Modeling Drinking Water Disinf

INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. U M I

films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the

copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send U M I a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by

sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced

xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact U M I directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Graduate School Form 9
(Revised 7/94)
PURDUE UNIVERSITY
GRADUATE SCHOOL
Thesis Acceptance
This is to certify that the thesis prepared
B y _______ Chun-Ming Chen_______________________________________________________
Entitled
Modeling Drinking Water Disinfection in Ozone Bubble-Diffuser

Contactors
Complies with University regulations and meets the standards of the Graduate School for originality
and quality
For the degree o f ______ Doctor o f Philosophy______________________________________
Signed by the final iniog commu
, chair
Approved by:
Department Head Date
CD is
This thesis E is not to be regarded as confidential.
Format Approved by:
________________________________________
Chair. Final Examining Committee
Z Format Advi
M ODELING D R IN K IN G W ATER DISINFECTION
IN OZONE BUBBLE-DEFFUSER CONTACTORS
A Thesis
Submitted to the Faculty
of
Purdue University
by
Chun-Ming Chen
In Partial Fulfillment o f the
Requirements for the Degree
of
Doctor o f Philosophy
August 1998
UMI Number: 9914465
UMI Microform 9914465

Copyright 1999, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized

copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
DEDICATED TO
My Dearest Friend, Foon-Yee Chong
M y Parents, Ping-Tsan Chen and Li-Ying Shien
ACKNOWLEDGMENTS
I thank all my committee members: Dr. Benito J. Marinas, Dr. Dennis A. Lyn, Dr.
Ronald F. Wukasch, Dr. Robert B. Jacko and Dr. Joseph F. Pekny for their support and
helpful suggestions. I would like to extend special thanks to my major advisor, Dr.
Marinas, and co-advisor, Dr. Lyn, for their encouragement, guidance, and support
throughout the period o f this study. I am also indebted to Dr. Chung-Fan Chiou for
building the pilot-scale contactor, and former environmental engineering students, Foon-
Yee Chong and Sharon Waller for their experimental assistance.
I would like to acknowledge The Metropolitan Water District of Southern
California, Purdue University Physical Plant, The Risk Reduction Engineering Laboratory
of the U.S. Environmental Protection Agency, and Camp Dresser & Mckee Inc. for their
financial assistance during the period of this research.
I cannot thank enough to my best friend ever, Foon-Yee Chong, she has provided
a tremendous amount o f support during my studies, and make a new colorful and
meaningful era of my life.
TABLE OF CONTENTS
Page
TABLE OF C O N TEN TS................................................................................................... iv
LIST OF TA B LES..............................................................................................................ix
LIST OF FIGURES.............................................................................................................xi
LIST OF SYMBOLS..........................................................................................................xv
ABSTRACT.......................................................................................................................xix
Chapter 1. IN TR O D U C TIO N ..............................................................................................1

II. Background and Problems.................................................................................. 1
1.1.1. Assessment o f Ozone Disinfection Efficiency............................................ 2
1.1.2 Application in Full-Scale Ozone Disinfection........................................... 3
1.2. Research Objectives.............................................................................................4
1.2.1. Model Development................................................................................... 4
1.2.2. Analysis o f Major Model Parameters......................................................... 5
1.2.3. Model Verification......................................................................................5
1.2.4. Reactor Design Optimization......................................................................5
Chapter 2. LITERATURE R E V IE W ................................................................................. 7

2.1. Regulations...........................................................................................................7
2.1.1. Treatment Requirements........................................................................... 8
2.1.2 CT Concept................................................................................................10
2.1.2.1. Ozone Concentration (C ).............................................................. 11
2.1.2.2. Contact Time ( T ) ........................................................................... 12
2.2 G iardia and Cryptosporidium........................................................................... 14
2.3. Ozone Contactors................................................................................................15
2.4 Contactor Hydrodynamics and Tracer Tests.......................................................17
2.4.1. Tracer Tests...............................................................................................18
2.4.2. Axial Dispersion M odel........................................................................... 20
2.4.3. Studies on Reactor Hydrodynamics.........................................................23
2.5. Ozone Mass Transfer..........................................................................................26
2.5 .1. Mass Transfer Kinetics.............................................................................27
2.5.2. Mass Transfer Coefficient KL................................................................... 31
2.5 3. Average Bubble Diameter and Terminal Rise Velocity........................ 33
2.5.4. Specific Interfacial A rea...........................................................................36
2.5.5. Ozone Solubility........................................................................................36
2.6. Reactivity of Aqueous Ozone.............................................................................38
2.6.1. Ozone Decomposition..............................................................................38
2.6.2. Ozone Reaction with Other Compounds.................................................40
2.6.3. Reactions in Natural Waters and Ozone Demand Concept................... 43
2.7. Ozone Disinfection.............................................................................................. 47
2 .7 .1. Experimental Methods.............................................................................50
2.7.2. Disinfection Kinetics and Models............................................................51
2.7.3. Parameters Affecting Disinfection Efficiency......................................... 55
2.8. Ozone Concentration Measurements................................................................. 57
2.8.1. Iodometric Method.................................................................................. 57
2.8.2. Ultraviolet Absorption Method................................................................ 58
2.8.3. Amperometric Method.............................................................................58
2.8.4. Indigo Trisulfonate Methods................................................................... 59
2.8.4.1. Indigo Trisulfonate Batch Method................................................59
2.8.4.2. Automated Indigo Method............................................................60
2.8.4.3. Modified Indigo Method............................................................ 60
2.9. Ozonation Models.............................................................................................. 61
2.10. Strength and Weakness o f Using C T .............................................................. 66
Chapter 3. A X IA L DISPERSION M O D EL................................................................... 69

3.1. Introduction........................................................................................................69
3.2. Ozone Distribution.............................................................................................70
3.2.1. Model for Ozonation Chambers.............................................................. 73
3.2.2. Model for Reactive Chambers................................................................ 78
3.3. Tracer Tests........................................................................................................79
3.4. Mass-Transfer.....................................................................................................81
3.5 Batch Ozone Decomposition............................................................................ 82
3.6. Disinfection....................................................................................................... 82
3.6.1. CT-Lag Model for Giardia Cyst Inactivation........................................ 82
3.6.2. Axial Dispersion Inactivation Model....................................................... 85
Chapter 4. M ATERIALS AND M ETHODS.................................................................. 89

4.1. General Use Chemicals and Equipment............................................................ 89
4.2. Analytical Methods and Experimental Equipment............................................ 90
4.2.1. Non-Standard Methods...........................................................................90
4.2.1.1. Tracer Concentration Measurement............................................. 90
4.2.1.2. Ozone Concentration Measurement............................................. 91
4.2.1.3. Bubble Size Measurement............................................................ 93
4.2.2. Standard Methods................................................................................... 94
4.3. Pilot-Scale Experiments.................................................................................... 96
4.3.1. Experimental Apparatus..........................................................................96
4.3.2. Experimental Matrix.............................................................................. 100
4.3.3. Ozonation Tests.....................................................................................102
4.3 .4. Tracer Tests........................................................................................... 103
4.3.5. Mass Transfer Tests.............................................................................107
4.3 .6. G iardia M u ris Inactivation Tests...........................................................110
4.4. Bench-Scale Ozone Decomposition Tests..................................................... I l l
4.4.1. Initial Fast Ozone Demand Tests..........................................................112
4.4.2. Ozone Self-Decomposition Tests.......................................................... 115
4.4.2.1. Syringe Indigo Method...............................................................116
4.4.2.2. Stopped Flow Method................................................................. 118
Chapter 5. THEORETICAL ANALYSIS OF A X IA L DISPERSION MODEL 121

5.1. Role of Contactor Hydrodynamics...................................................................121
5.2. Role of Mass-Transfer..................................................................................... 128
5.2.1. Effect o f Mass Transfer on Ozone ConcentrationProfiles.....................128
5.2.2. Effect o f Pressure Changes along the Water Column...........................132
5.3. Significance o f Ozone Reaction........................................................................136
Chapter 6. PILOT-SCALE EXPERIMENTAL RESULTS AND DISCUSSION 142

6.1. Bubble Size Distribution.................................................................................. 142
6.1.1. Method Verification...............................................................................144
6.1.2. Results o f Bubble Size Measurement.....................................................145
6.2. Contactor Hydrodynamics................................................................................ 155
6.3. CSTR-ADR-CSTR model................................................................................ 168
6.4. Ozone Mass-Transfer........................................................................................170
6.5. Ozone Decay Kinetics...................................................................................... 175
6.6. Ozone Concentration Profiles...........................................................................183
6.6.1. Significance o f the Non-Ideal Conditions.............................................. 183
6.6.1.1. Effect of Bubble Size on Ozone Concentration Profile.............183
6.6.1.2. Effect o f Local High Mixing........................................................ 188
6.6.2. Ozone Concentration Profiles................................................................ 191
6.6.3. Effect o f Model Parameters and Operating Conditions........................194
6.7. Inactivation o f G iardia muris Cysts.................................................................195
Chapter 7. REACTOR DESIGN O P TIM IZA TIO N ..................................................... 203

7.1. Model Verification......................................................................................... 204
7.2. Theoretical Analysis and Optimization........................................................... 212
7.2.1. Predictions for Current Popular Designs..............................................213
7.2 .2. Alternatives o f Current Popular Designs.............................................217
Chapter 8. SUM M ARY A ND CO NCLUSIO N............................................................. 224

8.1. Model and Method Development.....................................................................224
8.2. Theoretical Analysis with Axial Dispersion Model.......................................... 225
8.3. Pilot-scale Experiments and Results.................................................................225
8.4. Full-Scale Application and Optimization.......................................................... 226
8.5. Weakness and Alternative o f CT Concept.......................................................228
LIST OF REFERENCES................................................................................................. 230
APPENDICES
APPENDIX A. COMPUTER PROGRAMS.......................................................... 245
A. 1. Programs for Ozonation and Disinfection Tests......................................246
A.2. Program for Tracer Test........................................................................... 250
APPENDIX B: PHOTOS OF EXPERIM ENTAL APPARATUS......................... 251
APPENDIX C: BUBBLE SIZE MEASUREMENTS............................................ 255
APPENDIX D: TRACER TE S TS .......................................................................... 259
APPENDIX E. MASS TRANSFER TESTS.......................................................... 271
APPENDIX F: OZONE REACTION TESTS........................................................ 281
APPENDIX G: O ZO N ATIO N TESTS...................................................................293
APPENDIX H: APPLICATIO N FOR DISPERSION COEFFICIENT................ 304
APPENDIX I: EXPER IM EN TA L DATA FOR SINGLE-DIFFUSER S ET 309
V IT A ................................................................................................................................. 329
ix
LIST OF TABLES
Table Page
1. Recommended Giardia Cysts and Cryptosporidium Oocysts Treatment

Levels Based on Source Water Qualities (U.S. E PA 1994)...................................9
2. Summary o f Expected Removal o f G iardia lamblia Cysts and Viruses by

Filtration and Recommended Level o f Disinfection (U.S. EPA 1991)...................9
3. Recommended CT (mg-min/L) Values for Inactivation of Giardia lamblia

Cysts and Viruses by Ozone at pH 6 - 9 (U.S. EPA 1991).................................. 10
4. Guidelines to Predict Ozone Concentration (C) Based on Chamber Inlet and

Outlet Ozone Concentrations (U.S. E P A 1991)................................................... 11
5. Recommended Procedures to Calculate Contact Time (T) for Ozone

Disinfection (U.S. EPA 1991)................................................................................13
6. Summary o f CT Value Ranges Corresponding to 99% Inactivation of

Various Microorganisms by Common Drinking Water Disinfectants at 5 °C
(Hoff; 1986)............................................................................................................ 48
7. Second Order Disinfection Rate Constants for Ozone Inactivation of

G iardia lam blia Cysts and Viruses (U.S. E P A 1991)......................................... 53
8. Development o f Mass Balance Expressions for Dissolved Ozone (Based on

the Axial Dispersion Model) in a Control Volume of a Bubble-Diffuser
Contactor Operated at Counter-Current Flow (see Figure 6)............................... 72
9. Closed Vessel Boundary Conditions....................................................................... 75
10. Standard Methods for the Measurement o f Selected Major Water Quality
Parameters (APHA et a l, 1992).............................................................................95
11. Major Water Quality Parameters in the Pretreated Water Fed to the Ozone
Contactor............................................................................................................... 143
X
12. Results for Calibration Procedure of Bubble Size Measurements Performed

with Glass Beads................................................................................................... 146
13. Operating Conditions and Experimental Results for Bubble Size

Measurement and Tracer Tests............................................................................. 152
14. Dispersion Coefficient and Oxygen Mass Transfer Coefficient Estimated

from Various Empirical Correlations for the Operating Conditions Used in
this Research..........................................................................................................165
15. Operating Conditions and Experimental Results from Ozone Reaction Tests
and Ozonation Tests..............................................................................................181
16. CT Values and Giardia Inactivation Efficiencies for MWDSC's Ozonation

Tests Estimated Based on Guidelines from U.S. EPA (1991) and Predicted
with ADR Cascade Model....................................................................................210
LIST OF FIGURES
Figure Page
1. Schematics o f Full-Scale Ozone Bubble-Difiiiser Contactors Located at (a)

Oakland, CA,and (b) Tucson, AR (U.S. EPA, 1991).............................................16
2. Determination o f Ao from Pulse-Input and Step-Input Normalized Tracer

Test Results............................................................................................................. 21
3. Ozone Decomposition o f Water from University Lake, Chapel Hill, NC

(Guittonneau et al., 1992)...................................................................................... 44
4. Typical Semi-Log Microorganism Survival Curves Observed in Disinfection

Experiments.............................................................................................................54
5. Schematics o f Selected Models for Bubble-Difiiiser Ozone Contactors: (a)

Axial Dispersion Model (Marinas et al., 1993); (b) Back-Flow Cell Model
(Smith and Zhou, 1994); (c) CSTR/Axial Dispersion Combination Model
(Sauze et al., 1992).................................................................................................63
6. Schematic o f an Infinitesimal Control Volume for a Bubble-Difiiiser

Contactor Operated in the Counter-Current Flow Configuration......................... 71
7. Axial Dispersion Model Predictions of Aqueous and Gaseous Ozone

Concentration Profiles in a Bubble Diffuser Contactor Chamber Operated in
Both Counter-Current and Co-Current Modes with the Assumption of
Negligible Ozone Self-Decomposition................................................................... 77
8. Model Predictions of Tracer Curve with Different Dimensionless Time and

Length Increments Used in the Finite Difference Approach..................................80
9. Inactivation o f G iardia muris Cysts in a Semi-Batch Reactor Operated at

15°C (Wickramanayake et al., 1985) and Corresponding CT-lag Model............. 83
10. Residence Time Distribution Curve for Viable Cysts from a Simulated
Pulse-Input Inactivation Test..................................................................................87
xii
11. Effect o f Sampling Time on the Observed Inactivation Efficiency o f Pulse-

Input G iardia muris Cyst Disinfection Test........................................................... 88
12. Schematic o f Pilot-Scale Ozone Bubble-Difiixser Contactor.................................97
13. Side and Top View o f Diffuser Systems with (a) One, and (b) Four
Diffusing Stones...................................................................................................... 99
14. Experimental Matrix for Operating Conditions....................................................101
15. Conceptual Schematic of Experimental Setup and Data Acquisition System

Used for Tracer Tests............................................................................................ 105
16. Schematic o f Experimental Apparatus Used for the Measurement o f Initial

Fast Ozone Demand...............................................................................................113
17. Schematic o f Experimental Set-up Used for the Measurement o f Ozone

Self-Decomposition by the Syringe Indigo Method............................................. 117
18. Theoretical Analysis of the Effect o f Dispersion Number on Ozone

Concentration Profiles and Giardia muris Cyst Inactivation Inside an Ozone
Bubble-Diffiiser Contactor Operated in the Counter-Current Mode................... 123
19. Theoretical Analysis of the Effect o f Dispersion Number on Ozone

Bubble-Diffiiser Contactor Operated in the Co-Current Mode........................... 124
20. Predicted Ratio o f tio Residence Time to Mean Residence Time

Corresponding to Dispersion Numbers up to 0.3 Based on the Assumption
o f Ideal Axial Dispersion Model.......................................................................... 126
21. Theoretical Analysis of the Effect o f Volumetric Mass Transfer Rate on

Ozone Concentration Profiles Inside an Ozone Bubble-Diffiiser Contactor
Operated in the Counter-Current Mode................................................................ 130
22. Theoretical Analysis o f the Effect o f Volumetric Mass Transfer Rate on

Ozone Concentration Profiles Inside an Ozone Bubble-Diffiiser Contactor
Operated in the Co-Current Mode.........................................................................131
23. Theoretical Analysis o f the Effect o f Varying Hydraulic Pressure on Ozone

Concentration Profiles Inside Ozone Bubble-Diffiiser Contactors Operated
in both Counter-Current and Co-Current Modes................................................. 135
24. Effect of Initial Fast Ozone Demand on Dissolved Ozone Concentration in

the Effluent o f a Contactor Operated in the Counter-Current Mode...................137
xiii
25. Simulated Decreases in Dissolved Ozone Concentration at the Effluent o f a

Counter-current Contactor as a Function o f Initial Fast Ozone Demand at
Various Second-Stage Ozone Self-Decomposition Rates................................... 139
26. Simulation o f Ozone Concentration Profiles in a Counter-Current Contactor

with Various Assumptions for Initial Fast Ozone Demand(at Constant Rate
o f 100 L/mg-min) and Subsequent Self-Decomposition Rate Constants 140
27. Bubble Size Distribution at Average Water Flow Rate o f 3.6 gpm...148
28. Bubble Size Distribution at Average Water Flow Rate o f 7.1 gpm...149
29. Bubble Size Distribution at Average Water Flow Rate o f 13 gpm....150
30. Empirical Correlation for Mean Bubble Diameter as a Function o f Standard

Gas Flow Rate....................................................................................................... 151
31. Comparison o f the Correlation Developed in this Study for the Sauter Mean
Bubble Diameter to Experimental Values Reported in the Literature for
Various Types o f Bubble Difiuser Stones............................................................. 156
32. Normalized Results from Selected Tracer Tests and Corresponding Model
Predictions............................................................................................................. 157
33. Hydrodynamics o f the Pilot-Scale Ozone Bubble-Diffixser Contactor and the

Corresponding Schematic of CSTR-ADR-CSTR Model.................................... 159
34. Plot of Liquid Phase Dispersion Coefficient Against Approach or Superficial

Gas Velocity.......................................................................................................... 162
35. Plot of Total Volume Ratio of CSTR Against Gas Flow Rate............................ 166
36. Representative Results from Mass-Transfer Tests Obtained with the Pilot-
scale Contactor Operated in the Counter-Current Flow Mode........................... 171
37. Representative Results from Mass-Transfer Tests Obtained with the Pilot-
scale Contactor Operated in the Co-Current Flow Mode................................... 172
38. Effect o f Applied Ozone Dosage on Ozone Decomposition Kinetics

Observed in Bench-Scale Batch Experiments..................................................... 177
39. Effect o f the Volumetric Ratio o f Ground Water Sample to Ozone Stock
Solution on Ozone Decomposition Kinetics Determined by Performing
Bench-Scale Batch Experiments............................................................................179
xiv
40. Plot of First-Order Ozone Decomposition Rate Constant and Initial Fast
Ozone Demand as a Function o f TO C................................................................ 182
41 Splitting o f Bubble Size Distribution Curve for Test 16-1025 to Study the
Effect o f Two Different Sizes o f Gas Bubbles.................................................... 185
42. Theoretical Analysis o f the Effect o f Gas Bubbles with Two Different Sizes
on Ozone Profiles Using Bubble Size Measurement Results from Test 16-
1025...................................................................................................................... 187
43. Model Predictions of Ozone and Viable Cyst Profiles with Greater Extent of
Back Mixing Assumed at The Bottom of Contactor Compared to the
Remaining o f the Water Column Inside the Contactor........................................ 189
44. Representative Ozonation Test Results obtained with the Pilot-Scale Ozone
Contactor Operated in Counter-Current and Co-Current Mode and
Corresponding CSTR-ADR-CSTR and ADR Model Predictions....................... 192
45. Effects o f Gas and Water Flow Rates on Average Ozone Concentration 196
46. Effects o f Gas and Water Flow Rates on / l0 Residence Times........................... 197
47. Experiment Results and Model Predictions Corresponding to Giardia Cyst

Inactivation Test Cyst-0.5..................................................................................... 200
48. Experiment Results and Model Predictions Corresponding to Giardia Cyst

Inactivation Test Cyst-1........................................................................................ 201
49. Ozonation Tests Performed with the MWDSC Demonstration Contactor

(Coffey eta l., 1995) Operated at 1.5 mg/L Applied Ozone Dose with
Different Gas Flow Rates and Gas-Phase Ozone Concentration.........................205
50. Ozonation Tests Performed with the MWDSC Demonstration Contactor

(Coffey et al., 1995) Operated at 2 mg/L Applied Ozone Dose with
Different Water and Gas Flow Rates................................................................... 208
51. Ozone and Giardia Cyst Profiles Predicted for The Ozone Contactors
Currently Used by Tucson, Los Angeles, and Metropolitan............................... 215
52. Theoretical Comparison for the Disinfection Performance of an Ozone

Contactor Between Counter- and Co-Current Operation Mode......................... 218
53. Theoretical Evaluation of the Effect o f Gas Distribution for Ozonation Test
3016F from Metropolitan......................................................................................220
XV
54. Theoretical Evaluation o f the Effect of Gas Distribution for Modified

Tucson Contactor with fast ozone decomposition (k j = 1.0 min'1) .....................222
55. Theoretical Evaluation o f the Effect of Gas Distribution for Modified

Tucson Contactor with slow ozone decomposition (kd = 0.1 min*1) ................... 223
LIST OF SYMBOLS
The following symbols are used in this thesis except where indicated separately:
A = contactor cross section area (cm2)
A, = initial absorbance o f indigo reagent at 600 nm (M 'lcm'1)
Am = absorbance o f sample and indigo reagent mixture at 600 nm (M 'lcm‘I
A 258 = absorbance o f sample at 258 nm (M 'lcm‘1)
A/46oo= absorbance difference at the wavelength of 600 nm (M 'lcm'1)
A O = applied ozone dosage (mg/L)
a = interfacial transfer area per unit volume (cm 1)
c = tracer concentration (mg/cm3)
Co = normalized tracer concentration
D = concentration o f fast ozone demand chemical (mg/cnr)
D-}o~ fast ozone demand measured within initial 30 seconds (mg/cmJ)
Dc - column inner diameter (cm)
Z)o3 = molecular diffusivity o f ozone in water(cm2/min)
d = E i/{U c L ) = dispersion number (dimensionless)
dB = average bubble diameter (mm)
do - Eu/(JJcU \-r)') = dispersion number for axial dispersion portion
xvii
E l = axial dispersion coefficient (cm2/min)
g = gravitational constant (cm/min2)
H = Henry’s law constant (dimensionless)
H a = Hatta number (dimensionless)
K l = overall mass transfer coefficient (cm/min)
ko = gas phase mass transfer coefficient (cm/min)
kL = liquid phase mass transfer coefficient (cm/min)
k j = ozone self-decomposition rate constant (cm3(','l)/mgn‘1/min)
kr —reaction rate constant between O and D (cm3/mg/min)
k„ = inactivation rate constant after initial time-lag (cm3/mg-min)
L = water column height (cm)
M md = molar absorptivity o f indigo reagent at 600 nm (M ^cm 1)
M 0 = molar absorptivity o f aqueous or gaseous ozone (N f‘cm'1)
M r = total mass of tracer added (mg)
N = microorganism density (number/cmJ)
N cstr ~ number of CSTR in cascade
N d = kd-L/Ui = slower ozone self-decomposition units (mg'"l/cm3(n' I))
N l = K co -L/U l = number o f transfer units (dimensionless)
N k = kn'L/Ui = number o f inactivation unit (dimensionless)
Nr = kr'L/U t - fast ozone demand units (mg/cm3)
O = dissolved ozone concentration (mg/cm3)
O = flow rate (cm3/min)
r = volume ratio o f CSTR in contactor
Reox = U 'dff ul = apparent gaseous or aqueous Reynolds Number
S = H- Ug/U l = stripping factor (dimensionless)
t = time (min)
/lo = time required for 10% o f tracer mass to reach the effluent for a pulse-input
tracer test (min)
t = mean residence time from tracer test result (min)
U = Q/A = apparent velocity (cm/min)
Vb = terminal rise velocity o f gas bubbles in static water (cm/min)
V* b = ^G
/ s ~^L
/ \ _ s ~ sl'P velocity (cm/min)
V, = indigo reagent volume (mL)
V0 = ozone solution volume (mL)
Vs = ozone sample volume (mL)
x = distance from top of column in downward direction (cm)
Y = gaseous ozone concentration (mg/cm3)
z = x/L = normalized distance from top of column in downward direction
Greek letters
e = gas hold-up (dimensionless)
S = diffuser pore size (cm)
p i = liquid phase density (kg/m3)
O = mass flux (mg/min)
r = interfacial tension (N/m )
a = standard deviation from normalized tracer result
9 = t jt - normalized time
vl = water kinematic viscosity (cmVmin)
Subscripts
B = values at the bottom o f the water column under a pressure PB
D = axial dispersion zone
E = CSTR zone effluent
G = gas phase
/ = CSTR zone influent
in = contactor influent
L = liquid phase
out = contactor effluent
S = complete mixing zone
top = located at the top o f contactor
1,2 = gas bubble groups assumed with different mean size
XX
ABSTRACT
Chen, Chun-Ming, Ph.D., Purdue University, August, 1998. Modeling Drinking Water
Disinfection in Ozone Bubble-Difiiiser Contactors. Major Professor: Benito J. Marinas.
Co-Advisor: Dennis A. Lyn.
Ozone is the most powerful biocide among the chemical disinfectants commonly
used for drinking water treatment. Its use generally results in better control o f disinfection
by-products associated with other disinfectants. However, because of a lack o f good
understanding about the chemistry o f ozone, the hydrodynamics and mass transfer
characteristics o f ozone contactors, and the inactivation kinetics o f target microorganisms,
a conservative approach based on the CT concept, the product o f disinfectant
concentration and contact time, is used for regulatory purposes. The objective o f this
research was to develop an alternative to the CT approach based on the use o f a reliable
model for assessing disinfection efficiency in ozone bubble-diffuser contactors. More
specific objectives were: (1) development of a mathematical model for assessing
disinfection efficiency achieved in ozone contactors; (2) theoretical evaluation o f contactor
CT and the effect o f model parameters on dissolved ozone concentration distribution and
inactivation efficiency; (3) evaluation o f model with experiments performed with a pilot-
scale bubble-diffuser contactor; (4) analysis of experimental data available for full-scale
contactors and development o f contactor design recommendations.
xxi
A model based on a combination of continuous flow stirred tank reactor (CSTR)
and axial dispersion reactor connected in series were found to provide good representation
of ozone residuals and disinfection efficiencies in bubble-diffuser ozone contactors. A CT-
lag model for the inactivation o f G iardia cyst was also developed to simulate kinetic
curves with shoulder-effect.
A pilot-scale ozone contactor was used to verify the disinfection model.
Experimental methods were developed to measure aqueous ozone concentration profiles,
and evaluate the mass-transfer characteristics of the pilot-scale contactor, and to measure
ozone decay kinetics in a bench-scale batch reactor. Empirical equations developed in this
study for major model parameters were compared with literature values.
The model developed was extended to represent the performance o f full-scale
ozone contactors and verified with experimental results reported in the literature. Ozone
contactor configurations currently used were theoretically evaluated with the model.
1
Chapter 1. INTRODUCTION
1.1. Background and Problems
Chlorine is the most commonly used chemical for drinking water disinfection in the
United States. However, chlorine also reacts with organic matter in water and forming
disinfection by-products (DBPs), primarily trihalomethanes (THMs), many o f which are of
public health concern (Krasner et al., 1989). The U.S. Environmental Protection Agency
(U.S. EPA) has proposed new more stringent regulations for disinfectants and DBPs
(D/DBP) (U.S. EPA, 1994). For instance, the D/DBP rule proposes to lower the
maximum contaminant level (M CL) for total THMs from the current value o f 0.1 mg/L to
ultimately 0.04 mg/L. Maximum Residual Disinfectant levels (MRDLs) for chlorine,
chloramines, and chlorine dioxide were also proposed as part of the D/DBP rule. Also,
free chlorine and chloramine have been found to be practically ineffective against
Cryptosporidium parvum oocysts, a new concern in the proposed Enhanced Surface
Water Treatment Rule (ESWTR) (U.S. EPA, 1994). In response to these forthcoming
regulations, the water industry is currently looking for alternatives such as the
enhancement of conventional clarification processes from the point o f view o f disinfection
precursor removal, and the exploration of alternative disinfection processes. Current
alternative chemical disinfectants include ozone, chlorine dioxide, and chloramines.
Among these, ozone is the most powerful biocide (H o ff and Geldreich, 1981) and
generally results in better control o f DBPs in the absence o f bromide. The ozone
disinfection process is the focus o f this research.
Direct monitoring o f protozoan parasites such as Giardia lamblia cysts or
Cryptosporidium parvum oocysts as the assessment o f inactivation efficiency achieved in
the disinfection process o f water treatment plants is not feasible primarily because
microorganisms are not generally present in raw waters, and the viability assessment
techniques available are complex. Consequently, the Surface Water Treatment Rule
(SW TR) (U.S. EPA, 1991) specifies the use of the “CT” parameter, the product of
disinfectant residual concentration and contact time, for demonstration of adequate
disinfection capability in water treatment plants. “C” and “T ” can be estimated from the
disinfectant concentrations distribution and tracer tests, respectively. In order to comply
with regulations, water treatment utilities have to demonstrate that CT parameters
achieved in their contactors are equal or greater than the minimum CT values required to
inactivate the fraction o f microorganisms not removed by clarification processes..
1.1.1. Assessment of Ozone Disinfection Efficiency
Among all chemicals used for water disinfection, ozone is the most unstable and
reactive under drinking water conditions. The hydrodynamics of ozone contactors are
generally more complex than other disinfection units because the involvement of phase
transfer. In addition, the experimental data for the inactivation of microorganisms
(especially for G iardia cysts and Cryptosporidium oocysts) with ozone are insufficient.
Because o f these unfavorable reasons, the assumptions associated with CT concept are
oversimplified and the CT specified in the SWTR for ozone disinfection includes high
safety factor. Consequently, overdose and the corresponding higher cost and greater
formation o f health-risk ozone disinfection by products (Ferguson et al., 1991) could
result. Therefore, EPA encourage water utilities to develop alternative methods for
assessing disinfection efficiency for ozone contactors.
The alternative approaches developed in this research include: (1) performing more
intensive measurement to provide more realistic ozone concentration distribution inside
contactor chamber and increasing CT credits; (2) developing more accurate mathematical
models to predict ozone concentration distribution and assess disinfection efficiencies
throughout the entire contactor; (3) using microbial surrogate indicators to simulate the
actual inactivation o f microorganisms.
1.1.2 Application in Full-Scale Ozone Disinfection
The most commonly used contactor design for ozone disinfection is bubble-
diffuser type. A typical bubble-diffiiser contactor is usually compartmentalized into
subsequent counter-current and co-current chambers by positioning bafflers so that the
water flow is forced in an over-under mode. However, the ozonation and inactivation
mechanisms in ozone contactor are not well-understood and difficult to assess because of
the unstable nature o f dissolved ozone, and the complex hydrodynamics partially resulted
from the involvement o f phase transfer. Consequently, the ozone contactor could be
inefficiently designed or operated. Therefore, it is necessary to develop a practical model
to represent the dissolved ozone and inactivation level achieved in ozone bubble-diffiiser
contactors, and to develop the corresponding approaches to assess model parameters.
1.2. Research Objectives
The objective o f this research was to develop an alternative to the CT approach
based on the use o f a reliable model for assessing disinfection efficiency in ozone bubble-
diffiiser contactors. More specific objectives were: (1) development o f a mathematical
model for assessing disinfection efficiency achieved in ozone contactors; (2) theoretical
evaluation of contactor CT and the effect o f model parameters on dissolved ozone
concentration distribution and inactivation efficiency; (3) evaluation o f model with
experiments performed with a pilot-scale bubble-diffiiser contactor; (4) analysis of
experimental data available for full-scale contactors and development o f contactor design
recommendations.
1.2.1. Model Development
The most commonly used models for studying bubble-diffuser contactor were
reviewed and evaluated. The model developed in this research is one dimensional, which
is based on steady-state operation, Iiquid-phase axial dispersion, gas-phase plug flow, two-
film mass transfer, fast ozone reaction with selected contaminants, slower ozone
decomposition, and mixed second order microorganism inactivation kinetics. Variables in
the model include dissolved ozone concentration, gas-phase ozone concentration, fast
ozone demand concentration, and microorganism density.
1.2.2. Analysis of Major Model Parameters
Hydrodynamics o f contactor, ozone mass transfer, and ozone reaction kinetics are
the major parameters affect the ozone residual distribution and disinfection efficiency. The
effects o f tnese parameters on dissolved ozone concentration profiles were theoretically
analyzed with the model developed in this research.
1.2.3. Model Verification
Several ozonation and two Giardia muris cyst inactivation tests have been
performed with a pilot-scale bubble diffuser contactor for the purpose of model
verification. Tracer, mass transfer, ozone demand, and ozone decomposition tests have
been performed with the pilot-scale contactor or a bench-scale batch reactor to determine
the parameters required for model predictions. Empirical correlations for the major model
parameters were estimated and compared with the experimental data reported in literature.
The non-ideal conditions observed in the pilot-scale experiments, and the effect of
operating condition on dissolved ozone concentration profile and contact time were also
evaluated.
1.2.4. Reactor Design Optimization
The model developed was applied to full-scale ozone contactors and verified with
experimental results reported in literature. The CT concept was evaluated. Several
popular full-scale ozone contactor designs were theoretically evaluated with the model
developed in this research. Contactor design and operating conditions were
recommended.
7
Chapter 2. LITERATURE R EVIEW
2.1. Regulations
The Surface Water Treatment Rule (SWTR) was promulgated by the U.S. EPA on
June 1989, and was approved and went into effect on December 1990 (U.S. EPA, 1991)
Regulations for filtration, disinfection, turbidity, G iardia lamblia, viruses, Legionella, and
heterotrophic bacteria were published and apply to all public water systems using surface
water or ground water sources under the direct influence o f surface water. The SWTR
specified the removal/inactivation requirements for G iardia lamblia cysts and viruses with
various disinfectants, and assumed that the level of treatment providing adequate
protection from G iardia cysts and viruses would also be sufficient for protecting against
most other pathogenic microorganisms. Monitoring o f microorganism and demonstration
o f inactivation by water utilities were not deemed to be feasible tasks primarily because
microorganisms were not generally present in raw water, and the analytical techniques
available for determining microorganism densities and viability are complex and require
specialized personnel and laboratories. Consequently, the SWTR is based on the approach
o f specifying treatment requirements instead o f the more common regulating approach of
establishing maximum contaminant levels (MCLs) for the target pathogenic
microorganisms.
The proposed Enhanced Surface Water Treatment Requirements (ESW TR) (U.S.
EPA, 1994) was issued to provide additional protection against disease-causing
pathogenic microorganisms. The ESWTR was proposed because: (1) the SWTR did not
address Cryptosporidium', (2) pathogen reductions specified in the SWTR may be
inadequate; (3) viruses CT values may be greater than assumed by the SWTR; and (4)
D/DBP rule may undermine pathogen control. The ESWTR will apply to all public water
systems using surface water or ground water under the influence o f surface water.
2.1.1. Treatment Requirements
According to the proposed ESWTR, five treatment alternatives were addressed to
control Giardia, Cryptosporidium, and/or viruses. Several options were included in each
alternative. Alternative A addresses enhanced treatment for Giardia. Alternatives B and
C address treatment for Cryptosporidium. Alternative D specifies disinfection
requirements for viruses. Alternative E maintains existing level o f treatment requirements
for Giardia and viruses. According to these options, all public water systems using
surface water sources or ground water under the influence of surface water might need to
remove/inactivate at least 99.9 percent of Giardia lamblia cysts, 99 percent of
Cryptosporidium parvum oocysts, and 99.99 percent of viruses. Low quality water
sources might be required to achieve greater removal/inactivation efficiencies.
Recommended overall treatment levels based on source water quality are listed in Table 1.
9
Table 1. Recommended Giardia Cysts and Cryptosporidium Oocysts Treatment Levels

Based on Source Water Qualities (U.S. EPA, 1994)
Daily Average
Cyst or Oocyst/100 L
(geometric mean) <1 1 -9 1 0 -9 9 100-999 >999
Giardia (log)
3A t Â 1
Removal/Inactivation (log) 5A1/4A2 6A1/5A2 6a,/6a2
Cryptosporidium
3 ^ / 2 83/ 2 s4 B2 ^ B3 ^ B 4 6B2,5B3,4B* 6B2/5B3/484
Removal/Inactivation (log)
AI A2 : Options 1 and 2 in Alternative A, proposed ESWTR, respectively

B2 3 3 3 4 . Qptjons 2 , 3j and 4 in Alternative B, proposed ESWTR, respectively
Actual disinfection requirements specified in the SWTR vary for different overall
treatment system design. Table 2 provides a summary o f the estimated removal
performance in various filtration systems and the corresponding recommended
supplementary minimum level of disinfection (U.S. EPA, 1991).
Table 2. Summary of Expected Removal of G iardia lam blia Cysts and Viruses by
Filtration and Recommended Level o f Disinfection (U.S. EPA, 1991)
Expected Removals Recommended Disinfection

(log removal) (log inactivation)
Filtration Process Giardia Cysts Viruses Giardia Cysts Viruses
Conventional 2.5 2.0 0.5 2.0
Direct 2.0 1.0 1.0 3.0
Slow Sand 2.0 2.0 1.0 2.0
Diatomaceous Earth 2.0 1.0 1.0 3.0
10
2.1.2 CT Concept
The U.S. EPA requires that water utilities demonstrate adequate disinfection by
estimating the CT parameter, the product o f disinfectant residual concentration in mg/L,
and contact time in minutes, o f the disinfection contacting system. CT values should be
equal or greater than the minimum parameters specified in the SWTR to ensure a
minimum level of inactivation. CT values required by the U.S. EPA for inactivation of
Giardia cysts and viruses by ozone at pH 6 - 9 and different temperature are presented in
Table 3 (U.S. EPA, 1991). No CT values has been set for Cryptosporidiumparvum
oocyst as yet.
Table 3. Recommended C T (mg-min/L) Values for Inactivation o f Giardia lamblia

Cysts and Viruses by Ozone at pH 6 - 9 (U.S. EPA, 1991)
Giardia Cysts Temperature (°C)

Inactivation < 1 5 10 15 20 25
0.5 - log 0.48 0.32 0.23 0.16 0.12 0.08
1.0 - log 0.97 0.63 0.48 0.32 0.24 0.16
1.5 - log 1.5 0.95 0.72 0.48 0.36 0.24
2.0 - log 1.9 1.3 0.95 0.63 0.48 0.32
2.5 - log 2.4 1.6 1.2 0.79 0.60 0.40
3.0 - log 2.9 1.9 1.43 0.95 0.72 0.48
Viruses Temperature (°C)
Inactivation < 1 5 10 15 20 25
2 -lo g 0.9 0.6 0.5 0.3 0.25 0.15
3 - log 1.4 0.9 0.8 0.5 0.4 0.25
4 -lo g 1.8 1.2 1.0 0.6 0.5 0.3
2.1.2.1. Ozone Concentration (C)
The ozone concentration used for the determination o f CT is defined as average
concentration Civc which may either equals the average concentration evaluated by direct
measurement through contactor sampling ports or estimated from Table 4. Most ozone
contactor chambers can be classified into four types based on flow configurations: Turbine
(Continuous-Flow Stirred-Tank Reactors), Counter-Current (gas flowing upward and
water flowing downward), Co-Current (both gas and water flowing in the upward
direction), and Reactive Flow (without gas input) chambers. CT credits are calculated
from contactor chamber outlet and inlet ozone concentrations, Coul and Cm, respectively.
Special considerations were used for the first chamber because the ozone residual may be
very sensitive to water impurities exerting fast ozone demand.
Table 4. Guidelines to Predict Ozone Concentration (C) Based on Chamber Inlet and
Outlet Ozone Concentrations (U.S. EPA, 1991)
Chamber Type Turbine Co-Current Counter-Current Reactive
First Chamber C
^ = '-'out
C Partial Credit1 Partial Credit1 N ot Applicable
Subsequent c
^ = c
ôut C = Cout or C = CJ 2 C = ôul
C
Chambers C = (C ^+C J/2
1-log o f virus inactivation providing that C ^ > 0.1 mg/L, and 0.5-log G iardia cysts
inactivation providing that Cout >0.3 mg/L
12
Some authors have proposed approaches to estimate the concentration C for
disinfection criteria different from the guidelines listed in Table 4. Lev and Regli (1992b)
proposed the use o f a one-third rule (Cout/3) as an estimate for C in counter-current
chambers to ensure that the contactor characteristic concentration would be conservative
for a broader range o f operating conditions. Zhou et al. (1994) recommended that the
concentration C for a counter-current chamber be obtained by averaging three values
measured at the top, bottom, and middle of the chamber. For co-current and reactive
flow, these authors indicated that an estimate of Clve by Cout or (Cout+Cin)/2 as suggested
by U.S. EPA appears to be representative under most operating conditions.
2.1.2.2. Contact Time (T)
The contact time used to calculate CT credit is lw. The t\a residence time must be
obtained by performing tracer tests. The selection of /10 as contact time is a conservative
measure which ensures a minimum exposure time for 90 percent of the water and
microorganisms entering a disinfection contactor to disinfectants.
The residence time tio was selected to provide a safety margin to compensate for
losses in disinfection efficiency resulting from hydrodynamic non-idealities inside
contactors, e.g., gas and water flow short-circuiting and high degree of back-mixing.
However, as contactor hydrodynamics approach Continuous-Flow Stirred-Tank Reactor
(CSTR) conditions, back-mixing approaches infinity and the safety margin fails to
compensate for decreases in disinfection efficiency. In this case, either the Segregated
Flow Analysis (SFA) or CSTR method should be used (U.S. EPA, 1991). The SFA, a
13
method accounting for micro-mixing in the disinfection contactor, is often used to
characterize chemical reactions (Levenspiel, 1972). The reactor is assumed to be divided
into several channels. Each channel with its own RTD curve given by its characteristic
tracer tests. Overall reactor performance is estimated by the summation of performances
in all channels.
The CSTR method assumes that the back-mixing inside a contactor is uniform and
ideally infinite. Guidelines to estimate contactor contact time based on tracer test results
are presented in Table 5.
Table 5. Recommended Procedures to Calculate Contact Time (T) for Ozone

Disinfection (U.S. EPA, 1991)
Conditions: t < H R T 1/ , < HRT/ , > H R T/

ho ^ / 3 ho ^ /3 ho * / 3
RIAC2 < 2.5-Log RIAC > 2.5-Log

II
II
H
H
O
O
Recommended SFA SFA SFA
Methods CSTR3 CSTR CSTR
HRT is hydraulic retention time, or reactor volume divided by water flow rate
Required level of inactivation in logs of either Giardia lamblia cysts or viruses
whichever value is greater
CSTR method is extremely conservative and should be avoided when alternative
approaches are possible
14
2.2 Giardia and Cryptosporidium
G iardia lamblia is the most frequently identified pathogenic intestinal protozoan
parasite, and the most commonly implicated agent in waterborne disease outbreaks
reported in the Unite States (Lin, 1985). Giardia cysts are presented in the stools of wild
animals, particularly beaver and other aquatic mammals, and they are specially prevalent in
cold and clear surface waters (Fogel et al., 1993). Cysts could remain infectious for
several months at cold temperatures. Contamination o f drinking water with Giardia cysts
has been a major concern. The non-human pathogen, Giardia muris, is often used in
disinfection studies as a surrogate indicator for Giardia lamblia.
Cryptosporidium oocysts have been identified recently in many surface waters
throughout the United States. The most notorious outbreak o f Cryptosporidium took
place in Milwaukee, Wisconsin in 1993. More than 400,000 people were affected (Solo-
Gabriele and Neumeister, 1996). Cryptosporidium has been known as a cause of disease
in cattle and poultry, and these animals also act as reservoirs for the organism.
Cryptosporidiosis has been identified as a very severe disease in humans with
compromised immune systems, in some cases leading to death (Bird and Smith, 1980).
There is no effective cure for cryptosporidiosis at this time. Accordingly, the ESWTR
includes Maximum Contaminant Level Goal (M CLG) of zero for Cryptosporidium
parvum. Based on laboratory studies, Cryptosporidium appears to be much more resistant
to common disinfectants compared to Giardia (Korich et al., 1990). Research is currently
underway to determine actual disinfection requirements.
15
2.3. Ozone Contactors
Ozone is the most insoluble o f all disinfectants used for drinking water treatment.
Furthermore, ozone decays rapidly and is consumed by many impurities typically present
in natural water. For these reasons, ozone gas is generally generated on site and the
ozonated gas is immediately introduced into disinfection contactors. Immediately after
transferring from gas bubbles into water, ozone starts to decompose by reacting with other
solutes and microorganisms. The effectiveness of ozone disinfection depends on the
concentration achieved for dissolved ozone. The distribution o f dissolved ozone
concentration inside a contactor depends on transfer efficiency, ozone demand, and
contactor hydrodynamics and configuration. Ozone transfer efficiency, in turn, also
depends on the method used for introducing ozonated gas bubbles into the contactor.
Ozone contactors commonly used in the United States include bubble diffusers, turbines,
deep U-tubes, static and sonic mixers, packed towers and plate columns, and injectors
(Masschelein, 1982; U.S. EPA, 1991). Bubble diffusers are the most widely used among
these configurations, and thus the focus o f this investigation.
Ozone bubble-diffiiser contactors are typically comprised of four to six flow
through chambers separated by baffles. These baffles are usually installed so that water
flows in an over-under mode to reduce the overall back-mixing inside contactors (Blank et
al., 1993). Examples o f typical bubble diffuser contactors are shown in Figure 1 (U.S.
EPA, 1991). Ozonated gas is introduced through porous diffusers located at the bottom
o f the contactor. Water and gas flows within a chamber are either counter-current
16
(a)
influent
effluent
=> \
= >
• 0 9 e a o
vA
(b)
'A 'A A Â effluent
=>
influent —
© « 9
<=> iy 7 VVV
“ L
Figure 1. Schematics o f Full-Scale Ozone Bubble-Diffuser Contactors Located at (a)

Oakland, CA, and (b) Tucson, AR (U.S. EPA, 1991)
17
or co-current. For a single chamber contactor, counter-current operation mode results in
higher ozone residual at the contactor effluent but lower average ozone concentration
throughout the water column, compared to co-current operation under the same
conditions. As depicted in Figure 1, multi-chamber contactors could also include reactive
chambers (chambers with no gas input) following contacting chambers to provide
additional disinfection by ozone residual.
Ozone contactors have some similarities in physical properties to bubble-column
slurry reactors used in chemical processes. Shah et al. (1982), Chaudhari and Shah
(1986), and Lee and Foster (1990) have extensively reviewed the hydrodynamics and mass
transfer characteristics o f bubble-column reactors, and determined the dispersion and
volumetric mass transfer coefficients, both of which have major effects on reaction
completion achieved in a reactor. Most dependencies o f the parameters in the empirical
equations developed for bubble-column slurry reactors could also be applied to bubble-
diffiiser ozone contactors. However, there are several notable differences between the
two applications that need to be considered: (1) full-scale ozone contactors are generally
much larger than slurry reactors; and (2) in general, superficial water flow velocities are
higher and superficial gas flow velocities are lower in ozone bubble-diffiiser contactors
compared to bubble-column slurry reactors.
2.4 Contactor Hydrodynamics and Tracer Tests
Mixing conditions inside ozone contactors are affected by ascending gas bubbles
and water flow configuration, and can range from almost continuous-flow stirred tank
reactor (CSTR) to plug flow reactor (PFR) conditions (Lev and Regli, 1992a). The axial
dispersion model and CSTR cascade are the most popular two classes o f single parameter
models, both can be used to represent mixing conditions in between these two extreme
ideal reactor conditions o f CSTR and PFR (Levenspiel, 1972). The extent o f dispersion
can be determined from residence time distribution (RTD) data obtained by means of
performing tracer tests.
2.4.1. Tracer Tests
There are two most commonly used methods o f performing a tracer test for
analyzing the overall hydrodynamic conditions inside a contactor: pulse-input and step-
input tests. Both methods are related and provide the same information (Levenspiel,
1972). For a pulse-input test, a spike of a pre-measured volume o f tracer is introduced to
the water immediately upstream from the inlet of the reactor, and monitoring of the tracer
concentration at the outlet is started simultaneously with the tracer injection. In contrast,
a tracer solution is added continuously at constant rate and concentration for a step-input
test. The RTD curve obtained from a step-input test is proportional to that obtained by
integration o f the RTD curve corresponding to a pulse-input test performed under
identical conditions.
The RTD data collected from a tracer test consist o f sampling time and tracer
concentration paired data series. For the convenience o f analysis and comparison between
tracer tests, the tracer concentration and time are then normalized to obtain dimensionless
values. Normalization should result in total tracer mass or ultimate tracer concentration o f
unity for a pulse-input or step-input test, respectively. In the case o f a pulse-input test, the
normalized tracer concentration, eg, and time, 0 , can be derived from experimental RTD
data with in EQ. 1 and EQ.2:
EQ.l
EQ.2
where c = tracer concentration (mg/L)
Ql = liquid flow rate (L/min)
t = time (min)
The mean residence time, t , total mass of tracer added, M r , and variance, cr, of the RTD
curve can be estimated from pulse-input tracer test results with the expressions:
EQ.3
^ t = Q l ' Z c< ‘ A/. EQ.4
EQ.5
The symbol A refers to a finite difference or increment, and the subscript, denotes each
data pair in the tracer data series.
In addition to the mean residence time F, the /io residence time is another
characteristic time variable of interest in disinfection (White, 1986) and has been used in
20
the definition o f CT concept as an indicator for inactivation of microorganisms (U.S. EPA,
1991) as discussed previously. The / 10 residence time is defined as the time interval
required for the outlet tracer concentration to achieve 10 percent of its ultimate response
for a step-input tracer test, or the time required for 10 percent of tracer mass to reach the
effluent for a pulse-input tracer test. The determination of t\o from pulse-input and step-
input tracer test data is illustrated in Figure 2.
2.4.2. Axial Dispersion Model
The axial dispersion model has been widely applied to characterize the
hydrodynamic behavior in bubble diffuser reactors (Houzelot et al., 1983; Lev and Regli,
1992a; Marinas et al., 1993; Roustan et al., 1993; Zhou et al., 1994). The mathematical
expression for a liquid phase undergoing axial dispersion can be obtained by performing a
mass balance for a conservative tracer corresponding to a contactor control volume of
infinitesimal thickness:
dc d 'c dc ^ ^
— = d — -------- EQ.6
36 3z~ 3z
where d = E L / (U L ■L ) - dispersion number (dimensionless)
z = x/L = normalized distance (dimensionless)
x = distance from the top of column (cm)
E l = axial dispersion coefficient (cmVmin)
Ul = 0 L / A = superficial water velocity (cm/min)
21
2.0
Normalized Pulse-Input Data
cn
c/i
JJ
C
o
*35
c
u
E
-5 10% Tracer Mass
u 0.5
0.0
cn
«
JJ
C Normalized Step-Input Data
o
’«
c
o
E
3 0.5
0.0
0 1 2 3 4
6 (dimensionless)
Figure 2. Determination o f /i0 from Pulse-Input and Step-Input Normalized Tracer Test
Results.
22
A = contactor cross sectional area (cm2)
The unidimensional axial dispersion model is based on the assumptions of infinite
radial dispersion, uniform axial dispersion throughout the water column, negligible gas
volume fraction, validity o f Fick's Law to represent axial dispersivity, and closed vessel
boundary conditions as defined by Levenspiel (1972). For a step-input tracer test, the
boundary conditions at the inlet and outlet o f the contactor, and the initial conditions are
shown as EQ.7, EQ.8, and EQ.9, respectively.
EQ.7
= 0 EQ 8
EQ.9
in which c,„ is the influent tracer concentration. A finite-differences approach can be used
to simulate a dimensionless step-input tracer test with EQs 6-9. Tracer concentrations
obtained from this step-input equation can then be differentiated to obtain the equivalent
dimensionless pulse-input concentrations.
The dispersion number for a closed vessel can be estimated from standard
deviation (a) and mean residence time ( f ) obtained from the RTD curve according to the
following expression (Levenspiel, 1972):
23
2 - d - 2 d z ( \ - t - ' /d) = ? ! = —— EQ.IO

x } t1 N rCSTR
The number o f reactors for an equivalent CSTR cascade (Ncsm) can also be
obtained from EQ. 10. Experimental tracer curves can be predicted with EQs.6-9 after
estimating the dispersion number with EQ. 10. Deviation between experimental and
predicted curves would give information on the non-ideal hydrodynamics of the system,
such as heterogeneous mixing and short-circuiting.
In addition to the dispersion number, the ratio t l0/ t is also commonly used as a
parameter to represent the extent of back-mixing inside a reactor. The extent of back-
mixing increases with decreasing t l0/ I in the range o f 1 (ideal plug flow) to approximately
0.1 (ideal CSTR).
2.4.3. Studies on Reactor Hydrodynamics
The back-mixing of gas-liquid bubble columns has been extensively studied and
reviewed by Shah et al. (1982) and Chaudhari and Shah (1986). Generally, back-mixing
in the liquid phase depends on the column diameter, gas distributor and gas velocity.
Hydrodynamic parameters are also dependent on the flow regime prevailing within the
reactor due to the effects of gas flow rate and column diameter. Three flow regimes have
been characterized on the basis of bubble size distribution, trajectory, and upward velocity
(Shah et a l 1982). “Quiescent” bubble flow regime is reported for superficial gas
velocities o f less than 5 cm/'s. Gas bubbles have nearly uniform size and are distributed
equally in the axial and radial directions. “Heterogeneous” or “Chum-turbulent” flow
24
regime was observed at greater superficial gas velocities with a fraction o f bubbles
coalescing into larger bubbles that rise at higher velocities. “Slug” flow regime
characterized by bubble slugs was observed in columns with diameter smaller than 15 cm.
Empirical dependence o f dispersion coefficient on interfacial gas velocity has been
suggested for different flow regimes. Full-scale ozone disinfection systems are typically
operated at the gas flow conditions corresponding to the “Quiescent” bubble flow regime
A first-order dependence o f the dispersion coefficient on interfacial gas flow velocity is
suggested by Shah et al. (1982) under this flow regime. Several researches studied ozone
contactor with smaller scale systems, and the conditions might fall in the more complicated
Chum-turbulent” or “Slug” flow regimes due to the limited size of such contactors and
fail to well model full-scale ozone contactors.
Good correlations for dispersion coefficients in bubble columns of various
diameters and operating interfacial gas flow velocities were found in several studies (Ohki
and Inoue, 1970; Kato and Nishiwaki, 1972; Deckwer et al., 1974; Hikita and Kikukawa,
1975; Baird and Rice, 1975). Empirical correlations reported for the dispersion
coefficient under the condition o f “Quiescent” bubble flow regime were:
Ohki and Inoue (1970): E t = 0.3 • D \ ■U ]r + 170 • 8 E Q .ll
Kato and Nishiwaki (1972): E L = EQ.12
Deckwer et al. (1974): E l = 2.7 • D'c4 • (7°'3 EQ. 13
25
Hikita and Kikukawa (1975): =0.037 + 0.188

jD '-g *ypL-* ^ D c gj
EQ.14
Baird and Rice (1975): E L = 0.35 • D f ■(UG ■g ) m EQ.15
where Dc (cm) is column inner diameter, Uq (cm/s) is superficial gas velocity, and 8 (cm)
is diffuser pore opening. However, these equations might not be applicable for ozone
contactors because of the differences in operating conditions and gas sparging devices.
The validity o f these equations will be evaluated in this research with pilot-scale tracer
experiments.
More recently, correlations have been reported for hydrodynamic parameters in
ozone bubble-diffiiser contactors. Marinas et al. (1993) reported the following
correlations for the dispersion number and t l0/ t as a function of liquid-phase Reynold
Number, Re/., and gas-phase Reynolds Numbers, Reo, under different operating
conditions for a 6-inches in diameter and 16-feet in water column height pilot-scale
bubble-diffuser ozone contactor:
EQ.16
EQ.17
where Re/; = UL-ds/ uL, Rec = Uq-cI b/ ul, and ul is water kinematic viscosity.
26
Several studies have focused on the occurrence o f high mixing zones in ozone
contactors. Sauze et al. (1992) studied the effect o f gas and water flow rates on axial
dispersion coefficients and ozone transfer efficiencies in a 15-cm diameter and 4.2-m water
column height pilot-scale ozone contactor. The results revealed that the mixing behavior
in this pilot unit could not be generally modeled by ideally uniform axial dispersion. Three
different mixing zones were differentiated to account for the observations: axial dispersion
in the middle portion and complete mixing at the top and bottom o f column.
Roustan et al. (1993) studied the results o f pulse-input tracer tests from two full-
scale bubble-diffuser contactors o f different treatment plants. A PFR-CSTR cascade
model (ideal plug flow followed by CSTRs in series) were used to simulate flow patterns
in the contactors. An adequate fit was obtained between tracer results and model
predictions. The mixing parameter t XQ/ t for a full-scale bubble-diffuser contactor with
two reactive chambers in series was found to be approximately 0.5. This number
contrasted with a value of 0.9 observed for a pilot system operated under similar
conditions.
2.5. Ozone Mass Transfer
Ozone is typically generated on site by passing pure oxygen or dry air gas between
electrodes across which a high electrical potential is generated. The output gas from the
ozone generator is then introduced into water through porous stone diffusers inside the
disinfection contactors.
27
2.5.1. Mass Transfer Kinetics
The two-film theory proposed by Lewis and Whitman (1924) is the most widely
used model for representing the phenomena o f mass transport across the gas-liquid phase
boundary. The two-film theory is based on the assumptions that very thin interfacial films
o f stagnant fluid occur on each side o f the phase boundary, and that the bulk of each fluid
phase beyond the corresponding film is well mixed, i.e., uniform concentration.
Therefore, mass transfer can be represented by molecular diffusion in the films.
Furthermore, the accumulation o f the diffusing in the films is negligible compared to that
in the bulk fluids. In accordance with these assumptions, the concentration distribution in
each film is linear at steady state conditions, and the resulting mass transfer coefficients are
directly proportional to the molecular diffusivities of the gas molecule in the liquid and gas
phases.
The direct proportionality between mass transfer coefficient and diffusivity has
been confirmed in a limited number o f studies (Akita and Yoshida, 1974; Deckwer, 1974).
However, more commonly, discrepancies have been found when investigating various gas
molecules under otherwise identical conditions (Calderbank and Moo-Young, 1961;
Yoshida and Akita, 1965; Hikita et al., 1981). There are many other mass transfer
theories that account for these discrepancies, such as the penetration theory (Masschelein,
1982) and the surface renewal theory (Danckwerts, 1951). These two approaches take
into consideration the hydrodynamic behavior and the unsteady-state activities of the two
28
phases near the interface. The penetration theory assumes that turbulent eddies penetrate
from the bulk solution into the interfacial film remaining there for a short period of time,
and that every surface element is exposed to the solute for the same length of time. In
contrast, instead o f a constant time of exposure, the surface renewal theory pictures the
liquid phase as completely disturbed, the solute being transported through numerous
infinitesimally small eddies at a rate controlled by the distribution of surface renewal rate.
The dependence o f the mass transfer coefficient on diffusivity suggested by these authors
is approximately to the order o f 0.5 to 1 (Kuo and Yocum, 1982). Stankovic (1988)
compared the mass transfer o f oxygen and ozone using experimental data and additional
information from the literature and found that oxygen transfers approximately twice as fast
compared to ozone.
Generally, the rate o f ozone mass transfer into water without reaction can be
represented by an overall transfer coefficient multiplying the interfacial area, and the
concentration difference between the two phases (Y/H - O) as the driving force. The
gas/liquid interfacial equilibrium is represented according to Henry’s law:
d0
— = V - na x
K, —-0 EQ 18
dt L V // J
where O and Y are ozone concentrations in the liquid- and gas-phase, respectively. K l is
the overall mass transfer coefficient, a is interfacial area per unit volume, and H is the
Henry’s law constant. The overall mass transfer coefficient combines information for the
mass transfer coefficients o f both the gas and liquid side (i.e., =^ +^ ^ ).
29
The overall mass transfer coefficient K i can be estimated by the liquid phase mass transfer
coefficient ki for gases with high volatility such as ozone.
The mass transfer parameter K L a can be determined by one of two methods: (1)
studying the rate at which the concentration o f dissolved ozone increases (i.e., transient) in
a batch system; and (2) assessing the mass of ozone transferred into water inside a flow
through reactor operated at steady state. The latter approach requires the application of a
known hydrodynamic model to account for back-mixing (e.g., axial dispersion model).
The overall mass transfer efficiency is estimated by solving simultaneously mass balances
on the liquid and gaseous phases.
Ozone has a tendency to decompose in natural waters and thus the effect of
chemical reaction on the local mass transfer must be evaluated. Danckwerts (1970)
suggested the use of the dimensionless Hatta number, Ha, for this purpose. Ha is the ratio
between the maximum chemical reaction rate developed in the film and the maximum
physical absorption rate:
, 2 D m -k d f o r '
H a 1 = ------ --— — EQ.19
(n + \ ) k L-
where n is the order o f the irreversible ozone decomposition reaction, kd is the reaction
rate of ozone decomposition, Den is the molecular diffiisivity o f ozone in liquid phase.
EQ. 19 is based on the assumption that there is no resistance to mass transfer in the gas
phase (i.e., high volatility compounds). The effect o f chemical reaction on mass transfer
rate can be neglected for 7/a2 « 1.
30
The mass transfer parameter, K L a, and gas bubble characteristics in bubble
columns and other agitated systems have been extensively studied by Shah et al. (1982),
Chaudhari and Shah (1986), and Lee and Foster (1990). In general, mass transfer was
found to increase with increasing gas superficial velocity, number o f bubbles, mixing
energy, and the column diameter. The mass transfer coefficient KL increased with
increasing temperature and decreased with increasing liquid viscosity. Shah et al. (1982)
found that column size affected mass transfer for column diameters of up to 0.15 m.
Because the mass transfer parameter is greatly dependent on system
characteristics, the applicability o f simplified correlations obtained for mass transfer from
bubble column experiments to ozone contactors is questionable (e.g., Kca=\.2Uc?S2 from
Deckwer et al., 1974, used by Zhou et al., 1994). There are several empirical correlations
for KL-a that were estimated from experiments performed with pilot-scale ozone bubble-
diffuser contactors. The contactor used by Xu and Liu (1990) had a 5-cm i.d. and water
column height of 4 m. Laplanche et a l (1991) used a unit with a 19-cm i.d. and 4-m water
column height. The ranges for Uq and Ul investigated in these two studies were
approximately 20 to 200 and 350 to 1200 cm/min by Xu and Liu, and 1.7 to 6.5 and
constant at 0.76 cm/min by Laplanche et al. Both groups of authors applied plug flow
model and correlated K c a to the superficial gas velocity only:
Xu and Liu (1990): K l a = 0.0788 • £ /°983 EQ.20
Laplanche et a l (1991): K l ■a = 0.0475 • EQ.21
31
The units used for K L-a and Ug were min'1and m/hr, respectively. Evaluation of these
empirical correlations with the ozone mass-transfer studies performed in this research is
presented in a subsequent section.
2.5.2. Mass Transfer Coefficient kL
In mass transfer studies with bubble columns, ^ is usually correlated with
superficial gas and liquid velocities, mean bubble diameter, molecular diffusivity of the
solute in liquid phase, and kinematic/dynamic viscosity (Shah et al., 1982). The
correlations developed by Hughmark (1967) and Akita and Yoshida (1974) were based on
experiments performed with wide ranges of operating conditions and system
characteristics that are appropriate for the application o f ozone.
^ • D 03
Hughmark (1967): EQ.22
y.:
D.0 3 V, g d t g-dl pL
Akita and Yoshida (1974): k L - 0.5 EQ.23
dB >■ ^ 03 ' \ V L J V T
Sh in EQ.22 is the Sherwood Number (dimensionless) which is given by the following
empirical equation:
1.61
1/3 > 0.072
0.4*4 rr 0.339 db g
Sh = 2 + 0.0187- Re -Sc EQ.24
D,023/3
where kL = mass transfer coefficient in liquid phase (cm/min)
32
D a3= molecular diffusivity o f ozone in water (cm2/min)
dB = average bubble diameter (cm)
Re = d B V* / v L = Reynolds Number (dimensionless)
Sc = v L / D01 = liquid-phase Schmidt Number (dimensionless)
g = gravitational constant (cm/min2)
v L = water kinematic viscosity (cm2/min)
V *b = ^ G
/ e ~ ^ L
/ \ - e = S^P velocity (cm/min)
Ug = superficial gas velocity (cm/min)
s - gas hold-up (dimensionless)
pL = liquid phase density (kg/m3)
r = interfacial tension (N/m)
Gas hold-up is defined as the volume fraction o f gas bubbles to the total volume of
fluids in the contactor. Under typical operating conditions for ozone disinfection, gas
hold-up is small because of low gas flow rate. Therefore, the slip velocity will be
approximately equivalent to the terminal bubble rise velocity in a batch reactor (Rakness et
al., 1988), or
Vg = ^ - ± = ^ ± U L =Vg EQ.25
s 1- s e
33
where VB is terminal rise velocity o f gas bubbles in a batch water column, which depend on
the diameter o f the bubble and the water temperature. The sign in front o f U l is positive
for co-current and negative for counter-current flow.
Masschelein (1982) developed an equation for the liquid phase mass transfer
coefficient in static water column into which ozonated gas is bubbled:
k L = 1 -1 3 ^ °3 EQ.26
Because ozone is slightly soluble, and the resistance to mass transfer occurs
primarily in the liquid phase, kt estimated from these correlations approximately equals
the overall mass transfer coefficient (K L). The correlation developed by Hughmark (1967)
were used by Roustan et al. (1981), Rakness et al. (1988), Stankovic, (1988), and
Marinas et al. (1993) to predict the ozone mass transfer in either pilot-scale or full-scale
ozone contactors.
2.5.3. Average Bubble Diameter and Terminal Rise Velocity
Bubble size affects mass transfer efficiency by inducing mixing as well as by
providing a certain contact area. The studies by Satterfield and H uff (1981) and Ahmad
and Farooq (1985) have clearly revealed the importance of precise knowledge of bubble
size in establishing the mass transfer characteristic o f bubble columns. Photographic
techniques have been used widely for the determination o f bubble sizes because of their
simplicity. Yamashita et al. (1979) have reported the comparison between photographic
34
and other methods such as using electrical and optical probes. All investigated methods
gave reliable results for homogeneous bubbles.
Akita and Yoshida (1974) determined the bubble size distribution in columns up to
0.3 m in diameter and for superficial gas velocities o f up to 0.07 m/s. The gas was
sparged through perforated plates and single orifices. It was concluded that the diameter
of the bubbles is mainly affected by column diameter and only slightly dependent on the
gas velocity. The orifice diameter had no effect on the final bubble diameter reached in the
bulk region away from sparger due to coalescence and breakup o f the bubbles. An
opposite observations were reported by Koide et al. (1979) for columns o f larger scale.
Results from those authors indicated that the bubble diameter was independent of column
diameter but affected by the diameter o f the nozzles used.
For the operating conditions typically applied in ozone bubble-diffuser contactors,
the average bubble size depends on the type of porous diffuser used and the gas flow rate
applied. The experiments performed by Stankovic (1988) with a pilot-scale column, 94
mm in diameter, were used to conclude that the mean bubble diameter is proportional to
the 0.2 order o f superficial gas velocity but it does not depend on the liquid flow rate.
Saxena et al. (1990) studied the bubble size distribution in two bubble columns
with different diameters o f 10.8 and 3.5 cm. Both columns were equipped with perforated
plates to introduce the gas. Their results suggested that the bubble size was (1)
independent of superficial gas flow for high superficial gas velocities in the range of 3 .6 to
9.2 cm/s, (2) dependent on column diameter with smaller bubbles being observed in the
narrower columns, and (3) smaller at the column wall than at the center line.
35
Marinas et al. (1993) evaluated the dependence of bubble diameter on gas flow
rate reported by Ahmad and Farooq (1985) and Stankovic (1988). An empirical
correlation for mean bubble diameter was developed for various diffusers:
d B = d B.o + 0.21 x U c EQ.27
dB.o (cm) was a value characteristic o f the porous diffuser used, which was defined as the
value o f bubble diameter when gas input tends to zero. This equation was valid for Ug ^
0.6 cm/s only.
The terminal rise velocity o f gas bubbles, VB, is reported to increase linearly with
bubble size, reaching a maximum velocity of 36 cm/s at 20°C for a diameter of
approximately 1.8 mm. VB tends to decrease gradually with larger diameters until reaching
a minimum velocity of 25 cm/s at a diameter o f approximately 5.5 mm (Masschelein,
1982). The experimental rising velocity curve was approximated by Marinas et al. (1993)
with the following expressions:
1.004
0.5mm< d B < 1.8mm VB = 2 \0 x .d B x -------- x 60 EQ.28
1.8mm < d g < 4mm EQ.29
where f i is the water viscosity in centipoise, dB is expressed in cm, and VB is in cm/min.
36
2.5.4. Specific Interfacial Area
The specific interfacial area (cm'1) is defined as the bubble surface area per unit
volume. This parameter can be determined from the bubble size and gas hold-up
according to:
6-e
a = —— EQ.30
dB
The gas hold-up can be estimated from the superficial water and gas velocities, and the
terminal bubble rise velocity. The resulting expression for the specific interfacial area is:
a = - --------------- EQ.31
( V . * U L) x d .
The sign in front of Ul is negative for counter-current and positive for co-current flow.
Empirical correlations for specific interfacial area and gas hold-up, and their
corresponding measurement methods were studied extensively by Shah (1982) and
Abraham and Sawant (1990).
2.5.5. Ozone Solubility
It is generally agreed in the literature that the solubility of ozone in water follows
Henry’s law, or that the dissolved ozone concentration is proportional to the partial
pressure or concentration o f the gas-phase ozone at a given temperature.
Even though the solubility of ozone has been the subject o f investigation for nearly
a century (Masschelein, 1982), discrepancies can be still observed in the data presented in
the literature. Factors contributing to this problem include the use of different
37
experimental methods and analytical procedures for the determination of the dissolved
ozone concentration and ozone decomposition.
Because ozone is unstable in the liquid phase,, the observed steady-state aqueous
ozone concentration might not be the real saturation (equilibrium) concentration.
Therefore, the mass transfer o f the gaseous ozone into the aqueous phase and the rate of
self-decomposition o f ozone are necessary to determine the true solubility characteristics
(Henry’s law constants) o f ozone in an ozonation system. Furthermore, some o f the
decomposition products might interfere with the analytical determination o f dissolved
ozone. Sotelo et al. (1987 and 1989) studied the absorption o f ozone in water in the
presence o f several buffer systems at various values o f ionic strength, temperature, pH,
gas flow rate, ozone partial pressure, and mixing intensity. With the assumptions of two-
film mass transfer, and second order ozone decomposition kinetics in the presence of
sodium phosphate solution (pH values from 2 to 8.5), the Henry's law constant was
estimated for the temperature range of 0 to 20 °C, and ionic strength of 10‘3 to 0.5 M:
EQ.32
where / is ionic strength in M , and H is expressed in kPa/mole-fraction. This equation is
only valid for temperature and pH values within the experimental ranges investigated, as
there is the possibility o f a shift in the ozone decomposition mechanism (Hoigne and
Bader, 1976; Sullivan, 1980) at higher pH and temperature values.
38
2.6. Reactivity o f Aqueous Ozone
Immediately after dissolving into water, ozone self-decomposes and undergoes fast
reactions with many components commonly present in water and wastewater. Some of
the intermediate compounds produced during ozone decomposition (e.g., hydroxyl
radicals) are much stronger oxidizing agents than ozone itself. Furthermore, the
decomposition o f ozone is also affected by free radical attack. Therefore, the reaction of
ozone with reactants can hardly be separated from ozone decomposition.
2.6.1. Ozone Decomposition
The half-life o f ozone in natural waters is generally on the order of seconds to
minutes. Langlais et al. (1991) reviewed the mechanisms proposed by Staehelin and
Hoigne (1982) and Tomiyasu et al. (1985) for the auto-decomposition of ozone in basic
aqueous solutions. Three types of reaction were identified: initiation, promotion, and
inhibition, each caused by different compounds (Staehelin and Hoigne, 1985; Hoigne and
Bader, 1985; Foml et al., 1982):
1. Initiators are compounds capable of inducing the formation of a superoxide radical
ion, 0 2\ from an ozone molecule. These compounds can be inorganic (e.g.,
hydroxide ions OFT, hydroperoxide ions H 0 2\ and some cations) and organic (e.g.,
glyoxylic acid CHOCOOH, formic acid HCOOH, and humic substances).
Furthermore, ultraviolet light radiation at 253.7 nm is also capable of initiating the
free radical process.
39
2. Promoters are compounds capable o f regenerating the superoxide anion, 02*, from the
hydroxyl radical, OH*. Common organic promoters include aryl groups, primary
alcohols, glyoxylic acid, formic acid, and humic acids. There are also inorganic
promoters such as phosphate species.
3. Inhibitors are compounds capable o f consuming hydroxyl radicals without forming
species that act as chain carriers. The most common inhibitors include carbonate and
bicarbonate ions, alkyl groups C J ^ + r, and tertiary alcohols. Such radical
scavengers thereby extend the lifetime o f ozone in water.
Experimental procedures typically followed for the measurement o f ozone
decomposition include the performance o f batch experiments in either temperature
controlled gas-tight reactors or directly performed in a thermostat stopped-flow
spectrophotometer cell. Stock ozone solutions are prepared with ozone demand-free
water buffered with various acid-base compounds and the pH adjusted by the addition of
acid or base. Ozonated gas (generated from pure oxygen or air) is then bubbled until the
desired dissolved ozone concentration is reached.
A review o f the literature revealed many discrepancies including different apparent
ozone decomposition kinetic order and rate constant, even if the experiments were
performed over the same pH and temperature ranges (Kuo et al., 1976; Sullivan and Roth,
1980; Gurol and Singer, 1982; Staehelin and Hoigne, 1982; Roth and Sullivan, 1983;
Tomiyasu et al., 1985; Sotelo et al., 1987). All studies agree that the overall ozone
decomposition rate increases with increasing temperature and pH, increases in the
40
presence of free radical initiators and promoters, and decreases in the presence of
inhibitors. In the presence of radical scavengers, the ozone decomposition rate is always
found nearly first order with respect to ozone.
2.6.2. Ozone Reactions with Other Compounds
Ozone may react with various organic or inorganic compounds present in an
aqueous solution. These reactions follow two major pathways: direct reaction with
molecular ozone, and indirect reaction with radical species (e.g., hydroxyl radicals) formed
during ozone decomposition in water. Some benzene derivatives may react with ozone or
hydroxyl radicals and form secondary radicals that also act as ozone self-decomposition
and reaction catalysts.
Free radicals are generally much more reactive than molecular ozone, and
therefore the reactions through this pathway are very fast. The main free radicals
generated from ozone decomposition are hydroxyl radicals which were first reported by
Weiss (1935). Hydroxyl radicals are among the most reactive oxidizing agents known to
occur in water. These species can oxidize most types o f organic and inorganic
contaminants present in solution. The reactions are so fast that the corresponding rates
usually approach the diffusion-control rate limit (Hoigne, 1982; van Sonntag, 1986).
Carbonate and bicarbonate ions commonly present in natural waters can be oxidized by
hydroxyl radicals into new radicals, carbonate or bicarbonate radicals, with lower
reactivity (Hoigne, 1982). For the treatment o f drinking water, alkalinity components are
generally present in excess so that the ozone reaction pathway through free radicals is
somewhat limited, and thus reactions with molecular ozone can be predominant.
Ozone molecules can be represented with the following structures in resonance:
5+
.0,
5- 5+ 5- EQ.33
This structure illustrates that ozone molecules could act as either dipoles, electrophilic
agents, or nucleophilic agents. The ozonation o f organic compounds with each of these
structures have been studied by Bailey (1978). It was concluded that the molecular ozone
reactions with organic solutes are extremely selective and limited to compounds with
unsaturated bonds, and aromatic or aliphatic compounds with specific functional groups.
Apparatuses commonly used for the measurement o f ozone reaction kinetics
include stopped-flow, gas-tight batch, and continuous-flow reactors. Each approach
suffers limitations, might only be adequate for a certain range o f reaction rates and might
be limited to specific types o f compounds. The stopped-flow technique is the simplest and
has the best time resolution (equal to the maximum sampling rate of the
spectrophotometer), which is suited to investigate fast reactions. However, direct
measurement o f ozone residual at 258-260 nm is less sensitive than Indigo or Iodometry
method, and cannot be used to measure the ozone reaction with solutes such as aromatic
hydrocarbons that absorb light at this wavelength. Batch reactors are the most precise and
widely used. Experimental conditions can be controlled generally well, except for fast
reactions. Volatilization o f dissolved ozone is a typical problem encountered in batch
42
experiments. Continuous flow reactors are more complex in design but can be used to
avoid some o f the batch approach problems. However, the reliability o f results is highly
dependent on the sensitivity and stability o f the spectrophotometer and the pumps used.
Furthermore, the dynamic range o f time scales is restricted for flow-through technique.
The kinetics for the reactions between molecular ozone and several organic and
non-metal inorganic drinking water contaminants have been reported by Hoigne and Bader
(1983a, 1983b, 1985) and Yao and Haag (1991). Bicarbonate and other radical
scavengers were added to solutions to block out the competing free radical reaction
pathway and lower the rate o f ozone decomposition. Generally, the reaction kinetics were
found to be mixed second order with respect to ozone and contaminants, and the
corresponding rate constants ranged from almost no reaction to 109 M~ls‘l. Most of the
reaction rates were found to increase with increasing pH corresponding to their degree of
dissociation or deprotonation.
Ozone also oxidizes the reduced form of metal species commonly present in
natural waters, e.g., Fe2* and M n2*. The reaction generally follows a first-order kinetic
law with respect to both ozone and the oxidizible compound, and is less dependent on pH.
Reaction rates vary for different species. For example, the reaction with iron is extremely
fast in contrast with the much slower reaction with manganese, the rate o f which could
diminish in the presence of other fast-reacting compounds (Stoebner and Rollag, 1981).
The ozone doses required to completely oxidize metal species are close to the theoretical
43
stoichiometric doses, for example, 0.43 mg-CVmg-Fe and 0.88 mg-(Vmg-Mn (Langlais
etal., 1991).
2.6.3. Reactions in Natural Waters and Ozone Demand Concept
Ozone reaction kinetics in natural waters are generally complex due to the possible
simultaneous/competing involvement in many reaction pathways and chain reactions
including those described previously in the preceding section. As a result, the apparent
ozone decomposition rate could vary widely with water quality and water source (Hoigne
and Bader, 1979). Guittonneau et al. (1992) investigated the ozone decay kinetics in
three different natural waters and observed significant differences in decay rates. The
results also illustrated that hydroxyl radical formation depended on the chemical
composition of the water. Therefore, ozone reaction kinetics in specific natural waters are
not predictable and thus need to be determined experimentally. This task is usually done
by performing batch ozonation tests (Hoigne, 1994). Batch experiments for ozonation of
natural water typically show initial fast decay or practically spontaneous ozone demand
followed by slower decomposition (Hoigne and Bader, 1994). An illustration of this trend
is presented in Figure 3 (Guittonneau etal., 1992). The data presented corresponded to
ozonation experiments performed with University Lake (Chapel Hill, NC) water with
various total organic carbon (TOC) values. The TOC level was changed by diluting the
raw water with distilled water and buffering with phosphate ions in order to keep the pH
constant. The experiments were performed in a continuous-flow reactor at a pH of 6.4,
temperature of 18 ± 1 °C, initial dissolved ozone (after
0
-2
TOC(mg/L) Slope Intercept.

-4
0.8 -0.016 -0.17
1.4 -0.034 -0.33
2.8 -0.137 -0.67
■5
0 10 20 30
Time (sec)
Figure 3. Ozone Decomposition o f Water from University Lake, Chapel Hill, NC

(Guittonneau etal., 1992).
45
mixing) concentration ([ 0 3 ]0) o f 0.66 mg/L, and ionic strength o f 10'3 M . The TOC, and
bicarbonate concentration in the raw water were 2.8 mg/L, and 40 mg/L as CaC0 3 ,
respectively. As depicted in Figure 3, very fast decay was observed at the beginning of
each test followed by a slower first-order decay that increased with increasing TOC
concentration.
The empirical concept o f "ozone demand" is typically used for modeling
convenience (APHA et al’, 1992). Ozone demand is defined as total ozone loss due to
consumption by water impurities and self-decomposition in a period o f time. Because the
overall reaction o f ozone with impurities in natural water is generally much faster than the
ozone self-decomposition, ozone demand is due primarily to reaction with impurities. The
ozone demand concept can be used to avoid the errors and difficulties incurred when
attempting to measure the reaction of ozone with impurities and the corresponding ozone
decomposition (Sotelo etal., 1987; Yao and Haag, 1991). However, this concept was
suspected to be an over-simplification for cases in which the ozone decay kinetics
corresponded to a slower initial fast reaction and longer transition between initial fast
reaction and the subsequent slower first order decomposition (Hull and Singer, 1994).
The initial fast ozone consumption depicted in Figure 3 appears to be inversely
proportional to the dilution ratio, meaning the initial ozone demand is directly proportional
to the concentrations o f the compounds causing this consumption. This result is important
when experimentally determining the amount o f initial fast ozone demand in natural waters
by mixing and thus diluting the sample with stock ozone solution. However, the rate of
46
the subsequent slower decay appeared to increase faster than linearly with respect to TOC
concentration.
Similar observations were found from the results of Singer et al. (1992). These
authors studied the kinetics o f ozone decomposition in synthetic solutions containing
phosphate buffer and Aldrich humic acid. Ozone decomposed very fast initially and then
at a much slower rate. Initial ozone demand and the subsequent reaction rate were found
to increase with increasing humic acid concentration. However, the reaction of initial
demand was found to be relatively slow presenting a transition to the following slower
decay. The slower subsequent ozone decomposition rate also increased with the
concentration o f humic acid added but it was not directly proportional. Hull and Singer
(1994) also model experimental results for the ozonation of synthetic solutions in a bench-
scale flow-through glass column. Various concentrations of bicarbonate and humic acid
were added to phosphate buffered solutions. The instantaneous ozone demand for this
TOC-containing solution was determined from the intercept o f ordinate and linear
regression o f data points against time elapsed after the start o f the experiment, similar to
that done for the batch data depicted in Figure 3.
Duguet et al. (1985) studied the effect of applying hydrogen peroxide during
drinking water ozonation. Three different natural waters were tested. The results
indicated that hydrogen peroxide also exerted ozone demand proportional to the hydrogen
peroxide concentration. However, the slower subsequent ozone decomposition rate was
found to be independent o f hydrogen peroxide dosage.
47
In summary, ozone decomposition kinetics in natural waters can be generally
modeled with two stages: an initial stage in which a given amount o f ozone (ozone
demand) is consumed by certain impurities practically instantaneously; followed by a stage
with much slower decay rate. Microorganisms present in natural waters are usually at
very low population densities, and therefore, generally exert a negligible contribution to
overall ozone demand. The parameters required for modeling are the “ozone demand” in
the first stage and the decomposition rate in the second stage. However, there is a need to
determine whether the assumption o f instantaneous initial demand is applicable for each
specific water.
2.1. Ozone Disinfection
Compared to other conventional water disinfectants, ozone is the most powerful
oxidizing agent, and also the most efficient disinfectant for most microorganisms present
in natural waters. The efficiencies o f free chlorine, chloramine, chlorine dioxide, and
ozone for inactivation of specific bacteria, viruses, and protozoa cysts in terms of CT
values determined in batch reactors are summarized in Table 6 (Hoff, 1986). Ozone can
inactivate 99% o f all types of microorganisms listed at much lower CT values compared to
other disinfectants.
The high biocidal efficiency o f ozone is due to a combination o f having a high
oxidation potential and being a small neutral molecule. Legeron (1982) explained the
relatively higher biocidal efficiency o f ozone by the fact that unlike halogen-based
compounds, ozone does not have a reversible inhibiting effect on intracellular enzymes.
Ozone can oxidize the chemical constituents o f bacteria cell walls before penetrating the
microorganisms, and subsequently attack essential components, such as enzymes, proteins,
DNA, and RNA (Farooq et a l., 1977a; Langlais et a l., 1991).
Table 6. Summary of CT Value Ranges Corresponding to 99% Inactivation of Various

Microorganisms by Common Drinking Water Disinfectants at 5 °C (Hoff,
1986).
Disinfectant
Free Preformed Chlorine
Microorganism Chlorine Chloramine Dioxide Ozone
pH 6 to 7 pH 8 to 9 pH 6 to 7 pH 6 to 7
E. Coli 0.034 - 0.05 95 - 180 0.4 - 0.75 0.02
io
0 .2 -6 .7
1
Polio 1 1.1 -2 .5 768 - 3740
o
0
Rotavirus 0.01 -0.05 3806 - 6476 0 .2 -2 .1 0.006 - 0.06
Bacteriophage f2 0.08-0.18 — -- —
G. lamblia cysts 4 7 -> 150 — -- 0.5 -0 .6
G. muris cysts 36 -630 — 7 .2 - 18.5 1.8-2.0
Reactions with molecular ozone are generally believed to be the predominant
pathway for the inactivation o f microorganisms. The cell liquid offers an approximately
neutral pH and a high concentration o f bicarbonate ions (Langlais etal., 1991); therefore,
49
it is possible that the radical reactions are inhibited. Also, the dispersed microorganisms
might not be able to compete for radicals with dissolved impurities. Labatiuk et al. (1994)
compared G iardia muris cysts inactivation by ozone in the presence o f bicarbonate and
hydrogen peroxide as hydroxyl radical scavenger and magnifier, respectively, in solutions
buffered with phosphate at pH near 6. Experimental results confirmed that reaction with
molecular ozone was the major pathway for cyst inactivation.
The response o f various microorganisms to disinfectants has been reviewed by
Sobsey (1989). Indicator bacteria such as coliforms and pathogens such as salmonella are
very sensitive to inactivation by ozone. Bacteria of sporular forms such as Mycobacteria
are always far more resistant than vegetative forms, and viruses are somewhere in-
between. Viruses are usually aggregated or associated with cell debris in environmental
waters. These shielded viruses were found to be more resistant to disinfection than free
viruses in laboratory experiments (Williams, 1985; Sobsey et a l., 1991). Protozoa cysts
and oocysts generally present the most resistance to disinfection. Wickramanayake et al.
(1984a, 1984b, and 1985) found that CT values o f 0.53 and 1.94 mg-min/L were required
for 99 percent inactivation o f G. lamblia and G. muris cysts, respectively, at 5 °C and
neutral pH. Peeters et al. (1989) and Langlais et al. (1991) suggested that
Cryptosporidium parvum oocysts are the human pathogens most resistant to disinfectants.
Korich et al. (1990) studied the effect of various disinfectants on Cryptosporidium
parvum oocyst viability. The results revealed that C. parvum oocysts were approximately
30 times more resistant to ozone than Giardia lam blia cysts exposed to these disinfectants
50
under the same conditions. CT values estimated for 99 percent inactivation of C. parviim
oocysts were between 5 and 10 mg-min/L for ozonation, approximately 80 mg-min/L for
chlorine dioxide, and several thousands mg-min/L for free chlorine and preformed
chloramines at neutral pH and 25 °C in phosphate buffer solutions. It was concluded that
with the possible exception of ozone and chlorine dioxide, the use of disinfectants alone
might not be practically feasible for the control o f C. parviim oocysts in drinking water.
2.7.1. Experimental Methods
Most ozone disinfection studies have been performed in batch reactors with dose-
response experimental protocols. Samples were withdrawn from a reactor containing
mixture o f ozone stock solution and microorganism suspension at selected time intervals
and analyzed for the microorganism survival fraction and ozone concentration.
Microorganism suspensions were prepared by adding microorganisms into ozone demand-
free buffer solution. Batch studies require a precise monitoring o f ozone concentration
because dissolved ozone would be consumed by impurities present in the microorganism
stock solution at the beginning o f the experiment followed by a subsequent slower
decomposition. Furthermore, ozone loss would also take place by volatilization and by
reacting with the microorganisms during the inactivation process. Because the ozone
concentration decreases during batch experiments, selection o f representative ozone
concentrations for the interpretation o f result is critical and should be regarded with
caution (Majumdare/ a/., 1973; Finch et a/., 1993b; Labatiuk etal., 1994).
51
Semi-batch tests is an alternative which can overcome some of the disadvantages
o f batch tests. Ozonated gas is continuously bubbled into the reactor to maintain a
constant dissolved ozone concentration prior to adding the microorganism suspension.
Therefore, the dose-response microorganism inactivation kinetics can be obtained without
the difficulty o f unsteady ozone concentration.
Inactivation tests performed with a CSTR are not very common due to the large
amount of microorganisms required for flow-through systems. Ozonated gas is
continuously bubbled into the CSTR, and microorganisms are fed at constant density.
Ozone can also be introduced by continuous input from stock solution (Farooq et al.,
1977a; Roy et al., 1980; Zhou and Smith, 1994) to avoid to the possible influences from
ozonated gas-bubbles (Farooq et al., 1977b). The system is allowed to reach steady state
prior to sampling for ozone concentration determination and microorganism viability
assessment
2.7.2. Disinfection Kinetics and Models
The Chick and Watson model for disinfection kinetics proposed in 1908 was based
on the similarity between chemical reactions and microbial inactivation:
dN
— = -k N C n EQ.34
dt
where N is the density o f viable microorganisms at contact time t, C is the disinfectant
concentration, n is the coefficient of dilution, and k is the disinfection rate constant.
52
Integration o f EQ.34 with the assumptions o f constant disinfectant concentration and
mixed second-order kinetics (n = 1) results in the first-order Chick-Watson’s expression:
EQ.35
where No is the density o f viable microorganisms at time zero, and T is exposure time.
According to EQ.35, CT values required to achieve a given inactivation level can be
predicted if the inactivation rate constant k is known. Disinfection rate constants k for
ozone inactivation of viruses and Giardia lamblia cysts, are shown in Table 7 (U.S. EPA,
1991). The k values presented in Table 7 have been estimated based on scarce data (Roy
etal., 1982; Wickramanayake eta l., 1985; Vaughn etal., 1987) and the assumption of
mixed second order kinetic (w = 1). Consequently, a safety factor of two was applied for
experimental k values obtained at 5 °C. k values at other temperatures were assumed to
double compared to the 5°C data for every 10 °C increase. Independence o f rate constant
on pH was also assumed.
The Chick-Watson inactivation model assumes that the microorganisms are a
single strain and that they are all genetically similar. The disinfection mechanism is also
defined to be of the single-hit and single-site type. However, microorganisms in real-life
are much more complex. If the reaction conforms to Chick’s law, a plot of the natural
logarithm of survival fraction versus CT values should result in a straight line with a slope
k. In practice, experimental results do not always follow a linear relationship. Two types
of deviation from Chick-Watson kinetics are commonly observed : (1) “shoulder" or time
lag curve with an initial lag period before first-order kinetics are observed; (2) “tailing” or
decreasing rate phenomenon with a rapid initial decline in viable population followed by
first-order kinetics. These commonly observed deviations are depicted in Figure 4 (Hoff,
1986). The “shoulder” effect could be explained by either the diffusion limitation model
(Haas, 1980) or the series-event model (Severin et a l, 1984). The tailing effect might be
due to the formation o f microorganism clumps, or could be explained by differences in
resistance to disinfectant by different strains, which can be described by the parallel-steps
model (Hacker and Lockowitz, 1976). A multiple multi-Poisson disinfection model
developed by Wei and Chang (1975) is able to explain both “shoulder” and “tailing”
effects by assuming the formation o f microorganism clumps.
Table 7. Second Order Disinfection Rate Constants for Ozone Inactivation o f Giardia
lamblia Cysts and Viruses (U.S. EPA, 1991).
Temperature, °C 0.5 5 10 15 20 25
Giardia lamblia cysts, L/mg-min 2.37 3.64 4.79 7.19 9.60 14.4
Viruses, L/mg-min 5.11 7.67 9.21 15.4 18.4 30.6
54
0
(Log Scale)
Survival Fraction
Contact Time
Figure 4. Typical Semi-Log Microorganism Survival Curves Observed in Disinfection

Experiments.
55
Zhou and Smith (1994) studied the inactivation of E. coli by ozone in a CSTR.
Five inactivation models were evaluated including Chick-Watson, series event, mixed-
strain with two resistances, and two modified models from the last two by the authors.
The modification involved replacing the originally first-order in ozone concentration with
an n-th order since the dominant importance of ozone concentration in determining its
inactivation efficiency over contact time was observed. However, the results indicated
that the Chick-Watson kinetics was still the most adequate in this case.
The shapes o f inactivation curves (time-lag or tailing) are not only dependent on
the nature o f the microorganism, but they might also be an artifact o f not accounting
correctly for system configuration and operating conditions. For example, inactivation of
Giardia cysts in batch experiments showed a tailing curve (Labatiuk et al., 1992b and
1994; Finch et al., 1993a). On the other hand, initial time lag inactivation curves were
found for semi-batch experiments (Wickramanayake etal., 1984a and 1985;
Wickramanayake and Sproul, 1988). Fauris et al. (1986) investigated chlorination of
poliovirus and found that the inactivation curves were altered by ionic strength. A straight
line and initial time lag were observed at high and low ionic strength solution, respectively.
The formation of virus aggregation at low ionic strength was explained by the author to be
the reason for a “shoulder” effect.
2.7.3. Parameters Affecting Disinfection Efficiency
The rate of microorganism inactivation generally increases with increasing
temperature. Wickramanayake and Sproul (1988) studied the inactivation of N. gruberi
56
cysts for temperatures in the range o f 5 to 30 °C at pH 7. The overall inactivation rate
was found to relate to the absolute temperature according to an Arrhenius-type
expression. In contrast, Langlais et al. (1991) found that increasing temperature does not
result in improved overall disinfection with ozone because increasing the temperature also
results in decreasing solubility o f ozone and increasing rate o f decomposition.
Changes in pH might produce changes in the physiology of the microorganism,
ozone solubility, and the decomposition rate. Significant discrepancies exist in the results
from different research groups. For instance, Roy et al. (1982) observed that increasing
the pH would result in greater initial kill of echovirus. In contrast, Farooq et al. (1977a)
observed that the inactivation o fM fortuitum yeast by ozone was independent of pH.
The work o f Wickramanayake et al. (1984a) indicated that controversy exist for the effect
of pH change on the inactivation o f Giardia muris and N. gruberi cysts by ozone, and
concluded that pH effects are organism-specific.
Findings for the effects o f turbidity on disinfection efficiency also show significant
discrepancies. Kaneko (1989) found that the addition o f kaolin as suspended solids
significantly reduced the rate o f poliovirus and T2-phage inactivation. On the contrary,
Scott et al. (1992) performed inactivation tests in pilot-scale studies with natural surface
water and found that turbidity had no influence on disinfection efficiency. Langlais et al.
(1991) concluded that these effects depended on the type of turbidity rather that the
turbidity level. The inhibiting effect of turbidity on disinfection efficiency observed in the
57
literature might have been due to the present o f ozone-demand exerting substances which
compete with microorganism and consume large quantities o f ozone.
Farooq et al. (1977b) studied the effect o f ozone bubbles on disinfection efficiency
under three different conditions: (1) presence of ozone bubbles and ozone residual, (2)
presence of ozone residual only, and (3) presence o f ozone bubbles only. The results
indicated that direct contact with ozone gas bubbles might be more significant than the
reaction with dissolved ozone for the inactivation o f C. parapsilosis and M fortuitam
2.8. Ozone Concentration Measurements
Ozone has a tendency to undergo continuous self-decomposition, volatilization
from solution, and reactions with various radicals and many organic and inorganic
contaminants commonly present in water. Consequently, the determination o f the
dissolved ozone concentration is difficult. The available methods for the measurement of
ozone concentration in either aqueous or gas phase includes iodometry, ultraviolet light
absorption, amperometry, indigo trisulfonate and other colorimetric methods (Stanley and
Johnson, 1982; Gordon et al., 1988; Langlais etal., 1991). These methods and their
limitations are summarized in this section.
2.8.1. Iodometric Method
This method is based on the oxidation of the iodide ion (I') to iodine (I2) by ozone
in an unbuffered potassium iodide solution. The solution is then acidified and the liberated
iodine ( I 3 ') is measured (APHA et al., 1992). However, this method determines the total
58
oxidizing capacity of the solution being analyzed, and thus other oxidizing agents and
ozone decomposition by-products present can interfere. Therefore, the iodometric
determination o f ozone in aqueous phase is not recommended.
The ozone/iodine stoichiometry for this reaction has been found to range from
0.65 to 1.5 (Gordon et al., 1988). Results could be affected by buffer composition and
concentration, pH, iodide ion concentration, sample collection and handling techniques,
and holding time prior to the measurement.
2.8.2. Ultraviolet Light Absorption Method
Ultraviolet light absorbance at 258-260 nm can be used for the measurement of
ozone concentration in both gas and aqueous phases. In the gas phase, the molar
absorptivity at 253.7 nm is generally accepted to be 3,000 ± 30 M 'lcm'1 at STP (Langlais
et al., 1991). However, less consistency is found for aqueous phase molar absorptivity
values reported in the literature which range from 2,500 to 3,600 M '1cm*1 (Chiou et al.,
1994). The potential interferences include organic and inorganic compounds which absorb
U V light at the wavelength range o f 240-300 nm, suspended particles, and color.
2.8.3. Amperometric Method
Ozone can be reduced to oxygen directly in solution with bare electrode
amperometers. The electrode is rotated to establish a diffusion layer, and the electrical
current measured is directly proportional to the concentration of dissolved oxidant
(Stanley and Johnson, 1982). Because o f the high potential for interferences from
59
electrode fouling agents and other strong oxidizing agents, the application o f gas-
permeable membranes has been used to increase selectivity and preventing electrode
fouling by providing extra “resistance” (Gordon et al., 1988). Amperometric methods are
applicable for continuous monitoring o f both gaseous and aqueous ozone concentrations.
2.8.4. Indigo Trisulfonate Methods
2.8.4.1. Indigo Trisulfonate Batch Method
The indigo colorimetric method developed by Bader and Hoigne (1981) is
currently the most widely used method for aqueous ozone concentration measurement
(Chiou et al., 1994) and that proposed in Standard Methods for the Examination of Water
and Wastewater (APHA et al., 1992). The analysis is conducted by mixing the sample in
a batch reactor containing potassium indigo trisulfonate in an acidified phosphate buffer
solution, the ozone decolorized solution is then measured spectrophotometrically at 600
nm. The method is very sensitive, precise, fast, and more selective for residual ozone than
other methods. At pH 2, most o f the ions do not decolorize indigo reagent except
chlorine, bromine, iodine, and oxidized forms of manganese. The interference of chlorine
and manganese can be solved by the addition o f malonic acid and glycine solution,
respectively (APHA et al., 1992).
Bader and Hoigne (1981) used the ultraviolet absorption method to standardize
the indigo method for ozone. An indigo reagent absorptivity o f 20,000 M '1cm‘1 at 600 nm
was obtained based on a molar absorptivity o f 2,950 M 'lcm'1 for aqueous ozone. It
60
should be noticed that if a different value for the U V molar absorptivity of ozone is used,
the indigo reagent absorptivity would change accordingly.
2.8A.2. Automated Indigo Method
Straka et al. (1985) proposed a gas diffusion flow injection analysis (GD-FIA)
technique using the indigo method that can be carried out directly at the location of
sampling. In GD-FIA, the volatile ozone diffuses through a gas-permeable hydrophobic
membrane into an acceptor stream where the reaction with indigo takes place and the
absorbance is measured. It eliminates most o f the interferences of the indigo method.
The GD-FIA indigo method exhibits an improvement over the manual method in
terms o f higher linear range, greater precision and accuracy, higher sampling frequency,
and lower reagent consumption. However, more studies should be conducted with
specific gas permeable membranes, particularly to evaluate the repeated and/or
continuous exposure to ozone solutions in a process control environment (Gordon et al.,
1988; Langlais et al., 1991).
2.8.4.3. Modified Indigo Method
The indigo method developed by Bader and Hoigne for aqueous ozone analysis
was modified by Chiou et al. (1994) to allow for both gaseous and aqueous ozone
determination. Gas or water samples can be extracted from various sampling ports of
specific reactors with a gas-tight syringe containing a known volume o f indigo reagent.
The sample can then be thoroughly mixed with the indigo reagent by shaking the syringe
61
immediately and the absorbance can be measured at 600 nm with a spectrophotometer.
By assuming that the molar absorptivity of gaseous ozone at 258 nm is the well accepted
value o f 3,000 NT'cm'1, the molar absorptivity o f indigo reagent was determined to be
23,150 ± 80 M 'Icm'1at 600 nm , higher than 20,000 NT‘cm*1suggested by Bader and
Hoigne. These authors also investigated the apparent molar absorptivity o f aqueous
ozone at the wavelength o f 258 nm by comparing to the modified indigo method. The
molar absorptivity was found to vary from 2,400 to 3,600 M *‘cm*1 in the investigated
ozone concentration range of 0.4 to 11 mg/L, the variation was verified to be caused by
the inherent interference o f unidentified ozonation by-products.
This method was able to minimize analytical errors associated with ozone
volatilization and decomposition during transportation, and to provide a more consistent
basis for gaseous and aqueous ozone determination. However, the accuracy o f the syringe
used becomes a critical factor.
2.9. Ozonation Models
Development o f mathematical models to assess dissolved ozone concentration in
flow-through ozone contactors under a given set o f operating conditions requires
knowledge of contactor hydrodynamics, the rate o f mass transfer, the chemical reactivity
o f dissolved ozone in water, and the role of various water quality parameters (Hoigne,
1994). The initial fast ozone demand and subsequent ozone decomposition rate can be
investigated in bench-scale batch reactors under well controlled conditions. Mass transfer
rates and contactor hydrodynamics can be determined by performing mass transfer and
tracer tests with the actual flow-through reactor or estimated by empirical correlations
developed from similar systems. The hydrodynamics and dissolved ozone distribution in
ozone bubble-diffuser contactors can be represented with models such as CSTR, PFR,
CSTR cascade, axial dispersion reactor, back-flow cell, and various combinations o f these
models. Conceptual schematics for the axial dispersion and back-flow cell models
together with an example of a model combining CSTR's and axial dispersion are shown in
Figure 5. Most studies presented in the literature have focused in describing contactors
hydrodynamics with few cases dedicated to combine all the parameters to model dissolved
ozone concentration distribution in contactors.
Lev and Regli (1992b) reported a theoretical analysis of the effect o f operating
conditions on ozone concentration assuming PFR conditions. The role of gas transfer,
first order ozone decomposition, and fast ozone demand were also investigated for both
counter-current and co-current flow configurations.
Sauze etal. (1992) studied the hydrodynamics in a pilot-scale ozone contactor.
The experimental results revealed three regions with different levels of dispersion as
described in Figure 5 (c). Ozonation tests were also performed but a predictive ozonation
model developed was not presented.
Rakness et al. (1988) developed a design model to predict overall ozone transfer
efficiency under given operating conditions in full-scale ozone contactors. Factors
considered in the model included diffuser depth, ozone demand, temperature, residence
63
Mass
Transfer
Gas Water Gas Water Gas Water

Flow Flow Flow Flow Flow Flow
Gas Water Gas Water Gas Water

Flow Flow Flow Flow Flow Flow
(a) (b) (C)
Figure 5. Schematics o f Selected Models for Bubble-Diffuser Ozone Contactors: (a)

Axial Dispersion Model (Marinas et al., 1993); (b) Back-Flow Cell Model
(Smith and Zhou, 1994); (c) CSTR/Axial Dispersion Combination Model
(Sauze et al., 1992).
64
time, and bubble size. An overall ozone mass balance equation and the empirical
expression developed by Hughmark (1967) for estimating volumetric mass transfer
coefficients were used to predict ozone contactor performance.
Marinas et al. (1993) reported experiments performed with two pilot-scale ozone
contactors. Each column had a water column height o f 16 feet and was six inches in inner
diameter. One of the contactors had eighteen sampling ports along the column for the
determination of ozone concentration profiles. An ozonation model for counter-current
flow was developed based on the axial dispersion model with the assumptions o f steady
state operation, ideal plug flow in the gas phase, and uniform distribution o f gas bubble
size and interfacial area throughout the column. The model included two-film mass
transfer theory, first order ozone decomposition kinetics, effect of gas flow rate on gas
bubble size, and correlations for the dependence of dispersion number and /10 residence
time on gas and water flow rates. Model predictions o f experimental ozone profiles were
fairly accurate throughout the water column except near the bubble diffuser at the bottom
o f the contactor, where the occurrence of a high mixing zone was not taken into
consideration as part o f the model.
Smith and Zhou (1994) applied both axial dispersion and back-flow cell models
(see Figure 5 (a) and (b)) to theoretically analyze ozone residual distribution in an ozone
contactor for the effects o f back-mixing and flow configuration (counter-current and co-
current). The assumptions made were similar to those used by Marinas et al. (1993). It
was demonstrated that both models predicted similar results for identical inputs.
Disinfection efficiencies were also predicted with the back-flow cell model based on the
assumption that the inactivation kinetics followed Chick-Watson law with pseudo-first
order reaction with respect to microorganism density. These authors did not use the axial
dispersion model to predict disinfection efficiency because o f mathematical difficulties.
However, the back-flow cell model requires an additional fitting parameter, back-flow
ratio, that could not be estimated from experimental data and thus this model was not
deemed practical for the prediction o f contactor performance.
Zhou et al. (1994) used the axial dispersion model previously developed by Smith
and Zhou (1994) to extend the theoretical analysis to assess the role o f back-mixing,
ozone decomposition, and gas-liquid mass transfer rates on the distribution o f dissolved
ozone in a bubble-diffuser contactor for both counter-current and co-current flow
configurations. In addition to the assumptions mentioned in the work by Marinas et al.
(1993), the pressure and corresponding ozone mass transfer rate were assumed to depend
inversely linearly with the column height. However, the necessary adjustments for the
corresponding changes in gas-phase ozone concentration due to pressure changes were
not recognized by the authors. Furthermore, based on the assumption o f decreasing
pressure for the rising bubbles, the interfacial area should increase theoretically due to the
increase in bubble size (if bubbles do not split or joint) and increasing mass transfer rate.
Experimental results from Sauze et al. (1992) were used for the evaluation o f the model.
Empirical equations for oxygen mass transfer in tall fast flow bubble column developed by
Deckwer et al. (1974) were used to estimate ozone mass transfer with the adjustment of
diffiisivity. Faster ozone decomposition rates were assumed within the top-half region of
the column to account for the presence o f ozone-demanding substances.
66
Hull and Singer (1994) also applied the axial dispersion model with the same
assumptions as Marinas et al. (1993) to predict the dissolved ozone concentration profiles
obtained with a bench-scale glass column approximately 100 cm tall and 10 cm in
diameter. Both counter-current and co-current experiments were performed and the feed
solutions were dosed with different concentrations o f total inorganic carbon and humic
acid. Ozone decomposition kinetics and contactor hydrodynamics were obtained by
performing batch experiments and tracer tests, respectively. The mass transfer parameters
were determined by matching model simulations to experimental data for the ozonation of
a pH 3 ozone demand-free solution which had a reaction rate constant o f essentially zero.
The instantaneous consumption of a fraction o f the ozone dosage at the liquid-phase inlet
was included in the model for the tests with humic acid addition. This effect was
represented by applying a negative initial ozone concentration at the liquid-phase inlet.
Generally, model predictions fitted the experimental results fairly well under all operating
conditions investigated except for some tests with humic acid addition. The authors’
explanation for the discrepancy was that the assumption o f ozone decomposition kinetics
according to an instantaneous initial consumption followed by first-order decay may have
been over-simplified.
2.10. Strength and Weakness o f Using CT
Simplicity and ease of use are major advantages o f the CT approach to evaluate
disinfection efficiency. Although most CT values are based on experimental data, the data
base is extremely limited. CT values included in the regulations were extrapolated from
67
very limited number o f experimental conditions to a wider range of water temperatures
and pH values. The validity o f CT values to represent actual disinfection efficiency is also
questionable because inactivation rates do not always obey Chick-Watson’s law, as
described in Figure 4. Kim et al. (1980) investigated the ozone inactivation of
bacteriophage £2 at various ozone dosages, 25 °C, and pH 7 in a semi-batch reactor after
30 s of contact time. Differences in inactivation rates of one hundred fold were observed
for two different initial ozone dosages.
The validity of using the CT concept to represent disinfection efficiency achieved
in ozone contactors is based on the following assumptions (Lev and Regli, 1992a):
A. disinfection kinetics follow ideal first-order Chick-Watson's law (see EQ.35);
B. dissolved molecular ozone is the primary disinfectant;
C. analyses of average disinfectant concentration and of contact time can be separated
(separation of variables), regardless of the mixing pattern taking place inside the
contactor.
However, the third assumption o f separation of variables C and T is mathematically
correct for ideal CSTR and PFR only, and the validity of applying this assumption to full-
scale ozone contactor is doubtful. Also, the first assumption has been challenged by Zhou
and Smith (1994) who studied the ozone disinfection of E. coli in a CSTR. The empirical
coefficient o f dilution factor was found to be as high as 3.3, suggesting that ozone
concentration might have been a much more important factor than contact time in
determining the efficiency o f the disinfection process.
68
The selection o f C and T based on the SWTR guidelines described in Tables 4 and
5, respectively, to effectively represent the average concentration and tl0 contact time o f a
full-scale ozone contactor might also be questionable. Lawler and Singer (1993)
recommended that the t t0 obtained directly from a single tracer study performed with a
series of reactors would be more precise and give far more disinfection credit than adding
together the individual t l0 values for each reactor. The selection of the average
disinfectant concentration recommended by the SWTR was also evaluated by Lawler and
Singer (1993). These authors reported significant differences between theoretically
predicted disinfection efficiencies and credits given by the SWTR. Zhou et al. (1994)
proposed a three-point average method to represent the average ozone concentration.
However, the occurrence o f fast initial ozone demand, an important factor that can alter
dissolved ozone concentration profile significantly (Hull and Singer, 1994), was not
considered by the former authors.
69
Chapter 3. A XIA L DISPERSION MODEL
3 .1. Introduction
The main objective o f this research is to develop a model for assessing disinfection
efficiency achieved in ozone bubble-diffuser contactors. The model approach presented in
this chapter is based on the occurrence of uniform finite axial dispersion with infinite radial
dispersion throughout the water column. Additional assumptions and simplifications
include a negligible volume fraction o f gas bubbles (i.e., gas holdup), and a uniform
distribution o f gas bubble size and interfacial area throughout the column. Deviations
from these conditions will be addressed and evaluated in later chapters. The axial
dispersion model was previously applied by Marinas et al. (1993), Zhou e ta l (1994), and
Hull and Singer (1994) to represent hydrodynamics and ozone distribution in bubble-
diffuser contactors. As discussed in the preceding chapter, the axial dispersion model has
been found to have advantages over other commonly used models that can generally
represent ozone contactor behaviors well with proper selections o f parameters. However,
several modifications are required so that the axial dispersion model can actually be used
for the prediction o f ozone profiles and disinfection efficiencies attained in full-scale ozone
contactors. Major tasks required includes: (1) developing an ozone decomposition model
that includes the consideration o f fast initial ozone demand and subsequent slower ozone
70
decay; (2) implementing a method for the estimation o f ozone mass transfer; and (3)
developing a disinfection model that could represent shoulder/tail inactivation kinetics.
3 .2. Ozone Distribution
The aqueous mass flux, O, o f a chemical, c, can be represented with the following
equation assuming axial dispersion hydrodynamics:
<& = - A £ l — + 0 L c EQ.36
Ax *“
where A is contactor cross section area, E l is the axial dispersion coefficient, and x is
distance in the axial direction. In this expression, the direction of the water flow is
consistent with the positive x axial (see Figure 6)
The derivation o f mass balance expressions for dissolved ozone in a control
volume o f infinitesimal thickness inside a bubble diffuser contactor operated in counter-
current mode is presented in Table 8. The schematic of the control volume in a bubble-
diffuser ozone contactor is illustrated in Figure 6. The first row in Table 8 represents the
overall mass balance equation in a conceptual format. The terms (Mass in) - (Mass out)
equal A<t>*A/. The expression in the second row of Table 8 was derived from the first row
with the assumptions o f axial dispersion, two-film mass transfer, and nth-order ozone
decomposition. The equation in the third row resulted from normalization as described
subsequently. Independent tracer, mass-transfer, and ozone decomposition experiments
are required to obtain the three parameters (d, NL, ND, and N r) in the normalized mass
balance equation.
71
gas out
A
water m water in
x, z=x L
control
volume
AX, AZ=A XL
T
high mixing
zone, z„,
gas m Inf
V
water out
Figure 6. Schematic o f an Infinitesimal Control Volume for a Bubble-Diffuser Contactor

Operated in the Counter-Current Flow Configuration.
Table 8. Developm ent o f Mass Balance Expressions fo r Dissolved Ozone (Based on the A x ia l Dispersion M o d e l) in a C o n trol
V olum e o f a B ub b le -D iffu ser C o n ta ctor Operated at C ounter-C urrent F lo w (see Figure 6).
(Mass Accumulation)2 (Mass In) - (Mass Out) + (Mass Transferred from Gas) + (Mass Reacted)
Mass Balance Tor - k d • On - A • Ax ■At

Dissolved Ozone a (0 • A • Ax) = +A AO-At +K, a / — - O ■A • Ax • At
L \H . -k'ODAAx-At
do_ d 'O 60
Normalize +d- +Nl - \ - - 0 - N d • On - N r • o • d
dQ dz1 dz L \H )
Corresponding
Experimental Tracer Test M ass-Transfer Test O zone D ecom position Test
Method
-j
to
73
3.2.1. Model for Ozonation Chambers
The model used in this research to simulate disinfection behavior by ozonation in
bubble-diffuser contactors is a modified form of the widely used axial dispersion model.
The original model has been expanded to include the concentration o f fast ozone-demand
chemical with the assumption o f mixed second-order reaction kinetics with ozone. The
modified model includes three major variables: dissolved ozone concentration (O),
gaseous ozone concentration (Y), and fast ozone-demand (D). Under steady state
operation and with the assumptions o f ideal plug flow for the gas phase (i.e., uniform
bubbles rising at constant velocity) and first order ozone autodecomposition rate, mass
balance expressions for these variables are:
, d 20 _ dO
d —— + — - N d -0 - Nr O-D = 0 EQ.37
dz z dr
0 EQ.38
, d 2D _ dD
d —^ + — - N r O- D = 0 EQ.39
where H = Henry's law constant (dimensionless)
O = dissolved ozone concentration (mg/cm3)
Y = gaseous ozone concentration (mg/cm3)
D = fast ozone-demand (mg/cm3 0 3)
N l = K l-o -L/U l = number o f transfer units (dimensionless)
No = kj-L/UL - slower ozone self-decomposition number (dimensionless)
Nr = k rU U L = fast ozone demand number (cm3/mg)
S = H- Uq/U l ~ stripping factor (dimensionless)
Ua - QcyA = approach gas velocity (cm/min)
0 G = gas flow rate (cmVmin)
K l = mass transfer coefficient (cm/min)
a = interfacial transfer area per unit volume (cm 1)
d = E l/(U l'L) = dispersion number (dimensionless)
z = x/L = normalized downward distance from top of column (dimensionless)
x = downward distance from top of column (cm)
E l = axial dispersion coefficient (cmVmin)
U L = Qi/A = approach water velocity (cm/min)
A = contactor cross sectional area (cm2)
0 L = liquid flow rate (cmVmin)
The overall rate constant for ozone reaction with fast ozone-demand chemicals is usually
very high so that as long as ozone is transferred, it will be consumed until the demand is
satisfied. The sign in front of dO/dz and dD/dz in EQ.37 and EQ.39 is negative for
counter-current (gas flowing upward and water flowing downward) and positive for co
current (both gas and water flowing in the upward direction) operation. Closed vessel
75
boundary conditions for both counter-current and co-current operation modes are
presented in Table 9. 0,„ , Ym, and D,„ listed in Table 9 are the concentrations o f aqueous
ozone, gaseous ozone, and fast ozone-demand, respectively, entering the reactor.
Table 9. Closed Vessel Boundary Conditions.
counter-current co-current
0\ 0 = 0 m = 0
m + d ddz
0
r= 0
dz r= 0
d0
= 0 o u , = o„ - d —
dr :=\ .-= !
dD
D \-.-0 - D ,n + d = 0
r= 0 - a r= 0
dD
= 0 AL = D. -
“ fdr
n.., = y.
J5.
II
«•
it
76
Predicting the performance of an ozone contactor under a given set of operating
conditions requires knowledge o f reaction kinetics as a function of water quality, mass
transfer, and contactor hydrodynamics. Ozone decomposition kinetics and fast ozone
demand, D , can be investigated in batch reactors under well controlled conditions.
Contactor hydrodynamics and mass transfer rate can be determined by performing tracer
and mass-transfer tests, respectively, in the actual flow-through reactor controlled under
identical operating conditions as those used for ozonation tests.
The nonlinear multi-variable ordinary differential equation system comprised of
EQs 37-39 and corresponding boundary conditions presented in Table 9 could be solved
numerically by employing the Generalized Newton-Raphson method (Chapra and Canale,
1988), after approximating the variables as a function o f distance with corresponding finite
differences. A computer program written in QBASIC for solving these equations is
presented in Appendix A. Illustrations of the output from the computer program are
presented in Figure 7. The following operating conditions and other assumptions made
were: water column height of 19.75 ft, column diameter o f 6 in, water flow rate of 14 gpm
(corresponding to a hydraulic detention time o f 2 min), ozonated gas flow rate of 8.7 scfh,
temperature o f 16 °C, applied ozone dosage of 2 mg/L (corresponding to input gaseous
ozone concentration o f 26 mg/L), instantaneous ozone demand chemical concentration of
0.3 mg/L, dispersion numbers (d) of 0.02, ozone mass transfer rate (K ca) of 1.5 min'1, and
negligible ozone self-decomposition.
Top of Contactor Top of Contactor
CO
15 -- 15
(ft)
Height
-C
op
‘53
X co-current
10 - - 10
c
Water Column
S co-current
_3
O
CJ
ka
OJ
cS counter-current
£
counter-current
5 -- 5
0
0.0 0.5 1.0 1.5 2.0 0 10 20 30
Dissolved 0 3 Cone (mg/L) Gaseous 0 3 Cone. (mg/L)
Figure 7. Axial Dispersion Model Predictions o f Aqueous and Gaseous. Ozone

Concentration Profiles in a Bubble Diffuser Contactor Chamber Operated
Both Counter-Current and Co-Current Modes with the Assumption of
Negligible Ozone Self-Decomposition.
78
Applied ozone dosage , AO, a widely used term to quantify the input to an ozone
contactor, can be calculated with the expression:
A O Q L =Yi n QG EQ.40
An actual mass balance can be expressed assuming that there is no ozone demand
due to self-decomposition:
( 0 „ - ( O m - D m) - Q L„ = Y m Qaj, - y „ EQ.41
the subscripts mt and denote the specific quantities leaving and entering the contactor,
respectively. The changes in total gas mass and pressure was assumed to be negligible in
the model, and therefore, water and gas flows should remain nearly constant along the
column. The aqueous and gaseous ozone concentrations at the bottom and top of the
contactor presented in Figure 7 satisfy EQ.41 for both counter-current and co-current
operations. The computer programs developed in this research for a flow-through
contactor (except tracer program) were checked with EQ.41 to ensure its validity.
3.2.2. Models for Reactive Chambers
Reactive chambers always come after at least one or more counter-current or co
current chambers, and thus fast ozone demand has generally been satisfied upstream. For
reactive chambers, EQ.38 is not applicable, and the gas transfer term in EQ.37 can be
dropped. The mass balance expression for dissolved ozone concentration then becomes.
d^ rr + - Nd O = 0 EQ.42
dz dz
79
This equation can be solved analytically with the two corresponding closed vessel
boundary conditions presented in Table 9.
3.3. Tracer Tests
Equations 1 to 10 based on the assumption o f ideal axial dispersion can be used for
data normalization, parameter calculations, and tracer response predictions. EQs.6 to 9
can be solved by finite differences after selecting a set of known operating conditions.
Theoretical tracer concentration at contactor outlet at various time could be obtained by
solving the tri-diagonal matrix resulting from applying finite differences by the Thomas
algorithm (Chapra and Canale, 1988). A QBASIC computer program listed in Appendix
A was written for this purpose.
The residence time distribution (RTD) curves predicted are affected by the
increments o f dimensionless time and length set in the computer program, as depicted in
Figure 8. The dispersion numbers assumed for the RTD curves predicted in Figure 8 are
0.2 and 0.02, which correspond to CSTR cascades o f 3 to 25 reactors, respectively.
Theses numbers were selected to simulate extreme back-mixing conditions typically
observed in full-scale ozone contactors. Prediction curves generated with various
increment of dimensionless time and length (1/200x1/200, 1/500x1/500, 1/1000x1/1000).
As depicted in Figure 8, model predictions at low back-mixing are more sensitive to the
resolution of the finite differential approach.
80
2.0
(dimensionless)
number of elements = 200

Normalized Tracer Concentration
0 .5
0.0
0 1 2 3 4
Normalized Time (dimensionless)
Figure 8. Model Predictions o f Tracer Curve with Different Dimensionless Time and
Length Increments Used in the Finite Difference Approach.
81
3.4. Mass-Transfer
Mass transfer experiments can be performed by, for example, stripping dissolved
oxygen from water previously equilibrated with air with pure nitrogen bubbles. In such a
case, EQ.39 is not applicable and the reaction terms in EQ.37 can be neglected. From
EQ.38, dissolved oxygen concentration, can be expressed as a function o f gaseous oxygen
concentration,
r s
0 = —------— x
4 Vh ) EQ.43
H Nl dz
where O and Y are aqueous and gaseous oxygen concentrations, respectively.
Substitution of EQ.43 into EQ.37 gives the following expression:
-d-S
HK \ d ± s + (S + 1) • A l M I - o EQ.44
NL J dz3 ^ N J
The sign in front of S/NL and 1 are positive and negative, respectively, for counter-current
flow, and negative and positive for co-current flow. EQ.44 can be solved analytically with
the three corresponding boundary conditions. For co-current flow operation, boundary
conditions were assumed to be the same as indicated in Table 9. For counter-current flow
operation, the same boundary conditions as listed in Table 9 were assumed at the bottom
o f contactor (z = 1) for dissolved and gaseous oxygen. However, the experimental
measured dissolved oxygen concentration at the top sampling port o f contactor can be
used as the open vessel boundary condition at top (z = 0). The reason o f this alternative
selection will be addressed in the Materials and Methods section. A sub-program written
82
in QB ASIC to solve the analytical solution is included in the ozonation program as listed
in Appendix A.
3.5 Batch Ozone Decomposition
Feed water samples can be analyzed in bench-scale batch reactors to determine
initial fast ozone demand and subsequent ozone self decomposition kinetics. The ozone
concentration measured at various times can be analyzed with the following mass-balance
equations:
^ - = - k r D O - k d On EQ.45
- =- k - D O EQ.46
dr
Ozone demand chemical concentration in the influent, D mj reaction rate constants, kr and
kj, as well as the order of ozone self-decomposition kinetics will be evaluated with these
equations and used in pilot-scale ozonation model predictions.
3.6. Disinfection
3 .6.1. CT-Lag Model for G iardia Cyst Inactivation
Bench-scale studies for the inactivation o f G iardia cysts in semi-batch ozone
reactors usually reveal a "shoulder" or time-lag deviation from ideal Chick-Watson's
kinetics as depicted for Giardia muris in Figure 9. Wickramanayake et al. (1985)
83
increase input to 6 times
100
CT-Iag
10 -
slope = kn/ ln( 10)
O Wickramanayake et al., 1985

Cyst Survival Data @ 15°C
CT-lag model prediction
0.0 0.1 0.2 0 .3 0.4 0.5
CT (mg-min/L)
Figure 9. Inactivation o f Giardia muris Cysts in a Semi-Batch Reactor Operated at 15°C

(Wickramanayake et al., 1985) and Corresponding CT-Iag Model.
84
speculated that this phenomenon was due to the resistance of the cyst wall to disinfectants.
Directly applying Chick-Watson's rule to simulate a kinetic with shoulder effect (as used
by U.S. EPA in SWTR) could result in an over/under-estimation o f disinfection efficiency
at low/high CT values. The time-lag model was reported to be able to well represent the
experimental data with shoulder effect (Wickramanayake and Sproul, 1988).
The effect o f ozone concentration, temperature, and pH on the inactivation of
Giardia muris had been studied by Wickramanayake and co-workers (1985, 1988). An
approximately first-order inactivation kinetics with respect to ozone was found, and the
corresponding cyst survival curves versus CT values shown in Figure 9 exhibited an
approximately independent relationship with ozone concentration. Therefore, the time-lag
model can be modified to the CT-lag model by assuming an initial CT value required to
damage the cyst wall prior to cysts inactivation. With additional contact time, ozone
would penetrate into the cyst and the inactivation kinetics would follow Chick-Watson's
law. In order to apply this model to represent the inactivation achieved in ozone
contactors with axial dispersion hydrodynamics, an alternative approach was used in this
research as indicated in Figure 9. The amount o f cysts spiked was increased to a fictitious
value obtained by extrapolating the Chick-Watson's inactivation kinetics to time zero. For
any given time and location, the corresponding predicted cyst survival ratio was obtained
by comparing the predicted viable cyst density between the original spike without
inactivation and the increased spike with inactivation. No inactivation was assumed if the
latter value was larger or for CT < CT-lag. The simulated inactivation curve (two straight
lines) from this model predicted the inactivation o f Giardia muris in a semi-batch reactor
at 15 °C reported by Wickramanayake et al. (1984) reasonably well as depicted in
Figure 9.
3.6.2. Axial Dispersion Inactivation Model
Microorganisms were assumed to have no influence on dissolved ozone
concentration because typical densities observed in drinking water are usually low. With
the axial dispersion model, the mass balance for viable microorganism, N, in the control
volume under transient conditions can be expressed as:
cN , d 2N dN
— = d ~ T :~ + — - /V L /U l EQ.47
CO oz dz
where 0 is normalized time. The sign in front o f dN/dz is negative for counter-current, and
positive for co-current operation mode. R n represents the inactivation kinetics by ozone
(e.g., Ru = kn-0 -N for mixed second-order Chick-Watson's kinetics, O was obtained from
the ozonation model, EQ.37 to EQ.39). Close vessel boundary conditions for N are the
same as O or D listed in Table 9.
For contactors with continuous seeding o f constant intensity of microorganisms in
the influent, water can be sampled at steady-state condition. Therefore, dN/30 equals zero
and N,„ is constant for 6 > 0. For disinfection tests with an initial spike o f microorganism
followed by effluent sample collection at various times, the procedure and result are
similar to pulse-input tracer test and N,„ equals zero for 9 > \ . A simulated residence time
86
distribution (R TD) curve from a pulse-input o f Giardia muris cysts is plotted in Figure 10
together with the predictions o f viable cysts from Chick-Watson’s rule and the CT-lag
inactivation model developed in this research. The operating conditions assumed for the
modified ozonation model are: Ol = 4.4 gpm, Oo = 0.0226 scfrn, L = 16 ft, T = 15 °C, and
0 3 dose = 0.3 mg/L. The k„ and CT-lag estimated from these operarting conditions are
approximately 12.6 L/mg-min and 0.025 mg-min/L (Figure 9), respectively. It is depicted
from this figure that the front o f the RTD curves from the prediction o f CT-lag model is
shortly overlapped with the no-inactivation curve, due to the CT-lag effect as expected.
Cysts survival ratios can be obtained at each time by dividing the viable cyst concentration
by that corresponding to the curve without inactivation (same as conservative tracer test
curve).
The effect o f sampling time on Giardia muris cyst inactivation efficiency achieved
in pulse-input disinfection tests at different applied ozone dosages is shown in Figure 11.
Operating conditions assumed in this figure are the same as those in Figure 10 except the
input gaseous ozone concentration is adjusted to achieve for the various ozone dosages
indicated. It is observed from Figure 11 that a difference o f 0.2 detention time (1 min)
while sampling at approximately one detention time could result in a difference of 0.5-log
or higher in G iardia muris cysts inactivation for applied ozone dosages greater than I
mg/L. The computer program for the estimation of microorganism density was appended
to the ozonation model program listed in Appendix A.
87
without inactivation
Chick's kinetic, kn = 12.6 L/mg-min
with initial lag, CT. = 0.025 mg-min/L
Normalized Number of Live Microorganisms
0 .5 -
0.0
0 1 2 3 4
Dimensionless Time
Figure 10. Residence Time Distribution Curve o f Viable Cysts for a Simulated Pulse-
Input Inactivation Test.
88
15
0 3 dose = 0.1 mg/L
0 3 dose = 0.5 mg/L
0 3 dose = 1.0 mg/L
0 3 dose = 1.5 mg/L
Cysts
0 3 dose = 2.0 mg/L
10
m u r is
G ia r d ia
L o g ,0-K ill of
0
0 1 2 3 4
Sampling Time / Hydraulic Detention Time (dimensionless)
Figure 11. Effect o f Sampling Time on the Observed Inactivation Efficiency of Pulse-
Input Giardia Muris Cyst Disinfection Test.
89
Chapter 4. MATERIALS AND METHODS
4.1. General Use Chemicals and Equipment
Deionized water used for rinsing and solution makeup was obtained from a
NANOpure water system (Bamstead/Thermolyne Corp., Dubuque, IA). The conductivity
of the water produced by this system was 18.3 megaohm-cm. All chemicals used were
analytical reagent grade compounds purchased from various manufacturers.
A digital balance Mettler model AT 200 (Highstown, NJ) with readability and
precision o f 0.0001 mg was used for weight measurements.
The pH was measured with a Fisher Scientific pH/ion meter Accumet model 925
(Pittsburgh, PA) with pH readability o f 0.001 and accuracy o f ±0.002. The pH probe used
was an Orion (Boston, M A ) glass body type, model 81-02, capable of providing a stable
reading to 0.01 pH in less than 30 s. A Fisher (Pittsburgh, PA) brand temperature probe
model 13-620-16 was used for automatic temperature compensation of the pH
measurement.
Ozone was produced by passing commercial-grade oxygen (99.5% purity) or
compressed air (99.9% purity) through an ozone generator PCI Model GL-1 (PCI Ozone
& Control Systems, West Caldwell, New Jersey). The output from the ozone generator
was highly dependent on the gas temperature and flow rate. Generally, the maximum
90
ozone concentration was approximately 8% by weight for pure oxygen feed and 4% by
weight for air feed at room temperature and with the cooling water valve o f the generator
fully open.
4.2. Analytical Methods and Experimental Equipment
4.2.1. Non-Standard Methods
4.2.1.1. Tracer Concentration Measurement
The tracer chemical used in this research was Rhodamine WT, which is a
fluorescent organic dye developed specifically for tracer test applications in environmental
systems. Rhodamine W T was developed to overcome the disadvantages o f the more
widely used Rhodamine B that tends to absorb on many materials and presents greater
toxicity. The maximum emission wavelength o f Rhodamine WT is about 560 run.
Background reading for natural water and wastewater at this wavelength is typically found
to be low.
A Turner Designs (Sunnyvale, CA) Model 10 fluorometer including a flow
through cell was used to measure the concentration o f the dye. This apparatus was
specified to provide a detectability of approximately 0.1 ppb Rhodamine W T in raw
sewage with a ± 1% accuracy. The fluorometer was allowed to warm up for a minimum
time period o f 15 min. A calibration curve including at least three different concentrations
o f the dye was developed prior to each measurement. Calibration standards were
prepared with the actual water tested and were kept at the same temperature as samples to
eliminate the influence o f temperature on measurements. Maintaining baseline signal
constant and remaining in the linear portion of the calibration curve for the duration o f a
tracer test was found to be more important than the accuracy o f the absolute values for
tracer concentration. The latter condition was not as important because all tracer
concentrations were normalized to dimensionless values as described in the previous
chapter. According to the calibration results, Rhodamine W T was found to give accurate
and reproducible linear response of output signal on the fluorometer within the
concentration range o f 0.1 to 160 pg/L, with an all time precision of less than 3% relative
percent difference and r-square correlation coefficient larger than 0.99.
4.2.1.2. Ozone Concentration Measurement
The modified indigo method developed by Choiu et al. (1994) was used in this
research to determine aqueous and gaseous ozone concentrations. The precision o f this
method was affected by the path length of the photo-cell used. The typical detection limits
observed with the 1-cm path quartz cell used in this research were 0.02 mg/L and 0.002
percent by weight at 20 °C for aqueous and gaseous ozone concentrations, respectively,
with a relative percent difference of less than ± 5%. Light absorbance at 600 nm of the
sample/indigo mixture was measured with a Shimadzu spectrophotometer Model U V
160U (Shimadzu Scientific Instruments, Columbia, M D ). Method 4500-03 in Standard
Methods for the Examination o f Water and Wastewater was followed for preparing the
indigo stock solution and two working solutions. The only exception was that the dilution
92
ratios were changed to approximately 20 and 4 in the preparation o f reagents I and II,
respectively, for the particular application o f this study.
Instead o f transporting aqueous sample into a batch reactor and mixing as
described in the Standard Method 4500-03, gaseous or aqueous samples and indigo
solution were extracted and mixed in a gas-tight syringe to minimize ozone losses by
volatilization and self-decomposition. The syringes used in this research were gas-tight
Hamilton Series 1000 (Reno, N V ) with capacities o f 10 and 25 mL, and a gas-tight 100
mL syringe model 100MAXLLGT (Scientific Glass Engineering Pty. Ltd., Ringwood,
Australia). The accuracy obtained by balance tests with pure water were approximately ±
0.1, 0.2, and 1 mL for the 10, 25, and 100 mL syringes, respectively.
The ozone concentrations were calculated with the following expressions:
AA
O = 48,000 x EQ.48
M tnd
V .(A —A )
for gaseous samples: A/leoo = 1------1------ — EQ.49
V - A - ( V + V )■ A
for aqueous samples: AA600 = —----- -p----- -------— EQ.50
S
where O = ozone concentration (mg/L)
M mj = molar absorptivity o f indigo reagent at 600 nm (M 'lcm 1)
93
AA6oq~ absorbance drop at 600 nm
V, = indigo reagent volume (mL)
Vs - ozone sample volume (mL)
A, = initial absorbance of indigo reagentat 600 nm
Am = absorbance o f sample and indigo reagent mixture at 600 nm
Molar absorptivity o f indigo solution used in this research was 23,150 M '1cm‘1as reported
by Chiou et al. (1994).
4.2.1.3. Bubble Size Measurement
The bubble size distribution was measured for each ozonation experiment run with
the pilot-scale ozone contactor. The images were taken with a Ricoh video camera
recorder Model R-86 Super (Tokyo, Japan), and displayed with a JVC S-VHS video
cassette recorder Model HR-S4700U (Tokyo, Japan). In order to minimize the parallel
image error between the nearest and farthest bubbles inside contactor with respect to the
camera, the camera was set far away from the contactor and the maximum zoom ratio
(12X) was used. Side spotlight was applied to increase the visibility of bubbles.
Enlarged images o f bubbles on television were compared against the stainless steel
sampling tube inside the ozone contactor, and the paused images were measured directly
from a T V screen. The bubble size was obtained from the ratio of bubble length to
sampling tube diameter times the pre-measured tube diameter, 0.65 mm, assuming bubbles
were spherical. Bubble dimensions on the horizontal direction were distorted by the
cylindrical contactor and therefore not measured.
94
A calibration procedure was performed in a cylindrical glass container with inner
diameter of one foot. Pre-measured glass beads with two major sizes were used instead of
gas bubbles to determine the accuracy and precision o f this experimental approach. The
beads were dropped evenly in the axial direction at distances o f 1/4, 2/4 (center), 3/4
diameter from the front and side walls of the cylindrical container with respect to camera
location to investigate the maximum optical distortion and perspective errors. Sizes of
beads were also determined from the images on T V by comparing to a piece o f sampling
tube located at the center o f contactor. The optimum equipment set-up conditions, such
as shutter speed and back light were also determined as part o f the calibration procedure.
4.2.2. Standard Methods
Major water qualities parameters which might affect ozonation results such as
alkalinity, pH, total organic carbon (TOC), and various metals and ions were analyzed
according to methods specified in Standard Methods fo r the Examination o f Water and
Wastewater, 18th Edition (APHA et al., 1992), as listed in Table 10.
TOC was measured with a total organic carbon analyzer Model DC-80
(Rosemount Analytical Inc., Santa Clara, CA). The detection limit of this equipment was
0.1 mg/L with a repeatability o f less than ± 5% difference. Other equipment used included
an Atomic Absorption Spectrophotometer Model Video 12 AA/AE (Thermo Jarrell Ash,
TJA Corporation, Waltham, M A ) with graphite furnace, and an Ion Chromatography
System 20001 with IONPAC AG4A guard column and AS4A analytical column (Dionex
Corporation, Sunnyvale, CA).
95
Table 10. Standard Methods for the Measurement o f Selected Major Water Quality
Parameters (APHA et al., 1992).
Method
Anions Fluoride 4110 B
Chloride 4500-cr F
Bromide 4500-Br C
Nitrate 4500-N03‘ C
Sulfate 4500-S042' B
Cations Sodium 3500-Na B, C
Potassium 3500-K B, C
Calcium 3500-Ca B, C
Magnesium 3500-Mg B, C
Other Metals Zinc 3500-Zn B, C
Manganese 3500-Mn B, C
Iron 3500-Fe B, C
Copper 3500-Cu B, C
Cobalt 3500-Co B, C
Other Parameters Alkalinity 2320 B
TOC 5310 B
pH 4500-H B
96
4.3. Pilot-Scale Experiments
4.3.1. Experimental Apparatus
A schematic o f the pilot-scale ozone bubble diffuser contactor used for this study
is shown in Figure 12. The unit included a pre-treatment system and the actual ozone
bubble-diffuser contactor. Ground water from a local dedicated well was pretreated in a
stainless steel aeration tank having a effective volume o f400 gallons, and subsequently
passed through a sedimentation tank. Iron and manganese present in the ground water
were expected to be oxidized and at least partially removed in these pre-treatment units.
The contactor had an internal diameter o f six inches and a height o f approximately
20.5 feet. The unit consisted o f seventeen coupled Pyrex glass pipe sections and a special
top section. The water and gas piping and valve systems were designed so that the
contactor could be operated in either counter-current, co-current, and recycle flow modes.
Under normal operating conditions, the water column was maintained at a constant height
of 19.75 feet by a water level control unit and water effluent valves. The materials used
for all parts of the experimental apparatus in contact with water and ozonated gas were
borosilicate glass, stainless-steel SS-316, and Teflon. Three photos corresponding to an
overview of the pilot-scale unit, the sampling of aqueous ozone, and a close-up view o f a
section of the contactor with gas bubbles inside are inculded in Appendix B.
A total o f seventeen aqueous sampling ports located throughout the water column
and an additional port located at the effluent line were used to withdraw water samples for
the determination of dissolved chemical concentration profiles inside the contactor.
97
off-gas
sam pling port waste
tracer injection p o rt
off-gas treatm ent
effluent
608.1 cm
tw o-w ay
valves
556.3 cm
508.1 cm
462.4 cm
sam pling ports
4 15.4 cm
w ater level
control unit 369.7 cm
effluent
-eM
cooling
w ater
322.5 cm
* 4- 276.8 cm
ozone
generator
229.8 cm
recycle
water
oxygen
flow | 184.1 cm
aeration input
meter
tank
gas
151.2 cm feed-gas y input
flow m eters '
135.9 cm
sedim entation
tank 118.6 cm
groundw ater
well 103.0 cm
v zf[ pa- 86.3 cm ozone

m onitor
70.1 cm
45.7 cm
ozone . feed-gas
quench sam pling port
chem ical 17.7 cm
0 cm ^7) pressure
fluorometer
effluent gage
Figure 12. Schematic o f Pilot-Scale Ozone Bubble-Diffuser Contactor.
98
The distance between sampling ports was shorter in the lower part of the contactor to
monitor sharper changes in dissolved ozone concentration near the diffuser. The aqueous
sampling ports along the column consisted o f 1/4-inch stainless steel tubing with the end
inside the contactor bend 90° upward to avoid bubble intrusion, and the external end
connected to a sampling valve. These sampling ports were adjustable so that the aqueous
samples could be withdrawn from either the center or at any other location along the
diameter o f the column. Two gaseous sampling ports were located at the gas inlet and
off-gas lines for measuring gaseous ozone concentrations.
Gas was introduced into the contactor by means of bubble diffusers. The diffusers
used were one-inch in diameter spherical diffusing stones purchased from Fisher Scientific
(Pittsburgh, PA). The diffusing stones were made of fused crystalline alumina grains and
had an average pore size o f 60 pm. Two diffusing devices were designed to investigate
the effect o f diffusing surface and diffuser distribution on overall contactor back-mixing
and gas-liquid mass transfer. Figure 13 shows these diffusing devices and their top view
indicating the relative positions to the bottom o f the column. One device was a single
diffuser located in the center and the other one included four diffusers distributed
symmetrically.
Three parallel rotameters were used to control and measure the gas flow rate in the
overall operating range o f 0.001 to 1 standard cubic feet per minute (scfm). Another
rotameter was used to control and monitor the water flow rate between 2.45 and 19.79
gpm. All rotameters were various Fischer & Porter (Warminster, PA) models with
99
Top View
lutlet
m & m m sm
Gas In
Water Oul
Side View
Figure 13. Side and Top View of Diffuser Systems with (a) One, and (b) Four Diffusing
Stones.
100
stainless steel rotary bead and casting. Calibration results by volumetric measurement
revealed a precision of less than 5% relative differences, and an accuracy of 95 to 105 %
volume recovery for all meters. Measurements for different gas input (air, nitrogen, and
ozonated gas) were adjusted by the gas densities (Perry et al., 1984).
The video camera recorder was set at a height o f 70 cm from the bottom of the
contactor, and at a distance o f approximately two meters away to take the images of gas
bubbles. As discussed previously, bubble sizes were estimated by comparing the images of
gas bubbles and the third sampling tube from the bottom o f contactor. Two side
spotlights were used to increase the visibility of bubbles.
4.3.2. Experimental Matrix
The pilot scale bubble-difiuser contactor was used to perform ozonation and
disinfection tests. Corresponding tracer and gas-liquid mass transfer tests performed were
under identical conditions to estimate the back-mixing and mass-transfer parameters
required for modeling. Operating parameters investigated included water flow rate, gas-
phase ozone concentration, ozone dosage, and diffuser type. The matrix of operating
conditions used for these experiments is presented in Figure 14. Feed-gas ozone
concentrations were controlled at target values of 1, 2, and 4 % by weight. Hydraulic
retention time was adjusted to 2, 4, and 8 minutes, corresponding to water flow rates of
14.3, 7.13, and 3.56 gpm. The gas flow rates were adjusted to achieve applied ozone
101
Figure 14. Experimental Matrix for Operating Conditions.
102
dosages o f 1, 2, and 4 mg/L. Both diffuser sets presented in Figure 13 were used for each
set of operating conditions shown in Figure 14.
4.3.3. Ozonation Tests
Ozonation tests were performed to determine dissolved ozone concentration
profiles throughout the water column inside the contactor. For each ozonation test run,
the liquid and gas flow rates were first controlled at the required levels, and then the
designed input ozone dosage was achieved by slow adjustment o f the voltage knob of the
ozone generator until the measured ozone concentration in the input-gas reached the
target value. Gaseous and aqueous ozone concentrations were determined after steady
state concentrations were reached. A membrane electrode connected to an ozone monitor
Orbisphere Model 27504 (Orbisphere Laboratories, Neuchatel, Switzerland) was used for
the continuous on-line monitoring o f feed-gas ozone concentration during experimental
runs. However, the membrane electrode monitor was only used as an indicator for system
stability. The reported ozone concentrations were always based on analyses by the
modified indigo method described previously. After changing operating conditions and
obtaining steady readings on the membrane electrode monitor, waiting periods of at least
three hydraulic retention times were allowed before starting sampling procedures.
Aqueous and gaseous ozone concentrations were determined by the modified
indigo method with the 10 and 25 mL gas-tight syringes, respectively, according to the
following procedures: (1) a volume V, mL of indigo reagent was first extracted into the
syringe; (2) a volume Vs mL o f gaseous or aqueous ozone sample was then extracted into
the syringe directly from the sampling port; (3) the syringe was shaken vigorously
immediately after sample extraction to mix reagent and sample for aqueous samples, or
absorb ozone gas into indigo solution for gaseous samples; (4) mixing was applied for 5 s
for aqueous samples, and 1 min for gaseous samples to provide sufficient reaction between
ozone samples and indigo reagents; (5) reacted indigo solution was sipped through Teflon
tubing into the quartz cell o f a spectrophotometer to measure the absorbance at 600 nm,
and the ozone concentration was calculated with EQs. 48 to 50. Volumes V, and Vs were
selected so that the decolorization of the indigo reagent was within the range of 50 to 80
% with respect to the initial color intensity. Feed-gas samples for applied ozone dosage
determination were taken at the beginning and end o f each experiment to check the
variation of the output from the ozone generator. At least duplicated samples were taken
for each ozone concentration measurement to ensure the precision requirements.
A wet scrubber unit containing sodium nitrite was used as an ozone quenching
chemical and Rhodamine W T as an indicator to destroy ozone residual in the off gas from
the contactor. Ozone residual in the contactor effluent was destroyed by mixing with a
quenching solution o f sodium nitrite or sodium thiosulfate and filtering through a GAC
column.
4.3.4. Tracer Tests
Tracer curves obtained were analyzed to ascertain the extent of back-mixing, and
the occurrence of short-circuiting or dead-water zones inside the contactor. Parameters
104
investigated in the analysis o f axial dispersion include water and gas flow rates, and
diffuser type.
A conceptual experimental schematic o f the setup used for tracer tests is illustrated
in Figure 15 All tests were performed by adding a pulse input of the tracer Rhodamine
WT at the influent line just upstream from the contactor inlet after the system was allowed
to reach steady state. Rhodamine W T concentration was measured on-line. A side stream
from the contactor effluent was connected to the flow-through cell of the Turner Series 10
fluorometer for this purpose. The side stream flow rate was controlled to at least 30
mL/sec to ensure plug flow condition, and the retention time of the side stream including
flow-through cell was less than 2 s to minimize interferences by additional back-mixing on
tracer test results.
Tracer data were collected with a microcomputer (Gateway 2000, 386/25) data
acquisition system in order to increase the sampling frequency and data accuracy.
Software was developed to collect and analyze tracer test data with a multifunction I/O
board Model PC-LPM-16 (National Instruments Corporation) immediately after
performing the test. The parameters determined were dispersion number, mean residence
time, /10 residence time, and total tracer mass. Hydrodynamic conditions inside the
contactor were modeled by assuming axial dispersion with closed vessel boundary
conditions (Levenspiel, 1972).
The data acquisition system was capable o f recording voltage signals from the
fluorometer once per second. A noise rejection threshold o f 0.02 volts corresponding to a
105
gas out tracer injection

1
water m
<
tracer plume
Dispersion number
Mean residence tim

program — ;
110residence time
Tracer mass
ooo
computer
I/O Control
Board
signal out
gas m
water out
oooo
Fluorometer
Figure 15. Conceptual Schematic of Experimental Setup and Data Acquisition System
Used for Tracer Tests.
106
Rhodamine W T concentration o f approximately 0.1 pg/L was selected. Signals weaker
than the noise rejection threshold were ignored in the tracer curve analysis.
Monitoring o f tracer concentration in the contactor effluent stream was started at
the time o f tracer addition and stopped after five hydraulic residence times (the time
required for tracer to be no longer detected was usually less than three hydraulic residence
times of the contactor). A total o f 300 data points were recorded for each tracer run
during the period o f five hydraulic residence times. Quality control for the tracer tests was
achieved by performing tracer mass balances and checking that the tracer mass recovered
was within five percent o f the mass added.
Factors affecting precision and accuracy of tracer concentration measurements and
the corresponding actions undertaken to minimize interferences were:
1 pH. The emission intensity o f Rhodamine W T decreased significantly below pH 5 ,
but stayed nearly constant above pH 6. The pH o f all standards and ground water for
tracer experiments were always above pH 6.
2. Temperature. The emission intensity of Rhodamine W T varied inversely with
temperature. Tracer tests were run at room temperature without special temperature
control. Standard solutions were placed in a water bath with temperature control to
ensure that the temperature o f the standards was within ± 0.5 °C of that for the
ground water at the time o f the experiment.
3. Consumption. Rhodamine W T is stable in the presence of oxygen and sunlight.
Although it could react with some strong oxidizing agents such as ozone and free
107
chlorine, it was demonstrated by performing mass-balances that no decomposition
took place in the absence o f these strong oxidizing agents.
4. Background Interference. The maximum emission wavelength o f Rhodamine WT is
about 560 nm. The background interference was minimized by preparing the
standard solution with the experimental ground water and suppressing the
background adsorption for each tracer experiment.
The fluorometer was calibrated with three Rhodamine W T standards, 10, 25, and
50 (ig/L prior to each tracer test. Standards were made up with ground water to suppress
background interferences, and were immersed in the feed water inside sedimentation tank
to control the temperature to that o f the experimental ground water. The injection volume
of tracer was selected so that the maximum concentration at the outlet would range from
30 to 50 pg/L.
4.3.5. Mass Transfer Tests
Mass transfer tests were performed with the pilot-scale contactor under identical
operating conditions as corresponding ozonation experiments to evaluate the overall mass
transfer efficiency. As mentioned previously, the mass transfer efficiency could be
assessed with an overall volumetric rate constant KL-a, the product of the mass transfer
coefficient K L and the interfacial area per unit volume a. In general, the mass transfer rate
increases with increasing superficial gas velocity, number o f bubbles, and temperature, and
108
decreasing liquid viscosity. The presence of fast ozone demand could also increase the
ozone mass transfer rate.
Ozone tends to be unstable in water, especially in the presence of reduced form o f
inorganic species and organic impurities and at higher pH. Consequently, alternative
chemicals of more stable properties such as carbon dioxide or oxygen are used to
determine the mass transfer characteristics o f reactors for a given set of operating
conditions. In this research, mass transfer tests were performed by stripping oxygen from
the feed-water into nitrogen gas.
The initial procedure was to turn up fully the air flow in the aeration tank of the
pre-treatment system to ensure the saturation o f dissolved oxygen in the feed-water. The
input gas in the pilot-scale contactor was switched to commercial grade nitrogen
(99.995% pure) applied at the same flow rate as that used for the corresponding ozonation
test, to strip out dissolved oxygen in the oxygen-riched water. Water samples were taken
from feed water and water column sampling ports into 300 mL BOD bottles with a Teflon
sampling tube after the system was allowed to reach steady state. One end of the sampling
tube was connected to the sampling port with the other end inserted all the way to the
bottom of the BOD bottle. Water was allowed to overflow for at least 600 mL, after
wasting the initial 50 mL extracted from the sampling port to ensure that the sample was
more representative. The BOD bottles were then fitted with an oxygen probe for
immediate dissolved oxygen concentration determination. A YSI Model 54A (Yellow
Springs Instrument Co., Inc., Yellow Springs, O H) oxygen meter with accuracy o f ± 0.1
mg/L, and a self-stirring BOD bottle probe Y SI G -05519-40 were used for this purpose.
109
The oxygen meter was allowed to warm up for at least 20 min, and was calibrated with an
air-saturated solution prior to each test. Three minutes were waited prior to recording the
oxygen meter reading (approximately one and a half minutes were required for the meter
to reach a steady state reading). All samples were taken in duplicate and the dissolved
oxygen in the feed water was measured before and after sampling from the contactor ports
for quality control.
The data obtained were initially analyzed with the axial dispersion model assuming
that the mass transfer parameter KL*a was constant throughout the column. Dissolved
oxygen concentration profiles were predicted by solving the oxygen mass-transfer
equations EQs.43 and 44. Although the dissolved oxygen concentrations in the feed water
were measured, the oxygen concentration in the top boundary condition (inlet boundary
condition for closed vessel) was unknown for counter-current flow because a fraction of
the dissolved oxygen in the feed water was stripped out at the interface with the head
space at the top o f contactor. Therefore, an open vessel boundary condition with the
dissolved oxygen concentration measured at the highest (seventeenth) sampling port was
used instead o f the corresponding close vessel boundary condition.. The mass transfer
rate parameter KL»a were then estimated by best-fit of the model to the experimental data.
A second approach to represent mass transfer data was used to account for
discrepancies between data and predictions with the simpler axial dispersion model. This
alternative approach based on the occurrence o f high mixing zones at the top and bottom
of the contactor and a central zone with axial dispersion conditions is described in Chapter
6 of this dissertation.
110
The estimated oxygen mass transfer rate constants were converted to ozone mass
transfer rate constants by applying a diffusivity correction factor. Various conversion
ratios found in the literature were investigated. For example, the ratio o f mass transfer
coefficient of oxygen to ozone was reported to be approximately 2 and 1.2 by Matter-
Muller (1981) and Gurol and Singer (1983), respectively. Stankovic (1988) compared the
mass transfers o f oxygen and ozone with the experimental results performed by the author
and reported in the literature and found that the rate of oxygen transfer was approximately
twice as fast compared to that for ozone.
4.3.6. Giardia M uris Inactivation Tests
Two pilot-scale disinfection tests were performed at a hydraulic detention time of
4 min, gas flow rate of 4.3 scfh, and applied ozone dosages of approximately 0.5 and I
mg/L. G iardia muris cysts were used as indicator microorganisms instead o f the
regulated human pathogen Giardia lamblia. The Giardia muris Robert-Thompson mouse
model strain has been maintained at Oregon Health Science University, Portland, Ore., in
Swiss-Wepster white mice from Simonsen Laboratories, CA.
The procedures were: (1) Run the ozonation test at target operating conditions
and ozone dosage; (2) Prepare Giardia muris cyst suspension at a target cyst population
density o f approximately 108 cysts per liter; (3) Wait for steady-state conditions of
ozonation test; (4) Pump cyst suspension at a steady flow rate o f approximately 150
mL/min to the inlet line o f the contactor; (5) Wait three additional hydraulic detention
times for system to reach steady-state and start collecting effluent sample. The total
Ill
number o f cysts collected was 3 x 106, sufficient to achieve a reasonable estimation of 2-
log inactivation. Seven liters of samples were required to collect this number of cysts. A
concentration o f 10 percent sodium thiosulfatc solution was delivered to the sampling line
at a flow rate o f 2 mL/min to quench residual ozone.
Viability o f Giardia muris cysts was determined by the in-vitro excystation method
suggested by the U.S. EPA (1989), modified slightly to include the counting procedures
proposed by Chiou (1994). The disinfection test was performed on day 4 after harvest,
and the collected cyst sample was excysted one day after performing the test. The
experimental procedures for assessing the viability of Giardia muris cysts inactivation
were described by Waller (1994).
4.4. Bench-Scale Ozone Decomposition Tests
Feed water samples were analyzed in bench-scale batch reactors to determine
initial fast ozone demand and subsequent ozone self decomposition kinetics. Ozone losses
via volatilization during transportation and reaction (Hart et a i, 1983) were minimized by
limiting the contact with air. A 100 mL gas-tight syringe was used as the reactor for this
purpose.
Stock ozone solution was prepared by continuous bubbling o f ozonated gas
through two liters o f acidified deionized water inside a reactor immersed in a water bath
controlled at the same temperature as that in the feed water. Dissolved ozone
concentrations were controlled by adjusting the output o f the ozone generator. Acidified
deionized water (pH < 5) was obtained by adding two drops o f concentrated sulfuric acid
112
into two liters o f deionized NANOpure water. The stock ozone solutions were not
buffered so that the pH values o f the water samples after mixing and the pH variations
during the test would be roughly the same as in the ozonation tests. Stock ozone
solutions were capped and ready to use after bubbling for at least 20 minutes. In order to
minimize contact o f the ozone stock solution with air, a rubber sealer was used on the cap
for needles to pierce through while withdrawing solution.
4.4.1. Initial Fast Ozone Demand Tests
The experimental design for the measurement o f initial fast ozone demand was
based on the assumptions that the initial fast ozone demand was primarily due to the
impurities which would react rapidly with ozone, and was directly proportional to the
concentrations o f those impurities (and therefore not affected by the dilution of samples)
It was also assumed that these rapid reactions would be completed within 30 s. Because
natural water usually contains very high concentration o f radical scavengers such as
carbonate, it was anticipated that the amount of dissolved ozone consumed by subsequent
ozone self-decomposition within the 30 s duration of the test would be negligible
compared to the initial fast ozone demand due to the relatively slow reaction rate.
A conceptual schematic of the experimental set-up used is depicted in Figure 16.
Measurement o f fast initial ozone demand was accomplished by comparing the
113
Sample Reference
W ater Bath Solution
Solution
Stock Ozone Solution
Sample Water
§
M ixin g 30 seconds ft
i Indigo Solution
§ I
I Stirring Bar I
A
3-W ay V alve
U V Cell
Waste
U V Light Absorbance at 600 nm
Figure 16. Schematic of Experimental Apparatus Used for the Measurement o f Initial Fast
Ozone Demand.
114
absorbance differences between sample and reference solutions reacted with indigo
solution at the wavelength o f600 run. Sample or reference solution were prepared by
drawing Va mL o f ozone stock solution with dissolved ozone concentration of
approximately 3 mg/L, and Vs mL of feed water sample (sample solution) or pure water
(reference solution) into a 100 mL syringe. These samples were mixed with magnetic
stirrers for thirty seconds, and then a volume of V, mL o f indigo solution was drawn and
mixed again and the resulting absorbance at 600 nm was measured. The initial fast ozone
demand in 30 sec, Z)30, was calculated with the following equation:
D 30 = 48,000 x '( F° + V> +V ') E q 5 1
where AA6 0 0 was the absorbance difference between sample and reference solutions at 600
nm. Because of the existence o f initial fast ozone demand in the sample solution, less
aqueous ozone would be available for reaction with the indigo solution compared to the
reference solution, and therefore, the sample absorbance at 600 nm should be higher
compared to the reference.
The experiments were performed in duplicate with the exception that the sequence
of sample and reference solutions preparation was reversed in order to minimize and
compensate any potential errors incurred during experimental procedures and aqueous
ozone losses in the ozone stock solution. An extra experiment was performed to
investigate the effect o f sample dilution. A set of sample and reference solutions with 85
mL of sample/water and 15 mL of ozone stock solution (volumes Vs = V 0 = 3 0 mL were
115
used in all other experiments) was used for this purpose. The effect o f ozone dosage on
D 3 0 measurement was also evaluated by varying the ozone concentration in stock ozone
solution.
An experiment was also designed to evaluate the accuracy of the Syringe Indigo
Method. The ground water sample used in the preparation o f sample solution was
substituted with indigo reagent I diluted to the absorbance o f 0.3 at 600 nm with pure
water. This solution could quench an aqueous ozone concentration of approximately 0 .62
mg/L for the same volume, or provide an ozone demand o f approximately 0.62 mg/L. A
reference solution was prepared similarly using pure water instead of dilute indigo
solution.
Potential interference o f ground water on indigo absorbance at 600 nm was
assessed by mixing Va mL o f deionized water, Vs mL o f pre-treated ground water and V,
mL of indigo solution.
4.4.2. Ozone Self-Decomposition Tests
Ozone self-decomposition tests were designed to investigate the kinetic of aqueous
ozone self-decomposition in water samples after the initial fast demand.
While applying (mixing) ozone to sample, the volume ratio of ozone stock solution
to ground water sample was kept as low as possible (ozone concentration in ozone stock
solution was maximized for this purpose) to minimize the dilution effect on kd. A
sample/ozone solution ratio o f 85/15 was used in this study. Volume ratios higher than
this value did not provide enough dissolved ozone for initial fast ozone demand. The
116
effect of applied ozone dosage and dilution ratio on kd was evaluated by varying ozone
concentration in stock solution and mixing ratio (50/50), respectively, and will be
discussed in a later chapter. A reference test using direct bubbling o f sample instead of
adding ozone solution to ozonation was also performed for comparison.
Two methods were designed to investigate residual ozone decomposition kinetics:
syringe and stopped flow methods. The Syringe Indigo Method is based on the use of a
gas tight syringe as a batch reactor and the determination o f the dissolved ozone
concentration by proposed Method 4500-03 in Standard Methods f o r the Examination o f
Water and Wastewater (APHA et al., 1992). The Stopped Flow Method was based on
the direct determination o f the aqueous ozone concentration by measuring the absorbance
at 256 nm inside a 1-cm quartz flow-through cell installed in the cell holder o f a
spectrophotometer.
4.4.2.1. Syringe Indigo Method
A conceptual schematic o f the experimental set-up used to investigate ozone self
decomposition kinetics is shown in Figure 17. Sample solutions were prepared by
withdrawing 15 mL o f ozone stock solution into a 100 mL syringe containing 85 mL of
groundwater sample and a small stirring bar. The syringe was capped and put in a water
bath controlled at the same temperature as in the corresponding ozonation test and mixed
for 30 s. The ozone residual was determined at various times by injecting 10 mL of the
solution with a needle into a 25 mL test tube containing 10 mL o f indigo reagent, then
117
Water Bath
85 mL Sample
+
15 mL Ozone
Mixing
Stimng Bar
Test Tubes with Indigo Reagent

Needle
p2\ tZZZZSfi
m !si
to
to to
//////////7777/777/77/////////7/7/77777777/ / / / / / / / / / / / / / / / / / / / / / / / / / tPi
Time (min) o.5 1.5 3 5 10 15 20 25 30

of Injection
Figure 17. Schematic of Experimental Set-up Used for the Measurement o f Ozone Self-
Decomposition by the Syringe Indigo Method.
118
measuring the absorbance at 600 nm, and using EQs.48 and 50 to calculate the aqueous
ozone concentration ( V\ = V S =10). The syringe was recapped and returned to the water
bath for continued mixing until the next measurement. The time specified for injecting
samples were 0.5, 1.5, 3, 5, 10, 15, 20, 25, 30 min from the time o f mixing the sample
with the stock ozone solution. An important point was that the injection point (tip of
needle) should be inserted all the way to the bottom o f the test tube containing the indigo
solution as illustrated in Figure 17 to avoid the volatilization o f un-quenched ozone.
This method had the advantage that the aqueous ozone concentration could be
accurately determined by indigo colorimetry, the best accepted analytical method for
ozone. However, procedure operating errors might occur because o f the frequent
handling o f the syringe and switching between syringe cap and needle. A reference
experiment using acidified ozone-demand free water instead o f ground water sample in the
syringe and following the same processes was performed to estimate the aqueous ozone
losses due to operating procedures.
4.4.2.2. Stopped Flow Method
A mixture o f ground water sample and ozone stock solution was inserted into the
flow-through cell of the spectrophotometer after mixing for 30 s. The volume drawn was
at least triple the volume required for the 1 -cm flow-through cell to reach a steady state
reading (approximately 3 mL) to ensure the solution inside the cell was representative.
The continuous monitoring of the absorbance of aqueous ozone at 258 nm was started
119
immediately after stopping the flow. Batch reactor conditions inside the stopped flow
quartz cell were assumed. The residual ozone concentration was estimated with the
expression:
0 = 48,000 x ^ L EQ.52
K
where A^s was the sample absorbance at 258 nm, and M a was the molar absorptivity o f
gaseous or aqueous ozone. M a for gaseous ozone is widely accepted to be 3,000 M^cm ' 1
at STP. In contrast, M a values reported for aqueous ozone vary as discussed in the
preceding chapter. In this study, M a was assumed to be constant at 3,000 M '1cm l.
Background absorbance at 258 nm was obtained by measuring solutions with 85 mL o f
groundwater sample and 15 mL o f pure water, this value was deducted in the calculation
of aqueous ozone concentration. Estimation o f ozone decomposition kinetics was
accomplished by the direct analysis o f changes in absorbance instead of transforming those
values to ozone concentration with EQ. 52.
The Stopped Flow Method is widely used to monitor ozone decomposition
kinetics. Ozone losses due to sample manipulation can in principle be avoided. This
method also has the optimum time resolution. However, the Stopped Flow Method is
usually used for monitoring ozonated pure water with buffer, the validity of using it for
monitoring the natural groundwater o f this study which contains some organic matter
needs to be evaluated.
Though U V light absorbance at 258-260 nm could be used for the determination
of ozone concentrations in both gaseous and aqueous phases, many water constituents and
impurities as well as their ozonated by-products would also tend to absorb light at this
wavelength range, and therefore interfere with aqueous ozone measurements (Watt,
1984). The extent o f this interference for ground water sample was assessed in this study
by comparing the measurements o f initial fast ozone demand between the Indigo Syringe
Method and the results from directly measuring the aqueous ozone absorbance at 258 nm.
For the Syringe Indigo Method, sample (or reference) solutions were prepared by
withdrawing 30 mL of groundwater sample (or pure water) and ozone stock solution, then
following the procedures required to obtain initial fast ozone demand. For the direct
spectrophometric method, instead o f withdrawing indigo reagent into the syringe, the
aqueous ozone absorbance of the sample and reference solutions were measured at 258
nm immediately after 30 s of mixing. The absorbance differences between the two
solutions were used as A 238 in EQ.52 to estimate initial fast ozone demand. For both
methods, measurements were also performed in duplicate following reversed order in
sample and reference preparation. The Stopped Flow Method should not be used if
interferences by constituents present in the ground water are significant.
121
Chapter 5. THEORETICAL ANALYSIS OF A X IA L DISPERSION MODEL
A primary objective of this research was the development o f a mathematical model
(Chapter 3) for assessing disinfection efficiency in ozone bubble-diffiiser contactors. The
role of major water quality parameters and operating conditions in contactor performance
is theoretically evaluated in this chapter. The experimental data and theoretical analysis
correspond to a pilot-scale ozone bubble-diffiiser contactor (see Chapter 4) are presented
subsequently in Chapter 6.
Baseline operating conditions and major model parameters assumed for model
simulations were: contactor diameter of 6 inches, water column depth of 20 feet; water
flow rate of 14 gpm, gas flow rate of 8 scfh, applied ozone dosage of I mg/L, temperature
of 15 °C, dispersion number o f 0.1, ozone mass-transfer rate of 0.5 min*1, ozone self
decomposition rate of 0.3 min*1, lack o f initial fast ozone demand, and Giardia muris
inactivation kinetics proposed by Wickramanayake (1984). Deviations from these baseline
values will be specified individually for each model simulation case.
5.1. Role of Contactor Hydrodynamics
The axial dispersion model presented in Chapter 3 was used to assess the effect of
varying dispersion number on predicted ozone profiles and G iardia muris cyst
122
inactivation. Results from these efforts are presented in Figures 18 and 19 for counter-
and co-current operations, respectively. The back-mixing was characterized by the
dispersion number according to the axial dispersion model. Dispersion numbers used were
0.01, 0.1, and 0.3, which corresponded approximately to equivalent CSTR cascades with
N c str (refer to EQ. 10 in Chapter 2) o f 50, 5, and 2, respectively. Other operating
conditions assumed for these predictions were those previously specified for baseline
conditions.
As depicted in Figures 18 and 19, the ozone concentration profiles in the gas phase
were not affected much by changing the extent o f back-mixing. In contrast, the
distribution of ozone in the aqueous phase changed significantly with varying dispersion
number. The predicted results were consistent with the expectation that an increase in the
dispersion number (increase in the extent o f back-mixing) would tend to distribute
dissolved ozone more evenly throughout the water column. High back-mixing also
resulted in concentration jumps for dissolved ozone at the water inlet, as more clearly
shown in Figure 18 for the dispersion number of 0.3. This effect was more pronounced
for co-current operation as shown in Figure 19.
Another interesting observation was that lower back-mixing resulted in higher
concentration o f dissolved ozone in the effluent o f the counter-current contactor. This
trend resulted because the concentration o f dissolved ozone reached further up inside the
contactor at high back-mixing, and therefore it was exposed to a longer net residence time
and the corresponding greater level o f self-decomposition. This phenomenon was found
123
G. m uris Cyst Survival Percent, %

0.1 1 10 100
20 • » » 1»«tL 20
Top of Contactor Top o f Contactor
Q l = 14 gpm; Qa = 8 scfh; D m = 0 mg/L

k j = 0.3 min'1 ; kn= 12.6 L/mg-min
KLa = 0.5 min'1; CT-lag = 0.03 mg-min/L
15 d rio CT Iog-kill 15
(mg-ZL) (min)(mg-min/L) (log)
0.01 0.18 1.69 0.304 1.55
(ft)
(ft)
0.10 0.24 1.10 0.264 1.54

0.30 0.34 0.75 0.256 1.63
Water Column Height

Height
Water Column
0.0 0.2 0.4 0.6 0.8 0 5 10 15 20

Dissolved 0 3 Cone. (mg/L) Gaseous 0 3 Cone. (mg/L)
Figure 18. Theoretical Analysis of the Effect of Dispersion Number on Ozone

Bubble-Diffuser Contactor Operated in the Counter-Current Mode.
124
G. m uris Cyst Survival Percent, %

10-2 10-1 10° 101
i » i i uttl
' Top o f Contactor Top o f Contactor
0 L = 14 g p m ; Oc = 8 sc fh ; D m = 0 m g L
k j = 0.3 m in '1 ; kn = 12.6 L /m g -m in
KLa = 0.5 m in ’1; C T -la g = 0 .0 3 m g-m in/L
d ^ jjv g fio CT log-kill

(m g /L ) (m in ) (m g -m in /L ) (lo g )
0.01 0 .5 9 1.69 0 .9 9 7 5.55
(ft)
(ft)
0 .1 0 0 .6 0 1.10 0 .6 6 0 3 .6 4
0 .3 0 0 .5 9 075 0 .4 4 3 2 55
Water Column Height

Height
Water Column
0.0 0.2 0.4 0.6 0 5 10 15 20

Dissolved 0 3 Cone. (mg/L) Gaseous 0 3 Cone. (m g/L)
Figure 19. Theoretical Analysis o f the Effect of Dispersion Number on Ozone

Bubble-Diffuser Contactor Operated in the Co-Current Mode.
125
to be more pronounced at higher ozone decomposition rates, and not as pronounced for
co-current operation. According to the SWTR (EPA, 1991), higher ozone concentrations
at the effluent would not only result in greater disinfection credit, but it would also
provide higher levels o f dissolved ozone for disinfection in subsequent chambers of a
multi-chamber contactor.
Although average dissolved ozone concentrations increased significantly with
increasing back-mixing for the counter-current operation, CT values did not improve
because o f the corresponding decrease in tio residence time with increasing dispersion
number. The predicted ratio o f /io residence time to mean residence time for dispersion
numbers in the range o f 0.01 to 0.3 are presented in Figure 20. The CT values
corresponding to the counter-current profile presented in Figure 18 would be 0.304,
0.264, and 0.256 mg-min/L for the dispersion numbers of 0.01, 0.1, and 0.3, respectively.
For co-current operation, lower back-mixing did not affect the average dissolved ozone
concentration much but it increased significantly the t io residence time, and therefore, the
CT credit improved greatly. The CT values predicted for the profiles in Figure 19 were
0.997, 0.660, and 0.443 mg-min/L for the dispersion numbers of 0.01, 0.1, and 0.3,
respectively. The /io residence time values used for the calculation of CT values were
obtained from the theoretical tracer response generated from axial dispersion models with
respect to each dispersion number.
The predicted inactivation levels achieved for co-current operation were more
consistent with predicted CT values compared to those for counter-current operation.
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
126
0 .9
0.8 O result from ADR model prediction
0 .7 fitted with 0.895*e'2242*/ 5 + 0.105
0 .5
0 .4
0 .3
0.2
for CSTR
1
0.0
0.00 0 .0 5 0.10 0 .1 5 0 .2 5 0 .3 0
Dispersion N um ber (dim ensionless)
Figure 20. Predicted Ratio o f tio Residence Time to Mean Residence Time Corresponding
to Dispersion Numbers up to 0.3 Based on the Assumption o f Ideal Axial
Dispersion Model.
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
127
This trend resulted because the dissolved ozone concentration was distributed more evenly
in the co-current operation, and also the CT-lag effect was less significant because o f the
higher CT values achieved.
An important consideration is that the trends depicted in Figures 18 and 19
represent only theoretical predictions for the effect o f back-mixing. However, in reality,
the dispersion number would be highly dependent on operating conditions, such as gas and
water flow rate, and should be analyzed simultaneously. Practically, changes in back-
mixing and corresponding changes in t\o and CT could be accomplished by varying water
and gas flow rates, changing average bubble size, inputting mixing power, inserting
baffles, or modifying the distribution o f gas diffusing devices. A more extensive analysis
o f the effects o f operating conditions on disinfection performance will be discussed in a
subsequent chapter.
Figures 18 and 19 depict typical ozone concentration profiles observed for pilot-
scale contactors operated in counter-current or co-current mode. Gaseous ozone
concentration profile were always similar for each flow configuration. As gas bubbles
entered and started contacting with water in the lower portion o f the contactor, gaseous
ozone rapidly transferred to the liquid phase and thus the gas phase concentration
decreased significantly in the upward direction for both counter-current and co-current
flow operation. The rate of transfer was gradually slower as the bubbles rose through the
water column because of the lower mass-transfer driving force. The gaseous ozone
128
concentration approached pseudo-equilibrium with the dissolved ozone concentration at
the top o f the contactor.
The shapes of the dissolved ozone concentration profiles were quite different
between the two flow configurations. For counter-current operation, the concentration of
dissolved ozone increased downward within the column, with the highest dissolved ozone
concentration always being reached at the bottom or liquid outlet of the contactor.
In contrast, the concentration o f dissolved ozone inside a co-current contactor
increased with increasing distance from the inlet until reaching a maximum and then it
decreased gradually until reaching the outlet at the top of the contactor. This trend was
consistent with the mass-transfer mechanism dominating first and then ozone decomposing
more rapidly than transferring. The maximum concentration usually occurred within the
bottom one-third of the column. Generally, the concentration o f dissolved ozone was
more uniformly distributed in co-current contactors than in counter-current units. In
contrast, dissolved ozone concentration in the effluent from co-current contactors was
usually lower than that in the counter-current units for identical water quality and
operating conditions.
5.2. Role of Mass-Transfer
5.2.1. Effect of Mass Transfer on Ozone Concentration Profiles
Sets o f three ozone concentration profiles simulated with the uniform axial
dispersion model for various volumetric mass transfer coefficients K L*a are presented in
129
Figures 21 and 22 for counter-current and co-current operation, respectively. The KL*a
values used were 0.5, 1.0, and 1.5 min'1, which cover typical mass transfer efficiencies
observed in ozone bubble-diffuser contactors. Other operating conditions assumed for
these predictions were the baseline conditions discussed previously.
In contrast to the relatively weaker effect o f the dispersion number on ozone
concentration profiles (see Figures 18 and 19), varying K i*a within a relatively narrow
range had a stronger influence on the shape of the ozone profiles for both the aqueous and
gas phases. At higher mass transfer rates, gaseous ozone transferred to the aqueous phase
more rapidly and thus the concentration of gaseous ozone decreased faster in the rising
bubbles. Most o f the overall ozone mass was transferred in a lower portion of the
contactor at higher transfer efficiency. For devices with lower mass transfer rate, the
ozone concentration would distribute more evenly throughout the water column and
would correspond to flatter ozone concentration profiles in both aqueous and gaseous
phases.
For the counter-current profiles presented in Figure 21, higher average dissolved
ozone concentrations were observed at lower K ^ a values because the ozone was
distributed further up in the water column. This prediction conflicts with those made by
Zhou et at. (1994). Lower K ^ a values also corresponded to lower aqueous ozone
residuals in the contactor effluent because the dissolved ozone was inside the contactor for
longer periods o f time and thus more decay took place. A reverse phenomenon was
observed for co-current operation as depicted in Figure 22: i.e., lower average ozone
130
0 L = 14 g p m ; Qa = 8 sc th ; D lfl = 0 m g/L
-i
k . = 0 .3 m in ; k_ = 12.6 L /m g -m in
d = 0 .1 ; C T -la g = 0.03 m g-m in/L
3avg "io C T lo g -k ill

(ft)
0 .2 9 1.10 0 .3 1 9 1.98
Height
0 .2 4 1.10 0 .2 6 4 1.54
0 .1 8 1.10 0 .1 9 8 1.12
Water Column
0.0 0.2 0 .4 0.6 0.8 0 5 10 15 20

Dissolved 0 3 Cone. (mg/L) Gaseous 0 3 Cone. (mg/L)
Figure 21. Theoretical Analysis o f the Effect of Volumetric Mass Transfer Rate on
Ozone Concentration Profiles Inside an Ozone Bubble-Diffuser Contactor
Operated in the Counter-Current Mode.
131

20
Ql = 14 g p m ; Oc = 8 sc fh ; Dm = 0 m g /L
kd = 0 .3 m in '1 ; kn = 12.6 L m g - m in
d= 0 .1 ; C T -la g = 0 .0 3 m g -m in /L
15 -
^jjivg rio CT lo g -k ill
(ft)
0J6 1.10 0 .3 9 6 2 .4 7
0 .6 0 1.10 0 .6 6 0 3 .6 4
0 .6 3 1.10 0 .6 9 3 3 .7 8
Water Column Height
10 - - 10 =
C3
0.0 0.2 0.4 0.6 0 5 10 15 20

Dissolved 0 3 Cone. (m g/L) Gaseous 0 3 Cone. (m g/L)
Figure 22. Theoretical Analysis o f the Effect o f Volumetric Mass Transfer Rate on
Ozone Concentration Profiles Inside an Ozone Bubble-Diffuser Contactor
Operated in the Co-Current Mode.
132
concentration, and higher effluent residual at lower K ca. Additional simulations not
shown revealed that increasing the ozone decomposition rate would result in less
pronounced differences in average dissolved ozone concentration and greater differences
in ozone residual in the effluent stream.
The dispersion number used for the simulations presented in Figures 21 and 22
was the same for all cases even if the K ^ a values varied. However, changes in mass
transfer typically require changes in operating conditions such as the gas or water flow
rate, which would also result in different hydrodynamics (changing dispersion number)
inside the contactor. A more comprehensive analysis o f the effects of changing operating
conditions on disinfection performance including the corresponding influences on the
major parameters will be discussed in a subsequent section.
5.2.2. Effect o f Pressure Changes along the Water Column
Changes in pressure along the water column height have usually been neglected for
ozonation models reported in the literature. However, typical full-scale ozone contactors
have water column heights in the range 15-25 feet, and thus the assumption o f constant
pressure might be inadequate. The hydraulic head in the contactor increases linearly with
the water column depth, hence the pressure P (atm) at normalized depth z becomes (under
standard atmospheric pressure conditions):
n i t 1 atm rn
P = \ + p g - L z ---------------- EQ.53
101,300 Pa
133
where p is the water density (« 1,000 kg/m3), g is the gravitational constant (~ 9.8 m/s2),
and L is the water column height inside contactor in m.
According to the ideal gas law, the volume o f certain mass of gas is inversely
proportional to its pressure. With the assumptions that (1) gas bubbles do not breakup or
coalesce; (2) the overall change in gas bubble size resulting from mass transfer is negligible
(i.e., total moles o f gas molecules transferred into and out o f the gas bubbles are
approximately equal); and (3) all gas bubbles are spherical in shape, the gaseous phase
ozone concentration, gas hold-up, gas flow rate, and interfacial area at normalized depth r
can be represented with the expressions:
EQ.54
EQ.55
where subscript b denotes the values at the bottom o f the contactor under a pressure P b.
A mass-balance for the gaseous ozone concentration in the control volume Ax
during an infinitesimal time At could be expressed as:
Assuming steady-state operation, dividing both sides of EQ.56 by (A/-Ar\4), and
introducing dimensionless parameters:
where MLB = KL-aB-L/U i and SB = H-U q.b ^ l- The mass balance equation, EQ.37, for
aqueous ozone concentration could also be modified as:
\ 73
d20 _ 1 dO „ (P . f Yb P'
d ^ - + — . — - N LB- \ ^ - • N d - 0 - M r - 0 D =0 EQ.58
dz: 1-e d: I H PB) D
The gas hold-up, s, is neglected in this study, i.e., 1 - e = 1. YB profile obtained from
solving EQs.57 and 58 give gaseous ozone profiles at constant pressure PB. The actual
gaseous ozone concentration at normalized depth z can then be calculated with EQ.54.
Ozone concentration profile sets for both phases simulated with EQs.57 and 58 are
presented in Figure 23. Except for mass transfer rate, the baseline conditions described at
the beginning of this chapter were used for the simulations. K L values were estimated with
the empirical correlation proposed by Hughmark (EQs.22 and 24) assuming dB = 2 mm at
the bottom of the contactor, and were varied with z according to corresponding changes in
bubble size. Y,„ was estimated from EQ.40 with a modified Qg.b A simulation from the
model (EQs.37 and 38) without the consideration of the effect of hydraulic pressure was
also presented in Figure 23 for comparison. As depicted in Figure 23, a consideration of
hydraulic pressure has an effect similar to increasing mass transfer rate for both counter-
current and co-current operation modes. The effect o f hydraulic pressure is significant
and need to be considered in pilot-scale or full-scale ozone bubble-diffuser contactors.
135

20
without pressure effect

— Y, O q, a are functions o f pressure
-- 15
(ft)
j=
Height
cp
'5
X
10 10 g
Column
3
o
U
Water
-■ 5
0.0 0.2 0.4 0.6 0.80 5 10 15 20 25

Dissolved O, Cone (mg/L) Gaseous O, Cone. (mg/L)
Figure 23. Theoretical Analysis of the Effect of Varying Hydraulic Pressure on Ozone
Concentration Profiles Inside Ozone Bubble-Diffuser Contactors Operated in
both Counter-Current and Co-Current Modes.
136
5.3. Significance o f Ozone Reaction
The overall kinetics o f ozone decay in natural water was simplified to include two
stages as described previously in the model development chapter: a fast ozone demand
predominating in the first stage, and a subsequent slower first order ozone self
decomposition predominating in the second stage. The uniform axial dispersion model
was used to simulate ozone concentration distributions assuming first order rate constants
of 100 L/mg-min and 0.1 min'1 for the first and second stage decomposition rate,
respectively, and a finite initial demand in the range of 0 to 0.7 mg/L as ozone. Figure 24
presents the effect o f initial fast ozone demand on the dissolved ozone concentration at the
effluent o f a counter-current contactor operated at applied ozone dosages of 1, 2 and 4
mg/L. Other parameters used for these simulations were the same as the baseline
conditions described previously in this chapter.
The plots presented in Figure 24 confirmed that the decrease in ozone residual at
the contactor outlet was approximately equal to the amount of initial fast ozone demand
and independent of the applied ozone dosage assumed. Understanding the effects of the
role o f operating parameters on the ozone residual concentration at the contactor outlet is
important because ozone concentrations are only reported for water samples taken from
contactor inlet and outlet for most ozonation tests performed with full-scale and pilot-
scale ozone contactors. However, this direct relationship between residual ozone drop at
137
5
(m g /L )
initial fast 0 3 demand = 0

—© — initial fast 0 3 demand = 0.1 mg/L
at Contactor Outlet
4 —B — initial fast O, demand = 0.4 mg/L

—V — initial fast 0 3 demand = 0.7 mg/L
3
Ozone Concentration
1
Dissolved
0
0 1 2 3 4 5
Applied Ozone Dosage (m g/L)
Figure 24. Effect o f Initial Fast Ozone Demand on Dissolved Ozone Concentration in the
Effluent o f a Contactor Operated in the Counter-Current Mode.
138
contactor outlet and initial fast ozone demand would be less significant if the subsequent
ozone decomposition rate was higher, as depicted in Figure 25. Furthermore, it would be
harder to distinguish between initial fast ozone demand and ozone self-decomposition
simply by analyzing the ozone residual at the contactor outlet. Except for the ozone self
decomposition rates listed, the operating conditions and parameters used for Figure 24
were also used for the simulations presented in Figure 25.
The apparent effect o f ozone decomposition on ozone concentration distribution
inside a contactor is not only a function o f the corresponding decomposition rate constant,
but also o f the operating conditions that affect the extent o f back-mixing and mass
transfer. Figure 26 presents two sets each with three ozone concentration profiles
simulated for a contactor operated in the counter-current mode. The set represented with
solid lines corresponds to a shorter hydraulic retention time o f 2 min and water and gas
flow rates of 14.25 gpm and 4.24 scfh, respectively. Ozone mass transfer rate coefficient
and dispersion number were estimated with empirical equations proposed by Hughmark
(1967, refer to EQs.22 and 24) and Marinas et al. (1993, refer to EQ. 16), respectively,
assuming dg.o = 0.1 cm (refer to EQ.27). The corresponding dispersion number had the
relatively low value o f approximately 0.015. Dotted line profiles were simulated for
hydraulic retention time o f 8 min, water and gas flow rates o f 3.56 gpm and 4.28 scfh,
respectively, and corresponding dispersion number of 0.14. The profiles for each set were
obtained by assuming different levels o f initial fast ozone demand (kr = 100 L/mg-
139
Decrease in Dissolved 0 3 Concentration at Contactor Outlet (mg/L)

0.8
0 3 self-decomposition rate = 0
• 0 3 self-decomposition rate = 0.1 min
0.6 - —Q — 0 3 self-decomposition rate = 0.5 min
0.4 -
0.2 -
or
0.0
0.0 0.2 0.4 0.6 0.8
Initial Fast 0 3 Demand (mg/L)
Figure 25. Simulated Decreases in Dissolved Ozone Concentration at the Effluent of a

Counter-current Contactor as a Function o f Initial Fast Ozone Demand at
Various Second-Stage Ozone Self-Decomposition Rates.
140
Top of Contactor
Ql = 14.25 gpm
Q0 = 8.53 scfh
d = 0.015. K La = 0.39 min
Ql = 3.56 gpm - 15
Q0 = 4.28 scfh
d= 0.14, K,a=* 0.34 min'1
j= i
00 J 00
'55 '53
X
1 10 - - 10
c c
E E
J3 k j = 0.3 m in '. =0 J3
O O
U O
C3
£
kd = 0.3 min'
D „ = 0.4 mg/L
0.0 0.5 1.0 1.5 0 10 20 30

Dissolved O, Cone. (mg/L) Gaseous O, Cone. (mg/L)
Figure 26. Simulation o f Ozone Concentration Profiles in a Counter-Current Contactor

with Various Assumptions for Initial Fast Ozone Demand(at Constant Rate of
100 L/mg-min) and Subsequent Self-Decomposition Rate Constants.
141
min for all cases) and subsequent slower decomposition rate constant: no initial demand or
subsequent decay (D,„ = 0 mg/L; k j = Q min'1); no initial demand (D,„ = 0 mg/L) and
subsequent decay at k j = 0.3 min*1; and D,„ = 0.4 mg/L and k j = 0.3 min'1. Other
operating conditions and parameters used were the same as those specified for the baseline
conditions. As depicted in Figure 26, contactor operation at a higher dispersion number
resulted in a broader spread distribution o f dissolved ozone and greater overall ozone
demand from the second-stage self-decomposition (i.e., less overall ozone demand from
Dm).
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
142
Chapter 6. PELOT-SCALE EXPERIM ENTAL RESULTS AND DISCUSSION
This chapter presents the experimental results from gas-bubble size measurements,
tracer tests, gas-liquid mass transfer tests, ozonation tests, and cysts inactivation tests
performed with the pilot-scale glass-column ozone bubble-diffuser contactor described in
Chapter 4 (Figure 12). Only the results for the experimental set performed with the four-
diffusers (Figure 13 (b)) as gas sparging device are discussed in this chapter because big
bubbles leaking from the single-diffUser (Figure 13 (a)) impaired the data analysis for the
set. The results for the experimental set performed with the single-diffUsers are included
in Appendix I. Ozonation and mass transfer tests corresponding to each set of operating
conditions were performed consecutively in a single experiment day. Tracer tests were
performed separately due to limitation o f available time during a single experimental day.
o
Water samples taken from contactor influent line were stored at 4 C for later analyses.
Water quality parameters for the pre-treated ground water are presented in Table 11.
6.1. Bubble Size Distribution
Bubble sizes were measured photographically with image taken by a Ricoh
(Tokyo, Japan) video camera recorder Model R-86 Super, and playback with a JVC
(Tokyo, Japan) S-VHS video cassette recorder Model HR-S4700U. Enlarged pause
143
Table 11. Major Water Quality Parameters in the Pretreated Water Fed to the Ozone
Contactor.
Test Anions (mg/L) Metals (mg/L) Other

No. F cr S 0 4‘2 Na* K* Mg*2 Ca*2 Alkal. TOC pH
1-1106 0.36 59.8 90.3 8.67 3.62 22.4 87.2 155 0.485 6.95
2-1030 0.38 56.3 83.2 9.31 3.49 22.2 82.1 169 0.498 6.82
3-1027 0.41 60.0 91.5 9.95 3.25 22.3 81.3 143 0.703 6.32
4-1118 0.36 59.9 90.9 8.67 3.55 22.2 85.7 150 0.643 6.88
5-1104 0.42 58.6 90.3 9.31 3.80 22.4 85.0 153 0.501 6.82
6-1101 0.39 59.0 90.3 9.50 3.50 22.0 86.1 161 0.367 7.00
8-1115 0.39 60.4 91.7 9.95 3.74 22.3 84.9 150 0.426 7.00
9-1112 0.43 60.1 90.9 10.6 3.86 22.6 81.1 134 0.634 7.24
10-1012 0.41 59.2 90.8 9.31 3.74 20.9 89.0 163 0.561 6.40
11-1007 0.41 59.2 92.2 9.31 3.55 22.0 89.1 166 0.775 6.49
12-1128 0.42 59.6 92.7 11.2 3.80 20.1 87.8 156 0.545 6.60
13-1020 0.42 59.1 91.3 9.95 3.49 20.5 85.9 142 0.822 6.26
14-1015 0.41 59.2 90.9 9.95 3.49 20.3 85.5 145 0.549 6.57
15-1010 0.41 58.8 91.9 9.95 3.62 21.1 84.7 146 0.616 6.32
16-1025 0.42 55.3 89.8 7.39 3.55 22.4 89.9 180 0.421 6.31
17-1022 0.41 59.4 91.4 9.31 3.55 22.6 80.1 125 0.684 6.48
18-1017 0.41 58.3 91.5 9.95 3.62 20.4 82.1 138 0.654 6.43
19-0927 0.42 58.2 91.2 9.31 3.62 22.2 88.2 167 0.473 7.52
20-1201 0.41 58.4 92.0 10.6 3.43 22.9 95.8 229 0.349 7.90
21-1210 0.43 58.0 91.0 9.95 3.55 21.9 90.8 182 0.388 7.50
22-0930 0.40 58.9 92.0 9.95 3.62 21.6 87.1 159 0.528 6.51
23-0925 0.41 58.3 91.5 9.95 3.62 22.4 85.5 157 0.586 7.24
24-1204 0.41 57.9 91.9 10.6 3.74 22.6 89.4 182 0.413 7.51
25-1004 0.43 58.5 91.3 10.6 3.68 20.6 89.0 170 0.467 7.72
26-0814 0.41 59.1 91.9 9.95 3.68 21.6 75.9 126 0.972 6.55
27-0819 0.41 38.3 86.5 10.6 3.80 22.0 74.9 128 0.687 6.97
0106-co 0.38 21.0 77.3 10.0 3.51 18.1 110.6 277 0.325 7.47
0109-co 0.39 21.1 77.6 10.2 3.50 18.0 105.1 265 0.410 7.38
0115-co 0.39 21.7 77.5 9.95 3.52 17.9 105.4 277 0.395 7.23
0118-co 0.40 21.9 78.2 10.2 3.52 17.8 108.6 273 0.155 7.32
Cyst 0.41 22.0 78.8 10.0 3.50 19.3 83.0 205 0.389 7.14
all values in mg/L except pH
144
images were directly measured on T V screen. Diameters o f gas-bubbles were obtained by
comparing to the quarter-inch stainless steel sampling tube located horizontally across the
center of contactor. As discussed in Chapter 4, only the size on the vertical direction of
bubble images were measured because the size on the horizontal direction would be
distorted by the cylindrical shape of the glass contactor.
6.1.1. Method Verification
A calibration procedure was performed to assure the accuracy of the method and
to obtain the optimum equipment setup. It was found for the best resolution, the camera
should be set at least 1.5 meters away from the contactor with the highest zoom ratio
(15X) to minimize parallax, but should be closer than 2 meter or the images would
become too small. The shuttle speed was set to 1000 to reduce the image error resulted
from the motion of bubbles and still be able to capture enough light. Two side spotlights
and black background were also used to improve the clarity of the pictures.
The calibration was performed in a cylindrical glass container with inner diameter
of one foot. Pre-measured glass beads with two major sizes distribution (roughly 6.2, and
3.6 mm in diameter) instead o f gas bubbles were used to determine the accuracy and
precision o f this experimental design. A total o f fifteen beads with similar sizes were
dropped evenly at distances o f 1/4, 2/4 (center), 3/4 diameter of the diameter from the
front wall of the cylindrical container to investigate the maximum optical distortion error.
145
Sizes o f beads were also determined from the images on T V by comparing to a piece of
sampling tube located at the center of contactor.
The results from calibration procedure are listed in Table 12. Generally, bead sizes
estimated from the measurement o f images were larger than the real size, and the error
were mostly within ± 10 % . It was also observed that the discrepancies do not have a
strong dependence to the locations of glass beads. However, the error slightly increases
with decreasing glass bead size. The overall average increment, 6.7%, were used to adjust
the data obtained in the measurements of bubble size in the pilot-scale bubble-diffiiser
contactor. Vertical resolution o f the TV screen was found to be extremely critical for the
accuracy of measurements, the detection limit for the T V (vertical resolution approximate
200 lines in 20 inches diagonal) used in the measurement o f bubble size was estimated to
be approximately 0.5 mm.
6.1.2. Results of Bubble Size Measurement
A total o f approximately three hundred to four hundred gas bubbles were
measured from ten random paused screens for each operating condition. Only the bubbles
close to the reference sampling tube and centered on the T V screen were measured to
avoid possible errors in the fringes of the optical lenses of the camera and the TV tube.
The video camera was set at a height of the third sampling port (approximately 70 cm
from the bottom) and two meters away from the contactor. Gas bubbles were always
found well distributed across the glass column from this point up.
146
Table 12. Results for Calibration Procedure o f Bubble Size Measurements Performed
with Glass Beads.
Position to o f Dropping Bead with Respect to Camera Location

Front Center Rear
Bead Image Error Bead Image Error Bead Image Error
Diameter Size Diameter Size Diameter Size
(mm) (mm) % (mm) (mm) % (mm) (mm) %
6.30 6.64 5.40 6.15 6.83 11.1 6.10 6.07 -0.49
6.20 6.45 4.03 5.65 6.26 10.8 6.10 6.45 5.74
6.10 6.45 5.74 6.15 6.64 7.97 6.10 6.83 12.0
6.30 6.83 8.41 6.00 6.45 7.50 6.15 6.45 4.88
5.85 5.69 -2.74 6.20 6.64 7.10 6.25 6.45 3.20
Average: 4.17 8.88 5.06
3.80 4.17 9.74 3.85 4.17 8.31 3.55 3.41 -3.94
3.45 3.79 9.86 3.50 3.79 8.29 4.05 4.55 12.4
3.60 3.98 10.6 4.05 4.17 2.96 3.45 3.79 9.86
4.00 4.17 4.25 3.50 3.79 8.29 3.60 3.98 10.6
3.40 3.79 11.5 3.40 3.41 0.29 3.55 3.79 6.76
Average: 9.17 5.63 7.11
R eproduced with perm ission o f the copyright owner. Further reproduction prohibited w ith o u t perm ission.
147
The normalized size distribution o f gas bubbles measured under various gas flow
rates for the approximate water flow rates o f 3.6, 7.1, and 13 gpm, are presented in Figure
27, 28 and 29, respectively. Each distribution curve was averaged from the experiments
operated under similar gas and water flow rates. The complete experimental results are
listed in Appendix C. It was found from these figures that an increase in the gas flow rate
broadens the bubble size distribution with possible multimodal distribution (more than one
peak or major distribution) at high gas flow rates. Similar observations have been
reported by Sundareasn and Varma (1990).
The widely used Sauter mean diameter, also known as volume-to-surface mean
diameter, is used for the calculation of average bubble size in this research. The equation
for Sauter mean bubble diameter is:
2X ,
d B = ^ ------ EQ.59
2 X ,
1=1
where dB.,, is the estimated diameter o f i-th bubble. The Sauter mean diameter was used
in most o f the empirical equations relevant to bubble columns, such as gas-liquid mass
transfer coefficient and superficial area. The estimated Sauter mean bubble diameter for
all tests were plotted in Figure 30 against gas flow rates and are also presented in Table 13
with the corresponding operating conditions. The Sauter mean diameter reported were
148
35
—a — average Qc = 0.67 scfh

£3 average Qc = 1.33 scfh
30 -
—9 — average 0 Q = 2.71 scfh
—O — average Oc = 5 50 scfh
—v — average QG = 11.2 scfh
25 -
(%)
Percent Counted
15 -
10 -
0 1 2 3 4 5
Bubble Diameter (mm)
Figure 27. Bubble Size Distribution at Average Water Flow Rate of 3.6 gpm.
149
40
35 -
—A — average 0 G= 1.33 scfh
£3 average Qa = 2.69 scfh
30 - —9 — average Oq = 5.53 scfh
—O — average Qa - 11.2 scfh
—v — average Oa = 23.3 scfh
•o
o
c
3
O
U
u
0.
10 -
0 1 2 3 4 5 6 7
Figure 28. Bubble Size Distribution at Average Water Flow Rate o f 7.1 gpm.
150
35
—a — average Q a = 2.64 scfh

30 - £3 average 0 G = 5.52 scfh
—e — average 0 G = 11.2 scfh
—w — average 0 G = 23.2 scfh
25 -
T3 20 -
U
♦-*
C
3
O
U
15 -
t>
cu
10 -
0 1 2 3 4 5 6 7
Figure 29. Bubble Size Distribution at Average Water Flow Rate o f 13 gpm.
151
4
Sauter Mean Diameter
dB = O.O72*0g + 2.243, r2= 0.942
3
Average QL = 3.6 gpm
Averaged Bubble Diameter, dH (mm)
Average QL = 7A gpm
Average 0 L = 13 gpm
-a
2
4
Arithmetic Mean Diameter
dB= O.O63*0C+ 1.998, r2 = 0.951
0 5 10 15 20 25
Gas Flow Rate, Qc (scfh)
Figure 30. Empirical Correlation for Mean Bubble Diameter as a Function o f Standard
Gas Flow Rate.
152
Table 13. Operating Conditions and Experimental Results for Bubble Size
Measurement and Tracer Tests
Test Ql Qg da MRT *10 d r d El El r

No. Sauter tracer ADR1 BF2 BF2 BF2 corl3 corl-f1
(gpm) (scfh) (cm) (min) (min) (cm2/s) (cm2/s)
1-1106 13.1 11.2 0.30 2.28 1.48 0.061 0.270 0.0185 50.98 52.49 0.267
2-1030 13.0 5.53 0.27 2.26 1.47 0.074 0.375 - - 33.06 0.337
3-1027 13.0 2.64 0.25 2.29 1.63 0.038 0.295 - - 23.00 0.265
4-1118 13.0 23.2 0.36 2.21 1.34 0.065 0.245 0.0360 98.44 91.12 0.259
5-1104 13.1 11.2 0.28 2.25 1.46 0.054 0.250 0.0220 60.62 52.49 0.266
6-1101 13.0 5.53 0.26 2.30 1.48 0.078 0.375 - - 33.06 0.341
8-1115 13.0 23.2 0.37 2.18 1.32 0.072 0.260 0.0345 94.34 91.12 0.266
9-1112 13.0 11.2 0.29 2.26 1.45 0.055 0.240 0.0255 69.73 52.26 0.274
10-1012 7.10 5.53 0.26 4.14 2.29 0.116 0.440 0.0165 24.64 33.06 0.422
11-1007 7.00 2.68 0.26 4.27 2.70 0.081 0.385 - - 23.14 0.340
12-1128 7.10 1.33 0.23 4.19 2.85 0.063 0.345 - - 18.25 0.296
13-1020 7.10 11.2 0.36 4.10 2.35 0.077 0.285 0.0425 63.47 52.26 0.315
14-1015 7.10 5.53 0.26 4.08 2.34 0.094 0.425 0.0140 20.91 33.06 0.397
15-1010 7.05 2.71 0.23 4.20 2.70 0.074 0.380 - - 23.23 0.329
16-1025 7.15 23.3 0.40 4.01 2.17 0.103 0.300 0.0480 72.19 91.58 0.250
17-1022 7.10 11.3 0.32 4.16 2.40 0.074 0.275 0.0415 61.98 52.72 0.302
18-1017 7.00 5.53 0.27 4.17 2.42 0.100 0.420 0.0110 16.20 33.06 0.380
19-0927 3.50 2.73 0.25 8.16 4.11 0.136 0.410 0.0425 31.29 23.32 0.441
21-1210 3.70 0.67 0.22 8.93 5.19 0.093 0.415 0.0125 9.73 15.82 0.385
22-0930 3.60 5.50 0.26 7.89 3.72 0.148 0.415 0.0540 40.89 32.97 0.445
23-0925 3.50 2.64 0.31 7.67 4.05 0.114 0.420 0.0320 23.56 23.00 0.423
24-1204 3.56 1.33 0.22 9.30 4.93 0.118 0.415 0.0310 23.21 18.25 0.437
25-1004 3.50 11.2 0.31 7.89 3.74 0.149 0.420 0.0505 37.18 52.54 0.359
26-0814 3.50 5.49 0.25 7.89 3.75 0.150 0.425 0.0490 36.07 32.92 0.437
27-0819 3.50 2.67 0.24 8.22 4.37 0.117 0.425 0.0295 21.72 23.09 0.419
0106-co 3.45 11.17 - 8.74 3.94 0.175 0.475 0.0465 33.74 51.23 0.409
0109-co 3.60 1.39 - 8.32 4.78 0.130 0.435 0.0085 6.44 19.03 0.380
0115-co 13.1 2.65 - 2.41 1.75 0.047 0.240 0.0065 17.91 23.18 0.228
0118-co 13.1 11.17 - 2.35 1.47 0.070 0.305 0.0200 55.11 51.23 0.310
dispersion number calculated with ideal axial dispersion model
obtained by linear-square fit to normalized tracer results using CSTR-ADR model
calculated from empirical correlation
estimated by fitting CSTR-ADR model prediction to tracer results with dispersion
number estimated from empirical correlation
153
Figure 30 includes the also commonly used arithmetic mean diameter. The
corresponding mathematical equation is:
id,..
d , = —------ EQ 60
n
where n is the total number o f gas bubbles measured. Average bubble diameters
corresponding to EQ.60 were also plotted in Figure 30 against gas flow rates. It was
observed from the figure that the arithmetic mean diameter was smaller than Sauter mean
diameter.
As depicted in Figure 30, there was no significant trends between bubble size
distribution and water flow rate. In addition, an approximately linear relationship between
gas flow rate and average bubble diameter was found for the operating conditions used in
this research.
In general, gas bubble size can change throughout the water column as a result of
varying hydraulic pressure (gas density), gas absorption and desorption, and the
coalescence and breakup of gas bubbles. Although some of these effects could be taken
into consideration for the purpose o f modeling, the approach followed in this research
assuming that the ascending gas bubbles are independent of each other and the volume of
gas absorption and desorption are equal. Therefore, bubble diameter (also gas flow rate
and concentration) along water column are exclusively depended on hydraulic pressure
154
(EQ.53 - 55). The Sauter mean diameters reported in Figure 30 and Table 13 were under
standard condition (1 atm) estimated based on this assumption.
The mean bubble size has been empirically correlated to superficial gas velocity
and column diameter in bubble columns used as slurry reactors. However, the use o f these
correlations in ozone bubble-diffuser contactors is doubtful due to differences in contactor
characteristics, gas sparging device, and operating conditions. For example, the bubble
size predicted for the operating conditions used in this research with the widely used
correlation proposed by Akita and Yoshida (1974) is overly high at approximately 8 mm.
Although average bubble diameter was generally correlated to superficial or
approach gas velocity, it is believed that dB is more closely related to the surface area of
the diffuser than the column because the gas hold-up in an ozone bubble-diffuser contactor
is typically several orders smaller than that in a bubble column. Accordingly, the following
correlation (r2 = 0.942) is proposed for the Sauter mean bubble diameter (cm) in this
study:
d B = 0.224 + 0.0708 x U Gds EQ .61
where Ugm is the gas velocity per diffuser surface area (gas flow rate divided by total
diffuser surface area). This equation is valid for Uadi values in the range of 0.068 to 2.365
cm/s, and diffusers with pore size(<5) of approximately 60 pm. Use of the correlation for
other experimental apparatus and operating conditions should be verified after correction
for the difference in diffuser surface area.
155
Figure 31 includes the average bubble diameters in ozone bubble-difiiiser
contactors reported by Ahmad and Farooq (1985), Stankovic (1988), Roustan et al.
(1996), and the correlation from this research as a function o f U gm - The diffuser pore size
(<5), if available, for the diffusers used are also included in the figure. The surface area of
the diffuser used by Stankovic is not available, and is assumed to be half o f the column
cross section area. As shown in Figure 31, the dB for the coarse diffuser used by Ahmad
and Farooq, which with similar porosity to that used in this study, agreed closely with the
correlation developed in this study. In addition, the correlation developed for the diffusing
stone system of this study could be used to represent most o f the data sets by changing the
intercept of the line with approximately the same slope. Therefore, if the characteristic
intercept value of a porous diffuser dB,o is known, dB corresponding to a gas flow rate can
be approximately estimated according to Figure 31. However, if Ugm is higher or lower
than the experimental values in the corresponding research, dB could be strongly affected
by column diameter or diffuser porosity, respectively.
6.2. Contactor Hydrodynamics
Dispersion number and t l 0 residence times obtained from tracer tests, based on the
assumption that the hydrodynamics for the entire water column can be represented with an
axial dispersion reactor (ADR) model, are summarized in Table 13 together with the
corresponding operating conditions. Examples of normalized experimental tracer results
and the corresponding fitting with ADR model (solid lines) are presented in Figure 32. A
complete set o f experimental tracer tests and model fitting are presented in Appendix D.
156
0.40
(cm)
Sauter Mean Bubble Diameter, d B
□ Data from Ahmad and Farooq (1985)

0.20 -
- O fine diffuser ( S = 4-5.5 pm)
- □ medium diffuser (5 = 10-15 pm)
O A coarse diffuser (<5= 40-60 pm)
0.1 5 - # Data from Stankovic (1988)

O Data from Roustan et al. (1996)
- Correlation tor this research (S - 60 pm)
0.10 I 1 1 1 ' | 1 1 ' ■ 1 |-■1 1 1 1 ! 1 1 1 1 | 1 1 1 1 |

0.0 0.5 1.0 1.5 2.0 2.5
Gas V e lo city per D iffuser Surface Area, ^ (cm/s)
Figure 31. Comparison o f the Correlation Developed in this Study for the Sauter Mean
Bubble Diameter to Experimental Values Reported in the Literature for
Various Types o f Bubble Diffuser Stones.
157
2
Test 1-1106, Ql= 13.1 gpm, Qc= 11.22 scfli
O Normalized Tracer Data

Predicted with A D R model, d = 0.061
Fitted with d = 0.0190, CSTR = 26.7%
1
(dimensionless)
0 qnnntmmmmu.imu.u-
2
Test 2-1030, QL= 13.0 gpm, Qc~ 5.53 scfh
O Normalized T racer Data

Predicted with A D R model, d = 0.074

Fitted withd = 0.0121, CSTR = 33.7%
1
0 "1 --------1 I I
2
Test 3-1027, QL= 13.0 gpm, Qc — 2.64 scfh

Predicted with ADR model, d = 0.038
1 — Fitted with d = 0.0084, CSTR = 26.5%
0 BiG-CO t -COr-
0 1 2 3 4 5
Figure 32. Normalized Results from Selected Tracer Tests and Corresponding Model
Predictions.
158
As depicted in Figure 32, significant discrepancies were found between ADR
model fitting and normalized experimental tracer curve especially at lower dispersion
numbers. For tests with higher dispersion, the residence time distribution curve for tracer
tests is less sensitive to the variation in back-mixing and therefore the effects of non-ideal
condition (i.e., lack o f uniformity in dispersion along the water column) are less
significant.
In comparison with ADR model predictions, the normalized experimental tracer
curves generally had a higher peak, longer tail, and slower initial detection time. These
discrepancies could be explained by the occurrence o f local high mixing. The high mixing
could have been the result of the injection of water stream and corresponding water
surface unstability at the top o f contactor, as depicted in Figure 33. An additional cause
could be the occurrence of eddies caused by the un-evenly distributed ascending gas
bubbles coming from the diffusers at the bottom o f contactor as also depicted in
Figure 33.
The axial dispersion reactor model could be used to represent the different levels
of back-mixing taking place at the different locations o f the reactor by using different
dispersion numbers for each back-mixing zone. However, because of the difficulty in
estimating more than one dispersion number from tracer test curves, it was decided that
the back-mixing could be represented with axial dispersion reactor model for the central
zone o f the water column and CSTR model to account for the high mixing end effects.
159
gas out
water in water in
high mixing
zone
b -o
control
volume
% =x/L(l-r)
A zD=A x/L(l-r)
T
local
circulation
zone
gas m nr
water out
Figure 33. Hydrodynamics o f the Pilot-Scale Ozone Bubble-Difiiiser Contactor and the
Corresponding Schematic of CSTR-ADR-CSTR Model.
160
An additional reason for this option was that the use of a CSTR will result in a
conservative prediction in microorganism inactivation and provide a safety margin for
design purpose. The discrepancies between mass transfer and ozonation tests results and
model prediction due to the assumption o f CSTR end-effects will be discussed later in this
chapter.
A CSTR-ADR model with the assumption o f a CSTR zone (all high mixings
located at the water inlet instead of considering high mixing both at the inlet and outlet as
illustrated in Figure 31) followed by axial dispersion hydrodynamics was developed to
evaluate the extent of high mixing in the ozone contactor. The residence time distribution
(RTD) curve of a step-input tracer test can be predicted with the following mass-balance
equation, essentially a modified version o f EQ.6 with the necessary adjustments for initial
and boundary conditions-
dc dc d 2c
+ d D -r -^ r EQ.62
d0 D dzD d± D
top B.C. bottom B.C. initial condition
dlrn=0 = 1 - =0 4 = 0
dzD
where c is the tracer concentration (mg/L), Bo is a dimensionless time (ration o f time to
the hydraulic retention time of the axial dispersion zone), zo is dimensionless distance in
the downward direction (see Figure 33), and do is the dispersion number in the axial
161
dispersion zone. A finite differences approach was used for integrating EQ.62 and 63.
The result were converted to pulse-input RTD curve by differentiation and compared with
the normalized tracer curve as shown for the examples presented in Figure 32.
There are two model parameters for the CSTR-ADR model, the fraction o f CSTR
volume to total reactor volume (r = r r + rB) and the dispersion number of the axial
dispersion zone (dD). These two parameters were estimated simultaneously by least-
square fitting o f the normalized experimental tracer results with the CSTR-ADR model.
The fitted do values were converted to dispersion numbers based on entire water column
height (d = dDx (l- r ) ) in order to compare with the dispersion numbers obtained for the
ADR model. The resulting values are presented in Table 13 together with corresponding r
values. Dispersion numbers estimated at values lower than 0.005 were not included in the
table because those values were found to be too small for the RTD curve to be accurately
predicted by finite difference methods. It was observed from Table 13 that the dispersion
number estimated for the ADR model were much higher than those for the CSTR-ADR
model. In general, the dispersion number increased with increasing gas flow rates and
decreasing water flow rates as anticipated.
The liquid phase axial dispersion coefficient E l (= dD x Ul * L ( \-r) ) estimated for
the ADR section of the CSTR-ADR model is plotted against superficial gas velocity in
Figure 34. Consistent with results reported by other research groups (Shah et al., 1982),
E l appeared to be independent o f water flow rate and to increase linearly with increasing
approach gas velocity. The data scatter in Figure 34 might have resulted from non-ideal
162
100
90 HRT = 2 min
Liquid Phase Axial Dispersion Coefficient, EL (cm2/s)
HRT = 4 min
80 HRT = 8 min
70
60
50
r‘ = 0.84
40
30
20
10
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Superficial Gas Velocity, UQ (cm/s)
l------ 1--------i--------i-------- 1 i----------- 1---- 1-------- 1------1---------- 1-------- 1--------1

0 2 4 6 8 10 12 14 16 18 20 22 24
Gas Fow Rate, Qc (scfh)
Figure 34. Plot o f Liquid Phase Dispersion Coefficient Against Approach or Superficial
Gas Velocity.
163
mixing conditions not taken into consideration such as the occurrence o f dead-water zones
and wall effects.
The axial dispersion coefficient have been reported to be a function of bubble
column diameter, superficial gas velocity, and physical properties o f the liquid such as
density, viscosity, and surface tension. For typical ozonation treatment facilities, the
physical properties of water are fairly consistent, and the dependencies of E l on column
diameter have been shown to range from Dc13 to Dc (Shah et al., 1982). Assuming E l
D cl s, the following correlation is obtained for the ADR zone in the pilot-scale ozone
contactor o f this study:
E l = (0.243 + 1.283 x f /c ) x D 'c 5 EQ.64
where Dc, Ua, and EL are in cm, cm/s, and cm2/s, respectively. The correlation coefficient
obtained was approximately (r2 =) 0.84. The EL calculated from EQ.64 for each operating
condition o f pilot-scale experiment is also included in Table 13, these values will be used
later for model predictions of ozonation and disinfection experiments. It should be kept in
mind that the E l « D c 1' 5 dependence in EQ.64 is a rough assumption which can only
provide general information for contactor design. More extensive hydrodynamic studies
with contactors of various diameters would be required to determine the actual effect of
contactor sizes on full-scale ozone contactor. A list o f several hydrodynamic studies from
the literature and corresponding predictions with EQ.64 is presented in Appendix H for
reference.
164
Similar to the previous discussion for dB, the use o f empirical correlations to
represent the axial dispersion coefficient might not be appropriate for use in ozone
contactors with different size, range o f operating conditions and sparging devices. For
example, E l calculated from the empirical equations proposed by Kato and Nishiwaki
(1972) and Hikita and Kikukawa (1975) are approximately 150 cm/s2 or much higher than
those for the correlation developed in this study; Deckwer et al. (1974) and Baird and
Rice (1975) are between 50 and 130 cm/s2, somewhat more consistent; and Ohki and
Inoue (1970) ranging from 2 to 70 cm/s2 are in good agreement. These empirical values
o f El are summarized in Table 14.
The total volume ratio o f CSTR (r) was re-estimated by least-square fit o f the
normalized experimental tracer data to the CSTR-ADR model but using the dispersion
number calculated from empirically correlated E l (EQ.64). The results are also plotted in
Figure 32 for comparison with the ADR model fitting. As depicted in the figure,
predictions from CSTR-ADR model fitted all normalized tracer tests better than the ADR
model. For a few tests with very low mixing, predicted peaks were somewhat lower than
experimental results possibly due to a resolution limitation o f the finite difference method
used for solving the mathematical equations.
The re-estimated r values are listed in Table 13 and plotted in Figure 35 against
corresponding gas flow rates, and will be used for subsequent predictions of ozonation
and disinfection experimental results. Generally, r ranged from 0.25 to 0.45 and increased
with decreasing water flow rate. This dependency could be affected by the
165
Table 14. Dispersion Coefficient and Oxygen Mass Transfer Coefficient Estimated from
Various Empirical Correlations for the Operating Conditions Used in this
Research
Test E-l (cm2/s) Kl (cm/min)
No. Ohki Kato Deckwer Hikita Baird Hughm. Akita Massch. Xu Laplanche
1-1106 30.2 154.4 98.4 154.6 103.1 0.98 0.96 2.83 1.31 0.22
2-1030 13.5 149.6 79.6 149.1 81.4 1.02 0.90 3.05 0.65 0.15
3-1027 6.17 146.8 63.8 146.2 63.7 1.10 0.88 3.32 0.32 0.10
4-1118 70.6 163.2 122.3 163.7 131.3 0.91 1.05 2.49 2.67 0.33
5-1104 30.2 154.4 98.4 154.6 103.1 1.01 0.93 2.95 1.31 0.22
6-1101 13.5 149.6 79.6 149.1 81.4 1.05 0.88 3.16 0.65 0.15
8-1115 70.6 163.2 122.3 163.7 131.3 0.90 1.07 2.44 2.67 0.33
9-1112 30.0 154.4 98.2 154.6 102.9 1.01 0.95 2.91 1.30 0.22
10-1012 13.5 149.6 79.6 149.3 81.4 1.05 0.89 3.17 0.65 0.15
11-1007 6.25 146.8 64.0 146.0 64.0 1.05 0.89 3.17 0.32 0.10
12-1128 3.29 145.3 51.9 144.2 50.7 1.12 0.83 3.48 0.16 0.07
13-1020 30.0 154.4 98.2 154.6 102.9 0.92 1.05 2.53 1.30 0.22
14-1015 13.5 149.6 79.6 149.2 81.4 1.04 0.90 3.10 0.65 0.15
15-1010 6.31 146.8 64.2 146.0 64.2 1.09 0.85 3.38 0.32 0.10
16-1025 71.1 163.3 122.5 164.0 131.6 0.89 1.10 2.37 2.68 0.33
17-1022 30.4 154.5 98.6 154.6 103.3 0.96 0.99 2.72 1.31 0.22
18-1017 13.5 149.6 79.6 149.2 81.4 1.03 0.91 3.07 0.65 0.15
19-0927 6.37 146.9 64.4 146.4 64.4 1.10 0.89 3.30 0.33 0.10
20-1201 3.39 145.4 52.5 144.2 51.3 1.10* 0.84* 3.41* 0.17 0.07
21-1210 2.01 144.4 42.2 143.5 40.3 1.18 0.83 3.67 0.08 0.05
22-0930 13.4 149.6 79.5 149.6 81.3 1.08 0.90 3.24 0.65 0.15
23-0925 6.17 146.8 63.8 146.3 63.7 1.01 0.98 2.86 0.32 0.10
24-1204 3.29 145.3 51.9 144.4 50.7 1.15 0.83 3.59 0.16 0.07
25-1004 30.2 154.4 98.4 155.0 103.2 1.01 0.98 2.86 1.31 0.22
26-0814 13.4 149.6 79.4 149.6 81.3 1.10 0.89 3.32 0.65 0.15
27-0819 6.22 146.8 64.0 146.4 63.9 1.13 0.87 3.44 0.32 0.10
0106-co 30.0 154.4 98.3 155.0 103.0 1.02* 0.98* 2.90* 1.30 0.22
0109-co 3.39 145.4 52.6 144.5 51.4 1.13* 0.85* 3.48* 0.17 0.07
0115-co 6.18 146.8 63.8 146.1 63.7 1.09* 0.86* 3.34* 0.32 0.10
0118-co 30.0 154.4 98.3 154.7 103.0 0.99* 0.97* 2.85* 1.30 0.22
Cyst 13.5 149.6 79.6 149.6 81.4 1.07* 0.91* 3.18* 0.65 0.15
calculated with dB estimated from empirical correlation developed in this study
166
0.5
HRT min
HRT mm
HRT mm
Total Volume Ratio of CSTR
0.4
0.3
0.2
0 5 10 15 20 25
Gas Flow Rate, 0 G(scfh)
Figure 35. Plot of Total Volume Ratio o f CSTR Against Gas Flow Rate.
water distribution device. The inflow water was introduced at the top o f the column by a
stick-out tube with several rows o f lateral holes. At high water flow, the strong water jet
might either directly hit the contactor wall instead o f water surface and reduced the
energy, or the higher water flow velocity might break the patterns o f small circulation cells
and waves resulting from bubble flow or water jets, and therefore reduce the length or
extent o f back-mixing. For a fixed water flow rate, the length of CSTR increased with
increasing gas flow rate until reaching a maximum at O g ~ 4.3 scfh, and then decreased
with subsequent increases in gas flow. Apparently, the bubble flow density and
distribution at Q g ~ 4.3 scfh resulted in maximum high-mixing at the ends of the pilot-
scale contractor used for this study (refer to Figure 33).
Sauze et al. (1992) measured the residence time distribution curves on four equal
parts o f a pilot-scale bubble-diffuser contactor, and frequently observed a high mixing
zone for the top 50% of their contactors. Similar as the observation from this research,
longer high mixing zones were found for low water flow rates in some circumstances.
However, the effect of Ql and Og on r could not be clearly characterized due to
inadequate resolution of the tracer tests performed.
Though the CSTR-ADR model is able to simulate reasonably well the RTD curves
from experimental tracer tests, the actual extent o f high mixing taking place in the
contactor was likely less than ideal complete mixing. Between end high-mixing and
middle ADR, there should also exist a relatively long transition zone since the dispersion
was very low in the middle section. Therefore, the total volumetric ratio of high-mixing
zone might be longer than the r values estimated from CSTR-ADR model. The
discrepancies between experimental concentration profiles and model predictions due to
the assumptions o f CSTR will be illustrated later in mass transfer and ozonation test
results and discussion.
6.3. CSTR-ADR-CSTR Model
In order to simulate the high mixing observed at the top and bottom of the
contactor (Figure 33), the contactor was divided into two complete mixing (CSTR) zones
at the ends and an axial dispersion (ADR) zone. The resulting CSTR-ADR-CSTR (SDS)
model was based on the assumptions of steady state operation, axial dispersion in the
middle zone, two-film mass transfer, first order ozone decomposition kinetics, ideal plug
flow in the gas phase, and constant mass transfer rate throughout the column.
Assuming bubble size, gas flow rate, and gaseous ozone concentration are
exclusively affected by hydraulic pressure, mass balance expressions for variables O, Yg,
and D along a control volume in the ADR zone (see Figure 13) of infinitesimal thickness
can be modeled with expressions of the form EQ.37-39 after including the effect of
hydraulic pressure (EQ .53-55):
p
o ------------ N d - 0 - N r 0 D =0 EQ.65
7 / PBJ
=0 EQ.66
H PaJ
169
EQ.67
where zD is a dimensionless distance (relative to total height o f ADR zone) ranging from 0
to 1 in the downward direction. An expression for viable microorganism would be similar
to EQ.67 but substituting N r for N u . Open vessel boundary conditions were used at the
outlet from the ADR zone (rD = 1 for counter-current and rD = 0 for co-current flow).
Boundary conditions at the inlet (zD = 0 for counter-current and Zd - 1 for co-current
flow) were essentially the CSTR effluent concentration for both O and D.
Mass-balance equations for dissolved ozone and fast ozone demand in each of the
CSTR ends were:
Oi - 0 E - r to
xporbottom
De xporbottom ■n r o e - d e = o
r to EQ 69
where r is the volume fraction o f the corresponding CSTR at the top or bottom of the
contactor. Subscript / denote the influent to the CSTR zone. This value equals contactor
O and D influent or ADR effluent for top or bottom CSTR, respectively, in the case of
counter-current flow operation. Subscript e denote the effluent from the CSTR zone,
which equals the concentration inside the CSTR for both O and D. Influent and effluent
gaseous ozone concentrations can be obtained by the integration o f the following mass-
balance equation:
where zs is a dimensionless distance (relative to the height of CSTR zone) ranging from 0
to 1 in the downward direction. YB is gaseous ozone concentration (varies with zD) at
bottom pressure. The actual Y value o f the corresponding water column pressure can be
obtained from EQ.54. Open vessel boundary condition is used for gaseous ozone at any
location throughout the water column. The system o f nonlinear multi-variable ordinary
differential equations (EQ.65-70) was linearized and solved numerically by employing the
generalized Newton-Raphson method.
6.4. Ozone Mass-Transfer
Representative results obtained from oxygen mass transfer experiments are
presented in Figures 36 and 37 for a set o f tests performed with the pilot-scale contactor
operated in counter-current and co-current flow configuration, respectively. The
complete set o f mass transfer experimental results and corresponding model predictions
are presented in Appendix E. The results shown in Figure 36 for counter-current flow
operation revealed that the oxygen profiles had a lower and flatter curve at the bottom of
the contactor with a clear boundary (concentration drop) at approximately 5 feet from the
bottom. This boundary though not as marked for relatively low gas flow rates, it was also
noticed for co-current flow operation as depicted in Figure 37. The high mixing
conditions at the bottom was believed to have resulted from the occurrence o f large
171
top o f contactor
25-1004 estimated complete''

mixing top zone for
18 K l = 0.033 ft/m in CSTR-ADR-CSTR^
0 L - 3.5 gpm L model „
On = 11.23 scfh
14
-c
ep
S
X
c 10
E
J3
O
O Q
24-1204
K l = 0.038 ft/m in
0 L = 3.56 gpm
O n = 1.33 scfh
0 1 2 3 4 5 6 7
Dissolved Oxygen Concentration (m g/L)
assume 5 ft o f C STR at the bottom o f contactor

fitted with uniform axial dispersion throughout water column
Figure 36. Representative Results from Mass-Transfer Tests Obtained with the Pilot-scale
Contactor Operated in the Counter-Current Flow Mode.
172
i i . | .
top o f contactor
. | r . _
: 1 1 1
_
co-0109
U>
00
o
>
n
II
o
o
- 4 0 L = 3.6 gpm
- TO
T 0 G = 1.39 scfh_
r
- r
-
12 I
cp \°
S -
X
c 10 a a
E
_3 - o \ a\
O
U 8
u.
<D -
C3 ° \
1
£
6
-
o A
4 o A
- co-0106 O A
A
= 0.034 ft/m in q A
2
:Q l = 3.45 gpm o A
- Q g = 11.17 scfh O A
0 t * t » i t i i r i i i
i
( i f f
i
i i i i
i
i i l l i t t i
i i
i i i i
r
i i i i
0 1 2 3 4 5 6 7 8 9
Dissolved Oxygen Concentration (m g/L)
assume 5 ft o f CSTR at the bottom of contactor

fitted with uniform axial dispersion throughout water column
Figure 37. Representative Results from Mass-Transfer Tests Obtained with the Pilot-scale
Contactor Operated in the Co-Current Flow Mode.
173
eddies formed by gas bubbles not uniformly distributed in the radical direction as
illustrated in Figure 33.
Because o f the complex hydrodynamics observed for the pilot contactor, the
estimation o f average K ca values by fitting mass transfer test results with the simplified
models under consideration was found to be inadequate. Instead, the empirical correlation
for mass transfer coefficients proposed by Hughmark (1967), as described in EQ.22 and
24, was used in this research because the author had better control of the experimental
conditions: using rigid spheres instead o f bubbles and investigating bubble swarms as well
as single bubble. A constant mass transfer coefficient along the entire water column was
assumed for each operating condition, and the value was calculated with the Sauter mean
bubble diameter estimated for the middle of the water column height This assumption is
made because of the variation in K l resulted from the change o f hydraulic pressure or
bubble size are usually less than 5%, and the effect o f this variation on dissolved ozone or
oxygen concentration profile is negligible.
Both ADR (EQ.57 and 58) and CSTR-ADR-CSTR (EQ .65-70) models without
demand and decay terms were used to predict the oxygen concentration profiles as shown
in Figures 36 and 37. A constant CSTR height o f 5 ft and varied length were assumed at
the bottom and top of the contactor, respectively, for the CSTR-ADR-CSTR model. The
total CSTR fraction r was that obtained from tracer tests (see Table 13). The
experimental data revealed that the hydrodynamics and mass transfer conditions at the
inlet o f the contactor were more complex than those predicted by the CSTR model.
Consequently, it was decided that only the parameters for the middle ADR section and the
174
outlet CSTR would be obtained by least-square fitting o f the experimental data with the
open boundary condition closer to the inlet as a fitting parameter.
For counter-current flow, a significant amount o f dissolved oxygen in the influent
was stripped out by the nitrogen gas present in the head space, and the inflow water jets
could have also increased the contact between water and nitrogen air in the head space
and thus increased the mass transfer significantly. Because the high-mixing pattern at the
top is different from that at the bottom of the contactor (ass Figure 33), the boundary at
the top might not have been as distinct with a clear drop in concentration, but the actual
length o f high mixing zone and its transition could be much longer than the CSTR zone
assumed in the CSTR-ADR-CSTR model (as depicted in Figure 36). Due to these
reasons, only data obtained from sampling ports located below port #14 were used for
model fitting in the case o f the counter-current flow experiments. For co-current flow, the
mass transfer rate at the bottom CSTR was likely higher compared to that at the ADR
section especially at high gas flow rates as depicted in Figure 37. Also, the large eddies at
the bottom CSTR could have resulted in a different distribution o f dissolved oxygen
concentration in the radial direction, and the samples taken from near the wall o f the
contactor could have had very different oxygen concentration compared to those taken
from the axial center. These unknown factors could affect the concentration profile
significantly because o f the high oxygen concentration. Therefore, only the experimental
data obtained above sampling port #8 (5 ft height) were used for model fitting.
Mass transfer coefficients estimated with the empirical correlation by Hughmark
(1967) were fairly constant, ranging form 0.89 to 1.18 cm/min for the range of operating
conditions used in this research. Therefore, it appeared that the mass transfer rate was
primarily dependent on interfacial area. Generally, Hughmark's correlation was able to
represent the K l o f the middle ADR section well. The oxygen concentration profile
predictions form the CSTR-ADR-CSTR model were able to simulate the concentration
drop observed for the lower sections o f the contactor. Both models were able to fit
experimental data reasonably well with only a few under-predictions for counter-current
flow (as depicted for Test 24-1204 in Figure 36), which might have resulted form
neglecting the higher mass transfer occurred at the bottom section as mentioned earlier.
K l values estimated with Hughmark's correlation are listed in Table 14 together with
predictions with other correlations reported in the literature (EQ.20-26). It was depicted
from Table 14 that only the K l prediction form Hughmark (1967) and Akita and Yoshida
(1974) were comparable and appropriate for use in this research.
6.5. Ozone Decay Kinetics
The light absorbance o f ground water at the wavelength of 600 nm was found to
be negligible before and after ozonation, and thus the ground water did not present
background interfere for the Indigo Method. In contrast, significant interferences by
organic mattes present in the feed ground water were observed when measuring the
aqueous ozone concentration by direct absorbance at 258 nm. Therefore, the Syringe
Indigo Method was selected to investigate the ozone decomposition kinetics. A quality
control experiment was performed by measuring the decomposition kinetic o f a stock
176
ozone solution using the Syringe Indigo Method. The dissolved ozone demand during 30
min o f contact time was always within ± 5% o f that determined initially.
The bench-scale apparatus used to determine the ozone decomposition kinetics
required mixing o f the sample with a stock ozone solution to achieve target initial
dissolved ozone concentration in the range of 1.5 to 3.5 mg/L. In contrast, ozone was
transferred from gas bubbles into water without any dilution effect in the case of the pilot-
scale ozonation tests. Consequently the role of initial ozone dosage and dilution factors in
the overall kinetics o f dissolved ozone decomposition was evaluated to ensure the validity
o f the bench scale parameters to represent the pilot scale conditions.
The effect o f applied ozone dosage on ozone decay kinetics was assessed by
varying the ozone concentration in the stock ozone solution. The results obtained for
experiments performed at varying initial ozone concentration but maintaining a constant
dilution ratio are presented in Figure 38. The initial ozone concentration (ozone dosage)
and D 30 were obtained from the corresponding initial fast ozone demand tests. D,„ and kd
were obtained by linear-square fitting o f the results presented in Figure 38 with the batch
reactor model expressions (see EQ.45 and 46). The rate (kr) for the reaction between
dissolved ozone (O) and ozone demand chemicals (D) were too high to be accurately
estimated with the ozone reaction experiments, and were assumed to be the high value of
20 L/mg-min for all tests. This value was estimated from matching ozonation model
predictions to dissolved ozone concentration profiles obtained from ozonation tests.
177
4 .0
3 .0
2.0
(mg/L)
Ozone Concentration
0.8
0 .7
0.6
0 .5
0 .4
0 .3
0 3 dose volume ratio in D in
,
0.2 (mg/L) mixture (sample/03) (mg/L)
O 0.97 85/15 0.363
□ 1.99 85/15 0.385
A 3.27 85/15 0.396
0.1 | 1 •' J 1 | 1 1 1
1 11I| L1 1‘ JI 'r 1
' 1' 1' 1
I '| 1 ‘ ' 1 i ' 1 ' '—
0 5 10 15 20 25 30
Time (min)
Figure 38. Effect o f Applied Ozone Dosage on Ozone Decomposition Kinetics Observed
in Bench-Scale Batch Experiments.
178
Because o f the slow rate o f subsequent ozone self-decomposition and the fairly complete
reaction between fast-demand chemicals and ozone within the initial thirty seconds, D 30
values obtained from initial fast ozone demand tests were approximately equal to the D,„
estimated from model fitting o f the batch experimental data. Three different initial ozone
concentrations o f approximately 1, 2, and 3 mg/L were investigated for both initial fast
ozone demand and ozone self-decomposition tests. The volumetric ratio o f ground water
sample to stock ozone solution was fixed at 85/15 in all three cases. As depicted in Figure
38, the applied ozone dosage did not have a significant effect on the estimation o f kd or D m
in this study.
The role o f ground water sample to stock ozone solution mixing ratio on ozone
decay kinetics was studied with experiments for which results are plotted in Figure 39.
Two volumetric sample/ozone stock ratios, 85/15 and 50/50, were used in the initial fast
ozone demand and ozone self-decomposition tests. An additional ozone self
decomposition test was performed by directly absorbing ozone gas into a ground water
sample inside a syringe. Ozone was allowed to absorb for 15 seconds. The air was then
ejected from the syringe prior to analyzing by the Syringe Indigo Method. As depicted in
Figure 39, the sample/ozone volumetric ratio did not have a strong effect on the D,„
estimation. For ozone self-decomposition tests, however, the volume of stock ozone
solution should kept as low as possible to minimize the effect o f sample dilution on kd.
179
3 .0
2.0
(mg/L)
1.0
0 .9
0.8
0 .7
Ozone Concentration
0.6
0 .5
0 .4
0 .3 0 3 dose volume ratio in Dm

(mg/L) mixture (sample/03) (mg/L)
0.2 O 100/0 (0 3 gas)
□ 1.99 85/15 0.385
A 2.30 50/50 0.378
0.1 j 1 i. . T i. . . . | ■ ■ i- . t ■ ■ i - i .r «. .i .....-----
0 5 10 15 20 25 30
Time (min)
Figure 39. Effect of the Volumetric Ratio of Ground Water Sample to Ozone Stock
Solution on Ozone Decomposition Kinetics Determined by Performing Bench-
Scale Batch Experiments.
180
The sample/ozone volumetric ratio o f 85/15 used in this study was found adequate to
represent the condition o f ozone introducing for pilot-scale ozonation tests as shown in
Figure 39.
Based on the results presented in Figures 38 and 39, the ozone decomposition
kinetic tests were accomplished by measuring aqueous ozone decay in feed-water sample
contain 15% by volume of stock ozone solution. The initial ozone residual was controlled
at 2 to 3 mg/L, and the drop in pH was observed to be generally within 0.2 pH units.
Good linear correlations were found for all ozone decomposition tests with first-order
decay fits having an r2 greater than 0.95. The Z)30 and kd values estimated for all tests are
listed in Table 15. Z)30 values were widely scattered from 0.18 to 0.98 mg/L with no
strong dependence on any water quality parameters. kd values were relatively low with
most of the values being lower than 0.04 min'1, and therefore, a weak influence on ozone
profile was anticipated. The results from all bench-scale initial fast ozone demand tests
and ozone self-decomposition test, and the corresponding model predictions are listed in
Appendix F.
The initial fast ozone demand and subsequent slower ozone decomposition rate
constant was reported by Hull and Singer (1994) to increase with TOC. The D m and kd
obtained in this research were plotted against TOC in Figure 40. Despite the data scatter
shown in the figure, both kd and D,„ had a tendency to increase with TOC.
181
Table 15. Operating Conditions and Experimental Results for Ozone Decomposition and
Ozonation Tests
Test Ql O g O 3 dose /10 Dm kd Average O 3 Cone. (mg/L)

No. (gpm) (scfh) (mg/L) (min) (mg/L) (min1) observed SDS1 model A D R model
1-1106 13.1 11.2 0.87 1.48 0.45 0.0194 0.065 0.122 0.061
2-1030 13.0 5.53 0.98 1.47 0.37 0.0155 0.069 0.166 0.083
3-1027 13.0 2.64 0.94 1.63 0.88 0.0271 0.029 0.049 0.023
4-1118 13.0 23.2 1.72 1.34 0.38 0.0385 0.287 0.413 0.299
5-1104 13.1 11.2 1.83 1.46 0.44 0.0289 0.201 0.389 0.232
6-1101 13.0 5.53 1.86 1.48 0.37 0.0153 0.171 0.409 0.230
8-1115 13.0 23.2 3.58 1.32 0.38 0.0272 0.849 1.082 0.870
9-1112 13.0 11.2 3.48 1.45 0.39 0.0696 0.471 0.911 0.632
10-1012 7.10 5.53 0.85 2.29 0.40 0.0176 0.074 0.124 0.058
11-1007 7.00 2.68 0.87 2.70 0.67 0.0164 0.070 0.066 0.025
12-1128 7.10 1.33 0.76 2.85 0.89 0.0243 0.031 0.018 0.010
13-1020 7.10 11.2 1.71 2.35 0.47 0.0298 0.242 0.356 0.253
14-1015 7.10 5.53 1.86 2.34 0.31 0.0288 0.245 0.439 0.286
15-1010 7.05 2.71 1.84 2.70 0.61 0.0345 0.168 0.318 0.153
16-1025 7.15 23.3 3.29 2.17 0.41 0.0155 1.277 1.001 0.916
17-1022 7.10 11.3 3.79 2.40 0.47 0.0724 0.706 1.013 0.791
18-1017 7.00 5.53 3.43 2.42 0.41 0.0498 0.483 0.862 0.594
19-0927 3.50 2.73 0.73 4.11 0.28 0.0121 0.055 0.124 0.066
20-1201 3.50 1.38 1.03 4.46* 0.23 0.0106 0.202 0.219 0.126
21-1210 3.70 0.67 0.76 5.19 0.47 0.0143 0.081 0.080 0.028
22-0930 3.60 5.50 1.57 3.72 0.39 0.0238 0.264 0.343 0.228
23-0925 3.50 2.64 1.66 4.05 0.26 0.0341 0.214 0.396 0.295
24-1204 3.56 1.33 1.57 4.93 0.33 0.0110 0.220 0.335 0.191
25-1004 3.50 11.2 3.34 3.74 0.34 0.0124 1.201 1.093 1.011
26-0814 3.50 5.49 3.64 3.75 0.39 0.1150 0.745 0.841 0.665
27-0819 3.50 2.67 3.58 4.37 0.40 0.0679 0.636 0.842 0.601
0109-co 3.60 1.39 3.25 3.94 0.20 0.0175 1.692 1.443 1.396
0115-co 13.1 2.65 1.57 4.78 0.33 0.0146 0.884 0.898 0.832
0118-co 13.1 11.17 0.86 1.75 0.18 0.0107 0.469 0.515 0.474
0118-co 13.1 8.50 1.89 1.47 0.21 0.0190 1.399 1.062 0.993
Cyst-0.5 7.2 5.53 0.46 2.32* 0.42 0.0108 0.039 0.037 0.022
Cyst-1 7.2 5.53 0.79 2.32* 0.42 0.0108 0.127 0.112 0.058
1 CSTR-ADR-CSTR model
data simulated with empirical correlation from this research
182
0.9
0.8
5 0.7
g 0.6
w 0.5
0.4
0.3
0.2
0.1
0.12
0.10
!S 0.08
E
~ 0.06
0.02
0.00
0.0 0.2 0.4 0.6 0.8 1.0 1.2
TOC (mg/L)
Figure 40. Plot o f First-Order Ozone Decomposition Rate Constant and Initial Fast
Ozone Demand as a Function o f TOC.
183
6.6. Ozone Concentration Profiles
6.6.1. Significance of the Non-Ideal Conditions
The gas bubble size was assumed to be unique at a given depth within the ozone
contactor for the model development efforts presented previously. However, in addition
to the size variation resulted from hydraulic pressure, which has been evaluated in
previous chapter, a certain distribution of gas bubble sizes were observed as discussed at
the beginning of this chapter. Standard deviations of up to 1.6 mm were observed based
the results from bubble size measurement, with the average values increasing with gas
flow rates. The existence o f bubbles o f different sizes could have had an important effect
on hydrodynamics and volumetric mass transfer, and could have also resulted in apparent
dispersion in the gas phase ozone because gas bubbles of different size rising at different
velocities.
The CSTR-ADR-CSTR model assumed complete mixing at the top and bottom of
the contactor. It was deemed necessary to assess the significance of this assumption on
the predicted ozone concentration and viable microorganism profiles, and the level of
discrepancy between model predictions and experimental results.
6.6.1.1. Effect of Bubble Size on Ozone Concentration Profile
In order to assess the maximum effect of having a gas bubble size distribution on
the ozone concentration profiles, Test 16-1025 was selected because it had the broadest
distribution of gas bubble sizes. The gas bubble size distribution observed for this test are
184
plotted in Figure 41. The Sauter mean diameter was 3.97 mm and the corresponding
standard deviation was 1.66 mm. The bubbles were split into two groups o f small (51%
of total) and large (49% of total) bubbles as depicted by the dotted line in Figure 41. The
small bubbles group was responsible for 32% of the total bubble volume inside the
contactor, and had a Sauter mean diameter o f 3.12 mm with a standard deviation of 0.76
mm. The large bubbles group corresponding to 68% o f the total bubble volume, and had
a Sauter mean diameter of 4.52 mm with a standard deviation o f 0.95 mm. The system
was then simplified by assuming the occurrence of two groups o f bubbles each with a
uniform size equal to the corresponding Sauter mean diameter.
The occurrence of two different sizes of bubbles in the ozone bubble diffuser
contactor was modeled by substituting EQ. 57-58 with the following system o f equations
for the ADR model:
p \ 73 f
r B
• o - EQ.71
T ) \
EQ.72
- N D - 0 = 0
EQ.73
185
— Test 16-1025, 0 L = 7.15 gpm

0 G= 23.3 scfh
dg (sauter) = 3.97 mm
20 -
(%)
15 -
Percent Counted
10 -
dB ( s a u te r) dB (s a u te r)
= 3 .1 2 m m = 4 .5 2 m m
1 2 3 4 5 6 7
Figure 41. Splitting o f Bubble Size Distribution Curve for Test 16-1025 to Study the
Effect o f Two Different Sizes of Gas Bubbles.
186
The subscripts i and 2 denoted gas bubbles o f each size group. Total gas flow rates, Og
(23.3 scfh), was also split (7.46 and 15.84 scfh) according to the volume ratio. Water
flow rate, temperature, input gaseous ozone concentration, and contactor configuration
were assumed to be the same as those specified for Test 16-1025. To simplify the
equation and maximize the effect o f different sizes o f bubbles, axial dispersion
hydrodynamics was assumed for the entire contactor (i.e., r =0), and k j and D,„ were
minimized to 0.02 min'1 and 0 mg/L, respectively. Bubble rise velocity Vb, and mass
transfer coefficient K l, were calculated with empirical equations proposed by Marinas et
al. (1993) and Hughmark (1967), respectively for each bubble group size or overall gas
flow. The dispersion number d was estimated from empirical correlations from Marinas et
al. (1993) using the overall mean bubble size.
Aqueous ozone concentration profiles predicted for a single group of bubbles with
the overall Sauter mean diameter (3.97 mm) and for the co-existence o f two groups of
bubbles with different diameters (3.12 and 4.52 mm, 32:68 volume ratio) are plotted in
Figure 42 for both counter-current and co-current operations. Predictions with small or
large bubbles only were also included in the figure for additional reference. As depicted in
the figure, predictions based on uniform bubbles with overall mean bubble diameter were
approximately the same as those corresponding to the occurrence of bubbles with two
different sizes for both counter-current and co-current operation. Therefore, the use of
Sauter mean diameter to represent a group o f non-homogeneous gas bubbles in an ozone
contactor was deemed appropriate.
187
Top o f Contactor
1 5 --
co-existence of
(ft)
3.12 and 4.52 mm.

32:68 volume ratio
Height
predicted with average

dB= 3.97 mm (Sauter)
10 -■
Water Column
small bubbles only

with dg = 3.12 mm
counter-currrent
large bubbles only
with dB = 4.52 mm
co-current
5 -
0
0 1
Dissolved 0 3 Cone. (m g/L)
Figure 42. Theoretical Analysis o f the Effect o f Gas Bubbles with Two Different Sizes on
Ozone Profiles Using Bubble Size Measurement Results from Test 16-1025.
188
6.6.1.2. Effect o f Local High Mixing
The focus o f this section is to study the effect o f high mixing at the bottom of
contactor on the overall shape of the ozone concentration profile and the disinfection
efficiency achieved in the pilot-scale ozone contactor operated in the counter-current
mode. Because for counter-current flow, most o f the dissolved ozone is present at the
bottom of the contactor, a change in dispersion at the bottom could result in significant
differences in the overall disinfection efficiency achieved. For co-current operation, on the
other hand, the effect of differences in high mixing at the contactor ends on disinfection
efficiency might not be as dramatic (refer to the preceding chapter) because the dissolved
ozone is more evenly distributed along the water column and changes only slightly with
dispersion.
To study the effect of high mixing at the contactor (5 ft above the bottom) bottom
zone, EQ.57-58 of the ADR model were solved by finite differences assuming various
dispersion numbers and close vessel boundary conditions. The baseline conditions listed at
the beginning o f the preceding chapter were used as the operating conditions. The
inactivation kinetics proposed by Waller (1994) for Giardia m uris cysts were used to
represent the inactivation kinetics term. EL was assumed to be 29.5 cm2/s for the water
column above 5 ft, and 29.5, 295, and 2066 cm2/s (N c str = 12, 2, and 1.1, respectively)
for the high mixing zone at bottom 5 ft of the contactor. The predicted ozone
concentration and viable cyst profiles are plotted in Figure 43. Also included in the figure
is the average dissolved ozone concentration, CT parameter, and Giardia muris
189
G. muris Cyst Survival Percent, %

0.0001 0.01 0.1 1 10 100
20 20
14 g p m ; Qa = 8 sc fh ; Dm = 0 m g /L
* , = 0 3 m in '1 ; kn= 5 0 .5 L /m g -m in
KLa = 0 .5 m in '1; C T -la g = 0 .0 9 m g -m in /L
15 - ^LMattam Q jng ll0 CT log-kill - 15

(c m 2/s ) (m g /L ) ( m in )(m g -m in /L ) (lo g )
2 9 .5 0 .0 9 9 1.61 0 .1 5 9 4 .06
(ft)
295 0 .1 3 2 1.52 0.2 0 1 3.18

2066 0 .1 6 5 1.47 0343 2.64
Water Column Height
sp
'5
X
10 -H 10 |
o
CJ
o
C8
O , P ro file
5 -- - 5
V iab le C y s t
P ro file
0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 5 10 15 20

Dissolved 0 3 Cone. (mg/L) Gaseous Q Cone. (mg/L)
Figure 43. Model Predictions o f Ozone and Viable Cyst Profiles with Greater Extent of
Back Mixing Assumed at The Bottom o f Contactor Compared to the
Remaining o f the Water Column Inside the Contactor.
190
cyst inactivation efficiency predicted by the ADR cascade model. The /to residence time
was estimated from a simulated RTD curve predicted separately.
The theoretical predictions in Figure 43 indicated that though increasing bottom
high mixing result in higher average ozone concentration and CT values, inactivation
efficiency is significantly decreased. Therefore, neglecting the end effects could lead to an
over-prediction for the inactivation o f microorganisms. It should be recognized that the
effect o f contactor bottom higher mixing on actual dissolved ozone profile could be much
less pronounced or at least different from the model prediction for the following reasons:
gradual transition effect - the experimental ozone profile will not have such a distinct
boundary or concentration jump but rather a gradual transition; high mass transfer effect -
the actual ozone profile in the contactor bottom zone could also be affected by the
occurrence of faster mass transfer rate as discussed in the preceding section on Ozone
Mass Transfer section; and non-uniform concentration effect - because of the eddies
created at bottom by bubble flow (as shown in Figure 33), the small amount o f sample
taken from the axial center o f the column might not be representative of the average
concentration (ozone concentration measured at the wall could be different from that at
center). Therefore, as long as an end-effect is detected from tracer tests, it is
recommended to assume a high dispersion zone or a conservative CSTR to model the end
effect even though the ozone concentration did not clearly indicate a flat profile.
191
6.6.2. Ozone Concentration Profiles
A total o f 32 ozonation tests were performed for this study. Four tests were
performed with the contactor operated in the co-current flow mode, and the remaining
tests were performed in the counter-current flow configuration. Two of the counter-
current experiments were at relatively low ozone dosage to investigate the inactivation of
Giardia muris cysts. Examples of ozonation test results corresponding to counter-current
and co-current experiments run under comparable operating conditions are presented in
Figure 44. Experimental data at each sampling port generally had a standard deviations
less than 0.1 mg/L. Dissolved ozone concentrations observed in the lower sampling ports
occasionally depicted larger variances. Uneven distribution o f ozone gas bubbles in the
radial direction and the occurrence of large eddies formed at the bottom o f the contactor
(see Figure 33) might have played a role in the variability observed.
Ozone curves predicted with the ADR and CSTR-ADR-CSTR model developed in
this research are also presented in Figure 44. The experimental results for all other
ozonation tests and corresponding model predictions are presented in Appendix G. EL and
K i used in model predictions were obtained from empirical correlations EQ.64 and EQ.22
(Hughmark, 1967), respectively. For the CSTR-ADR-CSTR model, r values listed in
Table 13 were used as 5 ft at the bottom zone o f the contactor (estimated form the results
of mass transfer tests) and the rest at the top of the contactor. kd and D mvalues were
obtained from corresponding bench-scale ozone decomposition experiments (Table 15).
In general, the predicted ozone curves usually fitted experimental results well
192

20 20
Counter-Current Co-Current
(25-1004) (0106-co)
Q l = 3.5 gpm
15 -- -- 15
Q0 - 11 scfh
0 3 dose = 3.3 mg/L
(ft)
El = 52 cm2/s
Height
SO KLa = 0.26 min'(25-l004)
'5
X = 0.24 min‘'(0106-co)
5 10 -- kd = 0.013 min'1(25-1004) -• 10
Water Column
c
3 = 0.018 min '(0106-co)
o
CJ Dm= 0.34 mg/L(25-1004)
CQ = 0.20 mg/L(0106-co)
£
5 --
S
A J*
.1
'■u
*
i- i- j 111*11
1 2 3 0 1 2 3
Dissolved 0 3Cone. (mg/L) Dissolved 03 Cone. (mg/L)
predictions with CSTR-ADR-CSTR model

predictions with ADR model
Figure 44. Representative Ozonation Test Results obtained with the Pilot-Scale Ozone
Contactor Operated in Counter-Current and Co-Current Mode and
Corresponding CSTR-ADR-CSTR and ADR Model Predictions.
193
along the water column as depicted in Figure 44, indicating that the ADR mode! by itself
or combined with other models, such as the CSTR-ADR-CSTR model is adequate for
representing the data. The introducing o f the fast ozone demand concept and the method
o f measuring D,„ and kd values discussed previously were also proven able to effectively
represent the ozone reaction kinetics in ozone contactors.
For counter-current operation, experimental ozone concentrations generally
depicted a higher value than the A DR model prediction below 5 ft especially for gas flow
rates greater than 6 scfh as plotted in Figure 44. CSTR-ADR-CSTR model predictions
were more consistent with the higher dissolved ozone concentrations observed at the
bottom zone o f the contactor as a result of higher back-mixing. Because o f the gradual
back-mixing transition zone, high mass transfer, and non-uniform concentration effects
discussed in the preceding section, it is anticipated that the experimental ozone profiles
could be less affected by the high mixing occurred at the contactor bottom zone compared
to model predictions as observed for several tests in this study (Appendix G). Therefore,
though several ozonation tests were apparently better predicted with the ADR model, the
CSTR-ADR-CSTR model is recommended in order to have a better representation of
microorganism inactivation as discussed in the next section.
For co-current operation with high gas flow rates as depicted in Figure 44, the
higher mass transfer and occurrence o f large eddies at the bottom (resulting from the
presence o f a large quantity of bubbles coming out of the diffuser) might have been the
cause for the experimental ozone concentration profile deviation from model prediction.
On the other hand, ozone concentration was well predicted by both model at low gas flow
194
rate (0109-co and 0115-co). Interestingly the ADR model gave better representation of
the ozone concentration profiles at the bottom of the contactor. Apparently, the mixing
conditions were different for co-current operation with both gas and liquid flowing in the
same direction. Nevertheless, fluctuation of the ozone concentration profile was still
observed under the water column height of 5 ft.
6.6.3. Effect o f Model Parameters and Operating Conditions
D,„ had the most direct impact on the aqueous ozone concentration profile. At
large kr, dissolved ozone was observed only after most o f the fast ozone-demand was
satisfied (see Appendix G). The overall dissolved ozone concentration profile was shifted
down approximately a value o f D,„ at the relatively high kr and low kd values of these
experiments. The effect o f kd on ozone residual was more significant at higher dissolved
ozone concentrations and for longer hydraulic residence time. For counter-current
operation with fixed ozone dose, the ozone residual in the contactor effluent for tests with
higher average ozone concentrations was usually a lower fraction o f the applied ozone due
to ozone decay.
The effects of mass transfer and dispersion on ozone profile were highly dependent
on operating conditions. The ozone residual was distributed more evenly throughout the
water column for tests with higher dispersion. For counter-current flow, lower mass
transfer rates also resulted in a more even distribution o f dissolved ozone inside the
contactor. However, these two parameters are somewhat interrelated and thus it would
195
be more realistic to analyze the overall effect o f operating conditions on contactor
performance.
Average dissolved ozone concentrations estimated from experimental and model
predicted profiles o f all ozonation tests are listed in Table 15. CT (mg-min/L) values were
then calculated by multiplying these average ozone concentrations by the corresponding / w
residence times also listed in Table 15. The effects o f gas and water flow rates on the
average ozone concentration and contact time, are illustrated in Figures 45 and 46,
respectively. The average ozone concentration were divided by the applied ozone dosage
in Figure 45 for normalization purposes. It was clearly observed that for a constant ozone
dosage, lower water flow rates resulted in greater average dissolved ozone concentrations
and contact time. Higher gas flow rates resulted in greater average dissolved ozone
concentration, but lower tl0 thus limiting the improvement in CT. It is interesting to
notice from Figure 46 that unlike the dispersion coefficient, which is also a characteristic
value for back-mixing, t l0/F (theoretical value from -ln0.9 for CSTR to 1 for PFR)
increased with increasing water flow rate and decreasing gas flow rate. However, the
effect of gas flow rate on t w/ t was only significant when Q c < 5 scfh.
6.1. Inactivation of Giardia muris Cvsts
Two disinfection tests, tests Cyst-0.5 and Cyst-1, were performed at applied ozone
dosages of approximately 0.46 and 0.79, respectively. More than 100 (Cyst-0.5) and 500
(Cyst-1) cysts, including intact cysts (ICs), partially excysted trophozoites (PETs),
trophozoites (TRs), and empty cyst walls (ECWs) were counted to estimate
196
0.45
o
eo
g 0.40
o
Q
r»1
O
J 0.35
TU
Q.
< 0.30
C
0
1 0.25
c
<D
c
cj 0.20
n"l
1 0.15
E
■A
0.10
« 0.05
<
0.00
0 5 10 15 20 25
Gas Flow Rate, Qc (scfh)
Figure 45. Effects of Gas and Water Flow Rates on Average Ozone Concentration.
197
0.80
0.75
0.70
0.65
0.60
0.55 £]
0.50
0.45
0.40
0.35
0.30
0 5 10 15 20 25
Gas Flow Rate, Qa (scfh)
Figure 46. Effects of Gas and Water Flow Rates on /i0 Residence Times.
198
the target inactivation efficiency. Inactivation efficiency was estimated with the following
expressions:
ECW + PET T /'? + PET

%excystation = ------------------------- x 100 = x 100 EQ 74
ECW + PET + IC T R ^ + PET + IC
% inactivation - 100 - eXpOSed % eXCy5ted x 100 EQ.75

control % excysted
No control was run prior to the experiment, and thus a viability o f 100% was assumed for
the cyst stock solution. This assumption should have resulted in a relatively small error if
the viability was comparable to that of approximately 85% reported for a pilot-study
(AW WA, 1991). The inactivation levels observed were approximately 1.44-log for test
Cyst-0.5, and 2.02-log for test Cyst-l. the actual efficiency would be somewhat lower if
the viability o f the cyst stock was lower than 100%. The corresponding CT values were
0.09 and 0.29 mg-min/L for tests Cyst-0.5 and Cyst-1, respectively.
Ozone concentration profiles used for the prediction o f disinfection efficiency were
obtained by linear-square fit o f experimental data with both the ADR and CSTR-ADR-
CSTR models. Although D,„ was determined experimentally to be 0.42 mg/L, this
parameter was allowed to change to optimize the fitting of the experimental profiles with
the two models. Other parameters used for model predictions were: E l estimated with
EQ.64; r = 0.4 (estimated from Figure 35) with 5 ft o f high mixing zone at the bottom of
the contactor for the CSTR-ADR-CSTR model; K l estimated with EQ.22 (Hughmark,
199
1967); and kd from Table 15. Dissolved ozone concentration profiles and the
corresponding model predictions with both ADR and CSTR-ADR-CSTR models are
presented in Figures 47 and 48 for tests Cyst-0.5 and Cyst-1, respectively.
The inactivation kinetics were estimated from the experimental results proposed by
Waller (1994) for the same strain o f Giardia muris cysts. The CT-lag model presented in
Chapter 3 was used for this purpose. The resulting inactivation rate constants as a
function of temperature within the range o f 10 - 20°C were estimated with the following
Arrhenius-type expression:
10102
In* = 38.98----------------- EQ.76
T + 273.15
where k„ is in L/mg-min. The initial density of cysts required to define the shoulder o f the
inactivation curve was 4.5 times the actual density. k„ was estimated to be approximately
57 L/mg-min at 16°C.
Model predictions for viable cyst profiles are plotted in Figures 47 and 48 for tests
Cyst-0.5 and Cyst-1, respectively, together with the experimental results. The
corresponding inactivation levels predicted for tests Cyst-0.5 and Cyst-1 were 1.41-log
and 4.71-log with the ADR model, and 0.78-log and 1.81-log with the CSTR-ADR-CSTR
model. As depicted in these figures, the shoulder effect typically observed in the
inactivation o f G iardia cyst can be simulated with the models developed in this study.
The low applied ozone dosage and corresponding variability in dissolved ozone
data for test Cyst-0.5 might have been responsible for the significant discrepancy
200
Viable Cyst Percent, %
0.01 0.1 1 10 100

20 I I t t±LI It 20
Top of Contactor
Test Cyst-0.5
0 3 conc., experimental
viable cyst, experimental
15 -• -- 15
least-square fitted 0 3 profile
(Dm= 0.50 mg/L) and
(ft)
predicted viable cyst profile

with CSTR-ADR-CSTR mode
Height
SO
least-square fitted 0 3 profile u
W
10 -- (Dm= 0.33 mg/L) and -- 10 c
Water Column
1
"o
with ADR model U
b.
2
viable cyst
profiles
-- 5
dissolved
ozone
profiles
0 • 1 1 * 1 •«
0.0 0.1 0.2 0.3 0.4
Dissolved 0 3 Conc. (mg/L)
Figure 47. Experimental Results and Model Predictions Corresponding to Giardia Cyst
Inactivation Test Cyst-0.5.
201
0.0001 0.001 0.01 0.1 1 10 100

» 1* 1»'"I ml i i i nm l t i i m tl
20
Top of Contactor
Test Cyst-1
0 3 conc., experimental
vaible cyst, experimental
least-square fitted 0 3 profile - 15
(Dm= 0.39 mg/L) and

(ft)
with CSTR-ADR-CSTR mode

Height
least-square fitted 0 3 profile so

'5
(Dm= 0.10 mg/L) and X
- 10 c
Column
E
with ADR model 3
O
U
Water
/i\ «
dissolved £
ozone ..j viable cysty
profiles l profiles/
-- 5
i i i t j i t i i 0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Dissolved O- Conc. (mg/L)
Figure 48. Experimental Results and Model Predictions Corresponding to Giardia Cyst
Inactivation Test Cyst-1.
202
observed between inactivation results and CSTR-ADR-CSTR model predictions. At the
CT value o f 0.09 mg-min/L, the conditions were close to the shoulder region o f the
inactivation curve, and thus the discontinuous CT-lag model might not have been
appropriate.
For test Cyst-1, the fitted D mvalue was comparable to the experimental value
when the ozone concentration profile was predicted with the CSTR-ADR-CSTR model.
However, the predicted ozone concentration profile did not fit the experimental profile
well as shown in Figure 48. This could have resulted from a combination o f the gradual
transitional boundary, high mass-transfer, and non-uniform radial concentration effects
discussed in a previous section. In contrast, the inactivation efficiency o f G. muris was
predicted well with the CSTR-ADR-CSTR model.. On the other hand, the ADR model
resulted in good representation of the experimental ozone profile but gross over-
prediction of inactivation efficiency.
The SWTR do not allow for CT credit in the first chamber of multichamber ozone
contactors. However, a partial credit of 0.5-log inactivation o f Giardici cysts is allowed if
the effluent ozone residual from a first chamber is higher than 0.3 mg/L. The inactivation
efficiency achieved for tests Cyst-0.5 and Cyst-1 were 1.44-Iog and 2.02-log, while the
ozone residual in the effluent were approximately 0.15 and 0.64 mg/L, respectively.
Apparently, the CT and cyst inactivation credit allowed by the SWTR were conservative
compared to the experimental results obtained in this study.
203
Chapter 7. REACTOR DESIGN O PTIM IZA TIO N
In this chapter, the ozonation and disinfection models developed were used to
predict the performance of over-under ozone bubble-diffuser contactors typically used in
ozone disinfection plant. The over-under contactors were modeled as ADRs in series.
End effects were not taken into consideration because they would depend on contactor
configuration conditions such as separation between diffuser stones, and dimensions of the
opening between chambers. Therefore, the discussion presented in this chapter should be
reviewed as an effort to compare different design options in the ideal case o f no end
effects but the inactivation efficiencies achieved in real contactors might be somewhat
lower.
The dispersion in each chamber was predicted with the empirical correlation
developed in this research (EQ.64) after extrapolation by correcting for the square root of
the cross-sectional area as D c. Dispersion number and Ao residence times were estimated
with the ADR model (see EQ.6 and Figure 20, respectively). Mean bubble size were
obtained from EQ.61 assuming the total surface area of diffusers was equal to 1/20 of the
cross-sectional area of each chamber. Mass transfer coefficients and rates were estimated
with the empirical correlation proposed by Hughmark (EQ.22).
204
7.1. Model Verification
The ADR cascade model was evaluated with several ozonation tests performed by
The Metropolitan Water District o f Southern California (MW DSC) with a 5.5 mgd
demonstration plant located at the F.E. Weymouth Filtration Plant in La Verne, CA
(Coffey et al., 1995). The contactor consisted of six main chambers of equal size, with
floor dimensions o f 3.67 ft by 7.0 ft and a design water column depth of 20 ft. At the
design water flow of 5.5 mgd, the theoretical contact time was 1.0 min per chamber, or a
total of 6.0 min for the entire contactor.
The tests selected to compare with model predictions were operated in August and
September 1994 using California State Project water (SPW), one of MWDSC's source
waters. The initial fast ozone decay kinetics for SPW had been studied by Guittonneau et
al. (1992). An and kr were estimated to average 0.55 mg/L and 7 L/mg-min, respectively.
kd for SPW was estimated from a series of experiments with a pilot-scale glass contactor
(AWWA, 1991). The rate constant estimated at 25°C and pH 8.3 was kd = 1.2 min'1. The
inactivation kinetics o f G iardia lamblia cysts used in model were the same as that
presented in the SWTR (U.S. EPA, 1991), as summrized in Table 7.
The results ftom two experiments (3016E and 3016F) designed to investigate the
effect of gas flow rate on CT are plotted in Figure 49 together with model predictions for
dissolved ozone concentration and viable Giardia lam blia cyst profiles. The water flow
rate was 5.2 mgd, and the entire ozone dose of 1.5 mg/L applied in the first (counter-
current) chamber for both experiments. kd for these tests was assumed to be about 2/3 of
205
0.8 11 " "T

100
3016E
o 0.6
e
o 10 *
u 1 s.
u u
g 0.4
\ _o
<5 \ 0.1 - i
-a / >
J: 0.2
o /r 0.01
tn
5 0.001
0.0
A 2 3 4 5
2 C = 6.5 scfm Hydraulic Retention Time, min
0.8
100
1 3016F
E
O- 0 .6 10
c
u 1
§ 0.4
'•j
jj
S 0 .1 " o
•a >
j> 0.2
0.01
S T -
Q ffi.-
0.0 0.001
Qc = 36.3 scfm Hydraulic Retention Time, min
Figure 49. Ozonation Tests Performed with the MWDSC Demonstration Contactor
(Coffey et al., 1995) Operated at 1.5 mg/L Applied Ozone Dose with Different
Gas Flow Rates and Gas-Phase Ozone Concentration.
206
the value reported at pH = 8.3 to account fot the lower pH o f these experiments (pH =
7.3). The empirical correlation proposed by Sullivan and Roth (1980) was used to obtain
this pH corrected value. It should be pointed out that though the contactor cross-sections
in Figure 49 were plotted with contactor walls and and over-under baffles, the purpose of
doing so was to indicate chamber configuration and flow direction. The horizontal (x-
axis) direction in the figure was actually representative o f the cumulative hydraulic
retention time or vertical distance (downward for counter-current and upward for co-
current operation) o f each chamber.
Ozone residuals measured in the first chamber generally exhibited substantial
variability. The large 95-percent confidence intervals (error bar) shown in Figure 49 could
have resulted from unevenly distributed gas bubble flow and dissolved ozone, as well as
the occurrence o f short-circuiting and large eddies in the first chamber. The confidence
intervals for both tests usually overlapped with model predictions, indicating that the
model was statistically appropriate. The discrepancies between model predictions and
experimental results might have resulted at least partially from the assumption of uncertain
conditions such as diffiiser characteristics and water quality (D,„ and kd). For tests with
relatively high gas flow rate (3016F and for others refer to Coffey et al., 1995), the ozone
residuals measured after the first chamber were higher than those predicted by the ADR
cascade model. This could have been the result o f ozone bubbles entering the second
chamber or excess ozone gas in the head space transfering through the water surface in
downstream chambers.
207
Similar to the experimental results observed for the pilot-scale ozonation
experiments (A PPEN D IX F) in this research, a steeper ozone residual gradient and higher
ozone residual in the effluent were usually observed for ozonation chamber with lower gas
flow rates (first chamber in 3016E). However, in contrast to the observation from this
research (Figure 45), average ozone residual achieved in the first chamber was higher for
the test with the lower gas flow rate (predicted average ozone residual was 0.27 mg/L and
0.22 mg/L for tests 3016E and 3016F, respectively). Thus the overall contactor CT is
much greatly improve because the contact time (/io) was also greater at lower gas flow
rate. The discrepancy between pilot and demonstration scale contactors could have been
the result of incomplete ozone transfer that occurred for the test with high QG (predicted
ozone mass transfer efficiencies were 92% and 62% for tests 3016E and 3016F,
respectively), and the much higher kd value for the SPW (kd effects see Figure 26).
MWDSC experiments 3020B, 3020D, and 3020F plotted in Figure 50 were
originally designed to investigate the effect of contact time. The input gaseous ozone
concentration was kept oonstant at 6.1% by weight for all three tests, and the water flow
rate was varied between 12.5 and 4.3 mgd. The gas flow rate (applied at the first chamber
only) was varied accordingly to control the applied ozone dosage at approximately 2
mg/L. The water temperature was 23°C and the pH was 8.3 for all three experiments, and
thus kd = 1.2 min'1 was used for model predictions.
Except for the over-prediction for the first chamber o f test 3020B, ozone residuals
predicted with the ADR cascade model agreed well with the experimental data as
208
0.8 "'HI
00 3020B 100
s \
4. \
o 0.6
c \ 10 0s
^
o /I- \
a
si
C . _o
I
o 0.4 ■i
N \
o 0.1 a
~o / r >
> 0.2 - /
/ 0.01
/ _1
o.o £ 0.001
o A i 2 3 4
Qa = 12.5 scfin J Hydraulic Retention Time, min
j 0.8 T
so 100
E
3020D
4 0.6
10 *
o
U
o
c 0.4 9-/ ‘ 1 >.
u
O
o
N
•a J
Ifro r 0.1 2
w
0.2 i ! >
\ 0.01
?
0.0 - 0.001
4 6 8 10 12
Hydraulic Retention Time, min
Qa = 6.5 scfm
_■ 0.8
100
3020F
g 0.6 10 v=
o
U
o 1 stJ
c
o 0.4 T* I
N
o v
A r 0.1 3
-a
iTc# \ . >
j> 0.2 •\
If N 0.01
0.0 & 0.001

6 9 12 15 18
Qa = 4.3 scfm Hydraulic Retention Time, min
Figure 50. Ozonation Tests Performed with the MWDSC Demonstration Contactor
(Coffey et al., 1995) Operated at 2 mg/L Applied Ozone Dose with Different
Water and Gas Flow Rates.
209
depicted in Figure 50. The ozone mass transfer efficiencies predicted with the model were
similar (88-90% ) for all three tests. The average ozone residuals in the first chamber
increased with water flow rate because less time was available for the dissolved ozone to
decay. Note that for the pilot-scale experiments performed in this research, opposite
results (Figure 45) were obtained because the kd value was much lower, and the effect on
dissolved ozone from self-decomposition was less significant.
Though the highest C value was predicted for test 3020B, the predicted CT in the
first chamber was lower compared to the two other tests (refer to Table 16) because of the
shorter contact time. In contrast to the general assumption that higher CT would be
achieved at lower water flow rates (refer to the preceding chapter and Scott et al., 1992),
the CT predicted for the entire contactor was the smallest for test 3020F. This
observation is consistent with the relatively high kd value, and corresponding greater
ozone decomposition at higher residence time.
As depicted in Figures 49 and 50, the first ozonation chamber with higher ozone
residual in the chamber outlet usually resulted in better overall inactivation efficiency for
the entire contactor because the ozone residual in the subsequent reactive chambers was
higher. Even though sometimes lower CT value were resulted in the first ozonation
chamber (i.e., tests 3020B, 3020D, and 3020F as listed in Table 16).
The guidelines proposed by the U.S. EPA (1991) for demonstrating compliance
with the SWTR suggested an alternative simplified method for assessing microorganism
inactivation based on CT. Because a comprehensive characterization o f ozone profiles
210
Table 16. CT Values and Giardia Inactivation Efficiencies for MWDSC's Ozonation
Tests Estimated Based on Guidelines from U.S. EPA (1991) and Predicted
with ADR Cascade Model.
Credited* Giardia Inactivation (log) Predicted** Giardia Inactivation (log)

and /CT (mg-min/L) and /CT (mg-min/L)
Test Chamber Overall Chamber Overall
1 2 3 1 2 3
3016E 0.5/- 0.79/0.13 0.67/0.11 2.24/- 1.18/0.15 1.84/0.24 0.84/0.09 4.52/0.55
3016F 0.5/- 0.98/0.16 0.79/0.13 2.86/- 0.90/0.08 0.90/0.10 0.31/0.03 2.27/0.23
3020B 0.5/- 0.38/0.06 0.45/0.07 1.52/- 1.33/0.18 1.81/0.26 0.63/0.08 4.08/0.54
3020D 0.5/- 0.48/0.08 0.15/0.02 1.20/- 1.63/0.23 1.48/0.20 0.26/0.02 3.41/0.46
3020F 0.5/- 0.24/0.04 0.070.01 0.85/- 1.76/0.25 1.20/0.15 0.11/0.01 3.08/0.41
T equal /io as calculated from dispersion number assuming ideal axial dispersion model for
both Credited and Predicted CT values.
* C is calculated based on experimental inlet and outlet ozone residuals.

G iardia inactivation is calculated based on CT and the disinfection rate constant
proposed by the U.S. EPA (1991).
**: Predictions with ADR cascade model
might not be feasible for many ozone contactors currently in use, the guideline also
proposed an approach for calculating C based on a minimal level of information: contactor
inlet and outlet ozone concentrations (Table 4). The CT values for MWDSC's tests
credited based on the experimental inlet and outlet ozone residuals of each chamber are
listed in Table 16, together with the CT predicted from the ADR cascade model. The rl0
211
values used for both methods were calculated from dispersion numbers (refer to Figure
20) with the assumption o f ideai axial dispersion in each chamber. Table 16 also includes
the inactivation o f Giardia lamblia cysts estimated by both methods. The credited
inactivation efficiency o f Giardia cysts were calculated directly from corresponding CT
values and disinfection rate constants proposed by the U.S. EPA (Table 7). Inactivation
efficiencies predicted with the ADR cascade model are also presented in Table 16.
It was observed from Table 16 that the inactivation efficiency credited for first
ozonation chamber might be arbitrary and too conservative, therefore the result could be
much less than the actual inactivation achieved. Except for test 3016F (which significant
discrepancies were found between measurement and prediction), Table 16 depicts that for
reactive chambers, C credited by the U.S. EPA were usually lower than the average
concentration from experiments or model predictions especially at high ozone residual and
kd. CT values based on ADR cascade model predictions were usually a good indication of
Giardia cyst inactivation (inactivation increased/decreased with increasing/decreasing
CT). However, the log survival ratios were not proportional to CT values but generally
larger than directly estimated from CT (multiplied by £,/ln l0 ) because the T values were
conservatively estimated with /io residence times.
Because o f the lack of understanding about the performance of ozone contactors,
the U.S. EPA developed conservative CT values for regulatory purposes. Consequently,
inactivation efficiencies credited with CT values specified in the SWTR tend to be much
less than the actual efficiencies achieved. As depicted in Table 16, overall inactivation
achieved in a typical full-scale ozone bubble diffuser contactor could be under-credited by
212
2-3 log scales. This under-credit situation could be even worse because the k„ values used
by the U.S. EPA were based on scarce data and included a safety factor of two.
Furthermore, ozone concentrations measured in a full-scale contactor could have
enormous variability even in reactive chambers (Coffey et al., 1995) and under well-
controlled measurement procedures, therefore, increasing the inaccuracy in the estimation
o f contactor CT. By contrast, the disinfection model developed in this research should be
able to reasonably represent the inactivation throughout the entire contactor with the
proper input of ozone decay and microorganism inactivation kinetics, which can be
accurately obtained with experiments performed separately in smaller contactors.
1.2. Theoretical Analysis and Optimization
In the first part o f this section, several popular designs of ozone bubble-diffuser
contactors currently used in water treatment plants will be theoretically analyzed and
compared with the ADR cascade model developed in this research. In the second part, the
ADR cascade model will be applied to optimize the performance of various configurations
o f bubble-diffuser ozone contactors.
The following baseline operating conditions and major model parameters will be
used for ADR cascade model simulations: hydraulic retention time of 12 min, input
gaseous ozone concentration o f 4 % by weight, applied ozone dosage of 1 mg/L,
temperature of 20°C, ozone self-decomposition rate of 1.0 min'1, and initial fast ozone
demand of 0.4 mg/L (kr = 20 L/mg-min). Deviations from these values will be specified
when necessary for specific cases. The assumptions and equations for mean gas bubble
213
size, ozone mass transfer, and dispersion were the same as those presented at the
beginning o f this chapter. The inactivation o f Giardia muris cysts was estimated with the
CT-lag model based on the inactivation data reported by Wickramanayake et al. (1984a,
1985):
1^^*7f\
5-15°C In *, = 49.89-------- - -------- EQ.77
T + 273.15
3897
15-25°C In * = 16.31------ - --------- EQ.78
T + 273.15
where k„ is in L/mg-min. The increase in initial cyst density required for the CT-lag model
was estimated to be 4 times the actual influent density.
7.2.1. Predictions for Current Popular Designs
Three full-scale ozone contactors were theoretically investigated for their
performances under the conditions described previously: Central Arizona Project Water
Treatment Plant operated by the City o f Tucson, Arizona (Tucson), the Los Angeles
Aqueduct Filtration Plant operated by the Los Angeles Department of Water and Power
(Los Angeles), and the demonstration ozone disinfection plant operated by the
Metropolitan Water District o f Southern California (Metropolitan). The Metropolitan
contactor was described previously in this chapter.
The Tucson plant has four ozone contactors in parallel, each unit is 25 ft wide and
has a hydraulic capacity o f 50 mgd. Each contactor consists of six down-flow main
chambers o f equaled length o f 12 ft, and four small up-flow chambers connecting the first
214
five down-flow chambers, each with a length o f 1.5 ft (see Figure 51). At a designed
water flow rate o f 30.2 mgd, the contactor has a hydraulic detention time of 16.74 min at
the water column height of 25 ft.
The Los Angeles plant also has four parallel ozone contactors, each 34 ft wide and
having a hydraulic capacity of 150 mgd. The contactor contains four chambers with
counter-current design at the first and third chamber. The length for these chambers are
30, 28.25, 30, and 8 ft from first to fourth (see Figure 51). At the design water flow rate
o f 145 mgd, the contactor has a hydraulic detention time of 5 min at the water column
height of 20 ft.
Assuming that the entire gas flow is applied into the first chamber, the aqueous
ozone concentration and viable Giardia cyst percentage profiles for all three contactors
under consideration are plotted in Figure 51. In the figure, the shoulder effect for the
inactivation o f Giardia cyst was clearly seen at the beginning o f contact (in the first
chamber). Because o f the shoulder effect, the dependence between CT and inactivation
efficiency is usually weaker for the first chamber than other chambers (e.g., In the first
chamber, CT = 0.0535, 0.0415, and 0.0815 mg-min/L, and inactivation = 0.40, 0.40, and
0.42 log for Tucson, Los Angeles, and Metropolitan, respectively).
As depicted in Figure 51, the Metropolitan contactor has the best disinfection
performance among the three designs, with Tucson slightly behind. However, an
additional theoretical prediction assuming a rearrangement of the configuration of the
215
100

Tucson: height; width; overall length
25 ft 25 ft 78 ft
4 6 8 10
2 =70 6 Hydraulic Retention Tim e (m in)
1• ' • ‘ ' ' 1 i .........

- 100

O i ' L
§ °-2 10
U
u
e 1 |
c
N | Los Angeles: height; width; overall length!
o 0.1 | 20 ft 34 ft 96.25 ft !
•o
o
i 1
0.1
4 6 8 10 12
Qr = 94.8 scfm Hydraulic Retention Time (m in)
100

Metropolitan
height; width; overall length j 10
20 ft 7 ft 22.8 ft
01
4 6 8 12
Qc = 4.62 scfm Hydraulic Retention Time (m in)
Figure 51. Ozone and G iardia Cyst Profiles Predicted for The Ozone Contactors
Currently Used by Tucson, Los Angeles, and Metropolitan.
216
Tucson's contactor with the baffle system for the Metropolitan contactor (i.e., six
chambers with equal length o f 13 ft), the result indicating lower inactivation efficiency
(1.49 log overall) compared to the original Tucson design. This order of disinfection
performance (Metropolitan > Tucson > Tucson with Metropolitan's baffle position » Los
Angeles) was always held even if the operating conditions and water quality parameters
were changed. Because ozone decay kinetics and mass transfer efficiency (~ 95%) were
similar for all these designs, the effect o f contactor configuration on disinfection
performance was attributed to dispersion.
Lowering the overall dispersion not only has the advantage o f increasing the
ozone-cyst contact time (see t \o trend shown in Figure 20), but also the ozone residual in
the first counter-current chamber effluent is usually increased (Figure 18) as well.
According to EQ.64, the dispersion coefficient decreased with chamber cross-sectional
area. Therefore, the disinfection performance can be improved by increasing the number
of baffles or reducing the size o f contactor as supported by the profiles in Figure 51. It
should be pointed out that the dependence of dispersion on contactor diameter is an
assumption, though it is generally reported in the literature and verified with several full-
scale experimental results (refer to the preceding chapter), actual dependencies
corresponding to the length/width o f chamber in full-scale contactors need to be further
investigated.
In addition to baffling and sizing, there are many other parameters that could affect
the hydrodynamics o f a contactor, and thus be critical for disinfection. Some of these
parameters are the type o f gas bubble diflusers and their distribution, and the design of
217
interchamber connections. A thorough consideration of these parameters in designing or
building an ozone contactor could improve disinfection performance.
7.2.2. Alternatives o f Current Popular Designs
Compared with counter-current operation, a co-current ozonation experiment
performed in a single-chamber pilot-scale bubble-diffuser contactor generally results in a
much higher average ozone residual (or CT), but less overall mass transfer efficiency and
lower ozone residual in the effluent. In full scale, because higher dispersion is generally
presented, the advantage o f high CT of the co-current chamber could be less significant or
even out-performed by counter-current chamber conditions. Furthermore, lower residual
in the effluent o f co-current chambers would have significant disadvantage from the point
of inactivation efficiency achieved in subsequent reactive chambers. Therefore, the design
of full-scale ozone contactor is usually in favor of counter-current chamber.
The Tucson ozone contactor was used to compare the differences between
counter- and co-current chamber as plotted in Figure 52. The placement o f over/under
baffles was reversed to account for the change in flow direction. It can be observed from
the comparison o f the first plot in Figures 51 and 52 that though the co-current chamber
(first chamber) had higher inactivation (0.47 vs. 0.40 log) than the counter-current
chamber, overall disinfection efficiency achieved for entire contactor was lower because
there were less ozone residual left for subsequent reactive chambers. Figure 52 also
includes another set of model predictions with the assumption o f lower kd. As
218
0.3
100
E 10
O 0.2 Ss
c i* s?
o 1
o Co-Current, kd = 1.0 min'* u
o
-a 0.1 •5
a> 0.1
> >
0.01
0.0 .3 ... 0.001

' 0 4 6 8 10 12
= 70.6 scfinjj Hydraulic Retention Time (min)
0.5 1e+2
y—S 1e+1
04 1e+0
E
Counter-Current,^ = 0.1 min’ 1e-1 >5
d - 3 ... 1e-2 r.
c 0.3
o 1e-3 t.
u
1e-4 O
w 1e-5 ^
0.2
1e-6 =
O 1e-7
t/i 0.1
co 1e-8
Q 1e-9
0.0 1e-10
4 6 8 10 12
Qa = 70.6 scfm Hydraulic Retention Time (min)
0.5 1e+2
1e+1
1> 0.4 1e+0
1 1G-1 .a
d Co-Current, kd = 0.1 : lin i 1e-2 t -
§ 0.3 1e-3 £,
U /
1e-4 u
c
u 0.2
■o 1e-5 Z
1*6-3
o 1e-7
in 0.1 i 1e-8
5 1e-9
0.0 1e-10
4 6 8 10 12
Qa = 70.6 scfm|j Hydraulic Retention Time (min)
Figure 52. Theoretical Comparison for the Disinfection Performance of an Ozone

Contactor Between Counter- and Co-Current Operation Mode.
219
anticipated, disinfection efficiencies for both operation modes were improved significantly
and the counter-current design was still better.
In addition to operating conditions and contactor configuration, distributing the
ozonated gas into different chambers could also affect the contactor performance. To
input the entire massive amount o f gas flow into one chamber could result in low mass
transfer efficiency as depicted in Figure 49 (3016F). In such case, distributing a portion of
the gas into subsequent chambers could increase transfer efficiency and improve contactor
performance significantly. For example, under identical operating condition as
Metropolitan's test 3016F except distributing 1/3 o f the ozonated gas into the second
chamber could increase overall mass transfer efficiency from the original 62% to 69%
(74% in 1st and 59% in 2nd chamber), and improve G iardia cyst inactivation by 1 log
scale as shown in the second plot o f Figure 53. The ozone residual predicted in the
second plot also fits the experimental results better, which supports previous assumption
that the elevated ozone residual in the second chamber might have resulted from entrained
gas bubbles. The predicted inactivation efficiency in the first plot of Figure 53 was
different than that in Figure 49 because different inactivation kinetics were used.
In the third plot o f Figure 53, the same overall flow o f ozonated gas was
distributed evenly into the first three chambers o f the Metropolitan contactor. The
predicted overall mass transfer increased to 77% (86% , 60%, and 85% for the 1st, 2nd
and 3rd chamber, respectively), and the overall disinfection efficiency was also improved
significantly.
220
0.8 100
Metropolitan
3016F
E
Ci 0.6
e
0
U
1 0 .4 0.1
<s
13 0.01 >
.> 0.2
o 0.001
8
a
o.o 0.0001
Qa = 36.3 scfm| Hydraulic Retention Tim e, min

0.8 T T 100
0.6
0 .4 0.1
0.01
0.2
0.001
0.0 0.0001
Qa = 24.3 scfmj 12 scfm Hydraulic Retention Tim e, min

0.8 i 100
•sL
E
J 0.6
c
o
U
§ 0 .4 0.1
<5
•o 0.01
> 0.2
o 0.001
Q
0.0 0.0001
Qa = 12.3 scfm 12 scfm 12 scfm _

Hydraulic Retention Tim e, min
Figure 53. Theoretical Evaluation of the Effect o f Gas Distribution for Ozonation Test
3016F from Metropolitan.
221
For contactors with fairly high ozone mass transfer efficiencies, changing the
distribution o f gas flow could still improve the contactor performances as depicted in
Figure 54. The contactor shown was a modification o f the Tucson contactor with eight
equal-size chambers separated with evenly distributred baffles. The predicted performance
was slightly better than the original Tucson design. The baseline operating conditions and
parameters were used for the predictions. As depicted in the figure, the partition o f gas
into other chambers also help increasing the inactivation efficiency. The distribution of
gas into chambers no. 1,3, and 5 (all counter-current) was slightly better than into
chambers no. 1,2, and 3 with respect to both mass-transfer efficiency (99% versus 92%
for the second ozonation chamber, others were similar) and inactivation efficiency. These
results were consistent with previous observations that counter-current chamber were
usually better than co-current in full-scale ozone contactors.
For water with relatively low ozone demand, spreading the gas input might not be
able to improve the contactor performance. The predictions in Figure 54 were repeated
for a relatively low ozone decay (k j = 0 .1 min'1) as shown in Figure 55. it was observed
that the first prediction with all of the gas input at the first chamber has the best
disinfection efficiency. Because of the slow decay o f ozone, the high residual in first
chamber effluent persists longer and has greater influence on the subsequent reactive
chambers. By contrast, distributing ozonated gas in the later chamber would result in
more ozone residual wasted in the contactor effluent.
222
0.4 T
/-N 100
£ 0 .3
o
c
o 10 t f
ui, 0.2 u
jj
-o
<u / r 1 <
.o
I 0.1 - /
8 /
,ti..
/
Y
0.0 ■■■a 1 0.1

0 2 4 6 8 10 12
Qa = 70.6 scfm |J
Hydraulic Retention Time (min)
0 .4 m,.,.
- 100
2
£ 0 .3
10
0.2 u
/ .. \r
O
13
/
/
-
■s
75 0.1 / \\ 1 >
: /
/
: / i j " —— 0.1
4 6 8 10 12
“ *T 2l-28scfm|J
= 28.24scfm[J ~ 7 1 U 21_28scfm
2
0 .4
100
0.2
0.1
0.0 0.1
Qa = 28.24scfml
Figure 54. Theoretical Evaluation o f the Effect of Gas Distribution for Modified Tucson
Contactor with fast ozone decomposition (k j = 1.0 m in1).
223
0.5 \
. . . . . . . i 1e+2
: /
i / 1e+1
: / 1e+0
|s,/ 04 ; /
: / 1e-1 ^
o : 1 - t L _ . 1e-2
§ 0.3 r 1 1e-3 £
OP*\ : /
1
- 1e-4 O
--------
o 0.2
: / 1e-5 3
-o
CJ
r /
1e-6 .2
i /
>
"o
Ul 0.1
:
-/
/ 1e-7 >
Eft V 1e-8
5 - 1e-9
1 *. . . i
0.0
.
1e-10
4 6 8 10 12
Qa - 70.6 scfin
Hydraulic Retention Time (m in)
0.5 1e+2
1e+1
0.4 / 1e+0
1e-1 .
1e-2 '
o 0.3 1e-3
U 1e-4 1
0.2 1e-5 •
T3
_O
1e-6 •
>
O 1e-7 '
0.1 1e-8
Q 1e-9
0.0 1e-10
Qc= 21.28scfm
Hydraulic Retention Time (m in)
O.b . . . , . . . . . . . . . i . . 1 1 1 1 1 1 1 1! 1 1 1 " 1 1 T ’* . . . .,
1e+2
1e+1
0.4
/ - 1e+0
/ __
L " / 1e-1 -o
/ 1e-2 t -
0.3 / 1e-3 S
C
/
CJ / 1e-4 &
0.2 ■ / 1e-5 2
-o
l
l
/ 1e«;S
o / '
U
O / 1e-7
tA
5/1
0.1 / 1e-8
s /
/ /
.t.
-i 1e-9
[e-10
8 10 12
0 if
. = 28.24scfmU * 2l.28scfm LI 6
21_28scfin ,[
Figure 55. Theoretical Evaluation of the Effect o f Gas Distribution for Modified Tucson
Contactor with slow ozone decomposition (kd = 0.1 min'1).
224
Chapter 8. SUMMARY AND CONCLUSIONS
The axial dispersion model, and a model based on the combination of CSTR and
axial dispersion reactor connected in series appeared to be a good approach for
representing the distribution o f dissolved ozone concentration and the disinfection
efficiency achieved in bubble-diffuser ozone contactors. The subsequent sections
summarize the conclusions drawn from modeling efforts and experiments performed in
this study, application o f the model to the optimization o f full-scale ozone contactors, and
evaluation o f the CT concept.
8.1. Model and Method Development
The axial dispersion model including the rate o f ozone reaction with impurities
generally present in natural water, and the consideration o f the effect o f hydraulic pressure
on mass transfer represented dissolved ozone concentration profiles well. A CT-lag model
was developed to represent the inactivation of G iardia cysts including a shoulder-effect.
Experimental methods were developed to assess aqueous ozone concentrations,
and evaluate mass-transfer in a pilot-scale contactor, and to obtain ozone decay kinetics in
a bench-scale batch reactor.
225
8.2. Theoretical Analysis with Axial Dispersion Model
The effects o f major model parameters such as dispersion number, mass transfer
rate constant and ozone decay rate constant on the distribution o f dissolved ozone
concentration inside an ozone bubble-diffuser contactor were theoretically studied with the
axial dispersion model. For counter-current operation, increasing dispersion number or
decreasing mass transfer rate resulted in decreasing ozone residual in the contactor
effluent, but an increasing average ozone concentration. For co-current operation, low
dispersion and high mass-transfer rate were more favorable for the set o f operating
conditions and water quality assumed in this study.
8.3. Pilot-scale Experiments and Results
A pilot-scale ozone contactor was used to verify the ozonation and disinfection
models developed. Though the axial dispersion model alone was able to reasonably
predict the results for ozonation tests, significant end-effects that might have resulted from
introducing gas and water was observed consistently for entire pilot-scale experiments
(tracer, mass-transfer, and ozonation tests). Ignoring these end-effects could have
resulted in an over-prediction of inactivation efficiency. A CSTR-ADR-CSTR combined
model was developed to simulate the end-effects. The model was able to predict
reasonably well the inactivation o f G iardia muris cysts.
Empirical equations for mean gas bubble size and dispersion coefficient were
developed in this study and compared with experimental data reported in the literature.
For mass-transfer coefficients, the empirical equation proposed by Hughmark (1967) is
226
recommended for use in ozone contactors. Ozone decay kinetics had a significant effect
on dissolved ozone profiles. Because o f site and seasonal variability of water quality, it is
recommended that the decay rate constant be measured for each specific water quality.
The effects of operating conditions were also investigated with the pilot-scale
contactor. For the ground water used in this study, the average ozone concentration was
found to increase with decreasing water flow rate and increasing gas flow rate. t l0/ t was
found to decrease with increasing gas flow rate and decreasing water flow rate.
8.4. Full-Scale Application and Optimization
Pilot-scale contactors are usually tested prior to designing full-scale ozone
contactors. However, it was found that the optimum operating conditions obtained from a
single-chamber pilot contactor were not always best for a multi-chamber operation.
Because o f the complex hydrodynamics usually present in a pilot-scale contactor, which
might also significantly differ from the full-scale contactor, the parameters that could be
obtained from a pilot-scale contactor are limited and might not be valid for full-scale.
Furthermore, the operation of a pilot-scale contactor is costly and the results could have
large errors if the experiments were not performed carefully. By contrast, the use of
appropriate models is cost-efficient, capable of reasonably simulating dissolved ozone and
predicting viable microorganism curves in a full-scale contactor. Models can also be
applied to develop optimal operating conditions. The parameters needed in the model can
be accurately obtained from well-controlled tracer tests or individual bench-scale
experiments.
227
In addition to providing better disinfection efficiency, well-designed contactors
could allow more complete mass-transfer and greater level o f reaction with ozone residual.
Generally, the contactor can be improved by changing the operating conditions or
contactor configuration. Several optimization principles and the corresponding
procedures concluded from this research are:
Reducing ozone demand: by practicing pre-ozonation or enhanced coagulation prior to
ozone contactor.
Reducing dispersion: by reducing size of contactor, adding baffles, smoothing water flow,
increasing diffuser area and distributing diffusers more evenly, or altering operating
conditions, such as increasing water flow rate or reducing gas flow rates (i.e.,
increasing ozone gas concentration).
Increasing mass transfer and chamber effluent ozone residual: by increasing diffuser
surface area or using diffusers with smaller pore size, or increasing water flow rates.
For contactor with low mass-transfer efficiency, decreasing gas flow rate, increasing
input-gas ozone concentration, or partition o f input-gas to several chambers could
improve the contactor performance significantly.
The counter-current configuration is usually better than co-current ozonation
chamber. At high kd, partition of input-gas into several chambers could also improve
contactor performance. At very low kd, applying all gas in the first chamber is
recommended.
Though increasing the water flow rate results in a decrease o f the dispersion and
an increase in ozone residual in the effluent o f an ozonation chamber, the contact time
228
would also be reduced. Generally, the selection of operating conditions is highly
dependent on water quality and contactor configuration, and should be evaluated with the
model for each specific system.
8.5. Weakness and Alternative of CT Concept
Limitations of the current CT concept include:
Inactivation kinetics: The inactivation rate constants suggested by guidelines were based
on scarce data and included a strong safety factor, therefore, the credited inactivation
level could be significantly lower compared to actual values.
Contact time: The use o f tio as contact time was originally a conservative measure, where
/ io values are obtained from tracer tests. However, the safety factor provided from /i0
might vary. For example, the results from pilot-scale experiments revealed the
presence o f a high mixing zone at the bottom of the contactor where the majority of
dissolved ozone was present for counter-current operation, thus the safety margin for
tio might fail and result in an over-estimation in inactivation.
Average ozone concentration: In guidelines proposed by U.S. EPA, C can be averaged
from the direct measurement of residuals along contactor, or calculated from the
residuals measured in the influent and effluent of each chamber. However, the direct
measurement is usually not feasible, and the guidelines based on in/outlet residual are
somewhat arbitrary and usually result in inaccurate estimations. Furthermore, the
ozone concentrations measured in a full-scale contactor could have great variability,
therefore increasing the inaccuracy in the estimation o f CT.
229
By contrast, the disinfection model developed in this research could reasonably
estimate the inactivation level achieved in an ozone contactor with the combination of
appropriate hydrodynamic and transfer information, and proper input of ozone decay and
microorganism inactivation kinetics, which can be accurately obtained with experiments
performed separately in smaller scale reactors. Therefore, with further extensive studies
to evaluate the model parameters in full-scale contactors, the model developed in this
research could be a promisingly better alternative to the current CT concept used in the
regulations.
LIST OF REFERENCES
230
LIST OF REFERENCES
Abraham, M ., and Sawant, S.B., "Hydrodynamics and Mass Transfer Characteristics of

Packed Bubble Columns" The Chemical Engineering Journal, 43, 95-105 (1990).
Ahmad, M . & Farooq, S., "Influence o f Bubble Sizes on Ozone Solubility Utilization and
Disinfection" Water Science & Technology, 17(6/7), 1081-1090 (1985).
Akita, K. and Yoshida, F., “Bubble Size, Interfacial Area, and Liquid Phase Mass Transfer
Coefficients in Bubble Columns”, Industrial & Engineering Chemistry, 13(1), 84-
91 (1974).
American Public Health Association (APHA), American Water Works Association

(AW W A), and Water Environment Federation (WEF), Standard Methods fo r
Examination o f Water and Wastewater, 18th ed. APHA, Washington, D C.
(1992).
AWWA, "Pilot-Scale Evaluation o f Ozone and PEROXONE", AWWA Reserach

Foundation, (1991).
Bader, H. and Hoigne, J., “Determination of Ozone in Water by the Indigo Method”,
Water Research, 15, 449-456 (1981).
Bailey, P.S., Ozonation in Organic Chemistry, Academic Press, Inc., New York, N Y
(1978).
Baird, M .H. and Rice, G., “Axial Dispersion in Large Unbaffled Column”, The Chemical
Engineering Journal, 9(2), 171-174(1975).
Bird, R.G. and Smith, M .D ., “Cryptosporidiosis in Man: Parasite Life Cycle and Fine
Structural Pathology”, J. Pathol., 132(3), 217-232 (1980).
231
Blank, B.D., Marinas, B.J., Corsaro, K., and Rakness, K.L., “Enhancement of Wastewater
Disinfection Efficiency in Full-Scale Ozone Bubble-Diffixser Contactors”, Ozone
Sciences & Engineering, 15(4), 295-310 (1993).
Boyce, D. S., Sproul, O.J., and Buck, C.E., "The Effect o f Bentonite Clay on Ozone
Disinfection of Bacteria and Viruses in Water", Water Research, 15, 759-767
(1981).
Calderbank, P.H. and Moo-Young, M .B .,”The Continuous Phase Heat and Mass Transfer
Properties of Dispersions”, Chemical Engineering Science, 16, 39-47 (1961).
Cerf, O., "Tailing o f Survival Curves o f Bacterial Spores." Journal o f Appied

Bacteriology, 42, 1-19 (1977).
Chapra, S.C. and Canale, R.P., Num erical Methods fo r Engineers, Second Ed., McGraw-
Hill, Inc., New York, N Y (1988).
Chaudhari, R. V. and Shah, Y .T., “Recent Advances in Slurry Reactors”, Chapter 6 in

Concepts and Design o f Chemical Reactors, II, Ed.by Whitaker, S. and Cassano,
A.E., Gordon and Breach Science Pub., New York, N Y (1986).
Chiou, C.F., Marinas, B.J., and Adams, J.Q., “Modified Indigo Method for Gaseous and
Aqueous Ozone Analyses”, Ozone Sciences & Engineering, 17, 329-344 (1995).
Chiou, C.F., "Theoretical Analysis o f Ozone Disinfection Kinetic Studies and

Development of Non-Biological Surrogate Indicators for Assessment of Ozone
Disinfection Efficiency", Ph.D. Thesis, Purdue University, West Lafayette, IN
(1994).
Coffey, B .M ., Graff, K.G., Mofidi, A.A., and Gramith, J.T., "On-Line Monitoring of
Ozone Disinfection Effectiveness within an Over/Under Baffled Contactor",
Proceedings AWWA Annual Conference, Anaheim, Calif. (1995).
Danckwerts, P.V., “Significance o f Liquid-Film Coefficients in Gas Absorption”,

In d u stria l & Engineering Chemistry, 43(6), 1460-1467 (1951).
Danckwerts, P.V., Gas-Liquid Reactions, McGraw-Hill, Inc., New York, N.Y. (1970).
232
Deckwer, W .D ., Burckhart. R , and Zoll, G., “Mixing and Mass Transfer in Tall Bubble
Columns”, Chemical Engineering Science, 29, 2177-2188 (1974).
Domingue, E.L., Tyndall, R.L., Mayberry, W.R., and Pancorbo, O.C., “Effects o f Three
Oxidizing Biocides on Legionella pneumophila Serogroup 1”, A pplied and
Environmental M icrobiology, 54(3), 741-747 (1988).
Duguet, J.P., Brodard, E., Dussert, B., and Mallevialle, J., "Improvement in the
Effectiveness o f Ozonation o f Drinking Water through the Use o f Hydrogen
Peroxide" Ozone Sciences & Engineering, 7(3), 241-258 (1985).
Farooq, S., Chian, E.S.K., and Engelbrecht, R.S., "Basic Concepts in Disinfection with
Ozone", Journal o f WPCF, 49(8), 1818-1831 (1977a).
Farooq, S., Engelbrecht, R.S., and Chian, E.S.K., "The Effect o f Ozone Bubbles on
Disinfection", Progress Water Technology, 9(1), 233-247 (1977b).
Fauris, C., Danglot, C., and Vilagines, R., “Inactivation of Viruses by Disinfection
Methods”, Water Supply, 4, 19-41 (1986).
Ferguson, D.W ., Gramith, J.T., and McGuire, M.J., "Apply Ozone for Organics Control
and Disinfection: A Utility Perspective", Journal o f AWWA, 5,32-39 (1991).
Finch, G.R., Smith, D.W ., and Stiles, M.E., "Dose-Response o f Escherichia C oli in
Ozone Demane-Free Phosphate Buffer", Water Research, 22(12), 1563-1570
(1988).
Finch, G.R. and Fairbaim, N., "Comparative Inactivation o f Poliovires Type 3 and MS2
Coliphage in Demand-Free Phosphate Buffer by Using Ozone", A pplied and
Environm ental M icrobiology, 57(11), 3121-3126 (1991).
Finch, G.R., Yuen, W.C., and Uibel, B.J., “Inactivation o f Escherichia co li Using Ozone
and Ozone-Hydrogen Peroxide”, Environmental Technology, 13, 571-578 (1992).
Finch, G.R., Black, E.K., Labatiuk, C.W., Gyurek, L., and Belosevic, M ., “Comparison of
G iardia Lam blia and G iardia M uris Cyst Inactivation by Ozone”, A pplied and
Environm ental M icrobiology, 59(11), 3674-3680 (1993a).
233
Finch, G.R., Daniels, C.W., Black, E.K., Schaefer HI, F.W ., and Belosevic, M., “Dose
Response of Cryptosporidium Parvum in Outbred Neonatal C D -I Mice”, Applied
and Environmental M icrobiology, 59(11), 3661-3665 (1993b).
Fogel, D., Isaac-Renton, J., Guasparini, R., Moorehead, W ., and Ongerth, J., “Removing
G iardia and Cryptosporidium by Slow Sand Filtration”, Journal o f AWWA,
85(11), 77-84(1993).
Foml, L., Bahnemann, D., and Hart, E.D., “Mechanism o the Hydroxide Ion Initiated
Decomposition o f Ozone in Aqueous Solution”, Journal o f Physical Chemistry,
86(2), 255-259 (1982).
Gordon, G., Pacey, G.E., Cooper, W.J., and Rice, R.G., "Current State-Of-The-Art
Measurements o f Ozone in the Gas Phase and in Solution", Ozone Science &
Engineering, 10(14),353-366 (1988).
Guittonneau, S., Glaze, W .H., Duguet, J.P., Wable, O., and Mallevialle, J.,
“Characterization ofNatral Water for Potential to Oxidize Organic Pollutants with
Ozone”, Ozone Science & Engineering, 14(3), 185-196(1992).
Gurol, M .D . and Singer, P.C., “Kinetics of Ozone Decomposition, a Dynamic

Approach”, Environmental Science & Technology, 16(7), 377-383 (1982).
Gurol, M . and Singer, P., “Dynamics o f the Ozonation o f Phenol - Mathematical

Simulation”, Water Research, 17(9), 1173-1181 (1983).
Haas, C.N., "A Mechanistic Kinetic Model for Chlorine Disinfection", Journal o f ES&T,
14(3), 339-340(1980).
Hart, E.J., Sehested, K , and Holcman, J., “Molar Absorptivities of Ultraviolet and Visible
Bands of Ozone in Aqueous Solutions”, Analysis Chemistry, 55, 46-49 (1983).
Hacker, D.S. and Lockowitz, T., "Use o f Ozone in The Disinfection o f Coliphage T-7
Virus", American Institute o f Chemical Engineers Symposium Series, 73(166),
242-251 (1976).
234
Helmer, R.D. and Finch, G.R., “Use o f MS2 Coliphage as a Surrogate for Enteric Viruses
in Surface Water Disinfected with Ozone”, Ozone Sciences & Engineering, 15(3),
279-293 (1993).
Henry, D.J. and Freeman, E.M ., "Finite Element Analysis and Tio Optimization o f Ozone
Contactors", Ozone Science & Engineering, 17, 587-606, (1995).
Hikita, H. and Kikukawa, H., “Dimensionless Correlation o f Liquid-Pahse Dispersion

Coefficient in Bubble Columns”, J. Chem. Eng. Japan, 8(5), 412-413 (1975).
Hikita, H., Asai, S., Tanigawa, K., Segawa, K., and Kitao, M .,”The Volumetric Liquid-
Phase Mass Transfer Coefficient in Bubble Columns”, Chemical Engineering
Journal, 22(1), 61-69 (1981).
Hoff, J.C. and Geldreich, E.E., "Comparison o f The Biocidal Efficiency of Alternative
Disinfectants", Journal o f AWWA, 73(1),40-44 (1981).
Hoff, J.C., "Inactivation o f Microbial Agents by Chemical Disinfectants", USEPA Project

Summary, EPA/600/S2-86/067 (1986).
Hoigne, J. and Bader, H., “The Role o f Hydroxyl Radical Reactions in Ozonation
Processes in Aqueous Solutions”, Water Research, 10(5), 377-386 (1976).
Hoigne, J. and Bader, H., “Ozonation of Water: Oxidation-Competition Values of

Different Types o f Water Used in Switzerland”, Ozone Sciences & Engineering,
1(4), 357-372 (1979).
Hoigne, J., “Mechanisms, Rates and Selectivities of Oxidations o f Organic Compounds

Initiated by Ozonation o f Water”, Chapter 12 in Handbook o f Ozone Technology
and Applications, Vol. I, Rice, R.G. & Netzer, A. Eds. Ann Arbor Science, Ann
Arbor, M I, Vol. 1 (1982).
Hoigne, J. and Bader, H., "Rate Constants o f Reactions o f Ozone with Organic and
Inorganic Compounds in Water-I: Non-Dissociating Organic Compounds", Water
Research, 17(2), 173-183 (1983a).
235
Hoigne, J. and Bader, H ., "Rate Constants o f Reactions o f Ozone with Organic and
Inorganic Compounds in W ater-II: Dissociating Organic Compounds", Water
Research, 17(2), 185-194 (1983b).
Hoigne, J. and Bader, H ., "Rate Constants o f Reactions o f Ozone with Organic and
Inorganic Compounds in W ater-Ill: Inorganic Compounds and Radicals”, Water
Research, 19(8), 993-1004 (1985).
Hoigne, J., “Characterization o f Water Quality Criteria for Ozonation Processes Part. I:
Minimal Set o f Analytical Data”, Ozone Sciences & Engineering, 16(2), 113-120
(1994).
Hoigne, J. and Bader, H ., “Characterization o f Water Quality Criteria for Ozonation

Processes Part II: Lifetime o f Added Ozone”, Ozone Sciences & Engineering,
16(2), 121-134 (1994).
Horn, L.W ., “Kinetics o f Chlorine Disinfection in an Ecosystem”, Journal o f the Sanitary

Engineering D ivision, ASCE, 98(SA1), 183-194(1972).
Houzelot, J.L., Thiebaut, M .F., Charpentier, J.C., and Shiber, J., “Hydrodynamic Study of
Bubble Columns”, Entropie, 19(113-4), 121-126(1983).
Hughmark, G.A., “Holdup and Mass transfer in Bubble Columns”, In d u stria l &
Engineering Chemistry, 6(2), 218-220 (1967).
Hull, C.S. and Singer, P.C., “Modeling o f Ozone Contactors”, Presented a t the American
Water Works Association Anm tal Conference, New York, N Y (1994).
Ishizaki, K., Sawadashi, K., Miura, K., and Shinriki, N., “Effect o f Ozone on Plasmid
DNA o f Escherichia c o li in situ'1, Water Research, 21(7), 823-827 (1987).
Kaneko, M ., "Effect o f Suspended Solids on Inactivation o f Poliovirus And T2-Phage by

Ozone", Water Science & Technology, 21(3), 215-219 (1989).
Kato, Y. and Nishiwaki, A., “Longitudinal Dispersion Coefficient o f a Liquid in a Bubble

Column”, International Chemical Engineering, 12(1), 182-187(1972).
236
Kim, C.K., Gentile, D .M ., and Sproul, O.J., "Mechanism o f Ozone Inactivation of

Bacteriophage f2", A pplied and Environmental M icrobiology, 39(1), 210-218
(1980).
Koide, J., Morooka, S., Ueyama, K., Matsuura, A.,Yamashita, F., Iwamoto, S., Kato, Y ,
Inoue, H., Shigeta, M ., Suzuki, S., and Akehata, T., “Behavior o f Bubbles in
Large Scale Bubble Columns”, J. Chem. Eng. Japan., 12(2), 98-104 (1979).
Korich, D.G., Mead, J.R., Madore, M.S., Sinclair, N.A., and Sterling, C.R., “Effects of
Ozone, Chlorine Dioxide and Monochloramine on Cryptosporidium Parvum
Oocyst Viability”, A pplied and Environmental M icrobiology, 56(5), 1423-1428
(1990).
Krasner, S.W., e ta l., "The Occurrence o f Disinfection By-products in US Drinking

W ater", Journal o f AWWA, 81(8), 41-53 (1989).
Kuo, C H., Li, K.Y., Wen, C.P., and Weeks, J.L. Jr., "Absorption and Decomposition of
Ozone in Aqueous Solutions", American Institute o f Chemical Engineers
Symposium Series, 73(166), 230-241 (1976).
Kuo, C.H. and Wen, C.P., "Ozonizations o f Formic Acid, Formaldehyde, and Methanol in
Aqueous Solutions", American Institute o f Chemical Engineers Symposium
Series, 73(166), 272-282 (1976).
Kuo, C.H. and Yocum, F.H., “Mass Transfer of Ozone into Aqueous System”, Chapter 5
in Handbook o f Ozone Technology and Applications, Vol. I, Rice, R.G. & Netzer,
A. Eds. Ann Arbor Science, Ann Arbor, M I, Vol. I (1982).
Labatiuk, C.W., Belosevic, M ., and Finch, G.R., “Factors Influencing the Infectivity of
G iardia m uris Cysts Following Ozone Inactivation in Laboratory and Nature
Waters”, Water Research, 26(6), 733-743 (1992a).
Labatiuk, C.W., Belosevic, M ., and Finch, G.R., “Evaluation o f High Level Ozone
Inactivation of G iardia m iris Using an Animal Infectivity Model”, Ozone Sciences
& Engineering, 14(1), 1-12 (1992b).
237
Labatiuk, C.W., Belosevic, M ., and Finch, G.R., “Inactivation o f G iardia rrturis Using
Ozone and Ozone-Hydrogen Peroxide”, Ozone Sciences & Engineering, 16(1),
67-68 (1994).
Langlais, B., Reckhow, D.A., And Brink, D.R., Ozone in Water Treatment Application
and Engineering, 2 Ed., Lewis Publishers, Michigan (1991).
Laplanche, A., Sauze, N. L., Martin, G., and Langlais, B., “Simulation o f Ozone Transfer
in Water: Comparison with a Pilot Unit”, Ozone Sciences & Engineering, 13(5),
535-558 (1991).
Lawler, D.F. and Singer, P C., “Analyzing Disinfection Kinetics and Reactor Disign: a
Conceptual Approach Versus the SWTR”, Journal o f AWWA, 85(11), 67-76
(1993).
Lee, J.H. and Foster, N.R., "Measurement of Gas-Liquid Mass Transfer in Multi-Phase
Reactors", Applied Catalysis, 63(90), 1-36 (1990).
Legeron, J., “Ozone Disinfection of Drinking Water”, Chapter 1 in Handbook o f Ozone

Technology and Applications, Vol. II, Rice, R.G. & Netzer, A. Eds. Ann Arbor
Science, Ann Arbor, M I (1982).
Lev, O. and Regli, S., "Evaluation of Ozone Disinfection Systems: Characteristic Time T"
Journal o f Environmental Engineering, ASCE, 118(2), 268-285 (1992a).
Lev, O. and Regli, S., "Evaluation of Ozone Disinfection Systems: Characteristic

Concentration C", Journal o f Environmental Engineering, ASCE, 118(4), 477-
494 (1992b).
Levenspiel, O., Chemical Reaction Engineering, 2 ed., John Wiley & Sons, Inc., New
York (1972).
Lewis, W.K. and Whitman, W.G., "Principles o f Gas Absorption", Industrial &
Engineering Chemistry, 16(12), 1215-1220 (1924).
Lin, S.D., “G iardia lam blia and Water Supply”, Journal o f AWWA, 77(2), 40-47 (1985).
238
Majumdar, S.B., Ceckler, W .H., and Sproul, O.J., "Inactivation o f Poliovirus in Water by
Ozonation", Jou rna l ofW PCF, 45(12), 2433-2443 (1973).
Marinas, B.J., Liang, S. and Aieta, E.M., "Modeling Hydrodynamics and Ozone Residual
Distribution in a Pilot-Scale Ozone Bubble Diffuser Contactor" Journal o f AWWA,
85(3), 90-99 (1993).
Masschelein, W.J., "Contacting o f Ozone with Water and Contactor Offgas Treatment",
Chapter 6 in Handbook o f Ozone Technology and Applications. Vol 1, Rice, R.G.
and Netzer A. Eds., Ann Arbor Science, M I (1982).
Matter-Muller, C., Gujer, W ., and Giger, W., “Transfer o f Volatile Substances from Water
to the Atmosphere”, Water Research., 15, 1271-1279(1981).
Mudd, J.R., “Reaction o f Ozone with Amino Acids and Proteins”, Atmos. Envir., 3, 669-
676(1969).
National Research Council. Committee on Medical and Biological Effects of

Environmental Pollutants, Ozone and Other Photochemical Oxidants, National
Academy o f Science, Washington, D C., 341-363 (1977).
Ohki, Y. and Inoue, H ., “Longitudinal Mixing o f the Liquid Phase in Bubble Columns”,
Chemical Engineering Science, 25, 1-16 (1970).
Peeters, J.E., Mazas, E.R., Masschelein, W.J., Maturana, I.V ., and Debacker, E., “Effect
of Disinfection o f Drinking Water with Ozone or Chlorine Dioxide on Survival of
Cryptosporidium parvum Oocysts”, Applied and Environm ental M icrobiology,
55(6), 1519-1522 (1989).
Perry, R.H., Green, D. W ., and Maloney, J.O., P e rry’s Chemical Engineers ’ Handbook,
6th ed., McGraw-Hill, New York, N Y (1984).
Rakness, K.L., Renner, R.C., Hegg, B.A., and Hill, A.G., "Practical Design Model for
Calculating Bubble Diffuser Contactor Ozone Transfer Efficiency", Ozone Science
& Engineering, 10(2), 173-214(1988).
239
Roth, J.A. And Sullivan, D.E., "Kinetics o f Ozone Decomposition in Water", Ozone
Science & Engineering, 5(1), 37-49 (1983).
Roustan, M ., Mallevialle, J., Roques, H., And Jones, J.P., "Mass Transfer o f Ozone to
Water: a Fundamental Study", Ozone Science & Engineering, 2(4), 337-344
(1981).
Roustan, M „ Beck, C., Wable, O., Duguet, J.P., and Mallevialle, J., “Modeling Hydraulics
o f Ozone Contactors”, Ozone Science & Engineering, 15(3), 213-216 (1993).
Roustan, M „ Wang, R. Y., and Wolbert, D ., "Modeling Hydrodyanmics And Mass

Transfer Parameters in a Continuous Ozone Bubble Column", Ozone Science &
Engineering, 18, 99-115 (1996).
Roy, D., Engelbrecht, R.S., Wong, P.K .Y., and Chian, E.S.K., "Inactivation of
Enteroviruses by Ozone", Progress Water Teachnology, 12, 819-836 (1980).
Roy, D., Chian, E.S.K., and Engelbrecht, R.S., "Kinetics o f Enteroviral Inactivation by
Ozone", Journal o f Environm ental Engineering, ASCE, 107(EE5), 887-901
(1981).
Roy, D., Engelbrecht, R.S., and Chian, E.S.K., "Comparative Inactivation o f Six
Enteroviruses by Ozone", Journal o f AWWA, 74(12), 660-664 (1982).
Setterfield, C.N. and Huff, G.A., “Mass Transfer Limitations in Fischer-Tropsch Slurry
Reactors”, Chemical Engineering Science, 36(4), 790-791 (1981).
Sauze, N.Le., Laplanche, A., Orta De Velasquez, M .T., Martin, G., Langlais, B., and
Martin, N., "The Residence Time Distribution o f the Liquid Phase in a Bubble
Column and Its Effect on Ozone Transfer." Ozone Science & Engineering, 14,
245-262 (1992).
Saxena, A.C., Rao, N.S., and Saxena, S.C., "Bubble Size Distribution in Bubble
Columns", The Canadian Journal o f Chemical Engineering, 68(2), 159-161
(1990).
240
Scott, K.N., Wolfe, R.L., and Stewart, M.H., "Pilot-Plant-Scale Ozone and Peroxone
Disinfection o f G iardia M uris Seeded into Surface Water Supplies", Ozone
Science & Engineering, 14(1),71-90 (1992).
Severin, B.F., Suidan, M .T ., and Engelbrecht, R.S., "Series-Event Kinetic Model for
Chemical Disinfection", Journal o f Environmental Engineering, ASCE, 110(2),
430-439 (1984).
Shah, Y .T., Kelkar, B.G., and Godbole, S.P., "Design Parameters Estimations for Bubble
Column Reactors", A IC hE Journal, 28(3), 353-379 (1982).
Shinriki, N., “Degradation o f Nucleic Acids with Ozone. HI. Mode o f Ozone-Degradation
o f Mouse Proline Transfer Ribonucleic Acid (tRNA) and Isoleucine tRNA”, Chen.
Phcrrm. B u ll, 31(10), 3601-3612 (1983).
Singer, P.C., Saravanan, K., And Hull, C.S., "Kinetics o f Ozone Decomposition: Effects
o f pH, Carbonate, and Humic Substances", Preprints o f Papers Presented at the
203rd ACS N ational Meeting, Elzerman, A.W . ed., Division o f Environmental
Chemistry, American Chemical Society (ACS), San Francisco, California, 32(1),
741-744 (1992).
Smith, D.W . and Zhou, H ., "Theoretical Analysis o f Ozone Disinfection Performances in a

Bubble Column." Ozone Science & Engineering, 16(5), 429-441 (1994).
Sobsey, M .D ., "Inactivation of Health-Related Microorganisms in Water by Disinfection

Processes", Water Science & Technology, 21(3), 179-195 (1989).
Sobsey, M .D ., Takashi, F., and Hall, R.M., “Inactivation o f Cell-Associated and Dispersed
Hepatitis A Virus in Water”, Journal o f AWWA, 83(11), 64-67 (1991).
Solo-Gabriele, H. and Neumeister, "US Outbreaks o f Cryptosporidiosis", Journal o f

AWWA, 88, 76-86(1996).
Sotelo, J.L., Beltran, F.J., Benitez, F.J., and Beltran-Heredia, J., "Ozone Decomposition in
Water: Kinetic Study", Industrial & Engineering Chemistry Research, 26(1), 39-
43 (1987).
241
Sotelo, J.L., Beltran, F.J., Benitez, F.J., and Beltran-Heredia, J., “Henry’s Law Constant
for the Ozone-Water System”, Water Research, 23 (10), 1239-1246 (1989).
Sproul, O.J., Pfister, R.M ., and Kim, C.K., “The Mechanism of Ozone Inactivation of
Water Borne Viruses”, Water Science & Technology, 14, 303-314 (1982).
Staehelin J. and Hoigne J., “Decomposition of Ozone in Water: Rate of Initiation by

Hydroxide Ions and Hydrogen Peroxide”, Environmental Science & Technology,
16(10), 676-681 (1982).
Staehelin J. and Hoigne J., “Decomposition of Ozone in Water in the Presence of Orgnic
Solutes Acting as Promoters and Inhibitors of Radical Chain Reactions”,
Environmental Science & Technology, 19(12), 1206-1213 (1985).
Stankovic, I., "Comparison o f Ozone and Oxygen Mass Transfer in a Laboratory and Pilot
Plant Operation", Ozone Science & Engineering, 10(3), 321-336 (1988).
Stanley, J.W. and Johnson, J.D., “Analysis of Ozone in Aqueous Solution”, Handbook o f
Ozone Technology and Applications, Vol. I, Rice, R.G. & Netzer, A. Eds. Ann
Arbor Science, Ann Arbor, M I, Vol. I, Chap. 8 (1982).
Stoebner, R. A., and Rollag, D.B., “Ozonation of a Municipal Groundwater Supply to

Reduce Iron, Manganese, and Trihalomethane Formation”, Aqua, 4, 291-305
(1981).
Straka, M.R., Gordon, G., and Pacey, G.E., “Residual Aqueous Ozone Determination by
Gas Diffusion Flow Injection Analysis”, Analysis Chemistry, 57, 1799-1803
(1985).
Sullivan, D.E. and Roth, J. A., "Kinetics of Ozone Self-Decomposition in Aqueous

Solution", American Institute o f Chemical Engineers Symposium Series, 76, 142-
149(1980).
Sundaresan, A. and Varma, Y.B.G., “Dispersed Phase Holdup and Bubble Size
Distributions in Gas-Liquid Cocurrent Upflow and Countercurrent Flow in
Reciprocating Plate Column”, The Canadian Journal o f Chemical Engineering,
68, 560-568 (1990).
242
Tomiyasu, H.,Fukutomi, H ., and Gordon, G., “Kinetics and Mechanism o f Ozone

Decomposition in Basic Aqueous Solution”, Inorganic Chemistry, 24(19), 2962-
2966 (1985).
U.S. EPA, "Determining Chloramine Inactivation o f Giardia for the Surface Water
Treatment Rule", Microbiological Treatment Branch, Risk Reduction Engineering
Laboratory, and Parasitology and Immunology Branch, Environmental Monitoring
Systems Laboratory, U.S. Environmental Protection Agency, Cincinnati, Ohio
(1989).
U.S. EPA, Office o f Drinking Water, Guidance M anualfo r Compliance w ith the
F iltra tio n and D isinfection Rewuirements fo r Public Water Systems Using Surface
Water Sources, American Water Works Association, Denver, CO (1991).
U.S. EPA, “National Primary Drinking Water Regulations: Enhanced Surface Water
Treatment Requirements; Proposed Rule”, Fideral Regster 59(145), 38832-38857
(1994).
van Sonntag, C., “Disinfection by Free Radicals and UV-Radiation”, Water Supply, 4, 11-
18 (1986).
Vaughn, J.M., Chen, Y.S., Lindburg, K., and Morales, D., "Inactivation of Human and
Simian Rotaviruses by Ozone", Applied and Environmental M icrobiology, 53(9),
2218-2221 (1987).
Vaughn, J.M., Chen, Y.S., Novotny, J.F., and Strout, D., "Effects o f Ozone Treatment on
The Infectivity o f Hepatitis A Virus." Canadian Journal o f M icrobiology, 36(8),
557-560 (1990).
Waller, S.A., "Flow Cytometry Light Scatter for Assessing G airdia Inactivation by
Ozone", Thesis, Purdue University, West Lafayette, IN (1994).
Watt, R.D., "Ozonation o f Dissolved Organic Matter in Substrate Varied Activated

Sludge Effluents”, Ph.D. thesis, Purdue University, West Lafayette, IN (1984).
Wei, J.H. and Chang, S.L., “A Multi-Poison Disinfection Model for Treating Disinfection
Data”, Chapter 2 in D isinfection Water and Wastewater, Ed. Johnson, J.D., Ann
Arbor Science Pub. Inc., Ann Arbor, M I (1975).
243
Weiss J., “Investigations on the radical HO 2 in Solution”, Transactions Faraday Society,

31, 668-681 (1935).
White, G.C., The Handbook o f Chlorination. 2nd ed., Van Nostrand Reinhold Co., New
York (1986).
Wickramanayake, G.B., Rubin, A.J., and Sproul, O.J., "Inactivation of Naegleria and
G ia rd ia cysts in water by ozonation", Journal ofW PCF, 56(8), 983-988 (1984a).
Wickramanayake, G.B., Rubin, A.J., and Sproul, O.J., "Inactivation of G iardia lamblia
Cysts with Ozone”, Applied and Environm ental M icrobiology, 48(3), 671-672
(1984b).
Wickramanayake, G.B., Rubin, A.J., and Sproul, O.J., "Effects o f Ozone and Storage
Temperature on Giardia Cysts", Journal o f AWWA, 77(8),74-77 (1985).
Wickramanayake, G.B. and Sproul, O.J., "Ozone Concentration and Temperature Effects
on Disinfection Kinetics", Ozone Science & Engineering, 10, 123-135 (1988).
Williams, F.P. Jr., “Membrane-Associated Viral Complexes Observed in Stools and Cell
Culture”, A pplied and Environmental M icrobiology, 50(2), 523-526 (1985).
Xu, F. and Liu, C., "Mass Balance Analysis o f ozone in a Conventional Bubble Column"
Ozone Science & Engineering, 12, 269-279 (1990).
Yamada, H. and Somiya, I., “Identification o f Products Resulting from Ozonation of

Organic Compounds in Water”, Ozone Science & Engineering, 2(3), 251-260
(1980).
Yamashita, F., Mori, Y ., and Fujita, S., “Sizes and Size Distributions of Bubbles in a
Bubble Column-Comparison between the Two Point Electric Probe Method and
the Photographic Method”, J. Chem. Eng. Japan, 12(1), 5-9 (1979).
Yao, D.C.C. and Haag, W.R., “Rate Constants for Direct Reactions of Ozone with
Several Drinking Water Contaminants”, Water Research, 25(7), 760-773 (1991).
244
Yoshida, F. and Akita, K., “Performance o f Gas Bubble Columns: Volumetric Liquid-
Phase Mass Transfer Coefficient and Gas Holdup”, American Institute o f
Chemical Engineers Symposium Series, 11(1), 9-13 (1965).
Zhou, H., Smith, D.W ., and Stanley, SJ., “Modeling of Dissolved Ozone Concentration
Profiles in Bubble Columns”, Journal o f Environmental Engineering, ASCE,
120(4), 821-840 (1994).
Zhou, H. and Smith, D.W., “Kinetics o f Ozone Disinfection in Completely Mixed

System”, Journal o f Environmental Engineering, ASCE, 120(4), 841-858 (1994).
APPENDICES
245
APPENDIX A: COMPUTER PROGRAMS
246
A .I. Programs for Ozonation and Disinfection Tests
• 1 1Inactivation units and CT-lag ■

T h e contactor are assumed to be CSTR+ADR-CSTR. this program are focused on kn - 12.6: lag - 03
‘ solving center ADR. B C o f AD R are based on M -B o f top and bottom CSTR nn - kn * d / u l
'Newton’s Linearization and Approxim ation Method were Used in M atrixO YD
T h e solving o f pressure were based oo ( I/Y . ug) * Pb/P * ( 1/Yb. ugb) GOSUB G e tfW ile
’ a • (P b /P )^ 67 • ab are optional that controlcd by varyaS C A L L VinisStepfgO. live)
* Yb (gfi+jn )) obtained were convened back to Y in subroutine ’calkill - -LOG flive) / LOG( 10)
CLOSE 42.44
OPTION BASE I GOSUB ResuHOutput
D E F IN T I-J NEXT jtest
DECLARE SUB OpenBC (g(). cb) PRINT 41.
DECLARE SUB VirmStep (g(). live) CLOSE 41
DECLARE SUB BentioElim (b<). g()) END
C O M M O N SHARED pi. dn. nl. nd. nr. s, a n . din. jn . jm , yin. him. dose
CO M M O N SHARED csum. cl. vm. lot. nn. lag. varyaS, pressS. typeS, st, atb. pb
pi - 3 141592653599 REDIM gCOO) GetProfile
cxa - 6 A 2 * p i / 4 / 144 cl * 19 75 atb * 25 ’—check an - 0
CLS jn - 40: typeS * "counter'* '—check jm - 3 * jn. REDIM g(jm )
OPEN "s-d-ssum" FOR OUTPUT AS 41 ' —check C A LL OpenBC(gO» cb)
9
FOR jtest - I TO 26 —check RETURN
READ testS. mp, ql. qg. tmp. yin. ltd. din. st. db
OPEN testS * ' sds" FOR OUTPUT AS 42 ResultOutpuc
•OPEN testS * " vir" FOR OUTPUT AS 44 • Print parameters and results ■
REM ’Calculate the Parameters f t Constants----------------------------------- PRINT INPUT INFO- ‘ .testS
IF st < atb THEN st • atb ♦ 01
• Pressure effect controller (atm) ■■■' PRINT 41. ............... testS. " • • • • • • • • -
pressS - "yes" varyaS * "yes" PRINT 4 1 ," flo w -". typeS, ‘ pressure-", pressS. " a_press-". varyaS
IF pressS - "yes" THEN pb = I - 98 • cl • 3048 / 1013 ELSE pb - I PRINT 41. USING " Q!(gpm>-####0 4#” . ql;
'1 1 Superficial velocities (ft/m in) (ql=gpm. qg-read) ■ PRINT 41. U S IN G " . 0 g (s d h )-4 # 4 #404". qg.
u! - ql / (cxa * 7 4814) ug * qg / cxa / 60 PRINT 41. U S IN G " . 0 3 dose-### 444"; dose
’ 1 Bubble sizcfcm) f t Applied Ozone Dosage(mg/L) ■■ ■ PRINT 41. U S IN G " u«fpm )-#0# 0##4*\ ul.
•db • 207 ♦ 213 • ug • 12 * 2 54 / 60 PR IN T41. U S IN G * . u g(fpm )-000####“ . ug.
db - db / I 067 PRINT 41. USING " . ugb 444 4444". ugb
1 Dispersion Coefficient (sqcm/s) and dn (whole c l) 1 PR IN T41, U S IN G " Tem pC O -44444". tmp.
el • 4 5718 • qg * 13 401 PRINT 41. U S IN G " . yin(ppm)-444 44". yin;
dn • el / (* » !• 1 2 * 2 5 4 /6 0 ) . (d • 12 * 2 54) PRINT 41. U S IN G ” . O Dinfppm )-## 4444". din
11 ,ni 11 ’ Standard Og (qg-scfh) from gauge pressure (psi) — " ■■■ PRINT 41. U S IN G " dn - 444 4444". dn.
mp - mp / 14 7 I PR IN T41. U S IN G " . CIO -4 4 4 4444". tIO;
IF pressS * "yes" THEN qg - qg • mp A 5 PR IN T4 1 .U S IN G * . U a -# 4 4 4444".k l • a
ug - qg / cxa / 60 ugb ■ ug pb dose * ugb • yin / ul PRINT 41, USING " avg.03 conc. - 44 444". csum;
■ Henry's Law Constant for ozone (ppm/ppm) — ■■■ PRINT 41. USING " . CT - 44 444". csum • tIO / jm
IF tmp > 1 5 G 0 T 0 2 PRINT 41. U S IN G " . k d ( l/m in ) -4 4 4 4444". kd
h im - 1794* 10 A1-840 / (tmp *2 7 3 15#)) GOTO 3 PR IN T41. U S IN G " kn-444 444". kn,
2 him - I 574# • 10 ' 6 • 10 ' (-1687/(tm p * 273 154)) PRINT 41. U S IN G " . Live % -444 444". liv e * 100;
■ Viscocity (CcntiPotse. cm2/s) PR IN T41. U S IN G " . dB(cm )-44 44 4 ".db
3 xxx - SQR(8078 4# * (tmp • 8 4354) A 2) PRINT 41. USING " 03-out - #4 4444". cb,
mu! - I / (214824 ■ (tmp ♦ 8 435# + xxx) - 120) PR IN T4 1 .U S IN G " .log-kill - 444 444". caOrill.
Oxygen difTusivity (sqcm/sec) « * « = — PRINT 41. USING " M - 444.444". him
do3 - 6 * 10 A (-8) • (tm p * 273 154) /(m u l • 100) — m an balance (must lu H ). enhanced w / ^density and -kla) - — "
* 111 Bubble rise velocity (ft/m in )— =— *mtn - yin * ugb
xxx - 010054/m u l * 6 0 / 2 544 / 12: vb - 210 • db • xxx I F typeS - "counter* THEN
IF db > 18 THEN vb * (33 84 • EXP(-4 84 • db) * 20) • xxx * mout • g(jn) * ul * g(jn ♦ I) * him • ug
■" 1 Gas/liquid interfanal area (1/ft) ■■■■ * ELSE
IF varyaS - "yes" THEN uga - ugb ELSE uga - ug ‘ mout - g ( l) • ul * g(jn + I) * him • ug
a - 6 • 2.544 • 12 • uga / ((vb * ul) • db) •END IF
IF typeS - "co" THEN ’PR IN T41. U S IN G " Mass-In - 4444 444444". min;
a - 6 * 2.544* 12 "u g a / ((vb * ul) • db) *PRINT 41. USING " . Mass-Out —44
EN D IF RETURN
'1 ■ ReynoldsSchmidLSherwood 4 f t Hughmark ki (ft/m w )r - —
sd - mul / d o i R E A D tcstSjnp. qL.qg.tmp. yinJuLdin. sufb—
rg - db • vb • 254# • 12 / (mul • 60) TJATA 0I06-CO.10 6.3.45.8.5.16 3. 12.7776.0 0175.0 2. 0 409.0 285
xxx - db • (980) A ( I ' 3) / (d o i) ' (2 / 3) D A T A 0109-CO.8 6, 3 6.1.1.16.1. 51 84.0 0146.0 33. 0380.0.217
s h n - 2 * 01874 * ( r g A 4840 • sd A 3390 • xxx A (7 2 / 1000)) A 1614 D A T A 0115-C0.8.6, 13 10.1.15.6. 54.3143.0 0107.0 18. 0 228.0 226
k l - shn • d o i • 60 / (db • 2 540 • 12) D A T A 0118-co. 10.6, 13 1.8 5.15 7. 28 2152.0 01895.0 2 1 .0 310.0 285
’ ozone transfer units and stripping factor at b D A T A cyst-8.9 6. 7 2.4 3.16. 7 6429,0 0108.0.50. 0 4 .0 247
n l - l d * a * c l / u l s « him * ugb / id D A T A cyst-25.9.6, 72.4 3.16, 13 0577.0 0108.0.39. 0 4,0.247
nd - kd * cl / ul: nr - 20 • cl / ul
247
R E M S D Y N A M IC FOR i - I TO 3
d(i. I ) - b t ( i )
R E M S STA TIC d(i. 2) • a(i. 2)
SUB Analytical (g(). pnntS) <Hl 3 ) - * i . 3)
N EXT i
REM ‘Solving lambda, x l. i l x 3 — GOSUB Determinant3X3
cnO ■ I Ctrl • I ctr2 - 1 ctr3 - I ‘swich o f config delta 1 - det
IF typeS - -c o -T H E N cnrO « - 1 ‘Calculate delta2 ■1
*x*3 +• a*x^2 +■ b*x - c - 0 *==*— FOR i ■ I TO 3
snl - s / nl d(i, I) * a(i. 1)
a - (dn * eaO • s n l)' (-dn * snl) < * i. 2 ) - b * i )
b * (-ctrO ♦ nd • snl - s) / (-dn • snl) d ( i.J ) - a < i.3 )
c * -nd / (-dn • snl) NEXT i
Reduced y*3 ♦ p*y * q - 0. QQ - (jV3)*3 - (<y2r2 — ■■■ GOSUB DetcfmmanfJX3
p --a *a /3 *b d d ta 2 -d e t
q -2 *a *a *a /2 7 -a » b /J *c Calculate ddta3 "
Q Q -p *p « p /2 7 *q » q .4 FOR i - I TO 3
IF Q Q > 0 T H E N d(i. I ) - a ( » , l )
PRINT "Q > 0 . Complex roots, no analytical soin‘ " • " <*t. 2 ) - * i . 2)
GOTO quitAnaiyiical dCi, 3 ) - b t ( i)
END IF NE XT i
*yI - 2*SQR(-pf3)*COS(r/3> — GOSUB Detcrmiiiant3X3
y2J - -2*SQR(-p/J)*COS(r/3 ♦- 60pi/l80) delta3 * det
*COS(r)“ -<^( 2*SOR(-<pf3)AJ ) ). r in radian y l • d d ta l / delta
‘x - y - a/3. x represent Imbda here” """' y 2 - d d ta 2 /d e lta
xxx - -2 • ABS(SQR(-p A 3 / 27)) / q y3 “ ddta3 / delta
r - pi - ATN(ABS(-SQR( I ♦ 27 • q A 2 / 4 / <p ' 3))) • xxx) RETURN
GOTO 23
2 2 r - ATN(ABS(-SQ R(I * 27 • q A 2 / 4 / ( p A .))))• xxx) REM Determinant o f third order 11
'
23 x l * 2 • ABS(SQ R(-p 3)) • COS(r / 3) ♦ a 3 Detennmant3X3
I F A B S ( x l A 3 - * a * x l A 2 * b * x I * c ) > l» T H E N G O T 0 22 d e t - d ( I . I) • <K2.2) • d(3 .3) ♦ d(2. I) • d(3.2) • d (1 .3)
x2 - -2 • ABS(SQ R(-p/ 3)) # C O S ( r / J - p i I ) - a / 3 det - d(3. I) • d ( l . 2) • d (2 .3) - d ( l. 3) • d (2 ,2) • d(3. I) * det
x3 « - 2 • ABS(S0R(-p / 3)) • COS(r / 3 > pi 3) - a / 3 d e t - d e t - d ( l. I) * d(2.3) * d (3 .2) - d (1 .2) • d(2. I) * < * 3 .3 )
RETURN
REM ‘Solving for 3 Boundary equations .................
IF typeS - -co* THEN quitAnalytical:
Ctrl - E X P (x l) ctr2 - EXP(x2): ctr3 - EXPU3) ‘switch o f cfg
END IF END SUB
y l. y2. y3 denote alpha, beta, gama "m" " n
’ *al I a l2 a !3 * *yl* 'b tl* SUB BenitoEIim (b(). gO)
* *a21 a22 a23* *y2* * ’bt2’
‘ *a31 a32 a33* *y3’ -bt3’ “ 6010 REM 'matrix inversion with accompanying solution o f linear equations
D IM a (3 .3). bt(3) REDIM idO'm)
a (l. I) • x l • (1 - snl• x l) • E X P (xl) / Ctrl 6020 DETERM -V
a (I. 2) - x2 • ( I - snl• x2) • EXP(x2) / ctrZ 6030 FOR i - I TO jm
a (I. 3) - xJ • ( I - snl• xJ) • EXP(x3) / ctr3 6040 id(i) - 0
a(2. I) - ( I - (snl * ctrO • dn) • x l ♦ ctrO • dn • snl • x l A 2) • Ctrl 6050 N E XT
a<2. 2) - ( I - (snl * ctiO • dn) • x2 «• ctiO • dn • snl • x2 A 2) • ctr2 6060 FOR K M N * I TO jm
a<2. 3) * ( I • (snl * cut) • dn) • xJ ♦ ctrO • dn • snl • x i A 2) • ctr3 6070 B M A X - I 1
a(3. 1) - E X P (x l) a(3. 2) - EXP(x2): * 3 . 3) - EXP(x3) 6080 FOR i - I TO jm
b t ( l ) - 0 - bt(2) ■ a n bt(3) - ytn / him 6090 IF id ( i ) O 0 THEN GOTO 6250
GOSUB ThreeEquations 6)00 B N E X T -O '
6)10 B T R Y -01
REM ‘Results and outputs-------------------------------------------- 6120 F O R j-IT O jm
•PRINT "y l.y 2 .y 3 -. y l. yZ. y3 6130 IF id(D o 0 THEN GOTO 6200
PRINT "check x l" . \ l . x l A 3 * - a * x l A 2 - b * x l * c 6140 IF ABS(b(i. j)) < - BNEXT THEN GOTO 6200
PRINT "check. x2". * 2 . x 2 A 3 * a « x 2 A 2 - b * x 2 - * - c 6150 BNEXT -A B S flrfi.j))
PRINT "check.x3". xJ. x3 A 3 * a • x3 A 2 * b • x3 »■c 6)60 IF BNEXT < - BTRY THEN GOTO 6200
‘Get dissolved and gas ozone concentration ■■■■ 6170 B N E X T -B T R Y
F O R i - I TO jn 6180 B T R Y -A B S (b (i.j))
z - i / jn 6190 jc - j
xxx - y2 • ( I - snl • x2) • EXP(x2 • z) 6200 NEXT
yyy - y3 • (1 - snl • x3) • EXP(x3 • z) 6210 IF BNEXT > - B M AX • BTRY THEN GOTO 6250
g(i ♦ jn ) - y l • EXP(xl • z) * y2 • EXP(x2 • z) * y3 • EXP(x3 • z) B M A X -B N E X T /B T R Y
6220
g(») “ y I • ( I - snl • * I ) • EXP(xI • z) * xxx * yyy 6230 IR O W - i
IF g (i) < 0 THEN g(i) * 0 6240 J C O L - jc
IF printS • "yes" THEN 6250 NE XT
PRINT *2. USrNG "*####### cl • ( ! - ( i - l ) / ( j n - 0). 6260 IF idQ'c) - 0 THEN GOTO 6290
PRINT *2. USrNG "### ####» g(i); g(i jn ) • him 6270 D E TE R M -0>
EN D IF 6280 GOTO 6590
+
csum “ csum g(t) 6290 id(JCOL) - 1
NE X T i 6300 IF JC O L -[R O W THEN GOTO 6410
IF typeS - "co" THEN Oout * g( 1) ELSE Oout - g(jn) 6310 F O R j - I T O j m
GOTO quitAnalytical 6320 SSAVE - b(!R OW ,j)
6330 b(IROW. j ) • b(JC0L, j)
REM ’ Solution for Three Linear Equations------ 6340 b (J C O L .j)-S S A V E
ThreeEquaaonx 6350 NE XT
REDIM d(3. 3) 6370 SSAVE - g(lROW)
'Calculate delta 6380 g(IROW ) - g(JC0L)
FOR i - I TO 3 6390 g(JCOL) - SSAVE
d(i. l ) - a ( i . I) 6410 F - I! / b(JCOL, JCOL)
d ( i. 2 ) - a ( i . 2) 6430 FOR j - 1 TO jm
d(i. 3) « afi. 3) 6440 b(JCOL. j) - b(JCOL j ) • F
NEXT i 6450 NEXT
GOSUB Determinam3X3 6470 g(JCOL) - g (K O L ) • F
delta - det 6490 F O R i - I TO jm
Calculate delta I * « * » * • 6500 IF i- J C O L THEN GOTO 6570
248
6510 F - bfi. JCOL) b f i * 2 * j n . i * 2 * j n - l ) - - l / <2 * detz)

6520 F O R j-irO jm bfi ♦ 2 • jn . i) - old / s / ap
6530 b fi, j) - bfi. j) - F • bfJCOL. j) b(t + 2 * j n , i ' » - 2 # jn + I) ■ I / (2 • detz)
6540 NEXT W i * 2 * j n . i * 2 * j n ) * - o ld /s / ap * p
6560 g fi) - gfi) - F • gfJCOL) NEXT i
6570 NEXT
6580 NE X T -B .C ^ z - l —*—
6590* IF pressS - "yes" TH EN p - ( I ♦ 0295 * d • ( I - x b » / pb ELSE p - I
IF varyaS • "yes" THEN ap • p ELSE ap • 1
ENDSUB xx • EXPf-nlb • p / s / ap). as ■ s • yin / him • ( I • xx)
bb ■ I ♦ ndb ■*■(! - x x ) * s / p : e e - nrb: gg - bb - as • ee
SUB OpenBC (g(). cb) *■■■■“ Parameters for B C o f dissolved ozone (& z - l ■ 111 ■
T h is sub solving middle ADR w/ matrix. CSTR w/ equation at result printout IF typeS - "counter* THEN
‘z direction fo r both counter • and co-current is from up to down bfjn. jn ♦ 3) • -dd / (detz A 2)
b(jn. j n - 2 ) “ 4 * d d / (detz A 2) - I / 2 / detz
R E D IM c(jm ) itmax - 10 iter * 0- detz - 1 (jn - I) ‘—check b (jn.jn - 1) • -5 • d d /(d e tz A 2) * 2 /detz
nit - nl • (st - atb): old » ni • ( I • st): nib - nl • atb b(jn. jn ) * 2 • dd / detz A 2 - 3 / 2 / detz - nld / ap - odd - nrd • c<2 • jn )
ndt • nd • (st - atb) ndd * nd • ( I - st): ndb - nd • atb b(jn, 2 • jn ) • -ard * cfjn) bfjn. 3 * jn ) * nld / ap • p
nrt ■ nr • (st - atb) nrd - nr • 11 • st) nrb * nr • atb gfjn) - -nrd • cfjn) • c(2 * jn )
dd • dn / (1 - st): pn * I ELSE
In itia l Guess [ 1 - Dts 03]fjn~ 03 Demand](2jn- Gas 0 3 ] ------------- fi ■ (ee • (din - cin) gg) A 2 * 4 * b b * e e * (aa * an)
FOR i * I TO jn cb ■ (ee • (cin -d in ) - gg * pn • fi A 5) / 2 / ee/ bb
IF typeS - "counter- THEN b fjn .jn • 2) " d d / 2 / d e tz b (jn .jn • I ) « - 2 • d d / detz
g(i) - dose / 2 jn • »: g fi ♦ jn ) = 9 • din • 5 • din / jn • i bfjn. jn ) • 1 3 * dd / 2 / detz gfjn) ■ cb
ELSE gfi) * dose ' 2: g(i ♦ jn ) - din / jn • i END IF
END IF ‘ 11 ■ 1 ■ Parameters for B C o f ozone demand (g) z» I 1 ■ ■ 11
gfi ♦ 2 • jn ) - yin him / jn • i b f 2 * j n , 2 * j n - 2 ) « b ( j n . j n - 2 ) : b(2 • jn . 2 • jo - 1) - b fjru jn - 1)
NEXT i IF tygieS • "counter* THEN
b(2 • jn , 2 • jn - 3) - bfjn. jn - 3)
InitialOpen. bf2 • jn , 2 • jn ) - 2 • d d ( d e t z A 2) - 3 / 2 / detz - nrd • cfjn)
F O R j - 1 T O jm bf2 • jn . jn ) ■ -nrd * c f 2 • jn): g(2 • jn ) ■ -nrd • cfjn) * c f 2 • jn)
cO) • sCi) ELSE
NEXTj d b " d in / ( I ee • g(jn))
*
iter * iter 1 REDIM b<jm. jm ). g(jm ) b f2 * jn . 2 * jn ) « b 0 'n .jn ) g(2 * j n ) » d b
END IF
’Defining matric parameters— -B C 2 Parameters far B C o f gas phase (g z*l
IF pressS - "yes" THEN p - 1 1 * 0295 * cl " (« - atb)) / pb ELSE p - I IF typeS ■ "counter* THEN
IF varyaS - "yes" THEN ap = p ELSE ap ■ I hh “ yin / him • xx: kk ■ ( I - xx) / p
* - 1 1111" Parameters for B C o f aqueous 03 at z -0 — ■■■■■■ f i « (ee • cf2 • jn ) - ee • cfjn) •*-gg)A 2 + 4 * b b * c e # (aa-*' cfjn))
IF typeS « "counter" THEN fcO - I / 2 / b b • fi A - 5 • (ee • cfjn) ♦ 2 • b b - g g - e e * c f 2 • jn ))
xx • EXP(-nlt • p - s / ap) fdO - I / 2 / bb • fi A - 5 • (ee • c(2 • jn ) «* gg - ee • cfjn))
b( I . I ) - I ♦ n d t - s * x x / p * s / p + n r t * c(jn ♦ I ) b(3 • jiu jn ) - k k / 2 / bb ♦ pn • kk • fcO
b f i. jn ♦ I) - nn • cf I) b( I. 2 • jn I) * s • (xx - I) b( 3 • jn . 2 • jn ) - -kk / 2 / bb ♦ pn • kk • fdO
g( 1) - nrt • c< 1) • cfjn * I ) * cm bf3 • jn . 3 • jn ) ■ -I
ELSE g(3 • jn ) ■ -hh pn • fcO • cfjn) + pn • fdO • cf2 • jn )
b( I . I ) - 2 • dd detz A 2 - 3 ' 2 •' detz ♦ nld ap -ndd - nrd * d jn ♦ I ) g(3 • jn ) • g(3 • jn ) ♦ kk / 2 / bb / ee • (gg - pn • f i A 5)
b (I. 2) - - 5 • d d ' detzA 2 - 2 / detz ELSE
b (l. 3) - 4 " dd / detz A 2 - 1 2 / detz b< I. 4) - -d d /d e tz A 2 b(3 • jn . 3 • j n ) « I
1 b<
b( I . jn ♦ I ) * -nrd * cf ) 1. 2 • jn I ) * nld * p / ap g(3 * jn ) “ (yin / him - gfjn) / p) • xx ♦ gfjn) / p
g ( l) - -nrd • c(I) • cfjn ♦ 1) END IF
END IF
s Parameten for B C o f O D i^ f O j ■■■■■■ Solving Matrix and Comparing R esult----------------------------------
IF typeS - "counter" THEN C A LL BenhoEIimfbf). g())
b(jn I . jn * I ) = I *• nrt • c( I ) b(jn + 1.1) -* nrt • c(jn * I ) FOR i • I T O jm
g(jn ♦ I ) * nrt * cf I ) • c(jn * I ) din IF ABSfgfi) - c fi)) > 005 A N D (iter < itmax) THEN GOTO InitialOpen
ELSE NEXT i
b(jn ♦ l . j n + l ) « 2 * d d / detz " 2 - 3 / 2 detz - nrd • cfjn - I ) csum - 0: j t - 4: jb ■ 4
*
b(jn -*• I. jn 2) ’ b f l. 2 ) Ujn+ l . j n - 3 ) b (l. 3) ■■ 1Print Profiles at top CSTR
*■
b(jn ♦ I. jn 4) ~ bCI - 4) bfjn * I. 1) - -nrd • c(jn ♦ I) IF typeS • "counter" THEN
g(jn ♦ I ) « -nrd • ct 11 • cfjn * 1) c t - g ( l ) d t - g ( jn ♦ I)
END IF ELSE
'' Parameters for B C o f gas 0 3 (fvz= 0 ■ p • ( I *■ 0295 • cl • (st - atb)) / pb: ap - p
‘
b(2 • jn ♦ I, 2 • ]n • I) • -3 (2 • detz)- nld / s / ap • p xx ■ EXP(-nIt * p / s / ap): aa • s • g(2 • jn I) • ( I - xx)
b(2 • jn ♦ 1 .2 • jn * 2) » 2 (detz) b b " l > n d t * ( l - x x ) * s / p e c “ n it:g g -b b -a a *ce
'
b(2 • jn ♦ 1 .2 • jn ►3) = -I (2 •detz) fi • (ee • (gfjn ♦ I) - g fl) ) ♦ gg) A 2 ♦ 4 • bb • «e • (aa ♦ g f l) )
b(2 • jn ♦ I. I) • nld ' j ap g(2 • jn ♦ I) “ 0 ct ■ (ee • (g( I ) - gfjn ■*•!))- g g - * - p n * f i A 5) / 2 / ee / bb
dt - gfjn ♦ I) / ( I + e e * a )
------------------ ADR Equation-------- END IF
FOR i - 2 TO (jn • I ) FOR i t - 1 TO j t + I
IF pressS - "yes" THEN IF pressS-"yes" THEN
p « (I 0295 • cl • (st - atb * ( I • st) • (i - I) • detz)) / pb p - ( I ♦ 0295 • cl • (st - atb) • ( it - l ) / j t ) / p b
ELSE p > I ELSEp- I
EN D IF END IF
IF varyaS ■ "yes" THEN ap - p ELSE ap * I IF varyaS - "yes* THEN ap - p ELSE ap - I
Parameters for Equations o f dissolved 0 3 ■■ xxx • EXPfnlt • p / s / ap • ((it - 1) / j t - 1))
IF typeS - "counter" THEN yy • I ELSE yy « -I y t - ( ( g ( 2 * jn + I ) - c t / p ) * xxx + ct / p) • p • him
b(i. i - I) - dd / (detz A 2) ♦ yy / (2 • detz) height - d - (cl • (st ♦ atb) • (it - 1) / jt)
b(i, i ♦ I ) - dd / (detz A 2) - yy / (2 • detz) PRINT *2. USING "### ##### ". height, c t y t dt
b(i. i) ■ -2 • dd / (detz A 2) - nld / ap • ndd • nrd * c(jn i) NEXT it
b(i, i ♦ jn ) ■ -nrd • c(i) bfi. i * 2 • jn ) - nld / ap • p 1 ■ ■ Print Profiles at middle A D R ■■■■■■■■
g(i) - -nrd * c f i ) • c(jn ♦ i) FOR i - 1 TO jn
parameter for equations o f 0 3 demand — IF preaS • "yes* THEN
bfjn + i. jn + i - I ) ■ b(i. i - 1) p - ( ! + 0295 * d • ( s t - a t b * ( l - s t ) * ( i - l ) * d e t z ) ) / p b
b(jn ♦ i. jn ♦ 0 ■ -2 • dd / (detz A 2) • nrd • c(i) ELSEp - 1
*
b(jn i, jn * i ♦ I) - b(i. i * I) bfjn i. i) - -nrd • c(jn i) END IF
g(jn i) - -nrd • cfi) • cfjn * i) cm - g(i): y m - p * g ( i + 2 * j n ) * h in t <kn- gfl ♦ jn )
s Parameters for equations o f gas phase ’IF gfi) < 0 TH EN gfi) - 0
249
height ■ cl - (cl • (st • atb M 1 - st) • («• I ) * detz)) < i) - di / (b(i) - ai • e(i - 1))
PRINT #2. USING "»## »#«## *; height, cm. ym; dm NE XT i
cxum - csum ♦ g(i)
N E XT i *Step Input to get virus density as a ftinc o f time t and depth z . -------
Print Profiles at bottom CSTR ■■■1■— •Obtain v(z)@t from the tracer curve o f previous time cJzXgijt-1.--- -----
F O R ib -O T O jb ’and output every v(t) @ z—n to virus(t) —■
IF pressS-"yes-TH E N F O R i - I T O ( j t * I)
p - ( l - 0295 • cl • ( I - (jb - ‘b) • a t b / j b ) ) / pb g ( l) - s p ik e
ELSEp - 1 g (2)-v(2)/dt
END IF « t B < l ) - < g ( l ) / b l - x l * g ( 2 ) / b l / d a ) / ( I - x l • a i / b l / di)
IF varyaS - "yes” THEN ap * p ELSE ap 9 I F O R j - 2 TO jz
IF typeS - "counter" THEN g ti) " *0 ")1dl
fi “ (ee • (g(2 • jn ) - g(jn)) * g g ) ' 2 * 4 * b b * e e * ( a a * g(jn)) en(i) - <*(i) - « * - 1)) f (b(j) •* * dj - 1))
cb - (ee • g(jn) - ee • g(2 • jn ) • gg ♦ pn • fi A 5)■' 2 / b b / e e NEXTj
db - g(2 * j n ) / (ee • c b * I) v (jz + I) - xn • c(jz - 1) * eta(jz) • xn • eta(jz - 1) - an * eta(jz)
ELSE v ( jz * I) - v ( jz * I) / (bn - an • e(jz) * xn • e<jz) • e(jz -1 ))
cb - cb db - db F O R j - j z TO I S T E P -I
END IF v(j) - etaO*) - e(j) • v(j * 1)
MX - EXP(nlb • p / s / ap • ( i b / j b - 1)) NEXTj
yb - ((yin / him - cb / p) • xxx *• cb / p) * p • him virusfi) - v (jz + 1)
height - d - (d * ( I - (jb - ib) • atb / jb )) NEXT i
PRINT #2. USING "*## #»### ". height: cb. yb: db LiveBug - vtrusCjSampleTune)
N E X T ib
RETURN
csum - ct * (st • atb) * csum / jn • ( I - st) *• cb • atb
IF iter - itmax THEN StepOutputScreen:
PRINT "Newton-Raphson method may not c o n v c rg !""” ’— Output Normalized Tracer data and Predicted Curve to Screen---------
END IF •SCREEN 9: COLOR 7
•WINDOW ( - 5 . - 5 M 5 .5 .3 3)
EndOpen: U N E (0. OHS. 0) LIN E (0. OHO. 3)
• F O R j - I TO j t
END SUB * LIN E (dt * j • dt. v u u s (j)H d t * j. virus(j * I)). 11
NEXTj
SUB VirusStcp (03<). live)
REM SDYNAM1C
T h e survival % arc tested independent with spike amount (good), and were
time - 4: icut - 10 jt * 500 jstep - 10 '---------check * tested decreasing with increasing finite resolution
jSompleTime - 5. d - dn jz - icut * jn - icut Note: Inacbviation results at a certain time (ex: take sample at I denaan
jSampleTime - jSamplcTimc / umc * j t ' time) are different between Pulse-Input and Step-Input
REDIM o(icut * jn ). v(jz ♦ 2)
'— reproduce ozone profile .......... END SUB
F O R j - I TO jn - I
‘
det03 - (0 3 (j * I ) * 03(j)) icut
FOR i - I TO icut
o(icut • j - i * I) - 03(j) * ('cut • i) • det03
NEXT i
NEXTj
o<jz + l ) - 0 3 ( j n )
•— calculate survival % at tim e -1 . s p ik e jin v ----------------

' Without inactivation - = —
vin - 10000 spike - vtn nnv - 0
GOSUB StepVirusTimeCurve
nolull - LivcBug
1 1 W ith Inactivation and consider CTIag
spike • vin • EXP(Iag • kn) nnv « nn
GOSUB StepVirusTimeCurve
live - LiveBug / nokill
*— Print Out Results-------------------------
F O R i - I TO jz * I STEP jstep
IF v(i) > vin TH EN v(i) - vtn
height - d »c! * (i • I) / jz
IF typeS - "co" THEN
PRINT #4, height, vijz * 2 • i) / vin • 100. o<i)
ELSE P R IN T #4. height. v(i) / vin * 100. o(i)
END IF
N EXT i
GOTO StepOutputScreen
StepVirusTimeCurve
'— in itial the variables-------------

d t - t i m e / j t : d z - I • jz
REDIM v<jz * 2). gOz * 2). b<jz ♦ 2). v(jz * 2)
REDIM e(jz * 2). eta(jz * 2). virus(jt * 2)
*— initial the coefficient in m a trix-------------
d l - - 2 * d / d E d i - * d / ( d z A 2) * I / 2 / d z
b l - I - * * 3 * d / 2 / d z bn • -3 / 2 / dz
F O R i - I TO jz ♦ I
b(0 * 2 * d / ( d z A 2 ) * I / <Jt ♦ nnv • o(i)
IF typeS * "co" THEN b(i) - 2 " d / d z A 2 * I / dt * nnv • o(jz ♦ 2 - i)
NE XT i
x l - d / 2 / d r xn - - I / 2 / d z
ai - -d / (dz • dz) - I / (2 • dz) an - 2 / dz
—calculate e( I ) e(n)-----------
e ( I) - ( d l / b l - x l • b(2) / b l / di) / ( I - x l # a i / b l / da)
FOR i - 2 TO (jz * I)
250
A.2. Program for Tracer Test
•----------------------------------- PREDICTION-----------------------------------------
T h is program analysis tracer tests, generate normalized curve, calculate OPEN nameS * " p r e " FOR OUTPUT AS #2
’dispersion coefficient, dispersion number, detention time. HO. t50. t90 time - 4 : jz - 1000; j t - 1000 ’------- check
Co, recovery percentage, and predict unit pulse system response curve dt - tim e / j t dz • I / jz
RE D IM C0(jx* X*(j * 2), 2 2)
DEFDBL A - R L-Z D IM dHjz* 2). cta#(jz ♦ 2)
OEFINT I-K D IM conc#(jt + 2). tbeta#(jt + 2). final#(jt * 2)
REM SDYNAM1C *— in itia l the coefficient in m a trix -------------
D IM t#(2000). c#<2000) '— check d l - - 2 * d / d z d » “ - d / (dz A 2) + 1 / 2 / dz
INPUT "Data filename & path(if not current) '' . nameS b l - l + 3 * d / 2 / d r b i « 2 * d / ( d z A 2) + l / d t b n - - l / 2 / d z
OPEN nameS FOR INPUT AS # I xl - d / 2 / d t x n - .J / 2 / d z
OPEN nameS ♦ ’ nmi" FOR OUTPUT AS *3 ai - -d / (dz • dz) - I / (2 • dz): an - 2 / dz
OPEN nameS + " sum" FOR OUTPUT AS *5 FOR i - 0 TO (jz ♦ I)
INPUT "Water flow rale (gpm)'r\ ql eta#(i) - 0: c#(i) - 0: t#(i) - 0
jn - a ql « ql • 3 7854 NEXT i
DO W H ILE NOT EO F(l) FOR i « O T O ( j t I)
jn - jn ♦ I conc#(i) “ 0: thcta#(0 - O final#(i) - 0
INPU T #1. t#(jn). c#(jn) NE XT i
LOOP ’— calculate e ( l ) ... e(401)-----------
’ initialize and boundanze element values------------ e # ( 0 ) - f r e # ( l ) - ( d l / b l - x l * b i / b l / d i ) / ( I - xl • a / b l di)
c l - c # ( l ) * ( t # < 2 ) - t # ( l ) ) / 2 e n - c # 0 n ) • ( t » ( jn ) - O K j n - 1)) 2 F O R t - 2 T O ( j z > I)
t l - ( t # ( l ) > t # ( 2 ) ) ' 2 tn » ( u w t # ( j n - D ) 2 dH i) • di / (hi - ai • e#(i - 1))
Sc - el ♦cn. Set - el M l * en • tn: Sett " e l M l A 2 * c n " m * 2 N E XT i
’ Integrate f t calculate momentum by central retangu. dement-------- t# (l)- 1 l)-0
FOR i - 2 TO jn - I 'Assume step input to get conc as a func o f time t and depth z --------
dt ■ (t# (i ♦ l » / 2 clem - c#(i) * dt •Calculate cftXgjz-n by compute c(z)(aft-i inorder to get t(zX£tPn+1 —
t - (O K i * U * 2 * t» (i) * t # ( i - I ) ) / 4 'and store c(t) to conc(t)----------------------------- ---------- -
Sc ■ Sc * d e n t Set * Set *• t • clem: Sett ■ Sett * t • t • elem FOR i • 2 TO ( jt ♦ 1)
NEXT i t#(2) - c#(2) / dt
* Get mean time, tracer mass. 110. tSO. vanence. f t s td ----------- e t a # ( l ) - ( t # ( l ) / b l - x l • t#(2) / b l / di) / (1 - x l * * / b l / d i)
mean * Set / Sc: m an 3 Sc " ql F O R j - 2 TO jz
var • Sett / Sc - mean ' 2 Std ■ var A 5# t#U) - c#(j)/ *
high - e l HO - <yt50 - a tW ** 0 cta#(j) - (t#0) - ai • eta#(j - 1)) / ( b i . at • e#(j • D)
FOR i - 2 TO jn - I NEXTj
low * high c#(jz I) - xn • flS(jz - 1) • cta#(jz) - xn *eta#(jz* I) - an * eta#(jz)
high - low ♦ c#(i) • (t#(i *■ I ) • t#(i - I)) / 2 aHjz* l ) - c # ( j z * I ) / (bn - an • e#(jz) ♦ xn • e#(jz) • dHjz
- I))
IF ( l o w - I * * Sc) * (high • I * • Sc) <■ 0 THEN F O R j - jz T O I S T E P -I
tIO - t # ( i - 1) ♦ (t#(i) - t#(i - I)) • ( I# • Sc • lo w) /(h ig h - low) c#(j) - eta#0) - «*(j) * c#(j * I )
ELSEIF (low - 5# • Sc) • (high - 5# • Sc) • - 0 T H E N NEXTj
1 5 0 - OKi - I) ♦ (t#(i) - t#(i - I)) * ( 5 # • Sc • lo w )/(h ig h • low) cono t(i) • c#(jz ♦ 1)
ELSEIF (low - o® • Sc) • (high - 9# • Sc) * 0 THEN NE XT i
t90 - OKi - I) ♦ (t#(i) - t#(i - I)) • ( 94 • Sc - lo w )/(h ig h • low) D ifferentiate step tnput(accumulated) concentration conc(t) to g e t-------
END IF ’ pulse input concentration flnal(t). then print out time period ---------
N E XT i ' theta(t) and final(t) <g z -jz ( outlet) to file " pre” -------------------
' — Use Newton method to find dispersion num ber----------- FORi - I T O jt
dtry - 01# d - I# fioaUKi) • (cooc#(i ♦ I ) - conc#(i)) / dt
DO W H ILE ABSfd dtry) > - 000001# thcta#(i) - dt * i - dt
d - dtry PRINT #2, USING " ## ######". theta#(i). final#(i)
deno —2 - 4 * d * 4 * d * EXPf-l / d) ♦ 2 • EXP(.| / d) NE XT t
dtry - d - ( 2 * d - 2 * d A 2 M I - EXP(-I / d)) • var / mean A 2) / deno CLOSE #2
LOOP END
‘— Find dimensionless ume .conc. and accumul mass-----------
>
x • 1 / mean: y - Set (Sc A 2) accu - c l • x • y
PRINT #3. USING " ## ######". t# (I) • x. c#( I) • y
FOR i - 2 T O j n - I
accu ■ accu ♦ c#(i) • (t#(i ♦ I ) • t#(i - I)) / 2 • x • y
PRINT #3. USING ’ ## ######*. t#(i) • x. c#(i) • y
N EXT i
PRINT #3, USING " *# ######". t#(jn) * * . c#(jn) • y
PRINT #5. test tm
names
PRINT #5. "Water flow rate (Ipm) - "; ql
PRINT #5. "Residence time (m m )- mean
PRINT #5. "Dispersion number ■ *; d
PRINT #5. “ 10% residence time tlO (m in) * tIO
PRINT #5. "50% residence time (50 (min) ■ t50
PRINT #5. "90% residence time t90 (min) • t90
PRINT #5. "Recovered tracer mass (g) - ". mass / (10 A 6)
C L O S E # !. #3. #5
251
APPENDIX B: PHOTOS OF EXPERIMENTAL APPARATUS
■ ht n\A/npr Further reproduction prohibited without permission.
Reproduced with permission of the copynght owner. Further repr
255
APPENDIX C: BUBBLE SIZE MEASUREMENTS
256
Os CO VO oo
2.6 19.6
CO ■<r
1.63 6.23
8.61
U VS vs OO r-
O % Os* cd ©*
Z ©' CN VO* cd
© © CO vs 00 VS 00 os CO
(mm)
34 2.93
o
2.88
8.56 ©
3.5
CN Os CN CN vs rd CN
l CN cd cd d
vs 55
CN CN vs Os Ov CN CO
99
so
% CN vo CN CN
r^
2.6 12.9
00
14.5
ro
3.73
29
u r** Os Tf
o % ©
rd CO ©*
ro Z
© © vs 00 VS OO
(mm)
o rO m
2-34
4.27
€91
© £
3.5
CN Os CN Os CN vs
vO
cK CN CN* <d cd
CN © CN Os
35
OS
% r-» r- CO
ro VO Os OS
3.9 1.12
2.6 11.5
2.93 5.58
27.5
U r^* 00 © r-
© % P^
© CN d ©*
© Z
CO VS OO vs 00
(mm)
00 ro
1.63
O ©
2'41
4.28
Os N
3.6
CN ©i Os os CN CN vs
CN 53 ©’ CN cd cd
CN CN ro r o vs Tf VS CN CO
% Tf P- CO
CN vs CN OO Tf Os
6.4
L s Os Os
© =*fc 00 CN ©' ovs
z m CN cd ©*
Ov
CN VO P- 00 Os
(mm)
1.24
1.55
2.23
Ov ©
3.5
CN N oo Tj*
2 .1 2
©i cd cd
p- 53 CN CN CN*
CN ro vo CN CN 00 CN
43
% ro VO CN
4.72]
CN CN CN CN P*-
10.4
2.6 14.2
U P-
O % ©*
Z CO CN ©
p-
CN OO CO vs 00
(mm)
1.63
1.95
O
2.35
© CO
Z.l‘3
Os
3.5
cn ©i N CN Os CN vs
Os 53 CN CN* cd cd
© © ©
30
45
22
% ro 3
1.43]
00 oo vs VO 00
24 1.24 11.4
16.7
SOI
U ro 00
© % CN
Os*
CN CN
CN ©*
00 Z
ro v> VO r- OO Os
(uiui)
2.07
On ©
3.56
901
CN ©i Os vs 00 cd cd
55 ©* CN CN CN
CN VS CN CN vs CN SO ro
=*fc vs VO CO CN
1.55 28]
20]
os r- CN OO N* 00
38 1.24 16.9
VS u
VO © % 00 vo r-* p^
o vd Os*
Tj* Z
CN ro P" 00 Os
(uiui)
Os ©
981
0.53
2.0 4
3.7
CN ©i VO Os rr r** cd
K ©' © CN CN CN*
Sampling Tube Size:
CN CN V) CO vs r- CN Os
% so CN CN
TV Bubble
Size (mm)
ea
Measured
u £ S 6 vs VO p- OO Os o CN
14
16
18
13
15
19
17
20
(N W
■©+* a a z
3
CB -i u S
cn O' O' H
P NO cn Tf
u 00* np-
1.63 1.96
13.4
18.6
O n cn ON
801
On
o % CN NO © J © p ON p- CN CN
Z © N* in cn CN o O d 257
«n
oCN CN 4> cn in oo cn «n OO m m OO N cn in 00
(uiui)
3.9
3.7
2.6
O N
7.15
ON CN ON CN p CN p 00 in p p
©
n
i cn CN CN cn cn n* N* Nf in in NO
in oo cn
53
OO O n ON NO tt
% CN P* NO 3 CN —
CN NO cn oo NO
1.94
13.6
2.6 18.3
NO CN cn
001
u CN p oo CN «n
O % © p^ ©
CN z
cn CN P «n © © o
O O aN
CN cn cn cn «n 00 °i cn in 00
(uiui)
2.99
OO
cn CN p oo
8.6
CN NO O n CN on CN p
p* 55 CN CN cn cn N*
p- O n o NO
20
CN cn cn CN
99
% — NO p- N* N" CN
U p- TT ON p p- o in
1.55 1.37
19.7
12.9
m
8.77
4.93
001
o OO OO CN m «n
Z cn piN* O O d
o
CN o p- OO wn p- 00
(uiui)
3-35
CN cn NO ON
CN p O N NO
OO TT ON
p^ cn cn P* O cn p ON CN P p
p-* 00 mi C/3 CN CN CN cn N* N* N* in in in NO*
«n CN CN VO p» CN «n 00 in 00 NO CN o CN
% NO p- so N* cn
U p- 00 00 m00
2.6 12.2
CN ON CN CN
o % cn cn ON N*
p» Z d OO cn CN OO
O aN cn cn •n OO 00
(mm)
3.9
2.52
3.25
© cn
4.3
CN p- NO ON CN ON p
00i 53 CN CN cn
«n
33
CN ON CN cn
zz
% ON p^ CN
«n 00
oU % 2.6 17.3 m cn
1.31
VO CN 00 cn © CN cn
m Z
© NO CN N* o
O in oo m 00 ON
© a
(mm)
cn
2.49
2.93
cn
4.3
CN NO ON CN CN p cn
p*‘
T•f cn CN cn cn
00 N*
53
CN ON cn
Sfc oo ON cn
00 N*
16.2
0.74
OO OO
u
o % cn NO cn - p^ oo
CN Z
© N* CN cn N*
N* in p* 00
£o>
(mm)
3.4
2.42
NO
N*. ON
p^ cn
4.3
CN © CN p OO P*
p**’
© cn CN CN CN cn
cn ocn «n CN
ZL
CN cn
% ON 3
p-
ou
2.3
CN p- p ON
% cn vq © cn NO* p
o Z CN m CN in
(mm)
N* m NO r^ OO ON
3.1
CN a cn
7.05
2.15
CN CN © N ON CN 00 P-
m 53 © CN CN CN
Tf OO cn ON P*- P*
96
=tfc in p- N*
2.6 15.9
1.59
U
o %
ON m
p- ON oo in P-
p^
NO in
Z © O n cn CN NO cn
p
oo* cn cn in 00
(mm)
3.25
cn 00
20
2.41
a
2.13
|mm
p- NO ON CN ON p
cn CN CN cn
00
40
CN cn
93
VO P* N*
% CN NO
p** cn 00
ou
2.28 12.2
ON NO ON
00
| 6.5
=*fc P- NO P-* O n cn
Z d CN CN N*
oo© a
n
1 20l \zvz
oo cn cn m
(mm)
3.25
no cn
90*1
N ON NO O n CN O n
p*’
CN cn
d CN
Sampling Tube Size:
70
76
20
CN cn
% cn —
TV Bubble
Size (mm)
Measured
Qc (scfh):
a
•o ?&
Test No.:
O CN m N* in p* 00
20
u OD — CN cn in VO p- 00 ON -
NO On
s -J
A
cn a
|
15.31
11.4
12.4
1.06
cn p 00
1.03 0.79
ON CN in o
o %
Z CN CN NO 258
(mm)
On p- in ON 00 CN NO
2.69
Size
tr
4.3
rz.
cn p~ p cn o P-
>
U CN CN CN cn cn cn
cn cn p» 00 00 rr
23
47
=*fc — n* 00 ON in
16.6
ON in TT OO CN m On
o % p* oo cn 00 © p m cn cn
Z © m oo rn CN p o*
cn m «o oo cn 00 p cn in oo CN
(mm)
2.6
3.25
3.39
©
Size
CN rn p* VO p CN p in cn* CN m 0( in
CN CN cn N* N* Tf
cn ON cn ON cn OO CN CN
85
67
46
% *r in ON in -
cn m P-* oo On
16.5
0.2
00 00 p-
661
12.5
13.7
0.2
o
15.1
b On
o % o ON in p p
Z <*■ «n oo rn o’
cn n oo p m m 00 CN cn in 00
2.6
O cn cn «n OO
CN rn s s NO ON CN ON CN in cn CN m 00 in in oo
m i ca E — — CN CN cn cn N* N" in in NO
00
N" oo cn NO p» in NO o in o
20
83
69
45
% CN NO P* in CN
13.2
00 oo cn ON
3.22
b O cn NO
o % o*
Z OO CN CN NO
i;
cn m oo
(mm)
3.9
•n oo NO
2.76
2.93
£91
Size
CN m •n p CN CN CN in
00* CN* cn cn
9-H
o p* 00 ON m
25
63
43 p* -N- CN
%
b 00 00 p- p- CN
O % NO P* ON NO P- 00 00 O n m
Z — p- CN O n 00 cn o o
rn cn m cn in 00 p cn m
(mm)
2.6
00
2.83
O O
Size
8.55
CN SO p CN p CN in cn CN in
rn
i CN* CN cn cn •n*
n Tf O p* in CN cn
09
24
49
% 00 cn CN
SO p- p- n oo oo
13.7
•N* oo o
15.6
8.94
La «n O in in CN CN CN
o % in
Z O cn p^ CN o CN CN O o
r— OO cn cn in 00 s o cn in 00 p cn m 00
20
in o 4* 9*
SO — «n N B CN NO ON CN CN ON CN in cn CN m oo
CN — oo 1 C/J S o *■" CN CN cn cn
in
CN CN CN p- ON o
56
77
49
Sfc cn ON CN
p» oo
10.7
OO m CN SO
0.3
5.95
w ON V) cn ON in in p
o %
Z CN NO* CN CN cn CN
o
6-1101
(mm)
OO cn cn «n OO cn in oo
2.6
3.9
I 24
Size
4.3
CN rn ON so p CN O n CN in
© CN CN cn* cn*
o NO O CN p*
22
36
98
__ 44
% 00 CN
1
La cn p* p- p^ ■n ; in 00 in cn CN
O ON 00 CN Os NO cn m
Z o CN CN CN O n in © o o o
2-103(
|°3
(mm)
cn in oo p cn m
2.6
CN OO cn cn in 00
Size
4.3
|mm
in cn ON no p CN p CN «n cn CN m
CN o CN CN* cn cn*
|
cn oo ON CN p» cn CN
80
47
% in P»* cn
o
9.6
CN in 00 CN CN O
14.6
811
La
| 6.5
o % oo cn in p rr NO NO cn
Z NO CN CN n - CN* o* o’ o*
o CN oo cn cn in 00 cn in oo p cn m 00
2.6
CN 4* S
CN rn cn — O N fi ON NO ON CN O n CN in cn CN m 00
CN CN ca E o — — CN CN cn cn
Sampling Tube Size:
cn
38
CN «n P- NO 00 CN CN O o
% CN p* oo N* cn
o ^
aa a s
4* ■o ?
u o 3 S — cn cn
NO P** 00 ON o
cn n NO p*
19
L a «n - CN tt 00
20
s 4» C
u
06 o
Z
ca s (A > h
4* A _) U A ii
s O' O' H
259
APPENDIX D: TRACER TESTS
260
Test 1-1106, Q l = 13.1 gpm, Q c = 8.55 scfh

Predicted Tracer Curve with d = 0.061
(dimensionless)
m m i U l ( ( l U U U y M l l ( l U W ( M U U U f U < ( U ( >-
Test 2-1030, Q L = 13.0 gpm, Qc = 4.30 scfh

86

WliUltlUHtttltmmWH XP
Test 3-1027, Q , = 13.0 gpm, Qc = 2.10 scfh

— Fitted with d = 0.0084, CSTR = 26.5%
0 ^ jSSJSSSHHBiQ-pOi— .— O -j— OQ— r-
261
Test 4-1118, Q l = 13.0 gpm, Qc = 17.0 scfh

1- Fitted with d = 0.0333, CSTR = 25.9%
(dimensionless)
^ lU t llllllllt p ggDOCh-OCDgpn
2 -


Fitted with d - 0.0190, CSTR = 26.6%
mammtnqmmuL'M PDgnp-o- I i i
Test 6-1101, Q l - 13.0 gpm, Qa - 4.30 scfh

Fitted w ith d = 0.0121, CSTR = 34.1%
262
2
(dimensionless)
0
2
Test 8-1115, Q l = 13.0 gpm, Q c = 17.0 scfh

0 g^ttttttit(fii[{7TTTrwiccmtrxiJiîi-XDa3Pc>
2

Fitted with d = 0.0191, CSTR =2 7.4%
0
0 1 2 3 4 5
263
Test 10-1012, Ql = 7.10 gpm, QG = 4_30 scfh

Fitted with d = 0.0221. CSTR = 42.2%
(dimensionless)
P T r p i i u 11 i,i i H f { ( i { { C ( ( { { ( ( n i n T P j i n n n R i n n i t
Test 11-1007, Q, = 7.00 gpm, Qc = 2.13 scfh


Test 12-1128, QL = 7.0 gpm, Qc =1.06 scfh

o -< ufUlffffffffffffirgnxzD-tinDp—
ctxifxmia
264
Test 13-1020, QL = 7.10 gpm, Qc = 8.50 scfh

(dimensionless)
1— ,— — 1— I-C
Test 14-1015, Ql - 7.10 gpm, Qc = 430 scfh


, TOw/»»frrrfmrrrKT|mTr)-r-n-
Test 15-1010, Ql = 7.05 gpm, QG=2.15 scfh

Fitted with d = 0.0157, CSTR = 32.'
265
2 -

(dimensionless)
««««(<««««( 1 l~“ l 1---1“


Fited with d = 0.0353, CSTR = 30.2%
PSSSHSSBflDi—O i— i— |—0 ~ i— i
2 -
Test 18-1017, Ql - 7.0 gpm, QG=4.30 scfh

[Win 11111111
o
266
2

Predicted Tracer Curve with d - 0.136
(dimensionless)
0 BjggggByu(m »(m ffl{tqtttttKnTm m r)-

2
0
2
Test 21-1210, Ql - 3.70 gpm, Qc = 0.53 scfh

0 ..... ymmmumae
0 2 3 4 5
267
2 -

(dimensionless)
B g n i [11111m y in H m im irtn H H H K K K r


m xm m u m trrpttTn- r w im —
Test 24-1204, Qt = 3.56 gpm, Qc = 1.06 scfh

268

(dimensionless)
-i— .— ,— |—,— ,— i— i—,—



1 2 3 4 5
269
2
Test co-0115, Ql = 13.1 gpm, Qc = 2.1 scfh

(dimensionless)
o -hmBBsiS- ««««««((«(<««««•’
2 -
Test co-0118, Ql = 13.1 gpm, Qa = 8.5 scfh


0 UMiiuitiHtttHtiittiiuiiiMMii rmrrr
2
0
0 2 3 4 5
270
2
Test co-0106, QL = 3.45 gpm, Qc = 8.5 scfh

(dimensionless)
0 mn|atTTttttn7[-)-rKTnrv
2
Test co-0109, QL = 3.60 gpm, Qc - 1.1 scfh

o ■wtitwimttKuqmnnmmi
2
0
0 2 3 4 5
271
APPENDIX E: MASS TRANSFER TESTS
272
Top o f Contactor Top of Contactor Top o f Contactor
1-1106 o 2-1030 3-1027 o

18
Ql = 13.1 gpm Ql = 13 gpm Ql = 13 gpm
Qa = 11.22 scfh O Qc = 5.53 scfh O Qc = 2.64 scfh O
16 " Kl = 0.032 ft/min Kl =0.034 ft/min Kl =0.036 ft/min
dB =0.30 cm ° dB =0.27 cm ° dB =0.25 cm O
14 _ E. = 52.5 cm2/s E, =33.1cm2/s E. = 23.0 cm2/s
O L O O
Water Column Height (ft)
12 ,: O
10
Q
4 Q
Oi
O
2
1
0
0 1I IH | 1.1 1 1 ^ 1 I t t ^ I I I I ^ I I I ■ ^ . r y J l i n i r ^ i i i ^ i i y n
1.0 1.5 2.0 2.5 3.0 3.5 2.5 3.0 3.5 4.0 4.5 3.5 4.0 4.5 5.0 5.5
Dissolved Oxygen Conc. (mg/L)
assume 5 ftof high-mixing zone at the bottom of contactor

fitted with uniform axial dispersion
273
Top o f Contactor Top o f Contactor T o p o f Contactor
4-1118 o _ 5-1104 o 6-1101 o

18
Ql = 13.1 gpm Ql = 13 gpm
O Qc = 11.22 scfh O Qg = 5.53 scfh O

16 Kl = 0.033 ft/min Kl = 0.035 ft/min
o dB =0.28 cm ° dB = 0.26 cm O
14 E. = 52.5 cm2/s E, =33.1 cm2/s
o L O O
(ft)
12 j :O / /°
$
Water Column Height
//o b
10
//° 6
8
f
f
lb jiO
4 0
t >3gpm
Q l=
01
d
■&
1
d
r-pO
o
/
P
OI Q = 23.17 scfh o/ O':

a /
Oi oi
0 Kl = 0.030 ft/min
2
q! dB = 0.36 cm °7
o 1 °
cj El = 9 1.1 c m 2/s
9 .1....................... ° lJj 11 1_t i l l MM t i l l >11
LI
0
0 .0 0.5 1.0 1.5 2 .0 2 .5 .0 1.5 2.0 2.5 3.0 3 .5 2 .5 3 .0 3 .5 4 .0 4.5
Dissolved Oxygen Cone. (mg/L)
assume 5 ft o f high-mixing zone at the bottom o f contactor

274
Top o f Contactor Top o f Contactor Top o f Contactor
8-1115 O 9-1112
18
Q l = 13 gpm Q l = 13 gpm
Q c = 23.17 scfh O Q c = 11.15 scfh
16 K l = 0 .0 30 ft/min K l = 0.033 ft/min
d B = 0 .3 7 cm O d B - 0.29 cm C
14 E. = 91.1cm 2/s E , = 52.3 cm2/s

O O
(ft)
12
Waler Column Height
10
4
\
0 1 1 1 l i . ,1 L J.il
0.0 0.5 1.0 1.5 2.0 2.0.0 0.5 1.0 1.5 2.0 2.5.0 1.5 2.0 2.5 3.0 3.5
assume 5 ft o f high-mixing zone at the bottom of contactor

—— fitted with uniform axial dispersion
275
Top o f Contactor_________ Top o f Contactor__________ Top o f Contactor
10-1012 c, 11-1007 o 12-1128 o

18
fit= 7.1 gpm Ql = 7 gpm f i t =7.1 gpm
Qa = 5.53 scfh O Qc =2.68 scfh O Qc= 1.33 scfh O
16 Kl =0.035 ft/min K l = 0.035 ft/min K l = 0.037 ft/min
dB = 0.26 cm ° dB =0.26 cm O dB = 0.23 cm O
14 E, = 33.1 cm2/s El = 23.1 cm2/s E, =18.2cm2/s
L O O O
(ft)
12 P
1? 1 f
Water Column Height
10 r P f
r°
8 °
f
P
° f °
6 ih
if P
1: r
Oi o ol
4 Oi 67 a
Oi ol ol
0 P! ol
2 G o G
4 of d
oi d
0 i4 t i i
:: j.. i l i i ii i i i — 1— C _1_J---- i 1 _ L 1 J _ l _ —I_ .L _ i _ i , i _ J ------- t i l t -------- t i l l -------- l—l—L
2 3 4 53 4 5 6 4 5 6 7
assume 5 ft of high-mixing zone at the bottom of contactor

276
18 -
13-1020 o_ 14-1015 o 15-1010 o_
Ql = 1.\ gpm Ql= 7.1 gpm Ql= 7.05 gpm
Qc = l 1.15 scfh O Qc = 5.53 scfh O g c = 2.71 scfh O
16 - Kl = 0.030 ft/min Kl = 0.034 ft/min Kl - 0.036 ft/min
dB = 0.36 cm O dB = 0 .2 6 cm O dB = 0.23 cm O
14 E. = 52.3 cm2/s E. = 3 3 .1 c m 2/s E, = 23.2 cm2/s

O O O
(ft)
12 9/
,/°
1:
If
Water Column Height
10 -
1 ° J l If
Ip
Ip
8- f
| llo
f li
6-
1lb
: 1°
f
of: rb of
ol o
ol Ol 0
Ol QI 0
Ol Ol
q
2-
a
q
j_i—i_i—i—i—i. i i i i i i i i q
7
- t i l l ----t i l l
?
O
1 i 1 1----1—l—L 1qHI J-------------- 1 « « 1----1 1 t
42 5 3 4

277
16-10253 17-1022 o 18-1017 c

18
Q l = 7A gpm QL = 7 gpm
O Qc = 11.28 scfh O Qc - 5.53 scfh O
16 - Kl = 0 .0 3 2 ft/min Kl = 0 .0 3 4 ft/min
o dB = 0.32 cm O dB = 0.27 cm O
14 E, = 52.7 cm2/s E. = 33.1 cm2/s
o O O
(ft)
12 /P 9/
I
Waier Column Height
lb
10 f r
p b P
c
8
9 1 1°
6 9 | io
© of W
9 0 t = 7.15gpm
4 oMlI vy —
q Q c = 23.20 scfh 01 ol
9
Cf) K . = 0 .0 2 9 ft/min
L 9
2 cl
j; dg = 0.39 cm
d 1
I E, = 9 1 .6 cm2/s J
0
9 •°i• I ........ . 1111 .......................
1 2 3 4 1
Dissolved Oxygen Cone. (m g/L)
assume 5 ft of high-mixing zone at the bottom o f contactor

278
Top o f Contactor T o p o f Contactor Top o f Contactor
19-0927 20-1201 21-1210

18 -
Q l = 3.5 gpm Q , = 3.5 gpm
Q g = 2.73 scfh g „ = 1.38 scfh
16 - K l = 0.036 ft/min K l = 0 .0 3 6 ft/min
d B = 0.25 cm d g = 0.23 cm
14 - E l = 23.3 cm*/s E. = 18.4 cm Is

O
(ft)
12
Water Column Height
10 -
8-
6-
4 - Q l = 3 7 8Pm
Qc = 0.67 scfh
K l =0 .0 39 ft/min
2- d g = 0.22 cm
E l = 15.8 cm /s
-i r
5 3 4 5 6 2
assume 5 ft of high-mixing zone at the bottom of contactor

- fitted with uniform axial dispersion
279
Top o f Contactor T o p o f Contactor Top o f Contactor
22-0930 o_ 23-0925 c 24-1204 o

Q l = 3.6 gpm Q l = 3.5 gpm = 3.56 gpm
Q g = 5.50 scfh O Qc = 2.64 scfh O Qc = 1.33 scfh O
" K l = 0.036 ft/min K l = 0.033 ft/min = 0.038 ft/min
dB = 0.26 cm O d B = 0.31 cm O dB =0 .2 2 cm O
_ E. = 33.0 cm!/s E, = 23.0 cm2/s El - 18.2 cm2/s
O O O
(ft)
Water Column Height
I:
r f '
f / : |
- 4
//o //
//
o
rl i ° -
9 1 °
Or
O' Of-
-?'■ 0 / f
a ol
9 Q/
0 Ql
1 o°
1
t> °1
q ........................................
i i i ! i i i i
1 2 3 4 21 3 4 5 3 4 5 6
......... assume 5 ft of high-mixing zone at the bottom of contactor

280
Top o f Contactor_________ Top o f Contactor_________ Top o f Contactor
25-1004 o 26-0814 o 27-0819 c

18
Q l = 3.5 gpm Q l = 3.5 gpm
O Q g = 5.49 scfh O Qc = 2.67 scfh O
16 K l = 0.036 ft/min K l = 0.037 ft/min
o dB = 0.25 cm O dB = 0 .2 4 cm O
14 E, = 33.0 cm2/s El = 23.1 cm2/s

o O O
s 12 if
JeZo
’o
= 10
O
U
k.
O
3
8
c
b //1 o ° -
6 - <p
r
d o f
a Q l = 3.5 gpm o
4 6
Q = 11-23 scfh
q g
o
0 - 0.033 ft/min C)
dB = 0.31 cm Q
2 - 9
9 E, = 52.5 cm7s
9 0
A ...........................................................
0 1 1 1 ! 1 1 1 1 1 1 1
0 1 2 3 4 1 2 3 4 2 3 4 5

- Fitted with uniform axial dispersion
281
APPENDIX F: OZONE REACTION TESTS
282
O Z O N E S E L F -D E C O M P O S IT IO N
T e s t: 1-1106 O 3 D e c a y R a te C o n st, k* ( m i n 1): -0.019422 f i t 1s t s t a g e
In d ig o , Am . 2.71 In te rc e p t: 0.562 (p p m ): 0 .4 5 k o o (m in '1): -20.000 X y
T im e A b so rb an c e Sam p + O j ln(O j) 0 0 .7 8 9 7
0.9
(m in ) @ 600 ru n C o n e . (m g /L (m g/L ) 0 .3 0 .5 6 1 8
<►
0 0.8250 2.2027 0.789664795 0 .5 0 .5 6 1 8
0.7
0.5 0.923 1.795324675 0.585185883 £ 0 .8 0 .5 6 1 8
1 0.937 1.737142857 0.552241727 1 0 .5 6 1 8
1.5 0.947 1.695584416 0.528027469 8 OS 1 .3 0 .5 6 1 8
0
u S a - V -
2 0.951 1.678961039 0.518175173 o’ 1 .5 0 .5 6 1 8
3 0.961 1.637402597 0.493111204 S 0J 2 0 .5 6 1 8
4 0.964 1.624935065 0.485467855 3 0 .5 6 1 8
♦
5 0.979 1.562597403 0.446349438 0.1 5 0 .5 6 1 8
8 0.981 1.554285714 0.441016092 0 10 20 30 7 .5 0 .5 6 1 8
Tune (mini
20 1.071 1.18025974 0.165734533 10 0 .5 6 1 8
S m p /D I (m L; 0 3 (m L ) In d ig o (m L ) A bs(DD A b s(S m p ) A b s(S m p ) A bs(D I) O D (m g /L )
30 30 20 0.206 0.282 0 .2 8 4 0.198 0.44883 2
T e s t: 2-1030 O j D e c a y R a te C o n st, k j (min*1): -0.0154934 f i t 1 :s t s t a g e

In d ig o , Aooo: 2.85 In te rc e p t: 0.881 Du, (p p m ): 0_37 ko o (min*1) : -20.000 X y
T im e A b so rb a n c e S am p + O j ln(O j) 0 1 .0 2 5 2
(m in ) @ 600 run C o n e . (m g/L ) (m g/L ) 0 .3 0 .8 8 1 2
0 0.7542 2.7877 1.025220524 i ---------------------------------------------------- 0 .5 0 .8 8 1 2
0.5 0.779 2.684675325 0.987559799 S t _______________________________ 0 .8 0 .8 8 1 2
5 08
1.5 0.844 2.414545455 0.881511052 ♦ 1 0 .8 8 1 2
3 0.847 2.402077922 0.876334164 i I0 0.6 1 .3 0 .8 8 1 2
u
5 0.879 2.269090909 0.819379271 6 0.4 ------------------------- . _ 1 .5 0 .8 8 1 2
10 0.938 2.023896104 0.705024418 e 2 0 .8 8 1 2
0.2
15 0.982 1.841038961 0.610330065 3 0 .8 8 1 2
29 1.06 1.516883117 0.416657649 --------------------------------------------------* _ 5 0 .8 8 1 2
45 1.126 1.242597403 0.217203868 0 20 40 60 7 .5 0 .8 8 1 2
■une iminj
56 1.179 1.022337662 0.022091831 10 0 .8 8 1 2
S m p /D I (m L ) O 3 (m L ) In d ig o (m L ) A bs(D I) A b s(S m p ) A b s(S m p ) A bs(D I) O D (m g /L )
30 30 20 0.315 0.388 0.38 0.318 0.374026
T e s t: 3-1027 O 3 D e ca y R a te C o n st, k^ (min*1): -0.0271394 f i t 1 s t s t a g e

In d ig o , A mm: 2.71 In te rc e p t: 0.758 Du, (p p m ): 0.88 k o o (min*1): -20.000 X y
T im e A b so rb an c e S am p + O j In(O j) 0 1 .1 0 3 9
(m in ) @ 600 a m C o n e. (m g/L ) (m g/L ) 0 .3 0 .7 5 8 4
1
0 0.6293 3.0158 1.103876539 0 .5 0 .7 5 8 4
0.5 0.839 2.144415584 0.76286706 I- - - - .. . ___________ 0 .8 0 .7 5 8 4
"g0.8
1 0.842 2.131948052 0.75703614 1 0 .7 5 8 4
1.5 0.858 2.065454545 0.725350321 I - V ....... ......................... 1 .3 0 .7 5 8 4
♦
2 0.875 1.994805195 0.690546399 1 .5 0 .7 5 8 4
6
3 0.891 1.928311688 0.656644847 £0.4 2 0 .7 5 8 4
4 0.897 1.903376623 0.643629479 3 0 .7 5 8 4
5 0.913 1.836883117 0.608070177 0.2 ------------------------------------- * ----------------- 5 0 .7 5 8 4
8 0.932 1.757922078 0.564132474 C 10 20 30 7 .5 0 .7 5 8 4
1mac (man)
20 1.057 1.238441558 0.213853781 10 0 .7 5 8 4
S m p /D I (m L ) O , (m L ) In d ig o (m L ) Abs(D I) A b s(S m p ) A b s(S m p ) AbsflDI) O D (mg/1 -)
30 30 20 0.32 0.477 0.468 0.307 0.881039
283
T est: 4-1118 O j D e ca y R a te C o n st. It* (min*1): -0.0384726 f i t 1 s t s ta g e
In d ig o , A**,: 3.63 In te rc e p t: 0.865 (p p m ): 0 3 8 koo (m in '1] : - 20.000 X y
T im e A b s o rb a n c e S a m p -<-03 ln( 0 3) 0 1.0134
(m in) @ 600 nm C o n e . (m g /L (m g/L ) 0 .3 0.8651
0 1.1521 2.7548 1.013361174 *
0 .5 0.8651
0 .5
0.5 1.232 2.422857143 0.884947481 * — 0.8 0.8651
1.5 1.243 2.377142857 0.865899286 ♦ 4 1 0.8651
2 0.4
3 1.308 2.107012987 0.745271299 O
O ♦
1.3 0.8651
5 1.377 1.82025974 0.598979205 0 * 1 .5 0.8651
10 1.412 1.674805195 0.515696857 e 0 2 0.8651
♦
15 1.502 1.300779221 0.262963485 3 0.8651
20 1.549 1.105454545 0.100256604 -0.4 5 0.8651
25 1.6 0.893506494 -0.11260168 0 10 20 30 7 .5 0.8651
tune (nun)
30 1.631 0.764675325 -0.26830395 10 0.8651
<
i
S m p /D I (mL] 0 3 (m L ) In d ig o (m L ) A bs(Sm p) A bs(D I) O D (m g/L )
a
A bs(D D
30 30 20 0.262 0.332 0.332 0.265 0.3795671
T est: 5-1104 O j D ecay R a te C o n st, kd (m in'*): -0.028879 f it 1 s t s ta g e

In d ig o , A«o: 2.71 In te rc e p t: 0.952 Do, (p p m ): 0.44 k o o ( m i n 1) - 20.000 X y
T im e A b s o rb a n c e Sam p + O s ln (O j) 0 1 .1 0 8 8
(m in) @ 600 nm C o n e . (m g/L ) (m g/L ) 0 .3 0 .9 5 1 7
0 0.6258 3.0306 1.108750033 0 .5 0 .9 5 1 7
0.5 0 .747 2.526753247 0.926935177 f 0 .8 0 .9 5 1 7
I 0.738 2.564155844 0.941629319 1 0 .9 5 1 7
8 ♦ 1.3 0 .9 5 1 7
1.5 0.755 2.493506494 0.91368995
V0
2 0 .774 2.414545455 0.881511052 o ' 0.6 1.5 0 .9 5 1 7
3 0 .785 2.368831169 0.862396656 c 2 0 .9 5 1 7
♦
4 0 .792 2.33974026 0.850039923 3 0 .9 5 1 7
5 0.82 2.223376623 0.799027042 0.2 5 0 .9 5 1 7
8 0.872 2.007272727 0.696776949 0 10 20 30 7 .5 0 .9 5 1 7
20 1.002 1.467012987 0.383228352 10 0 .9 5 1 7
S m p /D I (m L) O 3 (m L ) In d ig o (m L ) A bs(D I) A b s(S m p ) A bs(Sm p) A bs(D I) O D (m g/L )
30 30 20 0.348 0.428 0.432 0.353 0.4405195
T est: 6-1101 O j D ecay R a te C o n st. k« ( m i n 1): -0.0153311 f it 1 s t s ta g e

In d ig o , Amo: 3.63 In te rc e p t: 0.881 Do, (p p m ): 0 3 7 koo ( m i n 1): • 20.000 X y
T im e A b s o rb a n c e Sam p + O 3 ln (O j) 0 1.0222
(m in) @ 60 0 n m C o n e . (m g/L ) (m g/L ) 0 .3 0 .8 8 1 2
0 1.1462 2.7794 1.022249612 0 .5 0 .8 8 1 2
0.5 1.205 2.535064935 0.930219252 £ *' 0 .8 0 .8 8 1 2
rw
1.5 1.214 2.497662338 0.915355229 1 0 .8 8 1 2
3 1.249 2.352207792 0.855354373 20.8 1.3 0 .8 8 1 2
0 *
0
5 1.304 2.123636364 0.753129885 0
♦ 1.5 0 .8 8 1 2
10 1.332 2.007272727 0.696776949 ’cO .6 - - --------------- * 4^
------------- 2 0 .8 8 1 2
15 1.368 1.857662338 0.61931889 ♦ 3 0 .8 8 1 2
<
20 1.378 1.816103896 0.59669349 0.4 5 0 .8 8 1 2
25 1.42 1.641558442 0.49564606 0 10 20 30 7 .5 0 .8 8 1 2
30 1.442 1.55012987 0.438338715 10 0 .8 8 1 2
S m p /D I (m L) O 3 (m L ) In d ig o (m L ) A bs(D D A bs(S inp) A bs(Sm p) A bs(D I) O D (m g/L )
30 30 20 0.337 0.408 0.398 0.337 0.3657143
284
O Z O N E S E L F -D E C O M P O S IT IO N |
T e s t; O j D ecay R a te C o n st, k* (min*1): #V A L U E ! fit 1 s t s ta g e
In d ig o , Atm: OD (m ^ L ): D„ (ppm ): k o o (m in '1): X y
T im e A b s o rb a n c e Sam p + O j In (O j) 0 #DIV/0!
(m in ) @ 60 0 ru n C o n e . (m g /L (m g/L ) 0 .3 #DIV /0!
0 .0 0 0 #NU M ! 0 .5 #DIV /0!
1
0 #N U M ! £ 0 .8 #DIV /0!
0 #NU M ! 1 #DIV/0!
0 #NU M ! ie o 8 1 .3 #DIV /0!
U
0 #N U M ! 6 1 .5 #DIV /0I
0 #NU M ! c 06 2 # 0 IV /0 !
0 #N U M ! 3 #DIV /0!
0 #NU M ! 04 5 #DIV /0!
0 #NU M ! 0 10 20 30 7 .5 #DIV /0!
Tine (min)
0 #N U M ! 10 #DIV/0!
S m p /D I (mL; O j (m L ) In d ig o (m L ) A bs(D I) A bs(Sm p) A b s(S m p ) A b s(D I) O D (m g/L )
60 15 25 0
T e s t: 8 -1 115 O j D ecay R a te C o n st, kd ( m in '1): -0.0271965 f i t 1 s t s t a g e

In d ig o , A mo: 2.71 I n te r c e p t: 0.901 Du, (ppm ): 0 J 8 k o o ( m in '1) : -20.000 X y
T im e A b s o rb a n c e Sam p + O j ln (O j) 0 1 .0 4 4
(m in ) @ 600 n m C o n e . (m g /L ) (m g/L ) 0 .3 0 .9 0 0 5
1.1
0 0.6715 2.8405 1.043987118 * 0 .5 0 .9 0 0 5
0.5 0.76 2.472727273 0.905321701 f 0.9 0 .8 0 .9 0 0 5
1 0.769 2.435324675 0.890080085 s 1 0 .9 0 0 5
1.5 0.781 2.385454545 0.869389691 ♦ 1 .3 0 .9 0 0 5
u1 07
2 0.8 2.306493506 0.835728409 6 1 .5 0 .9 0 0 5
3 0.808 2.273246753 0.821209097 2 0j 2 0 .9 0 0 5
4 0.825 2.202597403 0.789637302 3 0 .9 0 0 5
♦
5 0.851 2.094545455 0.739336563 0.3 5 0 .9 0 0 5
8 0.876 1.990649351 0.688460892 0 10 20 30 7 .5 0 .9 0 0 5
20 1.01 1.433766234 0.360304712 10 0 .9 0 0 5
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D Q A bs(Sm p) A b s(S m p ) A bs(D D O D (m g /L )
30 30 20 0.244 0.31 0.321 0.25 0.3795671
T e s t: 9-1112 O j D ecay R a te C o n st, kd ( m in '1): -0.0695858 f i t 1 s t s t a g e

In d ig o , A mo: 2.71 I n te r c e p t: 0.892 Du, (ppm ): 0 .3 9 koo ( m in '1) : -20.000 X y
T im e A b so rb a n c e Sam p + O j ln (O j) 0 1 .0 3 8 2
(m in ) @ 600 n m C o n e. (m g /L ) (m g/L ) 0 .3 0 .8 9 1 6
0 0.6755 2.8241 1.038185049 0 .5 0 .8 9 1 6
0.5 0.775 2.41038961 0.879788398 0 .8 0 .8 9 1 6
! "
1 0.799 2.310649351 0.837528589 v . 1 0 .8 9 1 6
1.5 0.835 2.161038961 0.770589107 ?0j ■ 1 .3 0 .8 9 1 6
<3 ♦
2 0.86 2.057142857 0.721318058 o' 1 .5 0 .8 9 1 6
3 0.876 1.9 9 0 6 4 9 3 5 1 0.688460892 £ 0 - 2 0 .8 9 1 6
4
5
8
0.892
0.952
1.012
1.924155844 0.654487349
1.674805195 0.515696857
1.425454545 0.354490742
■0J-■ 0
------------------------------------A-----------------
10 20 3(
3
5
7 .5
0 .8 9 1 6
0 .8 9 1 6
0 .8 9 1 6
Time (min)
20 1.21 0.602597403 -0.50650596 10 0 .8 9 1 6
S m p /D I (m L ) O j (m L ) In d ig o (m L ) A bs(D D A bs(Sm p) A bsfS m p) A bs(D D O D (mg/1 -)
30 30 20 0.23 0.302 0.309 0 .242 0 .385108 2
285
T e s t; 10-1012 O j D e c a y R a te C o n st, k* (m in '1): -0.0175609 f it 1 s t s ta g e
In d ig o , A mo'. 2.74 In te r c e p t: 0.890 D I , (p p m ) : 0 .4 0 k o o (m in '1) : -20.000 X y
T im e A b s o rb a n c e Sam p + O j ln (O j) 0 1 .0 4 3 8
(m in) @ 600 n m C o n e . (m g /L (m g/L ) <* 0 .3 0 .8 9 0 2
0 0.6866 2.8401 1.043823232 I 0 .5 0 .8 9 0 2
\
0.5 0.775 2.472727273 0.905321701 ^ __________________ __________ 0 .8 0 .8 9 0 2
IBf ou 9
I 0.78 2.451948052 0.896882832 1 0 .8 9 0 2
1.5 0.782 2.443636364 0.893487243 g o8 1 .3 0 .8 9 0 2
e
U ♦ ♦
2 0.799 2.372987013 0.864149505 A 07 1 .5 0 .8 9 0 2
3 0.815 2.306493506 0.835728409 e 2 0 .8 9 0 2
0.6
4 0.835 2.223376623 0.799027042 3 0 .8 9 0 2
♦
5 0.862 2.111168831 0.747241743 0.5 5 0 .8 9 0 2
8 0.86 2.119480519 0.751171021 0 10 20 30 7 .5 0 .8 9 0 2
Time (min)
20 0.952 1.737142857 0.552241727 10 0 .8 9 0 2
S m p /D I (mL] 0 3 (m L ) In d ig o (m L ) A bs/D I) A bsCSm p) A bs(S m p) A bs(D D O D (m g /L )
30 30 20 0.414 0 .4 8 6 0.498 0.424 0.40450: 12
T e s t: 11-1007 0 3 D e c a y R a te C o n st, kd (m in*1): -0.0164349 f it 1 s t s ta g e

In d ig o , Afoo: 3.12 In te rc e p t: 1.094 Dtm ( p p m ): 0.67 k o o (m in -1) : -20.000 X y
T im e A b s o rb a n c e Sam p + O j In(O j) 0 1 .2 9 5 4
(m in) @ 600 ru n C o n e . (m g /L ) (m g/L ) 0 .3 1 .0 9 3 5
0 0.6811 3.6525 1.295421003 0 .5 1 .0 9 3 5
0.5 0.839 2.996363636 1.097399432 0 .8 1 .0 9 3 5
* ...
1 0.842 2.983896104 1.093229864 1 1 .0 9 3 5
1.5 0.858 2.917402597 1.070693699 8 1 .3 1 .0 9 3 5
o *
2 0.866 2.884155844 1.059232255 S 0.9 - - 1 .5 1 .0 9 3 5
3 0.881 2.821818182 1.037381423 e 2 1 .0 9 3 5
4 0.89 2.784415584 1.024038007 ♦ 3 1 .0 9 3 5
5 0.903 2.73038961 1.004444313 0.7 5 1 .0 9 3 5
8 0.932 2.60987013 0.959300461 0 10 20 30 7 .5 1 .0 9 3 5
i m e (mini
20 1.041 2.156883117 0.768664178 10 1 .0 9 3 5
S m p /D I (m L ) O j (m L ) In d ig o (m L ) Abs(DD A b s(S m p ) A bs(S m p) A b s(D I) O D (m g /L )
30 30 20 0.146 0.2 8 0.299 0.192 0.6677056
T e s t: 12-10 0 6 (1 1 2 8 ) O j D e c a y R a te C o n st, kd ( m in '1): -0.0242904 fit 1 s t s ta g e

In d ig o , A**,: 3.63 In te rc e p t: 0.714 (p p m ): 0.89 k o o (m in ‘): -20.000 X y
T im e A b s o rb a n c e Sam p + O j ln (O j) 0 1 .0 7 4 4
(rain) @ 600 n m C o n e . (m g /L ) (m g/L ) 0 .3 0 .7 1 3 7
0 1.1104 2.9281 1.074352301 0.9 0 .5 0 .7 1 3 7
0.5 1.278 2.231688312 0.802758389 i , 0 .8 0 .7 1 3 7
1.5 1.335 1.994805195 0.690546399 1 0 .7 1 3 7
3 1.342 1.965714286 0.675855683 I 0-5
1 .3 0 .7 1 3 7
e
U ♦ ♦
5 1.367 1.861818182 0.621553527 $ 0 3 - 1 .5 0 .7 1 3 7
10 1.455 1.496103896 0.402864326 e ♦ 2 0 .7 1 3 7
~ 0.1 ------------------------------------------- ,
15 1.467 1.446233766 0.368962775 3 0 .7 1 3 7
20 1.533 1.171948052 0.158667366 -0.1 - 5 0 .7 1 3 7
25 1.563 1.047272727 0.046189382 0 10 20 3( 7 .5 0 .7 1 3 7
Time (min)
30 1.553 1.088831169 0.085104799 10 0 .7 1 3 7
S m p /D I (m L ) O j (m L ) In d ig o (m L ) A bs(DD A b s(S m p ) A b s(S m p ) A bs(D D O D (mg/1 .)
30 30 20 0.139 0.311 0.294 0.146 0.886580 1
286
T e s t: 13-1020 O j D e c a y R a te C onst. k* (m in '1): -0.0298323 f i t 1 s t s ta g e
In d ig o , A mo: 3.63 In te rc e p t: 0.847 D «, (p p m ): 0.47 koo (m in ' 1) : -20.000 X y
T im e A b s o rb a n c e S a ra p -f O j ln( 0 ,) 0 1.03
(m in ) @ 600 n m Cone. (m g/L ' (m g/L ) 0 .3 0 .8 4 7 1
0 1.1410 2.8011 1.030008765 l 1 ►-------- - - - -------- 0 .5 0 .8 4 7 1
0.5 1.178 2.647272727 0.973529951 w 0.8 0 .8 4 7 1
*B u08
1.5 1.253 2.335584416 0.848262145 ♦ ♦ 1 0 .8 4 7 1
3 1.303 2.127792208 0.75508492 2 06
0 1 .3 0 .8 4 7 1
0 ♦
5 1.315 2.077922078 0.731368393 6 04 1 .5 0 .8 4 7 1
10 1.415 1.662337662 0 .508224842 e ♦ 2 0 .8 4 7 1
15
02
1.486 1.367272727 0.312818046 ♦ 3 0 .8 4 7 1
20 1.513 1.255064935 0.227187312 0 5 0 .8 4 7 1
25 1.534 1.167792208 0 .155114964 0 10 20 30 7 .5 0 .8 4 7 1
l in e (nun)
30 1.581 0.972467532 -0.02791859 10 0 .8 4 7 1
S m p /D I (m L ) O j (m L ) Indigo (m L ) A bs(D D A b s(S m p ) A bs(Sm p) A bs(D D O D (m g /L )
30 30 20 0.279 0.362 0.366 0.28 0 .4 6 8 2 2 >1
T e s t: 14-1015 0 3 D e ca y R a te C onst, (m in '1): -0.0287756 f it 1 s t s ta g e

In d ig o , Aftoo: 3.12 In te rc e p t: 0.885 Z) (p p m ): 0.31 Icqd ( m i n 1): -20.000 X y
T im e A b s o rb a n c e S am p + O j In(O j) 0 1 .0 0 5 7
(m in) @ 600 n m C one. (m g/L ) (m g/L ) 0 .3 0 .8 8 5 2
0 0.9022 2.7338 1.005694563 _ I >---------------------------------------------------- 0 .5 0 .8 8 5 2
0.5 0.966 2.468571429 0.903639614 *B 0.8 -4 ♦ 4F 0.8 0 .8 8 5 2
1.5 1.021 2.24 0.806475866 1 0 .8 8 5 2
3 1.033 2.19012987 0.783960843
8 0.6 1 .3 0 .8 8 5 2
O
O ♦
5 1.041 2.156883117 0.768664178 6* 0.4 1 .5 0 .8 8 5 2
10 1.121 1.824415584 0.601259708 c ♦ 2 0 .8 8 5 2
0.2
15 1.164 1.645714286 0.498174506 3 0 .8 8 5 2
20 1.238 1.338181818 0.291311841 0 5 0 .8 8 5 2
25 1.266 1.221818182 0.200340062 0 10 20 30 7 .5 0 .8 8 5 2
30 1.324 0.980779221 -0.0194079 10 0 .8 8 5 2
S m p /D I (m L ) O j (m L ) Indigo (m L ) A bs(D D A b s(S m p ) A bs(Sm p) A bs(D D O D (m g/L )
30 30 20 0.026 0.087 0.084 0.033 0.310303
T e s t: 15-1010 O j D e c a y R a te C onst. k d (m in '1): -0.0344461 f it 1 s t s ta g e

In d ig o , Ata,: 3.63 In te rc e p t: 1.039 Du, (p p m ): 0.61 koo (m in '1): -20.000 X y
T im e A b s o rb a n c e S am p + O j ta (O j) 0 1 .2 3 4 7
(m in ) @ 600 nm C one. (m g/L ) (m g/L ) < 0 .3 1 .0 3 9 5
0 0.9879 3.4373 1.234680941
1.1 0 .5 1 .0 3 9 5
0.5 1.16 2.722077922 1.001395531 io .9 ________________________________ 0.8 1 .0 3 9 5
E ♦
1.5 1.185 2.618181818 0 .962480114 ~ 0.7 ♦ 1 1 .0 3 9 5
3 1.205 2.535064935 0.930219252 8 ♦ 1 .3 1 .0 3 9 5
^ 0.5
5 1.257 2.318961039 0 .8 4 1 1 19257 ♦ 1 .5 1 .0 3 9 5
- 0.3
10 1.308 2.107012987 0.745271299 c 2 1 .0 3 9 5
♦
15 1.399 1.728831169 0.547445555 0.1 3 1 .0 3 9 5
<
20 1.465 1.454545455 0.374693449 -0.1 ■ 5 1 .0 3 9 5
25 1.532 1.176103896 0.162207193 0 10 20 31) 7 .5 1 .0 3 9 5
Time (min)
30 1.579 0.980779221 -0.0194079 10 1 .0 3 9 5
S m p /D I (m L ) O 3 (m L ) Indigo (m L ) A bs(D I) A b s(S m p ) A bs(Sm p) A bs(D I) O D (mg/1 D
30 30 20 0.402 0 .514 0.505 0.397 0.609523 8
287
O Z O N E S E L F -D E C O M P O S T T IO N
T e s t: 16-1025 O j D ecay R a te C o n st, k^ (m in '1): -0.0154518 fit 1st stage
In d ig o , Amo | 2.85 I n te r c e p t: 1.172 O t . (ppm ) 0.41 k o o ( m i n 1): -20.000 X y
T im e A b s o rb a n c e Sam p + O j In(O j) 0 1.292
(m in) @ 600 n m C o n e . (m g /L (m g/L ) 0 .3 1.1716
i►
0 0.5491 3.6400 1.291995765 0.5 1.1716
0.5 0.645 3.241558442 1.176054215
i-i L 0 .8 1.1716
♦ ♦
1.5 0.692 3.046233766 1.113905997 1 1.1716
8 ---------- ---------------
3 0.662 3.170909091 1.154018326 e ♦ ------------- 1.3 1.1716
U
5 0.694 3.037922078 1.111173755 A 1.5 1.1716
♦
10 0.775 2.701298701 0.993732658 e 0.8 > 2 1.1716
15 0.784 2.663896104 0.979789752 3 1.1716
♦ 1.1716
20 0.871 2.302337662 0.833924982 0.6 5
25 0.9 2.181818182 0.780158558 G 10 20 30 40 7.5 1.1716
Time (min)
35 0.97 1.890909091 0.637057714 10 1.1716
S m p /D I (mL;
<
I
1
O j (m L ) In d ig o (m L ) A bs(D I) A bs(Sm p) A b s(D D O D (m g/L )
30 30 20 0.297 0.381 0.367 0.3 0 2 0.4128139
T e s t: 17-1022 O j D ecay R a te C o n st. k^ ( m in '1): -0.072395 fit 1st stage

In d ig o , A<ao: 2.85 I n te rc e p t: 1.006 O f (ppm ): 0.47 k o D ( m i n 1) : -20.000 X y
T im e A b so rb a n c e Sam p + O j ta (O j) 0 1.1646
(m in) @ 600 n m C o n e . (m g /L ) (m g/L ) 0 .3 1.0057
0 0.6539 3.2047 1.164630006 * 0 .5 1.0057
0.5 0.784 2.663896104 0.979789752 * ns ♦♦ 0.8 1.0057
1.5 0.831 2.468571429 0.903639614 ♦ 1 1.0057
Io --------------- 4 ----------------- ---------------
3 0.9 2.181818182 0.780158558 1.3 1.0057
u ♦
5 0.975 1.87012987 0.626007878 r? -OS 1.5 1.0057
10 1.126 1.242597403 0.217203868 Jg 2 1.0057
-1 ----- - 4 - - -
15 1.184 1.001558442 0.001557228 3 1.0057
20 1.253 0.714805195 -0.33574523 -1.5 5 1.0057
25 1.328 0.403116883 -0.90852873 c 10 20 30 40 7.5 1.0057
i unc own)
35 1.338 0.361558442 -1.01733159 10 1.0057
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D I) A bs(Sm p) A bs(S m p) A b s(D I) O D (m g/L )
30 30 20 0.287 0.372 0.379 0.2 9 4 0.4709957
T e s t: 18-1017 O j D ecay R a te C o n st, k^ ( m in '1): -0.0497915 fit 1st stage

In d ig o , A ao: 3.63 I n te rc e p t: 0.914 D f (ppm): 0.41 k o o (m in '1): -20.000 X y
T im e A b so rb an c e Sam p + O j ln (O j) 0 1.0676
U
(m in) @ 600 n m C one. (m g /L ) (m g/L ) 0.3 0.9145
0 1.1152 2.9083 1.067553623 - l 1► 0 .5 0.9145
0.5 1.261 2.302337662 0.833924982 £ 0.8 0.9145
1.5 1.282 2.215064935 0.795281719 ~ 0.5 ---------- ------------------ ------------ 1 0.9145
3 1.309 2.102857143 0.743296964 io ♦ 1.3 0.9145
U 0
5 1.347 1.944935065 0.665228591 o* ♦ 1.5 0.9145
10 1.414 1.666493506 0.510721722 ♦- 2 0.9145
~<
15 1.512 1.259220779 0.2304931 3 0.9145
20 1.61 0.851948052 -0.16022973 -1 5 0.9145
25 1.648 0.694025974 -0.36524589 C 10 20 30 7.5 0 .9 1 4 5
i unc immj
30 1.677 0.573506494 -0.55598602 10 0 .9 1 4 5
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D D A bs(Sm p) A bs(S m p) A b s(D D O D (m g /L )
30 30 20 0.285 0.351 0.354 0.271 0.4128139
288
T e s t: 19-0927 O j D e c a y R a te C o n st. It* (m in '1): •0.0121214 f it 1 s t s ta g e
In d ig o , A**: 3.12 In te r c e p t: 0.964 D m (p p m ) : 0.28 k o o (m in '1): - 20.000 X y
T im e A b so rb an c e S a m p -t- O j ln (O j) 0 1 .0 6 5 2
12
(m in) @ 600 n m C o n e . (m g /L (m g/L ) 0 .3 0 .9 6 3 8
0 0.8619 2.9014 1.065184262 <► 0 .5 0 .9 6 3 8
0.5 0.918 2.668051948 0.981348599 £ K --~ ~ 0.8 0 .9 6 3 8
1.5 0.934 2.601558442 0.956110666 ^ ♦ 1 0 .9 6 3 8
3 0.957 2.505974026 0.918677492 8
0
0.8 1 .3 0 .9 6 3 8
u ♦ •
5 0.97 2.451948052 0 .8 % 8 8 2 8 3 2 <5 ♦ 1.5 0 .9 6 3 8
10 0.987 2.381298701 0.867646012 e 0.6 ► 2 0 .9 6 3 8
15 1.061 2.073766234 0.729366391 3 0 .9 6 3 8
20 1.054 2.102857143 0 .7 4 3 2 % 9 6 4 0.4 - 5 0 .9 6 3 8
25 1.079 1.998961039 0.692627565 G 10 20 30 7 .5 0 .9 6 3 8
tune (nun;
30 1.131 1.782857143 0.578217214 10 0 .9 6 3 8
S m p /D I (mL] O j (m L ) In d ig o (m L ) A bs(D D A b s(S m p ) A b s(S m p) A bs(D I) O D (m g /L )
30 30 20 0.113 0 .1 7 0.156 0.112 0.2798268
T e s t: 2 0 -0 9 16(1201) O j D e c a y R a te C o n st, k,, (m in '1): -0.0105457 f it 1 s t s ta g e

In d ig o , A<oo: 3.7 In te r c e p t: 1.056 Dm (p p m ): 0.23 koD (m in '1): -20.000 X y
T im e A b so rb an c e Sam p + O j ln( 0 ,) 0 1 .1 3 3 4
(m in) @ 600 tu n C o n e . (m g /L ) (m g/L ) 0 .3 1 .0 5 6 5
0 1.1026 3.1062 1.133404612 0 .5 1 .0 5 6 5
* -
0.5 1.149 2.913246753 1.069268182 0.8 1 .0 5 6 5
1.5 1.164 2.850909091 1.047637923 1 1 .0 5 6 5
3 1.18 2.784415584 1.024038007
g 0.9 -
0 1 .3 1 .0 5 6 5
u
5 1.193 2.73038% 1 1.004444313 eS ♦ 1.5 1 .0 5 6 5
10 1.243 2.522597403 0.925289086 e 0.75 <► 2 1 .0 5 6 5
15 1.252 2.485194805 0.910351049 3 1 .0 5 6 5
20 1.287 2.33974026 0.850039923 0.6 5 1 .0 5 6 5
25 1.308 2.252467532 0.8 1 2 0 2 6 2 % c 10 20 30 7 .5 1 .0 5 6 5
1m e min)
30 1.354 2.061298701 0.723336222 10 1 .0 5 6 5
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D I) A b s(S m p ) A b s(S m p) A bs(D D O D (m g /L )
30 30 20 0.091 0.1 2 9 0.135 0.09 0.2299567
T e s t: 21 -0 9 14(1210) O j D e c a y R a te C o n st. It* (m in '1): •0.0142802 f i t 1 s t s ta g e

In d ig o , A«o: 3.7 In te r c e p t: 1.061 Dm (p p m ): 0.47 k o o ( m in 1) : -20.000 X y
T im e A b so rb an c e Sam p + O j l n ( 0 ,) 0 1 .2 1 3 2
U
(m in) @ 600 n m C o n e . (m g /L ) (m g/L ) 0 .3 1 .0 6 1 4
0 1.0405
1.2 1 t
3.3641 1.213169335 0 .5 1 .0 6 1 4
0.5 1.152 2.900779221 1.064979398 ■&1.1 i- 0 .8 1 .0 6 1 4
1.5 1.179 2.788571429 1.025529432 Z 1 *♦« 1 1 .0 6 1 4
♦
3 1.17 2.825974026 1.038853093 * 1 .3 1 .0 6 1 4
______ _____________ 4 . . ^ _________________ .
*
5 1.213 2.647272727 0.973529951 1 .5 1 .0 6 1 4
—0.8
10 1.253 2.481038961 0.908677408 s ♦ 2 1 .0 6 1 4
15 1.269 2.414545455 0.881511052 0.7 - ------------------------------------------------------------------
------ 3 1 .0 6 1 4
20 <
1.327 2.173506494 0.776341759 0.6 5 1 .0 6 1 4
25 1.368 2.003116883 0.694704409 10 20 30 i 7 .5 1 .0 6 1 4
30 1.399 1.874285714 0.628227634 10 1 .0 6 1 4
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D D A b s(S m p ) A b s(S m p) A bs(D I) O D (m g /L )
30 30 20 0.256 0.3 4 2 0.321 0.236 0.4737662
289
T e s t; 22-0930 O j D e c a y R a te C o n st, k^ (m in '1): -0.0238113 f i t 1 s t s ta g e
In d ig o , A«o: 3.12 In te rc e p t: 0.641 ( p p m ): 0 3 9 koo (m in * ):-20.000 X y
T im e A b s o rb a n c e S a m p -t- O j In(O j) 0 0 .8 2 7 8
—
(m in) @ 600 nm C o n e . (m g /L (m g/L ) 0 .3 0 .6 4 0 6
) --------------- ------------------------
bo
0
o
1.0094 2.2882 0.827772471 0 .5 0 .6 4 0 6
0.5 1.0181 2.252051948 0.811841778 £ 0 .8 0 .6 4 0 6
o\
o
_E
1.5 1.126 1.803636364 0.589804829 1 0 .6 4 0 6
8o _____ ♦ ........... ..........................
o
3 1.146 1.720519481 0.542626269 ♦ 1 .3 0 .6 4 0 6
U
5 1.163 1.64987013 0.5006% 576 <5 1 .5 0 .6 4 0 6
Kj
o
10 1.189 1.541818182 0.432962358 e 2 0 .6 4 0 6
o
15 1.224 1.396363636 0.333871455 3 0 .6 4 0 6
►
20 1.265 1.225974026 0.203735651 5 0 .6 4 0 6
K»
o
25 1.313 1.026493506 0.026148632 () 10 20 30 7 .5 0 .6 4 0 6
itmetmmj
30 1.345 0.893506494 -0.11260168 10 0 .6 4 0 6
S m p /D I (mL] O j (m L ) In d ig o (m L ) A bs(D I) A b s(S m p ) A b sfS m p ) A bs(D I) O D (m g /L )
30 30 | 20 0.112 0 .1 7 4 0.191 0.112 0.39064! H
T est: 23-0925 O j D e c a y R a te C o n st. Ic* (m in '1): -0.0340669 f i t Is st s ta g e

Indigo, A^ kj: 3.7 In te rc e p t: 0.798 Du, (p p m ): 0.26 k o D (m in '1) : -20.000 X y
T im e A b s o rb a n c e Sam p + O j ln(O j) 0 0 .9 0 9
(m in ) @ 600 n m C o n e . (m g/L ) (m g/L ) 0 .3 0 .7 9 8 2
I
0 1.2528 2.4819 0.90903309 > 0 .5 0 .7 9 8 2
0.5 1.305 2.264935065 0.81754609 £ 0.8 0 .8 0 .7 9 8 2
E
1.5 1.356 2.052987013 0.719295812 ~ 0.6 ^ ___________________________ 1 0 .7 9 8 2
i ♦
3 1.374 1.978181818 0.682178149 1 .3 0 .7 9 8 2
^ 0.4
5 1.404 1.853506494 0.617079247 ♦ 1 .5 0 .7 9 8 2
~ 0.2
10 1.439 1.708051948 0.535353509 n ♦ 2 0 .7 9 8 2
15 1.521 1.367272727 0.312818046 0 3 0 .7 9 8 2
20 1.584 1.105454545 0.100256604 -0.2 ♦ -> 5 0 .7 9 8 2
25 1.637 0.885194805 -0.12194754 0 10 20 30 7 .5 0 .7 9 8 2
Time (min)
30 1.651 0.827012987 -0.18993488 10 0 .7 9 8 2
S m p /D I (m L ) O j (m L ) In d ig o (m L ) A bs(D I) A b s(S m p ) A b s(S m p ) A bs(D I) O D (m g /L )
30 30 20 0.112 0.168 0.177 0.139 0.2604329
T est: 2 4 -0 9 1 8 (1 2 0 4 ) O j D e c a y R a te C o n st, k< (m in '1): -0.0109615 f i t 1 s t s ta g e

Indigo, A «o: 3.7 In te rc e p t: 0.124 Du, (p p m ): 0.33 koD (m in '1): -20.000 X y
T im e A b s o rb a n c e S a m p -1- O j ln(O j) 0 0 .3 8 1 1
0.6
(m in) @ 600 nm C o n e. (m g/L ) (m g/L ) 0 .3 0 .1 2 4 3
0 1.4977 1.4639 0.381135188 0 .5 0 .1 2 3 5
0.5 1.55 1.246753247 0.22054277 0 .8 0 .1 2 3 5
1.5 1.579 1.126233766 0.118879116 ~ 0.2 r 1 0 .1 2 3 5
3 1.581 1.117922078 0.111471675 £e 1 .3 0 .1 2 3 5
U 0
5 1.599 1.043116883 0.042213234 ------------- ^ - V ............................ 1 .5 0 .1 2 3 5
6
10 1.606 1.014025974 0.01392852 ♦ ♦ 2 0 .1 2 3 5
5-0.2
15 1 .6 19 0.96 -0.04082199 3 0 .1 2 3 5
20 1.638 0 .8 8 1 0 3 8 % ! -0.12665343 -0.4 5 0 .1 2 3 5
25 1.64 0.872727273 -0.13613217 0 10 20 3C 7 .5 0 .1 2 3 5
Time (min)
30 1.652 0.822857143 -0.19497267 10 0 .1 2 3 5
S m p /D I (m L ) O j (m L ) In d ig o (m L ) Abs(DD A b s(S m p ) A b sfS m p ) A bs(D R O D (mg/1 .)
30 30 20 0.078 0.135 0.162 0.099 0.332467 5
290
T e s t: 25-1004 O j D ecay R a te C o n st, k* ( m i n 1): -0.0124806 fit 1st stage
In d ig o , Am : 3.7 In te rc e p t: 0.479 £>(. (ppm ): 0.34 koo ( m i n " ) : -20.000 X y
T im e A b so rb an c e Sam p + O j ln (O j) 0 0 .6 6 9 2
(m in) @ 600 n m C o n e. (m g /L (m g/L ) 0 .3 0 .4 7 9 1
0 1.3802 1.9526 0.669156928 0 .5 0 .4 7 9 1
0.5 1.395 1.890909091 0.637057714 * u . 0 .8 0 .4 7 9 1
&
1.5 1.46 1.620779221 0.482907034 1 0 .4 7 9 1
3 1.468 1.587532468 0.462180904 2 c15 1 .3 0 .4 7 9 1
,©
5 1.478 1.545974026 0.435654149 o* 1 .5 0 .4 7 9 1
10 1.525 1.350649351 0.300585477 c C5 2 0 .4 7 9 1
♦ ♦ v
15 1.537 1.300779221 0.262963485 3 0 .4 7 9 1
>
20 1.545 1.267532468 0.237072072 c1 ♦ 5 0 .4 7 9 1
25 1.579 1.126233766 0.118879116 a 10 20 30 7 .5 0 .4 7 9 1
lime (mtn)
30 1.566 1.18025974 0.165734533 10 0 .4 7 9 1
S m p /D I (mL] O , (m L ) In d ig o (m L ) A bs(D D A bs(Sm p) A bs(S m p) A b s(D I) O D (m g/L )
30 30 20 0.121 0.187 0.194 0.138 0.338001 17
T e s t: 2 6-1003(0814) 0 3 D ecay R a te C o n st, k d (min*1): -0.1149698 fit 1jst stage

In d ig o , A«oo: 3.7 In te rc e p t: 0.694 D& (ppm ): 0 3 9 koo ( m in '1): -20.000 X y
T im e A b so rb an c e Sam p + O j in (O j) 0 0 .8 6 9 8
i
(m in) @ 600 n m C o n e. (m g/L ) (m g/L ) <► 0 .3 0 .6 9 3 8
0 1.2758 2.3864 0.86980371 ~ 0.5 ----------------------------------------------- J 0 .5 0 .6 9 3 8
0.5 1.373 1.982337662 0.684276786 £ ♦ ♦ 0 .8 0 .6 9 3 8
e o
1.5 1.45 1.662337662 0.508224842 1 0 .6 9 3 8
8 -0.5 ♦
3 1.527 1.342337662 0.294412618 1 .3 0 .6 9 3 8
u
5 1.572 1.155324675 0.144381409 6 -i 1 .5 0 .6 9 3 8
10 1.687 0.677402597 -0.3894895 e 2 0 .6 9 3 8
♦
-1.5
15 1.762 0.365714286 -1.00590289 3 0 .6 9 3 8
♦ •
20 1.804 0.191168831 -1.65459831 -2 5 0 .6 9 3 8
25 1.788 0.257662338 -1.35610532 0 10 20 30 7 .5 0 .6 9 3 8
Tfanc (min)
30 1.808 0.174545455 -1.74557009 10 0 .6 9 3 8
S m p /D I (m L ) O j (m L ) In d ig o (m L ) A bs(D D A bs(S m p) A bs(S m p) A b s(D I) O D (m g/L )
30 30 20 0.127 0.185 0.203 0.122 0.3851082
T e s t: 27-0929(0819) O j D ecay R a te C o n st, k j (m in " ): -0.0679146 fit 1st stage

In d ig o , Atoo: 2.85 In te rc e p t: 1.163 Du, (p p m ): 0.40 koo ( m in " ): -20.000 X y
15
(m in) @ 600 n m C o n e. (m g/L ) (m g/L ) 0 .3 1 .1 6 2 8
<
0 0.5586 3.6007 1.281134164 0 .5 1 .1 6 2 8
s
0.5 0.686 3.071168831 1.122058216 5
,§
1 ♦ 0 .8 1 .1 6 2 8
1.5 0.723 2.917402597 1.070693699 1 1 .1 6 2 8
♦
3 0.831 2.468571429 0 .9 0 3 6 3 % 14 8 05 1 .3 1 .1 6 2 8
o
u
5 0.894 2.206753247 0.791522316 1 .5 1 .1 6 2 8
6
10 0.98 1.849350649 0.614834577 5 0 ________ - ♦ . _____________ 2 1 .1 6 2 8
16 1.17 1.05974026 0.05802384 ♦ * ♦ 3 1 .1 6 2 8
20 1.234 0.793766234 -0.2 3 0 % 6 2 8 -05 5 1 .1 6 2 8
25 1.214 0.876883117 -0.13138157 ) 10 20 30 40 7 .5 1 .1 6 2 8
1tine (mini
35 1.23 0.81038961 -0.21024015 10 1 .1 6 2 8
S m p /D I (m L ) O j (m L ) In d ig o (m L ) A bs(D D A bs(Sm p) A bs(S m p) A b s(D I) O D (m g/I .)
30 30 20 0.133 0.201 0.206 0.129 0.401731 6
291
O Z O N E S E L F -D E C O M P O S m O N |
T e st: 0106-co O j D ecay R a te C o n st. 1^ (m in*'): -0 .0175195 f i t 1 s t s ta g e
In d ig o , A n ,: 3.62 I n te r c e p t: 1.018 Du, (p p m ) : 0.20 ko o (min*1):-2 0 .0 0 0 X y
T im e A b so rb an c e Sam p + O j ln (O j) 0 1 .0 8 8
i.i, ►
(m in) @ 600 n m C o n e . (m g /L (m g/L ) 0 .3 1 .0 1 8 3
0 1.0957 2.9684 1.088020822 \ ________ 0 .5 1 .0 1 8 3
0.5 1.133 2 .813506494 1.034431568 £ 0 .8 1 .0 1 8 3
1 1.152 2.734545455 1.005965226 ~ 0.9 — A ♦ - 1 1 .0 1 8 3
1.5 1.162 2.692987013 0.990650991 8e 1 .3 1 .0 1 8 3
W US
2 1.164 2.684675325 0.987559799 1 .5 1 .0 1 8 3
3 1.187 2 .589090909 0.951306814 I m 2 1 .0 1 8 3
4 1.192 2.568311688 0.943248752 ♦ 3 1 .0 1 8 3
5 1.214 2.476883117 0.907000962 0.6 5 1 .0 1 8 3
8 1.205 2.514285714 0.921988753 0 10 20 30 7 .5 1 .0 1 8 3
20 1.345 1.932467532 0.658797701 10 1 .0 1 8 3
S m p /D I (mL] O j (m L ) In d ig o (m L ) A bs(D Q A bs(S m p) A bs(Sm p) A bs(D I) O D (m g /L )
60 15 25 0.274 0.328 0.348 0.288 0.197 4 0 2 6
T e s t: 0109-co O j D ecay R a te C o n st, leu (min*1): -0 .0146057 f i t 1 s t s ta g e

In d ig o , Aoo: 3.62 I n te r c e p t: 0 .800 Du, (p p m ): 0_33 k o o (min*1): -20.000 X y
(m in) @ 600 n m C o n e. (m g /L ) (m g /L ) 0 .3 0 .8 0 0 5
0 1.1948 2.5566 0.938691454 _ 0.9 0 .5 0 .8 0 0 5
0.5 1.261 2.281558442 0.824858736 £ 0 .8 0 .8 0 0 5
*
1 1.269 2.248311688 0 .810179574 ~ 08 1 0 .8 0 0 5
1.5 1.278 2.210909091 0.793403784 8© 1 .3 0 .8 0 0 5
♦♦♦
W U./ ----------* -
2 1.294 2.144415584 0.76286706 £ 1 .5 0 .8 0 0 5
3 1.306 2.094545455 0.739336563 s UK 2 0 .8 0 0 5
4 1.312 2.06961039 0.727360372 3 0 .8 0 0 5
5 1.314 2.061298701 0.723336222 OS * 5 0 .8 0 0 5
8 1.333 1.982337662 0.684276786 c 10 20 30 7 .5 0 .8 0 0 5
20 1.408 1.670649351 0.513212384 10 0 .8 0 0 5
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D I) A bs(S m p) A bs(Sm p) A bs(D I) O D (m g /L )
60 15 25 0.246 0.329 0.38 0.272 0.330 7 3 5 9
T e s t: 0115-co O j D ecay R a te C o n st, k^ (min*1): -0.0 1 0 6 8 9 f it 1 s t s ta g e

In d ig o , A«oo: 3.62 In te r c e p t: 0.910 Du, (p p m ): 0.18 koo (m in " ): -2 0 .0 0 0 X y
1
(m in) @ 600 n m C o n e. (m g /L ) (m g /L ) <► 0 .3 0 .9 1 0 2
0 1.1688 2.6647 0.980103756 ♦ 0 .5 0 .9 1 0 2
£0.9
0.5 1.214 2.476883117 0.907000962 0 .8 0 .9 1 0 2
,6
1 1.199 2.539220779 0.931857254 — ♦ 1 0 .9 1 0 2
1.5 1.227 2.422857143 0.884947481 80.8 1 .3 0 .9 1 0 2
e
u
2 1.229 2.414545455 0.881511052 <5 1 .5 0 .9 1 0 2
3 1.238 2.377142857 0.865899286 Is 0.7 --------------- ---------- ------------- 2 0 .9 1 0 2
4 1.244 2.352207792 0.855354373 3 0 .9 1 0 2
5 1.245 2.348051948 0.853586026 0.6 5 0 .9 1 0 2
8 1.255 2.306493506 0.835728409 10 20 30 7 .5 0 .9 1 0 2
20 1.327 2.007272727 0.696776949 10 0 .9 1 0 2
S m p /D I (m L) O j (m L ) In d ig o (m L ) A bs(D R A bs(S m p) A bs(S m p) A bs(D I) O D (m g /L )
60 15 25 0.249 0.314 0.304 0.266 0.178355
292
T e s t: 0118-co O j D e c a y R a te C o n st. kd (m in '1): •0.01895 f i t 1 s t s ta g e
In d ig o , A*,*,: 3.62 In te rc e p t: 0.747 D fe (p p m ) 0.21 koo ( m i n 1):-2 0 .0 0 0 X y
T im e A b s o rb a n c e Sam p + O j In(O j) 0 0 .8 4 1 8
(m in) @ 600 nm C o n e. (mg/L! (m g/L ) 0 .3 0 .7 4 6 9
0 1.2516 2.3205 0.841783251 0.8 - 0 .5 0 .7 4 6 9
0.5 1.3 2 . 119480519 0.751171021
If 0i
V 0 .8 0 .7 4 6 9
1 1.31 2.077922078 0.731368393 1 0 .7 4 6 9
1.5 1.314 2.061298701 0.723336222 I 0.6 -------- ‘ . . a 1 .3 0 .7 4 6 9
u
2 1.333 1.982337662 0.684276786 1 .5 0 .7 4 6 9
3 1.332 1.986493506 0.686371027 e 2 0 .7 4 6 9
0.4
4.5 1.343 1.940779221 0.663089553 3 0 .7 4 6 9
5.5 1.357 1.882597403 0.63265242 0J ► 5 0 .7 4 6 9
9.5 1.367 1.841038961 0 .610330065 0 to 20 30 7 .5 0 .7 4 6 9
Time (min)
21.5 1.477 1.383896104 0.324902785 10 0 .7 4 6 9
S m p /D I (m L ) 0 3 (m L ) In d ig o (m L ) A bs(D D A b s(S m p ) A b s(S m p ) A bs(D D O D (m g /L )
60 15 25 0.26 0.317 0.338 0.273 0 .211 2 5 5 4
T e s t: 1007-cyst 0 3 D e c a y R a te C o n st, kd (m in-1): -0.0107732 f i t 1 s t s ta g e

In d ig o , Amo: 2.76 In te rc e p t: 0.739 Du, (p p m ): 0.42 koo ( m i n 1) : -20.000 X y
T im e A b s o rb a n c e Sam p + O j In (O j) 0 0 .9 2 1 5
(m in) @ 600 nm C o n e . (m g/L ) (m g/L ) 0 .3 0 .7 3 8 6
0 0.7753 2.5130 0.921494712 0.9 A 0 .5 0 .7 3 8 6
0.5 0.871 2.115324675 0.749208312 ^ 08 0 .8 0 .7 3 8 6
1.5 0.864 2.144415584 0 .76286706 1 0 .7 3 8 6
3 8 0.7 -------- - - -
0.902 1.986493506 0 .686371027 0 1 .3 0 .7 3 8 6
u
5 0.9 1.994805195 0.690546399 £ 0.6 _ _ 4 ----------- 1 .5 0 .7 3 8 6
e ♦
10 0.954 1.77038961 0.571199641 ♦ 2 0 .7 3 8 6
0J - -
15 0.942 1.82025974 0.598979205 3 0 .7 3 8 6
20 0 .974 1.687272727 0.523113455 0.4 ---------- < 5 0 .7 3 8 6
25 0.987 1.633246753 0.490569907 0 to 20 30 7 .5 0 .7 3 8 6
lune (mint
30 1.019 1.50025974 0.405638253 10 0 .7 3 8 6
S m p /D I (m L ) O j (m L ) In d ig o (m L) A bs(D I) A b s(S m p ) A b s(S m p ) A bs(D D O D (m g /L )
60 15 25 0.242 0 .364 0.385 0.267 0.41 5 5 8 4 4
293
APPENDIX G: OZONATION TESTS
294
Top o f Contactor T op o f Contactor Top o f Contactor
1-1106 2-1030 3-1027

18 - -
Q ,. = 13.1 gpm Q l = 13 gpm

O Q l = 13 8Pm
Qa = 11.22 scfh O g = 5.53 scfh Qa = 2.64 scfh
16 -
0 3 dose = 0.87 mg/L ( ^ O j dose = 0.98 m g /L ^ O j dose = 0.94 mg/L
O
D m = 0.449 mg/L D m = 0.374 mg/L D m = 0.881 mg/L
14 d =0.0190 d =0.0121 =0.0084
CSTR 26.7% C STR 33.7% CSTR 26.5%
(ft)
Water Column Height
_l-------- !-----!-------- I I I ! ........I 1 ‘I

0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6
Dissolved Ozone Cone. (mg/L)
assume 5 ft o f CSTR exist at the bottom o f contactor

assume all high mixing occurred at the top o f contactor
— assume axial dispersion througout water column
295
4-1118 <3 5-1104 6-1101

18
Q l = 13 gpm Ql = 13.1 gpm Q l = 13 gpm

O O
Qa = 23.17 scfh Qu = 11.22 scfh Qa = 5.53 scfh
16 - i
0 3 dose = 1.72 mg/L ( Oj dose = 1.83 m g/lA O j dose = 1.86 mg/1
(5
D m = 0.380 mg/L Dm = 0.441 mg/L D m = 0.366 mg/L
14 - d =0.0333 d =0.0190 d =0.0121
CSTR 25.9% CSTR 26.6% CSTR 34.1%
(ft)
Water Column Height
n r
0.0 0.5 1.0 1 .50.0 0.5 1.0 1.50.0 0.5 1.0 1.5

— assume axial dispersion throughout water column
__________________________________________________
296
8-1115 9-1112
Q l = 13 gpm Q l = 13 gpm
Qr = 23.17 scfh Qa = 11.15 scfh
Oj dose = 3.58 mg/I^ j 0 3 dose = 3.48 mg/1
Dm = 0.380 mg/L D m = 0.385 mg/L
d =0.0333 d =0.0191
CSTR 26.6% CSTR 27.4%
(ft)
Water Column Height
"I I I I I TT
0 .5 1.0 1.5 0 1 2 30

297
Top o f Contactor Top o f Contactor T op o f Contactor
10-1012 _o 11-1007 12-1128

18 -
Q l = 7.1 gpm Q l = 7 gpm ^ = 7.1 gpm
D
Q g = 5.53 scfh Q g = 2.68 scfh Q a = 1.33 scfh
16
0 3 dose = 0.85 mg/L Oj dose = 0.87 mg/L( ^ O j dose = 0.76 mg/1
D
D m = 0.405 mg/L D m = 0.668 mg/L D m = 0.887 mg/L
14 d =0.0221 d =0.0157 d = 0 .0 1 2 2
O
CSTR 42.2% CSTR 34.0% C STR 29.6%
(ft)
12 Ci -D
Water Column Height
O
10
O O
6 -D -i>
i
tP O.......
...
\O o
4 -
V\ O
° o
o
o
2 -
V o o
V 0
,,, A, ■
0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6 0.0 0.2 0.4 0.6

assume axial dispersion throughout water column
298
13-1020 14-1015 15-1010

18 -
Q l = 1.\ gpm Q l = 7.1 gpm Q l = 7.05 gpm
Qa = 11.15 scfh Qa = 5.53 scfh Qa = 2.71 scfh
16 -
0 3 dose =1.71 mg/L 0 3 dose = 1.86 mg/L, 0 3 dose = 1.84 m g/l
Dm = 0.468 mg/L Dm = 0.310 mg/L D m = 0.61 mg/L
14 -L d = 0.0350 d =0.0221 d =0.0157
CSTR 31.5% CSTR 39.7% CSTR 32.9%
(ft)
J3
Water Column Height
10
a
8 -
Pi
\\
6 - o\
\\
V o
4 - \°
** £
o
2 -
n i i i i i i r
0.0 0.5 1.0 1 .5 0 .0 0.5 1.0 1 .5 0 .0 0.5 1.0 1.5

299
16-1025 17-1022 18-1017

18 -IO
j Q Ql ~ 7-15 gpm QL = 7 .l gpm Q l = 7 gpm
()
Q(. = 23.30 scfh Q g = 11.28 scfh Qa = 5.53 scfh
16 -
q 0V3 dose = 3.29 mg/| ^ O, dose = 3.79 mg/L^ ^ 0 3 dose = 3.43 mg/L
dm = 0.413 mg/L Dm =0.471 mg/L Dm = 0.413 mg/L
14 Ofi =0.0609 d =0.0353 d =0.0225
CSTR 25.0% CSTR 30.2% CSTR 38.0%
(ft)
12 -
Water Column Height
10 -
8 -
6 -
4 -
w o
n r
0 1 2 30
assume 5 ft o f CSTR. exist at the bottom o f contactor

300
Top o f Contactor Top o f Contactor T op o f Contactor
O 19-0927 20-1201 21-1210
Q l = 3.5 gpm Q l = 3.5 gpm Q l = 3.7 gpm

C) O
Qa = 2.73 scfh Qa = 1.38 scfh Q a = 0.67 scfh
16
0 3 dose = 0.73 mg/L 0 3 dose = 1.03 mg/L 0 3 dose = 0.76 m g/l
C
D m = 0.280 mg/L D m = 0 2 3 0 mg/L D m = 0.474 mg/L
d = 0 .0 31 7 d = 0.0250 = 0 .0 2 0 3
C STR 44.1% CSTR 43.0% C S T R 38.5%
(ft)
Water Column Height
_i-------- !---------1— ----- i i r ~n i 1* * * | i i

0.0 0 .2 0.4 0.6 0.0 0.2 0 .4 0 .6 0 .8 0 .0 0 .2 0 .4 0.6

301
22-0930 24-1204
18
Q l = 3.6 gpm Q , = 3 5 8Pm Q l = 3.56 gpm
Qa = 5.50 scfh = 2.64 scfh Qa = 1.33 scfh
16
0 3 dose = 1.57 m g/L( j Oj dose = 1.66 mg/L^ O j dose = 1.57 m g/l
Dm = 0.391 mg/L Dm = 0.260 mg/L Dm = 0.332 mg/L
14 d =0.0435 d =0.0312 d = 0 .0 24 4
CSTR 44.5% CSTR 42.3% CSTR 43.7%
(ft)
12
Water Column Height
10
0
0 .0 0.5 1.0 1.50.0 0.5 1.0 1.50.0 0.5 1.0 1.5

302
25-1004 P 26-0814 27-0819

18 -
Ql = 3.5 gpm Q l = 3.5 gpm Q l = 3.5 gpm
lO
Qa = 11.23 scfh Qa = 5.49 scfh Qg = 2.67 scfh
16 -T
t q Oj dose = 3.34 m g/L( j Oj dose = 3.64 m O j dose = 3.58 mg/l
D = 0.338 mg/L Dm = 0.385 mg/L D m = 0.402 mg/L
14 - i J =0.0314
Od =0.0714 o d =0.0447
j CSTR 35.9% CSTR 43.7% CSTR 41.9%
(ft)
12 -
\ O
Water Column Height
\ \o
10 - \ A
8 -
\
6 - \Q
o
o
o
\ o
W Q
WO
\C>
w
t ----- 1------ 1------ r
2 30 1 2 30
Dissolved Ozone Cone. (m g/L)

303
c o -0 1 0 9 c o -0 1 1 5 c o -0 1 1 8
Q, = 3.6 gpm Q l = 13.1 gpm Ql = 13.1 gpm

Q(i = 1.39 scfh Q(. = 2.65 scfh Q a = 11.17 scfh
0 3 dose = 1.57 mg/Ll O j dose = 0.86 mg/L O j dose = 1.89 mg/Ij
Dm = 0.33 mg/L D = 0.18 mg/L D m = 0.21 mg/L
d =0.0243 =0.0083 d =0.0190
CSTR 31.0%
(ft)
CSTR 38.0% CSTR 22.8%

Water Column Height
d?
i ------------1----------- 1------------1 i i r
0.0 0.5 1.0 0.0 0.2 0.4 0.6 0.0 0.5 1.0 1.5

—- assume axial dispersion throughout water column
304
A P P E N D IX H : A P P L IC A T IO N F O R D IS P E R S IO N C O E F F IC IE N T
305
176.7 cnr G2 G3 G1 G4
Ql Qa N cstr | d N cstr d N cstr d N cstr d
L/h L/h exp Jim | sect. exp Jim sect. exp Jim sect. exp Jim sect.
1400 210 5 3.4 | 0.176 10 3.4 0.176 5 3.4 0.176 ~U "~ 3.4 0.176
1400 150 3 4.1 | 0.144 3 4 1 " 0.144 6 4.1 0.144 ~ (T ~ 4.1 0.144
1400 70 na 5.5 j 0.101 na 33 0.101 6 5.5 0.101 — 0 ~ 33 0.101
1400 0 5 8.3 0.064 7 r i o ' 0.064 5 8.3 0.064 10 8.3 0.064
900 210 1 2.5 0.274 5 2.5 0.274 5 2.5 “0 2 7 4 2 2.5 0.274
“ 900 150 6 2.9 0.224 5 2.9 0.224 5 2.9 0.224 — o- 2.9 0.224
1*1
oc
900 70 na 0.158 na r " 3 3 ' 0138" 7 3.8 0.158 5 3.8 0.158
900 0 2 5.6 0.1 3 5.6 0.1 6 5.6 0.1 6 5.6 0.1
325 210 0 1.5 0.758 1 | 1.5 0.758 3 1.5 0.758 2 j 1.5 0.758
325 150 I 1.6 0.62 1 1.6 0762“ 4 1.6 0.62 1 1.6 0.62
"325" 70 na 1.9 0.437 na 1.9 U 1 3 T na 1.9 0.437 0 1.9 0.437
325 0 4 2.5 0.276 7 2.5 0.276 3 2.5 0.276 7 2.5 0.276
Sauze, N.Le.. Laplanche, A., Orta De Velasquez, M .T., M artin, Ci., Langlais, 3., and Martin, N.
"The R<:sidence Time Distribution o f the Liquid Phase in a Bubble Column an d Its Effect on Ozone Transfer.'
Ozone S cien ce & E n gineering, 14,245-262 (1992).
counter, tine co-curr, tine, packed counter, tine

Ql QC i NcSTR d Q l ! Qa ^CSTR d Ql Qg I N cstr d
gpm cfm | exp Jim section gpm elm exp Jim section gpm cfm mm exp Jim section
7.84 0.0086 | 21.4 41.4 0.0122 7.84 0.0086 88.2 41.4 U.01'22 1.5 0.167 1.4 1.21 1.2 2.0989
©
O
m
rf
7.84 31.7 0.016 7.84 0.0342 85.3 31.7 0.016 1.5 0.167 5.4 2.81 1.7 0.5441
7.84 0.0687 ! 10.4 24.1 0.0212 7.84 0.0687 70.9 24.1 0.0212 1.5 0.167 12.4 3.07 2.8 0.237
3.92 0.0086 | 13 21.0 0.0244 3.92! 0.0086 89.8 21.0 0.0244 1.5 0.167 15.7 3.27 3.3 0.1872
3.92 0.0342 !17.15 16.1 0.0321 3.92 j 0.0342 60 16.1 0.0321
3.92 0.0687 | 4.51 12.3 0.0424 3.92 | 0.0687 37 12.3 0.0424
2.61 0.0086 i 4.08 14.1 0.0367
---- ---------
276r 0.0342 j 2.95! 10.9 0.0482”
2.61 0.0687 j 2.6 j 8.4 0.0637 1
1
1.96 U.0U86 ■2.73 10.8 0.0489 j
1.96 0.0342 I 2.5 8.3 0.0642

1.96 0.0687 | 1.96 6.4 0.0848 I
1
Area= 182 cm2 i
i
i 1 1
Manr las, B.J., Liang, S. and Aieta, E.M.
"Mod eling Hydrodynamics and Ozone Residual Distribution in a Pilot-Scale Ozone Bubble Diffuser Contactoi
Jou m al o f A WWA. 85(3), 90-99 (1993).
306
TotaTT "T=— T T T i r T=~ TO " TT T T " "3 3 — r T^- TTJ TT" Total
73.9 ft A= 313 ft- A= 370 ft2 A= 625 ft2 A= 540 ft2
Ql Qg ~ L d Pm El —d Pm El — d PTn El 7 W Ncstr N,CSTR
mgd kcfh cm2/s sect. Jim cm2/s sect. Jim cn^/s sect. Jim cm2/s sect. Jim Jim exp
22.5 11.3 0 5 7 " ? 33 TT UM T T fTTTt t t fl w 3.14 m m "53CT 7-m.55—rronnr
f f i t n t t f i “3537“ TOT m m I F 7TT
3.36 mmm 6967 8 . “T41I f f f f r r r t f f 6243 6.51 aH aH aH Ha Ha
iiim n ii
22.5 15 4142 .5 ft ft ft tif f 4702 4.0 6.5
22.5 20 4510 2.72 ##### 5119 3.66 mmm 7585 9.16 mmm 6797 7.09 iHHHta
r m r r r r r r 3.9 6.4
10677 1 4 * 1 f r t r r r r t t r 9568 9.36 mmm 3.9
• A | U lllfJ IM
24 45 6348 3.59 ##### 7206 4.83 ftH IT
HHHH
IT Tt IT 4.5
25 35 5613 3.05 m m 6371 4.1 mmm 9440 10.3 mmm 8460 7.94 mmm 3.9 5.6
25 60 7452 4.05 mmm 8458 AA
J ,* t * T
HHHHH
TTTTTTTTTT 12532 13.6 mmm 11231 10.5 ttH ttH ttH Utt Htt 3.9 4.9
28.5 60 7452 3.55 mmm 8458 4.77 mmm 12532 11.9 mmm 11231 9.25 mmm 3.9 5.1
32 50 6716 2.85 mmm 7623 3.83 mmm 11295 9.59 mmm 10122 7.43 mmm 3.9 6.4
2Ao mf II II I I i I II
ttttfwtf 6788
aaaaa 95? tth tth tth hTthtt 9014 6.41 mmm 4.0 6.5
33 40 5981 3.31 ffffft f tf f 10058 8 . ^ O
40 36 5687 I1 ttf ttttttttf

rrrr f t rr 6455 2.6 mmm 9564 6.5 ItH ftH ITU HTTTHT 8571 5.03 ttU ttH Htt ftH ttl l 4.0 11
42 45 6348 2.05 ##### 7206 2.76 mmm 10677 6.91 mmm 9568 5.35 mmm 4.0 7.5
42 51 6790 9 9 rf rf rt rt ttrt rt rt tr fr 7707 2.95 h h h u ii
tt tt tt tt IT 11419 7.39 mmm 10233 5.72 mmm 4.0 9.3
42 60 7452 9 4 1I rr ffftr r trftrffff rrr 8458 3.24^ A
H t t f t w t 12532 8.11 mmm 11231
iiin iiiil
6.28 mmm 4.0 8.3
45 22.5 4694 tfc rIfr rtilt
I1 .*<19 r r r ft Ml rr 5328 1.9 mmm 7894 4 .77 a«<7 iiiiiiitu
ft tt ft tt ft 7074 3.69
a a aa a
4.1 7.2
45 30 5245 1.58 ^0
If M l M l
HH
5954 2.13 mmm 8822 5.33 mmm 7906 4.12 THt tHt Htt THTttH 4.1 7.3
45 40 5981 i. 8 i mmm 6788 2.43 tfjnttm
It TTTttTtr 10058 6.07 mmm 9014 4.7 mmm 4.0 7.8
original chamber configuration
Total L t f — n r 1T 1 = — TO " T T L =1"8.33— !T p = n n r IT Total

73.9 ft A= 313 ft2 A= 370 ft2 A= 208 ft2 A= 270 ft2
6 i 6c ~ L d 7W El d T W ~El d rW ~E~l d 7W N cstr ^ cstr
mgd kcfh cm2/s sect. Jim cm2/s sect. Jim cm2/s sect. Jim cm2/s sect. Jim Jim exp
I 2 3 T T 3 7537" 2TT--njnnnr
mJJ t ttttttfTT T 3 S T 7. MI^ Hill!IIIF
rrrrrrT frr 1 8 3 2 " 1.15 ##### 3465 1.81 T9— TOT
22.5 15 4142 9 J IffI Iffff ffff 4702
I I t TTTT 3.36 mmm 3055 i.23 mmm 3712 1.94 mmm 7.8 11.3
22.5 20 4510 .72
1111
i i ii ii
5119 3.66 f r r w r r t r 3326 4042 2.11 mmm 7.8 10.9
24 60 7452 4 9 9 rrftrffff r tr rffft
rrr 8458 5.67 mmm 5496 2.07 mmm 6678 3.27 mmm 7.4 8
25 45
M I I 11 r I / 1 11
6348 3, */ l♦ Js frtr frrr rfrr rrrt rftf 7206 A f u* T1 fHHHHH
“ .U ff f ff f ff f 4682 1.7*7 Ii Ii Ii»I I<«I IiItI I ii
5689 2.67 mmm 7.5 8.7
26 35 5613 2.93 mmm 6371 3.94 mmm 4140 1I . <L4 ? ♦ fl tl lf lt trfr tr tr ft tl 5030 2.27 mmm 7.7 8.4
26 45 6348 3 ^4 AI fnf frfP i i i i ff
i i ff
f f rr f r f f f f
n
7206 4.46 mmm 4682 1.63 mmm 5689 2.57 mmm 7.6 8.4
28.5 60 7452 8458 4.77 ffffffff ft 5496 1.75 mmm 6678 2.75 7.5 8.2
3.55 mmm
rrrr f f f f ff
32 50 6716 2.85 mmm 7623 3.83 ftH ItH ItH HII Htt 4954 1.4 mmm 6019 2.21 mmm 7.7 9
5981 2.46 mmm 6788 4411 1.21 mmm
im m iii
33 40 3.31 TTTTTTTTTT
h h h h h
5360 1.91 f f f f f f f f f f 7.9 9.2
40 36 5687 1.93 mmm 6455 7. UPi IfIt fIfI fIfIflfiffft 4194 0.95 mmm
^ 5096 1.5 ffff ffffff 8.2 12.4
42 45 6348 2.05 mmm 7206 2.76^ f* II II I I I I II
ffffflfw 4682 l.oi mmm 5689 1.59 mmm 8.1 9.4
42 51 6790 2.2 ff rt fi rt rnf ri ri ri rl 7707 9 Q
a .7^ S rrrrrrrrrr 5008 1 O S fIIII
I .vO f fr fr Itrrfillrr 6085 1.7 mmm 8.0 10.1
42 60 7452 2.41 rlirrIIII I!II 8458
r ff f f f f 3 A J iliiJ iiiJ l
ffftftffff 5496 i . i 8 mmm 6678 1.87 ititititii
r r f t rr f t f t 7.9 10.7
45 22.5 4694 1i 4 9 ftftttttttt
f f f ff f f f t 5328 1.9 mmm 3462 0•7i fffffffitll tftrrrtrr 4206 1.1 mmm 8.8 8.4
45 30 5245 1
^ O aa a a ii
5954 2.13 f f f t f t f t f t 3869 0. 19.
iiiM iim
/ O f»f f»f f»f » f f f»f 4701 1.23 mmm 8.5 11.2
45 40 2931 o.88 mmm 3271 1.17 mmm 2252 0.45 mmm 2666 0.7 mmm 10.0 13.5
m o d if ie d c h a m b e r c o n f ig u r a t io n ( b a t t le s a d d e d )
Blank, B.D., Mannas, 6.J., Corsaro, K.., and Rakness, K..L.

"Enhancement o f Wastewater Disinfection Efficiency in Full-Scale Ozone Bubble-Diffuser Contactors."
Ozone Science & E ngineering, 15(4), 295-310(1993).
307
Test wide Depth Q c Lcngpi E l

Ql d H i */N Qc u=*h E l d # / a/N Lcn0h El d # f ‘/N Ncsnt '^CSTR
ft
mgd ft cfm ft e n r/s cfm ft cnF/s ft ctrF/s Jim exp
MwdSC 2.75 7 19.7 0 3.6/ 467 0.2 6 mm 0 1 176 0 0 0.0 1 176 0 0 (T.IT 23.03 12.26
Contra 22 14 20 12 12 2274 0.6 I ### o 12 1908 0.5 2 mm 4 837 0.1 i mm 6.258 3.313
Tucson 30.2 25 25 0 12 2948 0.8 6 ### 0 I 457 0 0 0.0 I 457 0 0 0.0 8.684 5.148
Henry, UJ. and freeman. b.M.

"Finite Element Analysis and TIO Optimization o f Ozone Contactors."
Ozone Science & Engineering, 17. 587-606, (1995).
Area(fF)= 25.7 Dc(ft)= 5.1 L= 19.5 A l 1.2 A5= 1 4 3 1= 2 3 B5= 15 M W D S C -i

1A m 2A 2U 3A 3B
T est Ql Qc El d I'lN Qc El d f'/N Q c El d t 'lN El d 1‘lfi El d iW N c sn t N c sn t
mgd cfm cnF/s cfm e n r/s cfm c n F /s cnF/s cnF/s Jim exp
07-3B5C 2.84 12.2 1061 0.3 ### 0 466 o.2 m 0 466 0.2 mm 466 03 mm 466 0.1 mm 19.4 16.5
08-3B5C 2.81 12.3 1066 0.3 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 19.3 13.2
19-3B5C 2.78 12.2 1061 0.3 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 19.1 17.2
20-3BSC 2.78 12.2 1061 0.3 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 19.1 19.9
25-1A5C 2.75 0 466 0.2 mm 3.8 3.16
1A5W 2.75 0 466 0.2 mm 3.8 3.01
3B5C 2.75 0 466 0.2 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 21.5 18.1
26-IA 5C 2.77 0 466 0.2 mm 3.8 3.17
1A5W 2.77 0 466 0.2 mm 3.8 3.48
3B5C 2.77 0 466 0.2 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 21.6 18.9
36-1A5C 2.77 29.9 1923 0.6 mm 1.6 2.95
1A5W 2.77 29.9 1923 0.6 1.6 2.1
3B5C 2.77 29.9 1923 0.6 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 17.5 14
43-1A5C 5.5 5.1 715 0.1 mm 4.8 1.73
1A5W 5.5 5.1 715 0.1 mm 4.8 3.53
1B5C 5.5 5.1 715 0.1 mm 0 466 0.1 mm 10.0 12.5
2A5C 5.5 5.1 715 0.1 mm 0 466 0.1 mm 0 466 0.1 mm 16.4 11.4
3 BSC 5.5 5.1 715 0.1 mm 0 466 0.1 mm 0 46 6 0.1 mm 466 0.1 mm 466 0 mm 37.0 14.7
47-1A 5C 5.53 30 1928 0.3 mm 2.3 2.99
1A5W 5.53 30 1928 0.3 mm 2.3 3.14
1B5C 5.53 30 1928 0.3 mm 0 466 0.1 mm 5.8 5.14
2A5C 5.53 30 1928 0.3 mm 0 466 0.1 mm 0 466 0.1 mm 10.9 6
3B5C 5.53 30 1928 0.3 mm 0 466 0.1 mm 0 466 0.1 mm 466 0.1 mm 466 0 mm 29.5 14.4
62-3B5C 5.56 30.1 1933 0.3 mm 0 466 0.1 mm 0 466 0.1 mm 466 0.1 mm 466 0 mm 29.6 17.6
66-3B5C 2.78 29.5 1904 0.6 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 17.5 12.2
72-3B5C 2.75 5.1 715 0.2 mm 0 466 0.2 mm 0 466 0.2 mm 466 0.2 mm 466 0.1 mm 20.2 22.8
308
Areafft2^ 25.7 Dc(ft)= 5.1 L= 19.5 A1 1.2 A5= 1 4 3 1= 2 3 B5= 15 M W U S C -2

IA IB -"2 X 2B 3A JB
T est Ql Qc El d r'/N Q c El d f’.Ti Q c E l d r'/N E l d r'/N El d
r'/N N cst* ■^CSTR
mgd cfm cnF/s cfm c n r /s cfm cmVs cnF/s cttF/ s Jim exp
63-1A5C 3.58 JO 1928 O J mm 2.3 J .1 J
IB5C 5.58 30 1928 0.3 mm 30 1928 0.4 m m 4.2 5.66
2 A IC 5.58 30 1928 0.3 mm 30 1928 0.3 m m 0 466 1.2 9.9 4.8 6.83
3B5C 5.58 30 1928 0.3 mm 30 1928 0.3 m m 0 466 0.1 m m 466 0.1 m m 466 o m m 23.5 19.9
73 -IA 5 C 5.53 5 710 0.1 mm 4.8 3.61
1B5C 5.53 5 710 0.1 mm 5 710 0.1 m m 8.7 5.84
2A IC 5.53 5 710 0.1 mm 5 710 0.1 m m 0 466 1.3 m m 10.2 6.47
3B5C 5.53 5 710 0.1 mm 5 710 0.1 m m 0 466 0.1 m m 466 0.1 7.0 466 o m m 40.7 20.1
76-1A 5C 2.76 30 1928 0.6 mm 1.6 1.49
1B5C 2.76 30 1928 0.6 mm 30 1928 0.8 m m 3.0 3.53
2 A IC 2.76 30 1928 0.6 mm 30 1928 0.6 m m 0 466 2.5 m m 3.4 3.92
3B5C 2.76 30 1928 0.6 mm 30 1928 0.6 m m 0 466 0.2 m m 466 0.2 3.8 466 o.i m m 16.5 14.6
64 -IA 5 C 2.78 5 710 0.2 mm 2.8 2.04
1B5C 2.78 5 710 0.2 mm 5 710 0.3 m m 5.1 4.25
2A1C 2.78 5 710 0.2 mm 5 710 0.2 m m 0 466 2.5 m m 5.9 4.71
3B5C 2.78 5 710 0.2 mm 5 710 0.2 m m 0 466 0.2 m m 466 0.2 3.8 466 o.i m m 22.6 16.8
Area(fl2)= 25.7 Dc(ft)= 5.1 L= 19.5 A1 12 A5= 1 4 3 1= 2 3 B5= 15 MWDSC-3

1A IB 2A 2B 3A.3U
T est Ql Qc El d r'/N Q c El d r'/N Q c E l d r'/N E l d r'/N El d
r'/N N cstr N cstk
mgd cfm cm2/s cfm crnVs cfm cnF/s cnrVs cnF/s Jim exp
70-IA 5C 5.5 0 466 0.1 m m 7 .0 " 5.75
1B5C 5.5 0 466 0.1 m m 30 1928 0.4 mm 6.9 6.32
2A1C 5.5 0 466 0.1 m m 30 1928 0.3 mm 30 1928 5.2 mm 7.2 5.45
2A5C 5.5 0 466 0.1 m m 30 1928 0.3 mm 30 1928 0.4 mm 8.5 7.25
2B IC 5.5 0 466 0.1 m m 30 1928 0.3 mm 30 1928 0.3 mm 466 0.7 m m 9.3 8.69
3B5C 5.5 0 466 0.1 m m 30 1928 0.3 mm 30 1928 0.3 mm 466 0.1 m m 466 0 m m 23.2 14.7
74-1A 5C 5.3 0 466 0.1 m m 6.8 4.48
1B5C 5.3 0 466 0.1 m m 5 710 0.2 mm 10.3 5.68
2A1C 5.3 0 466 0.1 m m 5 710 0.1 mm 5 710 2 mm 11.6 6.75
2A5C 5.3 0 466 0.1 m m 5 710 0.1 mm 5 710 0.2 mm 14.5 8.17
2B1C 5.3 0 466 0.1 m m 5 710 0.1 mm 5 710 0.1 mm 466 0.7 m m 16.5 9
3 BSC 5.3 0 466 0.1 m m 5 710 0.1 mm 5 710 0.1 mm 466 0.1 m m 466 0 m m 33.4 24.3
77-1A 5C 2.77 0 466 0.2 m m 3.8 5.64
I BSC 2.77 0 466 0.2 m m 30 1928 0.8 mm 4.6 3.62
2A1C 2.77 0 466 0.2 m m 30 1928 0.6 mm 30 1928 10 m m 4.7 3.24
2A5C 2.77 0 466 0.2 m m 30 1928 0.6 mm 30 1928 0.9 m m 5.9 5.07
2B1C 2.77 0 466 0.2 m m 30 1928 0.6 mm 30 1928 0.6 m m 466 1.3 m m 6.3 5.33
3B5C 2.77 0 466 0.2 m m 30 1928 0.6 mm 30 1928 0.6 m m 466 0.2 m m 466 0.1 m m 14.6 11.4
71-1A 5C 2.74 0 466 0.2 m m 3.8 4.33
1B5C 2.74 0 466 0.2 m m 5 710 0.3 mm 6.0 4.6
2A1C 2.74 0 466 0.2 m m 5 710 0.2 mm 5 710 3.9 mm 6.7 4.6
2A5C 2.74 0 466 0.2 m m 5 710 0.2 mm 5 710 0.3 mm 8.6 6.21
2B1C 2.74 0 466 0.2 m m 5 710 0.2 mm 5 710 0.2 mm 466 1.4 m m 9.7 7.03
3B5C 2.74 0 466 0.2 m m 5 710 0.2 mm 5 710 0.2 mm 466 0.2 m m 466 0.1 m m 19.0 18.7
309
A P P E N D IX I: E X P E R IM E N T A L D A T A F O R S IN G L E -D IF F U S E R S E T
310
W ater HRT Gas MRT Tie Dispersion Number Dispersion T racer T racer Recovery
Flow Rate V/Q l Flow Rate (Tracer) Number of CSTR CoefT. Mass In. Mass Re. Percent
Date (gpm) (cfm) (min) (cfh) (cfm) (min) (min) (ft2/min) (mg) (mg) %
12/20/93 4.70 0.628 6.06 4.00 0.067 5.898 2.575 0.1932 3.203 12.213 2.5 2.557 102.3
12/22/93 3.25 0.435 8.77 4.00 0.067 8.679 3.683 0.1970 3.156 8.609 2 1.947 97.4
12/23/93 3.40 0.455 8.38 2.00 0.033 8.456 4.212 0.1296 4.431 5.927 2 1.930 96.5
2/9/94* 14.00 1.872 2.04 8.56 0.143 2.194 1.431 0.0479 10.955 9.026 3 3.143 104.8
2/10/94* 14.00 1.872 2.04 4.30 0.072 2.169 1.443 0.0424 12.320 7.979 3 3.138 104.6
2/10/94* 14.00 1.872 2.04 2.12 0.035 2.290 1.582 0.0531 9.945 9.996 3 3.129 104.3
2/11/94* 13.75 1.838 2.07 17.25 0.288 2.112 1.271 0.0877 6.247 16.225 2.5 2.446 97.8
2/13/94* 14.00 1.872 2.04 4.20 0.070 2.022 1.396 0.0415 12.579 7.808 1.5 1.493 99.5
2/17/94* 7.25 0.969 3.93 2.15 0.036 4.066 2.640 0.0600 8.864 5.851 2 2.032 101.6
2/16/94* 7.25 0.969 3.93 4.10 0.068 4.082 2.496 0.0668 8.245 6.511 2 2.069 103.5
2/27/94* 3.50 0.468 8.14 0.53 0.009 8.219 4.878 0.0721 7.472 3.395 2 1.967 98.4
3/2/94* 3.50 0.468 8.14 1.10 0.018 7.675 4.617 0.0614 8.672 2.892 2.5 2J81 95.2
3/4/94* 3.50 0.468 8.14 8.56 0.143 8.430 4.054 0.1446 4.042 6.805 3 2.965 98.8
3/6/94* 3.50 0.468 8.14 4.26 0.071 8.383 4.414 0.1094 5.133 5.147 2.5 2.455 98.2
3/9/94* 3.50 0.468 8.14 2.12 0.035 8.268 4.654 0.0871 6.289 4.099 2.5 2 J9 6 95.8
• experiments with b ig bubbles observed
294 21 0723 594 5 3230 894 1 0750 1194 0 1977 1494 0 0323
12/ 20/93 300 20 8591 600 5 1400 900 09961 1200 0 1809 1500 0 0365
6 306 20 4099 606 4 8881 906 0.9701 1206 0 1876 1506 0 0256
-00063
12 •0 0004 312 19 5688 612 4 7345 912 0 9591 1212 0 1650 1512 0 0256
318 19 0802 618 4 6194 918 0 9247 1218 0 1734 1518 0 0264
18 00105
24 0 0139 324 18 6285 624 4 4499 924 09012 1224 0 1255 1524 0 0273
30 330 184413 630 4 3055 930 0 8760 1230 0 1557 1530 00180
•0 0130
36 336 18 1936 636 42282 936 0.8458 1236 0 1415 1536 0 0306
•0.0046
342 178217 642 4 0242 942 0.8257 1242 0 1599 1542 0 0424
42 -00004
348 17 4255 648 3 9277 948 0.7845 1248 0 1356 1548 00441
48 •00063
34 354 16 4835 654 3 9143 954 0 7694 1254 0 1339 1554 00441
•0.0029
60 360 16 0663 660 3 8295 960 0.7249 1260 0 1280 1560 0 0458
-0 0038
66 00071 366 15 7271 666 3 7010 966 0 7434 1266 0 1297 1566 0 0340
72 372 15 3846 672 3 5952 972 0.6922 1272 01305 1572 00231
0 0634
78 378 15 0698 678 34181 978 0.6880 1278 0 1045 1578 00407
0 3824
84 29690 384 14 7440 684 3 3627 984 06393 1284 0 1054 1584 00189
90 33434 390 14 3923 690 3 1822 990 06091 1290 0 1188 1590 0 0273
96 4 8998 396 13 9373 696 30982 996 06057 1296 0.0936 1596 0 0315
102 67938 402 13 7551 702 3 0605 1002 0 5847 1302 00911 1602 0 0332
108 408 13 2614 708 29589 1008 0 5570 1308 0 0810 1608 0 0323
88515
414 13 1347 714 2 8321 1014 0 5209 1314 00827 1614 0 0239
114 107630
120 420 12 7989 720 28044 1020 0.5234 1320 0 0785 1620 0 0038
134613
426 12 4757 726 26130 1026 0 5058 1326 00869 1626 0 0239
126 140917
132 16 7656 432 12 0828 732 25081 1032 0.4873 1332 00902 1632 0 03)5
138 180114 438 11 7176 738 24669 1038 04714 1338 0 0793 1638 0 0222
144 18 8300 444 11 3390 744 2 4308 1044 04756 1344 00802 1644 0 0113
130 19 4067 450 10 9133 750 23653 1050 0.4622 1350 00735 1650 00155
156 456 106111 756 2.3041 1056 0 4386 1356 00827 1656 00281
203562
162 21 4829 462 10 3030 762 22520 1062 0.4336 1362 00785 1662 0 0365
168 21 8741 468 10 1494 768 21555 1068 03816 1368 00625 1668 0 0248
174 474 9 6700 774 20195 1074 03732 1374 00583 1674 00130
22.5642
180 23 0998 480 9 3988 780 I 9817 1080 03664 1380 00676 1680 4)0080
186 23.4406 486 9 1235 786 1 9011 1086 03664 1386 0 0567 1686 -0 0231
492 8 9564 792 1 8583 1092 03446 1392 0 0533 1692 •0 0071
192 23 8688
198 498 8 5795 798 1 7483 1098 0J455 1398 0.0500 1698 00063
24 1492
204 240367 504 8 2562 804 1 7038 1104 0.3396 1404 0 0533 1704 0 0029
210 24 1878 510 7 9901 810 1 6811 1110 03161 1410 00516 1710 -0 0038
216 516 7 5695 816 1 6215 (116 0 3110 1416 0.0491 1716 4)0147
24 1315
222 24 2927 522 7 5099 822 1 5393 1122 0.2859 1422 OOS33 1722 •00113
228 528 7 3689 828 1 4981 1128 0 2968 1428 00416 1728 4)0206
24 1483
234 240190 534 7 1304 834 1 4847 1134 0.2699 1434 00424 1734 4)0164
240 237143 540 6 9248 840 1 4402 1140 0.2741 1440 00365 1740 -00122
246 23 4834 546 6 7208 846 I 4016 1146 0.2548 1446 00432 1746 •00164
252 23 3348 552 6 5344 852 1 3403 1152 02397 1452 0 0432 1752 •00113
258 23 0897 558 6 3069 858 1 2983 1158 0.2237 1458 0.0424 1758 4)0180
264 22.8546 564 6 1205 864 1 2656 1164 0.2221 1464 0.0424 1764 0 0021
270 226196 570 5 8477 870 1 2186 1170 0.2153 1470 0.0416 1770 4)0281
276 220378 576 5 6957 876 I 1707 1176 0.1985 1476 0.0407 1776 4)0130
282 21 7935 582 5 5925 882 I 1472 1182 0.1851 1482 0.0382 1782 4)0231
288 21 5853 588 5 4682 888 1 1103 1188 0 1860 1488 00281 1788 00021
311
9 3448 1328 0 6999 2024 0.0557 264 160919 960 23127

1794 -0.0239 63 2
640 1336 0 6516 2032 0.0754 272 166838 968 22112
1800 *0 0130 9 0702
648 8 74 4 0 1344 0 6573 2040 0 0713 280 17 1903 976 21977
656 X 6440 1352 0 6384 2048 00623 288 17 5271 984 21034
664 8 45 7 2 1360 0 6786 2056 00516 296 178936 992 20343
672 8 0875 1368 0 6376 2064 00451 304 180067 1000 1 8762
680 7 8974 1376 06180 2072 0.0705 312 18.2699 1008 1 8412
688 7 65 4 8 1384 0 5868 2080 0 0484 320 18 4208 1016 1 7846
12/22/93 696
704
7 5007
7 3081
1392
1400
0 6114
0 5540
2088
2096
0.0467
0 0541
328
336
18.5241
18.6732
1024
1032
1.7352
1 6795
8 •0 0066 1040 16131
712 7 1008 1408 0 5253 2104 00664 344 18.6543
16 -0 0033 1048 1 5610
720 6 9459 1416 0 5212 2112 0.0369 352 18.7710
24 00074 1 4945
728 6 7369 1424 0 5122 2120 0.0213 360 186291 1056
32 00000 1064 1 4388
736 65115 1432 0 4819 2128 0.0393 368 18.7064
40 00156 1 4253
744 6 43 2 8 1440 0 4745 2136 00393 376 18 6776 1072
48 0 0025 1080 13804
7 52 60091 1448 0 4655 2144 00320 384 18.5474
36 •0.0074 I 3239
760 5 9280 1456 0 4352 2152 0.0279 392 18.3552 1088
64 •0 0066 1.2439
768 5 7886 1464 0 4368 2160 00303 400 180247 1096
72 -0 0025 1104 1 1909
776 5 56 4 9 1472 0 4467 2168 00229 408 179169
80 0 0049 1112 1 1694
7 84 5 5075 1480 0 4229 2176 0.0254 416 17 7382
88 0 0049 1 1487
7 92 5 21 8 2 1488 0 4270 2184 00123 424 174)04 1120
% 0.0631 17 3206 1128 1 0957
800 5 0076 (496 0 3959 2192 002 29 432
104 03254 1136 1 0903
8 08 4 93 9 6 1504 0 3877 2200 0 0270 440 16 9424
U2 0 9335 1144 1 0499
816 4 89 6 9 1512 0 3942 2208 00303 448 16 6416
120 1 8342 1152 0 9799
824 4 7289 1520 0 3532 2216 00205 456 16.5275
128 3 1078 16.2284 1160 09466
832 4 5503 1528 03565 2224 0.0270 464
136 4 6675 15 8377 1168 0 9520
840 4 47 4 0 1536 0 3098 2232 00164 472
144 57198 1176 0 8811
8 48 4 3314 1544 0 3385 2240 00254 480 153572
152 7 7351 15 1569 1184 0 8451
8 56 4 1724 1552 0 JI88 2248 00156 488
160 8 8555 1192 0 7957
8 64 4 1175 1560 0 3196 2256 0.0172 496 14 8462
168 10 1930 1200 0 7670
872 4 0 0 )2 1568 0 3147 2264 00131 504 14 4995
176 10 9847 0 7571
880 3 8971 1576 0.3147 2272 00148 512 143819 1208
184 I I 5404 1216 0 7320
8 88 3 83 6 4 1584 03123 2280 0 0418 520 139570
192 I I 9944 1224 0 7167
8 96 3 6676 1592 0 2934 2288 0.0164 528 13 7424
200 12.4534 06862
9 04 3 5996 1600 0 2819 2296 0.0156 536 13 3355 1232
208 128861 1240 0 6538
912 3 4963 1608 0 2770 2304 0 0082 544 119951
216 13 5131 1248 0 6170
920 3 3938 1616 0 2573 2312 0.0041 552 115263
224 14 0770 1256 06009
928 3 27 1 7 1624 0 2639 2320 00172 560 120997
232 14 6580 1264 0 5748
936 3 1693 1632 0 2319 2328 0.0123 568 11 8320
240 15 1236 1272 0 5550
944 3 1283 1640 0 2246 2336 00123 576 1) 4952
248 15 4112 I I 2384 1280 0 5281
952 3 0341 1648 0 2246 2344 0.0049 584
256 16 0218 10 9474 1288 0 5245
960 2 ‘>062 1656 02172 2352 0.0090 592
264 16 1070 10 7255 1296 0 4580
‘>68 2 8193 1664 0 2041 2360 00385 600
272 16.3464 1304 04634
976 2 69 9 7 1672 02131 2368 00164 608 103025
280 16 5758 1312 0 4149
984 2 66 2 8 1680 0 1983 2376 00205 616 10 1624
288 16 7283 9 7798 1320 0 3934
992 2 5497 1688 0 1844 2384 00123 624
296 169365 9 5149 1328 0 4158
1000 2 5144 1696 0 1647 2392 00180 632
304 17 0848 0 3871
1008 2 4604 1704 0 1836 2400 00156 640 9 0586 1336
312 17 1241 1344 0 3727
1016 2 2825 1712 0 1688 648 8.9185
320 17 3684 0 3341
1024 2 25 2 2 1720 0 1418 656 87613 1352
328 17 4233 1360 03161
1032 2 1637 1728 0 1574 664 8 4937
336 17 3962 83239 1368 0 2946
1040 2 1325 1736 0 1492 672
344 17 4503 1376 0 2937
1048 20186 1744 0 1393 680 8 1362
352 17 3454 1384 0 2937
1056 20014 1752 0 1606 688 7 7114
360
368
17 2741
17 0946
(0 6 4 1 939! 1760 0 1418 12/23/93 696 7 4958
70522
1392
1400
0 2784
02982
1072 1 89 8 9 1768 0 1656 00117 704
376 17 0578 8 0 2838
1080 1 86 3 7 1776 0 1442 712 6 8627 1408
384 16 9487 16 •0.0198
1088 1 7998 1784 0 1483 720 66669 1416 0 2281
392 16 5931 24 00018
1096 1 7047 1792 0 1320 728 66031 1424 0 2335
400 16 4496 32 •00045
1104 1 7063 1800 0 1369 736 63687 1432 0 1886
408 16 2431 40 0 0099
1 1 )2 I 5982 1808 0 1385 744 60391 1440 0 1859
416 160153 48 0 0314
1120 1 5900 1816 0 1459 752 5 8945 1448 0 1868
424 157710 56 0 0234
1128 I 5424 1824 0 1360 760 56331 1456 0 1751
432 15 5907 64 00099
1136 1 4760 1832 0 1303 768 55110 1464 0 1832
440 15 4727 72 00054
1144 1 4892 1840 0 1467 776 53053 1472 0 1419
448 15 0924 80 0.0234
1152 1 43 5 9 1848 0 1246 784 5 2092 1480 0 1706
456 149154 88 0.0027
1160 I 3769 1856 0 1238 792 5 0278 1488 0 1491
464 14 5367 96 0.0135
1168 1 3367 1864 0 1156 800 48508 1496 0 1509
472 14 3253 104 0 0036
1176 1 2933 1872 0 1213 808 4 7116 1504 0 1374
480 14 0327 112 0.0269
1184 1 2638 1880 0 1090 816 45940 1512 0 1177
488 13 8573 120 0.0503
1192 1 2015 1888 00902 824 44880 1520 00961
496 13 4500 128 0.0665 0 0979
1200 1 1818 1896 0 0861 832 4 2060 1528
504 13 2263 136 04455
1208 1 1572 1904 0 0951 840 40847 1536 00988
512 13 0468 144 06745
1216 1 1146 1912 0 0852 848 38764 1544 00898
520 128050 152 0 8739
1224 I 0941 1920 0 0820 856 37614 1552 00943
528 12.4386 160 17298
1232 1 05 7 2 1928 00697 864 3 6 (5 0 1560 00916
536 12.2329 168 15759
1240 1 0466 1936 00664 872 3 3994 1568 0 0817
544 12.0961 176 3 4785
1248 1 0015 1944 0 0770 880 33384 1576 00665
552 11 8412 184 49595
1256 0 95 2 3 1952 0 0754 888 32144 1584 0 0665
560 I I 5281 192 75884
1264 0 9540 1960 0 0836 896 3 IIII 1592 0 0548
568 I I 2052 200 8.4748
1272 0 ‘>073 1968 0 0754 904 3 0824 1600 00602
576 109069 208 11.3965
1280 0 85 5 6 1976 0 0574 912 28651 1608 00692
584 106749 216 117625
1288 0 8310 1984 00697 920 28112 1616 0.0647
592 105479 224 119403
1296 0 8188 1992 00869 928 27043 1624 0 0566
600 10 1578 232 13 5735
1304 0 7819 2000 0 0779 936 25920 1632 0 0727
608 100070 240 14.1196
1312 0 7753 2008 0 0729 944 25345 1640 0.0584
616 98250 248 15.1138
1320 0 7483 2016 0Q721 952 24133 1648 0.0305
624 9 4939 256 15.4372
312
1656 0 0341 2352 •00198 148 28 9375 322 0 1503 496 •0 0082 56 05797
1664 0.0440 2360 -0 0404 150 28 0145 324 0 1247 498 -0.0238 58 1 5687
1672 0.0422 2368 -0 0269 152 27 4270 326 0.1228 500 -0 0009 60 1 7805
1680 0.0620 2376 -0 0368 154 26 7634 328 0 1082 502 00220 62 2.3378
1688 0 0781 2384 4)0359 156 25 3601 330 00871 504 001 0! 64 4 7383
1696 00692 2392 •0 0413 158 24 3005 332 0.0944 506 00119 66 6 5036
1704 00557 2400 4) 0323 160 22.9137 334 0 1118 508 00128 68 69069
1712 00683 162 20 942! 336 0.0898 510 •0.0092 70 80417
1720 00745 164 194078 338 0 1017 512 -0 0284 72 12.9096
1728 00638 166 19 1053 340 00926 514 •0 0247 74 13 6383
1736 0 0341 168 18 2932 342 0 1017 516 -0.0247 76 14.2277
1744 0 0395 170 172840 344 00816 518 -0 0147 78 15 729!
1752 00539 172 15 6864 346 00495 520 -0.0119 80 21 6621
1760 0 0431 2/9/94 174 15 0567 348 00431
0.0403
522
524
-0 0027
-0 0018
82
84
22.4229
22-9206
1768 00368 i 4)0073 176 14 5435 350
1776 0047b 178 13 0632 352 00412 526 •0 0055 86 25 871!
4 J) 0055
1784 0 0557 180 12.5746 354 0.0385 528 -0 0055 88 31.5576
6 -0 0247
1792 0.0413 182 12.3547 356 00339 530 00037 90 34 9627
8 4)0348
1800 0 0467 184 11 9312 358 00257 532 •0010! 92 34 4393
10 0 0046
1808 0 0494 186 1! 2383 360 00247 534 •0.0211 94 336061
12 0 0037
1816 0.0386 188 10 6755 362 0.0321 536 -0 0339 96 34 4806
14 4)0037
1824 00449 190 10 2300 364 00302 538 -0 0082 98 35 6905
16 4) 0027
1832 00350 192 10 0082 366 00257 540 -0 0156 100 37 6281
18 4)0082
1840 00413 194 94518 368 00183 542 -0 0284 102 38 7088
20 00312
1848 0 0395 196 8 7928 370 00073 544 •00128 104 39 7335
0 0156 39 8261
1856 00377 198 8 4775 372 00119 546 -00174 106
24 4)0027
1864 0 0225 200 83024 374 0.0092 548 -0 0257 108 40 0085
26 0 0082 40 5254
1872 0 0314 202 7 4216 376 0.0055 550 •0 0302 110
28 0 0312
1880 0 0431 204 7 2016 378 -0.0037 552 -0.0247 112 40 5612
30 0 0321
1888 00296 206 7 0339 380 0 0046 554 •00339 114 40 8692
32 0 0330 40 6519
1896 00180 208 6 5939 382 •0.0018 556 -0 0302 116
34 0 0202
1904 0.0225 210 5 7653 384 00156 558 •0 0202 118 40 4384
36 0 0055
1912 0.0117 212 5 4876 386 •0.0009 560 4)0137 120 40 1111
38 4) 0064
1920 00207 214 5 2804 388 -0.0009 562 -0 0009 122 39 5658
40 -00101
1928 00296 216 4 9459 390 00009 564 0 0073 124 39 0030
42 4)0082
1936 00162 218 4 7067 392 •0.0101 566 00192 126 38.1881
44 0 0119
1944 0 0207 220 4 5903 394 -0.0082 568 0 0220 128 38 0342
46 0 0183
0 0296 3 9175 396 •0.0119 570 0 0055 130 37 7876
1952 48 0 1742
00171 224 3 6838 398 •00128 572 -0 0275 132 36 8344
1960 50 0 3941
1968 Q 0305 226 3 4904 400 •00174 574 -00330 134 33 8454
32 0 5362
1976 0 0377 228 3 2768 402 -00128 576 -0 0302 136 31 3083
54 0 6462
1984 0 0395 230 3 0559 404 •0 0119 578 •0 0302 138 30 6566
56 0 9459
1992 00323 232 2 7314 406 -0.0229 580 4)0275 ;*c 30 6621
58 1 7571
2000 0.0054 234 24546 408 •0 0183 582 4)0247 142 29 2643
60 3 1228 28 2258
2008 00162 236 24161 410 -00082 584 4)0174 144
62 4 2566 27 992!
2016 00269 238 23080 412 00082 586 -0 032! 146
64 6 8689
2024 00135 240 22236 414 00302 588 4)0293 148 26 9215
66 7 2419
-0 0099 242 20596 416 00137 590 4)0284 150 25 7547
2032 68 8 1594
2040 0 0027 244 t 9175 418 •0.0018 592 4)0385 152 24 2606
70 10 5288 228949
2048 00000 246 ! 8222 420 0 0119 594 -0 0385 154
72 12 8679
2056 0 0018 248 1 7443 422 00211 596 4)0422 156 21 4980
74 14 1695
2064 -0 0108 250 1 5802 424 00202 598 4)0376 158 21 1717
76 17 0842 19 9509
2072 •0 0135 252 1 3474 426 0 0312 600 4)0440 160
78 18 7341
2080 -00189 254 1 3400 428 0.0000 162 186823
80 20 3372
2088 -0.0036 256 I 2805 430 -00293 164 18 7813
82 23 8413
2096 0 0584 258 1 2108 432 00082 166 182781
84 25 3555
2104 0 0467 260 1 1072 434 •00018 168 17 6594
86 26 7276
2112 00359 262 1.0247 436 00119 170 173221
88 29 1465
2120 0 0368 264 1 0238 438 00156 172 162405
2128 0 0341
90
92
30 5819
32 5342
266 0 9138 440 0 0165 2/10/94 174
176
14 6842
14 5494
2136 00584 268 08194 442 0.0027 2 0 0105
94 35 2317 13 9610
2144 0 0431 270 0 7727 444 00046 178
96 37 5552 4 00023
00323 272 0 7397 446 -00073 180 13.4129
2152 98 38 7908 6 -0 0023
2160 0 0341 274 07186 448 -00046 182 11 2048
100 39 3105 8 -0 0032
2168 Q.0431 276 0 6279 450 •0.0312 184 137144
102 40 1199 10 00050
2176 00180 278 0 5894 452 -00348 186 10 3313
104 40 3765 12 0 0060
2184 00144 280 0 5774 454 -0.0321 188 94083
106 40 5543 14 00041
2192 0 0521 282 0 5719 456 •0 0247 190 8 7978
ton 41 2638 16 4)0005
2200 0 0278 284 0 5655 458 •0.0330 192 7 9674
110 41 4801 18 •0 0041
2208 0.0323 286 0 5380 460 -00110 194 7 6127
112 41 4984 20 4)0353
2216 00063 288 0 4977 462 00046 196 7 3487
114 41 7193 4)0005
2224 -0 0090 290 0 4051 464 0.0082 198 70050
116 41 5552 24 00023
2232 -0 0216 292 03712 466 0.0027 200 7 0508
118 41 4278 26 4)0014
2240 -0 0045 294 03318 468 00073 202 6 2727
120 40 9)45 28 4)0023
2248 -00153 296 Q 3006 470 00257 204 56870
122 40 5332 30 4)0050
2256 -0 0018 298 02961 472 00192 4)0069 206 55513
124 39 5855 32
2264 -00135 300 0 273! 474 •0 0009 208 49784
126 38 9439 34 O.OOU
2272 0 0027 302 0 2557 476 0.0147 210 46714
128 38 3508 36 -0 0078
2280 0 0063 304 02759 478 00165 212 4 6246
130 37 0575 38 00032
2288 00000 306 0 2475 480 00018 214 4 3671
132 36 2940 40 -00142
2296 •00153 308 0 2310 482 0.0073 216 43002
134 35 8605 42 •00197
2304 •0.0368 310 02264 484 0.0037 218 39152
136 34 942! 44 4)0160
2312 •0 0216 312 0 207! 486 00055 220 3 5000
138 34 1061 46 0.0014
2320 -0 0027 314 0 1943 488 0.0000 222 3 2983
140 32711! 48 •00041
2328 -00296 316 0 1842 490 0.0073 224 3 0664
142 32 2647 SO 00032
2336 -0.0009 318 0 1641 492 0.0284 226 28886
144 31 1923 52 0.0270
2344 -0.0153 320 0 1677 494 0.0247 228 27979
146 30 294! 54 0.0481
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
313
230 18492 404 -0 0325 578 -0 0316 138 36.5731 312 0.2644 486 0 0188
232 16274 406 •0 0243 580 4)0270 140 34.3037 314 02745 488 0 0307
234 12901 408 -0 0(51 582 4)0390 142 31 5842 316 0 2498 490 0 0270
236 12168 410 4)0124 584 4)0380 144 30.6822 318 0 2333 492 00298
238 10545 412 -0 0335 586 4)0225 146 30 9792 320 0 2342 494 00197
240 1 9033 414 •O 0069 588 4)0316 148 310168 322 02140 496 00261
242 1 6301 416 -00133 590 -0 0344 150 30.1946 324 0 1957 498 00344
244 I 5302 418 •0 0105 592 4)0252 152 29.3440 326 0 1966 500 0 0362
246 1.4523 420 -00179 594 -00380 154 28.0223 328 0 1856 502 0 0454
248 I 3753 422 •0 0206 596 -0 0325 156 25.6071 330 0.1636 504 00463
250 1.2525 424 -00151 598 4)0399 158 23.4027 332 0 1444 506 00518
252 1 1810 426 •0 0069 600 4)0445 160 21.5860 334 0.1544 508 0 0335
254 1.1269 428 -0 0 )6 0 162 21.3431 336 0 1462 510 00307
256 1.0454 430 4)0087 164 21 3532 338 0 1370 512 0037!
258 0.9913 432 0 0005 166 198537 340 0 1214 514 00408
260 09015 434 -0 0014 168 193560 342 0 1343 516 0 0252
262 0 8639 436 0 0050 170 18.5952 344 0 1159 5IS 00408
264 08153 438 0 0078 172 16.8106 346 0 1104 520 0 0472
266 0 7695 440 4) 0005 2/10/94 174 16.0380 348
350
00994
0 1141
522
524
00463
0 0527
268 0 7071 442 0 0014 ■» 176 15.1067
0 0280
270 0 6054 444 4)0005 178 142506 352 0 1178 526 0 0619
4 00197
272 0 5550 446 4)0142 ISO 132689 354 0 1068 528 0 0591
6 0 0069
274 0 5422 448 4)0160 182 12.5008 356 00994 530 0 0500
8 00096
276 0 5376 450 0 0197 (84 122057 358 00830 532 0 0481
10 0 0335
278 0.5101 452 4)0170 186 120077 360 00637 534 0 0527
12 0 0234
280 04588 454 4)0280 188 10.8867 362 0.0674 536 0 0518
14 4)0041
282 0.4083 456 4)0289 190 103350 364 00912 538 0 0545
16 0 0325
284 0.3928 458 4)0298 192 10.5403 366 00939 540 0 0435
18 0 0344
286 03607 460 4) 0289 194 10.4046 368 0 0866 542 00481
20 0 0371
288 0 3625 462 4) 0335 196 97896 370 0 081! 544 0 0362
4)0032
290 0 3268 464 4)0270 198 9 6695 372 00793 546 0 0380
24 -0 QQ5Q
292 0.3148 466 4)0243 200 8.9472 374 00600 548 0 0353
26 0 0)24
294 0.2709 468 4)0298 202 8.6310 376 0 0435 550 0 04)7
28 0 0124
296 0 2516 470 4)0399 204 83991 378 00362 552 0 0380
30 00197
298 0 2415 472 0 0445 206 8.1251 380 0.0637 554 0 0390
32 •0 0050
300 0.2058 474 -0 0380 208 76393 382 0.0408 556 0 0325
34 •00151
302 0 2021 476 4)0417 210 70481 384 0 0426 558 0 0371
36 -00115
304 0 1856 478 4)0481 212 6.6283 386 0 0545 560 0 0417
38 4)0115
306 0.1590 480 4)0564 214 6 2699 388 00692 562 0 0417
40 4)0005
308 0 1444 482 -0 0445 216 5.9848 390 00802 564 0 0426
42 00041
310 0 1398 484 4 ) 0280 218 5 2974 392 00555 566 0 0573
44 0 0142
312 0 1370 486 4) 0527 220 4.7530 394 0 0555 568 0 0454
46 0 0380
314 0 1260 488 4)0408 222 4.5302 396 0 0738 570 0 0463
48 0 0335
316 0 1416 490 • 0 0417 224 4 2470 398 0 0930 572 0 0371
50 00188
318 0 1159 492 0 0005 226 4 0435 400 00976 574 0 0536
52 0 0225
320 0 1086 494 0 0060 228 3 8309 402 0 0976 576 0 0390
54 0 0)70
322 0.1059 496 4)0014 230 35779 404 0 1343 578 0 0362
56 0 0307
324 0 08o6 498 0 0060 232 3 0793 406 0 1159 580 0 0481
58 0 0105
326 00912 500 4)0078 234 29784 408 00628 582 0 0390
60 0 0225
328 00765 502 -0 0069 236 27255 410 00S18 584 0 0307
62 0 0353
330 0 0637 504 4)0105 238 25596 412 00417 586 0 0261
64 0 0747
332 00610 506 -0 0069 240 24450 414 00280 588 0 0399
66 0 2479
334 00545 508 4) 0032 242 23772 416 0 0252 590 00408
68 0 5367
336 0 0481 510 4)0206 244 22516 418 00197 592 00124
70 0 8996
338 0 0536 512 4)0188 246 20463 420 0.0133 594 0 0307
72 1 5678
340 0 0399 514 4)0325 248 20142 422 00252 596 00289
74 2 8355
342 00243 516 -0 0298 250 1 8373 424 0 0234 598 0 0289
76 3 8758
344 00243 518 0 0133 252 1.6897 426 0 0353 600 0 0280
78 7 7099
346 0 0280 520 0 0234 254 1 6815 428 00362
80 10 9436
348 0 0 (1 5 522 -0 0280 256 I 6549 430 0 0353
82 13 1736
350 0.0170 524 4)0115 258 1.5779 432 0 0545
84 15 3872
352 -00105 526 0 0014 260 I 4276 434 0 048)
86 19 9380
3S4 00151 528 4)0215 262 1.4936 436 0 0564
88 24 8555
530 -0 1)087 264 I 4587 438 00490
356
358
00206
0 0087 532 4 ) 0050
90
92
25 6227
30 6978 266 1.3607 440 00555 2/11/94
360 00041 534 00133 268 1 2305 442 00564 0 0049
94 34 7876
362 -0 0096 536 0 0041 270 I 1416 444 00463 4 00022
96 38 2972
364 0 0087 538 0 0014 272 1 0215 446 0 0325 6 00102
98 43 2486
366 0 0023 540 4)0032 274 0.9381 448 0 0435 8 0 0076
100 45 6235
368 -00160 542 0 0023 276 0.8080 450 00435 10 0 0093
102 49 4402
370 -0 0069 544 0 0014 278 0 7539 452 00371 12 00129
104 486134
372 00005 546 •0 0087 280 06879 454 00463 14 00129
106 46 7976
374 -00124 548 4 ) 0060 282 0.6631 456 00518 16 0 0022
108 480167
376 •00188 550 00096 284 0 6522 458 0 0518 18 0 0022
110 49 5098
378 -00170 552 0 0179 286 0.6292 460 0.0426 20 00084
112 49 8169
380 -0 0087 554 0 0133 288 0 5724 462 0.0344 00102
114 49 3091
382 •00124 556 00133 290 0.5486 464 00197 24 00262
116 49 2467
384 -00188 558 00060 292 05073 466 00206 26 00155
118 48 7866
386 •00160 560 0 0078 294 0.4578 468 00307 28 00111
120 48 1157
388 •0 0243 562 4)0124 296 04349 470 0 0325 30 00182
122 47 2999
390 -0 0188 564 •0 0179 298 0.4221 472 0.0289 32 00173
124 45 7619
392 •0 0298 566 •0 0316 300 03799 474 00325 34 0 0378
126 44 1194
394 -0 0344 568 4)0380 302 0.3708 476 0.0362 36 00777
128 41 7793
396 -0 0316 570 •0 0481 304 0.3543 478 00170 38 0 1515
130 38 8692
398 -0 0289 572 4)0362 306 0.3293 480 00335 40 0 1674
132 38 8444
400 -00380 574 -0 0298 308 0.2993 482 00344 42 02305
134 37 1891
402 -0 0316 576 4)0399 310 0.2699 484 00335 44 07662
136 37 4530
314
46 09989 220 2 3198 3 94 00440 568 00253 128 24 2967 302 00660
48 1 4839 y>? 2 1244 396 0 0529 570 0 0635 130 23 4755 304 0 0750
SO 1 8490 224 20444 3 98 0 0537 572 0.0644 132 21 8205 306 00623
52 2 .4353 226 I 9538 400 0 0582 574 00662 134 20 5511 308 0 0515
54 3 0380 228 I 8188 402 0 0520 576 0.0466 136 18 8617 310 00407
56 3 9605 230 1 7673 404 0 0671 578 0.0724 138 18.5862 312 00407
58 50096 232 1 6554 406 0 0564 580 00733 140 16.4487 314 00343
60 6 .8 8 9 3 2J4 1 5843 408 00537 582 0.0831 142 15.5787 316 0 0479
62 89466 2J6 I 4653 410 0 0546 584 00866 144 15 0917 318 00425
64 98300 238 I 3613 412 0 0502 586 0.0946 (46 14 1215 320 00352
66 10 1476 240 1 2805 414 0 0591 588 00911 148 13 4782 322 00443
68 11 0 6 1 6 242 I 2423 416 0 0582 590 0.0706 ISO 13 1647 324 0 0425
70 11.8753 244 1 1570 418 0 0431 592 00466 152 12.7193 326 0 0398
72 13 6 4 0 4 246 1 1170 420 0 0431 594 00395 154 124763 328 00361
74 15 6 3 3 3 248 1 0335 422 00466 596 00573 156 12.2496 330 0 0235
76 17 3 8 9 9 250 0 9492 424 0 0564 598 00440 158 1! 8945 332 0 0316
78 18 1565 252 0 9207 426 00493 600 00457 160 11 3479 334 0 0343
80 189080 254 0 8577 4 28 0 0324 162 10 6523 336 0 0370
82 198194 256 0 8088 430 0 0262 164 100027 338 0 0298
84 20 3 8 6 2 258 0 7742 432 0 0209 166 8.7849 340 0 0578
86 21 6351 260 0 7431 434 00404 168 75906 342 00407
88 2 2 .6 1 0 7 262 0 7102 436 00475 170 6 7061 344 0 0407
90 23 0 6 9 7 264 0 6933 438 0 0511 172 6 6474 346 0 0416
92 23 9 1 7 2 266 0 6889 440 00617 2/13/94 174 6 2851
57630
348
350
0 0370
0 0434
94 24 2592 268 0 6089 442 00662 2 0.0280 176
96 24 6749 270 0 5681 444 0 0475 178 55254 352 0 0524
4 00190
98 24 8410 272 0 5485 446 0 0529 180 5 4730 354 0 0307
6 00343
100 24 9823 274 0 4970 448 0 0564 182 4.9941 356 0 0253
8 00163
102 25 1004 276 0 4819 450 0 0537 184 4 5885 358 0 0398
10 00163
104 24 9965 278 0 4490 452 00466 186 4 1585 360 0 0280
12 0.0217
106 2 4 7913 280 0 4375 454 0 0546 188 39606 362 0 0244
14 0 0262
108 246767 282 04117 456 0 0475 190 3 6254 364 0 0524
16 00154
HO 24 5 3 1 0 284 0 3762 4 58 0 0431 192 3 3318 366 00280
18 00208
!t2 24 2681 286 0 3620 460 0 039S 194 3 0454 368 0 0244
20 0 0280
114 23 7893 288 0 3353 462 0 0457 196 16624 370 0 0407
22 0.018!
116 23 3 0 6 0 290 0 3185 464 0 0635 198 2.2947 372 0 0235
24 00208
118 22 9063 292 0 3140 466 0 0751 200 11953 374 0 0370
26 0 0316
120 2 2 .6 5 4 0 294 0 2927 468 0 0582 202 1 9803 376 00325
28 00370
2 2.3351 296 0 2918 470 0 0555 204 1 9153 378 0 0289
122 30 00289
124 2 1 .9 0 8 7 298 0 2705 472 0 0671 206 I 7626 380 0 0370
32 0.0316
126 21 1812 300 0 2838 474 0 0635 208 1 6316 382 0 0334
34 0 0208
128 2 0 5594 302 0 2483 476 0 0733 210 1 5494 384 0 0307
36 00163
130 20 0584 304 0 2350 478 0 0795 212 1 4419 386 0 0343
38 0.0244
132 19 2491 306 0 2110 480 0 0546 214 I 3777 388 0 0343
40 00027
134 187162 308 Q 2083 482 00617 216 1 2982 390 00425
42 -0.0043
136 180979 310 02110 484 0 0644 218 I 2756 392 0 0289
44 0 0036
138 17 5 8 3 6 312 0 1905 486 0 0591 220 1 1437 394 0 0307
46 0 0117
140 17 0 2 3 0 314 0 1870 488 0 0546 222 1 0407 396 0 0361
48 00334
142 16 6 1 8 9 316 0 1834 490 0 0546 224 0 9116 398 0 0298
50 0 1265
144 160370 318 0 1817 492 0 0591 226 0 8736 400 00316
52 0 2222
146 15 3 8 0 6 320 0 1595 494 0 0680 228 08104 402 0 0361
54 0 3252
148 14 6131 322 0 1S79 496 0 0680 230 0 7679 404 0 0253
56 04318
ISO 13 8 9 6 2 324 0 1692 498 0 0635 232 0 6758 406 0 0398
58 0 592b
152 13 2 8 2 4 <26 0 1648 500 0 0662 234 0 6721 408 0 0253
60 0 8077
154 13 0 5 6 8 328 0 1559 502 0 0573 236 06261 410 0 0325
62 24194
156 128676 330 0 1248 5 04 0 0555 238 0 5881 412 0 0280
64 25983
158 12 4 8 2 9 332 0 1239 5 06 0 0431 240 0 5240 414 0 0307
66 28340
160 11 8 4 8 7 334 0 1355 508 00360 242 05195 416 0 0307
68 3 7664
162 11 0501 336 0 1532 510 0 0271 244 04860 418 0 0334
70 54648
164 10 5 4 6 4 338 0 1417 512 0 0306 246 0 4626 420 0 0163
72 7 0829
166 9 9965 340 0 1346 514 0 0422 248 0 4219 422 0 0325
74 10 5782
168 9 6972 342 0 1319 516 0 0333 250 03858 424 0 0145
76 14 2642
170 9 2948 344 0 1319 5 18 0 0324 252 03659 426 00154
78 170585
172 8 7636 346 0 1230 520 0 0289 254 0 3343 428 0 0081
80 18.5997
174 8 1897 348 0 1159 522 0 0200 256 03108 430 0 0045
82 193442
176 7 8904 350 0 1186 524 0 0209 258 02665 432 -0 0072
84 21 3471
178 7 5821 352 0 1133 526 0 0102 260 0.2403 434 •0 0307
86 24.5587
180 7 2153 354 0 1115 5 28 •0 0 0 2 2 262 02231 436 -0 0009
88 25 6871
182 66192 356 0 0955 530 00084 264 0 2096 438 0 0126
90 26 5508
184 6 3687 358 0 0697 532 00191 266 0 2069 440 00181
92 28 4923
186 6 1022 360 0 0449 5 34 0 0191 268 02033 442 00117
94 28.8825
188 5 8562 362 0 0635 5 36 00155 270 02006 444 0 0072
96 29 2457
190 5 6527 364 0 0671 538 0 0262 272 0 1924 446 00117
98 29 7507
192 54138 366 0 0777 540 0 0191 274 0 1680 448 00108
100 30.2982
194 5 .1 5 2 6 368 0 0902 542 00182 276 0 1590 450 001 8!
102 30 8249
196 4 8275 370 0 0911 5 44 0 0138 278 0 1364 452 0 0126
104 31 3389
198 4 4260 372 0 0875 546 0 0076 280 0 1247 454 0 0081
106 31 2477
200 4 1533 374 0 0724 5 48 0 0120 282 0 1265 456 00154
108 31.2070
202 39330 376 0 0724 550 0 0271 284 0 1075 458 0 0063
110 31.1574
204 3 7793 378 0 0688 552 0 0262 286 00967 460 0 0045
112 30 6885
206 35483 380 0 0697 554 0 0076 288 Q.Q88S 462 0 0145
114 30.2594
208 3 3014 382 0 0795 556 0 0173 290 0.0587 464 00181
116 29 9260
210 3 0225 384 0 0777 5 58 00182 292 00660 466 0 0045
118 28.9792
212 28786 386 0 0520 560 00342 294 00632 468 •0 0045
120 28.2492
214 2 7640 388 0 0608 5 62 00466 296 00632 470 0 0036
122 27 2753
216 27000 390 0 0626 5 64 00315 298 0.0425 472 0 0027
124 26.2563
2 18 25597 392 0 0582 566 0 0342 300 0 0515 474 •00117
126 25.3664
315
54 0 01 1 5 315 7 78 9 9 576 0 .0 6 0 3 837 00222 177 2 2 .9 0 7 5

476 0 0027
57 -0 0 1 8 6 318 7 3423 579 0 .0 7 9 8 840 0 .0 0 0 9 180 2 3 5381
478 0 .0 1 1 7
-0 0 0 4 4 6 8974 582 0 1099 843 0 0248 183 23 8427
480 00063 60 321
324 6 58 2 8 585 0 1055 846 0 0319 186 24 0233
482 00018 63 -0 0 1 6 0
327 6 4028 588 0 1055 849 00372 189 2 4 .5 3 3 5
484 -0 0 0 7 2 66 -0 00 3 5
69 00018 330 6 0917 591 00612 852 90647 192 24 8483
486 -0 0244
-0 0071 333 59198 594 00470 8 55 00682 195 2 5 .0 5 8 5
488 - 0 02 0 8 72
336 5 77 9 7 597 00629 858 00780 198 2 5 170 6
490 - 0 02 0 8 75 -0 0 0 1 8
339 600 00620 861 0 .0 8 1 5 201 25 2289
492 •0 .0 2 3 5 78 4 )0 0 3 5 5 4562
5 4305 603 0 .0 5 4 1 864 00886 204 25 3428
494 -0 0235 81 -0 0 1 3 3 342
00550 867 0 .0 7 4 5 207 25 2974
496 -0 0 3 1 6 84 -0 0 0 1 8 345 5 0202 606
4 8899 609 0 .0 7 (8 870 00700 2 (0 2 5 .2 4 0 0
498 -0 0 2 4 4 87 -0 0 0 9 7 348
0 0301 44742 612 0 .0 8 6 0 873 0 .0 7 2 7 213 2 5 .1 3 3 5
500 -0 0298 90 351
00771 87 6 00665 216 24 8650
502 •0 .0 3 0 7 93 0 1055 354 4 4671 615
357 618 0 .0 8 0 7 87 9 00629 219 2 4 .8 4 4 6
504 -0 0 1 4 5 96 0 1578 4 29 4 3
00594 882 00736 222 24 9 2 0 6
506 -0 0 1 3 6 99 0 1587 360 3 9858 621
624 00922 885 00665 225 2 4 6881
508 -0 .0 1 9 9 102 0 3785 363 3 .8 7 5 9
3 5639 627 0 08 5 1 888 0 .0 5 2 3 228 2 4 1538
510 •0 0 1 6 3 105 0 72 3 2 366
369 3 4726 630 0 0425 891 0 0496 231 23 6881
512 -0 0 1 3 6 108 1 5945
3 22 1 8 633 0 .0 2 3 0 894 0 0337 234 23 0251
514 -0 0 1 0 8 III 2 1148 372
636 0 .0 4 4 3 89 7 00346 237 2 2 8241
516 -0.0 1 8 1 114 3 1801 375 3 1021
378 2 9054 639 0 0558 900 0030! 240 2 2 .5 7 2 3
518 -0 0 1 0 8 117 3 1121
3 4141 3 81 2 7503 642 0 0532 243 22 0463
520 •0 0 0 9 0 120
645 0 .0 7 7 ! 246 21 6991
522 0 0000 123 4 0877 384 2 5331
387 24516 648 00869 249 21 3 1 8 5
524 00063 126 4 2668
651 0 .0 7 9 8 252 20 7934
526 00045 129 4 8083 390 23177
5 4881 393 22140 654 0 .0 4 9 6 255 20 4962
528 00108 132
0 0434 258 19 7 6 0 0
530
532
00009
00136
135
138
6 0687
6 8052
396
399
21485
20191
657
660 0 .0 4 6 1 2/16/94 2 61 19 3 9 8 9
190739
141 402 1 9632 663 00133 264
534 00081 7 4132 3 -0 0051
1 8808 666 00248 267 18 5 4 7 0
536 -0 0018 144 8 0939 405 6 -0 0 0 3 2
1 8471 669 0 .0 2 3 9 270 17 7 9 9 7
538 -0 0 1 0 8 147 9 1629 408 9 •0 .0 1 2 5
9 9419 00275 273 17 2 7 3 8
540 4 )0 0 6 3 150 411 1 75 7 6 672 12 00134
1 6849 00080 276 17 0 5 9 9
542 4 )0 0 4 5 153 I I 7625 414 675 15 Q 01 8 1
678 00284 279 16 6 2 8 4
544 •0 0 1 1 7 156 12 0 6 3 8 417 1 6025 18 0 0144
420 1 4775 681 0 .0 2 0 4 282 1 6 .2 6 4 5
546 0 0009 159 12 77 0 2 21 00051
1 4359 684 0 .0 3 6 3 285 15 6 4 8 7
548 0 0018 162 15 9 8 0 9 423 24 -0 0 0 1 4
687 0 0417 288 15 2 3 6 6
550 -0 0018 165 14 79 5 5 426 1 39 3 3 27 0 0236
429 690 0 .0 3 7 2 291 14 7 2 7 3
552 •0 0045 168 15 4221 1 34 6 3 30 0 0208
1 2887 693 00638 294 143616
554 0 0090 171 16 3935 432 33 0 0579
696 00213 297 13 9 2 6 3
556 0 0018 174 16 9 7 5 8 435 1 1930 36 00866
1 1407 699 00133 300 13 4 6 3 4
558 0 0063 177 17 4 8 4 6 438 39 0 0838
4 41 1 0920 702 00328 303 13 0 4 0 2
560 0 0036 180 17 5794 42 0 0S60
705 0 .0 3 9 0 306 123818
562 0 0000 183 180111 444 0 9927 45 0 0523
00319 309 120679
564 -0 0 1 1 7 186 18 5508 447 0 9758 708 48 0 0338
189 450 0 9431 711 00284 312 11 8 7 6 2
566 4 )0 1 3 6 19 0 5 7 8 51 0 0569
714 00097 315 I I 0206
568 4 )0 0 3 6 192 19 4 5 4 0 453 0 9395 54 00449
717 00151 318 10 827 1
570 4 )0 1 9 9 195 19 6 1 3 6 456 0 8925 57 00440
459 720 00292 32! 10 4 0 2 0
572 4 )0 2 7 1 |9 8 19 84 5 8 0 80 9 2 60 00162
0 7472 723 00275 324 1 0 .1 7 7 0
574 4 )0 2 4 4 201 20 0762 462 63 0 0051
0 71 7 0 726 00168 327 9 6455
576 0 000*) 204 20 2641 465 66 00236
729 0 .0 2 4 8 330 9 3899
578 4 )0 0 1 8 2 07 20 2 6 5 0 468 0 70 4 6 69 00468
00195 333 90159
580 00009 210 2 0 1303 471 0 6701 732 72 0 1051
0 0115 336 8 8547
582 4 )0 1 3 6 213 20 2021 474 0 6364 735 75 0 1162
738 00168 339 8 5630
584 4 )0 1 9 9 216 20 0124 477 0 6045 78 0 1199
|0 9 2 4 7 0 5859 741 00071 342 8 2436
586 4 )0 0 2 7 219 480 8! 0 2403
744 00168 345 79287
5 88 0 0000 222 19 5648 483 0 53 8 0 84 0 3977
747 0 .0 3 2 8 348 7 7084
590 4 )0 0 6 3 2 25 19 3432 486 0 4866 87 0 5061
489 750 00230 351 7 4917
592 4 )0 2 1 7 228 18 9 1 0 7 0 4662 90 06172
0 4441 753 00106 354 70481
594 4 )0 1 3 6 231 18 6 2 0 0 492 93 0 8153
756 00239 357 6 7639
596 •0 0361 234 18 0 6 2 5 495 04148 96 1 26 0 7
17 84 0 9 498 04113 759 00363 360 6 4861
598 4 )0 1 4 5 137 99 1 79 0 4
SOI 04175 762 00230 363 6 2175
600 4 )0 1 2 6 240 17 4 9 5 2 102 2.4071
504 765 00399 366 59860
243 17 2 0 2 7 0 4042 105 3 2386
0 .0 4 2 5 369 5 8434
2 46 16 4 8 4 8 507 0 4086 768 108 3 7137
249 15 80 9 4 510 0 35 2 8 771 00514 372 5 3712
11 4 1535
03014 774 00408 375 5 1628
252 15 4 8 5 0 513 114 4 72 4 8
516 0 3031 777 0 .0 4 4 3 378 47971
2 55 15 2945 117 5 6073
519 780 00815 381 4 5637
2 58 14 9 6 5 7 0 2810 120 7 0583
2/17/94 261 14 7 3 0 8
14 26 9 9
522 0 3164
03155
783
786
00718
00771
123 7 6380
384
387
4 4387
4 1767
264 525 126 8 0491
3 0 .0 0 0 0 4 0841
267 13 8693 528 0 2969 789 00576 390
0 0053 129 8 4297
6 3 .8 0 2 6
270 13 4 3 0 5 531 0 2898 792 00558 393
9 4 )0 0 0 9 132 98159
534 0 3022 795 0 .0 5 5 0 396 3 5525
2 73 12 9 5 9 0 135 10 66 1 3
12 4 )0 0 6 2 399 3 4016
276 12 7534 537 0 27 7 4 798 00496 138 12.8809
15 -0 .0 1 0 6 33127
279 12 25 3 5 540 0 2588 801 00434 402
4 )0 0 8 9 141 13 6 2 1 7
18 405 3 2303
282 120957 5 43 0 2065 804 00727 144 14 7 2 5 5
2i 00009
546 0 2349 807 00886 408 3 0544
285 11 70 4 0 147 15 5 0 8 9
24 4 )0 1 5 1 2 .9 6 0 8
288 11 0 6 7 6 549 0 23 4 9 810 00833 411
27 4 )0 1 6 8 ISO 1 6 .3 3 0 2
0 1879 813 00594 414 28627
291 10 6 5 7 2 552 153 17 1571
30 4 )0 3 1 0
294 9 95 4 4 555 0 1462 816 00319 417 27442
33 -0 0 1 3 1 156 1 8.3683
297 9 89 0 5 5 58 0 1409 819 00399 420 24701
36 4 )0 1 9 5 159 19 5 6 3 7
3 00 9 3862 561 0 1480 822 00230 423 24275
39 4 )0 1 5 1 162 204490
303 9 1841 564 0 1427 825 00213 426 23552
42 4 ) 00 5 3 165 21 1425
306 9 09 2 8 567 0 1294 828 00292 429 21043
45 0 0071 168 21 9 8 9 8
309 570 0 1134 831 0 .0 4 0 8 432 20339
4 )0 0 0 9 8 7622 171 226102
48 20747
312 8 3271 5 73 0 1161 834 0 .0 3 9 9 435
51 00053 174 2 2 .7 1 7 6
316
438 20052 699 0 0588 104 0 0350 800 4 1227 1496 00135 2192 •00154
441 1 8293 702 0 0644 112 00098 808 37453 1504 00061 2200 •0 0070
444 t 7700 705 0 0421 120 00107 816 3 6827 1512 00285 2208 •00154
447 1 7441 708 0 0255 128 0 0051 824 3 5342 1520 00416 2216 •0 0079
1 6580 711 0 0310 136 •0 0070 832 30177 1528 0 0248 2224 •0 0079
430
453 1 5691 714 0 0394 144 0 0294 840 3 1120 1536 00061 2232 •00135
456 1.4672 717 0 0468 152 C 1182 848 2-9261 1544 •00135 2240 00135
459 1 4089 720 00181 160 0 5880 856 2.7514 1552 •00107 2248 00154
462 1.2950 723 0 0403 168 09168 864 2.6207 1560 00079 2256 0 0388
465 1 3015 726 0 0523 176 1 0046 872 2-3759 1568 00061 2264 0 0248
468 1 2561 729 0 0681 184 10943 880 2.3367 1576 00304 2272 00107
471 1 1848 732 0 0218 192 1 6697 888 2.3423 1584 00191 2280 0 0210
474 1 1366 735 0 0292 200 38313 896 2.1237 1592 00126 2288 00406
477 1 0829 738 0 0495 208 4 3899 904 1 9266 1600 •0 0481 2296 00079
480 1 0505 741 0 0579 216 54650 912 1.8640 1608 -00145 2304 00322
0 9663 744 0 0727 224 58592 920 1 8491 1616 •0 0238 2312 00117
483
486 09477 747 o 0532 232 6 4935 928 1.6613 1624 •00145 2320 0 0388
489 08848 750 0017! 240 8 2450 936 1 5707 1632 •0 0369 2328 00117
08616 753 0 0505 248 8 6541 944 1 4913 1640 -0 0304 2336 •0 0014
492
495 0 7607 756 0 0644 256 10 3430 952 1 4166 1648 •0 0304 2344 •0 0163
498 06912 759 0 0634 264 11.3079 960 I 3643 1656 -00135 2352 •0 0033
SOI 0.7098 762 0 0634 272 I I 9179 968 1 3456 1664 •00163 2360 0 0229
504 06903 765 0 0282 280 13 2032 976 1 2148 1672 •00126 2368 0 0005
507 0 6542 768 0 0357 288 14 7698 984 1 0345 !680 •00107 2376 •0 0070
0 6209 771 0 0477 296 15 1994 992 1.0794 1688 -0 0079 2384 -0 0229
S10
513 06125 774 0 0347 304 15 8561 1000 1 0224 1696 -0 0079 2392 •0 0051
516 0 5857 777 0 0329 312 17 7757 1008 0.9776 1704 0 0033 2400 •0 0313
519 0 5514 780 0 0069 320 18 2269 1016 09000 1712 -00154
522 0 5070 783 0 0069 328 18 9256 1024 0.7403 1720 00173
525 0 4866 786 0 0273 336 20 4193 1032 0.7692 1728 00182
528 0 480! 789 0 0227 344 21 3431 1040 0.7562 1736 00182
531 0 4848 792 0 0088 352 2! 3515 1048 0 7216 1744 00061
06366 1752 00079
534
537
0.4616
0 4403
795
798
0 0273
0 0273
360
368
21 6476
220867
1056
1064 05983 1760 00098 3/2/94
540 0 4218 801 0 0301 376 22 1539 1072 0 5675 1768 0 0070 8 •0 0246
543 0.3959 804 0 0181 384 22 5033 1080 0.5460 1776 00332 16 -0 0291
546 0 3792 807 -0 0144 392 226509 1088 0 5040 1784 00481 24 •0 0344
549 0 3727 810 -0 0116 400 228928 1096 04965 1792 00238 32 •0 0362
552 0 3292 813 0 0310 408 229292 1104 0.4853 1800 00201 40 •0 0389
555 0 3329 816 0 0338 416 228545 1112 0.4386 1808 0 0397 48 •0 0416
558 03283 819 0 0394 424 22 7359 1120 0 3741 1816 00341 56 •0 0451
561 0 2792 822 0 0523 432 22.5294 1128 03554 1824 00089 64 •0 0514
564 02579 825 0 0607 440 223333 1136 03339 1832 00014 72 -0 0291
567 02755 828 0 0366 448 22 1483 1144 0 3255 1840 00135 80 •00192
570 0 2227 8J1 0 0690 456 21 6374 1152 0 2835 1848 00135 88 -0 0523
573 02218 834 0 0940 464 21 3314 1160 0.3013 1856 0 0210 96 •0 0532
576 0 2412 837 0 0958 472 20 9284 1168 0 2387 1864 0 0079 104 •0 0219
579 0 2542 840 0 0458 480 20 6024 1176 02331 1872 00173 112 •0 0210
582 0 2375 843 0 0162 488 20 0279 1184 02293 1880 0 0145 120 •0 0040
585 0 2264 846 0 0384 496 192152 1192 0 1957 1888 0 0145 128 -0 0407
588 0 2125 849 0 0449 504 18 7033 1200 0 1910 1896 0 0276 136 0 0031
591 0217! 852 0 0468 512 186062 1208 0.1845 1904 00126 144 0 0559
594 0 2384 855 0 0468 520 t8 1195 1216 0 1574 1912 0 0117 152 0 1855
597 0 2070 858 0 0338 528 17 5646 1224 0 1453 1920 00061 160 0 7389
600 0 2005 861 0 0477 536 170845 1232 0 1200 1928 00163 168 0 9955
603 0 2014 864 0 0505 544 160177 1240 0 1144 1936 0 0332 176 1 4264
606 02033 867 0 0477 552 15 7141 1248 0 1387 1944 0 0332 184 21085
609 0 1903 870 0 0329 560 15 3844 1256 0 1191 1952 0 0285 192 3 0454
612 0 1894 873 0 0394 568 15 0930 1264 0 1182 I960 00201 200 4 6851
615 0.1736 876 0 0384 576 14 4578 1272 0 1023 1968 00079 208 6 5652
618 0 1838 879 0 0162 584 13 8525 1280 0 1116 1976 0 0201 216 84757
621 0 1912 882 0 0162 592 129996 1288 0.1303 1984 0 0201 224 9 6477
624 0 1866 885 0 0051 600 124130 1296 0.1210 1992 00294 232 10 2637
627 0 1884 888 0 0097 608 12 3056 1304 0 1107 2000 0 0033 240 127946
630 0 1718 89! 0 0051 616 I I 6040 1312 0.1070 2008 00014 248 13 7646
633 0.1570 894 0 0199 624 10 9548 1320 0 0976 2016 00023 256 16 7524
636 0 0764 897 -0 0171 632 10 7886 1328 00892 2024 -0 0098 264 168570
639 0 0736 900 -0 0255 640 10 4971 1336 0 0724 2032 -00107 272 19 1913
642 00995 648 9 7302 1344 0 0687 2040 0 0033 280 19 3504
645 0 1255 656 9 3808 1352 0 0836 2048 -0.0444 288 203767
648 0 1218 664 88998 1360 0.0752 2056 •0 0481 296 2! 3593
651 0.1208 672 86915 1368 00649 2064 -0 0107 304 23 9751
654 00949 680 8 1833 1376 00724 2072 0 0005 312 24 8817
657 0 092! 688 7 9750 1384 0.0677 2080 •0 0285 320 26 0921
660 0 0764 2/27/94 696 7 3034 1392 00640
00360
2088
2096
•0 0416
•0 0369
328 263755
663 0 0727 8 -0 <)070 704 7 0736 1400 336 274913
666 0 0903 712 6 8344 1408 0.0434 2104 •00500 344 27 7264
16 -0 006!
669 00875 720 64188 1416 00276 2112 •00126 352 28 1877
24 •0 0089
672 0.0477 728 62161 1424 0.0248 2120 -0 0276 360 28 4371
32 •0 0248
675 00366 736 58462 1432 0.0304 2128 -0 0313 368 28 6231
40 -0 0005
678 00514 744 5 4090 1440 00182 2136 -00191 376 28 6133
48 -0 0266
681 0 0329 752 5 3137 1448 00210 2144 -0 0201 384 28 5972
56 -0 0089
684 0 0644 760 5 1484 1456 •0.0023 2152 •0.0033 392 28 6803
64 0 0154
687 0 0931 768 4 8812 1464 -0.0079 2160 00033 400 28 4434
72 0 0369
690 0 1032 776 46122 1472 00005 2168 0 0117 408 28 1752
SO 0 0313
693 0.1125 784 4 1984 1480 0 0023 2176 -0.0042 416 27 7443
88 0 0145
696 0 1440 792 4 1956 1488 0.0126 2184 00042 424 27 2177
96 001 82
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
317
432 26 8735 1128 0 2856 1824 •0 0094 64 -00047 760 82974 1456 0 3775
440 26.3335 1136 0 2517 1832 •00(92 72 •00009 768 8 0168 1464 03468
448 25 8454 1144 0 1953 1840 •0 0049 80 0.0140 776 78164 1472 03486
436 25.1651 1152 0 1712 1848 00013 88 00718 784 7 4388 1480 0 3412
464 24 9031 1160 0 1998 1856 0 0013 96 0 1193 792 7 2533 1488 03319
472 23 9519 1168 0 1694 1864 0 0308 104 03048 800 7 0305 1496 03179
480 23 5773 1176 0 1676 1872 -00112 112 0.7746 808 6.7248 1504 0 2983
488 22.8469 1184 0 1560 1880 -0 0040 120 12706 816 64899 1512 03039
496 22.2425 1192 0 1345 1888 •0 0004 128 1.9548 824 63985 1520 03058
304 214880 1200 0 1202 1896 -0 0121 136 17276 832 6.1515 1528 02852
512 20 7075 1208 0 1211 1904 •00130 144 39320 840 59287 (536 02638
520 20 1050 1216 0 1238 1912 00013 152 5.4440 848 5 7227 1544 0 2731
328 19 6079 1224 0 1140 1920 -0 0040 160 66996 856 56090 1552 02219
536 192324 1232 0 1345 1928 0 0067 168 8.2937 864 5 4356 1560 0 2442
544 184224 1240 0 1059 1936 -0 0013 176 9.7954 872 5 2985 1568 02312
552 175445 1248 00961 1944 00174 184 11 6682 880 50413 1576 0 2097
560 162822 1256 0 0845 1952 •0 0040 192 116470 888 48147 1584 0 1846
568 15 1807 1264 0 0934 i960 -0 0076 200 13.9259 896 4 6712 1592 0 2079
576 144879 1272 0 0693 1968 -0 0040 208 15 7884 904 4 5081 1600 0 1911
584 13 7611 1280 0 0890 1976 •0 0308 216 165332 912 43160 1608 0 1725
592 13 5116 1288 0 0890 (984 •0 0201 224 17 9670 920 42265 1616 0 1725
600 12.7025 1296 0 0755 1992 •0 0299 232 19 2692 928 4 0792 1624 0 1659
608 12.1205 1304 0 0729 2000 •0 0004 240 202452 936 3 9142 1632 0 1445
616 11 5725 1312 0 0550 2008 0 0282 248 21 4151 944 3 7558 1640 0 1585
624 11 0728 1320 0 0371 2016 •00112 256 211758 952 36616 1648 0 149!
632 10 4335 1328 0 0434 2C24 -0 0273 264 218320 960 3 5274 1656 0 1482
640 9 8015 1336 0 0344 2032 0 0058 272 23.9227 968 13978 1664 0 1342
648 8 8985 1344 0 0353 2040 •0 0058 280 24 7766 976 3 2729 1672 0 1510
656 8 7233 1352 0 0505 2048 00317 288 25 1811 984 3 1303 1680 0 1268
664 8.1923 1360 0 0362 2056 0 0076 296 25.7759 992 3 0203 1688 0 1277
672 7 9080 1368 0 0443 2064 •0 0049 304 260304 1000 2.9271 1696 0 1314
680 7 5986 1376 0 0693 2072 -0 0425 312 26.5440 1008 2.8208 1704 0 1650
688 7 2527 1384 0 0764 2080 -0 0353 320 26.6820 1016 2.6819 1712 0 1268
696 7 0551 1392 0 0630 2088 •0 0201 328 270865 1024 2.5514 1720 0 1165
704 6 7163 1400 0 0639 2096 •0 0210 336 27 1192 1032 14973 1728 0 1100
712 63104 1408 0 0371 2104 -0 0237 344 27 4277 1040 13836 1736 0 1147
720 5 7963 1416 0 0237 2112 -0 0291 352 273941 1048 13333 1744 0 1081
728 5 5379 1424 0 0577 2120 •0 0389 360 274454 1056 12540 1752 0 1044
736 5 3878 1432 0 0746 2128 -0 0291 368 27 3391 1064 11580 1760 00848
744 5 0024 1440 0 0434 2136 -0 0326 376 27 2814 1072 10797 1768 0 0914
752 4 5465 1448 0 0595 2144 •0 0434 384 27 0903 1080 19921 1776 0 0680
760 4 4553 1456 0 0273 2152 -0 0425 392 26 8814 1088 1 9660 1784 0 0942
768 40557 1464 0 0308 2160 •0 0210 400 26 6111 1096 1 8541 1792 00969
776 3 7705 1472 0 0353 2(68 -0 0237 408 26 2504 1(04 I 8289 1800 0 1016
784 35640 1480 0 0344 2176 -0 0255 416 25 9465 1112 1 7572 1808 0 0988
792 35068 1488 0 0630 2184 •00156 424 25 6388 1120 1 6928 1816 0 0802
800 3 2761 1496 0 0550 2192 •0 0425 432 252697 1128 1 6472 1824 0 0839
808 3 1250 1504 0 0514 2200 •0 0451 440 24 8036 1136 1 5894 1832 0 0792
816 2.8961 1512 ■0 0013 2208 •0 0058 448 24.4615 1144 1 4962 1840 0 0783
824 27647 1520 0 0192 2216 •00353 456 24 0439 1152. 1 4207 1848 00690
832 24813 1528 0 0085 2224 -0 0550 464 23 7288 1160 1 3694 1856 0 0503
840 23964 1536 0 0013 2232 •0 0603 472 23 2646 1168 1 3405 1864 0 0625
848 22793 1544 0 0165 2240 -0 0595 480 214685 1176 1 3200 1872 0 0550
856 21443 1552 0 0067 2248 -0 0586 488 211515 1184 1 2678 1880 0 0830
864 20030 1560 0 0130 2256 -0 0514 496 21.6528 1192 I 2379 1888 0 0755
872 1 8966 1568 00004 2264 •0 0505 504 21 1979 1200 1 1680 1896 0.0373
880 I 8510 1576 0 0416 2272 -0 0237 512 20.8222 1208 1 1177 1904 0 0401
888 1 6481 1584 00183 2280 •0 0505 520 20.2191 1216 1 0729 1912 0 0485
896 I 5855 1592 0 0335 2288 •0 0541 528 19 7745 1224 1 0142 1920 0 0569
904 1 4711 1600 0 0174 2296 -0 0523 536 19 4584 1232 0 9564 1928 0 0597
912 1 3531 1608 0 0058 2304 •0 0085 544 188898 1240 0 9527 1936 0 0336
920 1 3120 1616 -0 0049 2312 •0 0344 552 183436 1248 0 9275 1944 0 0261
928 1 2154 1624 0 0058 2320 0 0058 560 17 9399 (256 0 8716 1952 0 0466
936 1 1975 1632 •0 0004 2328 00040 568 17 4859 1264 0 8278 I960 0 0S69
944 I 1314 1640 0 0076 2336 -0 0013 576 16.9509 1272 08231 1968 00419
952 1 0867 1648 -0 0067 2344 0 0058 584 165715 1280 0 8035 1976 0 0531
960 0 9830 1656 -00192 2352 00013 592 16 1240 1288 0 7383 1984 0 0298
968 0 9418 1664 -0 0121 2360 •0 0031 600 156915 1296 0.7206 1992 00186
976 08641 1672 -0 0076 2368 -0 0478 608 152282 1301 0 7196 2000 0 0224
984 0 7952 1680 -0 0004 2376 -0 0487 616 149355 1312 06907 2008 -0 0019
992 0 7469 (688 0 0058 2384 -0 0559 624 14 4209 1320 06693 2016 -00149
1000 0 7317 1696 00040 2392 -0 0746 632 14 0658 1328 06124 2024 •0 0280
1008 0 64S0 1704 00004 2400 •0 0639 640 13 5465 1336 0 5919 2032 0 0280
1016 0 6397 1712 0 0067 648 132678 1344 0 5686 2040 0 0410
1024 0 5923 1720 00165 656 117775 1352 05733 2048 0 0457
1032 0 5619 1728 00094 664 114130 1360 0.5472 2056 •0 0028
1040 0.4877 1736 00094 672 110644 1368 0.5220 2064 0 0214
1048 0 4778 1744 0 0022 680 114575 1376 04941 2072 0 0336
1056 0 4582 1752 00183 688 109895 1384 05090 2080 00121
1064 0 4412 1760 -0 0013 3/4/94 696 10.6092
103799
1392
1400
0 4941
0 4745
2088
2096
0 0261
4)0047
1072 03974 1768 -0 0013 g -0 0009 704
1080 0 3464 1776 4)0004 712 10.1869 1408 04502 2104 0.0261
16 00121
1088 0 3482 1784 0 0031 720 99091 1416 04409 2112 00056
24 00214
1096 03437 1792 -0 0(03 728 96649 1424 04474 2120 -00131
32 0 0354
1104 0.3375 1800 •0 0067 736 9.4141 1432 0.4363 2128 •00112
40 0 0354
1112 0 3232 1808 00004 744 8.9742 1440 04186 2136 •00121
48 0.0158
1120 0.3035 1816 -0 0103 752 8 6619 1448 04130 2144 00280
56 00009
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
318
2152 -0.0009 392 24 5130 1088 13237 1784 00271 24 00439 720 75989
2160 00149 400 24 1826 1096 12703 1792 00299 32 00211 728 72861
2168 •00112 408 23 5367 1104 1 2179 1800 00299 40 0 0341 736 7 1073
2176 00103 416 23 4557 11(2 t 1774 1808 0.0308 48 0 0219 744 6.7563
2184 •0.0019 424 23 1658 1120 1 123! 1816 0.0336 56 00796 752 6 4972
2192 -0 0093 432 22 9367 1128 1 079S 1824 0.0170 64 00487 760 6 1836
2200 •0 0009 440 22 5346 1136 1 0154 1832 0.0170 72 00309 768 5 7749
2208 00047 448 22 1665 1144 0 9887 1840 00087 80 00138 776 5 5385
2216 00009 456 220018 1152 09446 1848 00189 88 00162 784 53695
2224 00056 464 21 5417 1160 09151 1856 •00087 96 0 0366 792 5 1079
2232 0 0252 472 21 1985 1168 0 8903 1864 •0.0032 104 00146 800 4 9966
2240 00103 480 20 7265 1176 08360 1872 0.0179 112 0.0975 808 46098
2248 0 0224 488 20 4320 1184 07909 1880 •00087 120 00861 816 44815
2256 00019 496 20 0041 1192 0 7660 1888 0.0133 128 0 1519 824 42304
2264 00047 504 19 5726 1200 0 7200 1896 0 0005 136 02998 832 4 1581
2272 00186 512 19 2790 1208 06924 1904 0.0014 144 0 5874 840 39883
2280 0 0317 520 18 4914 1216 0 6391 1912 •0 0198 152 0 8677 848 3 8128
2288 0 0028 528 17 9908 1224 0 6225 1920 00032 160 1 2455 856 3 5837
2296 0 0075 536 17 6761 1232 0 5839 1928 -00198 168 1 8345 864 3 4619
2304 00103 544 17 2224 1240 0 5636 1936 -0 0336 176 2.2570 872 3 2969
2312 -0 0028 552 16 9436 1248 0 5719 1944 •0.0299 184 3 1369 880 3 1166
2320 0 0037 560 164311 1256 0 5213 1952 •0.0143 192 3 9030 888 30231
2328 00000 568 16 0041 1264 0 4928 I960 •0.0271 200 5 6847 896 28793
2336 4)0075 576 15 3563 1272 0 4605 (968 •0.0327 208 6.6662 904 2 7445
2344 4)0112 584 14 8981 1280 04311 1976 -00170 216 8 3049 912 25673
2352 -0 0084 592 14 5825 1288 0 4293 1984 00124 224 9 5715 920 24950
2360 -0 0 U 2 600 14 0212 1296 0 3796 1992 00124 232 10 7877 928 24187
2368 0 0168 608 13 3844 1304 0 3915 2000 00124 240 11 7367 936 23041
2376 00084 616 13 1277 1312 0 3501 2008 0 0262 248 IZ 7 0 I9 944 22253
2384 0 0270 624 12 7780 1320 03133 2016 0 0244 256 133697 952 21733
2392 00606 632 12 3621 1328 0 3060 2024 •0 0014 264 139693 960 20661
2400 00625 640 12 0133 1336 03023 2032 00005 272 15 5755 968 1 9808
648 I 7879 1344 0 2995 2040 00051 280 17 0858 976 1 9011
656 11 1162 1352 03124 2048 0 0262 288 18 1355 984 I 7720
664 10 9128 1360 0 3041 2056 •0.0023 296 18 7497 992 1 6736
672 10 4214 1368 02866 2064 00060 304 19 8717 1000 1 5916
680 10 1279 1376 0 2461 2072 •0.0087 312 20 9206 (008 1 5128
688 9 8868 1384 0 2415 2080 0.0244 320 219541 1016 I 4380
3/6/94 696 9 6292 1392 02185 2088
2096
0 0189
00244
328
336
228234
23 4628
1024
1032
1 3544
1 2569
00336 704 92151 1400 0 2075
8 1040 1 2008
712 8 8038 1408 0 2139 2104 •0 0041 344 23 7195
16 0 0400
720 8 5618 1416 0 2342 2112 00041 352 242135 1048 I 1155
24 0 0363 1056 10984
728 8 2323 1424 0 2075 2120 •0 0106 360 24 5263
32 0 0511 1064 I 0237
736 7 9535 1432 02112 2128 •0.0032 368 25 1121
40 00124 0 9879
744 7 4806 1440 0 1928 2136 -00179 376 25 1267 1072
48 00189
752 7 3619 1448 0 1689 2144 -0 0354 384 25 1478 1080 0 9441
56 00345 1088 0 8961
760 7 0867 1456 0 1458 2152 •0 0060 392 25 3176
64 00244 08604
768 6 8162 1464 0 1698 2160 0.0290 400 25 3802 1096
72 00244 0 7930
776 6 6193 1472 0 1689 2168 -00161 408 25 1421 1104
80 0 0354 0 7759
784 6 2687 1480 0 1753 2176 000 4! 416 25 0178 1112
88 0.0465 1120 0 7531
792 6 1619 1488 0 1412 2184 00032 424 24 8797
96 00612 24 5019 1128 07117
800 5 8758 1496 0 1403 2192 •0 0005 432
104 0 0824
808 5 5160 1504 0 1468 2200 00087 440 24 17B5 1136 0 6540
112 0 0952 1144 05988
816 5 2997 1512 0 1661 2208 -0 0087 448 23912!
120 0 2581 0 5923
824 5 1157 1520 0 1274 2216 •0 0078 456 23 4953 1152
128 05516 05606
832 4 8535 1528 0 1100 2224 •00152 464 23 0005 1160
136 1 0642 0 5533
840 4 5516 1536 0 1054 2232 -0 0290 472 223700 1168
144 1 6963
848 4 4863 1544 0 1155 2240 •00216 480 21 9654 1176 05086
152 2.2337 04907
856 4 3446 1552 00916 2248 -0.0225 488 21 1579 1184
160 2 7840
864 4 1017 1560 0 1026 2256 •0 0225 496 20 8377 1192 0 4720
168 3 7953 04444
872 3 9655 1568 0 1008 2264 0.0014 504 19 9522 1200
176 5 1139
880 3 8054 1576 0 0824 2272 •0.0400 512 195232 1208 04168
184 60644
888 3 6720 1584 0 0805 2280 •0 0281 520 190154 1216 0 4298
192 66552 1224 0 4314
896 3 5312 1592 0 0897 2288 -0 0041 528 18 8830
200 78311 04306
904 3 3959 1600 0 0952 2296 -0 0345 536 18.0884 1232
208 87256 0 4249
912 3 3158 1608 0 1090 2304 •00133 544 17 7992 1240
216 9 8463
920 3 1603 1616 0 0925 2312 0.0005 552 168705 1248 0 4062
224 1) 6315
928 29H 0 1624 0 0870 2320 0.0060 560 164408 1256 04135
232 12.8847
936 2 8098 1632 0 0851 2328 •0.0317 568 16.1353 1264 0 3794
240 140874
944 2 7536 1640 0 0962 2336 •0.0031 576 15.3805 1272 03754
248 15 2468
952 2 7122 1648 0 0722 2344 00060 584 14 5478 1280 0 3022
256 16.9786
960 2 6690 1656 0 069S 23S2 •0 0023 592 13 9782 1288 0 2705
264 179015
968 2 5438 1664 0 0778 2360 •0 0051 600 13 6971 1296 0 2681
272 19 0499
976 2 4178 1672 0 0759 2368 00078 608 13 4201 1304 02592
280 199222 0 2291
984 2 3092 1680 00888 2376 00)33 616 129293 1312
288 20 6584
992 2 1767 1688 0 0548 2384 •00005 624 125272 1320 0 2332
296 21 2537
1000 2 1362 1696 0 0548 2392 0.0032 632 11 5076 1328 0 2397
304 21 8629
1008 2 0727 1704 0 0566 2400 •0.0133 640 I I 3930 1336 0 1917
312 22.4453
1016 1 9558 1712 0 0363 648 10 8982 1344 0 1836
320 23 1244
1024 1 7976 1720 00198 656 104164 1352 0 1690
328 239508
1032 1 7304 1728 0.0253 664 9.9517 1360 0.1495
336 240557
1040 1 6568 1736 00161 672 96462 1368 0.14U
344 24 1661
1048 1 5804 1744 0 0244 680 93781 1376 0.1479
352 243796
1056 1 5436 1752 0 0207 688 8 7793 1384 0 1349
360
368
24.6262
24 7467 1064 1 4957 1760 0 0262 3/9/94 696 8.4869
8 2886
1392
1400
0 1373
0 1332
1072 1 4516 1768 0 0327 0.0707 704
376 24 8304 8
1080 1 3752 1776 0 0290 712 8 0644 1408 0 1251
384 24 7053 16 00479
319
1416 0 I I 13 2112 -0 0325

1424 0 1276 2120 •0 0496
1432 0.1235 2128 -0 0252
1440 0 1284 2136 •0 0081
1448 0 1316 2144 •0 0081
1456 0 1414 2152 -0 0171
1464 0 1430 2160 •0 0268
1472 0 1072 2168 -0 0057
1480 0.1072 2176 0 0032
1488 0 1024 2184 0 0097
1496 0 1129 2192 00138
1504 008 29 2200 -0 0252
1512 0 0804 2208 •0 0008
1520 00585 2216 •0 0024
1528 0 0626 2224 •0 0081
1536 0 0374 2232 -0 0171
1544 00406 2240 -0 0081
1552 0 0431 2248 •0 0252
1560 00154 2256 •0 0195
1568 0 0317 2264 -0 0203
1576 0 0260 2272 •0 0106
1584 0 0252 2280 -00073
1592 00260 2288 -0 0122
1600 00154 2296 4)0292
1608 -0 0097 2304 •0 0106
1616 0.0073 2312 0 0016
1624 0 01 7) 2320 •00146
1632 0 0479 2328 •00130
1640 0.0455 2336 •0 0049
1648 00691 2344 -0 0162
1656 0.0552 2352 -0 0268
1664 0 0320 2360 -0 0431
1672 00439 2368 0 0032
1680 0 0041 2376 0 0195
1688 001 06 2384 •0 0081
1696 0 0268 2392 0 0114
1704 0 0398 2400 •0 0284
1712 0.0309
1720 0 0301
1728 0 0341
1736 0 0276
1744 0 0357
1752 0 0244
1760 0 0032
1768 0 0 3 (7
1776 0 0374
1784 0 0325
1792 0 0179
1800 0 0236
1808 0 0000
1816 -0 0032
1824 -0 0065
1832 -0 0024
1840 -0 0422
1848 -0 0114
1856 -00146
1864 -00114
1872 0 0057
1880 -0 0325
1888 -0 0041
1896 -0 0349
1904 -0 0219
1912 4)0073
1920 0 0073
1928 -00260
1936 -0 0073
1944 0 0211
1952 -0 0341
I960 -0 0357
1968 -0 0317
1976 -0 0455
1984 -00292
1992 -0 0203
2000 -00195
2008 -0.0219
2016 -00179
2024 4) 0292
2032 0 0171
2040 4)0203
2048 0 0081
2056 0 0195
2064 4)0162
2072 4)0349
2080 4)0081
2088 -0 0236
2096 0 0024
2104 4)0471
320
iTest 1-0209 1 P ort Vmn VsMP Abs.1 Abs.2 Abs_3 A vfcO j S td-D er. D.O.1 D .0 2 A vr.D .O .
in 12 8 1.66 1.38 1.42 13.6623 0.472
W ater Q l (gpm ): 14 OfT 10 10 0.578 0572 0.573 0.0533 0.007
Gas Qc (cfh): 9 In 12 8 1.42 1.66 1.38 13.6623 0.422
Abs. Indigo I: 0.6 1 6 4 0.219 0.21 02 3 0.7290 0.052 1.28 1.28
Abs. Indigo II: 5.87 2 6 4 0.209 0.22 0.23 0.7290 0.055 157 1.37
W ater Tem o. *C: 0.6 3 6 4 0.216 0.23 0.235 0.6909 0.051 153 1.33
Gas T em o."C : 5.87 ______ <L_ ____ i - 2 2 V - - J U 6 & - -2 2 5 2 - 2 2 5 7 2 - ___ 0381__ _ ia « .__________ - - L 3 8 ___
Pressure, psi: 10.000 5 5 5 0.18 0.176 0.191 0.4890 0.032 15 1.50
6 5 5 0.154 0.141 0.178 05915 0.078 1.43 1.43
O j Dose (m g/L): 1.09509 7 5 5 0.216 0248 0.265 0.2369 0.103 155 1.55
Feed O j (% ): 0.580 8 5 5 0.29 0.292 0.289 0.0402 0.006 1.78 1.78
Transfer EHIcL: 99.6096 9 5 5 0.289 0.288 0.292 0.0429 0.009 1.9 1.90
10 5 5 0.287 0.286 0.285 0.0582 0.004 2.1 110
O ther Parameters: 11 5 5 0.285 0.29 0.287 0.0526 0.010 25 130
O.D. (m g/L): — U — ______ S__ — 5 — 2 2 V - __ Q 2 8 i_ - 2 2 V -2S2522- — O J U 1 -. -2 ,5 5 .__________ — 2.5$— .
Kd (m in 1): 13 4 6 0.229 0229 0.0381 0.000 19 190
K m (m in 1): 14 4 6 0.233 02X2 -0.0606 0.120 3.13 3.13
Dispersion #: 15 4 6 0.231 0.23 0.0329 0.002 35 3.30
16 4 6 0.234 0.231 0.0260 0.007 3.61 3.61
17 6 0.231 0.226 00398 0.012 3.93 2.91
ITest 2 -0 2 1 0 1 P ort V in o V-iMF Abs.1 Abs.2 A bs5 A ve.O i S id. Dev. D .O .I D .0 2 Ave-D.O .
In 15 6 0 .6 0 9 0 .7 6 1 053 275939 1 .1 3 6
W ater Q l (g p m ): 14 o ir 10 10 0 .5 7 1 0572 0559 0 .0 4 7 1 0 .0 1 5
GasQc; (cfh ): 4 .3 5 In 15 6 0536 0 .1 5 8 0 .5 0 8 2 7 .8 2 5 1 1 .1 0 4
Abs. Indigo 1: 0 .5 9 1 6 4 0 .2 0 4 0 .1 9 2 0 .1 9 8 0 .8 1 0 4 0 .0 3 1 25 150

Abs. Indigo II: 5 .8 4 2 6 4 0 .1 9 2 0219 0 .2 1 7 0 .7 5 1 5 0 .0 7 8 279 179
W ater Temo. *C: 13 3 6 4 0 .2 5 1 0 .2 4 7 0 .1 6 1 0 .6 9 7 8 0264 178 178
Gas Temo. "C : 2 0 .5 - - i - - ______ f t - - ____ — 2256. -J L 2 S I- -2 2 5 9 .-0 5 5 7 1 - — a m . . -2 ,8 ft.________ ____2.86—
Pressure, psi: 9 .0 0 0 5 5 5 0213 0246 0 .2 3 9 02590 0 .0 7 2 29 190
6 5 5 0 .2 3 1 0 .2 0 7 0 .2 3 4 02951 0 .0 6 2 3 .0 3 3 .0 3
O j Dose ( m g /L ): 1 .0 7 79 7 7 5 5 0 .2 8 2 0288 0 .2 7 6 0 .0 5 4 0 0 .0 2 5 3.21 3.2 1
Feed O j (% ): 1.273 8 5 5 0 .2 9 0 .2 8 8 0 .2 9 2 0 .0 2 0 8 0 .0 0 8 3 .3 3 3 .3 3
T ransfer EfTkrL: 9 9 8307 9 5 5 0 .2 9 0288 0 .2 8 8 0 .0 2 6 3 0 .0 0 5 3 .4 7 3 .4 7

10 5 5 0 .2 8 5 0283 0 .2 8 3 0 .0 4 7 1 0 .0 0 5 3 .6 5 3 .6 5
O ther Parameters: 11 5 5 0 .2 8 7 0285 0 .2 8 4 0 .0 4 0 2 0 .0 0 6 3.91 3.91
O .D. (m g/L): __U _________5 _ _ 5 — 22 8 8 . -JL233.- - 2 2 ® . _ fl4 R 9 1 _ — J5S0Q... -1 2 9 . ________ ____ 1.29--.
Kd (m in 1): 13 4 6 0 .2 2 2 0227 0 .0 3 9 8 0 .0 1 2 4.1 4 .1 0
KlM(m in 1): 14 4 6 0 .2 2 4 0 .2 3 1 0 .0 2 9 4 0 .0 1 7 45 4 .3 0
Dispersion 1: 15 4 6 023 0 .2 2 9 0 .0 2 2 5 0 .0 0 2 4 .4 9 4 .4 9
16 4 6 0 .2 2 8 0228 0 .0 2 7 7 0.000 4 .9 4 .9 0
17 4 6 0 .2 3 0 .2 2 7 0 0260 0 .0 0 7 5 1 5 10
ITest 3-02 IOC 1 Port Vmn Vssir Abs.1 Abs2 A bsJ A v tO j Std-Dev. D .O .I D .0 2 Ave.D .O .
In 17 1.015 0593 0.413 53.4557 2.183
W ater Q l (gpm ): 14 OIT 10 10 0.689 0.693 0.685 0.0374 0.008
GasQ n (cfh ): 2.15 In 17 0.804 1.014 0.184 53.4780 2372
Abs. Indigo 1: 0.707 1 6 0.263 0271 0.285 0.7855 0.058 4.05 4.05
Abs. Indigo II: 8.24 2 6 0.318 0293 0.328 05777 0.094 425 4.25
W ater Temo. *C: 12 3 6 0.326 0556 0.313 0.4807 0.1 IS 4.39 4.39
Gas Temo. "C : 18.8 — i — ______ f t - - . - iU S L — Q v U i- - 2 2 V i. — & P fil_ . __4J___________ __ i3 P ___
Pressure, psi: 8.000 5 5 0.297 0286 0.329 02057 0.093 4.4 4.40
6 5 0.321 0.298 0.298 0.1988 0.055 4.42 4.42
O ] Dose (m g/L): 1.01399 7 5 0.341 0.35 0.344 0.0353 0.019 459 4.59
Feed Os (% ): 2.568 8 5 0.353 0.35 0.347 0.0145 0.012 4.8 4.80
T ransfer EfTlcL: 999301 9 4 6 0.31! 0.316 0.284 -0.0723 0.060 5.05 5.05
10 4 6 0.283 0.28 028 0.0062 0.006 5.15 5.15
O tb cr Parameters: 11 4 6 028 0277 0.285 0.0074 0.014 5.4 5.40
O .D . (m g/L): — 52— ______ 4_ _ __ ft ___ 0,28- — <U 7ft_ -22 13 .- O Q lftft- — P J » Z _ . - J J ___________ — 5.59—
K d (m in 1): 13 3 7 0.209 0.209 0.0092 0.000 5.8 5.80
K lb (m in 1): 14 3 7 0.209 0212 0.0047 0.006 5.9 590
Dispersion ff: 15 3 7 0.21 0213 0.0018 0.006 5.95 5.95
16 3 7 0214 0214 -0.0056 0.000 622 6.22
17 3 7 0.20ft 0.21 0.0122 0 008 — 622 _ . .. 6.32
321
ITest 4-0211 1 Port Vnm VsMP Abs.1 Abs3 A bsJ A rf.O i Std-Der. D.O.1 D .0 3 A re.D .O .
In 10 12 1.67 1.49 134 6.9668 0.161
W ater Q l (gpm): 13.8 OfT 10 10 0.673 0.688 0.695 0.0450 0.023
Gas Qt; (cfh l: 17 12 In 10 12 1.28 1.25 1.21 73439 0.322
Abs. Indigo 1: 0.707 I 6 4 0.299 0.278 0.279 0.7214 0.062 0.75 0.75
Abs. Indigo II: 5.59 2 6 4 0.292 0.297 0.296 0.6712 0.014 0.69 0.69
W ater T em o.'C : 12 3 6 4 0.319 O J Il 0.314 03690 0.021 0.65 0.65
Gas Temo. *C: 16.25 __ 3 __ ______ 6— ____ i - - 0 J U L __<U 21_ -0 2 1 4 ,_ 0 3 3 M _ ___ Q 0 1 L -. -0 0 1 __________ ____0 .6 1 -_ .
Pressure, psi: 13 000 5 5 5 0.205 0.214 0.229 0.5714 0.050 063 0.63
6 5 5 0.219 0.231 0.224 03354 0.025 0.6 0.60
O j Dose (m g/L): 1.12048 7 5 5 0.26 0.255 0.257 03997 0.010 0.7 0.70
Feed Os (% ): 0.274 8 5 5 0.305 0321 0.317 0.1628 0.035 0.75 0.75
Transfer EITid.: 99 3785 9 4 6 0.266 0.251 0.258 0.0847 0.026 0.9 0.90
10 4 6 0.272 0.264 0.275 0.M 32 0.020 0.85 0.85
O ther Parameters: II 4 6 0.284 0383 0.287 •0.0065 0.007 0.97 0.97
O .D. (m g/L): __ 12__ ______ 4__ 6 . 0 J 8 L . .J iJ S L . _ 0.285 - 0 0 0 5 3 - ___Q0.1I___ _ J J __________ ____L IP ___
Kd (m in 1): 13 r 3 7 0.22 0.209 -0.0071 0.023 13 1.20
Kea (m in 1): 14 3 7 0.212 0314 -0.0027 0.004 1.63 1.63
Dispersion f : 15 3 7 0.209 0314 0.0018 0.010 1.89 1.89
16 3 7 0.216 0312 -0.0056 0.008 112 112
17 3 7 Q JM 0.214 0.0077 0.019 3.45 3.45
ITest 5-0212 1 P ort Abs.1 Abs3 A hsJ A rt.O i Std-Der. D .O .I D .0 3 A re.D .O .

In 15 6 0.711 0.148 0.607 25.9810 1356
W ater Q l (gpm): !3.9 O ff 10 10 0.674 0.676 0.624 0.1122 0.061
GasQ c (cfh): 8.6 In 15 6 0.258 0.66 0.368 26.1368 1.210
Abs. Indigo I: 0.712 I 6 0.089 0.121 0.131 1.6287 0.114 13 1.30
Abs. Indigo II: 5.49 2 6 0.177 0.147 0.127 1.4383 0.131 1.41 1.41
W ater Temo. "C: 12 3 6 0.208 0.224 0.227 1.0781 0.053 131 131
Gas Temo. "C : 16.3 - - 4 - - ______ ft— — _ 3 — a a . __0 2 2 2 - -0305 ._ 10071 __ .J im ... - 1 1 1 __________ - - L 6 5 ___
Pressure, psi: 13.000 5 5 0.154 0.097 0.15 0.9240 0.132 1.69 1.69
6 5 0.11 0.117 0.097 1.0306 0.0*2 1.6 1.60
O j Dose (m g/L): 2.01625 7 5 0.23 0.221 03 03777 0.064 1.61 1.61
F c c d O j(% ): 1.019 8 5 0.261 0.276 0.203 0.4544 0.160 1.83 1.83
Transfer E iffe l; 99.5707 9 4 6 0.247 0.247 0.251 0.1263 0.008 2.05 105
10 4 6 0.276 0.284 0.278 0.0189 0.014 2.1 110
O ther Parameters: 11 4 6 0.283 0381 0.286 0.0051 0.009 239 129
O.D. (m g/L): - . U - - _______4— ____fi --0287_ _0282- -0283 .-00062__ .0 0 0 0 .-. - 2 0 2 .__________ -_ 1 4 2 _ _ .
K d (m in ')-. 13 3 7 0319 0.215 -0.0101 0.008 2.72 172
K u (m in 1): 14 3 7 0.212 0.21 0.0077 0.004 3.03 3.03
Dispersion #: 15 3 7 0.212 0.215 0.0003 0.006 3.12 3.12
16 3 7 0.212 0312 0.0047 0.000 3.49 3.49
17 3 7— 0313 0.209 0.0077 0.008 4 4.00
iTest 6-0213 1 P ort V ino Vsnir Abs.1 Abs3A bsJ A v r.O j Std-Der. D.O.1 D .0 3 A rg,D .O .
In 17 5 0.408 0.0940.313 563958 1.138
W ater Q l (gpm): 13.9 O ff 10 10 0.669 0.6710.658 0.0873 0.015
GasQu (cfh): 424 In 17 5 0.486 1.1 0.8 54.4459 3551
Abs. Indigo I: 0.708 I 6 4 0.106 0.1370.117 1.5834 0.082 237 2.37
Abs. Indigo II: 834 2 6 4 0.205 0.1490.164 13098 0.151 2.62 362
W ater Temo. *C: 115 3 6 4 0.214 0.1980.215 1.1210 0.050 2.76 376
Gas Temo. *C: 17 — i . . ______ ft— . . . i __ 0 2 6 7 - __Q 258_ --Q .1 4 - 0 0 8 2 1 - — J U C L -. - 2 0 1 - ____________15?___
Pressure, psi: 9.000 5 5 5 0.211 0.176 0.18 0.68S7 0.080 2.82 382
6 5 5 0.189 0.136 0.151 0.8118 0.114 2.8 380
Os Dose (m g/L): 107074 7 5 5 0.251 0.266 0.277 03713 0.054 2.88 388
Feed Os (% ): 2.574 8 5 5 0.254 0.281 0.271 0.3546 0057 3.2 3.20
T ransfer E ffk i.: 99.8397 9 4 6 0.277 0.266 0.276 0.0353 0.021 3.3 3.30
10 4 6 0.281 0.277 0.278 0.0157 0.007 3.4 3.40
O ther Parameters: II 4 6 0.278 0.278 0.283 0.0122 0.010 3.48 3.48
O .D. (m g/L): — 12__ ______ 4 — 6 . 0 2 8 6 . . . U S l . -0 2 8 3 . - 0 0 0 0 5 - _ _ 0 0 0 9 — . -2 6 7 __ ____________167___
Kd (m in S: 13 3 0.208 0.21 0.0101 0.004 3.92 3 92
K ls (m in '): 14 3 0.211 0.21 0.0056 0.002 4.5 4.50
Dispersion 8: 15 3 0.214 0.212 -0.0018 0.004 4.3 4.30
16 3 0.212 0313 •0.0003 0.002 485 4.85
17 3 0.213 0.208 00056 . . 0010 5.33 5.33
322
ITest 7-0214 1 Port Vism V*MT Abs.1 A b sJ A bsJ A r tO i StrLDee. D.O.1 D.O-2 A re.D .O .
In 8 13 0.946 0.229 0.416 26.8783 0.476
W ater Q l (cpm l: 13.9 OIT 10 10 0.903 0.907 0.891 0.4565 0.017
GasQu (cfh): 17 In 8 13 0.01 0.07 0.06 27.1875 0.454
Abs. Indigo I: 1 12 1 7 3 0.206 0.212 0.226 3.9434 0.071 055 0.55
Abs. Indigo I I: 21.55 2 7 3 0.263 0.207 0.259 3.7472 0.216 05 050
W ater Temo. "C : 11.8 3 7 3 0.277 0.301 0.328 33385 0.177 055 0.55
Gas Temo. *C: 16.2 __ i ____ 4 __ QAIX. . j u a t L . -JM7Z8 ._3£25i_ ___ flJDJ.— —P5 _______ __Q.JP—
Pressure, psi: 1X000 5 6 4 0.101 0.112 a 124 29074 0.060 0.6 0.60
6 6 4 0.122 0.089 0.112 2.9316 0.088 055 0.55
O j Dose (m g/L): 4.14584 7 6 4 0.203 0.173 0.19 25108 0.078 05 9 059
Feed Os (% ): 1.093 8 6 4 0.29 0.235 0.309 20468 0.200 0.62 0.62
Transfer E fllcL : 98.3211 9 5 5 0.206 0.173 0.178 15557 0.074 0.8 0.80
10 5 5 0.212 0.243 0.239 13659 0.070 0.82 0.82
O ther Parameters: tl 4 6 0.156 0.167 0.147 1.0089 0.035 0.97 0.97
O.D. (m g/L): __U __ ____ 4__ __ fi _ -_QJJ22_ -£'237. ,_ £ J 3 8 S _ .—UR._________ __ LQf*___
Kd (m in 'i: 13 3 7 0.168 0.165 05032 0.006 1.25 1.25
Km (m in '): 14 3 7 0.217 0221 03473 0.008 1.76 1.76
Dispersion #: 15 3 7 0.272 0.281 0.1766 0.019 1.71 1.71
16 3 7 0.303 0303 0.0980 0.000 2.05 205
17 3 7 0.326 0J26 0 0297 0000 2.45 2.45
Test 8-0215 1 Port V IN I* Abs.1 Abs.2 A bs3 A i l Oi Std-Der. D.O.1 D .0 2 A re.D .O .

In 12 9 1.46 1.87 203 53.7859 0.815
W ater Ql (gpm ): 13.9 OIT 10 10 1.007 1.015 1.026 0.1745 0.020
Gas Qc (cfh): 8.5 In 12 9 1.42 1.58 0.7 545524 1.283
Abs. Indigo I: 1.1 1 6 4 0.089 0.066 0.079 3.0234 0.060 1.4 1.40
Abs. Indigo II: 21.2 2 6 4 0.089 0.151 0.15 27532 0.184 1.6 1.60
W ater Temo. "C : 12 3 6 4 0.257 0.221 0.173 23013 0.219 155 1.55
Gas Temo. "C : 18.3 _ _ 4 _ _ _____ ft._ __ _ 4 — 0J82. ._2<QJ32_ _ 1 $ L __________ _ _ L 6 J ___
lo
C
1
Pressure, psi: 10.000 5 5 5 0.073 0.07 0.117 1.9255 0.109 1.61 161
6 5 5 0.119 0.025 0.083 1.9713 0.197 1.6 1.60
Os Dose (m g/L): 4.15936 7 5 5 0.156 0.191 0.122 1.6360 0.143 1.65 1.65
Feed O j (% ): 2.372 8 5 5 0.291 0266 0.251 1.1664 0.084 1.78 1.78
T ransfer EITid.: 99.68 9 4 6 0.279 0.276 0.277 05633 0.005 2 200
10 4 6 0.34 0.351 0.339 0.3348 0.023 2.1 210
O ther Parameters: 11 4 6 0.377 0.4 0.392 0.1743 0.040 2.29 2.29
O .D. (m g/L): __12__ ______rL_ ___fi __£5L7__ _QAJ J. _ -S U V . . a m i . ___ PJ22__ - 2 4 1 _________ __ 147___
Kd (m ln 'l: 13 3 7 032 0.32 0.0297 0.000 2.7 270
K in (m in '): 14 3 7 0332 0327 0.0015 0.010 288 288
Dispersion #: 15 3 7 0.324 0.326 0.0148 0.004 3.2 3.20
16 3 7 0.324 0.328 0.0119 0.008 3.7 3.70
17 3 7 QJ21- Q.22 0.0252 .Q.QQ6 395 3.95
ITest 9-0216 1 P ort V inu V « ,r Abs.1 A b sJ Abs5 Arg.O s Std-Der. D.O.1 D .0 2 A re.D .O .
In 14 7 0.98 0.978 0.832 132987 0.353
W ater Ql (gpm ): 7.2 OIT 10 10 1.017 1.022 1.029 0.1191 0.013
Gas Qu (cfh): 4.3 In 14 7 0.914 0.717 0.321 13.8791 1.041
Abs. Indigo I: 1.08 1 6 4 0.533 0518 0533 0.6234 0.045 205 205
Abs. Indigo II: 4.13 r 6 4 0538 0543 0.544 05524 0.017 223 223
W ater Temo. "C : 12 3 6 4 0.582 0557 0563 0.4190 0.068 23 230
Gas Temo. *C: 20 __d__ _____ <L_ _ _ _ i — M t t . __ Q ^ 7 & - . a SIS, ___ P -fttL _ . ____________i t f —
Pressure, psi: 9.000 5 5 5 0.472 0.472 0.474 0.2798 0.005 244 244
6 5 5 0.482 0.489 0.474 0.2424 0.031 25 250
O j Dose (m g/L): 1.03349 7 5 5 0.501 0.495 0.487 0.1898 0.029 254 254
Feed O j(%): 0.618 8 5 5 0.497 0502 0.485 0.1884 0.036 259 2.59
Transfer E fflcL: 99.1416 9 4 6 0.414 0.433 0.431 0.0208 0.036 3.01 3.0!
10 4 6 0.43 0.428 0.427 0.0127 0.005 3.25 3.25
O.D. (m g/L): __12__ _____ 4 . . ___fi — 052?_ — 0 5 1 _ ._fl£U 2_ — J L flO i- _ JJ__ __________ 18P ___
Kd (m in '1): 13 3 7 0.325 0542 -0.0282 0.036 4.42 4.42
K ls (m in 14 3 7 0.318 0522 0.0119 0.008 45 4.50
Dispersion #: 15 3 7 0.321 0.32 0.0104 0.002 4.74 4.74
16 3 7 0.32 0.324 0.0059 0.008 5.38 5.38
17 3 7 0.32 0.323 0.0074 0.006 6.Q 22.. 6.03
323
ITest 10-0217 1 P ort Vmo V smt Abs.1 Abs.2 A hsJ A « r.O i Std-Dev. D.O.1 D .0 5 A i».D .O .
In 8 12 152 2.07 0.436 285708 1.152
W ater Ql (JEpm): 7.2 oa 10 10 1.085 1.089 1.087 05140 0.004
GasQn (cfh ): 2.13 In 8 12 3.8 3.64 2.81 26.8338 1.796
Abs. Indigo I: 1.19 1 6 4 0.583 0567 0543 0.7775 0.105 353 3.33
Abs. Indigo II : 21.75 2 6 4 0.578 0577 0569 0.7238 0.026 352 352
W ater Tesno. "C : 12 3 6 4 0.618 0589 0.637 05160 0.126 3.75 3.75
Gas Temn. *C : 15.7 ______ 6— ___ 4 — 0.649_ _ -0 .6 4 1 - -2)2(44 -2 )1 5 6 2 - — 5 W 1 1 -. - 1 1 1 __________ — 169—
Pressure, psi: 8.000 5 5 5 05 2 0505 0527 05228 0.047 4 4.00
6 5 5 0534 0529 0538 05549 0.019 3.95 3.95
O3 Dose (n s /L ): 0.98978 7 5 5 0553 055 0.547 0.1870 0.012 4 4.00
Feed O3 (% ): 1.349 8 5 5 055 0547 0551 0.1898 0.009 4.09 4.09
Transfer E fflcL: 99 2024 9 4 6 0.471 0.458 0.467 0.0369 0.023 4.37 4.37
10 4 6 0.473 0.477 0.474 0.0046 0.007 4.6 4.60
OJ>. (m g/L): __12 _____ 4— 6 . .2)484. -0 .4 7 8 - -2)482 ._ 4)2)2.73- ___ ( u u a _ . —iQO ________ — 5.09__
K d (min S: 13 3 0.362 0.374 -0.0327 0.025 5.19 5.19
K ts (rain*1): 14 3 0.358 0553 0.0045 0.010 5.7 5.70
Dispersion t : 15 3 0.351 0.36 0.0045 0.019 552 5.52
16 3 0.354 0555 0.0074 0.002 6.05 6.05
17 3 0.357 0.347 0.0148 0.021 65 6.50
ITest 11-0218 1 P ort VsMr Abs.1 Abs5 Abs3 Ave-O i StrLDev. D.O.1 D .0 5 Ave-D-O.
In 12 8 3.43 152 352 58.0571 3.521
W ater Q l (Kptn): 7.2 OfT 10 10 1.095 1.088 1.089 0.2064 0.008
Gas Qr. (cfh): 1.06 In 12 8 8.83 8.05 8.15 49.4545 9.719
Abs. Indigo I: 1.19 I 6 4 0567 0547 0.536 0.8519 0.082 458 458
Abs. Indigo I I: 21.45 2 6 4 0.609 0584 0566 0.6632 0.112 4.85 4.85
W ater Temn. "C : 12.2 3 6 4 0.639 0.629 0.648 0.3913 0.049 5 5.00
Gas Temn. "C : 198 — 4 ______ .6— ____ i — 0 6 5 2 . — 0 « i_ — 0.69 -2 )1 7 2 1 - — O ft4i_. — i f l i __________ — 5.05—
Pressure, psi: 7.500 5 5 5 0519 0528 0.536 0.2798 0.035 5.11 5.11
6 5 5 0.552 0546 0.547 0.1939 0.013 5.13 5.13
O ] Dose (m g/L): 0.9078 7 5 5 058 0566 0.556 0.1150 0.050 553 5.23
Feed O j (56): 2.858 8
, 5 5 0.592 0.57 0.579 0.0610 0.046 5.38 5.38
Transfer E flk i: 99.5826 9 4 6 0.474 0.473 0.473 0.0092 0.002 5.75 5.75
10 4 6 0.476 0.472 0.473 0.0081 0.007 5.88 5.88
O.O. (m g/L): — U — _____ 4— ____ 6 -2 ) 4 1 4 - — -04Z- -2)477 -0 0 Q B 1 - — P5U2— - 2 2 8 .__________ — 6.28—
Kd fm in '1): 13 3 7 0.353 0.367 -0.0089 0.029 655 6.35
K l* (rain '): 14 3 7 0.352 0558 0.0059 0.013 6.65 6.65
Dispersion V: 15 3 7 0.355 0.353 0.0089 0.004 6.8 6.80
16 3 7 0.358 0.363 -0.0104 0.010 7 7.00
17—. 1 7 0.351 0.346 0.0222 0010 7.52 7.52
ITest 12-0219 1(1128) P ort Vino Vswr Abs.1 Abs.2 AbsJ A vg-O i Std.D ev. D.O.1 D .0 5 Ave.D.O.
In 14 7 057 0.706 0.733 14.0481 0.363
W alcrQ L (gpm ): 7.1 OfT 10 10 1.075 1.092 1.1 0.2203 0.027
GasQn (cfh): 8.56 In 14 7 0.735 0.634 0.782 13.9498 0.322
Abs. Indigo 1: 1.195 I 6 4 0.43 0.429 0.426 1.4996 0.011 0.94 0.94
Abs. Indigo II: 4.05 2 6 4 0.447 0.44 0.436 1.4338 0.029 1 1.00
W ater Temn. *C: 12.2 3 6 4 0.471 0.472 0.479 1.2623 0.023 1.02 1.02
Gas Temn. "C : 175 _ _ 4— __________ f t — ____4 — a ii_ — 0 5 0 6 - — S>J — 0026— . — 1 0 4 ._________ — L 0 L -.
Pressure, psi: 10.000 5 5 5 0.324 0549 0.366 1.0(38 0.088 1.1 110
6 5 5 054 0565 0523 1.0590 0.088 1.08 1.08
Os Dose (m g/L): 2.09697 7 5 5 0576 0555 0.392 0.9274 0.077 1.12 1.12
Feed Os (56): 0.621 8 5 5 0.359 0.402 059 0.8887 0.092 1.15 1.15
T ransfer EUlci.: 98.4211 9 4 6 0.364 0.366 0.395 0.3567 0.060 15 1.50
10 4 6 0.403 0.409 0.408 0.2470 0.011 1.71 1.71
O.D. (m g/L): __ 12— _____ 4— ______ 5 — 0565- — 0 4 6 i_ _D4«J - 0 0 4 9 6 - — 51008— . - 2 5 1 __________ — 113—
Kd (m in ') : 13 3 0.359 057! -0.0193 0.025 2.82 2.82
K m (m in 1): 14 3 0557 0559 0.0015 0.004 3.13 3.13
Dispersion >: 15 3 0.358 0554 0.0074 0.008 3.25 3.25
16 3 0.354 0561 0.0030 0.015 4.45 4.45
17 1 .U S 0.352 0.0178 0.002 iB 1.80
324
IT o t 13-0220 1 P ort Vnm Vsmt Abs.1 A bs3 Abs3 Ave.O> Std-Dev. D.O.1 D.O-Z Ave.D.O.
In 8 12 2.3 1.95 138 273302 0.643
W ater Ql (gpm ): 7 125 OfT 10 10 1.129 1.129 1.125 0.1295 0.005
G asQ c (d h ): 4.26 In 8 12 0.422 0.73 0.35 28.4832 1.134
Abs. Indigo I: 1.19 I 6 4 0391 0358 0.406 1.7091 0.128 1.98 1.98
Abs. Indigo II: 21.75 2 6 4 0.424 0.402 0.413 13636 0.057 2.21 121
W ater Tem n. *C: 118 3 6 4 0.419 0.439 0.432 1.4753 0.053 2-29 129
Gas Temn. *C: 16.2 __4-_ ____ ft______4 -_ 0 .4 2 ? _ _ _ 0 .4 2 _ — 0.46 .- 1 1 3 4 6 - —J3JQ 1- - 2 1 2 — ___________ 122___
Pressure, psi: 9.000 5 5 5 0.304 0353 0.41 0.9946 0.220 238 138
6 5 5 0.338 0.298 036 1.0930 0.131 23 130
O j Dose (m g/L): 2.12335 7 5 5 0345 0.402 0.395 0.8907 0.129 23 130
Feed Os (% ): 1 261 8 5 5 0.351 0.396 0.351 0.9517 0.108 2.4 140
Transfer E flic L : 99.5453 9 4 6 0.416 0.425 0.426 0.1859 0.019 2.97 197
10 4 6 0.476 0.457 0.453 0.0485 0.043 3.2 3.20
O ther Parameters: 11 4 6 0.474 0.47 0.465 0.0219 0.016 336 3.36
O .D. (m g/L): _tf__ ______ .4 .. __ ft - .0 4 8 5 - _ -0 .4 7 5 - -0 4 8 5 ftA 2Q ._. - 1 0 2 .- ___________ 169—
Kd (m in 1): 13 3 7 0.363 0.37 -0.0282 0.015 4.21 4.21
Kla (m in ‘): 14 3 7 0361 0366 -0.0193 0.010 4.6 4.60
Dispersion #: 15 3 7 0.363 0358 -0.0104 0.010 4.81 4.81
16 3 7 03 6 0.36 -0.0089 0.000 5.29 5.29
17 5 7 0147 0.346 0.0312 0.002 6.52 6.52
ITest 14-0221 1 P ort V im s VsMr Abs.1 Abs.2

Abs3 A vv.O , Std-Dev. D.O.1 D .0 3 Ave.D.O.
In 12 8 4 3.59 3.91 55.8442 0.672
W ater Q l (gpm ): 7.2 OfT 10 10 1.083 1.09
1.083 0.1344 0.008
G asQ c (d h ): 2.13 In 12 8 5.3 3.75 539 54.3688 2426
Abs. Indigo I: 1.15 1 6 4 0.298 0.292
0.309 10277 0.045 335 3.25
Abs. Indigo I I: 21.75 2 6 4 0.371 03940.371 1.6173 0.069 3.7 3.70
W ater Tetun. "C : 11.9 3 6 4 0.449 0.426 0.44 13074 0.060 3.73 3.73
Gas Temn. "C : 16.3 - _ 4 _ - ______ ft. ____ 4 __01Q3- _-0.485- _ 042.4 .-1 1 3 9 4 - — JU15— . - 1 1 1 - _________ 173—
Pressure, psi: 8.000 5 5 5 0.311 0.32 0.312 1.0833 0.021 3.9 390
6 5 5 0.342 0.29 0.283 1.1221 0.134 3.9 3.90
Os Dose (m g/L): 100542 7 5 5 0.464 0.39 0.367 0.6982 0.211 3.9 3.90
Feed Os (% ): 2.659 8 5 5 0.428 0.413 0.507 0.5223 0.210 4 4.00
T ransfer E flicL : 99 7529 9 4 6 0347 0.436 0.44 -0.0496 0.218 434 4.54
10 4 6 0.456 0.454 0.458 0.0139 0.007 4.69 4.69
O .D. (m g/L): _ _ u _ _ _____ 4. ___ 6 --0.45 8- - -0.45ft. -0457. -0 0 1 0 4 - ___ 0*003__ -4 1 4 — _________ 4,74—
Kd (m in 13 3 7 0.348 0.349 -0.0104 0002 53 520
K m (m in 1): 14 3 7 0.343 0.35 •0.0045 0.015 53 5.50
Dispersion ( : 15 3 7 0.347 9344 -0.0015 0.006 5.7 5.70
16 3 7 0.342 0345 0.0045 0.006 6.4 6.40
17 3 . .. 7 0.338 0.347 0.0074 0.019 7.16 7.16
ITest 15-0222 1 P ort VlND VsMF Abs.1 Abs-2 Abs3 Avg.Os Std-Dev. D.O.1 D.O.2 Ave.D.O.
In 14 7 0.648 0.741 0.694 13.9041 0.193
W ater Q l (gpm ): 7.115 O ff 10 10 0.469 0.486 0.487 13700 0.021
GasQ c (d h ): 172 In 14 7 0.703 0.602 0.63 14.0066 0.215
Abs. Indigo I: 1.14 1 7 3 0.386 0366 0.384 2.9045 0.076 039 0.39
Abs. Indigo II : 4.01 2 7 3 0.384 0347 034 3.0545 0.164 035 0.35
W ater Temn. *C: 11.6 3 7 3 0.367 0.373 0.372 2.9599 0.022 035 0.35
Gas Temp. "C : 14.6 — 4 _ _ _____ ft. - — 4 - - 0 J 6 7 - — 0.104- -0 1 7 3 - 2 0 4 5 2 - ___ P -fiB l — - - P .4 - - ___________0,40—
Pressure, psi: 13.000 5 6 4 0.191 0.183 0.212 2.5385 0.078 0.4 0.40
6 6 4 0.193 0.201 0.184 2-5524 0.044 0.38 0.38
Os Dose (m g/L): 4.21584 7 6 4 0.194 0.191 0.212 23195 0.059 0.4 0.40
Feed Os (% ): 0.543 8 6 4 0.216 0.235 0.22 2.3913 0.052 0.4 0.40
Transfer E ftlc i: 90.2186 9 5 5 0.137 0.135 0.153 1.7801 0.041 05 0.50
10 5 5 0.173 0.199 0.164 1.6263 0.076 058 0.58
O.D. (m g/L): — L2_ ______ 5__ 5 0168- - - Q 2 7 I - -0 1 7 2 - 1 2 4 5 4 - - 0 0 0 2 . - . - S J — ___________a s p . . .
Kd fm in 'S : 13 4 6 0.157 0.134 1.0407 0.007 1.27 1.27
K m (m in 1): 14 4 6 0.199 0.206 0.8779 0.017 2.63 163
Dispersion ff: 15 4 6 0.274 0.255 0.6632 0.047 2.9 190
16 4 6 0.33 0325 0.4450 0.012 3.3 3.30
17 3 7 0.242 0.223 01250 0.040 5.2 . . 5.20
325
ITest 16-0223 1 Port V|*n V*MT Abs.1 A bsJ A bsJ A vv.O , Std-Dev. D.O.1 D .0 5 A ve.D .O .
In 8 12 1.71 1.67 255 27.1146 0.688
W ater Q l (gpm ): 7.25 OfT 10 10 0.521 0527 056 1.0681 0.044
GasQ c (cfh ): 8.53 In 8 1.92 256 1.381 27.1305 0.676
Abs. Indigo I: 1.05 I 7 0.236 0.257 0.299 35623 0.222 0.9 0.90
Abs. Indigo II: 21.55 2 7 0.323 0.25 0.272 3.1400 0.259 1.02 1.02
W ater Tem n. “C: 12.2 3 7 0.31 0516 0.302 29483 0.049 1.05 1.05
Gas Tem o. *C: 18 __ 4 _ ______ 7_ „ ££SL — < U 7 I_ -£ ,2 6 5 24692 _ — - J J ___________ __ L lf l___
Pressure, psi: 10.000 5 6 0.157 0.174 0.12 2.4918 0.143 1.02 1.02
6 6 0.185 0.167 0.157 25913 0.074 1.1 1.10
O j Dose (m g/L): 3.97995 7 6 0.211 0.198 0.163 25823 0.129 1.11 1.11
Feed O j (% ): 1.199 8 6 0.27 0.194 0.223 2.0831 0.199 1.05 I 05
T ransfer E flic L : 96.0633 9 5 0.206 0.221 0.252 15412 0.097 1.4 1 40
to 5 0.262 0.279 0.273 1.0542 0.036 1.68 1.68
O ther Parameters: 11 5 0.363 0.354 0.347 0.7079 0.033 1.91 1.91
O.D. (m g/L): „ u__ ______ 5. _——5 --£ 4 0 5 L ___ 0 4 1 - -£ 4 1 .4 - £ 4 7 3 2 - _ _ J U U I__ _ 2 £ 7 .__________ __ 2.07___
K d (m ln'S : 13 4 6 0.274 0.308 0.4468 0.083 2.82 2.82
K u i (m in '): 14 4 6 0.316 0529 05377 0.032 3.02 3.02
Dispersion #: 15 4 6 0.366 0.357 0.2026 0.022 35 3.20
16 4 6 0.403 0589 0.0831 0.034 4.15 4.15
17 3 7 JLM 0.313 0.0134 0.010 55 5.20
ITest 174)224 1 P ort V ino Abs.1 A bs5 A h«J A v r.O , Std-Dev. D.O.1 D .0 5 A ve.D .O .
In 12 8 2.16 2.8 57 575403 1.073
W ater Q l (xpm ): 7.125 OfT 10 10 1.395 1.4 157 05528 0.033
GasQ c (d h ): 4.23 In 12 8 2.77 566 2.04 57.4390 1.036
Abs. Indigo I: 1.51 1 6 0526 0503 0.207 3.6052 0.064 1.8 1.80
Abs. Indigo II: 20.95 2 6 0.316 0.249 0.262 35745 0.185 508 508
W ater Tem n. *C: 12 3 6 0.318 0571 0.342 59212 0.138 51 51 0
Gas Temn. *C: 15.7 — 4— __________ f t - - — £4<K— <2.422- — <222 ._2<fiQ9ft_ — J L U 1 -. - J U L . ___________ 2.20___
Pressure, psi: 9.000 5 5 0.168 0.153 0.156 54769 0.033 52 520
6 5 0.125 0.191 0.152 2.4894 0.138 s ■> 520
Os Dose (m g/L): 4.25178 7 5 0.361 0.265 0.319 1.8286 0.200 2.25 525
Feed O , (% ): 2.610 8 5 0.359 0.416 0.36 1.5654 0.136 258 538
Transfer ETflcL: 99.5599 9 4 6 0.41 0.408 0.438 0.6418 0.058 2.89 589
10 4 6 0.477 0515 0.506 0.3625 0.069 3.1 3.10
O ther Parameters: 11 4 6 052 0521 0545 05609 0.049 3.29 3.29
O.D. (m g/L): -_ 1 2 _ _ ______ t . — ft __ £529_. . _QJt22_ _ £228 .-£< 2424- — JU227— . - 3 4 7 .- ___________ 2.49—
Kd (m in 1): 13 3 7 0.422 0.403 0.1202 0.040 4 4.00
K m (m in '): 14 3 7 0.423 0.416 0.0994 0.015 4.6 4.60
Dispersion <: 15 3 7 0.448 0.456 0.0030 0.017 4.45 4.45
16 3 7 0.454 0.451 0.0015 0.006 557 5.27
17 3 7 0.443 0.443 00297 0.000 5.6 5.60
Test 184)225 1 Port V in o V>Mr Ahs.1 Abs.2 A bsJ A v r.O i Std-Dev. D.O.1 D .0 2 A v*.D .O .
In 14 7 0.638 0.495 0.688 14.1008 0.416
W ater Q l (xpm ): 3.53 O ff 10 10 1.39 1.36 1.39 0.2701 0.036
GasQ c (d h ): 2.13 In 14 7 1.92 1.84 1.92 11.4279 2.942
Abs. Indigo I: 1.51 1 4 6 0.425 0.412 0.432 0.6268 0.035 2.3 23 0
Abs. Indigo II: 4 2 4 6 0.437 0.448 0.435 05680 0.024 243 243
W ater Tcm o. “ C: 12.5 3 4 6 0.48 0.48 0.472 0.4387 0.016 26 26 0
Gas Temn. "C : 19.9 — 4 — ______.4.. _ — 4 — £429- _-0.494- — 0.49 -£ < 2 7 9 8 - — flJ U ft- -2 4 8 — ___________ 2.58-_.
Pressure, psi: 8.000 5 3 7 0.373 0556 0.356 0.2711 0.029 267 267
6 3 7 0.379 0.39 0.392 0.1959 0.021 289 28 9
O j Dose (m x/L): 0.85976 7 3 7 0.398 0588 0.385 0.1860 0.020 28 28 0
Feed Os (% ): 0.684 8 3 7 0.396 0.407 0.4 0.1544 0.017 2.79 279
Transfer E d ld .: 97.6362 9 3 7 0.444 0.449 0.444 0.0218 0.009 3.1 3 10
10 3 7 0.453 0.453 0.447 0.0059 0.010 3.64 3.64
O.D. (m g/L): — U — ___________ 1- 7 -_ £ 4 4 S - — 0.44ft. -£■447 - £ £ l 7 a _ ____ P J 2 0 1 - . — U — _____________________ 410 ______
Kd (m in 1): 13 3 7 0.449 0.461 -0.0059 0.025 4.7 4.70
K ta (m in '): 14 3 7 0.447 0.449 0.0148 0.004 5.2 5.20
D ispersion*: 15 3 7 0.452 0.45 0.0059 0.004 5.4 5.40
16 3 7 0.446 0.448 0.0178 0.004 5.65 5.65
17 3 7 0.448 0.444 Q.Q2Q8. 0008 6.29 6.29
326
ITest 19-0226 1 P ort V im VsMP Abs.1 A bsJ A b sJ A vv.O , Std-Dev. D.O.1 D .O J Ave.D.O.
In 8 12 0501 0.443 0.343 28.9815 0.111
W ater Ol (gpm ): 3.6 on 10 10 0.14 0.144 0.143 28211 0.004
G asQ c (d h ): 1.13 In 8 12 0.506 0.582 0.535 28.9039 0.113
Abs. Indigo I: 15 1 4 6 0.385 0.38 0.396 0.7377 0 028 3.15 3.15
Abs. Indigo II: 21.35 2 4 6 0.444 0.427 0.393 0.6188 0.090 3.4 3.40
W ater Temo. *C: 12.3 3 4 6 0.461 0.459 0.443 05045 0.034 353 3.53
Gas Temo. *C: 16.4 — 4— ____ t
- __ 6 - 0 4 2 5 - _ -4.492 _ - 0 4 ® ..0 3 6 9 4 - — J U f f li- - H i . .
1
1
1
1
1
1
1
1
Pressure, psi: 7.500 5 3 7 0.35 0541 0.341 05147 0.015 361 3.61
6 3 7 0568 056 0.357 0.2622 0.017 3.85 3.85
Os Dose (m g/L): 1.1312 7 3 7 0.378 0.405 0547 05177 0.086 3.79 3.79
Feed Os (% ): 1.417 8 3 7 0.389 0.375 0.389 0.1949 0.024 3.79 3.79
Transfer EdlcL: 90.2396 9 3 7 0.427 0.427 0.423 0.0722 0.007 4.19 4.19
10 3 7 0.45 0.445 0.444 0.0109 0.010 4.7 4.70
O th er Parameters: 11 3 7 0.444 0.446 0.437 0.0228 0.014 4.8 4.80
O.D. (m g/L): __ 12__ ______ 3_ _ ___ 2 — a « _ — 0.442. .0 4 4 5 .-00Q 72- ___ R 0 0 2 -. - 2 2 S .- __________i ® _ _ .
Kd (m in 1): 13 3 7 0.448 0.433 0.0282 0.031 5.45 5.45
K ta (m in 1): 14 3 7 0.451 0.451 41.0030 0.000 6.04 6.04
D ispersion#: 15 3 7 0.448 0.45 0.0030 0.004 6.01 6.01
16 3 7 0.452 0.45 -0.0030 0.004 6.62 6.62
17 3 7 ___SLHJL 0,446 0.0193 0.010 7.02 7.02
ITest 20-0227 1(1201) Port Vim Vsmt Abs.1 A bsJ A b sJ A ve.O , Std-Dev. D.O.1 D .0 2 Ave.D.O.
In 12 8 1.54 0.7 158 625766 1.549
W ater Ql (gpm): 3.55 OfT 10 10 1.47 1.45 1.4 0.1247 0.075
GasQ c (d h ): 0.53 In 12 8 0.74 1.85 3.39 61.4545 3.058
Abs. Indigo 1: 1.5 I 4 6 0.241 0278 0.276 1.1602 0.072 4.85 4.85
Abs. Indigo II: 21.35 2 4 6 0.338 0507 0.333 0.9489 0.058 528 5.28
W ater Temn. *C: 122 3 4 6 0.404 0563 0.401 0.7296 0.079 5.49 5.49
Gas Temo. *C: 17.2 _ _ 4 _ _ ______ 4 -_____ fi — 04HL _ -0.451- .0 4 6 0 ._ 0 5 0 7 2 . P J 0 2 - . -- 5 5 __ __ 2,60___
Pressure, psi: 7.500 5 3 7 0.296 0522 0.313 0.4146 0.039 5 66 5.66
6 3 7 0.341 0.346 0.339 05206 0.011 5.65 5.65
Os Dose (m g/L): 1.14396 7 3 7 0.395 0561 057 02216 0.052 5.65 565
Feed Os (% ): 3 052 8 3 7 0.374 0.373 0.365 0.2355 0.015 5.45 5.45
Transfer E flicL : 99.797! 9 3 7 0.422 0.428 0.45 0.0495 0.044 5.75 3.34 4.55
10 3 7 0.442 0.451 0.445 0.0119 0.014 6 3.8 4.90
O ther Parameters: 11 3 7 0.454 0.452 0.451 -0.0069 0.005 5.89 4.3 5.10
O.D. (m g/L): — 12— ______ 3______ 1 — 0 4 i L _-0.442- -0 4 S J ._i)0Q (0_ — J U S 1 - . - i - z i — -4 ,2 — — 127___
Kd (m in '): 13 3 7 0.449 0.466 -0.0223 0.036 5.65 4.26 4.96
K u i (m in '): 14 3 7 0.444 0.446 0.0148 0.004 5.97 4.98 5.48
Dispersion V: 15 3 7 0.445 0.447 0.0119 0.004 6.1 5.58 5.84
16 3 7 0.451 0.45 -0.0015 0.002 6.8 6 6.40
17 3 7 0.441 0.441 0.0267 0.000 7.29 6.6 6.25
ITest 21-0228 1(1210) Port V im Vwn- Abs.1 A bsJ A bsJ Ave.Os S id.Dev. D.O.1 D .O J Ave.D.O.
In 7 0.647 0.702 0.628 13.6353 0.160
W ater Q l (gpm): 3.6 O if 10 1.29 1.31 1J8 0.4294 0.032
G asQ c (d h ): 4.4 In 7 0.902 1.035 1.085 12.9115 0.837
Abs. Indigo I: 1.5 1 5 0.421 0.381 0.376 1.4850 0.103 1.09 1.09
Abs. Indigo II: 3.94 2 5 0.373 0397 0.412 1.4795 0.082 1.2 1.20
W ater Temo. "C : 125 3 5 0.418 0.418 0.421 13756 0.007 1.33 1.33
Gas Temo. *C: 19.2 — 4— ___ ___ 5 — 044?- — 0.472- _ 0-457 — U V 2 _ - J U B L - - 1 2 1 ______ --L 2 7 ___
Pressure, psi: 9.000 5 6 0.268 0.256 0.268 1.1636 0.024 1.32 1.32
6 6 0.282 0.261 0.293 1.1128 0.056 1.38 1 38
Os Dose (m g/L): 1.9676 7 6 0.307 0315 0.311 1.0009 0.014 1.49 1.49
Feed Os (% ): 0.632 8 6 0.336 0.358 0.339 0.8854 0.041 1.49 1.49
Transfer E flicL : 96.674 9 0.186 0.207 0.186 0.7629 0.036 1.65 1.65
10 0.309 0.324 0.293 0.4195 0.046 2.01 101
O ther Parameters: U 0.372 0.382 036 0.2335 0.033 231 131
O.D. (m g/L): — 12— 0402- — 0.405. - 0 4 U ._0vU?fi_ ___ 0 JU 1 -. -2 S L _________ — 167___
Kd (m in*1): 13 0.407 0.43 0.0935 0.048 3.02 3.02
K u i (m in*1): 14 0.426 0.421 0.0787 0.010 3.86 3 86
Dispersion #: 15 0.447 0.443 0.0119 0.004 3.78 3.78
16 0.453 0.455 -0.0119 0.004 437 4.57
17 0.446 0.441 0.0193 .0,010. 5.35 5.35
327
ITest 22-0229 1 Port V ini> Vsfcir Abs.1 A bsJ A bsJ A,e.O< S ld.D e,. D.O.1 D .0 2 A ,e.D .O .
In 8 12 2.03 152 231 265235 0.388
W ater Q l (gpm ): 3.5 OfT 10 10 1.27 1.33 134 0.2632 0.079
G asQ c (cfh ): 2.13 In 8 12 301 2.8 4.21 25.8401 1.032
Abs. Indigo I: 144 1 5 5 0.289 0.32 0.308 1.7219 0.065 2.18 218
Abs. Indigo II: 21.5 2 5 5 0.264 0.268 0.276 1.8729 0.025 2.48 248
W ater Tem o. aC: 12 3 5 5 0.348 0.307 0.343 1.6097 0.093 2.45 245
Gas Tem o. "C : 16.1 - _ d - _ ______ 5______ 5 — 3U55- -0228 .-1 5 6 2 2 - ___QJ08._. — 22 L — __________ 25J___
Pressure, psi: 8.000 5 4 6 0.216 0.232 0.226 1.2167 0.028 25 250
6 4 6 0.2 0.2 0.224 12745 0.048 259 259
Oj Dose (m g/L): 1.96071 7 4 6 0.313 0316 0.292 0.9316 0.045 2.7 270
Feed O j (%): 1.268 8 4 6 0.284 0.284 0.333 0.9547 0.098 2.7 270
T ransfe r E d ic t: 98.9814 9 3 7 0.255 0.307 027 0.4591 0.079 333 3.33
10 3 7 0394 0.385 0391 0.1247 0.014 3.6 3.60
O th er Parameters: 11 3 7 0398 0.407 0.408 0.0821 0.016 3.81 3.81
O .D . (m g/L): _ _ u _ _ ______ 3______ 1 0,309- _-0.402. - 0 2 U .-0 0 6 5 2 - ___B003__ _2.28.__ __________ 129___
Kd (m in*1): 13 3 7 0.418 0.441 0.0074 0.048 4.1 4.10
K u i (m in*1): 14 3 7 0.437 0.428 -0.0015 0.019 4.66 466
Dispersion 1: 15 3 7 0.426 0.435 0.0045 0.019 4.72 4.72
16 3 7 0.433 0.442 -0.0163 0.019 5.4 5.40
17 3 7 0.426 0.432 00089 0.013 63 6.30
>■
ITest 23-0302 1 Port Abs.1 A bsJ A bsJ Std-Dev. D.O.1 D .O J Ave-D.O.
9
P
In 12 8 212 3.38 24 58.9610 2062
W ater Q l (g p m ): 3.5 OfT 10 10 1.12 1.09 1.19 0.8035 0.107
G asQ c (d h ): 1 06 In 12 8 1.27 0.78 1.83 61.0494 2830
Abs. Indigo I: 1.52 I 5 5 0.332 0.328 055 1.7593 0.049 35 3.50
Abs. Indigo 11: 21.55 2 5 5 0.371 0565 0.381 1.6111 0.034 3.7 3.70
W ater Tem n. "C : 12 3 5 5 0.431 0.441 0.472 IJ96 6 0.089 3.82 3.82
Gas Tem n. * C : 172 ______ 5______ 5 - - 0 ,4 9 - ___052 _ . a m __ 11.678- -J J — _______ISP__
Pressure, psi: 7 500 5 4 6 055 0519 0.335 0.9466 0.054 4 4.00
6 4 6 0.361 0583 0.386 0.8012 0.047 4.23 4.23
O j Dose (m g/L): 2.3053 7 4 6 0.394 0.42 0.41 0.6926 0.045 4.3 4.30
Feed O j (% ): 2.877 8 4 6 0.431 0.431 0.454 05864 0.046 4.4 4.40
T ransfe r EfTlcL: 98.6839 9 3 7 0.35 0.332 0.301 0.3810 0.074 4.49 4.49
10 3 7 0.389 0.399 0.402 0.1761 0.020 4.91 4.91
O th e r Parameters: U 3 7 0.433 0.427 0.425 0.0821 0.012 5.33 5.33
O .D . (m g/L): ..1 2 .. ______ 3______ 1 ___i)445_ _ -0.438- -0-445. .-0 0 4 J 6 - ___ 001 1__ —50— __________ &6P____
K d (m in ') : 13 3 7 0.45 0.46 0.0030 0.021 5.82 5.82
K ls (m in '): 14 3 7 0.457 0.46 -0.0074 0.006 6.2 6.20
D ispersion t i 15 3 7 0.466 0.458 -0.0178 0.017 6.2 6.20
16 3 7 0.458 0.458 -0.0059 0.000 6.82 6.82
17 3 7 0.454 0.458 0.0000 0.002 7.13 7.13
ITest 24-0304 1(1204) Port V INI) VsMF Abs.1 A bsJ A bsJ A ve.O j S td.D c,. D.O.1 D .O J Avg.D .O ,
In 13 8 1.29 15 153 13.0788 0.070
W ater Q l (gpm ): 3.5 o rr 10 10 0.95 0.94 0.97 1.1152 0.032
G asQ c (sd h ): 8.56 In 13 8 1.22 1.32 1.24 13.1576 0.149
Abs. Indigo 1: 1 49 I 6 4 0.424 0.418 0.422 2.4554 0.016 0.49 0.49
Abs. Indigo II : 5.18 2 6 4 0.42 0.433 0.449 25896 0.075 051 051
W ater Tem o. "C : 124 3 6 4 0.469 0.438 0.479 2.2442 0.111 0.42 0.42
Gas Tem o. "C : 16.5 — 3 _ _ ______ <L_ — 4 -_ U 4 5 4 _ _ -0 .4 2 2 . -0 4 S 2 ..u r n .. ___ OSSA... -S>J ___________ — a s > _ - .
Pressure, psi: 10.500 5 5 5 0.257 0.248 0.237 2.0682 0.042 0.48 0.48
6 5 5 0.252 0.227 0.242 2.0973 0.052 052 0.52
O j Dose (m g/L): 4.01227 7 5 5 0.288 0.298 0.272 1.9075 0.055 0.49 0.49
Feed O j (% ): 0.565 8 5 0.275 0.264 0.309 1.9214 0.097 0.57 0.57
T ra n sfe r EITid.: 91.5246 9 4 6 0.218 0.2 0.158 1.3991 0.107 0.7 0.70
10 4 6 0.244 0J37 0.255 1.2144 0.031 0.9 0.90
O th e r Param eters: 11 4 6 0.314 0507 05 1.0009 0.024 1 1.00
O.D. (m g/L): _ _ u _ _ ______ ____ 6 . . a m . — <U 76_ -0 5 6 8 . . a m i . — - 0 0 1 i_ . - 1 2 2 . _________ - - L 2 9 — .
K d (m in 1): 13 3 0.217 0.217 0.6827 0.000 152 1.52
K ia (m in 1): 14 3 0.263 0.236 05863 0.057 1.87 1.87
D ispersion*: 15 3 0.311 0511 0.4037 0.000 2.19 2.19
16 3 0547 0.363 OJ731 0.034 2.66 2.66
17 3 0.383 0.406 0.1558 0.048 4.1 4.10
328
Ite s t 25-0306 1 P ort Vmm V sm, Abs.1 A b sJ A bsJ A*».Oi Std.Dev. D.O.1 D .O J A»i.D.O.
In 9 11 4.43 4.26 4.42 29.1230 0.162
W ater Q l (xpm ): 3.5 on 10 10 1.01 1.02 1.01 0.9905 0.012
GasQ c (cfh ): 425 In 9 U 3.9 3.69 3.55 29.6812 0.648
Abs. Indigo I: 1.49 1 6 4 0.284 0.313 0.297 3.0961 0.075 1.18 1.18
Abs. Indigo II: 21.5 2 6 4 0.306 0.294 0.293 3.0978 0.038 133 1.23
W ater Tem n. *C: 122 3 6 4 0.358 0368 0.372 27429 0.037 131 1.31
Gas Tem o. *C: 18.5 _4 __ ______ IL _ __ 4 — 0 3 85. __ 0-381- -.0.405 ._2.fi 165- — f i0 f il_ . — U S .__________ __L28___
Pressure, psi: 9000 5 5 5 0.195 0.168 0.162 23688 0.073 139 1.39
6 5 5 0.188 0218 0.198 22594 0.063 139 1.39
O j Dose (tn g/L): 4.49377 7 5 5 0.287 0.248 0.245 20156 0.097 136 136
Feed O j (% ): 1.344 8 5 5 0.286 0.289 0.266 1.9311 0.052 135 135
Transfer E fflc i: 96.663 9 4 6 0.182 0.199 0.205 13876 0.O1I 1.78 1.78
to 4 6 0.294 0337 0.321 0.9651 0.075 202 202
O ther Parameters: 11 4 6 0.414 0.407 0.364 0.6961 0.094 232 232
O J). (m g/L): _12_ ____ 4__ __ fi — 0 4 5 3 . _ -0 .4 5 2 - _ 0-438 .-0 4 1 3 2 - __ P J 2 2 -. . 2 2 4 . __________ __
Kd (m in *1: 13 3 7 0.326 0.32 0.3681 0.013 304 3.04
Kl B (m in 1): 14 3 7 0.351 0358 0.2746 0.015 3.61 3.61
Dispersion #: 15 3 7 0.388 0387 0.1766 0.002 3.78 3.78
16 3 7 0.428 0.422 0.0653 0.013 43 4.30
17 3 7 0.455 0.444 -0.0074 0023 5.12 5.12
iTest 26-0309 1(08141 P ort Vrnn V w Abs.1 A bsJ A b sJ A re .O i Sld-Der. D.O.1 D .O J Ave.D .O .
In 12 8 27 3.6 3.11 56.6130 1.404
W ater Q l (Kpm): 3.56 O ff 10 10 03 0351 0363 1.8327 0.070
GasQ c (d ll) : 2.12 In 12 8 1.56 1.68 24 583714 2489
Abs. Indigo 1: 1.42 1 6 4 0.219 0.166 0.232 33576 0.182 244 244
Abs. Indigo II: 21.3 2 6 4 0.24 0J48 0.292 3.0753 0.145 249 249
W alcr Tem n. *C: 12 3 6 4 0.303 0347 0.306 27706 0.128 271 271
Gas Tem o. *C: 17 5 _ _ 4 _ _ ______ ft— ____4 - - 0 J 4 . L - - Q 3 4 1 - -0-388 -£ 5 fi? Z _ — JU 38— - 2 6 8 .__________ __ *68___
Pressure, psi: 8 000 5 5 5 0.132 0.123 0.137 24076 0.029 2.79 279
6 5 5 0.219 0.201 0.208 20807 0.038 296 296
O i Dose (m g/L): 4.34891 7 5 5 0.237 0.242 0.219 1.9837 0.050 2.98 298
Feed O j (56): 2.706 8 5 5 0.318 0.298 0.286 1.7011 0.067 29 290
Transfer E d ic i.: 96.871 9 4 6 0.2 0.188 0.179 13126 0.036 3.13 3.13
10 4 6 0 J9 2 0.309 0.308 0.9177 0.033 3.6 3.60
O.D. (m g/L): ______ 4— . .fi — 0.44*. __0.424- -0-417. 0-4832- — JU ttt— - 4 0 1 __________ __ 4.04___
Kd (m in '1): 13 3 0.318 0322 03147 0.008 433 4.53
Kls (m in *): 14 3 0.358 0365 0.1915 0.015 4.75 4.75
Dispersion #: 15 3 0.383 0377 0.1365 0.013 4.94 4.94
16 3 0 J9 4 0.407 0.0757 0.027 5.42 5.42
17 3 0.417 0.422 0.0193 0.010 5.89 ..502
VITA
329
V IT A
Chun-Ming Chen was bom on August 31,1964 in Chiai, Taiwan, Republic of China. He
received his Bachelor degree in Agricultural Engineering from National Taiwan
University in June, 1987, and Master degree o f Civil Engineering at Purdue University in
August, 1991. Before he came to Purdue University, he was an engineer in Sinotech
Engineering Consultants, Inc., Taipei, Taiwan, R.O.C..
IMAGE EVALUATION
TEST TARGET ( Q A - 3 )
1.0 £
I - t ii 1 1 2 .2
!r
£ L i 2.0
l.l
.8
1.25 1.4 1.6
150m m
<5>
A P P L I E D A IIV U G E . In c
1653 East Main Street
- Rochester, NY 14609 USA
A
_ = Phone: 716/482-0300
= Fax: 716/288-5989
/ / 0 1993. Applied Image. Inc.. All Rights Reserved
>
o/

Modeling Drinking Water Disinf

Uploaded by

Copyright:

Available Formats

Modeling Drinking Water Disinf

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modeling Drinking Water Disinf

Uploaded by

Copyright:

Available Formats

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. U M I

The quality of this reproduction is dependent upon the quality of the

Oversize materials (e.g., maps, drawings, charts) are reproduced by

Photographs included in the original manuscript have been reproduced

This is to certify that the thesis prepared

B y _______ Chun-Ming Chen_______________________________________________________

Modeling Drinking Water Disinfection in Ozone Bubble-Diffuser

For the degree o f ______ Doctor o f Philosophy______________________________________

Signed by the final iniog commu

Format Approved by:

IN OZONE BUBBLE-DEFFUSER CONTACTORS

Submitted to the Faculty

In Partial Fulfillment o f the

Requirements for the Degree

UMI Microform 9914465

This microform edition is protected against unauthorized

My Dearest Friend, Foon-Yee Chong

M y Parents, Ping-Tsan Chen and Li-Ying Shien

Yee Chong and Sharon Waller for their experimental assistance.

I would like to acknowledge The Metropolitan Water District of Southern

financial assistance during the period of this research.

meaningful era of my life.

TABLE OF C O N TEN TS................................................................................................... iv

Chapter 1. IN TR O D U C TIO N ..............................................................................................1

Chapter 2. LITERATURE R E V IE W ................................................................................. 7

Chapter 3. A X IA L DISPERSION M O D EL................................................................... 69

Chapter 4. M ATERIALS AND M ETHODS.................................................................. 89

Chapter 5. THEORETICAL ANALYSIS OF A X IA L DISPERSION MODEL 121

Chapter 6. PILOT-SCALE EXPERIMENTAL RESULTS AND DISCUSSION 142

Chapter 7. REACTOR DESIGN O P TIM IZA TIO N ..................................................... 203

Chapter 8. SUM M ARY A ND CO NCLUSIO N............................................................. 224

LIST OF REFERENCES................................................................................................. 230

1. Recommended Giardia Cysts and Cryptosporidium Oocysts Treatment

2. Summary o f Expected Removal o f G iardia lamblia Cysts and Viruses by

3. Recommended CT (mg-min/L) Values for Inactivation of Giardia lamblia

4. Guidelines to Predict Ozone Concentration (C) Based on Chamber Inlet and

5. Recommended Procedures to Calculate Contact Time (T) for Ozone

6. Summary o f CT Value Ranges Corresponding to 99% Inactivation of

7. Second Order Disinfection Rate Constants for Ozone Inactivation of

8. Development o f Mass Balance Expressions for Dissolved Ozone (Based on

9. Closed Vessel Boundary Conditions....................................................................... 75

12. Results for Calibration Procedure of Bubble Size Measurements Performed

13. Operating Conditions and Experimental Results for Bubble Size

14. Dispersion Coefficient and Oxygen Mass Transfer Coefficient Estimated

16. CT Values and Giardia Inactivation Efficiencies for MWDSC's Ozonation

1. Schematics o f Full-Scale Ozone Bubble-Difiiiser Contactors Located at (a)

2. Determination o f Ao from Pulse-Input and Step-Input Normalized Tracer

3. Ozone Decomposition o f Water from University Lake, Chapel Hill, NC

4. Typical Semi-Log Microorganism Survival Curves Observed in Disinfection

5. Schematics o f Selected Models for Bubble-Difiiiser Ozone Contactors: (a)

6. Schematic o f an Infinitesimal Control Volume for a Bubble-Difiiiser

7. Axial Dispersion Model Predictions of Aqueous and Gaseous Ozone

8. Model Predictions of Tracer Curve with Different Dimensionless Time and

9. Inactivation o f G iardia muris Cysts in a Semi-Batch Reactor Operated at

11. Effect o f Sampling Time on the Observed Inactivation Efficiency o f Pulse-

12. Schematic o f Pilot-Scale Ozone Bubble-Difiixser Contactor.................................97

14. Experimental Matrix for Operating Conditions....................................................101

15. Conceptual Schematic of Experimental Setup and Data Acquisition System

16. Schematic o f Experimental Apparatus Used for the Measurement o f Initial

B y _ Chun-Ming Chen_________________________________________________

For the degree o f Doctor o f Philosophy________________________________