See discussions, stats, and author profiles for this publication at: [Link]

net/publication/288004057

In-sewer Treatment of Domestic Wastewater

Thesis · April 1996

CITATIONS READS

3 768

1 author:

Murtaza Malik

35 PUBLICATIONS   193 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Policy Formulation View project

Rwanda water and sanitation View project

All content following this page was uploaded by Murtaza Malik on 09 April 2016.

The user has requested enhancement of the downloaded file.

IN-SEWER TREATMENT OF

DOMESTIC WASTEWATER

by

MURTAZA MALIK

BSc (Civil Eng) (lions), MSc (Environmental Eng) (Distinction)

Thesis submitted for the degree of

Doctor of Philosophy

April 1996

Division of Environmental Engineering

Department of Civil Engineering
University of Newcastle upon Tyne
NE 17 RU
United Kingdom
 ii

Dedicated
to
my
parents
 ACKNOWLEDGEMENTS

This work would not have been completed without support from many quarters. First
and foremost, I owe a great debt of gratitude to my supervisors Prof. G. K. Anderson and
Mr. D. J. Elliott for their invaluable help and guidance given during the course of this
study. I am also indebted to Prof. M. B. Pescod, Mrs L. M. Evison, Dr. Tom Curtis, Dr.
A. V. Metcalf and Prof. G. P. Noone, for their help at various stages of this study. The
assistance of administrative and technical staff in the Department of Civil Engineering,
including Pat Johnston, Amanda Allan, John Hamilton, Johan Baugh and Lez Hepple, is
gratefully acknowledged.

Dr. Staurt McGulshan of BOC, Mr. Merve Ogston of CIG Australia, Dr. Y. S. Cao of
Delft, The Netherlands, Dr. Kama Raunkjaer of Aalbarg University, Denmark, Dr. Adem
Ozer of Dokuz Eylul University, Turkey, Dr G. Zeeman of Agricultural University
Wagningen, The Netherlands, and Mr Bryan Hancock of Irvine Ranch water District,
California, provided useful literature concerning various aspects of this research, and I
would like to extend my sincere thanks to all of them for their valuable assistance.

I am grateful to Tameez Ahmad, Mohesni Bandpi, Raja Rafhat Iqbal, Amer Islam Sheikh,
M. Shoaib Razzaq, Dr. Khawaja Farooq Ahmad, Md. Mehbub Hasan, Ali Mandavi, Dr.
Muna Hindiye, Zaini Ujang and Dr Noel Craine for their assistance, friendship and
encouragement during my stay at Newcastle.

My grateful thanks are due to the Secretary General Water Authority of Jordan, for
allowing me to use research facilities at the Authority's Central Laboratories in Amman.
Special thanks are due to Dr. Raja Gedeon, Abdual Wahb Matr, Nawal Surma, Suzan-al-
Kilani, Omar Khitab, Dr. Muwafiq Saqqar, and all the staff at Wadi Seer Laboratories in
Amman.

I wish to thank the Ministry of Science of Technology, Government of Pakistan, for their
financial support and the Department of Communication & Works, Government of
NWFP, Pakistan, for granting study leave. Thanks are also due to the Committee of Vice-
ACKNOWLEDGEMENTS iv

Chancellors and Principals of the Universities of the UK for an ORS award, and the
University of Newcastle upon Tyne for the TULIP award and for supporting research
visits in the UK and abroad.

Last but not least, I would like to express my deep appreciation to my mother, sisters, and
brothers Zaffar Iqbal Malik and Mustafa Kamal Malik. This work would not have been
possible without their undiminished love and countless sacrifices.
 ABSTRACT

Urban sewerage systems, which are normally used for the transport of wastewater from its

origin to a wastewater treatment plant (WWTP), could be used as a treatment facility

because (i) they contain heterotrophic bacteria capable of oxidising organic matter both

suspended within the body of the flowing wastewater and attached to the surface of the

wetted perimeter, and (ii) they provide retention times which are often comparable to

those in a conventional activated sludge aeration tank and which, in some cases, may be

equal to the hydraulic retention time in a WWTP. Using sewers as a treatment facility

could be an economical method of alleviating the load on an existing WWTP or reducing

the size of the proposed WWTP.

The current study was undertaken to investigate the feasibility of using urban sewerage

systems as suspended growth biological reactors for the treatment of domestic

wastewater. The flow in a linear gravity sewer was simulated using a batch reactor fed

with raw domestic wastewater. A comparison of simulated aerobic and anaerobic gravity

transport indicated that aerobic treatment would be the most favoured method of in-sewer

biological treatment. The soluble COD (SCOD) removal efficiencies over a retention

period of 8 hours averaged 36 and 6% under aerobic and anaerobic conditions,

respectively, at an average temperature of 22°C. The corresponding total COD removal

averaged 8 and 11%, respectively. When the effluent samples, taken from the batch

reactors after a retention period of 6 hours, were settled in a bench-scale settling column

for one hour, the average suspended solids removal under aerobic conditions was 29%

greater than those under anaerobic conditions.

ABSTRACT vi

Under aerobic conditions, the removal of soluble organic matter during simulated gravity

transport was found to be strongly influenced by the strength of the incoming wastewater.

To investigate the effect of wastewater influent soluble COD (SCOD 0) and influent

suspended solids (SS ()) on in-sewer aerobic treatment, 27 individual wastewaters collected

from the inlets to three wastewater treatment plants were subjected to batch tests at 20°C.

The SCOD over a retention period of 8 hours at 20°C averaged 48, 40 and 61% for

wastewaters having low SCODo and low SS 0, high SCOD0 and low SS 0, and high SCOD0

and high SSo, respectively . The corresponding soluble BOD 5 removal efficiencies

averaged 64, 59, and 81%. A statistical analysis of soluble COD data revealed that, over a

retention period of one to three hours, the soluble COD removal is only significantly

influenced by SSo. At higher retention periods, the soluble COD removal was found to be

significantly affected by both SCOD 0 and SSo. Soluble COD removal was found to follow

first-order kinetics with respect to time. The oxygen uptake rate of the individual

wastewaters varied widely and did not appear to show any clear relation with the SCOD0

or SSo.

An increase in the suspended biomass of the wastewater by the addition of activated

sludge, at a concentration as low as 100 mg VSS/1, at the inlet of the simulated aerobic

gravity sewer resulted in a significant increase in the removal of soluble organic matter.

The soluble COD removal in the seeded wastewater was found to increase almost linearly

with the increase in seed concentration in the range of 100-1000 mg/l. The effect of seed

concentration on soluble COD removal however, appeared to diminish with the increase

in retention time. SCOD removal in the seeded wastewater appeared to follow second-

order kinetics with respect to time. At an initial seed concentration of 100-1000 mg/1, the

batch reactor's effluent after a retention period of 6 hours showed satisfactory settling
ABSTRACT vii

characteristics. The oxygen uptake rate of the seeded wastewater did not show any

specific trend over time at seed concentrations of 100 and 250 mg/1, while at higher seed

concentrations it was similar to that observed in a typical plug flow activated sludge

aeration tank.

The results of the case study in which the wastewater collected from the inlet of the

Greater Amman Siphon (GAS) was maintained aerobic in a batch reactor, showed that by

maintaining aerobic conditions in the GAS, average SCUD and soluble BOD 5 removal

efficiencies of 60 and 78%, respectively, could be achieved over 8 hours at an average

temperature of 25°C. The average oxygen demand of the wastewater was estimated to be

30 mg/l.h.

The result of the current study suggest a strong possibility of using urban sewerage

systems as an aerobic biological reactor for the removal of soluble organic matter during

transit.
 viii

CONTENTS

ACKNOWLEDGMENTS

ABSTRACT

LIST OF FIGURES xvi

LIST OF TABLES xix

ABBREVIATIONS AND NOMENCLATURE xxii

CHAPTER 1. INTRODUCTION 1

CHAPTER 2. LITERATURE REVIEW 6

2.1. Introduction 6

2.2. Research in the field of in-sewer treatment 6

2.2.1. Gravity sewers 6

(i) Self-purification in gravity sewers, and 6

(ii) Laboratory studies on in-sewer treatment in gravity sewers 7

2.2.2. In-sewer treatment in pressure mains 8

2.2.3. Enhanced treatment in sewers 11

2.2.4. Models for in-sewer treatment 14

(i) Gravity sewers, and 15

(ii) Pressure mains 16

2.3. Oxygen requirements of domestic wastewater 16

2.3.1. Oxygen uptake rate of suspended biomass 17

2.3.2. Biofilm (or slime layer) oxygen uptake rate 19


CONTENTS ix

2.3.3. Total oxygen requirements 21

2.4. Solubility of oxygen in sewers 23

2.5. In-sewer aeration 25

2.5.1. Pressure mains 25

(i) Direct injection of compressed air, 25

(ii) U-tube aeration, 26

(iii) Air lifts, and 28

(iv) Venturi aspirators 28

2.5.2. Gravity sewers 29

(i) Surface aeration and ventilation, 29

(ii) Water falls, 30

(iii) Aeration at the points of high turbulence, and 30

(iv) In-sewer aeration devices 30

2.6. In-sewer oxygenation 32

2.6.1. Injection of oxygen in pressure mains 34

(i) Direct injection 34

(ii) U-Tube oxygenation 35

(a) Single U-tube injection, 35

(b) Vitox-II multiple U-tube dissolver, and 36

(iii) Side-stream dissolvers 38

2.6.2. Injection of oxygen in gravity sewers 38

(i) Side-stream venturi dissolver, 38

(ii) Counter-current technique, 40

(iii) U-tube and multiple U-tube dissolvers, and 40


CONTENTS x

(iv) Other techniques 42

2.7. Adverse effects of in-sewer aeration or oxygenation 43

2.7.1. Gas Locking 43

2.7.2. pH change 45

2.8. Cost of in-sewer aeration and oxygenation 45

2.9. Summary 48

CHAPTER 3. AIMS OF RESEARCH 50

CHAPTER 4. EXPERIMENTAL DESIGN, MATERIALS AND METHODS 52

4.1. Introduction 52

4.2. Experimental design 52

4.2.1. Phase-I studies 52

(i) Part-I: Preliminary studies, 52

(ii) Part-II: Addition of oxygen to raw wastewater, and 53

(iii) Part-III: Addition of oxygen to seeded wastewater 53

4.2.2. Phase-II studies 53

4.3. Materials and methods 54

4.3.1. Phase-I studies 54

(i) Feed wastewater, 54

(ii) Seed sludge, 54

(iii) Reactors, 55

(a) Batch reactors employing magnetic stirrer 56

(b) Impeller mixed batch reactor 56

(iv) Reactor operation and feeding, 58


CONTENTS xi

(v) Measurement of dissolved oxygen, 58

(vi) Analytical methods, 58

(vii) Measurements of settleability, 59

(viii) Measurement of oxygen uptake rate of suspended biomass, and 59

(ix) Calculation of removal efficiencies 60

4.3.2. Phase-II studies 62

(i) Description of the Greater Amman Siphon, 62

(ii) The Al-Samra wastewater stabilization pond system, 63

(iii) Feed wastewater, 63

(iv) Seed sludge, 64

(v) Experimental methodology, and 64

(vi) Analytical methods 64

CHAPTER 5. PRELIMINARY STUDIES 65

5.1. Introduction 65

5.2. Aerobic vs. anaerobic conditions in sewers 66

5.2.1. Soluble COD removal 66

5.2.2. Total COD removal 68

5.2.3. Effect of redox conditions on the settleability of solids 68

5.3. Effect of turbulence on in-sewer aerobic treatment 69

5.4. Air vs. oxygen for in-sewer aerobic treatment 70

5.5. Discussion 72

CHAPTER 6. ADDITION OF OXYGEN TO RAW WASTEWATER 76

6.1. Introduction 76
CONTENTS

6.2. Wastewater characteristics 76

6.3. Removal of soluble organic matter 77

6.3.1. Removal of soluble organic matter at 10°C 78

6.3.2. Removal of soluble organic matter at 20°C 80

6.3.3. Removal of soluble organic matter at 30°C 84

6.3.4. Effect of wastewater strength on SCUD removal 86

6.3.5. Temperature Dependency of SCUD removal from the wastewater 89

6.4. Total COD removal 91

6.5. Effect of oxygen addition on SS in the wastewater 92

6.6. Oxygen uptake rate 94

6.7. Discussion 98

6.7.1. SCUD removal 98

6.7.2. Total COD removal 103

6.7.3. SS results 104

6.7.4. Oxygen uptake rate 106

CHAPTER 7. ADDITION OF OXYGEN TO SEEDED WASTEWATER 109

7.1. Introduction 109

7.2. Characteristics of wastewater and seed sludge 109

7.3. Removal of soluble organic matter 112

7.4. Effect of seed concentration on removal of soluble organic matter 118

7.5. Effect of oxygen addition on biomass 123

7.6. Effluent settleability 124

7.7. Oxygen uptake rate 125

CONTENTS

7.8. Discussion 129

7.8.1. Removal of soluble organic matter 129

7.8.2. Biomass growth 135

7.8.3. Effluent settleability results 137

7.8.4. Oxygen uptake rate 139

CHAPTER 8. CASE STUDY-THE GREATER AMMAN SIPHON 142

8.1. Introduction 142

8.2. Characteristics of the wastewater and seed sludge 142

8.3. Removal of soluble organic matter 144

8.3.1. Effect of wastewater strength on SCOD removal 148

8.3.2. Kinetics of SCOD removal 148

8.3.3. Effect of seeding on SCOD removal 150

8.4. Oxygen uptake rate of suspended biomass 153

8.5. Discussion 157

8.5.1. Removal of soluble organic matter 157

8.5.2. Oxygen uptake rate 159

8.5.3. Prospects of using the GAS as a treatment facility 160

CHAPTER 9. GENERAL DISCUSSION 163

9.1. Introduction 163

9.2. Network layout and characteristics 163

9.2.1. Pressure main versus gravity sewers 164

9.2.2. Long, single pressure pipe 165

9.2.3. Loop system 165

CONTENTS xiv

9.2.4. Upstream wastewater treatment plant 166

9.2.5. Upstream wastewater treatment plant on an unconnected sewerage

system 166

9.3. Role of the slime layer in in-sewer treatment 166

9.4. In-sewer treatment for developing countries 167

9.5. Operational problems associated with in-sewer treatment 168

9.6. Economics of in-sewer treatment 168

CHAPTER 10. CONCLUSIONS 170

10.1. Aerobic vs. anaerobic conditions in sewers 170

10.2. Effect of mixing speed on in-sewer aerobic treatment 170

10.3. Effect of dissolved oxygen source on in-sewer aerobic treatment 171

10.4. Addition of oxygen to raw wastewater during simulated gravity transport

at 20°C 171

10.5. Addition of oxygen to raw wastewater added with activated sludge, during

simulated gravity transport at 20°C 172

10.6. Case study-the Greater Amman Siphon 174

CHAPTER 11. RECOMMENDATIONS FOR FURTHER WORK 176

REFERENCES 178

APPENDICES

Appendix A-1. The Greater Amman Siphon 194

Appendix A-2. Experimental set-up for phase-II studies 195

Appendix B-1. Results of ANOVA tests for the effect of mixing speed 196

CONTENTS XV

Appendix B-2. Results of ANOVA tests for the effect of dissolved oxygen source 197

Appendix C-1. SCOD removal efficiencies at various retention times for unseeded

wastewater at 20°C 198

Appendix C-2. Results of regression analyses of the data shown in Appendix C-1 199

Appendix C-3. Rates of reaction 201

Appendix D-1. SCOD removal efficiencies at an added seed concentration range of

0-1000 mg/1 199

Appendix D-2. Results of multiple regression analysis of the data shown in Appendix

D-1 204

Appendix D-3. Results of multiple regression analysis of SOUR data 205

Appendix E-1. SCOD results for Greater Amman Wastewater 207

Appendix E-2. Regression analysis of the data shown in Appendix E-1 208
 xvi

LIST OF FIGURES

Figure 2.1. The Irvine Ranch Water District PPT system. 12

Figure 2.2. Direct injection of compresses air 26

Figure 2.3. U-tube aeration 27

Figure 2.4. Tom Maguire V02 venturi aspirator 29

Figure 2.5. The BOC Primox oxygen injection system 35

Figure 2.6. The CIG multiple U-tube dissolver 37

Figure 2.7. The Vitox sidestream venturi dissolver 39

Figure 2.8. The counter-current techniques 41

Figure 2.9. Biocone system 42

Figure 2.10. The non-porous membrane system 43

Figure 4.1. Pictorial view of magnetic stirred batch reactors 57

Figure 4.2. Experimental set-up for impeller mixed batch reactor 57

Figure 4.3. The SOUR reactor 60

Figure 5.1. Variation in SCOD of domestic wastewater under aerobic conditions 67

Figure 5.2. Variation in SCOD of domestic wastewater under anaerobic conditions 67

Figure 6.1. Variation in SCOD of domestic wastewater at 10°C 78

Figure 6.2. Variation in SCOD of Broomhaugh wastewater at 20°C 80

Figure 6.3. Variation in SCOD of Sacriston wastewater at 20°C 81

Figure 6.4. Variation in SBOD 5 of Broomhaugh and Sacriston wastewater at 20°C 81

Figure 6.5. Variation in SCOD and SBOD 5 of Durham wastewater at 20°C 82

Figure 6.6. Variation in SCOD of domestic wastewater at 20°C 82

Figure 6.7. Variation in SCOD of domestic wastewater at 30°C 85

LIST OF FIGURES xvii

Figure 6.8. The effect of SCODo and SS° on SCOD removal at 20°C 88

Figure 6.9- Comparison of SCOD removal in the wastewater at 10 and 20°C 90

Figure 6.10- Comparison of SCOD removal in the wastewater at 15 and 30°C 90

Figure 6.11. Variation in SOUR of domestic wastewater at 10°C 95

Figure 6.12. Variation in SOUR of the Broomhaugh wastewater at 20°C 95

Figure 6.13. Variation in SOUR of the Sacriston wastewater at 20°C 96

Figure 6.14. Variation in SOUR of Durham wastewater at 20°C 96

Figure 6.15. Variation in SOUR of domestic wastewater at 30°C 97

Figure 7.1. Variation in SCOD of the wastewater at an initial added seed

concentration of 100 mg/1 at 20°C 112

Figure 7.2. Variation in SCOD of the wastewater at an initial added seed

concentration of 250 mg/1 at 20°C 113

Figure 7.3. Variation in SCOD of the wastewater at an initial added seed

concentration of 500 mg/1 at 20°C 113

Figure 7.4. Variation in SCOD of the wastewater at an initial added seed

concentration of 750 mg/I at 20°C 114

Figure 7.5. Variation in SCOD of the wastewater at an initial added

seed concentration of 1000 mg/tat 20°C 114

Figure 7.6. The average variation in SCOD of seed and unseeded wastewater 119

Figure 7.7. Effect of seed concentration on SCOD removal 121

Figure 7.8. Effect of retention time on SCOD removal 122

Figure 7.9. Effect of seed concentration on SW and SS of the supernatant 124

Figure 7.10. SOUR profile of the wastewater at an initial added seed

concentration of 100 mg/1 126

LIST OF FIGURES xviii

Figure 7.11. SOUR profile of the wastewater at an initial added seed

concentration of 250 mg/1 126

Figure 7.12. SOUR profile of the wastewater at an initial added seed

concentration of 500 mg/1 127

Figure 7.13. SOUR profile of the wastewater at an initial added seed

concentration of 750 mg/I 127

Figure 7.14. SOUR profile of the wastewater at an initial added seed

concentration of 1000 mg/1 128

Figure 7.15. Effect of influent SCOD on the effluent SCOD concentration 133

Figure 8.1. Variation in SCOD of GAS wastewater 145

Figure 8.2. Variation in SCOD of GAS wastewater at an initial added seed

concentration of 100 mg/1 145

Figure 8.3. Variation in SCOD of GAS wastewater at an initial added seed

concentration of 250 mg/1 146

Figure 8.4. Variation in SCOD of GAS wastewater at an initial added seed

concentration of 500 mg/1 146

Figure 8.5. Effect of seed concentration on SCOD removal 151

Figure 8.6. Effect of retention time on SCOD removal 151

Figure 8.7. SOUR profile of GAS wastewater 154

Figure 8.8. SOUR profile of GAS wastewater at an initial added seed

concentration of 100 mg/I 154

Figure 8.9. SOUR profile of GAS wastewater at an initial added seed 155
concentration of 250 mg/1 155
Figure 8.10. SOUR profile of GAS wastewater at an initial added seed

concentration of 500 mg/1 155

 xix

LIST OF TABLES

Table 2.1. Solubility (mg/1) of atmospheric and pure oxygen in pure water 23

Table 2.2. CIG "Primox" in-line dissolver design parameters 35

Table 2.3. CIG U-tube dissolver design parameters 36

Table 2.4. Typical costs for direct compressed air injection into a force main

for H2S control 46

Table 2.5. Typical costs for direct compressed oxygen injection into a force main

for H2S control 47

Table 2.6. Costs of various oxygen dissolving techniques 47

Table 4.1. Analytical methods and sensors or meters used 59

Table 4.2. Salient features of AWSP system 63

Table 5.1. Soluble COD removal efficiencies under different redox conditions 66

Table 5.2. Effect of redox conditions on the removal of SS 69

Table 5.3. Effect of mixing speed on SCOD and SS removal under aerobic

conditions 70

Table 5.4. Effect of dissolved oxygen source on SCOD and SS removal 71

Table 6.1. Characteristics of the wastewater used in simulation study 77

Table 6.2. Average SCOD removal efficiencies at 10°C 79

Table 6.3. Wastewater characteristics corresponding to the data

shown in Table 6.2 79

Table 6.4. Average SCOD removal efficiencies at 20°C 83


LIST OF TABLES XX

Table 6.5. Wastewater characteristics corresponding to the data

shown in Table 6.4 83

Table 6.6. Average soluble BOD5 removal efficiencies at 20°C 83

Table 6.7. Wastewater characteristics corresponding to the data shown in Table 6.6 84

Table 6.8. Average SCOD removal efficiencies at 30°C 85

Table 6.9. Wastewater characteristics corresponding to the data shown in Table 6.8 86

Table 6.10. Average TCOD removal efficiencies 92

Table 6.11. Average changes in SS of wastewater 93

Table 6.12. SS removal efficiencies 93.

Table 7.1. Average characteristics of wastewater 110

Table 7.2. Average characteristics of wastewaters for which BOD 5 tests were

carried out 110

Table 7.3. Average characteristics of seed sludge 111

Table 7.4. Average soluble COD removal efficiencies 115

Table 7.5. Average soluble BOD 5 removal efficiencies 115

Table 7.6. Substrate removal efficiencies after a retention period of 8 hours 115

Table 7.7. Values of constants a and b for different seed concentrations 118

Table 7.8. Empirical equations showing the effect of seed concentration, SCODo,

and SS on SCOD removal 120

Table 7.9. Settleabffity test results 125

Table 8.1. Average characteristics of GAS wastewater 143

Table 8.2. Average characteristics of seed sludge 143

Table 8.3. Average SCOD removal efficiencies 147

LIST OF TABLES xxi

Table 8.4. Average soluble BOD 5 removal efficiencies after a retention period

of 8 hours 147

Table 8.5. SOUR of suspended biomass 156

Table 8.6. Average oxygen demand and SOUR values 156

 ABBREVIATIONS AND NOMENCLATURE

°C Degrees Celsius
atm Atmospheres (pressure)
ATU Allylthiourea
B Biomass or seed concentration
b {1+B/100}
BOC British Oxygen Company
BOUR Oxygen uptake rate of the biofilm or slime layer
CFSTR Continuous flow stirred tank reactor
CIG Commonwealth Industrial Gases (Australia)
d Day
DO Dissolved oxygen
h Hour
1SAT In-sewer aerobic treatment
kg Kilograms
km Kilometers
1 Litres
m Meter
mg Milligrams
min Minute
mwg Meters water gauge
n Number of replicate experiments
02 Oxygen
p Minimum level of significance at which an
hypothesis can be rejected
RAS Returned activated sludge
rpm Revolutions per minute
RSCOD SCOD removal in mg/1
SBOD5 Five days soluble biochemical oxygen demand
SCOD Soluble chemical oxygen demand
ABBREVIATIONS AND NOMECLATURE xxiii

SCOD0 Influent SCOD

SD Standard deviation
sec Second
SOUR Oxygen uptake rate of suspended biomass
SS Suspended solids
SS° Influent SS
SSVI Stirred sludge volume index
SW Sludge volume index
t Retention time
VSS Volatile suspended solids
WAJ Water Authority of Jordan
VVWTP Wastewater treatment plant
 1

CHAPTER ONE

INTRODUCTION

The possibility of using the sewerage system to achieve treatment of a wastewater during

its transit would be an attractive option in any wastewater management strategy. In

general, urban sewerage systems are designed purely to transport wastewater from its

origin to wastewater treatment works. If it was possible, however, to treat the wastewater

on route to its destination, then the size and cost of any required treatment works would

be greatly reduced.

The main objectives of any domestic wastewater treatment plant are, in essence, the same.

In the United Kingdom, these objectives are embodied by the EC Urban Wastewater

Treatment Directive. To protect the environment, the function of any wastewater

treatment plant is to reduce both the organic strength of the wastewater and the quantity of

suspended solids in the effluent, to levels prescribed by the Directive. If these prime

objectives and the methods of achieving them are considered, it begins to become clear

that the prevailing conditions in a sewer may be well suited to providing at least partial

treatment of the wastewater.

In general, a typical sewer acts as a plug flow reactor (Hemmings et al., 1983) with

biomass both suspended within the body of the flowing wastewater, and attached on the

wetted perimeter. The retention time in the sewer is dictated by the sewer length and the

flow rate of the wastewater. In the case of a very long sewer, the wastewater transit time

may be in excess of the retention time at any downstream treatment works. Additionally,
CHAPTER 1: INTRODUCTION 2

the natural flow in the sewers provides ideal conditions of mixing for bacterial growth and

flocculation. Without engineering the sewer in any special way, it is clear that some of the

basic requirements of a biological treatment process are satisfied by simple flow in a

sewer. By actively promoting other conditions in the sewer, the potential for in-sewer

biological treatment becomes greatly increased.

Of the various methods of biological treatment, it is likely that aerobic biological

treatment will prove to be the most feasible because of the fact that in-sewer air and

oxygen injection is widely practised for sulphide control, and the potential odour and

corrosion problems associated with anaerobic conditions may rule out treatment by

anaerobic methods.

Several alternative methods of aerobic treatment exist that may be applicable to in-sewer

situations. First, the indigenous biological population, present in the wastewater and in the

naturally occurring wall slimes, will contribute to the natural degradation of organic

matter. Given suitable conditions and sufficient time, a significant level of treatment may

be feasible without introducing additional microorganisms.

If higher rate methods are desired, however, it is possible that either suspended growth or

fixed film processes may be applicable. Obviously, these methods require some

'engineering' of the sewer environment to be feasible. For example, suspended growth

systems require a suitable source of seeding culture and fixed film methods require an

extended surface area.

CHAPTER 1: INTRODUCTION 3

A small treatment works located near the head of the sewer leading to a regional treatment

works could serve as a feasible source of seeding (Greenfield, 1980). The excess activated

sludge instead of being wasted to the digester could be added to the head of the sewer if

an upstream treatment plant is not readily available, a portion of the main flow could be

drawn off and treated. Alternately, it may be feasible to recycle the sludge from a

downstream plant via an aerated pressure main and then use a downhill gravity sewer as

the mixed liquor tank (Pomeroy and Lofy, 1977).

The surface area in a pressure main could be increased by using a number of small

diameter pipes instead of a large pipe or by filling one large pipe with a number of small

pipes (Newcombe et al., 1979). In gravity sewers, the surface area could be increased by
lining the lower half of the sewer with plastic sheet having vertical vanes projecting into

the stream, aligned parallel to the axes of the pipe (Pomeroy and Lofy, 1977).

The suspended growth in-sewer treatment system appears to be more practicable

considering the fact that increasing the surface area would result in increased friction

losses in pressure mains and higher capital costs for both pressure mains and gravity

sewers (Pomeroy and Lofy, 1977; Newcombe eta!., 1979).

The advantages of in-sewer treatment methods are numerous. As stated, the size and cost

of the downstream treatment works may be greatly reduced by the promotion of treatment

within the sewer. This may prove of particular benefit in the UK, for example, where

many new or upgraded plants are required in coastal areas and land use may be restricted,

or especially sensitive, to the development of a treatment plant.

In situations where an existing treatment plant is overloaded, in-sewer treatment may be

especially beneficial. Treatment in the sewer may not only reduce the organic loading on
CHAPTER 1: INTRODUCTION 4

the plant but also increase the treatability of the wastewater by discharging it at the plant

in a truly aerobic state.

Under favourable conditions i.e. longer retention times and higher temperatures

(Hemmings et al., 1983), or by seeding the sewers with active biomass (Stoyer, 1970;

Green et al., 1985), an aerobic sewer followed by a sedimentation tank could prove a

cost-effective alternative to conventional high rate carbonaceous removal facilities.

The problems of odour and corrosion associated with the formation of sulphides in the

sewerage system have been widely reported (Laughlin, 1964; Boon et. al., 1977;

Pomeroy, 1992) and much work has been undertaken solely with the intention of

preventing sulphide build up. In these instances partial treatment of the wastewater has

been seen merely as an additional benefit. In a situation, however, where in-sewer

treatment was engineered to achieve the most practicable degree of treatment, odour and

corrosion control would become significant additional benefits of the scheme.

Some work on in-sewer treatment has been undertaken in the last twenty five years which,

in turn, has caused a growing interest in potential applications but there is little evidence

to indicate that this work has developed into more than a handful of full-scale schemes.

This apparent indifference does not necessarily indicate that in-sewer treatment is not

feasible. It should be noted that sewerage systems only grow through urbanisation and as

a move away from on-site wastewater disposal methods such as septic tanks for

individual houses. At the outset, the sewer is only seen as a method of transporting the
CHAPTER 1: INTRODUCTION 5

wastewater to a single point, but once a sewerage system is in place and growing, the

emphasis is fixed firmly upon achieving efficient treatment at the final point of collection

with the effect that the majority of collective efforts have been focused on the wastewater

treatment plants themselves.

Furthermore, the improved availability of funds for more conventional treatment methods,

and greater process understanding and development efforts to squeeze the existing assets,

were key drivers which reduced the needs to refine in-sewer treatment (Noone, 1995).

It is also interesting to note that work in the field of in-sewer treatment has progressed in

a piece-meal fashion. Even the most recent work has tended not to refer to previous

studies and, as a result, there has been considerable duplication of effort.

The intention of this study is to bring together the up-to-date information in the field of

in-sewer treatment, and to undertake laboratory studies to assess the general feasibility of

using the sewers as a suspended growth aerobic biological reactor to achieve partial or

full treatment of the wastewater during transit. Through the combination of this work, an

assessment of the general feasibility of in-sewer treatment is presented and parameters

influencing the in-sewer treatment have been outlined.

 6

CHAPTER TWO

LITERATURE REVIEW

2.1. INTRODUCTION

The use of sewers for the dual purpose of transportation and treatment of wastewater has

been the subject of several investigations over the past 25 years. In this chapter, an attempt

has been made to bring together up-to-date information in the field of in-sewer treatment.

Following a summary of previous investigations, various aspects of dissolving oxygen in

sewers including the oxygen requirements of the wastewater, factors affecting the dissolution

of oxygen in sewers, and various methods of supplementing dissolved oxygen in pressure

mains and gravity sewers are discussed.

2.2. RESEARCH IN THE FIELD OF IN-SEWER TREATMENT

A brief description of various studies undertaken to investigate the use of the sewer as an

aerobic treatment facility is given in the following sections.

2.2.1. Gravity sewers

(i) Self-purification in gravity sewers

In gravity sewers, especially where the amount of oxygen dissolved by turbulence produced

at junctions and other structures is sufficient to maintain aerobic conditions, a significant

degree of in-sewer purification could occur (Pomeroy and Parkhurst, 1972; Pomeroy and

Lofy, 1977). Pomeroy and Parkhurst (1972) observed that the BOD 5 in a trunk sewer, which

drained a hilly area, was reduced from an average of 192 to 141 mg/1 (26%) during a

retention period of about 4 hours. Thomas et al. (1985) noted a 20% COD removal in a
CHAPTER 2: LITERATURE REVIEW 7

sewer in which dissolved oxygen was enhanced by a high degree of ventilation and the air

dissolved at falls and eddies. Raunkjaer et al. (1995) reported removals of 25 per cent and 14

per cent in dissolved and total COD respectively, under aerobic conditions at 15°C, in a

gravity sewer having a diameter of 500 mm and a retention time of approximately 3 hours.

The dissolved COD of the influent wastewater varied from 200-300 mg/l.

(ii) Laboratory studies on in-sewer treatment in gravity sewers

Laboratory studies, to assess the feasibility of in-sewer aerobic treatment, were originally

undertaken at the University of Newcastle upon Tyne (UNUT) at the request of Anglian

Water plc. These laboratory studies, which were carried out by this author during the initial

stages of the work described in this dissertation, have been reported as UNUT (1994). Batch

tests, undertaken to simulate the aerobic transport of domestic wastewater, showed that the

addition of oxygen to raw wastewater could result in soluble COD and soluble BOD5

removal efficiencies of 30 and 43% respectively, over a retention period of 3 hours. The

corresponding removal efficiencies over a retention period of 6 hours averaged 46 and 78%,

respectively. The influent soluble COD and soluble BOD 5 ranged from 152-206 and 58-74

mg/1, respectively.

Batch studies to investigate the relative contribution of suspended biomass and biofilm

during the biodegradation of substrate in dual-phase systems (sewers and drains) have been

presented by Cao et a/. (1992), and Cao and Alaerts (1995). It was noted that the biofilm

played a significant role in biodegradation when the true wetted area per unit volume of

water was large, and the suspended biomass concentration was below a certain value (12

mg/1). The specific activity of suspended biomass with respect to the oxygen consumption

rate was found to be 50 to 100 per cent of that of the biofilm.

CHAPTER 2: LITERATURE REVIEW 8

Manandhar and Schroder (1995) used an 87.12 m long, 0.20 m deep and 0.08 m wide

galvanised steel channel to investigate the possibility of treating a wastewater by circulating

it to its point of origin. The slime area in the channel varied from 20.1-34.81 m 2, while the

inflow to the sewer was maintained at 0.36 m3/h. The average filtered COD removal in the

system ranged from 49-63% at a temperature of 29-33°C, when surface aeration was the only

source of dissolved oxygen. The corresponding influent COD and SS varied 46-61 mg/1, and

69-86 mg/1, respectively. The authors also developed a design equation for calculating the

length of a loop sewerage system to achieve a given degree of in-sewer treatment.

2.2.2. In - sewer treatment in pressure mains

The idea of using pressure pipes for in-sewer treatment appears to have emerged when the

use of force main aeration for sulphide control resulted in a decrease in the BOD 5 of

wastewater reaching a downstream treatment plant. Pomeroy (1959) noted that air injection

to control sulphide in a 61 cm diameter, 8.8 km long force main, which had an average flow

of 300 m3/h, reduced the average BOD5 of the wastewater from 146 to 64 mg/1 (i.e. 44%).

Laughlin (1964), based on a study of air injection in a 1.6 km long force main which had a

diameter of 76 cm and an average flow rate of 237 m 3/h, reported that in addition to the

control of sulphides, air injection resulted in a 20-50% reduction in BOD 5 of wastewater

during the night when the retention period was long. No significant reduction in BOD5

however, was observed during the day.

Tanaka and Takaneka (1995) studied the removal of organic matter in a force main having a

retention time of 6-7 hours, with air being injected to control sulphides. Reductions of 30-55
CHAPTER 2: LITERATURE REVIEW 9

and 30-75 per cent were noted for total and soluble BOD 5 respectively, at 25°C. The soluble

BOD5 of the wastewater at the inlet of the sewer varied from 70-240 mg/l.

With the advent of pure oxygen, the use of in-sewer oxygenation for sulphide control also

became widespread. Due to higher solubility (approximately 100% oxygen as compared to

20% oxygen in air), oxygen injection enabled higher dissolved oxygen concentrations to be

maintained within the sewer, which in addition to alleviating sulphide problems, resulted in a

significant degree of in-sewer treatment and also improved primary sedimentation at the

downstream treatment works. A typical example of such a case is an 8 km long pressure

main in Bath, UK, where oxygen injection, initially used to control sulphide, led to the use of

the sewer for the reduction of load on a 40% overloaded biological filtration plant. This Bath

study has been described by Pullin (1977), Boon et al. (1977), Newcombe et al. (1979),

Carne el al. (1982) and Toms and Booth (1982).

The Bath pressure main had an average retention time of about 3 hours. The oxygen

requirement was estimated to be 14 mg/l.h and 700 mg/m 2.h for the wastewater and slime

layer respectively. Oxygen was injected directly into the sewer on the discharge side of the

pump using a BOC Primox injector (Figure 2.5). The removal of BOD 5 settled wastewater,

which included the removal of soluble BOD5 and the effect of improved settleability,

averaged 68 kg/h and 26 kg/h at mean wastewater temperatures of 21°C and 11°C

respectively. These removal figures represented reductions of 50 per cent and 30 per cent

respectively in the daily BOD5 load of settled wastewater. Oxygen injection also resulted in

the elimination of sulphide problems and reduced surge pressures. Although oxygen injection

required an increase by approximately 50 per cent in pumping energy due to gas locking at
CHAPTER 2: LITERATURE REVIEW 10

high points in the main, an economic appraisal of the system showed that the use of in-main

oxygen injection with the dual objectives of controlling sulphides and in-sewer treatment

would be economical when compared to the extension of the existing plant by conventional

methods and when using chemicals for sulphide control.

In Australia, research in the field of in-sewer treatment was pioneered by Commonwealth

Industrial Gases (CIG) Ltd. After laboratory- and pilot-scale studies (Greenfield et al., 1980;

Hutchinson, 1981), a full-scale in-sewer treatment facility aimed at achieving complete

carbonaceous treatment of wastewater was commissioned at Boulder Bay (Hemmings et al.,

1983). It consisted of a two-step feed, 6 km long sewer having three pressure mains with

step-flow increases of 35 and 40%. The total retention time in the sewer at dry weather flow

was 17 hours in summer (due to holiday population increases) and 28 hours in winter, with

step flows joining the main flow after retention times of 6.1 and 11.4 hours in summer and

9.1 and 17.5 hours in winter, respectively. The oxygen uptake rate varied from 14 mg/1.h at

24-28°C, and 3-5 mg/I.h at 16-20°C. A primary settling tank was installed at the end of the

sewer, with the effluent being discharged to the ocean. Oxygen was dissolved at three points

along the sewer using direct injection at the first point and CIG multiple U-tube dissolvers

(Figure 2.6) at the others. During a 12-months trail, the Boulder Bay in-sewer treatment

system was found to achieve an average BOD5/NFR (non filterable residue) effluent of 69/65

mg/I in the non-holiday winter period and 76/69 mg/1 in the higher loading summer period.

The influent BOD 5 and NFR ranged from 130-184 mg/I and 241-314 mg/I, respectively.

Hemmings et al. (1983) claimed that the in-sewer treatment could be linked to an ultra high-

rate activated sludge process because it relies on a high growth rate of micro-organisms. In a
 CHAPTER 2: LITERATURE REVIEW 11

conventional activated sludge process, growth rate is low due to limiting substrate

conditions, in spite of high MLSS levels. In an in-sewer aerobic treatment system, high

substrate concentrations would be maintained along the most of the length of the reactor,

resulting in a high cell growth which would compensate for any low cell population.

Presently, the Bath system is not in operation while the Boulder Bay system is used as a

sulphide control facility (Ogston, 1994). The main reasons that the above systems were

abandoned were the improved availability of funds for more conventional treatment methods,

greater process understanding, and development efforts to squeeze the existing assets

(Noone, 1995).

The use of nitrates for control of sulphides has also been reported to result in a significant

degree of in-sewer treatment in pressure mains. Bentzen et al. (1995), who injected nitrates

in a rising main with an average retention time of 5.4 hours, found that controlled dosing of

nitrates, in addition to effective elimination of sulphides, resulted in a soluble BOD 5 removal

of 35 per cent at a temperature of 12-22°C. The soluble BOD 5 of the wastewater at the inlet

of the main averaged 137 mg/1. •

2.2.3. Enhanced treatment in sewers

When compared with the activated sludge process, which contains a high number of active

microoroganisms, the active biomass is low in raw wastewater. The treatment of wastewater

in sewers, under normal conditions may therefore require significantly higher retention

periods (Shaw, 1981). In-sewer treatment however, could be improved by seeding the sewers

with activated sludge to increase its active biomass concentration (Pomeroy and Lofy, 1977;

Shaw, 1981). The above concept has been studied by a number of researchers as described

below.
 CHAPTER 2: LITERATURE REVIEW 12

The earliest investigations on the use of pressure pipe for enhanced treatment of

wastewater were undertaken at the Irvine Ranch Water District (IRWD), California, and

have been documented by Stoyer (1970), Stoyer and Scherfig (1972) and Anon. (1973).

Laboratory- and subsequently large-scale pilot studies were conducted to investigate the

possibility of using a long pressure main employing activated sludge addition at the head of

the main, and oxygen or air injection at several points along the main (Figure 2.1).

FLOTATOR SAND
SCREEN
FILTER
WET WELL
PIPELINE
Raw sewage .-- Liquid .
1"--'
t t .t \*"..
Water
effluent
(for reuse)

Oxygen or air
Solids

AEROBIC SLUDGE ,
ACTIVATOR Excess sludge
Recycle activated sludge 4 (to land disposal) . a

Figure 2.1- The Irvine Ranch Water District PPT system (Anon., 1973)

The main objective of the IRWD system referred to as the "pressure pipe treatment (PPT)

system" was to use the pressure main for both treatment and transportation of the

wastewater so that it could be reused upstream. Pilot-scale 'studies on two mains having

lengths of 5.1 km and 4.57 km and diameters of 61 cm and 10 cm, respectively, showed

that an average BOD 5 removal efficiency of 90 per cent or more was possible, within a

retention time of about 6 hours at 27.5°C. The influent BOD 5 averaged 125 mg/l. The

effluent quality was comparable to that of a conventional activated sludge plant. Based on

a theoretical quantitative comparison of completely mixed and plug flow reactors, and the

results of pilot-scale studies, it was argued that a residence time of about 1 hour would be
CHAPTER 2: LITERATURE REVIEW 13

sufficient for pipeline treatment as compared with the conventional completely mixed

process, due to the acceleration of BOD 5 removal as a result of high initial BOD 5 values.

The PPT system, however, did not get much beyond large-scale pilot plant testing (Hancock,

1994). According to Scherfig (1994), bench- and small-scale pilot plant testing showed

encouraging results but significant problems were encountered with solids separation (using

flotation) during large-scale pilot testing. The main reason that the project could not be

pursued further, however, was that the demand for the reclaimed water increased

dramatically in the area between the downstream treatment plant and the upstream water

reservoir. "The increase in demand made the pipeline very useful as both a direct conveyance

line and also as a back-bone line in the reclaimed water distribution system" Scherfig (1994).

Pilot-scale studies undertaken to investigate the use of a plug-flow pressure aeration basin

followed by a conventional secondary clarifier, for the treatment of wastewater have been

described by Moser et al. (1977). The aeration basin was constructed from a reinforced

asbestos concrete tube with a diameter of 1.5 m and a length of 27 m, and was designed for a

pressure of 10 bars. At a loading rate of 3 kg BOD 5/m3 .day and at a sludge concentration of

4-5 kg MLSS/m3 in the basin, the effluent BOD5 was found to be 25 mg/1 or lower. The

BOD5 of the feed was up to 280 mg/l. The corresponding sludge volume index varied from

50 to 60 ml/g. It was concluded that the pressure plug flow aeration basin would give

excellent BOD5 removal and smooth operation.

Laboratory studies to investigate the possibility of using a U-shaped sewerage system in

Greater Tel Aviv, Israel, for the treatment of wastewater have been presented by Green and
 CHAPTER 2: LITERATURE REVIEW 14

Shelef (1980) and Green et al. (1985). A batch reactor was used to simulate the 39 km long

step-fed sewer with a retention time of 10 hours. The study considered the sewer as a step-

fed plug flow reactor with sludge being recycled to the head of the sewer from the

downstream treatment plant and air injected at several points along the sewer. The removal

efficiencies were investigated at two initial VSS concentrations (after the first wastewater

injection) i.e. 4760 mg/1 and 1430 mg/l. The corresponding final VSS concentrations (as a

result of dilution due to four feedings and biomass growth) were found to be 1000 and 500

mg/l. The corresponding influent soluble CODs of the synthetic wastewater (weighted

average of all the steps) averaged 590 and 667 mg/1 respectively. The dissolved COD and

BOD5 removals averaged 79 and 93% respectively at an initial biomass concentration of

4760 mg VSS/1 and at temperatures of 20-23°C. The corresponding COD removal for an

initial biomass concentration of 1430 mg VS S/1 averaged 73%.

The laboratory studies undertaken during the earlier stage of the current study (UNUT, 1994)

also showed that addition of activated sludge to raw wastewater at a concentration of 500 mg

VSS/1 could result in soluble COD and BOD 5 removal efficiencies of 52 and 78%,

respectively over a retention period of 3 hours. The corresponding removal efficiencies over

a retention period of 6 hours averaged 64 and 93%, respectively.

2.2.4. Models for in-sewer treatment

A number of models have been developed to describe various processes occurring in sewers

e.g. degradation of organic matter in suspension and biofilm, hydrolysis of organic matter

and biomass growth. A brief description of these models follows.

CHAPTER 2: LITERATURE REVIEW 15

(i) Gravity sewers

A reactive sewer model to simulate the flow routing though the sewer, oxygen transfer

though the air-water interface, and biological reactions occurring under aerobic, anoxic and

anaerobic conditions, has been presented by Gall et al. (1995). The model considers the

sewer as a series of completely mixed tanks while the aerobic biological reactions are

described in a way similar to that in General Activated Sludge Model (Dold, 1990).

The MOUSE TRAP model, which describes surface run off quality, sediment transport,

advection and dispersion, and water quality in sewers has been discussed by Garsdal et al.

(1995). The main processes incorporated in the water quality part of the above model include

the degradation of organic matter, bacterial fate, surface aeration, and oxygen demand from

eroded sewer sediments.

Ozer and Kasigra (1995) used a heterogen system, which takes into account the substrate flux

rate though the biofilm surface as well as diffusion and biochemical reactions in

concomitance, to develop a model to predict the substrate removal in gravity sewers. The

model did not consider the liquid phase and was based on the assumption that substrate

transfer in sewer is characterised by molecular diffusion and substrate consumption which

converges to the Monod Equation. Based on the results of the model, Ozer and Kasigra

(1995) showed that a 23 km long sewer of 100 cm diameter would be required for a 70 per

cent COD reduction of a weak wastewater (having soluble COD=170 mg/1) with an assumed

velocity of 0.5 m/s.

CHAPTER 2: LITERATURE REVIEW 16

(ii) Pressure mains

Koch and Zandi (1973) presented a theoretical model to describe the reactions occurring in

pressure pipelines to be used for aerobic biological treatment. Considering the pipeline

reactor as a series of completely mixed reactors, equations were developed to describe cell

growth, substrate utilisation, dissolved and gaseous oxygen, and pressure losses along the

reactor. Using the model, Koch and Zandi (1973) found that a 43 km length (retention time

5.8 hours) of pressure pipeline with either three air injections, or a single oxygen injection,

would be required to achieve a 30 per cent BOD 5 reduction using a seed concentration of 10

mg/l. A sensitivity analysis of the model showed that increasing the seed concentration by a

multiple of 5 and 10 would result in a decrease in the length of the pipeline required to

achieve a 30 per cent reduction by factors of one-third and one-fifth, respectively. The

oxygen requirements, however, would remain the same but the spacing of the aerators would

be decreased.

In addition to the models described above, models dealing with various aspects of in-sewer

treatment i.e. oxygen consumption in dual-phase systems (Cao, 1994), and oxygen transfer in

sewers (Jensen, 1995) have also been developed.

2.3. OXYGEN REQUIREMENTS OF DOMESTIC WASTEWATER

The critical dissolved oxygen (DO) concentration required for an aerobic process is generally

thought to be about 0.5 mg/1 although a higher value (1 mg/1) is generally considered more

acceptable. A DO level of greater than 1 mg/1 has been recommended to be maintained in

sewers to prevent the formation of sulphides (Boon and Lister, 1975; Hollerbach, 1985;

USEPA, 1985). Although the oxidation process has been found to be independent of the DO
CHAPTER 2: LITERATURE REVIEW 17

concentration above the critical value (Rickard and Gaudy, 1968; Kalinski, 1971; Pomeroy

and Parkhurst, 1972), a concentration of 2 mg/1 is usually used as a design value for aerobic

biological reactors (Reynolds, 1982). The amount of oxygen to be added to flowing

wastewater to maintain aerobic conditions is governed by the oxygen utilisation rates of

suspended biomass, and the biofilm or slime layer on the wetted perimeter of the sewer.

2.3.1. Oxygen uptake rate of suspended biomass

Under aerobic conditions, the oxygen utilisation rate of suspended biomass in the wastewater

(SOUR) depends upon the age of wastewater, amount of biomass, concentration of soluble

organic matter, temperature of wastewater, concentrations of substances inhibitory to

biochemical oxidation and pH (Boon and Lister, 1975; Boon et al., 1979; Newcombe et al.

1979; Simpson, 1979; Vivona, 1979; Ozer and Kasigra, 1995).

Pomeroy and Parkhurst (1972) undertook extensive field and laboratory studies to investigate

the oxygen uptake rate of wastewater in the Los Angeles County Sanitation District sewerage

systems. It was found that the SOUR varied widely and did not bear any correlation with the

BOD5 of the wastewater. Under aerobic conditions, domestic wastewater was found to have a

low SOUR (2 to 3 mg/1.h) near its point of origin which increased for several hours reaching

as high as 20 mg/1.h, after which it declined. The duration at which peak SOUR occurred

varied greatly, being higher (6 to 10 hours) for a fresh domestic wastewater and much less (2

hours) for samples collected after being transported for about 10 hours under low DO

conditions. Several wastewaters also exhibited a pronounced secondary SOUR peak

following the main peak. The reasons for this secondary peak could not be specified. The

SOUR tended to be lower for small sewers than in large ones. Tributary sewers were found
CHAPTER 2: LITERATURE REVIEW 18

to bring in wastewater that had a lower SOUR than the main line flow, producing a saw-

tooth profile of SOUR.

Boon et al. (1977) also reported SOUR profiles of fresh domestic wastewater, similar to that

observed by Pomeroy and Parkhurst (1972). Marwood (1984) as quoted by Dack and

Nadebaum (1989) described the SOUR of an oxygenated, previously anaerobic wastewater

as a three phase process i.e. the first phase in which rapid oxidation of sulphide, and other

minor chemical oxidation and biological uptake occurs, the second phase characterised by a

low oxygen uptake which is then followed by the third phase in which SOUR increases

rapidly due to acclimatisation of facultative aerobic bacteria. Matos and de Sousa (1991)

suggest that an increase in SOUR of wastewater with age may be due to sloughing of

biologically active slimes, and gradual transformation of complex organics into easily

biodegradable compounds. A decrease in SOUR after the peak value has been attributed to

the depletion of readily biodegradable organic matter (Loldcegaard et al., 1995).

Ozer and Kasigra (1995), on the basis of their laboratory studies, reported that the SOUR is

strongly affected by substrate concentration and suspended biomass. They developed the

following empirical equation showing correlation between SOUR and soluble organic

matter.

RES = 0.077 +0.00064 S (2.1)

Where; RES (mg/[Link]) and S (mg/I) represent the respiration rate of the wastewater and the

biodegradable part of the filtered COD (mg/1) respectively.

CHAPTER 2: LITERATURE REVIEW 19

2.3.2. Biofilm (or slime layer) oxygen uptake rate

The oxygen uptake rate of the biofilm on the wetted perimeter of the sewer (BOUR) depends

upon the amount of readily biodegradable organic matter in the wastewater, dissolved

oxygen (DO) concentration of the wastewater, temperature, velocity, and specific surface

area of the sewer. Each of these parameters is discussed below.

Thistlethwayte (1972) states that the slime activity would be higher for a strong wastewater

due to greater availability of nutrients. In contrast to this, Boon (1995) reports BOUR to be

independent of BOD5 of the wastewater (within a range of 90-780 mg/1). BOUR increases

with temperature due to increased slime activity, and reduced kinematic viscosity of the

wastewater which in turn decreases the thickness of the boundary layer thus increasing the

diffusion of nutrients to the biofilm (Thistlethwayte, 1972).

Pomeroy and Purkhurst (1973), on the basis of their field studies, report that in sewers in

which slime is an effective oxygen sink, the BOUR varies directly in proportion to the

oxygen concentration (Equation 2.2). It is however, indicated that the BOUR would not be

proportional to DO in well aerated streams of low oxygen demand.

ill'U
BOUR =5.302 r (2.2)

Where: BOUR = biofilm oxygen uptake rate (mg/l.h),

02 = dissolved oxygen concentration (mg/1),

S = slope of energy line (m/m),

 CHAPTER 2: LITERATURE REVIEW 20

U = velocity (m/sec), and

r = is the hydraulic radius of the stream (m).

On the other hand however, Boon et al., (1977) argue that under aerobic conditions, the

BOUR remains almost constant in sewers, probably due to more readily available DO and

substrate level. Matos and de Sousa (1990) also concluded that BOUR did not show any

evident correlation with the DO concentration in small sewers with relatively low flows and

relatively high DO concentrations (above 3 mg/1). Nielsen et al. (1992) reported that the

BOUR of the laboratory biofilms grown in biofilm reactors designed to simulate the sewer

slimes, was almost independent of DO concentration above 0.5 mg 02/1.

The BOUR increases with an increase in the velocity of wastewater (Pomeroy and Parkhurst,

1972; Matos and de Sousa, 1991; Nielsen et al., 1992), probably due to increased transfer of

organic matter by molecular diffusion into the slime layer, caused by increased turbulence at

higher velocities. Pomeroy and Parkhurst (1972) conclude that slimes are not likely to be

absent in sewers unless the velocity exceeds 2.2 m/sec or abrasive material is present.

Generally for pressure mains, slime activity is assumed uniform over the whole surface area

of the pipe line. In gravity sewers however, the active slime area varies with diurnal

variations in wastewater level due to changes in flow. In this case, an estimate of active slime

area could be made based on the minimum, average and maximum value of the flow rate

(Newcombe et al., 1979). Considering the variations in wall shear, diameter and alignments,

pipe jointing details, and wastewater level, Thistlethwayte (1972) adopted an active slime

growth of 15% of the wetted surface area as a design criterion for sulphide control. He also
 CHAPTER 2: LITERATURE REVIEW 21

presented design charts for assessing approximate values of slime area under various

influencing factors.

2.3.3. Total oxygen requirements

The total oxygen uptake rate of wastewater in a particular sewer would be the sum of the

SOUR and the BOUR. The daily total oxygen requirement would be determined by the

wastewater flow, while the maximum DO concentration to be achieved at a given point along

the sewer would depend upon the retention time in the sewer at the minimum flow.

To obtain a reliable estimate of oxygen input for design purposes, it has been recommended

that wastewater should be tested several times for diurnal as well as seasonal sulphide

concentrations, oxygen uptake rate, and rate of sulphide oxidation (US EPA, 1974;

Newcombe et al., 1979; Wernicke, 1987). Samples must be tested immediately after

collection to have a better simulation of real conditions in the sewer. A detailed procedure for

estimating the total oxygen requirements for sulphide control has been presented by USEPA

(1985).

Boon and Lister (1975) have proposed the following equation for calculating the DO

concentration to be introduced at the bottom of a rising main, to maintain aerobic conditions.

2.8 70 2 / 5
Co = (SOUR +—)-1.07 T-1 (2.3)
d 4a
 CHAPTER 2: LITERATURE REVIEW 22

Where: C o dissolved oxygen concentration to be introduced at a point (mg/1)

SOUR = oxygen uptake rate of suspended biomass (mg/l.h)

diameter of the rising main (m)

1 length of the rising main

Qh = average flow rate of the wastewater (m3/h)

temperature of the wastewater (°C) within the range of 10-20°C.

Equation 2.3, which has been developed from empirical data (Kite and Garrett, 1983),

assumes a BOUR of 700 mg/m 2 of the wetted area of the sewer per hour, at 15°C.

According to Boon (1995), the average SOUR of a typical domestic wastewater (BOD 5 about

400 mg/1) would be about 14 mg/1.h at 15°C. The corresponding value for a wastewater

containing a significant portion of industrial discharges or having a significantly lower BOD5

then domestic wastewater, would be 6 mg/l.h. In contrast, Speece et al. (1990), have reported

SOUR and BOUR values of 10 and 4 mg/1.h, respectively at 20°C.

From the oxygen uptake rates given above, it appears that the total oxygen requirement for

sewers with a long retention time, especially at higher temperature, may be very high. In this

case, it may not be economically feasible to dissolve such a high DO concentration at a

single point and hence injection facilities would have to be located at many points along the

sewer.
 CHAPTER 2: LITERATURE REVIEW 23

2.4. SOLUBILITY OF OXYGEN IN SEWERS

The maximum dissolved oxygen concentration that can be achieved in the flowing

wastewater downstream from an injection point depends upon the pressure and temperature

of wastewater, concentration of the dissolved impurities in the wastewater and the total

uptake rate of oxygen along the sewer (Boon et al., 1977).

Table 2.1- Solubility (mg/I) of atmospheric and pure oxygen in pure water

Temperature 1 atm 2 atm 5 atm 10 atm

a
°C AO POb AO PO AO PO AO PO

10 11.3 54 22.4 107 56 268 112.4 537

20 9.2 43 18.6 87 46.6 217 92.9 434

30 7.6 36 15.2 72 37.8 179 75.8 359

Notes: a AO = atmospheric oxygen (Sawyer et al., 1994)

b
PO = pure oxygen (Pomeroy and Lofy, 1977)

At any given temperature, the solubility of oxygen in wastewater increases with an increase

in atmospheric pressure. Therefore, more efficient aeration or oxygenation can be achieved

in pressure mains than in gravity sewers where the flow is under atmospheric pressure. The

solubility of oxygen decreases with an increase in temperature. Table 2.1 shows the

solubility of pure and atmospheric oxygen in pure water at different temperatures and

pressures.
 CHAPTER 2: LITERATURE REVIEW 24

The solubility of oxygen is not significantly affected by the dissolved impurities in

wastewater except nitrogen (Pomeroy and Lofy, 1977). Raising the dissolved oxygen content

of wastewater by air injection increases the dissolved nitrogen in wastewater which interferes

with the dissolution of further oxygen thus reducing the efficiency of dissolution. Although

the use of pure oxygen helps alleviate this problem, nitrogen interference will reduce the

transfer efficiency at higher DO concentrations, especially under ambient pressures.

Wastewater contains dissolved nitrogen in an amount equal to its air saturation concentration

(approximately 15 mg/1 at 1 atm and 20°C). At normal wastewater temperatures, the amount

of pure oxygen that would completely dissolve at equilibrium with atmospheric pressure is

the same as that for water in equilibrium with air (9.2 mg/1), because the gas phase must

consist of 79% nitrogen. If attempts are made to achieve higher DO concentrations by

applying pure oxygen, some oxygen will remain undissolved, and some nitrogen will strip

out of solution.

The stripping of nitrogen causes two problems (Speece et al., 1990); (i) it decreases the

oxygen transfer rate because of the dilution of oxygen concentration in the gas phase by the

stripped nitrogen, and (ii) it reduces the oxygen-absorption efficiency potential because the

nitrogen diluted oxygen must be wasted if higher DO concentrations are to be achieved.

Raising the DO to 30 mg/1 at 25 °C and 1 atmospheric pressure by single injection, for

instance, will result in 29% wastage of oxygen, although the solubility of oxygen is about 40

mg/1 (Pomeroy and Lofy, 1977). This wastage could be minimised if oxygen is supplied in

more than one step or by removing the dissolved nitrogen from wastewater by nitrogen

stripping (Charnock, 1979; Shaw, 1980).

 25

The nitrogen stripping potential of water decreases with an increase in pressure (Speece et

al., 1990). At a pressure of 4 atmospheres, for example, the gas phase must consist of 20%

nitrogen, as compared to 79% at ambient pressure. The saturation oxygen concentration at

4 atmospheres will be 141 mg/1 DO (80% oxygen composition), while the corresponding

value for ambient pressure is 9.2 mg/1 (20% oxygen composition).

The oxygen transfer efficiency in pressure mains, in addition to the factors discussed

above, also depends upon the bubble diameter, length and diameter of the pipe and

velocity in the main (Cadee, 1984; Sewards and Carpenter, 1989).

2.5. IN-SEWER AERATION

2.5.1. Pressure mains

Air has been used as an effective method for controlling sulphides in pressure mains since

1942 (Pomeroy, 1959). Noticeable BOD 5 reductions have been reported when air was

injected for sulphide control (Section 2.2.2). A brief description of various in-sewer

aeration methods used for pressure mains is given in the following sections.

(i) Direct injection of compressed air

The direct injection of compressed air is the most widely used in-sewer aeration method

for controlling sulphides in pressure mains (Laughlin, 1964; Vivona, 1979, 1980;

Rudoloph, 1981; Vivona and Whalen, 1982; Kameda and Odauchi, 1984; Bacquet et al.,

1992; Tanaka and Takenake, 1995). This method (Figure 2.2) involves injecting

compressed air directly in the main at a low point. In a continuously upward sloping pipe,
 CHAPTER 2: LITERATURE REVIEW 26

turbulence produced by the bubbles passing up the pipe results in efficient dissolution of

oxygen.

Force Main

Controls

I Pressure
I CheCk
uge Control
Flexible I G a I Valve
Valve
Coupling!
Pressure I
Air Relief
Compress° Valve

Wet
Well

Figure 2.2-Direct injection of compressed air (USEPA, 1985).

Air injection is not always feasible especially for a nearly level main or where the sewer has

an irregular profile (US EPA, 1974; Pomeroy, 1992). In a flat main the air moves easily

and turbulence produced is not enough to achieve sufficient oxygen dissolution. In a

pressure main with irregular profile, the injection of air results in excessive gas locking

problems (Section 2.7.1).

(ii) U-tube aeration

U-tube aeration (Figure 2.3) has been used as an efficient method for controlling sulphides

in pressure mains (Mitchell, 1973; USEPA, 1974). As shown in Figure 2.3, air is

introduced in fluid in the descending leg of the U-tube either through a venturi aspirator

placed in a pipeline leading to the U-tube, or by injecting compressed air through a diffuser

collar. Injection of air under pressure is necessary for situations where the force main is

nearly level and the discharge pressure is low (USEPA, 1974). The U-tube produces

efficient dissolution of oxygen due to the relatively long contact time of the gas with water
CHAPTER 2: LITERATURE REVIEW 27

and increased hydrostatic pressure caused at the bottom of the U-tube due to the water

column.

Various operating parameters of the U-tube include the air/water ratio, DO concentration

in the inlet, depth of U-tube, water velocity through U-tube, and disperser depth, and have

been discussed at length by Speece et al. (1969), Speece and Orosco (1970), Mitchell,

(1973), Speece eta!. (1981) and USEPA (1985). Although the use of U-tube aeration has

been successful for controlling sulphides in pressure mains, it is thought unlikely that U-

tubes could be used for the incremental increase in DO that would be required by

biological treatment (Pomeroy and Lofy, 1977).

Aeration Device:
Aspirator
Compressed Air
Gaseous Oxygen

Discharge to
Gravity Sewer

Upleg

Expanded Section
Downleg

Pump
Discharge

Figure 2.3-U-tube aeration (USEPA, 1985).

 28

(iii) Air lifts

An air lift is similar to a U-tube except the air is injected into the rising leg. It has been

indicated that air lifts, if used for pumping of wastewater where there is relatively low lift,

would add to the dissolved oxygen supply due to the energy dissipated by the rising air

through the wastewater (Pomeroy and Lofy, 1977). Although, the pumping efficiency of air

lift is low, it would appear to be improved if air lift is looked upon as an air pumping and

oxygen dissolving device. In relatively flat sewers, a series of low-lift stations using air lifts

would achieve DO concentrations sufficient to control sulphides and accomplish significant

in-sewer purification. Also such a system may be economical compared with conventional

pumping stations (Pomeroy and Lofy, 1977).

(iv) Venturi aspirators

A venturi aspirator (Figure 2.4) consists of an inlet convergent nozzle, a throat, and a

divergent barrel. Liquid flowing through a high velocity nozzle creates suction at the

discharge side, thus allowing air to be drawn in from the air entraining pipe. In the throat,

where velocity is maximum and pressure is minimum, the entrained air is combined with the

liquid and is compressed to the discharge pressure by the momentum derived from the

pumped jet. The intimate mixture of air and liquid is ejected via the divergent barrel into the

bulk of the liquid where high turbulence causes intimate mixing of air and water.

The venturi aspirator has been used for sulphide control in force mains and lift stations in the

USA (USEPA, 1985). In the UK, there appears to be no documentary evidence regarding the

use of venturi aspirators for sulphide control in force mains. However a venturi system

marketed by Tom Maguire and Company Ltd. under the name V02, could possibly be used

for this purpose. V02 has been used for the treatment of sludges, and
CHAPTER 2: LITERATURE REVIEW 29

domestic and industrial wastes. An oxygen transfer efficiency of up to 30% has been

claimed (ERC, 1985; Tom Maguire, 1986).

•••///4..
smack..
••••••••
Ptim 5Qtax ''''''''''
(Liquid affluent)

Figure 2.4-Tom Maguire V0 2 venturi aspirator (ERC, 1986).

2.5.2. Gravity sewers

A comprehensive review of various sources of dissolved oxygen and tentative in-sewer

aeration devices for gravity sewers, was presented by Pomeroy and Lofy (1977), and is

summarised as follows:

(i) Surface aeration and ventilation

The main sources of oxygen supply in gravity sewers include normal surface aeration, and

increased surface aeration at points of high turbulence. The amount of oxygen dissolved in

the water stream through surface aeration is proportional to (su)mb, where s is the slope

of the energy line of the stream (m/m), u is the velocity (m/sec) and, b is the surface width

of the stream (m). In very small sewers with swift flow, the amount of oxygen dissolved

may be sufficient to maintain aerobic conditions. However, for a sewer more than a few

centimeters deep in warm climates or more than 1 m deep where wastewater temperatures

are low, oxygen supply from surface aeration will be insignificant. Considering the fact

that, in large systems the wastewater is collected into larger trunk sewers with flatter

slopes. and deeper flows, the 'authors concluded that normal surface aeration will not
 CHAPTER 2: LITERATURE REVIEW 30

provide sufficient oxygen for in-sewer treatment. It is also concluded that ventilation is not a

realistic method of supplementing DO for biological treatment.

(ii) Water falls

Laboratory experiments designed to study aeration in gravity sewers from a waterfall,

revealed that greater oxygen absorption could be achieved by dissipating a part of the

available difference between two points along a proposed sewer in a fall or a series of falls,

as compared with providing steep slopes and higher velocities. It is also illustrated that a

series of low falls rather than one high fall would be more effective provided there is

sufficient time for oxygen depletion between the falls.

(iii) Aeration at the points of high turbulence

Pomeroy and Lofy (1977) conclude that junctions, drops, hydraulic jumps or other such

points of turbulence will not provide sufficient oxygen to accomplish a significant degree of

biological treatment in large trunk sewers. In smaller sewers however, the oxygen supply

from these sources may be quite significant, and hence efforts should be made to maximise

this effect during the design of a system.

(iv) In-sewer aeration devices

A number of tentative in-sewer aeration devices, including surface aerators, jet aeration,

venturi aspirators and diffused aeration has been considered by Pomeroy and Lofy (1977).

The use of mechanical aerators is discounted as their installation within a sewer will not only

be difficult, but costly. In addition, as each unit only provides a small increase in dissolved

oxygen, many aerators would be required. The use of jet nozzles for in-sewer aeration would
CHAPTER 2: LITERATURE REVIEW 31

be more applicable and hence deserves further investigation. Venturi aspirators would not be

favoured unless specific advantages are demonstrated in the future.

Pomeroy and Lofy (1977) concluded that the diffused aeration seems to have the highest

prospects of being used for in-sewer aeration. Small or large bubble diffusers may be

appropriate for installation in gravity sewers provided there is sufficient water depth for

reasonable submergence of the diffuser. Small bubble diffusers may transfer up to 5 to 10%

of oxygen available in the air supplied while large bubble diffusers would transfer less. Large

bubble systems, however, are thought to be most appropriate because of their lower operating

and maintenance requirements. The configuration for diffused aeration favoured by Pomeroy

and Lofy (1977) is the use of a perforated pipe as a source of diffused air, placed in each

manhole, with a compressed air supply line running parallel to the sewer. These diffusers

could lie longitudinally in the sewer although a preferred arrangement would be to place the

diffuser immediately downstream a weir built for this purpose, where high velocity and

turbulence will help enhance the oxygen dissolution.

The use of U-tubes, air lift and pressure tank dissolvers for sulphide control in gravity sewers

has also been evaluated in the USA (USEPA, 1985). These systems were installed at the end

of the force mains or gravity sewers which then discharged into other gravity sewers. Data

reported were not sufficient to draw any general conclusions regarding the effectiveness of

these methods for supplementing DO in gravity sewers. No such installation appears to be

presently in operation.
CHAPTER 2: LITERATURE REVIEW 32

2.6. IN-SEWER OXYGENATION

Commercial oxygen for controlling sulphides in both the pressure mains and gravity sewers,

is used world-wide. There are more than 80 full-scale systems using oxygen injection to

prevent septicity in the UK (Boon, 1995). The corresponding number of installations in

Australia is 157 (Ogston, 1994). The principal advantages of in-sewer oxygenation, as

compared with in-sewer aeration, are outlined below:

(i) the solubility of pure oxygen is approximately five times greater than oxygen, which

enables higher concentrations of dissolved oxygen to be achieved rapidly and efficiently.

This not only eliminates the sulphide problems, but also results in a significant degree of

in-sewer treatment and improved primary settling characteristics (Newcombe et al., 1979;

Hemmings et al., 1983; Wernicke, 1987). In contrast with this, air systems are

characterised by low transfer efficiencies, and are unable to economically achieve high

DO concentrations (Dack and Nadebaum, 1989; Sewards and Carpenter, 1989; Speece et

al., 1990).

(ii) pure oxygen produces less inert gases which results in fewer gas locking problems. Air

systems, on the other hand, increase the dissolved nitrogen in wastewater which not only

interferes with further dissolution of oxygen, but also results in increased gas locking

(Boon eta!., 1977),

(iii) the use of pure oxygen for both pressure mains and gravity sewers is well established.

On the other hand, in-sewer aeration has been found to be economically feasible only for

continuously upward sloping pressure mains (Section 2.5.1), and

(iv) oxygen injection systems, especially liquid oxygen, are simple and easy to install and

have low maintenance requirements (Mozell, 1985).

CHAPTER 2: LITERATURE REVIEW 33

The main advantages claimed for pure oxygen over air, from the biological treatment point of

view, include the capability of meeting higher oxygen demands thus enabling high MLSS

concentrations to be maintained, improved settleability (Chapman et al. 1976; Miller, 1978;

Shelef and Green, 1980; Andreakis, 1987; Gould, 1990), less sludge production (Williamson

and Nielsen, 1981; Jeffries, 1983; Gould, 1990), and greater BOD 5 removal rates (Crossley

and Macdonald, 1972; Sidwick and Lewandowski, 1975). The last two advantages however,

are not supported by many comparative studies and it has been recommended that pilot-plant

studies be undertaken and an economic evaluation made before adopting a pure oxygen

system for biological treatment (White, 1987; Gray, 1989).

The various disadvantages of oxygen compared with air systems include the need for

sophisticated equipment for oxygen generation and the delivery of liquid oxygen, the

additional cost of oxygen production, corrosion, skilled supervision (especially where the

oxygen is to be produced at the site) and, an increased fire hazard. A major disadvantage of a

covered, pure oxygen activated sludge process is the decrease in pH of the mixed liquor due

to an accumulation of carbon dioxide in the wastewater, which in turn may cause inhibition

of the nitrification process (Kalinske, 1971; Benefield and Randall, 1980). This effect

however, is less serious for wastes having a high alkalinity (White, 1987).

A brief description of various in-sewer oxygenation devices is given in the following

sections. Most of these techniques are developed and marketed by the International BOC

group, represented by Commonwealth Industrial Gases (CIG), Afrox, Airco Industrial Gases,

and BOC in Australia, South Africa, United States, and the rest of the world respectively,
CHAPTER 2: LITERATURE REVIEW 34

while use of in-sewer oxygenation techniques developed and marketed by L' Air Liquide

(France), and Linde have also been reported.

2.6.1. Injection of oxygen in pressure mains

The various methods used for pure oxygen injection to pressure main are described in the

following sections:

(i) Direct injection

Direct injection is the most commonly used oxygen injection method for pressure mains

(Boon and Lister, 1975; Boon et al., 1977; Newcombe et al., 1979; Hollerbach, 1985;

Mozell, 1985). This method involves direct dosing of a metered quantity of oxygen to a

pressure main at the discharge side of the pump where turbulence, pressure and velocity are

at its maximum thus promoting efficient transfer of oxygen. Generally the oxygen supply is

interlocked with pumps so that oxygen is supplied only when pumps are in operation. The

correct design of the injector and the bubble size are essential for achieving efficient

dissolution of oxygen. A nozzle injector is used at locations where oxygen solution rates are

critical (i.e. pumping velocities greater than 0.6 m/s and pumping pressures greater than 15 m

head). Under more critical conditions (velocities less than 0.6 m/s), it has been recommended

to use diffusers made of a porous material in order to promote the formation of fine bubbles

(Hollerbach, 1985).

A direct injection system developed and marketed by the BOC group under the name Primox

(Figure 2.5) has also been used for in-sewer treatment (Newcombe et al., 1979). The

principal design parameters of a CIG "Primox" in-line nozzle dissolver are given in Table

2.2.
CHAPTER 2: LITERATURE REVIEW 35

Ornen
siorage
Elects/valve

Control tx4
panel 0 0
lor pumps--4-sn-. Flowrneter•

Pressure
gat91

Iniedxrt
WasIewalef
reception tank

Figure 2.5. The BOC Primox oxygen injection system (Boon eta!., 1977).

Table 2.2-CIG "Primox" in-line dissolver design parameters (Ogston, 1994)

Nozzle velocity range (m/sec) 7-9

Minimum nozzle diameter (mm) 100

Maximum head loss (mwg) 4.0

Oxygen transfer efficiency (% Cs) <60

Notes: (i) mwg = meters of water gauge

(ii) Cs = theoretical saturation oxygen concentration at a given temperature and pressure

The main advtages of Primox oxygen injection system include its simplicity to install and

operate, and less supervision and maintenance requirements (Garrett, 1984).

(ii) U-Tube oxygenation

(a) Single U-tube injection

Single U-tube (Section 2.5.1) oxygenation for sulphide control in sewers has been

evaluated in the USA (USEPA, 1985). The U-tube provides an efficient dissolution of

oxygen with low unit energy consumption. A 30 m deep U-tube employing off-gas recycle,
CHAPTER 2: LITERATURE REVIEW 36

could result in more than 80% transfer efficiency at a through put velocity of 2.5 m/s (Speece

et al., 1990).

(b) Vitox II Multiple U-tube dissolver

The multiple U-tube dissolver (Figure 2.6) has been developed and marketed by the BOC

group and used successfully for odour control and in-sewer treatment (Hemmings et al.,

1983; Cadee et al., 1984; Sewards and Barkley, 1992). Its main advantages include the

ability to achieve up to 75% of the saturation oxygen concentration at an oxygen utilisation

efficiency of more than 90%, freedom from gas locking problems, and minimum attention

and maintenance requirements because of the absence of moving parts (Cadee et al., 1984;

Hollerbach, 1985). Table 2.3 lists the main design criteria of the CIG U-tube dissolver.

Table 2.3-CIG U-tube dissolver design parameters (Ogston, 1994)

Minimum velocity: single and three stage (m/sec) 0.45

Maximum velocity: single and three stage (m/sec) 1.2

Oxygen transfer efficiency: (% Cs)

- single stage 75

-three stage 60

The world's largest U-tube dissolver is located at Munster Pumping station, upstream of

Woodman WWTP, Perth, Australia, and is quoted as being capable of dissolving more than

53 mg/1 of oxygen at 30° C (Cadee et al., 1984). The dimensions of this dissolver are as

follows:

CHAPTER 2: LITERATURE REVIEW 37
CHAPTER 2: LITERATURE REVIEW 38

(i) pipe diameter = 1100 mm (ii) height = 7500 mm (iii) width = 6970 mm (iv) phase

separator diameter = 2150 mm (v) phase separator length = 6270 mm (vi) oxygen return line

diameter = 200 mm.

(iii) Side stream dissolvers

Direct injection of pure oxygen results in efficient dissolution of oxygen if the velocity in the

main is sufficient to maintain two-phase flow. For force mains, where the wastewater

velocity is insufficient to maintain bubbles in suspension, use of sidestream dissolvers

(Section 2.6.2) has been recommended (Speece et al., 1990). Sidestream DO concentrations

of up to 30-60 mg/1 can be achieved depending upon the static head. Alternately, where

sufficient hydrostatic head is not available, the sidestream could be pressurised up to 35 mwg

using a throttling valve on the effluent of the oxygenation chamber, to achieve a sidestream

DO level of 130 mg/l.

2.6.2. Injection of oxygen in gravity sewers

The use of pure oxygen for increasing the DO level in gravity sewers has been found to be

very successful and is now well developed. The commonly used oxygen injection techniques

are described in the following sections.

(i) Sidestream venturi dissolver

The most commonly used sidestream system for both pressure mains and gravity sewers is

the one developed and marketed by the BOC group under the name Vitox (Forrest, 1985;

Garrett and Jeffries, 1984; Dack and Nadebaum, 1989). The Vitox system (Figure 2.7) draws

a portion of main flow which is then pressurised and delivered to a venturi where oxygen is
CHAPTER 2: LITERATURE REVIEW 39
CHAPTER 2: LITERATURE REVIEW 40

injected. The sidestream injected with oxygen is then returned to main flow through a high

velocity nozzle. A substantial portion of oxygen is dissolved under pressure. The discharge

of sidestream through the high velocity nozzle forms micro bubbles and causes rapid mixing

and further dissolution. The Vitox system has been reported to achieve an oxygen transfer

efficiency of 40-60% in gravity sewers (Hollerbach, 1985).

The design and operating parameters of a sidestream venturi system include the nozzle

diameter, sidestream flow, pressure, and, the target DO concentration (Dack and Nadebaum,

1989; Holder and Leow, 1994). Optimum values of these parameters for a particular site are

usually based on field tests. It has been shown that higher transfer efficiencies could be

achieved for smaller sidestream flows and small nozzle diameters (Dack and Nadebaum,

1989).

(ii) Counter-current technique

As already discussed in Section 2.4, nitrogen stripping inhibits the efficient transfer of

oxygen, especially at ambient pressures. One way to alleviate this problem is to use counter-

current technique (Figure 2.8). A DO concentration of up to 30 mg/1 can be achieved at

ambient pressures (Shaw, 1981). A transfer efficiency of up to 70% has been reported

(Hollerbach, 1985).

(iii) U-tubes and multiple U-tube dissolver

Single U-tubes (Section 2.5.1) have been successfully used for sulphide control in gravity

sewers (USEPA, 1985). The Multiple U-tube dissolver described in Section 2.6.1 can also


CHAPTER 2: LITERATURE REVIEW 41

.•• • •J 'PI

• I
II

...11
t ii
• it%

''t • •n •••

4— AlddnS

....., n••••n •n ••••n •••••••,"

• c=, c=x=r+ •="c=, ca=,
CHAPTER 2: LITERATURE REVIEW 42

be used for gravity sewers by employing an in-line booster pump (Dack and Nadebaum,

1989).

(iv) Other techniques

In addition to the various oxygen injection methods described above, the use of the

Biocone system (Wernicke, 1987; Weinsaepfel and Rabergeau, 1991), and non-porous

membranes (Sekoulov and Muller, 1986) for dissolving pure oxygen in both pressure

mains and gravity sewers is also documented. The Biocone system is developed and

marketed by L' Air Liquide and is shown in Figure 2.9.

Figure 2.9-Biocone system (Mozzel, 1985).

The non porous membrane system (Figure 2.10) employs flexible thin-walled tubes of

either a simple or an annular pipe arrangement. Oxygen is allowed to diffuse into

wastewater under a pressure of 4 bars. A mass transfer rate Of 10 mg/m 2.h has been

claimed. The main advantages of this system, as compared with conventional systems

include no loss of oxygen or gas stripping, and freedom from clogging. Oxygen injection
CHAPTER 2: LITERATURE REVIEW 43

into a hydraulic water fall in gravity sewers has been reported by USEPA (1985) and

Speece et al. (1990).

02 circulation
Free swimming
hose arrangement

(a)
. ta524/0 ..02 transfer
• Membrane hose )
n.•

- am=4) Wastewater flow-

/7772771 '
77-17-77777 r y. • •
'

(b) •

Perforated disc -
..- 1"
\.*. .'. ... . . '.- . .+. J
Wastewater flow Ems3ditiej \_ Guide hose .
'Membrane hose
,/..,/,,,, : f ;•„.,,,,,,,-,y. - -,2/ . . ' / • .

Figure 2.10-The non-porous memberane system (Sekoulov and Muller, 1986).

(a) Enclosed gas circulation; (b) Single hose system

2.7. ADVERSE EFFECTS OF IN-SEWER AERATION OR OXYGENATION

2.7.1. Gas locking

As discussed in Section 2.4, attempts to achieve higher concentrations of dissolved oxygen

in sewers by injecting air or oxygen result in the production of residual gases. In a pressure

main of even gradient, these undissolved gases are carried along the pipe soffitt at

approximately 80% of flow velocity (Carrie et al., 1982). However, in a pressure main of

irregular profile, the bouancyj of the residual gases tends to resist flow in the downward

gradient, thus causing accumulation of bubbles at high points i.e. "gas-locking" (USEPA,
CHAPTER 2: LITERATURE REVIEW 44

1974; Dawson, 1980; Wernicke, 1987). The degree of gas locking is mainly affected by the

rate of gas injection, degree of oxygen dispersion at the injection points, wastewater velocity,

pump and pipeline characteristic, and the effectiveness of air valves along the main (Came et

al. 1982).

The adverse effects of gas locking include (USEPA, 1974; Boon et al., 1977) :

(i) an increase in the static head of the system and an increased friction losses causing a

reduction in pumping rate and an increase in power consumption,

(ii) pipe-line restriction and hence reduction in the capacity of the main,

(iii) prevention of the operation of siphons in the main, which may initiate water hammer

with a risk of fracture of the main, and

(iv) odour nuisance when residual gases are released to the atmosphere.

The various measures taken to minimise gas locking include:

(i) multiple injection points and release of residual gases at high points. Residual gases can

be vented at high points by providing air release valves. In those cases where a very high

stand pipe is required, the use of a return pipe taking a continuous bleed of wastewater

and air from the high point to a suitable point in the wastewater collection system has

been recommended (USEPA, 1974),

(iii) maintaining the velocity of wastewater above 0.6 m/s and a Reynold's Number of at

least 300,000 to ensure that undissolved gases remain entrained in the wastewater

(Newcombe et al., 1979),

(iv) using high efficiency dissolvers, and

(v) injecting air or oxygen after down grades.

CHAPTER 2: LITERATURE REVIEW 45

2.7.2. pH change

The injection of air or oxygen in pressure mains (which is essentially a closed reactor) to

achieve high concentrations of dissolved oxygen, may cause a decrease in the wastewater pH

due to an accumulation of CO2 (Vivona, 1980). According to Pomeroy and Lofy (1977) it is

unlikely that this effect would cause any significant problem.

2.8. COSTS OF IN-SEWER AERATION AND OXYGENATION

The total cost of air injection consists of capital recovery cost, equipment maintenance cost,

and the cost of energy for air compression and drying (Vivona, 1982; Speece et al., 1990).

Electricity, and hence the energy costs, vary widely depending upon the location and

demand. Typical energy efficiency of force main aeration has been reported to be about 0.3

to 1 kg of 02 per kw-h of electrical energy (Pomeroy and Lofy, 1977) and it has been shown

that the cost of air systems increases exponentially with an increase in the DO level (Speece

eta!., 1990).

As compared with air which is free, the cost of oxygen injection also includes the cost of

oxygen generation and varies with the location of the site and the demand. Oxygen can be

either delivered to the site in liquid form or can be generated on-site using the Pressure

Swing Absorption System for higher demands (USEPA, 1985).

In the UK, typical on-site generation costs of oxygen (production and dissolution) supplied

by BOC varies from £20 to £45 per tonne (BOC, 1994). The equivalent cost for cryogenic

oxygen delivered at site as a liquid was £37 to £80 per tonne varying as a function of size of

the demand, transportation distance, and availability for delivery. The oxygen transfer power
CHAPTER 2: LITERATURE REVIEW 46

has been claimed up to 10 kg 02/kw-h. The cost of oxygen may be higher outside the UK,

especially where the technology is not so well developed. Typical costs (budget level

estimates: +30%, -15%) for direct air injection and direct oxygen injection in the USA are

given in Table 2.4 and 2.5 respectively. If separate dissolution system is required, then this

cost (Tables 2.4 and 2.5) would increase depending on the cost of the dissolution system.

Table 2.6 shows typical costs of some of the oxygen dissolution systems marketed by CIG,

Australia.

Table 2.4-Typical costs for direct compressed air injection into a force main for H2S

control (USEPA, 1985)

Condition Capital cost *( 1984 US $)

Flow = 37, 850 m3/day

Pipe: diameter = 25.4 cm

Length = 1, 600 m 21, 000

Pressure = 158 kPa

Air flow = 2.5 m3/min

Flow = 37, 850 m3/day

Pipe: diameter = 61 cm

length = 1, 600 m 54, 000

Pressure = 158 kPa

Air flow = 15.3 m3/min

Note: *includes concrete pad, compressor, piping, valves, start-stop, controls, pre-fab buildings and

installation.


CHAPTER 2: LITERATURE REVIEW 47

Table 2.5-Typical costs for direct oxygen injection into a force main for H 2S control (US
EPA, 1985)

Capital cost *( 1984 US $)

Condition Capital cost ($) Oxygen ($/year)
Flow = 37, 850 m3/day
Pipe: diameter = 36 cm 20, 000 9, 000 a
length = 1, 600 m
02 required = 83 kg/day
Flow = 37, 850 mi/day
Pipe: diameter = 76 cm 50, 000 20, 000 b
length = 1, 600 m
02 required = 310 kg/day
Notes: *includes concrete pads, liquid 0 2 storage vessel, vaporiser, piping, start-stop controls and installation.
a Based on 02 cost of $0.3 I/kg
b
Based on 0 2 cost of $0.18/kg

Table 2.6-Costs of various oxygen dissolving techniques (Ogston, 1994)

Capital Cost
Dissolver Type
(1994 Aust. $)

Direct Injection Double Injection Assembly/ dry well pump station 4332

Double Injection Assembly/wet well pump station 3800
Sidestream Venturi : Type D inclu. pump and nominal pipe work 37400
Single Stage U-Tube : 300 mm 17500
400 mm 25900
500 mm 36000
900 mm 90000
3-Stage U-Tube : 300 mm 33000

400 mm 44200

500 mm 66000

900 mm 180000

In-Line Nozzle : 300/100 6000

500/300 7300
CHAPTER 2: LITERATURE REVIEW 48

The selection of air or oxygen for a particular system would depend upon a number of factors

including the objective of DO enhancement, total oxygen demand, location of site, power

costs, and characteristics of the sewerage system. The final decision regarding the use of air

or oxygen system would be based on a detailed technical and economical evaluation of both

systems and safety considerations. If the objective is to control sulphides only, then in many

situations the overall cost of an air system may be less than that of an oxygen system. The

use of oxygen however, would be economically attractive, if the additional benefits of

oxygen addition, i.e. in-sewer treatment, is considered (Pomeroy and Lofy, 1977; Mozell,

1985; Speece et al., 1990). Considerable savings in investment have been reported when

oxygen injection was used for sulphide control and in-sewer treatment (Newcombe et al.,

1979; Hutchinson, 1981).

2.9. SUMMARY

The following main conclusions can be drawn from this literature survey on in-sewer

treatment:

(i) by maintaining aerobic conditions in sewers, a significant degree of in-sewer treatment

could be achieved,

(ii) the in-sewer purification could be enhanced either by seeding the sewers by the addition

of activated sludge or by increasing the slime supporting area. The former method

however, would be favoured, because of the operational problems and high cost

associated with the latter,

(iii) the oxygen requirements of flowing wastewater within a sewer vary greatly depending

upon the age and temperature of wastewater, concentration of biomass, concentration of

organic matter, and the wetted area. Although shallow gravity sewers may maintain
CHAPTER 2: LITERATURE REVIEW 49

themselves aerobic, anaerobic conditions would prevail in pressure mains and deep

gravity sewers, unless the dissolved oxygen is supplemented artificially,

(iv) both air and oxygen injection are widely used for the control of sulphide in pressure

mains. In-sewer oxygen injection techniques for both pressure mains and gravity sewers

are well developed, and

(v) the field of in-sewer treatment has been investigated to a limited extent. Also, due to

limited operating experience, a thorough understanding of the process and hence, detailed

design criteria of in-sewer treatment system have not been developed. A great deal of

research and operating experience would be required before detailed design criteria for

such a system could be developed.

 50

CHAPTER THREE

AIMS OF RESEARCH

The review of literature has shown that there exists a strong possibility of using urban

sewerage systems as aerobic biological treatment facilities for reducing the load on

existing overloaded wastewater treatment plants or reducing the size of a proposed

plant. Under favourable conditions i.e. where the retention time is long and the

temperature is high (Hemmings et aL, 1983), or where the wastewater is seeded with

activated sludge (Green et al., 1985), the degree of treatment which could be achieved in

sewers may be comparable to that found in a conventional secondary treatment plant.

Despite the installation of full-scale in-sewer treatment systems in Bath, UK, and in

Boulder Bay in Australia, there have been few scientific investigations on in-sewer

treatment. Most of the studies undertaken so far either involved direct in-situ

measurements of BOD 5, or were focused merely on the removal of organic matter in a

particular system which had a specified retention time. No study appears to have been

undertaken to fully investigate the treatment potential of sewers while taking into

account various operational parameters of an in-sewer treatment system. Furthermore,

many studies seem to have been undertaken without any reference to previous

investigations resulting in the field of in-sewer treatment remaining in its infancy.

Consequently, to gain a better understanding of in-sewer treatment systems, there is a

need to explore, in-depth, the treatment potential of sewers and to identify and optimise

the various parameters affecting in-sewer treatment.

CHAPTER THREE : AIMS OF RESEARCH 51

This study, therefore, will be undertaken with the broad objectives of investigating the

possibility of using urban sewerage systems for the treatment of domestic wastewater

during its transit. The particular objectives of this study are:

(i) to investigate the degree of in-sewer treatment which may be achieved by the

suspended biomass normally present in the wastewater while aerobic conditions are

maintained in the sewer,

(ii) to study the possibility of enhancing in-sewer aerobic treatment by increasing the

suspended biomass of the flowing wastewater by the addition of activated sludge at

the inlet of the sewer,

(iii) to study the oxygen requirements of various treatment alternatives stated under

objectives (i) and (ii), and

(iv) to study the settling characteristics of the effluent from the treatment systems stated

under objective (i) and (ii).

 52

CHAPTER 4

EXPERIMENTAL DESIGN, MATERIALS AND METHODS

4.1. INTRODUCTION

To achieve the research objectives, the current study was divided into two phases. The

first phase was undertaken in the Environmental Engineering Laboratory, University of

Newcastle upon Tyne, while the second phase was undertaken at the Central Laboratories

of the Water Authority of Jordan, Amman, Jordan. A description of the experimental

design, materials and experimental methods used in both phases is given in the following

sections.

4.2. EXPERIMENTAL DESIGN

4.2.1. Phase-I studies

In the first phase, which comprised three parts, a general study of the possibility of using

sewers as a treatment facility was investigated. The work undertaken in each part of the

Phase-I is as follows:

(i) Part-1: Preliminary studies

(a) a comparative study of aerobic and anaerobic conditions in sewers on in-sewer

removal of COD, and SS at the downstream wastewater treatment plant,

(b) a comparative study of the effect of the wastewater velocity (mixing speed) in the

sewer on in-sewer removal of soluble COD, and SS at the downstream wastewater

treatment plant, and

 CHAPTER 4: MATERIALS AND METHODS 53

(c) a comparative study of the effect of the dissolved oxygen source (air or oxygen) on

in-sewer removal of soluble COD, and SS at the downstream wastewater treatment

plant.

(ii) Part-II: Addition of oxygen to raw wastewater

(a) the removal of organic matter in sewer under aerobic conditions while taking into

account the effect of variations in the wastewater strength, and

(b) the effect of temperature variations on in-sewer removal of soluble COD.

(iii) Part-III: Addition of oxygen to seeded wastewater

In this part of the study, the possibility of enhancing in-sewer aerobic treatment by the

addition of activated sludge, at an initial concentration of 100-1000 mg VSS/1, at the inlet

of a sewer, was investigated.

4.2.2. Phase-II studies

In this phase of the current study, a case-study was undertaken to assess:

(i) the degree of treatment that could be achieved in the Greater Amman Siphon (GAS)

under aerobic conditions without any seeding, and

(iii) the degree of treatment that can be achieved in the GAS under aerobic conditions

with the addition of activated sludge, at an initial concentration of 100-500 mg VSS/1, at

the inlet.
 CHAPTER 4: MATERIALS AND METHODS 54

4.3. MATERIALS AND METHODS

4.3.1. Phase-I studies

(i) Feed wastewater

The raw wastewater (screened) used in this study was collected from the inlets to three

local wastewater treatment plants i.e. Durham, Sacriston, and Broohaugh, all treating

mainly domestic wastewater. Because of the length of the time required to carry out a

batch experiment, it was not possible to use the wastewater immediately after collection.

The grab samples of the wastewater were therefore stored at 4°C just after arrival in the

laboratory and were used for batch tests within 24 hours of collection.

(ii) Seed sludge

The activated sludge used for seeding the raw wastewater was the returned sludge

collected from Durham Wastewater Treatment Plant. The returned activated sludge (RAS)

was allowed to settle for one hour just after arrival in the laboratory and the supernatant

decanted in order to increase the suspended solids concentration. The sludge was aerated

overnight before being used. As the sludge was collected from a wastewater treatment

plant treating mainly (98%) domestic wastewater, the effect of acclimatisation was

considered to be negligible.

Before the start of an experiment, the suspended solids (SS) and volatile suspended solids

(VSS) of the returned sludge were determined and the volume of RAS to be added to the

wastewater to achieve a given concentration was calculated by the mass balance equation

as follows:
CHAPTER 4: MATERIALS AND METHODS 55

(v. + va )x vss,. = Vw x (vss.)+ va x (vss a ) (4.1)

Where:

V, = volume of the wastewater in the batch reactor (litres),

Va = volume of returned activated sludge (litres) to be added to V„ to achieve an active

biomass concentration of VSS, (mg/1) in the batch reactor,

VSS, = volatile suspended solids of the raw wastewater (mg/1), and

VSSa = volatile suspended solids of returned activated sludge (mg/1).

It should be noted that in Equation 4.1 the value of the VSS of the raw wastewater prior to

the addition of sludge (VSS, ) was assumed to be zero.

(iii) Reactors

A sewer conforms perfectly to plug flow conditions (Hemmings et al., 1983). The

transport of the wastewater in a sewer, therefore, could be simulated by a batch reactor

with a retention time equivalent to that of the retention time of the wastewater in the

sewer (Green et al., 1985). The retention time in a sewer is related to its length as follows:

t=— (4.2)
V

Where:

t = retention time in the sewer (hours),

L = length of the sewer (m), and

V = wastewater velocity (m/s).

 CHAPTER 4: MATERIALS AND METHODS 56

In the present study two different types of batch reactors were used to simulate the

transport of the wastewater in a linear gravity sewer. The 5-litre magnetic-stirred glass

rectors were used to compare the aerobic and anaerobic gravity transport in sewers

(Section 5.2) while in order to have a greater sample volume, maintain a constant

temperature and achieve better mixing, especially at high seed concentrations, it was

decided to use the impeller mixed reactors for the rest of experimental work. A brief

description of both types of reactor is given in the following sections.

(a) Batch reactors employing magnetic stirrer

These reactors (Figure 4.1) consisted of glass aspirators each having a working volume of

5 litres. The anaerobic reactor was kept air tight and was injected with nitrogen at the start

of the experiment to exclude the entrained air. In the aerobic reactor, the dissolved oxygen

(DO) was maintained well above 2 mg/1 by intermittent injection of oxygen through a

porous diffuser placed at the bottom of the reactor. Both the reactors were mixed using a

magnetic stirrer and maintained at ambient temperature (average 22°C).

(b) Impeller mixed batch reactor

This reactor consisted of a 20-litre cylinder which had a working volume of 10-15 litres.

The reactor contents were mixed using a variable speed mechanical stirrer. All the

experiments except those designed to investigate the effect of mixing speed on in-sewer

treatment (Section 5.3) were undertaken at a mixing speed of 100 rpm. A constant

temperature was maintained by placing the reactor in a controlled-temperature water bath.

Aerobic conditions were maintained in the reactor by intermittent supply of oxygen

through four porous diffusers placed at the bottom of the reactor. A schematic drawing of

the experimental set-up for impeller mixed batch reactor is shown in Figure 4.2

CHAPTER 4: MATERIALS AND METHODS 57

Figure 4.1. Pictorial view of magnetic mixed batch reactors

1. Aerobic reactor; 2. Anaerobic reactor; 3. Oxygen cylinder; 4. Nitrogen cylinder;

5. Magnetic mixer.

10

5
4
L9

Figure 4.2- Experimental set-up for impeller-mixed batch reactors

1. Batch reactor; 2. DO probe; 3. Impeller; 4. Water bath; 5. Diffuser; 6. Oxygen cylinder;

7. Oxygen supply line. 8. Valve; 9. Temperature controller; 10. DO meter; 11. Support
slab
 CHAPTER 4: MATERIALS AND METHODS 58

(iv) Reactor operation and feeding

The containers of wastewater were held in a water bath before feeding to equilibrate with

the operating temperature. After the wastewater had reached the desired temperature, it

was fed to the batch reactor manually, the reactor contents were mixed and an influent

sample was taken. The oxygen supply was then switched on and effluent samples were

taken at specified intervals of time. For tests involving the addition of active biomass to

the wastewater (Chapters 7 and 8), a specified quantity of the RAS (Equation 4.1) was

added to the reactor just before the oxygen supply was turned on. The reactor contents

were emptied and disposed of at the end of each individual test. The reactor was then

washed thoroughly with water to ensure that no biomass growth occurred on the walls of

the reactor.

(v) Measurement of dissolved oxygen

The dissolved oxygen (DO) in the reactor was measured using a dissolved oxygen probe

(YSI, Model 57). The DO meter was calibrated at the start of each experiment by reading

the value shown on the meter against a sample of known DO concentration (determined

by Azide modification of Winkler method), and was zeroed by placing the probe in a

solution of sodium sulphite and cobalt chloride.

(vi) Analytical methods

Influent and effluent samples were analysed immediately or stored at 4°C. All the analysis

were carried out within 24 hours. Unfiltered samples were used for the determination of

total COD and suspended solids (SS) while filtered samples (using Whatman GF/A) were

used for the determination of soluble COD and BOD 5 . All analyses were carried out

CHAPTER 4: MATERIALS AND METHODS 59

according to Standard Methods (1989). A description of various tests is given in Table

4.1.

Table 4.1-Analytical methods and sensors or meters used (Standard Methods, 1989)

Parameter Method

BOD5 (ATU) Dilution method, DO by modified Winkler method

DO Electrode

Soluble COD Closed reflux, titrimetric Method

SOUR See text

SS Total suspended solids dried at 103-105°C

SVI Static method

Total COD Open reflux, titrimetric method

V S S Fixed and volatile solids ignited at 5500

(vii) Measurement of settlability

A one-litre glass cylinder (Appendix A) was used as a settling tank. For experiments

involving the addition of activated sludge to the wastewater, the settleability was

measured using Sludge Volume Index (Standard Methods, 1989).

(viii) Measurement of oxygen uptake rate of suspended biomass

The oxygen uptake rate of suspended biomass (SOUR) was measured using an SOUR

reactor consisting of a one-litre air tight plastic container which was fitted with a

dissolved oxygen meter (Figure 4.3). The contents of the SOUR reactor were kept mixed
 CHAPTER 4: MATERIALS AND METHODS 60

by a magnetic stirrer. At specified intervals of time, the SOUR reactor was filled with

wastewater or mixed liquor from the main batch reactor and the DO level increased to 6-8

mg/I if required. The change in the DO level in the reactor over time was monitored with

a DO probe. The contents of the SOUR reactor were transferred to the batch reactor

afterwards. The gradient of DO concentration was taken as SOUR (mg/1.h).

J
o 4 0

Figure 4.3. The SOUR reactor

1. Plastic container; 2. DO probe; 3. Stir bar; 4. Magnetic stirrer

(ix) Calculation of removal efficiencies

The soluble COD or BOD5 removal efficiencies for a given retention time were calculated

on the basis of the soluble COD or BOD 5 of the wastewater at the start of the experiment

and after that retention time, without settling the effluent. A similar method was used for

the calculation of total COD removal efficiencies. The removal efficiency R is given by:
 CHAPTER 4: MATERIALS AND METHODS 61

S —S
R= 100( ° 1) (4.3)
So

Where:

R = COD or BOD5 removal efficiency (%) after t hours from the start of the experiment,

So = COD or BOD5 concentration (mg/1) at the start of the experiment, and

St = COD or BOD5 concentration (mg/1) after t hours from the start of the experiment.

The amount of soluble COD removal (mg/1) over a given retention time (t hours) was

found as follows:

RSCOD = (So-S t) (4.4)

Where: RSCOD = soluble COD removal (mg/1) after a given retention time t (hours).

For experiments involving the addition of actived sludge to the wastewater, the removal

efficiency was calculated using Equation 4.3, except the influent soluble CODs and

soluble BOD 5s were calculated by applying mass balance to the CODs and BOD 5s of the

raw wastewater and the activated sludge as follows:

Saa +Sr/
Ow
SO= (4.5)
Vw+Va

Where:

So = influent SCOD or SBOD 5 for seeded experiments (mg/1),

Sa = SCOD or SBOD 5 (mg/I) of the activated sludge at the start of the experiment,
 CHAPTER 4: MATERIALS AND METHODS 62

Va = Volume of returned activated sludge (litres) added to the wastewater, and

V, = volume of the wastewater in the batch reactor (litres).

It must be noted that terms "removal efficiencies" or "treatment efficiencies" used

throughout this dissertation refer to the treatment efficiencies achieved by the suspended

biomass only, as the effect of slime layer on the sewer wall was not considered.

4.3.2. Phase-II studies

(i) Description of the Greater Amman Siphon

The Greater Amman Siphon (GAS) conveys the screened and degritted wastewater and

septage, from the city of Amman from Ain Ghazal pre-treatment facility, to Al-Samra

wastewater stabilisation ponds system (AWSP). The GAS (Appendix A-1) is a pressure

gravity pipe line and consists of 52 grade steel with welded joints. It is 38.6 km long and

has a diameter of 1228 mm. Its elevation at the inlet near Ain Ghazal is 688 m while at

the discharge point the elevation is 580 m. The elevation at the lowest point of the GAS,

which is at a distance of about 26 km from the inlet, is 460 m. The design pressure at this

point is approximately 23 bars. The retention time in the GAS averages 8 hours.

The total flow at the inlet of the GAS averages 115,000 m 3/day, with average domestic

wastewater and the septage flows amounting to 110,000 m 3/day and 5000 m3/day,

respectively. A step flow of about 30,000 m 3/day is pumped into the siphon at Zarqa, at a

distance of about 25 km from the head of the siphon. For the purpose of the current study

however, the step-flow to the GAS was ignored, as a separate WWTP is being planned for

Zarqa area.
 CHAPTER 4: MATERIALS AND METHODS 63

(ii) The Al-Samra wastewater stabilisation ponds system

The Al-Samra wastewater stabilisation ponds system, the largest WSP system in the

Middle East, consists of three trains, each comprising of anaerobic, facultative, and

maturation ponds. Salient features of the AWSP system are given in Table 4.2. The

AWSP system in addition to treating and disposing the major portion of domestic

wastewater, septage and industrial wastewater from the city of Amman, also handles and

treats the wastewater from Zarqa, Ruseifa and Hashimiya.

Table 4.2-Salient features of AWSP system (Al-Salem, 1987)

Effective area 181 hectares

Effective volume 2.8 million m3

Design flow-Average dry weather 68,000 m3/day

Peak wet weather 148,000 m3/day

The AWSP system, which was commissioned in 1985, was designed to treat an average

dry weather flow of 68,000 m3/day. The system however, is currently overloaded by

150%, with the average daily flow amounting to 170,000 m 3/day. The Water Authority of

Jordan (WAJ) intends to upgrade the AWSP system and studies are underway to

determine the various alternative options for extension.

(iii) Feed wastewater

The wastewater used in this study was collected from inlet to the GAS, at AM Ghazal.

The wastewater consisted of a mixture of domestic wastewater and septage, both of which

were separately pre-treated at Ain Ghazal pre-treatment works comprising of screens and
 CHAPTER 4: MATERIALS AND METHODS 64

aerated grit chambers. The grab samples of the wastewater were collected at Ain Ghazal

between 8 and 9 am and were subjected to batch testes at WAJ laboratories at Wadi Seer,

within three hours of collection.

(iv) seed sludge

The returned activated sludge used for seeding the raw wastewater was collected from

Abu Nusair extended aeration activated sludge plant. The quantity of sludge added to the

raw wastewater to attain a given activated sludge concentration in the reactor was

calculated using Equation 4.1.

(v) Experimental methodology

The impeller-mixed batch reactors used in this study (Appendix A-2) were similar to

those described in Sections 4.3.1, except the temperature of the wastewater was

maintained at ambient. The experimental methods used for this study were essentially the

same as described in Sections 4.3.1, except for some of analytical methods as described in

the following section.

(vi) Analytical methods

Filtered samples (using Whatman GF/C) were used for the determination of soluble COD

and BOD5 . As practised in WAJ laboratories, no chemical (i.e. ATU) was added to the

dilution water to arrest the nitrification during the BOD 5 test. All analyses were carried

out according to Standard Methods (1989).

 65

CHAPTER 5

PRELIMINARY STUDIES

5.1. INTRODUCTION

A sewer constitutes a dual phase system wherein the heterotrophic bacteria are present in

suspended form in the flowing wastewater and in attached form in the slime layer on the

walls of the sewer.

During the transit of wastewater to treatment plants, the dissolved oxygen in the

wastewater is depleted as a result of the respiration of the indigenous biomass. Although

shallow gravity sewers may maintain themselves aerobic, anaerobic conditions prevail in

deep gravity sewers and pressure mains unless the dissolved oxygen is supplemented

(USEPA, 1974). The time for the onset of anaerobic conditions could be as short as 10

minutes in small pipes, and 20 to 30 minutes in large pipes (Pomeroy, 1992).

Although a few studies have been undertaken in the past to investigate the removal of

organic matter in sewers under aerobic conditions (Section 2.2), there does not appear to

be much information in the literature regarding the effect of anaerobic conditions in sewers

on the removal of organic matter (COD or BOD) and the subsequent settleability at the

downstream treatment works.

The main objective of the work described in this chapter was to compare the removal of

soluble organic matter in the wastewater and the subsequent primary settleability, under

aerobic and anaerobic conditions in sewers. In addition to this, an investigation was carried

out on the effect of turbulence (wastewater velocity), and the dissolved oxygen source (air
 CHAPTER 5: PRELIMINARY STUDIES 66

or oxygen) on the removal of soluble COD during the simulated aerobic gravity transport,

and the subsequent primary settleability, in order to establish the optimum operating

conditions for further experimental work.

5.2. AEROBIC VERSUS ANAEROBIC CONDITIONS IN SEWERS

To compare the aerobic and anaerobic transport of domestic wastewater in a gravity sewer,

two identical 5-litre batch reactors (Figure 4.1) were run in parallel at ambient temperature

(average 22°C). A detailed description of the experimental methodology used for this

study is presented in Section 4.3.1.

5.2.1. Soluble COD removal

Table 5.1 summarises the average SCOD removal efficiencies under different redox

conditions. The variation in SCODs of the individual wastewaters over a retention period

of 8 hours is illustrated in Figures 5.1 and 5.2.

Table 5.1-Soluble COD removal efficiencies under different redox conditions

Retention time Redox condition SCOD removal (%)

2 Anaerobic 12±4*

Aerobic 15±3

4 Anaerobic 13±4

Aerobic 25±3
6 Anaerobic 11±6
Aerobic 35±3
8 Anaerobic 6±7
Aerobic 36±6
Notes (i) * average-±SD (iii) average influent SCOD=291±33 mg/I
(ii) average influent SS=366±75 mg/I (iv) average temperature= 22°C
 CHAPTER 5: PRELIMINARY STUDIES 67

350 _
SS (mg/I'
• 290 • 380
I
300 _ • 330 • 465
,---, x x average best fit (average)
-61)
E
' 250 •
a)
0 U
U
cip •
• •
200 _
•

•
150 •

0 2 4 6 8
Retention time (hours)

Figure 5.1-Variation in SCOD of domestic wastewater under aerobic conditions.

SS (mg/I)
350 _
• 290 • 380
A 330 • 465
x average best fit (average)

0 2 4 6 8
Retention time (hours)

Figure 5.2-Variation in SCOD of domestic wastewater under anaerobic conditions.

 CHAPTER 5: PRELIMINARY STUDIES 68

As can be seen from Figure 5.1, the SCOD in the wastewater under aerobic conditions

showed a gradual decrease with the passage of time. The SCOD removal over 8 hours

ranged from 80-140 mg/l. The net SCOD removal rate averaged 13 mg/1.h over a retention

period of 8 hours. The SCOD removal rate was at a maximum during the first two hours of

oxygen addition and showed a decrease with increase in retention time.

Under anaerobic conditions, the SCOD in the reactor showed a decrease initially which

was then followed by an increase (Figure 5.2). The net SCOD removal ranged from 30-

mg/1 over the first four hours from the start of the experiment. During the next four hours,

the SCOD in the reactor increased by an average of 25 mg/1 (n=3). The net SCOD removal

over 8 hours thus decreased to 5-10 mg/l.

5.2.2. Total COD removal

The influent total COD (TCOD) of the wastewater averaged 769 mg/1 (SD=55). The

TCOD removal efficiency as calculated on the basis of the TCOD of unsettled effluent

(Equation 4.3) after a retention period of 8 hours averaged 8% (SD=8) and 11% (SD=6)

under aerobic and anaerobic conditions, respectively.

5.2.3. Effect of redox conditions on the settleability of solids .

To study the effect of a range of redox conditions on the settleability of SS downstream of

the sewer, effluent samples taken from batch reactors after a retention time of 6 hours were

settled in a 1 litre cylinder (Section 4.3.1) for 1 hour and the SS of the supernatant was

determined. The results of the settleability tests are shown in Table 5.2.
 CHAPTER 5: PRELIMINARY STUDIES 69

Table 5.2-Effect of redox conditions on the removal of SS

Redox condition Effluent SS Supernatant SS Solids removal
mg/1 mg/1

Anaerobic 335±48* 145±18 56±6

Aerobic 380±76 105±9 72±4
Notes: (i) * average ± SD
(ii) Influent SS = 370±75 mg/1
(iii) n = 4

In the aerobic reactor, the raw wastewater appeared to assume the form of a dilute

suspension of activated sludge after the first hour of oxygen addition. The net changes in

SS of the raw wastewater over a retention period of 6 hours varied from 0 to +6%

(average=+3, SD=2) under aerobic conditions, and 0 to -18% (average -7%, SD=8) under

anaerobic conditions. A comparison of the settleability results for both redox conditions

(Table 5.2) showed that the average solids removal under aerobic conditions was 29%

higher than that under anaerobic condition.

5.3. EFFECT OF TURBULENCE ON IN-SEWER AEROBIC TREATMENT

Velocities in sewers vary greatly due to the variation in wastewater flow. To study whether

these changes in velocity (turbulence) would have any effect on the removal of soluble

organic matter and primary settleability under aerobic conditions, and subsequently to

select an appropriate mixing speed to be used for further experimental work, four identical

impeller-mixed batch reactors (Figure 4.2), were run in parallel at different mixing speeds,

at a constant temperature (20°C). Three different wastewaters were subjected to batch tests

at mixing speeds of 50, 100, 200 and 300 rpm. Effluent samples, taken after a retention

time of 6 hours were settled in a 1 litre cylinder (Section 4.3.1) for 1 hour and the SS of the
 CHAPTER 5: PRELIMINARY STUDIES 70

supernatant was determined, while those taken after a retention period of 8 hours were

analysed for SCUD. Table 5.3 summarises the average SCUD and SS removal efficiencies

at different mixing speeds.

Table 5.3- Effect of mixing speed on SCOD and SS removal under aerobic conditions

Mixing speed (rpm) 50 100 200 300

SCUD removal (%) 41±4 42±4 39±3 42±5

Solids removal (%) 74±12 77±10 74±10 47±3
Notes: (i) average influent SCUD = 252±113 mg/1
(ii) average influent SS = 437±163 mg/I
(iii) n = 3

To determine statistically whether the changes in mixing speed had significantly affected

the SCUD removal, one-way analysis of variance (ANOVA) tests (Metcalf, 1994) using

Minitab 9.2, were carried out on the raw data. The results of the ANOVA tests (Appendix

B-1) indicated that at 5% significance level, there was no significant difference (p > 0.05)

in terms of SCOD removal, due to changes in mixing speed in the range of 50-300 rpm,

over a retention period of 8 hours. Similarly, no significance difference was noted in SS

removal efficiencies at mixing speeds of 50-200 rpm. An increase in mixing speed beyond

200 rpm however, was found to significantly affect the amount of SS removed (Table 5.3

and Appendix B-1) possibly due to increased turbulence in the batch reactor.

5.4. AIR VERSUS OXYGEN FOR IN-SEWER AEROBIC TREATMENT

Both air and oxygen are widely used for sulphide control in pressure mains, while in-sewer

oxygenation for gravity sewers has also been reported (Section 2.6). In-sewer oxygenation,

as compared to in-sewer aeration, has been claimed to result in an improved primary

 CHAPTER 5: PRELIMINARY STUDIES 71

settleability (Newcombe et al., 1979; Mozell, 1985). To compare the removal of soluble

organic matter and primary settleability for in-sewer oxygenation and in-sewer aeration

systems, two identical impeller mixed batch reactors (Figure 4.2), one using pure oxygen

and other air as a source of dissolved oxygen, were operated in parallel at 20°C. Effluent

samples, taken after a retention period of 6 hours were settled in a 1 litre cylinder for 1

hour and the SS of the supernatant was determined, while those taken after a retention

period of 8 hours were analysed for SCOD. Table 5.4 summarises the average SCOD and

SS removal efficiencies for air and oxygen systems.

Table 5.4-Effect of dissolved oxygen source on SCOD and SS removal

Dissolved oxygen source Air Oxygen

SCOD removal (%) 36±5 39±5

SS removal (%) 66±4 70±4

Notes: (i) average influent SCUD = 286±28 mg/I

(ii) average influent SS = 308±33 mg/1
(iii) n = 4

From the results shown in Table 5.4, it appears that the use of air when compared with

oxygen resulted in 3 and 4% higher SCOD and SS removal, respectively. To assess

statistically the effect of dissolved oxygen source on SCOD and SS removal however, one-

way ANOVA test was applied to the raw data collected during this study. The results of

the ANOVA tests (Appendix B-2) revealed that at a significance level of 5%, there was no

significant difference (p > 0.05) in soluble COD removal, for both air and oxygen systems,

over a retention period of 8 hours. A similar effect was found for the suspended solids

removal efficiencies after a retention period of 6 hours (Appendix B-2).

 CHAPTER 5: PRELIMINARY STUDIES 72

5.5. DISCUSSION

The results shown in Table 5.1 and Figures 5.1 and 5.2, demonstrate that the greatest

removal of soluble organic matter would occur under aerobic conditions. The higher

removal rate during the first two hours under aerobic conditions is possibly due to

oxidation of the readily biodegradable fraction of soluble organic matter while a

subsequent decrease in the soluble COD removal rate may be attributed to an increase in

soluble COD due to hydrolysis of the particulate organic matter (Section 6.7.1).

The results of the present study indicate that under anaerobic conditions, the soluble COD

of the wastewater, in general, decreases during the first four hours which is then followed

by an increase in SCUD with the passage of time, thus offsetting the early decrease. An

increase in soluble COD of the wastewater during the latter stages of the experiments

could be due to hydrolysis of particulate organic matter.

When the results of one experiment not conforming to the general trend (SS = 465 mg/1,

Figure 5.1) were omitted, the SCOD removal over 8 hours under aerobic conditions

averaged 95 mg/l. The corresponding SCUD removal under anaerobic conditions averaged

8 mg/1 or one twelfth of that achieved under aerobic conditions. This is in reasonable

agreement with the argument of Boon (as quoted by Pullin, 1977) that, under anaerobic

conditions in sewers, the degradation rate of the wastewater would be about one fourteenth

of that under aerobic conditions.

Greenfield et al. (1980) have reported a soluble BOD 5 removal pattern under anaerobic

conditions similar to the SCUD removal pattern found in the present study. Nielsen et al.
 CHAPTER 5: PRELIMINARY STUDIES 73

(1992) reported that, under anaerobic conditions, the changes in dissolved protein of the

wastewater were not significant over a retention period of 20 hours. Based on field studies

involving a pressure main, Hvitved-Jacobson et al. (1995) concluded that, under anaerobic

conditions, the reduction in soluble COD did not show any clear pattern. Kaijun et al.

(1995), found that SCOD removal of the raw wastewater under anaerobic conditions varied

widely during the hydrolysis phase. According to them, the main processes under such

conditions during the first two weeks are hydrolysis and acidification.

The total COD results (Section 5.2.2) indicate that under aerobic conditions, the changes in

total COD were significantly low as compared with soluble COD. This is in keeping with

the argument of Newcombe eta!. (1979) and Greenfield eta!. (1980) that little changes are

expected in total COD during aerobic transport in sewers because of the transformation of

soluble and colloidal organic matter into additional biomass, which results in an increase

in the total organic matter of the wastewater.

Results shown in Table 5.2 suggest that maintaining aerobic conditions in the sewers could

result in a significant improvement in primary sedimentation, in addition to the removal of

soluble organic matter. This observation is in accordance with the experience from the

field investigations. Newcombe et al. (1979) and Hemmings et al. (1983) found that

injection of oxygen in sewers resulted in significant reduction in load and improved

primary sedimentation the downstream treatment works. According to Hutchinson (1981),

the improvement in primary sedimentation as a result of oxygen addition is due to the fact

that under aerobic conditions, biomass tends to flocculate as it is formed into rudimentary

flocs similar to those found in activated sludge.

 CHAPTER 5: PRELIMINARY STUDIES 74

The overall results of the current study demonstrate that aerobic treatment, when compared

with anaerobic treatment, would be the most promising method for in-sewer biological

treatment, and could result in a significant removal of soluble organic matter and increased

primary sedimentation. Also, in-sewer anaerobic treatment is less likely to be favoured due

to potential odour and corrosion problems associated with anaerobic conditions.

The results of the present study indicate that changes in mixing speed (in the range of 50-

200 rpm) would not significantly affect the soluble COD removal and the subsequent

settleability. According to Pomeroy and Parkhurst (1972), the mixing speed does not make

a measurable difference in oxygen uptake rate of wastewater, as long as it is sufficient to

keep the organic solids suspended. Newcombe eta!. (1979) also noted a similar affect.

The velocity of wastewater in sewers however, has been reported to influence strongly the

dispersion of air or oxygen injected in the sewer (Newcombe et al. 1979; Speece et al.,

1990). For in-sewer oxygenation to be effective, and to prevent gas locking (Section 2.4),

Newcombe et al. (1979) recommend a wastewater velocity of above 0.6 m/s and a

Reynolds's Number of at least 300,000 to be maintained in the sewer.

From the results of the current study, in-sewer removal of soluble organic matter and the

subsequent solids removal appear to be independent of the dissolved oxygen source (air or

oxygen). In-sewer oxygenation however, may offer other advantages, as discussed in

Section 2.6.
CHAPTER 5: PRELIMINARY STUDIES 75

Based on the results of the preliminary studies, it was decided to focus further research

work on the use of sewers as an aerobic treatment facility. A rotational speed of 100 rpm

(corresponding to an average linear velocity of 1 m/s in the reactor) was selected to be

used for the rest of the work, while in order to have a better operational control in the

laboratory and avoid excessive turbulence, oxygen was chosen to be used as the source of

dissolved oxygen.
 76

CHAPTER 6

ADDITION OF OXYGEN TO RAW WASTEWATER

6.1. INTRODUCTION

As emphasised in Chapter 5, heterotrophic micro-organisms are present in sewers in

suspended form in the wastewater and in attached form in the slime layer on the wetted

area of sewer walls. It has been indicated that, under aerobic conditions, these micro-

organism would increase at a high rate thus resulting in a significant degree of in-line

purification (Pomeroy and Parkhurst, 1972; Boon eta!., 1977).

This section describes the results of the extensive experimental work undertaken to

investigate, in general, the degree of in-sewer treatment which could be achieved by the

suspended heterotrophs normally present in the wastewater, under aerobic conditions.

To simulate the aerobic transport of wastewater in a linear gravity sewer, the raw

domestic wastewater was maintained aerobic in a batch reactor (Figure 4.2) by the

addition of oxygen, at atmospheric pressure. A detailed description of the experimental

methodology adopted for this study is presented is Section 4.3.1.

6.2. WASTEWATER CHARACTERISTIC

The raw (screened) wastewater used in this study was collected from inlets to three local

wastewater treatment plants (WWTPs) namely Broomhaugh, Sacriston and Durham, all

treating mainly domestic wastewater. Table 6.1 summarises the ranges of the

characteristics of the wastewaters collected from the three WWTPs.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 77

Table 6.1-Characteristics (mg/1) of the wastewater used in simulation study

WWTP Broomhaugh Sacriston Durham

SCOD 70-215 255-390 250-400

SS 150-420 230-500 540-740
TCOD 280-750 540-1170 920-1480

As can be seen from Table 6.1, a considerable variation was noted in the strength of

wastewater collected from the different WWTPs. The samples collected from

Broomhaugh represented a wastewater having typically low soluble COD (SCOD) and

low suspended solids (SS). Samples from Sacriston depicted a wastewater with typically

high SCOD and low SS while those from Durham represented a wastewater with a high

SCOD and high SS.

6.3. REMOVAL OF SOLUBLE ORGANIC MATTER

The removal of soluble organic matter from the wastewater was studied at three different

temperatures i.e. 10 0 , 20° and 30°C. Initially a minimum of eight wastewaters were

studied at each temperature. A wide variation was observed in the strength of the

wastewater investigated at each temperature, and the removal of soluble organic matter in

the wastewater was found to vary widely depending upon the initial suspended solids

(SS0) and the initial soluble COD (SCOW of the wastewater. Consequently, it was

decided to investigate more wastewaters at 20°C in order to have wide range of SS () and

SCOD0 concentrations, so that sufficient data for carrying out the statistical analysis to

assess the effect of SS 0 and SCOD0 on the SCOD removal, could be generated. SCOD

was chosen to be the principal parameter for the measurement of organic matter in this

study because of the large number of samples involved.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 78

6.3.1. Removal of soluble organic matter at 10°C

Figure 6.1 shows the variation in the SCODs of the individual wastewaters over a

retention period of 8 hours. Table 6.2 summarises the average SCOD removal efficiencies

at 10°C while the corresponding average SCOD 0 and SS 0 concentrations are shown in

Table 6.3.

Influent SS (mg11)
400
—,--.610 —x-360 _*_ 180 _9_600 _o_ 390
-- 150 —e-610 .._270 _x_ 150 _4_420
350

300

150
x x x
100 x
x
50

T
o 2 4 6 8
Retention time (hours)

Figure 6.1-Variation in SCOD of domestic wastewater at 10°C. (Results of

wastewaters having a SCOD 0 of less than 150 mg/1 were omitted while calculating the

average SCOD removal efficiencies (Table 6.2) to minimise the scatter in the results).

From Figure 6.1, it can be seen that the SCOD removal is highly dependent on the SS0

and the SCOD 0 of the wastewater. The SCOD removal over 8 hours for Broomhaugh and

Sacriston wastewaters was found to be 76 mg/1 (SD=22) and 87 mg/I (SD=1) respectively.

The corresponding SCOD removal for Durham wastewaters averaged 196 mg/1 (SD=9).
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 79

Table 6.2-Average SCOD removal efficiencies at 10°C

Retention time SCUD removal (%)

(hours) Broomhaugh Sacriston Durham

1 15±1* 10±3 20±7

2 31±17 13±9 24±5
3 31±17 16±9 32±7
4 34±17 21±6 35±10
6 36±20 21±4 45±6
8 43±11 26±0 53±4
Note: *average±SD.

Table 6.3-Average wastewater characteristics corresponding to the data shown in

Table 6.2

WWTP SCOD 0 (mgil) SS() (mg/1) n

Broomhaugh 175±7 390±42 2

Sacriston 338±4 330±85 2
Durham 372±20 607±6 3
Note : n = number of individual samples studied

For Broomhaugh wastewaters, the influent soluble BOD5 (SBOD5) and SBOD5 removal

over 8 hours were found to be 60 mg/1 (SD=5) and 50 % (SD=15) respectively. The

corresponding values for Sacriston wastewaters amounted to 138 mg/1 (SD=3) and 41%

(SD=5). The influent SBOD 5 for Durham averaged 152 mg/1 (SD=7) while the removal

efficiency over 8 hours averaged 64% (SD=11).

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 80

6.3.2. Removal of soluble organic matter at 20°C

Figures 6.2 to 6.5 illustrate the variation of SCOD and SBOD 5 with time, for individual

wastewaters collected from different WWTPs, during the simulated aerobic gravity

transport at 20°C. The corresponding average effluent SCODs are plotted in Figure 6.6.

Table 6.4 summarises the average SCOD removal efficiencies for different wastewaters,

over a retention period of 1 to 8 hours. The SCOD0 and SS 0 values corresponding to the

results shown in Figure 6.6 and Table 6.4 are shown in Table 6.5. Table 6.6 shows the

average SBOD5 removal efficiencies with corresponging influent wastewater

characteristics summarised in Table 6.7.

Inf11.01
230
--o— 410 —0— 270 --A— 330 300
210 -NE__ 420 _0_160 ---F— 400
340

190

170

5 150

0 130_
C-)
C/D
1 10 _

90

70

2 4 6 8
Retention time (hours)

Figure 6.2-Variation in SCOD of Broomhaugh wastewater at 20°C (Results of

samples having SS() of 160, 420 and 340 mg/1 have been omitted while calculating the
average SCOD removal efficiencies shown in Table 6.4, to minimise the scatter).
CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 81

Influent SS (mg/I)
400

_o__ 280 • 270 —A— 230

380 380 400
350 310 [3 SOO --e-- 470
• 310 U 380 - -0- - 240

bO 300

(21
° 250

200

150

0 2 4 6 8
Retention time (hours)

Figure 6.3-Variation in SCOD of Sacriston wastewater at 20°C.

210 _
Influent SS (mg/I)
190 380 __x_ 380 340 0 238100
170 . --e— _o__ 340
25 70 0 27700
4 400
150 +

130

110

0 90
Pc1
70

50

30

10

2 4 6 8
Retention time (hours)

Figure 6.4-Variation in SBOD 5 of Broomhaugh and Sacriston wastewater at 20°C (B

represents samples from Broomhaugh WWTP).



CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 82

375

325 Influent SS (mg/I)

540 _0_ 660 640 ___x__ 630
x 740 _0_ 580 550 660
640 740 _s__ 580

SCOD

25

0 2 4 6 8
Retention time (hours)

Figure 6.5-Variation in SCOD and SBOD 5 of Durham wastewater at 20°C.

375
SCOD = 323 .37e-"64t
325 R2 = 0.99
o Broomhaugh
Sacriston
275 0 Durham

225
a
0
175 SCOD = 275.51e4).I7t
0 R2 = 0.96
125
COD = 157.37em.°74t
R2 = 0.94
75

0 2 4 6 8
Retention time (hours)

Figure 6.6-Variation in SCOD of domestic at 20°C (Average of n samples (Table 6.5)

are plotted, t = retention time)

CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 83

Table 6.4-Average SCOD removal efficiencies at 20°C

Retention time COD removal (%)

(hours) Broomhaugh Sacriston Durham

1 14±8 8±3 21±8

2 23±7 13±4 31±4
3 27±7 19±5 39±6
4 32±10 25±6 44±7
6 38±6 35±5 50±3
8 48±6 40±5 61±6

Table 6.5-Wastewaters characteristics corresponding to the data shown in Table 6.4

WWTP SCOD0 (mg/I) SS() (mg/1) n

Broomhaugh 176±17 353±61 5

Sacriston 329±48 346±87 12
Durham 306±30 620±70 7

Table 6.6-Average soluble BOD 5 removal efficiencies at 20°C

Retention time Average SBOD5 removal (%)
(hours) Broohaugh Sacriston Durham
2 26±5 19±6 31±11
4 42±5 35±10 57±7
6 57±6 47±11 72±3
8 64±8 59±10 81±4
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 84

Table 6.7-Wastewaters characteristics corresponding to the data shown in Table 6.6

WWTP SCOD0 (mg/1) SBOD5 (mg/1) SS() (mg/1) n

Broomhaugh 185±19 90±32 328±49 4

Sacriston 346±47 155±38 374±84 8
Durham 324±18 149±18 655±66 4

The results shown in Figures 6.2-6.6 demonstrate that that the SCOD removal pattern in

the wastewater at 20°C would strongly depend upon the initial concentration of soluble

organic matter and the suspended solids present in the wastewater.

The SCOD removal over a retention period of 8 hours averaged 84 mg/1 (SD=13) for

Broomhaugh wastewater. The corresponding SCOD removal for Sacriston and Durahm

wastewaters averaged 130 (SD=25), and 186 (SD=27) mg/1, respectively. The average

SBOD5 removal over a retention period of 8 hours averaged 56 (SD=17), 88 (SD=15) and

120 (SD=18) mg/1, for Broomhaugh, Sacriston and Durham wastewaters, respectively.

6.3.3. Removal of soluble organic matter at 30°C

Figure 6.7 shows the changes in the SCOD of the individuals wastewater over a retention

period of 8 hours. Data summarising the average SCOD removal efficiencies for different

wastewaters are listed in Table 6.8, while the corresponding influent wastewater

characteristics are shown in Table 6.9.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 85

450 _

400 InfluertSS (mg/I)

540 320 300 --x— 280

350 _
--Ne— 700 --e— 245 —I-- 165 _e— 400
300

250 _
0
C..) 200

150 _

100

50

0 2 4 6 8

Retention time (hours)

Figure 6.7-Variation in SCOD of domestic wastewater at 30°C (Results for

wastewaters having SCOD 0 of less than 150 mg/1 were omitted while calculating the

average SCUD removal efficiencies).

Table 6.8-Average SCOD removal efficiencies at 30°C

Retention time SCUD removal efficiency (%)

(hours) Broomhaugh Sacriston Durham
1 23±11 7±2 19±5
2 36±13 14±10 35±5
3 39±7 23±11 40±5
4 41+5 33±1 43±8
6 51+7 41±3 51+4
8 53±5 48±0 57±8
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 86

Table 6.9-Wastewaters characteristics corresponding to the data shown in Table 6.8

WWTP SCOD (mg/1) SS (mg/1) n

Broornhaugh 167±26 327±64 2

Sacriston 349±54 283±53 2
Durham 340±82 620±113 3

The SCOD removal over a retention period of 8 hours averaged 89 mg/1 (SD=17), 166

mg/1 (SD=24) and 191 (SD=19) mg/1 respectively for Broomhaugh, Sacriston and

Durham wastewaters. For Broomhaugh wastewater, the influent SBOD 5 and SBOD5

removal over 8 hours, were found to be 51 mg/1 (SD=12, n=2) and 75% (SD=5),

respectively. The corresponding values for Sacriston wastewaters amounted to 142 mg/1

(SD=30, n=2) and 73% (SD=3). The influent SBOD 5 for Durham averaged 123 mg/1

(SD=15, n=2) while the removal efficiency over 8 hours averaged 87% (SD=4).

6.3.4. Effect of wastewater strength on SCOD removal

As already indicated in Sections 6.3.1-6.3.3, the SCOD removal from the wastewater at

all temperatures investigated in this study, was found to vary as a function of the initial

SS (SS0) and initial SCOD (SCOD0) of the wastewater.

At 10°C, although the data collected did not cover a wide range of wastewater SCODs

and SSs (Figure 6.1) in order to statistically assess the effects of SCOD 0 and SS° on

SCOD removal, the SCOD removal, in general, appeared to increase with an increase in

SS0 and SCODo.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 87

At 10°C, the SCOD removal over 4 and 8 hours for Broomhaugh wastewaters was found

to be 38 and 60 mg/1, respectively, for a sample having SCOD 0 of 170 and SS ° of 360

mg/l. The corresponding SCUD removals increased to 83 and 91 mg/1, when the SCOD0

and SS0 of the wastewater were 180 and 420 mg/1 respectively. A very low SCUD

removal was observed over 8 hours when the SCOD 0 was in the range of 85-95 mg/1 and

the SS ° was 150 mg/l. Similarly the SCUD removal appeared to level off and no further

removal was observed when the SCUD had reached a value of 90 mg/1, for a wastewater

having an SCOD0 of 180 mg/1 (Figure 6.1).

A comparison of the average SCOD removal efficiencies for Sacriston and Durham

wastewaters at 10°C indicates that, on average, a 34 mg/1 (10%) increase in SCOD 0 and a

277 mg/1 (84%) increase in SS 0 increased the amount of SCUD removal for Durham

wastewater by 86 and 125%, over a retention period of 4 and 8 hours respectively. From

these results, it appears that the SS° (approximating to initial biomass concentration) plays

a major role in the removal of SCUD at 10°C.

Figure 6.8 illustrates the changes in average SCUD removals versus average SS ° and

SCOD0 values for different wastewaters at 20°C. The results shown in Figure 6.8 indicate

that, in general, the SCUD removal in wastewater increases with an increase in SS ° or

SCOD0 of the wastewater at 20°C. For the Broomhaugh wastewater, the SCUD removal

over 8 hours varied from 75-100 mg/1, when the SCODo and SS 0 were in the range of

155-200 and 270-410 mg/1, respectively. The corresponding SCUD removal however,

decreased to 15 mg/1 for a dilute wastewater having an SCOD 0 and SS ° of 70 and 160

mg/1, respectively (Figure 6.2).

CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 88

700 -
p SS-0 • SCOD-0
600 -
• RSCOD-4 0 RSCOD-8

--,30. 500 -
.—

T
-5: 300
0
c9 200
C...)
100 -

0 ,
Broomhaugh Sacriston Durham
WWTP

Figure 6.8-The effect of SS° and SCOD0 on SCOD removal at 20°C. Average of n

samples (Table 6.4) are plotted along with standard deviation (SD).

In order to delineate the effects of SCOD 0 and SS () on SCOD removal at 20°C, the SCOD

removals for 27 wastewaters over a retention period of 1 to 8 hours were analysed using

multiple regression analysis. Appendix C-1 shows how the data was arranged to be used

for regression analysis. Computer outputs exhibiting the regression equations for SCOD

removals and their summary statistics including the estimated coefficients for the

independent variables involved, their standard deviations and t-ratios, are shown in

Appendix C-2. An inspection of the summary statistics of the regression analyses revealed

that at a significance level of 5%, the SCUD removal was significantly affected (p < 0.05)

by SS 0 over a retention period of 1 to 3 hours. For higher retention periods (  4 hours), the

SCOD removal was significantly affected by both the SS 0 and the SCOD0 of the

wastewater.
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 89

Based on all the SCOD data collected at 20°C, the following empirical relationship for

determining the amount of SCOD removal at a given retention period under aerobic

gravity transport, was formulated

RSCODt = -74.4 + 0.156 SS0 + 0.146 SCOD0 + 12.9 t (R2 = 0.8) (6.1)

Where: RSCODt = SCOD removal (mg/1) over a retention period oft (1-8) hours, and SSo

and SCOD0 are, respectively, influent SS and SCOD concentrations (mg/1), and R 2 =

correlation coefficient.

A statistical analysis of the results of 7 wastewaters investigated at 30°C showed that the

SCOD removal over the first three hours would follow a trend similar to that at 20°C. The

SCOD removal after a retention period of 4 hours however, was in general significantly

affected only by SCOD 0 concentration.

6.3.5. Temperature dependency of SCOD removal from the wastewater

Higher removal rates are associated with higher temperatures due to the increased

metabolic activity at high temperatures. Generally the bacterial reaction rate

approximately doubles for each 10°C rise in temperature (Newcombe et al., 1977). To

estimate the temperature sensitivity of the SCOD removal rates in the wastewater,

experiments were run in parallel on the same wastewater at (i) 10° and 20°C (Figure 6.9),

and (ii) 15° and 30°C (Figure 6.10). A total of 8 wastewaters (4 at 10°C and 20°C, and 4

at 15° and 30°C) were subjected to batch tests.

CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 90

o 2 4 6 8
Retention time (hours)

Figure 6.9-Comparison of SCOD removal in the wastewater at 10 0 and 20°C

(n = 4, SS 0=540±250 mg/I).

400
380
360 _
340
.---,
-6b 320 _
E
300 _
M
0 280
C..)
Cl) 260

240 _
220 _
200

0 1 2 3 4 5 6

Retention time (hours)

Figure 6.10-Comparison of SCOD removal in the wastewater at 15° and 30°C

(n=4, SS 0=350±83 mg/I).

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 91

The average first order SCOD removal rate coefficients were used to calculate the

temperature rate coefficients from the van't Hoff-Arrhenius relationship which is

expressed by the following equation (Benfield and Randall, 1980) :

K2 = K i 0
(12-T1)
(6.2)

Where K 1 and K2 are the first order reaction rate coefficients at temperatures T 1 and T2,

respectively while 0 is the temperature coefficient.

From the first order SCOD removal rates, K 1=0.039 Ii i at 10°C, and K 1 = 0.091 Iii at

20°C (Figure 6.9), 0 was found to be 1.088. The K 1 at 15° and 30°C (Figure 6.10)

amounted to 0.0352 and 0.0799 If% respectively. The corresponding value of 0 was found

to be 1.056.

6.4. TOTAL COD REMOVAL

To estimate the changes in total COD (TCOD) in the wastewater during simulated aerobic

gravity transport, the TCOD of unsettled wastewater samples was determined before and

after 8 hours of oxygen addition. The average TCOD removal efficiencies over a retention

period of 8 hours are summarised in Table 6.10.

The average TCOD removal over 8 hours amounted to 50 mg/1 (SD=12, n=4) at 10°C,

when the influent SS was more than 200 mg/l. For samples having a SS concentration of

less than 200 mg/1, one wastewater did not show any change in TCOD over 8 hours while

another exhibited an 11% decrease over 8 hours.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 92

Table 6.10-Average TCOD removal efficiencies*

Temperature Influent SS Influent TCOD TCOD removal

°C (mg/1) (mg/1) (%)

10 468±166 1140±335 5±3 4

20 431±153 871±250 15±6 14

30 428±185 890±349 15±4 5

Note: * : average results for samples showing a decrease in TCOD over 8 hours

The TCOD removal over 8 hours at 30°C averaged 136 mg/1 (SD=74, n=5). The

corresponding TCOD removal at 20°C amounted to 110 mg/1 (SD=51, n=14). At 20°C,

one wastewater did not demonstrate any change in TCOD over 8 hours, while the effluent

TCOD of two samples was slightly higher than the corresponding influent. At all the

temperatures investigated, the TCOD removal over 8 hours did not appear to show any

relation to the influent TCOD and SS concentrations for the range investigated in this

study.

6.5. EFFECT OF OXYGEN ADDITION ON SS IN THE WASTEWATER

To study the changes in SS concentration (biomass concentration) in the wastewater

during simulated aerobic gravity transport, the SSs of the wastewater were measured

before and after six hours of oxygen addition in the batch reactor. The average changes in

SS of the wastewater, as a result of oxygen addition for 6 hours, are given in Table 6.11.
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 93

Table 6.11-Average changes in SS of wastewater

Temperature Influent SCOD Influent SS Change in SS

° C (m g/1) (mg/1) (%)

10 243±126 374±187 5±7 10

20 271±86 417±153 5±11 25

30 245±116 369±174 10±12 8

A portion of the effluent from the batch reactor, after a retention period of 6 hours was

settled in a one-litre cylinder for one hour and the SS of the supernatant was determined.

The corresponding SS removal efficiencies at different temperatures are summarised in

Table 6.12. The results shown in Table 6.12 indicate that the SS removal efficiencies in

the bench-scale settling column are significantly higher than those achieved in a typical

primary sedimentation tank (50-70%), thus reflecting the improved primary settleability

due to aerobic conditions during the simulated gravity transport.

Table 6.12-SS removal efficiencies

Temperature Effluent SS Supernatant SS SS removal

°C (mg/1) (mg/1) (%)

10 444±224 96±49 77±6 7
20 406±216 73±39 80±10 7
30 343±192 79±53 78±4 6

From Table 6.11, it can be seen that the changes in SS (biomass growth) as a result of

oxygen addition showed a wide variation. All but one wastewater collected from Durham

WWTP showed an increase in SS at all the temperatures investigated. The wastewaters

collected from other WWTPs however, did not always display a similar trend. Out of 10
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 94

wastewaters investigated for changes in SS at 10°C, 30% either did not exhibit any

change in SS or showed a decrease in SS as a result of oxygen addition for 6 hours. The

corresponding figures at 20 and 30°C were 29% (7 out of 27) and 25% (2 out of 8). The

overall decrease in SS ranged from 3-28%. The change in SS of the wastewater did not

appear to show any obvious relation to the changes in TCOD. It appeared however that, in

general, dilute wastewaters having low SCODs and low SS (especially those collected

from Broomhaugh) were more likely to exhibit a decrease in SS as a result of oxygen

addition. The VSS of the wastewater also followed a trend similar to the SS with regard to

changes due to oxygen addition.

For the wastewaters which showed an increase in SS (or VSS) over 6 hours, the yield

coefficient, calculated from changes in VSS of the wastewaters and the corresponding

SCOD removals (Appendex D-1) varied from 0.1-0.35 at 10°C, 0.1-0.72 at 20°C and 0.1-

0.5 mg VSS/mg SCOD at 30°C. The value of the yield coefficient averaged 0.3 mg

VSS/mg SCOD (SD=0.17, n=15) at 20°C. This value is lower than the literature values of

the yield coefficient of 0.4-0.8 Mg VSS/mg SCOD removed for the activated sludge

process (Metcalf and Eddy, 1991).

6.6. OXYGEN UPTAKE RATE

To estimate the oxygen uptake rate of the suspended biomass present in the wastewater

(SOUR), samples of the oxygenated wastewater were taken from the batch reactor at

specified intervals of time and were subjected to SOUR tests (Section 4.3.1). Figures

6.11-6.15 show the variation in SOUR of the wastewaters at the different temperatures

investigated.
CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 95

Retention time (hours)

Figure 6.11- Variation in SOUR of wastewater at 10°C (SCOD0 and SS ° are expressed
in mg/I).

SCOD0 SS0
175 300
215 340
160 420
70 160

0 1 2 3 4 5 6 7
Retention time (hours)

Figure 6.12- Variation in SOUR of the Broomhaugh wastewater at 20°C.

CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 96

18

16 _
SCOD 0 SS0
14 _ 390 380
• 12 385 400
.-; .
-0- 375 310
10-

dfA rg
--X- 365 500
• 8

O 6
k. -o-
-s-
300
305
470
240
4

0
_

•Air -o-
-e-
255
290
280
270

0 1 2 3 4 5 6 7 8

Retention time (hours)

Figure 6.13-Variation in SOUR of the Sacriston wastewater at 20°C.

25

SCOD0 SSD
20 _ 250 540
- 8- 315 740

'.7--
-7
013 15 _ 350 580
- 300 550
- m- 310 640
10
0
C/D

5 _

0

0 2 3 4 5 6 7 8

Retention time (hours)

Figure 6.14-Variation in SOUR of the Durham wastewater at 20°C.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 97

45

40 _ SCOD0 SS0
35 _ 311 320
139 300
30 -
189 280
25 -
= 398 700
5 20 387 245
015 84 165

10

5-

0 2 3 4 5 6 7 8
Retention time (hours)

Figure 6.15-Variation in SOUR of domestic wastewater at 30°C.

From the data presented in Figures 6.11-6.15, the following three distinct types of time

profiles of SOUR could be identified:

(i) Type 1: SOUR is low initially, increases to a peak value and then declines,

(ii) Type 2: SOUR is at a maximum initially, then declines with the passage of time,

and

(iii) Type 3: the SOUR does not show any general trend with respect to time.

When compared to the samples collected from Broomhaugh and Durham WWTPs which

displayed both Type-2 and Type-3 profiles, most of the samples taken from Sacriston

appeared to show a regular trend (Type-1 profile) with respect to time. The initial (after

about 1 hour from the start of the test) and peak values of SOUR at 20°C ranged from 2-8

and 8-17 mg/l.h, respectively. The corresponding value of SOUR after a retention period

of about 6 hours varied from 1-6 mg/l.h. The initial SOUR value at 30°C was found to be

about 13 mg/1, while peak and final values varied from 28-42, and 8-12 mg/1.h,
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 98

respectively. The retention time at which the peak value of SOUR occurred varied from 2-

5 hours.

As can be seen from Figures 6.11-6.15, the SOUR values for wastewaters collected from

individual WWTPs varied widely, and did not appear to show any clear relation with the

corresponding SS0 or SCOD0 concentrations. The SOUR values however, were quite low

when the SCOD0 of the wastewater was _110, 70, and  189 mg/1 at 10, 20 and 30°C,

respectively. This indicates that, at a given temperature, dilute wastewaters (having

SCOD0 concentrations similar to the above) are likely to exhibit lower SOUR values.

Maximum SOUR values of 13, 25 and 40 mg/l.h were observed at 10, 20 and 30°C,

respectively, indicating that the SOUR of the wastewater would be significantly affected

by temperature.

6.7. DISCUSSION

6.7.1. SCOD removal

The results of the present study clearly demonstrate that, by maintaining aerobic

conditions in sewers, a significant degree of soluble organic matter could be achieved.

These results support the conclusions of previous field investigations (Pomeroy and

Parkhurst, 1972; Newcombe et al., 1979) that sewers could be used to effect at least partial

treatment of wastewater during transit.

The results presented herein indicate that the removal of soluble organic matter in

wastewater in sewers would be strongly affected by the strength of the wastewater. At a

given temperature, a wastewater with a high initial SS (SS 0) and SCOD (SCOD0) was
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 99

found to achieve a high degree of SCOD removal while compared to a wastewater having

low SCOD0 and SSo. The results also demonstrate that, if the sewers are to be used as an

aerobic treatment facility, the effluent quality of such a system may fluctuate greatly,

depending upon the diurnal and seasonal variations in the influent concentrations.

During the first 3 to 4 hours from the start of the batch experiments, the SCUD removal

rates at 20°C and 30°C were found to be greatly affected by the SS ° concentration and to

be independent of the SCOD 0 concentration. A possible explanation for this phenomenon

may be that during the first few hours of oxygen addition, the surface of the bacteria is

completely saturated with substrate and all enzymes are in a complexed state (Sundstrom

and Klei, 1979). An increase in SCUD removal with an increase in initial SS concentration

of the wastewater was expected due to the high number of micro-organisms present in the

wastewater which were able to metabolise the organic matter. The dependency of the

SCUD concentration on SCUD removal during the latter stages of the experiments

confirms the suggestion of Benfield and Randall (1980) that at low substrate

concentrations in batch cultures, the rate of substrate utilisation would be directly

proportional to the substrate concentration, because very little substrate is available to

complex with the enzymes.

The removal of soluble organic matter from the wastewater flowing in the sewers, while

taking into account the effect of changes in wastewater strength, has not been reported

previously. Nevertheless, the findings of the present study are in reasonable agreement

with those few reported in the literature. Lokkeggard et al. (1995) reported a soluble COD

removal of 35 mg/1 over 8 hours, for a wastewater having a SCOD 0 and SS 0 of 140 and

160 mg/1, respectively, at 17°C. The present study noted a similar trend for dilute
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 100

wastewaters. Henze (1992) based on laboratory studies, reported a soluble COD removal

of about 50 mg/1 over a retention period of 3 hours, for a freshly produced wastewater

having a SCOD 0 of about 175 mg/l. This removal rate is of the same order of the

magnitude as that found in present study for a wastewater having a SCOD 0 of 175 mg/1

and a SS 0 of 300 mg/1 (Figure 6.2).

The SCOD results (Section 6.3) showed that the SCOD removal in the wastewater was at

a maximum during the initial hours of oxygen addition. Of all the wastewaters

investigated at 20°C, 89 percent showed a maximum SCOD removal during the first hour

of oxygen addition, while the corresponding time for the remaining 11 percent was 2

hours. The SCOD removal rate during the first 2 hours averaged 19, 25 and 47 mg/1.h for

Broomhaugh, Sacriston and Durham watewaters, respectively. The corresponding rate

during the next 6 hours averaged 10, 15 and 15 mg/l.h, respectively. These results are in

good agreement with those reported by Lokkegaard et al. (1995). They, based on batch

studies undertaken to investigate the kinetics of the wastewater, found that the readily

biodegradable organic matter in most of the wastewaters studied was depleted during the

first two hours under aerobic conditions.

A rapid decrease in SCOD during the first few hours could be due to the high degradation

rate associated with readily biodegradable organic matter (Henze, 1992) while the

subsequent decrease in SCOD removal rate may be attributed to the hydrolysis of slowly

biodegradable organic matter (Benfield and Randall, 1980; Raunkjaer et al. 1993). Henze

(1992) reported a biodegradable SCOD fraction equal to 35% of the total SCOD for a

wastewater with a SCOD 0 of 170 mg/l. For a similar wastewater, Lokkegaard et al. (1995)

reported the readily biodegradable fraction to be 18% of the initial SCOD. Lesouef et al.
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 101

(1992) have reported the readily biodegradable fraction to be 25 and 30% for two

wastewaters having total CODs of 400 mg/l. In the present study, an average SCOD

removal of 23 % (SD=7) was observed at 20°C for Broomhaugh wastewater (Table 6.4)

having an average SCOD of 176 mg/1 (SD=17). This removal efficiency is reasonably

consistent with the ranges of the readily biodegradable fractions of SCOD reported in the

above studies.

When the change in SCOD over 8 hours for individual wastewaters investigated at 10°-

30°C was described as zero- and first-order reactions (Appendix C-3), the first-order fits

yielded higher correlation coefficients than the corresponding zero-order values. This

suggests that the changes in SCOD in the wastewater during simulated aerobic gravity

transport would be best described by first-order kinetics (Figure 6.6). This observation is

in agreement with other published results for the activated sludge processes (Benfield and

Randall, 1980; Argaman, 1991).

The results of the current study suggest that the retention time required to achieve a certain

degree of soluble organic matter removal in the wastewater under aerobic conditions

would depend upon the strength of the wastewater (SCOD 0 and SS0) and the temperature.

This is in contrast to the observation by Shaw (1981) that a retention period of about 20

hours would be required to achieve a high degree of self-purification in sewers.

Ozer and Kasigra (1995) pointed out that, when compared with low substrate

concentrations, higher substrate concentrations in wastewater would result in higher

reaction and respiration rates, which would eventually result in the adequacy of shorter

lengths of sewers to achieve an equal decrease in concentration. The present study also
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 102

demonstrated this effect. For Durham wastewater (having a high SCOD 0 and SS0), the

SCUD removal over a retention period of 1 hour averaged 64 mg/I, at 20°C. The

corresponding retention time required to achieve a similar amount of SCUD removal for

Broomhaugh wastewater (low SS 0 and low SCOD0) was found to be 6 hours.

From Figures 6.1-6.5 it can be seen that a few wastewaters, especially those collected from

Broomhaugh (Figure 6.2) showed a slight increase in SCOD over the course of the time,

thus offsetting to some extent the early decrease in SCOD, and producing a saw-toothed

profile of SCOD variation over time. A similar effect has been reported by Raunkjaer et

al. (1995). They, while measuring the dissolved COD of raw wastewater at four stations

along a gravity sewer, found that a few samples showed an increase in dissolved COD

between two intermediate sampling stations.

A possible reason for the increase in SCOD mentioned above could be the contribution of

the SCUD as a result of hydrolysis of the slowly biodegradable fraction. When compared

with the readily biodegradable fraction of SCUD, which can pass directly through the cell

wall for synthesis and oxidative metabolism by organism, the slowly biodegradable

substrate is removed from the wastewater by adsorption onto the floc surface and by

enmeshment in the floc structure. This slowly biodegradable material is then hydrolysed

thus causing an increase in soluble COD (Benfield and Randall, 1980).

For Durham wastewater (which had a high SS 0 and high SCOW the SBOD 5 removal

efficiency over a retention period of 8 hours averaged 64%, 81% and 87% at 10°, 20° and

30°C, respectively. When compared with Broomhaugh and Sacriston wastewaters, a high
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 103

removal efficiency for Durham wastewater at all temperatures studied suggests that by

increasing the SS° concentration of the wastewater in sewers (by seeding it with activated

sludge), a removal efficiency comparable to that of secondary treatment may be achieved.

The temperature dependency of the SCOD removal rate in the raw wastewater does not

seem to have been reported previously. The values of temperature coefficients obtained in

the present study (1.088 for 10°-20°C and 1.056 for 15°-30°C) however, compare

favourably to those for normal aerobic biological processes reported in the literature.

According to Tucek et al. (1971), values of temperature coefficient varying from 1.035-

1.109 (and even more in the lower temperature region) have been confirmed by other

researchers. Benfield and Randall (1980) have reported temperature coefficients of 1.135

and 1.056 for a temperature range of 4-20°C and 20-30°C, respectively.

6.7.2. Total COD removal

The results shown in Table 6.10 indicate that, when compared with SCOD removal, the

TCOD removal in the wastewater during aerobic transport would be much lower. Previous

studies have reported a similar effect. Based on field studies, Newcombe et al. (1979)

reported that during aerobic transport in a pressure main, the changes in total BOD 5 were

not significant. Greenfield et al. (1980) also reported similar results. Bentzen et al. (1995)

found that controlled dosing of nitrates to a rising main to control septicity resulted in an

increased removal of soluble organic matter, but the corresponding changes in total BOD5

were not significant. The above mentioned studies argued that the concentration of total

organic matter during the aerobic transport would not change significantly due to the
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 104

transformation of soluble organic matter into additional biomass which would contribute

to total organic matter.

Henze (1992) reported a total COD removal pattern in the wastewater similar to that of

SCOD removal. From the results of the present study, there was no evidence to suggest

that the TCOD removal in the wastewater would display a removal trend similar to that of

the SCOD.

6.7.3. SS results

The SS results (Section 6.5) showed that, in general, the addition of oxygen to the

wastewater for 6 hours resulted in an increase in SS (or VSS) concentration. This increase,

however fluctuated considerably varying from negligible to 30%, and did not show any

clear relation with the initial SCOD to initial VSS ratio. The corresponding yield

coefficient varied widely and was, in general, significantly lower than typical values for

aerobic processes.

Unexpectedly, 12 out of 45 wastewaters studied within a temperature range of 10°-30°C,

either showed a decrease in SS or did not exhibit any change at all. The exact reason for

this phenomenon could not be explained. It may be that the shear force produced by the

impeller and by the bubbling of oxygen in the reactor had caused some of the solids to

dissolve in the wastewater. A decrease in the SS content of the wastewater due to

attachment of SS to the walls of the reactor was ruled out as such a phenomenon was not

observed.
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 105

A low growth or a zero change in biomass concentration observed in batch reactors could

also be explained by the ratio of initial substrate (SCOD0 or So) to initial biomass

concentration (X0) i.e. S 0/X0 in the batch reactor. The S 0/X0 or F/M (food/micro-organism

or VSS) ratio in the present study varied from 0.46-1.6 mg VSS/mg SCOD. According to

Chudoba et al. (1992), the S 0/X0 ratio is the most important factor effecting the growth

rate in batch reactors as it determines whether or not cell multiplication will take place. At

low S 0/X0 ratios, the initial energy level is low, and the increase in cell mass reflects the

increase in molecular polymer content in the biomass. Consequently, the weight changes

may not reflect similar changes in cell number. Based on batch experiments involving a

mixture of peptone and glucose, Chudoba et al. (1992) noted that at So/X0 ratios of less

than 3, the substrates were removed linearly, indicating no cell multiplication. In another

study (Chudoba, 1985), it was indicated that in batch cultivation of activated sludge,

significant growth would occur only when S 0/X0 is 10 or more.

The results with respect to changes in SS (Table 6.11) agree with those of many previous

studies. Newcombe et al. (1979) noted that no significant changes in SS occurred as a

result of the addition of oxygen to wastewater in an 8.2 km long pressure main. They

found that over a retention period of 2.4-2.8 hours, the SS of the wastewater increased by

1% in a temperature range of 10-12°C and decreased by 2-3% when the temperature

varied from 19-20°C. Greenfield et al. (1980) observed a two to five-fold increase in the

number of viable organisms over 8 hours as a result of the addition of oxygen to raw

wastewater in a batch reactor, although the corresponding changes in SS did not show any

clear pattern. Barber et al. (1980) reported that the addition of oxygen to wastewater in

batch reactors resulted in an increase in SS concentration together with higher bacterial

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 106

plate counts. They however, noted that a higher number of micro-organisms would

account for only a minor fraction of the increase in SS.

The findings of the present study however, are in contrast to the results of Bentzen et al.

(1995) who reported a 54% increase in SS as a result of the addition of nitrates to a rising

main having a retention period of 5.4 hours in a temperature range of 12-22°C. They

attributed the increase in SS to the transformation of soluble organic matter into a

suspended component. As compared with the results of Bentzen et al. (1995), the

maximum increase in SS observed in the present study amounted to 11, 19 and 30% at 10,

20 and 30°C, respectively.

6.7.4. Oxygen uptake rate

A wide variation in SOUR values of the individual wastewaters investigated in the present

study may be attributed to the differences in nature and origin of the wastewaters as well

as the diurnal changes in transportation time (age of the wastewater), and the prevailing

redox values and climatic conditions in sewers.

Previous studies have also reported a great variation in SOUR. Boon and Lister (1975)

reported SOUR values of 10-18 mg/l.h for domestic wastewater at a temperature of 14-

16°C. The total COD of the wastewater varied from 180-1470 mg/l. Barber et al. (1980)

reported that the results of SOUR tests were not reproducible and values ranged from 8-18

mg/1 with an average of 13 mg/1.h (at 25.5°C ). Pescod and Price (1981) reported that the

SOUR values of the wastewater entering a treatment plant were found to vary with time of

the day, and ranged from 1.5-19.9 mg/1.h.

 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 107

The SOUR of a domestic wastewater has been found to vary with the age of the

wastewater. The SOUR of fresh domestic wastewater under aerobic conditions is low near

its origin (2-3 mg/l.h at 15°C), which may increase to a peak value of as high as 20 mg/l.h

as the wastewater ages during transportation in sewer, and then declines (Section 2.3.1).

Considering the above and the fact that the wastewater samples used in the current study

were collected from the inlet to the WWTPs, different SOUR profiles were expected

because the age of the individual wastewater samples was likely to change greatly due to

changes in the retention time in sewers caused by flow variations.

The majority of wastewaters collected from Sacriston exhibited Type-1 profiles which are

identical to the SOUR profiles for young wastewaters reported by Pomeroy and Parkhurst

(1972) and Newcombe et al. (1979). These results indicate that when compared with

wastewaters collected from other WWTPs, the samples taken from Sacriston were

relatively young. On the other hand, samples which displayed Type-2 profiles could be

relatively aged and might have passed the peak SOUR during transportation to the

WWTP. Pomeroy and Parkhurst (1972) also reported SOUR profile, identical to the Type-

2 profile observed in the current study, for a wastewater which was collected after being

transported in a sewer for about 3.5 hours.

From Figures 6.11 and 6.15, it can be seen that several wastewaters exhibited pronounced

secondary SOUR peaks after high initial SOUR values. Similar results were reported by

Pomeroy and Parkhurst (1972) and Marwood (1984). Marwood (1984) attributed the

initial high SOUR of the oxygenated, previously anaerobic wastewater, to rapid oxidation

of sulphide and other minor chemicals and also to biological uptake. He suggested that the
 CHAPTER 6: ADDITION OF OXYGEN TO RAW WASTEWATER 108

rapid increase in SOUR after the high initial SOUR was due to acclimatisation of

facultative aerobic bacteria.

A possible reason for the secondary peaks in SOUR may also be the increase in oxygen

demand due to the release of readily biodegradable SCOD to the reactor as a result of the

hydrolysis of particulate organic matter (Section 6.7.1). A comparison of the SOUR and

SCOD profiles of the wastewaters (Figures 6.1-6.5) indicated that, in general, an increase

in SOUR of the wastewater was consistent with an increase in SCOD in the reactor.

Wastewaters which produced a zig-zag SOUR profile, also generally exhibited a similar

trend with respect to SCOD, thus supporting the hypothesis that the secondary peaks were

possibly caused by hydrolysis of the particulate organic matter.

As already mentioned, the SOUR values did not relate clearly with initial SS or SCOD

values of the wastewater. This is in agreement with the findings of Pomeroy and Parkhurst

(1972), who noted that 1-hour SOUR of wastewater did not bear any relationship to the

BOD 1 and BOD 5 of the wastewater. Boon and Lister (1975) and Barber et al. (1980) have

also reported SOUR results which did not show any clear relation with the total COD or

BOD of the wastewater. The SOUR data presented by Boon and Lister (1975), when

regressed by this author, against the corresponding total COD values, yielded a correlation

coefficient of 12%. Similarly the correlation coefficient between the SOUR and BOD5

data reported by Barber et al. (1980) amounted to 2.7%. The poor correlation of SOUR

with COD or BOD 5 of the wastewater implies that there are other factors (Section 2.3.1)

which also affect the SOUR of the wastewater.

 109

CHAPTER SEVEN

ADDITION OF OXYGEN TO SEEDED WASTEWATER

7.1. INTRODUCTION

The results of the experimental work described in Chapters 5 and 6 have clearly

demonstrated that by maintaining aerobic conditions in sewers, the indigenous suspended

biomass present in the flowing wastewater could be exploited to achieve a significant

degree of in-sewer treatment so that wastewater would reach the end of the sewer at least

partially treated. In cases, where sufficient retention time is not available however, or

where a high degree of in-sewer treatment is required, it has been suggested that active

biomass be added to the sewer to enhance the in-line purification (Koch and Zandi, 1973;

Pomeroy and Lofy, 1977; Shaw, 1981).

The work described in this chapter was designed to explore the possibility of enhancing in-

sewer aerobic treatment by the addition of the activated sludge at the inlet of the sewer. To

simulate an aerobic linear gravity sewer with activated sludge being added at the inlet, raw

wastewater was added with a specified concentration of activated sludge (Equation 4.1),

and maintained aerobic in a batch reactor at 20°C and at atmospheric pressure. The

experimental methodology used for this study has been described in detail in Section 4.3.1.

7.2. CHARACTERISTICS OF WASTEWATER AND SEED SLUDGE

The removal of soluble organic matter during simulated aerobic transport was studied at

five different initial seed (activated sludge) concentrations i.e. 100, 250, 500, 750 and 1000

mg VSS/1. A minimum of six wastewaters were investigated at each seed concentration.

The wastewater used in this study was collected from the inlet to Sacriston WWTP while
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 110

the returned activated sludge used for seeding purposes was collected from Durham

WWTP. The average characteristics of the wastewaters studied at different seed

concentrations are shown in Table 7.1. Due to large number of samples involved, it was

not possible to undertake BOD 5 analyses for all the experimental runs. Table 7.2 shows the

average characteristics of the wastewaters for which BOD 5 analyses were carried out. The

average characteristics of seed sludge are summarised in Table 7.3.

Table 7.1-Average characteristics of wastewater

Seed concentration Wastewater characteristics (mg/1) Initial SCUD in

(mg VSS/l) reactor (mg/1)
SS SCUD

100 477±126* 303±69 298±67 9

250 477±126 303±69 289±61 9
500 411+73 324±67 298±60 6
750 382±80 323±67 300±53 6
1000 548±69 339±101 312±79 6
Notes: (i)* average±SD
(ii) SCOD=Soluble COD
(iii) n=number of wastewaters studied

Table 7.2-Average characteristics of wastewaters for which BOD 5 tests were carried
out

Seed concentration SS Initial SBOD5 in

(mg VS S/1) reactor (mg/1)
(mg/1)

100 486±170* 139±29 5

250 486±170 133±27 5
500 403±95 152±18 3
750 375±96 138±17 4
1000 548±69 137±40 6
Note: SBOD 5 = Soluble BOD5

CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 111

Table 7.3-Average characteristics of seed sludge

Seed concentration Seed sludge characteristics (mg/1) II

(mg VSS/l)
VSS SCOD
100 3081±873 162±46 9
250 3081±873 162±46 9
500 3425±1214 183±50 6
750 3928±875 228±100 6
1000 4098±418 210±35 6

As a result of the addition to the wastewater of activated sludge which had a low SCUD,

the initial SCUD (or SBOD 5) in the batch reactor at the start of the test was lower than the

corresponding SCOD (or SBOD 5) of the raw wastewater, depending upon the volume of

activated sludge added to the reactor. The initial SCODs and soluble BOD 5 (SBOD5)

concentrations in the reactor, corresponding to each seed concentration, are also shown in

Table 7.1 and 7.2, respectively. The characteristics of the individual wastewaters

investigated at each seed concentration are shown in Appendix D-1. Although, when

compared with the experimental investigation described in Chapter 6 the wastewaters used

in this study were collected from a single WWTP in such a way as to minimise the effect

of variation in influent SS and SCUD on SCUD removal, it can be seen from Table 7.1 and

Appendix D-1 that the characteristics of the wastewater varied greatly during the study

period.
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 112

7.3. REMOVAL OF SOLUBLE ORGANIC MATTER

Figures 7.1 to 7.5 illustrate the variation in soluble COD (SCUD) of the seeded

wastewaters over time. The average SCUD and SBOD 5 removal efficiencies at different

seed concentrations are shown in Tables 7.4 and 7.5.

The soluble COD removal efficiencies at different seed concentrations over a retention

period of 8 hours and the corresponding "mg of SCUD removed per mg of total VSS

(summation of seed concentration and the VSS of the raw wastewater prior to the addition

of activated sludge)" are given in Table 7.6.

425 Influent SS (mg/I'

740 580 ._x___ 550
375 390 —0-390 -I-. 380
350 470 _0_440
325

-c-d 275

a) 225

0 175
CID

125

75

0 2 4 6 8

Retention time (hours)

Figure 7.1-Variation in SCOD of the wastewater at an initial added seed

concentration of 100 mg/1 at 20°C (Influent SS indicate the SS concentration of the

wastewater prior to the addition of activated sludge).

CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 113

Influent SS (mg/I)
_.- 740 .--580 --x-- 550
___Ne._ 390 -.- 3 90 --1— 380
—350 —e_ 470 -.- 440

0 2 4 6 8

Retention time (hours)

Figure 7.2-Variation in SCOD of the wastewater at an initial added seed

concentration of 250 mg/1 at 20°C.

Influent SS (mg/I)
.__.__ 400 _Ay_ 310 ___A_ 500
, 350 -*-470 _._440
,-

0cn

0 2 4 6 8
Retention time (hours)

Figure 7.3-Variation in SCOD of the wastewater at an initial added seed

concentration of 500 mg/I at 20°C.

•
CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 114

400

350

300 Influent SS (mg/J)

—4.-400 _6_310 _6_500

• 250
—*-- 350 440 290

O 200
C)
150

100 _

50

0 2 4 6 8

Retention time (hours)

Figure 7.4-Variation in SCOD of the wastewater at an initial added seed

concentration of 750 mg/1 at 20°C.

400 _

Influent SS (mg/I)

_4_680 530 __air_ 560

r. 250 - 520 __NE_ 520 480
=

O 200
C_)

50

0 2 4 6 8
Retention time (hours)

Figure 7.5-Variation in SCOD of the wastewater at an initial added seed

concentration of 1000 mg/1 at 20°C.
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 115

Table 7.4-Average SCOD removal efficiencies (%)

100 250 500 750 1000

-3t
1 25±9 36±10 42±13 50±11 54±6
2 33±10 44±9 51±8 56±8 63±3
3 40±9 52±7 56±5 61±5 65±4
4 45±7 56±6 59±3 62±2 63±5
6 54±9 62±6 66±4 66±3 67±5
8 61±8 66±7 67±4 68±6 67±6
Notes: (i) B=seed concentration (mg VSS/I)
(ii) t=retention time (hours)

Table 7.5-Average soluble BOD5 (SBOD5) removal efficiencies (%)

100 250 500 750 1000

'------------ t----3------„,
E

2 48±10 62±9 59±10 74±8 75±9

4 65±6 79±6 80±2 9215 86±5
6 79±6 90±4 89±1 93±3 90±5
8 84±5 92±5 94±3 95±3 94±3

Table 7.6- Substrate removal efficiencies after a retention period of 8 hours

Seed concentration RSCOD-8 (mg/1) mg RSCOD/mg VSS

100 181+51 0.47±0.12

250 192±50 0.3±0.08
500 200±43 0.23±0.05
750 205±48 0.19±0.05
1000 206±70 0.14±0.05
Note: RSCOD-8=SCOD removal after 8 hours
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 116

The food to micro-organisms (F/M) ratio, as computed on the basis of influent total VSS

(summation of VSS in the raw wastewater prior to the addition of the seed, and seed

concentration), averaged 0.6 (SD=0.2), 0.45 (SD=0.12), 0.35 (SD=0.07), 0.28 (SD=0.05),

0.21 (SD=0.05) g COD/g VSS at seed concentrations of 100, 250, 500, 750 and 1000 mg/1,

respectively. The corresponding F/M ratios, when the non-degradable fraction of raw

wastewater was neglected, averaged 0.7 (SD=0.01), 0.6 (SD=0.14), 0.42 (SD=0.09), 0.32

(SD=0.06), 0.24 (SD=0.06), respectively. For the above calculations, the non-

biodegradable fraction was assumed to be 40% of the VSS of the raw wastewater, as

suggested by McKinney (1970).

The effluent SCODs over 8 hours averaged 116 (SD=36), 97 (SD=25), 99 (SD=23) and 98

(SD=21) mg/1 at seed concentrations of 100, 250, 500 and 750 mg/1, respectively. The

corresponding effluent BOD5 values averaged 21 (SD=7), 10 (SD=7), 10 (SD=5), 9

(SD=4) and 8 (SD=3) mg/1, respectively.

When the variation in SCODs of the individual wastewaters over time (all the individual

curves shown in Figures 7.1-7.5) was described by the first and second order reactions

(Appendix C-3), the second order fits yielded higher correlation coefficients for all the

wastewaters investigated at seed concentrations of 100 and 250 mg/l. This suggests that

the SCUD removal at these seed concentrations, with respect to time, would be best

described by second order kinetics.

At 500 mg/1 seed concentration, the SCUD removal for all wastewaters but one displayed

second order kinetics (Correlation coefficient; R2 =0.9-0.97). For a wastewater which had
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 117

an initial SCUD (SCOW of 205 mg/I, the SCUD removal kinetics were second order

during the first 2 hours. The SCUD removal appeared to become asymptotic afterwards as

can be observed from Figure 7.3.

At 700 mg/1 seed concentration, 50% of the wastewaters investigated showed second order

kinetics with regard to SCUD removal over 8 hours. For the remaining 50% wastewaters,

the SCUD removal mechanism was second order only over first 2 hours, as very little

COD removal was subsequently observed (Figure 7.4).

At a seed concentration of 1000 mg/1, the correlation coefficients (R2) for second order

kinetics reaction rates, although higher than the corresponding first order values, were low

and varied from 0.2-0.84. The second order model fitting exercise showed that during the

first three hours, the SCUD degradation rate mainly followed second order kinetics with

the correlation coefficient varying from 0.8-0.92. Beyond 3 hours however, the SCUD

removal appeared to level off for most of the wastewaters investigated, and a little SCUD

removal was observed thereafter.

In order to empirically describe the changes in SCUD over time in seeded wastewater,

regression analysis was carried out on the data shown in Figures 7.1-7.5. The best-fit

relationship for the description of SCUD removal over time at an initial added seed

concentration of 100 mg/1 (all the data shown in Figure 7.1) is as follows:

SCOD, 2.65
= l loge on (R2= 0.84) (7.1)
SCOD0 (1+ t) •
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 118

where: SCODt = soluble COD (mg/1) in the seeded wastewater after a retention time t

(1-8 ) hours, and

SCOD0 = soluble COD (mg/1) in the seeded wastewater at the start of the oxygen

addition.

The SCOD variation over time at a seed concentration of 250-1000 mg VSS/1 (Figures 7.2-

7.5) was found to be described by the following best-fit relationship:

SCOD, a +b
(R2= as shown in Table 7.7) (7.2)
SCOD0 — (1+ t)

Where a and b are constants, with values for individual seed concentrations shown in

Table 7.7.

Table 7.7-Values of constants a and b (Equation 7.2) for different seed concentrations

Seed concentration (mg VSS/l) a b R2

250 0.72 0.28 0.86
500 0.74 0.24 0.92
750 0.75 0.21 0.91
1000 0.76 0.18 0.89

7.4. EFFECT OF SEED CONCENTRATION ON REMOVAL OF SOLUBLE

ORGANIC MATTER

The average effluent SCODs for different initial seed concentrations are plotted in Figure

7.6. For comparison purposes, the average results of all the unseeded wastewatdrs (Chapter

6) have also been included.


CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 119

Seed concentration (mg VSS/I)

100
_er_ 250 x 500
-*- 750 _o_ 1000

1 2 3 4 5 6 7 8
Retention time (hours)

Figure 7.6-The average variation in SCOD of seeded and unseeded wastewater.

From the average results shown in Figure 7.6, it can be seen that an increase in seed

concentration from 0-100 and again from 100-250 mg/1 led to a marked reduction in

SCOD over a retention period of 1 to 8 hours. The effect of increasing the seed

concentration beyond 250 mg/1 on SCOD removal however, appears to be dictated by the

retention period, and seems to be noticeable only when the retention period was about 3

hours.

In efforts to further delineate the effect of seed concentration on SCOD removal while

taking into account the variation in SCODs and SS concentrations of the individual

wastewaters prior to the addition of activated sludge (Table 7.1), multiple regression

analysis was used. For this purpose, the SCOD removal efficiencies (RSCOD) over a

retention period of 1 to 8 hours for all the wastewaters investigated at seed concentrations

of 100-1000 mg/1 (36 wastewaters) and those studied without any seed concentration (26
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 120

wastewaters; Chapter 6), were analysed together with their corresponding SS 0 and SCOD0

values. Appendix D-1 shows how the above data were arranged so that multiple regression

analysis could be carried out. The empirical models for SCOD removal efficiencies over a

retention period of 1 to 8 hours, obtained from multiple regression analysis, are shown in

Table 7.8.

Table 7.8-Empirical equations showing the effect of seed concentration, SCOD 0 and
SS 0 on SCOD removal (seed concentration = 0-1000 mg/I)

Retention time RSCOD Correlation

(hours) (mg/1) coefficient (R2)
1 -11.8 + 0.093 SS0 + 0.0655 SCOD 0 + 0.134 B 0.80
2 -31.8 + 0.104 SS° + 0.196 SCOD 0 + 0.129 B 0.81
3 -43.4 + 0.11 SS ° + 0.307 SCOD 0 + 0.113 B 0.81
4 -52.9 + 0.0999 SS° + 0.409 SCOD 0 + 0.095 B 0.81
6 -52 + 0.0914 SS° + 0.497 SCOD0 + 0.0821 B 0.82
8 -61.7+0.123 SS° + 0.553 SCOD0 + 0.0675 B 0.81
Notes: (i) SS 0 = SS concentration in the raw wastewater (mg/1) prior to the addition of activated sludge
(ii) SCOD 0=soluble COD in the reactor at the start of the experiment (mg/1)
(iii) B = seed concentration added to the reactor at the start of the experiment (mg VSS/1)
(iv) RSCOD = SCOD removal (mg/1)

From the equations presented in Table 7.8, it can be observed that fairly high correlation

coefficients (R2 0.8) were obtained. This indicates that the operating variables selected

for the prediction of the response i.e. RSCOD are adequate, and the above equations (Table

7.8) could be used to predict the effect of seed concentration on SCOD removal efficiency.

The empirical models shown in Table 7.8 were used to deduce the SCOD removal

efficiencies, at different seed concentrations, for a wasewater with SCOD 0 and SS°
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 121

concentrations of 300 and 450 mg/1, respectively. These values of SS 0 and SCOD0 were

assumed considering the average SCOD 0 and SS° concentrations of 300 and 450 mg/1

respectively, for 36 individual wastewaters investigated at a seed concentration of 100-

1000 mg/l. Based on the results of the empirical models (Table 7.8), theoretical curves

shown in Figure 7.7 were generated to illustrate how the seed concentration would affect

the SCUD removal at constant SCOD 0 and SS 0 concentrations.

240
220
200
180
160
• 140
O
▪ 120
0 100
Cl) 80
60
40

0 200 400 600 800 1000
Seed concentration (mg VS S/1)

Figure 7.7-Effect of seed (activated sludge) concentration on SCOD removal.

Based on the data shown in Figure 7.7, it appears that the SCUD removal in seeded

wastewater under aerobic conditions increases linearly with an increase in the seed

concentration in the range of 100-1000 mg/l. The effect of seed concentration on SCUD

removal however, apperas to diminish with an increase in the retention period, as indicated

by a decrease in the slope of lines shown in Figure 7.7.

 •
CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 122

Figure 7.8 shows the results presented in Figure 7.7 but re-plotted to show more clearly the

effect of retention time on SCOD removal at individual seed concentrations. The results

shown in Figure 7.8 reveal that the SCUD removal rate is at a maximum during the first

four hours for seed concentrations of 100-500 mg/1, and first two 3 hours for higher seed

concentrations. Thereafter, the SCUD removal rate is relatively low and appears to become

constant.

240
220
200
bi)
E 180
▪ 160
2 140
• 120

0 100

CI) 80
60
40

1 2 3 4 5 6 7 8
• Retention time (hours)

Figure 7.8-Effect of retention time on SCOD removal/

When multiple regression analysis was carried out on all the SCUD data shown in Figures

7.1-7.5, to statistically determine the effect of SCOD0, SS, retention time and seed

concentration on SCUD removal efficiency, the following empirical equation was

obtained:
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 123

RSCOD = - 110 + 0.479 SCOD0 + 0.0457 SS0 + 17.2 t + 11.2 b - 1. 11 bt (R2 = 0.77)

(7.3)

Where: RSCOD = SCOD removal (mg/1) after a retention time t (1-8) hours from the start

of the experiment, and

b = {l+seed concentration in mg per litre/100}.

The summary statistics of Equation 7.3, including the estimated coefficients for the

independent variables involved, their standard deviations and t-ratios, is shown in

Appendix D-2. An inspection of the summary statistics of the regression analysis revealed

that at a significance level of 5%, the SCOD removal in the seeded wastewater was

affected significantly by SCODo, SSo, retention time and seed concentration (in the range

of 100-1000 mg/1). The effect of SS 0 however, was found to be least significant.

Furthermore, the effect of interaction between seed concentration and retention time was

also found to be significant.

7.5. EFFECT OF OXYGEN ADDITION ON BIOMASS

To estimate the changes in the VSS (biomass) of the seeded wastewater (mixed liquor) as a

result of oxygen addition, the VSS of the mixed liquor were measured before, and after six

hours of oxygen addition in the batch reactor.

The VSS of the mixed liquor at all seed concentrations investigated, in general, showed a

decrease over 6 hours although a few wastewaters also showed a slight increase. The

overall change in mixed liquor volatile suspended solids (MLVSS) after six hours of

oxygen addition ranged from -12 to +23, -10 to 5, -18 to 9, -13 to 2, and -6 to -16%,
CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 124

respectively, at seed concentration of 100, 250, 500, 750 and 1000 mg/l. The MLVSS

results also showed that the decrease in MLVSS tended to increase with an increase in the

seed concentration.

7.6. EFFLUENT SETTLEABILITY

To evaluate the settling characteristics of the solids in the effluent at different seed

concentrations, a sample of mixed liquor after a retention period of 6 hours was taken from

the reactor and subjected to a SVI test. A sample of the mixed liquor was collected from

the batch reactor at the same time and settled in a bench-scale settling column for one hour

and the SS of the supernatant was determined. The results of SVI tests and the SS of the

supernatant from the settling column are plotted in Figure 7.9 and summarised in Table

7.9.

150
_e_ S VI (ml/g)
130 _
—0— Supernatant SS (mg/1)

110 .
cip
cA
L. 90 .
o
1.-;.,
CD
70 _

50 .

30

o 250 500 750 1000
Seed concentration (mg/1)

Figure 7.9-Effect of seed concentration on SVI and SS of the supernatant.

 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 125

Table 7.9- Settleability test results

Seed concentration SVI (mug) Supernatant SS (mg/1)

100 76±10* (9) 72±15 (5)
250 98±13 (9) 64±21 (5)
500 118±19 (6) 47±12 (6)
750 130±21 (6) 44±16 (6)
1000 149±13 (6) 47±13 (6)
Notes: (i) * average-±SD
(ii) the values in parenthesis indicate the number of experimental runs

The results shown in Table 7.9 and Figure 7.9 indicate that the SVI increases steadily with

an increase in seed concentration up to 1000 mg/1. The SS of the effluent from the settling

column however, appear to decrease gradually with an increase in seed concentration up to

500 mg/l. Beyond 500 mg/1, the effluent SS appears to be independent of the seed

concentration, as can be seen from Figure 7.9.

7.7. OXYGEN UPTAKE RATE

To estimate the variations in oxygen uptake rate of the seeded wastewater (SOUR),

samples of the oxygenated mixed liquor were taken from the batch reactor at specified

intervals of time and the SOUR was determined as described in Section 4.3.1. Figures

7.10-7.14 illustrate the variation in SOUR of the wastewater at seed concentrations of 100-

1000 mg/1, at 20°C.

 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 126

25
•

20 SCOD0 SS0
215 340
315 740
15 • 350 580
300 550
5 10 • 385 380
CID • --x— 210 350
-qv -.414:1,4100...
5- •
•
—o— 300 470
—x— 305 440
•

1 2 3 4 5 6 7
Retention time (hours)

Figure 7.10-SOUR profile of the wastewater at an initial added seed concentration of

100 mg/I (SCOD0 and SS () are expressed in mg/1).

30

SCOD0 SS0
25 _

—o— 215 340

,.."Es 20 —o— 315 740
—0— 350 580
- -NC- 300 550
—A— 185 390
—i— 385 380
--x— 210 350
—e— 300 470
5 305 440
—A—

1 2 3 4 5 6 7
Retention time (hours)

Figure 7.11-SOUR profile of the wastewater at an initial added seed concentration of

250 mg/l.
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 127

40

35 - SCOD0 SS°

30 _ --a— 385 400

--x--- 375 310
-'-: 25 _
z._, --s— 370 500
oi)
E 20 —13— 210 350
--6— 300 470
15 - —0— 305 440
0
cn
10 _

0
0 1 2 3 4 5 6 7
Retention time (hours)

Figure 7.12-SOUR profile of the wastewater at an initial added seed concentration of

500 mg/I.

Retention time (hours)

Figure 7.13-SOUR profile of the wastewater at an initial added seed concentration of

750 mg/i.
CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 128

140

120

100
--a:
--<'
t:in 80
e
5
0
v)
60

40 _

20

2 3 4 5 6 7 8
Retention time (hours)

Figure 7.14-SOUR profile of the wastewater at an initial added seed concentration of

1000 mg/i.

To evaluate the effect of SS of the wastewater prior to the addition of activated sludge, the

initial SCOD in the reactor, and the retention time, on the SOUR at different seed

concentrations, multiple regression analysis was performed on the data shown in Figures

7.10-7.14. The empirical equations describing the SOUR of the seeded wastewater over

time, and their summary statistics are shown in Appendix D-3.

From the empirical equations presented in Appendix D-3 and from their summary

statistics, it was inferred that:

(i) at a seed concentration of 100 mg/1, the SOUR of the wastewater did not appear to

show any significant correlation with SSo, SCOD 0 or the retention time. Also, none

of the above parameters were found to have a significant influence on SOUR.

(ii) at a seed concentration of 250, 500 and 750 mg/1, only the effect of retention time

on SOUR was found to be significant, and

 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 129

(iii) at a seed concentration of 1000 mg/1, the SOUR was significantly affected both by

SCOD0 and retention time.

7.8. DISCUSSION

7.8.1. Removal of soluble organic matter

The results of the present study demonstrate that, when compared with the soluble organic

matter removal in the unseeded wastewater (Chapter 6), the addition of activated sludge to

the wastewater, at a concentration as low as 100 mg VS 5/1, would result in a considerable

increase in soluble organic matter removal and hence a significant reduction in the time

required to achieve a given degree of in-sewer treatment. For unseeded wastewaters

(Chapter 6), the SCUD and SBOD5 removal efficiencies (for 24 wastewaters) over a

retention period of 8 hours averaged 48% and 66%, respectively. When compared to the

this, the SCUD and SBOD 5 removal efficiencies at a seed concentrations of 100 mg VSS/

averaged 54 and 79% respectively, over a retention period of 6 hours (Tables 7.4 and 7.5).

The average SCUD and SBOD 5 removal efficiencies at a seed concentration of 250 mg

VSS/1 amounted to 56 and 79%, respectively, over a retention period of 4 hours.

The results shown in Tables 7.4 and 7.5, and Figures 7.6-7.8 illustrate that the effect of an

increase in seed concentration on substrate removal efficiency tends to diminish with the

increase in retention time. Moreover, the improvement in SCUD removal efficiency as a

result of increasing the seed concentration was not in proportion to the concentration of

seed added. Over a retention period of 8 hours, for instance, an increase in seed

concentration of 300% (250 to 1000 mg/1), increased the SCUD removal (predicted from
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 130

Table 7.8) by 66, 34, 25 and 19%, when the retention period was 2, 4, 6 and 8 hours,

respectively. Similarly, increasing the seed concentration from 750 to 1000 mg/1 resulted

in an increase of 15, 9, 7 and 5% in SCUD removal, over a retention period of 2, 4, 6 and 8

hours, respectively.

The results of the present study with respect to the effect of seed concentration on SCUD

removal at higher retention periods support the findings of Green et al. (1985) who, based

on laboratory studies undertaken to simulate a step-fed sewer having a retention period of

10 hours, reported that decreasing the initial seed concentration by about 70% resulted in

only a small decrease (6%) in the dissolved COD removal.

A diminished effect of seed concentration on substrate removal efficiency with an increase

in retention time may be due to the fact that at higher seed concentrations, more active

biomass and more substrate is present initially, thus resulting in a considerably higher

SCUD removal during the first few hours. The removal rate subsequently decreases as

soon as the limiting substrate conditions are reached and thus little removal is observed

with an increase in retention period.

A relatively small improvement in SCOD removal efficiency with an increase in seed

concentration in the range of 250-1000 mg/l, at higher retention periods (Table 7.4 and

Figures 7.6-7.8) could also be explained by the sludge activity theory advanced by Tench

(1968, 1994). According to Tench (1994) the rate of oxidation of wastewater is

proportional to the number of viable micro-organisms in the activated sludge and the

concentration of the substance adsorbed on the floc surface. He, based on experimental
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 131

trials with full-scale activated sludge plants, noted that "the active biomass is a decreasing

proportion of the sludge as its concentration is increased". Tench (1994) also suggested

that there is an optimum sludge concentration for each plant and that the sludge could be

varied greatly without significantly affecting the plant treatment efficiency. An increase in

MLSS beyond the optimum concentration caused only a small reduction in effluent

quality. Based on full-scale experimental results, Tench (1994) developed an empirical

equation linking sludge viability and its concentration, which he used to demonstrate that a

100% increase in MLSS in the aeration tank would result in only about 10% increase in

the active biomass concentration of the sludge.

From the results shown in Tables 7.4 and 7.5, it can be seen that the soluble BOD 5 removal

efficiencies in excess of 90% were achieved over 8 hours, at seed concentrations of 250-

1000 mg/l. The BOD5 removal efficiency in a conventional activated sludge plant (ASP)

which is usually designed to operate at a mixed liquor suspended solids concentration of

1200-3500 mg/1, ranges from 85-95% (Horan, 1991). Considering the above, the substrate

removal efficiencies found in the current study are quite comparable to those found in a

typical ASP, inspite of considerably lower MLSS concentrations than those found in

conventional ASPs. According to Grutsch and Mallatt (1976) as quoted by Capps et al.

(1995), the critical MLSS concentration, the concentration of MISS below which the

activated sludge treatment becomes difficult, is about 700 mg/l. Shao et al. (1992) have

reported an excellent effluent quality from a high-rate activated sludge plant employing a

MLSS concentration of 950 mg/l. From the results shown in Tables 7.4 and 7.5, it appears

that, within the range of operating parameters investigated in the present study, a sewer

with a seed concentration of about 500-1000 mg/1 being added at its inlet may prove an
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 132

alternative to a conventional aeration tank. The optimum concentration of seed to be added

to an in-sewer treatment system to achieve a given degree of treatment however, would

depend upon the strength of the wastewater, the retention time available, temperature, and

the characteristics of the sewerage system.

High substrate removal efficiencies at low biomass concentrations , as is the case in the

present study (Table 7.6), could also be explained by the sludge activity theory as

described by Green and Shelef (1980). According to them, the sludge activity in a

conventional activated sludge plant is low (less than 20%), because of the low net growth

caused by the limiting substrate conditions in the reactor. On the other hand, an increase in

the food to micro-organism ratio (> 0.3 per day) causes higher net growth which results in

steadily increasing sludge viability. This high sludge viability therefore, compensates for

low MLVSS concentration, thus resulting in a high substrate removal.

According to Moser et al. (1977), tubular (pipe) reactors are likely to exhibit some

instability against step changes in inflow or concentration, which could be overcome by

the control of return sludge depending upon the BOD5 concentration of the influent. In the

present study, the effluent SCOD at seed concentrations of 100 mg/1 was found to fluctuate

with variation in influent concentrations similar to that observed for unseeded wastewaters

(Chapter 6). This instability in effluent concentration however, appeared to dampen out

with an increase in seed concentration as shown in Figure 7.15.

 •

133

riD
riD
—
tal)

8
0

— 0

cf,

C.
)

CC

CC

cO)

00

▪0

C.)

C.)

cZ

C:=1

riD

-r1

tr)
§
(v2111) OS
8 tr)
 134

Most of the studies undertaken to investigate the SCUD or BOD 5 removal kinetics in

wastewater treatment report that the substrate removal in an activated sludge plant follows

first order kinetics (Wolfbauer et al., 1978; Newbry eta!., 1988). Wolfbauer et al. (1978),

on the other hand, demonstrated that the process follows a zero order reaction. Contrary to

the finding of the above studies, the SCUD removal rate in the seeded wastewaters

investigated in the present study appeared to follow mainly second order kinetics. This

observation however, is in accordance with the results of Tucek and Chudoba (1969),

who, based on laboratory and field work, concluded that the course of BOD 5 removal in

batch or plug flow systems could be described by a second order reaction rate.

All the wastewaters studied at a seed concentration of 100-1000 mg/1 showed a maximum

SCUD removal during the first hour of oxygen addition. The SCUD removal over a

retention period of 1 hour averaged 25, 36, 42, 50 and 54% (Table 7.4) at a seed

concentration of 100, 250, 500,750 and 1000 mg/1, respectively. The corresponding

values at a retention period of 8 hours were found to be 61, 66, 67, 68, 67%, respectively.

A high SCUD removal rate,. initially, may be due to the oxidation of readily

biodegradable material (Braha and Hafner, 1986), and simultaneous physical adsorption

of particulate organics and biochemical absorption of soluble organics (Rickert and

Hunter, 1971).

The SCUD removal for wastewaters studied at a seed concentration of 1000 mg/1 was

found to approach nearly a stationary value after a retention period of 2-4 hours (Figure

7.5). The SCUD removal averaged 63 and 67% over a retention period of 2 and 8 hours,

respectively. These observations are consistent with those of Rickert and Hunter (1971),
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 135

who, while investigating the effect of aeration time on SCOD removal in a plug-flow

aeration tank, showed that soluble COD removal reached nearly a stable level after one

hour of aeration. They found that the soluble COD removal averaged 77% over 1 hour

and 78% over 6 hours. Stoyer and Scherfig (1972), based on a theoretical quantitative

comparison of completely mixed and plug flow reactors, demonstrated that the theoretical

time required to achieve a given degree of treatment in a plug flow reactor is only about

one-third of that required in a completely mixed system. They, based on pilot-scale tests

involving the addition of activated sludge to a pressure pipe, showed that most of the

organic matter in a plug-flow in-pipe treatment system was removed in about 1 hour.

From the SCUD profiles shown in Figure 7.5, it can be seen that at a seed concentration

of 1000 mg/1, the SCUD reduces up to 3 hours, then shows an increase and is again

reduced to a stationary value. This phenomenon may be attributed to the generation of

soluble microbial products (Germali et al., 1991). According to Boero et al. (1991), a

major fraction of the soluble organic carbon in biological wastewater effluents consists of

soluble microbial products, which may be either substrate utilisation associated products

(those resulting from intermediates or end products of substrate degradation, cell

metabolism, or cell growth) or biomass associated products which result from cell lysis

and decay.

7.8.2. Biomass growth

The change in biomass in the seeded wastewater, measured as the change in MLVSS after

a retention period of 6 hours (Section 7.5), varied widely and appeared to decrease with

respect to the MLVSS concentration in the reactor at the start of oxygen addition.
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 136

Considering the fact that a significant degree of SCOD removal occurred over 6 hours at

all the seed concentrations investigated (Table 7.4), a decrease in VSS as a result of

oxygen addition was unexpected. The results of the current study with regard to biomass

growth are not consistent with the sludge viability theories cited in previous section.

These results are also contrary to the findings of Green et al. (1985), who reported 3.1 and

5.2 fold increases in sludge viability, in a simulated step-fed plug flow sewer at initial

seed concentrations of 4760 and 1430 mg VSS/1, respectively.

As regards the changes in VSS as a result of oxygen addition, the unseeded wastewaters

(Section 6.7.3) exhibited a trend similar to that observed in the current study. The possible

explanations for this phenomenon were outlined in Section 6.7.3. In addition to the

reasons stated in Section 6.7.3, a probable reason for the decrease in VSS observed in the

present study may be related to the use of VSS for the measurement of biomass.

According to Benfield and Randall (1980) the VSS measurement for biomass "is suspect"

and ,"inconsistencies do arise when VSS is used as a measure of active biomass when the

raw wastewater contains a large fraction of non-biological suspended solids material", as

was the case in the current study.

It should be noted that the substrate concentrations used in the present study (average

SCOD=300 mg/1) were considerably lower than the substrate concentrations used by

Green et al. (1985) (SCOD=590-634 mg/1). Moreover, they used synthetic wastewater,

prepared by the addition of yeast to settled domestic wastewater, which may not truly

simulate the behaviour of the raw wastewater flowing in sewers. Also, the sludge viability
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 137

was measured using substrate removal and oxygen utilisation rates (Green and Shelef,

1981).

A decrease in VSS observed in the present study may also be associated with the way in

which the sludge was handled before it was added to the wastewater. The sludge used in

the current study originated from a CFSTR activated sludge plant, and was kept under

anaerobic conditions for about 1 hour during transportation to the laboratory, and for

another one hour during which it was allowed to settle to increase the SS concentration.

Due to the length of time required to undertake an experiment, it was not possible to use

the sludge on the same day. The sludge was therefore, kept aerated over night without any

supply of substrate.

Chudoba et al. (1991) investigated the effect of anaerobic stabilisation of returned

activated sludge in a continuous system on biomass production under batch conditions at

various F/M ratios. It was found that at low S 0/X0 (substrate/biomass ratios) cell

replication did not take place during substrate removal. At higher S0/X0 ratios (>2)

however, a significant increase in biomass concentration was observed. They noted that

under anaerobic conditions (for 3 hours) in the absence of substrate, micro-organisms are

subjected to physiological shock caused by the lack of oxygen and food. Under the above

conditions, ATP is used as a source of energy. When the micro-organisms are returned to

aerobic conditions and supplied with exogenous substrate, they rebuild energy reserves at

the expense of growth. The anaerobic conditions however, did not effect the system's

assimilative capacity.
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 138

7.8.3. Effluent settleability results

In the current study, SVI (sludge volume index) rather than SSVI (stirred sludge volume

index) was used to evaluate the settling characteristics of the mixed liquor because of the

fact that the mixed liquor suspension was too dilute, especially at low seed concentrations,

to form an interface in a stirred cylinder. SVI however, is widely used as an operational

parameter (Echeverria et al., 1992). A detailed discussion about the comparison of the

working ranges of SVI and SSVI for the range of solids concentrations normally

occurring in activated sludge plants has been presented by Rachwal et al. (1982).

Although there seems to be no information available in the literature about the

comparison between quiescent and stirred tests for the MLSS ranges used in the current

study, Vesilind (1971), based on laboratory studies undertaken to evaluate the effect of

stirring on settling velocity of activated sludges noted that at relatively dilute

concentrations, stirring seemed to have no beneficial effect on settling velocity,

irrespective of the size of cylinder.

Activated sludge exhibits a good settleability if the SVI is less than 100, while a SVI

value of greater than 150 has been reported to indicate settleability problems and possible

bulking (Gray, 1989). Considering the above values of SVI, it can be observed from Table

7.9 that the average SVI values found in the present study, at seed concentration of 100-

1000 mg/1, are lower than 150 mug thus representing satisfactory settleability

characteristics.

As can be seen from Table 7.9 and Figure 7.9, the settleability (SVI) tends to shift from

"moderate" to "poor" (Gray, 1989) as the seed concentration increased from 100-1000
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 139

mg/1 (average F/M= 0.6-0.21 g COD/g MLVSS, Section 7.3). This is in accordance with

the findings of Echeverria et al. (1992) who, based on pilot-scale studies reported that

SVI increases with a decrease in F/M ratio in the range of 0.41-0.1 g BOD 5/g MLVSS. In

contrast, Grutsch and Mallatt (1976) as quoted by Capps et al. (1995) suggest that lower

MLSS concentrations (less than 700 mg/1) in activated sludge plant result in poor sludge

settling and therefore poor clarifier performance. Shao et al. (1992) based on comparative

studies on full-scale activated sludge plants (ASP) reported that the SVI of high rate ASP

(MLSS=900 mg/1) was higher than the conventional ASP (210 ml/g vs. 147 mug). Pilot-

scale studies on a pressure pipe treatment system at Irvine Ranch Water District

California (Section 2.2) also encounted significant problems with solids separation using

flotation (Scherfig, 1994).

A probable reason for the increase in SVI with an increase in seed concentration observed

in the current study may be the addition of starved sludge to the wastewater for seeding

purposes. As mentioned earlier (Section 7.8.2), the seed sludge was kept aerated overnight

without feeding any substrate. Horan and Shanmugan (1986), who investigated the effect

of nutrient starvation on the settling properties of activated sludge, found that nutrient

starvation resulted in a loss of settleability. They attributed this effect to extensive cell

lysis which caused pin flocs, and possible high growth rate of filamentous organisms at

low nutrient concentrations.

From Table 7.9 and Figure 7.9, it can be observed that inspite of low SVIs at seed

concentrations of 100 and 250 mg/1, the suspended solids removal in the bench-scale

settling column was lower than those found at higher seed concentrations. From these
 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 140

results it appears that suspended solids removal at these seed concentrations may be poor,

possibly due to lack of adequate bioflocculation because of low biomass concentration.

7.8.4. Oxygen uptake rate

The results presented in Figures 7.10 and 7.11 show that the SOUR at seed concentrations

of 100 and 250 mg/1 varied widely and did not show any specific trend. Also the SOUR

values at these seed concentrations correlated poorly with SS of the wastewater prior to

the addition of activated sludge, initial SCOD, and the retention time. SOUR profiles

similar to the above were also observed for unseeded wastewaters (Section 6.6), and a

number of possible explanations for the variation in SOUR have been outlined in Section

6.7.4.

The results shown in Figures 7.12-7.14 indicate that, when compared with unseeded

wastewater (Section 6.6), the SOUR of the seeded wastewater (at a seed concentration of

>250 mg/1) would, in general, be at a maximum following the addition of seed to the

wastewater and then would decrease with time. A similar oxygen profile has been

reported for a plug-flow activated sludge process by other researchers. Kesseler and

Nicholas (1935) reported SOUR values of up to 137 mg/l.h at the inlet and 28 mg/l.h after

2.5 hours of aeration. Randall et al. (1991), while investigating the SOUR of raw

wastewater added with activated sludge in a batch reactor, noted a peak SOUR of 15 mg

02/g MLSS.h which decreased to 7 mg 0 2/g MLSS.h after a retention time of 6 hours.

A sharp increase in SOUR following the addition of seed to the wastewater may be

attributed to the rapid utilisation of readily biodegradable organic matter (RBOM). The

SOUR decreases as the RBOM is depleted, and is subsequently dominated by growth on

 CHAPTER 7: ADDITION OF OXYGEN TO SEEDED WASTEWATER 141

substrate released by hydrolysis and less by endogenous respiration (Kappeler and Gujer,

1992).

The results presented in Figures 7.12-7.14 show that the initial SOUR of the wastewater

increases with an increase in seed concentration. This is in agreement with the results

reported in Water Pollution Research Report (1964). Schulze and Kooistra (1969) quoting

Water Pollution Research Report (1964) reported inlet SOUR values of 70, 100, and 140

mg/1.h for domestic wastewater at MLSS concentration of 2000, 4000, and 6000 mg/1,

respectively. The corresponding values at the end of the aeration tank were found to be 5-

20 mg/1.h.

The maximum value of SOUR observed at a temperature of 20°C (Section 6.6) amounted

to 25 mg/l.h. As compared to the above, SOUR values of up to 24, 28, 38, 57, 134 mg/l.h

were observed at seed concentrations of 100, 250, 500, 750, and 1000 mg/1, respectively.

In practice higher values of OUR may be found in sewers due to the activity of slime

layer. The SOUR results cited above indicate that seeding sewers with activated sludge (at

a concentration of 250 mg/1) would result in a considerable increase in the total oxygen

requirements of the wastewater, especially at the point of addition of seed. Considering

the fact that an increase in the dissolved oxygen concentration of the wastewater would

result in a reduction in the dissolution efficiency of oxygen in sewers (Section 2.4), the

results of the current study suggest that the concentration of seed to be added to a given

sewerage system may be governed by the ability to dissolve the required amount of

oxygen.
 142

CHAPTER EIGHT

CASE STUDY-THE GREATER AMMAN SIPHON

8.1. INTRODUCTION

This chapter describes the results of experimental work carried out at the Central

Laboratories of the Water Authority of Jordan, Amman, to study the feasibility of using

the 39 km long Greater Amman Siphon (GAS) as an aerobic treatment facility. Batch tests

were undertaken, on raw wastewater collected at the inlet of the GAS (Appendix A-1), to

simulate the removal of organic matter under aerobic conditions with and without the

addition of activated sludge at the inlet of the GAS. A description of the GAS and the

experimental methodology adopted for this study are given in Chapter 4 (Section 4.3.2).

8.2. CHARACTERISTICS OF THE WASTEWATER AND SEED SLUDGE

The removal of soluble organic matter from the GAS wastewater was studied using an

initial added seed concentration of 0 (no seeding), 100, 250 and 500 mg VSS/1. A total of

four individual grab samples of the wastewaters were subjected to batch tests at each of

the above seed concentrations. In the first stage, two batch reactors, the first containing 10

litres of raw wastewater and the second containing 10 litres of raw wastewater seeded

with 100 mg VS S/1 of activated sludge were run in parallel. In the second stage, two

reactors, the first containing raw wastewater seeded with 250 mg VSS/1 of activated

sludge and the second containing raw wastewater seeded with 500 mg VSS/1 of sludge

were operated in parallel. The average characteristics of the wastewater studied at

different seed concentrations are shown in Table 8.1, while Table 8.2 summarises the

average characteristics of the seed sludge. As a result of the addition of activated sludge,
CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 143

which had a very low soluble COD (SCOD), the SCOD and soluble BOD 5 (SBOD5) in

the batch reactor at the start of the test were lower than the corresponding SCODs or

SBOD5 values of the raw wastewater. The initial SCODs and SBOD 5 concentrations in

the reactor corresponding to each seed concentration are also shown in Table 8.1. The

total COD of the influent wastewater averaged 1428 mg/1 (SD=328) at initial seed

concentrations of 0 and 100, and 1524 (SD=195) mg/1 at initial seed concentrations of 250

and 500 mg/l.

Table 8.1-Average characteristics of GAS wastewater

Seed Wastewater characteristics Initial concentration n

concentration
SS SCOD SBOD5 SCOD SBOD5

0 850±142 395±27 205±26 395±27 205±26 4

100 850±142 395±27 205±26 386±26 200±26 4
250 855±164 360±20 191+24 341±21 185±26 4
500 855±164 360±20 192±24 324±19 171+24 4
Notes: (i) concentration in batch reactor at the start of test
(ii) all parameters except "n" are expressed in mg/I

Table 8.2-Average characteristics of seed sludge

Seed concentration Seed sludge characteristics (mg/1) n

(mg VS S/1)
VSS SCOD SBOD5
100 3820±184 41±9 8±2 4
250 4645±574 40±9 9±2 4
500 4645±574 40±9 9±2 4
 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 144

The temperature of the GAS wastewater or the mixed liquor at the start of the experiments

ranged from 25-26.5°C at seed concentrations of 0 and 100 mg/1, and 24-30°C at seed

concentrations of 250 and 500 mg/l. The temperature of the wastewater in the batch

reactors was found to decrease during the test period, depending upon the ambient

conditions. The final temperature (after 8 hours) in the batch reactor varied from 20-23°C

at seed concentrations of 0 and 100 mg/1, and 20-28°C at seed concentration of 250 and

500 mg/l. The average daily temperatures at each seed concentration however, amounted

to approximately 25°C.

8.3. REMOVAL OF SOLUBLE ORGANIC MATTER

Figures 8.1-8.4 show the variation in SCOD over time from the wastewater/mixed liquor

at a seed concentration of 0-500 mg/l. The average SCOD and SBOD 5 removal

efficiencies at different seed concentrations are summarised in Tables 8.3 and 8.4,

respectively.

The average effluent SCODs, after a retention period of 4 hours, amounted to 213

(SD=33), 184 (SD =25), 141 (SD=27), and 110 (SD=16) mg/l. The corresponding values

over a retention period of 8 hours were found to be 160 (SD=26), 130 (SD=14), 136

(SD=19), and 114 (SD=16) mg/1, respectively. The corresponding effluent SBOD5

concentrations, after a retention period of 8 hours, averaged 45 (SD=3), 36 (SD=4), 23

(SD=8), and 12 (SD=10) mg/1, respectively.


CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 145

450 ..

400
Influent SS (mg/I)
350 _ '4\144444444444444444iiimmiiiiiii....
mimmemmilimmommiw
___._ 750 -g- 800
___A_. 790 -x- 1060

0 250
U
v)
200

150 _

100

0 2 4 6 8

Retention time (hours)

Figure 8.1-Variation in SCOD of GAS wastewater.

450

400 Influent SS (mg/I)

.._750 -.--800
350
..._790 -.*-1060

200 -

150 _

100
0 2 4 6 8
Retention time (hours)

Figure 8.2-Variation in SCOD of GAS wastewater at an initial added seed

concentration of 100 mg/1 (Influent SS indicate the SS of the raw wastewater prior to the

addition of activated sludge).

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 146

400

350 Influent SS (mg/I)

_4_710 .__.__ 960
,--, 300 __A_ 1030 _x__. 720
...ico
/
E
'—' 250
12t
0
U
cn 200

150
•••—IIII

100

0 2 4 6 8
Retention time (hours)

Figure 8.3-Variation in SCOD of GAS wastewater at an initial added seed

concentration of 250 mg/l.

375

Influent SS (mg/I)
325
—0_ 710 _8_960

,--, 275 - ____ 1030 __x__. 720

E
'—' 225 _
a)
0
U
cn 175_

125 _

75
0 1 2 3 4 5 6 7 8
Retention time (hours)

Figure 8.4-Variation in SCOD of GAS wastewater at an initial added seed

concentration of 500 mg/l.


CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 147

Table 8.3-Average SCOD removal efficiencies

Seed concentration Retention time SCOD removal SCOD removal

(mg/1) (hours) (mg/1) (%)

4 183±30 46±7
0
8 235±18 60±5
4 203±25 52±6
100
8 256±14 66±2
4 200±26 59±7
250
8 205±9 6014
4 214±18 66±4
500
8 210±13 65±4

Note : average influent SCOD and SS concentrations are shown in Table 8.1

Table 8.4-Average soluble BOD 5 removal efficiencies after a retention period of 8

hours

Seed concentration (mg/I) SBOD5 removal (mg/1) SBOD5 removal (%)

160±25 78±3

100 164±21 82+0

250 161±24 87±4

500 160±24 93±1
Note : average influent SBOD 5 and SS concentrations are shown in Table 8.1

Considering the fact that the use of ATU to arrest nitrification during the BOD 5 test is not

practised in Jordan, the effluent BOD 5 concentrations for seeded experiments could be

high as they may include the nitrogenous oxygen demand as well.

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 148

8.3.1. Effect of wastewater strength on SCOD removal

For unseeded experiments (Figure 8.1), the SCUD removal over 8 hours for individual

wastewaters, in general, was to found to increase with an increase in SCOD 0 or SS0 of the

wastewater. At approximately the same SCOD 0 concentration, i.e. 370 mg/1, an increase

in SS° of the wastewater, from 790 to 1060 mg/1 (34%), increased the SCUD removal

over 8 hours from 215 to 250 mg/1 (16%). For a similar 5S0 concentration, i.e. 800 mg/1,

an increase in initial SCUD from 370 to 410 mg/1 (11%) increased the SCUD removal

over 8 hours by 35 mg/1 (16%). Within the range of values observed during the period of

current study, the SCUD removal over 8 hours for seeded experiments did not appear to

show any obvious relation with the SS 0 of the wastewater prior to the addition of

activated sludge or the initial SCUD of the wastewater.

8.3.2. Kinetics of SCOD removal

To find the degradation rate of SCUD in the GAS wastewater at various seed

concentrations, first- and second-order reaction rate models were fitted to all the

individual curves shown in Figures 8.1-8.4. At seed concentrations of 0 and 100 mg/1,

second-order fits yielded higher correlation coefficients than the first-order fits, indicating

that SCUD degradation with respect to time would follow second-order kinetics at these

seed concentrations.

At seed concentration of 250 mg/1, two wastewaters studied exhibited second order

kinetics with respect to SCUD removal over time, while the SCUD removal for the

remaining two wastewaters did not show any removal after a retention period of 4 hours

(Figure 8.3). A trend similar to the above was also shown by three wastewaters studied at
 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 149

a seed concentration of 500 mg/1 (Figure 8.4). The variation over time in the GAS

wastewater at seed concentrations of 250 and 500 mg/1 displayed a behaviour similar to

that exhibited by wastewater samples collected from Sacriston WWTP at seed

concentrations of 750 and 1000 mg/1 (Section 7.3; Figures 7.4 and 7.5). Although no

SCUD data for the GAS wastewaters were collected between a retention period of 0 and 4

hours, based on the results presented in Section 7.3, it is suggested that the SCUD

removal at seed concentrations of 250 and 500 mg/1 in the GAS followed second order

kinetics with respect to time, during a retention period of 4 hours.

Multiple regression analysis was used to formulate a relationship for describing SCUD

removal in the GAS wastewater over time at various seed concentrations. The best-fit

empirical equation for SCUD for unseeded wastewater (Figure 8.1) is as follows:

SCOW SCOD0 = loge (2.7/(1+0° 27 ) (R2 = 0.97) (8.1)

Where: SCODt SCUD in the batch reactor at any time t (1-8 ) hours from the start of the

test, and

SCOD0 = SCUD in the batch reactor at the start of the test.

Equation (8.1) shows a SCUD removal trend similar to that represented by the best-fit

equation obtained for Sacriston wastewater at an added seed concentration of 100 mg/1

(Equation 7.1). This may reflect the effect of the high strength of the GAS wastewater and

the high temperature on SCOD removal pattern.

CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 150

The following best-fit relationships were obtained for the description of SCUD variation

over time at seed concentrations of 100-500 mg/1:

Seed concentration = 100 mg/1:

SCUDt / SCODo= loge (2.7/(1+t)°31) 2_O98) (8.2)

Seed concentration =250 mg/1:

SCUDt / SCOD0 = (0.3+0.70/(1+t)) (R2 = 0.98) (8.3)

Seed concentration = 500 mg/1:

SCODUSCODO = (0.23+0.771(1+W (R2 =0.98) (8.4)

A comparison of the best-fit equation at a seed concentration of 100-500 mg/1 for

Sacriston wastewater (Equation 7.2), with Equations 8.2-8.4 indicated that at similar seed

concentrations, the SCUD removal over time for Sacriston and the GAS wastewater

generally showed a similar trend. Consequently, it can be concluded that Equation 7.2 is a

reasonable approximation of the SCUD removal over time in a seeded wastewater.

8.3.3. Effect of seeding on SCOD removal

Figure 8.5 illustrates the effect of seed concentration on SCUD removal at retention

periods of 4 and 8 hours. Figure 8.6 shows the data shown in Figure 8.5 but replotted to

illustrate the effect of retention time on SCUD removal efficiency at various seed

concentrations.
CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 151

400

350
g___.___ influent SCOD
__a_ SCOD removal after 4 hours
al 300 _
0 -A__ SCOD removal after 8 hours
U
cn
a) 250 .
to
al
;-.
a)
>
"c4 200

150

0 100 200 300 400 500
Added seed concentration (mg/1)

Figure 8.5-Effect of seed concentration on SCOD removal.

260
250 _ Seed concentration (mg/I)
_._ 0 __ 100
240
et 230
'-cit'
_f
r__ 250 ___ 500

> 220 -
1 210
I-,
a) 200
0
U
CID 190 -

180 -
170

3 4 5 6 7 8 9
Retention time (hours)

Figure 8.6-Effect of retention time on SCOD removal.

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 152

The average improvement in SCUD removal as a result of the addition of 100 mga seed,

as compared with unseeded wastewaters, averaged 11 and 10% over retention periods of 4

and 8 hours, respectively (Figure 8.5). The average increase in SCUD removal when the

seed concentration was increased from 250 to 500 mg/1 amounted to 7 and 2.4%, over

retention periods of 4 and 8 hours, respectively.

From Figure 8.6, it can be seen that at seed concentrations of 0 and 100 mgfi, the SCUD

removal appears to increase linearly with an increase in the retention time from 4 to 8

hours. In contrast, the SCUD removal at seed concentrations of 250 mg/1 did not show

any noticeable removal after 4 hours, while, at a seed concentration of 500 mg/1, it

showed a decrease when the retention period was increased from 4 to 8 hours, due to an

increase in the SCUD in the reactor due possibly to generation of soluble microbial

products (Section 7.8.1).

It must be noted that a direct comparison of SCUD removal efficiencies for all the seed

concentrations investigated can not be made from the results of this study because all the

experiments were not run in parallel (Section 8.2), and also the wastewater strength for

the seed concentrations of 250 and 500 mg VS S/1 was significantly lower than that of 0

and 100 mgVSS/1 (Table 8.1). Multiple regression analysis was, therefore, used to

evaluate the effect of seed concentration on SCUD removal while taking into account the

effect of variation in initial SCUD, and SS concentrations of the individual wastewaters

prior to the addition of activated sludge (Appendix E-1). The results of the multiple

regression analysis are as follows:

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 153

RSCOD = - 44.3 + 0.464 SCOD0 - 0.0342 SS° + 16.7 t+ 25.1 b -3.2 bt (R2 = 0.6)

(8.5)

Where : RSCOD = SCOD removal (mg/1) after a specified retention time t (hours),

b = (1 + 0.01 x seed concentration in mg VSS/l),

SCOD0 = SCUD concentration in the reactor at the start of the experiment, and

SS0 = SS concentration in the wastewater prior to the addition of seed.

The summary statistics for Equations 8.1 is shown in Appendix E-2. Inspection of

summary statistics revealed that at a confidence interval of 5%, the SCUD removal in the

GAS wastewater was significantly affected by SCODo, retention time and seed

concentration, and interaction between the seed concentration and the retention period.

The effect of SS0 however, was not found to be significant.

8.4. OXYGEN UPTAKE RATE OF SUSPENDED BIOMASS (SOUR)

The SOUR profiles of the of wastewater at various seed concentrations, over a retention

period of 8 hours, are shown in Figures 8.7-8.10. Table 8.5 shows, for different seed

concentrations, the ranges of SOUR observed just after the start and before the end of

batch experiments as well as the peak values.

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 154

75 -
SCOD0 SS 0 ADT
70 -
65 - --e- 410 750 24
60 _ -43- 425 800 24.5

-9
-k- 370 790 25
C 50 - 375 1060 26
g 45 _
40 _ 370 1030 23
P4 35 - -0-- 325 720 23
0 30 _
cn 25 .
20 . --
15 .
10 -
5
2 3 4 5 7 8
Retention time (hours)

Figure 8.7-SOUR profile of GAS wastewater (ADT=average daily temperature °C;

SCODO and SSO are experssed in mg/1).

80 _

70 -
SCOD, SS 0 ADT
60 . , -e- 400 750 24
-e- 415 800 24.5
-79-- 360 790 25
370 1060 26

20 _

10 _

0 i

0 I 2 3 4 5 6 7 8 9
Retention time (hours)

Figure 8.8-SOUR profile of GAS wastewater at an initial added seed concentration

of 100 mg/1.


CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 155

100
90 -
80 -
70 -
60
50
40 _
30 -
20 -
10 -
0
0 1 2 3 4 5 6 7 8
Retention time (hours)

Figure 8.9-SOUR profile of GAS wastewater at an initial added seed concentration

of 250 mg/l.

100 _
90 -
SCOD 0 SS0 ADT
80 - 330 710 29
..c: 335 960 23
Ti ) 60
335 1030 23
50_
295 720 23
0 40 -
30 _
20 _
10

0 1 2 3 4 5 6 7 8
Retention time (hours)

Figure 8.10-SOUR profile of GAS wastewater at an initial added seed concentration

of 500 mg/l.

CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 156

Table 8.5-SOUR of suspended biomass

Seed concentration SOUR range (mg/1.h)

(mg VSS/l) Initiala Maximum Final
0 31-38 53-67 11-18
100 45-51 51-69 10-27
250 47-69 56-91 20-28
500 58-95 65-93 20-28
Notes: a during the first hour from the start of oxygen addition
between seven and eight hours from the start of oxygen addition

The total oxygen demand of a given wastewater over 8 hours was determined by

computing the area under the SOUR curve for that particular wastewater. Since no SOUR

data were collected at 0 and 8 hours, the SOUR curves were therefore extrapolated to

include the area between these retention periods. The average SOUR values were

determined by dividing the total oxygen demand by the total retention period (8 hours).

Table 8.6 summarises the average oxygen demand and SOUR values of the wastewater at

different seed concentrations.

Table 8.6-Average oxygen demand and SOUR values

Seed concentration (mg VSS/I) 0 100 250 500

Total oxygen demand (mg/1) 240.3±17.9 244±20.9 268.2±34.3 279.6±47.8

Oxygen uptake rate (mg/1.h) 30±2.2 30.5±2.6 33.5±4.3 35±6

From the results shown in Figures 8.7-8.10, it can be seen that the SOUR at seed

concentrations of 0-500 mg/1, was, in general, low just after the start of the oxygen

addition, it increased to a maximum value afterwards, and then showed a gradual

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 157

decrease. The peak values of SOUR appeared to have occurred when the retention period

was between 1-2.5 hours for seed concentrations of 0 and 100 mg/1, and < 1.5 hours at

seed concentrations of 250 and 500 mg/l.

From the results presented in Table 8.5, it can be seen that an increase in seed

concentration tends to increase the initial SOUR of the wastewater. It was, however, not

possible to isolate the effect of seed concentration on overall SOUR (average over 8

hours) due to the variation in SCODs and temperature of the wastewater.

8.5. DISCUSSION

8.5.1. Removal of soluble organic matter

The results of the current study demonstrate that by maintaining aerobic conditions in the

GAS, SCOD and SBOD 5 removals of 60 and 78% respectively, could be achieved at an

average temperature of 25°C and over a retention period of 8 hours without seeding the

GAS. There was also evidence to suggest that the corresponding substrate removal

efficiencies could improve to 66 and 82% respectively if the GAS wastewater were to be

seeded with activated sludge at a concentration as low as 100 mg VSS/1. From the results

shown in Table 8.3 it also appears that the time required to achieve SCOD reductions

equivalent to those obtained over 8 hours with unseeded wastewater may be reduced to

4 hours when the seed concentrations is increased to 250-500 mg/l. Moreover, average

soluble BOD5 removal efficiencies of 87 and 93% were achieved at seed concentration of

250 and 500 mg/1 respectively (Table 8.4), which are comparable to those found in a

typical activated sludge plant.

 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 158

SCOD removal efficiencies over 4 hours for the unseeded wastewater collected from

Durham WWTP (Chapter 6) averaged 134 mg/1 and 144 mg/1 at 20 and 30°C,

respectively. The corresponding removal efficiencies over 8 hours averaged 186 and 191

mg/1, respectively. When compared to the Durham wastewater (having a high SCOD 0 and

high SS0), the SCOD removal efficiencies for the unseeded GAS wastewater averaged

183 and 235 mg/I over a retention period of 4 and 8 hours, respectively, at an average

temperature of 25°C. These values are 37 and 27% higher than the corresponding values

over 4 hours for the Durham wastewater at 20° and 30°C, respectively. Similarly, the

SCOD removals in the GAS over 8 hours were 26 and 23% higher than the corresponding

values found for Durham wastewaters at 20 and 30°C, respectively. Also, the SCOD

removal over time for unseeded GAS wastewater, when compared with unseeded

wastewater studied at Newcastle (Chapter 6), was found to follow second-order kinetics.

The high SCOD removal efficiencies for the GAS wastewater and second order SCOD

removal kinetics may be attributed to the high strength of the wastewater, especially the

high SS (average 850 mg/I) concentration. These results demonstrate that the strength of

the wastewater would be a major parameter affecting in-sewer treatment.

The results of the Amman study showed that the effect on SCOD removal of seeding the

wastewater with activated sludge in the range of 100-250 mg/1 was marked only when the

retention period was 4 hours, possibly due to the fact that, for seeded experiments, most

of the SCOD was removed over this period. A diminished effect of seeding with an

increase in seed concentration and retention time is consistent with the findings of

previous investigations which involved wastewaters taken from Sacriston WWTP, in

Newcastle (Chapter 7). A number of possible explanations for the diminished effect of

seed at higher retention periods have been outlined in Chapter 7 (Section 7.8.1).
 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 159

Considering the fact that under limiting substrate conditions, the amount of SCOD

removal in a batch reactor is proportional to the amount of SCOD remaining in the

reactor, one of the reasons for low SCOD removal efficiencies at seed concentrations of

250 and 500 mg/1 noted in the current study may be the low initial SCOD concentrations

at these seed concentrations.

8.5.2. Oxygen uptake rate

When compared with the SOUR profiles for unseeded wastewaters studied in Newcastle

(Section 6.6), the SOUR profiles for unseeded GAS wastewater showed a regular pattern

with respect to time (Figure 8.7). This may be due to the fact that the samples of the GAS

wastewater used for batch tests were relatively young as they were collected at the inlet to

the sewer rather than the inlet to the wastewater treatment works, as was the case for

Newcastle study. The SOUR results observed in the current study support the view

expressed by other researchers (Pomeroy and Parkhurst, 1972; Boon and Lister, 1975)

that the SOUR of a fresh wastewater would exhibit a bell-shaped profile (Figure 8.7) over

time.

The maximum values of SOUR observed for wastewaters studied in Newcastle (Section

6.6) were found to be 25 and 40 mg/l.h at 20° and 30°C respectively. When compared to

the above, peak values ranged from 53-67 mg/l.h for the unseeded GAS wastewater, while

the SOUR over 8 hours averaged 30 mg/l.h at a temperature of 25°C. At a similar

temperature (24-28 °C), Hemmings et al. (1983), based on field studies, reported an

average wastewater oxygen uptake rate of 14 mg/l.h. Considering the above, a high

SOUR value for the GAS wastewater may be attributed to its high strength.
 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 160

The average daily temperature of the wastewater did not vary greatly except for the

experiments conducted at seed concentrations of 250 and 500 mg/l. Although the

variation in temperature did not appear to show any noticeable effect on SCOD removal

over retention periods of 4 and 8 hours, it did significantly affect the SOUR of the

wastewater as can be seen from Figures 8.9 and 8.10. At a seed concentration of 250 mg/1,

the total oxygen demand of the wastewater over 8 hours increased from 235-273 mg/1 to

322 mg/1 when the average daily temperature increased by 6°C (from 23°-29°C).

Similarly, a 6°C rise in temperature, at a seed concentration of 500 mg/1, increased the

total oxygen demand from 235-273 to 322 mg/l. These results demonstrate that the

diurnal variation in the temperature of the GAS wastewater could produce marked

changes in the oxygen requirements of the wastewater, and thus provision should be made

in the design to accommodate such variation.

8.5.3. Prospects of using the GAS as a treatment facility

The data collected during the course of this study, although limited, have produced

evidence which demonstrates that maintaining aerobic conditions in the GAS could result

in a considerable degree of in-sewer treatment, and may also alleviate the serious odour

problems being encountered presently at the inlet of AWSP system. From the data

presented herein it also appears that, seeding the GAS (at a concentration of about 250 mg

VSS/1) at the inlet could also result in removal efficiencies comparable to those obtained

in an aeration tank of a conventional activated sludge plant. In this case, the existing

abandoned activated sludge plant at the inlet of the GAS could be re-activated to serve as

a feasible source of seeding. A portion of flow could be drawn and treated in the above

treatment plant, and the excess sludge then could be added to the inlet of the sewer.
 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 161

Considering an average SOUR value of 30 mg/l.h (Table 8.6), the total oxygen demand of

the wastewater (including the oxygen uptake rate of the slime layer) as calculated from

Equation 2.3 (Boon and Lister, 1975), would be approximately 34.5 mg/l.h or 276 mg/1

over a retention period of 8 hours at 25°C. At an average daily flow of 115,000 m3/day

(Section 4.3.2), the total oxygen demand of the wastewater (for the unseeded option)

would amount approximately to 32 tons/day.

When compared to the UK where the cost of liquid oxygen delivered at site varies from

07-£80 per ton (BOC, 1994), the cost of liquid oxygen delivered to site in Jordan varies

from f200-£500 per ton. The cost of oxygen injection for unseeded option therefore, may

exceed f6400-£16000 per day or f2,336,000-£5,840,000 per year. It is however, likely

that this cost may be reduced by on-site generation of oxygen. The use of GAS as a

seeded treatment facility with a view to enhancing the in-sewer treatment, when compared

with the unseeded option, would entail higher capital and running costs including the cost

of increased oxygen requirements and the cost of seed generation.

Considering a high oxygen demand of the GAS wastewater, and the fact that the

dissolution efficiency of oxygen decreases with an increase in temperature and the target

dissolved oxygen (Sections 2.4 and 2.8), it appears that the use of oxygen as a source of

DO for the GAS would be favoured in the place of air. Also, it seems that the oxygen

injection facilities have to located at many points along the GAS as it would not be

economically feasible to dissolve such a high DO concentration (276 mg/I) at a single

point. A high pressure at the bottom of the siphon may however, help minimise the

number of oxygen injection points along the rising leg of the siphon.
 CHAPTER 8: CASE STUDY-THE GREATER AMMAN SIPHON 162

As mentioned earlier, the cost of oxygen in Jordan is considerably higher than that in the

UK or other countries where in-sewer oxygen injection technology is well developed.

Also, no in-sewer oxygen injection technology is currently available in Jordan and the use

of the GAS as an aerobic treatment facility would therefore, require sophisticated

equipment to be imported from abroad, which may not be economically viable.

Considering the high cost of oxygen, the absence of in-sewer oxygen injection technology

and the lack of world-wide experience in the field of in-sewer treatment, it appears that

the use of GAS as a treatment facility, for the time being, may not be favoured, when

compared with the proposed short-term extension of the existing wastewater stabilisation

pond system. Subsequent to the signing of the peace agreement between Israel and Jordan

however, huge capital investment is being made in various sectors in Jordan by a number

of European and American companies. This is likely to solve the problem of oxygen

injection technology and bring down the cost of oxygen generation, and hence, may in the

long-term, improve the prospects of using the GAS as a treatment facility.

The results of the present study however, have advanced sufficient evidence in support of

the use of GAS for in-sewer treatment to warrant further research work, using more

precisely designed experiments. A detailed technical and economic evaluation of the GAS

as a treatment facility would require further experimental work aimed at investigating,

both the seeded and unseeded options, the effect of diurnal and seasonal variation in

wastewater strength and temperature on in-sewer treatment, oxygen uptake rate of the

wastewater and slime layer, settling characteristics of the effluent, and sludge production.
 163

CHAPTER 9

GENERAL DISCUSSION

9.1. INTRODUCTION

The results of the current study have demonstrated that sewers may be used as aerobic

biological reactors to effect at least partial treatment of wastewater during transit, and

could under certain circumstances, replace the aeration units of an activated sludge plant.

The main factors affecting in-sewer treatment have been found to be wastewater strength,

retention time, temperature, and the seed concentration (in those cases where it is used).

The design of an existing sewerage system or a proposed system for in-sewer treatment

however, would require a number of factors to be considered in addition to the parameters

referred to above and some of which may be involved in assessing the potential of a

sewerage system for in-sewer treatment are discussed in the following sections.

9.2. NETWORK LAYOUT AND CHARACTERISTICS

The type, layout and characteristics of a sewerage system would be a critical element in

assessing its feasibility for in-sewer treatment, either technical or economic, in any given

situation. There are many scenarios, with consideration being given to its potential for use

from partial to complete secondary treatment. The following are a few such scenarios

which may be considered:

 CHAPTER 9: GENERAL DISCUSSIONS 164

9.2.1. Pressure mains versus gravity sewers

All the field studies on in-sewer treatment conducted so far have involved pressure mains

(Section 2.2.2). The major reasons for the selection of pressure mains over gravity sewers

could be the well-developed air or oxygen injection technology and the ability to attain

high dissolved concentrations under a high pressure. Also, better operational control

possibly favours the use of pressure mains while the situation in gravity sewerage systems

is more complex because of flow variations and the presence of tributaries. When

compared with pressure mains, the changes in flow in gravity sewers result in changes in

the effective slime area in contact with the wastewater and hence the degree of treatment

that can be achieved. An estimate of the treatment in gravity sewers however, could be

made based on minimum, average and maximum flow rates (Newcombe et al. 1979) .

In addition to the advantages described above, it has been suggested that the use of a

pressure main for in-sewer treatment may result in a higher degree of in-sewer treatment

that in gravity sewers. Based on batch studies, undertaken to simulate a step-fed sewer,

Green et al. (1985) reported that the average soluble COD removal efficiency increased

by 7% when the reactor was operated under pressure (2 atmospheres) for 3 hours out of a

total retention period of 10 hours. They attributed increased substrate removal to the

decreased floc size as a result of pressure, which could cause an increase in substrate

availability to the inner parts of the floc, and also provides a larger surface area. Lin and

Huang (1993), based on laboratory studies, also claimed that the efficiency of an activated

sludge process could be significantly improved by pressurised aeration.

 CHAPTER 9: GENERAL DISCUSSIONS 165

9.2.2. Long, single pressure pipe

A long single pressure pipe could offer a significant potential for partial treatment of a

wastewater, even without seeding, and may even offer treatment potential comparable to

that in a secondary treatment plants when the temperature is high. Typical examples of

such systems include a 40 km long pressure main in Kuwait, and a 40 km long siphon in

Amman, Jordan.

An alternative configuration of the long, single pressure pipe would be the use of a

specifically constructed pressure pipe which treats the wastewater while it is being

transported to an upstream point where there is a potential for its re-use. Such a system,

proposed initially by Stoyer (1970), may be particularly applicable to some current

situations in the UK. It may be possible to supplement low river flows, which are

becoming an increasing concern of the Environmental Agency, by returning a wastewater

to an upstream point. Not only would this increase the potential for in-sewer treatment, it

would also prevent the 'loss' of water to coastal areas.

9.2.3. Loop system

The sewerage system may be such that the location of the downstream wastewater

treatment plant is in reasonably close proximity to the head of the system and a return

activated sludge pipe could readily and cheaply be constructed to introduce activated

sludge at the start of the system. Such a sewerage system exists in Tel Aviv, Israel, where

the treatment work is at a distance of only 8 km from the inlet of a 37 km long sewer

(Green et al., 1985). Manandhar and Schroder (1995) have proposed the use of loop
 CHAPTER 9: GENERAL DISCUSSIONS 166

sewerage system as a circulating reactor wherein the wastewater could be recycled to its

point of origin to enhance self-purification.

9.2.4. Upstream wastewater treatment plant

It may be viable for a small upstream treatment plant to be constructed which may draw

off a portion of the wastewater flow from the sewer for treatment. The excess activated

sludge could then be returned to the sewer as the source of biomass for in-sewer

treatment.

9.2.5. Upstream wastewater treatment plant on unconnected sewerage system

Due to a number of factors including population distribution, topography, historical

development etc., it may be possible for a treatment plant to be located close to the head

of another sewerage system. The trunk sewer being considered for in-sewer treatment

could use the excess sludge as the required biomass, as well as being used for disposal of

the sludge.

9.3. ROLE OF THE SLIME LAYER IN IN-SEWER TREATMENT

As already indicated, sewers constitute a dual-phase system with suspended biomass

present in the liquid phase and in the slime layer on the wetted surface of the sewer wall.

The present study focused on the removal of soluble organic matter in the liquid phase

only. The in-situ substrate removal rate, therefore, may be higher than those observed in

the current study due to the activity of the micro-organisms present in the slime layer.
 CHAPTER 9: GENERAL DISCUSSIONS 167

The contribution of the slime layer to the removal of organic matter in sewers is

significant only for shallow sewers (Pomeroy and Parkhurst, 1972) and is governed by the

ratio of the area of wetted surface (A) to the wastewater volume (V), and on the

suspended solids concentration in the wastewater (Cao et al. 1992). Based on laboratory

studies designed to investigate a shallow river system, Cao et al. (1992) demonstrated that

an increase in AN resulted in a corresponding increase in the contribution of the biofilm

to the removal of organic matter, relative to that of the liquid phase. It was however, noted

that at an AN ratio of 2.1 mil_the-biofilm played a major role only when the suspended

solids concentration was 12 mg/l. The specific activity of suspended biomass with

respect to oxygen consumption rate was found to be 50 to 100 per cent of that of the

biofilm. On the basis of laboratory studies, Raunkjaer (1993) demonstrated that at an A/V

ratio of 37 m-1 , the removal from the biofilm accounted for 70% of the total organic

matter removal from the liquid phase and the biofilm, at 16°C.

9.4. IN-SEWER TREATMENT FOR DEVELOPING COUNTRIES

Much of the developing world is currently without sewerage or any form of wastewater

treatment. Moreover, the problems of wastewater collection, treatment and disposal is

generally magnified in developing countries since they are frequently in warm climates

and the wastewater strength is often significantly greater than that in the UK.

The potential of in-sewer treatment for developing countries appears to be great

considering that, in many cases, completely new sewerage systems are being developed

which could be "engineered" to optimise in-sewer treatment. Another aspect which is

likely to improve the prospects of in-sewer treatment in developing countries is the high
 CHAPTER 9: GENERAL DISCUSSIONS 168

temperature (greater than 25°C) and the need anyway to maintain higher DO levels for

sulphide, odour and corrosion control. Many large, new schemes are also being

constructed in areas with little or no natural fall, thereby increasing the retention time and

increasing the need for oxygen supplementation.

9.5. OPERATIONAL PROBLEMS ASSOCIATED WITH IN-SEWER

TREATMENT

Although the use of sewers as a treatment facility would prove beneficial in certain

situations, it could adversely effect nutrient removal at a downstream treatment plant

incorporating such a system.

Another adverse effect of in-sewer treatment would be the "decentralisation" of operation

and maintenance since much of the aeration and pumping equipment would be located at

points remote from the main wastewater treatment plant. This, however, does not detract

from the overall attractiveness of the concepts discussed in this dissertation.

9.6. ECONOMICS OF IN-SEWER TREATMENT SYSTEM

Compared with an aeration basin in a conventional secondary treatment system, the land-

take of an in-sewer aerobic treatment (ISAT) system, in the case of an existing sewer,

would be negligible, while the corresponding running costs would be mainly associated

with the cost of oxygen injection. The overall economics of ISAT, however, will be

governed by a number of factors including the degree of treatment which can be achieved,

the availability of adequate oxygen injection techniques, the cost of sludge recycling, in

addition to the factors outlined in Section 9.2, and would be highly site-specific.
CHAPTER 9: GENERAL DISCUSSIONS 169

In-situ studies have demonstrated that a considerable saving in investment could be

achieved by using sewers for partial or full aerobic treatment. Newcombe et al. (1979)

concluded that by using an 8 km long rising main to upgrade a plant overloaded by 40 %

would result in a saving of £188,000 in NPV (net present value) over 20 years, while

producing an effluent quality of 50/30 (SS/BOD 5) in winter and 30/20 in summer. The

overall saving was found to be as a result of the prevention of septicity and savings in the

cost of conventional extensions.

The use of a 6 km long pressure main followed by a sedimentation tank as a full treatment

facility in Australia was found to produce a 50/50 (SS/BOD) effluent (Hutchinson, 1981;

Hemmings et al., 1983). It was estimated that, in exchange for an additional ongoing

running cost of Aust. $20,000 per annum, largely associated with oxygen injection, there

would be a saving of Aust. $700,000 in initial capital outlay for conventional secondary

treatment facilities to achieve a similar effluent quality. Stoyer (1970) and Green et al.

(1985) also estimated that by using a sewer followed by a sedimentation tank for full

treatment instead of a conventional activated sludge plant, a saving of more than 50 %

could be expected in the construction costs.

 170

CHAPTER 10

CONCLUSIONS

The following main conclusions can be drawn from the results of this simulation study

designed to investigate the in-sewer treatment of domestic wastewater:

10.1. AEROBIC VS. ANAEROBIC CONDITIONS IN SEWERS

• A comparison of simulated aerobic and anaerobic gravity transport of raw domestic

wastewater demonstrated that aerobic conditions in sewers, when compared with

anaerobic conditions, could result in a significantly greater removal of soluble organic

matter together with improved settleability of primary solids at the downstream

treatment works. This would suggest that aerobic treatment would be the most

favoured method of in-sewer biological treatment.

10.2. EFFECT OF MIXING SPEED ON IN - SEWER AEROBIC TREATMENT

• During the simulated aerobic gravity transport of wastewater, the effect of mixing

speed (which is in turn related to the wastewater velocity), in the range 50-300 rpm,

did not appear to have a significant effect on the removal of soluble COD nor on

suspended solids. At higher mixing speeds (>200 rpm) however, there was a

statistically significant reduction in suspended solids removal in a bench-scale settling

column, due to increased turbulence in the reactor which caused the the flocs to break-

up.
CHAPTER 10: CONCLUSIONS 171

10.3. EFFECT OF DISSOLVED OXYGEN SOURCE ON IN-SEWER AEROBIC

TREATMENT

• A comparative study designed to investigate the effect of dissolved oxygen source (air

or oxygen) on the reduction in soluble COD from raw wastewater during simulated

aerobic gravity transport, and on suspended solids removal at the downstream

treatment works, showed that the effect of source would not be statistically significant.

10.4. ADDITION OF OXYGEN TO RAW WASTEWATER DURING

SIMULATED GRAVITY TRANSPORT AT 20°C

• At a given temperature, the removal of soluble organic matter from wastewater was

found to be strongly influenced by the suspended solids and soluble organic matter

present in the wastewater as well as by the hydraulic retention time. Over a retention

period of 8 hours at 20°C, the soluble COD removal efficiencies averaged 48, 40, and

61% for wastewaters having average influent soluble COD concentrations of 176, 329

and 306 mg/1, and influent suspended solids concentrations of 353, 346 and 620 mg/1,

respectively.

• At 20°C, the soluble COD removal from the wastewater was found to be significantly

influenced by the influent suspended solids (but not the influent soluble COD) over

short retention periods i.e. 1-3 hours. For higher retention periods (  4 hours), the

soluble COD removal was significantly affected by both the influent suspended solids

and the influent soluble COD of the wastewater.

 CHAPTER 10: CONCLUSIONS 172

• The soluble COD removal from the wastewater, within a temperature range of 10-

30°C, mainly followed first-order kinetics with respect to time. Based on the

experimental results, an empirical relationship for calculating the soluble COD

removal efficiency for a given strength of wastewater was formulated.

• The temperature coefficient for soluble COD removal rate from wastewater averaged

1.088 and 1.056 within the temperature ranges of 10°-20°C and 15°-30°C,

respectively.

• Over a retention period of 8 hours, the total COD removal from the raw wastewater

(unsettled) varied from 6-23% at 20°C, and did not appear to show any relation to the

influent total COD or influent suspended solids. These results suggest that, when

compared with soluble COD removal, the total COD removal from the raw wastewater

would be much lower.

• Over a retention period of 6 hours at 20°C, the changes in suspended solids of the

wastewater did not show any clear pattern, ranging from -28 to +19%.

10.5. ADDITION OF OXYGEN TO RAW WASTEWATER ADDED WITH

ACTIVATED SLUDGE, DURING SIMULATED GRAVITY TRANSPORT

AT 20°C

• Addition of activated sludge to the wastewater in order to increase the suspended

biomass concentration, at initial levels of as low as 100 mg VSS/1, resulted in a

CHAPTER 10: CONCLUSIONS 173

significant increase in soluble organic matter removal, and hence a significant

reduction in the retention time required to achieve a given degree of treatment.

• The soluble COD removal was found to increase almost linearly as the activated sludge

concentration added to the wastewater increased in the range of 100-1000 mg/l. The

effect of increasing the seed concentration on soluble COD removal however, appeared

to diminish with the increase in retention time.

• The soluble COD and suspended solids concentrations present in the wastewater prior

to the addition of activated sludge, the concentration of activated sludge added to the

wastewater and the hydraulic retention time were all found to have significant effect on

soluble COD removal from the wastewater. In addition, the interaction between

activated sludge concentration and the hydraulic retention time was also found to have

a significant effect on soluble COD removal.

• Within a range of activated sludge concentrations of between 100-1000 mg/1, the

soluble COD removal from the wastewater appeared to follow second order kinetics

with respect to time. Based on the experimental data collected in the current study,

empirical equations were developed for the description of variation in soluble COD

over time.

• At activated sludge concentration of 100-1000 mg/1, the effluent from the batch reactor

after a retention period of 6 hours, showed satisfactory settling characteristics i.e.

having average SVI values of less than 150 ml/g. The SVI of the mixed liquor,
CHAPTER 10: CONCLUSIONS 174

however appeared to increase with an increase in seed concentration. The settleability

results also suggested that, despite low SVI values, the solids removal at activated

sludge concentrations of 100-250 mg/1 may be poor due to lack of adequate

bioflocculation, although the nature of the SVI test may be a significant factor in this

case.

• The oxygen uptake rate of the seeded wastewater did not show any specific trend over

time at activated sludge concentrations of 100 and 250 mg/1 while at higher

concentrations it displayed a pattern similar to those observed in the aeration tank of a

plug flow activated sludge plant. The oxygen uptake rate, especially just after the

addition of the activated sludge, increased significantly with an increase in activated

sludge concentration. At activated sludge concentrations of 250 mg/1, the oxygen

uptake rate was found mainly to be influenced by the hydraulic retention time. At a

sludge concentration of 1000 mg/1 however, the effect of influent soluble COD was

also found to be significant.

10.6. CASE STUDY-THE GREATER AMMAN SIPHON

• It has been demonstrated that, by maintaining aerobic conditions in the Greater

Amman Siphon, average soluble COD and soluble BOD 5 removal efficiencies of 60%

and 78% respectively, could be achieved over a retention period of 8 hours at 25°C.

• The oxygen uptake rate of the unseeded Amman wastewater averged 30 mg/l.h. The

total oxygen demand of the wastewater was estimated to be approximately 32 tons/day

at an average flow rate of 115,000 m3/day.

CHAPTER 10: CONCLUSIONS 175

• There was evidence to suggest that by seeding the Amman wastewater, by the addition

of activated sludge at the inlet of the Greater Amman Siphon, at a level of 250-500 mg

VSS/1, the soluble organic matter removal comparable to that found in the aeration

tank of a typical activated sludge plant, could be achieved.

• The overall results of the case study indicate that the Greater Amman Siphon has a

great potential for being used as a long term solution to a reduction in the load on the

Al-Samra wastewater stabilisation ponds system. Further studies however would be

required before a detailed technical and economic evaluation of the siphon as a

treatment facility could be undertaken, recommendations have been made in this

regard.
 176

CHAPTER 11

RECOMMENDATIONS FOR FURTHER WORK

Based on the results of the current study, the following recommendations are made for

future research on in-sewer treatment:

1. Pilot-scale studies using fresh raw domestic wastewater should be undertaken to

investigate the treatment potential of sewers with and without seeding. For this purpose, a

pipe could be laid parallel to an existing sewer, and fed with fresh wastewater using a

pump. The length of the pipe could be minimised by laying the pipe in the form of a loop

in which wastewater may be circulated for the desired retention period. The main

parameters of interest would include:

(i) the effect of diurnal and seasonal variation in wastewater strength and temperature

on in-sewer treatment and on the oxygen uptake rate of the wastewater and the slime

layer,

(ii) settling characteristics of the effluent after discharge from the sewer,

(iii) sludge production rates and biological sludge viability,

(iv) effect of changes in velocity and flow variations,

(v) effect of the type of the sewer i.e. pressure vs. gravity, and linear vs. branched,

(vi) bacterial die off rates,

(vii) use of chemicals instead of air or oxygen,

(viii) relation between the oxygen uptake rate and wastewater strength,

(ix) relative organic matter removal rates in the liquid phase as well as by the slime

layer,
CHAPTER 11: RECOMMENDATIONS FOR FURTHER WORK 177

(x) continuous vs. intermittent air or oxygen injection, and

(xi) the effect of in-sewer treatment on the efficiency of nutrient removal unit processes.

2. A comparative study should be undertaken to investigate the performance of various in-

sewer aeration or oxygenation devices for in-sewer treatment.

3. The treatment potential of sewers should be considered at the design stage and attempts

should be made to design and lay the new sewerage systems such as to optimise the in-

sewer treatment.

4. A survey of existing sewer systems should also be undertaken to identify those with

potential for in-sewer treatment. The economics of in-sewer treatment in comparison

with the costs of extension of existing overloaded wastewater treatment plants should

also be assessed.

5. A mathematical model should be developed for the design and operation of in-sewer

treatment system.
 178

REFERENCES

Al-Salem, S. S. (1987) Evaluation of the As-Samra waste stabilisation pond system and

its suitability for unrestricted irrigation, paper presented at Food and Agricultural

Organisation's Consultant Review Meeting on the "Treatment and Reuse of Sewage

Sludge for Irrigation", Rome, Italy.

Andreakis, A. D. (1987) A comparative study of the air and oxygen activated sludge

Systems. Environmental Technology Letters. 8 (5), 209-220.

Anon. (1973) In-pipe wastewater cleanup. Chemical Engineering. 80 (22), 32 & 34.

Appleton, B. (1976) Acid tests for Middle East drain brains. New Civil Engineer. 181, 20-

23.

Arceivala, S. J. (1981) Wastewater Treatment and Disposal: Engineering and Ecology in

Pollution Control. Marcel Dekker Inc., New York.

Argaman, Y. (1991) Chemical reaction engineering and activated sludge. Water Research.

25(12), 1583-1586.

Bacquet, P., Derville, A. and Seneliery, Y. (1992) Counteracting the formation of

hydrogen sulphide in sewer systems; treatment with compressed air. Techniques,

Sciences, Methodes. 87 (1), 25-35.

Barber, P. D. G., Gray, R. B., Lever, M. L. and Quinn, G. F. (1980) Oxygen Injection into

the Eagle Farm-Luggage Point Rising Mains to Control Sulphide. Unpublished

Report, Brisbane City Council Australia, Department of Water Supply and

Sewerage.
REFERENCES 179

Benefield, L. D. and Randall, C. W. (1980) Biological Process Design for Wastewater

Treatment. Prentice-Hall.

Bentzen, G., Smith, A. T., Bennett, D., Webster, N. J., Reinholt, F., Sletholt, E., and

Hobson, J. (1995) Controlled dosing of nitrate for prevention of H 2S in a sewer

network and the effects on the subsequent treatment processes. Water Science and

Technology. 31 (7), 293-302.

Berger, M. (1978) Oxygen injection in rising mains. Umwelt & Technik. 11, 34-36.

Bland, C. (1992) Hydrogen sulphide prediction and control in sewer systems. Pipes and

Pipeline International. 37 (1), 24-27.

BOC (1994 ) Personal Communications. British Oxygen Company, London

Boero, V. J., Eckenfleder, W. W, Jr. and Bowers, A. R. (1991) Soluble microbial product

formation in biological systems. Water Science and Technology. 23 (4-6), 1067-

1076.

Boon, A. G. and Wheatland, A. B. (1979) Aeration and oxygenation in sewage treatment-

some recent developments. Progress in Water Technology. 11 (3), 171-179.

Boon, A. G., and Lister, A. R. (1975) Formation of sulphide in rising main sewers and its

prevention by injection of oxygen. Progress in Water Technology. 7 (2), 289-300.

Boon, A. G., Skellett, C. F., Newcombe, S., Jones, J. G. and Forster, C. F. (1977) The use

of oxygen to treat sewage in a rising main. Water Pollution Control. 76 (1), 98-112.

Boon, A. G., and Skellett, C. F. (1979) Control of septicity by oxygen injection.

Proceedings IWPC conference on "Septic sewage: problems and solutions".

Maidstone, Kent, 102-111.

Boon, A. G. (1995) Septicity in sewers: causes, consequences and containment. Water

Science and Technology. 31(7), 237-253.

 REFERENCES 180

Bowker, R. P. G., Smith, J. M. and Webster, N. A. (1989) Odour and Corrosion Control

in Sanitary Sewerage Systems and Treatment Plants, Hemisphere Publishing Co.,

NY, USA.

Braha, A. and Hafner, F. (1986) Use of lab batch reactors to model biokinetics. Water

Research. 21(1), 73-81.

Cadee, K. (1984) Predicted oxygen transfer characteristics of riser pipes at wastewater

pumping stations. Water (Australia). 11(2), 23-27.

Cadee, K., Long, T. E., and Wain, R. E. (1984) Oxygen dissolving facilities: the Munster

pumping station. Water (Australia). 11 (2), 28-30.

Cao, Y. S. and Alaerts, G. J., Hooijmans, C. M., Lubberding, H. J. (1992) Using batch

reactor to study oxygen consumption and organics decomposition in drainage

systems with attached biofilm. Water Science and Technology. 26 (3/4), 683-692.

Cao, Y. S. (1994) Aerobic Heterotrophic Biodegradation in Polluted Drains and Sewers-

The drain and sewer as a dual-phase reactors, PhD thesis Delft University of

Technology, A. A. Balkema Pub. Rotterdam, Netherlands.

Cao, Y. S. and Alaerts, G. J. (1995) Aerobic biodegradation and microbial population in a

channel with suspended and attached biomass. Water Science and Technology. 31

(7), 181-189.

Capps, R. W., Mantelli, G. N., and Bradford, M. L. (1995) Design concepts for biological

treatment of industrial wastewater, Journal of Environmental Progress. 14 (1), 1-8.

Carrie, R., Clough, G. F., and Newcombe, S. (1982) Oxygen injection into a rising main ;

its effect on pumping. Water Pollution Control. 81 (3), 399-411.

 REFERENCES 181

Chapman, T. D., Matsch, L. C., and Zander, E. H. (1976) Effect of high dissolved oxygen

concentrations on activated sludge systems. Journal of Water Pollution Control

Federation. 48(11), 2486-2510.

Charnock, G. (1979) NWA shows the way to saving oxygen. New Civil Engineer. 9, 29-

30.

Chudoba, J. (1985) Quantitative estimation in COD units of refractory organic

compounds produced by activated sludge microorganisms. Water Research. 19 (1),

37-43.

Chudoba, P., Chevalier, J. J., Chang, J. and Capdeville, B. (1991) Effect of anaerobic

stabilisation of activated sludge on its production under batch conditions at various

So/X,, ratios. Water Science and Technology. 23 (4/6). 917-926.

Chudoba, P., Capdeville, B., and Chudoba, J. (1992) Explanation of biological meaning

of the S0/X0 ratio in batch cultivation. Water Research. 26 (3-4), 743-751.

Crossley, T. V and Macdonald, F. W. (1972) Air vs. oxygen in activated sludge process.

Public Works. 103 (12), 58-61.

Dack, P. J. and Nadebaum, P. R. (1989) Odour control on the Latrobe Valley Outfall

sewer. Water (Australia). 16 (1), 34-41 & 48.

Dawson, P. (1980) The effects of air in a rising main. Public Health Engineer. 8 (3), 107-

111 & 125.

Diagger, G. T., and Grady, C. P. L. Jr. (1982) The dynamics of microbial growth on

soluble substrates. Water Research. 16, 365-382.

Dold, P. L. (1990) Incorporation of biological excess phosphorus removal in a general

activated sludge model. Proceedings 13th International Symposium on Wastewater

Treatment, Montreal, Canada, 83-113.

 REFERENCES 182

Echeverria, E., Seco, A., and Ferrer, J. (1992) Study of the factors affecting activated

sludge settling in domestic wastewater treatment plants. Water Science and

Technology. 24 (4-5), 273-279.

Edeline, F. (1989) Submerged biological modules for augmenting the self-purification

capacity of streams. Tribune de l'Eau. 42 (540), 7-11.

ERC (1985) Aeration of industrial and municipal liquors. Electricity Research Council's

Technical Information. EC-4443/11.86.

ERC (1986) Technical Information, EC-4949/9.86.

Forrest, D. (1985) Preventing odour and corrosion problems in wastewater systems using

pure oxygen. Water Sewage and Effluent. 20, 5-9.

Gall, B., Takacs, I. and Patry, G. (1995) The effect of organic reactions in a collection

system on wastewater treatment plant performance. Water Science and Technology.

31 (7), 25-31.

Garrett, M .E. and Rio, P. D. (1980) Improved wastewater treatment for holiday resort

towns on the coast. Progress in Water Technology. 12 (4), 723-730.

Garret, M. E. (1984) Odour problems at effluent treatment plants. Water Services. 8

(1062), 338-339 & 341.

Garsdal, H., Mark, 0., Dorge, J. and Jepsen. S. E. (1995) Mousetrap: Modelling of water

quality processes and the interaction of sediments and pollutants in sewers. Water

Science and Technology. 31(7), 33-41.

Germirli, F., and Orhon, D. and Artan, N. (1991) Assessment of the initial inert soluble

COD in industrial wastewaters. Water Science and Technology. 23 (4-6), 1067-

1076.
 REFERENCES 183

Giona, A. R. and Annesini, M. C. (1979) Oxygen uptake rate in the activated sludge

process, Journal Water Pollution Control Federation. 51 (5), 1009-1016.

Gonzalez-Martinez, S. and Norouzian, M. (1984) Aerobic and anaerobic intermediate

storage of activated sludge: efficiency study. Journal of the Water Pollution Control

Federation. 5 (11),1173.

Gould, F. J. (1990) Upgrading of wastewater treatment plants using pure oxygen. Water

Science and Technology. 22 (7/8), 301-302.

Gray, N. F.(1989) The Biology of Wastewater Treatment. Oxford University Press, New

York.

Green, M., and Shelef, G. (1980) Sewer trunk lines can be treatment facilities. Water and

Sewage Works. 8, 26-27 & 52-53.

Green, M. and Shelef, G. (1981) Sludge viability in a biological reactor. Water Research.

15, 953-959.

Green, M. and Shelef, G. (1981) Step-fed plug flow reactor for biological wastewater

treatment versus continuous flow stirred tank reactor (CFSTR). Journal of

Environmental Sciences. 24 (4), 25-26.

Green, M., Shelef, G., and Messing, A. (1985) Using the sewerage system main conduits

for biological treatment. Water Research. 19 (8),1023-1028.

Greenfield, P. F., Dobson, K., and Jones, R. (1980) In-main Addition of Oxygen to

Sewage. Unpublished Report, Department of Chemical Engineering, University of

Queensland, Australia.

Grutsch, J. F. and Mallat, R. C. (1976) Optimise the effluent system-part 1: activated

sludge process. Hydrocarbon Processing. 3, 105-112.

 REFERENCES 184

Hancock, B. (1994) Personal Communications. Irvine Ranch Water Authority, California,

USA.

Haung, J. Y. C. and Mandt, M. G. (1978) Pure oxygen activated-sludge system transfers

oxygen rapidly. Water and Sewage Works. 125 (6), 98-100 & 102-103.

Hegemann, W. (1977) Recommendations on the application of pure oxygen in the

activated sludge process. Progress in Water Technology. 8 (6), 137-139.

Hemmings, R., Anderson, R., and Shaw, R. (1983) In-main oxygen treatment of domestic

sewage; boulder bay sewerage system. Water (Australia). 10 (2), 29-32.

Henze, M. (1992) Characterisation of wastewater for modelling activated sludge process.

Water Science and Technology. 25 (6), 1-15.

Holder, G. A. and Leow, J. M. (1994) In-sewer oxygenation of wastewater using venturi

side stream dissolvers. Proceedings 17th IAWQ biennial conference. Budapest,

Hungary, 163-172.

Hollerbach, G. H. Jr. (1985) Controlling sulphides in sewage by direct injection of

oxygen. Public Works. 116 (3), 82-83.

Horan, N. J. and Shanmugan, P. (1986) Effects of starvation and nutrient depletion on the

settling properties of activated sludge. Water Research. 20 (5), 661-666.

Horan, N. J. (1991) Biological Wastewater Treatment-Theory and Operation. John Wiley

& Sons, England.

Hutchinson, D. M., (1981) In-sewer Treatment with Oxygen. Unpublished Report,

Commonwealth Industrial Gases Ltd. Australia.

Hvitved-Jacobsen , T., Raunkjaer, K. and Nielsen, P. H. (1995) Volatile fatty acids and

sulphide in pressure mains. Water Science and Technology. 31 (7), 169-179.

 REFERENCES 185

Jeffries, C. (1983) Review of the use of commercial oxygen in wastewater treatment. The

Public Health Engineer. 11 (4), 39-43.

Jensen, N. A. (1995) Empirical modelling of air-to-water oxygen transfer in gravity

sewers. Water Environment Research. 67 (6), 979-991.

Kaijun, W., Zeeman, G. and Lettinga, G. (1995) Alteration in sewage characteristics upon

Aging. Water Science and Technology. 31(7), 191-200.

Kalinske, A. A. (1971) Effect of dissolved oxygen and substrate concentration on the

uptake rate of microbial suspensions. Journal Water Pollution Control Federation,

43 (1), 73-80.

Kameda, Y. and Odauchi, T. (1984) Preliminary study of air injection into force main.

Journal of Japan Sewage Works Association. 21 (237), 14-23.

Kappler, J. and Gujer, W. (1992) Estimation of kinetic parameters of heterotrophic

biomass under aerobic conditions and characterisation of wastewater for activated

sludge modelling. Water Science and Technology. 25 (6), 125-139.

Kesseler, L. H. and Nicholas, M. S. (1935) Oxygen utilisation by activated sludge.

. Sewage Works. Journal. 7(5), 810.

Kite, 0. A., and Garrett, M. E. (1983) Oxygen transfer and its measurement. Water

Pollution Control. 1, 21-28

Koch, C. M. and Zandi, I. (1973) Use of pipelines as aerobic biological reactors. Journal

of the Water Pollution Control Federation. 45 (12), 2537-2548.

Laughlin, J. E. (1964) Studies in force main aeration. Journal of Sanitary Engineering

Division, ASCE, 90 (5A6), 13-24.


REFERENCES 186

Lesouef, A., Payraudeau, M., Rogalla, F., and Kleiber, B. (1992) Optimising nitrogen

removal reactor configurations by on-site calibration of the IAWPRC Activated

Sludge Model. Water Science and Technology. 25 (6), 105-123.

Lin, S. H. and Huang, C. M. (1993) Effect of pressurised aeration on the activated sludge

processes. Journal of Environmental Science and Health. 28A, 1473-1482.

Lokkegaard, H., Havited-Jacobson, T., Teichgraber, B., and te Heesen, D. (1995)

Experimental procedures characterising the transformation of wastewater organic

matter in the Emscher River, Germany. Water Science and Technology. 31 (7), 201-

212.

Manandhar, U. K., and Schroder, H. (1995). Sewage circulating reactor-an approach to

recirculating wastewater in sewers. Environmental Technology. 16 (3), 201-212.

Marwood, P. (1984) Sulphide Control in Latrobe Valley Outfall Sewer by Oxygen

Injection, unpublished report, LVOSB, Australia,

Matos, J. S. and de Sousa, E. R. (1991) Dissolved oxygen in small wastewater collection

systems. Water Science and Technology. 23 (10-12), 1845-1851.

McKinney, R. E. (1970) Design and Operation of Complete Mixing Activated Sludge

Systems. Environmental Protection Control Service report. 1 (3), (quoted by

Benefield and Randall, 1980).

Metcalf and Eddy Inc. (1991), Wastewater Engineering: treatment, disposal and re-use,

3rd Edition (Edited by Tchobanoglous, G., and Burton, F. L) McGraw-Hill.

Metcalf, A. V. (1994) Statistics for Engineering-a practical approach. Chapman & Hall,

London.

Miller, M. A. (1978) Air vs. oxygen: two activated sludge systems compared. Water &

Waste Engineering. 15 (4), 58-60 and 62-65.

REFERENCES 187

Mitchell, R. C. (1973) U-Tube Aeration. USEPA Technology Transfer Series EPA 670/2-

73-031, Washington DC.

Moser, F., Wolfbauer, 0., and Taibinger, P., (1977) A novel aeration basin for biological

wastewater treatment. Progress in Water Technology. 8 (6), 235-237.

Mozell, E. (1985) Oxygen injection cuts odours. Water/Engineering and Management.

132 (1), 40-42.

Newbry, B. W., Diagger, G. T, Taniguchi-Dennis, D. (1988) Unit process trade-offs for

combined trickling filter and activated sludge processes. Journal Water Pollution

Control Federation. 60 (10), 1813-1821.

Newcombe, S., Skellett, C. F. and Boon, A. G. (1979) An appraisal of the use of oxygen

to treat sewage in a rising main. Water Pollution Control. 78 (4), 474-504.

Nielsen, P. H., Raunkjaer, K., Norsker, N. H., Jensen, N. A. and Hvited-Jacobson, T.

(1992) Transformation of wastewater in sewer systems-a review. Water Science and

Technology. 25 (6), 17-31.

Noone, G. P. (1995) Personal Communications. Severn Trent Water plc.

Oakley, H. R. and Preston, J. R. (1979) Development of septicity in sewerage systems.

Effluent and Water Treatment Journal. 12, 607-617.

Ogston, M. (1994) Personal Communications. CIG Ltd, Australia.

Ozer, A. and Kasigra, E (1995) Substrate removal in long sewer lines. Water Science and

Technology. 31 (7), 213-218.

Pescod, M. B. and Price, A. C. (1981) Fundamentals of sewer ventilation as applied to the

Tyneside sewerage scheme. Water Pollution Control. 80, 17-33

REFERENCES 188

Peters, G. M., Hamaide-deplus, M., Monaux, V. and Nuns, J. (1975) Quantitative

determination of the population and biological activity of biological sludges used in

the activated sludge wastewater treatment process. Water Research. 9, 901-911.

Pomeroy, R. (1959) Generation and control of sulphides in filled pipes. Sewage and

Industrial Wastes. 31 (9), 1082-1095.

Pomeroy R. D. and Parkhurst, J. D. (1972) Self purification in Sewers. Proceedings of the

6th International conference on Water Pollution Research. Ed: Jenkins, S. H.,

Pergammon Press.

Pomeroy, R. D. and Lofy, R. J. (1977) Feasibility Study on In-Sewer Treatment Methods.

National Technical Information Service, US dept. of Commerce, Springfield,

Report No. PB-271 445.

Pomeroy, R. D. (1992) The Problem of Hydrogen Sulphide in Sewers, (Edited by Boon,

A. G), 2nd Ed. , Clay Pipe Development Association, London.

Poon, C. P. C. and Wang, K. K (1972) Oxygenation and high rate biological treatment

process. Journal Water Pollution Control Federation. 44 (2), 265-273.

Pullin, J. (1977) Treatment in the sewer cuts the works' load. Surveyor. 150 (4440), 21-22

& 27.

Rachwal, A. J., Johnstone, D. W. M., Hanbury, M. J. and Critchard, D. J. (1982), The

application of settleability tests for the control of activated sludge plants. Bulking of

Activated Sludge (Ed. Chambers, B and Tomlinson, E. J), Ellis Horwood Ltd.

Chichester.

Randall, E. W., Wilkinson, A. and Elcma, G. A. (1991) An instrument for the direct

determination of oxygen utilisation rate. Water SA. 17 (1), 11-18.

REFERENCES 189

Raunkjaer, K. (1993) Characterization and transformation of wastewater organic

material in sewer systems, PhD thesis, Aalborg University, Denmark

Raunkjaer, K., Havited-Jacobson, T. and Nielsen, P. H. (1995) Transportation of organic

matter in a gravity sewer. Water Environment Research. 67 (2), 181-188.

Reynolds, T. D. (1982) Unit Operations and Processes in Environmental Engineering

B/C Engg. Div. Massachusetts.

Rickard, M. D., and Gaudy, A. F. (1968) Effect of oxygen tension on 0 2 uptake and

sludge yield in completely mixed heterogeneous populations. Proceedings 23rd

Industrial Waste Conference, Purdue University. Ext. set. 132, S83-893.

Rickert, D. A. and Huneter, J. V. (1971). Effects of aeration time on soluble organics

during activated sludge treatment. Journal Water Pollution Control Federation. 43

(1), 134-138.

Rudolph, K. U. (1981) Aeration of pressure mains for elimination of corrosion and foul

odours. Korrespondenz Abwasser. 28 (11), 789-794.

Sawyer, C. N., McCarty, P. L and Parkin, G. F. (1994) Chemistry for Environmental

Engineering. 4th Ed. McGraw Hill.

Scherfig, J. (1994) Personal Communications. Department of Civil Engineering,

University of California, Irvine, USA

Schulze, K. L. and Kooistra, R. D. (1969) Oxygen demand and supply in an activated

sludge plant. Journal Water Pollution Control Federation. 41 (10), 1763-1774.

Sekoulov, I. and Muller, N. (1986) New method of oxygen injection into sewer pipelines

with the aid of non-porous membranes. GWF-Wasser/Abwasser. 127 (12), 646-649.

REFERENCES 190

Sewards, G. J and Carpenter, A. N. (1989) Development of a long term corrosion and

odour control strategy in sewerage system. Proceedings Australian Water and

Wastewater Association, 13th Federal Convention, Canberra. 428-432.

Sewards, G. J. and Barkley, D. S. (1992) Update on odour control in Geelong's

wastewater system. Water (Australia). 6, 24-25 & 29.

Shao, Y. J., Crosse, J., Keller, E., and Jenkins, D. (1992) High rate air activated sludge

operation at the city of Los Angeles Hyperion Wastewater Treatment Plant Water

Science and Technology. 25 (4-5), 75-87.

Shaw, R. G. (1981) Chemical engineering advances in WWT. Water Engineering and

Management. 128 (9), 86-92.

Shelef, G. and Green, M. (1980) Kinetics of pure oxygen activated sludge. Journal of

Environmental Engineering Division, ASCE. 106 (EE6), 1099-1110.

Sidwick, J. M. and Lewandowslci, T. P. (1975) An economic study of the Unox and

conventional aeration system. Water Pollution Control. 74, 645-654.

Simpson, J. R. (1979) Trends in Treatment and Disposal of Sewage. Effluent and Water

Treatment Journal. 8, 619-626.

Sollfrank, U., Kappler, J. and Gujer, W. (1992) Temperature Effects on Wastewater

Characterisation and the Release of Soluble Inert Organic Material. Water Science

and Technology. 25 (6), 33-41.

Speece, R. E (1969) U-Tube strewn re-aeration. Public Works. 100 (8),111-113.

Speece, R. E., Adams, J. L., and Wooldridge, C. B., (1969) U-tube aeration operating

characteristics. Journal of Sanitary Engineering, ASCE. 95 (SA3), 563-574.

Speece, R. E. and Orosco, R. (1970) Design of u-tube aeration system. Journal of

Sanitary Engineering, ASCE. 96 (SA3), 715-725.

REFERENCES 191

Speece, R. E, Gallagher, D., Krick, C. and Thompson, R. (1981) Pilot performance of

deep u-tubes. Water Science and Technology. 13 (1), 395-407.

Speece, R. E., Nirmalakhandan, N. and Tchobanglous, G. (1990) Commercial oxygen use

in water-quality management. Water Environment & Technology. 2 (7), 54-61.

Standard Methods for Examination of Water and Wastewater (1989) 17th Ed. APHA,

AWWA and WEF, New York, N.Y.

Stoyer, R. and Scherfig, J. (1972) Wastewater treatment in a pressure pipeline. The

American City. 87 (10), 84,86,90 & 93.

Stoyer, R. L. (1970) Pressure pipe wastewater treatment system. Proceedings 2nd Annual

Sanitary Engineering Research Laboratory Workshop on Wastewater Reclamation

and Use. Tahoe City, California, 45-47.

Sundstrom, D. W. and Klei, H. E. (1979) Wastewater Treatment. Prentice-Hall, USA

Tanaka, N. and Takenaka, K. (1995) Control of hydrogen sulphide and degradation of

organic matter by air injection into wastewater force main. Water Science and

Technology, 31(7), 273-282.

Templeton, L. L. and Grady, C. P. L. (1988) Effect of culture history on the determination

of biodegradation kinetics by batch and fed-batch techniques. Journal Water

Pollution Control Federation. 60 (5), 651-658.

Tench, H. B. (1968). Sludge activity and the activated sludge process, Water Pollution

Control. 67, 408-420.

Tench, H. B. (1994) A theory of the operation of full scale activated sludge plants. Water

Research. 28 (5), 1019-1024.

Thistlethwayte, D. K. B. (1972) The Control of Sulphides in Sewerage Systems.

Butterworths, Sydney.
 REFERENCES 192

Thomas, 0., Blake, G. , Wittenburg, M., Mazas, N. and Oudari, I. (1985) A new role of

sewers : purification in situ. Tribune du Cebedeau. 38 (499/500), 13-19.

Tom Maguire (1986) The Maguire V02 Venturi System. Tom Maguire & Co. Ltd. UK.

Toms, R. G. and Booth, M. G. (1982) The use of oxygen in sewage treatment. Water

Pollution Control. 81(2), 151 - 165.

Tucek, F. and Chudoba, J. (1969) Purification efficiency in aeration tanks with complete

mixing and piston flow. Water Research. 3, 559-570.

Tucek, F., Chudoba, J. and Madera, V. (1971) Unified basis for design of biological

aerobic treatment processes. Water Research. 5, 647-680.

UNUT (1994) Preliminary study on in-sewer treatment of wastewater-on behalf of

Anglinan Water Services Ltd. Department of Civil Engineering, University of

Newcastle upon Tyne, UK.

Upton, J., and Norman, M., and Crabtree, H. E. (1985) The Vitox process : its oxygen

utilisation efficiency and the optimal mode of deployment in the uprating context.

Tribune du Cebedeau. 38 (501), 29 - 45.

USEPA (1974) Process Design Manual for Sulphide Control in Sanitary Sewerage

Systems. EPA 625/1 - 74 - 005, Technology Transfer, Washington DC.

USEPA (1985) Odour and Corrosion Control in Sanitary Sewerage Systems and

Treatment Plants. EPA 625/1-85/018, Washington DC.

Vesiland, P. A (1971) Shearing effect on settling of activated sludge. Journal of Sanitary

Engineering Division, ASCE, 97 (SA4), 544-547.

Vivona, M. A. (1979) Force main odour control by air injection. Public Works. 110 (12),

70-72.
REFERENCES 193

Vivona, M. A. (1980) Designing for force main odour control. Public Works. 111 (7), 74-

76.

Vivona, M. A. and Whalen, G. A. (1982) Controlling sulphides and odours in sewers: part

1. Public Works. 113 (3), 73-76.

Vivona, M. A. and Whalen, G. A. (1982) Controlling sulphides and odour in sewers: part

2. Public Works. 113 (4), 69-72.

Vosloo, P. B. B. (1973) Oxygen requirements in the activated sludge process. Water

Pollution Control. 72, 209-212.

Weinzaepfel, B. and Rabergeau, G. (1991) The use of pure oxygen to reduce the oxygen

deficit of industrial and urban wastewaters; case studies. Eau, Industrie Nuisances.

150, 83-85.

Wernicke, J. C. (1987) Sewers get a breath of fresh air. Civil Engineering. 8, 67-69.

White, J. B. (1987) Wastewater Engineering. 3rd ed. Edward Arnold, London.

Williamson, K. J. and Nielsen, P. 0. (1981) Influence of dissolved oxygen on activated

sludge viability. Journal of Water Pollution Control Federation. 53 (10), 1533-

1540.

Wolfbauer, 0., Klettner, H., and Moser, F. (1978) Reaction engineering models of

biological wastewater treatment and the kinetics of the activated sludge process.

Chemical Engineering Science. 33, 953-960.

 194

to..
1.•n•
Om,
..Le:

‘..0
N
 195

APPENDIX A-2

Experimental set-up for Phase-II studies

1. Batch reactor; 2. Impeller; 3. Oxygen cylinder; 4. Do meter; 5. SOUR reactor; 6.

Magnetic mixer; 7. Settling column.


196

..•n•••

+ + 0
nn••nnn

0
ILl •3c
.-
)
U 0
+ +0
u)
+ o
H H ci")
C.) o 0
•zr (-1 0
r.74
1-4
al 0
0 ••n••

1-1 fai
Z + +0 + In
O 0

• 41
0U)
Z 4 •n•n••

> %.0 r) N VD 0 > VD N CO N

41 0 t.0 I—I N ...... • 03 41 •;I,in (-) co H
A mp r-I En I.- OV "Zr • 0 03 •zt. cv co fa
Ei • • • • • s-- r.n CA 0 pi • • • • >H
u) r) .1 CV •zr ... Z LOCO u] 1-1 GI 01 03 0 fa
U) H >
ai En 0)0
I $4
. g a)
a $4
Z o cn r- r-• (-1 . z o r- z el en re) r) H 0
Cf) o\c, 0 Ca • • • 4 0 VI \t, l0 cr ..--.. 0 • rn • F:C r1 t'l C'l r) 0) U U)
G l 0 CI tO tO
i CO ov, N • CO 41 ci ci r) co al u) CO
z • • • • ...- 41 co co Qo (N z • • • • .
H N c0 H Cl 0 C/3 N •:1' cr cr r- •:/, N co II II
Tr •:14 r) •zr Ci] Z v—I %Co N N r- r- .:14
rn 4 0 En
I H QC/3
P4 IX C.) I
II 4
> 44 ri Z Cl co Cl Cl
II0 g
P4
Z Cl Cl Cl Cl
> >1 0 r) co i-i
44 ca ILl —-I
n 3 o
El a) El '-
U) CI)
0 H• •
O 0000 0 u) 41 140000
4.1tn000 41 A >10 f24 1-4 41ul000 u)
> r-i c-N1 r) 1-1 1-1 IX 0 4 > 1-4 CN1 c•-) a)
41 0 III 4 04 El 41 0 -ii
1-4 0 P ZOki240 0 0
ai Z 4 U) 41 cIl Z
 197

cz
CD

..•n•••
0
çi

••n•n.

0 +0 0
Cx-1 >
Tr4 0 Cr-1
U) Cr)
E-I
rx.T Cr) H U]
0 Sa4 H a .4,
cs%
E-1 • :1* o\c, r-1 rxe fra
1-1 lr) 1-1
O al 0 C\0
ft4 0 0
(N CT-4 a4 O4
41
Z Z
c=4 00 •nn••
CLI Nzr 00

X X tO r-1
H
0 H rx) H
U) 0 cn
41 0 Z + +0 Z ...-.
1-i I....-T I. H 0:1
0
4 Cl) r-t
• cn 0 r- > r- CI . C/1 l0 0 > CNI IN
. ---- • • rx/ r- cn Z • • 14 o T-1
r=1 (1) o\o c0 C.- 0 cr 0 . - 0 a) CI (-) .1
E-1 C.) --- ,--I (•1 E-1 • •
a) cka Tr r-I El • •
0 0) In LC) 0 ..... U) zr
• '4'
0 }-1
0 C..) 0 U)
Cl] Cl) 0 Cr)
.. Z II) g
.
•• ZU)000 ,Z00 0 Z U) k0 01 1/40 Z r--Ln CO
Cr.4 -•-n QC/) • • • 4 0 0 k) . Or.!) • 4 0 •H• Ln
0
0\0 03 V3 Tr 41 0 0 ni 0 el •zr Cs) tO r-1 cn
s•-• [4 H to co Z • • • 0\0 çm] cr 1-1 lfl
T-1 T-I
z • •
0 0 r-i r-4 t.0 ON tn ....... C.) Ln 0 •zr
0 Z el el Z k0 N
U p:4 Cl) FtC
Crl H Ul H
g g g g
Cr-1 •• 4 II . 4 II
> rx-T H LO C.-- z •zr .4. > 1=4 H 1/40 r- Z •zr •cr
>, 0 > >I 0 >
(a P. Cr.1 r a r
C -1 41

3 0 A 3 0 0
a) E-1 a) E-I
Cr) CI) Z VI cn
0 H 0 H
cr) W al 44 r-1 C•1 0 Ul 41 W 1-1 H C4 0
A >400
0 g 1-1 cz1 CLI >4 C.) 0 g ;4
A w pz
i- g $4 0 0'4 > 1-1 )-1 g 0 4 > $-1
CO ex4 gE t rxl 0 0] 4 g E l p.i 0
E-4 Z0 0O i-1 0 E-I Z 0 0 g 0 L-1 0
X 4 ri) Ul ft.1 E-n al X 4 U] u) rx7 E-4 P-1
 198

APPENDIX C-1

SCOD removal efficiencies at various retention times for unseeded wastewater at

20°C.

WWTP SS0 SCOD0 RSCOD RSCOD RSCOD RSCOD RSCOD RSCOD

lhour 2 hours 3hours 4hours 6hours 8hours
Sacriston 230 275 15 10 25 40 80 110
Sacriston 380 380 20 45 65 80 105 120
Sacriston 380 390 50 65 90 115 155 190
Sacriston 400 385 30 50 65 95 130 145
Sacriston 310 375 30 60 50 80 125 155
Sacriston 500 365 25 30 60 65 125 145
Sacriston 470 300 30 50 80 100 100 110
Sacriston 310 290 20 45 65 75 135 135
Sacriston 380 340 40 60 70 95 120 130
Sacriston 240 305 15 40 55 80 105 110
Sacriston 280 255 20 30 40 65 90 100
Sacriston 270 290 40 50 80 95 105 115
Durham 540 250 65 85 95 120 125 165
Durham 660 320 75 80 125 140 155 195
Durham 640 310 60 100 85 130 150 195
Durham 630 295 90 105 140 150 155 190
Durham 740 315 80 105 135 140 160 185
( 110 190 23 10 (
Durham 580 350 20 90 130
Durham 550 300 55 95 115 90 140 145
Broomhaugh 420 160 70 60 75 75 90 85
Broomhaugh 340 215 35 25 65 20 30 95
Broomhaugh 300 175 20 40 50 65 70 95
Broomhaugh 160 70 -5 10 15 10 15 15
Broom haugh 330 200 45 60 75 80 80 100
Broomhaugh 410 155 35 45 45 60 70 80
Broom haugh 400 185 10 30 35 30 55 70
Broomhaugh 270 165 15 25 35 45 55 75
Notes: (i) all units in mg/I
(ii) RSCOD = SCOD removal after a specified retention period
 199

APPENDIX C-2

Results of regression Analysis of the data shown in Appendix C-1

Notes: A variable is not significant if p> 0.05

(a) Retention time = 1 hour

The regression equation is: RSCOD = - 9.0 + 0.126 SS ° - 0.0204 SCODo

Predictor Coef Stdev t-ratio p

Constant -8.96 12.00 -0.75 0.463
SS() 0.12571 0.02286 5.50 0.000
SCOD0 -0.02036 0.04064 -0.50 0.621

s = 16.09 R-sq = 57.9% R-sq(adj) = 54.4%

(b) Retention time = 2 hours

The regression equation is : RSCOD =-16.6-1- 0.151 SS° +0.0353SC0D0

Predictor Coef Stdev t-ratio p

Constant -16.58 11.32 -1.46 0.156
SS0 0.15070 0.02156 6.99 0.000
SCOD0 0.03532 0.03833 0.92 0.366

s = 15.17 R-sq = 72.6% R-sq(adj) = 70.3%

(c) Retention time = 3 hours

The regression equation is : RSCOD = - 18.1 + 0.177 SS ° + 0.0658 SCODo

Predictor Coef Stdev t-ratio P

Constant -18.11 13.37 -1.35 0.188
SS0 0.17677 0.02546 6.94 0.000
SCOD0 0.06584 0.04526 1.45 0.159

s = 17.92 R-sq = 73.7% R- sq (adj) = 71.5%

(d) Retention time = 4 hours.

The regression equation is : RSCOD = - 30.4 + 0.169 SS0 + 0.169 SCODo

Predictor Coef Stdev t-ratio P

Constant -30.44 16.86 -1 . 81 0.083
SS0 0.16880 0.03210 5.26 0.000
SCOD0 0.16924 0.05708 2.97 0.007

s = 22.59 R-sq = 70.0% R-sq(adj)-= 67.5%

 200

APPENDIX C-2-continued

Results of regression Analysis of the data shown in Appendix D-1

(e) Retention time = 6 hours

The regression equation is: RSCOD = -31.7 + 0.136 SS° + 0.304 SCODo

Predictor Coef Stdev t-ratio p

Constant -31.72 14.43 -2.20 0.038
SS 0 0.13648 0.02749 4.97 0.000
SCOD0 0.30393 0.04887 6.22 0.000

s = 19.34 R-sq = 80.7% R-sq(adj) = 79.1%

(f) Retention time = 8 hours

The regression equation is : RSCOD = - 32.0 + 0.175 SS () + 0.324 SCODo

Predictor Coef Stdev t-ratio p

Constant -32.01 15.84 -2.02 0.055
SS() 0.17477 0.03016 5.79 0.000
SCOD0 0.32445 0.05362 6.05 0.000

s = 21.22 R-sq = 82.4% R-sq(adj) = 80.9%

 201

APPENDIX C-3

Rates of Reaction (Benfield and Randall, 1985)

1- Zero-order reaction

C - Co = - kt

2- First-order reaction

C = Co e-kt

3- Second-order reaction

1 1
Z' — —
Co = kt

Where:

C = concentration of reactant at time t

Co = the constant of integration which is calculated as C = Co at t = 0
k = the reaction rate constant
 202

APPENDIX D-1

SCOD Removal efficiencies at an added seed concentration range of 0-1000 mg/I

' SS° SCOD0 RSCOD RSCOD RSCOD - RSCOD RSCOD RSCOlY seed
1 hour 2 hours 3 hours 4 hours 6 hours 8 hours con. (B)
540 250 65 85 95 120 125 165 0
660 320 75 80 125 140 155 195 0
640 310 60 100 85 130 150 195 0
630 295 90 105 140 150 155 190 0
740 315 80 105 135 140 160 185 0
580 350 20 90 130 170 190 230 0
550 300 55 95 115 90 140 145 0
300 175 20 40 50 65 70 95 0
280 255 20 30 40 65 90 100 0
270 290 40 50 80 95 105 115 0
330 200 45 60 75 80 80 100 0
230 275 15 10 25 40 80 110 0
410 155 35 ' 45 45 60 70 80 0
270 165 15 25 35 45 55 75 0
380 390 50 65 90 115 155 190 0
400 385 30 50 65 95 130 145 0
310 375 30 60 50 80 125 155 0
500 365 25 30 - - 125 145 0
470 300 30 50 80 100 100 110 0
310 290 20 45 65 75 135 135 0
380 340 40 60 70 95 120 130 0
240 305 15 40 55 80 105 110 0
420 160 70 60 75 75 90 85 0
160 70 -5 10 15 10 15 15 0
400 185 10 . 30 35 30 55 70 0
340 215 35 25 65 20 30 95 0
740 310 120 150 155 155 200 210 100
580 340 50 120 170 170 210 255 100
550 300 75 115 125 150 180 180 100
390 180 30 45 50 65 75 95 100
390 370 50 50 85 120 145 180 100
380 380 90 110 160 180 215 245 100
350 210 70 95 95 110 120 130 100
470 290 90 85 135 135 170 185 100
440 305 70 90 115 110 145 160 100
740 310 115 140 175 185 215 225 250
580 340 135 165 175 210 200 240 250
550 300 85 125 140 160 195 185 250
390 180 55 70 80 105 100 105 250
390 370 70 95 150 170 185 215 250
380 380 90 150 170 205 220 260 250
350 210 105 115 125 125 135 135 250
470 290 _ 125 140 160 150 175 180 250
 203

APPENDIX D-1-continued

SCOD Removal efficiencies at an added seed concentration range of 0-1000 mg/1

SSO SCCADO RSCCH) RSCCM) Rsccn) RSCOD RSCOD RSC013 seed

1hour 2houn 3hours 4houn 6hours 8hours con. (B)
440 300 120 130 160 145 185 180 250
400 355 70 135 170 200 225 230 500
310 355 150 180 190 220 240 255 500
500 325 125 160 185 180 230 225 500
350 205 120 125 125 125 125 143 500
470 250 120 140 150 155 160 165 500
440 300 130 160 170 170 200 180 500
400 345 105 160 200 210 230 250 750
310 345 180 180 225 225 240 255 750
500 305 140 160 185 190 205 215 750
350 200 120 120 120 120 125 130 750
440 295 155 175 160 180 185 170 750
290 310 180 210 210 195 210 210 750
680 350 195 215 225 235 230 350 1000
530 355 150 225 230 230 255 260 1000
560 195 * 115 115 120 120 125 1000
520 390 235 250 270 260 270 260 1000
520 350 205 240 235 245 250 250 1000
480 230 125 140 155 115 140 135 1000
Note: All units in mg/.
 204

APPENDIX D-2

Results of multiple regression analysis of the data shown in Appendix D-1

The regression equation is

RSCOD = - 110 + 0.479 SCOD0 + 0.0457 SS 0 + 17.2 t + 11.2 b - 1.11 bt

Predictor Coef Stdev t-ratio P

Constant -109.65 12.76 -8.60 0.000
scod0 0.47904 0.02989 16.03 0.000
ss 0.04569 0.01650 2.77 0.006
t 17.170 1.514 11.34 0.000
b 11.156 1.085 10.28 0.000
bt -1.1117 0.2359 -4.71 0.000

s = 26.40 R-sq = 76.7% R-sq(adj)= 76.2%

 205

APPENDIX D-3

Results of multiple regression analysis of SOUR data (Figures 7.10-7.14)

(i) Seed concentration = 100 mg/1

The regression equation is

SOUR = 10.4 - 0.868 t + 0.0115 SS° - 0.0176 SCOD0

Predictor Coef Stdev t-ratio p

Constant 10.390 5.880 1.77 0.087
t -0.8682 0.4571 -1.90 0.067
SS0 0.011479 0.008585 1.34 0.191
SCOD0 -0.01758 0.01973 -0.89 0.380

s = 5.959 R-sq = 15.2% R-sq(adj) = 7.3%

(ii) Seed concentration = 250 mg/1

The regression equation is

SOUR = 7.01 - 1.63 t + 0.0006 SCOD0 + 0.0134 SSo

Predictor Coef Stdev t-ratio p

Constant 7.011 5.001 1.40 0.169
t -1.6266 0.4643 -3.50 0.001
SCOD0 0.00060 0.01765 0.03 0.973
SS0 0.013443 0.008842 1.52 0.137
s = 6.267 R-sq = 29.6% R-sq(adj) = 24.1%

(iii) Seed concentration = 500 mg/1

The regression equation is

SOUR = 25.2 - 3.10 t - 0.0089 SCOD0 - 0.0036 SSo

Predictor Coef Stdev t-ratio p

Constant 25.19 10.85 2.32 0.031
t -3.0982 0.6718 -4.61 0.000
SCOD0 -0.00894 0.02280 -0.39 0.699
SS0 -0.00362 0.02103 -0.17 0.865

s = 6.756 R-sq = 51.9% R-sq(adj) = 44.7%

 206

APPENDIX D-3-Continued

Results of multiple regression analysis of SOUR data (Figures 7.10-7.14)

(iv) Seed concentration = 750 mg/I

The regression equation is

SOUR = 28.1 - 4.43 t- 0.0017 SCOD0 + 0.0016 SSo

Predictor Coef Stdev t-ratio p

Constant 28.10 16.79 1.67 0.108
t -4.4347 0.9793 -4.53 0.000
SCOD0 -0.00172 0.03893 -0.04 0.965
SS0 0.00159 0.03101 0.05 0.960

s = 11.38 R-sq = 48.5% R-sq(adj) = 41.5%

(v) Seed concentration = 1000 mg/I

The regression equation is

SOUR = 41.0 + 0.0872 SCOD 0 - 0.0182 SS ° - 8.27 t

Predictor Coef Stdev t-ratio p

Constant 41.03 25.00 1.64 0.111
SCOD0 0.08722 0.03131 2.79 0.009
SS() -0.01817 0.04545 -0.40 0.692
t -8.266 1.290 -6.41 0.000

s = 16.42 R-sq = 61.2% R-sq(adj) = 57.3%

 207

APPENDIX E-1

SCOD results for Greater Amman wastewater

t SS0 SCOD0 RSCOD b bt

4 750 410 160 1.0 4
4 800 425 215 1.0 4
4 790 370 200 1.0 4
4 1060 375 155 1.0 4
4 750 400 220 2.0 8
4 800 415 215 2.0 8
4 790 360 210 2.0 8
4 1060 370 165 2.0 8
4 710 355 230 3.5 14
4 960 350 210 3.5 14
4 1030 350 170 3.5 14
4 720 310 190 3.5 14
4 710 330 210 6.0 24
4 960 335 239 6.0 24
4 1030 335 208 6.0 24
4 720 295 198 6.0 24
8 750 410 225 1.0 8
8 800 425 250 1.0 8
8 790 370 215 1.0 8
8 1060 375 250 1.0 8
8 750 400 270 2.0 16
8 800 415 265 2.0 16
8 790 360 240 2.0 16
8 1060 370 250 2.0 16
8 710 355 215 3.5 28
8 960 • 350 210 3.5 28
8 1030 350 195 3.5 28
8 720 310 200 3.5 28
8 710 330 215 6.0 48
8 960 335 225 6.0 48
8 1030 335 200 6.0 48
8 720 295 199 6.0 48
 208

APPENDIX E-2

Regression analysis of data shown in Appendix E-1

The regression equation is:

RSCOD = - 44.3 + 0.464 SCOD0 - 0.0342 SS° + 16.7 t + 25.1 b - 3.20 bt

Predictor Coef Stdev t-ratio p

Constant -44.29 70.08 -0.63 0.533
SCOD0 0.4645 0.1549 3.00 0.006
SS0 -0.03424 0.02572 -1.33 0.195
t 16.706 3.309 5.05 0.000
b 25.096 6.199 4.05 0.000
bt -3.2010 0.9070 -3.53 0.002

s = 19.33 R-sq = 60.0% R-sq(adj) = 52.3%

View publication stats