rates in the two IACM tanks obtained, under both steady-state and transient-state conditions, were significantly lower than the corresponding rates in the 3-Bardenpho aerobic and anoxic zones, respectively. 205 The con observations are: 1) the amount of nitrification in the low ORP process increased as the external substrate dosages increased; 2) at the high acetate dosage (100 mgCOD/L), the bleeding of VFAs into the low ORP IACM tank caused nitrification to be complete in that tank and a decrease in the amount of the unaccounted for nitrogen loss; 3) methanol addition at the high dosage caused a decrease in the amount of the unaccounted for nitrogen loss in both high and low ORP processes; and 4) a significant amount of the unaccounted for nitrogen loss was observed in the 3-Bardenpho aerobic zone, in which DO concentration was maintained at a concentration of 3 mg/L. In summary, considering that all of the favorable conditions to the anoxic microzone denitrification were provided in the two-stage process, it cannot be precluded as a cause for the unaccounted for nitrogen losses observed in this study. However, it was unlikely to be the main cause. 4.3.5 Aerobic Denitrification and Heterotrophic Nitrification As described in the literature review, many studies have reported that denitrification can occur under fully aerobic conditions (e.g., Robertson et al., 1984; Kugelman et al., 1991; Gupta et al., 1994; Bang et al., 1995; Kshirsagar et al., 1995). The results obtained from most of the activated sludge studies generally confirmed the mechanism postulated by Robertson et al, 1988; 1991). However, Kugelman et al. (1988; 1991) gave a different explanation for aerobic denitrification; it was an inhibition of oxygen respiration by nitrite. Due to the inhibition, the biomass uses N O x as an electron acceptor even at a high DO concentration. They also noted that the inhibitory effect persisted at levels of nitrite as low 206 as 1.0 mg/L. This option can be rejected without further investigation because the nitrite levels in various bioreactors of the three experimental systems used in this study were very low. For example, the nitrite concentrations in both zones of the two 2-stage processes were less than 0.15 mgN/L during the entire experiments. Such low nitrite concentrations would not exert a significant inhibitory effect on oxygen respiration. Also described in the literature review, many heterotrophs are capable of performing both nitrification and denitrification under aerobic conditions, thus converting ammonia directly into gaseous products (Robertson et al., 1988; 1990; van Niel, 1991). It was suggested (van Niel, 1991) that heterotrophic nitrifiers only have a significant role in the oxidation of ammonia to nitrite or nitrate at a BOD/N ratio of 6.9 or above. Furthermore, low DO concentration and shorter retention times also favour heterotrophic, rather than autotrophic, nitrification. For aerobic denitrification, it was noted (van Niel, 1991; Neef et al., 1996) that aerobic denitrifiers appear to have an ecological advantage in niches with fluctuating aerobic/anaerobic periods, compared to anaerobic denitrifiers. Furthermore, some specific substrates, such as methanol, were found to be effective in selecting some species of aerobic denitrifiers, such as Paracocci and Hypomicrobium (Mechsner and Hamer, 1983; Bang et al. 1993; Neef et al., 1996). According to the working model proposed by Robertson et al. (1988; 1991), heterotrophic nitrification and aerobic denitrification, in Thiosphaera pantotropha, can occur to overcome redox problems in the cytochrome chain, and PHB synthesis will also serve the same purpose. As a consequence of heterotrophic nitrification and PHB synthesis, there would be less pressure on Thiosphaera pantotropha to denitrify 207 aerobically. Therefore, unstable aerobic denitrification by Thiosphaera pantotropha was expected (van Niel, 1991). It was also noted by Robertson et al. (1988; 1991) that both aerobic denitrification and heterotrophic nitrification rates decreased as the DO concentration increased. The working model fits all the observations listed in Section 4.3.1.4, especially, those conflicting with the diffusion limitation theory. Consider the following: 1) Based on the critical DO concentration of 0.1 mg/L in the innermost cells, at which anaerobic denitrification is completely inhibited by DO, it was suggested by Wuhrmann (1964) and Mueller et al. (1966) that full penetration of bacterial floes by DO is expected at a bulk liquid DO concentration range of 0.6-2.5 mg/L or above, depending on the floe size. Based on that critical DO range, denitrification in the anoxic microzone could not occur in the 3-Bardenpho aerobic zone, in which DO concentration was maintained at a constant concentration of 3 mg/L. However, denitrification mediated by aerobic denitrifiers can occur under fully-aerobic conditions, thus, explaining the unaccounted for nitrogen loss observed in the 3-Bardenpho aerobic zone. 2) As discussed in Section 4.1.3.5, the increasing trend of nitrification with increasing acetate and methanol dosages, observed in the low ORP IACM tank, indicated that the nitrification in this tank had to be a heterotrophic one. 3) The decreases in the amount of the unaccounted for nitrogen loss at the high acetate and methanol dosages were associated with the carbon storage in the IACM tank, which was induced by the bleeding of acetate or the presence of methanol at relatively high 208 levels. Many observations in this study, as discussed in Sections 4.1.4.2, 4.1.4.3 and 4.1.5.2, suggested that both acetate and methanol were stored as PHB in the IACM tanks. As predicted by the working model (Robertson et al., 1988; 1991), the heterotrophic nitrification and the PHB synthesis in the IACM tanks would reduce the capacity of aerobic denitrification in the tanks. 4.4 Roles of ORP Control Range and External Substrates in Selection of Bacteria Population Although the heterotrophic nitrification and aerobic denitrification can explain most of the observations in this study without conflicts, it does not mean that the other biological reactions, such as autotrophic nitrification and anaerobic denitrification, did not exist in the three processes. As described in the literature review, the coexistence of autotrophic nitrifiers with aerobic denitrifiers in mixed cultures, under oxygen limiting concentrations, was possible; the mixture of the two could convert almost 100% of ammonia present to gaseous forms of nitrogen (Bock et al., 1995). Other than that, Gupta et al. (1994) showed that Thiosphaera pantotropha (aerobic denitrifiers) could stably coexist with autotrophic nitrifiers under highly aerobic conditions in an RBC system. Furthermore, aerobic denitrifiers and autotrophic nitrifiers could be naturally selected and coexisted under fully aerobic conditions (DO>5 mg/L) in a lab-scale RBC system treating wastewater containing polyvinyl alcohol (PVA) (Bang et al., 1995). The results from the above three studies suggested that the nitrification in their systems mainly relied on autotrophic nitrifiers, while the denitrification relied on aerobic denitrifiers. 209 Therefore, it was possible that autotrophic nitrifiers, aerobic denitrifiers (also heterotrophic nitrifiers), and other heterotrophs (including anaerobic denitrifiers and PAOs), all existed in the current three activated sludge systems. However, the distribution of the bacteria population selected in different processes would be different, depending on the environmental conditions, such as the ORP control range and the external substrate dosage. The complete nitrification in the 3-Bardenpho and high ORP processes at various substrate dosages appeared to suggest that autotrophic nitrifiers dominated the nitrifying population in these two processes. The increasing trend of nitrification, with the increasing external substrate dosage, observed in the low ORP process, suggested that heterotrophic nitrifiers became dominant in that process at the relatively high substrate dosages. On the other hand, the total amount of N O x removal in the 3-Bardenpho anaerobic and anoxic zones contributed the major part of total N O x removal in the entire process; this indicates that anaerobic denitrifiers would dominate the denitrifying population in the process. In the low ORP process, most N O x was removed in the IACM tank; this suggests that aerobic denitrifies would dominate the denitrifying population in the process. In the high ORP process, the N O x removal in both zones were significant; this suggests that both aerobic denitrifiers and anaerobic denitrifiers were presented in the process (Section 4.1.3.5). The nitrification in both the wide and narrow ORP processes was considered to be mediated by the autotrophic nitrifiers, because the external substrate dosages added into these two processes might be too low to select for heterotrophic nitrifiers. The better 210 nitrification in the wide ORP process than in the narrow ORP process is considered to be due to a higher average DO level maintained in the wide ORP process. The denitrifying population in the narrow ORP process would be about the same as that in the high ORP process; that is, both aerobic and anaerobic denitrifiers would exist in significant numbers. Since aerobic denitrifiers appear to have an ecological advantage in niches with fluctuating aerobic/anaerobic periods, the longer aeration cycle used in the wide ORP process, as well as the presence of zero DO period in the aeration cycle, would select more anaerobic denitrifiers in the wide ORP process than in the other three two stage processes. The much better performance of N O x removal achieved in the wide ORP process, than in the narrow ORP process, is considered to be due to the higher rate of anaerobic denitrification, compared to aerobic denitrification. 4.4.1 Role of ORP Control Range Between the two 2-stage processes, the only difference was the ORP control range. Therefore, different performances between the two processes may not be directly dependent on, but ultimately are related to, ORP control range. The most important difference between the high and low ORP processes was the nitrification in the IACM tank. Since the nitrification in the high ORP IACM tank was always completed, regardless of the changes in external substrate dosages, the ORP control range, or more exactly the ORP average level, is believed to directly control the selection of autotrophic nitrifiers. The difference in nitrification resulted in. many differences in performance between these two processes. For example, significant 211 differences were observed for the following parameters: the N O x removal in the anaerobic zone, the N O x removal in the IACM tank, the ortho-P release and uptake in the acetate addition runs, overall T N removal in methanol addition runs, and overall phosphorus removal in all runs. The most significant difference between the wide and narrow ORP processes was the amount of N O x removal in the IACM tank; this suggests that wide ORP control range, as well as the zero-DO period, selected the anaerobic denitrifiers. Due to this, the wide ORP process even achieved better nitrogen removal than the 3-Bardenpho process, under the specific conditions. As described in the literature, in addition to oxygen, organic substrate was generally considered as the controlling factor for aerobic denitrification (Ottow and Fabig, 1983; Richardson and Ferguson, 1992; Bang et al., 1995; Neef et al., 1996). Further, the working model (Robertson et al., 1988; 1990) implies that the bulk liquid ORP level could be another factor to control aerobic denitrification and heterotrophic nitrification, if a direct relationship between the intracellular redox and the bulk liquid redox exists. 4.4.2 Role of Organic Substrate Experimental results in this study indicated that carbon reserves, such as PHB, emerge as playing a key role in explaining the mechanisms of aerobic denitrification and heterotrophic nitrification. Acetate addition runs. As indicated by the zero V F A concentrations in the anaerobic zones of the three systems, the carbon reserves stored by PAOs in the anaerobic zone, 212 rather than acetate added or present in the influent, were actually available to the 3-Bardenpho anoxic zone and the two IACM tanks. The increasing trends of phosphorus release with increasing acetate dosage (Figure 4.31) suggested that the amount of carbon reserves stored in the anaerobic zone would increase with increasing acetate dosage. Therefore, the increasing trends of the N O x removal in the anoxic zone (Figure 4.15) and in the two IACM tanks (Figure 4.21), with respect to the acetate dosage, suggested that it was the increase in the amount of carbon reserves, rather than the increase in acetate dosage itself, that actually improved the denitrification in these three reactors. Therefore, the simultaneous improvements on the denitrification and the phosphorus uptake in the 3-Bardenpho anoxic zone, by increasing acetate dosage, suggested that a subpopulation of PAOs is denitrifiers, which utilize carbon reserves for denitrification and synthesis of polyphosphate. In the same way, the simultaneous improvements on the denitrification, nitrification and the phosphorus uptake in the IACM tank, suggested that the subpopulation of PAOs is aerobic denitrifiers and heterotrophic nitrifiers. The significant difference in the amount of ortho-P released in the anaerobic zone between the high and low ORP processes suggests the amount of carbon reserves stored in the anaerobic zone between these two processes was also different. Therefore, the significant difference in the amount of N O x removal in the IACM tank between the high and low ORP processes may result from the difference in the amount of available carbon reserves between these two tanks, rather than the difference in the ORP control range. However, it should be noted that ultimately, it was the ORP control range, which governed the difference in the amount of carbon reserves between these two processes. 213 The increasing trend of nitrification, observed in the low ORP tank, with increasing acetate dosage, suggested that the increase in the amount of carbon reserves improved the nitrification in this tank. As suggested by Robertson et al. (1988; 1991) and van Niel (1991), aerobic denitrifiers are also heterotrophic nitrifiers. Therefore, it is expected that both the nitrification and denitrification in the low ORP IACM tank would be enhanced by the increasing amount of the carbon reserves at the same time. However, the bleeding of acetate at a concentration of 10 mg/L into the IACM tank caused a significant decrease in the N O x removal and a significant increase the nitrification in the low ORP IACM tank. The bleeding of acetate also resulted in a less than expected amount of phosphorus uptake in the IACM tank; this was believed to be caused by phosphorus release with concomitant PHB storage in the aeration tank. The PHB metabolism reviewed by Dawes and Senior (1973) have been recognized by many researchers. The metabolism suggests that PHB is degraded in the presence of oxygen, when the external carbon sources are limited and that PHB should not be degraded, if both oxygen and an external carbon sources are present. According to this mechanism, the bleeding of acetate in the IACM tank would have blocked the degradation of the carbon reserves stored in the anaerobic zone, thus causing a decrease in N O x removal in the IACM tank, as well as a decrease in the amount of ortho-P uptake. However, this mechanism cannot explain the improvement in nitrification, by the bleeding of acetate. The working model proposed by Robertson et al. (1988; 1991) can easily explain all the above observations, resulted from the bleeding of acetate into the IACM tank, 214 provided that a subpopulation of PAOs is aerobic denitrifiers and heterotrophic nitrifiers. That is, acetate presented was used for heterotrophic nitrification and PHB storage with concomitant phosphorus release; therefore, it would not be available for aerobic denitrification. Methanol addition runs. Many researchers (e.g., Mechsner and Hamer, 1983; Neef et al., 1996) suggested that methanol is a preferred substrate for aerobic denitrification; methanol addition selected for aerobic denitrifiers under alternating anoxic/aerobic environment. As described in Sections 4.1.3.5 and 4.1.3.6, direct addition of methanol, at the low and medium dosages, into the two IACM tanks, improved the denitrification in these two tanks and the nitrification as well. The above results further confirmed the conclusions from literature. The decrease in the amount of N O x removal, in both IACM tanks, observed at the high methanol dosage, suggested that the presence of methanol at high dosage would have the same effect on aerobic denitrification as the bleeding of acetate, that is, PHB production from methanol under DO or nitrogen limiting conditions caused the decrease in the aerobic denitrification in the IACM tank. However, it should be noted that the role of PAOs, in aerobic denitrification or heterotrophic nitrification, during the methanol runs, was not clear. 4 . 4 . 3 Practical Implications of SND This research has investigated the role and mechanisms of SND in BNR processes and identified the conditions favorable to SND. The results from this research, in line with 215 the results from other studies, have the following implications in BNR processes design and operation. 4.4.3.1 Implications for Mainstream BNR Processes Denitrification in the fine bubble aerobic zone of BNR plants could vary from about 20% of the total nitrogen in the influent to all the nitrogen not captured in the sludge (Randall et al, 1992). However, when using a surface aeration system, this figure was between 50 and 90% (Barnard, 1997). The results reported in Section 4.1.3.6 indicated that the SND in the 3-Bardenpho aerobic zone always occurred at a degree of 10-20% of the influent T K N at a DO concentration of 3 mg/L. Furthermore, during the first two steady-state experimental runs, one of the three experimental systems was used as a 3-stage Bardenpho process with intermittent aeration control in its aerobic zone. It was found that the SND in the aerobic zone of this system could reach 35% (results not shown in this thesis). This nitrogen loss may be incorporated into BNR plant design, such as by reducing the size of anoxic zone for three-stage processes and by completely eliminating the secondary anoxic zone for five-stage processes. Otherwise, it could lead to other problems, such as secondary release of phosphorus (Randall, 1992), since it could result in not enough nitrate being recycled back to the anoxic zone. It was reported by Rabinowitz and Barnard (1995) that controlling DO levels in the aeration basin led to the conversion of the Kelowna, B.C. plant from a 5-stage Bardenpho process configuration to a 3-stage Bardenpho configuration; that is, the secondary anoxic zone was completely removed 216 from the process train. The improvements in process performance by this measure were also reported. 4.4.3.2 Implications for a Single-Stage IA Process Obviously, phosphorus removal is not required for this process configuration. The deterioration in the denitrification side of SND, by the bleeding of acetate, observed under frequently intermittent aeration and low DO conditions, implies that aerobic denitrification may not work in a single-stage IA process, if the feed contains some VFAs or readily biodegradable COD. On the other hand, it was also observed that heterotrophic nitrification would not play a significant role in ammonium oxidation under low DO conditions, if the feed contains no readily biodegradable COD. Therefore, it appears that heterotrophic nitrification and aerobic denitrification cannot function in a single IA process at the same time, since they require different influent feeds. However, methanol addition at a low dosage is a choice for single-stage IA processes, because methanol would enhance both heterotrophic nitrification and aerobic denitrification at the same time. It was observed in this study that a relatively longer aeration cycle improved total nitrogen removal, by utilizing alternate nitrification and denitrification (i.e., autotrophic nitrification and anaerobic denitrification). This suggests that a relatively long aeration cycle, with a relatively high DO concentration (but still lower than 1 mg/L) during the air-on period, could be used for a single-stage IA process. The results of many previous studies supported this conclusion. However, the air-off period should be short enough to prevent the IA tank from proceeding to deeply anaerobic conditions, which is favorable to 217 a shift in organism population to facultative anaerobes, at the expense of true denitrifiers (Manoharan et al., 1989). 4.4.3.3 Implications for Two-Stage IA processes SND is the major process to nitrogen removal in a two-stage process. Since partial nitrification in the process is performed by heterotrophs, for the same levels of nitrogen and phosphorus removal, the two-stage process requires more organic carbon, but less oxygen, compared to mainstream BNR processes, in which nitrification is solely performed by autotrophs. Furthermore, compared to separate stage nitrification and denitrification, SND occurred at a reduced rate, no matter which mechanism it was. This research demonstrated that the SNRs in the IACM tank were only 8-38% of the those (3.4-8.1 mgNHVN/gMLVSSxh" 1) in the 3-Bardenpho aerobic zone, operated at a DO concentration of 3 mg/L. This means that the two-stage processes, using SND, needs larger tankage in total. It was suggested from the results of this study that maximizing PHA storage in the anaerobic zone is the key factor in optimizing the nitrogen and phosphorus removal in the two-stage process. Therefore, design and operation of a two-stage process should provide enough V F A and avoid returning N O x into the anaerobic zone. To reduce the N O x level in the return sludge, low DO conditions should be maintained in the IACM tank to inhibit autotrophic nitrification and promote heterotrophic nitrification, since the latter is always accompanied by aerobic denitrification and leaves no N O x accumulation in the process. In the case where not enough V F A is present in the influent, acetate can be supplemented or 218 produced on-site by fermentation. On the other hand, V F A addition into the anaerobic zone should avoid bleeding of acetate into the IACM tank. In the case where VFAs are present in excess in the anaerobic zone, measures should be taken to reduce the negative impacts of bleeding of acetate on aerobic denitrification and phosphorus uptake in the IACM tank. One effective measure is to raise the DO level (or ORP level) in the IACM tank to promote autotrophic nitrification. The accumulated N O x by autotrophic nitrification, through the sludge return flow, will be recycled back to the anaerobic zone, thus allowing oxidation of VFAs to occur in the anaerobic zone. 4.4.3.4 Implications for a Novel BNR Process As described in the literature review, many studies proved that methanol addition induces PHB storage. In this study, PHA concentration in the dry sludge was measured during the run with methanol addition at a dosage of 15 mgCOD/L (run 11). As shown in Table 4.10, the PHA content in the sludge was consistently higher in the IACM tank than in the preceding anaerobic zone, although the data appear to be scattered, somewhat. Table 4.10 The PHA content in the sludge obtained from run 11. Run 11 Low ORP process High ORP process Date Anaerobic zone (mgPHA/gMLSS) IACM tank (mgPHA/gMLSS) Anaerobic zone (mgPHA/gMLSS) IACM tank (mgPHA/gMLSS) May 17 4.1 4.1 35.2 NA May 19 2.4 11.6 2.9 8.2 May 21 1.9 7.2 4.6 6.2 May 25 2.0 5.9 3.6 5.1 May 27 2.3 10.8 3.9 7.9 May 29 2.1 7.6 3.1 8.6 219 As discussed in Section 4.4.2, the results from the acetate addition runs suggested that a sub-population of PAOs are aerobic denitrifiers (aerobic denitrifying PAOs) and heterotrophic nitrifiers. If this is still the case with methanol, a novel process (Figure 4.67) for both nitrogen and phosphorus removal can be conceptualized for wastewater containing no VFAs (or readily biodegradable COD). Methanol addition Influent Internal recycle IACM Anoxic deration •Effluent Sludge return Figure 4.67 The configuration of the proposed BNR) process using methanol. As shown in Figure 4.67, this novel process would include three tanks: an IACM tank with low DO, an anoxic tank, and an aeration tank with relatively high DO. In the IACM tank, methanol addition and intermittent aeration with low DO would promote heterotrophic nitrification and aerobic denitrification, as well as PHB storage. The DO level and the methanol dosage could be used as two control parameters to regulate the amount of methanol flows used for aerobic denitrification and PHB storage. In the anoxic tank, the aerobic denitrifying PAOs would use nitrates as an electron acceptor to oxidize the PHB stored in the preceding IACM tank and to take up ortho-P, since aerobic denitrifiers work even better under anoxic conditions. Other denitrifying organisms would 220 use the slow biodegradable COD. In the final aeration tank, the remaining ammonia would be further oxidized by autotrophic nitrifiers. The internal recycle would provide the nitrates to the anoxic zone, while the sludge return recycles the biomass back to the process. As shown in Figure 4.67, the major difference between the new process and classic BNR process is that the anaerobic zone is replaced by the IACM tank in the new process; the rationale here is that the new process uses a different mechanism to produce PHB. In classic BNR processes, the energy for producing PHA (or PHB) is generated from the hydrolysis of polyphosphate; therefore, the processes involve ortho-P release. In the new process, the PHB storage could be considered as a side reaction involving aerobic denitrification; the energy to drive this reaction may be generated from extra reducing power available; therefore, the process may not involve ortho-P release. It should be noted that PHB storage can be produced in a large amount, although it is considered as a side reaction of aerobic denitrification. In fact, methanol has been used as a substrate solely for PHB production, as described in the literature review. At this concept stage, it is impossible to assess the real advantages and disadvantages of the new process. However, the following advantages appear to be possible: 1) Currently, many wastewater treatment plants in the U.S. still use methanol for nitrogen removal. This process uses methanol as an external substrate for both nitrogen and 221 phosphorus removal; therefore, there is a great potential to use this technology to retrofit existing methanol-supplemented plants for phosphorus removal. 2) Since methanol is a cheaper external carbon source, compared to acetate, this process can be used in the case where VFAs are not presented in the feed and cannot be produced on-site. 3) The negative impact of N O x in the return sludge can be eliminated in the new process. Finally, it should be remembered that this process is based on a pure assumption that a sub-population of aerobic denitrifiers and heterotrophic nitrifiers selected by methanol are PAOs. If this assumption is not true, the proposed process will not work for phosphorus removal. However, it can still improve nitrogen removal. 222 Chapter Five CONCLUSIONS AND RECOMMENDATIONS 5.1 Summary and Conclusions The main objective of this study was to demonstrate the feasibility of achieving carbon (C), nitrogen (N) and phosphorus (P) removals from domestic sewage, in a two-stage BNR process under the conditions favorable to simultaneous nitrification and denitrification (SND). The results obtained in this study have also contributed towards a better understanding of the mechanism of SND. A bench-scale, steady-state, evaluation was conducted for the two-stage process, in direct comparison with the performance of a three-stage Bardenpho process with fine bubble aeration operated at a DO concentration of 3 mg/L. The three systems, operated at a 15 day SRT at 20 °C with a pH of 7.0-7.4, were primarily fed with domestic sewage and supplemented with acetate and methanol. Based on the results of the experimental studies, the following conclusions are made. 5.1.1 Conclusions Related to the Overall Process Performance 1) The steady-state evaluation on process performance clearly showed that the two-stage process achieved similar levels of nitrogen and phosphorus removals. It is possible to reduce the influent T N concentration from 24-32 mgN/L to 6-12 mgN/L and to achieve an 223 effluent TP concentration less than 1 mgP/L at the same time, in the two-stage process. Complete T K N removal was achieved in the high ORP process and in the three-stage Bardenpho process, but not in the low ORP process. 2) Compared to the three-stage Bardenpho process, and even though the two-stage process removed the same amount of total COD from the influent (75-90%), the two-stage process produced more solids, containing a higher volatile content. The average 480 mg/L more solids production, with consistently 1-3% higher volatile content in the two-stage process, indicated that less endogenous oxidation of bacteria cells occurred in their two IACM tanks; this resulted from the intermittent aeration and low DO conditions. 3) Compared to the three-stage Bardenpho process, the two-stage process produced sludge with higher SVIs. However, the sludge with higher SVIs did not lead to wash-out of solids in the two-stage process due to the use of large secondary clarifiers (nominal HRT = 5 hours). The high SVIs observed in the two-stage process were believed to be related to the low DO conditions and the intermittent aeration mode. 4) The unaccounted for nitrogen loss (i.e., the amount of SND) in the aeration tank accounted for up to 50% of the influent T K N for the two-stage process under low DO conditions and near a constant of 15% for the three-stage Bardenpho process with fine bubble aeration. The 15% unaccounted for nitrogen loss should be incorporated into full-scale mainstream BNR plant design. 224 5.1.2 Conclusions Related to the Process Control ORP control range, through regulating the aeration and methanol addition, controlled the DO level in the mixed liquor and the aeration cycle in the two-stage process. Significant differences in the process performance were observed when different ORP control ranges were used. 1) Compared to high ORP control, low ORP control resulted in incomplete nitrification in the IACM tank and shifted the N O x removal from the anaerobic zone to the IACM tank; this led to a significant higher amount of orthophosphate release in the anaerobic zone, and subsequently higher overall TP removal. 2) While the differences in DO concentration and in aeration cycle between the high and low ORP IACM tanks were not significant, the low ORP IACM tank achieved a higher amount of SND. 3) Compared to narrow ORP control, wide ORP control resulted in higher N O x removal in the IACM tank and higher overall T N removal. The above three differences, as well as the relationship between DO and ORP, proved that absolute ORP control range can be used as a control parameter for the two-stage process. 4) It was also noticed that ortho-P was released in the IACM tank during the air-off period, at a rate of approximately 0.34 mgP/gMLVSSxh"1 in the presence of N O x and at a rate of approximately 0.8 mgP/gMLVSSxh"1, when N O x concentration was less than 0.5 225 mgN/L. Therefore, for a two-stage process, the optimal aeration cycle time will probably be determined by the requirement for phosphorus removal, rather than nitrogen removal. 5.1.3 Conclusion Related to the Role of External Substrate The results obtained from steady-state evaluations showed that both acetate and methanol additions, at relatively low dosages, were effective in improving both overall nitrogen and phosphorus removal in all three processes. Based on the discussion in Section 4.4, the carbon reserves stored in the anaerobic zone are effective in improving both aerobic denitrification and heterotrophic nitrification. Therefore, the key factors to optimize both nitrogen and phosphorus removal in a two-stage process are to maximize carbon storage in the anaerobic zone and to avoid the bleeding of VFAs or the presence of methanol at relatively high dosages in the IACM tank. 5.1.4 Conclusion Related to the SND Mechanism Many SND hypotheses were examined based on the results obtained from this study. The working model postulated by Robertson and Kuenen (1988; 1991) for aerobic denitrification and heterotrophic nitrification appeared to fit all of the observations in this study. However, anoxic microzone denitrification can also be a cause for part of the unaccounted for nitrogen loss in the aeration tank. 5.1.5 Conclusions Related to the Transient-State Experiments The following conclusions are drawn based on the process responses to the ammonium and nitrate shock loads, observed during the transient-state experiments. 226 1) The transient-state experiments, as independent experiments, more clearly showed that nitrification and denitrification occurred simultaneously in the two IACM tanks. 2) Under the ammonium shock loading conditions, significant amount of ammonium (up to 15 mgN/L) could disappear from the aerobic liquid phase instantly. However, it is uncertain what specific conditions cause such instantaneous drop and whether or not it can be utilized in nitrogen removal processes. 3) The results from the steady-state evaluation confirmed that the maximum rates (both nitrification and denitrification), obtained by the proposed technique, were representative of continuous-flow conditions. 4) Compared to the corresponding rates in the three-stage Bardenpho aerobic and anoxic zones, the maximum specific nitrification and denitrification rates in the IACM tank were considerably lower. 5.2 Recommendations Further research work is recommended in the following areas. 1. To optimize the two-stage IA process studied, the effects of the following parameters, such as process SRT, HRT of IACM tank, ORP control range, and intermittent aeration cycle, should be further investigated. 2. It was found in this study that significant SND in the IACM tank was usually accompanied by incomplete nitrification and a deterioration in sludge settleability; as such, adding a small reaeration tank after the IACM tank of the two-stage process is 227 recommended. This would enhance process performance in two aspects, such as stripping out N2 and CO2 to improve sludge settleability and oxidizing the small amount of ammonia left from the IACM tank. Therefore, an investigation into the modified configuration should be conducted. 3. Unaccounted for nitrogen loss should be further investigated in a sealed reactor, for obtaining the true nitrogen balance and gas composition. 4. Pilot-scale or full-scale experiments should be conducted to validate the results obtained from this study. 5. Further study should be conducted to confirm the phenomenon of instantaneous ammonia drop and to investigate its role in nitrogen removal processes. 6. Further study should be conducted to confirm the phenomenon of methanol induced carbon storage and to investigate its implications in BNR processes. 7. It was suggested in this study that a subpopulation of phosphorus accumulating organisms is also aerobic denitrifiers or heterotrophic nitrifiers. Further study by microbiologists should be conducted to confirm this suggestion. 8. The possibility of using the proposed experimental technique to study other reactions, especially the reactions with non-zero-order kinetics, should be investigated. 228 6. REFERENCES Alefounder P. R. and Ferguson S. J. (1981) The location of dissimilatory nitrite reductase and the control of dissimilatory nitrate reductase by oxygen in Paracoccus denithficans. Biochem. Journal., 1 9 2 , 231-240. Alexander M . , Marshall K. C. and Hirsch P. (1960) Autotrophy and heterotrophy in nitrification. Trans. Int. Congr. Soil Sci., 7th. Madison, Wise, 586-591. Alleman J. E . and Irvine R. L. (1980) Nitrification in the Sequencing Batch Biological Reactor. J. Water Pollut. Control Fed., 5 2 , 2747-2754. APHA (1992) Standard methods for the examination of water and wastewater, 18th edition. American Public Health Association, Washington D.C. Artola A., Balaguer M . D. and Rogola M . (1997) Heavy metal binding to anaerobic sludge. Water Res., 3 1 (5), 997-1004. Bakti N. A. K. and Dick, R. I. (1992) A model for a nitrifying suspended-growth reactor incorporating intraparticle diffusional limitation. Water Res., 2 6 (12), 1681-1690. Balakrishnan S. and Eckenfelder W. W. (1970) Nitrogen removal by modified activated sludge process. J. of Environ. Eng., Am. Soc. Civ. Eng., 9 6 , 501-512. Bang D. -Y., Watanabe Y. and Noike T. (1995) An experimental study on aerobic denitrification with polyvinyl alcohol as a carbon source in biofilms. Water Sci. Technol, 2 2 (8), 235-242. Barnard J. L. (1973) Biological nutrient removal without the addition of chemicals. Water. Res., 9 (5), 485-490. Barnard J. L. (1974) Cut P and N without chemicals. Water and Wastes Engineering, 1 1 , 33-36. Barnard J. L. (1984) Activated primary tanks for phosphate removal. Water SA, 1 0 , 121-126. Barnard J. L. (1997) A personal communication. Batchelor B. (1982) Kinetic analysis of alternative configurations for single-sludge nitrification/denitrification. J. Water Pollut. Control Fed., 54(11), 1493-1504. Beccari M . , Di Pinto A. C , Ramadori R. and Tomei M . C. (1992) Effects of dissolved oxygen and diffusion resistance on nitrification kinetics. Water Res., 2 6 (8), 1099-1104. Bell L. C , Richardson D. J. and Ferguson S. J. (1990) Periplasmic and membrane-bound respiratory nitrate reductases in Thiosphaera pantotropha: the periplasmic enzyme catalyses the first step in aerobic denitrification. FEBS lett, 2 6 5 , 85-87. 229 Benefield L. and Molz F. (1983) A model for the activated sludge process which considers wastewater characteristics, floe behavior, and microbial population. Biotechnol. Bioengng., 26, 352-361. Betlach M . R. and Tiedje J. M. (1981) Kinetic explanation for accumulation of nitrite, nitric oxide and nitrous oxide during bacterial denitrification. Appl. Environ. Microbiol, 42, 1074-1084. Bock E. , Schmidt I.., Stuven R. and Zart D. (1995) Nitrogen loss caused by denitrifying Nitrosomonas cells using ammonia or hydrogen as electron donors and nitrite as electron acceptor. Arch. Microbiol., 163, 16-20. Bock E. , Wilderer P. A. and Freitag A. (1988) Growth of Nitrobacter in the absence of dissolved oxygen. Water Res., 22 (2), 245-250. Booker N. A., Cooney E. L: and Priestley A. J. (1996) Ammonia removal from sewage using natural Australian zeolite. Water Sci. Technol., 34 (9), 17-24. Brodisch K. E. U. (1985) Interaction of different groups of microorganisms in biological phosphorus removal. Water Sci. Technol., 17, 89-97. Brodisch K. E. U. and Joyner S. J. (1983) The role of microorganisms other than Acinetobacter in biological phosphate removal in activated sludge processes. Water Sci. Technol, 15(3/4) 117-125. Burdick C. R., Refling D. R. and Stensel H. D. (1982) Advanced biological treatment to achieve nutrient removal. J. Wat. Pollut. Control Fed., 54, 1078-1086. Carley B. N. and Mavinic D. S. (1991) The effects of external carbon loading on nitrification and denitrification of a high-ammonia landfill leachate. Res. J. Wat. Pollut. Control Fed., 63, 51-59. Carter J. P. Hsiao Y. H. , Spiro S. and Richardson D. J. (1995) Soil and sediment bacteria capable of aerobic nitrate respiration. Appl. Environ. Microbiol, 61 (8), 2852-2858. Carucci A., Ramadori R., Rossetti S. and Tomei M . C. (1996) Kinetics of denitrification reactions in single sludge systems. Water Res., 30 (10), 51-56. Castignetti D. and Hollocher T. C. (1984) Heterotrophic nitrification among denitrifiers. Appl. Environ. Microbiol, 47, 620-623. Castignetti D., Palutsis D. and Turkey J (1990) An examination of proton translocation and energy conservation during heterotrophic nitrification. FEMSMicrobiol. Lett., 66, 175-182. Charpenitier J., Florentz M. and David G. (1987) Oxidation-reduction potential (ORP) regulation: a way to optimize pollution removal and energy saving in the low load activated sludge process. Water Sci. Technol, 19 (3/4), 645-655. 230 Charpenitier J., Godart H., Martin G. and Mogno Y. (1989) 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 Sci. Technol., 21 (10/11), 1209-1223. Chen Y. S. and Bungay H. R. (1981) Microelectrode studies of oxygen transfer in trickling filter slimes; Biotechnol. Bioengng., 23, 781-792. Christensen M . H. and Harremoes P. (1977) Biological denitrification of sewage: a literature review. Prog. Water Technology, 8 (4/5), 509-555. Claus G. and Kutzner H. J. (1985) Denitrification of nitrate and nitric acid with methanol as carbon source. Appl. Microbiol. Biotechnol., 92, 378-381. Comeau Y. (1989) The Role of Carbon Storage in Biological Phosphate Removal from Wastewater. Ph.D. Thesis, Univ. of British Columbia, Vancouver, BC. Comeau Y., Hall K. J., Hancock R. E. W. and Oldham W. K. (1985) Biological model for enhance biological phosphorus removal. Proc. Internal Conf. on New Directions and Research in Waste Treatment and Residuals Management, Vancouver, B.C. June, 1985. Comeau Y., Oldham W. K. and Hall K. J. (1987) Dynamics of carbon reserves in biological dephosphatation of wastewater. Biological Phosphate Removal from Wastewater, Proc. of an IAWPRC Specialized Conference. Rome, Pergamon Press, Oxford, 39-56. Cooper R. L. and Catchpole J. R. (1973) The biological treatment of carbonization effluents. IV. Water Res., 7, 1137-1153. Dawes E. A. and Senior P. J. (1973) The role and regulation of energy reserve polymers in microorganisms. Adv. Microb. Physiol, 10, 135-266. de la Menardiere M . , Charpentier J., Vachon A. and Martin G. (1991) ORP as a control parameter in a single sludge biological nitrogen and phosphorus removal activated sludge system. Water SA, 17(2), 123-132. Downes M . T. (1988) Aquatic nitrogen transformations at low oxygen concentrations. Appl. Environ. Microbiol, 54, 172-175. Downing A. L. , Painter H. A. and Knowles G. (1964) Nitrification in the activated sludge process. J. Proc. Inst. Sew. Purif., p. 130. Downing A. L . , Tomlinson T. G. and Truesdale G. A. (1964) Effect of inhibitors on nitrification in the activated-sludge process. J. Inst. Sew. Purif, 537-554. Drews R. J. L. C. and Graeff A. M . (1973) Nitrogen elimination by rapid alternation of aerobic/anoxic conditions in orbal activated sludge plants. Water Res., 7 (8), 1183-1194. Ekama G. A., Marais G. v. R., Siebritz I. P. and et al. (1984) Theory, Design and Operation of Nutrient Removal Activated Sludge Processes. A collaborative 231 information document prepared for the Water Research Commission by the University of Cape Town, City Council of Johannesburg and the National Institute for Water Research of the C SIR. Focht D. D. and Chang A. C. (1975) Nitrification and denitrification processes related to wastewater treatment. Adv. Appl. Microbiol., 19, 153-186. Focht D. D. and Verstraete W. (1977) Biological ecology of nitrification and denitrification. In: Adv. Microb. Ecol., 1, (Ed. M . Alexander), Plenum Press, New York, 135-214. Gabb D. M . D., Ekama G. A., Jenkins D., Wentzel M . C , Casey T. G. and Marais G. v. R. (1996b) Filamentous organism bulking in nutrient removal activated sludge systems. Paper 5: Experimental examination of aerobic selectors in anoxic-aerobic systems. Water SA, 22(2), 139-146. Gabb D. M . D., Ekama G. A., Jenkins D., Wentzel M . C , Casey T. G. and Marais G. v. R. (1996a) Filamentous organism bulking in nutrient removal activated sludge systems. Paper 4: System configurations and operating conditions to develop low F / M filament bulking sludges at laboratory-scale. Water SA, 22 (2), 127-137. Gerber A., de Villiers R. H., Mostert E. S. and van Riet C. J. J. (1987) The phenomenon of simultaneously phosphate uptake and release, and its importance in biological nutrient removal. In: Biological Phosphate Removal from Wastewater, Proc. of an IAWPRC Specialized Conf. (Ed. Ramadori R.), Rome, Pregamon Press, Oxford. Gerber A., Mostert E. S., Winter C. T. and de Villiers R. H. (1986) The effect of acetate and other short-chain carbon compounds on the kinetics of biological nutrient removal. Water SA, 12(1), 7-12. Gersberg R. M . and Allen D. W. (1985) Phosphorus uptake by Klebsiella pneumoniae and Acinetobacter calcoaceticus. Water Sci. Technoh, 17, 113-118. Gibb A. J. (1995) Optimization of Biological Nutrient Removal in a Pilot-Scale Trickling Filter-Activated Sludge Process. Ph.D. Thesis, Univ. of British Columbia, Vancouver, B.C. Gupta S. K., Raja S. M . and Gupta A. B. (1994) Simultaneous nitrification-denitrification in a rotating biological contactor. Environ. Technol. Lett, 15, 145-153. Hallin S., Rothman M . and Pell M . (1996) Adaptation of denitrifying bacteria to acetate and methanol in activated sludge. Water Res., 30, 1445-1450. Hailing-S<|>rensen B. and J(|)rgensen S. E. (1993) The Removal of Nitrogen Compounds from Wastewater. Elservier Science Publishers, Amsterdam. Hanaki K., Wantawin C. and Ohgaki S. (1990a) Effects of the activity of heterotrophs on nitrification in a suspended-growth reactor. Water Res., 24 (3), 289-296. 232 Hanaki K., Wantawin C. and Ohgaki S. (1990b) Nitrification at low levels of dissolved oxygen with and without organic loading in a suspended-growth reactors. Water Res., 24 (3), 297-302. Hao O. J. and Huang J (1996) Alternating aerobic-anoxic process for nitrogen removal: process evaluation. Water Environ., Res., 68 (1), 83-93. Hao O. J., Richard M . G. and Jenkins D. (1983) The half-saturation coefficient for dissolved oxygen: a dynamic method for its determination and its effect on dual species composition. Biotech. &Bioengin., 25, 403-416. Harrison D. E. F. (1972) Physiological effects of dissolved oxygen tension and redox potential on growing populations of microorganisms. J. Appl. Chem. Biotechnol, 22 (3), 417-440. Heduit A. and Thevenot D. R. (1989) Relation between redox potential and oxygen levels in activated-sludge reactors. Water Sci. Technol., 21, 947-956. Heduit A. and Thevenot D. R. (1992) Elements in the interpretation of platinum electrode potentials in biological treatment. Water Sci. Technol., 26 (5/6), 1335-1344. Heduit A , Duchene P. H. and Sintes L. (1990) Optimization of nitrogen removal in small activated sludge plants. Water Sci. Technol., 22, 123-130. Hernandez D. and Rowe J. J. (1988) Oxygen inhibition of nitrate uptake is a general regulatory mechanisms in nitrate respiration. J. Biol. Chem., 263, 7937-7939. Hoffmann R. and Marais G.v.R. (1977) Phosphorus Removal in the Modified Activated Sludge Process. Research Report W 22, Department Of Civil Engineering, University Of Cape Town. Hollocher T. C , Tate M . E. and Nicholas D. J. D. (1981) Oxidation of ammonia by Nitrosomonas europaea. J. Biol. Chem., 256, 10834-10836. Huang C. -S. and Hopson N. E. (1974) Nitrification and denitrification rate in biological processes. J. of Environ. Eng., Am. Soc. Civ. Eng., 100, 409-422. Huang J. (1993) Process Evaluation and Mathematical Modeling of the Completely Mixed Alternating Aerobic-Anoxic Activated Sludge Process. Ph.D. dissertation, Univ. Of Maryland, Maryland, USA. Huang M . Y. and Bungay H. R. (1973) Microprobe measurements of oxygen concentrations in mycelial pellets. Biotechnol. Bioengng, 15, 1193-1197. IAWPRC Task Group on Mathematical Modeling for Design and Operation of Biological Wastewater Treatment (1986) Activated Sludge Model, IAWPRC Scientific and Technical Report No. 1. Ip S. Y., Bridger J. S. and Mills N. F. (1987) Effect of alternating aerobic and anaerobic conditions on the economics of the activated sludge system. Water Sci. Technol., 19 (5/6), 911-918. 233 Isaacs S. H., Henze M , S(t>eberg H. and Kummel M (1994) External carbon source addition as a means to control an activated sludge nutrient removal process. Water Res. 28(3), 511-520. Jayamohan S., Ohgaki S. and Hanaki K. (1988) Effect of DO on kinetics of nitrification. Water Supply, 6, 141-150. Jones P. H., Tadwalker A. D. and Hsu C. L. (1987) Enhanced uptake of phosphorus by activated sludge—effect of substrate addition. Water Res., 21, 301-308. Kang C. K., Lee H. S. and Kim J. H. (1993) Accumulation of PHA and its copolyesters by Methylobacterium sp. K C T C 0048. Biotechnol. Lett, 15 (10), 1017-1020. Kelly D. P. (1978) Bioenergetics of chemolithotrophic bacteria. In: Companion to microbiology (Eds. A T . Bull and P. M . Meadow), Longman, London, 363-386. Kerrn-Jespersen J. P. and Henze M . (1993) Biological phosphorus uptake under anoxic and aerobic conditions. Water Res., 27 (4), 617-624. Kiff R. J. (1972) The ecology of nitrification-denitrification systems in activated sludge. Wat. Pollut. control, 71, 475-484. Killham K. (1986) Heterotrophic nitrification. In: Nitrification. (Ed. Prosser J. I.), HAL Press. Oxford, England, 117-126. Kjaergaard L. (1977) The redox potential: Its use and control in biotechnology. In: Advances in Biochemical Engineering, (Eds. Ghose T.K., Fiechter A. and Blakebrough), Springer-Verlag, Berlin, 131-150. Kleiner D. (1985) Bacterial ammonium transport. FEMS Microbiology Review. 32, 87-100. Knowles R. (1982) Denitrification. Microbiol. Rev., 46, 43-70. Koch F. A. (1997) A personal communication Koch F. A. and Oldham W. K. (1985) Oxidation-reduction potential - a tool for monitoring, control, and optimization of nutrient removal systems. Water Sci. Technol., 17 (1 1/12), 259-281. Koon J. H. and Kaufman W. J. (1975) Ammonia removal from municipal wastewaters by ion exchange. J. Wat. Pollut. Control Fed., 47 (3). Krul J. M . (1976) The relationship between dissimilatory nitrate reduction and oxygen uptake by cells of an Alcaligenes strain in floes and in suspension and by activated sludge floes. Water Res., 10 (4), 337-341. Kshirsagar M . , Gupta A. B. and Gupta S.'.K. (1995) Aerobic denitrification studies on activated sludge mixed with Thiosphaera pantotropha. Environ. Technol. Lett., 16, 35-43. 234 Kucera I. and Dadak V. (1983) The effect of uncoupler on the distribution of the electron flow between the terminal acceptors and nitrite in the cells of Paracoccus denitrificans. Biochem, Biophys. Res. Comm., 117, 252-258. Kuba T., Smolders G. J. F., van Loosdrecht M . C. M . and Heijnen J. J. (1993) Biological phosphorus removal from wastewater by anaerobic-anoxic sequencing batch reactor. Water Sci. Technol., 27,241-252. Kuba T., van Loosdrecht M . C. M . and Heijnen J. J. (1996) Phosphorus and nitrogen removal with minimal COD requirement by integration of denitrifying dephosphatation and nitrification in a two-sludge system. Water Res., 30, 1702-1710. Kuba T., van Loosdrecht M . C. M . , Brandse and Heijnen J. J. (1997) Occurrence of denitrifying phosphorus removing bacteria in modified UCT-type wastewater treatment plants. Water Res., 31, 777-786. Kuenen J, G. and Robertson L! A. (1987) Ecology of nitrification and denitrification. In: The Nitrogen and Sulfur Cycles (Eds. J. A. Cole and S. J. Ferguson), Cambridge University Press, 161-218. Kuenen J. G. and Robertson L. A. (1994) Aerobic and anaerobic nitrification/denitrification processes. In: Water Quality International '94, Proc. IA WQ 17th Biennial Conf, July 19-24, 1994, Budapest, Hungary, 91-99. Kugelman I. J. and Spector M . (1988) Nutrient removal in staged biological reactors. In: Proc. of 3rd WPCF/JSWA Joint Technical Seminar on Sewage Treatment Technology, Tokyo, May 12-14, 3-27. Kugelman I. J. Spector M . , Harvilla A. and Parees D. (1991) Aerobic denitrification in activated sludge. In: Proc. of the 1991 Environ. Eng. Special Conf, Amer. Soc. Civ. Eng., Reno, Nev., 312-318. La Motta E. J. and Shieh W. K. (1979) Diffusion and reaction in biological nitrification. J. of Environ. Eng., Am. Soc. Civ. Eng., 105 (4), 655-673. Landine, R. C. (1991). Some Factors Affecting the Performance of an IACM Process in Nitrification and Denitrification. M.S. Thesis, Univ. of New Brunswick, Fredericton, Canada. Lau A. O., Strom P. F. and Jenkins D. (1984a) Growth kinetics of Sphaerotilus natans and a floe former in pure and dual continuous culture. J. Wat. Pollut. Control Fed, 56, 41-51. Lau A. O , Strom P. F. and Jenkins D. (1984b) The competitive growth of floc-forming and filamentous bacteria: a model for activated sludge bulking. Growth kinetics of Sphaerotilus natans and a floe former in pure and dual continuous culture. J. Wat. Pollut. Control Fed., 56, 52-61. Lee S. A., Koopman B., Bode H. and Jenkins D. (1983) Evaluation of alternative sludge settling indices. Water Res., 17 1421-1426. 235 Lefevre F., Audic J. M . and Bujon B. (1993) Automatic regulation of activated sludge aeration - single-tank nitrification-denitrification. Water Sci. Technol, 28 (10), 289-298. Lie E. and Welander T. (1994) Influence of dissolved oxygen and oxidation-reduction potential on the denitrification rate of activated sludge. In: Water Quality International '94. IAWQ 17th Biennial International Con/., July, 1994. Budapest, Hungary, 73-82. Lin K. -C and Tsang K. (1989) Nitrogen removal in an intermittently aerated completely mixed reactors. Environ. Technol. Lett., 10 (1), 1-8. Lloyd D., Boddy L. and Davies J. P. (1987) Persistence of bacterial denitrification capacity under aerobic conditions: the rule rather than the exception. FEMS Microbiol. Ecol, 45, 185-190. Lo C. K., Yu C. W., Tarn N. F. Y. and Traynor S. (1994) Enhanced nutrient removal by oxidation-reduction potential (ORP) controlled aeration in a laboratory scale extended aeration treatment system. Water Res., 28 (10), 2087-2094. Lotter L. H. (1985) The role of bacterial phosphate metabolism in enhanced phosphorus removal from the activated sludge process. Water Sci. Technol., 17 (11/12), 127-138. Lotter L. H. and Murphy M . (1985) The identification of heterotrophic bacteria in an activated sludge plant with particular reference to polyphosphate accumulation. Water SA, 11 (4), 179-184. Lotter L. H. and Pitman A. R. (1992) Improved biological phosphorus removal resulting from the enrichment of rector feed with fermentation products. Water Sci. Technol., 26 (5/6), 943-953. Ludzack F. J. and Ettinger M . B. (1962) Controlling operation to minimized activated sludge effluent nitrogen. J. Water Pollut. Control Fed., 34 (9), 920-931. Malnou D., Meganck M . , Faup G. M . and du Roster M . (1984) Biological phosphorus removal: study of the main parameters. Water Sci. Technol., 16, 173-185. Manoharan R. (1988) Readily Biodegradable COD as an Indicative Parameter in Estimating the Efficacy of Sewage for Biological Excess Phosphorus Removal. Ph.D. Thesis, Univ. of British Columbia, Vancouver, B.C. Manoharan R., Liptak S., Parkinson P., Mavinic D. S. and Randall C. W. (1989) Glucose vs. methanol as carbon sources for denitrification in biological treatment of leachate. Environ. Technol. Lett, 10 (8), 707-716. Matsche N. F. (1972) The elimination of nitrogen in the treatment plant of Vienna-Blumenthal. Water Res., 6, 481-486. Matsche N. F. and Spatzierer G. (1975) Austrian plant knocks out nitrogen. Water and waste Engineering, 12 (1), 19-24. 236 McCarty P. L. , Beck L. and Amant P. S. (1969) Biological denitrification of wastewaters by addition of organic materials. 24thIndustrial Waste Conf., Purdue Univ., Lafayett, 1271-1285. Mechsner K. and Hamer G. (1983) Denitrification by methanotrophic/methylotrophic bacterial associations in aquatic environments. In: Denitrification in the Nitrogen Cycle (Ed. Golterman H. L ) , Plenum Publishing Corp., New York, 257-271. Meiberg J. B. M . , Bruinenberg P. M . and Harder W. (1980) Effect of dissolved oxygen tension on the metabolism of methylated amines in Hyphomicrobium X in the absence and presence of nitrate: evidence for 'aerobic' denitrification. J. Gen. Microbiol., 120, 453-463. Mercer B. W., Ames L. L . , Touhill C. J., Van Slyke W. J. and Dean R. B. (1970) Ammonia removal from secondary effluents by selective ion exchange. J. Wat. Pollut. Control Fed., 42 (2) Part 2. Mines R. O. Jr. (1996) Assessment of AWT systems in Tampa Bay area. J. of Environ. Eng., Am. Assoc. Civ. Eng., 122 (7), 605-611. Montgomery D. C. (1984) Design and Analysis of Experiments. John Wiley & Sons Inc., Toronto, Canada. Moore S. F. and Schroeder E. D. (1970) An investigation of the effects of residence time of anaerobic bacterial denitrification. Wat. Res., 4, 685-694. Moriyama K , Takahashi M . and Harada Y. (1993) Retrofitting and operation of small extended aeration plants for advanced treatment-Some experiences in Japan. Water Sci. Technol, 28 (10), 377-385. Mostert E. S., Gerber A. and van Riet C. J. J. (1988) Fatty acid utilisation by sludge from full-scale nutrient removal plants, with special reference to the role of nitrate. Water SA. 14 (4), 179-184. Mueller J. A., Boyle W. C. and Lightfoot E. N. (1966) Oxygen diffusion through a pure culture floe of Zoogloea ramigera. Ind. Waste Conf., Purdue University, 2, 964-995. Munch E. , Lant P. and Keller J. (1996) Simultaneous nitrification and denitrification in bench-scale sequencing batch reactors. Water Res., 30 (2), 277-284. Murphy K. L. and Dawson (1972) The temperature dependency of biological denitrification. Water Res., 6, 71-83. Nakanishi H , Fukagawa M . , Takeuchi, M . , Ishikawa M . , Ukita M . , and Murakami, S. (1990) Biological nitrogen removal in a complete mixing type aerator with ORP control. Water Sci. Technol, 22 (7/8), 269-270. Nakis N. Rebhun M . and Sheindorf C. (1979) Denitrification at various carbon to nitrogen ratios. Water Res., 13, 93-98. 237 Neef A., Zaglauer A., Meier H., Amann R., Lemmer H. and Schleifer K. -H. (1996) Population analysis in a denitrification sand filter, conventional and in situ of Paracoccus spp. in methanol-fed biofilms. Appl. Environ. Microbiol, 62 (12) 4329-4339. Nelson L. M . and Knowles R. (1978) Effect of oxygen and nitrate on nitrogen fixation and denitrification by Azospirillum brasilence grown in continuous culture. Can. J. Microbiol, 24, 1395. Nelson P. O., Chung A. K. and Hudson M . C. (1981) Factors affecting the fate of heavy metals in the activated sludge process. J. Wat. Pollut. Control Fed., 53, 1323-1333. Nicholls H. A. and Osborn D. W. (1979) Bacteria stress: a prerequisite for biological phosphorus removal. J. Water Pollut. Control Fed, 51 (3), 557-569. Nicholls H. A., Osborn D. W. and Pitman A. R. (1987) Improvement of the stability of the biological phosphorus removal process at the Johannesburg Northern Works. Proc. IAWPRC Spec. Con/., Biological Phosphorus Removal from Wastewaters, Rome, Italy, Sept., 261-272. Nurse G. R. (1980) Denitrification with methanol: microbiology and biochemistry. Water ites., 14, 531-537.. Nyberg U., Aspegren H., Andersson B. and Jansen J. la Cour (1992) Full-scale application of nitrogen removal with methanol as carbon source. Water Sci. Technol, 26 (5/6), 1077-1086. Oldham W. K. (1985a) Full scale optimization of biological phosphorus removal at Kelowna, Canada. Water Sci. Technol, 17, 243-257. Oldham W. K. (1985b) Operating Experiences with the Kelowna Pollution Control Center - Treatment Results. In: Technology Transfer Seminar on Biological Phosphors Removal in Municipal Wastewater Treatment, Penticton, B.C. Oldham W. K. (1997) A personal communication Ottow J. C. G. and Fabig W. (1983) Influence of oxygen aeration on denitrification and redox level in different bacterial batch cultures. In: Planetary Ecology (Eds. Caldwell D E . , Brierley J. A. and Brierley C. L.), Van Norstrand Reinhold Company, New York. 427-443. Painter H. A. (1970) A review of literature on inorganic nitrogen metabolism in microorganisms. Water Res., 4, 393-450. Palm J. C , Jenkins D. and Kaufman W. J. (1980) Relationship between organic loading, dissolved oxygen concentration and sludge settleability in the completely mixed activated sludge process. J. Water Pollut. Control Fed, 52, 2484-2493. Payne W. J. (1981) Denitrification. John Wiley & Sons, Inc. N. Y. 238 Porter K. S. (1975) Nitrogen Phosphorus Food Production, Waste and the Environment. Ann Arbor Science Publishers, Inc. Poth M. (1986) Denitrogen production from nitrite by a Nitrosomonas isolate. Appl. Environ. Microbiol., 52, 957-959. Poth M . and Focht D. D. (1985) 15Nitrification kinetic analysis of N 2 0 production by Nitrosomonas europaea: an examination of nitrifier denitrification. Appl. Environ. Microbiol., 49, 1134-1141. Prakasam T. B. S. and Loehr R. C. (1972) Microbial nitrification and denitrification in concentrated wastes. Water Res., 6, 859-869. Prosser J. I. (1990) Mathematical modeling of nitrification processes. In: adv. Microb. Ecol. (Ed. Marshall, K. C ) , Plenum Press, New York, N.Y. 263-304. Rabinowitz B. (1985) The Role of Specific Substrate in Excess Biological Phosphorus Removal. Ph.D. Thesis, Univ. of British Colombia, Vancouver. B.C. Rabinowitz B. and Barnard J. L. (1995) Biological nutrient removal in western Canada. Water Res., 1995 year book, 22-28. Randall C. W., Barnard J. L. and Stensel H. D. (1992) Design and retrofit of wastewater treatment plants for biological nutrient removal. Water Quality Management Library -Volume 5. Technomic Publishing Company, Inc. Lancaster, Pennsylvania. Richardson D. J. and Ferguson S. (1992) The influence of carbon substrate on the activity of the periplasmic nitrate reductase in aerobically grown Thiosphaera pantotropha. Arch. Microbiol., 157, 535-537. Rittmann B. E. and Langeland W. E. (1985) Simultaneous denitrification with nitrification in single-channel oxidation ditches. J. Wat. Pollut. Control Fed., 57 (4), 300-308. Robertson L . A. and Kuenen J. G. (1984a) Aerobic denitrification - old wine in new bottle? Antonie van Leeuwenhoek, 50, 525-544. Robertson L. A. and Kuenen J. G. (1984b) Aerobic denitrification: a controversy revived. Arch. Microbiol., 139, 351-354. Robertson L. A. and Kuenen J. G. (1988) Heterotrophic nitrification in Thiosphaera pantotropha: oxygen uptake and enzyme studies. J. Gen. Microbiol., 134, 857-863. Robertson L. A. and Kuenen J. G. (1990) Combined heterotrophic nitrification and aerobic denitrification in Thiosphaera pantotropha and other bacteria. Antonie van Leeuwenhoek, 57, 139-152. Robertson L. A. and Kuenen J. G. (1992) Nitrogen removal from water and waste. In: Microbial Control of Pollution, (Eds: Fry J. C , Gadd G. M . , Herbert R. A., Jones C. W. and Watson-Craik I. A ) , Cambridge Univ. Press, Great Britain. 239 Robertson L. A., Cornelisse R., de Vos P., Hadioetomo R. and Kuenen J. G. (1989a) Aerobic denitrification in various heterotrophic nitrifiers. Antonie van Leeuwenhoek, 56,289-299. Robertson L. A., van Niel E. W. J., Torremans R. A. M . and Kuenen J. G. (1988). Simultaneous nitrification and denitrification in aerobic chemostat cultures of Thiosphaera pantotropha. Appl. Environ. Microbiol., 54 (11), 2812-2818. Schon G., Geywitz. S. and Mertena F. (1993) Influence of dissolved oxygen and oxidation-reduction potential on phosphate release and uptake by activated sludge from sewage plants with enhanced biological phosphorus removal. Water Res., 27 (3), 349-354. Schuldiner S., Piersma B. J. and Warner T. B. (1966) Potential of a platinum electrode at low partial pressures of hydrogen and oxygen. J. Electrochemical Society, 113 (6), 573-577. Schulze-Rettmer R. (1991) The simultaneous chemical precipitation of ammonium and phosphorus in the form of magnesium-ammonium-phosphate. Wat. Sci. Technol, 23, 659-667. Sen D., Randall C. W. and Grizzard T. J. (1990) Developing a multiple parameters nutrient removal systems, Proc. 1990 Amer. Soc. Of Civil Eng. Environ. Eng. Spec. Conf, Washington D.C. Sezgin M . , Jenkins D. Parker D. S. (1978) A united theory of filamentous activated sludge bulking. J. Water Pollut. Control Fed., 50, 362-368. Shoda M , Ohsumi T. and Udaka S. (1980) Screening for high phosphate accumulation bacteria. Agric. Biol. Chem., 44, 319-324. Siebritz J., Ekama G. A. and Marais G. v. R. (1983) A parametric model for biological excess phosphorus removal. Water Sci. Technol, 15 (3/4), 127-152. Simpkin T. J. and Boyle W. C. (1988) The lack of repression by oxygen of the denitrification enzymes in activated sludge. Water'Res., 22 (2), 201-206. Simpkins M . J. and McLaren A. R. (1978) Consistent biological phosphate and nitrate removal in an activated sludge plant. Prog. Water Technol, 10, 433-442. Sorm R., Bortone G., Saltarelli R., Jenicek P., Wanner J. and Tilche A. (1996) Phosphate uptake under anoxic conditions and fixed-film nitrification in nutrient removal activated sludge system. Water Res., 30, 1573-1584. Steinmueller W. and Bock E . (1977) Enzymatic studies on autotrophically, mixotrophically and heterotrophically grown Nitrobacter agilis with special reference to nitrite oxidase. Arch. Microbiol, 115, 51-54. Stensel H. D. (1981) Biological nitrogen removal system design. In: Water-1980, AIChE Sympo. Ser., 77, 327-338. 240 Stensel H. D., Loehr R. C. and Lawrence A. W. (1973) Biological kinetics of suspended growth denitrification. J. Water Pollut. Control Fed., 45 (2), 249-261. Stenstrom M . K. and Poduska R. A. (1980) The effect of dissolved oxygen concentration on nitrification. Water Res., 14 (6), 643-649. Stenstrom M . K. and Song S. S. (1991) Effect of oxygen transport limitation on nitrification in the activated sludge process. Res. J. Wat. Pollut. Control Fed., 63 (3), 208-219. Stouthamer A. H. , Van't Riet J. and Oltman L. F. (1980) Respiration with nitrate as acceptor. In: Diversity of Bacterial Respiration Systems, Vol. II, (Ed. Knowles C. J.), CRC Press, West Palm Beach, FL. Strand S. E . , McDonnell A. J. and Unz R. F. (1985) Concurrent denitrification and oxygen uptake in microbial films. Water Res., 19 (3), 335-344. Sundermeyer-Klinger H. , Meyer W., Warninghoff B. and Bock E. (1984) Membrane bound nitrite oxidoreductase of Nitrobacter. evidence for a nitrate reductase system. Arch Microbiol., 140, 153-158. Suresh N. R. W., Timmerman M . , Wells J., Cbccia M . , Roberts M. F. and Halvorson H. O. (1985) New strategies for the isolation of microorganisms responsible for phosphate accumulation. Water Sci. Technol., 17 (11/12), 99-111. Suzuki T., Yamane T. and Shimizu S. (1986a) Kinetics and effect of nitrogen source feeding on production of poly-P-hydroxybutyric acid by fed batch culture. Appl. Microbiol. Biotechnol., 24, 366-369. Suzuki T., Yamane T. and Shimizu S. (1986b) Mass production of poly-P-hydroxybutyric acid by fed-batch culture with controlled carbon/nitrogen feeding. Appl. Microbiol. Biotechnol., 24, 370-374. Tarn N. F. Y., Wong Y. S. and Leung G. (1992) Effect of exogenous carbon sources on removal of inorganic nutrient by the nitrification-denitrification process. Water Res., 26 (9), 1229-1236. Timmermans P. and van Haute A. (1983) Denitrification with methanol: Fundament study of the growth and denitrification capacity of Hyphomicrobium sp. Water Res., 17 (10), 1249-1255. Toerien D. F., Gerber A., Lotter L. H. and Cloete T. E. (1990) Enhanced biological phosphorus removal in activated sludge systems. In: Adv. Microb. Ecol. (II), (Ed. K. C. Marshall), Plenum Press, New York, NY. Turk O. (1986) The Feasibility of a Shortened Pathway for Nitrogen Removal from Highly Nitrogenous Wastes. Ph.D. Thesis, Univ. of British Columbia, Vancouver, B.C. U.S. EPA (1975) Process Design Manual for Nitrogen Control. Office of Technology Transfer, Cincinnati, Ohio. 241 U.S. EPA (1976) Process Design Manual for Phosphorus Removal. U.S. EPA Technology Transfer, 625/1-76-00la, Washington, D.C. U.S. EPA (1987) Technology Transfer Design Manual—Phosphorus Removal. USEPA/625/1-87/001 CERI Cinn., OH 45268. U.S. EPA (1993)Manual-Nitrogen Control. EPA/625/R-93/010. Washington, DC. Ueda S., Matsumoto S., Takagi A. and Yamane T. (1992) Synthesis of poly (3-hydroxybutyrate-co-3-hydroxyvalerate) from methanol and w-amyl alcohol by the methylotrophic bacteria Paracoccus denitrificans and Methylobacterium extorquens. Appl. Environ. Microbiol, 58 (11), 3574-3579. van der Geest A. T. and Witvoet W. C. (1977) Nitrification and denitrification in a carrousel system. Prog. Water Technol, 8 (4/5), 653. van der Graaf A. A., de Bruijn P., Robertson L. A., Kuenen J. G. and Mulder A. (1991) Biological oxidation of ammonium under anoxic conditions: Anammox process. Proc. of the Int. Symp. on Environ. Biotechnol, II, (Eds. Verachtert H. and Verstraete W.), The royal Flemish Society of Engineers, 661-610. van der Graaf A. A., Mulder A., Slijkhuis H., Robertson L. A. and Kuenen J. G. (1990) Anoxic ammonium oxidation. In: Proc. of the 5th Euro. Cong, on Biotechnoly 1, (Eds. Christiansen C , Munck L. and Villadsen J), Munksgaard International Publisher, Copenhagen, 388-391. van Huyssteen J. A., Barnard, J. L. and Hendriksz J. (1990) The Olifantsfontein nutrient removal plant. Water Sci. Technol, 22 (7/8), 1-8. van Niel E. W. J. (1991) Nitrification by Heterotrophic Denitrifiers and Its Relationship to Autotrophic Nitrification. Ph.D. Thesis, Delft University of Technology, Delft. Verstraete W. (1975) Heterotrophic nitrification in soil and aqueous media - a review. Bull. Acad. Sci. USSR Biol. Ser., 4, 515-530. Vlekke G. J. F. M . , Comeau Y. and Oldham W. K. (1988) Biological phosphate removal from wastewater with oxygen or nitrate in sequencing batch reactors. Environ. Technol Lett., 9, 791-796. von Schulthess, Wild D. and Gujer W. (1992) Nitric and nitrous oxide from denitrification activated sludge at low oxygen concentration. In: Water quality international '94, Proc. IAWQ 17th Biennial Conference, July, 1994, Budapest, Hungary, 93-102. Wadsworth H. M . and Stephens K. S. Editors. (1986). Modern Methods for Quality Control and Improvement. John Wiley and Sons, New York. Waki T., Kawato Y., Shimatani Y. and Ichikawa Y. (1980b) Dynamic characteristics of Paracoccus denitrificans in alternate aerobic-anaerobic continuous cultivation. J. Ferment. Technol, 58 (3), 251-257. 242 Waki T., Murayama K., Kawato Y. and Ichikawa Y. (1980a) Transient characteristics of Paracoccus denitrificans with changes between aerobic and anaerobic conditions. J. Ferment. Technol, 58 (3), 243-249. Wareham D. G. (1992) The Use of Oxidation-Reduction Potential (ORP) as a Process Control Parameter in Wastewater Treatment Systems. Ph.D. Thesis, Univ. of British Columbia. Vancouver, BC. Watanabe S., Baba K. and Nogita S. (1985) Basic studies of an ORP/External carbon source control system for the biological denitrification process. In: Instrumentation and Control of Water and Wastewater Treatment and Transport Systems, proceedings 4th IAWPRC Workshop, 27 April-4 May, Houston and Denver, USA., 641-644. Watanabe Y., Masuda S. and Ishiguro M . (1992). Simultaneous nitrification and denitrification in micro-aerobic biofilms. Water Sci. Technol., 26 (3/4), 511-522. Wentzel M . C , Ekama G. A. and Marais G. v. R. (1991) Kinetics of nitrification denitrification biological excess phosphorus removal system - A review. Water Sci. Technol, 23 (5/6), 555-565. Wentzel M . C , Ekama G. A. and Marais G. v. R. (1992) Processes and modeling of nitrification denitrification biological excess phosphorus removal systems - a review. Water Sci. Technol, 25 (6), 59-82. Wentzell M . C , Dold P. L. , Ekama G. A. and Marais G. v. R. (1984) Kinetics of Biological Phosphorus Release, Proc. 12th Conf. of the IAWPRC, Post-Conference Seminar on Enhanced Biological Phosphorus Removal from Wastewater, Paris, Sept., 1984. Wild D., Schulthess R. v. and Gujer W. (1992) Synthesis of denitrification enzyme in activated sludge, modeling with structured biomass. Water Quality International ^94, Proc. IAWQ 17th Biennial Conference, July, 1994, Budapest, Hungary, 83-92. Wilkinson T. G. and Hamer G. (1972) Some growth characteristics of a Hypomicrobium sp. in batch culture. J. Appl. Bacteriol, 35, 577. Wilkinson T. G. and Harrison D. E. F. (1973) The affinity for methane and methanol of mixed cultures grown on methane in continuous culture. J. Appl. Bacteriol, 36, 309. Williams D. R. (1978) Denitrifying Pseudomonas aeruginosa, some parameters of growth and active transport. Appl. Envir. Microbiol, 36, 257-263. Wimpenny J. W. T. (1969) The effect of Eh on Regulatory Processes in Facultative anaerobes. Biotechnol. Bioeng., 11 (4), 623. Wood L. B., Hurley B. J. E. and Matthews P. J. (1981) Some observations on the biochemistry and inhibition of nitrification. Water Res., 15, 543-551. Wuhrmann K. (1964) Microbial aspects of water pollution control. Adv. Appl. Microbiol, 6,119-151. 243 Xue H. B., Stumm W. and Sigg L. (1988) The binding of heavy metals to algal surfaces. Water Res., 22, 917-926. Ye O., Ohtake H. and Toda K. (1988) Phosphorus removal by pure culture and mixed culture of microorganisms. J. Ferment. Technol., 66 (6), 207-212. Zheng H., Hanaki K. and Matsuo.T. (1994) Production of nitrous oxide gas during nitrification of wastewater. In: Water quality international '94, Proc. IAWQ 17th Biennial Conference, July, 1994, Budapest, Hungary, 103-111. 244 APPENDIX A Mass Balance Calculation for an Individual Reactor The calculating methods below are referred to one of the three processes used in this research, the three-stage Bardenpho process. A schematic of this process train is shown in Figure 3.1 in Chapter 3. Qi„f= influent flow rate (L/d) Qret = sludge return flow rate (L/d) Qi„t = internal recycle flow rate (L/d) Cinf= concentration of substrate C in the influent (mg/L) Cana = concentration of substrate C in the anaerobic zone (mg/L) Cano = concentration of substrate C in the anoxic zone (mg/L) Caer = concentration of substrate C in the aerobic zone (mg/L) Cda = concentration of substrate C in the clarifier overflow (i.e., effluent) (mg/L) Cret = concentration of substrate C in return sludge (mg/L) (for soluble forms = Ccu) Anaerobic Substrate Removal (mg per liter feed) _ ( C j n f Q n f + CretQret)- Cana(Q[n{ + Qret) y. 245 Anoxic Substrate Removal (mg per liter feed) = [Cgna (flnf + Qre, ) + QntQmt ] - CJjQ* + Qm + Qmt ) Gnf Aerobic Substrate Removal (mg per liter feed) _ (Qwo(flnf + &, + gmt)-Qer(qnf + + qnt) Clarifier Substrate Removal (mg per liter feed) 246 APPENDIX B Calculation Methods for the Amount of Nitrification and the Amount of Unaccounted for Nitrogen Loss The following assumptions are considered to be necessary for calculating the amounts of nitrification and the amount of unaccounted for nitrogen loss for this study. 1) Ammonia was the primary source for assimilation and the ammonia uptake for assimilation had priority over ammonia oxidation e.g., nitrification; 2) Assimilation of nitrate or dissimilation of nitrate to ammonia were not considered because ammonia was plenty in the sewage; 3) Ammonia and organic nitrogen were first oxidized to nitrite and nitrate, subsequently, removed from liquid through denitrification of NOx to gaseous forms of nitrogen; and 4) Ammonia was not considered as electron donor for denitrification, in other word, nitrification occurs only under aerobic conditions. 1. Nitrification NIT = TKN mf - TKNejj - TKRwas (A2.1) where NIT = the amount of nitrification in aeration tank (mg per liter feed) TKNinf= influent TKN concentration (mg/L) TKNeff= effluent TKN concentration (mg/L) TKRwas = the amount of TKN removed due to sludge wastage (mg per liter of feed) 247 2. Unaccounted for Nitrogen Loss in the Entire Process The amount of unaccounted nitrogen loss (UNL) is based on the T N balance on the entire process train (including sludge wastage) and the NOx balances on individual reactors, such as the anaerobic zone, the anoxic zone, and the clarifier for the three-stage Bardenpho process. TNRove = TKNinf-TKNeff-NOxeff (A2.2) UNLml = TNRtot - TNRwas - DeNana - DeNano - DeNcla (A2.3) where TNRove = amount of overall TN removal (mg per liter of feed) TKNinf = influent TKN concentration (mg/L) TKNejr= effluent TAW concentration (mg/L) NOxeff = effluent NOx concentration (mg/L) UNLtot = the amount of unaccounted nitrogen loss (mg per liter of feed) TNRwas = the amount of TN removal due to sludge wastage (mg per liter of feed) DeNa„a = the amount of NOx removal in the anaerobic zone (mg per liter of feed) DeNano = the amount of NOx removal in the anoxic zone (mg per liter of feed) DeNcia = the amount of NOx removal in the clarifier (mg per liter of feed) 248 APPENDIX C The Results of the f-Tests Table c l The results of the paired t-test on the effluent TN concentration obtained from the runs under the wide/narrow ORP control regime. Effluent TN concentration (mg/L) Calculated /^Significant? Carbon source 3-B* Wide* Narrow* . 3-B/Wide Wide/Narrow 3-B/Narrow Sewage 7.6 14.1 14.7 -9.0/yes -1.0/no -13.4/yes Acetate 7.7 5.6 13.1 6.0/yes -11.8/yes -7.4/yes Methanol 7.7 5.6 10.0 -2.0/yes -4.7/yes -2.1/yes •Representing the 3-Bardenpho process, the wide ORP process and the narrow ORP process, respectively. 'Critical t at a=0.05 are 1.8-1.9 depending on the number of data pairs. Table C.2 The results of the independent t-test on the overall TN removal efficiency among different acetate and methanol dosages. Acetate Addition Runs Dosages (mgCOD/L) Between 30 and 50 Dosages (mgCOD/L) Between 50 and 100 Dosages (mgCOD/L) Between 30 and 100 Process Difference in TN Removal (%) /-value/ Significant? Difference in TN Removal (%) r-value/ Significant? Difference in TN Removal (%) /-value/ Significant? 3-Bardenpho 8.1 4.3/Yes 1.5 0.2/No 9.6 6.9/Yes Low ORP 11.1 2.1/Yes -4.3 -1.1/No 7.9 1.9/Yes High ORP 6.7 1.6/No 3.4 1.1/No 10.1 3.1/Yes Methanol Addition Runs Dosages (mgCOD/L) Between 15 and 30 Dosages (mgCOD/L) Between 30 and 60 Dosages (mgCOD/L) Between 15 and 60 3-Bardenpho 0.6 0.2/No 7.6 5.2/Yes 8.2 3.5/Yes Low ORP 13.8 3.4/Yes -0.5 -0.1/No 13.3 3.0/Yes High ORP 6.7 2.1/Yes -11.6 -3.8/Yes -4.9 -1.5/No •Critical t values at a=0.05 are in a range of 1.7 to 1.8 depending on the number of data (n) 249 Table C.3 The results of the paired r-test on the percentage of NO* removal among the three aeration tanks. Mean percentage of NOx removal (%) Significant? Dosage 3-B* Low High* 3-B/Low Low/High 3-B/High 0 15.2 5.8 yes 0 12.7 11.2 no k yes Acetate 30 24.0 . 35.7 25.2 yes" yes* no 50 20.1 44.4 28.7 yes yes yes 100 14.5 29.2 36.0 yes yes yes Methanol 15 15.8 32.3 25.4 yes yes yes 30 13.6 48.9 27.9 yes yes yes 60 9.8 45.9 17.5 yes yes yes •Representing the 3-Bardenpho process, the low ORP process and the high ORP process, respectively. #Indicateing that the results are obtained from the independent t-test. Table C.4 The results of the independent t-tests on the percentage of NO* removal in an aeration tank among different dosages. Acetate Dosages between Dosages between Dosages between Addition Runs 0 and 30 (mgCOD/L) 30 and 50 (mgCOD/L) 50 and 100 (mgCOD/L) Difference in f-value/ Difference in f-value/ Difference in f-value/ Process NOx Removal Significant? NOx Removal Significant? NOx Removal Significant? (%) (%) (%) 3-Bardenpho 5.8 1.76/no 3.9 1.2/no 2.9 0.96/no Lower ORP 16.3 2.9/yes 11.5 2.1/yes -15.5 3.0/yes Higher ORP 19.4 6.9/yes 0 0.003/No 7.3 2.2/yes Methanol Dosages between Dosages between Dosages between Addition Runs 0 and 15 (mgCOD/L) 15 and 30 (mgCOD/L) 30 and 60 (mgCOD/L) 3-Bardenpho 19 3.7/yes 20.1 5.9/yes 4.8 1.7/no Lower ORP 19 3.7/Yes 13.7 3.7/yes -0.7 0.2/no Higher ORP 21.7 4.6/Yes 4.1 0.7/no -15.3 3.5/yes •Critical t values at ct=0,05 are in a range of 1.7 to 1.8 depending on the number of data points (n). 250 APPENDIX D RAW DATA FROM THE VARIOUS EXPERIMENTAL RUNS 251 Raw Data from Run 1 Run 1 started on May 31, 1995 Raw Data of Ammonia (mg/L) from Run 1 Date Influent Anaerobic Anoxic Aerobic Effluent A* B* c * A B A B c A B C Jun. 15 6.6 5.0 5.3 5.2 1.4 1.5 0.4 1.0 0.4 0.2 1.5 0.3 Jun. 19 7.1 5.1 5.7 3.7 1.2 1.5 0.5 0.4 3.6 0.4 0.3 2.2 Jun. 28 4.1 2.6 2.2 2.6 0.9 0.6 0.1 0.1 0.1 0.1 0.1 0.1 Jul. 3 4.3 4.3 4.3 3.9 1.3 1.1 0.3 0.3 0.3 0.2 0.2 0.2 Jul. 5 7.5 7.1 6.6 7.0 2.5 1.8 0.6 0.5 0.6 0.5 0.2 0.4 Jul. 7 9.9 5.8 5.3 4.5 2.2 1.5 0.5 0.5 0.7 0.5 0.4 0.4 Jul. 10 7.1 6.6 4.8 5.0 2.2 1.2 0.6 0.5 0.7 0.4 0.5 0.4 Jul. 11 6.5 5.4 5.6 5.5 2.7 3.1 0.6 2.4 1.0 1.3 2.2 0.8 Average 6.6 5.2 5.0 4.7 1.8 1.5 0.4 0.7 0.9 0.4 0.7 0.6 STDEV 1.9 1.4 1.3 1.3 0.7 0.7 0.2 0.7 1.1 0.4 0.8 0.7 Median 6.9 5.2 5.3 4.7 1.8 1.5 0.5 0.5 0.6 0.4 0.3 0.4 A, B, C represent the three experimental systems. Raw Data of NOx (mg/L) from Run 1 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jun. 15 0.2 0.3 0.2 0.5 6.6 4.2 10.3 7.5 13.4 11.0 4.8 13.1 Jun. 19 0.2 0.4 0.3 0.4 7.2 6.7 11.1 10.4 1.2 10.7 9.2 4.0 Jun. 28 0.1 1.2 0.7 0.9 6.0 5.6 8.7 7.3 10.8 8.3 6.7 10.3 Jul. 3 0.1 1.7 0.7 1.1 7.1 5.8 10.2 8.2 11.3 9.6 7.2 11.0 Jul. 5 0.1 1.7 0.8 1.4 7.5 6.4 10.3 8.1 10.8 10.3 7.7 9.8 Jul. 7 0.1 1.3 0.5 1.1 6.7 5.7 8.9 7.2 11.2 9.2 7.0 10.5 Jul. 10 0.0 2.6 1.7 3.1 8.8 7.9 11.5 9.0 13.0 11.2 8.5 12.7 Jul. 11 0.0 1.3 0.2 0.8 7.1 1.9 10.3 2.9 10.9 10.1 3.0 10.8 Average 0.1 1.3 0.6 1.2 7.1 5.5 10.1 7.6 10.3 10.0 6.7 10.3 STDEV 0.1 0.7 0.5 0.8 0.8 1.8 1.0 2.2 3.8 1.0 2.0 2.8 Median 0.1 1.3 0.6 1.0 7.1 5.8 10.3 7.8 11.1 10.2 7.1 10.7 Raw Data of Nitrite (mg/L) from Run 1 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jun. 15 0.12 0.09 0.05 0.08 0.16 0.07 0.17 0.11 0.05 0.12 0.16 0.06 Jun. 19 0.06 0.08 0.10 0.08 0.68 0.17 0.31 0.10 0.09 0.26 0.12 0.12 Jun. 28 0.10 0.09 0.31 0.09 0.06 0.25 0.15 0.52 0.15 0.08 0.31 0.03 Jul. 3 0.03 0.51 0.08 0.15 0.46 0.19 0.32 0.09 0.10 0.23 0.07 0.09 Jul. 5 0.02 0.18 0.07 0.17 0.60 0.20 0.65 0.17 0.23 0.23 0.15 0.60 Jul. 7 0.03 0.31 0.07 0.13 0.65 0.15 0.69 0.08 0.12 0.56 0.07 0.12 Jul. 10 0.03 0.30 0.09 0.24 0.59 0.14 0.58 0.10 0.06 0.49 0.10 0.13 Jul. 11 0.03 0.32 0.05 0.12 0.73 0.09 0.69 0.08 0.26 0.55 0.13 0.25 Average 0.05 0.23 0.10 0.13 0.49 0.16 0.44 0.16 0.13 0.31 0.14 0.17 STDEV 0.04 0.15 0.08 0.05 0.25 0.06 0.23 0.15 0.08 0.19 0.08 0.18 Median 0.03 0.24 0.08 0.12 0.60 0.16 0.45 0.10 0.11 0.25 0.12 0.12 Raw Data of Orth-P (mg/L) from Run 1 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jun. 15 3.3 8.9 8.6 5.6 3.7 3.5 2.9 2.1 2.7 3.0 2.2 3.1 Jun. 19 3.8 9.2 8.4 6.8 1.7 2.7 1.0 2.9 3.5 1.4 1.9 3.3 Jun. 28 2.5 1.8 2.8 2.7 2.3 3.1 2.5 3.1 2.8 2.2 2.9 2.6 Jul. 3 2.5 2.1 2.0 3.4 1.6 1.7 1.7 1.6 2.3 1.4 1.5 2.1 Jul. 5 2.3 2.5 2.7 2.4 2.2 2.2 2.3 2.2 2.1 1.9 2.1 2.1 Jul. 7 2.0 2.1 2.0 2.3 2.1 2.1 2.1 3.1 2.3 2.2 2.2 2.1 Jul. 10 2.7 2.4 2.3 2.7 2.5 2.6 2.5 2.6 2.8 2.3 2.5 2.7 Jul. 11 2.5 2.4 2.8 3.1 1.8 2.3 1.8 2.1 2.9 1.8 2.4 2.8 Average 2.7 3.9 3.9 3.6 2.2 2.5 2.1 2.5 2.7 2.0 2.2 2.6 STDEV 0.6 3.2 2.8 1.7 0.7 0.6 0.6 0.5 0.5 0.5 0.4 0.5 Median 2.5 2.4 2.7 2.9 2.1 2.5 2.2 2.4 2.8 2.1 2.2 2.7 252 Raw Data from Run 1 Run started on May 31, 1995 Raw Data of T K N (mg/L) from Run 1 Date Influent Aerobic Effluent A * B" C * A B C Jun. 15 27.0 128 150 160 1.9 9.2 1.9 Jun. 19 41.4 150 213 248 1.7 6.3 7.0 Jun. 28 21.3 170 227 174 2.9 0.9 2.1 Jul. 3 23.2 158 158 189 1.5 1.8 9.4 Jul. 5 23.0 159 185 142 1.9 1.6 4.2 Jul. 7 24.2 192 198 196 4.8 3.9 3.1 Jul. 10 27.6 146 197 184 5.3 3.5 9.4 Jul. 11 26.1 155 165 164 4.0 9.8 4.1 Average 26.7 157 187 182 3.0 4.6 5.2 S T D E V 6.3 19 27 32 1.5 3.4 3.1 Median 25.1 156 191 179 2.4 3.7 4.1 A, B, C represent the three experimental systems. Raw Data of TP (mg/L) from Run 1 Date Influent Aerobic Effluent A B C A B C Jun. 15 4.6 65 64 57 2.7 1.9 2.5 Jun. 19 9.3 88 99 92 1.4 1.4 2.6 Jun. 28 20.2 80 107 70 1.9 3.1 3.0 Jul. 3 5.6 85 74 74 1.9 1.6 3.6 Jul. 5 1.9 76 88 56 2.0 2.2 1.5 Jul. 7 4.3 83 75 60 2.9 2.8 2.9 Jul. 10 3.6 63 73 54 2.1 2.8 2.5 Jul. 11 4.1 60 63 49 1.7 3.5 3.1 Average 6.7 75 80 64 2.1 2.4 2.7 S T D E V 5.8 11 16 14 0.5 0.8 0.6 Median 4.4 78 74 58 1.9 2.5 2.7 Raw Data of COD (mg/L) from Run 1 Date Raw Influent Anaerobic Anoxic Aerobic Effluent T C O D T C O D A B C A B A B C A B C Jun. 15 na 334 126 274 145 32 43 35 24 19 16 19 19 Jun. 19 360 368 189 191 208 76 70 61 61 61 48 49 29 Jun. 28 397 200 86 91 120 54 48 53 48 51 40 28 25 Jul. 3 260 210 86 84 148 58 na 35 33 36 33 21 30 Jul. 5 na 271 116 126 126 89 68 81 81 75 68 61 63 Jul. 7 280 230 97 112 na 68 68 63 51 53 41 33 33 Jul. 10 na 183 100 120 108 75 50 63 47 73 38 33 50 Jul. 11 na 213 106 118 136 74 65 62 56 65 38 33 33 Average 324 251 113 139 142 66 59 57 50 54 40 35 35 S T D E V 65 67 34 63 32 17 11 15 17 19 15 14 14 Median 320 221 103 119 136 71 65 61 49 57 39 33 31 Raw Data of Suspended Solids (mg/L) from Run 1 Date M L S S M L V S S % Effluent SS A B C A B C A B C Juu. 16 2556 2452 2836 73.0% 75.6% 76.8% 4.8 6.6 12.7 Jun. 17 2896 2912 3228 74.0% 74.7% 76.0% 5.0 4.5 10.0 Jun. 19 3266 3570 3616 71.3% 74.5% 76.0% 10.4 6.2 8.9 Jun. 23 3388 3414 3584 na na na 7.8 13.0 8.9 Jun. 26 2920 3324 3202 na na na 10.1 10.2 13.1 Jun. 27 2788 3140 2740 na na na 15.8 16.7 14.2 Jun. 28 3116 3036 2796 72.0% 74.0% 75.0% 15.0 11.0 11.0 Jul. 1 2624 2844 2614 na na na 11.3 10.0 11.0 Jul. 2 2806 2654 2486 na na na 10.6 6.7 13.8 Jul. 3 2478 2736 2478 70.1% 72.8% 73.7% 10.0 10.0 18.0 Jul. 5 2734 2636 2590 70.0% 72.5% 72.6% 7.8 7.8 10.2 Jul. 7 2686 2740 2648 69.5% 71.7% 72.1% 11.0 5.2 8.7 Jul. 10 2298 2558 2496 70.5% 71.9% 72.0% 13.5 4.7 5.6 Jul. 11 2660 2668 2500 68.6% 72.0% 72.1% 10.8 4.3 5.1 Average 2801 2906 2844 71.0% 73.3% 74.0% 10.3 8.4 10.8 S T D E V 300 343 401 1.7% 1.4% 1.9% 3.2 3.6 3.4 Median 2761 2792 2694 70.5% 72.8% 73.7% 10.5 7.3 10.6 253 Raw Data from Run 2 Run 2 started on July 12, 1995 Raw Data of Ammonia (mg/L) from Run 2 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jul. 20 9.5 7.7 9.9 10.4 3.8 9.8 1.2 9.2 9.5 0.7 9.6 9.6 Jul. 22 6.7 6.6 7.3 7.9 3.2 5.4 0.5 4.8 7.0 0.4 4.1 6.3 Jul. 25 7.0 5.8 6.8 7.3 2.5 6.6 0.6 6.9 6.5 0.3 6.1 6.9 Jul. 29 7.5 5.2 6.8 6.7 1.7 6.6 0.6 6.6 5.6 0.3 6.0 5.9 Jul. 31 6.1 4.4 5.3 6.4 1.4 5.9 0.7 4.8 6.6 0.3 5.2 6.6 Aug. 3 5.2 3.7 5.0 3.9 2.8 5.0 0.6 4.5 4.4 0.3 4.5 4.5 Aug. 5 10.4 9.3 12.2 10.9 3.8 11.6 0.8 10.2 10.1 0.5 10.7 8.5 Average 7.5 6.1 7.6 7.7 2.7 7.3 0.7 6.7 7.1 0.4 6.6 6.9 STDEV 1.8 2.0 2.6 2.4 1.0 2.5 0.2 2.3 2.0 0.1 2.5 •1.7 Median 7.0 5.8 6.8 7.3 2.8 6.6 0.6 6.6 6.6 0.3 6.0 6.6 Raw Data of NOx (mg/L) from Run 2 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jul. 20 0.1 0.9 .0.2 0.2 5.4 0.1 9.4 0.1 0.2 8.6 1.6 2.1 Jul. 22 0.1 0.8 0.2 0.3 5.1 0.3 9.4 1.1 3.2 8.9 2.3 5.0 Jul. 25 0.1 1.2 0.2 0.2 5.4 0.2 9.1 0.2 0.2 8.1 1.7 1.5 Jul. 29 0.1 1.5 0.5 0.2 6.7 0.1 10.6 0.1 0.7 9.9 1.4 2.0 Jul. 31 0.1 0.9 0.2 0.2 6.4 0.1 10.0 0.2 0.1 9.5 1.5 1.9 Aug. 3 0.3 1.3 0.6 0.5 7.1 0.1 10.4 0.2 1.6 9.9 1.8 3.5 Aug. 5 0.2 1.2 0.2 0.2 6.0 0.2 9.8 0.1 2.7 9.2 1.3 4.4 Average 0.2 1.1 0.3 0.3 6.0 0.2 9.8 0.3 1.2 9.2 1.7 2.9 STDEV 0.1 0.3 0.2 0.1 0.7 0.1 0.6 0.3 1.3 0.7 0.3 1.4 Median 0.1 1.2 0.2 0.2 6.0 0.1 9.8 0.2 0.7 9.2 1.6 2.1 Raw Data of Nitrite (mg/L) from Run 2 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jul. 20 0.03 0.15 0.04 0.04 0.42 0.02 0.51 0.02 0.02 0.87 0.52 0.46 Jul. 22 0.03 0.14 0.05 0.05 0.32 0.05 0.46 0.05 0.05 0.37 0.52 0.90 Jul. 25 0.03 0.21 0.04 0.06 0.43 0.02 0.46 0.03 0.03 0.33 0.48 0.43 Jul. 29 0.05 0.20 0.07 0.05 0.41 0.03 0.37 0.03 0.06 0.27 0.53 0.61 Jul. 31 0.03 0.21 0.07 0.07 0.50 0.02 0.46 0.04 0.03 0.39 0.68 0.57 Aug. 3 0.04 0.22 0.10 0.06 0.44 0.03 0.39 0.03 0.08 0.32 0.82 0.60 Aug. 5 0.04 0.20 0.04 0.04 0.44 0.02 0.49 0.01 0.06 0.36 0.40 0.59 Average 0.03 0.19 0.06 0.05 0.42 0.03 0.45 0.03 0.05 0.41 0.56 0.59 STDEV 0.01 0.03 0.02 0.01 0.06 0.01 0.05 0.01 0.02 0.20 0.14 0.15 Median 0.03 0.20 0.05 0.05 0.43 0.02 0.46 0.03 0.05 0.36 0.52 0.59 Raw Data of Orth-P (mg/L) from Run 2 Date Influent Anaerobic Anoxic Aerobic Effluent A B C A B A B C A B C Jul. 20 2.5 2.5 3.5 4.0 3.3 1.8 3.4 1.0 3.1 3.0 1.0 3.1 Jul. 22 2.8 2.3 4.0 3.3 1.9 0.6 1.7 0.3 2.7 1.5 0.2 2.6 Jul. 25 2.3 1.8 2.0 3.3 1.8 0.8 1.5 0.7 2.8 1.4 0.7 2.8 Jul. 29 2.9 2.3 4.8 3.8 2.6 3.6 2.6 3.3 3.1 2.3 3.4 3.0 Jul. 31 2.7 2.8 4.0 4.2 2.9 1.7 3.0 0.8 3.5 2.6 0.7 3.5 Aug. 3 2.5 2.3 4.7 3.0 3.1 5.9 3.2 5.3 3.1 2.9 5.6 3.1 Aug. 5 3.1 2.4 4.9 4.0 2.4 2.2 2.3 1.0 2.8 2.2 3.0 3.1 Average 2.7 2.3 4.0 3.7 2.6 2.4 2.5 1.8 3.0 2.3 2.1 3.0 STDEV 0.3 0.3 1.0 0.4 0.6 1.8 0.7 1.8 0.3 0.6 2.0 0.3 Median 2.7 2.3 4.0 3.8 2.6 1.8 2.6 1.0 3.1 2.3 1.0 3.1 254 Raw Data from Run 2 Run 2 started on July 12, 1995 Raw Data of TKN (mg/L) from Run 2 Date Influent Aerobic Effluent A B C A B C Jul. 20 27.9 102 169 144 4.2 12.8 1L0 Jul. 22 30.3 134 199 191 4.2 9.6 13.9 Jul. 25 24.3 108 134 172 1.7 15.0 15.0 Jul. 29 21.1 88 147 143 1.3 12.3 12.3 Jul. 31 20.8 104 128 151 1.0 11.8 14.3 Aug. 3 23.1 86 115 173 1.6 11.5 10.9 Aug. 5 22.3 118 205 182 2.3 11.6 8.8 Average 24.2 106 157 165 2.3 12.1 12.3 STDEV 3.6 16 35 19 1.4 1.6 2.2 Median 23.1 104 147 172 1.7 11.8 12.3 Raw Data of TP (mg/L) from Run 2 Date Influent Aerobic Effluent A B C A B C Jul. 20 3.5 44 66 55 4.0 2.1 3.5 Jul. 22 5.3 57 69 55 2.5 0.9 3.0 Jul. 25 4.5 62 59 58 2.5 4.2 2.0 Jul. 29 3.6 54 61 47 2.5 3.7 3.6 Jul. 31 4.7 60 64 52 3.1 1.6 2.5 Aug. 3 5.1 54 59 60 3.5 5.6 2.1 Aug. 5 4.9 67 88 59 2.6 2.7 3.1 Average 4.5 57 67 55 3.0 3.0 2.8 STDEV 0.7 7 10 5 0.6 1.6 0.6 Median 4.7 57 64 55 2.6 2.7 3.0 Raw Data of COD (mg/L) from Run 2 Date Raw Influent Anaerobic Anoxic Aerobic Effluent TCOD TCOD A B C A B A B C A B C Jul. 20 275 174 102 98 88 68 56 47 53 34 22 34 34 Jul. 22 302 265 115 95 107 98 52 53 45 48 34 33 26 Jul. 25 na 191 86 69 100 53 41 46 56 33 33 16 16 Jul. 29 276 170 111 111 95 59 65 48 48 43 49 35 37 Jul. 31 na 212 132 112 114 74 68 56 53 55 44 44 44 Aug. 3 na 188 120 134 129 86 74 58 55 51 34 36 31 Aug. 5 294 237 134 129 129 80 77 68 68 49 49 51 54 Average 287 205 114 107 109 74 62 54 54 45 38 35 34 STDEV 14 35 17 22 16 16 13 8 7 9 10 11 12 Median 285 191 115 111 107 74 65 53 53 48 34 35 34 Raw Data of Suspended Solids (mg/L) from Run 2 Date MLSS MLVSS% Effluent SS A B C A B C A B C Jul. 18 1830 2448 2452 na na na 13.7 7.1 4.7 Jul. 20 1686 2434 2376 72.0% 74.0% 74.0% 12.4 7.9 7.9 Jul. 22 1738 2408 2274 71.0% 74.0% 74.0% 11.3 5.7 3.7 Jul. 25 2032 1866 2360 na na na 9.7 88.4 2.4 Jul. 29 1554 2128 2298 71.0% 75.0% 75.0% 8.5 6.4 17.2 Jul. 31 1972 2434 2410 75.0% 76.0% 76.9% 5.8 2.9 5.4 Aug. 3 1562 2310 2444 73.1% 76.7% 76.8% 4.5 1.7 7.5 Aug. 5 2114 2402 2516 74.0% 77.0% 77.0% 5.4 3.3 4.6 Average 1811 2304 2391 72.7% 75.5% 75.6% 8.9 15.4 6.7 STDEV 212 206 81 1.6% 1.3% 1.5% 3.4 29.6 4.6 Median 1784 2405 2393 72.6% 75.5% 75.9% 9.1 6.1 5.1 255 Raw Data from Run 3 Run 3 started on Aug. 6, 1995 Raw Data of Ammonia (mg/L) from Run 3 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Aug. 17 na 10.3 7.0 7.5 7.2 1.4 0.5 0.3 0.3 1.9 2.2 0.3 Aug. 20 na 9.7 7.7 7.4 6.3 3.6 0.7 0.5 0.5 0.5 0.2 0.3 Aug.24 na 8.5 6.9 6.0 6.0 2.7 0.5 0.4 0.3 0.2 0.3 0.3 Aug. 26 na 13.5 10.5 9.9 9.7 1.3 0.3 0.3 0.2 0.2 0.2 0.3 Aug. 28 na 18.4 12.4 12.9 13.0 2.3 0.6 0.4 0.4 0.4 0.0 0.0 Aug. 30 na 15.4 10.0 11.0 11.8 1.9 0.1 0.0 0.0 0.0 0.0 0.0 Sep. 6 13.2 10.6 6.2 6.4 8.0 1.4 0.0 0.0 0.5 0.0 0.0 0.4 Sep. 9 10.3 8.2 4.8 5.0 5.5 1.5 0.1 0.0 0.0 0.0 0.0 0.0 Average 11.8 11.8 8.2 8.3 8.4 2.0 0.3 0.2 0.3 0.4 0.4 0.2 STDEV 2.1 3.6 2.5 2.7 2.8 0.8 0.3 0.2 0.2 0.6 0.7 0.2 Median 11.8 10.4 7.4 7.5 7.6 1.7 0.4 0.3 0.3 0.2 0.1 0.3 Raw Data of NOx (mg/L) from Run 3 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Aug. 17 na 0.2 0.2 0.7 0.1 3.2 7.0 13.0 12.2 5.6 12.7 11.5 Aug. 20 na 0.2 0.5 1.9 1.6 4.9 9.0 15.2 14.9 8.4 14.6 14.1 Aug.24 na 0.1 0.4 1.1 1.6 3.0 6.8 12.5 13.5 5.7 12.2 12.9 Aug. 26 na 0.2 0.5 1.2 2.1 2.7 6.5 12.6 14.0 5.0 11.9 13.6 Aug. 28 na 0.0 0.5 0.4 0.6 1.4 5.6 12.6 12.5 5.8 12.0 11.9 Aug. 30 na 0.5 0.6 0.6 1.0 2.4 6.5 12.4 12.2 6.0 12.1 13.2 Sep. 6 0.280 0.3 0.7 0.7 1.3 3.2 7.1 10.5 11.5 6.3 10.1 11.5 Sep. 9 0.190 0.3 0.3 0.3 0.8 2.2 5.7 9.5 11.2 5.5 9.4 11.2 Average 0.24 0.21 0.5 0.9 1.1 2.9 6.8 12.3 12.8 6.0 11.9 12.5 STDEV 0.06 0.16 0.2 0.5 0.6 1.0 1.1 1.7 1.3 1.0 1.6 1.1 Median 0.24 0.17 0.5 0.7 1.1 2.9 6.6 12.5 12.4 5.7 12.0 12.4 Raw Data of Nitrite (mg/L) from Run 3 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Aug. 17 na 0.03 0.07 0.12 0.22 0.34 0.42 0.05 0.09 0.23 0.02 0.10 Aug. 20 na 0.05 0.16 0.18 0.27 0.42 0.51 0.08 0.12 0.35 0.02 0.12 Aug.24 na 0.02 0.01 0.09 0.17 0.33 0.33 0.02 0.06 0.25 0.03 0.09 Aug. 26 na 0.04 0.04 0.10 0.25 0.35 0.35 0.03 0.09 0.27 0.02 0.08 Aug. 28 na 0.02 0.03 0.03 0.08 0.15 0.32 0.08 0.09 0.20 0.03 0.10 Aug. 30 na 0.02 0.02 0.04 0.16 0.24 0.30 0.03 0.06 0.14 0.04 0.10 Sep. 6 0.05 0.04 0.12 0.07 0.25 0.33 0.20 0.03 0.07 0.16 0.05 0.19 Sep. 9 0.04 0.03 0.04 0.05 0.16 0.15 0.15 0.03 0.05 0.07 0.04 0.08 Average 0.04 0.03 0.06 0.08 0.19 0.29 0.32 0.04 0.08 0.21 0.03 0.11 STDEV 0.01 0.01 0.05 0.05 0.06 0.10 0.11 0.02 0.02 0.09 0.01 0.03 Median 0.04 0.03 0.04 0.08 0.20 0.33 0.32 0.03 0.08 0.21 0.03 0.10 Raw Data of Orth-P (mg/L) from Run 3 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B c Aug. 17 na 2.7 3.1 1.8 2.5 1.5 1.0 0.6 2.0 1.6 0.7 2.0 Aug. 20 na 3.0 2.8 2.6 3.0 2.2 2.0 2.1 2.8 1.9 2.0 2.7 Aug.24 na 2.9 3.5 2.2 3.1 2.0 2.1 1.3 2.6 1.6 1.1 2.5 Aug. 26 na 2.9 3.1 2.2 3.2 1.3 0.9 1.0 2.9 1.3 0.7 2.7 Aug. 28 na 3.6 13.0 12.5 5.7 5.3 4.1 3.6 3.2 1.7 2.0 2.5 Aug. 30 na 3.4 11.6 10.1 4.9 4.6 2.9 3.8 3.3 2.1 3.1 2.7 Sep. 6 3.6 2.3 1.9 2.5 2.6 1.5 1.9 1.8 2.5 1.0 1.5 2.4 Sep. 9 3.4 3.1 10.1 9.5 5.2 3.9 2.9 3.1 3.1 2.5 3.5 3.0 Average 3.5 3.0 6.1 5.4 3.8 2.8 2.2 2.2 2.8 1.7 1.8 2.6 STDEV 0.1 0.4 4.6 4.5 1.3 1.6 1.0 1.2 0.4 0.5 1.0 0.3 Median 3.5 3.0 3.3 2.5 3.1 2.1 2.0 1.9 2.8 1.6 1.7 2.6 256 Raw Data from Run 3 Run 3 started on Aug. 6, 1995 Raw Data of TKN (mg/L) from Run 3 Date Influent Aerobic Effluent A B C A B C Aug. 17 26.4 167 215 193 1.2 1.5 3.6 Aug. 20 34.3 147 205 190 1.1 1.5 0.7 Aug. 24 31.3 139 212 198 1.2 4.3 1.0 Aug. 26 34.4 194 276 210 1.0 0.9 1.0 Aug. 28 30.4 194 244 222 1.2 2.9 4.0 Aug. 30 29.5 190 246 220 1.9 2.1 1.9 Sep. 6 29.9 198 302 218 2.6 2.4 2.7 Sep. 9 25.6 208 243 192 2.5 2.3 2.6 Average 30.2 180 243 205 1.6 2.2 2.2 STDEV 3.2 25 34 14 0.6 1.0 1.2 Median 30.1 192 243 204 1.2 2.2 2.3 Raw Data of TP (mg/L) from Run 3 Date Influent Aerobic Effluent A B C A B C Aug. 17 6.3 79 95 56 1.6 1.2 2.4 Aug. 20 4.9 71 91 55 1.8 2.2 2.7 Aug. 24 4.6 66 86 58 1.6 2.1 2.6 Aug. 26 4.6 86 111 62 1.4 1.2 3.0 Aug. 28 4.1 90 92 68 1.7 2.5 2.8 Aug. 30 5.7 94 103 68 2.0 3.1 2.9 Sep. 6 5.0 100 119 65 0.8 1.7 2.4 Sep. 9 4.5 97 95 57 2.3 3.5 3.7 Average 5.0 85 99 61 1.6 2.2 2.8 STDEV 0.7 12 11 5 0.5 0.8 0.4 Median 4.8 87.8 95.3 60.2 1.7 2.2 2.8 Raw Data of COD (mg/L) from Run 3 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Aug. 17 315 na 237 143 144 110 122 81 73 65 60 49 37 49 Aug. 20 372 na 436 159 143 97 135 80 64 48 48 45 31 24 Aug. 24 362 na 314 147 147 105 157 67 64 59 50 37 30 30 Aug. 26 426 206 376 216 207 184 178 156 170 162 170 162 124 170 Aug. 30 na na 345 154 100 108 117 80 77 70 77 69 60 64 Sep. 6 347 190 965 113 99 109 na 90 88 80 81 69 66 61 Sep. 9 295 232 292 187 111 111 na 90 92 83 83 65 77 74 Average 353 209 424 160 136 118 142 92 90 81 81 71 61 67 STDEV 46 21 247 33 38 30 25 29 37 38 42 42 33 49 Median 355 206 345 154 143 109 135 81 77 70 77 65 60 61 Raw Data of Suspended Solids (m g/L) from Run 3 Date TSS vss% MLSS MLVSS% Effluent SS Raw Influent Raw Influent A B C A B C A B C Aug. 10 na 87 na 89.0% 2022 2314 2514 73.8% 75.0% 77.5% 5.3 3.9 5.5 Aug. 16 86 48 84.7% 83.3% 1986 2584 2346 76.0% 77.2% 78.9% 3.4 59.8 4.1 Aug. 17 80 74 75.0% 86.5% 2360 2814 2040 78.6% 76.8% 75.9% 4.3 3.2 2.2 Aug. 20 144 255 79.6% 73.7% 2060 2790 2332 75.5% 76.4% 78.8% 5.7 7.6 7.6 Aug. 24 124 153 80.6% 85.6% 2642 2902 2410 74.5% 76.7% 79.2% 5.2 39.6 7.4 Aug. 26 128 162 89.7% 84.4% 2484 3724 2556 77.1% 78.0% 80.9% 4.4 8.8 5.0 Aug. 28 123 138 76.4% 83.3% 2880 3104 2740 76.8% 78.0% 80.1% 3.8 6.4 10.6 Aug. 30 143 151 78.5% 86.8% 2712 3176 2908 77.3% 78.2% 80.0% 4.8 7.0 6.4 Sept. 6 123 543 80.6% 85.6% 3440 3332 2824 78.7% 78.7% 81.3% 3.0 7.0 5.6 Sep. 9 100 121 na na 3224 3444 2784 77.7% 79.5% 81.7% 0.6 6.0 4.6 Average 117 173 80.6% 84.2% 2581 3018 2545 76.6% 77.5% 79.4% 4.1 14.9 5.9 STDEV 23 142 4.7% 4.3% 501 422 272 1.6% 1.3% 1.8% 1.5 19.0 2.3 Median 123 144 80.1% 85.6% 2563 3003 2535 77.0% 77.6% 79.6% 4.4 7.0 5.6 257 Raw Data from Run 4 Run 4 started on Sept. 10, 1995 Raw Data of Ammonia (mg/L) from Run 4 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Sep. 17 na 8.6 4.7 5.9 6.6 1.0 0.0 0.0 0.0 0.0 0.0 0.2 Sep. 19 13.5 10.6 6.0 6.2 6.9 1.8 0.0 0.0 1.4 0.0 0.0 1.4 Sep. 22 18.6 12.3 10.4 10.3 9.6 1.9 0.0 0.0 0.1 0.0 0.0 0.1 Sep. 25 14.2 12.3 7.1 7.3 8.8 2.1 0.0 0.0 5.1 0.0 0.0 3.8 Sep. 29 11.4 13.0 7.4 7.5 9.4 2.0 0.0 0.0 5.3 0.1 0.0 5.5 Oct. 2 13.2 10.6 6.2 6.4 8.0 1.6 0.0 0.0 4.3 0.0 0.0 4.2 Oct. 4 16.4 15.6 11.3 9.2 9.6 1.8 0.0 0.0 0.2 0.0 0.0 0.3 Oct. 7 10.5 8.0 5.7 6.2 6.0 1.6 0.0 0.0 5.8 0.0 0.0 4.5 Average 14.0 11.4 7.3 7.4 8.1 1.7 0.0 0.0 2.8 0.0 0.0 2.5 STDEV 2.8 2.5 2.3 1.6 1.4 0.4 0.0 0.0 2.6 0.0 0.0 2.2 Median 13.S 11.4 6.6 6.8 8.4 1.8 0.0 0.0 2.8 0.0 0.0 2.6 Raw Data of NOx (mg/L) from Run 4 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Sep. 17 na 0.22 1.05 0.54 0.55 0.35 2.99 8.74 4.27 2.74 8.95 4.11 Sep. 19 0.19 0.22 0.21 0.69 0.60 0.53 3.34 8.36 3.19 2.67 8.08 2.51 Sep. 22 0.25 0.30 0.32 0.54 0.31 0.72 3.11 8.69 5.10 2.73 8.13 4.76 Sep. 25 0.68 0.25 0.52 0.33 0.34 0.93 4.27 10.54 3.30 4.23 10.25 0.95 Sep. 27 0.28 0.25 0.26 0.54 0.54 1.35 4.44 9.97 0.16 4.09 9.93 0.91 Sep. 29 0.67 0.63 0.72 0.75 0.74 1.62 4.66 10.26 0.54 4.13 9.93 1.20 Oct. 2 0.63 0.70 0.73 0.75 0.70 1.51 3.84 8.52 0.72 3.75 8.58 1.40 Oct. 4 0.68 0.58 0.70 1.19 0.86 1.00 3.36 8.28 7.39 3.00 8.11 6.99 Oct. 7 0.34 0.22 0.30 0.31 0.39 0.28 2.78 6.99 0.52 2.62 5.81 1.38 Average 0.47 0.37 0.53 0.63 0.56 0.92 3.64 8.93 2.80 3.33 8.64 2.69 STDEV 0.22 0.20 0.28 0.27 0.19 0.49 0.68 1.13 2.51 0.70 1.36 2.14 Median 0.5 0.3 0.5 0.5 0.6 0.9 3.4 8.7 3.2 3.0 8.6 1.4 Raw Data of Nitrite (mg/L) from Run 4 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Sep. 17 na 0.03 0.03 0.03 0.14 0.02 0.05 0.03 0.07 0.04 0.04 0.17 Sep. 19 0.04 0.02 0.03 0.04 0.06 0.03 0.05 0.02 0.08 0.04 0.04 0.18 Sep. 22 0.03 0.03 0.04 0.04 0.05 0.03 0.04 0.02 0.05 0.04 0.04 0.09 Sep. 25 0.04 0.04 0.06 0.06 0.07 0.06 0.06 0.03 0.03 0.05 0.05 0.20 Sep. 27 0.18 0.03 0.03 0.04 0.06 0.03 0.05 0.02 0.03 0.04 0.03 0.21 Sep. 29 0.05 0.04 0.03 0.04 0.05 0.05 0.05 0.03 0.03 0.04 0.03 0.25 Oct. 2 0.04 0.03 0.03 0.05 0.06 0.04 0.05 0.03 0.04 0.04 0.03 0.29 Oct. 4 0.03 0.03 0.03 0.20 0.06 0.05 0.08 0.03 0.13 0.03 0.03 0.36 Oct. 7 0.06 0.03 0.02 0.03 0.04 0.02 0.03 0.05 0.03 0.05 0.03 0.37 Average 0.06 0.03 0.03 0.06 0.06 0.04 0.05 0.03 0.05 0.04 0.03 0.23 STDEV 0.05 0.01 0.01 0.05 0.03 0.01 0.01 0.01 0.03 0.01 0.01 0.09 Median 0.04 0.03 0.03 0.04 0.06 0.03 0.05 0.03 0.04 0.04 0.03 0.21 Raw Data of Orth-P (mg/L) from Run 4 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B C A B C Sep. 17 na 2.0 18.1 10.6 3.8 3.9 0.4 0.7 1.9 0.2 0.7 2.0 Sep. 19 2.66 1.2 18.4 11.5 4.1 2.8 0.2 0.6 1.7 0.1 0.6 1.7 Sep. 22 2.58 2.0 17.9 12.0 4.5 1.8 0.1 0.6 1.5 0.0 0.4 1.4 Sep. 25 3.50 2.8 17.1 5.3 5.9 0.9 0.1 0.4 3.0 0.0 0.4 2.7 Sep. 27 3.62 2.6 15.3 8.4 4.0 0.9 0.0 0.6 4.0 0.0 0.5 3.7 Sep. 29 2.76 2.7 14.5 9.0 5.1 0.9 0.1 0.3 0.4 0.0 0.3 0.3 Oct. 2 2.90 2.2 14.4 8.9 4.1 1.1 0.1 0.3 0.1 0.1 0.3 0.3 Oct. 4 2.94 2.2 15.5 1.5 4.4 1.3 0.1 0.2 0.1 0.1 0.1 0.1 Oct. 7 2.36 1.9 11.6 9.8 6.4 1.8 0.1 0.5 0.4 0.1 0.5 1.3 Average 2.9 2.2 15.9 8.6 4.7 1.7 0.1 0.5 1.5 0.1 0.4 1.5 STDEV 0.4 0.5 2.2 3.3 0.9 1.0 0.1 0.2 1.4 0.1 0.2 1.2 Median 2.8 2.2 15.5 9.0 4.4 1.3 0.1 0.5 1.5 0.1 0.4 1.4 258 Raw Data from Run 4 Run 4 started on Sep. 10, 1995 Raw Data of TKN (mg/L) from Run 4 Date Influent Anaerobic Aerobic Effluent A B C A B C A B c Sep. 19 20.1 311 261 315 212 242 226 1.3 1.1 2.6 Sep. 22 20.8 184 229 222 191 239 206 3.9 0.7 1.2 Sep. 25 24.0 257 218 303 315 289 224 2.7 0.7 7.0 Sep. 27 26.7 270 293 219 274 290 290 1.0 0.9 15.0 Sep. 29 26.7 244 347 252 266 276 241 0.7 3.1 7.1 Oct. 2 27.4 263 312 233 283 280 213 0.7 1.8 6.8 Oct. 4 23.8 236 300 253 175 250 218 0.5 3.0 3.3 Oct. 7 22.7 228 288 206 170 247 201 1.6 1.6 6.1 Average 24.0 249 281 250 236 264 227 1.6 1.6 6.1 STDEV 2.8 37 43 40 56 22 28 1.2 1.0 4.2 Median 23.9 251 291 243 239 263 221 1.1 1.3 6.4 Raw Data of TP (mg/L) from Run 4 Date Influent Anaerobic Aerobic Effluent . #1 #2 #3 #1 #2 #3 #1 #2 #3 Sep. 19 3.2 127 92 75 106 101 62 0.0 0.8 1.5 Sep. 22 4.8 92 88 59 102 105 58 0.1 0.8 0.6 Sep. 25 3.0 110 87 65 143 115 72 0.0 0.2 1.8 Sep. 27 3.9 126 115 49 141 109 55 0.0 0.0 2.1 Sep. 29 3.3 113 122 50 122 113 50 0.0 0.0 0.0 Oct. 2 2.6 121 112 49 134 118 43 0.0 0.0 0.0 Oct. 4 2.8 107 107 52 95 106 50 0.0 0.0 0.0 Oct. 7 2.7 114 110 45 98 110 47 0.0 0.0 0.0 Average 3.3 114 104 55 118 110 54 0.0 0.2 0.8 STDEV 0.7 11 13 10 20 6 9 0.0 0.4 0.9 Median 3.1 114 109 51 114 109 52 0.0 0.0 0.3 Raw Data of COD (mg/L) from Run 4 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble #1 #2 #3 #1 Ul #2 #3 ttl #2 #3 Sep. 17 na na 293 112 64 96 160 40 64 35 45 45 32 32 Sep. 22 310 162 283 102 111 125 143 96 88 88 80 82 75 74 Sep. 25 326 184 286 111 83 85 83 54 21 26 38 na 14 17 Sep. 27 344 190 313 92 94 103 136 76 77 69 107 61 57 66 Sep. 29 293 169 294 107 91 104 120 82 80 80 83 67 68 73 Oct. 2 263 145 301 87 86 116 140 68 62 67 67 58 51 56 Oct. 4 283 169 274 112 na na na na na na na 71 71 79 Oct. 7 289 161 275 94 108 112 140 102 96 98 104 91 83 83 Average 301 169 290 102 91 106 132 74 70 66 75 68 56 60 STDEV 27 15 13 10 16 13 24 22 25 27 27 15 23 24 Median 302 169 293 107 89 103 138 72 71 68 73 64 57 66 Raw Data of Suspended Solids (mg/L) from Run 4 Date SS VSS% MLSS MLVSS% Effluent SS Raw Influent Raw Influent #1 #2 #3 #1 #2 #3 #1 #2 #3 Sep. 17 153 134 80.0% 80.3% 2924 3424 2812 72.6% 75.5% 81.5% 4.8 3.8 13.9 Sep. 19 101 119 86.0% 84.7% 2996 3344 2924 75.8% 77.7% 82.4% 6.1 4.9 13.3 Sep. 22 91 130 77.0% 82.5% 2896 3124 3016 75.0% 76.0% 80.5% 5.6 3.6 8.4 Sep. 25 112 133 78.0% 86.0% 3868 3356 3236 75.9% 77.7% 82.3% 5.8 4.4 10.0 Sep. 27 131 167 86.8% 86.0% 3596 3368 3132 74.9% 77.4% 83.0% 4.9 3.9 8.0 Sep. 29 137 150 76.6% 81.3% 3868 3580 3152 76.0% 77.4% 81.7% 5.8 5.2 15.8 Oct. 2 116 148 94.0% 91.2% 3380 3432 2804 75.3% 77.3% 83.4% 4.9 6.4 12.3 Oct. 4 101 133 87.1% 89.5% 3720 3440 2840 74.9% 77.6% 82.4% 6.2 5.3 na Oct. 7 127 163 79.5% 82.8% 3424 3468 2660 75.5% 78.7% 82.7% 6.7 6.4 15.2 Average 119 142 82.8% 84.9% 3408 3393 2953 75.1% 77.3% 82.2% 5.6 4.9 12.1 STDEV 20 16 6.0% 3.7% 391 124 193 1.0% 1.0% 0.9% 0.7 1.1 3.0 Median 116 134 80.0% 84.7% 3424 3424 2924 75.3% 77.4% 82.4% 5.8 4.9 12.8 259 Raw Data from Run 5 Run 5 started on Oct. 8,1995 Raw Data of Ammonia (mg/L) from Run 5 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B c A B C Oct. 20 20.0 19.6 12.2 .11.2 12.2 3.3 0.1 0.0 1.9 0.1 0.0 2.0 Oct. 22 23.7 17.9 10.7 10.9 11.7 2.9 0.0 0.3 4.0 0.1 0.0 4.0 Oct. 24 17.8 19.8 11.9 13.5 10.7 2.6 0.0 0.0 3.9 0.1 0.0 4.2 Oct. 26 19.5 20.0 10.8 10.8 13.4 2.7 0.0 0.0 3.2 0.1 0.0 3.1 Oct. 30 25.4 15.4 13.2 15.0 na 2.3 0.0 0.0 na 0.1 0.0 na Nov. 1 14.2 12.4 7.2 7.7 7.9 2.6 0.0 0.0 1.3 0.0 0.0 1.4 Nov. 3 13.1 12.8 7.3 8.8 7.7 2.8 0.0 0.0 1.5 0.0 0.0 1.3 Nov. 5 17.7 13.3 7.8 10.1 8.3 2.5 0.0 0.0 1.1 0.0 0.0 0.9 Nov. 7 26.6 21.2 9.4 12.1 10.8 3.6 0.9 0.8 3.1 0.6 0.6 2.1 Average 19.8 16.9 10.0 11.1 10.3 2.8 0.1 0.1 2.5 0.1 0.1 2.4 S T D E V 4.7 3.5 2.2 2.2 2.1 0.4 0.3 0.3 1.2 0.2 0.2 1.3 Median 19.5 17.9 10.7 10.9 10.7 2.7 0.0 0.0 2.5 0.1 0.0 2.0 Raw Data of NOx (mg/L) from Run 5 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Oct, 20 0.3 0.2 0.3 0.4 0.3 0.3 4.2 8.9 2.5 3.3 9.0 2.9 Oct. 22 0.5 0.2 0.3 0.9 1.0 0.3 4.1 9.0 0.8 3.6 9.0 1.0 Oct. 24 0.5 0.3 0.3 0.6 0.3 0.6 3.8 7.9 0.8 3.0 7.9 1.0 Oct. 26 0.2 0.2 0.2 0.6 0.3 0.3 4.0 8.9 0.9 3.1 8.6 1.1 Oct. 30 0.3 0.2 0.3 0.6 na 0.3 3.9 9.0 na 3.5 9.0 na Nov. 1 0.3 0.2 0.3 0.6 0.6 0.3 4.1 9.5 2.9 3.7 9.3 3.5 Nov. 3 0.2 0.3 0.3 0.5 0.5 0.3 4.0 9.1 2.2 3.6 8.9 2.3 Nov. 5 0.2 0.2 0.3 0.6 0.3 0.5 3.8 8.5 2.0 3.6 8.6 2.4 Nov. 7 0.2 0.2 0.3 0.3 0.3 0.2 3.7 8.3 2.3 3.3 8.1 2.1 Average 0.3 0.2 0.3 0.6 0.4 0.3 4.0 8.8 1.8 3.4 8.7 2.0 S T D E V 0.1 0.0 0.0 0.2 0.2 0.1 0.2 0.5 0.8 0.2 0.4 1.0 Median 0.3 0.2 0.3 0.6 0.3 0.3 4.0 8.9 2.1 3.5 8.9 2.2 Raw Data of Nitrite (mg/L) from Run 5 Date Raw Influent Anaerobic Anoxic Aerobic Effluent #1 #2 #3 #1 #1 #2 #3 #1 #2 #3 Oct. 20 0.06 0.04 0.05 0.07 0.05 0.05 0.10 0.18 0.06 0.12 0.11 0.16 Oct. 22 0.06 0.04 0.04 0.11 0.05 0.05 0.08 0.05 0.05 0.09 0.09 0.13 Oct. 24 0.06 0.04 0.04 0.08 0.04 0.08 0.12 0.08 0.05 0.12 0.10 0.15 Oct. 26 0.11 0.04 0.04 0.09 0.05 0.05 0.10 0.11 0.06 0.12 0.11 0.18 Oct. 30 0.05 0.04 0.05 0.12 na 0.05 0.07 0.10 na 0.06 0.07 na Nov. 1 0.05 0.04 0.04 0.09 0.06 0.03 0.06 0.07 0.08 0.05 0.08 0.12 Nov. 3 0.05 0.04 0.04 0.06 0.05 0.04 0.06 0.05 0.07 0.05 0.07 0.16 Nov. 5 0.05 0.04 0.04 0.06 0.05 0.04 0.06 0.06 0.07 0.05 0.07 0.14 Nov. 7 0.05 0.04 0.05 0.08 0.07 0.04 0.07 0.07 0.08 0.05 0.07 0.14 Average 0.06 0.04 0.04 0.08 0.05 0.05 0.08 0.08 0.07 0.08 0.08 0.15 S T D E V 0.02 0.00 0.01 0.02 0.01 0.01 0.02 0.04 0.01 0.03 0.02 0.02 Median 0.05 0.04 0.04 0.08 0.05 0.05 0.07 0.07 0.07 0.06 0.08 0.14 Raw Data of Orth-P (mg/L) from Run 5 Date Raw Influent Anaerobic Anoxic Aerobic Effluent #1 #2 #3 #1 #1 #2 #3 #1 #2 #3 . Oct. 20 2.6 2.4 6.5 2.8 3.7 1.8 0.9 1.6 0.6 1.1 1.3 0.5 Oct. 22 2.8 2.4 2.2 2.5 3.6 0.9 0.7 1.4 0.5 0.6 1.2 0.6 Oct. 24 3.0 2.4 3.1 2.4 4.2 1.7 1.5 1.3 0.6 1.8 1.2 0.6 Oct. 26 2.8 2.5 3.0 2.6 4.3 1.3 0.9 1.2 0.5 0.8 1.2 0.6 Oct. 30 3.1 2.1 2.2 2.2 na 0.7 0.5 1.2 na 0.3 1.0 na Nov. 1 2.7 2.4 2.3 2.2 2.4 0.7 0.4 1.3 0.2 0.5 1.0 0.0 Nov. 3 2.7 2.3 2.4 10.7 3.4 0.8 0.4 0.4 0.1 0.3 0.6 0.0 Nov. 5 2.9 2.1 2.3 7.6 2.8 1.3 0.9 0.6 0.2 0.9 0.8 0.1 Nov. 7 3.0 2.1 2.1 4.0 3.2 1.1 0.7 0.3 0.2 0.5 0.3 0.1 Average 2.8 2.3 2.9 4.1 3.4 1.1 0.8 1.0 0.4 0.8 0.9 0.3 S T D E V 0.2 0.2 1.4 3.0 0.6 0.4 0.3 0.5 0.2 0.5 0.3 0.3 Median 2.8 2.4 2.3 2.6 3.5 1.1 0.7 1.2 0.4 0.6 1.0 0.3 260 Raw Data from Run 5 Run 5 started on Oct. 8, 1995 Raw Data of TKN (mg/L) from Run 5 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Oct. 20 25.7 358 269 235 211 219 270 1.5 3.2 3.2 Oct. 22 24.3 212 248 257 215 239 242 0.6 1.1 8.3 Oct. 30 24.7 310 284 na 272 219 245 0.1 3.0 3.7 Nov. 1 26.4 273 257 262 226 221 246 0.3 4.3 2.1 Nov. 3 26.3 271 549 296 283 223 231 1.9 1.7 2.3 Nov. 5 25.1 287 582 296 238 209 245 1.4 1.1 1.4 Nov. 7 26.3 268 537 264 225 222 234 1.2 1.4 5.1 Average 2S.S 283 389 268 239 222 245 1.0 2.3 3.7 STDEV 0.8 44 157 24 28 9 13 0.7 1.2 2.4 Median 25.7 273 284 263 226 221 245 1.2 1.7 3.2 Raw Data of TP (mg/L) from Run 5 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Oct. 20 4.1 101 87 79 90 82 97 1.2 1.3 0.7 Oct. 22 4.0 104 92 89 111 84 90 0.5 1.5 0.5 Oct. 30 3.9 131 93 77 108 80 74 0.2 1.0 2.7 Nov. 1 3.4 117 82 69 109 82 71 0.6 1.4 0.4 Nov. 3 3.9 120 178 87 119 74 72 0.5 1.1 0.4 Nov. 5 3.9 114 183 75 108 73 83 0.9 1.0 0.4 Nov. 7 3.4 117 177 88 108 74 85 0.8 0.5 0.4 Average 3.8 115 127 81 108 78 82 0.7 1.1 0.8 STDEV 0.3 10 49 7 9 5 10 0.3 0.3 0.8 Median 3.9 117 93 79 108 80 83 0.6 1.1 0.4 Raw Data of COD (mg/L) from Run 5 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Oct. 22 312 218 313 105 92 108 103 80 83 78 76 73 71 66 Oct. 24 258 156 348 95 93 126 110 68 78 58 0 0 20 36 Oct. 26 301 160 279 85 63 93 83 59 20 17 19 15 8 10 Oct. 30 320 184 345 96 66 77 125 56 32 48 96 27 19 61 Nov. 1 267 137 315 95 73 91 114 52 42 40 28 40 34 35 Nov. 3 320 169 272 114 111 167 107 86 68 71 81 56 58 52 Nov. 5 343 191 289 102 100 164 110 83 77 72 79 59 59 66 Nov. 7 299 168 305 92 102 153 118 81 84 76 79 65 64 62 Average 303 173 308 98 88 122 109 71 61 58 57 42 42 49 STDEV 28 25 28 9 18 35 12 14 25 21 36 26 24 20 Median 307 169 309 96 93 117 110 74 73 65 78 48 46 57 Raw Data of Suspended Solids (mg/L) from Run 5 Date TSS VSS% MLSS MLVSS% Effluent SS Raw Influent Raw Influent A B C A B C A B C Oct. 20 107 147 91.6% 83.7% 2896 3084 3256 77.2% 79.6% 80.1% 11.3 5.5 4.9 Oct. 22 107 132 82.2% 82.6% 3220 3084 3276 74.3% 77.4% 78.3% 4.4 7.6 5.0 Oct. 24 92 165 83.7% 86.8% na 3076 3312 75.6% 79.5% 80.3% 3.3 4.6 4.2 Oct. 26 100 142 89.0% 87.0% 2928 3024 3392 76.5% 79.9% 79.8% 5.4 6.4 5.2 Oct. 30 134 138 81.3% 84.8% 3484 2984 3220 75.0% 79.0% 82.6% 4.9 5.3 9.5 Nov. 1 108 111 84.3% 93.7% 3584 3088 3192 75.9% 80.8% 81.8% 2.4 5.8 5.0 Nov. 3 127 120 82.7% 85.0% 3500 2820 3132 75.7% 80.0% 80.8% 5.0 5.4 5.4 Nov. 5 117 142 82.1% 82.4% 3448 2876 3180 76.0% 79.7% 80.3% 5.3 5.4 3.9 Nov. 7 116 139 77.6% 79.9% 3392 2908 3192 75.4% 78.4% 78.7% 13.6 6.0 3.9 Average 112 137 83.8% 85.1% 3307 2994 3239 75.7% 79.4% 80.3% 6.2 5.8 5.2 STDEV 13 16 4.2% 3.9% 265 103 79 0.8% 1.0% 1.4% 3.7 0.8 1.7 Median 108 139 82.7% 84.8% 3420 3024 3220 75.7% 79.6% 80.3% 5.0 5.5 5.0 261 Raw Data from Run 6 Run 6 started on Nov. 7, 1995 Raw Data of Ammonia (mg/L) from Run 6 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Nov. 16 19.6 17.3 10.0 10.9 10.6 3.3 0.8 0.4 2.0 0.6 0.7 1.4 Nov. 18 18.4 17.4 10.8 12.1 11.5 3.4 1.1 1.0 2.6 1.1 0.7 2.0 Nov. 20 21.6 15.5 11.5 11.6 12.0 3.4 0.8 1.1 4.0 0.7 0.7 3.9 Nov. 22 18.6 14.9 10.6 10.9 11.5 3.6 1.1 0.6 4.6 0.6 1.6 4.6 Nov. 24 18.1 19.2 12.0 11.3 12.6 2.7 0.0 0.0 4.1 0.2 0.0 3.6 Nov. 26 14.1 13.0 8.7 10.0 8.7 1.8 0.0 0.0 1.5 0.0 0.0 1.5 Nov. 27 17.8 14.3 10.8 11.2 11.6 1.9 0.0 0.1 1.1 0.0 0.0 0.8 Average 18.3 15.9 10.6 11.2 11.2 2.9 0.5 0.5 2.9 0.5 0.5 2.5 STDEV 2.3 2.1 1.1 0.7 1.3 0.7 0.5 0.5 1.4 0.4 0.6 1.5 Median 18.4 15.5 10.8 11.2 11.5 3.3 0.8 0.4 2.6 0.6 0.7 2.0 Raw Data of NOx (mg/L) from Run 6 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Nov. 16 0.4 0.4 0.5 0.6 0.8 0.4 3.8 8.2 4.5 3.4 8.3 5.7 Nov. 18 0.2 0.1 0.3 0.3 0.4 0.3 3.6 7.8 2.3 3.4 7.9 2.8 Nov. 20 0.2 0.2 0.3 0.3 0.3 0.3 3.6 6.9 0.9 3.5 6.9 1.2 Nov. 22 0.2 0.2 0.2 0.5 0.6 0.5 3.5 3.7 0.6 3.3 1.8 0.8 Nov. 24 0.5 0.3 0.5 0.4 0.9 0.4 3.6 5.1 0.6 4.2 4.6 0.9 Nov. 26 0.6 0.3 0.4 0.7 0.5 0.7 3.2 6.0 2.0 3.0 5.8 1.8 Nov. 27 0.3 0.4 0.4 1.0 0.4 1.4 3.4 6.2 1.7 3.2 6.3 1.8 Average 0.3 0.3 0.4 0.6 0.5 0.6 . 3.5 6.3 1.8 3.4 5.9 2.1 STDEV 0.1 0.1 0.1 0.2 0.2 0.4 0.2 1.6 1.4 0.4 2.2 1.7 Median 0.3 0.3 0.4 0.5 0.5 0.4 3.6 6.2 1.7 3.4 6.3 1.8 Raw Data of Nitrite (mg/L) from Run 6 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Nov. 16 0.12 0.03 0.03 0.03 0.06 0.03 0.06 0.08 0.10 0.05 0.05 0.24 Nov. 18 0.04 0.03 0.03 0.05 0.04 0.03 0.06 0.03 0.08 0.05 0.05 0.13 Nov. 20 0.06 0.03 0.03 0.04 0.03 0.04 0.05 0.05 0.05 0.04 0.04 0.10 Nov. 22 0.03 0.02 0.03 0.04 0.05 0.06 0.06 0.05 0.05 0.05 0.10 0.09 Nov. 24 0.05 0.03 0.03 0.03 0.05 0.04 0.07 0.06 0.05 0.11 0.07 0.09 Nov. 26 0.04 0.03 0.04 0.09 0.04 0.11 0.06 0.04 0.08 0.05 0.03 0.09 Nov. 27 0.04 0.03 0.05 0.16 0.04 0.27 0.05 0.05 0.07 0.05 0.03 0.09 Average 0.05 0.03 0.03 0.06 0.04 0.08 0.06 0.05 0.07 0.06 0.05 0.12 STDEV 0.03 0.00 0.01 0.05 0.01 0.09 0.01 0.02 0.02 0.02 0.03 0.06 Median 0.04 0.03 0.03 0.04 0.04 0.04 0.06 0.05 0.07 0.05 0.05 0.09 Raw Data of Orth-P (mg/L) from Run 6 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Nov. 16 2.3 2.4 2.4 4.2 2.5 0.9 0.5 0.6 0.2 0.4 0.5 0.0 Nov. 18 2.3 2.3 2.3 4.0 2.4 0.5 0.5 0.4 0.1 0.3 0.6 0.1 Nov. 20 2.2 2.4 3.4 3.1 3.6 1.1 0.6 1.4 0.1 1.1 2.1 0.1 Nov. 22 2.8 2.1 2.1 3.9 3.0 1.2 0.9 1.1 0.1 0.8 1.3 0.2 Nov. 24 27 2.6 10.4 9.1 10.5 2.5 0.6 2.4 1.3 2.5 1.9 1.1 Nov. 26 2.1 17 1.9 2.0 2.1 1.7 1.8 1.6 0.7 1.6 1.6 0.9 Nov. 27 2.1 1.6 1.6 1.9 2.0 1.5 1.7 1.8 1.3 1.6 1.7 1.1 Average 2.4 2.2 3.4 4.0 3.7 1.4 0.9 1.3 0.5 1.2 1.4 0.5 STDEV 0.3 0.4 3.1 2.4 3.0 0.6 0.6 0.7 0.6 0.8 0.6 0.5 Median 2.3 2.3 2.3 3.9 2.5 1.2 0.6 1.4 0.2 1.1 1.6 0.2 262 Raw Data from Run 6 Run 6 started on Nov. 7, 1995 Raw Data of TKN (mg/L) from Run 6 Date Influent Anaerobic Aerobic Effluent A B C A B C A B c Nov. 16 28.4 206 324 312 235 225 243 3.2 1.7 3.1 Nov. 18 25.3 201 369 260 173 206 243 2.2 1.1 2.0 Nov. 20 23.5 228 236 202 191 223 209 2.9 0.9 5.2 Nov. 22 22.2 230 196 202 166 206 229 2.7 3.5 5.8 Nov. 24 22.5 188 269 208 152 207 225 2.6 0.7 3.3 Nov. 26 19.7 169 194 247 167 202 188 1.2 1.3 2.8 Nov. 27 19.0 254 236 207 153 184 217 1.8 0.6 1.4 Average 22.9 211 261 234 177 208 222 2.4 1.4 3.4 STDEV 3.2 29 65 42 29 14 20 0.7 1.0 1.6 Median 22.5 206 236 208 167 206 225 2.6 1.1 3.1 Raw Data of TP (mg/L) from Run 6 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Nov. 16 4.4 94 107 86 100 84 74 0.0 0.6 0.1 Nov. 18 4.3 95 149 85 89 83 85 0.0 0.6 0.2 Nov. 20 4.4 109 82 72 98 84 79 1.0 2.2 0.0 Nov. 22 3.9 106 87 74 85 82 88 0.3 1.3 0.1 Nov. 24 4.4 92 92 77 74 75 92 2.5 2.1 1.3 Nov. 26 3.5 84 69 94 86 80 84 1.9 1.7 0.9 Nov. 27 3.4 128 86 80 81 74 101 1.1 1.9 1.2 Average 4.0 101 96 81 88 80 86 1.0 1.5 0.5 STDEV 0.4 15 26 8 9 4 9 0.9 0.7 0.5 Median 4.3 95 87 80 86 82 85 1.0 1.7 0.2 Raw Data of COD (mg/L) from Run 6 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Nov. 16 333 190 380 116 72 111 139 42 22 30 28 14 11 0 Nov. 18 411 232 284 154 116 142 118 60 50 62 59 40 39 28 Nov. 20 343 177 298 123 110 143 122 62 70 71 55 75 72 48 Nov. 22 323 166 277 92 111 123 128 90 86 92 89 101 86 80 Nov. 24 304 168 314 168 105 114 106 80 72 65 68 72 61 64 Nov. 26 278 172 280 98 93 86 84 74 64 57 67 58 44 49 Nov. 27 252 145 212 81 36 43 30 33 61 15 25 23 15 15 Average 321 179 292 119 92 109 104 63 61 56 56 55 47 41 STDEV 51 27 50 32 29 35 37 20 20 26 23 31 28 28 Median 323 172 284 116 105 114 118 62 64 62 59 58 44 48 Raw Data of Suspended Solids (m g/L) from Run 6 Date TSS vss% MLSS MLVSS% Effluent $ S Raw Influent Raw Influent A B C A B C A B c Nov. 16 120 154 84.2% 84.4% 3096 3052 3128 74.7% 78.4% 81.2% 7.8 7.5 7.3 Nov. 18 148 141 76.4% 79.4% 3256 2976 3104 74.7% 75.0% 77.2% 10.6 7.6 3.0 Nov. 20 128 133 80.5% 80.5% 2816 3040 3116 74.6% 78.4% 77.8% 7.9 6.7 2.9 Nov. 22 129 128 75.2% 75.8% 2668 2884 3228 73.8% 77.1% 76.7% 14.0 2.6 2.8 Nov. 24 113 117 77.0% 79.5% 2640 2856 3064 74.1% 77.3% 76.1% 13.7 5.2 3.0 Nov. 26 118 156 74.6% 75.0% 2612 2916 2980 72.7% 76.3% 75.0% 20.0 6.2 3.8 Nov. 27 119 131 80.7% 78.6% 2616 2884 2972 72.9% 77.0% 75.5% 11.0 5.4 1.4 Average 125 137 78.4% 79.0% 2815 2944 3085 73.9% 77.1% 77.1% 12.1 5.9 3.5 STDEV 12 14 3.5% 3.1% 260 79 89 0.8% 1.2% 2.1% 4.2 1.7 1.8 Median 120 133 77.0% 79.4% 2668 2916 3104 74.1% 77.1% 76.7% 11.0 6.2 3.0 263 Raw Data from Run 7 Run 7 started on Nov. 28,1995 Raw Data of Ammonia (mg/L) from Run 7 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B c A B C Deo. 6 13.1 12.2 7.2 7.1 9.1 2.4 0.0 0.0 7.7 0.0 0.1 6.6 Deo. 8 16.3 14.3 7.9 9.7 11.1 2.6 0.1 0.0 7.0 0.1 0.0 5.9 Dec. 10 18.1 18.5 12.1 12.9 13.4 5.1 2.8 3.3 9.0 1.9 2.8 7.8 Dec. 12 22.1 21.2 13.1 15.3 20.6 3.0 0.1 3.6 8.8 0.0 3.7 8.2 Deo. 14 25.6 26.7 16.2 17.2 23.9 2.8 0.1 2.0 10.1 0.0 1.9 9.8 Deo. 17 14.2 13.2 7.6 7.7 12.0 2.4 0.1 0.0 8.6 0.0 0.0 8.3 Dec. 19 13.7 12.6 8.7 8.2 11.5 3.7 1.9 0.6 8.4 1.4 0.3 7.6 Dec. 20 18.4 17.7 13.5 12.7 19.3 2.5 0.2 0.6 7.4 0.4 0.3 7.2 Dec. 21 22.4 19.1 14.3 17.2 19.4 2.4 0.1 3.7 7.0 0.0 2.8 6.9 Average 18.2 17.3 11.2 12.0 15.6 3.0 0.6 1.5 8.2 0.4 1.3 7.6 S T D E V 4.4 4.8 3.3 4.0 5.2 0.9 1.0 1.6 1.0 0.7 1.5 1.1 Median 18.1 17.7 12.1 12.7 13.4 2.6 0.1 0.6 8.4 0.0 0.3 7.6 Raw Data of NOx (mg/L) from Run 7 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Dec. 6 0.4 0.3 0.3 1.4 0.3 2.4 4.6 9.0 0.0 4.7 8.6 2.0 Deo. 8 0.3 0.2 0.4 0.5 0.3 1.6 4.2 8.0 0.3 4.5 8.6 1.8 Dec. 10 0.2 0.2 0.3 0.4 0.4 0.3 3.3 7.0 0.1 3.7 6.8 1.6 Dec. 12 0.3 0.3 0.4 0.4 0.4 2.6 6.7 7.0 0.2 6.3 6.4 1.4 Deo. 14 0.4 0.2 0.4 0.5 0.4 5.8 9.2 10.0 0.0 7.5 9.2 1.2 Dec. 17 0.2 0.2 0.4 0.4 0.3 2.7 5.6 8.8 0.3 5.7 8.8 1.6 Dec. 19 0.2 0.3 0.4 0.4 0.3 0.4 2.5 7.1 0.2 2.9 7.1 1.4 Dec. 20 0.3 0.2 0.5 0.4 0.5 2.0 5.3 6.9 0.2 4.7 6.6 1.2 Dec. 21 0.2 0.2 0.3 0.3 0.3 2.2 5.3 1.9 0.2 5.1 3.3 1.1 Average 0.3 0.2 0.4 0.5 0.3 2.2 5.2 7.3 0.2 5.0 7.2 1.5 S T D E V 0.1 0.0 0.1 0.3 0.1 1.6 2.0 2.3 0.1 1.4 1.8 0.3 Median 0.3 0.2 0.4 0.4 0.3 2.2 5.3 7.1 0.2 4.7 7.1 1.4 Raw Data of Nitrite (mg/L) from Run 7 Date Raw Influent Anaerobic Anoxic Aerobic Effluen A B C A A B C A B C Dec. 6 0.04 0.03 0.03 0.25 0.03 0.48 0.14 0.05 0.02 0.13 0.08 0.39 Dec. 8 0.05 0.04 0.04 0.05 0.05 0.39 0.17 0.04 0.03 0.16 0.05 0.37 Dec. 10 0.04 0.03 0.04 0.06 0.06 0.04 0.17 0.10 0.03 0.21 0.11 0.30 Dec. 12 0.05 0.03 0.04 0.05 0.05 0.47 0.23 0.12 0.04 0.17 0.14 0.15 Dec. 14 0.06 0.03 0.03 0.05 0.05 0.35 0.14 0.09 0.03 0.14 0.19 0.11 Dec. 17 0.04 0.04 0.04 0.04 0.05 0.55 0.18 0.05 0.04 0.13 0.05 0.16 Dec. 19 0.05 0.10 0.06 0.06 0.05 0.04 0.17 0.07 0.04 0.12 0.13 0.16 Dec. 20 0.05 0.05 0.05 0.06 0.06 0.20 0.23 0.06 0.05 0.18 0.12 0.15 Dec. 21 0.05 0.05 0.06 0.05 0.06 0.19 0.20 0.06 0.03 0.11 0.16 0.13 Average 0.05 0.04 0.04 0.07 0.05 0.30 0.18 0.07 0.03 0.15 0.11 0.21 S T D E V 0.01 0.02 0.01 0.07 0.01 0.19 0.03 0.03 0.01 0.03 0.05 0.11 Median 0.05 0.04 0.04 0.05 0.05 0.35 0.17 0.06 0.03 0.14 0.12 0.16 Raw Data of Orth-P (mg/L) from Run 7 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Dec. 6 2.6 2.2 6.6 2.3 6.2 1.2 0.3 1.3 0.4 0.4 1.4 0.1 Dec. 8 2.6 2.5 9.2 5.8 6.5 2.0 0.6 1.6 0.4 0.5 1.4 0.0 Deo. 10 2.9 2.2 13.2 7.7 6.4 5.3 3.3 2.8 0.3 1.5 2.1 0.1 Deo. 12 3.4 2.3 6.5 4.8 4.3 0.8 0.4 1.1 0.1 0.2 1.0 0.0 Dec. 14 2.7 2.0 10.5 8.8 7.4 0.6 0.5 0.5 1.2 0.4 0.5 0.1 Dec. 17 2.3 1.9 6.6 4.0 5.3 0.9 0.7 0.8 0.0 0.7 1.1 0.0 Dec. 19 2.5 2.2 15.1 7.8 11.5 5.3 2.6 1.1 0.2 1.1 0.7 0.0 Dec. 20 2.5 2.1 10.9 7.5 9.6 1.2 0.1 0.5 0.0 0.2 0.7 0.3 Dec. 21 2.2 2.3 7.0 2.4 7.7 0.7 0.0 0.7 0.1 0.1 0.6 0.2 Average 2.6 2.2 9.5 5.7 7.2 2.0 0.9 1.2 0.3 0.6 1.1 0.1 S T D E V 0.4 0.2 3.2 2.4 2.2 1.9 1.2 0.7 0.4 0.5 0.5 0.1 Median 2.6 2.2 9.2 5.8 6.5 1.2 0.5 1.1 0.2 0.4 1.0 0.1 264 Raw Data from Run 7 Run 7 started on Nov. 28, 1995 Raw Data of TKN (mg/L) from Run 7 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Dec. 6 24.7 161 301 233 137 170 202 2.2 1.5 9.4 Dec. 8 24.3 171 279 252 140 170 194 1.2 1.3 8.6 Dec. 10 29.8 222 191 279 132 171 199 3.5 11.9 13.0 Dec. 12 28.2 176 197 227 169 182 218 2.0 7.1 17.5 Dec. 14 29.4 168 272 275 141 178 245 2.3 3.8 12.3 Dec. 17 24.2 163 165 243 163 180 213 2.0 1.3 14.0 Dec. 19 24.8 171 171 215 142 193 213 4.3 1.5 13.9 Dec. 20 26.4 206 175 234 167 206 217 1.5 1.8 11.8 Dec. 21 23.4 192 175 220 165 178 210 2.4 5.5 13.3 Average 26.1 181 214 242 151 181 212 2.4 3.9 12.6 STDEV 2.4 21 54 23 15 12 15 1.0 3.7 2.6 Median 24.8 171 191 234 142 178 213 2.2 1.8 13.0 Raw Data of TP (mg/L) from Run 7 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Dec. 6 3.6 71.3 78.7 68.0 54 62 64 0.7 1.5 0.6 Dec. 8 3.2 67.8 84.7 65.0 64 58 51 0.7 1.5 0.5 Dec. 10 3.4 85.0 60.9 65.7 59 55 47 1.6 1.6 0.5 Dec. 12 3.5 74.8 65.7 48.3 71 64 55 0.7 0.6 0.0 Dec. 14 3.2 61.5 83.8 68.6 70 60 62 0.4 3.3 0.9 Dec. 17 3.5 74.1 61.1 78.6 80 70 74 1.8 1.5 0.5 Dec. 19 3.6 79.9 63.6 94.6 74 71 74 3.2 1.8 0.5 Dec. 20 3.8 93.6 63.2 83.6 80 72 78 0.6 1.1 1.2 Dec. 21 3.6 99.5 65.0 79.5 81 72 75 0.6 1.2 0.5 Average 3.5 79 70 72 70 65 64 1.1 1.5 0.6 STDEV 0.2 12 10 13 10 7 11 0.9 0.7 0.3 Median 3.5 75 65 69 71 64 64 0.7 1.5 0.5 Raw Data of COD (mg/L) from Run 7 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Dec. 6 318 197 300 120 86 103 86 34 26 33 41 33 17 34 Dec. 8 358 157 265 124 53 71 na 35 25 5 21 0 0 14 Dec. 10 344 224 339 101 100 145 214 69 46 51 93 51 41 57 Dec. 14 299 182 355 202 134 130 113 68 58 44 45 50 22 32 Dec. 17 365 333 346 280 158 105 126 88 116 79 131 102 53 56 Dec. 19 342 194 368 166 126 153 126 81 79 65 77 48 68 48 Dec. 20 364 204 328 154 112 121 149 102 83 71 53 58 71 47 Dec. 21 na 215 235 257 106 121 124 63 55 na 71 58 36 36 Average 341 213 317 176 109 119 134 68 61 50 67 50 39 41 STDEV 25 53 47 66 32 26 40 24 31 25 35 28 25 14 Median 344 201 334 160 109 121 126 69 57 51 62 51 39 42 Raw Data of Suspended Solids (mg/L) fromRun 7 Date TSS VSS% MLSS MLVSS% Effluent SS Raw Influent Raw Influent A B C A B C A B C Dec. 6 116 129 82.8% 84.5% 2236 2552 2824 72.3% 75.1% 76.2% 4.2 5.9 3.3 Dec. 8 145 112 77.2% 77.7% 2216 2416 2644 70.9% 74.3% 76.4% 8.6 6.8 4.7 Dec. 11 na 124 na 84.7% 2532 2788 2948 74.2% 76.5% 79.2% 7.6 5.8 12.6 Dec. 13 113 110 85.0% 82.7% 2370 2615 na 73.0% 75.9% 79.4% 5.4 8.6 8.4 Dec. 17 105 141 74.3% 79.4% 2284 2844 3004 73.7% 76.1% 77.5% 14.0 7.0 7.8 Dec. 19 118 116 74.6% 79.3% 2452 2784 3040 72.3% 74.0% 74.7% 11.6 7.7 7.7 Dec. 20 137 117 80.3% 79.5% 2528 2668 2968 73.1% 75.3% 74.7% 6.2 7.8 6.6 Dec. 21 na 117 na na 2756 2768 3024 na na na 8.1 7.4 6.2 Average 122 121 79.0% 81.1% 2422 2679 2922 72.8% 75.3% 76.9% 8.2 7.1 7.2 STDEV 15 10 4.4% 2.8% 183 145 142 1.1% 0.9% 1.9% 3.2 1.0 2.8 Median 117 117 78.8% 79.5% 2411 2718 2968 73.0% 75.3% 76.4% 7.9 7.2 7.2 265 Raw Data from Run 8 Run 8 started on Dee. 23, 1995 Raw Data of Ammonia (mg/L) from Run 8 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B c A B c Jan. 4 27.9 29.5 23.3 19.4 19.2 9.5 8.3 0.9 4.2 7.9 0.7 2.4 Jan. 7 28.9 25.7 21.1 19.7 18.3 8.5 6.7 0.0 0.2 6.0 0.2 0.0 Jan. 10 34.5 30.2 21.6 20.7 16.5 7.1 4.4 0.0 0.3 3.4 0.0 0.2 Jan. 13 17.9 20.4 11.0 12.0 10.7 2.1 0.3 0.2 0.2 0.0 0.0 0.0 Jan. 15 16.7 13.4 9.2 8.0 9.5 2.3 0.0 0.0 0.0 0.0 0.0 0.0 Jan. 17 16.0 15.9 10.8 10.1 9.3 2.4 0.0 0.3 1.1 0.0 0.2 0.3 Jan. 19 19.0 14.1 8.9 8.6 7.6 2.2 0.0 1.1 0.2 0.0 0.3 0.0 Jan. 24 20.0 16.7 11.2 11.2 11.8 3.8 1.2 0.0 0.1 1.1 0.0 0.1 Jan.26 22.5 18.8 12.4 12.0 12.2 4.1 1.6 0.0 0.1 1.4 0.1 0.0 Average 22.6 20.5 14.4 13.5 12.8 4.7 2.5 0.3 0.7 2.2 0.2 0.3 S T D E V 6.4 6.4 5.8 5.0 4.2 2.9 3.2 0.4 1.3 2.9 0.2 0.8 Median 20.0 18.8 11.2 12.0 11.8 3.8 1.2 0.0 0.2 1.1 0.1 0.0 Raw Data of NOx (mg/L) from Run 8 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Jan. 4 0.8 0.3 0.6 0.6 0.5 0.5 3.1 13.3 5.9 3.4 13.2 8.8 Jan. 7 0.0 0.4 0.5 0.6 0.6 0.5 2.8 13.1 11.0 3.3 12.9 11.0 Jan. 10 0.0 0.4 0.5 0.5 0.6 0.5 3.6 11.0 7.3 4.0 10.5 7.0 Jan. 13 0.2 0.3 0.5 0.6 0.6 0.7 5.3 8.4 9.3 4.9 8.8 9.5 Jan. 15 0.0 0.3 0.5 0.4 0.6 0.6 4.2 7.8 9.2 3.7 7.9 8.9 Jan. 17 0.0 0.2 0.6 0.6 0.6 0.7 4.2 7.4 5.0 3.8 6.9 5.5 Jan. 19 0.9 0.7 0.8 0.8 0.8 0.8 4.3 7.6 3.7 3.8 7.6 5.7 Jan. 24 0.5 0.7 0.8 0.8 0.8 0.9 4.6 8.4 8.9 3.9 8.2 8.5 Jan.26 0.0 0.6 0.8 0.7 0.7 0.8 4.1 8.1 8.0 3.8 8.1 8.2 Average 0.3 0.4 0.6 0.6 0.6 0.7 4.0 9.4 7.6 3.8 9.3 8.1 S T D E V 0.4 0.2 0.1 0.1 0.1 0.1 0.8 2.4 2.3 0.5 2.3 1.8 Median 0.0 0.4 0.6 0.6 0.6 0.7 4.2 8.4 8.0 3.8 8.2 8.5 Raw Data of Nitrite (mg/L) from Run 8 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Jan. 4 0.31 0.06 0.09 0.07 0.08 0.06 0.16 0.09 0.16 0.19 0.10 0.22 Jan. 7 0.06 0.08 0.07 0.07 0.08 0.06 0.15 0.09 0.13 0.23 0.05 0.08 Jan. 10 0.07 0.08 0.09 0.08 0.09 0.06 0.25 0.07 0.12 0.29 0.04 0.09 Jan. 13 0.12 0.07 0.07 0.08 0.07 0.06 0.40 0.07 0.11 0.20 0.07 0.07 Jan. 15 0.22 0.06 0.06 0.07 0.07 0.05 0.27 0.06 0.06 0.14 0.05 0.06 Jan. 17 0.17 0.07 0.07 0.07 0.07 0.05 0.31 0.05 0.10 0.18 0.11 0.08 Jan. 19 0.06 0.06 0.07 0.06 0.06 0.05 0.32 0.04 0.08 0.19 0.16 0.06 Jan. 24 0.07 0.04 0.05 0.04 0.05 0.04 0.47 0.03 0.04 0.31 0.03 0.04 Jan.26 0.06 0.03 0.05 0.03 0.04 0.06 0.49 0.02 0.06 0.34 0.03 0.04 Average 0.13 0.06 0.07 0.06 0.07 0.05 0.31 0.06 0.09 0.23 0.07 0.08 S T D E V 0.09 0.02 0.01 0.02 0.02 0.01 0.12 0.02 0.04 0.07 0.04 0.06 Median 0.07 0.06 0.07 0.07 0.07 0.06 0.31 0.06 0.10 0.20 0.05 0.07 Raw Data of Orth-P (mg/L) from Run 8 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Jan. 4 3.1 3.0 22.3 16.4 19.3 6.6 0.3 0.4 0.1 0.0 0.3 0.0 Jan. 7 3.1 2.8 24.3 17.8 19.7 7.4 0.1 0.4 0.1 0.0 0.3 0.0 Jan. 10 3.2 2.8 23.1 17.4 20.6 5.9 0.1 0.3 0.1 0.0 0.2 0.0 Jan. 13 2.9 2.7 23.8 18.8 20.6 2.9 0.1 0.5 0.1 0.0 0.2 0.0 Jan. 15 3.1 2.5 24.0 20.0 19.9 3.2 0.3 0.4 0.5 0.1 0.3 0.1 Jan. 17 3.1 2.7 24.5 18.4 20.7 2.8 0.1 0.4 0.2 0.1 0.2 0.0 Jan. 19 2.4 2.5 24.1 18.7 20.7 2.2 0.2 0.4 0.1 0.1 0.2 0.1 Jan. 24 2.1 2.1 24.0 17.9 18.6 1.5 0.1 0.5 0.1 0.3 0.4 0.0 Jan.26 2.9 1.9 25.0 19.3 19.1 2.1 0.1 0.4 0.1 0.1 0.3 0.0 Average 2.9 2.6 23.9 18.3 19.9 3.8 0.1 0.4 0.2 0.1 0.3 0.0 S T D E V 0.4 0.4 0.8 1.1 0.8 2.2 0.1 0.1 0.1 0.1 0.1 0.0 Median 3.1 2.7 24.0 18.4 19.9 2.9 0.1 0.4 0.1 0.1 0.3 0.0 2 6 6 Raw Data from Run 8 Run 8 started on Deo. 23, 1995 Raw Data of TKN (mg/L) from Run 8 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Jan. 4 36.6 276 272 239 249 240 247 13.4 2.8 4.1 Jan. 7 35.5 331 287 405 286 330 263 11.6 1.8 1.0 Jan. 10 30.2 342 315 282 254 256 290 6.3 2.8 2.4 Jan. 13 35.4 244 269 365 276 250 282 2.8 1.9 1.9 Jan. 15 35.0 241 291 308 245 257 267 2.8 2.0 2.1 Jan. 17 31.2 257 282 433 240 276 292 3.1 3.8 4.4 Jan. 19 29.5 221 279 286 246 273 270 2.2 2.8 1.6 Jan. 24 30.5 224 278 276 240 249 227 3.3 1.3 1.0 Jan. 26 24.2 205 263 289 155 242 253 3.3 2.1 1.4 Average 32.0 260 282 320 243 264 266 5.4 2.4 2.2 S T D E V 4.0 48 15 65 37 28 21 4.2 0.7 1.3 Median 31.2 244 279 289 246 256 267 3.3 2.1 1.9 Raw Data of TP (mg/L) From Run #8 Date Influent Anaerobic Aerobic Effluent A B C A B C A B C Jan. 4 6.2 129 107 103 123 105 112 0.5 0.6 0.4 Jan. 7 5.5 157 122 152 140 128 121 0.4 0.6 0.2 Jan. 10 5.5 160 128 123 132 127 135 0.2 0.4 0.4 Jan. 13 4.3 121 119 155 141 120 129 0.3 0.6 0.3 Jan. 15 5.6 116 126 133 123 122 127 0.9 1.0 0.5 Jan. 17 5.4 129 122 176 122 129 137 0.4 0.4 0.1 Jan. 19 5.1 117 121 126 131 129 131 0.3 0.4 0.2 Jan. 24 4.1 116 121 120 130 121 113 0.6 0.6 0.3 Jan. 26 4.2 118 120 121 114 95 115 0.4 0.3 0.4 Average 5.1 129 121 134 128 120 124 0.4 0.5 0.3 S T D E V 0.7 17 6 22 9 12 10 0.2 0.2 0.1 Median 5.4 121 121 126 130 122 127 0.4 0.6 0.3 Raw Data of COD (mg/L) From Run #8 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Jan. 7 442 227 440 194 144 144 160 107 88 85 90 58 79 53 Jan. 10 402 209 392 170 122 120 161 68 68 59 68 34 27 32 Jan. 13 383 218 410 170 120 142 168 75 167 50 68 47 83 47 Jan. 15 355 233 392 165 138 150 137 92 137 78 80 67 42 67 Jan. 17 353 268 350 182 138 121 138 67 71 43 61 43 43 19 Jan. 19 364 203 365 153 130 142 117 70 70 52 67 35 20 37 Jan. 24 304 156 255 118 96 135 118 74 68 59 59 52 39 42 Jan. 26 368 198 267 87 95 104 113 101 73 90 73 50 78 31 Average 371 214 359 155 123 132 139 82 93 65 71 48 51 41 S T D E V 40 32 66 35 19 16 22 16 38 17 10 11 25 15 Median 366 214 379 168 126 139 138 75 72 59 68 49 43 40 Raw Data Of Suspended Solids (m g/L) FromRun #8 Date SS V S S % M L S S M L V S S % E ffluent £ Raw Influent Raw Influent A B C A B C A B c Jan. 4 149 170 80.5% 80.6% 3712 3448 3452 73.1% 74.5% 74.9% 6.8 7.8 5.2 Jan. 7 139 196 84.2% 82.1% 3712 3720 3904 75.0% 75.5% 76.5% 5.7 7.4 5.1 Jan. 10 133 146 88.0% 90.4% 3548 3788 3920 74.9% 75.6% 76.1% 4.6 5.2 6.4 Jan. 13 137 141 87.6% 86.5% 3496 3736 3876 74.6% 76.8% 76.8% 6.1 5.1 5.8 Jan. 15 116 157 82.8% 83.4% 3436 3788 3812 75.6% 76.3% 76.8% 12.1 7.6 6.1 Jan. 17 135 137 85.2% 85.4% 3512 3748 3948 76.9% 78.2% 77.8% 10.2 4.7 6.8 Jan. 19 120 145 84.2% 80.0% 3744 3804 3724 75.1% 76.2% 76.3% 15.0 6.0 14.0 Jan. 24 94 105 83.0% 83.8% 3100 3600 3636 75.4% 76.6% 77.6% 15.1 4.0 7.2 Jan. 26 138 127 83.3% 85.8% 2960 3444 3504 74.6% 77.1% 77.3% 9.2 5.9 8.2 Average 129 147 84.3% 84.2% 3469 3675 3753 75.0% 76.3% 76.7% 9.4 6.0 7.2 S T D E V 16 26 2.4% 3.2% 274 143 185 1.0% 1.1% 0.9% 4.0 1.4 2.7 Median 135 145 84.2% 83.8% 3512 3736 3812 75.0% 76.3% 76.8% 9.2 5.9 6.4 267 Raw Data from Run 9 Run 9 started on Jan. 26, 1995 Raw Data of Ammonia (mg/L) from Run 9 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Mar. 5 14.6 12.2 8.0 12.4 10.3 2.5 0.5 8.4 4.3 0.4 8.6 4.4 Mar. 6 13.9 12.0 7.6 11.9 8.5 2.2 0.4 7.7 2.3 0.3 7.9 2.5 Mar. 7 13.9 12.4 7.9 11.8 8.8 2.2 0.4 7.7 1.7 0.3 7.8 1.7 Mar. 12 10.7 9.4 4.5 7.8 7.1 1.7 0.3 6.2 1.8 0.2 6.3 1.6 Mar. 13 10.7 9.0 6.5 8.3 7.2 1.8 0.3 6.1 1.6 0.2 6.0 1.4 Mar. 14 13.6 13.3 8.8 12.1 9.0 2.7 0.4 9.8 3.5 0.3 9.3 2.9 Mar. 15 13.8 12.3 8.5 12.4 9.7 3.0 0.6 9.2 3.9 0.4 9.0 3.6 Mar. 17 13.4 11.2 6.6 9.5 8.4 2.3 0.6 9.1 4.0 0.3 8.6 3.7 Average 13.1 11.5 7.3 10.8 8.6 2.3 0.4 8.0 2.9 0.3 7.9 2.7 STDEV 1.5 1.5 1.4 1.9 1.1 0.4 0.1 1.4 1.2 0.1 1.2 1.1 Median 13.7 12.1 7.7 11.9 8.6 2.2 0.4 8.1 2.9 0.3 8.2 2.7 Raw Data of NOx (mg/L) from Run 9 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Mar. 5 0.3 0.2 0.3 0.3 0.6 4.0 7.0 0.1 0.8 6.6 0.2 0.7 Mar. 6 0.3 0.3 0.4 0.4 0.3 4.2 7.5 0.1 2.1 7.1 0.2 2.1 Mar. 7 0.2 0.2 0.3 0.3 0.5 3.1 6.2 0.1 1.7 5.9 0.2 2.0 Mar. 12 0.2 0.6 0.6 0.6 0.7 3.9 6.7 0.1 0.7 6.1 0.6 1.0 Mar. 13 0.6 0.3 0.5 0.5 0.5 2.6 5.9 0.0 0.7 5.7 0.8 1.0 Mar. 14 0.4 0.3 0.4 0.3 0.3 1.8 5.5 0.2 0.4 5.2 0.7 0.7 Mar. 15 0.3 0.2 0.3 0.3 0.2 1.9 5.6 0.2 0.4 5.4 0.6 0.7 Mar. 17 0.2 0.2 0.4 0.3 0.3 1.8 4.4 0.0 0.5 4.5 0.9 1.0 Average 0.3 0.3 0.4 0.4 0.4 2.9 6.1 0.1 0.9 5.8 0.5 1.1 STDEV 0.1 0.1 0.1 0.1 0.2 1.0 1.0 0.1 0.6 0.8 0.3 0.6 Median 0.3 0.3 0.4 0.3 0.4 2.9 6.0 0.1 0.7 5.8 0.6 1.0 Raw Data of Ort h-P (mg/L) from Run 9 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B c Mar. 5 3.1 2.4 7.7 17.2 13.2 0.7 0.3 0.4 1.1 0.3 0.8 0.6 Mar. 6 2.2 2.1 9.4 16.5 8.5 0.6 0.3 0.1 0.8 0.0 0.5 0.4 Mar. 7 2.5 2.1 5.5 15.6 11.1 0.3 0.1 0.4 0.2 0.0 0.5 0.5 Mar. 12 2.7 2.2 12.5 18.6 18.1 0.6 0.3 0.2 0.6 0.1 0.6 0.3 Mar. 13 2.7 2.3 15.2 18.8 18.4 0.8 0.2 0.1 0.4 0.1 0.5 0.2 Mar. 14 2.1 2.7 13.5 18.1 17.4 0.7 0.2 0.2 0.4 0.1 0.7 0.3 Mar. 15 2.3 2.5 13.9 15.9 16.1 0.5 0.2 0.2 0.5 0.1 0.6 0.3 Mar. 17 2.3 2.2 14.7 18.9 18.4 1.0 0.4 0.3 0.5 0.2 0.8 0.3 Average 2.5 2.3 11.5 17.5 15.1 0.6 0.2 0.2 0.5 0.1 0.6 0.4 STDEV 0.3 0.2 3.6 1.3 3.8 0.2 0.1 0.1 0.3 0.1 0.1 0.1 Median 2.4 2.2 13.0 17.7 16.8 0.6 0.2 0.2 0.5 0.1 0.6 0.3 268 Raw Data from Run 9 Run 9 started on Jan. 26, 1995 Raw Data of TKN (mg/L) from Run 9 Date Influent Aerobic Effluent A B C A B C Mar. 5 25.0 203 251 260 2.0 13.3 6.0 Mar. 6 35.0 144 248 261 1.7 9.8 4.0 Mar. 7 25.5 168 269 231 1.5 13.5 3.1 Mar. 12 27.1 172 255 255 2.2 12.4 4.0 Mar. 13 24.6 149 239 237 1.7 12.9 3.7 Mar. 14 25.9 139 229 259 2.3 11.1 4.6 Mar. 15 24.8 139 236 243 1.9 13.9 4.4 Mar. 17 26.4 131 220 308 2.3 13.5 5.9 Average 26.8 156 243 257 1.9 12.6 4.5 STDEV 3.4 23.9 15.6 23.6 0.3 1.4 1.0 Median 25.7 147 244 257 2.0 13.1 4.2 Raw Data of TP (mg/L) from Run 9 Date Influent Aerobic Effluent A B C A B C Mar. 5 5.1 117 116 155 0.5 2.3 1.7 Mar. 6 8.1 103 113 148 0.4 0.5 1.0 Mar. 7 4.7 110 121 133 0.7 1.2 1.8 Mar. 12 5.1 106 117 138 1.0 0.7 0.7 Mar. 13 5.3 103 110 133 0.8 1.7 0.8 Mar. 14 5.5 98 119 137 1.0 0.9 0.8 Mar. 15 5.4 99 115 131 1.0 1.7 0.5 Mar. 17 6.0 95 109 165 1.0 1.1 1.7 Average 5.7 104 115 143 0.8 1.3 1.1 STDEV 1.1 7 4 12 0.2 0.6 0.5 Median 5.3 103 116 138 0.9 1.2 0.9 Raw Data of COD (mg/L) from Run 9 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Mar. 5 307 165 311 145 126 136 158 137 61 64 108 51 81 97 Mar. 6 475 168 648 188 92 153 95 61 46 46 64 31 31 35 Mar. 7 384 188 518 128 104 112 121 52 52 54 66 38 35 52 Mar. 12 326 174 362 146 99 136 148 74 66 49 74 49 71 49 Mar. 13 348 163 304 133 75 100 133 63 51 30 53 41 41 23 Mar. 14 272 179 348 149 100 166 141 68 63 75 100 83 75 91 Mar. 15 551 187 328 150 100 109 100 67 67 63 63 50 35 42 Mar. 17 386 182 446 137 87 129 103 69 45 27 51 29 29 34 Average 381 176 408 147 98 130 125 74 56 51 72 47 50 53 STDEV 92 10 122 18 15 23 24 26 9 17 21 17 22 27 Median 366 177 355 146 100 133 127 68 57 52 65 45 38 46 Raw Data of Suspended Solids (mg/L) from Run 9 Date SS vss% MLSS MLVSS% Effluent SS Raw Influent Raw Influent A B C A B C A B C Mar. 5 94 147 86.2% 84.4% 2584 3592 4028 71.4% 73.7% 81.0% 6.3 5.2 3.7 Mar. 6 147 94 84.4% 86.2% 2656 3664 4088 71.4% 74.8% 71.6% 7.8 7.0 6.8 Mar. 7 136 118 86.8% 88.1% 2736 3596 3912 71.8% 74.0% 72.0% 31.0 6.6 5.2 Mar. 12 111 156 85.6% 86.5% 2728 3768 4040 60.0% 73.9% 73.0% 6.6 4.1 4.2 Mar. 13 104 132 87.5% 86.4% 2380 3452 3968 72.4% 74.9% 73.7% 14.4 3.2 3.8 Mar. 14 83 109 97.6% 94.5% 2500 3456 3872 74.2% 76.7% 74.2% 11.2 2.9 3.0 Mar. 15 236 119 86.9% 83.2% 2320 3276 3656 72.1% 75.2% 74.7% 11.7 3.3 5.2 Mar. 18 na na na na 2608 3832 4012 71.8% 76.6% 74.0% na na na Average 130 125 87.9% 87.0% 2558 3543 3938 70.5% 74.7% 74.3% 12.7 4.6 4.6 STDEV 52 22 4.4% 3.6% 165 162 145 4.7% 1.0% 3.2% 8.6 1.7 1.3 Median 111 119 86.8% 86.4% 2596 3594 3990 71.8% 74.9% 73.9% 11.2 4.1 4.2 269 Raw Data from Run 10 Run 10 started on Mar. 21, 1996 Raw Data of Ammonia (mg/L) from Run 1 0 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Apr. 23 19.4 17.4 17.4 11.4 16.6 11.2 10.5 1.5 7.9 9.5 1.0 7.0 Apr. 24 20.4 18.8 14.4 13.2 12.8 10.6 10.1 4.6 4.7 9.0 3.5 4.0 Apr. 25 21.6 17.6 14.8 13.2 12.5 10.8 9.6 5.6 6.4 9.2 5.2 5.5 Apr. 26 20.4 18.8 15.2 14.0 14.8 11.6 10.3 6.9 8.2 9.5 6.7 7.6 Apr. 29 23.6 20.0 16.0 16.8 16.8 9.4 8.8 9.6 14.0 8.0 9.0 13.8 Apr. 30 17.0 14.0 10.8 10.8 11.4 6.7 6.2 6.5 6.1 5.7 6.4 6.4 Average 20.4 17.8 14.8 13.2 14.2 10.0 9.3 5.8 7.9 8.5 5.3 7.4 STDEV 2.2 2.1 2.2 2.1 2.3 1.8 1.6 2.7 3.3 1.5 2.8 3.4 Median 20.4 18.2 15.0 13.2 13.8 10.7 9.9 6.1 7.2 9.1 5.8 6.7 Raw Data of NOx (mg/L) from Run 1 0 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Apr. 23 0.3 0.2 0.2 0.3 0.4 0.3 1.4 3.0 0.1 1.7 3.7 0.6 Apr. 24 0.2 0.2 0.3 0.2 0.3 0.3 1.5 0.3 0.5 1.9 0.6 1.7 Apr. 25 0.2 0.2 0.3 0.2 0.2 0.2 1.6 0.2 0.2 1.9 0.5 0.7 Apr. 26 0.0 0.2 0.2 0.2 0.2 0.2 1.8 0.2 0.2 2.1 0.5 0.7 Apr. 29 0.2 0.2 0.2 0.3 0.2 0.2 1.1 0.2 0.2 1.7 0.5 0.6 Apr. 30 0.2 0.1 0.2 0.2 0.2 0.3 1.4 0.2 0.2 1.9 0.5 0.3 Average 0.2 0.2 0.2 0.2 0.2 0.3 1.5 0.7 0.2 1.9 1.0 0.8 STDEV 0.1 0.0 0.1 0.0 0.1 0.0 0.2 1.1 0.1 0.1 1.3 0.5 Median 0.2 0.2 0.2 0.2 0.2 0.3 1.5 0.2 0.2 1.9 0.5 0.7 Raw Data of Nitrite (mg/L) from Run 1 0 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Apr. 25 0.09 0.04 0.05 0.04 0.04 0.07 0.32 0.03 0.03 0.37 0.16 0.15 Raw Data of Orth-P (mg/L) from Run 1 0 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B c A B C Apr. 23 2.4 2.4 13.0 21.6 16.9 3.3 0.4 0.6 2.5 0.0 0.2 1.7 Apr. 24 2.4 2.5 13.0 21.5 15.7 3.2 0.1 0.5 0.7 0.1 0.9 0.4 Apr. 25 2.7 2.3 14.2 22.2 15.2 3.3 0.1 0.6 1.3 0.1 1.2 0.8 Apr. 26 2.4 2.6 15.7 23.0 17.5 3.9 0.1 0.4 0.9 0.0 0.8 0.9 Apr. 29 3.0 2.6 16.7 23.4 17.7 0.2 0.1 0.3 2.6 0.0 0.3 5.0 Apr. 30 2.9 2.3 17.3 21.0 15.3 0.2 0.1 0.2 0.2 0.0 0.2 0.1 Average 2.6 2.5 15.0 22.1 16.4 2.4 0.1 0.4 1.4 0.0 0.6 1.5 STDEV 0.3 0.1 1.9 0.9 1.1 1.7 0.1 0.2 1.0 0.0 0.4 1.8 Median 2.6 2.5 15.0 21.9 16.3 3.3 0.1 0.5 1.1 0.0 0.5 0.9 270 Sheet2 Raw Data from Run 10 Run 10 started on Mar. 21, 1996 Raw Data of TKN (mg/L) from Run 10 Date Influent Aerobic Effluent A B C A B C Apr. 23 25.4 184 233 234 12.4 2.0 8.5 Apr. 24 30.0 173 236 236 12.6 4.3 4.7 Apr. 25 26.0 179 239 250 11.0 5.6 7.3 Apr. 26 26.6 176 241 235 12.0 7.2 8.1 Apr. 29 26.1 210 261 243 8.8 9.2 8.2 Apr. 30 26.4 190 244 227 7.2 7.2 6.6 Averag 26.8 185 242 238 10.7 5.9 7.2 STDEV 1.6 14 10 8 2.2 2.5 1.4 Median 26.3 181 240 236 11.5 6.4 7.7 Raw Data of TP (mg/L) from Run 10 Date Influent Aerobic Effluent A B C A B C Apr. 23 5.3 74 117 103 0.0 0.3 2.5 Apr. 24 3.7 71 126 108 0.1 1.0 0.5 Apr. 25 4.0 69 117 104 0.2 1.2 1.0 Apr. 26 4.0 78 122 109 0.2 0.8 0.9 Apr. 29 3.6 93 133 105 0.2 0.4 4.5 Apr. 30 4.0 90 132 94 0.1 0.4 0.2 Averag 4.1 79 124 104 0.1 0.7 1.6 STDEV 0.6 10 7 5 0.1 0.4 1.6 Median 4.0 76 124 105 0.1 0.6 0.9 Raw Data of COD (mg/L) from Run 10 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Apr. 23 351 186 310 132 102 102 102 85 59 42 51 42 34 42 Apr. 24 298 166 345 153 118 119 113 83 66 43 99 55 33 35 Apr. 25 323 185 320 121 109 118 101 84 47 51 37 37 39 25 Apr. 26 328 172 303 123 101 135 109 67 50 47 42 47 34 25 Apr. 29 311 197 328 143 100 144 118 66 66 57 64 61 33 46 Apr. 30 295 170 300 123 113 98 98 66 57 46 43 41 30 34 Averag 318 179 318 133 107 119 107 75 58 48 56 47 34 35 STDEV 21 12 17 13 7 18 8 10 8 6 23 9 3 9 Median 317 179 315 128 106 119 106 75 58 47 47 45 34 35 Raw Data of Suspended Solids (mg/L) from Run 10 Date SS VSS% MLSS MLVSS% Effluent SS Raw Influen Raw Influen A B C A B C A B C Apr. 8 na na na na 2536 3688 3624 na na na 31.0 7.0 7.4 Apr. 11 na na na na 2476 3484 3536 na na na na na na Apr. 23 117 135 88.0% 87.0% 2604 3316 3472 79.0% 80.0% 78.0% 18.4 8.2 7.0 Apr. 24 96 154 85.0% 81.0% 2508 3404 3328 80.0% 80.0% 19.8 7.7 10.6 Apr. 25 99 127 92.0% 89.0% 2472 3404 3276 79.0% 78.0% 79.0% 14.9 8.4 9.9 Apr. 26 111 130 82.0% 72.0% 2604 3372 3260 76.0% 76.0% 77.0% 14.2 8.6 8.0 Apr. 29 96 131 86.0% 86.0% 3152 3632 3312 75.0% 78.0% 80.0% 25.9 10.2 4.4 Apr. 30 103 112 92.0% 75.0% 2800 3512 2984 74.0% 77.0% 79.0% 27.2 8.8 4.8 Averag 104 135 86.6% 83.0% 2622 3471 3401 77.8% 78.0% 78.8% 20.7 8.4 7.9 STDEV 10 11 3.7% 6.8% 240 139 142 2.2% 1.6% 1.3% 6.6 1.1 2.2 Median 101 131 87.0% 83.5% 2570 3444 3320 77.5% 78.0% 79.0% 20 8 7 271 Raw Data from Run 1 1 Run 11 started on May 1, 1996 Raw Data of Ammonia (mg/L) from Run 11 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C May. 15 18.0 15.2 9.6 13.0 10.4 1.1 0.1 8.0 2.6 0.1 6.7 1.8 May. 17 16.3 14.0 8.4 13.6 9.1 1.3 0.2 5.9 1.4 0.1 5.7 1.2 May. 19 17.8 16.5 9.2 11.8 9.9 1.2 0.1 7.3 0.4 0.0 7.0 0.3 May. 21 16.9 16.0 11.5 12.1 11.6 1.0 0.1 7.8 0.2 0.0 7.7 0.3 May. 23 18.0 15.6 9.6 12.4 10.8 1.3 0.1 6.7 1.2 0.1 7.0 0.3 May. 25 18.2 15.6 11.9 13.2 9.2 1.3 0.2 7.6 0.5 0.1 7.4 0.5 May. 27 21.0 19.2 11.2 14.8 11.2 2.1 0.3 9.2 0.2 0.0 8.6 0.1 May. 29 21.5 19.0 13.2 16.6 13.4 2.1 0.5 9.6 0.2 0.3 9.1 0.1 Average 18.5 16.4 10.6 13.4 10.7 1.4 0.2 7.8 0.9 0.1 7.4 0.6 STDEV 1.8 1.8 1.6 1.6 1.4 0.4 0.1 1.2 0.8 0.1 1.1 0.6 Median 18.0 15.8 10.4 13.1 10.6 1.3 0.2 7.7 0.5 0.1 7.2 0.3 Raw Data of NOx (mg/I from Run 11 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C May. 15 0.2 0.2 0.3 0.3 0.3 1.2 3.9 0.2 2.8 2.5 0.5 4.5 May. 17 0.1 0.1 0.1 0.1 0.1 1.4 4.1 0.2 4.9 3.6 0.6 6.0 May. 19 0.1 0.1 0.2 0.2 0.2 7.0 9.9 0.3 11.5 7.8 0.7 8.6 May. 21 0.1 0.2 0.3 0.3 0.3 1.3 4.0 0.4 11.2 3.9 0.7 9.0 May. 23 0.3 0.2 0.4 0.3 0.4 4.8 7.3 0.2 10.1 4.2 1.1 11.2 May. 25 0.4 0.3 0.4 0.3 0.4 4.6 6.9 0.2 8.2 5.2 1.0 7.6 May. 27 0.3 0.2 0.4 0.3 0.3 3.6 5.6 0.4 12.3 4.0 1.1 11.7 May. 29 0.2 0.1 0.3 0.3 0.1 1.7 4.0 0.2 12.2 3.9 1.1 11.9 Average 0.2 0.2 0.3 0.2 0.2 3.2 5.7 0.3 9.2 4.4 0.8 8.8 STDEV 0.1 0.1 0.1 0.1 0.1 2.1 2.2 0.1 3.6 1.6 0.2 2.7 Median 0.2 0.2 0.3 0.3 0.3 2.7 4.9 0.2 10.7 4.0 0.9 8.8 Raw Data of Nitrite (mg/L) from Run 11 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C May. 23 0.15 0.04 0.05 0.07 0.05 0.32 0.30 0.06 0.14 0.20 0.17 0.12 Raw Data of Orth-P (mg/L) from Run 11 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B C A B C May. 15 2.5 2.8 7.6 12.6 6.6 1.7 1.2 0.1 0.3 1.1 0.1 0.16 May. 17 2.5 2.8 3.2 13.9 2.5 1.8 1.5 0.1 1.0 1.2 0.1 0.59 May. 19 3.3 2.7 2.4 4.1 2.7 1.6 1.4 0.3 1.5 1.3 0.4 1.48 May. 21 3.1 3.1 14.2 13.6 8.7 1.9 1.3 0.4 0.8 1.2 0.8 1.18 May. 23 3.2 3.0 3.5 7.0 3.1 0.8 0.7 0.3 0.8 0.6 0.8 0.77 May. 25 2.8 3.0 4.0 7.5 3.8 1.0 0.9 0.4 0.4 1.1 0.9 0.57 May. 27 2.8 3.0 3.7 5.9 2.6 1.6 1.7 0.5 0.6 1.4 0.8 0.42 May. 29 2.8 2.8 14.2 16.4 10.8 2.7 1.9 1.0 1.8 2.0 2.0 2.08 Average 2.9 2.9 6.6 10.1 5.1 1.6 1.3 0.4 0.9 1.2 0.7 0.9 STDEV 0.3 0.2 4.9 4.5 3.2 0.6 0.4 0.3 0.5 0.4 0.6 0.6 Median 2.8 2.9 3.8 10.1 3.5 1.7 1.3 0.4 0.8 1.2 0.8 0.7 272 Raw Data from Run 11 Run 11 started on May 1, 1996 Raw Data of TKN (mg/L) from Run 11 Date Influent Aerobic Effluent A B C A B C May. 15 27.3 246 260 240 1.0 8.8 1.8 May. 17 29.0 231 233 275 1.7 7.0 2.6 May. 19 25.1 193 267 235 2.6 9.1 0.4 May. 21 29.6 206 254 211 1.0 13.7 1.6 May. 23 24.4 193 276 235 2.2 8.0 0.6 May. 25 25.2 221 272 224 0.7 10.6 0.5 May. 27 29.2 173 277 241 1.2 9.9 1.3 May. 29 28.6 227 284 269 2.1 10.2 2.7 Average 27 211 26S 241 1.6 9.7 1.4 STDEV 2 24 16 21 0.7 2.0 0.9 Median 28.0 214 269 237 1.4 9.5 1.4 Raw Data of TP (mg/L) from Run 11 Date Influent Aerobic Effluent A B C A B C May. 15 6.4 115 104 91 1.5 0.1 0.4 May. 17 4.8 98 109 101 2.1 0.4 0.9 May. 19 5.6 91 108 93 1.4 0.4 1.6 May. 21 6.2 98 121 98 1.3 1.1 1.2. May. 23 5.1 103 123 99 0.7 0.9 0.9 May. 25 6.8 107 138 106 1.4 0.8 0.7 May. 27 4.8 93 129 100 1.4 1.0 0.5 May. 29 6.1 88 126 113 3.0 1.7 2.5 Average 5.7 99 120 100 1.6 0.8 1.1 STDEV 0.8 9 12 7 0.7 0.5 0.7 Median S.9 98 122 99 1.4 0.9 0.9 Raw Data of COD (mg/L) from Run 11 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C May. 15 311 156 353 128 76 121 99 34 34 34 25 34 17 25 May. 17 324 156 347 117 84 132 87 65 42 50 50 27 25 34 May. 19 406 234 373 108 91 99 108 66 46 36 36 25 30 25 May. 21 421 235 468 192 111 135 138 67 52 59 54 50 50 42 May. 25 373 215 351 141 96 145 116 56 48 48 50 32 32 32 May. 27 356 na 352 130 96 104 96 62 62 53 48 35 32 32 May. 29 358 205 422 164 90 120 95 44 41 67 41 33 28 31 Average 364 200 381 140 92 122 106 56 46 50 43 34 31 32 STDEV 40 36 47 29 11 17 17 13 9 12 10 8 10 6 Median 358 210 353 130 91 121 99 62 46 50 48 33 30 32 Raw Data of Suspended Solids (mg/L) from Run 11 Date SS vss% Anaerobic MLSS Aerobic MLSS MLVSS% Effluent S! > Raw Influent Raw Influent A B C A B C A B C A B C May. 15 97 142 85.0% 85.0% na na ma 3024 3372 3460 75.0% 78.0% 78.0% 7.7 14.0 8.2 May. 17 100 185 86.0% 81.0% na na ma 2948 3428 3328 76.0% 78.0% 79.0% 19.5 11.4 6.1 May. 19 143 198 87.0% 83.0% 3604 3568 4036 2960 3488 3376 75.0% 79.0% 78.0% 15.4 11.4 5.8 May. 21 128 190 84.0% 80.0% 4388 3760 3676 2992 3652 3552 76.0% 78.0% 79.0% 12.6 7.0 3.6 May. 23 126 233 81.0% 76.0% 3420 4300 3752 3080 2968 3540 75.0% 77.0% 77.0% ISO 5.6 3.4 May. 25 102 158 84.0% 83.0% 3408 4012 4152 2996 3720 3612 76.0% 77.0% 77.0% 21.0 5.8 3.2 May. 27 123 165 85.0% 81.0% 3272 3352 3744 2852 3764 3524 75.0% 77.0% 77.0% 34.7 11.0 7.6 May. 29 120 163 86.0% 82.0% 3380 3716 3724 2912 3788 3692 76.0% 77.0% 77.0% 42.0 5.8 8.2 Average 117 179 84.8% 81.4% 3579 3785 3847 2971 3523 3511 75.5% 77.6% 77.8% 21 9 6 STDEV 16 29 1.8% 2.7% 411 334 196 70 274 120 0.5% 0.7% 0.9% 11 3 2 Median 122 175 85.0% 81.5% 3414 3738 3748 2976 3570 3532 75.5% 77.5% 77.5% 19 9 6 273 Raw Data from Run 12 Run 12 started on Jun. 1, 1996 Raw Data of Ammonia (mg/L) from Run 12 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A . B C Jun. 8 18.6 17.2 12.0 14.2 12.9 2.0 0.1 6.9 0.4 0.0 6.7 0.2 Jun. 12 19.5 15.3 9.9 10.7 10.9 1.7 0.1 1.4 0.1 0.0 1.3 0.1 Jun. 13 16.8 14.2 9.1 7.0 9.2 1.1 0.0 0.0 0.0 0.0 0.0 0.0 Jun. 14 18.0 18.2 8.9 14.0 9.5 2.2 0.1 0.1 0.1 0.0 0.2 0.1 Jun. 16 17.4 17.5 10.6 11.6 11.8 1.9 0.1 0.1 0.1 0.0 0.2 0.1 Jun. 17 19.6 16.0 10.4 11.3 10.5 2.0 0.1 0.8 0.1 0.0 0.5 0.1 Jun. 18 20.6 19.6 12.5 11.0 12.5 2.1 0.1 0.3 0.1 0.0 0.2 0.1 Average 18.6 16.9 10.5 11.4 11.0 1.9 0.1 1.4 0.1 0.0 1.3 0.1 STDEV 1.3 1.8 1.4 2.4 1.4 0.4 0.0 2.5 0.1 0.0 2.4 0.1 Median 18.6 17.2 10.4 11.3 10.9 2.0 0.1 0.3 0.1 0.0 0.2 0.1 Raw Data of NOx (mg/L) from Run 12 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B c A B c Jun. 8 0.05 0.2 0.3 0.3 0.3 8.1 9.1 0.4 8.3 7.9 0.6 8.0 Jun. 12 0.16 0.2 0.3 0.2 0.2 6.7 7.4 1.4 6.5 6.7 0.6 5.7 Jun. 13 0.30 1.5 1.1 0.3 0.3 4.7 7.1 7.7 10.0 6.9 5.9 10.2 Jun. 14 0.24 0.2 0.2 0.2 0.2 3.6 6.1 6.3 10.6 5.0 6.0 9.8 Jun. 16 0.18 0.1 0.3 0.2 0.2 4.3 6.6 5.4 11.6 6.8 5.3 10.6 Jun. 17 0.10 0.1 0.2 0.2 0.3 5.4 8.2 2.3 10.4 6.8 2.7 9.9 Jun. 18 0.18 0.2 0.2 0.2 0.2 3.5 5.8 4.4 10.8 5.2 3.4 10.1 Average 0.2 0.4 0.4 0.2 0.2 5.2 7.2 4.0 9.7 6.5 3.5 9.2 STDEV 0.1 0.5 0.3 0.0 0.1 1.7 1.2 2.7 1.8 1.0 2.3 1.7 Median 0.2 0.2 0.3 0.2 0.2 4.7 7.1 4.4 10.4 6.8 3.4 9.9 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B C A B c Jun. 12 0.15 0.04 0.06 0.06 0.05 0.17 0.13 0.13 0.10 0.08 0.15 0.07 Jun. 16 0.09 0.05 0.05 0.05 0.05 0.18 0.10 0.16 0.08 0.06 0.13 0.07 jun. 18 0.16 0.06 0.05 0.06 0.06 0.09 0.08 0.12 0.08 0.06 0.14 0.06 Average 0.13 0.05 0.05 0.06 0.05 0.15 0.11 0.13 0.08 0.07 0.14 0.07 STDEV 0.04 0.01 0.01 0.01 0.00 0.05 0.03 0.02 0.01 0.01 0.01 0.00 Median 0.15 0.05 0.05 0.06 0.05 0.17 0.10 0.13 0.08 0.06 0.14 0.07 Raw Data of Orth-P (mg/L) from Run 12 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Jun. 8 2.8 2.2 8.4 7.3 9.4 1.2 0.9 1.1 1.2 0.5 0.8 0.6 Jun. 12 3.2 2.4 3.0 4.2 4.9 2.3 2.3 2.0 2.3 2.1 1.9 2.1 Jun. 13 2.9 2.6 4.0 3.1 3.6 1.5 1.3 1.0 1.2 1.4 1.1 1.4 Jun. 14 2.2 2.8 3.9 3.2 3.9 1.5 1.1 1.1 1.4 1.5 1.2 1.4 Jun. 16 2.0 2.3 4.7 4.1 3.9 2.3 1.9 1.6 1.9 1.8 1.5 1.9 Jun. 17 2.2 2.3 2.6 2.9 2.5 2.0 1.8 1.5 1.9 1.8 1.3 1.7 Jun. 18 2.3 2.5 5.0 5.1 4.0 1.5 1.2 1.2 1.4 2.1 2.0 2.4 Average 2.5 2.4 4.5 4.3 4.6 1.8 1.5 1.4 1.6 1.6 1.4 1.7 STDEV 0.5 0.2 1.9 1.5 2.2 0.4 0.5 0.4 0.4 0.6 0.4 0.6 Median 2.3 2.4 4.0 4.1 3.9 1.5 1.3 1.2 1.4 1.8 1.3 1.7 274 Raw Data from Run 12 Run 12 started on Jun. 1, 1996 Raw Data of T K N (mg/L) from Run 12 Date Influent Aerobic Effluent A B C A B C Jun. 8 28.4 263 179 240 1.5 7.5 1.2 Jun. 12 25.1 227 232 242 2.2 2.0 0.4 Jun. 13 22.0 203 189 233 0.8 1.0 0.3 Jun. 14 26.5 228 223 250 0.4 2.1 0.3 Jun. 16 25.6 243 232 260 2.7 0.7 1.4 Jun. 17 26.7 239 304 236 0.7 0.9 0.7 Jun. 18 25.2 229 227 243 0.4 0.5 1.3 Average 25.6 233 226 243 1.3 2.1 0.8 S T D E V 2.0 18 40 9 0.9 2.5 0.5 Median 25.6 229 227 242 0.8 1.0 0.7 Raw Data of TP (mg/L) from Run 12 Date Influent Aerobic Effluent A B C A B C Jun. 8 5.3 111 74 113 0.4 1.0 0.6 Jun. 12 4.7 103 98 99 2.6 2.3 2.5 Jun. 13 4.2 97 92 97 1.9 0.9 1.5 Jun. 14 4.9 94 93 102 1.9 1.2 1.6 Jun. 16 4.2 101 99 101 1.7 1.6 2.2 Jun. 17 4.6 97 98 95 2.1 1.4 1.9 Jun. 18 4.0 94 96 96 2.3 2.3 2.9 Average 4.6 99 93 100 1.8 1.5 1.9 S T D E V 0.5 6 9 6 0.7 0.6 0.8 Median 4.6 97 96 99 1.9 1.4 1.9 Raw Data of COD (mg/L) from Run 12 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Jun. 8 345 191 na 142 108 125 120 75 52 50 50 32 33 28 Jun. 12 325 • 172 318 130 87 90 87 49 41 41 38 21 33 28 Jun. 13 320 185 623 131 95 105 98 51 49 49 57 36 33 34 Jun. 14 382 220 355 186 93 102 88 51 47 34 37 20 34 25 Jun. 16 395 227 390 179 130 138 146 81 57 65 62 41 49 32 Jun. 17 360 211 368 144 104 120 101 70 46 62 42 38 32 40 Jun. 18 394 223 366 157 99 117 117 53 43 51 40 28 31 36 Average 360 204 403 153 102 114 108 61 48 50 47 31 35 32 S T D E V 31 21 110 22 14 16 21 13 5 11 10 8 6 5 Median 360 211 367 144 99 117 101 53 47 50 42 32 33 32 Raw Data of Suspended Solids (mg/L) from Run 12 Date TSS V S S % Anaerobic M L S S M L S S M L V S S % E ffluent S S Raw Influent Raw Influent A B C A B C A B C A B C Jun. 8 130 87% 4144 2488 3648 3344 2388 3352 78% 80% 79% 9.3 3.8 7.6 Jun. 12 116 169 83% 4212 3232 4064 3248 3300 3468 78% 79% 79% 9.8 5.0 8.2 Jun. 13 . 116 364 86% 76% 3596 3052 3592 3244 3080 3376 79% 79% 79% 9.0 1.4 14.0 Jun. 14 133 178 81% 79% 3956 3092 3560 3388 3152 3492 78% 78% 79% 6.8 3.8 12.3 Jun. 16 124 166 83% 83% 4048 3372 3924 3460 3348 3468 77% 79% 79% 10.2 5.1 7.0 Jun. 17 110 165 85% 82% 3708 3104 3456 3224 3220 3328 78% 79% 79% 8.0 4.6 12.2 Jun. 18 115 134 85% 84% 3864 3340 3696 3044 3216 3356 78% 79% 79% 8.6 4.7 8.6 Average 121 196 8 4 % 8 1 % 3933 3097 3706 3279 3101 3406 7 8 % 7 9 % 7 9 % 8.8 4.1 10.0 S T D E V 9 84 2 % 3 % 225 296 214 135 326 68 1 % 1 % 0 % 1.1 1.3 2.8 Median 116 168 8 5 % 8 2 % 3956 3104 3648 3248 3216 3376 7 8 % 7 9 % 7 9 % 9.0 4.6 8.6 275 Raw Data from Run 13 Run 13 started on Jun. 20,1996 Raw Data of Ammonia (mg/L) from Run 13 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Jul. 8 22.1 19.3 12.7 13.2 12.6 3.1 0.0 0.8 0.1 0.0 0.8 0.2 Jul. 10 20.1 18.3 12.2 13.8 12.0 3.3 0.0 4.1 0.1 0.0 3.5 0.1 Jul. 11 18.8 17.7 11.7 12.2 11.2 3.2 0.0 1.1 0.1 0.0 0.8 0.1 Jul. 12 22.2 19.5 13.5 13.8 11.8 2.9 0.0 5.3 0.1 0.0 4.1 0.1 Jul. 13 18.8 16.4 11.6 10.9 10.2 3.0 0.0 1.2 0.0 0.0 1.4 0.2 Jul. 14 19.0 17.6 12.3 12.6 12.0 2.9 0.0 0.1 0.1 0.0 0.1 0.1 Jul. 15 20.8 19.7 13.6 15.4 13.8 2.8 0.1 2.4 0.1 0.0 2.5 0.2 Jul. 16 29.0 28.4 19.0 18.7 16.9 2.6 0.0 3.2 0.1 0.0 2.7 0.1 Average 21.4 19.6 13.3 13.8 12.6 3.0 0.0 2.3 0.1 0.0 2.0 0.1 STDEV 3.4 3.7 2.4 2.4 2.0 0.2 0.0 1.8 0.0 0.0 1.4 0.0 Median 20.4 18.8 12.5 13.5 12.0 3.0 0.0 1.8 0.1 0.0 1.9 0.1 Raw Data of NOx (mg/1 ^ ) from Run 13 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B c A B c Jul. 8 0.76 0.2 0.3 0.2 0.4 0.3 3.9 8.2 9.0 2.6 7.0 7.6 Jul. 10 0.07 0.2 0.3 0.3 0.1 0.2 4.3 2.8 8.2 3.1 3.2 6.2 Jul. 11 0.08 0.2 0.3 0.2 0.3 0.2 3.3 4.4 9.8 2.7 3.7 7.0 Jul. 12 0.07 0.2 0.2 0.2 0.3 0.3 3.1 2.5 10.9 2.8 3.1 9.7 Jul. 13 0.27 0.2 0.3 0.3 0.3 0.3 4.1 7.2 8.1 3.6 6.0 7.2 Jul. 14 0.04 0.3 0.2 0.3 0.3 0.3 4.1 10.2 10.1 3.3 8.2 9.1 Jul. 15 0.00 0.2 0.2 0.3 0.2 0.2 4.1 8.1 9.8 3.2 6.3 8.8 Jul. 16 0.00 0.2 0.3 0.3 0.4 0.3 4.0 6.6 10.0 3.5 6.2 9.5 Average 0.2 0.2 0.2 0.3 0.3 0.3 3.9 6.3 9.5 3.1 5.5 8.1 STDEV 0.3 0.0 0.0 0.1 0.1 0.0 0.4 2.8 1.0 0.4 1.9 1.3 Median 0.1 0.2 0.3 0.3 0.3 0.3 4.0 6.9 9.8 3.2 6.1 8.2 Raw Data of Nitrite (mg/L) from Run 13 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B C A A B C A B C Jul. 10 0.08 0.03 0.04 0.05 0.03 0.03 0.04 0.17 0.06 0.03 0.18 0.05 Jul. 13 0.09 0.04 0.03 0.05 0.03 0.02 0.04 0.20 0.08 0.03 0.32 0.08 Jul. 16 0.13 0.03 0.04 0.04 0.04 0.02 0.04 0.18 0.08 0.04 0.37 0.05 Average 0.10 0.03 0.03 0.05 0.03 0.02 0.04 0.18 0.07 0.03 0.29 0.06 STDEV 0.03 0.01 0.00 0.01 0.00 0.01 0.00 0.02 0.01 0.01 0.10 0.01 Median 0.09 0.03 0.04 0.05 0.03 0.02 0.04 0.18 0.08 0.03 0.32 0.05 Raw Data of Orth-P (mg/L) from Run 13 Date Raw Influent Anaerobic Anoxic Aerobic Effluent A B c A A B c A B c Jul. 8 3.1 2.2 6.6 3.5 19.0 1.0 0.0 0.2 0.2 0.2 0.2 0.1 Jul. 10 2.9 2.4 18.0 6.5 16.1 3.3 0.0 0.2 0.2 0.1 1.1 0.1 Jul. 11 2.7 2.5 4.4 3.7 18.1 2.2 0.0 0.1 0.1 0.2 0.1 0.1 Jul. 12 2.8 2.5 2.8 4.0 16.5 0.4 0.1 0.3 0.2 0.1 0.2 0.1 Jul. 13 2.6 2.4 7.5 5.5 19.1 2.0 0.0 0.1 0.2 0.1 0.7 0.1 Jul. 14 2.9 2.7 4.0 3.0 16.4 0.6 0.1 0.1 0.1 0.1 0.1 0.1 Jul. 15 3.0 3.0 8.4 7.3 20.6 1.4 0.1 0.5 0.2 0.1 0.9 0.1 Jul. 16 3.4 2.9 4.5 4.3 17.2 0.5 0.0 1.4 0.1 0.0 1.9 0.1 Average 2.9 2.6 7.0 4.7 17.9 1.4 0.1 0.4 0.2 0.1 0.6 0.1 STDEV 0.2 0.3 4.8 1.5 1.6 1.0 0.0 0.5 0.0 0.1 0.6 0.0 Median 2.9 2.5 5.5 4.1 17.7 1.2 0.0 0.2 0.2 0.1 0.5 0.1 276 Raw Data from Run 13 Run 13 started on Jun. 20, 1996 Raw Data of TKN (mg/L) from Run 13 Date Influent Aerobic Effluent A B C A B C Jul. 8 27.6 280 265 237 1.5 2.2 2.3 Jul. 10 26.4 236 308 291 1.5 3.7 1.4 Jul. 11 28.5 278 161 245 1.7 2.2 1.4 Jul. 12 29.0 251 291 241 2.2 5.5 1.9 Jul. 13 26.9 285 296 342 2.2 2.2 2.4 Jul. 14 31.7 229 271 265 1.0 1.7 1.5 Jul. 15 27.3 245 293 244 1.0 3.7 1.8 Jul. 16 31.3 250 242 205 1.0 5.0 1.5 Average 28.6 257 266 259 1.5 3.3 1.8 S T D E V 2.0 21 47 42 0.5 1.4 0.4 Median 28.1 250 281 244 1.5 2.9 1.6 Raw Data of TP (mg/L) from Run 13 Date Influent Aerobic Effluent A B C A B C Jul. 8 4.3 115 111 123 0.0 0.2 0.3 Jul. 10 3.9 108 125 133 0.2 1.4 0.2 Jul. 11 2.8 111 76 123 0.5 0.4 0.3 Jul. 12 4.3 106 122 122 0.6 0.4 0.4 Jul. 13 3.5 125 134 155 0.3 0.8 0.2 Jul. 14 5.4 114 134 155 0.3 0.3 0.3 Jul. 15 4.4 119 141 139 0.2 1.0 0.1 Jul. 16 3.8 124 130 109 0.2 2.6 0.1 Average 4.0 115 122 132 0.3 0.9 0.3 S T D E V 0.8 7 21 17 0.2 0.8 0.1 Median 4.1 114 127 128 0.3 0.6 0.3 Raw Data of COD (mg/L) from Run 13 Date Raw Influent Anaerobic Anoxic Aerobic Effluent Total Soluble Total Soluble A B C A A B C A B C Jul. 8 532 244 448 172 101 136 170 84 84 50 67 34 0 17 Jul. 10 307 169 348 136 82 119 116 34 9 17 17 51 17 0 Jul. 12 338 196 406 149 135 122 142 74 34 51 34 14 18 27 Jul. 13 324 195 316 149 86 105 110 34 41 32 47 24 29 31 Jul. 14 307 172 327 118 74 83 74 33 36 25 33 44 8 16 Jul. 15 284 157 314 132 74 66 74 17 25 -25 25 8 8 8 Jul. 16 315 218 334 145 103 105 103 63 40 76 na na na na Average 344 193 356 143 94 105 113 48 38 39 37 29 13 17 S T D E V 85 30 51 17 22 24 35 25 23 21 18 17 10 12 Median 315 195 334 145 86 105 110 34 36 32 34 29 13 17 Raw Data of Suspended Solids (mg/L) from Run 13 Date TSS V S S % Anaerobic M L S S M L S S M L V S S % E ffluent S S Raw Influent Raw Influent A B C A B C A B C A B C Jul. 8 123 156 86% 83% 4164 3576 4568 3228 4064 3292 79% 80% 78% 6.8 3.8 8.0 Jul. 10 88 139 85% 71% 3624 3272 3672 3392 4308 3512 78% 79% 77% 9.2 3.8 9.0 Jul. 12 107 150 89% 87% 3752 3696 3932 3332 3876 3596 78% 79% 79% 8.2 4.2 10.2 Jul. 13 71 153 80% 78% 3668 4128 3928 3280 4188 3528 78% 80% 77% 8.2 4.8 10.1 Jul. 14 121 121 93% 87% 3380 3928 3748 3388 3424 3544 78% 77% 77% 9.0 2.6 7.2 Jul. 15 109 140 86% 85% 3604 3676 3620 3424 3612 3432 77% 77%> 77% 7.2 9.2 9.4 Jul. 16 94 122 80% 84% 3580 3748 3468 3424 3960 3592 77% 77% 76% 10.8 8.2 8.6 Average 102 140 8 6 % 8 2 % 3682 3718 3848 3353 3919 3499 7 8 % 7 8 % 7 7 % 8.5 5.2 8.9 S T D E V 19 14 5 % 6 % 241 269 358 76 313 107 1 % 1 % 1 % 1.3 2.5 1.1 Median 107 140 8 6 % 8 4 % 3624 3696 3748 3388 3960 3528 7 8 % 7 9 % 7 7 % 8.2 4.2 9.0 277 Raw Data of Volatile Fatty Acids (VFAs) Run 4 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Anoxic *Acet. *Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Sep. 17 na na 16.5 0.0 23.4 0.0 47.8 0.0 10.8 0.0 na na Sep. 19 na na 11.5 0.0 8.5 0.0 20.7 0.0 5.1 0.0 na na Sep. 22 21.7 8.9 0.0 0.0 0.9 0.0 1.3 0.0 13.4 0.0 0.0 0.0 Sep. 25 17.9 6.7 0.0 0.0 0.0 0.0 0.0 0.0 14.6 0.0 0.0 0.0 Sep. 27 22.4 11.0 0.0 0.0 0.0 0.0 0.0 0.0 20.0 0.0 0.0 0.0 Sep. 29 21.6 3.5 12.0 0.0 7.0 0.0 11.0 0.0 19.6 0.0 12.0 0.0 Oct. 2 13.9 4.3 0.0 0.0 0.0 0.0 3.3 0.0 9.3 0.0 0.0 0.0 Oct. 4 12.6 4.5 1.5 0.0 1.8 0.0 1.5 0.0 8.5 0.0 1.4 0.0 Oct. 7 7.7 2.8 1.9 0.0 1.0 0.0 2.1 0.0 7.8 0.0 1.2 0.0 Average 16.8 6.0 4.8 0.0 4.7 0.0 9.7 0.0 12.1 0.0 2.1 0.0 STDEV 5.6 3.0 6.6 0.0 7.7 0.0 15.8 0.0 5.2 0.0 4.4 0.0 Median 17.9 4.5 1.5 0.0 1.0 0.0 2.1 0.0 10.8 0.0 0.0 0.0 *Acet. and Prop, refer to acetic acid and propionic acid, respectively. Run 5 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Anoxic Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Oct. 20 13.5 7.9 0.0 0.0 1.2 0.0 0.0 0.0 1.1 0.0 1.0 0.0 Oct. 22 11.5 5.6 1.2 0.0 1.1 0.0 1.0 0.0 1.4 0.0 0.0 0.0 Oct. 24 6.0 4.2 1.2 0.0 0.9 0.0 1.0 0.0 0.0 0.0 0.9 0.0 Oct. 26 19.8 6.2 4.8 0.0 5.2 0.0 4.9 0.0 7.3 0.0 5.8 0.0 Oct. 30 17.4 5.1 5.9 0.0 5.8 0.0 7.7 0.0 6.7 0.0 5.3 0.0 Nov. 1 18.5 5.1 4.0 0.0 4.9 0.0 7.0 0.0 5.7 0.0 4.6 0.0 Nov. 3 12.0 2.1 2.3 0.0 2.5 0.0 7.2 4.2 6.8 0.0 4.4 0.0 Nov. 5 16.8 4.5 3.6 0.0 4.3 0.0 7.6 4.0 4.8 0.0 3.5 0.0 Nov. 7 14.6 4.4 4.4 0.0 4.5 0.0 3.1 0.0 2.5 0.0 2.1 0.0 Average 14.5 5.0 3.0 0.0 3.4 0.0 4.4 0.9 4.0 0.0 3.1 0.0 STDEV 4.3 1.6 2.0 0.0 2.0 0.0 3.2 1.8 2.8 0.0 2.1 0.0 Median 14.6 5.1 3.6 0.0 4.3 0.0 4.9 0.0 4.8 0.0 3.5 0.0 Run 6 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Anoxic Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Nov. 16 18.2 7.2 6.6 0.0 1.8 0.0 2.0 0.0 1.8 0.0 1.6 0.0 Nov. 18 8.3 2.2 1.0 0.0 1.9 0.0 1.5 0.0 2.2 0.0 1.6 0.0 Nov. 20 14.1 4.8 1.3 0.0 1.2 0.0 1.3 0.0 2.1 0.0 1.4 0.0 Nov. 22 13.8 4.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Nov. 24 11.4 4.2 12.8 4.7 2.1 0.0 4.3 0.0 2.7 0.0 0.0 0.0 Nov. 26 10.9 2.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Nov. 27 11.5 3.7 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Average 12.6 4.2 3.3 0.7 1.0 0.0 1.3 0.0 1.3 0.0 0.7 0.0 STDEV 3.2 1.6 4.7 1.8 1.0 0.0 1.6 0.0 1.2 0.0 0.8 0.0 Median 11.5 4.2 1.3 0.0 1.2 0.0 1.3 0.0 1.8 0.0 0.0 0.0 278 Run 7 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Anoxic Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Dec. 6 14.7 4.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Dec. 8 18.8 7.6 5.1 0.0 1.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Dec. 10 13.9 1.9 1.4 0.0 3.1 0.0 8.0 0.0 6.7 0.0 0.0 0.0 Dec. 12 15.2 3.3 1.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Dec. 14 13.4 4.1 0.0 0.0 2.7 0.0 3.2 0.0 6.7 0.0 0.0 0.0 Dec. 14 18.8 4.2 8.6 0.0 2.5 0.0 2.9 0.0 0.0 0.0 0.0 0.0 Dec. 17 14.1 0.0 0.0 0.0 0.0 0.0 5.2 0.0 0.0 0.0 0.0 0.0 Dec. 17 15.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Dec. 19 14.0 0.0 7.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.4 0.0 Dec. 20 21.2 6.5 0.0 0.0 0.0 0.0 0.0 0.0 na na 0.0 0.0 Dec. 21 14.7 3.2 0.0 0.0 0.0 0.0 0.0 0.0 na na 0.0 0.0 Average 15.3 2.9 2.7 0.0 1.1 0.0 2.2 0.0 1.5 0.0 0.4 0.0 STDEV 2.0 2.6 3.5 0.0 1.4 0.0 2.9 0.0 3.0 0.0 1.1 0.0 Median 14.7 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Run 8 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C An( )xic Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Jan. 4 na na 0.0 0.0 na na 19.3 0.0 15.0 0.0 na na Jan. 7 22.0 4.1 0.0 0.0 0.0 0.0 8.7 0.0 0.0 0.0 0.0 0.0 Jan. 10 28.2 8.1 0.0 0.0 0.0 0.0 5.6 0.0 0.0 0.0 0.0 0.0 Jan. 13 16.3 3.2 0.0 0.0 0.0 0.0 6.5 0.0 0.0 0.0 0.0 0.0 Jan. 15 23.3 6.3 0.0 0.0 0.0 0.0 9.0 0.0 0.0 0.0 0.0 0.0 Jan. 17 22.1 7.7 • 0.0 0.0 0.0 0.0 9.7 0.0 0.0 0.0 0.0 0.0 Jan. 19 21.4 8.3 0.0 0.0 0.0 0.0 11.1 0.0 0.0 0.0 0.0 0.0 Jan. 24 10.5 2.8 0.0 0.0 0.0 0.0 11.4 0.0 0.0 0.0 0.0 0.0 Jan. 26 7.4 0.0 0.0 0.0 2.2 0.0 15.7 0.0 4.9 0.0 0.0 0.0 Average 18.9 5.1 0.0 0.0 0.3 0.0 10.8 0.0 2.2 0.0 0.0 0.0 STDEV 7.0 3.0 0.0 0.0 0.8 0.0 4.3 0.0 5.0 0.0 0.0 0.0 Median 21.7 5.2 0.0 0.0 0.0 0.0 9.7 0.0 0.0 0.0 0.0 0.0 Run 9 (mg/L) Date Raw Sewage Influent Acet. Prop. Acet. Prop. Mar. 5 10.8 5.7 0.0 0.0 Mar. 6 7.2 0.0 0.0 0.0 Mar. 7 13.7 6.2 0.0 0.0 Mar. 12 12.1 4.8 0.0 0.0 Mar. 13 10.8 4.3 0.0 0.0 Mar. 14 13.4 5.5 6.3 0.0 Mar. 15 17.6 5.9 0.0 0.0 Mar. 17 23.9 8.3 0.0 0.0 Average 13.7 5.1 0.8 0.0 STDEV 5.1 2.4 2.2 0.0 Median 12.8 5.6 0.0 0.0 279 Run 10 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Apr. 23 19.3 6.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Apr. 24 14.6 4.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Apr. 25 31.5 12.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Apr. 26 32.5 8.5 0.0 0.0 2.8 0.0 2.5 0.0 0.0 0.0 Apr. 29 12.9 4.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Apr. 30 12.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Average 20.5 6.0 0.0 0.0 0.5 0.0 0.4 0.0 0.0 0.0 STDEV 9.2 4.3 0.0 0.0 1.2 0.0 1.0 0.0 0.0 0.0 Median 17.0 5.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Run 11 (mg/L) Date Raw Sewage Influent 12 (mg/L) Acet. Prop. Acet. Prop. May. 15 18.5 5.7 0.0 0.0 May. 17 17.0 4.2 0.0 0.0 May. 19 27.2 8.4 0.0 0.0 May. 21 29.1 8.4 3.1 0.0 May. 23 18.3 4.8 0.0 0.0 May. 25 23.0 5.6 0.0 0.0 May. 27 30.1 5.1 0.0 0.0 May: 29 30.5 7.0 7.1 0.0 Average 24.2 6.1 1.3 0.0 STDEV 5.7 1.6 2.6 0.0 Median 25.1 5.6 0.0 0.0 Run ) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Jun. 12 21.6 2.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jun. 13 19.2 5.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jun. 14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jun. 16 17.2 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jun. 17 19.1 5.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jun. 18 29.8 7.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Average 17.8 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 STDEV 9.8 2.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Median 19.2 4.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Run 13 (mg/L) Date Raw Sewage Influent Anaerobic A Anaerobic B Anaerobic C Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Acet. Prop. Jul. 10 19.0 3.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jul. 11 25.9 8.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jul. 12 8.6 3.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jul. 13 16.9 4.3 0.0 0.0 2.8 0.0 2.5 0.0 0.0 0.0 Jul. 14 19.2 5.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Jul. 15 18.3 3.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Average 18.0 4.9 0.0 0.0 0.5 0.0 0.4 0.0 0.0 0.0 STDEV 5.6 2.2 0.0 0.0 1.2 0.0 1.0 0.0 0.0 0.0 Median 18.7 4.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 280 s s OS a .2 o a I D r- NO > a rs r-H 00 eu OJD ett l« f i rs O > o o oo NO 3 ^> ON 00 o r ~ NO 00 ON • n >r> r-~ Jul IT) r- r~ CN v-> m ) 3 r~ NO o NO 3 i n r - NO <N •«t 00 ci i n NO NO 3 i—> CN o ON o C >n NO NO 3 i—> ON "* m •* a NO NO 3 r—t f> c • n NO NO 3 eu * « * 1 Da < 09 s eu i-a. fi U pa M)| £ e s > w Q H 1/3 I-eU IT; C/5 -J B m e 3 eu > w o H i/) eu i _ eu > NO GO NO pa © H -OH] -e s PS c s c s GO NO GO «3N C S 90 e> rs c s NO CN co 00 c s ON r -c s o NO CS CN CN o ro CN 00 CN II •o eu "> w o H GO eu l l t/5 O40| e Irs O rs 00 rs i<3 ir; o rs £1 iS I o ro o o oo CN O o INO | c N IO 131 00 00 I PI ON CN ro ON CO CM IS 181 lo NO ro o CN CN CJ t/5 M 3 vo -M | E s vo t/3 -M e s CZ3 -,W>| S s ae PS -M | a Median CJ 90 CN CJ CN o CJ CN STDEV V i CN OS OS Average 00 r~ 1-H CJ CN 1/5 CN CN Jun. 1 © Os i r i CN CN OS CN CN May. 31 CN o o . — i r~ co CN CO CN Median 1/3 O CN CJ O CN CO -* CN Median CN CN CN © 00 OS CN CN May. 28 o CN i—1 co CN •<* CN STDEV SO i—1 O CJ TT lH STDEV Os SO »-H Os ; MLSS) May. 25 SO «—< CN LIZ co CN Average •* O CN CN OS SO -* CN Average CJ CN CN CS 90 1"H CJ <N 1 (mL/j May. 22 00 C N CN 00 CO CN Jun. 17 CN O CN Os OS OS CO CN Jul. 15 SO CN CN t -Os OS CN CN Run 1 CN 1 co SO CN o CN ! MLSS; Jun. 16 CO O CN SO r» CO CO CN ;MLSS; Jul. 14 OS C N ©' 00 t—i m C N May. 20 -* 00 CN LIZ rS Jun. 14 >/-> CN CN SO C N r-•* CN 9*1 £ Jul. 13 V) CN 00 CN 00 C N CN OS CN co CN CN OS Run 1 Jun. 13 SO O CN SO CN OS CN Run 1 C N CN CO CN r - OS O CN 00 CO s sO oo co co CN CO CN Jun. 12 V) t> r-O CN 00 CO CN "3 O CN OS r -CN SO CN May. 171 OS co CN CN CN CN Jun. 7 CN CN 00 CO r> CN Jul. 8 -* CO CN 00 CN C N 1 Date 1 < 03 1 Date < ea u 1 Date < pa 

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.831.1-0050308/manifest

Comment

Related Items