Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effects of varying environmental conditions on performance of completely autotrophic nitrogen removal.. Wu, Chumeng 2012

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

Item Metadata

Download

Media
ubc_2012_fall_wu_chumeng.pdf [ 14.92MB ]
Metadata
JSON: 1.0050794.json
JSON-LD: 1.0050794+ld.json
RDF/XML (Pretty): 1.0050794.xml
RDF/JSON: 1.0050794+rdf.json
Turtle: 1.0050794+rdf-turtle.txt
N-Triples: 1.0050794+rdf-ntriples.txt
Original Record: 1.0050794 +original-record.json
Full Text
1.0050794.txt
Citation
1.0050794.ris

Full Text

EFFECTS OF VARYING ENVIRONMENTAL CONDITIONS ON PERFORMANCE OF COMPLETELY AUTOTROPHIC NITROGEN REMOVAL FROM DEWATERED SLUDGE LIQUORS      by   Chumeng Wu  B.Eng., South China University of Technology, 2010   A THESIS SUBMITED IN PARTIAL FULLFILMENT OF  THE REQUREMENTS FOR THE DEGREE OF   MASTER OF APPLIED SCIENCE  in  THE FACULTY OF GRADUATE STUDIES  (Civil Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)   September 2012  ©Chumeng Wu, 2012   ii Abstract Centrate, an ammonia rich wastewater, usually contains high concentrations of ammonia (around 1000 mg/L) and accounts for about 20 % of total nitrogen loading to wastewater treatment plants. However, the conventional nitrification/denitrification approach for centrate treatment is costly to operate and releases a large quantity of carbon dioxide, due to the addition of organic compounds for denitrification. A brand new approach for nitrogen management is a novel process called anaerobic ammonium oxidation (ANAMMOX). ANAMMOX offers a cost effective solution for centrate treatment. Compared to the conventional biological nitrogen removal process, ANAMMOX can save 63 % oxygen and 100 % exogenous carbon source usage.   In this research, a 400 L pilot-scale CANON SBR was studied at the J.A.M.E.S. WWTP in Abbotsford, B.C. and tested for treating on-site centrate. During 317 days of reactor running period, maximum ammonia removal rate (ARR) of 98.9 % was achieved with an average ARR of 91.0 %. The reactor also achieved a maximum of 0.81 kgN/m3•d ammonia loading rate (ALR) with a minimum hydraulic retention time (HRT) of 1.46 d. The performance of the CANON SBR was highly influenced by environmental conditions. Better reactor performance was achieved at the highest tested temperature of 32 ºC and maximum air flow rate of 14 L/min. Testing under 26 ºC would reduce ARR by 12% and generate high nitrite concentration peaks. Compared to an airflow rate of 10 L/min, the ARR was increased by 9% and HRT was reduced by 10%, at 14 L/min. Higher initial ammonia concentrations in the reactor could also lead to better reactor performance. A maximum nitrogen removal rate of 0.36 kgN/m3•d and minimum HRT of 1.79 d was achieved at the highest initial ammonia concentration of 455 mg/L. The feeding rate did not affect the results of nitrogen removal. However, slow feeding with 500 mL/min did not create any nitrite concentration peak during the reaction. Solids concentrations in the reactor could be effectively controlled by sludge settling time. Sludge settling times between 4 to 10 min were recommended, for achieving purposes of sludge enrichment or system optimization.    iii Table of Contents Abstract ......................................................................................................................................... ii Table of Contents ......................................................................................................................... iii List of Tables ................................................................................................................................. v List of Figures .............................................................................................................................. vi Nomenclature ............................................................................................................................... ix Acknowledgements ....................................................................................................................... x Dedication ..................................................................................................................................... xi 1. Introduction .............................................................................................................................. 1 1.1 Background ........................................................................................................................... 1 1.2 Objectives of the study ......................................................................................................... 3 1.3 Literature review ................................................................................................................... 3 1.3.1 Conventional nitrogen removal process ........................................................................ 3 1.3.2 Principal of ANAMMOX process ................................................................................. 6 1.3.3 Anammox bacteria ......................................................................................................... 8 1.3.4 Metabolism of the ANAMMOX process .................................................................... 10 1.3.5 ANAMMOX involved process .................................................................................... 12 1.3.6 Start up of an ANAMMOX reactor ............................................................................. 16 1.3.7 Operational concerns of ANAMMOX reactor ............................................................ 21 1.3.8 Applications of ANAMMOX ...................................................................................... 23 2. Materials and Methods .......................................................................................................... 25 2.1 Description of the sludge treatment facility ........................................................................ 25 2.2 Influent media ..................................................................................................................... 32 2.3 Experimental set-up ............................................................................................................ 33 2.3.1 Pilot-scale SBR ............................................................................................................ 33 2.3.2 CANON SBR system .................................................................................................. 34 2.3.3 Control of the reactor ................................................................................................... 36 2.4 Process operations .............................................................................................................. 38 2.4.1 Inoculation and start up ............................................................................................... 38 2.4.2 Sludge enrichment ....................................................................................................... 39 iv 2.4.3 System optimization .................................................................................................... 39 2.5 Sample analysis ...................................................................................................................... 40 3. Performance of the CANON SBR ......................................................................................... 42 3.1 Start-up period .................................................................................................................... 46 3.2 Sludge enrichment period ................................................................................................... 47 3.3 System optimization period ................................................................................................ 50 3.4 Cycle test ............................................................................................................................ 52 4. Kinetics of CANON SBR ....................................................................................................... 54 4.1 Effects of temperature ........................................................................................................ 54 4.2 Effects of aeration ............................................................................................................... 59 4.3 Effects of ammonia loading ................................................................................................ 64 4.4 Effects of feeding rate ......................................................................................................... 69 4.6 Effects of sludge settling time ............................................................................................ 73 5. Summary and Conclusions……..………………………………………………………...…75 6. Recommendations for Future Research………………………………………………....…78 References .................................................................................................................................... 79 Appendices .................................................................................................................................. 85 Appendix A: Data of reactor performance ............................................................................... 85   Appendix B: Data of cycle tests for system optimiation .......................................................... 90 v List of Tables Table 1.1 Fraction of un-ionized ammonia in aqueous solution at different pH values and temperatures……………..………………………………………………………………………..4  Table 1.2 Energy produced by ammonium oxidation reactions ……………………………….…6 Table 1.3 Parameters of aerobic and anaerobic ammonia oxidation (Jetten et al., 2001)..……….7 Table 1.4 Start-up ANAMMOX reactors with different seeding sludge………………………...21 Table 1.5 Summary of studies that investigated inhibitions on ANAMMOX process………….21 Table 1.6 Summary of the applications of ANAMMOX process…………………………….....23 Table 2.1 Main characteristics of centrate feeding to the ANAMMOX reactor……………...... 32 Table 2.2 Summary of tested conditions for system optimization……………………....………40 Table 2.3 Sample preparation and analysis methods…………………………………………….41  vi List of Figures Figure 1.1 Ultrastructure of Candidatus “Brocadia anammoxidans”. A: anammoxosome; N: bacterial nucleoid (Schmidt et al., 2003)…………………………….……………………………9 Figure 1.2 A possible metabolic pathway for anaerobic ammonia oxidation (van de Graaf et al., 1997)……………………………………………………………………………………………..10 Figure 1.3 Mechanism of anaerobic ammonium oxidation in Candidatus Brocadia Anammoxidans (Jettan et al., 2001)……………………………………………………………...11 Figure 1.4 Minimum sludge retention time for ammonium and nitrite oxidizers at different temperatures (Hellinga et al., 1998)…………………………………...………...………………13 Figure 1.5 Flow diagram of SHARON-ANAMMOX process (Khin and Annachhatre, 2004)....13 Figure 1.6 A typical sequencing batch reactor set up …………………………………….….….17 Figure 1.7 Gas lift reactor set up (Depena-Mora et al., 2004)…………………………………...19 Figure 1.8 Schematic diagram of a biofilm ANAMMOX reactor (Kim et al., 2009)...………....20 Figure 2.1 Picture of J.A.M.E.S. WWTP ..................................................................................... 25 Figure 2.2 Primary grit separation tank of J.A.M.E.S. WWTP .................................................... 26 Figure 2.3 Primary sedimentation tank of J.A.M.E.S. WWTP .................................................... 26 Figure 2.4 Decant of primary sedimentation tank of J.A.M.E.S. WWTP .................................... 27 Figure 2.5 Skimming removal tanks of J.A.M.E.S. WWTP ........................................................ 27 Figure 2.6 Top view of the Trickling Biological Filter………………………………………….28 Figure 2.7 Close up of water distributor and trickling biofilter media ………………….………28 Figure 2.8 Solids contact tank…..……………………………………………………………….29 Figure 2.9 Full view of secondary clarifier..…………………………………………………….29 Figure 2.10 Close view of secondary clarifier…..……………………………………………….30 Figure 2.11 Roof of anaerobic digester..………………………………………………………...30 Figure 2.12 Centrifuge…………………………………………………………………………...31 Figure 2.13 Stabilized bio-solids after anaerobic treatment and centrifuging…………………...31 Figure 2.14 Nitrification tanks…………………………………………………………………...32 Figure 2.15 Sequencing batch reactor (SBR) for the CANON process…………………………34 Figure 2.16 Schematic diagram of CANON-SBR system …........................................................35 Figure 2.17 Control panel for the CANON-SBR reactor………………………...………….…..37  vii Figure 2.18 Four main parts of the control panel: a) feed controller; b) decant controller; c) pH controller; d) aeration and mixing controller…...………………………………………………..38 Figure 3.1 Variation of influent and effluent ammonia-nitrogen concentration and ammonia- nitrogen removal rate (ARR) during reactor running period…………………………………….42 Figure 3.2 Concentration variations of the influent and effluent nitrite-nitrogen and nitrate- nitrogen during reactor running period…………………………………………………………..43 Figure 3.3 Variation of the TSS and VSS of influent, effluent and reactor during reactor running period………………………………………………………………………................………….44 Figure 3.4 Variation of the influent and effluent pH and alkalinity during reactor running period …………………………………………………………………………………………………...45 Figure 3.5 Variation of the cycle length and hydraulic retention time (HRT) during reactor running period…………………………………………………………………...……………….45 Figure 3.6 Picture of flocculent nitrifying sludge (a) and granular anammox sludge (b)…..…...48 Figure 3.7 Picture of the size of granular anammox sludge …………..…………………….…..49 Figure 3.8 Variation of ammonia-nitrogen loading rate during reactor running period…………50 Figure 3.9 Centrate with sludge layer from J.A.M.E.S. WWTP ….…………………………….51 Figure 3.10 Cycle profile for cycle test of CANON SBR under baseline conditions .................. 52 Figure 4.1 Comparison of the effects of temperature on pH variation in the reactor ...………....55 Figure 4.2 Comparison of the effects of temperature on ammonia concentration in the reactor..56 Figure 4.3 Comparison of the effects of temperature on nitrite concentration in the reactor …...67 Figure 4.4 Comparison of the effects of temperature on nitrate concentration in the reactor …..58 Figure 4.5 Comparison of the effects of temperature on the ratio of NO3-N generation to NH3-N consumption ……………………………………………………………………………………..58 Figure 4.6 Comparison of the effects of airflow rate on pH variation in the reactor..….……….60 Figure 4.7 Comparison of the effects of airflow rate on the hydraulic retention time ………….61 Figure 4.8 Comparison of the effects of airflow rate on the variation of ammonia-nitrogen.…..61 Figure 4.9 Comparison of the effects of airflow rate on the variation of nitrite-nitrogen ………63 Figure 4.10 Comparison of the effects of airflow rate on the variation of nitrate-nitrogen……..63 Figure 4.11 Comparison of the effects of ammonia concentrations on the variation of ammonia- nitrogen in the reactor within cycles…………………………………………………………..…64 Figure 4.12 Comparison of the effects of ammonia concentrations on the variation of pH in the reactor within cycles …………………………………………………………………………….66  viii Figure 4.13 Comparison of the effects of ammonia concentrations on the HRT, ALR and NRR …………………………………………………………………………………………………...66 Figure 4.14 Comparison of the effects of ammonia concentrations on the variation of nitrite- nitrogen in the reactor within cycles ……………………………………..……………………...67 Figure 4.15 Comparison of the effects of ammonia concentrations on the variation of nitrate- nitrogen in the reactor within cycles ……………………………………………..……………...68 Figure 4.16 Comparison of the effects of feeding rate on the variation of pH in the reactor …...70 Figure 4.17 Comparison of the effects of feeding rate on the variation of ammonia-nitrogen concentrations in the reactor within cycles …………………………………………………..….70 Figure 4.18 Comparison of the effects of feeding rate on the variation of nitrite-nitrogen concentrations in the reactor within cycles ……………………………………………………...71 Figure 4.19 Comparison of the effects of feeding rate on the variation of nitrate-nitrogen concentrations in the reactor within cycles ……………………………………………….……..72 Figure 4.20 Comparison of the effects of sludge settling time on TSS and VSS in the effluent..73 Figure 4.21 Comparison of the effects of sludge settling time on TSS and VSS in the reactor...74          ix Nomenclature Adenosine Triphosphate                                                                                                            ATP Ammonia-nitrogen                                                                                                                 NH3-N Ammonia Loading Rate                                                                                                             ALR Ammonium Oxidation Bacteria                                                                                                AOB Ammonia Removal Rate                                                                                                           ARR Anaerobic Ammonia Oxidation                                                                                   ANAMMOX Biochemical Oxygen Demand                                                                                                  BOD Complete Autotrophic Nitrogen Removal Over Nitrite                                                      CANON Chemical Oxygen Demand                                                                                                       COD Continuously Stirred Tank Reactor                                                                                         CSTR Dissolved Oxygen                                                                                                                        DO Electronic Microscopic                                                                                                                EM Fluorescence On-site Hybridization                                                                                          FISH Hydraulic Retention Time                                                                                                         HRT Initial Ammonia-nitrogen Concentration                                                                              NH3-Ni Membrane Aerated Biofilm Reactor                                                                                      MABR Nitrite-nitrogen                                                                                                                      NO2-N Nitrate-nitrogen                                                                                                                      NO3-N Nitrite Oxidation Bacteria                                                                                                         NOB Nitrogen Removal Rate                                                                                                             NRR Polymerase Chain Reaction                                                                                                       PCR Sequencing Batch Reactor                                                                                                         SBR Sludge Retention Time                                                                                                               SRT Single Reactor System for High Activity Ammonium Removal Over Nitrite                 SHARON Total Suspended Solids                                                                                                               TSS Upflow Anaerobic Sludge Blanket                                                                                         UASB Upflow Biofilm Reactor                                                                                                            UBF Volatile Suspended Solids                                                                                                          VSS Waste Water Treatment Plant                                                                                                WWTP  x Acknowledgements Here by, I would like to thank my research supervisor, Prof. Donald Mavinic, Department of Civil Engineering for his supervision, guidance and financial support to my research project and thesis writing. Also, I would appreciate Dr. Babak Rezania, for supplying the process design, his co-supervision, guidance and support through out my research project and thesis preparation work.   I would like to thank the City of Abbotsford for supplying the pilot trailer used in this research study. I would also appreciate Ron Bernier, manager of J.A.M.E.S. WWTP and other related staff for their kind support to this research project during my stay at the J.A.M.E.S. WWTP.   My thanks also go to Fred Koch, Manager of Environmental Engineering pilot plant, Department of Civil Engineering for his support during reactor assembly at the pilot plant. Also, I thank Paula Parkinson and Tim Ma, Environmental Engineering Laboratory, Department of Civil Engineering, for their kind support for my laboratory analysis work.   I would also like to thank my colleagues of Department of Civil Engineering: Parssa Hassan, Ian Chan, Hui zhang, Chaoyang Yue, Xiaoyu Zheng for their supporting and encouragement during my study and research time in the Department of Civil Engineering. Also, I thank my family and friends I love for their endless support and encouragement.  xi Dedication    To my parents     1 1. Introduction 1.1 Background  In the natural environment, nitrogen exists in the form of protein, ammonia-nitrogen (NH4, NH3), nitrite-nitrogen (NO2 -), nitrate-nitrogen (NO3 -), and nitrogen gas (N2). In fresh water, nitrogen is primarily in organic form like proteinaceous matter and urea. Decomposition of organic nitrogen by microorganisms changes the nitrogen to ammonia. With the presence of oxygen, ammonia can be oxidized to nitrites and nitrates, which can be used by plants and animals for protein generation. However, the death and decomposition of plants and animals returns ammonia back to the environment for a nitrogen balance in the environment.  In secondary wastewater treatment, nitrogen is always present as ammonia-nitrogen, nitrite- nitrogen, and nitrate-nitrogen. Since nitrogen and phosphorus are both fundamental elements for the growth of microorganism, part of nitrogen in wastewater is removed through secondary wastewater treatment processes. The ideal nutrient balance for activated sludge treatment process is BOD : N : P = 100 : 5 : 1. For a typical wastewater with BOD5 = 150 mg/L and 30% BOD removal in primary treatment process, only 5 ~ 6 mg/L nitrogen could be theoretically removed through a secondary treatment process. However, sewage usually contains about 40 mg/L NH4-N and 20 mg/L organic nitrogen, which is excessive to microorganisms in the activated sludge of secondary treatment processes. Insufficient removal of nitrogen from wastewater, before discharging, would lead to adverse effects to aquatic systems. Excessive amount of soluble ammonia is toxic to aquatic lives and also leads to eutrophication in water bodies (Robert and Russo, 1981).   Ammonium can be removed from wastewater through physical, chemical and biological processes. Among them, biological nitrogen removal processes are the most cost-effective technologies and have been widely adopted in wastewater treatment facilities globally.  One typical biological nitrogen removal process is known as biological nitrification and denitrification process, in which nitrifying bacteria oxidize ammonia to nitrite and nitrate under aerobic conditions, and then denitrifying bacteria convert nitrite/nitrate to nitrogen gas under anoxic conditions (Sliekers, et al., 2002). Denitrification bacteria are known as heterotrophic  2 microorganisms and require biodegradable organic compounds (COD) as a carbon source for their reproduction. However, wastewaters commonly contain insufficient COD to achieve complete denitrification. Thus, additional organic compounds, such as methanol, are required to add as an exogenous carbon-source during conventional biological denitrification processes. In addition, conventional biological nitrogen removal process is not suitable for some specific wastewaters such as landfill leachate and dewatered sludge liquids with high concentrations of ammonia and low BOD/COD ratio. The wastewater generated during digested sludge dewatering normally contributes to 15–20 % of total nitrogen loading to wastewater treatment plants; proper treatment of dewatered sludge liquids would therefore, reduce nitrogen loading to most wastewater treatment facilities.   To overcome the limitations of conventional biological nitrification and denitrification processes, several novel nitrogen removal processes have been developed, including the ANAMMOX process, SHARON process, CANON process and OLAND process (Hellinga et al., 1998; Kuai and Verstraete, 1998; Third et al., 2001; Vlaeminck et al., 2009).  Among them, OLAND and CANON processes use anammox in a single bioreactor. Reviewing different anammox systems described in the literature, it became clear that:  • ANAMMOX based processes are still under investigation in lab-scale devices, and there is limited experience with pilot-scale reactors. In this research, performance of the CANON process was tested in a pilot-scale sequencing batch reactor (SBR). Throughout the thesis, we will use the CANON SBR to describe the one stage ANAMMOX in a single bioreactor. • The ANAMMOX research is new to Canada and there is very limited research on dealing the local centrate with the ANAMMOX process. • Most of the experimental knowledge on autotrophic nitrogen removal is restricted to lab scale reactors using synthetic feed. • There is limited research on studying the effects of environmental conditions on the CANON process       3 1.2 Objectives of the study The objectives of this study were:  1) Perform a field study for dealing with on-site dewatered sludge liquor with high ammonia concentration on-site centrate, using a CANON SBR system, and evaluate the performance of the pilot scale CANON SBR.  2) Study the effects of variable environmental conditions including temperature, aeration, initial ammonia concentration, feeding speed and sludge settling time on the performance and microbial kinetics of CANON SBR.    1.3 Literature review 1.3.1 Conventional nitrogen removal process In order to removal nitrogen from waste water, a variety of physical, chemical and biological methods have been developed, including air stripping, ion exchange, and biological nitrification and denitrification process. 1.3.1.1 Air Stripping Air stripping is a controlling process based on the principle of ion equilibrium with ammonia and hydrogen ions exist in wastewater:   NH4 + NH3 + H +                                                                                                                          [1]  The equilibrium moves towards the right side when pH is increased above 7.0. The higher the pH level and temperature, the more ammonium is converted to gas form ammonia. According to the calculation of Emerson et al. (1975) in Table 1.1, the fraction of ammonia is about 85% in aqueous ammonium solution of pH 10 at temperature of 25 ºC. Bubbling through wastewater can strip part of the dissolved ammonia gas from wastewater to some degree, which could be done in an air-stripping tower.    4 However, air stripping is considered as economically unfavorable. Alfred and Culp (1967) reported a cost estimation of ammonia air stripping in a Tahoe water reclamation plant. The cost to achieve 95% ammonia removal was $13.85/Mgal and $25/Mgal for 98% ammonia removal. This cost did not include the additional cost of cleaning and maintaining the stripping tower. Due to the high pH, scales will form rapidly on the packed media within the tower and needs to be washed frequently by acid (Reeves, 1972).  Table 1.1 Fraction of un-ionized ammonia in aqueous solution at different pH values and temperatures pH 8 10 12 14 16 18 20 22 24 26 28 30 32               7.0 .0016 .0018 .0022 .0025 .0029 .0034 .0039 .0046 .0052 .0060 .0069 .0080 .0093 7.2 .0025 .0029 .0034 .0040 .0046 .0054 .0062 .0072 .0083 .0096 .0110 .126 .150 7.4 .0040 .0046 .0054 .0063 .0073 .0085 .0098 .0114 .0131 .0150 .0173 .0198 .0236 7.6 .0063 .0073 .0086 .0100 .0116 .0134 .0155 .0179 .0206 .0230 .0271 .0198 .0236 7.8 .0099 .0116 .0135 .0157 .0182 .0211 .0244 .0281 .0322 .0370 .0432 .0482 .0572 8.0 .0156 .0182 .0212 .0247 .0286 .0330 .0381 .0438 .0502 .0574 .0654 .0743 .0877 8.2 .0245 .0286 .0332 .0385 .0445 .0514 .0590 .0676 .0772 .0884 .0998 .1129 .1322 8.4 .0383 .0445 .0517 .0597 .0688 .0790 .0904 .1031 .1171 .1326 .1492 .1678 .1948 8.6 .0593 .0688 .0795 .0914 .1048 .1197 .1361 .1541 .1737 .1950 .2178 .2422 .2767 8.8 .0909 .1048 .1204 .1376 .1566 .1773 .1998 .2241 .2500 .2774 .3062 .3362 .3776 9.0 .1368 .1565 .1782 .2018 .2273 .2546 .2846 .3140 .3456 .3783 .4116 .4453 .4902 9.2 .2008 .2273 .2558 .2861 .3180 .3512 .3855 .4204 .4557 .4909 .5258 .5599 .6038 9.4 .2847 .3180 .3526 .3884 .4249 .4618 .4985 .5348 .5702 .6045 .6373 .6685 .7072 9.6 .3868 .4249 .4633 .5016 .5394 .5762 .6117 .6456 .6777 .7073 .7358 .7617 .7929 9.8 .5000 .5394 .5778 .6147 .6499 .6831 .7140 .7248 .7692 .7933 .8153 .8351 .8585 10.0 .6131 .6498 .6844 .7166 .7463 .7735 .7983 .8207 .8408 .8588 .8749 .8892 .9058 Adopted from Emerson et al., 1975  1.3.1.2 Ion exchange Ion exchange is a chemical process that the specified ions in wastewater are displaced by other ions from an insoluble exchange material based on ion exchange capacity of the material. Acid or cationic resins can exchange cation such as Ca2+ and NH4 +, while base or anionic resins can exchange anions like OH- or NO3 - (Reeves, 1972). The selectivity order is Ca2+ < K+ < NH4 + <  5 Na+ and SO4 2- < HPO4 2- < NO3 - < Cl-. Thus, NH4 + in wastewater can be exchanged by Na+ ion and NO3 - can be exchanged by Cl- ion, in theory.  1.3.1.3 Biological nitrification and denitrification Biological nitrification and denitrification process is the most widely used method for nitrogen removal from wastewater. The process includes two separate steps of bacterial activities: first oxidize ammonia to nitrates (i.e. nitrification) and then convert the nitrates to nitrogen gas (i.e. denitrification).  Nitrification, which converts ammonia to nitrates, is also a two steps biological process. Ammonium and ammonia in raw wastewater are converted to nitrite by bacteria called Nitrosomnas. Then, another type of bacteria, Nitrobacter, converts the nitrite to nitrate (Johnson and Schroepfer, 1964). The overall reactions of the two steps are shown in Equation 2 and 3.  NH4 + + 1.5O2 = 2H + + H2O + NO2 -                                                                                              [2]  NO2 - + 0.5O2 = NO3 -                                                                                                                       [3]  Both Nitrosomnas and Nitrobacter, know as “nitrifiers”, are strict “aerobes”. This means they need DO to grow and maintain nitrification under aerobic condition at dissolved oxygen (DO) levels of 1.0 mg/L or above (below DO level of 0.5 mg/L, the growth rate is minimal). It can be calculated from the above Equation 2 and 3, that for completely oxidizing 1 g of ammonia to nitrate, nitrifiers need to consume 4.18 g of oxygen. Nitrification is an alkalinity consuming process; it can be calculated that 7.14 g of alkalinity as calcium carbonate (CaCO3) need to be consumed for oxidizing 1 g of ammonia to nitrite.  In comparison to nitrification, biological denitrification is an anaerobic (anoxic) process that converts the nitrite and nitrate generated by the nitrification process to nitrogen gas (N2), through several facultative heterotrophic bacteria. These bacteria include species of Pseudomonas, Achromobacter, Bacillus, and Micrococcus (Reeves, 1972). The reactions of biological denitrification processes are presented in Equation 4 and 5.    6 2NO3 - + 10H+ + 10e- = N2 +2OH - +4H2O                                                                                 [4]  2NO2 - + 6H+ + 6e- = N2 + 2OH - + 2H2O                                                                                   [5]   These heterotrophic bacteria need electron donors to provide electrons for carrying out denitrification reaction. However, many wastewaters do not contain enough organic compounds to act as the electron donors for the denitrification process. Quality electron donors such as methanol, acetate, ethanol, lactate and glucose should be added for complete denitrification (Khin and Annachhatre, 2004). Among them, methanol is relatively cheap and is commonly used in denitrification processes.   Denitrification can occur under anoxic conditions with the DO level less than 0.5 mg/L, ideally less than 0.2 mg/L (Vit Mateju, 1992). Denitrification is an alkalinity production process. From Equation 4, it can be calculated that approximately 3.0 to 3.6 g of alkalinity as CaCO3 is produced per gram of nitrite being converted to nitrogen gas.  1.3.2 Principal of ANAMMOX process The history of ANAMMOX started in 1977, when E. Broda (1977) published his hypothesis of two kinds of lithotrophs missing in nature based on thermodynamically calculations. Ammonium oxidation reactions (as shown in Table 1.2), that use nitrite as electron acceptor, should be able to occur automatically since their Gibbs Free Energies (G0) are negative. This hypothesis was not proved until Mulder et al. (1995) observed NH4 + disappearing with a rate of 0.4 kg•N/m3•d from a denitrifying fluidized bed reactor treating the effluent of a methanogenic reactor. Both nitrate and ammonium consumption in the reactor increased with denitrogen gas production. They gave the name “ANAMMOX” to this novel biological process.  Table 1.2 Energy produced by ammonium oxidation reactions  Electron acceptor Chemical reaction G0 Possibility O2 2NH4 ++3O2=2NO2 -+2H2O+4H + -241 Possible NO2 - NH4 ++NO2 -=N2+2H2O -355 Possible NO3 - NH4 ++3NO3 -=4N2+9H2O+2H + -278 Possible Fe3 + NH4 ++6Fe3 +=N2+6Fe2 ++8H+ -100 Possible SO4 2- 8NH4++3SO42-=4N2+3H2S+12H2O+5H + -22 Possible  7   Later, van de Graaf et al. (1995) identified that ANAMMOX is a process using nitrite as an electron acceptor. The reaction can thus be expressed as:  NH4 + + NO2 - = N2 + 2H2O                                                                                                            [6]  Strous et al. (1998) estimated the ANAMMOX stoichiometry based on mass balance principles; they found out that the ratio of converted ammonia and nitrite is 1:1.32 anoxically, as presented in Equation 7:  NH4 + +1.32NO2 - + 0.066HCO3 - + 0.13H+ = 1.02N + 0.26NO3 - + 0.066CH2O0.5N0.15 + 2.03H2O                                                                            [7]  Parameters of this novel anaerobic ammonia oxidation process are compared with the conventional aerobic nitrification process (Table 1.3). Obviously, the ANAMMOX process does not require additional aeration and organic carbon sources.   Up to 90% of operational cost can be saved in comparison to conventional biological nitrification denitrification and other typical nitrogen removal processes (Jetten et al., 2001). The advantages of ANAMMOX make it a promising technology, to replace the traditional nitrification and denitrification technologies for nitrogen removal.   Table 1.3 Parameters of aerobic and anaerobic ammonia oxidation (Jetten et al., 2001) Parameter Nitrification ANAMMOX Unit Free energy -275 -357 kJ/mol Biomass yield 0.08 0.07 mol/mol C Aerobic rate 200-600 0 nmol/min/mg protein Anaerobic rate 2 60 nmol/min/mg protein Growth rate 0.04 0.003 /h Doubling time 0.73 10.6 Days    8 1.3.3 Anammox bacteria 1.3.3.1 Characteristics of anammox bacteria The so-called anammox bacteria that involved in the ANAMMOX process were found to be planctomycete-like. The low 16S RNA sequence of anammox organisms present over 90% similarities to other genera of the Planctomycetes such as gemmate, Isosphaera, or Planctomyces, indicating that the anammox should be in a second order within the Planctomycetes (Schmid et al., 2005).  Three typical types of anammox organisms were later identified by their 16rRNA sequences. Candidatus “Brocadia anammoxidans” was identified by Jetten et al. (2001). Two other species: Candidatus “Kuenenia Stuttgartiensis” and Candidatus “Scalindua sorokinii” were also found in wastewater treatment plants and marine sediment (Penton et al., 2006). A typical ultrastructure profile of the anammox bacteria Candidatus “Brocadia anammoxidans” is showed in Figure 1.1. The anammox bacteria are red coccoid bacteria with a diameter of less than 1 µm (Niftrik et al., 2004). The anammox bacteria are slow growing microorganism, with average doubling time of 11 days. Like other Planctomycetes, these anammox bacteria have a proteinaceous cell wall and are lacking in peptidoglycans. But they are distinct for autotrophic and anoxic. Electronic microscopic (EM) studies indicated that all anammox bacteria have one separate, ribosome-free cytoplasm intracellular compartment, termed as anammoxosome (Lindsay et al., 2001). The anammoxosome is bounded by a laddered lipids structure membrane, which is unique in nature and was only found in anammox bacteria (Niftrik et al., 2004). The bacterial nucleoid is located on the outside of the anammoxosome membrane. The anammoxosome can provide a site for catabolism, generating proton motive force for adenosine triphosphate (ATP), as well as protecting the rest of cell from the proton diffusion and toxic intermediates of the ANAMMOX process (Lindsay et al., 2001; Niftrik et al., 2004).  9  Figure 1.1 Ultrastructure of Candidatus “Brocadia anammoxidans”. A: anammoxsome; N: bacterial nucleoid (Schmidt et al., 2003)  1.3.3.2 Detection of anammox bacteria The detection methods of anammox bacteria include tracer experiments, PCR amplification and FISH analysis.  Tracer experiment with 15N-labeled ammonium and nitrite is the most basic method to indicate the presence and activity of anammox bacteria. Through anammox reactions, labeled [15N] ammonium reacts with unlabeled [14N] nitrite, generating 29N2 ( 14N15N). This method is especially widely used for studying anammox activities and its mechanism principles (van de Graaf et al., 1997; Jettan et al., 2001)  Polymerase Chain Reaction (PCR) amplification, with general 16S rRNA gene-targeted primers, is an advanced method to detect the presence of anammox organisms if the cell counts are too low for fluorescence on-site hybridization (FISH). PCR amplification can amplify selected 16S rRNA gene fragments from mixed DNA and compare them with gene libraries to rapidly identify anammox organisms (Amann et al., 1995). However, anammox organisms are still underrepresented in general 16S rRNA gene libraries and PCR amplification is not quantitatively representative in sample investigation (Amann et al., 1995).   10  Fluorescence on-site hybridization (FISH) is becoming a standard method for the detection of anammox organisms and was described by Schmidt et al. (2005). FISH has been intensively used to provide both qualitative and quantitative results for anammox bacteria in water and wastewater samples (Sliekers et al., 2002; Pynaert et al., 2003; López et al., 2008).  1.3.4 Metabolism of the ANAMMOX process In the ANAMMOX reaction, 1 mol ammonium with 1.32 mol nitrite can be converted to 1.02 mol nitrogen gas and 0.26 mol nitrate. ANAMMOX bacteria have been found to be metabolically flexible. Several possible alternative metabolic pathways have been proposed. The first metabolic pathway of ANAMMOX was proposed by van de Graaf et al. (1997) based on 15N studies with the dominant species of Candidatus Brocadia anammoxidans (Figure 1.2) and considered hydroxylamine (NH2OH) as an critical intermediate of nitrite reduction. There are five steps involved in this metabolic pathway. Ammonium is first oxidized by hydroxylamine (NH2OH) to form hydrazine (N2H2), which is step 1. Equivalents formed by N2H2 then react with nitrite to form finial product N2 and NH2OH (step 2, 3, and 4). In the end, parts of the nitrite would convert to nitrate for biomass growth (step 5).   Figure 1.2 Possible metabolic pathway for anaerobic ammonia oxidation (van de Graaet al., 1997)  A similar mechanism theory of ANAMMOX in Candidatus Brocadia Anammoxidans (Figure 1.3) was postulated by Jettan et al. (2001) through the investigation using 15N-labelling experiments. In Figure 1.3, NR is a nitrite-reducing enzyme (NH2OH is the product); HH is  11 hydrazine hydrolase that condenses hydrazine out of ammonia and hydroxylamine; HZO is a hydrazine-oxidising enzyme, which may be equivalent to hydroxylamine oxidoreductase. The results showed that nitrite as the electron acceptor was reduced to hydroxylamine and reacted with the electron donor ammonium, generating the ultimate products as denitrogen gas.  Interestingly, none of the clones analysed was closely relat- ed to B. anammoxidans. However, several environmentally retrieved sequences formed a clearly separated mono- phyletic cluster (with about 90% 16S rDNA sequence similarity to B. anammoxidans) within the B. anammoxidans line of descent. FISH with specific probes for this cluster showed that these bacteria dominated the microbial biofilm communities of the investigated trickling filter [8•] and rotating biological contactor (M Schmid, M Wagner, unpublished data), whereas B. anammoxidans could not be detected. Consequently, these bacteria, provisionally named Candidatus Kuenenia stuttgartiensis [8•], represent a novel genus of anammox bacteria. As this genus is genetically only distantly related to B. anammoxidans, we expect that there will be physiologi- cal differences between these two anammox genera that should be characterised in future studies (see Update). Furthermore, we are currently investigating, by using a combination of FISH and microautoradiography [9], whether the anammox phenotype is restricted to these two genera or is more widespread within the bacterial domain. Mechanisms of anaerobic ammonium oxidation in Candidatus Brocadia anammoxidans The possible metabolic pathway for anaerobic ammonium oxidation was investigated using 15N-labelling experiments. These experiments showed that the electron acceptor nitrite is reduced to hydroxylamine and that hydroxylamine somehow reacts with the electron donor ammonium, lead- ing to the ultimate production of dinitrogen gas [1•]. In batch experiments with excess hydroxylamine and ammoni- um, we observed a transient accumulation of hydrazine, indicating that hydrazine is the intermediate of this final step. We postulated that the oxidation of hydrazine to dinitrogen gas generates the electrons for the initial reduction of nitrite to hydroxylamine (Figure 2). As far as we know, the occurrence of free hydrazine in microbial nitrogen metabolism is rare, if not unique [10]. Cell-free extracts of anammox cultures showed a strong absorption at 468 nm in reduced cytochrome spectra. We purified the enzyme associated with this peak [11•]. The enzyme appeared to have some similarity to the hydroxyl- amine oxidoreductase (HAO) of Nitrosomonas europaea. HAO of N. europaea has a similar peak at 460 nm. Both enzymes are capable of oxidising both hydroxylamine and hydrazine. Apart from the different absorption maxima the enzymes also differed in size, as shown by polyacrylamide gel electrophoresis, HAO from B. anammoxidans being the smaller enzyme. The amino acid sequence of several polypeptides obtained from B. anammoxidans HAO digests showed no homology to any other sequences in the databases. The two enzymes did, however, share the presence of eight c-type cytochromes per enzyme subunit (and three identical subunits per enzyme). Using polyclonal antibodies raised against the purified enzyme, HAO has been detected with immunogold labelling in cells of B. anammoxidans [12]. HAO was found to be present only inside a cytoplasmic membrane-bound- ed region, which makes up about 30–60% of the cell volume. As this ‘organelle’ seems to have an important role in the catabolism of B. anammoxidans, we have named it the ‘anammoxosome’. The anammoxosome is completely surrounded by a compartment containing the nucleoid and ribosomes, and so resembles structurally the pirellulosome in other plancto- mycetes [12]. The pirellulosome is a major internal cell compartment containing the nucleoid and surrounded by a single membrane. It was first described in chemo- heterotrophic planctomycetes in the genus Pirellula, but later found to be a structural feature shared by all planctomycetes examined. Brocadia cells also have an outer compartment containing cytoplasm termed the ‘paryphoplasm’, which is separated from more interior compartments of the cell by an intracytoplasmic membrane. They share this feature  with other planctomycetes [12]. The exact function of the compartments in B. anammoxidans is currently under investigation. One possible function would relate to membrane potential generation internally across the anammoxosome membrane during ammonium oxi- dation. Some features of the compartmentalisation of the cell in B. anammoxidans are exactly analogous to the organisation of compartments in cultured heterotrophic planctomycete genera, such as Gemmata obscuriglobus, Planctomyces maris, Isosphaera pallida, Pirellula marina and Pi. staleyi. These features include the nucleoid- and ribosome-containing compartment that is equivalent to the pirellulosome, the outer paryphoplasm compartment, and the intracytoplasmic membrane that separates the The anaerobic ammonium oxidation (‘anammox’) process Jetten et al. 285 Figure 2 Mechanism of anaerobic ammonium oxidation. NR is a nitrite-reducing enzyme (N H2 O H is the assumed product); H H (hydrazine hydrolase) condenses hydrazine out of ammonia and hydroxylamine; H Z O is a hydrazine-oxidising enzyme (which might be equivalent to hydroxylamine oxidoreductase). 4e– N2NH3 NO2– 4H+ 5H+ HZO NR HH Cytoplasm Current Opinion in Biotechnology Anammoxosome N2H4 NH2OH  Figure 1.3 Mechanism of anaerobic ammonium oxidation in Candidatus Brocadia Anammoxidans. (Jettan et al., 2001)  Another slightly different ANAMMOX metabolic mechanism, was proposed by Strous et al. (2006), through the study of another species of anammox bacteria, Candidatus Kuenenia stuttgartiensis. Nitric oxide (NO) was postulated to be the intermediate for nitrite reduction instead of hydroxylamine. They found there were two unique enzymes that involved in this ANAMMOX process. Hydrazine hydrolase (hh) is the enzyme that produces hydrazine from nitric oxide and ammonium, while hydrazine dehydrogenase (hd) is the enzyme that can transfer electrons from hydrazine to ferredoxin.  All three hypotheses agreed that hydrazine (N2H2) is an important intermediate for the ANAMMOX process, since a transient accumulation of hydrazine was observed. The enzyme which can help the bacteria to oxidize hydrazine to denitrogen gas was identified as a hydroxylamine-oxidoreductase-like protein (HAO) of (Schalk et al., 2000). HAO was found to be present only insid  anammoxosome of the anam ox bacteria cells.   12 1.3.5 ANAMMOX involved process The ANAMMOX process has been carried out with ammonium to nitrite ratio of 1:1.32 (Strous et al., 1998). For most wastewaters, however, there is not enough nitrite to complete the ANAMMOX process (Zhang et al., 2008). NO2 - should be added additionally unless ANAMMOX is combined with other supplemental process like partial nitrification, that would generate additional nitrite.  1.3.5.1 Partial Nitrification-ANAMMOX Partial nitrification is a nitrification process that partially oxidizes ammonium to nitrite, but not to nitrate (Equation 8).   NH4 + +1.5O2 NO2 - + 2H+ + H2O                                                                                              [8]  To achieve partial nitrification, conventional subsequent oxidation process of nitrite to nitrate must be prevented. Nitrite oxidation can be prevented in two ways. First, ammonium oxidation is a more temperature dependent process, since activate energies of ammonium oxidation and nitrite oxidation are different, which is 68 kJ mol-1 and 44 kJ mol-1 respectively (Schmidt, et al., 2003). The ammonium oxidation bacteria (AOB) and nitrite oxidation bacteria (NOB) are also quite sensitive to temperature, and their growth rates (or sludge age) change with temperatures (Figure 1.4). At temperatures higher than 15 °C, ammonium oxidizers grow faster than nitrite oxidizers. NOB are outcompeted by AOB at temperatures above 25 °C (Khin and Annachhatre et al., 2004). At a temperature of 35 °C, the minimum sludge age of NOB is approximately only half of that of AOB (0.5 and 1 day, respectively). When the hydraulic retention time is controlled higher than the sludge retention time of nitrite oxidizers, but lower than that of ammonia oxidizers, nitrite oxidizers would be selectively washed out and further nitrite oxidation would be prevented. The second way to prevent nitrite oxidation is to use the principle that NOB are unable to grow at low oxygen concentration (i.e. less than 0.4 mg/L or 5% air saturation) with surplus ammonium (Schmidt et al., 2003). Nitrites become the stable end product of nitrification under this condition. However, the mechanism of this NOB inhibition by ammonium and low- level dissolved oxygen is not fully understood.    13 Zhang et al. / J Zhejiang Univ Sci B   2008 9(5):416-426 418 their specificities (Shimamura et al., 2007). Cyto- chromes were important electron transporters be- tween enzymes (Huston et al., 2007). Their abun- dance was closely related to bacterial activity. Ex- periments had shown a colour alteration from khaki to brownish and red with the increment of nitrogen re- moval rate (Trigo et al., 2006).  The ANAMMOX bacteria are known to anabo- lize CO2 taking advantage of energy generated from oxidation of nitrite to nitrate (Fig.2). Both stable carbon isotopic fractionation and genome detection proved that inorganic carbon was fixed through the acetyl-CoA pathway (Schouten et al., 2004; Strous et al., 2006).                     ANAMMOX-INVOLVED PROCESSES  The ANAMMOX process was carried out with 4 2NH / NO     close to 1. Wastewaters, however, do not always fulfill this characteristic. Thus, several novel processes were developed to ensure the optimal sub- strate ratio.  Partial nitritation-ANAMMOX Partial nitritation-ANAMMOX process is based on two biotechnologies. Firstly, ammonium is partly nitrified to nitrite in the partial nitritation stage, and then the produced nitrite is denitrified with the re- sidual ammonium in the ANAMMOX. The nitritation of ammonium is conducted by aerobic ammonium- oxidizing bacteria (AOB) that differentiate greatly from nitrite-oxidizing bacteria (NOB) in physiology. Therefore, selective retention of AOB is important in the partial nitritation stage. In practice, the following five factors are considered as the priorities. 1. Temperature The two groups of bacteria are quite sensitive to temperature (Hellinga et al., 1998). Increased tem- perature facilitates AOB to outcompete NOB (Fig.3). The optimal temperatures for pure cultural AOB and NOB are 38 and 35 °C, respectively (Grunditz and Dlhammar, 2001). Hellinga et al.(1998) have ob- tained a stable nitritation process at temperature over 35 °C based on a two years’ experiment. This value is not fixed. Recently, the nitritation process was also successfully started up and maintained between 15 and 30 °C (Yamamoto et al., 2006). However, the system performance deteriorated dramatically below 15 °C, which agreed well with the theoretical value in Fig.3.                 2. Sludge residence time (SRT) At higher temperature stated above, doubling time of AOB is shorter than that of NOB. Thus, SRT should be properly mediated in a limited range that enables retention of AOB but washing out of NOB. Full scale experience in Utrecht and Rotterdam wastewater treatment plants suggested that SRT be- tween 1 and 2.5 d was acceptable (van Kempen et al., 2001). Nevertheless, SRT as long as 5 d in an SBR (sequential batch reactor) also created a favourable environment for AOB to outgrow NOB (Galí et al., 2007). Fig.2  Metabolic pathway of Candidatus Kuenenia stuttgartiensis by environmental genomics analysis Diamond 1: Cytochrome; Diamond 2: Ferredoxin; Line L1: Reduction; Line L2: Oxidation 32NO  23NO  nar nir NO 4NH  hh N2H4 +0.43 V 4e� 1e� +0.38 V 3e � +0.34 V 8H+ 8H+ 6H+ N2 Q ret ret bc1 hd �0.75 V 4e� 1 1 1 2 L1 L1 L1 L1 L1 L2 L2 L2 L2 L2 Fig.3 Minimum residence time for ammonium and nitrite oxidizers at different temperatures (Hellinga et al., 1998) 0 10 <15 20 30 35 40 T (°C) M in im um  sl ud ge  a ge  (d ) 6 4 2 0 Residence time for nitrite route 4NH -oxidizers 2NO -oxidizers  Figure 1.4 Minimum sludge retention time for ammonium and nitrite oxidizers at different temperature (Hellinga et al., 1998)   Hellinga et al. (1998) designed a partial nitrification process for treating dewatered sludge liquor with high ammonia concentration (about 1000 mg/L) and named it SHARON (single reactor system for high activity ammonium removal over nitrite). The SHARON process is operated at temperature above 30 °C. High specific growth rates can be enabled under such a high temperature, so that no sludge retention is required for the SHARON process; this means that the sludge age (SRT) equals the hydraulic retention time (HRT). In such a system, the effluent concentration is controlled directly by the bacteria growth rate (1/S T) instead of influent concentration. Figure 1.5 shows the flow di gram of a SHARON-ANAMMOX process.  Figure 1.5 Flow diagram of SHARON-ANAMMOX process (Khin and Annachhatre, 2004)   14 Effluent from a SHARON process is usually used as feed for the ANAMMOX reactor directly. To make an appreciate feed for the ANAMMOX reactor, ammonium and nitrite ratio should be maintained around 1.0 within the effluent of the SHARON process, which means only 50% of the ammonium needs to be converted to nitrite by the SHARON process (Equation 9):  NH4 + + HCO3 - + 0.75 O2  0.5 NH4 + + 0.5 NO2 - + CO2 + 1.5 H2O                                           [9]  Van Dongen et al. (2001) reported a successful 53% oxidation of ammonium to nitrite in a lab scale continuously stirred tank reactor (CSTR) feed by sludge liquor from the Rotterdam WWTP at 1.2 kgN/m3•d. The temperature in the reactor was maintained within the range of 30~40 °C and the HRT was 1.0 day.   Fux et al. (2002) set up a pilot scale (3.6 m3) SHARON process with a CSTR reactor and operated at 30 °C without sludge retention. A conversion rate of 58% ammonium to nitrite was achieved over a half of year operation, resulting in nitrite production of 0.35 kgN/m3•d. They also recommended operating the SHARON ANAMMOX reactor in a weak alkaline condition. The nitrite/ammonium ratio dropped drastically when pH increased higher than 7.2.   Van Kempen et al. (2001) reported a SHARON® system experience constructed at the Utrecht WWTP and at the Rotterdam WWTP. The ammonia concentration in the feed of the SHARON system is quite concentrated, ranging from 0.5 to 1.5 g N/L. Stable nitrogen conversion rates of 90% were achieved in both WWTPs.  1.3.5.2 Completely autotrophic nitrogen removal over nitrite (CANON) When partial nitrification and ANAMMOX process are integrated into one reactor, the nitrogen removal process is named as completely autotrophic nitrogen removal over nitrite (CANON) process. Both AOB and anammox bacteria grow and cooperate within the CANON process reactor to achieve nitrogen removal. The AOB take oxygen to oxidize ammonium to nitrite (Equation 10), producing an oxygen depletion condition for anammox bacteria to convert nitrite and the ammonium remained in the reactor to nitrogen gas (Equation 11). The overall CANON process reaction is shown in Equation 12.   15 NH4 + + 1.5O2  NO2 - + 2H+ + H2O                                                                                           [10]  NH4 ++1.32NO2 - + 0.13H+  1.02N2 + 0.26NO3 - +2H2O                                                          [11]  NH4 + + 0.85O2  0.44N2 + 0.11NO3 - +1.08H+ + 1.43H2O                                                       [12]  As shown in the reaction, the CANON process does not require a carbon source and saves up to 63% oxygen (Khin and Annachhatre, 2004) in comparison to the traditional nitrification- denitrification nitrogen removal process (Equation 2, 3, 4 and 5). With nitrogen removal achieved in a single reactor with a small amount of aeration, the CANON process can greatly reduce the space and energy requirement, outcompeting the conventional nitrification and denitrification process or novel SHARON-ANAMMOX process, from economical points of view.   Oxygen and ammonia concentrations are critical for balancing the competitions between AOB, NOB, and anammox bacteria in CANON reactors. Since NOB can compete for oxygen for nitrite oxidation with AOB and compete for nitrite with anammox bacteria, the activity of NOB should be inhibited. It can be simply achieved by maintaining anoxic conditions in CANON reactors with DO level lower than 0.5 mg/l (Strous et al., 1997; Khin and Annachhatre, 2004). When the DO level is higher than 0.5 mg/l (or 0.5% air saturation), anammox bacteria are reversibly inhibited and will be eventually washed out from the reactor (Strous et al., 1997). The balance of the three groups of bacteria may also be destroyed by ammonium limited feeding to the reactor. For ammonium limited influent, ammonium in the reactor is exhausted rapidly by the AOB. The accumulated nitrite will facilitate the growth of NOB but inhibit anammox bacteria, leading to poor performance of the CANON the process for nitrogen removal. Third et al. (2001) reported a limit of effective and stable nitrogen removal (92% total nitrogen removal) in a CANON system as loading rate of 0.1 kgN/m3•d. Lower than this critical point, a third of the bacteria developed was later identified as Nitrobacter and Nitrospira species that resulted in nitrogen removal decreasing from 92% to 57%, temporarily.   The CANON process is usually operated with a sequencing batch reactor (SBR) or gas-lift reactor. Sliekers et al. (2002) reported a CANON system using a 2 L SBR, which was performed two cycles per day (11.5h filling period, 0.25h settling period, and 0.25h drawing period). The  16 reactor was inoculated with sludge from an ANAMMOX-SBR-reactor, in which 80% of the biomass consisted of anammox bacteria. The reactor achieved a nitrogen removal rate of 0.16 kgN/m3•d and maximum anammox activity of 0.15 kg NH4+–N per kg dry waste per day, after 70 days of operation under oxygen-limited conditions. Third et al. (2005) investigated an operation strategy of CANON SBR system with intermittent aeration. AOB and anammox bacteria were thus activated alternatively, under this intermittent aeration strategy. The reactor achieved nitrogen conversion rate of 0.08 kgN/m3•d. Sliekers et al. (2003) choose a 1.8 L gas-lift reactor for carrying out the CANON process under an oxygen concentration around 0.5 mg/L. The reactor achieved N-removal rate of 1.5 kgN/m3•d, which was much higher than previously reported.  A recent study also indicated that membrane-aerated biofilm bioreactor (MABR), equipped with non-woven fabrics, is also feasible for the CANON process treating ammonium-rich wastewater (Gong et al., 2007). The oxygen was forced to penetrate through the non-woven fabrics, generating an oxygen concentration gradient in the biofilm reactor. Partial nitrification that consumed dissolved oxygen occurred in the inner layer of fabrics.  ANAMMOX occurred in the outer layer of fabrics that was absent of oxygen. The MABR achieved 0.77 kgN/m3•day  of N conversion after 83 days.  1.3.6 Start up of an ANAMMOX reactor 1.3.6.1 Reactor configuration Since anammox bacteria are slow growth autotrophs with maximum specific growth rate of 0.0027 h-1 and doubling time of 11 days (Strous et al., 1999), reactors carrying out the ANAMMOX process need to have effective retention of biomass, preventing biomass washed out from the reactors, and also should be suitable for long term enrichment, cultivation and quantification.   A fluidized-bed reactor was first take into consideration since ANAMMOX process was originally discovered within a fluidized-bed reactor (Mulder et al., 1995). Anammox microorganisms should be able to grow as biofilms on the sand particles in the reactor (Strous et al., 1998). Strous et al. (1997) studied the feasibility of fluidized-bed reactor to hold anammox sludge with a 2.5 L (height 70cm, diameter 7cm, liquid volume 2.25L) glass column reactor  17 inoculated with anammox sludge. After 100 days, the sludge granules changed from brownish to red and anammox bacteria became the dominant bacteria in the reactor. The reactor achieved 82% ammonium removal efficiency and 99% nitrite removal efficiency. However, two out of 15 attempts in total ended by complete washout of the sludge and activity loss. This was mainly due to a lack of complete bulk mixing in a fluidized bed reactor, causing the biofilm structure to change over different reactor runs. If some areas of the reactor were unable to receive substrate, the biomass would starve and thus decrease anammox activity.  Sequencing batch reactor (SBR) seems to be more feasible and also widely used for carrying out the ANAMMOX process (Strous, et al., 1998; Arrojo, et al., 2006; Chamchoi and Nitisoravut, 2007; López, et al., 2008; Jin, et al., 2008). This is mainly because of the efficient biomass retention of SBR. In comparison with stoichiometry predicted results, Strous et al. (1998) reported a 90% biomass retention efficiency of their SBR for anammox sludge accumulation. This means that only 10% of biomass would be washed out from the reactor. A large amount of anammox sludge can thus be enriched in the reactor after a period of time.   In addition, SBR is an easy control process and is feasible to be optimized by changing its operational strategies. A typical SBR cycle includes three stages: mixed filling with substrate, settling and drawing, which can be controlled by PLC controllers (Figure 1.6).   INFLUENT EFFLUENT SBR PLC  Figure 1.6 A typical sequencing batch reactor set up  18  Depena-Mora et al. (2004) slightly modified their SBR cycle by adding a 30 minutes mixing stage with substrate supply between mixed fill stage and settling stage. The modification helped overcome the floatation problems in SBR and significantly improved the settling property of anammox sludge in the SBR reactor; the sludge volume index (SVI, volume of 1g suspension sludge after 30 min settling) decreased from 108 to 60 cm3/g. This is due to nitrogen gas bubbles can be better released during this mixing period without substrate supplying and because further nitrogen gas production was prevented by consuming remained nitrite in the reactor within the 30 min mixing stage. When carrying out a lab-scale ANAMMOX process, the anaerobic environment of SBR can be maintained by flushing with argon gas (Jin et al., 2008). While conducting the CANON or OLAND process, which requires an anoxic environment with DO level around 0.5 mg/L, the SBR can be modified with a pH-controlled aeration system where aeration is activated when pH is high in the reactor and off when pH drops, due to nitrification in the reactor (Kuai and Verstraete; Wett, 2006).  One limitation to the nitrogen conversion rates of the reactors is the efficiency of oxygen transfer through gas and liquid mass phase (Sliekers et al., 2002). Gas-lift reactor (Figure 1.7) was supposed to increase the N-conversion rate of Anammox related process due to its relatively high gas-liquid mass transfer of oxygen (Sliekers et al., 2003). Sliekers et al. (2003) proved that a gas-lift reactor was suitable for carrying out both ANAMMOX and CANON process; N-removal rates of 8.9 kgN/m3•d and 1.5 kgN/m3•d were achieved, respectively. The removal rate was almost 20 times higher than which was achieved in previous SBR laboratory reactors of 0.07 kgN/m3•d (Sliekers et al., 2002). However, a decrease of sludge settling ability was observed when the applied nitrogen load rate (NLR) exceeded the specific anammox activity rate of the biomass. This means that the NLR of gas-lift reactor may be limited by the specific anammox activity in the reactor.  19 INFLUENT EFFLUENT Argon DO pH ORP  Figure 1.7 Gas lift reactor set up (Depena-Mora et al., 2004)  According to Jin et al. (2008), an anaerobic sludge blanket (UASB) reactor is a more stable reactor configuration to tolerate substrate concentration shocks than both SBR and upflow biofilm reactor (UBF), when carrying out the ANAMMOX process. Since granular sludge is easily achievable, UASB reactor might be more suitable to carry out high-loaded ANAMMOX process than other reactors. Super high nitrogen removal rates (NRR) of 74.3–76.7 kgN/m3•d were achieved by Tang et al. (2011), which were almost three times higher than other previous reported NRR values.   Recent studies indicated that biofilm reactors (Figure 1.8), such as upflow biofilm reactor (UBF) and Membrane-aerated biofilm reactor (MABR), are also feasible to carry out the ANAMMOX process; the nonwoven fiber material used within the reactor was proved to be effective to attach biomass in anammox cultivation. Compare with SBR, UBF might be a better reactor in terms of reducing start-up time. Jin et al. (2008) reported that it took 57 days to start up an ANAMMOX SBR reactor, while UBF reactor needed only 31 days. However, a biofilm reactor is not feasible in tolerating high nitrogen loading because of the limitation of surface area. Kim et al. (2009) reported that the nitrogen-loading rate of biofilm ANAMMOX reactor cannot be higher than 1000 mg TN/L.  20 INFLUENT EFFLUENT Gas Collector  Figure 1.8 Schematic diagram of a biofilm ANAMMOX reactor (Kim et al., 2009)  1.3.6.2 Reactor inoculation and start up Anammox bacteria are widely found in the environment. Based on PCR analysis, anammox bacteria were identified in different water and wastewater treatment units including membrane bioreactors, denitrifying tanks, nitrifying RBCs, oxidation ditches, anaerobic digesters and aeration tanks (Zhang et al., 2008). Variety types of conventional sludge have been used as inocula for carrying out the ANAMMOX process in certain reactor configurations (Table 1.4). However, even in small-scale lab reactors, starting an ANAMMOX reactor with conventional sludge as inocula may take a period over 100 days. For example, start-up experience from a full scale UASB ANAMMOX reactor in Rotterdam, NL, with nitrifying sludge as inocula, indicated that the start-up times for full-scale reactors might be longer than 3 years (van de Star et al., 2007). In order to achieve a fast reactor start-up, the addition of mature anammox sludge from exotic sources might be necessary. Ni et al (2011) reported a successful start-up of a 50 L pilot- scale UASB granular reactor within 2 weeks, by using sludge containing mature anammox bacteria.     21 Table 1.4 Start-up ANAMMOX reactors with different seeding sludge Sludge Start-up time (d) Reactor used References Upflow blanket anaerobic sludge  150 SBR Chamchoi and Nitisoravut, 2007 Activate sludge 150 SBR Chamchoi and Nitisoravut, 2007 Anaerobic digestion sludge 150 SBR Chamchoi and Nitisoravut, 2007 Nitrifying sludge 105 UASB Zheng et al., 2004 Denitrification sludge  100 UASB Zheng et al., 2004  1.3.7 Operational concerns of ANAMMOX reactor In order to control and optimize the ANAMMOX process, parameters that might affect the performance of the ANAMMOX reactor had been widely investigated and are summarized in Table 1.5.  Table 1.5 Summary of studies that investigated inhibitions on ANAMMOX process Factors References Amount Effects Ammonium Strous et al., 1999 Bettazzi et al., 2010 Dapen-Mora et al., 2007 Ni and Meng et al., 2011 1g N L-1 90 mg NH4 +- N L-1 55mM 988.3 mg NH4 +- N L-1 No effect No effect 50% inhibition 50% inhibition Nitrate Strous et al., 1999 Bettazzi et al., 2010 1g N L-1 57 mg NO3  - N L-1  No effect No effect  Nitrite Strous et al., 1999 Bettazzi et al., 2010 Egli et al., 2001 Dapen-Mora et al., 2007 100 mg NO2  - N L-1 60 mg NO2  - N L-1 185 mg NO2  - N L-1 25 mM Complete inhibition Activity decrease Complete inhibition 50% inhibition Methanol Isaka et al., 2008 5 mM 70% inhibition DO level Strous et al., 1997 Jung et al., 2007 >0% O2 >0.2 ppm Complete inhibition Activity decrease  Anammox bacteria are strictly anaerobic. The anammox activity is complete inhibited by the presence of oxygen (Strous et al., 1997). However, the inhibition of oxygen is reversible, making it possible to combine partial nitrification and the ANAMMOX process within one reactor. Ni and Meng (2011) observed a 50% inhibition of anammox activity with ammonia concentrations  22 as high as 988.3 mg NH4 +-N/L. The inhibition can be fully recovered by decreasing the substrate concentration. Jung et al. (2007) suggested a DO level below 0.2 ppm when anammox bacteria were cultivated in the continuous culture reactor.  Concentration of nitrite in the ANAMMOX reactor is considered as a critical factor that may cause severe inhibition to the activity of anammox bacteria even at a low concentration level. Although nitrite is an electronic acceptor in the ANAMMOX process, nitrite accumulation should be avoided when carrying out the ANAMMOX process in any reactor. Strous et al. (1999) detected an activity loss of anammox bacteria in a SBR at concentration higher than 100 mg/L NO2-N, while Egli et al (2001) reported a complete activity loss at concentration at 185 mg/L NO2-N within a RBC system. Adding intermediate compounds (hydroxylamine and hydrazine) could help to recover the anammox bacteria activity after complete or partial inhibition by nitrite toxin, however, permanent recovery is hard to achieve (Bettazzi et al., 2010). The concentration of ammonia, as substrate of ANAMMOX reaction, could also inhibit anammox bacteria. However, compared with nitrite, the toxicity concentration of ammonia is relatively high.    Nitrates would be accumulated within the reactor as simultaneous products of ANAMMOX reaction. The concentration of nitrate compounds would not inhibit to anammox bacteria (Bettazzi et al., 2010). However, the production of nitrate should be limited in a SHRAON or CANON process, to protect AOB from partial nitrification and limit nitrite oxidation.   Organic compounds, especially methanol, may contaminate the ANAMMOX reactor and inhibit anammox bacteria, since nitrate would be produced in the ANAMMOX process and might be removed by the heterotrophic denitrification process. Methanol would be added as hydrogen donor and recycled to ANAMMOX reactor with effluent from heterotrophic denitrification process to dilute high ammonium concentration or to avoid nitrite accumulation in the ANAMMOX reactor. Batch experiments done by Isaka et al. (2008) demonstrated that the anammox activity decreased about 70% when 5 mM methanol was added. The inhibition effect was found irreversible, suggesting that it is important to avoid methanol contamination in an ANAMMOX reactor.   23 1.3.8 Applications of ANAMMOX Since the ANAMMOX process uses nitrite instead of organic carbon as an electronic acceptor, it is more suitable to be applied to treat wastewater with low carbon to nitrogen ratio. Recent research indicated that the ANAMMOX process could be successfully applied to treat sludge digester effluent (centrate), landfill leachate, and coke-oven wastewater (Table 1.6).   Table 1.6 Summary of the applications of ANAMMOX process Wastewater Process NRRa Scale Reference Centrate Partial nitrification-ANAMMOX Partial nitrification-ANAMMOX Partial nitrification-ANAMMOX 0.71 9.50 1.10 10 L 70 m3 1 L van Dongen et al., 2001 van der Star et al., 2007 Gali et al., 2007 Landfill leachate Partial nitrification-ANAMMOX 14 12 L Liang and Liu, 2008 Coke-oven WW ANAMMOX 0.062 1 L Toh and Ashbolt, 2002 a Nitrogen removal rate, unit: kg N/(m3•d)  Centrate from a municipal plant is a typical wastewater with a very low carbon to nitrogen ratio. In addition, the pH value (7.0 ~ 8.5) and temperature (35 ºC ~ 40 ºC) of centrate are commonly good for the growth of anammox bacteria. Thus, it is very suitable to use the ANAMMOX process to treat centrate. Van Dongen et al. (2001) first established a 10 L lab-scale Partial nitrification-ANAMMOX process treating centrate from Rotterdam WWTP, Netherlands. The SHARON process used a continued stirred tank reactor (CSTR), with hydraulic retention time (HRT) of 1 d. The ANAMMOX process applied a granular SBR. Over 80% of NH4 + could be converted to N2 in this case. Based on the lab results, a full-scale SHARON-ANAMMOX process was built in Rotterdam WWTP (van der Star et al., 2007). Stable NRR of 9.5 kg N/(m3•d) was finally reached after 3.5 years of start-up operation.  Landfill leachate is also characterized by low biodegradable organic contents and high concentrations of ammonia-nitrogen. Liang and Liu (2008) already indicated that leachate could be successfully treated with an ANAMMOX-soil infiltration system. Landfill leachate obtained from a municipal solid waste (MSW) sanitation landfill in Beijing, China with 1400~2800 mg/L NH4 +-N and 1100-2600 mg/L CODCr was treated with a partial nitrification-ANAMMOX system. Both partial nitrification and ANAMMOX were carried out in a fixed bio-film reactor.  24 Results showed that stable removal efficiencies of 97% NH4 +-N and 89% CODCr could be achieved by this system, indicating the process is feasible for leachate treatment.  Coke-oven wastewater, which is different from centrate and landfill leachate, contains not only high concentration of ammonia-N (330 ~ 650 mg/L) and organic matters (2000 ~ 2500 mg/L COD), but also toxic compounds like phenol (300 ~ 800 mg/L), cyanides (10 ~ 90 mg/L) and thiocyanates (300 ~ 500 mg/L) (Toh et al., 2002). Toh and Ashbolt (2002) reported a successful application of ANAMMOX process treating synthetic coke-ovens wastewater. Phenol was added stepwise to the influent from 50 to 550 mg/L to avoid instant toxic effect. After 15 months of enrichment, the NRR reached 0.062 kg N/(m3•d) in this system.      25 2. Materials and Methods 2.1 Description of the sludge treatment facility The research project was carried out at the J.A.M.E.S. wastewater treatment plant (WWTP). The J.A.M.E.S. WWTP (Figure 2.1), located on the bank of the Frasier River in the city of Abbotsford, BC, is the third largest secondary treatment plant in British Columbia. Serving over 240,000 residents the plant treats an average of 81,700 m3/d (18 MGD). The plant consists of both primary and secondary treatment, producing a high quality effluent with an average biochemical oxygen demand (BOD) of 45 mg/L and total suspended solids (TSS) of 45 mg/L.    Figure 2.1 Picture of J.A.M.E.S. WWTP  The influent is discharged into the plant pre-treatment area by two screw pumps. Large particles in the influent like rags, sticks and plastics are screened out from the influent. Water would then proceed through a separator system (Figure 2.2). Next, the particles are allowed to settle down in primary sedimentation tanks (Figure 2.3 and Figure 2.4) with a residence time of 2 to 7 hours. Lighter particles float to the surface and are collected using skimmers (Figure 2.5), while heavier particles sink to the bottom of the tanks where they are raked to a sump, and pumped to the anaerobic digesters for treatment. The particles collected by the screens and grit tanks are dewatered and transported off site to the regional landfill.   26   Figure 2.2 Primary grit separation tank of J.A.M.E.S. WWTP   Figure 2.3 Primary sedimentation tank of J.A.M.E.S. WWTP  27  Figure 2.4 Decant of primary sedimentation tank of J.A.M.E.S. WWTP   Figure 2.5 Skimming removal tanks of J.A.M.E.S. WWTP  The wastewater is then pumped into two biological trickling filters (Figure 2.6), where the water is distributed over plastic media plates (Figure 2.7) (covered by growing biomass) after being mixed with return sludge from the clarifier and effluent from the solids contact tank. On the surface of the filter, bacterial colonies would form, producing bio-films that adsorb particles and  28 consume dissolved organics in the wastewater, resulting in a cloudy effluent that flows to solids contact tanks.  Figure 2.6 Top view of the Trickling Biological Filter    Figure 2.7 Close up of water distributor and trickling biofilter media   In the solids contact tanks (Figure 2.8), the biological trickling filter effluent is combined with sludge from the secondary clarifiers. This results in a TSS of about 2000 mg/L. With aeration, further removal of nitrogen and dissolved organic compounds is achieved. After residing for 1-2  29 hours, the water is pumped to the secondary clarifiers (Figure 2.9 and Figure 2.10) where the sludge and treated wastewater is separated. The clarifier effluent has an average BOD of 45 mg/L and a TSS of 45 mg/L. As clarified secondary effluent enters the final chlorine contact tank, a chlorine solution is injected as disinfectant. Before the effluent flow is discharged into the Fraser River, a sulphur dioxide solution is injected to neutralize any residual chlorine.  Figure 2.8 Solids contact tank   Figure 2.9 Full view of secondary clarifier   30  Figure 2.10 Close view of secondary clarifier Sludge from the secondary clarifiers is pumped back to the primary sedimentation tank, where it is mixed with the sludge from the raw effluent. This sludge, now consisting of the skimming and suspended solids from the primary sedimentation tanks, and the sludge from the secondary clarifiers, is pumped into a thermophilic (high temperature) pasteurization process tank to achieve full pasteurization. Solids from the pasteurization process are then pumped to three anaerobic digesters (Figure 2.11) for solids stabilization. Organic contents and odours are greatly reduced in stabilized solids. Methane gas, produced as a by-product in this process, is combusted in the Treatment Centre’s boilers for digester heating and building areas.   Figure 2.11 Roof of anaerobic digester   31 Once the sludge is digested, the stabilized biosolids are pumped to a dry, soil-like consistency centrifuge (Figure 2.12) for dewatering, producing a dry cake of solids (Figure 2.13) and are stored on-site. The high strength dewatered sludge liquor (centrate) from the centrifuges, contains relatively high concentration of ammonia (850 mg/L to 1200 mg/L) and is pumped into nitrifying tanks (Figure 2.14). Treated effluent from the nitrifications tanks is returns back to the head of the plant and mixed with incoming sewage.   Figure 2.12 Centrifuge   Figure 2.13 Stabilized bio-solids after anaerobic treatment and centrifuging   32  Figure 2.14 Nitrification tanks  2.2 Influent media  The pilot reactor used in this project was fed by on-site centrate collected from the dewatered sludge liquor of centrifuge process in J.A.M.E.S. WWTP. Main characteristics of the centrate as feed to ANAMMOX reactor is summarized in Table 2.1  Table 2.1 Main characteristics of centrate feeding to the ANAMMOX reactor Parameter Abbreviate Unit Minimum Maximum Average Ammonia-nitrogen NH4-N mg/L 713.1 1177.0 910.5 Nitrite-nitrogen NO2-N mg/L 0.04 4.5 0.6 Nitrate-nitrogen NO3-N mg/L 0 4.0 0.3 Phosphate PO4-P mg/L NA NA 122.5 pH pH NA 7.6 8.7 8.2 Total suspended solids TSS mg/L 73 500 280 Volatile suspended solids VSS mg/L 33 435 218.1 Chemical oxygen demand COD mg/L NA NA 429 Biochemical oxygen demand BOD5 mg/L 9 33 18 Alkalinity ALK mg/L 2460 3870 3019  33  The centrate was mainly ammonium concentrated liquid, with an average ammonia-nitrogen concentration of about 910 mg/L. There was almost no nitrite in the centrate (Avg. = 0.6 mg/L), which means partial nitrification process is necessary to partially convert ammonium to nitrite for carrying the ANAMMOX process. Biochemical oxygen demand (BOD5) of the centrate was relatively low (Avg. = 18 mg/L), thus conventional biological nitrification and denitrification would not work under such a low C/N ratio. The Chemical Oxygen Demand (COD) was as high as 429 mg/L, which was mainly due to the addition of polymer in the sludge dewatering process. Total suspended solids (TSS) in centrate (73 mg/L ~ 500 mg/L) were also highly depended on the dosage of polymer added in the sludge dewatering process. The dosage of polymer used at J.A.M.E.S. WWTP may range from 0.65 L/s to 0.85 L/s. Increasing the dosage of polymer would reduce TSS in the centrate directly. The alkalinity of the centrate is around 3000 mg/L and the ratio of alkalinity and ammonium-nitrogen was around 3.0, indicating alkalinity was enough for carrying out the CANON process. Ortho-P levels averaged about 122 mg/L.  2.3 Experimental set-up 2.3.1 Pilot-scale SBR A 400 L, pilot-scale, sequencing batch reactor (SBR) was designed for the slow growth of the anammox bacteria (Figure. 2.15). SBR has been proved to be an effective reactor for carrying out ANAMMOX reaction, due to its ability of biomass retention (Kuai and Verstraete et al., 1998; Strous, et al., 1998; Arrojo, et al., 2006; Chamchoi and Nitisoravut, 2007; López, et al., 2008; Jin, et al., 2008). Two, 500-Watt Electronic control heaters (HMA-500, JSK Merchandising Inc.) were used to control the temperature of the reactor. Isolate capsules were also used to wrap the reactor for maintaining reactor temperature. A mechanical stirrer was installed within the SBR to achieve a complete mix condition in the reactor. Mixing is important to help mass transfer in the reactor and sludge granulation (Arrojo et al., 2006). The formation of well-settling, anammox granular biomass would not only improve anammox sludge retention and  also increase its tolerance to shock loading (van der Star et al., 2007; Fernandez et al., 2008). However, granular anammox sludge may break up when exposed to high shear force conditions, at high stirring speeds. Thus, a stirring speed of 15 rpm was chosen to provide sufficient mixing to the reaction without breaking the granular sludge.  34   Figure 2.15 Sequencing batch reactor (SBR) for the CANON process  2.3.2 CANON SBR system A CANON SBR system was designed for treating the on-site centrate at the J.A.M.E.S. WWTP. The schematic diagram of this CANON system was presented in Figure 2.16. Centrate was collected from the sampling spot of sludge dewatering effluent and pumped to a 550 Gal (or 2082 L) centrate storage tank. Centrate was then pumped periodically into the CANON reactor through a submersible feeding pump (No.49, Zoeller Pump Co.) for treatment. A mechanical impeller with a mixing speed controller was used to ensure a complete mixing condition in the reactor. An air compressor (HP-40-0110, Hiblow Technology Co. LTD.) was used to maintain a certain DO level in the reactor. The aeration condition in the reactor is adjustable through an air flow meter (0 ~ 40 STD L/min) located after the air compressor. A time controller fixed on the aeration control panel was able to adjust the aeration time. A decant valve on the reactor was used to discharge centrate, after being treated. The volume of feed and decant was controlled by level switchers that connected to the feed and decant control panel.  35 Centrate Storage Tank Feed Pump Feed and Decant Control Panel Mixer Low level switcher High level switcher Decant Pump Compressed Air Pump Aeration Control Panel pH Probe S  Figure 2.16 Schematic diagram of CANON-SBR system  The CANON SBR reactor worked continuously with fresh substrate feed. Each cycle of the nitrogen removal process consisted of five stages: feeding stage, anaerobic protection stage, anoxic reaction stage, settling stage, and decanting stage. In the feeding stage, a certain volume of centrate in the storage tank was pumped into the reactor at a certain feeding speed. The volume of feeding centrate can be easily adjusted through changing the level of the feeding controller within the reactor. After feeding substrate, the reactor went through an anaerobic protection stage (typically 1 to 2 h) that only stirrer works in this stage. The purpose of this stage is to protect anammox bacteria from instant nitrite toxic at the beginning of the cycle and also avoid nitrite accumulation in the reactor. In the following anoxic reaction stage, the air compressor supplied oxygen to the reactor, maintaining anoxic condition (DO level less than 0.5 mg/L) for the CANON SBR. When the reaction was finished, 4 to 10 min sludge settling time was applied to the settling stage to ensure anammox bacteria would settle and remain in the  36 reactor. After sludge settling, the supernatant was decanted from the effluent valve on the lower part of the reactor. Decant volume could be easily be adjusted by changing the level of the decanting controller within the reactor.   2.3.3 Control of the reactor The SBR is designed to be automatically controlled by the system control panels (Figure 2.17). The control is based on real-time SBR control, as described by Yu et al. (2000), Kim et al. (2004) and Wu et al. (2007). The automatically control system (ACS) consisted of four different controllers: a pH controller, a feed controller, a decant controller, an aeration and mixing controller (Figure 2.18).   Figure 2.17 Control panel for the CANON-SBR reactor   37  Figure 2.18 Four main parts of the control panel: a) feed controller; b) decant controller; c) pH controller; d) aeration and mixing controller  The core part of the system is the pH and air controller, since pH and aeration was used for real- time control of the SBR cycle.  The pH would drop in parallel with nitrogen removal, which is the basis of the control in the CANON SBR. An industry pH probe (from Cole-Parmer Instrument Co.) (for continuous pH measurements) was used to detect pH variation in the CANON SBR and was calibrated once every three months for quality control. Detected signals were transported to the pH controller and the pH value was shown instantly on the LED panel. A pH level of 6.0 was set as the set point of the pH controller. When pH in the reactor was higher than the set point, the air compressor would start to work and initiate the CANON reaction (pH controlled intermittent aeration). The intermittent aeration can be controlled by setting the on and off timer of the aeration controller (Figure 2.18d). When pH dropped below the set point, the aerator and mixer would be turned off automatically for sludge settling. Sludge settling time was controlled by the decant controller (as shown in Figure 2.18b). The settling time was adjustable,  38 allowing anammox bacteria to settle within a period of time (Typically 4 ~ 10 min). After settling, a solenoid decant valve would be turned on automatically and start decanting. When the liquid level within the reactor reached the low level switcher, the decanting valve turned off automatically and the feed pump started to work. The feeding pump will be turned off automatically, until the liquid level reached the high level switcher. When the pH level increased above the set point due to the feeding of centrate, the pH controller and aeration controller started to work again.  2.4 Process operations The CANON SBR reactor was running continuously for 317 days. Within this period of time, the running of the reactor was conducted under three experimental periods: (1) start-up stage; (2) sludge enrichment and (3) system optimization.  2.4.1 Inoculation and start up The start-up of the reactor took 115 days. The start up period of the reactor can be separated into two different stages. During the first stage (from day 1 to day 24), the CANON SBR was inoculated by anaerobic sludge from the digester of the Lulu Island wastewater treatment plant (WWTP). 30 L anaerobic sludge was added into the reactor and was mixed with 300 L water and 30 L centrate. Feeding and decanting of the system was operated manually, since the reactor was extremely unstable at the beginning. The airflow rate was maintained at 5 L/min, with an intermittent period of 1 hour on and 5 hours off. DO in the reactor was measured by a DO meter (HQ30d, from HACH Company), calibrated once every two months. The concentrations of nitrite-nitrogen (NO2-N), nitrate-nitrogen (NO3-N), and ammonia-nitrogen (NH3-N) in CANON SBR were measured daily with quick test kits (from API co.). No sample was collected from the reactor for analysis because the nitrogen removal was done by biological nitrification and denitrification, instead of the ANAMMOX process, during this period. Since the DO level was below 0.5 mg/L and temperature was above 30 ºC in the reactor, nitrite oxidation process could be inhibited (Schmidt et al., 2003), leading to the outgrowth of AOB over NOB. Partial nitrification could be achieved to provide appropriate amount of nitrites as electron acceptors for the ANAMMOX process.    39 During the second stage of start up period (from day 25 to day 115), 10 L of mature anammox flocculent sludge (from the Environmental Engineering lab at the of University of British Columbia) was added into the reactor. The aeration flow rate was still maintained at 5 L/min but the intermittent duration was reduced to 1 h on/ 9 h off, to avoid nitrite inhibitory effects to anammox bacteria. The aeration intermittent period was then slowly increased from 1 h on/ 9 h off to 15 min on/ 15 min off, to increase the reaction speed. Starting from day 76, samples of influent, decant effluent and mixed liquor in the reactor were collected, respectively, for the analysis of NH3/NH4 +-N, NO2-N, and NO3-N.  2.4.2 Sludge enrichment In the second period for sludge enrichment (from day 115 to day 148), the system was adjusted for automatic running and continuous aeration. The airflow rate was increased step-wisely from 5 L/min to 10 L/min, for improving the growth rate of AOB. Once the airflow rate was increased, the nitrite concentration in the reactor was checked with quick test kits. The aeration could be further improved if only nitrite concentration in the CANON SBR was maintained below the safety limit (considered as 20 mg/L).   2.4.3 System optimization During the last period (from day 148 to day 317), the CANON system was optimized under variable environmental conditions. Since the reactor was controlled by an automatically controlled system, all parameters of the process could easily be adjusted and tested for system optimization. Operational parameters include: feeding rate, initial ammonia concentration temperature, airflow rate and sludge settling time. Tests under variable environmental conditions are summarized in Table 2.2. The optimization tests were conducted at the J.A.M.E.S. WWTP with on-site centrate collected directly from on-site sludge dewatering process for the purpose of pilot testing. Since centrate quality may vary from plant to plant, the optimization results of this research project may only be suitable, accurate and appropriate for dealing with the centrate at the J.A.M.E.S. WWTP. Whether the optimization results are universally compatible for treating centrate from other WWTPs may need further investigation.     40 Table 2.2 Summary of tested conditions for system optimization Tests No. Test time Feeding volume (L) Feeding time (min) Reactor volume (L) Temp (ºC) Airflow rate (L/min) Settling time (min) pH set point 1 148d~150d 70 1 350 28 8 10 6 2 151d~153d 70 1 350 30 8 10 6 3 154d~158d 70 1 350 26 8 10 6 4 158d~165d 70 1 350 32 8 10 6 5 166d~229d 70 1 350 32 10 10 6 6 230d~240d 70 1 350 32 10 8 6 7 241d~244d 90 1 350 32 10 8 6 8 245d~253d 90 1 350 32 12 8 6 9 254d~260d 120 1 350 32 12 8 6 10 261d~271d 120 1 350 32 14 8 6 11 272d~281d 190 1 380 32 12 8 6 12 282d~290d 190 1 380 32 12 4 6 13 291d~297d 190 380 380 32 12 6 6 14 298d~317d 190 380 380 32 12 4 6 2.5 Sample analysis For evaluating the performance of the CANON SBR, samples from influent centrate, reactor mixed liquor and decanting effluent were collected four times per week. In order to describe the nitrogen removal within cycles, mixed liquor samples from the reactor were also taken once per hour from the beginning to the end of representative cycles.  Before being analyzed, samples were preserved in the laboratory refrigerator at 4 ºC. Samples were analyzed for ammonia-nitrogen (NH3-N), nitrite-nitrogen (NO2-N), nitrate-nitrogen (NO3- N), phosphate (PO4-P), Biochemical Oxygen Demand (BOD5), Chemical Oxygen Demand (COD), total suspended solids (TSS), volatile suspended solids (VSS), alkalinity, and pH. All analyses were conducted at the Environmental Engineering Laboratory of the Department of Civil Engineering, University of British Columbia. Methods for each analysis were summarized in Table 2.3. NH3-N, NO2-N, NO3-N and PO4-P were analyzed by flow injection analysis of spectrophotometry (Quickchem 8000, Lachat). COD was determined colorimetrically by using a Hach DR/2000 direct spectrophotometer at 600 nm. To ensure the accuracy of analysis of NH3- N, NO2-N, NO3-N and PO4-P, samples were centrifuged under 25000 rpm for 10 min before  41 being analyzed. Only supernatant was used for analysis, since particles in the samples would significantly influence on the accuracy of the results. In addition, blanks prepared by distilled water were used to calibrate the results.  Table 2.3 Sample preparation and analysis methods Parameter Method a Chemical oxygen demand (COD) Standard Methods 5220 Biochemical oxygen demand (BOD5) Standard Methods 5210 Ammonia-nitrogen(NH3-N) Standard Methods 4500 – NH3 Nitrite-nitrogen (NO2-N) Standard Methods 4500 – NO2 Nitrate-nitrogen(NO3-N) Standard Methods 4500 – NO3 Phosphate(PO4-P) Standard Methods4500 – P  Total suspended solids (TSS) Standard Methods 2540D Volatile suspended solids (VSS) Standard Methods 2550 Alkalinity (as CaCO3) Standard Methods 2320 Hardness (as Ca2 +) Standard Methods 2340 a Adapted from Standard Methods for the Examination of Water and Wastewater (APHA,1989)  Base on the analytical results of influent ammonia concentration (NH3 Inf) and effluent ammonia concentration (NH3 Enf), ammonia removal rate (ARR) could be calculated through Equation 13:     Hydraulic retention time (HRT) of the reactor could be determined by measuring cycle length time (T), liquid volume of the reactor (VR) and feeding volume (Vf), as presented in Equation 14:     Ammonia loading rate (ALR) could be calculated with Equation 15:    42 3. Performance of the CANON SBR In order to evaluate the performance of the CANON SBR system for nitrogen removal from centrate, concentrations of influent and effluent ammonia-nitrogen (NH3-N), nitrite-nitrogen (NO2-N), nitrate-nitrogen (NO3-N) were measured from day 76 to day 317. Figure 3.1 shows the variation of influent and effluent NH3-N concentration and ammonia-nitrogen removal rate (ARR). The concentrations of influent and effluent NO2-N and NO3-N are shown in Figure 3.2. TSS and VSS, as presented in Figure 3.3, are important to evaluate sludge enrichment and the sludge holding ability of the reactor. Since the CANON process is an alkalinity consuming process, the differences between influent and effluent pH value, as well as alkalinity difference between influent and effluent, is presented in Figure 3.4. The influent alkalinity varied from 2460 mg/L to 3870 mg/L, while alkalinity in the effluent changed from 15 mg/L to 134 mg/L, indicating most of alkalinity was consumed by the CANON process. The consumption of alkalinity would lead to pH decreasing. pH of feeding centrate varied from 7.56 to 8.76, with an average of 8.19. Effluent pH could be as low as 5.53 to 6.41, with an average value of 6.04 after the reaction. Figure 3.5 shows the cycles length of CANON cycles and HRT of the reactor, which are critical parameters for the evaluation of system stability and future design of a full- scale reactor. ! Figure 3.1 Variation of influent and effluent ammonia-nitrogen concentration and ammonia-nitrogen removal rate (ARR) during reactor running period  43     Figure 3.2 Concentration variations of the influent and effluent nitrite-nitrogen and nitrate-nitrogen during reactor running period  44   Figure 3.3 Variation of the TSS and VSS of influent, effluent and reactor during reactor running period  45  Figure 3.4 Variation of the influent and effluent pH and alkalinity during reactor running period  Figure 3.5 Variation of the cycle length and hydraulic retention time (HRT) during reactor running period  46 3.1 Start-up period Since no samples were collected before day 76, the performance of the early stage SBR start-up could only be evaluated by the results of quick test kits. During the first stage of start-up period (from day 1 to day 24), when anammox sludge was not yet inoculated, nitrite concentration varied from 20 mg/L to 100 mg/L even at low aeration (5 L/min, 1 h on/ 5 h off). At the same time, nitrate concentration ranged from 50 mg/L to 200 mg/L. At the end of this stage, nitrite concentration decreased and remained at around 10 mg/L, which would not inhibit the growth of anammox bacteria (Bettazzi et al., 2010).   During the second stage of start-up period from (day 25 to day 115), mature anammox sludge was inoculated to the reactor. Daily test of nitrite and nitrate concentration through the quick test kits showed that the nitrite concentration dropped constantly from 10 mg/L, on day 25, to less than 2 mg/L on day 75, while the nitrate concentration drop from 200mg/L to 50 ~ 100 mg/L. The reduction of nitrite concentration could be explained by the utilization of ANAMMOX process, indicating the success of the anammox bacteria inoculation. The decreasing nitrate concentration could be caused by partial nitrification, since ammonia oxidizing bacteria could outcompete the nitrite oxidizing bacteria, under the condition of DO below 0.5 mg/L and temperature above 25ºC.   From day 76, samples were collected for analysis. The results showed that the concentrations of NO2-N in the feeding centrate were both below 0.5 mg/L, which could be ignored in the CANON process. The nitrite concentrations in the decant effluent were also kept below 0.5 mg/L. Thus, the inhibition effects of nitrite to anammox bacteria could be ignored too. However, the ARR of the reactor was found to decrease from 89.5%, on day 76, to 79.4%, on day 88 (Figure 3.1), while the effluent nitrate concentrations increased from about 114 mg/L to 182 mg/L (Figure 3.2). The unexpected decreasing of ARR could be explained by the decreasing of TSS and VSS in the reactor. The TSS dropped from 2230 mg/L, on day 76, to 1500 mg/L, on day 88, while VSS dropped from 1875 mg/L to 1200 mg/L (Figure 3.3), indicating less anammox bacteria available in the reactor. The drastic increase of NO3-N also indicated that the anammox bacteria were out competed by NOB.    47 An incident occurred on day 97 that almost caused failure during the start-up of the reactor. The electronic controlled decanting valve broke and failed to close when feeding started, causing most of sludge in the reactor to be washed out. TSS in the reactor dropped drastically from around 2000 mg/L to 593 mg/L (Figure 3.3). Until the end of start-up period (day 115), the TSS and VSS in the reactor were still 1509 mg/L and 1345 mg/L, indicating the slow regrowth of sludge occurring in the reactor. However, the failing seemed to improve the performance of the reactor. After day 97, effluent NH4-N, and NO3-N concentrates were much lower than before. Effluent NH4-N increased from 44.0 mg/L on day 97 to about 93.2 mg/L on day 104 and then decreases to about 47 mg/L on day 115, which is the end of start-up period. Similar concentration variations also applied to effluent NO3-N that increased from 28.2 mg/L on day 97 to about 85.0 mg/L, on day 105, then decreased to 55.4 mg/L, on day 115. The ARRs were above 90% and ended with 94.7% on the last day of start-up period. During the 115 days of start- up period, cycle lengths were relatively long and unstable. The cycle length decreased from 46 h on day 76 to 25 h on day 115, while the HRT also decreased from 9.6 d to 5.2 d (Figure 3.5). The decreasing cycle length and HRT indicate the acceleration of the CANON reaction, which could be caused by the growth of both nitrifiers and anammox bacteria.  3.2 Sludge enrichment period After successful start-up the reactor, the system was operated under steady condition for sludge enrichment, from day 115 to day 148. Mixed liquor volume was 350 L, feeding volume was 70 L, and reactor temperature was kept at 32 ºC. The reactor was continuously aerated with an airflow rate starting from 5 L/min on day 115. The aeration was increased to 7 L/min on day 125 and was further increased to 10 L/min on day 137. An increase in effluent nitrite concentrations from 0.92 mg/L to 1.75 mg/L was observed right after changing the airflow rate from 7 L/min to 10 L/min. The NO2-N concentration kept increasing to about 9.4 mg/L during the following several days, indicating that the activity of AOB could be stimulated by increasing the aeration, thus generating surplus nitrite for the anammox reaction.   NH3-N and NO3-N concentrations in the effluent were kept relatively constant during the sludge enrichment period. The NH3-N varied from 33.8 mg/L to 56.3 mg/L (Figure 3.1) while effluent NO3-N changed from 37.8 mg/L to 60.2 mg/L (Figure 3.2). The ARR remained above 92%, and  48 varied from 92.8% to 95.7% during day 115 to day 148. The stable NH3-N and NO3-N concentrations in the effluent and the relatively unchanged ARRs indicated that the reactor could be considered as “stable running”.   For the purpose of sludge enrichment, the settling time before decanting was kept at 10 min, allowing most of sludge to settle and remain in the reactor. Thus, the TSS and VSS in the effluent were relatively low. Effluent TSS was around 400 mg/L, while VSS was around 350 mg/L. Since sludge was maintained in the reactor, TSS and VSS in the reactor increased about 3 times during this period. TSS increased from about 1509 mg/L on day 115 to 4050 mg/L on day 148, while VSS increased from 1345 mg/L to 3175 mg/L. Red anammox granular sludge was observed in the reactor, which is obviously different from the rest of flocculent nitrifying sludge (Figure 3.6). In addition, the anammox granules trended to grow denser and bigger as shown in Figure 3.7. The average diameter (Φ) of anammox granular sludge increased slowly from 0.8 mm on 105 d to 2.0 mm on 174 d, and remained at 2.0 mm until day 308. Since the size of anammox sludge granule may be influenced by mixing speed (Arrojo et al., 2006), 2.0 mm might be considered as the maximum average diameter of anammox granule under a certain mixing speed (15 rpm) of this research project. Sludge granules with a diameter larger than 2.0 mm might break up due to shear forces. Since anammox granules settled much faster than flocculent nitrifiers, almost all anammox bacteria could be preserved in the reactor, while a part of nitrifiers could be lost through decanting (leading to a healthier balance in reactor). The sludge enrichment of anammox bacteria was considered as successful on day 148, at the end of this period.   Figure 3.6 Picture of flocculent nitrifying sludge (a) and granular anammox sludge (b)  49   Figure 3.7 Picture of the size of granular anammox sludge  Due to enriched sludge, the cycle length reduced significantly from 25 h on day 115 to 10 h on day 148 (Figure 3.5). Due to the faster reaction speed, the HRT also decreased from 5.21 d on day 115 to 2.08 on day 148. In addition, the ammonia-nitrogen loading rate (ALR) increased from 0.17 kgN/m3•d on day 115 to 0.45 kgN/m3•d on day 148 (Figure 3.8). Without decreasing the ARR, higher ALR of the reactor indicated a higher treatment ability of CANON process. More centrate could thus be treated, within a certain period of time.   50  Figure 3.8 Variation of ammonia-nitrogen loading rate during reactor running period  3.3 System optimization period  From day 148 to day 317, the CANON SBR was tested under different environmental conditions, as summarized in Table 2.3. Under each condition, the reactor was allowed to run at least 5 cycles for taking enough samples. The environmental conditions during sludge enrichment period were adopted as an initial baseline for comparison. The initial conditions were set as: temperature 32 ºC, 70 L feeding, 1 min feeding time, 350 L reactor mixed liquid, 8 L/min airflow, 10 min settling time and pH set point of 6.0. Since the ARR under the baseline condition was already as high as 92.8%, other optimization conditions could not be expected to achieve significant ARR improvement. Reactor running under most conditions could achieve around 90% ARR, which was still much higher than other conventional biological ammonia removal processes.  It should be noted that on day 168 the system was moved to the J.A.M.E.S. WWTP and the change of centrate quality could influence on the performance of the reactor. Different from Lulu  51 Island WWTP, centrate from J.A.M.E.S. WWTP usually contains a 1~2 mm a sludge layer on the top of the clear centrate layer (Figure 3.9). Adding centrate with sludge layer directly into the CANON SBR might contaminate the reactor, since the added sludge may compete for oxygen and substrate with the original existed AOB and anammox bacteria. For this reason, the performance of the CANON SBR decreased significantly during the first month in the J.A.M.E.S. WWTP. The effluent NH3-N increased from 29 mg/L on day 171 to about 205 mg/L on day 212 (Figure 3.1). The NO3-N concentration in the effluent also increased from 22.1 mg/L on day 171 to about 132 mg/L on day 212 (Figure 3.2). In addition, the ARR of CANON SBR decreased drastically from 96.6% on day 171, to 78.2% on day 212 (Figure 3.1). This situation was a bit like the one occurring during day 76 to day 88, indicating the nitrite for ANAMMOX reaction was used up by NOB for nitrate generation, due to the outcompetition of anammox bacteria by NOB. The cycle length also increased for 20% from 10 h on day 168 to 12 h on day 212 (Figure 3.5), showing the decrease of the ANAMMOX reaction.    Figure 3.9 Centrate with sludge layer from J.A.M.E.S. WWTP  The situation of centrate contamination was not noticed until day 213. When feeding substrate was changed back to clear centrate, with the sludge layer being wasted, a significant improvement in reactor performance was then observed. The effluent NH3-N concentration decreased continuously from about 205 mg/L on day 212 to 76.0 mg/L on day 226 (Figure 3.1),  52 while effluent NO3-N also decreased from about 132 mg/L to 52.3 mg/L on day 226 (Figure 3.2). The ARR also increased to 91.7% on day 226 (Figure 3.1).   During the optimization period, TSS in the reactor varied from 2500 mg/L to 4500 mg/L, while VSS was usually 500 mg/L less than TSS (Figure 3.3). This was caused by adjusting the sludge settling time. The sludge settling time was changed stepwise from 10 min to 4 min. Reduced sludge settling time would decrease TSS and VSS in the reactor, by withdrawing more sludge through decanting.  3.4 Cycle test   Within the system optimization period, the reactor was running under variable environmental conditions. For more detailed study, the performance of the reactor under each condition, cycle test was conducted and cycle profiles were generated for each cycle. Data of cycle tests are summarized in Appendix A. As an example, the cycle profile for the baseline condition (32 ºC, 70 L feeding, 1 min feeding time, 350 L reactor mixed liquid, 8 L/min airflow, 10 min settling time and pH set point of 6) is presented in Figure 3.10.  Figure 3.10 Cycle profile for cycle test of CANON SBR under baseline condition  53  The cycle profiles reflect the variations in DO level, pH values, NH3-N, NO2-N and NO3-N concentrations within the reactor, from beginning to the end of a whole cycle. For example, in the profile of cycle test under baseline conditions (Figure 3.10), it shows that the cycle length of CANON SBR, under the baseline condition was 9.5 h. A whole cycle consisted of 2 hours of an anaerobic phase at the beginning, followed by 7.5 hours of anoxic aeration phase. DO levels were below 0.2 mg/L under the anaerobic phase, and increased to around 0.5 mg/L ~ 0.6 mg/L, when air was introduced. Right after feeding the centrate, the initial NH3-N concentration in the reactor was about 214 mg/L and the initial pH was 7.63 due to the dilution. NH3-N and pH remained unchanged during the first two hours of anaerobic phase. The two-hour anaerobic phase was designed to consume the remaining nitrite in the reactor. Thus, anammox bacteria could be protected from nitrite inhibition. As a result, the NO2-N concentration was decreased from 2.1 mg/L to 1.57 mg/L, after the anaerobic phase. During this period of time, nitrite was converted to nitrate and nitrate was converted to nitrogen gas. Due to the denitrification process, NO3-N concentration dropped for about 12% from 36.1 mg/L to 31.9 mg/L, after 2 hours. When aeration started and the CANON reaction was initiated, the NH3-N concentration started to drop. After 7.5 hours of aeration, the NH3-N concentration decreased to about 51.2 mg/L, and the pH decreased to the set point of 6.0. NO2-N slightly increased to around 3.0 mg/L in the reactor, while NO3-N kept increasing to 38.3 mg/L by the end of the cycle. The nitrate generation to ammonia reduction ratio was 0.08, which was lower than the theoretical ratio of 0.11 for CANON reaction (Equation 12).  54 4. Kinetics of CANON SBR The CANON system requires the cooperation of both ammonia oxidation bacteria and anammox bacteria while inhibiting the competition between anammox bacteria and nitrite oxidation bacteria. A study on the effects of variety parameters that may influence on the performance of the reactions would help better control and optimize the reactor. In this section, the effects of temperature, aeration, ammonia loading, feeding rate, and sludge settling time were investigated, to the extent possible.  4.1 Effects of temperature Temperature is usually critical for most biological reactions. The temperature for optimum growth of the nitrifying bacteria is between 30-35 ºC, while optimum temperature for anammox bacteria is around 35 ºC (Schmidt et al., 2003; Zhang et al., 2008). However, in the real case of centrate treatment, centrate temperature may vary from 30 ºC to 37 ºC. If there is no extra heating device to heat the centrate during the process, temperature within the reactor could be even lower. Thus, the effects of temperature ranging from 26 ºC to 32 ºC were investigated and the results are presented in Figures 4.1 to Figures 4.5. Temperatures below 26 ºC are too low to maintain the advantage of AOB, to outcompete NOB (Khin and Annachhatre et al., 2004). Except for the difference in temperature, all cycle tests were done under the same condition: 70 L feeding in one minute, 70 L decanting with 10 min sludge settling time, 350 L reactor mixed liquid volume, continuously aeration with 8 L/min, and pH set point of 6.0.  Since the system was controlled by pH variation, the cycle length was directly determined by the speed of pH decreasing (or alkalinity consumption) in the reactor. In other words, ammonia- nitrogen concentration in the effluent was not controlled. As shown in Figure 4.1, cycle lengths were slightly different under temperature conditions of 26 ºC, 28 ºC, 30 ºC and 32 ºC. All four sets of cycle tests started at almost the same pH level of 7.6, which is reasonable due to the same volume of feeding and the stable pH of centrate. A pH of 6.0 was the set point, which means a cycle will be completed when the pH in the reactor dropped to 6.0. For cycles at 26 ºC, 28 ºC, 30 ºC and 32 ºC, time to accomplish the cycle was 9 h, 11 h, 10 h, and 9.5 h, respectively, including  55 the first two hours of anaerobic protection period. The cycle length at 26 ºC was the shortest, while the cycle length under 28 ºC was the longest. However, the differences were not significant.   Figure 4.1 Comparison of the effects of temperature on pH variation in the reactor   Although the cycle length tested at 26 ºC was the shortest, the performance of ammonia removal was the worst. As shown in Figure 4.2, the NH3-N in the reactor was reduced from about 223 mg/L to 79 mg/L, which only achieved 64% ammonia removal. The ammonia removal rate increased at higher temperature. The ARRs at temperature of 28 ºC, 30 ºC and 32 ºC were 68%, 70%, and 76%, respectively. Due to the increase in temperature from 28 ºC to 32 ºC, the ARR was increased significantly by 12%, from 64% to 76%. This was attributed to the activity of anammox bacteria were inhibited at lower temperature, while the influence of temperature on AOB activity was not that significant.    56  Figure 4.2 Comparison of the effects of temperature on ammonia concentration in the reactor   The inhibition of anammox bacteria activity can be observed in Figure 4.3, which presents the effects of temperature on nitrite concentrations in the reactor, within cycles. As shown in Figure 4.3, the effect of the two-hour anaerobic protection period was obvious. Although the initial concentrations at the beginning of the cycle might be as high as 13.4 mg/L at 26 ºC, nitrite- nitrogen concentrates dropped blow 2.0 mg/L after two hours of anaerobic phase. During the cycle, nitrite concentration would increase due to a partial nitrification process. At temperature of 32 ºC, the level of NO2-N was relatively stable. The NO2-N increased from 1.6 mg/L to the maximum level of 4.0 mg/L, and then decreased to 2.1 mg/L at the end of the cycle, indicating anammox bacteria were robust and were capable of utilizing most of nitrite in the reactor generated by AOB. However, the nitrite concentration in the reactor increased with temperature decreasing. At 26 ºC, NO2-N concentration increased dramatically from 1.6 mg/L right after the anaerobic phase, to 15.7 mg/L at the end of the cycle. Compared with the relatively stable nitrite concentration at temperature of 32 ºC, the accumulated nitrite in the reactor at 26 ºC indicated that the activity of anammox bacteria was significantly inhibited at lower temperature; thus, nitrite generated by AOB could not be completely utilized by anammox bacteria. Although  57 nitrite levels at 26 ºC was still lower than the safety threshold value of 20 mg/L, more nitrite in the reactor would stimulate the growth of NOB and lead to more nitrate generation in the reactor, as shown in Figure 4.4. Due to the denitrification process within the anaerobic phase, nitrate concentrates were slightly reduced. At a temperature of 26 ºC, the nitrate-nitrogen increased from 31.9 mg/L, after the anaerobic phase, to 54.3 mg/L at the end of cycle, while NO3-N increased from 26.2 mg/L to 38.3 mg/L, at a temperature of 32 ºC. At 26 ºC, the NO3-N increased for 70.2%, which was much higher than the increased rate of 46.1%, at 32 ºC, since the activity of anammox was inhibited at 26 ºC; thus, some nitrites were utilized by NOB to generate nitrates.  Figure 4.3 Comparison of the effects of temperature on nitrite concentration in the reactor   In addition, the ratio of nitrate production to ammonia consumption (N/A ratio) was higher at a lower temperature, as shown in Figure 4.5. The N/A ratio was 0.16, 0.14, 0.10 and 0.08, respectively at temperatures of 26 ºC, 28 ºC, 30 ºC and 32 ºC. Compared to a theoretical N/A ratio of 0.11 for the CANON process, the N/A ratio of 26 ºC was much higher, indicating partial nitrification and a more unbalanced anammox process at lower temperatures.   58  Figure 4.4 Comparison of the effects of temperature on nitrate concentration in the reactor    Figure 4.5 Comparison of the effects of temperature on the ratio of NO3-N generation to NH3-N consumption  59 From the discussions above, it can be concluded that temperature is a factor that can change the performance of the CANON SBR reactor. A higher operational temperature of 32 ºC was recommended for the CANON SBR, since the ammonia removal rate was the highest. Temperatures as low as 26 ºC would lower the ARR to about 12 %, from 76% at 32 ºC to 64% at 26 ºC. At the mean time, nitrite increased drastically at 26 ºC, possibly leading nitrite accumulation in the bulk liquid of the reactor and stimulating the outgrowth of NOB over anammox. The higher level of accumulated nitrite in the bulk liquid would also require longer anaerobic protection time to consume the nitrite, before the CANON process could be started. For full-scale devices, heating the reactor could be costly. However, the temperature of the centrate is around 38 ºC ~ 40 ºC. Increase the feeding volume of the high-temperature, on-site centrate could lift the temperature of the reactor and dilute accumulated nitrite in the bulk liquid, if there is no external heater.  4.2 Effects of aeration Aeration is one of the most critical factors in the CANON process, since the reaction consumes 0.85 mol oxygen to convert 1 mol ammonia to 0.435 mol of nitrogen gas and 0.11 mol nitrate- nitrogen (Equation 12). Thus, insufficient aeration would slow the CANON reaction. However, excessive amount of oxygen supply would also stimulate the growth of AOB and generate superfluous nitrite that could not be used up by anammox bacteria, leading to nitrite accumulation and NOB reproduction. The effects of airflow rate of 10 L/min, 12 L/min, and 14 L/min were investigated and the results are presented in Figures 4.6 to Figures 4.10. Except for the aeration rate, all three sets of cycle tests were conducted under the same condition of 120 L feeding, 120 L decanting with 8 min sludge settling time, 350 L reactor mixed liquid, temperature of 32 ºC and pH set point of 6.0. In addition, before the continuous aeration, an hour of anaerobic protection period was set to remove possible accumulated nitrite from the reactor.   As shown in Figure 4.6, increasing the airflow rate from 10 L/min to 12 L/min significantly reduced the cycle length from 16 h to 14 h. However, a further increase in airflow rate from 12 L/min to 14 L/min made only a slightly difference, reducing the cycle length time from 14 h to 13.5 h. For all three sets of tests, the pH increased during the first hour of anaerobic protection period, since denitrification occurred in this period of time and generated extra alkalinity. When air was continuously introduced after the anaerobic period, the pH still increased for one or two  60 more hours. This was attributed to the air stripping effect that released parts of unionize ammonia and carbon dioxide from the bulk liquid phase of the reactor, leading to a slightly pH increase. After that, the pH started to decrease due to alkalinity consumption by the CANON process and ended at 6.0 as the set point.   Figure 4.6 Comparison of the effects of airflow rate on pH variation in the reactor   Increasing the airflow rate also lead to the reduction of hydraulic retention time (HRT) of the reactor (Figure 4.7). HRT is a critical parameter, especially for the design of bioreactors. With a smaller HRT, a certain volume reactor can deal with a larger volume of wastewater. On the other hand, a certain amount of wastewater can be treated in a smaller volume of reactor.  In this case, HRT reduced from 1.94 d to 1.76 d when the airflow was increased from 10 L/min to 14 L/min, indicating a 10% HRT reduction could be achieved at the cost of 40% increase in aeration. Reducing the HRT for 10% may lead to a 10% smaller reactor or 10% more treatment capacity, while the increase in 40% aeration may result in 40% more cost for aeration of the CANON system. Hence, a cost benefit optimization would be needed before a finial design package could be recommended.  61  Figure 4.7 Comparison of the effects of airflow rate on the hydraulic retention time   Figure 4.8 Comparison of the effects of airflow rate on the variation of ammonia-nitrogen in the reactor   62   For the performance of ammonia removal, the influence of airflow rate was not that significant (Figure 4.8). With 120 L feeding centrate, initial ammonia in the bulk liquid of the reactor was around 300 mg/L. With 10 L/min aeration, NH3-N was reduced from about 308 mg/L to 94.2 mg/L. At 12 L/min, the NH3-N was reduced from about 284 mg/L to 72.5 mg/L. With an airflow rate of 14 L/min, the NH3-N was further reduced from about 278 mg/L to 60.4 mg/L. The ammonia removal rate was slightly increased along with the increase in airflow rate. The ARR of 10 L/min aeration was 69%. A 6% ARR improvement from 69% to 75% was achieved, when aeration was increased to 12 L/min. However, only a 3% ARR increase, from 75% to 78%, was achieved when aeration was further increased to 14 L/min.  For nitrite in the reactor, a slight increase in nitrite concentration was observed when aeration was increased. For all three sets of tests, the nitrite concentration decreased drastically within the first hour of anaerobic protection period and then increased significantly when air was introduced. The nitrite concentration then kept decreasing after 2 ~ 3 hours of aeration, due to the activity of anammox bacteria. At the end of the cycle, nitrite concentrations increased again, probably because the activity of anammox bacteria was reduced at lower pH. Although nitrite production was stimulated by increased airflow, the overall nitrite concentrations were still kept at a low level. Even at an airflow rate of 14 L/min, the NO2-N concentration would not be higher than 3.5 mg/L, which would not inhibit the performance of anammox bacteria. An hour of anaerobic period, before aeration, could consume most of accumulated NO2-N in the bulk liquid.  Increasing the airflow rate had almost no effect on nitrate production in the reactor. Except for the first hour of anaerobic period, nitrate concentration increased continuously until the end of the cycle (Figure 4.10). With an airflow rate of 10 L/min, the NO3-N increased from 43.4 mg/L to 74.4 mg/L. With an aeration of 12 L/min, the NO3-N increased from 49.5 mg/L to 74.1 mg/L. While at 14 L/min aeration, NO3-N increased from 47.9 mg/L to 70.8 mg/L. For all three sets of cycles, the nitrate production to ammonia consumption (N/A) ratios were 0.12, very close to the theoretical N/A ratio of 0.11 for the CANON process. The same N/A ratio of 0.12 for cycles with 10, 12 and 14 L/min of aeration indicated increasing airflow rate did not affect the balance of partial nitrification and ANAMMOX reaction.    63  Figure 4.9 Comparison of the effects of airflow rate on the variation of nitrite-nitrogen   Figure 4.10 Comparison of the effects of airflow rate on the variation of nitrate-nitrogen   In conclusion, increasing the airflow rate could improve the performance of the reactor, since higher ARR and shorter HRT were achieved. However, the improvement was not that significant. Even at the cost of 40% more aeration, ARR could only be improved by 9%, while HRT could be reduced by 10%. Although the aeration requirement for the CANON process is 63% less than conventional biological ammonia removal process, as the effects of increasing the airflow rate to 14 L/min were still not impressive. Compared with an airflow rate of 10 L/min, it is more acceptable to maintain aeration at 12 L/min, for achieving 6% in ARR improvement and 6% HRT reduction, at a reasonable 20% more aeration cost.   64 4.3 Effects of ammonia loading Ammonia is one of the main substrates for the CANON process, so increasing the initial concentration of ammonia in the reactor should be able to stimulate the CANON process. However, excessive amounts of ammonia could also inhibit the ANAMMOX process. As reported by Ni and Meng et al. (2011), 50% inhibition was observed under NH3-N concentration of about 988 mg/L.   To investigate the effects of ammonia loading on the performance of the CANON SBR, the reactor was tested under three conditions of different feed to reactor liquid volume (F/R) ratios: F/R = 0.2 (70 L feeding centrate, 350 L reactor liquid), F/R = 0.34 (120 L feeding centrate, 350 L reactor liquid), and F/R = 0.5 (190 L feeding centrate, 380 L reactor liquid). Except for F/R ratios, all three sets of cycle tests were conducted under the same conditions: 32 ºC, 12 L/min continuously aeration with one hour anaerobic protection period at prior, pH set point of 6.0 and 4 min of sludge settling time.    Figure 4.11 Comparison of the effects of ammonia concentrations on the variation of ammonia- nitrogen in the reactor within cycles  65 Due to the differences in F/R ratios, initial NH3-N concentrations (NH3-Ni) in the reactor were significantly different (Figure 4.11). A higher F/R ratio would lead to significantly higher initial NH3-N concentrations in the reactor. NH3-N concentrations started with about 215 mg/L, 289 mg/L, and 455 mg/L, respectively, for F/R ratio of 0.2, 0.34 and 0.5. The NH3-N concentrations dropped smoothly to 84.1 mg/L, 72.5 mg/L and 72.4 mg/L respectively for cycles with NH3-Ni of 215 mg/L, 285 mg/L and 455 mg/L. It should be noted that effluent NH3-N concentrations did not change significantly, due to the increase of initial ammonia concentrations, indicating that the effluent NH3-N concentration was independent of F/R ratio or initial NH3-N concentration. However, a higher ammonia removal rate was observed at a higher initial NH3-N, probably because the activity of anammox bacteria was increased at higher ammonia concentration. 61% NH3-N removal was achieved at NH3-Ni = 215 mg/L, while 72% NH3-N was removed at NH3-Ni = 285 mg/L. At NH3-Ni =455 mg/L, the ARR was further improved to 84%.  At the same time, higher ARR was achieved at the cost of a longer cycle length (Figure 4.12) since the initial pH was also raised by an F/R increase. The initial pH level was 7.1 at NH3-Ni = 215 mg/L, and was increased to 7.22 at NH3-Ni = 285 mg/L. At NH3-Ni = 455 mg/L, initial pH was even increased to 7.39. Obviously, longer cycle length time was required for the cycle with higher initial pH to reach the finial pH set point of 6.0. Differences of cycle length times, between different initial ammonia concentrations, were remarkable. The cycle length increased from 10 h to 14 h when NH3-Ni was increased from 215 mg/L to 285 mg/L. When NH3-Ni was further increased from 285 mg/L to 455 mg/L, the cycle length was extended to 21.5 h.  Although cycle length was the longest at NH3-Ni of 455 mg/L, the HRT was actually the shortest, as shown in Figure 4.13. The HRT was 1.79 d for cycle with NH3-Ni = 455 mg/L and 1.82 d with NH3-Ni = 285 mg/L. The HRT was slightly reduced when NH3-Ni was increased from 285 mg/L to 455 mg/L. Compared with NH3-Ni = 285 mg/L and 455 mg/L, the HRT of the cycle under NH3-Ni = 215 mg/L was as long as 2.08 d, indicating smaller ammonia treatment capacity of the reactor under lower initial ammonia concentration. The nitrogen removal rate (NRR) was the lowest with NH3-Ni = 215 mg/L, which was 0.27 kgN/m 3•d. At NH3-Ni = 285 mg/L, the NRR improved to 0.3 kgN/m3•d. The maximum NRR of 0.36 kgN/m3•d was achieved at the highest NH3-Ni of 455 mg/L.  66  Figure 4.12 Comparison of the effects of ammonia concentration on the variation of pH in the reactor within cycles   Figure 4.13 Comparison of the effects of ammonia concentration on the HRT, ALR and NRR  67  The improvement in nitrogen removal ability indicated that the CANON process was stimulated by higher initial NH3-N concentration. More nitrite was generated during the CANON process at higher NH3-Ni (Figure 4.14). At lower NH3-Ni = 215 mg/L, nitrite increased from 0.2 mg/L to 1.1 mg/L after three hours of aeration. Nitrite-nitrogen increased from 1.6 m/L to 2.7 mg/L at NH3-Ni = 285 mg/L and also increased from 1.4 mg/L to 4.0 mg/L at NH3-Ni = 455 mg/L. The increasing nitrite concentration indicated that the partial nitrification process of AOB was stimulated by higher initial NH3-N concentration. This stimulation effect can be explained by the Monod Equation:  Where, S is the substrate concentration, µ is specific growth rate, µmax is the maximum specific growth rate and Ks is the Monod coefficient. Supplied by higher concentration substrate, both AOB and anammox bacteria should have higher specific growth rates that accelerate the process of partial nitrification and ANAMMOX. Anammox bacteria in the reactor could quickly consume the nitrite generated by AOB, thus no nitrite accumulation problem would occur, even at high initial ammonia concentration of 455 mg/L.  Figure 4.14 Comparison of the effects of ammonia concentration on the variation of nitrite- nitrogen in the reactor within cycles  68  For nitrates in the reactor, more nitrates would also be generated at higher initial ammonia concentrations (Figure 4.15). 18.3 mg/L nitrate-nitrogen was generated at NH3-Ni = 215 mg/L while 24.1 mg/L nitrate-nitrogen was generated at NH3-Ni = 285 mg/L, and more nitrates, at 45.6 mg/L, were produced at higher NH3-Ni = 455 mg/L. Most of the nitrates produced at higher initial ammonia concentration should come from the CANON reaction (Equation 12), since the nitrate production to ammonia consumption (N/A) ratio were 0.12 for both cycles with NH3-Ni of 285 mg/L and 455 mg/L, which were very close to the theoretical N/A ratio of 0.11. It indicated that the NOB was inhibited at a surplus ammonia condition, which was also shown by Schmidt et al. (2003). At lower NH3-Ni = 215 mg/L, the N/A ratio was increased to 0.17, indicating some nitrite might be used by NOB for extra nitrate production at lower ammonia concentration conditions.    Thus, the CANON SBR could achieve better performance with higher, initial ammonium concentration. At NH3-Ni = 455 mg/L, the reactor had the lowest HRT of 1.79 d and highest NRR = 0.36 kgN/m3•d.    Figure 4.15 Comparison of the effects of ammonia concentration on the variation of nitrate- nitrogen in the reactor within cycles  69  4.4 Effects of feeding rate The effects of feeding rate (FR) on the performance of the CANON SBR was investigated under the following conditions: 190 L feeding, 380 L reactor mixed liquid, temperature of 32 ºC, 12 L/min continuously aeration flow rate, one hour anaerobic protection period, pH set point of 6.0 and 4 min of sludge settling time. Two FR strategies were tested separately: 1) FR = 500 mL/min, so that the feeding period of 190 L centrate took 380 min (or 6.3 h), 2) FR = 190 L/min, so that the feeding 190L of centrate was rapidly completed within one minute.   The cycle lengths for both cycles with FR = 500 mL/min and FR = 190 L/min were 21.5 h, indicating the reaction speed was almost the same for two tested cycles. For the cycle with FR = 190 L/min, that 190L of centrate was fed to the reactor within one minute, and the pH in the reactor was increased from 6.0 to 7.39. Due to the denitrification process over the anaerobic period and air stripping by aeration, the pH was further increased from 7.39 to 7.68 on the first 4 hours of cycle. The pH then kept dropping and ended with 6.0, due to the alkalinity consumption by the CANON process. For the cycle with FR = 500 mL, the centrate was slowly pumped into the reactor within 380 min (or 6.3 h); this lead to a pH increase from 6.0 at the beginning of the cycle to 7.65 over 6 hours. As presented in Figure 4.16, the pH decreasing curve of the cycle with FR = 500 mL/min almost matched the pH curve of FR = 190 L/min, after 6 hours of reaction, when centrate feeding was completed. This proves that the CANON reaction was not influenced by substrate feeding rate.  For the performance of ammonia removal within the two cycles, effluent ammonia-nitrogen concentrations were also very close to each other as presented in Figure 4.17. NH3-N concentration was reduced from about 455 mg/L to 72.4 mg/L for the cycle with FR = 190 L/min. For the cycle with FR = 500 mL/min, NH3-N concentration increased from 69.3 mg/L to about 338 mg/L, during the first 6 hours of feeding, and decreased to 70.2 mg/L at the end of cycle, which was only 2.2 mg/L less than the effluent NH3-N concentration of the fast feeding one. This means that the feeding rate had almost no effect on effluent NH3-N concentration or nitrogen removal ability of the CANON SBR.       70  Figure 4.16 Comparison of the effects of feeding rate on the variation of pH in the reactor    Figure 4.17 Comparison of the effects of feeding rate on the variation of ammonia-nitrogen concentration in the reactor within cycles  71 As for nitrite generation in the reactor, NO2-N concentrations were much more stable during the slow feeding cycle with FR = 500 mL/min, than in the fast feeding cycle with FR = 190 L/min. As shown in Figure 4.18, NO2-N concentrations varied around 2.0 mg/L within cycle of FR = 500 mL/min and ended at 1.8 mg/L. For the cycle with fast feeding rate of 190/min, NO2-N concentration was increased drastically from 1.4 mg/L to 4.0 mg/L after first three hours of aeration. This was probably because of the partial nitrification process, where, AOB converted ammonia to nitrite and was stimulated at higher initial ammonia concentrations in the reactor. However, due to consumption of anammox bacteria, nitrite concentration decreased to 1.6 mg/L at the end of the cycle. Thus, no nitrite accumulation would occur in the reactor.   Figure 4.18 Comparison of the effects of feeding rate on the variation of nitrite-nitrogen concentration in the reactor within cycles  The variation of nitrate concentrations within cycles is shown in Figure 4.19. The effect of feeding rate on nitrate generation was also not obvious. With a feeding 190 L of centrate at 190 L/min, NO3-N in the reactor was instantly diluted to 20.2 mg/L, and then slowly increased to 65.8 mg/L at the end of cycle. The N/A ratio was 0.12, which was very close to the theoretical N/A ratio of 0.11, indicating almost all nitrate was generated by the CANON process. With a slow feeding rate of 500 mL/min, NO3-N concentration was slowly diluted from 64.4 mg/L to  72 32.0 mg/L over 9 hours, and then increased to 64.4 mg/L at the end of the cycle, which is very close to the effluent NO3-N concentration of cycle feed with FR = 190 L/min.   Figure 4.19 Comparison of the effects of feeding rate on the variation of nitrate-nitrogen concentration in the reactor within cycles  It could therefore be concluded that feeding rate had almost no effect on the overall performance of CANON SBR. Regardless of fast feeding rate (190 L/min) or slow feeding rate (500 mL/min), both cycles were completed within the same time, meaning the HRTs of both cycles were the same. In addition, ammonia, nitrite and nitrate concentrations in the effluents of both cycles were very close to each other. The concentration variation curves also matched to some degree, indicating the reaction speed of the CANON process was actually not affected by centrate feeding speed. However, nitrite concentrations were more stable at lower FR = 500 mL/min than higher FR = 190 L/min. A slow feeding rate might be more optimal for a CANON SBR, running under conditions like lower temperature or higher aeration airflow rate, that may tend to produce higher concentration peaks of nitrite within cycles and cause nitrite accumulation problems.   73 4.6 Effects of sludge settling time Since anammox bacteria are slow growth bacteria and, thus, must remain in the reactor, the sludge settling time should be long enough for most of anammox sludge granules to settle to the bottom of the reactor. At the same time, nitrifiers grow much faster than anammox bacteria. Thus, the over populated AOB and NOB should be wasted through decanting at the end of each cycle. Adjusting the sludge settling time could effectively change TSS and VSS of both the effluent and reactor mix liquor. Three different sludge settling time settings of 10 min, 8 min, and 4 min were tested over the periods of sludge enrichment and system optimization. Settling time shorter than 4 min would lead to the loss of anammox granular sludge, due to insufficient settling of anammox.    Figure 4.20 Comparison of the effects of sludge settling time on TSS and VSS in the effluent  As shown in Figure 4.20, average TSS and VSS in the effluent increased (an expected result) with the reduction of sludge settling time, indicating more nitrifiers would be wasted through decant with longer sludge settling time. With a sludge settling time of 10 min, a minimum amount of solids were wasted. Effluent TSS ranged from 53.3 mg/L to about 276 mg/L, with an  74 average value of about 172 mg/L. Most solids in the effluent were volatile, since the VSS concentrates were close to TSS, ranging from 40 mg/L to about 226.7 mg/L, with an average value of about 133 mg/L. With a sludge settling time of 8 min, average TSS and VSS in the effluent was slightly increased to about 328 mg/L and 271 mg/L, respectively. For the shortest sludge settling time of 4 min, effluent TSS and VSS increased drastically to about 873 mg/L and 743 mg/L, respectively.   Figure 4.21 Comparison of the effects of sludge settling time on TSS and VSS in the reactor   Sludge settling time also affected TSS and VSS in the reactor. As shown in Figure 4.21, TSS and VSS in the reactor decreased with sludge settling time reduction. At a sludge settling time of 10 min, TSS and VSS concentrations in the reactor were maintained at the highest levels. TSS varied from 2800 mg/L to 5275 mg/L, with an average value of 3828 mg/L and VSS ranged from 2425 mg/L to 4100 mg/L with average value of 3231 mg/L. Since almost no sludge was wasted at a sludge settling time of 10 min, sludge enrichment could be achieved within the shortest of time. With a sludge settling time of 8 min, TSS and VSS in the reactor were slightly reduced to 3256 mg/L and 2797 mg/L, respectively. The average TSS and VSS concentration in the reactor were further reduced to 2560 mg/L and 2212 mg/L with a sludge settling time of 4 min. Since most of granular anammox sludge settled within 4 min and could be maintained in the reactor, the reduction of TSS and VSS was achieved by wasting nitrifiers.  75  5. Summary and Conclusions In this research, a 400 L CANON SBR pilot system was built and tested for treating on-site, digester centrate with high ammonium concentrations. The reactor was able to maintain over 90% ammonia removal rates during the test period of 318 days. Concentrations of nitrite- nitrogen in the effluent of the reactor were kept below 5.0 mg/L, which was a safety level that would not inhibit the growth of anammox bacteria.   The reactor was inoculated with digested sludge. Mature anammox sludge was later added for a faster start up.  It took 115 days to achieve a complete start-up of the reactor with an ammonia removal rate (ARR) of 94.7% and HRT of 5.2 d. The start-up period could have been much shorter if a failure event had not occurred in the reactor. Since the growth rate of anammox bacteria was low, TSS and VSS concentration was still as low as 1509 mg/L and 1345 mg/L until the end of the start-up period. Thus, a sludge enrichment period was necessary to increase TSS and VSS concentration in the reactor for better reactor performance. After 33 days of sludge enrichment, TSS and VSS levels increased to 4050 mg/L and 3175 mg/L, respectively. At the same time, red granular anammox sludge was observed in the reactor, which was good for sludge settling and resistance to shock loading. Due to the granulation of anammox sludge, the ammonia loading rate (ALR) was increased from 0.17 kgN/m3•d to 0.45 kgN/m3•d, without ARR decreasing. During the same period of time, the cycle length and HRT of the reactor was reduced to 10 h and 2.08 d, respectively.   It was found out that the quality of on-site centrate, especially the sludge solids concentration in the centrate, was critical to affect the performance of the reactor. Solids in the centrate would introduce extra digested sludge into the reactor, and disturb the existing balance among bacteria groups within the reactor. Continuously feeding centrate, with a 1~2 mm sludge layer, reduced ARR significantly from 96.6% to 78.2%, since anammox bacteria were outcompeted by exogenous bacteria coming from sludge in the centrate. The ARR was recoverable when the sludge layer was removed from the feeding centrate. After 13 days of continuous feeding of clear centrate, the ARR increased to 91.7%, since the digested sludge was probably washed out from the reactor through decanting. Thus, for centrate containing too many sludge solids, pre- treatment for sludge removal might be necessary before centrate could be fed to the CANON SBR.  76  Through process kinetic studies of the CANON SBR under variety environmental conditions, aeration was found to be one of the most important factors that control the reactor performance. Higher ARR and lower HRT of the reactor could be achieved through increasing the airflow rate. Compared to an aeration airflow rate of 10 L/min, a 40% airflow increase, from 10L/min to 14 L/min, would lead to a 9% ARR increase and a 10% HRT reduction.   Temperature was also found to be a critical control factor for the CANON SBR. The highest ARR in the CANON SBR was achieved at 32 ºC, compared with lower tested temperatures of 30 ºC, 28 ºC and 26 ºC. At the lowest tested temperature of 26 ºC, ARR was reduced by 12 % and peak nitrite concentrations of 15.7 mg/L was observed in the reactor. The additional nitrite , which could not be completely consumed by anammox bacteria, might be utilized by NOB for nitrate production, stimulating the growth of NOB and disturbing the balance among microorganisms in the reactor.   Except for raising the temperature of the reactor, better CANON SBR performance was also observed at higher initial ammonium concentration in the reactor. Among tested initial ammonia concentrations of 215 mg/L, 285 mg/L and 455 mg/L, a cycle with NH3-Ni = 455 mg/L achieved the highest NRR of 0.36 kgN/m3•d and lowest HRT of 1.79 d; this indicated that both nitrification and ANAMMOX process were stimulated and more nitrogen could be removed under higher initial ammonia concentrations.   The effect of feeding rate was also studied but no significant difference was observed. The cycle length or HRT was the same, no matter if the feeding rate was 190 L centrate at 190 L/min or at 500 mL/min, indicating that the feeding rate did not affect the reaction speed. In addition, the effluent ammonia and nitrate concentrates were similar for both cycles tested under the two different feeding rates. However, the reactor under a fast feeding rate of 190 L/min, tended to generate more nitrite than a slow FR = 500 ml/min. A high peak concentration of nitrite was observed during the cycle under FR = 190 L/min, while nitrite concentrations remained relatively constant (around 2.0 mg/L) during the cycle with FR = 500 mL/min. Thus, when the CANON SBR was operated under low temperature or high aeration rate conditions, that tended to generate high peak concentrations of nitrite, the operational strategy of a slow feeding rate might be preferred.   77  Changing the sludge settling time of the CANON SBR was found to be an effective biomass wasting strategy for solid concentrations (TSS and VSS) control in the reactor. Adjusting the sludge settling time from 4 min to 10 min could effectively control the amount of nitrifying bacteria in the reactor, without losing the anammox granular sludge. For the purpose of reactor start-up and sludge enrichment, the sludge settling time could be set at 10 min or longer, allowing most of the sludge to remain in the reactor. During the sludge enrichment period, with a sludge settling time of 10 min, relatively high TSS over 5000 mg/L and VSS over 4000 mg/L was achieved in the reactor. With a sludge settling time of 4 min, during system optimization period, TSS and VSS in the reactor dropped below 2500 mg/L.                         78 6. Recommendations for Future Research Since anammox is novel technology in nitrogen removal field, more research needs to be done to make it more applicable. Following are some recommendations for future research:   • Since anammox bacteria grow slowly, establishing proper kinetic models for CANON SBR process could be helpful to do better system optimization.  • The function of bacteria in the CANON SBR, including anammox bacteria, ammonia oxidation bacteria, and nitrite oxidation bacteria, could be further investigated. Knowing how to control the activities of these bacteria in the CANON SBR could be very helpful.  • Future research could also try to combine the CANON SBR with other treatment facilities, like MBR, to achieve further nitrogen removal and phosphorus removal.            79 References Alfred, F., and Culp, G. L. (1967). Water Reclamation Studies at the South Tahoe Public Utility District. Water Environmental Federation, 39, 787–814. Amann, R. I., Ludwig, W., and Schleifer, K.-H. (1995). Phylogenetic Identification and On-site Detection of Individual Microbial Cells without Cultivation. Microbiological Reviews, 59(1), 143–169. Arrojo, B., Mosquera-Corral, A., Campos, J. L., and Méndez, R. (2006). Effects of mechanical stress on anammox granules in a sequencing batch reactor (SBR). Journal of Biotechnology, 123(4), 453–463. Astrid, A., Van de, G., Peter, B., Lesley, A., Mike, S. M., J., and Gijs, K. (1997). Metabolic pathway of anaerobic ammonium oxidation on the basis of 15N studies in a fludized bed reactor. Microbiology, 145, 2415–2421. Bettazzi, E., Caffaz, S., Vannini, C., and Lubello, C. (2010). Nitrite inhibition and intermediates effects on anammox bacteria: A batch-scale experimental study. Process Biochemistry, 45(4), 573–580.  Broda E. (1977). Two kinds of lithotrophs missing in nature. Microbiol., 17(6), 497~493.  Chamchoi, N., and Nitisoravut, S. (2007). Anammox enrichment from different conventional sludges. Chemosphere, 66(11), 2225–2232.  Dapena-Mora, A., Fernández, I., Campos, J. L., Mosquera-Corral, A., Méndez, R., and Jetten, M. S. M. (2007). Evaluation of activity and inhibition effects on ANAMMOX process by batch tests based on the nitrogen gas production. Enzyme and Microbial Technology, 40(4), 859–865.  Dapena-Mora, A., Campos, J. L., Mosquera-Corral, A., Jetten, M S M, and Méndez, R. (2004). Stability of the ANAMMOX process in a gas-lift reactor and a SBR. Journal of Biotechnology, 110(2), 159–170.  Egli, K., Fanger, U., Alvarez, P., Siegrist, H., van der Meer, J. R., and Zehnder, A. J. B. (2001). Enrichment and characterization of an anammox bacterium from a rotating biological contactor treating ammonium-rich leachate. Archives of Microbiology, 175(3), 198–207.  Fernández, I., Vázquez-Padín, J. R., Mosquera-Corral, A., Campos, J. L., and Méndez, R. (2008). Biofilm and granular systems to improve anammox biomass retention. Biochemical  80 Engineering Journal, 42(3), 308–313.  Fux, C., Boehler, M., Huber, P., Brunner, I., and Siegrist, H. (2002). Biological treatment of ammonium-rich wastewater b partial nitritation and subsequent anaerobic ammonium oxidation (anammox) in a pilot plant. Biotechnology, 99, 295–306. Gali, A., Dosta, J., Vanloosdrecht, M., and mataalvarez, J. (2007). Two ways to achieve an anammox influent from real reject water treatment at lab-scale: Partial SBR nitrification and SHARON process. Process Biochemistry, 42(4), 715–720.  Gong, Z., Yang, F., Liu, S., Bao, H., Hu, S., and Furukawa, K. (2007). Feasibility of a membrane-aerated biofilm reactor to achieve single-stage autotrophic nitrogen removal based on anammox. Chemosphere, 69(5), 776–784.  Hellinga, C., Schellen, A. A. J. C., Mulder, J. W., van Loosdrecht, M. C. M., and Heijnen, J. J. (1998). The SHARON process: an innovative method for nitrogen removal from ammonium-rich waste water, 1–8. Isaka, K., Suwa, Y., Kimura, Y., Yamagishi, T., Sumino, T. and Tsuneda, S. (2008). Anaerobic ammonium oxidation (anammox) irreversibly inhibited by methanol. Microbiol Biotechnol, 81, 379–385. Jetten, Mike SM, Wagner, M., Fuerst, J., van Loosdrecht, M., Kuenen, G., and Marc, S. (2001). Microbiology and application of the anaerobic ammonium oxidation (“ANAMMOX”) process. Environmental Biotechnology, 12, 283–288. Jin, R. C., Zheng, P., Hu, A. H., Mahmood, Q., Hu, B. L., and Jilani, G. (2008a). Performance comparison of two ANAMMOX reactors: SBR and UBF. Chemical Engineering Journal, 138(1-3), 224–230.  Jin, R. C., Hu, B. L., Zheng, P., Qaisar, M., Hu, A. H., and Islam, E. (2008b). Quantitative comparison of stability of ANAMMOX process in different reactor configurations. Bioresource Technology, 99(6), 1603–1609.  Johnson, W. K., and Schroepfer, G. J. (1964). Nitrogen Removal by Nitrification and Denitrification. Water Pollution Control Federation, 36(8), 1015–1036. Jung, J. Y., Kang, S. H., Chung, Y. C., and Ahn, D. H. (2007). Factors affecting the activity of anammox bacteria during start up in the continuous culture reactor. Water Science and Technology, 55(1-2), 459–468.  Khin, T., and Annachhatre, A. P. (2004). Novel microbial nitrogen removal processes. Biotechnology Advances, 22(7), 519–532.  Kim, J. H., Chen, M., Kishida, N., and Sudo, R. (2004). Integrated real-time control strategy for  81 nitrogen removal in swine wastewater treatment using sequencing batch reactors. Water Research, 38(14–15), 3340–3348. Kim, I., Lee, H. H., Chung, Y. C., and Jung, J. Y. (2009). High-strength nitrogenous wastewater treatment in biofilm and granule ANAMMOX processes. Water Science and Technology, 60(9), 2365.  Kuai, L. and Verstraete, W. (1998). Ammonium removal by the oxygen-limited autotrophic nitrification-denitrification system. Applied and Environmental Microbiology, 64(11), 4500– 4506. Liang, Z., and Liu, J. (2008). Landfill leachate treatment with a novel process: Anaerobic ammonium oxidation (ANAMMOX) combined with soil infiltration system. Journal of Hazardous Materials, 151(1), 202–212.  Lindsay, M., Webb, R., Strous, M., Jetten, M., Butler, M., Forde, R., and Fuerst, J. (2001). Cell compartmentalisation in planctomycetes: novel types of structural organisation for the bacterial cell. Archives of Microbiology, 175(6), 413–429.  López, H., Puig, S., Ganigué, R., Ruscalleda, M., Balaguer, M. D., and Colprim, J. (2008). Start- up and enrichment of a granular ANAMMOX SBR to treat high nitrogen load wastewaters. Journal of Chemical Technology and Biotechnology, 83(3), 233–241. Mulder, A., van de Graaf, A. A., Robertson, L. A., and Kuenen, J G. (1995). Anaerobic ammonium oxidation discovered in a denitrifying fluidized bed reactor. Microbiology Ecology, 16, 177–184. Ni, S. Q., and Meng, J. (2011). Performance and inhibition recovery of ANAMMOX reactors seeded with different types of sludge. Water Science and Technology, 63(4), 710.  Ni, S. Q., Gao, B.Y., Wang, C. C., Lin, J. G., and Sung, S. (2011). Fast start-up, performance and microbial community in a pilot-scale ANAMMOX reactor seeded with exotic mature granules. Bioresource Technology, 102(3), 2448–2454. Elsevier Ltd.  Niftrik, L. A., Fuerst, J. A., DamstÃ, J. S. S., Kuenen, J. G., Jetten, M. S. M., and Strous, M. (2004). The anammoxosome: an intracytoplasmic compartment in anammox bacteria. FEMS Microbiology Letters, 233(1), 7–13. Penton, C. R., Devol, A. H., and Tiedje, J. M. (2006). Molecular Evidence for the Broad Distribution of Anaerobic Ammonium-Oxidizing Bacteria in Freshwater and Marine Sediments. Applied and Environmental Microbiology, 72(10), 6829–6832. Pynaert, K., Smets, B. F., Wyffels, S., beheydt, D., Siciliano, S. D., and Verstraete, W. (2003). Characterization of an autotrophic nitrogen removing biofilm from a highly loaded lab-scale  82 rotating contactor. Applied and Environmental Microbiolgy, 69(6), 3626–3635.  Reeves, T. G. (1972). Nitrogen removal: A literature review. Water envronmental federation, 44(10), 1895–1908. Robert V. T. and Russo, R.C. (1981). Ammonia Toxicity to Fishes. Effect of pH on the toxicity of the un-ionized ammonia species. American Chemical Society, 15 (7), 837–840. Schmid, M. C., Maas, B., Dapena, A., van de Pas-Schoonen, K., van de Vossenberg, J., Kartal, B., van Niftrik, L., et al. (2005). Minireview: Biomarkers for on-site detection of anaerobic ammonium oxidizing (ANMMOX) bacteria. Applied and Environmental Microbiology, 71(4), 1677–1684.  Schmidt, I., Sliekers, O., Schmid, M., Bock, E., Fuerst, J., Kuenen, J. G., Jetten, M. S. M. (2003). New concepts of microbial treatment processes for the nitrogen removal in wastewater. FEMS Microbiology Reviews, 27(4), 481–492. Sliekers, A. O., Derwort, N., Gomez, J. L. C., Strous, M., Kuenen, J. G., and Jetten, M. S. M. (2002). Completely autotrophic nitrogen removal over nitrite in one single reactor. Water Research, 36, 2475–2482. Sliekers, A. O., Third, K. A., Abma, W., Kuenen, J. G., and Jetten, M. S. M. (2003). CANON and ANAMMOX in a gas-lift reactor. FEMS Microbiology Letters, 218(2), 339–344.  Strous, M., Heijnen, J. J., Kuenen, J. G., and Jetten, M. S. M. (1998). The sequencing batch reactor as a powerful tool for the study of slowly growing anaerobic ammonium-oxidizing microoganisms. Appl Microbiol Biotechnol, 50, 589–596. Strous, M., Kuenen, J. G., and Jetten, M.S.M. (1999). Key Physiology of Anaerobic Ammonium Oxidation. Applied and Environmental Microbiology, 65(7), 3248–3250. Strous, M., van Gerven, E., Kuenen, J. G., and Jetten, M. S. M. (1997a). Effects of Aerobic and Microaerobic Conditions on Anaerobic Ammonium-Oxidizing (ANAMMOX) Sludge. Applied and Environmental Microbiology, 2446–2448. Strous, M., van Gerven, E., Kuenen, J. G., and Jetten, M .S. M.  (1997b). Ammonium removal from concentrated waste streams with the anaerobic ammonium oxidation (ANAMMOX) process in different reactor configurations. Water Research, 31(8), 1955–1962. Tang, C. J., Zheng, P., Wang, C. H., Mahmood, Q., Zhang, J. Q., Chen, X. G., Zhang, L., (2011). Performance of high-loaded ANAMMOX UASB reactors containing granular sludge. Water Research, 45(1), 135–144.  Third, K. (2001). The CANON System (Completely Autotrophic Nitrogen-removal Over Nitrite) under Ammonium Limitation: Interaction and Competition between Three Groups of  83 Bacteria. Systematic and Applied Microbiology, 24(4), 588–596.  Third, K. A., Paxman, J., Schmid, M., Strous, M., Jetten, M. S. M., and Cord-Ruwisch, R. (2005). Enrichment of anammox from Activated Sludge and Its Application in the CANON Process. Microbial Ecology, 49(2), 236–244. Toh, S. K., and Ashbolt, N. J. (2002). Adaptation of anaerobic ammonium-oxidising consortium to synthetic coke-ovens wastewater. Applied Microbiology and Biotechnology, 59(2-3), 344– 352.  Toh, S. K., Webb, R. I., and Ashbolt, N. J. (2002). Enrichment of Autotrophic Anaerobic Ammonium-Oxidizing Consortia from Various Wastewaters. Microbial Ecology, 43(1), 154–167.  Tran, H. T., Park, Y. J., Cho, M. K., Kim, D. J., and Ahn, Dae-Hee. (2006). Anaerobic Ammonium Oxidation Process in an Upflow Anaerobic Sludge Blanket Reactor with Granular Sludge Selected from an Anaerobic Digestor. Biotechnology and Bioprocess Engineering, 11, 199–204. van de Graaf, Astrid A., Mulder, A., Peter, D., Jetten, M. S. M., Robertson, L. A., and Kuenen, J. G. (1995). Anaerobic Oxidation of Ammonium is a Biologically Mediated Process. Applied and Environmental Microbiology, 61(4), 1247–1251. van der Star, W. R. L., Abama, W. R., Blommers, D., Mulder, J. W., Tokutomi, T., Strous, M., Picioreanu, C., et al. (2007). Startup of reactors for anoxic ammonium oxidation: Experiences from the first full-scale ANAMMOX reactor in Rotterdam. Water Research, 41, 4149–4163.  van Dongen, U., Jetten, M S M, and van Loosdrecht, M. C. M. (2001). SHARON-ANAMMOX process for treatment of ammonium rich. Water Sciencen and Technology, 44(1), 153–160. van Kempen, R., J W, M., C A, U., and M C M, L. (2001). Overview: full scale experience of the SHARON process for treatment of rejection water of digested sludge dewatering. Water Science and Technology, 44(1), 145–152. Vit, M. (1992). Biological water denitrification--A review. Enzyme Microb Technol, 14, 170– 183. Vleaminch, S. E., Cloetens, L. F. F., Clippeleir, H. De, Carballa, M. and Verstraete, W. (2009). Granular biomass capable of partial nitrification and anammox. Water Science and Technology, 58(5), 1113–1120. Wu, C. Chen, Z., Liu, X., and Pend, Y. (2007). Nitrification-denitrification via nitrite in SBR using real-time control strategy when treating domestic wastewater. Biochemical  84 Engineering Journal, 36(2), 87–92. Yu, R., Liaw, S., Cheng, W. and Chang, C. (2000). Performance enhancement of SBR applying real-time control. Journal of Environmental Engineering, 126(10), 943–948. Zhang, L., Zheng, P., Tang, C. J., and Ren-cun, J. (2008). Anaerobic ammonium oxidation for treatment of ammonium-rich wastewaters. Journal of Zhejiang University SCIENCE B, 9(5), 416–426.  Zheng, P., Lin, F. M., Hu, B. L., and Chen, J. S. (2004). Performance of anammox granular sludge bed reactor started up with nitrifying granular sludge. Journal of Environmental Sciences, 16(2), 339–342.  85 Appendices Appendix A: Data of reactor performance Date Day Inf NH4, mg/L Inf NO2, mg/L Inf NO3, mg/L Eff NH4, mg/L Eff NO2, mg/L Eff NO3, mg/L Re TSS, mg/L Re VSS, mg/L Cycle length, h 25-Aug-11 76 1107.675 0.414 0.285 116.225 0.158 111.892       26-Aug-11 77                   27-Aug-11 78 1141.5 0.378 0 130.875 0.158 135.592 2230 1875 46 28-Aug-11 79                   29-Aug-11 80 1107.975 0.386 0 152.525 0.159 129.791 2128.6 1957.1 46 30-Aug-11 81 1042.65 0.419 0 163.475 0.161 153.739 3000 1928.6   31-Aug-11 82                   1-Sep-11 83 1068.45 0.556 0.239 195.675 0.163 168.037 1855.6 1355.6 48 2-Sep-11 84 1093.725 0.4 0 196.975 0.164 175.836 2414.3 2142.9 30 3-Sep-11 85 1028.225 0.419 0 194 0.175 179.925     26.5 4-Sep-11 86                   5-Sep-11 87 1008.625 0.375 0 206.225 0.178 189.572     38.3 6-Sep-11 88 994.275 0.42 0 213.475 0.17 186.78 1500 1200 25.5 7-Sep-11 89 967.425 0.468 0 199.175 0.549 182.351 1980 1730 26 8-Sep-11 90                 22 9-Sep-11 91                 34 10-Sep-11 92                   11-Sep-11 93                   12-Sep-11 94                   13-Sep-11 95             1045.45 809.09   14-Sep-11 96                   15-Sep-11 97 924.5 0.048 0.254 44.01 0.55 28.18 593.3 513.3 47 16-Sep-11 98 955.4 0.275 0.026 60.31 0.58 45.11     31.5 17-Sep-11 99                   18-Sep-11 100 953.2 0.244 0.089 71.1 0.62 55.91 782.1 667.9 36 19-Sep-11 101 890.45 0.288 0.018 88.4 0.65 40.5 675 600 24 20-Sep-11 102 1020.1 0.198 0.022 85.2 0.96 72.58 954.54 763.63 26.5 21-Sep-11 103 917.35 0.247 0.004 83.86 0.52 73.99 1250 1183.3 25 22-Sep-11 104 1049.2 0.204 0.14 93.2 0.3 79.97 1245.45 1127.3 21 23-Sep-11 105 1002.4 0.14 0.325 90.13 0.32 84.99     22 24-Sep-11 106 949.35 0.106 0.062 84.49 0.15 83.31     23.5 25-Sep-11 107                 24 26-Sep-11 108 1083 0.227 0.464 76.25 0.38 78.73 1572.7 1463.6 24 27-Sep-11 109 1007.45 0.273 0.381 79.97 0.38 75.64     24 28-Sep-11 110 1010.15 0.329 0.428 74.63 0.36 73.95 1909.1 1709.1 24.5 29-Sep-11 111 1036.8 0.434 0.697 60.86 0.59 59.41     23 30-Sep-11 112 840.05 0.226 0.026 74.06 0.28 67.29 1791.6 1616.7 22 1-Oct-11 113 885.9 0.151 0.089 60.66 0.37 65.94     22.5 2-Oct-11 114 901.1 0.124 0.14 53.39 0.39 59.39     25 3-Oct-11 115 900.35 0.316 0.05 47.46 0.46 55.44 1509.1 1345.5 25 4-Oct-11 116 873.25 0.543 0.3 46.55 0.27 48.08     19 5-Oct-11 117 789.5 0.729 0.27 42.75 0.33 48.86 1623.1 1438.5 18 6-Oct-11 118 859.3 0.177 0.1 48.14 0.39 50.69     19 7-Oct-11 119 789.7 0.082 0 47.33 0.39 52.72 1007.1 900 19  86 Date Day Inf NH4, mg/L Inf NO2, mg/L Inf NO3, mg/L Eff NH4, mg/L Eff NO2, mg/L Eff NO3, mg/L Re TSS, mg/L Re VSS, mg/L Cycle length, h 8-Oct-11 120 790.8 0.083 0.02 36.97 0.4 48     20 9-Oct-11 121 840.5 0.087 0.08 43.23 0.67 45.67     20 10-Oct-11 122 830.5 0.119 0.37 38.19 0.4 44.07 2133.3 1766.6 20 11-Oct-11 123 779.05 0.097 0.04 39.69 0.38 43.65     19 12-Oct-11 124 782.75 0.088 0 36.52 0.4 49.04 2200 2069.2   13-Oct-11 125 778.5 0.106 0 33.86 0.4 45.23     19 14-Oct-11 126 825.85 0.082 0 36.54 0.39 47.53 2277.8 1966.7 19 15-Oct-11 127 713.1 0.123 0.03 32.66 0.38 48.6     19 16-Oct-11 128                   17-Oct-11 129 923.25 0.313 0.063 42.08 0.46 39.04 2985.7 2671.4 19 18-Oct-11 130 936.4 0.142 0.063 43.84 0.34 41.3       19-Oct-11 131 977.55 0.148 0 45.94 0.46 47.14 2750 2375 17 20-Oct-11 132 907.65 0.437 0 57.34 0.83 53.32     16 21-Oct-11 133 923.6 0.611 0.023 52.78 0.44 60.22 3062.5 2637.5 16 22-Oct-11 134 904.7 1.409 0.04 46.24 0.35 53.89     16 23-Oct-11 135                   24-Oct-11 136 894.6 0.232 0 40.38 0.92 37.76 3033.3 2677.8 17 25-Oct-11 137 856.5 0.318 0 41.73 1.75 33.12     16 26-Oct-11 138 779.9 0.169 0.086 46.76   52.09 3055.6 2577.8 14.5 27-Oct-11 139 805.5 0.271 0.095 35.61 2.11 51.88     14.5 28-Oct-11 140 868.45 0.072 0.034 38.09 5.64 45.49 2666.7 2311.1 14.5 29-Oct-11 141 802.75 0.767 0     53.23     14.5 30-Oct-11 142                   31-Oct-11 143 793.8 2.712 1.345 43.13 0.14 40.73 3155.6 2777.8 12.5 1-Nov-11 144 784.5 3.279 0.119 38.96 0 39.59     12.5 2-Nov-11 145 971.95 0.05 0.01 68 0.38 40.95     12.5 3-Nov-11 146 935.85 0.078 0 56.33 0 33.31     12.5 4-Nov-11 147 972.05 0.095 0 52.83 0.04 30.16 4050 3175 12 5-Nov-11 148 936.15 0.154 0 67.06 0 35.01     10 6-Nov-11 149                   7-Nov-11 150 905.25 0.376 0 77.89 2.22 52.74 4975 3975 10 8-Nov-11 151 867.05 0.285 0.06 46.89 2.96 31.5     10 9-Nov-11 152 1095.5 0.339 0 60.82 1.94 43.29 5275 4100 10.5 10-Nov-11 153 892.1 0.642 0 52.34 1.41 40.36     10.5 11-Nov-11 154 937.9 0.302 0 56.44 1.68 35.95 3666.7 3058.3 10.5 12-Nov-11 155                   13-Nov-11 156                   14-Nov-11 157 931.9 0.531 0.155 61.2 6.75 37.87 3942.1 3545.5 9.5 15-Nov-11 158 921 0.651 0.04 62.51 5.13 41.72     9.5 16-Nov-11 159 983.75 0.191 0.006 58.84 3.93 39.49 2441.7 1983.3 9.5 17-Nov-11 160 1006.6 0.214 0.032 60.18 3.07 40.44     9.5 18-Nov-11 161 994.45 0.16 0 48.79 2.43 30.36 3450 2912.5 9.5 19-Nov-11 162                   20-Nov-11 163                   21-Nov-11 164                   22-Nov-11 165                   23-Nov-11 166                   24-Nov-11 167 851 0.171 3.994 9.57 1.38         25-Nov-11 168                   26-Nov-11 169                   27-Nov-11 170                   28-Nov-11 171 856.1 0.487 3.348 28.97 2.04 22.06 3800 3225               87 Date Day Inf NH4, mg/L Inf NO2, mg/L Inf NO3, mg/L Eff NH4, mg/L Eff NO2, mg/L Eff NO3, mg/L Re TSS, mg/L Re VSS, mg/L Cycle length, h 29-Nov-11 172 820.1 0.377 2.584 38.43 2.01 29.08    30-Nov-11 173                   1-Dec-11 174 921.4 0.512 2.404 54.05 2.06 44.31 2800 2480 10.5 2-Dec-11 175 791.8 0.766 2.283 49.24 2.67 38.42     11 3-Dec-11 176                   4-Dec-11 177                   5-Dec-11 178 865.4 2.752 2.614 58.32 2.25 49.22 2255.6 2022.2 11 6-Dec-11 179                   7-Dec-11 180 907.2   2.945 57.36 5.9 50.79 2887.5 2425 11 8-Dec-11 181                   9-Dec-11 182 885.5   2.927 58.81 6 51.27 2455.6 2066.7 11 10-Dec-11 183                   11-Dec-11 184                   12-Dec-11 185 809.75 1.932 0.498 63.45 0.84 55.44 3314.3 2700   13-Dec-11 186                   14-Dec-11 187 735.3 1.437 0.165 55.75 0.91 50.21       15-Dec-11 188                   16-Dec-11 189 816.15 0.923 0.091 57.58 0.71 48.57 3242.9 2871.5 12 17-Dec-11 190                   18-Dec-11 191                   19-Dec-11 192 786.4 0.632 0.01 56.59 0.52 48.42 2215.4 1976.3 12 20-Dec-11 193 829.35 0.573 0.076 63.86 0.58 57.39     11.5 21-Dec-11 194 825.1 0.766 0.1 68.21 3.29 58.66 3600 3077.8 11 22-Dec-11 195                   23-Dec-11 196 841.65 0.682 0.15 61.26 0.57 55.02       24-Dec-11 197                   25-Dec-11 198                   26-Dec-11 199                   27-Dec-11 200 865.05 0.964 0.107 77.37 0.73 67.87 2633.3 2300   28-Dec-11 201                   29-Dec-11 202                   30-Dec-11 203                   31-Dec-11 204 902.95 0.717 0.003 71.26 0.62 62.04       1-Jan-12 205                   2-Jan-12 206 929.75 0.225 0.002 144.12 0.87 127.78 4316.7 3683.3   3-Jan-12 207                   4-Jan-12 208 945.8 0.805 0.153 165.17 0.35 126.62     12 5-Jan-12 209 946.7 0.35 0.091 184.29 0.33 128.57     12 6-Jan-12 210 955.3 0.419 0.067 188.82 0.39 128.42 4532 3990 12 7-Jan-12 211                   9-Jan-12 212 941.55 1.026 0.258 205.15 0.39 132.29       10-Jan-12 213                   11-Jan-12 214 961.55 0.094 0.058 209.86 0.44 119.57       12-Jan-12 215 791 0.501 0.038 109.87 0.26 140.22     11 13-Jan-12 216 841.75 0.221 0.036 110.36 0.33 119.49 4250 3700 10.5 14-Jan-12 217                   15-Jan-12 218                   16-Jan-12 219                   17-Jan-12 220                   18-Jan-12 221                   19-Jan-12 222                   20-Jan-12 223                   21-Jan-12 224                    88 Date Day Inf NH4, mg/L Inf NO2, mg/L Inf NO3, mg/L Eff NH4, mg/L Eff NO2, mg/L Eff NO3, mg/L Re TSS, mg/L Re VSS, mg/L Cycle length, h 22-Jan-12 225                   23-Jan-12 226 918.3 0.055 0.19 75.99 0.12       16 24-Jan-12 227 865.85 0.087 0.137 134.36 0.13 109.26     15 25-Jan-12 228                   26-Jan-12 229                   27-Jan-12 230 995.7 0.642 0.157 151.92 0.18 104.5 4037.5 3275 11 28-Jan-12 231                   29-Jan-12 232                   30-Jan-12 233 965.85 0.042 0 133.04 0.12 87.72 3187.5 2762.5 9.5 31-Jan-12 234                   1-Feb-12 235 1040.6 0.118 0.065 136.28 0.075 87.7     10 2-Feb-12 236                 10.5 3-Feb-12 237 974.1 0.1 0.778 130.87 0 71.7 2850 2583.3 12 4-Feb-12 238                   5-Feb-12 239                   6-Feb-12 240 873.05 0.082 0.074 126.29 0.47 94.94 3180 2800 16 7-Feb-12 241                   8-Feb-12 242 887.45 2.853 0.062 129.12 0.13 96.94     15 9-Feb-12 243 876.15 0.659 0.229 136.82 0.04 103.26     14 10-Feb-12 244 821.95 1.005 0.07 111.8 0.06 84.09 2790 2160 15 11-Feb-12 245                   12-Feb-12 246                   13-Feb-12 247                   14-Feb-12 248 884 0.058 0.151 90.95 0.22 68.3 3620 3320 12 15-Feb-12 249 856.15 0.387 0.119 125.27 0.07 96.55     11 16-Feb-12 250 851.7 0.086 0.101 123.135 0.84 95.77     10 17-Feb-12 251 881.85 0.423 0.037 128.73 0.43 96.4 3500 3100 10 18-Feb-12 252                   19-Feb-12 253                   20-Feb-12 254 1031 0.183 0.126 124.9 0.48 77.35 4500 3885.7 10 21-Feb-12 255                   22-Feb-12 256 1082.8 0.107 0.107 110.66 1.54 81.4     15 23-Feb-12 257 1078.95 0.355 0.075 118.14 2.49 82.21     15 24-Feb-12 258 1101.4 0.841 0.099 117.15 2.42 82.36 4611.1 3944.4 14 25-Feb-12 259                   26-Feb-12 260                   27-Feb-12 261 1089.5 0.513 0.282 103.1 2.3 74.32 3777.7 3266.7 12 28-Feb-12 262                   29-Feb-12 263 1007.5 0.112 0.112 89 3.13 70.91     12 1-Mar-12 264 1020.5 0.128 0.128 96.05 0.39 71.75     12 2-Mar-12 265 966.75 0.295 0.236 87.86 2.92 68.42 3144.4 2388.9 12 3-Mar-12 266                   4-Mar-12 267                   5-Mar-12 268 1177 0.452 0.102 88.99 2.85 68.76     12 6-Mar-12 269                   7-Mar-12 270 1101 0.205 0.603 57.63 3.07 44.37       8-Mar-12 271 914.45 1.836 0.346 55.93 2.43 43.87       9-Mar-12 272 922.1 0.715 0.299 58.68 2.31 53.15 4050 3570   10-Mar-12 273                   11-Mar-12 274                   12-Mar-12 275 1029.75 0.428 0.428 68.8 2.58 59.11     21.5 13-Mar-12 276                   14-Mar-12 277                 24  89 Date Day Inf NH4, mg/L Inf NO2, mg/L Inf NO3, mg/L Eff NH4, mg/L Eff NO2, mg/L Eff NO3, mg/L Re TSS, mg/L Re VSS, mg/L Cycle length, h 15-Mar-12 278 782.4 1.46 0.37 90.36 3.465 67.56     24 16-Mar-12 279 740.63 1.33 0.762 75.25 3.7 77     26 17-Mar-12 280                   18-Mar-12 281                   19-Mar-12 282 816.2 0.46 0.26 72.39 2.316 73.07 3688.9 2911.1 24 20-Mar-12 283                   21-Mar-12 284 798.6 0.44 0.247 74 2.202 70.13     24 22-Mar-12 285 785.55 0.43 0.226 74.81 2.167 70.99     23 23-Mar-12 286 778.1 0.61 0.273 63.27 2.984 62.2 3143.7 2681.25 23 24-Mar-12 287                   25-Mar-12 288                   26-Mar-12 289 891.7 0.2 0 82.6 0.92   2561.5 2184.6 21 27-Mar-12 290                   28-Mar-12 291 859.8 0.8 0.27 72.55 0.92         29-Mar-12 292 862.6 4.5 0.38 65.675 0.8         30-Mar-12 293 889.5 2.2 0.38 72.475 0.8   2550 2280   31-Mar-12 294                   1-Apr-12 295                   2-Apr-12 296 927.6 0.24 0 80.55 1.06         3-Apr-12 297                   4-Apr-12 298 808.55 1.803 0.278 75.27 1.165 66.01 3030.7 2653.8   5-Apr-12 299 831.75 1.201 0.278 76.7 1.7 63.91     24 6-Apr-12 300 864.05 1.145 0.246 72.02 1.115 66.785     23 7-Apr-12 301                   8-Apr-12 302                   9-Apr-12 303 798.5 0.187 0 83.34 0.885 77.78 3228.6 2700   10-Apr-12 304                   11-Apr-12 305 816.1 0.908 0.141 75.13 1.5 70.755       12-Apr-12 306 780.45 2.415 0.364 70.91 1.51 70.79     21.5 13-Apr-12 307 748.55 3.559 0.523 60.41 1.125 54.01 2811.7 2425.3 21.5 14-Apr-12 308                   15-Apr-12 309                   16-Apr-12 310 788.65 0.811 0.099 67.75 0.075 50.885     22 17-Apr-12 311 805.75 1.153 0.06 75.02 0.39 60.43     22 18-Apr-12 312                   19-Apr-12 313 795.95 1.187 0.317 73.08 0.3 56.52     22 20-Apr-12 314 798.45 0.924 0.346 63.74 0.285 51.225 1953.3 1733.3 22 21-Apr-12 315                   22-Apr-12 316                   23-Apr-12 317 765.1 2.815 0.106 69.11 0.245 57.2     22  90 Appendix B: Data of cycle tests for system optimization Test No. Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2011-10-26 0 30.7 0.13 7.59 618 162.32 0.06 24.03 T=32 °C 1 31.1 0.46 7.58 580.6 146.52 0.64 27.31  Decant V= 70L 2 31.3 0.39 7.5 550 137.15 0.49 28.69 Air flow = 8L/min 3 31.6 0.38 7.41 500 128.12 0.64 28.68   4 31.8 0.33 7.33 391.7 114.01 0.41 30.54   5 32 0.45 7.22 335 101.04 0.52 31.71   6 32.2 0.51 7.13 295.8 85.16 0.31 31.22   7 32.3 0.42 6.99 250 73.53 0.44 34.1   8 32.4 0.47 6.86 210 65.59 0.33 33.91   9 32.5 0.38 6.57 125.7 50.79 0.47 36.02   10 32.6 0.4 6.28 77.5 41.5 0.46 36.94   10.5 32.6 0.48 6 54 37.13 0.43 38.37s                   2011-10-31 Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) T=30 °C 0 28.7 0.59 7.7 695 180.78 0.715 25.085  Decant V= 70L 1 28.9 0.65 7.66 655.6 193.58 0.82 30.05 Air flow = 8L/min 2 29.1 0.51 7.57 483.3 158.22 0.94 28.345   3 29.3 0.82 7.45 418.2 130.42 0.705 30.13   4 29.5 0.77 7.32 362.5 113.27 0.68 31.46   5 29.6 0.74 7.17 300 89.06 0.52 30.55   6 29.6 0.68 6.98 226.3 76 0.645 35.57   7 29.7 0.71 6.69 140 59.8 0.05 37.225   8 29.8 0.94 6.22 70 45 0.815 41.12   8.5 29.8 0.8 6 44 42.31 1.555 42.405                   2011-11-05 Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) T=28 °C 0 27.3 0.13 7.59 780 228.3 6.71 40.615  Decant V= 70L 1 27.7 0.13 7.65 890 230.97 0.38 39.79 Air flow = 8L/min 2 27.7 0.13 7.67 766.7 224.05 0.25 35.72   3 27.8 0.86 7.72 706.25 209.75 1.75 38.435   4 27.9 0.89 7.66 600 196.23 1.89 38.39   5 28 0.8 7.58 552.6 178.1 2.83 41.18   6 28.1 0.79 7.42 440.9 150.96 2.7 45.84   7 28.2 0.83 7.32 386.4 139.65 3.79 47.92   8 28.3 0.88 7.15 284.6 115.76 1.92 49.6   9 28.3 0.91 6.92 222.6 103.12 3.98 51.36   10 28.3 1.24 6.5 102.5 77.01 5.46 56.63   11 28.3 3.92 6 48 72.79 5.975 61.77 2011-11-06  Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L)  0 29.1 0.16 7.57 720 231.68 8.465 51.505 T=30°C 1 29.3 0.13 7.64 840 270.7 0.645 49.715  Decant V= 70L 2 29.3 0.14 7.65 770 255.37 0.335 42.485 Air flow = 8L/min 3 29.6 0.81 7.72 660 212.88 1.84 44.105   4 29.7 0.79 7.66 600 197.21 3.16 45.37   5 29.8 0.68 7.55 517.6 181.23 3.875 47.39   6 29.8 0.76 7.42 425 145.71 3.9 48.75   7 29.9 0.64 7.22 330.4 131.12 4.24 53.33   8 29.9 0.7 7.04 240 106.83 3.655 54.71   9 29.9 0.79 6.77 150 86.65 3.325 56.91   10 29.9 3.02 6 56 68.95 4.215 61.54           91  Test No.  Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2011-11-11 0 31.2 0.12 7.6 820 161.87 1.45 20.28 T=32°C 1 31.4 0.11 7.64 918.2 211.69 1.49 25.39  Decant V= 70L 2 31.5 0.09 7.65 820 255.08 1.68 29.38 Air flow = 8L/min 3 31.7 0.55 7.72 620 150.34 1.55 18.34   4 31.8 0.64 7.69 547.7 173.83 2.13 25.83   5 31.9 0.68 7.63 500 185.69 2.12 32.5   6 32 0.61 7.53 430.8 189.08 2.29 40.31   7 32.1 0.58 7.39 354.5 121.42 1.99 32.38   8 32.2 0.41 7.22 270.4 100.68 1.45 31.74   9 32.2 0.44 6.97 126.5 119.94 1.29 50.48   10 32.2 0.45 6.57 100 60.68 1.36 36.94   10.5 32.2 0.9 6 50.6 47.39 1.49 37.93   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2011-11-15 0 24.3 0.17 7.64 742.9 222.8 13.36 36.05 T=26°C 1 24.7 0.15 7.71 700 222.08 6.42 33.13  Decant V= 70L 2 24.9 0.14 7.76 700 222.26 1.61 31.88 Air flow = 8L/min 3 25.2 0.59 7.71 633.3 210.3 4.17 34.95   4 25.4 0.57 7.6 565 184 6.89 37.95   5 25.6 0.66 7.44 405 153.17 7.91 39.7   6 25.7 0.65 7.28 332 139.17 9.56 43.49   7 25.7 0.74 7.08 266.7 114.26 10.99 47.05   8 25.7 0.75 6.74 112 88.69 12.8 46.58   9 25.7 1.74 6 56 79.37 15.66 54.33   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2011-11-17 0 27.3 0.15 7.63 ––– 213.96 2.06 29.72 T=32°C 1 28.3 0.22 7.72 ––– 210.24 1.46 27.14 Redo one 2 28.6 0.14 7.72 ––– 208.48 1.57 26.24  Decant V= 70L 3 29.3 0.56 7.71 ––– 201.23 3.24 28.17 Air flow = 8L/min 4 29.9 0.52 7.62 ––– 175.76 4.01 29.85   5 30.4 0.59 7.5 ––– 148.07 3.26 31.59   6 30.8 0.62 7.34 ––– 124.71 3.17 31.99   7 30.9 0.69 7.17 ––– 111.59 3.26 35.43   8 31.1 0.55 6.93 ––– 89.27 3.2 37.06   9 31.2 0.58 6.54 ––– 65.78 2.29 38.12   9.5 31.2 0.6 6 ––– 51.18 2.13 38.27   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-02-02 0 30.3 0.14 7 ––– 203.83 0.46 52.23 T=32°C 1 30.8 0.09 7.05 ––– 197.27 0.16 43.24 Air flow = 10L/min 2 31 0.42 7.15 ––– 191.74 0.42 44.39  Decant V= 70L 3 31.2 0.4 7.16 ––– 188.84 1.08 45.82   4 31.3 0.38 7.13 ––– 174.66 0.64 46.89   5 31.3 0.36 7.08 ––– 160.26 0.43 47.44   6 31.3 0.38 7 ––– 135.04 0.6 48.41   7 31.3 0.32 6.88 ––– 111.08 0.5 50.21   8 31.4 0.43 6.73 ––– 108.43 0.84 51.76   9 31.6 0.45 6.58 ––– 92.06 0.38 52.88   10 31.6 0.47 6.31 ––– 82.55 0.33 53.46   11 31.6 0.58 6 ––– 73.99 0.34 54.68   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-02-09 0 27.6 0.16 7 ––– 322.71 2.735 86.905 T=32°C 1 28.1 0.15 7.01 ––– 296.39 3.105 84.885 Increase decant volume to 90L 2 28.8 0.22 7.18 ––– 293.2 2.83 88.92 Air flow = 10L/min 3 29.2 0.2 7.2 ––– 277.86 2.645 93.53   4 29.9 0.2 7.16 ––– 242.39 2.455 94.315  92 Test No.  Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L)  5 30.2 0.36 7.11 ––– 238.62 2.495 95.435   6 30.4 0.24 7.05 ––– 222.79 5.015 96.795   7 30.8 0.23 6.97 ––– 215.04 4.91 103.92   8 31 0.31 6.86 ––– 198.06 2.88 108.63   9 31.1 0.17 6.72 ––– 177.11 1.535 112.275   10 31.2 0.21 6.56 ––– 163.58 1.29 117.055   11 31.3 0.32 6.39 ––– 151.63 0.34 118.58   12 31.3 0.18 6 ––– 137.36 0.19 122.84   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-02-15 0 28 0.17 7 ––– 282.175 1.81 72.78 T=32 1 28.7 0.11 7.04 ––– 288.225 3.45 69.99 Increase air flow to12 L/min 2 29 0.38 7.21 ––– 269.7 2.44 74.61 Decant volume = 90L 3 29.2 0.35 7.24 ––– 249.525 2.88 72.91   4 29.4 0.2 7.18 ––– 228.675 3.44 76.13   5 29.8 0.17 7.13 ––– 221.55 4.13 80.88   6 30 0.24 7.07 ––– 202.275 4.17 81.45   7 30.5 0.17 6.97 ––– 185.425 5.89 83.52   8 30.9 0.18 6.85 ––– 175.825 3.97 86.41   9 31.3 0.24 6.69 ––– 160.85 2.81 87.96   10 31.4 0.27 6.49 ––– 142.325 2.41 93.02   11 31.4 0.39 6 ––– 116.75 0.8 95.96   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-02-23 0 27.8 0.04 7.22 1005 284.97 3.12 50.13   1               Air flow = 12 L/min 2 28.4 0.36 7.37 ––– 264.21 2.71 50.76 T=32 3 28.9 0.25 7.4 ––– 259.89 2.52 53.04  Decant volume = 120L 4 29.5 0.3 7.37 ––– 238.77 2.28 53   5 30 0.32 7.31 730 225.26 2.24 56.13   6 30.3 0.23 7.25 ––– 205.54 1.97 59.84   7 30.6 0.18 7.16 ––– 194.81 1.66 60.84   8 30.8 0.29 7.06 ––– 178.14 1.61 62.8   9 31 0.14 6.96 403.3 152.25 1.55 64.21   10 31.2 0.25 6.85 ––– 148.99 1.29 65.14   11 31.2 0.29 6.74 ––– 124.52 0.9 68.66   12 31.3 0.34 6.57 ––– 106.91 0.62 69.3   13 31.4 0.3 6.31 ––– 89.02 0.33 71.1   14 31.5 0.16 6 68 78.49 0.57 73.17   15 31.5 0.23 5.5 30 72.51 2.15 74.06   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-03-01 0 26.2 0.17 7.13 944 278.03 3.51 46.24 T=32 1 26.7 0.16 7.2 1092 272.91 2.16 45.07 Air flow = 14 L/min 2 27.3 0.36 7.41 1002 253.93 3.28 47.87 Decant volume = 120L 3 27.8 0.25 7.46 950.6 242.44 2.19 48.69   4 28.2 0.38 7.46 873.3 236.53 1.85 50.69   5 28.5 0.45 7.42 810 193.09 1.38 52.29   6 28.9 0.33 7.36 740 188.17 1.59 53.5   7 29.3 0.31 7.29 650.3 178.99 1.61 54.45   8 29.5 0.27 7.2 564.5 166.57 1.45 58.05   9 29.8 0.39 7.09 470 143.14 1 59.98   10 30.2 0.54 6.94 360 124.5 0.72 59.98   11 30.5 0.49 6.79 284 108.68 0.58 61.05   12 30.8 0.33 6.61 200 94.34 0.51 61.69   13 30.8 0.42 6.33 132.6 84.7 0.41 69.53   13.5 30.8 0.28 6 64.6 66.71 1.6 70.28   14.5 30.8 0.38 5.5 38 60.38 2.46 70.82  93 Test No.  Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-03-12 0 26.8 0.08 7.39 1500 455.15 2.38 20.2 T=32 1 27.3 0.09 7.44 ––– 452.74 1.37 18.81 Decant volume = 190L 2 27.9 0.43 7.6 ––– 434.4 2.34 19.53 Air =12L/min 3 28.3 0.45 7.67 ––– 399.84 3.32 20.65   4 28.8 0.48 7.68 ––– 356.23 3.97 21.97   5 29 0.53 7.67 ––– 324.98 3.86 23.6   6 29.4 0.47 7.65 ––– 310.35 3.71 26.68   7 29.9 0.42 7.6 ––– 299.03 3.27 28.77   8 30.1 0.54 7.55 ––– 275.43 2.6 31.67   9 30.2 0.51 7.5 ––– 258.15 1.96 32.69   10 30.4 0.47 7.44 823 244.88 1.88 37.15   11 30.5 0.45 7.37 ––– 222.55 1.94 39.23   12 30.5 0.48 7.32 ––– 192.6 2.1 46.18   13 30.6 0.47 7.25 ––– 187.33 2.36 49.78   14 30.8 0.42 7.18 ––– 175.95 2.4 52   15 30.8 0.36 7.12 423.3 167.2 2.43 56.38   16 30.9 0.42 7.05 ––– 148.25 1.5 58.64   17 30.9 0.35 6.93 ––– 132.3 1.22 60.63   18 31 0.4 6.8 ––– 118.75 0.93 62.43   19 31 0.44 6.65 ––– 109.75 1.13 63.47   20 31.2 0.42 6.46 ––– 94.99 1.35 64.78   21.5 31.2 0.63 6 66.7 72.38 1.64 65.82   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-03-28 0 30.5 0.09 6.03   69.34 2.11 64.39 Decant volume = 190L 1 30 0.13 6.76   134.49 1.77 53.44 Air = 12L/min 2 29.8 0.27 7.16   154.86 1.83 52.4 Feeding rate= 500 ml/min 3 29.5 0.29 7.3   173.54 1.94 48.82   4 29 0.24 7.47   221.89 2.3 40.06   5 29.2 0.31 7.58   293.86 2.2 38.2   6 29.6 0.44 7.65   338.56 1.73 35.61   7 30.1 0.48 7.61   322.23 2.15 33.64   8 30.5 0.44 7.56   288.35 1.78 32.45   9 30.5 0.38 7.5   266.29 2.28 31.95   10 30.6 0.39 7.44   249.22 1.86 33.7   11 30.7 0.42 7.38   228.12 1.95 36.24   12 30.8 0.48 7.32   219.87 1.86 41.7   13 31 0.49 7.26   193.36 2.34 46.45   14 31.1 0.37 7.2   163.27 1.87 49.99   15 31.2 0.44 7.12   158.69 1.79 50.65   16 31.2 0.46 7.03   149.2 1.89 55.49   17 31.2 0.38 6.93   137.43 2.03 58.58   18 31.2 0.42 6.8   127.66 1.85 59.01   19 31.2 0.38 6.66   101.88 2.2 61.73   20 31.2 0.33 6.48   83.06 1.72 61.91   21.5 31.2 0.39 6   70.24 1.77 64.39   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-05-01 0 29.5 0.1 7.1   215.4 0.4 46.82 T=32 1 29.8 0.08 7.16   210.33 0.22 44.385 Airflow=12L/min 2 30 0.63 7.26   200.2 0.64 44.81 V=70L 3 30.3 0.51 7.24   184 0.705 49.22   4 30.5 0.61 7.16   161.59 1.06 50.34   5 30.7 0.48 7.07   149.42 0.405 50.035   6 30.9 0.41 6.96   131.69 0.382 51.71   7 31 0.52 6.81   123.95 0.355 53.59   8 31.1 0.6 6.62   119.38 0.54 57.01   9 31.2 0.61 6.35   100.21 0.54 60.5  94 Test No.  Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-05-05 0 30.3 0.14 7 ––– 203.83 0.46 52.23 T=32°C 1 30.8 0.09 7.05 ––– 197.27 0.16 45.24 Air flow = 10L/min 2 31 0.42 7.15 ––– 191.74 0.42 44.39  Decant volume = 70L 3 31.2 0.4 7.16 ––– 188.84 1.08 45.82   4 31.3 0.38 7.13 ––– 174.66 0.64 46.89   5 31.3 0.36 7.08 ––– 160.26 0.43 47.44   6 31.3 0.38 7 ––– 135.04 0.6 48.41   7 31.3 0.32 6.88 ––– 111.08 0.5 50.21   8 31.4 0.43 6.73 ––– 108.43 0.84 51.76   9 31.6 0.45 6.58 ––– 97.02 0.38 52.88   10 31.6 0.47 6.31 ––– 92.06 0.33 53.46   11 31.6 0.58 6 ––– 89.43 0.34 54.68   Time (h) Temp (°C) DO (mg/L) pH ALK (mg/L) NH4-N (mg/L) NO2-N (mg/L) NO3-N (mg/L) 2012-05-10 0 30 0.13 7.2   308.82 1.14 45.94 T=32 1 30.2 0.14 7.31   325.22 0.18 43.43 Air=10L/min 2 30.5 0.52 7.56   316.87 0.26 46.59 Decent volume =120L 3 30.8 0.48 7.55   305.19 0.82 48.22   4 31 0.46 7.5   305.25 1.46 50.48   5 31 0.49 7.45   282.66 0.82 52.01   6 31.1 0.5 7.4   267.43 0.91 53.89   7 31.1 0.44 7.34   249.67 0.73 55.1   8 31.2 0.48 7.28   224.15 0.52 56.32   9 31.2 0.55 7.22   206.02 0.2 57.7   10 31.2 0.53 7.14   198.31 0.19 58.55   11 31.3 0.52 7.04   164.06 0.16 59.55   12 31.6 0.62 6.9   161.31 0.2 63.26   13 31.8 0.69 6.75   132.11 0.16 63.97   14 32 0.7 6.6   117.22 0.31 66.42   15 32 0.65 6.34   105.01 0.52 71.38   16 32 0.75 6   94.2 1.63 74.43      

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 26 0
Canada 19 3
China 16 4
Singapore 6 0
Germany 5 1
Poland 5 0
Spain 4 0
Russia 3 0
Ukraine 1 0
Republic of Korea 1 0
Indonesia 1 0
United Kingdom 1 0
Hong Kong 1 0
City Views Downloads
Unknown 23 1
Lewes 9 0
Toronto 6 0
Vancouver 6 3
Shenzhen 5 3
Singapore 4 0
Beijing 3 0
Ashburn 3 0
Hangzhou 3 0
Vieux-Saint-Laurent 3 0
Barcelona 3 0
Absecon 2 0
Guangzhou 2 1

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

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.24.1-0050794/manifest

Comment

Related Items