Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analysis of design factors influencing the oxygen transfer efficiency of a Speece Cone hypolimnetic aerator Kowsari, Assieh 2008

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

Item Metadata

Download

Media
24-ubc_2008_fall_kowsari_assieh.pdf [ 2.11MB ]
Metadata
JSON: 24-1.0063099.json
JSON-LD: 24-1.0063099-ld.json
RDF/XML (Pretty): 24-1.0063099-rdf.xml
RDF/JSON: 24-1.0063099-rdf.json
Turtle: 24-1.0063099-turtle.txt
N-Triples: 24-1.0063099-rdf-ntriples.txt
Original Record: 24-1.0063099-source.json
Full Text
24-1.0063099-fulltext.txt
Citation
24-1.0063099.ris

Full Text

ANALYSIS OF DESIGN FACTORS INFLUENCING THE OXYGEN TRANSFER EFFICIENY OF A SPEECE CONE HYPOLIMNETIC AERATOR  by Assieh Kowsari B.Sc. Sharif University of Technology, Iran, 1994  A THESIS SUBMITRTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in The Faculty of Graduate Studies (Civil Engineering) UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  October 2008  © Assieh Kowsari  ABSTRACT  The objective of this research was to characterize the performance of a downflow bubble contact (DBCA) hypolimnetic aerator  —  Speece Cone-. The effect of two key design  factors, inlet water velocity and the ratio of gas flow rate to water flow rate on four standard units of measure was examined: (a) the Oxygen Transfer Coefficient, KLa, corrected to 20°C,  KLa2O  (hr ) 1 , (b) the Standard Oxygen Transfer Rate, SOTR 2 (g0 . hr’)  (c) the Standard Aeration Efficiency, SAE 2 (gO k Whr’), and (d) the Standard Oxygen Transfer Efficiency, SOTE (%). Two sources of oxygen, Pressure Swing Adsorption (PSA) oxygen (87% purity) and air, were compared. KLa2Q, SOTR, and SAE increased with an increase in the ratio of gas flow rate to water flow rate for both air and oxygen, over a range of 0.5% to 5.0%; while SAE deceased. An increase in inlet water velocity resulted in a decrease in KLa, corrected to 20°C, SOTR, and SAE, but an increase in the SOTE. Treatments on air showed similar, but much less dramatic effect of the gas flow rate to water flow rate ratio and water inlet velocity on KLa2O, SOTE, SAE, and SOTE, when compared to treatments on PSA oxygen. The best performance was achieved with an inlet water velocity of 6.9-7.6 ms 1 and oxygen flow rate to water flow rate ratio of about 2.5%. At this combination, the SOTE was about 66-72%.  11  TABLE OF CONTENTS ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  v  LIST OF FIGURES  xiii  ACKNOWLEDGEMENTS  xvii  DEDICATION  xviii  CHAPTER 1: INTRODUCTION  1  1.1. Literature Review of Hypolimnetic Aeration  1  1.1.1.  Background  1  1.1.2.  History of Hypolimnetic eration  3  1.1.3.  Air! oxygen injection methods for hypolimnetic aeration  4  1.2. Rationale for Current research  11  1.3. Objectives  12  CHAPTER 2: EQUIPMENT AND METHODS 2.1. System Design  13 13  2.1.1. Aeration Tank  13  2.1.2. Downflow Bubble Contact Aerator (Speece Cone)  13  2.1.3. Sources of Water and Oxygen  16  2.1.4. Measuring, Monitoring and Recording Equipment  17  2.3. Test Procedure  22  2.4. Experimental Design  26  2.4.1. Preliminary Experiments  In  26  2.4.2. Main Experiments  .26  2.5. Parameter Estimation  31  Chapter 3. Results  36  3.1. Group 1- Water inlet velocity of 8.5 m.s oxygen 3.2. Group 2- Water inlet velocity of 8.0 m.s’ oxygen 3.3. Group 3- Water inlet velocity of 7.6 m.s 1 oxygen 3.4. Group 4- Water inlet velocity of 7.4 m.s 1 oxygen 3.5. Group 5- Water inlet velocity of 6.9 m.s 1 oxygen 3.6. Group 6- Water inlet velocity of 6.4 m.s 1 oxygen 3.7. Group 7- Water inlet velocity of 5.7 m.s oxygen  treatment on air and PSA 36  treatment on Air and PSA 38  treatment on Air and PSA 41  treatment on air and PSA 44  treatment on Air and PSA 46 treatment on air and PSA 49 treatment on air and PSA 52  3.8. Inlet Water Velocity  55  3.9. Hydraulic Residence Time  59  3.10. Oxygen vs. air  66  Chapter 4: Discussion  84  4.1. Gas-Liquid Transfer Theory  84  4.2. Gas Flow Rate to Water Flow rate Ratio  89  4.3. Inlet water velocity  91  4.4. Optimization  92  4.5. PSA Oxygen vs. air  93  Chapter 5: Conclusions and Recommendations  95  5.1. Conclusions  95  5.2. Recommendations for Future Work  97  References  98  iv  LIST OF TABLES  Table 1.1: Oxygenation Systems based on 5 tons/d systems installed in various locations (Beutel et al, 2002)  7  Table 2.1: Experimental combination for Group 1  27  Table 2.2: Experimental combination for Group 2  28  Table 2.3: Experimental combination for Group 3  28  Table 2.4:Experimental combination for Group 4  28  Table 2.5:Experimental combination for Group 5  28  Table 2.6: Experimental combination for Group 6  29  Table 2.7: Experimental combination for Group 7  29  Table 2.8: Summary of treatments for Speece Cone aeration experiments  30  Table 2.9: Summary of energy calculations  35  Table 3.1.a: Least squares means  1 and (± STDEV), for inlet water velocity 8.5 m.s =  gas flow rate/water flow rate Table 3.1.b: Least squares means  0.5%  66  (± STDEV), for inlet water velocity 8.5 m.s’ and =  gas flow rate/water flow rate Table 3.1.c: Least squares means  =  =  1.0%  66  1 and (± STDEV), for inlet water velocity 8.5 m.s  gas flow rate/water flow rate  =  =  1.5%  v  67  Table 3.1.d: Least squares means  (± STDEV), for inlet water velocity 8.5 m.s and  gas flow rate/water flow rate Table 3.1.e: Least squares means  67  =  3.0%  67 =  =  3.5%  68  1 and (± STDEV), for inlet water velocity 8.5 m.s =  =  4.0%  68  1 and gas (± STDEV), for inlet water velocity 8.5 m.s =  flow rate/water flow rate Table 3.2.a: Least squares means  2.5%  1 and (± STDEV), for inlet water velocity 8.5 m.s  gas flow rate/water flow rate Table 3.1.i: Least squares means  =  1 and (± STDEV), for inlet water velocity 8.5 m.s  gas flow rate/water flow rate Table 3.1.h: Least squares means  67 =  gas flow rate/water flow rate Table 3.1.g: Least squares means  2.0%  1 and (± STDEV), for inlet water velocity 8.5 m.s  gas flow rate/water flow rate Table 3.1.f: Least squares means  =  4.5%  68  1 and (± STDEV), for inlet water velocity 8.0 m.s  gas flow rate/water flow rate = 0.5% Table 3.2.b: Least squares means  1 and (± STDEV), for inlet water velocity 8.0 m.s =  gas flow rate/water flow rate Table 3.2.c: Least squares means  68  =  1.0%  69  (± STDEV), for inlet water velocity 8.0 m.s’ and =  gas flow rate/water flow rate  1.5%  69  Table 3.2.d: Least squares means (± STDEV), for inlet water velocity = 8.0 m.s 1 and gas flow rate/water flow rate  =  2.0%  vi  69  Table 3.2.e: Least squares means  1 and (± STDEV), for inlet water velocity 8.0 m.s =  gas flow rate/water flow rate Table 3.2.f: Least squares means  70  =  4.0%  70 =  =  4.5%  70  1 and (± STDEV), for inlet water velocity 8.0 m.s =  =  5.0%  71  1 and (± STDEV), for inlet water velocity 7.6 m.s =  0.5%  71  1 and (± STDEV), for inlet water velocity 7.6 m.s =  =  1.0%  71  1 and (± STDEV), for inlet water velocity 7.6 m.s =  gas flow rate/water flow rate  Table 3.3.d: Least squares means  3.5%  1 and gas (± STDEV), for inlet water velocity 8.0 m.s  gas flow rate/water flow rate  Table 3.3.c: Least squares means  =  =  gas flow rate/water flow rate  Table 3.3.b: Least squares means  70  1 and (± STDEV), for inlet water velocity 8.0 m.s  gas flow rate/water flow rate Table 3.3.a: Least squares means  3.0% =  flow rate/water flow rate Table 3.2.j: Least squares means  =  1 and (± STDEV), for inlet water velocity 8.0 m.s  gas flow rate/water flow rate Table 3.2.1: Least squares means  69 =  gas flow rate/water flow rate Table 3.2.h: Least squares means  2.5%  (± STDEV), for inlet water velocity 8.0 m.s’ and  gas flow rate/water flow rate Table 3.2.g: Least squares means  =  =  1.5%  71  1 and (± STDEV), for inlet water velocity 7.6 m.s  gas flow rate/water flow rate  =  =  2.0%  vii  72  Table 3.3.e: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate Table 3.3.f: Least squares means  —  3.5%  72  =  4.0%  73 =  =  4.5%  73  1 and (± STDEV), for inlet water velocity 7.6 m.s =  =  5.0%  73  1 and (± STDEV), for inlet water velocity 7.4 m.s =  =  0.5%  73  1 and (± STDEV), for inlet water velocity 7.4 m.s  gas flow rate/water flow rate Table 3.4.c: Least squares means  72  1 and gas (± STDEV), for inlet water velocity 7.6 m.s  gas flow rate/water flow rate Table 3.4.b: Least squares means  3.0%  =  gas flow rate/water flow rate Table 3.4.a: Least squares means  =  1 and (± STDEV), for inlet water velocity 7.6 m.s  flow rate/water flow rate Table 3.3.j: Least squares means  72  =  gas flow rate/water flow rate Table 3.3.i: Least squares means  2.5%  1 and (± STDEV), for inlet water velocity 7.6 m.s  gas flow rate/water flow rate Table 3.3.h: Least squares means  7.6 m.s 1 and  (± STDEV), for inlet water velocity 7.6 m.s and  gas flow rate/water flow rate Table 3.3.g: Least squares means  =  =  =  1.0%  74  (± STDEV), for inlet water velocity 7.4 m.s and  gas flow rate/water flow rate  =  =  1.5%  Table 3.4.d: Least squares means (± STDEV), for inlet water velocity  gas flow rate/water flow rate  =  2.0%  viii  74 7.4 m.s’ and 74  Table 3.4.e: Least squares means  (± STDEV), for inlet water velocity 7.4 m.s’ and =  gas flow rate/water flow rate Table 3.4.f: Least squares means  (± STDEV), for inlet water velocity 7.4 m.s’ and =  gas flow rate/water flow rate Table 3.4.g: Least squares means  74  2.5%  =  3.0%  75  (± STDEV), for inlet water velocity 7.4 m.s’ and =  gas flow rate/water flow rate  =  3.5%  75  Table 3.4.h: Least squares means (± STDEV), for inlet water velocity = 7.4 m.s’ and gas flow rate/water flow rate  =  4.0%  Table 3.4.1: Least squares means (± STDEV), for inlet water velocity flow rate/water flow rate  =  75 =  7.4 m.s’ and gas  4.5%  75  Table 3.4.j: Least squares means (± STDEV), for inlet water velocity = 7.4 m.s 1 and gas flow rate/water flow rate  =  5.0%  76  Table 3.5.a: Least squares means (± STDEV), for inlet water velocity = 6.9 m.s 1 and gas flow rate/water flow rate  =  0.5%  Table 3.5.b: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate  =  =  =  6.9 m.s 1 and  1.0%  Table 3.5.c: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate  76  76 6.9 m.s and  1.5%  76  Table 3.5.d: Least squares means (± STDEV), for inlet water velocity = 6.9 m.s and gas flow rate/water flow rate  =  2.0%  ix  77  Table 3.5.e: Least squares means (± STDEV), for inlet water velocity  6.9 m.s 1 and  gas flow rate/water flow rate = 2.5% Table 3.5.f: Least squares means  (± STDEV), for inlet water velocity 6.9 m..s and =  gas flow rate/water flow rate Table 3.5.g: Least squares means  =  3.0% =  =  3.5% =  —  4.0%  Table 3.5.i: Least squares means (± STDEV), for inlet water velocity  flow rate/water flow rate  =  78  =  5.0%  78  1 and (± STDEV), for inlet water velocity 6.4 m.s =  0.5%  78  1 and (± STDEV), for inlet water velocity 6.4 m.s =  =  1.0%  79  (± STDEV), for inlet water velocity 6.4 m.s’ and  gas flow rate/water flow rate Table 3.6d: Least squares means  6.9 1 m.s and gas  =  gas flow rate/water flow rate Table 3.6.c: Least squares means  =  1 and (± STDEV), for inlet water velocity 6.9 m.s  gas flow rate/water flow rate Table 3.6.b: Least squares means  78  4.5%  gas flow rate/water flow rate Table 3.6.a: Least squares means  77  1 and (± STDEV), for inlet water velocity 6.9 m.s  gas flow rate/water flow rate  Table 3.5.j: Least squares means  77  1 and (± STDEV), for inlet water velocity 6.9 m.s  gas flow rate/water flow rate Table 3.5.h: Least squares means  77  =  1.5%  79  1 and (± STDEV), for inlet water velocity 6.4 m.s  gas flow rate/water flow rate = 2.0%  x  79  Table 3.6.e: Least squares means  (± STDEV), for inlet water velocity 6.4 m.s’ and =  gas flow rate/water flow rate Table 3.6.f: Least squares means  2.5%  79  1 and (± STDEV), for inlet water velocity 6.4 m.s =  gas flow rate/water flow rate Table 3.6.g: Least squares means  =  =  3.0%  80  1 and (± STDEV), for inlet water velocity 6.4 m.s =  gas flow rate/water flow rate = 3.5% Table 3.6.h: Least squares means  1 and (± STDEV), for inlet water velocity 6.4 m.s =  gas flow rate/water flow rate Table 3.6.1: Least squares means  =  =  =  5.0%  81  1 and (± STDEV), for inlet water velocity 5.7 m.s =  =  81  0.5%  (± STDEV), for inlet water velocity 5.7 m.s and =  =  1.0%  81  1 and (± STDEV), for inlet water velocity 5.7 m.s =  gas flow rate/water flow rate Table 3.7.d: Least squares means  80  1 and (± STDEV), for inlet water velocity 6.4 m.s  gas flow rate/water flow rate Table 3.7.c: Least squares means  80  4.5%  gas flow rate/water flow rate Table 3.7.b: Least squares means  4.0% =  gas flow rate/water flow rate Table 3.7.a: Least squares means  =  1 and gas (± STDEV), for inlet water velocity 6.4 m.s  flow rate/water flow rate Table 3.6.j: Least squares means  80  =  1.5%  81  (± STDEV), for inlet water velocity 5.7 m.s and  gas flow rate/water flow rate  =  =  2.0%  xi  82  Table 3.7.e: Least squares means  1 and (± STDEV), for inlet water velocity 5.7 m.s =  gas flow rate/water flow rate Table 3.7.f: Least squares means  =  3.0%  82  (± STDEV), for inlet water velocity 5.7 m.s’ and =  3.5%  82  (± STDEV), for inlet water velocity 5.7 m.s and =  gas flow rate/water flow rate Table 3.7.i: Least squares means  82 =  gas flow rate/water flow rate Table 3.7.h: Least squares means  2.5%  1 and (± STDEV), for inlet water velocity 5.7 m.s  gas flow rate/water flow rate Table 3.7.g: Least squares means  =  =  4.0%  83  (± STDEV), for inlet water velocity 5.7 m.s and gas =  flow rate/water flow rate = 4.5% Table 3.7.j: Least squares means (± STDEV), for inlet water velocity  gas flow rate/water flow rate  83 5.7 m.s and  5.0%  83  Table 4.1 Relationship between KLa (C 1 CL), the operational factors, and the -  experimental variables (Ashley, 2002) Table 4.2  —  Calculated inlet diameter (m) for various inlet velocities (ms’)  xii  89 93  LIST OF FIGURES  Figure 1.1. Schematic representation of airlift aerator LPA1 (from Bums et al, 2002)  8  Figure 1.2. Photograph of experimental Speece Cone  9  Fig 2.1- Schematic diagram of aeration tank  14  Figure 2.2. Schematic of Speece Cone aerator with probe locations  15  Fig 2.3- Schematic diagram of manifold board with data logger and flow meters (from Ashley, 2002)  19  Figure 2.4- PT4 Monitor, Oxyguard probes and thermister probe (from Point Four Systems Inc. website)  20  Fig. 3.1.a. KLa2O for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  37  Fig. 3.1.b. SOTR for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  37  Fig. 3.1.c. SAE for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  38  Fig. 3.1.d. SOTE for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  38  Fig. 3.2.a. KLa2O for Water Inlet Velocity of 8.0 m.s 1 on PSA and Air  39  Fig. 3.2.b. SOTR for Water Inlet Velocity of 8.0 m.s 1 on PSA and Air  40  Fig. 3.2.c. SAE for Water Inlet Velocity of 8.0 m.s 1 on PSA and Air  40  Fig. 3.2.d. SOTE for Water Inlet Velocity of 8.0 m.s 1 on PSA and Air  41  Fig. 3.3.a.  42  KLa2O  for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  Fig. 3.3.b. SOTR for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  xiii  42  Fig. 3.3.c. SAE for Water Inlet Velocity of 7.6 m.s on PSA and Air  43  Fig. 3.3.d. SOTE for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  43  Fig. 3.4.a. KLa2O for Water Inlet Velocity of 7.4 m.s 1 on PSA and Air  44  Fig. 3.4.b. SOTR for Water Inlet Velocity of 7.4 m.s 1 on PSA and Air  45  Fig. 3.4.c. SAE for Inlet Velocity of 7.4 m.s 1 on PSA and Air  45  Fig. 3.4.d. SOTE for Water Inlet Velocity of 7.4 m.s’ on PSA and Air  46  Fig. 3.5.a. KLa2O for Water Inlet Velocity of 6.9 m.s on PSA and Air  47  Fig. 3.5.b. SOTR for Water Inlet Velocity of 6.9 m.s 1 on PSA and Air  47  Fig. 3.5.c. SAE for Water Inlet Velocity of 6.9 m.s 1 on PSA and Air  48  Fig. 3.5.d. SOTE for Water Inlet Velocity of 6.9 m.s 1 on PSA and Air  48  Fig. 3.6.a. KLa2O for Water Inlet Velocity of 6.4 m.s 1 on PSA and Air  49  Fig. 3.6.b. SOTR for Inlet Velocity of 6.4 m.s on PSA and Air  50  Fig. 3.6.c. SAE for Inlet Velocity of 6.4 m.s 1 on PSA and Air  51  Fig. 3.6.d. SOTE for Water Inlet Velocity of 6.4 m.s 1 on PSA and Air  51  Fig. 3.7.a. KLa2O for Water Inlet Velocity of 5.7 m.s 1 on PSA and Air  53  Fig. 3.7.b. SOTR for Water Inlet Velocity of 5.7 m.s 1 on PSA and Air  53  Fig. 3.7.c. SAE for Water Inlet Velocity of 5.7 1 m.s oh PSA and Air  54  Fig. 3.7.d. SOTE for Water Inlet Velocity of 5.7 m.s on PSA and Air  54  Figure 3.8. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on  55  KLa2O  xiv  Figure 3.9. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on SOTR  56  Figure 3.10. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on SOTE  56  Figure 3.11. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on SAE  57  Figure 3.12. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on KLa2O  57  Figure 3.13. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on SOTR  58  Figure 3.14. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on SOTE  58  Figure 3.15. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on SAE  59  Figure 3.16. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on KLa2O  60  Figure 3.17. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on SOTR  60  Figure 3.18. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on SOTE  61  Figure 3.19. The effect of hydraulic residence time and the ratio of oxygen flow rate to water  flow rate on SAE  61  xv  Figure 3.20. The effect of hydraulic residence time and the ratio of air flow rate to water  flow rate on KLa2O  62  Figure 3.21. The effect of hydraulic residence time and the ratio of air flow rate to water  flow rate on SOTR  62  Figure 3.22. The effect of hydraulic residence time and the ratio of air flow rate to water  flow rate on SOTE  63  Figure 3.23. The effect of hydraulic residence time and the ratio of air flow rate to water  flow rate on SAE  63  Figure 3.24. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet  water velocity  64  on KLa2O  Figure 3.25. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet water velocity on SOTR  64  Figure 3.26. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet  water velocity on SAE  65  Figure 3.27. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet water  velocity on SOTE  65  xvi  Acknowledgements I would like to express my deep and sincere gratitude to my supervisor, Professor Don Mavinic. His understanding, encouraging and personal guidance have provided a good basis for the present thesis. I am grateful to my supervisors, Dr. Ken Ashley, for his detailed and constructive comments and Dr. Ken Hall, for his important support throughout this work.  I wish to extend my warmest thanks to all those who have helped me with my work. Special thanks to Dr. Speece and Dr. Lawrence for their excellent advice and friendly help. Thanks to Susan and Paula, in environmental lab and Harold, Doug, Bill, Scott and John in workshop of civil department, for their technical support throughout this project.  Warm thanks to my friends, who have always lend a supportive hand whenever I asked for. Thanks to Ali A., Ali N., Ali R., Jeff, Margaret, Parvez, and Raphael.  My special gratitude is due to my brother, Reza, for his loving support.  Finally, I wish to acknowledge my late father, Taghi, and my mother, Zari as being the source of inspiration in my studies.  Things would remain incomplete without thanking the Almighty.  xvii  tv evorj o- vvj -2t’ier  vvotliey  \NttV Lo’e cic{ etervcL  xviii  CHAPTER 1: INTRODUCTION 1.1. Literature Review of Hypolimnetic Aeration 1.1.1. Background  Thermal stratification occurs in north and south temperate zone lakes and reservoirs during summer months. The water temperature of the lake surface increases due to several factors, including the magnitude of solar radiation, wind intensity and duration, air temperature, and relative humidity. Lakes stratify due to a vertical thermal regime and the force of gravity acting on density differences within the lake. The colder, denser water accumulates at the lake bottom forming the hypolimnion. The warmer surface water layer is the epilimnion. There is a region of sharp temperature change  —  the thermocline  between the epilimnion and hypolimnion. Oxygen in the hypolimnion is consumed by aerobic biodegradation of deposited plankton biomass and sediment oxygen demand (Beutel, 2001). The hypolimnion is minimally influenced by epilimnion or atmosphere, and there is no photosynthesis due to the lack of the sunlight in this layer. Thus, the hypolimnion cannot be re-aerated under natural conditions and the amount of the dissolved oxygen is often reduced to zero in mesotrophic and eutrophic lakes with a high hypolimnetic oxygen demand (Cooke and Carlson, 1989; Speece, 1971). Hypolimnetic anoxia negatively affects the water quality of the lakes and reservoirs. It increases internal recycling of nutrients that may stimulate algal growth, and ultimately increases the oxygen demand. It may cause solubilisation of reduced iron (Fe ) and 2 manganese (Mn ) accumulated in the sediments, and can lead to the production of 2 ammonia (NH ) and hydrogen sulfide 2 3 (H S ) gas. These compounds cause taste, odour, and color problems in the water. Therefore, water treatment costs will increase, due to the  increase of the oxidant demand to remove these compounds from the water, if the water is treated for potable use. The existence of certain organics in the raw water may lead to production of disinfection by-products (i.e. trihalomethanes) if chlorine is added as a primary disinfectant (Tate and Arnold, 1990). Low dissolved oxygen in hypolimnion limits the distribution of cold-water species of fish (Cooke et al. 2005) and reduces the space where they can survive (Muller and Stadelmann, 2004). Recent studies have shown that hypoxia causes disruption of endocrine and impairs reproduction in fish (Wu et al, 2003). To improve the water quality in a stratified water body, oxygen can be artificially added to the water to maintain aerobic conditions. Various techniques have been employed to add oxygen to oxygen-deficient hypolimion. The techniques are generally grouped into two categories: destratification and hypolimnetic aeration (Speece, 1971; Fast et a!., 1975). Destratification is defined as mixing the layers of the lake by artificial circulation, while in hypolimnetic aeration the hypolimnion is oxygenated without changing the the thermal profile of the water body. Aerating the hypolimnion would be beneficial if properly designed. It would improve water quality water by increasing dissolved oxygen concentrations and preventing an increase in the concentration of some chemicals, such as iron, manganese and hydrogen suiphide, causing color, taste and odour in the hypolimnion. Hypolimnetic aeration has several advantages over destratification: 1) it does not increase the temperature of hypolimnetic water; 2) It does not transport the nutrients from the hypolimnion into the epilimnion to support algal growth; 3) Less volume of water is involved, thus less energy  2  is required to obtain the same level of DO, compared to artificial mixing of the whole lake (Speece, 1971). The cold water can then be used for cooling of water supplies, municipal water supplies and serving as a cold-water habitat for fish species such as trout and salmon (Speece, 1971).  1.1.2. History ofHypolimnetic aeration Hypolimnetic aeration is a very important technique for water quality improvement and fisheries enhancement (Ashley, 1985). It is defined as oxygenation of the hypolimnion while maintaining the thermal-density gradient associated with stratification (Kortmann et al., 1994). Mercier and Perret (1949) described one of the earliest hypolimnetic aeration methods in Lake Bret, Switzerland. In this system, hypoxic water was withdrawn from the hypolimion into a shore based splash basin, where water was aerated utilizing mechanical agitation, and the aerated water was then returned to the same depth. This system has been successful in a limited number of cases (Cooke and Carison, 1989). The other method of hypolimnetic aeration is injecting either air or pure oxygen into the water. This method was first suggested by Bemhardt (1967) in Germany, and then developed by Speece (1971) and Fast (1971). Since then, there have been numerous projects completed on hypolimnetic aeration systems and a few theoretical models have been proposed (Taggart and McQueen, 1981; Ashley, 1985; McQueen and Lean, 1984; Little, 1995; Nakamuru, 1996). Reviews on hypolimnetic aeration have been published by Fast and Lorenzen (1976), Lorenzen and Fast (1977), McQueen and Lean (1986) and Singleton and Little (2006)  3  1.1.3. Air/oxygen injection methods for hypolimnetic aeration The techniques of oxygen/air injection used in lakes and reservoirs, can be classified into three primary groups: airlift aerators, bubble-plume diffuser, and submerged contact chamber (Speece cone) (Fast and Lorenzen, 1976; Lorenzen and Fast, 1977; Beutel and Home, 1999; Singleton and Little, 2006). Full or Partial Airlift Aerator: A typical full airlift hypolimnetic diffuser consists of a vertical pipe known as riser tube, a diffuser unit, an air/water separation chamber, and one or two return pipes known as down-comers (McQueen and Lean, 1986; Little, 1995; Ashley, 2002). Compressed air or oxygen is injected into the water through a diffuser, which is located near the bottom of the riser tube. The buoyant air-water mixture travels up the riser tube and enters a separation chamber at the top of the riser tube. In the separation chamber, most bubbles escape to the atmosphere, but some are entrained in the water flow entering the downcorner. The aerated water returns to the hypolimnion via the down-corner (Fig 1.1). In a partial lift aerator same process occurs except that the air-water mixture rises to in the vertical pipe to a given depth in the lake where gas bubbles escape to the atmosphere through a pipe (Burns et al, 2002). Deep Oxygen Bubble Injection (DOBI): This system aerates the water body by dispersing injected air or oxygen at a relatively low gas flow rate through a group of diffusers located in hypolimnion (McGinnis and Little, 2002). Diffusers can be arranged either linearly or circularly. Gas bubbles injected through diffusers into the water body mix with the water and create a positive buoyancy  4  flux. The plume will rise until the depth in which the plume momentum reaches zero. The plume will then move downward to the equilibrium depth, and then disperses horizontally (McGinnis et al, 2004). Downflow Bubble Contact Aerator (DBCA): The system consists of an oxygen/air supply, a conical bubble contact chamber and a submersible pump that draws water from the hypolimnion into the top of the cone (Figure 1.2). Water enters the cone at the top with a velocity greater than rise velocity of the bubbles. The design of the cone leads to an increase in the cross sectional area that causes a decrease in water velocity. Therefore, water exits the cone at the bottom with a velocity less than the rise velocity of the bubbles.  Thus, the bubbles are trapped inside the  chamber and the water flows downward through them (Speece et al, 1971). Prolonged contact time and high turbulence between the bubbles and water improves the mass transfer of oxygen, thus increasing aeration efficiency. Beutel et al, (2002) summarized the costs and advantages and disadvantages for various hypolimnetic aeration systems installed in various locations (Table 1.1). Capital costs are based on a 5 tons/d system (Beutel et al., 2002). They reported that the Speece Cone has the lowest operational cost, due to its extremely high oxygen transfer efficiency. Submerged contact chamber oxygenation systems such as DBCA have been installed in two lakes, Newman lake, Washington and Camanche Reservoir, California (Speece, 1994, Home, 1995), to improve the water quality; the objectives are to avoid a capital investment for additional taste and odor treatment, and to increase dissolved oxygen in the hypolimnion and prevent fish mortalities due to low oxygen. A Speece Cone has been operating in Newman Lake and Camanche reservoir since 1992 and 1994, respectively,  5  and has consistently maintained oxic conditions. Pair of large Speece cones has been successfully operating in Savannah Harbor since 2007 (www.eco2tech.com). The Speece Cone is recently being used for a new technology, named SuperOxygentaion, developed by Dr. Speece, to aerate water and wastewater in municipal systems (www.eco2tech.com). Site Stream Injection (SSI): In this system, water is drawn from hypolimnion and pumped onto shore, where oxygen is injected into the water, and then the water is discharged into the hypolimnion (Fast et al. 1975; Fast et a!. 1977). This system is not commonly used because of the high energy costs associated with the maintenance of a pressurized chamber on shore (Beutel et al., 2002).  6  Table 1.1. Oxygenation Systems based on 5 tons/d systems installed in various locations(Beutel et al, 2002)  Table 1.1.has been removed due to copyright restrictions. The information removed is a summary of the costs and advantages and disadvantages for various hypolimnetic aeration systems installed in various locations. From Beutel et al., 2002, AWWA 2002 Annual Conference  7  Figure 1.1. has been removed to copyright restrictions. The information removed is a Schematic representation of airlift aerator LPA1. From Burns, V.L., McGirmis D.F., and. Little J.C (2002) Predicting oxygen transfer and water flow rate in airlift aerators, Water Research 36(18): 4605-4615.  Figure 1.1. Schematic representation of airlift aerator LPA1 (from Bums et al, 2002)  8  Figure 1.2. Photograph of experimental Speece Cone  9  Advantages ofthe Speece Cone Hypolimnetic aerationloxygenation is a lake management technique designed to prevent hypolimnetic anoxia and its negative consequences on water quality. It can be achieved with a variety of designs using either air or oxygen injection. There are three types of air/oxygen injection systems; airlift aerator, bubble-plume diffusers, and Speece cone (Singleton and Little, 2006).The Speece Cone has several advantages over the two other techniques. First, in contrast to bubble plume and airlift systems, the Speece Cone injects oxygen near the bottom of the lake, where it is most beneficial. The highly oxygenated water is dispersed horizontally over the sediments of the reservoir, assuring a welloxygenated sediment-water interface (Fast, 2002; Beutel, 2002). Speece Cones do not operate on air because of energy costs associated with using air and recirculation of water that leads to lower efficiency (Ashley, 2002). Second, a Speece Cone with a properly designed horizontal dispersion system can overcome the limitations of a bubble plume or diffuse deep water system. These include accidental destratification caused by oxygen bubbles rising through the thermocline (Speece, 1994) and localized anoxia as a result of limited oxygen dispersion within the hypolimnion (Fast and Lorenzen, 1976). Third, hypolimnetic aeration is not recommended for the lakes with maximum depths less than 12 to 15 m or with a relatively small hypolimnetic volume (Cooke and Carison, 1989). In shallow lakes, the density gradient may not be sufficient to resist thermocline erosion that can results in unintentional destratification. However, a Speece Cone can be used for oxygenation of lake as shallow as 9 m (Doke et al, 1995). Beutel et al (2002) summarized costs, as well as advantages and disadvantages, for  10  various oxygenation systems, based on 5 tons/d systems installed in various locations (see Table 1.1). A Speece Cone has the lowest operating cost among others because of its extremely high efficiency in transferring dissolved oxygen into the hypolimnetic water, because of prolonged oxygen bubbles contact time and high rates of oxygen transfer (Speece et al., 1990; Beutel, 2002; Ashley, 2008). 1.2. Rationale for Current research  Numerous works have been completed and published about the physical, chemical, and biological effects of hypolimnetic aeration and oxygenation in lakes and reservoirs. These studies have been thoroughly reviewed by Fast and Lorenzen (1976), Lorenzen and Fast (1977), McQueen and Lean (1986), Beutel and Horn (1999), and Singleton and Little (2006). Although there are a large number of hypolimnetic aeration and oxygenation studies, only a few have examined the parameters that affect the performance of the system and few comprehensive analysis of design factors have been published on Speece Cone (Ashley, 2008). Hypolimnetic aeration is one of the most effective in-lake techniques for improving raw water quality. The Speece Cone is a promising technique of aerating/oxygenating of hypolimnion. However, there is very limited published information documenting the technical aspects of this device. Ashley (2002) investigated design factors that impact oxygen transfer in two different designs of hypolimnetic aeration systems; full lift and Speece Cone. The Speece Cone was improperly designed and the performance of the experimental Speece Cone was less than optimal. At present, a re-engineered Speece Cone was designed and manufactured. In a Speece Cone, water enters the cone at the top of the cone , energizing zone, with a  11  sufficient velocity to entrain injected oxygen gas and provide the energy to maintain a two-phase bubble swarm in the cone. The bubble swarm, the gas transfer zone, is a turbulent area that ensures a very high bubble to water interface and results in a high gas transfer rate. As the cross sectional area of the cone increases, the downward water velocity decreases to a velocity less than buoyant velocity of bubbles (i.e. 0.3 mIs). Thus, the key design parameters of a Speece Cone are inlet water velocity, gas flow rate to water flow rate ratio to achieve a complete gas dissolution, and proper cone dimensions to provide sufficient residence time for 100% gas transfer and maintaining all three zones simultaneously.( Speece at al., 1990; Ashley et al., 2008). The current research was conducted to provide a detailed analysis of design factors influencing oxygen transfer capabilities in a scaled —down Speece Cone. 1.3. Objectives  The objectives of the current research were: 1. To determine the effect of the ratio of gas flow rate to water flow rate on: o  KLa2O, the oxygen transfer coefficient at 20CC;  o  SOTR, the Standard Oxygen Transfer Rate;  o  SAE, the Standard Aeration efficiency and  o  SOTE the Standard Oxygen Transfer Efficiency. ,  2. To determine the effect of inlet water velocity, hence residence time on  KLa2O,  SOTR, SAE, and SOTE. 3. To compare alternative sources of oxygen: oxygen and air.  12  Pressure Swing Adsorption (PSA)  CHAPTER 2: EQUIPMENT AND METHODS 2.1. System Design  2.1.1. Aeration Tank Experiments were performed in a tank, located in the hydraulics laboratory of the Civil Engineering Department, at the University of British Columbia (UBC). The height of tank was 3.205 m with an internal width of 0.425 m and an internal height of 0.905 m. The water depth was kept 3.105 m during all the tests, so the operational volume was consistently 1,194 litres. The tank was equipped with two large (300 cm x 30 cm x 2.54 cm) windows of clear acrylic polymer at the front, to allow visual observations to be made during the experiments, and a self-sealed porthole with a diameter of 16 cm at one side; this could be opened from the outside when the tank was empty. A valve at the bottom of the other side was used for drainage (Fig 2.1). In order to minimize the deflection of the front panel and the sides of the tank, to minimize leaking when the tank was full to capacity, the tank was surrounded by a steel I-beam (7 mm) attached vertically in front; there were also two flat steel bars (12 mm thickness) that had been connected by two, threaded steel rods (with diameter of 8 mm) attached horizontally to the sides at the height of 82 cm, and 174 cm. A nylon band, with a width of 5 cm, was placed around the entire tank at the height of 1.5 m and tightened as much as possible with a cam ratchet. 2.1.2. Downflow Bubble Contact Aerator (Speece Cone) Aerator Dimensions The Speece Cone consisted of a cone-shaped contact chamber, with a height of 1.60 m, and a base outer diameter of 0.33 m ( Fig 2.2). The base of the cone was connected to a cylinder, which acted as a gas-liquid separation chamber, with height of 0.15 m and a  13  470  905  300  0  0 0 Cr)  1 -.  ..  .  -  Drain Vavie  Still IDW6t WOIlK  Fig 2.1- Schematic diagram of aeration tank  diameter the same as the base diameter of the cone. A submersible pump drew water from the bottom of the tank and sent it to the top of the cone via a 0.6 m vinyl hose with an inner diameter of 0.06 m that was connected to a pipe with the same diameter. The intake port consisted of seven replaceable throat inserts, with the inlet diameter of 0.06 m and outlet diameter of 0.016, 0.020, 0.023, 0.026, 0.028, 0.030, and 0.033 m respectively. The throat insert was connected by two flanges to the pipe at the top and to the inlet of the cone at the bottom. For each set of experiments, with same water inlet velocity, one of the intake throats was used. The outlet port was a pipe with an inner diameter of 0.06 m connected horizontally to the cylinder at the bottom of the cone. The water was discharged through the outlet  14  AWorPSAO2 Inlet  Throat insert Til,.  ,Cfl..,  Tk:,-I.  c  A1A  Vinyl Con nection Tubing  Probe 2 Current Meter  Probe I  Figure 2.2. Schematic of Speece Cone aerator with probe locations  15  port at the same depth as the water intake. The whole unit was installed on a base plate and was lowered to the bottom of the tank. 2.1.3. Sources of Water and Oxygen City water was used for these experiments. This water usually has total dissolved solids of about 22 mg/L and a pH of about 6.1, and overall a high quality for water. The manifold board was designed with three inputs, one for air coming straight from compressor, one for oxygen-rich air with 86-89% oxygen content coming from the AirSep unit, and one for oxygen gas, with a purity of 99.99% coming from the certified oxygen cylinder, and one output to the cone. This design of the manifold board allowed the system to work on any of the three gases, by adjusting the proper valves on the manifold and routing the desired gas to the cone. Compressed air was supplied by a rotary screw compressor located in the basement of the hydraulics lab and was distributed throughout valves located at several points of the laboratory. A 10 m compressed air hose, with an internal diameter of 1.9 cm, was used to deliver compressed air from the nearest distribution valve to the system. This compressed air was supplied either directly to the cone for the tests running on air, or to the oxygen generator for the tests running on oxygen (see Fig 2.3). The oxygen, used in the tests using oxygen, was supplied by an AirSep oxygen generator Model AS-20, that uses the Pressure Swing Adsorption (PSA) air separation process for producing oxygen. This process uses the fact that under pressure, nitrogen is adsorbed by a zeolite resin, and as the pressure is reduced, the nitrogen gas is desorbed (released). This unit was equipped with two solid beds. Nitrogen was adsorbed as pressurized air passed through one of the adsorbent beds. The gas coming out of the unit is enriched in  16  oxygen. While one of the solid beds produces oxygen, the other bed is regenerated by reducing the pressure and releasing the adsorbed nitrogen; then it was ready for another cycle of producing oxygen. The gas coming out from the unit contained 86-89% oxygen. The oxygen gas probe (number 7) was used to continuously measure the purity of the oxygen gas coming out from the AirSep oxygen generator and entering the experimental system. This probe was calibrated using the gas from a cylinder of certified oxygen gas with more than 99.99% purity, as the reference. 2.1.4. Measuring, Monitoring and Recording Equipment  Gas Flow Meter Distribution of various gases and regulation of their flow and pressure took place through a custom-designed, manifold board, manufactured by Pont Four Systems Inc. (Fig 2.3). Compressed air supplied to the system was first passed through a series of filters. This consisted of two filters: a Wilkerson 5.0  ji  particulate filter and a 0.01 micron filter.  Filtered, compressed air was then directed either to the AirSep unit, in order to produce oxygen for the tests with oxygen, or directly to the cone, for the tests with air via two different set of tubing; each was controlled by an on-off ball valve (Fig 2.3). The pressure of either the oxygen-rich air, inflowing from the AirSep  oxygen generator or the  compressed air, was adjusted by a single stage regulator, equipped with a pressure gauge. Three gas flow meters were mounted on the manifold board to measure and control the flow of the feed gas (either air or oxygen). They adjusted the gas flow via regulating valves. The first one was a Brooks Sho-rate flow meter, with a 150 mm scale used for indicating coarse range flow rates ranging from 2 to 46 L.min’; the second one was another Brooks Sho-rate flow meter, with a 150 mm scale used for indicating medium  17  range flow rates ranging from 2 to 12 L.min’ L.min ; and the last one was a Key 1 Instruments flow indicator, with a 80 mm scale used for indicating fine range flow rates, ranging from 0 to 3.5 L.min’. The first two units were designed to operate at 3.2 kg.cm 2 (45 psig), and the last one was designed to operate under 3.5 kg.cm 2 (50 psig), (while it operated under 3.2 kg.cm 2 (45 psig)). Therefore, a specific pressure correction factor (i.e. was applied for the readings on the fine coarse scale flow indicator to compensate for the lower operating pressure (i.e. 3.2 versus 3.5 bar) (Ashley, 2002). All of the readings on all of three flow meters were corrected by a density correction factor (i.e.  q  1.105/1.0) when operated on compressed air. No temperature correction factor was  applied, since all of the flow meters were calibrated for 21 °C (Ashley, 2002). A small portion of the inflowing gas was sent to a probe cup, in which an oxygen probe (probe 7) was installed. The purity of the oxygen in the inflowing gas was measured by this probe and the low-flow gas exited into an 11 cm cylinder, filled with water; thus, the gas flow could be adjusted on a low constant level by observing the bubbles. Certified oxygen gas (>99.99% oxygen) was used as the reference gas for calibrating this probe. This gas passed through a two-stage regulator, before contacting the probe.  18  Fig 2.3- Schematic diagram of manifold board with data logger and flow meters (From Ashley, 2002)  19  Dissolved Oxygen Meter and Data Logger TM probes were used for oxygen measurement in the system. OxyGuard probes OxyGuard have membrane-covered polarographic sensors that generate an electrical current when they measure oxygen, and a built-in temperature compensator (Fig 2.4). The current produced is proportional to the oxygen content, and is converted into a millivolt signal, which is then transmitted to the data logger. The probes were calibrated on a daily basis, calibration procedure will be described in section 2.3. Test Procedure. Probe 1 was installed at the inlet of the main pump, probe 2 was installed at the outlet of the Cone, and four probes were situated in the tank in four different depths (Figure 2.2)  Figure 1.1. has been removed to copyright restrictions. The information removed is a picture of PT4 Monitor, Oxygaurd probes and thermister probe (from Point Four Systems Inc. website: http://www.pointfour.com/)  Figure 2.4- PT4 Monitor, Oxygaurd probes and thermister probe (from Point Four Systems Inc. website)  20  To collect the operating data, a PT4 Monitor microprocessor-based, data logger, with eight channels was used (Figure 2.4). The unit was equipped with a temperature probe for measuring water temperature and seven OxyGuard probes  -  six for measuring dissolved  oxygen of the water used in the experiments and one for measuring the purity of the oxygen in the inlet gas. The maximum memory capacity for each channel was 894 readings. Sample averaging time, which is the period in which channel values are read and accumulated in the internal memory and the average value taken, and log interval, which is the time interval between one set of channel values being logged and the next set, can both be defined by the user. This data logger was programmed to log data within a time interval of 10 seconds and an averaging time of 5 seconds, for all the experiments. Water Velocity Measurement A Marsh McBirney Model 2000 flow meter was used to measure the velocity of the water. The flow meter measured the velocity of the discharge water over an averaging time of 120 sec using a sensor located at the outlet of the cone. The flow of the water was then calculated based on the measured water velocity using Equation 2.1 (pers. comm. Marsh McBirney Inc.).  (2.1)  Q=Vx(Ao—Am)  where:  Q  =  water flow rate, m 1 s 3  V= water velocity at the outlet (measured), ms ; 1 =  Am  area of the outlet, m 2 area of the probe (provided by manufacturer) = 2.63 cm 2  21  =  0.000263 m 2  The velocity of water at the inlet throat of the Speece Cone and the cone residence time were then calculated based on this flow rate (Equation 2.2. and Equation 2.3.) (2.2)  inlet water velocity (m.s ) 1  =  outlet water velocity (m.s’) x (outlet area (m )/inlet area 2 2 (m ) ) (2.3)  residence time (s)  =  volume of the cone (m ) / water flow rate 3 3 (m . s’)  Water Temperature Measurement Two instruments were used for measuring the water temperature. One was a stainless steel thermister probe, connected to the PT4 Monitor (Fig 2.1.3), and the other one was the temperature probe of an YSI meter.  Electrical Measurements The line voltage and the electrical current for the main pump of Speece Cone and the PSA unit were measured with a Fluke multi-meter (Model # 70-3) in voltage from 0  —  600 v, and 0- 40 amps.  2.3. Test Procedure  The following is a description of sequential steps that were performed during the execution of all the experiments: Each morning, the tank was filled with city water to the desired level within 50 minutes. The water in the tank was then circulated by using four pumps; two Little Giant pumps (4548L.hrjlocated at the bottom corners at one side of the tank, one pump(4548L.hr  22  1)  at  the top of the other side of the tank, and the main pump (15,600 L.hr’) of the Speece Cone. Dye tests were performed before the main experiments had been started, to ensure that the tank was totally mixed and there were no unmixed zones. In this test, red food dye was added to the tank, and it was observed that, in less than two minutes after the pumps were turned on, the 1,200 litres of water in the tank were completely mixed. Then, a PT4 monitor/data logger and a Toshiba laptop computer were turned on for monitoring, logging, transferring and storing the data. The compressed air valve and consequently the PSA oxygen generator unit were turned on, in order to produce oxygen and store it in the PSA storage unit, to a maximum pressure of 60 psig. Probe number 7, which was dedicated to measure the DO content of the gas inflowing to the system, was then calibrated to 99.99%, using oxygen gas of a standard cylinder with oxygen purity of more than 99.9%, as a reference. Probes numbered 1 to 6 monitored the amount of the dissolved oxygen, in percentage of saturation. In order to calibrate these probes, both the water DO content and the daily oxygen saturation were needed. In the first set of experiments (with throat insert inner diameter of 16 mm), tank water was sampled at the top in two BOD bottles and the dissolved oxygen of the two samples was measured by the Winkler Method, as described in Standard Methods (2004). The amount of dissolved oxygen used for calibrating the probes was the average of the two readings. For the remaining six sets of the experiments, using the Winkler Method was limited to once in two weeks. In these tests, a YSI meter was used to measure the dissolved oxygen of the water in the tank using a YSI dissolved oxygen /temperature probe in the tank. This YSI meter was calibrated each day using the air saturation  23  protocol and it was checked against the Winkler test, bi-weekly. Oxygen saturation was calculated daily using equation 2.4 (Colt, 1984):  * —  *  BP-PH20 760.0-PH20  where: C;  =  the dissolved oxygen air-solubility value (mg.L’) for the ambient barometric  pressure, temperature and vapor pressure of water; 760 = the dissolved oxygen air solubility value (mg.L’) for the barometric pressure C; equal to 760.0 mm Hg and ambient temperature; BP= barometric pressure in mm Hg; PH20 =  vapor pressure of water in mm Hg for the ambient temperature.  760 and C;  PH2O  were read from the reference tables in Colt (1984), based on ambient  temperature, measured by a thermometer installed in the lab by the tank. The barometric pressure was read from the website of the Weather Office of Environment Canada at www.weatheroffice.co.rn. The percentage of oxygen saturation of the test water was then calculated by dividing the water dissolved oxygen by the calculated dissolved oxygen air-solubility value (C) and multiplying the result by 100. The calculated value was used to calibrate probes number 1 to6. Since the tank was well-mixed, all the probes, except probe number 7, that was used for 24  measuring the percentage of oxygen in the inlet gas, and probe number 8 which was a ,  thermometer, were calibrated to this calculated number on the PT4 monitor, using a single point calibration method (Point Four Systems, 1997). Test water was de-oxygenated prior to each oxygenation procedure. The water was de oxygenated by adding sodium sulfite 3 SO and cobalt chloride 0 2 (Na ) .6H 2 (C0CI ). The chemical reaction is (Beak, 1977):  (2.5)  SO + 02 2 Na 3  -  SO 2 Na 4  The test water was deoxygenated with 0.1 mg.L 1 cobalt chloride as a catalyst and 10.0 mg.L’ of sodium sulfite for each 1.0 mg.L 1 of dissolved oxygen present in the volume of the tank. The cobalt chloride was added first and thoroughly mixed into the test water. Sodium sulfite was then added to a 1 L flask and the slurry was added to the tank. The tank was fully mixed by the pumps. Water was deoxygenated within 2-3 minutes to below 2% of oxygen saturation. Experiments were started when the tank was mixed for 8 minutes after this point. The air or oxygen supply was opened and its flow was adjusted to the desired level, via a flowmeter located on the manifold board. The PT4 monitor/logger was adjusted to monitor and log the data every five seconds and send them to a computer for storage. The inflow of the gas to the cone was terminated when the probe located in the inlet of the cone (Probe 1) showed 75% saturation. The main pump of the cone was turned off and the gas remained in the cone was released through an air relief system, to make certain that no additional oxygen would be added to the water in the tank. Then, the main pump  25  was turned on again and the system mixed for more two minutes, until the tank became homogenized. During this two-minute period, no new oxygen was introduced; thus, these 2 minutes were not counted in the time component of the oxygen transfer calculations. After termination of each test, the tank was mixed for 8 minutes before starting a new one (including deoxygentaion). For each tank of water, six tests were allowed to be conducted in order to prevent the interference from sodium sulfite accumulation in the tank (ASCE, 1992). Subsequently, the tank was drained, using the drainage valve of the tank, and refilled for the next set of six experiments.  2.4. Experimental Design 2.4.1. Preliminary Experiments  In order to identify the hydraulic characteristics of the cone, a set of preliminary tests was performed. In these tests, for each individual water flow rates, the gas flow rate was changed from 0.5% ratio of gas flow rate to water flow rate, up to where the gas bubbles started to leave the cone through the outlet port. The cone became positively buoyant at the high ratios of gas flow rate to water flow rates (i.e.> 5-6%). Based on these sets of experiments a range of 0.5% to 5% was chosen for the ratio of gas flow rate to water flow rate, to be tested in the main experiments.  2.4.2. Main Experiments  Experiments were conducted to evaluate the effect of two experimental factors on the performance of the Speece Cone:  26  1- Ratio of oxygen flow rate to water flow rate 2- Inlet velocity The performances of the cone, using two alternative sources of oxygen (i.e. PSA generated oxygen and air) were compared to each other. The tests were arranged into seven main groups, based on seven different inlet water velocities. Each group consisted of two subgroups based on the source of the oxygen: one running on air, and one running on PSA oxygen. The operational depth was the same for all the groups, and the gas mixture was the same for tests running on PSA generated oxygen. Each sub-group included ten treatments, based on the ten different ratios of gas flow rate to water flow rate; this ranged from 0.5% to 5%, with increments of 0.5%. There were four replications for each treatment, as 3 was the minimum number of replicates recommended for non-steady reaeration tests (ASCE 1992). In each group, the tests were conducted on a random basis, designed by the Random Number Generation analysis tool of Microsoft Excel (in order to avoid errors that might occur during any given treatment day). The treatments groups are shown in Tables 2.1 to 2.7.  Table 2.1 Inlet N O•  —  Experimental combination for Group 1  water  velocity (m.s’)  Gas composition  Gas flowrate/water flowrate  (%)  No. of combinations  1A  8.5  Air  O.5,1.O,1.5,...,4.5,5.O  10  lx  8.5  PSA  0.5,1.0,1.5,.. .,4.5,5.0  10  27  Table 2.2 Inlet N O•  —  Experimental combination for Group 2  water  velocity (m.s’)  Gas composition  Gas flowrate/water flowrate  (%)  No. of combinations  2A  8.0  Air  0.5,1.0,1.5,...,4.5,5.0  10  2X  8.0  PSA  0.5,1.0,1.5,.. .,4.5,5.0  10  Table 2.3 Inlet N O•  —  Experimental combination for Group 3  water  velocity (m.s ) 1  Gas composition  Gas flowrate/water flowrate  (%)  No. of combinations  3A  7.6  Air  0.5,l.0,1.5,...,4.5,5.0  10  3X  7.6  PSA  0.5,1.0,1.5,.. .,4.5,5.0  10  Table 2.4 NO  Inlet water velocity (m.s’)  —  Experimental combination for Group 4  Gas composition  Gas flowrate/water flowrate (%)  No. of combinations  4A  7.4  Air  0.5,1.0,1.5,...,4.5,5.0  10  4X  7.4  PSA  0.5,1.0,1.5,.. .,4.5,5.0  10  Table 2.5— Experimental combination for Group 5  NO  Inlet water velocity (m.s’)  5A  6.9  Air  0.5,1.0,1.5,...,4.5,5.0  10  5X  6.9  PSA  0.5,1.0,1.5,.. .,4.5,5.0  10  Gas composition  Gas flowrate/water flowrate (%)  28  No. of combinations  Table 2.6— Experimental combination for Group 6 N  Inlet water velocity (m.s’)  6A  6.4  Air  0.5,l.0,1.5,...,4.5,5.0  10  6X  6.4  PSA  0.5,1.0,1.5,.. .,4.5,5.0  10  Gas composition  Table 2.7 N O•  Inlet water velocity (m.s’)  —  Gas flowrate/water flowrate  (%)  No. of combinations  Experimental combination for Group 7  Gas composition  Gas flowrate/water flowrate  (%)  No. of combinations  7A  5.7  Air  0.5,1.0,1.5,...,4.5,5.0  10  7X  5.7  PSA  0.5,1.0,1.5,. ..,4.5,5.0  10  A total of 560 individual re-aeration experiments were completed (Table 2.8). Each test was named uniquely to avoid any repetition. The termination of the experiments has already been described in test procedure, Section 2.3.  29  Table 2.8  —  Summary of treatments for Speece Cone aeration experiments  No.  No. of combinations  Replicates  Total number of tests  1A  10  4  40  lx  10  4  40  2A  10  4  40  2X  10  4  40  3A  10  4  40  3X  10  4  40  4A  10  4  40  4X  10  4  40  5A  10  4  40  5X  10  4  40  6A  10  4  40  6X  10  4  40  7A  10  4  40  7X  10  4  40 Total  30  560  2.5. Parameter Estimation The performance of the Speece Cone was evaluated on the basis of the amount of oxygen transferred per unit of air/PSA oxygen introduced to the water. The overall oxygen transfer coefficient, KLa, was determined using the log deficit technique, for all the experiments. In this method, dissolved oxygen (DO) deficit was plotted vs. time on a semi-logarithmic graph, and the value of KLa was found from the slope of the line. Equation (2.6) was used to calculate the value of KLaT.  (2.6)  KLaT=  (Ln [(Cs*_ 1 C ) ! (Cs*_ 2 C ) ])! (t2 -ti)  where:  KLaT= oxygen transfer coefficient at the temperature of the wastewater (hrj; C’= dissolved oxygen saturation concentration, (mg.L ) for the ambient barometric 1 pressure, temperature and vapour pressure of water; = dissolved oxygen concentration at point 1 (mg.Lj; 1 C = dissolved oxygen concentration at point 2, 1 2 C (mg.L ) ; t time at point = 2 t  1 (hr);  time at point 2 (hr).  Point 1 and point 2 were chosen as the points at which the measured oxygen concentration was  10% and 75% of the saturation value for the test water, corrected for  temperature, vapour pressure and barometric pressure, respectively. KLa was then converted to KLa2O according to: (2.7)  KLa2O = KLaT  (20-T)  where:  31  T= test water temperature (°C) KLaT= oxygen transfer coefficient at temperature T (hr’); KLa2O= oxygen transfer coefficient at 20°C (hrj. The typical value of 0 for aeration devices is 1.024 (Tchobanoglous et al, 2003).  SOTR was then calculated as: (2.8)  SOTR= KLa2O C 20 V  where: SOTR= standard oxygen transfer rate (mg 0 .hf’); 2 KLa2O  Cs20  =  =  oxygen transfer coefficient at 20°C (hf ); 1 dissolved oxygen concentration (mg.Lj at saturation for 20°C and standard  pressure (760 mm Hg);  V = volume of water in the tank (m ). 3  Standard Aeration Efficiency (SAE) was calculated according to: (2.9)  SAE  =  SOTR / power input  where: SAE  =  standard aeration efficiency (mg 02. (kWhr ); 1  SOTR= standard oxygen transfer rate (mg 0 .hr’); 2 power input  =  total delivered power (kW)  Total delivered power was calculated as described in the next section.  32  Power calculations The power input was divided to delivered blower power and wire power. Delivered blower power is the “theoretical power required at a blower discharge to deliver a given mass flow of gas at a given discharge pressure, calculated based upon adiabatic compression” (ASCE, 1992). The delivered blower power was estimated considering an adiabatic compression (PV= constant) (ASCE, 1992):  (2.10)  -1] 283 P= wRTj /29.7ne [(p2/pl)°  where: P= power input (kW); w= weight of air flow (kgs’); R= gas constant for air  8.314 kJ(k mol0K);  = absolute inlet temperature before compression; 1 T pi=  absolute inlet pressure before compression;  P2  absolute inlet pressure after compression;  n = 0.283 for air (and oxygen); 29.7  =  constant for SI conversion;  e = compressor efficiency= 0.80.  The wire power was calculated according to:  33  (2.11)  P=I.V  where: P = wire power (W); I  current (A) and V = potential difference (V)  For the experiments running on air, the power input was separated into three compartments; two types of delivered blower power and one wire power. The first one was the delivered blower power, considering adiabatic compression, from the manifold board regulator (i.e. 4.2 2 kg.cm ) . The second one was the same adiabatic compression formula adjusted to the absolute ambient hydrostatic pressure of gas release (m). The third one was the calculated wire power of the main pump of the cone, based on the measured current passing through, and the measured potential difference. For experiments on PSA oxygen, the power input was separated into four components; two types of delivered blower power: one, the delivered blower power required to deliver the mass flow of gas at the absolute minimum pressure requirement of PSA unit (i.e. 7.36 kg.cm ) 2 , and the second one, the same adiabatic compression formula adjusted to the absolute ambient hydrostatic pressure of the gas release (m); and two types of wire power: one, power of the main pump of the cone and the other one, the power of the PSA unit. The sum of the components was equal to the total delivered power (Table 2.9).  34  Table 2.9. Summary of energy calculations Treatment  PSA Oxygen  Air  adiabatic compression  adiabatic compression  2 @ 7.36 kg/cm  2 @ 4.2 kg/cm  adiabatic compression  adiabatic compression  El  E2 @ m depth  m depth  Average measured wire power Average  measured  wire  E3 power of main pump  of main pump Average measured wire power E4 of PSA unit  Where applicable, all data is presented on a “log’ basis, rather than “Ln” basis, for ease of comparison to previously published values for transfer efficiencies. SOTE was finally calculated as: SOTE  (2.12)  =  SOTR / WO2  Where: SOTE  =  standard oxygen transfer efficiency (%)  SOTR = standard oxygen transfer rate (mg 0 .hr’) 2 02 w  =  mass flow rate of oxygen in the gas flow stream (mg 1 .hr 2 0 )  [Note: wo2 was  calculated using the standard procedure as reported by Ashley (2002)j  35  Chapter 3. Results This study was performed in seven groups. For each group, all of the results are plotted and the average of each combination, including four replicates, is defined. The average values of each group was plotted to determine the trend. Among all the 560 sets of data obtained, there were 4 outliers. One set was not applicable, due to “cone float” in the middle of the experiment. Overall, less than 1% of data were outliers. Data points in the figures illustrated in Sections 3.1 to 3.7 are a mean value of four replicates, with the standard deviation of less than 5% of mean value. This indicates the high accuracy of the experiments.  3.1. Group 1- Water inlet velocity of 8.5 m.s 1 treatment on air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters  KLa2O,  SOTR,  SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.l.a, 3.1.b, 3.1.c, and 3.1.d, respectively. For PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 2.0% ratio, the average KLa2O and SOTR were increased 3.8 fold, while the average SAE was increased 33.3 fold; after that, over the range of 2.0% to 3.0% ratio, the slope of the curve decreased to 1.9 for KLa2O and SOTR, and to 16.6 for SAE. Finally, over the range of 3.0% to 5.0% ratio of gas flowrate to water flowrate, the slope of the curve increased to 3.8 for KLa2O and SOTR, and to 38.9 for SAE. For air, the average KLa2O and SOTR were increased 0.7 fold, with increasing the ratio of gas flowrate to water flowrate, while the SAE increased 6 fold. The SOTE decreased, with a slope of -6.6 for oxygen and -4.0 for air, as the ratio of gas flowrate to water flowrate increased.  36  30  PSA Average 25 -  —  A A  PSA  A -  -  Air Average  A  20 •  A  Air  I  I..  £15 C (.4  A  Cu -J  A  10 A  5  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  Gas flow rate! Water flow rate  4.5  5.0  5.5  (%)  Fig. 3.1.a. KLa2O for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  300 PSA Average  A A  250 -  .c  PSA  A -  -  -  A  Air Average  -  200 •  Air  0  ,j 150  --  A  A  I  g  A  :z::::::r:  1::  --  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  (%)  Fig. 3.1.b. SOTR for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  37  5.5  250 PSA Average 200 -  -  A A  PSA  A -  -Air Average  A  b..  A  •  150  Air  ---1---A  0)  100 A  Co A  50  ILIihI  0  -----  0.0  0.5  1.0  1.5  -  2.0  2.5  3.0  3.5  Gas flow ratel Water flow rate  4.0  4.5  5.0  5.5  (%)  Fig. 3.1.c. SAE for Water Inlet Velocity of 8.5 m.s 1 on PSA and Air  140 PSA Average  120 100 80 Q Lii I g60 40 20  0  5  _  Gas flow ratel Water flow rate  (%)  Fig. 3.1.d. SOTE for Water Inlet Velocity of 8.5 m.s’ on PSA and Air  38  3.2. Group 2- Water inlet velocity of 8.0 m.s 1 treatment on Air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters  KLa2O,  SOTR,  SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.2.a, 3.2.b, 3.2.c, and 3.2.d, respectively. For PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 2.0% ratio, the average KLa2Q and SOTR were increased 5.9 fold, while the average SAE was increased 51.0 fold; after that, over the range of 2.0% to 4.0% ratio, the slope of the curve decreased to 2.2 for KLa2O and SOTR and to 17.6 for SAE. Finally, over the range of 4.0% to 5.0% ratio of gas flowrate to water flowrate, the slope of the curve increased to 6.4 for KLa2O and SOTR, and to 53.6 for SAE. For air, the average KLa2O and SOTR were increased 0.9 fold, with increasing the ratio of gas flowrate to water flowrate, while the SAE increased 8.2 fold. The SOTE decreased with a slope of 8.3 for oxygen and -2.2 for air, as the ratio of gas flowrate to water flowrate increased.  35  --___________________  PSA Average  -  -  -  -  a-  -  -  -  -  -  -t  -  o 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  Gas flow rate! Water flow rate (%)  Fig. 3.2.a. KLa2O for Water Inlet Velocity of 8.0 m.s 1 on PSA and Air 39  5.5  _-  350  -----------  -----  -  --  ---  -—-------  --  ——-  PSA Average  -  A  —  —  —  —  —  4—  —  —  —  —  -I—  —  —  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  Gas flow rate! Water flow rate (%)  Fig. 3.2.b. SOTR for Water Inlet Velocity of 8.0 m.s on PSA and Air  300  —  PSA  Averagel  Gas flow rate! Water flow rate (%)  Fig. 3.2.c. SAE for Water Inlet Velocity of 8.0 m.s 1 on PSA and Air  40  5.5  140 PSA Average 120  A  PSA  40 20 0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  5.5  (%)  Fig. 3.2.d. SOTE for Water Inlet Velocity of 8.0 m.s’ on PSA and Air  3.3. Group 3- Water inlet velocity of 7.6 m.s treatment on Air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters KLa2O, SOTR, SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.3.a, 3.3.b, 3.3.c, and 3.3.d, respectively. For PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 2.0% ratio, the average  KLa2O  and SOTR were increased 9.2 fold, while  the average SAE was increased 79.6 fold; after that, over the range of 2.0% to 3.5% ratio, the slope of the curve decreased to 4.2 for KLa2O and SOTR and to 34.5 for SAE. Finally, over the range of 3.5% to 5.0% ratio of gas flowrate to water flowrate, the slope of the curve increased to 5.3 for KLa2O and SOTR, and to 42.8 for SAE. For air, the average KLa2O and SOTR were increased 1.1 fold, with increasing the ratio of gas flowrate to water flowrate, while the SAE increased 10.4 fold. The SOTE decreased with a slope of 41  -  7.7 for oxygen and -1.8 for air, as the ratio of gas flowrate to water flowrate increased.  45  -  —  PSA Average  40  -  •1..  .t 0 CN  CD  -J  to  1.52M  233.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  5.5  (%)  Fig. 3.3.a. KLa2O for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  450 PSA Average  A  400  A  PSA  A  350 -  1  -  -  A  Air Average  300 •  CN  Air  A  0 250 0)  I  0  Cl)  200 150 100 50  -.  .1I  -  -.-----  -  i_.___  t-----I-  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  (%)  Fig. 3.3.b. SOTR for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  42  5.5  350 PSA Average  A  1250  —  —  0 0.0  .  a— 1.0  0.5  —  —  —  1.5  —.  —  —  2.0  2.5  3.0  3.5  Gas flow ratel Water flow rate  4.0  4.5  5.0  5.5  (%)  Fig. 3.3.c. SAE for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  140 PSA Average 120  A  100  PSA AirAverage  : 40  -  -  I-1---i --‘----!-  ----  20  -  0  -_________________________________  0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  (%)  Fig. 3.3.d. SOTE for Water Inlet Velocity of 7.6 m.s 1 on PSA and Air  43  5.5  3.4. Group 4- Water inlet velocity of 7.4 1 m.s treatment on air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters  KLa2O,  SOTR,  SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.4.a, 3.4.b, 3.4.c, and 3.4.d, respectively. For PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 3.0% ratio, the average KLa2O and SOTR were increased 7.4 fold, while the average SAE was increased 62.8 fold; after that, over the range of 3.0% to 4.0% ratio, the slope of the curve decreased to 2.1 for KLa2O and SOTR and to 14.9 for SAE. Finally, over the range of 4.0% to 5.0% ratio of gas flowrate to water flowrate, the slope of the curve increased to 3.8 for KLa2O and SOTR, and to 41.3 for SAE. For air, the average KLa2O and SOTR were increased 1.2 fold, with increasing the ratio of gas flowrate to water flowrate, while the SAE increased 9.8 fold. The SOTE decreased with a slope of 10.1 for oxygen and -1.6 for air, as the ratio of gas flowrate to water flowrate increased.  40  —  PSA Average 35  -  PSA  A  30---  25  -  -  -  A  A  A A  -Air Average  •  A-----A--  Air  A  £  20  -  A  A  A  A  A  1510  A  I----  5--  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  Fig. 3.4.a.  KLa2O for Water Inlet  4.0  4.5  5.0  (%)  Velocity of 7.4 m.s 1 on PSA and Air 44  5.5  -  450 PSA Average 400  A  PSA  A  --  350  A -  -  -  -  Air Average  V 300 .  o  •  Air  A  I  Cl)  A  A  250  A  0)  0  A  A  200  A  A  A  -  -  -  -A  A  150 A  100  --I  50 0 0.0  1.0  0.5  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  5.5  (%)  Fig. 3.4.b. SOTR for Water Inlet Velocity of 7.4 m.s on PSA and Air  350  —________________________________________________________  PSA Average 300  -A-  PSA  A  EI 50  I—  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  (%)  Fig. 3.4.c. SAE for Inlet Velocity of 7.4 m.s 1 on PSA and Air  45  5.0  5.5  140  -___________  -  -  —  PSA Average 120  -  A  100  PSA AirAverage  Gas flow ratel Water flow rate  (%)  Fig. 3.4.d. SOTE for Water Inlet Velocity of 7.4 m.s 1 on PSA and Air  3.5. Group 5- Water inlet velocity of 6.9 m.s 1 treatment on Air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters, SOTR, SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.5.a, 3.5.b, 3.5.c, and 3.5.d, respectively. For, PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 3.0% ratio, the average  KLa2O  and SOTR were increased 12.3 fold, while  the average SAE was increased 104.4 fold; after that, over the range of 3.0% to 5.0% ratio, the slope of the curve decreased to 1.6 for KLa2O and SOTR and to 9.4 for SAE. For air, the average  KLa2O  and SOTR were increased 1.2 fold, with increasing the ratio of gas  flowrate to water flowrate, while the SAE increased 9.8 fold. The SOTE decreased with a slope of -8.3 for oxygen and -3.2 for air, as the ratio of gas flowrate to water flowrate increased.  46  -  60  _____  —  PSA Average 50  PSA  A  A  A -  -  -  -  Air Average  40— •  Air  I .  A  30-  -  0 C4  A  A A  A  -J  -A----  20-  10  0.0  1.0  0.5  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  5.5  (%  Fig. 3.5.a. KLa2O for Water Inlet Velocity of 6.9 m.s on PSA and Air  700 PSA Average  600 500-  o  -  -  -  A  A  -AirAverage  •  400  -  PSA  A  Air  I-.  o U)  A  A  A A A  200  - - —--- -  •  100  -.--.II---I---I---.---.---I---.  1  0 0.0  0.5  -  -  ¶— 1.0  -  1.5  —--2.0  2.5  3.0  3.5  4.0  4.5  5.0  Gas flow rate! Water flow rate (°/  Fig. 3.5.b. SOTR for Water Inlet Velocity of 6.9 m.s 1 on PSA and Air  47  5.5  500 -  PSA Average  450  I Gas flow rate! Water flow rate (°/  Fig. 3.5.c. SAE for Water Inlet Velocity of 6.9 m.s 1 on PSA and Air  140 PSA Average 120  A  100  20  PSA AirAverage  f  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  Gas flow rate! Water flow rate (°/  Fig. 3.5.d. SOTE for Water Inlet Velocity of 6.9 m.s 1 on PSA and Air  48  5.5  3.6. Group 6- Water inlet velocity of 6.4 m.s 1 treatment on air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters  KLa2O,  SOTR,  SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.6.a, 3.6.b, 3.6.c, and 3.6.d, respectively. For PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 2.0% ratio, the average KLa2O and SOTR were increased 12.0 fold, while the average SAE was increased 102.1 fold; after that, over the range of 2.0% to 4.0% ratio, the slope of the curve decreased to 3.4 for  KLa2O  and SOTR and to 26.0 for SAE.  Finally, over the range of 4.0% to 5.0% ratio of gas flowrate to water flowrate, the slope of the curve increased to 11.5 for KLa2O and SOTR, and to 91.9 for SAE. For air, the average KLa2O and SOTR were increased 1.2 fold, with increasing the ratio of gas flowrate to water flowrate, while the SAE increased 9.8 fold. The SOTE decreased with a slope of -7.3 for oxygen and -3.0 for air, as the ratio of gas flowrate to water flowrate increased.  49  80  —  PSA Average 70  -  60 -  -  A  PSA  A -  -  •--  Air Average  I Gas flow rate! Water flow rate (°/  Fig. 3.6.a. KLa2O for Water Inlet Velocity of 6.4 m.s’ on PSA and Air  800 r PSA Average  A  700 PSA  A 600  -  -  -  -  Air Average  I: Gas flow rate! Water flow rate (°/  Fig. 3.6.b. SOTR for Inlet Velocity of 6.4 m.s 1 on PSA and Air  50  600 PSA Average  50O4  PSA  A -  I  -  A  -  -  Air Average  400 •  Air A  300—  --  -  A  0) Iii  Co  200  A  A  3.0  3.5  A  —  A  A  100  0 0.0  0.5  1.0  1.5  2.0  2.5  4.0  4.5  5.0  5.5  Gas flow rate! Water flow rate (°/  Fig. 3.6.c. SAE for Inlet Velocity of 6.4 m.s’ on PSA and Air  140 PSA Average 120  PSA  A  100  -  -  -AirAverage  80 Lii  I 60 40 20 0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  Gas flow rate! Water flow rate  4.0  4.5  5.0  (%  Fig. 3.6.d. SOTE for Water Inlet Velocity of 6.4 m.s’ on PSA and Air  51  5.5  3.7. Group 7- Water inlet velocity of 5.7 m.s 1 treatment on air and PSA oxygen The effect of the ratio of gas flow rate to water flow rate on the parameters  KLa2O,  SOTR,  SAE and SOTE, for air and PSA oxygen, is shown in Figures 3.7.a, 3.7.b, 3.7.c, and 3.7.d, respectively. For PSA oxygen, with increasing the ratio of gas flow rate to water flow rate, over the range of 0.5% to 2.5% ratio, the average KLa2O and SOTR were increased 10.0 fold, while the average SAE was increased 87.5 fold; after that, over the range of 2.5% to 3.5% ratio, the slope of the curve decreased to 4.3 for KLa2O and SOTR and to 32.4 for SAE. Finally, over the range of 3.5% to 5.0% ratio of gas flowrate to water flowrate, the slope of the curve increased to 11.2 for KLa2O and SOTR, and to 88.5 for SAE. For air, the average KLa2O and SOTR were increased 1.0 fold, with increasing the ratio of gas flowrate to water flowrate, while the SAE increased 8.3 fold. The SOTE decreased with a slope of 7.1 for oxygen and -3.8 for air, as the ratio of gas flowrate to water flowrate increased.  52  -  80  --______  PSA Average 70  —  PSA  A  60--  -  50—  -  -  •  A A  ---------  Air Average Air  I  A  40 C -J  A  A 30  -  A  20  0.5  1.0  1.5  4.0  5.5  Gas flow ratel Water flow rate (°/  Fig. 3.7.a. KLa2O for Water Inlet Velocity of 5.7 m.s 1 on PSA and Air  800 PSA Average A  700 PSA  A -  500  -  -  -  •  Air Average Air  0 .  A  A  600  A  A  4.5  5.0  400 A 300  U)  A 200 100  •  -  0 0.0  0.5  1.0  -  I 1.5  -  -  •-  2.0  -  I  -  2.5  -  3.0  3.5  4.0  Gas flow ratel Water flow rate (°/  Fig. 3.7.b. SOTR for Water Inlet Velocity of 5.7 m.s 1 on PSA and Air  53  5.5  600 PSA Average 500  PSA  A -  -  A  -  -  Air Average  400 •  Air A  6’  300  A  A A  -  LU  200  100  I---. 0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  Gas flow rate! Water flow rate (°/  Fig. 3.7.c. SAE for Water Inlet Velocity of 5.7 m.s 1 on PSA and Air  140 PSA Average 120 100  PSA  A -  •  -  -  -  Air Average  80  w  I-. 60 40 20 0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  Gas flow rate! Water flow rate (°,Q Fig. 3.7.d. SOTE for Water Inlet Velocity of 5.7 m.s’ on PSA and Air  54  5.5  3.8. Inlet Water Velocity  The effect of inlet water velocity on  KLa2O,  SOTR, SAE, and SOTE are shown in Figure  3-8 to 3-15 for experiments running on PSA oxygen and on air, respectively. According to the figures, as the inlet water velocity increases, KLa2O, SOTR, and SAE decreased for any ratio of gas flowrate to water flowrate, in the range of 0.5% to 5.0% of gas flowrate / water flowrate, for both oxygen and air. The rate of change of KLa2O, SOTR, and SAE became slower as the inlet water velocity increased. SOTE was increased as the water velocity increased at any given ratio of gas flowrate to water flowrate, in the range of 0.5% to 5.0%.  Water Velocity ——-5.7 rn’s -+—- 6.4rn’s  —A—6.9 rn/s 0 CN  .—a---7.4 rn/s  -J  —.--—7.6 rn/s —e---8.Ornfs —%--8.5 rn’s 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flowrate  Figure 3.8. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on KLa2O  55  Water velocity 5.7 m/s —+—6.4rrils  0  —A—-- 6.9 rn’s  I  —a——7.4 rn/s  0)  0 U)  —-—7.6mls —-—8.0 rns  —x—8.5 rn’s 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowrate!Water flowrate  Figure 3.9. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on SOTR. 120 Water velocity 100  —.--—5.7rns  —+----6.4rrVs  80  6.9 rn/s  60  7.4 rn’s 40-  —-—7.6rWs  20-  -  —o-—8.Orn’s ——8.5 rn’s  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flowrate  Figure 3.10. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on SOTE.  56  450 Water velocity  400  —s-—5.7 rTVs  350  r 8  —m-—.6.4 rn’s  300 250  —.k-— 6.9rrVs  200  —ci-—7.4rnfs  uJ Cl)  150  ———7.6rn’s  100 —o—8.0 m’s 50 —)— 8.5 in’s  0 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flowrate  Figure 3.11. The effect of inlet water velocity and the ratio of oxygen flow rate to water flow rate on SAE. 7  Water Velocity  6  —--- 5.7 m/s 5 —+-. 6.4rrVs ‘;.I_  4  s 1 —k-— 6.9 rn  0 c’1  —.c— 7.4 mIs —.-—7.6 ms 2  —.0.-— 8.Orn’s —x—8.5 m’s 0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % Air flowrate/Water flowrate  Figure 3.12. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on KLa2O  57  Water velocity —e-— 5.7m/s 60 —±---6.4 rn/s 50 —*—6.9 rn’s  0  0) 40 —G—-- 7.4m/s  I  0 U)  —.—7.6 rn/s —<--—8.0 rn/s  —z— 8.5 mis 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % Air flow rate!Water flowrate  Figure 3.13. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on SOTR.  60  Water velocity ——5.7 rn/s  50  —+----6.4  40-----  w  rn’s  —a—-6.9 rn/s 30  0  —u—7.4  U)  20  rn/s  -  ——8.0 m’s  10  —---8.5 rn/s  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  %Air flowrateIWater flowrate  Figure 3.14. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on SOTE.  58  Water velocity —.—5.7 m s 1 —+—- 6.4nVs I..  ——6.9 rn/s  x  6  —o—7.4 m’s  w Cl)  —.-—7.6 m’s —c-—8.O rn’s ——8.5 rn/s 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % Air flowratelWater flowrate  Figure 3.15. The effect of inlet water velocity and the ratio of air flow rate to water flow rate on SAE.  3.9. Hydraulic Residence Time The effect of hydraulic residence time on  KLa2O,  SOTR, SAE, and SOTE are shown in  Figure 3-16 to 3-23 for experiments running on PSA oxygen and on air, respectively. According to the figures, as the hydraulic residence time decreases, KLa2O, SOTR, and SAE increased for any ratio of gas flowrate to water flowrate, in the range of 0.5% to 5.0% of gas flowrate / water flowrate, for both oxygen and air. The rate of change of KLa2O, SOTR, and SAE became slower as the hydraulic residence time increased. SOTE was increased as the hydraulic residence time increased at any given ratio of gas flowrate to water flowrate, in the range of 0.5% to 5.0%.  59  HRT  (s) —.—  13  —t--—  14  —*—- 15  I.. 0 c’1  —tD— 17 —.—  21  —o— 26 ——38  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flow rate  Figure 3.16. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on KLa2Q  600 HR1  (s) 500  —13 —+— 14  .  400  •1  —A— 15 300 —0. 17  200 —0—26  100  —z— 38 0 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flowrate  Figure 3.17. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on SOTR.  60  120 HRT  (s) —s-— 13  100  .—+— 14  80 LU I  60 —.0—— 17  0  Cl)  40 ———26  20  —x—38  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flowrate  Figure 3.18. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on SOTE. 450 HRT  (s)  400 —.——  13  ——  14  350 300 I..  —k—— 15  250 200  —0—- 17  -  LU U)  150 100  —0’— 26  —x— 38 0 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % PSA Oxygen flowratelWater flowrate  Figure 3.19. The effect of hydraulic residence time and the ratio of oxygen flow rate to water flow rate on SAE.  61  7 HRT  (s) 6  •  13  —+— 14  5  —*—- 15 ,I.  —0--— 17  0 C”  J3  • 2  21  —0--— 26 —— 38  0 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % Air flowratelWater flowrate  Figure 3.20. The effect of hydraulic residence time and the ratio of air flow rate to water flow rate on KLa2O  80  HRT  (s)  70  —.1--— 13  60 —+— 14 ‘ri-  —A— 15  0  0) 40  I-  o C,)  —0— 17  30 —4-— 21  20 —o-—26  10  —--38  0 0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % Air flowratelWater flowrate  Figure 3.21. The effect of hydraulic residence time and the ratio of air flow rate to water flow rate on SOTR.  62  60  HRT (s)  50  40  --  —+— 14 -  uJ I 0 U)  20  -  —.—21 10 —x—38  0 0.0  0.5  1.0  1.5  2.0  2.5  3.0  4.0  3.5  4.5  5.0  5.5  %Air flowratelWater flowrate  Figure 3.22. The effect of hydraulic residence time and the ratio of air flow rate to water flow rate on SOTE.  60 HRT  (s) —4--— 13  50  —+— 14 I  —*— 15  3o  --.0—17  w  —e-— 21  20  10 —-- 38 0  0.5  1.0  1.5  2.0  2.5  3.0  3.5  4.0  4.5  5.0  5.5  % Air flowratelWater flowrate  Figure 3.23. The effect of hydraulic residence time and the ratio of air flow rate to water flow rate on SAE.  63  8.5 • 8.0 07.6 KLa2o  ) 1 (hr  O  7.4  • 6.9 El 6.4 • 5.7 Inlet Water Velocity (ms ) 1 U)  Oxygen flow ratelWater flow rate (%  Figure 3.24. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet  water velocity on 1 K 2 a 0  c 8.5 • 8.0 O 7.6  SOTR Ihr) 2 (g O  O 7.4  • 6.9 D 6.4 • 5.7 In let Water Velocity (ms ) 1 Oxygen flow rate/Water flow rate (%) Figure 3.25. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet  water velocity on SOTR  64  D 8.5  • 8.O D7.6  SAE (g O2IKWhr)  O 7.4  •6.9 5 D6.4 • 5.7  Inlet Water Velocity (ms ) 1 Oxygen flow ratelWater flow rate (%)  Figure 3.26. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet water velocity on SAE  •8.O  SOTE  o 7.6  (%)  07.4 •6.9 5.7  c)  06.4 •5.7  Inlet Water Velocity (ms ) 1 It)  Oxygen flow ratelWater flow rate (%)  Figure  3.27. Effect of the ratio of PSA Oxygen flowrate to water flowrate and inlet water velocity on SOTE  65  3.10. Oxygen vs. air The calculated mean values and standard deviation for KLa2O, SOTR, SAE, and SOTE for experiments on PSA oxygen and air for all the seven groups are shown in Tables 3.1 to 3.7, for comparative purposes. Each group has ten sub-groups. For each group, inlet water velocity was constant, and for sub-groups a, b, c, d, e, f, g, h, i, and  j,  the  percentage of gas flow rate to water flow rate was 0.5%, 0.1%, 1.5%, 2.0%, 2.5%, 3.0%, 3.5%, 4.0%, 4.5%, and 5.0%, respectively. As observed, the mean values obtained for  KLa2O, SOTR, SAE, and SOTE for the tests on air were significantly lower than for the tests on PSA oxygen. Table 3.1.a: Least squares means (± STDEV), for inlet water velocity = 8.5 m.s 1 and gas flow rate/water flow rate = 0.5%  Treatment KLa2O  SOTR  SAE  SOTE  (hr’)  (g0 ! 2 hr)  (gO k 2 W!hr)  (%)  Air  0.50(0.05)  5.41(0.54)  4.47(0.45)  50.10(5.04)  4  Oxygen  4.63(0.32)  50.13(3.43)  40.83(2.80)  96.67(7.44)  4  N  Table 3.1.b: Least squares means ( ± STDEV), for inlet water velocity = 8.5 m.s 1 and gas flow rate/water flow rate 1.0% KLa2o  SOTR  SAE  SOTE  (hr’)  (g02/hr)  (gO k 2 W/hr)  (%)  Air  0.97(0.27)  10.52(2.92)  8.68(2.41)  53.12(14.76)  4  Oxygen  5.48(0.08)  59.39(0.85)  48.26(0.69)  70.23(1.00)  4  Treatment  66  N  Table 3.1.c: Least squares means (± STDEV), for inlet water velocity flow rate/water flow rate = 1.5%  8.5 m.s 1 and gas  SOTR  SAE  SOTE  (i)  (g02/hr)  kWIhr) 2 (gO  (%)  Air  1.51(0.23)  16.35(2.45)  13.49(2.02)  56.77(8.52)  4  Oxygen  8.39(3.26)  90.90(35.28)  73.71(28.56)  79.38(23.84)  4  Treatment KLa2O  N  Table 3.1.d: Least squares means ( ± STDEV), for inlet water velocity = 8.5 m.s and gas flow rate/water flow rate Treatment  =  2.0%  KLa2O  SOTR  SAE  SOTE  (hr’)  (g02/hr)  (gO2kWfhr)  (%)  Air  1.65(0.34)  17.88(3.70)  14.70(3.04)  45.16(9.34)  4  Oxygen  10.02(0.76)  108.64(8.28)  87.79(6.69)  64.23(4.89)  4  N  Table 3.1.e: Least squares means ( ± STDEV), for inlet water velocity = 8.5 m.s 1 and gas flow rate/water flow rate  =  2.5%  o 2 Kja  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  kWIhr) 2 (gO  (%)  Air  1.81(0.76)  19.57(8.28)  16.06(6.80)  40.28(17.04)  4  Oxygen  10.93(1.28)  118.44(13.90)  95.48(8.25)  57.06(6.70)  4  Treatment  N  Table 3.1.f: Least squares means ( ± STDEV), for inlet water velocity = 8.5 m.s 1 and gas flow rate/water flow rate  =  3.0%  SOTR  SAE  SOTE  hr 1)  (gO2fbr)  kWfhr) 2 (gO  (%)  Air  2.12(0.31)  22.95(3.38)  18.79(2.77)  38.63(5.70)  4  Oxygen  11.99(1.41)  129.93(15.25)  104.43(12.26)  51.21(6.01)  4  Treatment KLa2o  67  N  Table 3.1.g: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate 3.5% Treatment KLa2O  =  8.5 m.s 1 and  SOTR  SAE  SOTE  (hr-i)  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  2.66(0.29)  28.80(3.09)  23.55(2.53)  42.03(4.52)  4  Oxygen  16.07(3.09)  174.11(33.46)  139.60(26.82)  59.49(11.43)  4  N  Table 3.1.h: Least squares means ( ± STDEV), for inlet water velocity = 8.5 m.s and gas flow rate/water flow rate = 4.0% SOTR  SAE  SOTE  (hr-i)  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  2.79(0.38)  30.27(4.07)  24.70(3.32)  38.21(5.13)  4  Oxygen  16.75(1.81)  181.51(19.59)  145.11(15.66)  53.66(5.79)  4  Treatment KLa2O  N  Table 3.1.i: Least squares means (± STDEV), for inlet water velocity = 8.5 m.s and gas flow rate/water flow rate = 4.5% Treatment KLa2O  SOTR  SAE  SOTE  (hr)  (g0 / 2 hr)  (gO k 2 Wfhr)  (%)  Air  3.52(0.35)  38.16(4.92)  31.09(2.79)  43.38(5.23)  4  Oxygen  20.57(0.83)  222.89(9.02)  177.77(7.20)  59.33(2.40)  4  N  Table 3.2.a: Least squares means ( ± STDEV), for inlet water velocity = 8.0 m.s 1 and gas flow rate/water flow rate = 0.5% Treatment KLa2O  SOTR  SAE  SOTE  (hri)  (g02/hr)  (gO k 2 W/hr)  (%)  Air  0.67(0.05)  7.22(0.49)  5.97(0.40)  53.49(3.63)  4  Oxygen  5.23(0.38)  56.73(4.08)  46.18(3.32)  98.39(7.08)  4  68  N  Table 3.2.b: Least squares means (± STDEV), for inlet water velocity flow = 1.0% gas  =  8.0 m.s’ and  rate/water flow rate  Kja2o  SOTR  SAE  SOTE  (hr’)  (g02/hr)  kWfhr) 2 (gO  (%)  Air  1.12(0.04)  12.13(0.38)  10.00(0.31)  44.94(1.41)  4  Oxygen  8.08(0.91)  87.58(9.87)  71.02(8.01)  75.94(8.56)  4  Treatment  N  Table 3.2.c: Least squares means ( ± STDEV), for inlet water velocity = 8.0 m.s 1 and gas flow rate/water flow rate = 1.5% KLa2O  SOTR  SAE  SOTE  (hr’)  (gO2Ihr)  (gO2kWfhr)  (%)  Air  1.57(0.01)  17.00(0.11)  13.97(0.09)  41.97(0.26)  4  Oxygen  11.62(1.01)  125.89(10.93)  101.71(8.83)  72.78(6.32)  4  Treatment  Table 3.2.d: Least squares means ( ± STDEV), for inlet water velocity gas  flow rate/water flow rate  =  =  N  8.0 m.s’ and  2.0%  KLa2o  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  kWIhr) 2 (gO  (%)  Air  2.05(0.01)  22.25(0.07)  18.24(0.06)  41.21(0.13)  4  Oxygen  13.88(1.13)  150.43(12.24)  121.09(9.85)  65.22(5.31)  4  Treatment  Table 3.2.e: Least squares means ( ± STDEV), for inlet water velocity  flow rate/water flow rate  =  =  N  8.0 m.s 1 and gas  2.5%  KLa2O  SOTR  SAE  SOTE  (hr’)  (g02/hr)  kWfhr) 2 (gO  (%)  Air  2.85(0.21)  30.84(2.25)  25.22(1.84)  45.68(3.33)  4  Oxygen  18.70(1.92)  202.70(20.82)  162.57(16.69)  70.31(7.22)  4  Treatment  69  N  Table 3.2.f: Least squares means (± STDEV), for inlet water velocity flow rate/water flow rate = 3.0%  =  8.0 m.s’ and gas  KLa2O  SOTR  SAE  SOTE  (hr1)  (g02/hr)  (gO k 2 W/hr)  (%)  Air  3.15(0.27)  34.14(2.93)  27.85(2.39)  42.15(3.62)  4  Oxygen  20.46(2.83)  221.77(30.64)  177.21(24.49)  64.10(8.86)  4  Treatment  Table 3.2.g: Least squares means ( ± STDEV), for inlet water velocity  gas flow rate/water flow rate  Treatment KLa2O (hr  1)  =  =  N  8.0 m.s 1 and  3.5%  SOTR  SAE  SOTE  (gO2fhr)  (gO k 2 WIhr)  (%)  N  Air  3.60(0.04)  39.06(0.43)  31.78(0.35)  41.33(0.46)  4  Oxygen  21.70(2.02)  235.17(21.84)  187.23(17.36)  58.26(5.41)  4  Table 3.2.h: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate = 4.0% Treatment KLa2O  8.0 m.s 1 and  SOTR  SAE  SOTE  (hrl)  (gO f 2 hr)  (gO k 2 W/hr)  (%)  Air  3.72(0.21)  40.35(2.32)  32.75(1.89)  37.36(2.15)  4  Oxygen  21.93(1.43)  237.71(15.45)  188.57(12.26)  51.53(3.35)  4  Table 3.21: Least squares means  ( ± STDEV), for inlet water velocity flow rate/water flow rate = 4.5%  =  N  1 and gas 8.0 m.s  Treatment KLa2O (hr-i)  SOTR  SAE  SOTE  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  4.54(0.42)  49.23(4.59)  39.86(3.72)  40.52(3.78)  4  Oxygen  26.26(1.28)  284.58(13.90)  224.94(10.99)  54.84(2.68)  4  70  N  Table 3.2.j: Least squares means (± STDEV), for inlet water velocity = 8.0 m.s 1 and gas flow rate/water flow rate = 5.0% Treatment KLa2O  SOTR  SAE  SOTE  (hr-i)  (gO2Ihr)  (gO k 2 W!hr)  (%)  Air  4.84(0.34)  52.48(3.64)  42.38(2.94)  38.87(2.70)  4  Oxygen  28.37(1.20)  307.43(12.98)  242.12(10.22)  53.32(2.25)  4  Table 3.3.a: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate = 0.5% Treatment  =  N  7.6 m.s and  KLa2o  SOTR  SAE  SOTE  (-i)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  1.62(1.49)  17.55(16.12)  6.80(0.20)  45.76(1.36)  4  Oxygen  5.95(0.36)  64.50(3.92)  52.44(3.19)  83.90(5.10)  4  N  Table 3.3.b: Least squares means ( ± STDEV), for inlet water velocity = 7.6 m.s 1 and gas flow rate/water flow rate = 1.0% Treatment  KLa2O  SOTR  SAE  SOTE  (hr-i)  (g02/hr)  (gO k 2 W/hr)  (%)  Air  1.45(0.08)  15.75(0.92)  12.96(0.75)  43.75(2.54)  4  Oxygen  11.03(0.19)  119.53(2.03)  96.70(1.64)  77.74(1.32)  4  Table 3.3.c: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 1.5%  =  N  1 and gas 7.6 m.s  SOTR  SAE  SOTE  (hr-i)  (gO2Ihr)  (gO k 2 Wfhr)  (%)  Air  2.16(0.13)  23.42(1.41)  19.20(1.15)  43.37(2.61)  4  Oxygen  14.45(0.34)  156.56(3.63)  126.03(2.92)  67.88(1.57)  4  Treatment KLa2O  71  N  Table 3.3.d: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate = 2.0%  SOTE  (g02/hr)  SAE kWfhr) 2 (gO  2.76(0.26)  29.94(2.83)  24.47(2.31)  41.59(3.92)  4  20.21(1.02)  218.98(11.07)  175.41(8.87)  71.21(3.60)  4  KLa2O  SOTR  (hr’) Air Oxygen  Treatment  7.6 m.s 1 and  (%)  Table 3.3.e: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 2.5% Treatment KLa2O  N  =  7.6 m.s’ and gas  SOTR  SAE  SOTE  (hr1)  (gO2Ihr)  kWIhr) 2 (gO  (%)  Air  3.41(0.04)  36.95(0.38)  30.09(0.31)  41.06(0.42)  4  Oxygen  23.10(0.84)  250.41(9.11)  199.61(7.26)  65.14(2.37)  4  Table 3.3.f: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 3.0%  N  7.6 m.s and gas  o 2 Kj,a  SOTR  SAE  SOTE  (hr’)  (gO2Ihr)  kWfhr) 2 (gO  (%)  Air  4.09(0.53)  44.34(5.72)  35.98(4.64)  41.05(5.30)  4  Oxygen  25.26(1.49)  273.81(16.15)  217.21(12.81)  59.36(3.50)  4  Treatment  Table 3.3.g: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate = 3.5%  N  7.6 m.s 1 and  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  kWfhr) 2 (gO  (%)  Air  4.74(0.53)  51.36(3.24)  41.55(2.62)  40.76(2.57)  4  Oxygen  26.54(1.82)  287.59(19.70)  227.04(15.55)  53.44(3.66)  4  Treatment KLa2o  72  N  Table 3.3.h: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate = 4.0%  7.6 m.s and  SOTR  SAE  SOTE  (hr’)  (g02/hr)  kWfhr) 2 (gO  (%)  Air  5.11(0.03)  55.36(0.34)  44.63(0.28)  38.44(0.24)  4  Oxygen  30.38(3.40)  329.23(36.86)  258.67(28.16)  53.53(4.53)  4  Treatment KLa2O  Table 3.3.1: Least squares means (± STDEV), for inlet water velocity flow rate/water flow rate = 4.5%  =  N  7.6 1 m.s and gas  KLa2O  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  kWIhr) 2 (gO  (%)  Air  5.59(0.196)  60.55(2.10)  48.65(1.69)  37.37(1.30)  4  Oxygen  31.81(440)  344.78(47.72)  269.59(37.32)  49.83(6.90)  4  Treatment  Table 3.3.j: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 5.0% Treatment KJa2o  N  7.6 m.s 1 and gas  SOTR  SAE  SOTE  (hr’)  (gO f 2 hr)  kWfhr) 2 (gO  (%)  Air  6.24(0.28)  67.65(2.99)  54.19(2.39)  37.59(1.66)  4  Oxygen  34.94(4.35)  378.63(47.10)  294.66(36.66)  49.25(6.13)  4  Table 3.4.a: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate = 0.5% Treatment KLa2o  N  7.4 m.s’ and  SOTR  SAE  SOTE  (hr’)  (g02/hr)  kWfhr) 2 (gO  (%)  Air  0.70(0.16)  7.62(1.72)  6.28(1.42)  35.26(7.97)  4  Oxygen  7.48(0.04)  81.12(0.40)  65.88(0.32)  87.93(0.43)  4  73  N  Table 3.4.b: Least squares means (± STDEV), for inlet water velocity = 7.4 m.s and gas flow rate/water flow rate = 1.0% KLa2O  SOTR  SAE  SOTE  (hr’)  (g02/hr)  (gO k 2 W/hr)  (%)  Air  1.37(0.29)  14.81(3.09)  12.17(2.54)  34.28(7.16)  4  Oxygen  12.36(4.53)  133.94(49.15)  108.13(39.68)  72.59(26.64)  4  Treatment  Table 3.4.c: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 1.5% Treatment  N  1 and gas 7.4 m.s  KLa2O  SOTR  SAE  SOTE  (hr-i)  (g02/hr)  (gO k 2 W/hr)  (%)  Air  2.67(0.28)  28.89(3.04)  23.64(2.49)  44.58(4.69)  4  Oxygen  15.24(2.59)  165.22(28.12)  132.61(22.57)  59.70(10.16)  4  N  Table 3.4.d: Least squares means ( ± STDEV), for inlet water velocity = 7.4 m.s and gas flow rate/water flow rate = 2.0% Treatment KLa2O  SOTR  SAE  SOTE  (hri)  (gO2fhr)  (gO k 2 WIhr)  (%)  Air  2.73(0.29)  29.64(3.10)  24.15(2.53)  34.31(3.59)  4  Oxygen  20.60(4.04)  223.22(43.79)  178.11(34.94)  60.49(11.87)  4  N  Table 3.4.e: Least squares means ( ± STDEV), for inlet water velocity = 7.4 m.s’ and gas flow rate/water flow rate = 2.5% Treatment  KLa2O  SOTR  SAE  SOTE  (hr-i)  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  3.46(0.24)  37.48(2.56)  30.42(2.08)  34.70(2.37)  4  Oxygen  22.03(3.34)  238.72(36.25)  189.37(28.76)  51.75(7.86)  4  74  N  Table 3.4.f: Least squares means (± STDEV), for inlet water velocity = 7.4 m.s’ and gas flow rate/water flow rate = 3.0% KLa2O  SOTR  SAE  SOTE  (hr’)  (gO2Ihr)  (gO k 2 Wfhr)  (%)  Air  4.00(0.64)  43.40(6.94)  35.09(5.61)  33.49(5.35)  4  Oxygen  26.64(5.04)  288.73(54.65)  227.72(43.10)  52.16(9.87)  4  Treatment  Table 3.4.g: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate Treatment  =  N  7.4 m.s 1 and  3.5%  KLa2O  SOTR  SAE  SOTE  (hr 1)  (gO2fhr)  (gO k 2 W/hr)  (%)  Air  4.90(0.52)  53.13(5.59)  42.78(4.50)  35.14(3.69)  4  Oxygen  27.70(1.67)  300.25(18.10)  235.45(14.19)  46.49(2.80)  4  _________  N  Table 3.4.h: Least squares means ( ± STDEV), for inlet water velocity = 7.4 m.s’ and gas flow rate/water flow rate Treatment  =  4.0%  KLa2o  SOTR  SAE  SOTE  (hr’)  (gO f 2 hr)  (gO k 2 Wfhr)  (%)  Air  5.25(0.48)  56.91(5.20)  45.64(4.17)  32.93(3.01)  4  Oxygen  28.71(9.33)  311.14(10.11)  242.60(78.84)  42.16(13.70)  4  Table 3.4.1: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 4.5% Treatment KLa2o  =  N  7.4 m.s’ and gas  SOTR  SAE  SOTE  (.i)  (gO2fhr)  (gO k 2 W/hr)  (%)  Air  5.73(0.66)  62.11(7.18)  49.62(5.73)  31.95(3.69)  4  Oxygen  31.33(1.80)  339.58(19.56)  263.27(23.09)  22.74(3.25)  4  75  N  Table 3.4.j: Least squares means (± STDEV), for inlet water velocity = 7.4 m.s’ and gas flow rate/water flow rate = 5.0%  SOTR  SAE  SOTE  (hr’)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  5.62(0.45)  60.90(4.85)  48.46(3.86)  28.19(2.25)  4  Oxygen  28.72(2.76)  311.32(29.95)  239.99(23.09)  33.74(3.25)  4  Treatment KLa20  Table 3.5.a: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate = 0.5% Treatment  =  N  6.9 m.s and  KLa2o  SOTR  SAE  SOTE  (hji)  (g0 / 2 hr)  (gO k 2 Wfhr)  (%)  Air  0.95(0.12)  10.27(1.32)  8.47(1.09)  43.24(5.57)  4  Oxygen  7.32(0.88)  79.30(9.54)  64.36(7.75)  78.14(9.41)  4  N  Table 3.5.b: Least squares means ( ± STDEV), for inlet water velocity = 6.9 m.s 1 and gas flow rate/water flow rate = 1.0% SOTR  SAE  SOTE  (hr-i)  (g0 / 2 hr)  (gO k 2 Wfhr)  (%)  Air  1.79(0.17)  19.40(1.86)  15.93(1.53)  40.98(3.93)  4  Oxygen  15.06(1.16)  163.24(12.60)  131.64(10.16)  80.73(6.23)  4  Treatment KJ,a2 0  Table 3.5.c: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 1.5% Treatment KLa2O  =  N  6.9 m.s’ and gas  SOTR  SAE  SOTE  ()  (g0 / 2 hr)  (gO k 2 W/hr)  (%)  Air  2.92(0.24)  31.54(2.57)  25.77(2.10)  44.36(3.61)  4  Oxygen  19.15(0.27)  207.60(2.87)  166.33(2.30)  68.36(0.95)  4  76  N  6.9 m.s —1 and  Table 3.5.d: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate 2.0%  SAE  SOTE  (hr’)  SOTR /hr) 2 (g0  (gO k 2 Wfhr)  (%)  Air  3.90(0.12)  42.25(1.31)  34.38(1.06)  44.62(1.38)  4  Oxygen  24.59(3.58)  266.54(38.84)  212.20(30.92)  65.91(9.61)  4  Treatment KLa2o  Table 3.5.e: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 2.5% Treatment KLa2O  =  N  6.9 m.s 1 and gas  SOTR  SAE  SOTE  (hr’)  (g0 / 2 hr)  (gO k 2 Wfhr)  (%)  Air  5.08(0.66)  55.01(7.15)  44.56(5.80)  46.48(6.04)  4  Oxygen  33.24(4.93)  360.22(53.46)  293.96(42.29)  71.26(10.58)  4  N  Table 3.5.f: Least squares means ( ± STDEV), for inlet water velocity = 6.9 m. .s 1 and gas flow rate/water flow rate = 3.0% Treatment KLa2O  SOTR  SAE  SOTE  (hr’)  (g0 / 2 hr)  (gO k 2 W/hr)  (%)  Air  4.97(0.42)  53.83(4.54)  43.41(3.66)  37.91(3.20)  4  Oxygen  34.42(11.17)  373.07(101.3)  293.27(95.17)  61.51(19.96)  4  N  Table 3.5.g: Least squares means ( ± STDEV), for inlet water velocity = 6.9 m.s 1 and gas flow rate/water flow rate 3.5% Treatment KLa2o  SOTR  SAE  SOTE  ()  (gO f 2 hr)  (gO k 2 W/hr)  (%)  Air  5.36(0.42)  58.11(4.53)  46.66(3.63)  35.08(2.73)  4  Oxygen  43.77(9.66)  474.42(104.42)  370.61(81.80)  67.05(14.80)  4  77  N  Table 3.5.h: Least squares means (± STDEV), for inlet water velocity = 6.9 m.s 1 and gas flow rate/water flow rate = 4.0%  SOTR  SAE  SOTE  (i)  (gO2fhr)  kW/hr) 2 (gO  (%)  Air  6.07(0.24)  65.77(2.65)  52.59(2.12)  34.74(1.40)  4  Oxygen  37.72(4.20)  416.96(45.48)  319.72(34.87)  41.26(4.50)  4  Treatment Kia2o  N  Table 3.5.i: Least squares means ( ± STDEV), for inlet water velocity = 6.9 m.s and gas flow rate/water flow rate = 4.5% Treatment  KLa2o (hr  1)  SOTR  SAE  SOTE  /hr) 2 (g0  kWfhr) 2 (gO  (%)  N  Air  6.60(1.13)  71.51(12.22)  56.94(9.73)  33.58(5.74)  4  Oxygen  36.84(4.49)  399.28(48.64))  308.05(37.53)  43.90(5.35)  4  Table 3.5.j: Least squares means ( ± STDEV), for inlet water velocity = 6.9 m.s 1 and gas flow rate/water flow rate = 5.0% o 2 Kja  SOTR  SAE  SOTE  (hri)  (gO2fhr)  kW/hr) 2 (gO  (%)  Air  6.33(0.70)  68.59(7.54)  54.38(5.97)  28.99(3.19)  4  Oxygen  38.47(4.20)  416.96(45.48)  319.72(34.87)  41.26(4.50)  4  Treatment  N  Table 3.6.a: Least squares means ( ± STDEV), for inlet water velocity = 6.4 m.s 1 and gas flow rate/water flow rate = 0.5% KIa2o  SOTR  SAE  SOTE  (hr’)  (gO2Ihr)  kWfhr) 2 (gO  (%)  Air  0.94(0.02)  10.22(0.26)  8.43(0.21)  40.57(1.03)  4  Oxygen  7.67(0.46)  83.13(4.99)  67.45(4.0)  77.24(4.64)  40  Treatment  78  N  Table 3.6.b: Least squares means (± STDEV), for inlet water velocity = 6.4 m.s’ and gas flow rate/water flow rate = 1 .0%s Treatment KIa2o  SOTR  SAE  SOTE  (hr-i)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  1.61(0.07)  17.41(080)  14.28(0.66)  34.54(1.59)  4  Oxygen  14.31(2.95)  155.04(27.12)  124.93(21.85)  72.02(12.60)  4  Table 3.6.c: Least squares means ( ± STDEV), for inlet water velocity • flow rate/water flow rate = 1.5% Treatment KLa2O  N  6.4 m.s and gas  SOTR  SAE  SOTE  (hri)  (gO2Ihr)  (gO k 2 Wfhr)  (%)  Air  2.82(0.10)  30.58(1.05)  24.97(0.86)  40.45(1.39)  4  Oxygen  18.40(2.95)  199.40(31.95)  159.57(25.57)  61.75(9.90)  4  N  Table 3.6.d: Least squares means ( ± STDEV), for inlet water velocity = 6.4 m.s 1 and gas flow rate/water flow rate = 2.0% Treatment KLa2O  SOTR  SAE  SOTE  (hr’)  (gO f 2 hr)  (gO k 2 Wfhr)  (%)  Air  3.64(0.32)  39.43(3.47)  32.04(2.82)  39.11(3.44)  4  Oxygen  26.24(7.81)  284.42(84.67)  226.06(67.30)  66.06(19.67)  4  N  Table 3.6.e: Least squares means ( ± STDEV), for inlet water velocity = 6.4 m.s 1 and gas flow rate/water flow rate = 2.5% Treatment KIa2o  SOTR  SAE  SOTE  (hr-i)  (gO2fhr)  (gO k 2 WIhr)  (%)  Air  4.05(0.17)  43.92(1.89)  35.53(1.53)  34.86(1.50)  4  Oxygen  25.73(1.73)  278.86(16.59)  220.15(13.09)  51.82(3.08)  4  79  N  Table 3.6.f: Least squares means (± STDEV), for inlet water velocity flow rate/water flow rate=3.0%  =  6.4 m.s 1 and gas  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  4.60(0.39)  49.86(4.22)  40.15(3.40)  32.98(2.79)  4  Oxygen  34.50(5.65)  373.92(61.24)  293.22(48.02)  57.90(9.48)  4  Treatment KLa2O  N  Table 3.6.g: Least squares means ( ± STDEV), for inlet water velocity = 6.4 m.s 1 and gas flow rate/water flow rate  =  3.5%  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  (gO2kWfhr)  (%)  Air  5.30(0.38)  57.41(4.10)  46.01(3.29)  32.54(2.32)  4  Oxygen  35.28(9.91)  382.39(107.41)  297.86(83.67)  50.75(14.26)  4  Treatment KLa2O  N  Table 3.6.h: Least squares means ( ± STDEV), for inlet water velocity = 6.4 m.s 1 and gas flow rate/water flow rate = 4.0% Treatment KLa2O  SOTR  SAE  SOTE  (hr 1)  (gO2Ihr)  (gO k 2 Wfhr)  (%)  Air  5.51(0.15)  59.68(1.61)  47.62(1.29)  29.61(0.80)  4  Oxygen  30.06(1.92)  325.84(20.81)  252.14(16.10)  37.84(2.42)  4  Table 3.6.i: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate  =  N  6.4 m.s 1 and gas  4.5%  KLa2O  SOTh  SAE  SOTE  (hr’)  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  5.76(0.08)  62.46(0.85)  49.60(0.68)  27.54(0.38)  4  Oxygen  48.73(18.56)  528.10(201.10)  405.97(154.59)  54.52(20.76)  4  Treatment  80  N  Table 3.6.j: Least squares means (± STDEV), for inlet water velocity flow rate/water flow rate = 5.0% Treatment KLa2O  6.4 m.s 1 and gas  SOTR  SAE  SOTE  (hr1)  (gO2fhr)  (gO k 2 Wfhr)  (%)  Air  6.21(0.01)  67.26(0.14)  53.17(0.11)  26.69(0.06)  4  Oxygen  41.56(2.68)  450.48(29.03)  344.04(22.17)  41.85(2.70)  4  N  Table 3.7.a: Least squares means ( ± STDEV), for inlet water velocity = 5.7 m.s 1 and gas flow rate/water flow rate = 0.5%  SOTR  SAE  SOTE  (hr’)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  2.59(2.68)  28.12(22.65)  23.17(2.33)  94.14(10.7)  4  Oxygen  8.88(0.21)  96.16(2.45)  78.05(1.46)  83.46(0.69)  4  Treatment KLa2O  Table 3.7.b: Least squares means ( ± STDEV), for inlet water velocity gas flow rate/water flow rate = 1 .0%s Treatment KLa2O  =  N  5.7 m.s 1 and  SOTR  SAE  SOTE  (hr-i)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  1.89(0.04)  20.50(0.41)  16.81(0.33)  37.97(0.75)  4  Oxygen  16.25(0.35)  176.15(2.14)  141.79(1.57)  76.37(0.84)  4  Table 3.7.c: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate 1.5% Treatment KLa2o  =  N  1 and gas 5.7 m.s  SOTR  SAE  SOTE  (hr-i)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  2.80(0.30)  30.33(3.25)  24.74(2.65)  37.45(4.01)  4  Oxygen  18.48(0.29)  200.24(3. 18)  160.01(2.54)  57.88(0.92)  4  81  N  Table 3.7.d: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate = 2.0%  SOTR  SAE  (hr’)  (gO2Ihr)  Air  3.51(0.10)  Oxygen  25.32(1.23)  Treatment KLa2o  5.7 m.s 1 and  N  (gO k 2 WIhr)  SOTE (%)  38.09(1.13)  30.91(0.92)  35.27(1.05)  4  274.47(13.29)  217.73(10.55)  59.50(2.88)  4  Table 3.7.e: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate = 2.5% Treatment KLa2O  =  5.7 m.s and gas  SOTR  SAE  SOTE  (hr1)  (g02/hr)  (gO k 2 Wfhr)  (%)  Air  4.15(0.39)  44.96(4.19)  36.31(3.38)  33.31(3.10)  4  Oxygen  29.33(3.91)  317.86(42.75)  250.33(33.67)  55.12(7.41)  4  N  Table 3.7.f: Least squares means ( ± STDEV), for inlet water velocity = 5.7 m.s’ and gas flow rate/water flow rate = 3.0% Treatment KLa2o  SOTR  SAE  SOTE  (hr’)  (gOilhr)  (gO k 2 Wfhr)  (%)  Air  4.77(0.56)  51.73(6.07)  41.57(4.88)  31.93(3.75)  4  Oxygen  28.72(5.47)  311.31(59.25)  243.42(46.33)  44.99(8.56)  4  N  Table 3.7.g: Least squares means ( ± STDEV), for inlet water velocity = 5.7 m.s and gas flow rate/water flow rate = 3.5%  SOTR  SAE  SOTE  ()  (g0 / 2 hr)  (gO k 2 Wfhr)  (%)  Air  5.26(0.48)  57.00(5.23)  45.58(4.18)  30.16(2.77)  4  Oxygen  33.60(16.78)  364.19(27.13)  282.75(21.06)  45.11(3.36)  4  Treatment KLa2o  82  N  Table 3.7.h: Least squares means (± STDEV), for inlet water velocity gas flow rate/water flow rate = 4.0%  =  5.7 m.s 1 and  SOTR  SAE  SOTE  (hr’)  (gO2fhr)  kW/hr) 2 (gO  (%)  Air  5.81(0.70)  62.94(7.55)  50.08(6.00)  29.14(3.49)  4  Oxygen  44.04(12.80)  477.26(138.74)  367(106.95)  51.73(15.04)  4  Treatment Kia2o  Table 3.7.1: Least squares means ( ± STDEV), for inlet water velocity flow rate/water flow rate 4.5%  =  N  5.7 m.s 1 and gas  KLa2O  SOTR  SAE  SOTE  (hr-i)  (gO2Ihr)  kW!hr) 2 (gO  (%)  Air  6.08(0.56)  65.95(6.09)  52.22(4.83)  27.14(2.51)  4  Oxygen  47.64(16.78)  516.37(181.88)  395.28(139.23)  49.75(17.52)  4  Treatment  N  Table 3.7.j: Least squares means ( ± STDEV), for inlet water velocity = 5.7 m.s 1 and gas flow rate/water flow rate = 5.0% Treatment Kia2o (i)  SOTR fhr) 2 (gO  SAE kWfhr) 2 (gO  SOTE  N  (%)  Air  6.14(0.58)  66.60(6.30)  52.48(4.96)  24.66(2.33)  4  Oxygen  51.12(9.92)  554.03(107.68)  421.16(81.86)  48.04(9.34)  4  83  Chapter 4: Discussion A brief review of the relationship between a gas and a liquid is required, before discussing the results of these experiments, as it will assist in interpreting the data. 4.1. Gas-Liquid Transfer Theory  There are a number of theories to describe the mass transfer of a soluble gas to water. Earlier works have been reviewed by Sherwood et a!. (1975), Aiba et a!. (1965), Eckenfelder and O’Connor (1961), and Mueller et a!. (2002). The “two-film gas theory” was first purposed by Nernst in 1904. Lewis and Whitman developed it in 1924, and then Ippen et a!. adapted and revised it in 1952, for the calculations of oxygen absorption rates in water (Mavinic, D.S.  -  Course Notes, CIVL 569). This theory is based on a model in  which two films exist at the gas-liquid interface. According to this theory, the gas passes through the gas and liquid films, respectively, by a slow rate of molecular diffusion. The mass transfer between the two phases occurs by a partial pressure gradient in the gas phase and a concentration gradient in the liquid phase, which is usually called the “driving force”. For any gas molecule to transfer through the two films, it has overcome the resistance from both films. For a gas with low solubility in water, such as oxygen, this resistance of the gas film is very little compared to the liquid film. Therefore, it can be assumed that the oxygen concentration in solution at the interface is the saturation concentration and that the entire resistance to the oxygen transfer into the water is due to the liquid film. The penetration theory, developed by Higbie in 1935, assumes a small element of fluid is in contact with the interface for a short time, where diffusion into the element occurs as a transient process. The contact time for bubble aeration is defined as the time for a single  84  bubble to travel with a given bubble velocity through liquid at a distance equal to its diameter. The penetration theory was expanded by Danckwertz (1951), by employing a wide spectrum of times instead of a single contact time. In this theory an element of fluid is exposed to the interface at the saturation concentration. O’Connor and Dobbins (1958) defined the surface renewal rate as a function of fluid turbulence parameters, a characteristic mixing length and vertical velocity fluctuation. Brumley and Jirka (1988) indicate that the above conceptual models are on the right track. The basic equation of gas transfer is derived from the combination of the classic two-film theory and the surface renewal model (Equation 4.1) (Mavinic and Bewtra, 1974).  dmIdtr=KLA(CICL)  (4.1)  where: dmldt KL=  =  rate of mass transfer (mg.hrj;  liquid film coefficient (m.hrj;  A= interfacial or absorbing surface area of air 2 (m ) ; C= saturation value of dissolved oxygen at the interface between liquid and air bubble (mg.L’) CL=  average concentration of dissolved oxygen in the bulk liquid (mg.Lj.  By introducing the volume of liquid, V, the equation of mass transfer can be expressed in concentration units: (4.2)  dC/dt= 1/V dmldt  1 CL) KLA/V (C -  Thus:  85  (4.3)  dC/dt  =  1 CL) KLa (C -  where: dC/dt = rate of oxygen transfer 1 (mg.L . hr’); a  =  A/V= the specific interfacial surface area of the air bubbles through which diffusion  can occur, generated by the particular aeration system per unit volume of water ) 3 / 2 (m m .  Equation (4.3) is the basic equation used to describe oxygen transfer in actual aeration systems (Mueller et a!., 2002). The maximum rate of oxygen transfer is obtained when the oxygen concentration in the liquid is equal to zero. The minimum transfer rate occurs when the concentration of oxygen in the liquid is equal to the saturation value. KLa, the overall oxygen transfer coefficient 1 (hr ) , is the product of the liquid film coefficient, KL, and the specific surface area, a. Measuring KL and a is impossible in most systems except the simplest ones. Thus, in practice, the measurable value of KLa is considered appropriate, when characterizing the performance of aeration systems (Nienow, 1980). 4.1 .1. Practical Application To Gas Transfer From the basic equation of gas transfer (Eq. 4.3) the main parameters affecting gas transfer are: 1)  Oxygen Saturation Concentration, C  2) Oxygen transfer coefficient, KLa 3) Dissolved oxygen concentration in bulk liquid, CL The factors affecting these parameters are discussed below.  86  1- Oxygen Saturation Concentration, C Based on Henry’s law, oxygen saturation concentration is a function of Henry’s constant and oxygen concentration in the gas phase. Henry’s constant is directly proportional to temperature, pressure and dissolved solids concentrations. Oxygen saturation increases as Henry’s constant increases. In open systems, the oxygen partial pressure is a function of liquid temperature, gas temperature, barometric pressure and increased hydrostatic pressure from aerator submergence. 2- Oxygen Transfer Coefficient, KLa Factors effecting both the liquid film coefficient, KL, and the interfacial area, a, are temperature, dissolved organics and the turbulence level in the system. Since temperature affects viscosity and surface tension, as well as KL directly, KLa increases as temperature increases. Dissolved organics (such as surface active agents) reduce the liquid film coefficient but, on the other hand, decrease the surface tension, resulting in an increase in surface area (Wagner and Poepel, 1995). According to Eckenfelder (1970), in low turbulence conditions, the bulk resistance to oxygen transfer is high. In intermediate turbulence conditions, the resistance to oxygen transfer reduces as the level of turbulence level increases, resulting in an increase in transfer rate. At high turbulence levels, oxygen transfer depends on surface renewal. Under these conditions, transfer rate increases due to increased interfacial area. 3-Dissolved Oxygen Concentration in Bulk Liquid, CL According to Mavinic and Bewtra (1974) and Ashley (2002) factors that effect oxygen transfer rate into water in diffused aeration systems are: water temperature, depth of gas release, contact time of the gas bubble in the water, size of the gas bubble, rate of gas  87  flow, type of diffuser, orifice diameter, oxygen concentration gradient, turbulence in and around the gas-liquid interface, position of the diffuser, geometry of the mixing tank, concentration of dissolved solids, elevation, barometric pressure, presence of surface active agents, and presence of oil contaminants in the air supply Some of these parameters, such as water temperature, concentration of dissolved solids, elevation, barometric pressure, and presence of surface active agents, cannot be controlled; some, such as the presence of contaminants in the air supply and geometry of the mixing tank, can be reduced or eliminated by proper design and maintenance. The three key factors controlling the mass transfer of oxygen in the basic equation (Eq. 4.3) (Ashley, 2002) are: 1) the hydrodynamics of the systems which influence KL; 2) the area of contact between the gas and liquid , a; 3) the concentration gradient between the gas and liquid phase (Cl-CL) (Bewtra and Mavinic, 1978, Nienow, 1980) Two specific parameters were chosen as key variables for these laboratory experiments involving the Speece Cone ; oxygen (gas) flow rate and inlet water velocity. Both of these variables influence all three key factors: KL, a, and (C -CL), by effecting operational 1 factors. The relationship between experimental variables and the operational factors are shown in Table 4.1.  88  Table 4.1 Relationship between KLa (C 1 CL), the operational factors, and the experimental variables (Ashley, 2002) -  Design factors  Effect on KLa (C 1 C ) 1  Operational design factors  Oxygen flow rate  KL, a, (C C1)  -Bubble contact time -Gas flow rate -Oxygen concentration gradient -Turbulence in and around the gas-liquid interface  Inlet water velocity  KL, a, (C CL)  -Bubble contact time -Turbulence in and around the gas-liquid interface  -  -  -  The effect of varying these factors on: O°, KLa2Q (hrj the oxygen transfer coefficient at 2 ,  SOTR (g0 /hr), the Standard Oxygen Transfer Rate, 2 SAE (gO kW/hr), the Standard Aeration Efficiency, and 2  SOTE  (%), the Standard Oxygen Transfer Efficiency,  is discussed below, as it pertains to the Speece Cone data base.  4.2. Gas Flow Rate to Water Flow rate Ratio The ratio of gas flow rate to water flow rate significantly influenced  KLa2O,  SOTR, and  SAE, while it had a less significant effect on SOTE. KLa2O, SOTR, and SAE responded positively to the increase of gas flow rate, and SOTE responded negatively. For all of the experimental groups, KLa2O and SOTR increased almost linearly as a function of the ratio of gas flow rate to water flowrate, up to a maximum in some cases, depending on the inlet velocity. As the ratio increases, the number of bubbles increases;  89  this means an increase in interfacial area available for oxygen transfer. Thus, higher gas flow rates and ratios results in a higher ratio of total interfacial area to total gas volume, a. As gas flow rate increases, shear forces at the gas-water interface also increase, causing higher turbulence. An increase in shear forces increases the renewal rate of liquid film, resulting in an increase in the liquid film coefficient, KL. Due to these reasons, KLa2O and SOTR both increased as the gas flow rate and ratios increased, both for air and oxygen. Numerous studies on gas transfer reported that  KLa2O  and SOTR responded positively to  increased gas flow rate (Bewtra et al, 1970; Speece, 1971; Schmit et a!, 1978; Ashley, 2002). For the oxygen tests, the increase in KLa2O and SOTR gradually decreased at the gas flow rate to water flowrate ratios higher than 2.5%, especially at higher water velocities. At higher oxygen flow rates, the oxygen from the bubbles does not dissolve quickly enough, and bubbles accumulate within the cone (McGinnis and Little, 1998). This gas holdup builds into a “gas pocket” that was observed while running the experiments on high gas flow rates. The “air pocket” resulted in the coalescence of bubbles inside the cone, decreasing the interfacial area between oxygen and water, and consequently decreasing the oxygen transfer rate. In the experiments running on high oxygen flow rates, it was observed that some of the bubbles that were not dissolved were swept out of the cone through the outlet pipe. These bubbles are also the gas bubbles that are not fully discharged of oxygen, meaning that oxygen was “wasted” at the higher ratios. The SAE also increased as the gas flow rate and ratios increased. Larger gas flow rates and higher ratios cause higher turbulence level in the system, resulting in increased rates of oxygen transfer. Thus, despite the additional power/energy required to deliver more air  90  or PSA oxygen to the system, the energy cost was more than offset by the increased oxygen transfer; therefore, SAE responded positively to the increase in gas flow rate, for both air and oxygen. The Standard Oxygen Transfer Efficiency (SOTE) decreased, as the gas flow rate and ratios increased. The effect of water velocities was also not as pronounced. When air or oxygen flow rate increases, the air bubbles become larger (Mavinic, 1973), thus the ratio of interfacial area to bubble volume reduces, resulting in less oxygen transfer, even in a Speece Cone setup. Also, as the size of bubbles increases, their rising velocities increase (Mavinic, 1973, Carver, 1955). McGinnis and Little (1998) developed a model for predicting the performance of a Speece Cone. They performed a preliminary analysis, using “assumed” cone dimensions and operational parameters, to validate their model. As a result, they reported that, as the bubbles size increased, the gas in those bubbles did not dissolve rapidly, and cone performance could be impaired. The results from this study confirm the preliminary analysis of McGinnis and Little, at least as far as SOTE is concerned.  4.3. Inlet water velocity  Increasing the inlet water velocity caused a decrease in oxygen transfer coefficient (KLa2O), and subsequently in SOTE and SAE.  Higher water flow rates increase the  turbulence at the gas-water interface, thus increasing the liquid film coefficient, KL. Also, higher turbulence generates smaller bubble size, meaning a higher ratio of bubble interfacial area to bubble volume, A!V=a. Thus, higher water flow rates are expected to increase the gas transfer. However, an increase in water flow rate also means a decrease  91  in contact time. As a result, any increase in the amount of oxygen transfer due to the greater turbulence is offset by the decrease in hydraulic residence time in the cone, meaning less bubble contact time. On the other hand, an increase in turbulence results in higher surface renewal rates and more oxygen transferred to the liquid, resulting in an increase in oxygen transfer efficiency (SOTE). The inlet water velocities used in these experiments were much higher than the inlet water velocity suggested by Dr. Speece (i.e. 3.0 ms’) (Speece et a!. ,1990). It has been calculated that an inlet choke with an inner diameter of 0.045 m (45 mm) must be used with this current sustem to achieve the design inlet velocity of 3.0 ms 1 (Table 4.2).  4.4. Optimization  Since the results of the oxygen experiments were significantly higher in value than the results of air experiments, only oxygen tests were considered for optimization. As shown in Figure 3.8, the rate of change in  KLa2O  gradually decreases for the gas flowrate to water  flowrate ratios larger than 2.5%, meaning that the oxygen transfer rate decreases at these higher ratios. This indicates that, at oxygen flow to water flowrate ratio higher than about 2.5%, oxygen was not being fully dissolved inside the Speece Cone, and thus the oxygen gas was being wasted. It was observed that oxygen bubbles were leaving the discharge port under higher ratios (i.e .>3 .5%). The slope of the curve of SOTE vs. oxygen flowrate to water flowrate ratio doesn’t show significant changes for the ratios larger than 2.5%,  at  most inlet velocities. In order to select the best combination of operating variables, the mean value of inlet water velocity, 6.9-7.6 ms , might be considered. At this combination, 1  92  the SOTE was about 66-72%, which is much higher than other similar systems reported and studied by Ashley (2002 and 2008).  Table 4.2 Calculated inlet diameter (m) for various inlet velocities 1 (ms ) diameter Outlet diameter Outlet velocity Intlet velocity —  Inlet (m)  (m)  (ms1)  (ms’)  0.016  0.06  0.6  8.5  0.020  0.06  0.9  8.0  0.023  0.06  1.2  7.6  0.026  0.06  1.4  7.4  0.028  0.06  1.6  6.9  0.030  0.06  1.7  6.6  0.033  0.06  1.8  5.7  0.035  0.06  1.8  5.0  0.039  0.06  1.8  4.0  0.042  0.06  1.8  3.5  0.045  0.06  1.8  3.0  0.050  0.06  1.8  2.5  4.5. PSA Oxygen vs. air  All combinations of gas flow rate and water flow rate were run both on PSA Oxygen, with oxygen purity of about 87%, and air with an oxygen purity of 21%. For the  93  experiments running on air, all performance parameters showed a similar behavior to the experiments running on PSA oxygen.  KLa2O,  SOTR, and SAE responded positively to the  increases in air/water flow flow ratios, due to increased interfacial area between gas and water, increased renewal rate of the liquid film and possibly increased driving force. SOTE decreased with the increase in air/water flow ratios, due to a decrease in contact time, amongst several related factors. According to the data summarized in Tables 3.1.a to 3.7.j  ,  values of KLa2O, SOTR, and  SAE in experiments running on air are about 6-7 fold lower, on a mean average basis of all air flow/water flow ratios, than the experiments running on oxygen. SOTE was also about 1.5 fold lower, on the same basis, in the tests running on air. An increase in the percentage of oxygen in the influent gas increases oxygen transfer and absorption efficiency (Speece, 1971). Higher oxygen solubility and higher system transfer efficiencies reduce the energy needed to deliver an equivalent amount of oxygen, using PSA oxygen rather than air. Thus, all of the experimental parameters, KLa2O, SOTR, SAE, and SOTE, were much higher in the tests using PSA oxygen rather than air; this was not a surprising result. In addition to better performance of the Speece Cone, there are several advantages to using oxygen, including: avoidance of hypolimnetic dissolved nitrogen saturation (Fast et al., 1975), low energy use (Speece, 1994), and low commercial oxygen costs (Beutel, 2002). The oxygen transfer efficiency in the treatments using air was relatively low. Practically speaking, however, a Speece Cone would rarely be operated on air, due to the 6-7 folds lower oxygenation performance, as shown in the experiments and noted by Ashley (2002).  94  Chapter 5: Conclusions and Recommendations 5.1. Conclusions Based on the results obtained in this study, the following conclusions can be drawn about the operation of a pilot-scale Speece Cone:  •  The Oxygen Transfer Coefficient, KLa, corrected to 20°C, increased with an increase in the ratio of gas flow rate to water flow rate, for both air and oxygen, over a range of 0.5% to 5.0%;  •  The Standard Oxygen Transfer Rate, SOTR, increased with an increase in the ratio of gas flow rate to water flow rate for both air and oxygen, over a range of 0.5% to 5.0%;  •  The Standard Aeration Efficiency, SAE, increased with an increase in the ratio of gas flow rate to water flow rate, for both air and oxygen. The increase in oxygen transfer was apparently, much larger than the increase in energy associated with delivery air or PSA oxygen to this system.  •  The Standard Oxygen Transfer Efficiency, SOTE, decreased with an increase in the ratio of gas flow rate to water flow rate, for both air and oxygen.  •  The ratio of gas flow rate to water flow rate, within a range of 0.5% to 5% , is extremely important in transfer of oxygen into the water, in a Speece Cone  95  design. At oxygen flow rate to water flow rate ratios higher than 2.5%, oxygen was not fully dissolved into the water, resulting in gas accumulation within the Cone and/or wasting oxygen through bubbles leaving the Cone.  •  An increase in inlet water velocity resulted in a decrease in the overall Oxygen Transfer Coefficient, KLa, corrected to 20°C, the Standard Oxygen Transfer Rate, SOTR, and the Standard Aeration Efficiency, SAE, but an increase in the Standard Oxygen Transfer Efficiency, SOTE.  •  In a Speece Cone system running on air, compared to a system running on oxygen, an increase in the ratio of gas flow rate to water flow rate had a similar, but much less dramatic effect on KLa, SOTR, SAE, and  SOTE.  Overall oxygen transfer efficiencies and performance were 6-7 folds lower.  •  For optimization of oxygen transfer in this investigation, the best combination was achieved at the inlet water velocity of 6.9-7.6 ms 1 and oxygen flow rate to water flow rate ratio of about 2.5%. At this combination, the SOTE was about 66-72%, a value much higher than previously published data on full lift hypolimnetic aerator operating on air.  96  5.2. Recommendations for Future Work  •  Pressure is one of the key factors in oxygen transfer as per Henry’s Law. Increasing pressure in a Speece Cone would enhance transfer efficiency. Studying the oxygen transfer efficiency under higher pressures is recommended. Higher pressure in a Speece Cone can be provided mechanically (i.e. by using a larger water pump) or by using higher hydrostatic pressure, via installing the device in deeper locations.  •  In order to select the most energy efficient design, it is suggested to investigate the use of alternative equipment such as: a main pump with lower energy requirement, different methods of oxygen generation, and a more efficient PSA oxygen generator unit that produces oxygen-rich air with more than 87% purity that was used in these experiments.  •  For higher ratios of gas flow rate to water flow rate, an accumulation of gas was observed in the top part of the cone. This “gas pocket” acts as a barrier to oxygen transfer, by gas bubbles. A special “gas release” system designed into the cone, may enhance oxygen transfer efficiency.  •  To achieve the design inlet velocity of 3.0 ms 1 suggested by Dr Speece (Speece et al., 1990), it is recommended to operate the current system using inlet throat with inner diameter of 0.045 m.  97  References Aiba, S., Humphrey A.E, and Milhis N.F. (1965) Biochemical Engineering, Academic Press, New York.  Ashley K. I. (1985) Hypohimnetic Aeration: Practical Design and Application, Water Research, 19(6): 735-740. Ashley K.I., Hall K.J. (1990) Factors influencing oxygen tyransfer in hypolimnetic aeration systems, Verh. mt. Verein. Limnol., 24: 179-183. Ashley K. I. (2002) Comparative analysis of oxygen transfer in full lift and downflow bubble contact hypolimnetic aerators. PhD thesis, Civil Engineering Department, The University of British Columbia, Vancouver, BC. Ashley K.I., Mavinic D.S., Hall K.J. (2008) Oxygenation performance of a laboratoryscale Speece Cone hypolimnetic aerator: preliminary assessment, Canadian Journal of Civil Engineering, 35(7): 663-675. Bewtra, J.K., Nicholas W.R and. Polkowski L.B.( 1970) Effect of temperature on oxygen transfer in water. Water Research 4(1): 115-122. Burns, V.L., McGinnis D.F., and. Little J.C (2002) Predicting oxygen transfer and water flow rate in airlift aerators, Water Research 36(18): 4605-4615. Beutel, M.W.. and Home A.J.(1999) A review of the effects of hypolimnetic oxugenation on lake and reservoir water quality, Reserv. Manage 15, 285-297. Buetel, M.W. (2001) Oxygen consumption and ammonia accumulation in the hypolimnion of Walker lake, Nevada Beutel, M.W., Brown and Caidwell (2002), AWWA 2002 Annual Conference Brumley, B.H. and Jirka G.H. (1988) Air-Water Transfer of Slightly Soluble Gases: Turbulence, Interfacial Processes and Conceptual Models, Physiochemical Hydrodynamics, 10(3): 295-319. Carver, C.E. Jr., Absorption of oxygen in bubble aeration (1955) Biological treatment of sewage and industrial waste, Vol 1, pp 149-171, Reinhold publishing Co, New York. Cooke, G. D. and Carison R. E.(1989) Reservoir Management for Water Quality and THM Precursor Control, 387 pp. , AWWA Research Foundation, Denver, CO.  98  Cooke, Welch, Peterson, Nihols (2005) Restoration and Management of Lakes and Reservoirs, 3’ edition, 459 PP., CRC press. Colt, J. (1984) Computation of dissolved gas concentrations in water as functions of temperature, salinity andnpressure. American Fisheries Society Special Publication No 14. Bethesda, Maryland, USA.. Danckwerts, P.V. (1951) Significance of liquid film coefficient in gas absorption. Industrial and Engineering Chemistry 43(6): 1460. Doke J.L., Funk W.H., S.T.J. Juul S.T.J. and Moore B.C.(1995) Habitat availability and bethnic invertebrate population changes following alum treatment and hypolimnetic oxygenation in Newman Lake, Washington. J. Fresh Water Ecol. 10: 87-102. Eckenfelder, W.W.,Jr. (1970) Water Quality Engineering for Practicing Engineers, Barnes & Noble, New York. Eckenfelder, W.W., O’Connor D.J.(1961) Biological Waste Treatment, Pergamon Press, Elmsford, NY. Fast, A. W., Doff V. A., and Rosen R.J.(1975) A submerged hypolimnion aerator. Water Resources Research 11: 287-293. Ippen, A.T., Campbell L.G. and Carver C.E.Jr. (1952) The determination of Oxygen Absorption in Aeration Processes. Masschussetes Instittute of Technology, Hydrodynamics Lab. Tech. Rep. No.7.  Kortmann, R.W., Knoecklein G.W., and Bonnell C.H.(1994) Aeration of stratified lakes: Theory and practice, Lake and Reservoir Management 8: 99-120.  Lewis, W.K. and Whitman. W.G. (1924) Principles of gas absorption. md. Eng. Chem. 16: 1215-1220. Little JC. (1995) Hypolimnetic aerators: predicting oxygen transfer and hydrodynamics. Water Res 29:2475-82 Mavinic, D.S. (1973) Optimization of oxygen transfer into water in a coarse-bubble, diffused aeration system. PhD thesis, Civil Engineering Department, The University of Windsor, Windsor, ON.  Mavinic, D.S. and Bewtra J.K.(1974) Mass transfer of oxygen in diffused aeration systems, Can. J. Civil Eng. 1:71-84. McGinnis, D.F., and Little, J.C. (1998) Bubbles dynamics and oxygen transfer in a speece  99  cone. Water Science and Technology, 37(2): 285-292 McGinnis, D.F., and Little J.C.(2002) Predicting diffused-bubble oxygen transfer rate using the discrete- bubble model. Water Res. 36(18)4627-4635. McGinnis, D.F, Lorke, A., Wuest, A., Little J.C. (2004 ) Interaction between a bubble plume and the near-field in a stratified lake, Water Resour. Res. 40 Mercier, P. and Perret J. (1949), Aeration station of Lake Bret, Monastbull, Schweiz, Ver. Gas. Wasser-Fachm, 29, 25. Muller, R., and Stadelmann P. (2004) Fish habitat requirements as the basis for rehabilitation of eutrophic lakes by oxygenation, Fisheries Management and Ecology 11:251-260. Nienow, A. W. (1980) The aeration of waste water: Basic physical factors. Paper No.1. Symp. on the Profitable Aeration of Waste Water (London, 1980) BHRA Fluid Engineering, Cranfield, Bedford, U.K. Point Four Systems (1997) PT4 users manual for UBC Civil Engineering Department. Point Four Sstems, Inc. Richmond, BC. Sherwood, T.K., R.L. Pigford, and C.R. Wilke. Mass Transfer, McGraw-Hill, mc, NY, 1975 Schmit, F.L., Wren J.D. and Redmon D.T. (1978) The effect of tank dimensions and diffuser placement on oxygen transfer. J. Water Poll. Control Fed. 50(7): 1750-1767 Speece, R.E. (1971) Hypolimnion Aeration, J Amer Water Works Ass, 63 (1): 6-9. Speece, R.E., Nirmalakhandan, N., and Tchobanoglous, G (1990) Commercial oxygen use in water-quality management. Water Environment and technology, 2(7): 54-6 1. Speece, R.E. (1994) Lateral thinking solves stratification problems. Wat. Qual. Tnt. 3:1215. Tate, C. H. and Arnold K. F.(1990) Health and aesthetic aspects of water quality, in Water Quality and Treatment, edited by F. W. Pontius, pp. 63-154, McGraw-Hill Inc., New York. Taggart, C.T., and McQueen, D.J. (1982) A model for the design of hypolimnetic aerators. Water Research, 16(6): 949-956 Wagner, M.R. and People H.J.(1995), Influence of surfactants on oxygen transfer, th WEFTEC’95- 68 Annual Conference of the Water Environment Federation, Miami Beach, FL, 297-306.  100  Wu, R.S., Zhou, B.S., Randall, D.J., Woo, N.Y., Lam, P.K.S.(2003) Aquatic hypoxia is an endocrine disruptor and impairs fish reproduction. Environmental Science and Technology 37: 1137-1141.  101  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0063099/manifest

Comment

Related Items