LIBRARY

Michigan State

University

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
To AVOID FINES return on or before date due.

 

DATE DUE

MTE DUE

DATE DUE

 

 

AW

 

. o ‘
@311; ~

 

 

MAX121352302

 

JUN 1 7 2003

 

 

1104 0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1m WWJGS-p.“

 

THE USE OF OZONE FOR THE REMOVAL OF ODOR FROM SWINE MANURE

By

Jiunn-Jer (Jerry) Wu

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Civil and Environmental Engineering

1998

IF

inc

ch.
fre
tin

Ill]

CC

pl

af

dL

ABSRTACT

THE USE OF OZONE FOR THE REMOVAL OF ODOR FROM SWINE MANURE

By

J iunn-Jer (Jerry) Wu

The ozonation of swine manure was investigated. Malodorous chemicals
including phenol, p-cresol, p-ethylphenol, indole, skatole, volatile fatty acids, ammonia,
and hydrogen sulfides were monitored during ozonation of swine manure. To
characterize the generation of these malodorous chemicals during manure storage, the
freshly collected manure was stored in a fermentation tank for 12 weeks. During this
time the concentrations of the target malodorous compounds along with odor and the
number of microorganisms were monitored. As part of this study, swine manure was
treated in lab-scale and pilot-scale reactors to investigate the ozonolysis of malodorous
compounds, to determine the operational parameters for reactor design, to understand the
physical and chemical factors affecting the ozonation of these target compounds, and to
predict the degradation of odorous substances during ozonation by using a mathematical
model.

Malodorous compounds accumulated rapidly in manure during storage. The odor
of both the fresh and stored (up to 12 weeks) manure was reduced to an acceptable level
after ozonation at a dosage of 1 g/L. The odor of ozone-treated manure did not recur

during subsequent storage, even after several months. The concentrations of the phenolic

Emit

an

020

and indolic compounds were reduced to nondetectable levels. The populations of
anaerobic and aerobic microorganisms were found to be slightly reduced after ozonation
at a dosage of 1 g/L. However, E. coli and total coliform were almost eliminated by
ozone. In the pilot study, a reactor having a capacity of 3 gallons was constructed to treat
swine manure. Using this design, the applied ozone dosage can be reduced to 0.5 g/L and
still reduce the odor to an acceptable level as determined by the olfactometry test. The
pilot-scale reactor has the advantages of a venturi injector which generates fine bubbles
and deep-reactor which increases the liquid-gas contact time. Also, a stirrer provides
more hydrodynamic mixing. This ensures almost complete absorption of ozone by the
manure slurry and serves to effectively reduce the concentrations of malodorous
compounds.

A mass-transfer model for ozone was developed to estimate the mass-transfer
coefficient and partition coefficient for a semi-batch reactor. A plug-flow pattern
described the rise of ozone bubbles from the bottom to the top of the reactor. In the
kinetic study, stoichiometric factors and rate constants for the reaction of ozone with
phenolic and indolic compounds were determined. Increases in the pH of the aqueous
solution resulted in a significant increase in the reaction rate constants of the phenolic and
indolic compounds. In a semi-batch experiment, it was shown that the applied ozone
dosage is a reliable operational parameter that can be used to describe the removal of
malodorous compounds. An oxidation model that combines mass transfer and chemical
oxidation has been developed to predict the degradation of malodorous compounds

during ozonation of either artificial or real swine manure.

for h:
dunn
aclm
Dr. I
and

me]

ACKNOWLEDGEMENTS

I would like to extend my appreciation to my major advisor, Dr. Susan J. Masten,
for her professional guidance and advice on my thesis research and academic activities
during these years in Michigan State University. Also, I wish to express my
acknowledgements to Dr. Mackenzie Davis, Dr. John Gerrish, Dr. Howard. Person, and
Dr. Melvin Yokoyama for serving on my committee and for their advice, encouragement,
and guidance. Many thanks should be addressed and directed to those people who gave
me much help during my research, including Dr. Susan Hengemuehle, Sung-bee Park,
Hyesoon Kim-Yang, Steven Innes, and Natasha Muju. I also wish to thank Dr. Pual
Loconto and Yang-Lang Pan for their assistance in using the analytical instruments in the
Engineering Research Complex. In addition, I want to acknowledge the manager of
MSU Swine Teaching and Research Facility, Alan Snedegar, for his help in using the
facilities at the farm and Dr. Paul Bartlett for his assistance with the statistical analysis of
the odor data.

Finally, I want to thank my wife, Angela, for her support during these years and
allowing me to focus on my research without worrying about other matters. Therefore,
she deserves the honor much credit for my Ph.D. degree. Also, my son, Steven, should

be addressed because he is such a joy and has taught me the responsibility of being a

father.

iv

I'll 1|“ I I ill] t

TABLE OF CONTENT

 

 

 

LIST OF TABLES ----------------- - - - - --------- - - -- -- Aviii
LIST OF FIGURES - - - -- .3:
CHAPTER 1 INTRODUCTION - -- “1
l.1 Environmental Significance .............................................................................. l
1.2 Original and Identification of Malodorous Substances in Livestock Waste ..... 2
1.3 Odor Measurement ............................................................................................ 3
1.4 Current Technologies to Control Odor .............................................................. 6
1.5 Background of Using Ozone Treatment ............................................................ 8
1.6 Research Hypotheses ......................................................................................... 11
1.7 Research Objective ............................................................................................ 12
1.7.1 Specific Aims: Phase I ............................................................................ 13
1.7.2 Specific Aims: Phase II .......................................................................... 13
1.7.3 Specific Aims: Phase III ......................................................................... 14
1.7.4 Specific Aims: Phase IV ......................................................................... 14
1 .8 References .......................................................................................................... l 5

CHAPTER 2 THE EFFECT OF STORAGE AND OZONATION ON

2.1
2.2

2.3

2.4
2.5

THE PHYSICAL, CHEMICAL, AND BIOLOGICAL

CHARACTERISTICS OF SWINE MANUR SLURRIES ......... 20
Introduction ........................................................................................................ 20
Materials and Methods ....................................................................................... 22
2.2.1 Manure Collection and Storage Test ....................................................... 22
2.2.2 Ozone and Ozonation Test ...................................................................... 23
2.2.3 Odor Evaluation - Olfactometry ............................................................ 24
2.2.4 Chemical Analysis .................................................................................. 24
Results and Discussion ....................................................................................... 29
2.3.1 Storage Experiment ................................................................................. 29
2.3.2 Ozonation Experiment ............................................................................. 34
Conclusions ........................................................................................................ 38
References .......................................................................................................... 4O

CW

'1)

CHAPTER 3 THE USE OF OZONE TO REDUCE THE CONCENTRATION
OF MALODOROUS METABOLITES IN SWINE MANURE

 

SLURRY BY USING A PILOT-SCALE SYSTEM 64

3.1 Introduction ........................................................................................................ 64
3.2 Materials and Methods ....................................................................................... 66
3.2.1 Manure Collection and Storage ............................................................... 66

3.2.2 Ozone and Ozonation Reactor ................................................................. 67

3.2.3 Odor Evaluation ...................................................................................... 68

3.2.4 Chemical Analysis .................................................................................. 69

3.3 Results and Discussion ...................................................................................... 71
3.3.1 Effect of Storage and Ozonation ............................................................. 71

3.3.2 Comparison of Stripping, Oxygenation, and Ozonation ............................ 78

3.3.3 Comparison of Varying Temperature and Using Hydrogen Peroxide ....... 79

3.3.4 Effect of Subsequent Storage of Manure Slurry after Ozone Treatment...80

3.3.5 Economic Analysis of Pilot-scale Ozonation System ............................. 82

3.4 Conclusions ........................................................................................................ 85
3.5 References .......................................................................................................... 86

CHAPTER 4 MASS TRANSFER OF OZONE IN A SEMI-BATCH

 

STIRRED REACTOR - --107

4.1 Introduction ......................................................................................................... 107
4.2 Materials and Methods ...................................................................................... 108
4.3 Model Development .......................................................................................... 110
4.3.1 Mixed-Flow Model ................................................................................. 111

4.3.2 Plug-Flow Model .................................................................................... 115

4.4 Results and Discussion ...................................................................................... 118
4.4.1 Self-decomposition of Ozone in Water and Headspace ......................... 118

4.4.2 Transfer of Ozone in DDI Water (without pH buffer) ........................... 120

4.4.3 Transfer of Ozone in DDI Water (pH=2 acidified by sulfuric acid) ....... 123
4.4.4 Transfer of Ozone in DDI Water (pH=6.7 buffered by phosphate salt) .125

4.4.5 Comparison of Experimental Data and Model Prediction ...................... 129
4.4.6 Sensitivity Analysis of Mass-Transfer Model of Ozone ........................ 130
4.4.7 Application of Mass-transfer Model in a Pilot-scale System ................. 130
4.5 Conclusions ........................................................................................................ 131
4.6 References .......................................................................................................... 133

CHAPTER 5 OXIDATION KINETICS OF PHENOLIC AND INDOLIC
COMPOUNDS IN AQUEOUS ENVIRONMENT BY OZONE...152

5.1 Introduction ........................................................................................................ 152
5.2 Materials and Methods ....................................................................................... 153
5.2.1 Preparation of Buffer and Chemical Solutions .......................................... 153

vi

5.2.2 Stoichiometric Factor and Reaction Rate Constant .................................. 154
5.2.3 Factors Affecting the Removal of the Odorous Target

 

Compounds by Ozone ............................................................................... 155
5.2.4 The Ozonation of Synthetic and Real Swine Manure in
a Semi-batch Reactor .............................................................................. 156
5.3 Model Development ............................................................................................ 159
5.4 Results and Discussion ....................................................................................... 161
5.4.1 Stoichiometric Factors and Rate Constants for Ozonation Reactions ...... 161
5.4.2 Factors Affecting the Oxidation of Phenolic and Indolic Compounds
by Ozone ................................................................................................... 164
5.4.3 Comparison between Experimental Result and Model Prediction ........... 167
5.4.4 Oxidation of Synthetic Manure and Stored Swine manure by Ozone ...... 170
5.5 Conclusions .......................................................................................................... 174
5.6 References ............................................................................................................ 176
CHAPTER 6 CONCLUSIONS AND RECOMMENDATIONS 226
6.1 Conclusions ........................................................................................................... 226
6.2 Recommendations ................................................................................................. 229
APPENDIX ..................................................................................... 231

vii

LIST OF TABLES

Table 1.1 Identified Odorous Compounds in Livestock Waste ........................... 4
Table 1.2 Oxidation-reduction Potentials for Water Treatment Agents ................. 9
Table 2.1 The Structure of the Phenolic and Indolic Compounds .............................. 26
Table 2.2 The Concentrations of Each Volatile Fatty Acid during 12 Weeks

of Storage Period ......................................................................................... 33
Table 2.3 The Concentrations of Each Volatile Fatty Acid for Stored and

Ozonated Swine Manure Slurry ................................................................ 37
Table 3.1 The Characteristics of Fresh and Stored Swine Manure ............................ 73
Table 3.2 Evaluation of the Odor of Fresh and Stored Manure before and

after Ozonation ........................................................................................... 74
Table 3.3 The Systematic Parameters of Ozonating Stored Manure ......................... 77

Table 3.4 Analysis of Stored Ozonated Manure (1 g/L) over a Period of

Four Weeks ............................................................................................... 81
Table 3.5 The Capital Cost for the Pilot-scale Ozonation System ............................ 83
Table 3.6 The Operational Cost for the Pilot-scale System (Based on the

calculation at 0.5 g/L ozone dosage) ......................................................... 84
Table 3.7 The Calculation of Total Cost for the Pilot-scale System ........................ 84

Table 4.1 Self-decomposition of Dissolved Ozone in DDI Water under
Different Temperature ............................................................................... 120

Table 4.2 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Inlet Ozone Concentrations .................................... 121

Table 4.3 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Temperatures in DDI Water .................................. 122

viii

Iwhi

kai

Iwk‘

ka

ka

Iwh

ram

Id:

Id

Ta

Ta

Ta

Table 4.4 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Flowrates in DDI Water ........................................ 123

Table 4.5 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Ionic Strength in DDI Water
(Acidified to pH=2) ................................................................................. 124

Table 4.6 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Temperatures and Flowrates in DDI
Water (Acidified to pH=2) ..................................................................... 126

Table 4.7 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Temperatures and Flowrates in DDI
Water (pH=6.7 Buffered by Phosphate System) .................................... 127

Table 4.8 Enhancement Factors for the DDI Water (pH=6.7) Buffered
with Phosphate Solution ........................................................................ 128

Table 4.9 Fitting Parameters Obtained from the Mass-transfer Equations.
Effect of Changes in Temperatures and Agitation Speeds in DDI

Water in a Pilot-scale Reactor .............................................................. 131
Table 5.1 The Composition of Stock Buffer Solution ........................................ 154
Table 5.2 Concentrations of Malodors in the Synthetic Manure ........................ 157

Table 5.3 The Stoichiometric Factor and Rate Constants for the Reaction
of Ozone with Phenolic and Indolic Compounds under Different
pH Conditions (C.L.= Confidence Limits) ......................................... 162

Table 5.4 The Enhancement Factor Calculated under Different
Flowrates and pH Values ................................................................... 168

Table 5.5 The Reaction Coefficients for Phenols and Indoles under
Different Flowrates at pH 6.7 ............................................................ 170

Table 5.6 The Reaction Coefficients for Synthetic Manure Determined
at Different Flowrates ....................................................................... 172

Table 5.7 Chemical Characteristics of Real Swine Manure Taken from
the Storage Pit ................................................................................. 173

ix

Fig

Fig

n

H

LIST OF FIGURES

Figure 1.1 Decomposition of Ozone in Wastewater ................................................. 19
Figure 2.1 System Setup for the Storage Experiment ............................................... 43
Figure 2.2 The Photo of the Fermentation Tank for Storage Experiment ................ 44
Figure 2.3 System Setup for the Ozonation Experiment .......................................... 45
Figure 2.4 The Photo of the Dynamic Dilution Binary Scale Olfactometer ............ 46
Figure 2.5 Effect of Time on the Release of Ammonia and Hydrogen Sulfide

from Stored Swine Manure ..................................................................... 47
Figure 2.6 Effect of Time on the pH in Stored Swine Manure Slurry ...................... 47

Figure 2.7 Effect of Time on the Concentrations of Ammonia Nitrogen in
Stored Swine Manure Slurry .................................................................. 48

Figure 2.8 Effect of Time on the Concentrations of BODs in Stored Swine
Manure Slurry ......................................................................................... 48

Figure 2.9 Effect of Time on the Concentrations of COD in Stored Swine
Manure Slurry ......................................................................................... 49

Figure 2.10 Effect of Time on the Concentrations of Total Solids in Stored
Manure Slurry ....................................................................................... 49

Figure 2.11 Effect of Time on the Ratios of BOD5 to COD in Stored Swine
Manure Slurry ....................................................................................... 50

Figure 2.12 Effect of Time on the Concentrations of Soluble Phosphate in
Stored Swine Manure Slurry ................................................................ 50

Figure 2.13 Effect of Time on the Concentrations of Soluble Sulfate in
Stored Swine Manure Slurry ................................................................ 51

Figure 2.14 Effect of Time on the Ratios of Total Sulfides to Total Solids in

Figure I

Figure

Figure

Figure

F1 gure

Figure

Figure

Figun

FigUr

Figul

Stored Swine Manure Slurry ................................................................ 51

Figure 2.15 Effect of Time on the Concentrations of Total VFAs in Stored

Swine Manure Slurry ............................................................................ 52
Figure 2.16 Effect of Time on the Concentrations of Phenolic and Indolic

Metabolites in Stored Swine Manure Slurry ........................................ 52
Figure 2.17 Effect of Time on the Odor Intensity in Stored Swine Manure

Slurry ..................................................................................................... 53
Figure 2.18 Relationships between Odor Intensity and Concentrations of

Ammonia Nitrogen in Stored Swine Manure Slurry ............................. 54
Figure 2.19 Relationships between Odor Intensity and Concentrations of

Sulfides/Total Solids in Stored Swine Manure Slurry .......................... 55
Figure 2.20 Effect of Time and Ozonation on the pHs in Stored Swine .................. 56

Figure 2.21 Effect of Time and Ozonation on the VFA Concentrations in
Stored Swine Manure Slurry ................................................................ 56

Figure 2.22 Effect of Time and Ozonation on the Concentrations of Total
Sulfides in Stored Swine Manure Slurry .............................................. 57

Figure 2.23 Effect of Time and Ozonation on the Sulfate Concentrations in
Stored Swine Manure Slurry ................................................................ 57

Figure 2.24 Effect of Time and Ozonation on the Odor Intensity of Stored
Swine Manure Slurry ........................................................................... 58

Figure 2.25 Effect of Subsequent Storage on the Odor Intensity of
Ozonated Swine Manure Slurry ............................................................ 58

Figure 2.26 Effect of Time and Ozonation on the BOD5 Concentrations in
Stored Swine Manure Slurry .............................................................. .59

Figure 2.27 Effect of Time and Ozonation on the COD Concentrations in
Stored Swine Manure Slurry ............................................................... 59

Figure 2.28 Effect of Time and Ozonation on the Concentrations of Total
Solids in Stored Swine Manure Slurry ................................................ 60

Figure 2.29 Effect of Time and Ozonation on the Concentrations of
Ammonia Nitrogen in Stored Swine Manure Slurry ........................... 60

xi

Figur

Figur

Figur

Figur

Figur
Figur
Flgu:
Flgm

Figur

Figm

Fié’Ur

Figure 2.30 Effect of Time and Ozonation on the Concentrations of Soluble

Phosphate in Stored Swine Manure Slurry .......................................... 61
Figure 2.31 Effect of Time and Ozonation on Aerobes and Anaerobes in

Stored Swine Manure Slurry ............................................................... 61
Figure 2.32 Effect of Time and Ozonation on Total Coliforrns in Stored

Swine Manure Slurry ........................................................................... 62
Figure 2.33 Effect of Time and Ozonation on E. Coli in Stored Swine

Manure Slurry ...................................................................................... 62
Figure 2.34 Effect of Time and Ozonation on Coliphage Plaques Produced

on E. Coli in Stored Swine Manure Slurry .......................................... 63
Figure 3.1 The Collection System for the Swine Manure ....................................... 90
Figure 3.2 The Photo of the Manure Collection System ......................................... 91
Figure 3.3 The Configuration of Pilot-Scale System for Ozone Treatment . . . ......92
Figure 3.4 The Photo of Pilot-Scale System for ozone treatment ........................... 93

Figure 3.5 Changes in Phenol Concentration during the Storage and
Ozonation of Swine Waste Slurry ......................................................... 94

Figure 3.6 Changes in p-Cresol Concentration during the Storage and
Ozonation of Swine Waste Slurry ......................................................... 95

Figure 3.7 Changes in p—Ethylphenol Concentration during the Storage and
Ozonation of Swine Waste Slurry ......................................................... 96

Figure 3.8 Changes in Skatole Concentration during the Storage and
Ozonation of Swine Waste Slurry ......................................................... 97

Figure 3.9 Changes in Number of Total Aerobes during the Storage and
Ozonation of Swine Waste Slurry ......................................................... 98

Figure 3.10 Changes in Number of Total Anaerobes during the Storage and
Ozonation of Swine Waste Slurry ........................................................ 99

Figure 3.11 Changes in Number of Coliphage during the Storage and
Ozonation of Swine Waste Slurry ..................................................... 100

Figure 3.12 Effect of Different Treatment Processes on the Removal of

xii

Phenol (The flowrate of ozone was 25 mg/L-min) ........................... 101

Figure 3.13 Effect of Different Treatment Processes on the Removal of

p-Cresol (The flowrate of ozone was 25 mg/L—min) 101
Figure 3.14 Effect of Different Treatment Processes on the Removal of
p-Ethylphenol (The flowrate of ozone was 25 mg/L-min) ................. 102
Figure 3.15 Effect of Different Treatment Processes on the Removal of
Skatole (The flowrate of ozone was 25 mg/L-min) 102
Figure 3.16 Effect of Temperature on the Removal of Phenol 103
Figure 3.17 Effect of Temperature on the Removal ofp-Cresol 103
Figure 3.18 Effect of Temperature on the Removal of p-Ethylphenol . . . . . 104
Figure 3.19 Effect of Temperature on the Removal of Skatole 104
Figure 3.20 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of Phenol ........................................................................... 105
Figure 3.21 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of p—Cresol ........................................................................ 105
Figure 3.22 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of p-Ethylphenol ............................................................... 106
Figure 3.23 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of Skatole .......................................................................... 106
Figure 4.1 The System Setup for the Semi-batch Ozonation Process . . . . ..137
Figure 4.2 The Photo of the System Setup for the Semi-batch
Ozonation Process ............................................................................... 138
Figure 4.3 First-order Decomposition of Dissolved Ozone in DDI Water .......... 139

Figure 4.4 Second-order Decomposition of Dissolved Ozone in DDI
Water (pH=6.7) ................................................................................... 140

Figure 4.5 System Setup for Determining the Self-decomposition rate of
Moisturized Gaseous Ozone ................................................................. 141

Figure 4.6 Method to Fit the Mass-transfer and Partition Coefficients by

xiii

Figure

Figure

Figun

Linear Regression .............................................................................

Figure 4.7 Relationship between Mass-transfer Coefficient of Ozone and

Superficial Velocity .........................................................................

Figure 4.8 Comparison of Dissolved Ozone Concentration Calculated by
Analytical and Numerical Solutions in Mixed-flow Model for

DDI Water at Flowrate 400 mUmin and Temperature 22 C .. . .

Figure 4.9 Comparison of Exit Ozone Concentration Calculated by
Analytical and Numerical Solutions in Mixed-flow Model
for DDI Water at Flowrate 400 mUmin and Temperature

22C .................................................................................................

Figure 4.10 Comparison of Dissolved ozone Concentration in DDI Water
Obtained Model Prediction with that Experimentally Determined

at Flowrate 400 mL/min and Temperature 22 C ............................

Figure 4.11 Comparison of Exit Ozone Concentration in DDI Water
Obtained Model Prediction with that Experimentally Determined

at Flowrate 400 mIJmin and Temperature 22 C ............................

Figure 4.12 Comparison of Dissolved Ozone Concentration in DDI Water
(pH 2.1) Obtained Model Prediction with that Experimentally

Determined at Flowrate 400 mIJmin and Temperature 22 C .......

Figure 4.13 Comparison of Exit Ozone Concentration in DDI Water
(pH 2.1) Obtained Model Prediction with that Experimentally

Determined at Flowrate 400 mIJmin and Temperature 22 C .......

Figure 4.14 Comparison of Dissolved Ozone Concentration in DDI Water
(pH 6.7) Obtained Model Prediction with that Experimentally

Determined at Flowrate 400 mUmin and Temperature 22 C .......

Figure 4.15 Comparison of Exit Ozone Concentration in DDI Water
(pH 6.7) Obtained Model Prediction with that Experimentally

...142

...142

143

...143

...144

...144

..145

..145

..146

Determined at Flowrate 400 mIJmin and Temperature 22 C ......... 146

Figure 4.16 Sensitivity Profile of Dissolved and Exit ozone Concentrations

By Changing the Mass-transfer Coefficient in DDI Water ..........

Figure 4.17 Sensitivity Profile of Dissolved and Exit ozone Concentrations

..147

By Changing the Partition Coefficient in DDI Water .................. 148

Figure 4.18 Sensitivity Profile of Dissolved and Exit ozone Concentrations

xiv

By Changing the Self-decomposition Coefficient in DDI Water 148

Figure 4.19 Sensitivity Profile of Dissolved and Exit ozone Concentrations
By Changing the Gas Hold-up in DDI Water ............................ 149

Figure 4.20 Comparison of Dissolved Ozone Concentration in DDI Water
in Pilot-scale System Obtained by Model Prediction and
Experimental Data at Flowrate 400 mIJmin and
Temperature 22 C ............................................................ 150

Figure 4.21 Comparison of Exit Ozone Concentration in DDI Water
in Pilot-scale System Obtained by Model Prediction and
Experimental Data at Flowrate 400 mUmin and
Temperature 22 C ............................................................ 151

Figure 5.1 The System Setup for the Semi-batch Ozonation Process ............... 179

Figure 5.2 The Effect of Flowrate on the Oxidation of Phenol in the Mixture
of Phenolic Compounds at pH 6.7 and Temperature 22 C ............... 180

Figure 5.3 The Effect of Flowrate on the Oxidation of p-Cresol in the Mixture
of Phenolic Compounds at pH 6.7 and Temperature 22 C ............... 180

Figure 5.4 The Effect of Flowrate on the Oxidation of p-Ethylphenol in
the Mixture of Phenolic Compounds at pH 6.7 and
Temperature 22 C .............................................................. 181

Figure 5.5 The Effect of Flowrate on the Oxidation of Indole in the Mixture
of Indolic Compounds at pH 6.7 and Temperature 22 C 181

Figure 5.6 The Effect of Flowrate on the Oxidation of Skatole in the Mixture
of Indolic Compounds at pH 6.7 and Temperature 22 C ................ 182

Figure 5.7 The Effect of Ozone Dosage on the Oxidation of Phenol in the
Mixture of Phenolic Compounds by Different Flowrates at pH
6.7 and Temperature 22 C ...................................................... 182

Figure 5.8 The Effect of Ozone Dosage on the Oxidation of p-Cresol in the

Mixture of Phenolic Compounds by Different Flowrates at pH

6.7 and Temperature 22 C ...................................................... 183
Figure 5.9 The Effect of Ozone Dosage on the Oxidation of p-Ethylphenol

in the Mixture of Phenolic Compounds by Different Flowrates at

pH 6.7 and Temperature 22 C .................................................. 183

Figure 5.10 The Effect of Ozone Dosage on the Oxidation of Indole in the

XV

isuri

F1901:

F1 gu

F1 gu

Hymin

Figure 5.12

Figure 5.13

Figure 5.14

Figure 5.15

Figure 5.16

Figure 5.17

fiymim

Figure 5.19

Figure 5.20

Figure 5.21

Mixture of Indolic Compounds by Different Flowrates at pH
6.7 and Temperature 22 C ..................................................... 184

The Effect of Ozone Dosage on the Oxidation of Skatole in the
Mixture of Indolic Compounds by Different Flowrates at pH
6.7 and Temperature 22 ...................................................... 184

Effect of pH on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at Flowrate 300 mL/min and
Temperature 22 C ............................................................ 185

Effect of pH on the Oxidation of p-Cresol in the Mixture of
Phenolic Compounds at Flowrate 300 mL/min and
Temperature 22 C ............................................................ 185

Effect of pH on the Oxidation of p-Ethylphenol in the Mixture of
Phenolic Compounds at Flowrate 300 ijmin and
Temperature 22 C ............................................................ 186

Effect of pH on the Oxidation of Indole in the Mixture of
Indolic Compounds at Flowrate 300 mUmin and
Temperature 22 C ............................................................ 186

Effect of pH on the Oxidation of Skatole in the Mixture of
Indolic Compounds at Flowrate 300 mIJmin and
Temperature 22 C ............................................................ 187

Effect of Temperature on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at Flowrate 300 mL/min and
Temperature 22 C ............................................................ 187

Effect of Temperature on the Oxidation of p-Cresol in the
Mixture of Phenolic Compounds at Flowrate 300 mIJmin and
Temperature 22 C ............................................................ 188

Effect of Temperature on the Oxidation of p-Ethylphenol in
the Mixture of Phenolic Compounds at Flowrate 300 mUmin
and Temperature 22 C ....................................................... 188

Effect of Temperature on the Oxidation of Indole in the Mixture of
Indolic Compounds at Flowrate 300 mL/min and
Temperature 22 C ............................................................ 189

Effect of Temperature on the Oxidation of Skatole in the Mixture of
Indolic Compounds at Flowrate 300 mL/min and

xvi

0:

a1: \

QL\

Fig

IA

Temperature 22 C ............................................................ 189

Figure 5.22 Effect of Dextran on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 190

Figure 5.23 Effect of Dextran on the Oxidation of p-Cresol in the Mixture of
Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 190

Figure 5.24 Effect of Dextran on the Oxidation of p-Ethylphenol in the Mixture
of Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 191

Figure 5.25 Effect of Dextran on the Oxidation of Indole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 191

Figure 5.26 Effect of Dextran on the Oxidation of Skatole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 192

Figure 5.27 Effect of Ammonia on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 192

Figure 5.28 Effect of Ammonia on the Oxidation of p—Cresol in the Mixture
of Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 193

Figure 5.29 Effect of Ammonia on the Oxidation of p-Ethylphenol in the
Mixture of Phenolic Compounds at pH 6.7, Flowrate 300
mIJmin, and Temperature 22 C ............................................ 193

Figure 5.30 Effect of Ammonia on the Oxidation of Indole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 194

Figure 5.3] Effect of Ammonia on the Oxidation of Skatole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 194

Figure 5.32 Effect of VFAs on the Oxidation of Phenol in the Mixture of

Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 195

xvii

Figu

Figu

Figure 5.33 Effect of VFAs on the Oxidation of p-Cresol in the Mixture of
Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 195

Figure 5.34 Effect of VFAs on the Oxidation of p-Ethylphenol in the Mixture
of Phenolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 196

Figure 5.35 Effect of VFAs on the Oxidation of Indole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 196

Figure 5.36 Effect of Ammonia on the Oxidation of Skatole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mIJmin, and
Temperature 22 C ............................................................ 197

Figure 5.37 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 2.1, Flowrate 300 mIJmin, and Temperature 22 C ................ 197

Figure 5.38 Comparison of p-Cresol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 2.1, Flowrate 300 mIJmin, and Temperature 22 C ................ 198

Figure 5.39 Comparison of p-Ethylphenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 2.1, Flowrate 300 mIJmin, and Temperature 22 C ................ 198

Figure 5.40 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 2.1, Flowrate 300 mIJmin, and Temperature 22 C ................ 199

Figure 5.41 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 2.1, Flowrate 300 mIJmin, and Temperature 22 C ................ 199

Figure 5.42 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 100 mIJmin, and Temperature 22 C ................ 200

Figure 5.43 Comparison of p—Cresol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 100 mIJmin, and Temperature 22 C ................ 200

Figure 5.44 Comparison of p-Ethylphenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at

xviii

Figure

Figure

Figure

figure

Figure

pH 6.7, Flowrate 100 mIJmin, and Temperature 22 C ................ 201

Figure 5.45 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 300 mIJmin, and Temperature 22 C ................ 201

Figure 5.46 Comparison of p-Cresol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 300 mIJmin, and Temperature 22 C ................ 202

Figure 5.47 Comparison of p-Ethylphenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 300 mIJmin, and Temperature 22 C ................ 202

Figure 5.48 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 500 mL/min, and Temperature 22 C ................ 203

Figure 5.49 Comparison of p-Cresol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 500 mIJmin, and Temperature 22 C ................ 203

Figure 5.50 Comparison of p—Ethylphenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 500 mL/min, and Temperature 22 C ................ 204

Figure 5.51 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 100 mIJmin, and Temperature 22 C ................ 204

Figure 5.52 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 100 mIJmin, and Temperature 22 C ................ 205

Figure 5.53 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 300 mL/min, and Temperature 22 C ................ 205

Figure 5.54 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 300 mIJmin, and Temperature 22 C ................ 206

Figure 5.55 Comparison of Indole Concentration in the Mixture of Indolic

Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 500 mIJmin, and Temperature 22 C ................ 206

xix

FumeS

HymS

Emmi

Emmi

Hflmf

Figure 1

Figure :
Figme :
53m:
ERR:
ERR:

Hfimf

Bfims

Figure 5.56 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction at
pH 6.7, Flowrate 500 mIJmin, and Temperature 22 C ................ 207

Figure 5.57 Effect of Temperature on the Oxidation of Phenol in Synthetic
Manure by Ozone at Flowrate 300 mIJmin, and
Temperature 22 C ................................................................. 207

Figure 5.58 Effect of Temperature on the Oxidation of p—Cresol in Synthetic
Manure by Ozone at Flowrate 300 mIJmin, and
Temperature 22 C ................................................................. 208

Figure 5.59 Effect of Temperature on the Oxidation of p-Ethylphenol in Synthetic
Manure by Ozone at Flowrate 300 mIJmin, and
Temperature 22 C ................................................................. 208

Figure 5.60 Effect of Temperature on the Oxidation of Indole in Synthetic
Manure by Ozone at Flowrate 300 mIJmin, and
Temperature 22 C ................................................................. 209

Figure 5.61 Effect of Temperature on the Oxidation of Skatole in Synthetic
Manure by Ozone at Flowrate 300 mIJmin, and
Temperature 22 C ................................................................. 209

Figure 5.62 Effect of Flowrate on the Oxidation of Phenol in Synthetic
Manure at Flowrate 300 mIJmin, and Temperature 22 C ............... 210

Figure 5.63 Effect of Flowrate on the Oxidation of p—Cresol in Synthetic
Manure at Flowrate 300 mIJmin, and Temperature 22 C ............... 210

Figure 5.64 Effect of Flowrate on the Oxidation of p-Ethylphenol in Synthetic
Manure at Flowrate 300 mIJmin, and Temperature 22 C ............... 211

Figure 5.65 Effect of Flowrate on the Oxidation of Indole in Synthetic
Manure at Flowrate 300 mIJmin, and Temperature 22 C ............... 211

Figure 5.66 Effect of Flowrate on the Oxidation of Skatole in Synthetic
Manure at Flowrate 300 mIJmin, and Temperature 22 C ............... 212

Figure 5.67 Effect of Ozone Dosage on the Oxidation of Phenol in Synthetic
Manure by Different Flowrates at Flowrate 300 mIJmin,
at Temperature 22 C .......................................................... 212

Figure 5.68 Effect of Ozone Dosage on the Oxidation of p-Cresol in Synthetic
Manure by Different Flowrates at Flowrate 300 mIJmin,

XX

F

>——

gure

Figure

Figure

Figure

Figure

Figure

FigUre

Figure

at Temperature 22 C .......................................................... 213

Figure 5.69 Effect of Ozone Dosage on the Oxidation of p—Ethylphenol in
Synthetic Manure by Different Flowrates at Flowrate 300 mL/min,
at Temperature 22 C .......................................................... 213

Figure 5.70 Effect of Ozone Dosage on the Oxidation of Indole in Synthetic
Manure by Different Flowrates at Flowrate 300 mIJmin,
at Temperature 22 C .......................................................... 214

Figure 5.71 Effect of Ozone Dosage on the Oxidation of Skatole in Synthetic
Manure by Different Flowrates at Flowrate 300 mIJmin,
at Temperature 22 C .......................................................... 214

Figure 5.72 Change of VFAs Concentration in Synthetic Manure by Ozonation
Flowrate 300 mIJmin and Temperature 22 C ............................. 215

Figure 5.73 Comparison of Phenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 100 mIJmin, and Temperature 22 C .......................... 215

Figure 5.74 Comparison of p-Cresol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 100 mIJmin, and Temperature 22 C .......................... 216

Figure 5.75 Comparison of p-Ethylphenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 100 mIJmin, and Temperature 22 C .......................... 216

Figure 5.76 Comparison of Indole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 100 mIJmin, and Temperature 22 C .......................... 217

Figure 5.77 Comparison of Skatole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 100 mIJmin, and Temperature 22 C .......................... 217

Figure 5 .78 Comparison of Phenol Concentration in the Synthetic Manure

between Experimental Data and Model Prediction at pH=6.7,

Flowrate 300 mIJmin, and Temperature 22 C .......................... 218
Figure 5.79 Comparison of p—Cresol Concentration in the Synthetic Manure

between Experimental Data and Model Prediction at pH=6.7,

Flowrate 300 mIJmin, and Temperature 22 C .......................... 218

Figure 5.80 Comparison of p-Ethylphenol Concentration in the Synthetic Manure

xxi

Figt

between Experimental Data and Model Prediction at pH=6.7,
Flowrate 300 mIJmin, and Temperature 22 C .......................... 219

Figure 5.81 Comparison of Indole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 300 mIJmin, and Temperature 22 C .......................... 219

Figure 5.82 Comparison of Skatole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 300 mIJmin, and Temperature 22 C .......................... 220

Figure 5.83 Comparison of Phenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 500 mL/min, and Temperature 22 C .......................... 220

Figure 5.84 Comparison of p-Cresol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 500 mIJmin, and Temperature 22 C .......................... 221

Figure 5.85 Comparison of p-Ethylphenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 500 mIJmin, and Temperature 22 C .......................... 221

Figure 5.86 Comparison of Indole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 500 mIJmin, and Temperature 22 C .......................... 222

Figure 5.87 Comparison of Skatole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 500 mIJmin, and Temperature 22 C .......................... 222

Figure 5. 88 Effect of Oxygenation on the Concentration of Phenolic and
Indolic Compounds in Real Swine Manure at Flowrate 300
mIJmin and Temperature 22 C ........................................... 223

Figure 5. 89 Effect of Ozonation on the Concentration of Phenolic and
Indolic Compounds in Real Swine Manure at Flowrate 300
mIJmin and Temperature 22 C ........................................... 223

Figure 5.90 Comparison of Phenol Concentration in the Real Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 300 mIJmin, and Temperature 22 C .......................... 224

Figure 5.91 Comparison of p-Cresol Concentration in the Real Manure
between Experimental Data and Model Prediction at pH=6.7,

xxii

humi‘

fing}

 

 

Flowrate 300 mIJmin, and Temperature 22 C .......................... 224

Figure 5.92 Comparison of p-Ethylphenol Concentration in the Real Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 300 mIJmin, and Temperature 22 C .......................... 225

Figure 5.93 Comparison of Skatole Concentration in the Real Manure
between Experimental Data and Model Prediction at pH=6.7,
Flowrate 300 mIJmin, and Temperature 22 C .......................... 225

xxiii

11 En‘
Ei
animal
intense
Surrou
11131 re
enter;
PTCSSL
the pr
iOrmi
I
dam2t
gOVer
induS
Opfira-
decisj

marlru'

Chapter 1

INTRODUCTION

1.1 Environmental Significance

Environmental pollution by livestock waste is a major nationwide concern of the
animal agriculture industry. As livestock production has moved to larger and more
intense production units, odor complaints have increased from the residents of
surrounding communities (Ritter, 1989). There is a growing public intolerance of odors
that result from modern farming practices. The increase in the size of livestock
enterprises and the increasing number of people living in rural areas will result in
pressure on livestock producers to take positive actions to reduce this odor. Therefore,
the problem of handling the large amounts of animal wastes generated on farms is a
formidable task and is of significant concern to agricultural community.

An increasing number of civil litigations, more restrictive township ordinances,
demands for the protection of public health and new Federal and State regulations
governing surface run-off and groundwater contamination necessitate that the livestock
industry take actions to control the environmental problems associated with livestock
operations. George et a1. (1985) summarized the US Midwest odor litigations and
decisions. They recommended that odor concerns should be addressed during initial

planning, site selection and design. They also concluded that day-to-day management

actions a
developr
The
the Light
nutrient
concent
the soil
inject t1.
101116 2
proble
condit
10 C011
\he le
applic
time;

at [in

1.2 t

SPTE

are

C125

actions and public relations might be the most significant determinants as to the
development and outcome of any conflict or litigation regarding an odor problem.

The application of animal wastes to agricultural lands poses significant problems to
the agricultural industry, despite all of the benefits of waste application in terms of
nutrient addition. Since animal wastes contain very high phosphorus and nitrogen
concentrations, the application of wastes can cause excessively high nutrient loadings to
the soil. The problem is exacerbated by the use of manure injection techniques which can
inject the manure below the plant root zone, thus, increasing potential nutrient migration
to the groundwater. Odors emanating from livestock wastes can also exacerbate the 1
problem as farmers are not necessarily able to apply manure to the land when the soil
conditions are optimal because of such constraints as wind direction. Innovative methods
to control, reduce or eliminate odors associated with livestock wastes would mean that
the less malodorous wastes (1) would not need to be injected and could be surface-
applied; as such, the nutrients would be available to the plant roots for a longer period of
time; and (2) could be applied to the soil at the time of optimal soil conditions, rather than

at times such as optimal weather conditions.

1.2 Original and Identification of Malodorous Substances in Livestock Waste

Odors from swine farms emanate from the ventilation air from the buildings, the
odors from the slurry handling near the piggery and in the storage pits, and from
spreading slurry on land. (Sneath et a1., 1992). Generally odors from the livestock wastes
are a result of anaerobic microbial decomposition. Mackie (1994) indicated that four

classes of odorous components, including volatile fatty acids (VFA), phenols and indoles,

ammonia and volatile amines, and volatile sulfur-containing compounds, are the normal
bacterial metabolic end-products from the breakdown of amino acids. Welsh et a1. (1977)
reported the anaerobic degradation of cellulose, lipids, proteins, and other complex
organic materials results in intermediate fermentation products which are responsible for
most odors in anaerobic storage of manure. Many researchers have identified specific
odorous compounds of livestock waste odors. Barth et a1. (1984) reported that more than
75 specific odorous compounds have been identified in livestock waste. O’Neill and
Phillips (1992) reported that 168 odorous compounds are present in livestock waste and
in the air surrounding these operations. Odor producing compounds have been measured
by gas chromatography, wet chemistry and organoleptic tests (Ritter, 1989). These
compounds are end products or intermediate products of biological reactions and include:
volatile organic acids, alcohol, aldehydes, fixed gases, carbonyls, esters, amines, sulfides,
mercaptans, and nitrogen heterocycles. Some of the specific compounds identified in

swine waste are listed in Table 1.1.

1.3 Odor Measurement
There are five basic approaches to odor detection and measurement. All are based

on the nose, wet chemistry and gas chromatography (Ritter, 1989). These methods are:

(a) identification of odor gases;
(b) measurement of odorant concentration;

(c) measurement of odor intensity by vapor dilution;

l..—

 

I’." £79153

Table 1.1 Identified Odorous Compounds in Livestock Waste

 

 

 
 

 

 

 

 

 

 

 

 

'.'Niir‘¢3érifcampouhdé ‘ [Volatile Fatty Acids . -5311fuf‘COIDPeWtdS ' ‘. , ,
mmonia ‘ , Methanoic acid Carbonylsulfide
ethylarmne Ethanoic acid Hydrogen sulfide
thylamine propanoic acid Methanethiol

Trimethylarnine Butanoic acid Dimethylsulfide

Triethylamine 2-Methylpropanoic acid Dimethyldisulfide

Pentanoic acid Dimethyltrisulfide

j Alcohols _ if, 3-Methylbutanoic acid Diethyldisulfide
ethanol I Hexanoic acid Propanethiol
thanol 4-Methylpentanoic acid Butanethiol

1-Propanol Heptanoic acid Dipropldisulfide

2-Propanol Octanoic acid 2-Methylthiophene

l-Butanol Nonanoic acid Propylprop-l-ethyldisulfide
-Butanol Phenylacetic acid 2,4-Dimethylthiophene
2-Methyl-1-Propanol 2-Phenylpropanoic acid 2-Methylfuran

3-Methyl- 1 -butanol

2-Ethoxy-1-propanol .. D’Arom‘atics

2-Methyl-2-pentanol Phenol

2,3-Butanediol 4-Methylphenol

4-Ethylphenol
l] i Aldehydcsi fi‘ Toluene
ethanal Xylene
thanal Indole
opanal Benzaldehyde
utanal Benzoic acid
entanal Methylphthalene
exanal Skatole
tanal Acetophenone
anal o-Aminoacetophenone

-Methyl- 1 -Propanal

Aneline

 

 

 

 

((1) measurement of odor intensity by liquid dilution; and
(e) ranking of odor intensities by arbitrary scales based upon the offensiveness of the
odon

Elliott et al. (1978) reviewed the analytical methods for detecting odors. They
concluded that quick and economical means of measuring odors from waste management
systems are needed in order to develop odor control technology. Although the nose can
be used to measure odor intensity, vapor and liquid dilution are more popular methods for
measuring odor intensity. However, vapor dilution methods are probably more popular
than liquid dilution techniques because they are easier tests and can be done both in the
laboratory and field (Ritter, 1989). A dynamic dilution binary scale olfactometer using
air/odorant vapor mixtures at 8 different concentrations has been developed by IIT
Research Institute to measure the odor intensity (Dravnieks, 1974). The liquid dilution
technique involves mixing odor-containing material with odor-free water. Several
dilutions are generally made and then the odorous dilution samples along with odor-free
samples are evaluated by a panel of observers. The odor intensity can be determined by
using odor intensity index which depends on the dilution ratio of odor-free water to
sample used. Sobel (1972) evaluated odor quality with a panel ranking the offensiveness
of the odor on a scale from 0 (non-offensiveness) to 10 (very strong offensiveness).
Williams ( 1984) indicated that odor offensiveness should be used rather than intensity to
evaluate the change of odors. In recent years, an electronic nose has been commercially
developed in the food and flavor industry for quality control. Gardner and Bartlett (1994)
defined an electronic nose system as an instrument which consists of an array of

electronic chemical sensors with partial specificity and an appropriate pattem-recognition

system c
material
metals.
Persaut
arrays
volatil
result;
Odon
slum
for n
has 1

6x“

Oil 2

Ci

system capable of recognizing simple or complex odors. They list a range of active
materials which have been used in sensor arrays, including sintered metal oxide, catalytic
metals, lipid layers, phthalocyanins, conducting polymers and organic semi-conductors.
Persaud et a1. (1996) have been developing an instrument (the Odorrnapper) with sensor
arrays based on conducting polymers which have been shown to respond well to the
volatile compounds found in the headspace above pig slurries and give reproducible
results over a three month period. Hobbs et al. (1995), using an earlier version of the
Odormapper, demonstrated its ability to discriminate between odors from pig and chicken
slurries, but reported low sensitivity of the instrument when compared with Olfactometry
for measuring odor concentration. Recent development in the electronic nose technique
has led to new applications in the assessment of odor from livestock waste. It is
expected, eventually, that the electronic nose will be favored over traditional
Olfactometry, as it can avoid the subjectivity associated with the human nose and

perception.

1.4 Current Technologies to Control Odor

A wide variety of biochemical and chemical products are being sold to treat and
prevent odors in feedlots, liquid waste storage tanks and lagoons. There are six
categories of odor controlling agents:
(1) masking agents, which are mixtures of aromatic oils that have a strong particular

odor of their own and are designed to cover up the waste odor with a more desirable

one;

(4)

Bu
“‘3
inc

Ch

de

(2) counteractants, which are mixtures of aromatic oils that cancel or neutralize the waste
odor so that the intensity of the mixture is less than that of the constituents;

(3) digestive deodorants, which contain bacteria or enzymes that eliminate odors through
modifications in biochemical digestive processes;

(4) adsorbents, which are products having a large surface area that may be used to adsorb
the odors before they are released to the environment;

(5) feed additives, which are compounds added to feed to improve animal performance
and reduce odors;

(6) chemical deodorants, which are strong oxidizing or germicidal agents that alter or
eliminate the bacteria responsible for odor production or chemically oxidize odorous
compounds.

A variety of masking agents and counteractants are available for odor control.
Burnett and Dondero (1970) evaluated 38 masking agents and counteractants on poultry
waste along with several digestive deodorants and chemical deodorants. The results
indicated that, in general, masking agents and counter agents are the most effective, while
chemical deodorants are moderately effective and digestive deodorants least effective.
Despite research data to indicate the limited success of digestive deodorants, digestive
deodorants are the most popular of the proprietary chemicals sold for odor control
because of aggressive marketing by some companies (Ritter, 1989). Williams and
Schiffman (1995) indicated that chemical deodorants did not significantly improve the
odor. Powers et al. (1995) also suggested that the odor of liquid dairy manure was not
improved after the use of commercial additives. Chemical deodorants are either

oxidizing agents or disinfectants. Oxidizing agents transform odorous compounds into

less offensive ones by chemical oxidation. Disinfectants alter or eliminate the bacterial
action that is responsible for odor production. Strong oxidizing agents also act as
disinfectants via their ability to degrade oxidatively enzymatic proteins (Ritter, 1989).

In addition to methods of controlling odor, two biological processes, anaerobic
digestion and aerobic lagoon, have been developed for manure management and odor
removal (Powers et al., 1995; Burton, 1992; Sneath et al., 1992; Williams et al., 1984;
Evans et al., 1983; Welsh et al., 1977). Although anaerobic digestion provides some
benefits including energy production, odor control, waste stabilization, and digester
residual as a feed component, there are very few farm digesters in operation in North
America. Major problems with the digester include waste handling, gas leakage, and
pipe and valve corrosion (Ritter, 1989). Although anaerobic digestion reduces odors, the
digested effluent is still a source of odor when it is spread on land (Voerrnans, 1986).
Aerobic treatment of animal slurries represents an increasingly popular option for
producers in the management of animal waste (Burton, 1992). Odors will not be
problematic if aerobic conditions are maintained. However, when malfunctions or other
circumstances cause aerobic systems to go anaerobic, odors will be released. As lagoons
or oxidation ponds require large land areas and the mechanically aerated systems used

have high energy requirements, aerobic systems are expensive to operate (Ritter, 1989).

1.5 Background of Using Ozone Treatment
Ozone, the triatomic allotrope of oxygen, has the formula 03 and generally exists as
a relatively unstable, reactive gas. Commercially, large quantities of ozone are produced

in specially engineered corona discharges (Carlins and Clark, 1982). Ozone is a very

powerful oxidant (E° = 2.07 volts), which is capable of reacting with numerous of organic
chemicals. It is more powerful than most of the other oxidants currently used in water

treatment (Rice and Browning, 1981) (see Table 1.2).

Table 1.2 Oxidation-reduction Potentials for Water Treatment Agents

   

Reactions Potential in Volts (E°) at 25 °C

 

F2 + 26' = 2F' 2.87
03 + 2H+ +26- = 02 +H20 2.07
H202 + 2H+ + 26' = 2H20 (acid) 1.76
MnO4' + 4H+ +36' = MnOz + 2H20 1.68
HC102 + 3H+ +46' = C1' + H20 1.57
HOCl + H+ + 26' = C1' + H20 1.49
C12 + 26' = 2Cl' 1.36
HOBr + H+ + 26' = Br' + H20 1.33
Brz + 26' = ZBr' 1.07
c102...“ + e' = C102' 0.95
H202 + 2H3O+ + 26' = 4H20 (base) 0.87

 

 

 

In an aqueous solution, ozone may react with compounds (M) either:
0 by direct reactions with molecular ozone, or
o by indirect reactions with the radical species that are formed when ozone
decomposes in water.
The resonance structures of the ozone molecule make it as a dipole, an electrophile, and
a nucleophile. Therefore, the direct reactions include three kinds of elementary
reactions, namely such as cyclo addition, electrophilic reaction, and nucleophilic reaction
(Langlais et al., 1990). Ozone can react with aromatic compounds directly or indirectly.

The direct reaction is selective, favoring compounds with electron donating groups (e.g.,

10

OH and NH;). Compounds with electron withdrawing groups (e.g., N 02 and C1) are not
easily oxidized by ozone. However, these compounds can be oxidized with ozone
indirectly using OH radicals.

Figure 1.1 illustrates the reaction of aqueous ozone in water (Masten and Davies,
1994). Hydroxide ions initiate ozone decomposition by reacting with ozone, producing
superoxide anions (0;). Ozone then reacts with the superoxide anion to form ozonide
radical ions and oxygen. The protonated ozonide ion rapidly decomposes to produce
oxygen and hydroxyl radicals. The hydroxyl radicals decompose ozone further,
producing more superoxide ions, which enter into the cyclic reaction. The stability of
dissolved ozone is affected by pH, ultraviolet (UV) light, ozone concentration, and the
concentration of radical scavengers.

Ozone is recognized as an effective means for the treatment of water, industrial
effluents and sewage. Ozonation is used extensively in Europe for drinking water
treatment; there are well over 1000 drinking water treatment plants using ozone in
Europe. More than 200 drinking water treatment plants in North America presently use
ozone (IOA News, 1997; Debevec, 1990; Fressonet-Chambarlhac et al., 1983; Glaze,
1987). Ozonation is used in over 40 wastewater treatment plants in the United States
(Masten and Davies, 1994). While ozone has been used for odor removal in municipal
wastewater treatment plants, it has not, to our knowledge, been used for odor control for
stored livestock waste facilities. Ozonation has the potential to reduce odors in stored
livestock waste by (1) controlling the rate of production of odorous metabolites by killing
the odor-causing bacteria and (2) oxidizing odorous metabolites produced as a result of

fermentation. The ozonation of agricultural wastes could, potentially, reduce the

ll

likelihood of contamination of groundwater and surface run-off by pathogenic bacteria

and viruses that are often found in agricultural wastes. As ozone is a very strong oxidant

which attacks the cell wall and the cytoplasmic membrane of the cell, pathogens can be

inactivated. Thus, ozone can be used not only to reduce the concentration of odor-

causing chemicals, some of which are harmful to the animals, but also to reduce the

numbers of pathogens present in the water.

1.6 Research Hypotheses

In this research we evaluated, in a bench-scale and pilot-scale systems, the

effectiveness of ozone for the treatment of swine manure slurry. The following

hypotheses were tested:

1.

The fermentation of manure slurries during storage results in an increase in the odor

emanating from swine waste.

Odorous gases, such as ammonia and hydrogen sulfide, are expected to be released

from stored manure waste.

Ozone, with its strong oxidation power, is very effective in removing malodorous
chemical compounds from fresh or stored manure waste, and the ozone-treated

manure waste will no longer be malodorous.

Ozone can effectively kill the bacteria and pathogens that normally exist in untreated

swine waste.

. The mass transfer or absorption of ozone in the liquid slurry can be enhanced by the

proper design of the pilot treatment system. In other words, the applied ozone dosage

can be specified that will treat manure waste to an acceptable level.

12

6. Temperature affects the rate of the reaction between ozone and malodorous
components. Thus, adjusting ozone dosages will be necessary to remove the odor at

different temperatures.

7. The combination of hydrogen peroxide with ozone (advanced oxidation process), thus
generating more reactive radicals, will remove odorous compounds more efficiently
than the ozonation process alone.

8. A model including mass transfer and oxidation kinetics can be developed and used to
predict the disappearance of malodorous compounds in the swine manure. The
concentration of COD (Chemical Oxygen Demand) can be an indicative parameter of

the consumption of ozone in the treatment system.

1.7 Research Objective

This study was conducted in four phases. The first phase focused on laboratory
studies to investigate the effect of ozonation on the physical, chemical, biological and
odor characteristics of the swine waste slurry. Changes in the characteristics of the
untreated manure waste during storage were studied. In the second phase, we designed
and constructed a pilot-scale ozonation system and tested it in a field situation to develop
and demonstrate the technology that is appropriate for commercial swine operation. In
the third phase, mass transfer theory was explored to develop the model that could be
used to describe the transport of ozone in the semi-batch reactor. In the fourth phase, the
kinetics and the reactions of ozone with five malodorous compounds that exist in the
swine manure (phenol, p-cresol, p-ethylphenol, indole, and Skatole) were investigated. In

addition, the effect of physical and chemical parameters on ozonation process was

13

determined at this phase. Finally, a model combining mass transfer and oxidation
kinetics was developed to predict the fate of phenolic and indolic compounds during

ozone treatment in a semi-batch reactor.

1. 7.1 Specific Aims: Phase I

(1) To characterize changes in the characteristics of the manure slurry that occurs
during the fermentation in the slurry storage pit. Changes in the number of
microorganisms, including aerobes, anaerobes, coliforms, E. coli. and coliphage,
and in the physicochemical characteristics of the stored manure slurry were
investigated.

(2) To study the effect that storage of the wastes (prior to treatment) has on the odor
of the waste and how length of storage affects the efficacy of ozonation.

(3) To determine the effect of ozone (applied dosage of l g ozone per liter waste) on
the physicochemical and biological characteristics and the odor of the swine

waste.

(4) To determine if the malodor reoccurs upon storage of the ozone-treated swine

manure.

1.7.2. Specific Aims: Phase II

(1) To design, construct and test the pilot-scale ozonation system. The mass-transfer

experiment was conduct to determine the mass transfer of ozone into the liquid.

l4

(2) To evaluate the effect of ozonation on the physicochemical and biological
characteristics of fresh and stored swine manure waste using the pilot-scale
system.

(3) To determine the minimum dosage of ozone required to achieve an acceptable
level of odor reduction.

(4) To compare the efficiency of using advanced oxidation process, such as combined
treatment of hydrogen peroxide and ozone, and ozonation process alone. The
optimal ratio of applied dosage between hydrogen peroxide and ozone was also
determined by observing the reduction of odor, concentrations of phenolic and
indolic metabolites, and numbers of microorganisms.

(5) To determine the effect of water temperature on the ozonation process.

(6) To conduct an economic evaluation of the treatment process.

I . 7.3 Specific Aims: Phase III
(1) To develop a mass-transfer model that can predict the mass-transfer coefficient and
partition coefficient of ozone in a semi—batch reactor.

(2) To determine the effect of temperature, flowrate, pH, and ionic strength on the

mass-transfer coefficient and partition coefficient.

1. 7.4 Specific Aims: Phase IV
(1) To determine the Stoichiometric factor and reaction rate constants for the reactions

of ozone with five target compounds, including phenol, p-cresol, p-ethylphenol,

indole, and skatole, at different pH values.

15

(2) To understand the effect that competition for ozone has on the reactivity of these
compounds.

(3) To investigate the effect of other substances that are naturally present in the swine
manure, and which have a low reactivity with ozone, on the oxidation of the target
compounds. These substances include ammonia and VFAs.

(4) To formulate a kinetic model for predicting the removal of target compounds in

synthetic and real manure.

1.8 References

1.

CL. Barth; L.F. Elliott; S.W. Melvin, “Using Odor Control Technology to Support
Animal Agriculture”, Transations of the ASAE, 27 :859—864 (1984).

WE. Burnett; N.C. Dondero, “Control of Odours from Animal Wastes”, Transactions
of the ASAE, 13:221-224 (1970).

CH. Burton, “A Review of the Srrategies in the Aerobic Treatment of Pig Slurry:
Purpose, Theory and Method”, Journal of Agricultural Engineering Research,
53:249-272 (1992).

J.J.Carlins; R.G. Clark, “Ozone Generation by Corona Discharge”, In: Handbook of
Ozone Technology and Application, Volume 1, Chapter 2, Ann Arbor Science, Ann
Arbor, MI, 41-75 (1982).

L. Debevec Jr., “A Review of Ozone Generating Facilities in some U.S. Water and
Wastewater Treatment Plants”, Ozone Sci. Eng., 5:103-112 (1990).

A. Dravnieks, “Instruction Manual - Dynamic Dilution Binary Scale Olfactometer”,

IIT Research Institute, Chicago, Illinois.

7.

10.

ll.

12.

13.

l4.

16

LP. Elliott; J .W. Doran; T.A. Travis, “A Review of Analytical Methods for Detecting
and Measuring Malodors from Animal Wastes”, Transactions of the ASAE, 21: 130-
135 (1978).

MR. Evans; E.A. Deans; R. Hissett; M.P.W. Smith; [F Svoboda; F.E. Thacker,
“The Effect of Temperature and Residence Time on Aerobic Treatment of Piggery
Slurry- Degradation of Carbonaceous Compounds”, Agricultural Wastes, 5:25-36
(1983).

B. Fressonnet-Charnbarlhac; Y. Richard; M. Pare, “Ozonation of Urban Wastewater
in Tourist Zone”, Ozone Sci. Eng., 5, 69-77 (1983).

J .W. Gardner; H.V. Shurmer; T.T. Tan, “Application of an Electronic Nose to the
Discrimination of Coeffee”, Sensors and Actuators B, 6:71-75 (1992).

IA. George; C.D. Fulhage; S.W. Melvin, “A Aummary of Midwest Livestock Odour
Court Actions”, In: Agricultural Waste Management and Utilization, Proceedings of
the 5th International Symposium on Agricultural Wastes, ASAE Publication, St.
Joseph, MI, 431-438 (1985).

W.H. Glaze, “Drinking Water Treated with Ozone”, Environ. Sci. Technol, 21:224-
230 (1987).

P.J. Hobbs; T.H. Misselbrook; B.F. Pain, “Assessment of Odoropus from Livestock
Wastes by a Photoionization Dector, an Electronic Nose, Olfactometry and Gas
Chromatography-Mass Spectrometry”, Journal of Agricultural Engineering
Research, 60: 137-144 (1995).

IOA News, International Ozone Association, Vol. 25, No. 4, Ashton, MD (1997).

17

15. B. Langlais; D.A. Reckhow; DR. Brink, Ozonation in Water Treatment: Application
and Engineering, Lewis Publishers, Chelsea, MI (1990).

16. RI. Mackie, “Microbial Production of Odor Components”, In: Proceedings of
International Round Table on Swine Odor Control, 18-19 (1994).

17. Masten, S.J.; Davies, S., “The Use of Ozonation to Degrade Organic Contaminants in
Wastewaters”, Environ. Sci. Technol, 28:180A-185A (1994).

18. DH. O’Neill; V.R. Phillips, “A Review of the Control of Odour Nuisance from
Livestock Building: Part 3, Properties of the Odorous Substances which have been
Identified in Livestock Wastes or in the Air around them”, Journal of Agricultural
Engineering Research, 53:23-50 (1992).

19. KC. Persaud; S.M. Khaffaf; P.J. Hobbs; R.W. Sneath, “Assessment of Conducting
Polymer Odour Sensors for Agricultural Malodour Measurements”, Chemcial Senses,
21:495-505 (1996).

20. WJ. Powers; A.C. Wilkie; H.H. Van Horn, “Effect of Anaerobic Digestion and
Commercial Additives on Odors from Liquid Dairy Manure”, Proceedings of
International Livestock Odor Conference, 44-49 (1995).

21. RC. Rice; M.E. Browning, “Ozone Treatment of Industrial Wastewater”, Noyes Data
Corporation, NJ, Section 6 (1981).

22. W.F. Ritter, “Odour Control of Livestock Wastes: State-of-the-Art in North
America”, Journal of Agricultural Engineering Research, 42:51-62 (1989).

23. R.W. Sneath; C.H. Burton; A.G. Williams, “Continuous Aerobic Treatment of
Piggery Slurry for Odour Control Scaled up to a Farm-size Unit”, Journal of

Agricultural Engineering Research, 53:81-92 (1992).

18

24. AT. Sobel, “Olfactory Measurement of Animal Manure Odor”, Transactions of the
ASAE, 696-699, 703, (1972).

25. J .A.M. Voermans, “Farm Experiments of Anaerobic Digestion to Control Odours
from Slurry”, In: Odour Prevention and Control of Organic Sludge and Livestock
Farming, V.C. Neilsen; J.H. Voorburg; P.L. Hermite (eds), Elsevier Applied Science
Publisher, London, England, 358-365 (1986).

26. F.W. Welsh; D.D. Schulte; E.J. Kroeker; H.M. Lapp, “The Effect of Anaerobic
Digestion upon Swine Manure Odors”, Canadian Agricultural Engineering, 19: 122-
126 (1977).

27. A.G. Williams, “Indicators of Piggery Slurry Odour Offensiveness”, Agricultural
Wastes, 10: 15-36 (1984).

28. A.G. Williams; M. Shaw; SJ. Adams, “The Biological Stability of Aerobically-
treated Piggery Slurry during Storage”, Journal of Agricultural Engineering
Research, 29:231-239 (1984).

29. CM. Williams; S.S. Schiffman, “Livestock Odor Control Additives: Effectiveness
and Laboratory-Scale Protocol for Evaluation”, Proceedings of International
Livestock Odor Conference, 50-53 (1995).

30. LL Wu; S.H. Park; S.M. Hengemuehle; M.T. Yokoyama; H.L. Person; J.B. Gerrish;
S.J. Masten, “The Use of Ozone to Reduce the Concentration of Malodorous
Metabolites in Swine Manure Slurry”, Accepted for Publication by Journal of
Agricultural Engineering Research, 1998.

31. A. Yasuhara, “Relation between Odor and Odorous Components in Solid Swine

Manure”, Chemosphere, 9:587-592 (1980).

 

 

The dark reactions that occur in “pure” water are shown in blue. The additional dark
reactions that may occur when reactive solutes are present are shown in red. D is a
compound that can react directly with ozone; I is a compound that, when it reacts with
ozone via an electron transfer reaction, produces the ozonide radical ion (i.e., it is an
initiator of ozone decomposition); RH is a promoter of ozone decomposition; and S is an
‘OH scavenger. Propagation of the chain reaction may occur by the reaction of RH with
'OH to from an organoperoxide radical, ROO'. Termination of the chain reaction may
occur via reaction of S with ‘OH to form a secondary radical (b, which does not
participate in the chain reaction. Additional reactions that occur in O3/H202 and/or
O3/UV system are shown in yellow.

Figure 1.1 Decompostion of Ozone in Wastewaters

Chapter 2

THE EFFECT OF STORAGE AND OZONATION ON THE PHYSICAL,
CHEMICAL, AND BIOLOGICAL CHARACTERISTICS OF SWINE MANURE

SLURRIES

2.1 Introduction

Environmental pollution by livestock waste is a major nationwide concern of the
animal agriculture industry. An increasing number of civil litigations, more restrictive
township ordinances, demands for the protection of public health and new Federal and
State regulations governing surface run-off and groundwater contamination necessitate
that the livestock industry take action to control the environmental problems associated
with livestock operations.

Odors emanating from livestock operations are the major source of complaints
against livestock farms. As livestock production has moved to larger and more intensive
production units, odor complaints from residents of surrounding communities have
increased (Ritter, 1989). Innovative methods to control, reduce or eliminate odors

associated with livestock operations must be developed and employed to ensure the future

viability of the industry.

20

21

Ozonation has the potential to reduce odors in stored livestock waste by (1)
controlling the rate of production of odorous metabolites by killing the odor-causing
bacteria and (2) chemically oxidizing of odorous metabolites produced during
fermentation. As ozone is a very strong oxidant that attacks the cell wall and the
cytoplasmic membrane of the cell to inactivate the organism, ozone can also be used to
reduce the concentration of odor-causing chemicals and to reduce the numbers of
pathogens present in the waste water, making it safer for reuse. The ozonation of
agricultural wastes has the potential to reduce the likelihood of contamination of
groundwater and surface run-off by pathogenic bacteria and viruses that are ubiquitous in
agricultural wastes.

In the study presented in this chapter no attempt was made to optimize the mass
transfer of ozone into the reactor. As such, a significant fraction of the applied ozone
dosage may be wasted. Manure slurries were stored anaerobically for 12 weeks. Over
these weeks the changes were measured in physical, chemical and biological
characteristics of the swine manure slurry. Each week a portion of the manure slurry was
ozonated and the effect of ozonation was evaluated. Since swine manure slurry is
typically held (prior to land application) for approximately 3-12 months, it was
imperative that we determine if ozonation could be applied anytime during storage and
how long odors remain surpressed during storage at the ozone treated swine manure

slurry.

22

2.2 Materials and Methods
2.2.1 Manure Collection and Storage Test

Fresh swine manure slurry was collected from the Michigan State University Swine
Research and Teaching Farm, East Lansing, MI. Four 100-kg finishing pigs, which were
fed a standard com-soybean finishing ration, were placed in a crate with a slotted floor,
which allowed for the collection of fresh manure. The manure was collected over a five-
day period by flushing a small amount of water to wash the manure, which had adhered to
the slotted floor into a collection tray. Upon collection, the manure was stored at 4 0C to
minimize any changes that might occur in its physicochemical and biological properties.
Approximately 150 liters of manure slurry was collected.

A tank was used to simulate the processes that occur during storage of the slurry in
the storage pits. The tank was constructed of PVC and had a capacity of forty gallons
(Figures 2.1 and 2.2). Four sampling ports were located along the side of the tank at
intervals of 21 cm, dividing the 90 cm-deep tank into 4 zones (with zone 1 being the
uppermost region and zone 4 the bottom). This allowed for the investigation of the effect
of the length of storage and tank depth on the physical, chemical and microbiological
properties of the waste. During storage in this tank the manure was maintained at 20-25
0C. The manure was stored for a period of 3 months and sampled once per week for
BOD, COD, phosphates, sulfides, sulfates, odor, solids, phenolic and indolic metabolites,
volatile fatty acid concentrations, total aerobic and anaerobic bacterial numbers, total
coliforms, and total coliphage numbers. To capture the ammonia and hydrogen sulfide as
it volatilized from the slurry, cylinder air was continually passed across the headspace of

the tank at a flow rate of 10-15 mIJmin. The effluent air was then passed into solutions

23

of boric acid and zinc acetate to trap the ammonia and hydrogen sulfide from the air

stream.

2.2.2 Ozone and Ozonation Test

Ozone was generated from dried pure oxygen by corona discharge using an ozone
generator (Model T-408, Polymetrics Inc., San Jose, CA), which produced an ozone
concentration of 1.3% (by volume) in the oxygen gas stream. The oxygen was dried using
a molecular sieve (SIP Brand #G5301-21). A 500 mL gas-washing bottle was used as the
ozonation reactor (Figure 2.3). Ozone was bubbled through a porous diffuser into 300
mL of the manure slurry. The manure slurry was treated for approximately 25 minutes at
a flowrate of 150 mL/min until an ozone dosage of 1 g ozone/L of manure was achieved.
The gaseous ozone concentration in the inlet stream was determined
spectrophotometrically. The absorbance of ozone was measured in a 2-mm flow-through
quartz cuvette at the wavelength 258 nm. An extinction coefficient of 3000 M'lcm'l was
used to convert absorbances to concentration units (Nowell and Hoigné, 1988). To reduce
foaming caused by the bubbling of ozone into the slurry, 1 mL of anti-foaming agent,
Anti-foam 289 (Sigma Co., St. Louis, MO) was added to the slurry prior to ozonation.
Between ozonation experiments the porous diffuser was cleaned by purging nitrogen
through the frit, which was submerged in a mixture of water and hydrogen peroxide

(85: 15).

24

2.2.3 Odor Evaluation — Olfactometry

A dynamic dilution binary scale olfactometer (ASTM E544-75, IIT Research Center,
Chicago, Illinois), which employs l-butanol as the standard odorant, was utilized to
quantify the odor intensity from untreated or ozonated manure samples (see Figure 2.4).
Six participants were asked to smell the samples, presented in a random order, and match
the intensity of the odor of the samples with that coming from sensory ports that

discharged l-butanol vapor at various concentrations.

2.2.4 Chemical Analysis

Chemical oxygen demand (COD) determinations were made using prepackaged
Hach vials (0-15,000 mg/L). Total solids and biological oxygen demand (BOD)
determinations were made according to Standard Methods for the Examination of Water
and Wastewater (1992). Phosphate and sulfate were determined using ion
chromatography. Samples were pretreated by centrifugation at 6,0000 G for 20 minutes
and filtration through a 0.22 pm glass filter to remove the solid material, thereby reducing
the possibility of clogging the ion chromatographic columns. Ammonia was measured
with an ammonium electrode connected to a pH meter. Quantification was accomplished
using the method of known addition. Sulfides were measured using a purge and trap
method (Barth and Polkowski, 1974), employing nitrogen gas to purge hydrogen sulfide
out of the solution. All sulfide species were first converted to hydrogen sulfide by the
addition of 10 mL concentrated sulfuric acid to lower the pH of the samples to 2. The
trap solution was prepared by adding 15 mL of zinc acetate (2N) to 135 mL deionized

water. After trapping, the sulfides were determined according to Standard Methods for

25

the Examination of Water and Wastewater (1992). The concentrations of volatile fatty
acids (namely acetate, propionate, butyrate, iso-butyrate, valerate, iso-valerate) were
determined by gas-liquid chromatography (Model 5840A, Hewlett-Packard Inc.,
Avondale, PA). One mL of concentrated hydrochloric acid was added to 10 ml of slurry
sample. After the mixture was agitated in a vortex mixer, it was centrifuged at 11,950 G
for 10 minutes to remove all particulate matter. The supernatant was injected directly
into a glass column packed with Supelco GP 10% SP-1000/ 1% H3PO4 on 100/ 120
Chromosorb W AW. Nitrogen was used as a carrier gas at a flow rate of 50 mIJmin.
The initial injection temperature was held 110 °C for 3 minutes; the temperature was
ramped at 5 °C/min to 170 °C for 30 minutes.

The bacterial degradation of tyrosine and tryptophan in the intestinal tract and during
storage of swine manure results in the production of several odorous volatile phenolic and
indolic metabolites. These include phenol, p-cresol, p-ethylphenol, indole, and Skatole
(see Table 2.1). The concentrations of these metabolites were quantitatively determined
in the swine waste slurry by gas-liquid chromatography (Yokoyama et. a1, 1982).

Samples (10 mL) were adjusted to pH 4.0 with concentrated sulfuric acid and extracted
twice with 30 mL of methylene chloride. The extract was evaporated down to a volume
of approximately 1 mL using a rotary evaporator, transferred to a screw cap vial, and
made up to 4.0 mL with methylene chloride. Two uL were analyzed directly by GC-FID,
using a stainless steel column (2 mx2 mm) packed with 21% Carbowax 4000 and WAW-

DMCS (60/80 mesh). The oven temperature was programmed as follows: held at 100 oC

26

Table 2.1 The Structure of the Phenolic and Indolic Compounds

1. Phenol

2. p-Cresol

3. p-Ethylphenol

4. Indole

5. Skatole

27

for the first 1 min after injection, ramped at 10 °C/min to 125 oC and held for 8 min, then
ramped at 2 °C/min to 155 0C. The total run for each sample was 26.5 min. The

Injection port and detector temperatures were maintained at 200 and 250 °C, respectively.

2.2.5 Microbiological Analysis

Total aerobic bacterial numbers were determined by the pour plate method using
Sterile Nutrient Agar. Manure samples were serially diluted in sterile 0.1% peptone
water and 0.1 ml of each dilution was inoculated on agar pour plates. Plates were
incubated at 39 0C for 48 hrs, then enumerated for emergent colonies. Total anaerobic
bacterial numbers were determined by the anaerobic roll tube technique of Hungate
(1950). A prereduced medium of GCSX (glucose, cellobiose, starch, and xylose) plus
clarified ruminal fluid (20%) was prepared with 1.8% agar and autoclaved. Nine mL of
the media were added to the tubes and the sterile media was stored at 45-50 0C until it
was used. Samples were serially diluted in anaerobic dilution media (mineral solution I,
11.2 mL; mineral solution II 11.2 mL; resazurin, 0.3 mL; distilled water, 256.3 mL).
Dilutions were inoculated into the tubed GCSX medium, immediately rolled under cold
water to solidify the agar, and allowed to set for 10 min. After the rolled agar tubes were
incubated at 39 0C for 48 hrs, the emergent colonies were enumerated.

Total coliforms and Escherichia coli were determined by the Most Probable
Numbers Method (Hungate and Fletcher, 1957). Slurry samples were serially diluted in
EC broth medium, containing fermentation gas tubes, and incubated at either room

temperature or in a 45 0C water bath. Growth in the highest dilution of the EC broth

28

medium (4 replicates) at room temperature was used to estimate total coliforrn numbers.
Growth at 45 0C with gas production was used as an estimate of E. coli numbers.
Coliphage determinations were made by the IPTG/X-gal method (Ijzerman and
Hagedom, 1992), which involved the use of E. coli C inoculated from ATCC strain
13706 into 10 mL sterile LB broth containing glycerol (10%) and incubated at 37 0C until
an CD. (at 520 nm) of 0.35 was obtained. One and one-half mL of E. coli culture was
added to 18 mL of modified LB agar at 50 0C (modified by adding ammonium nitrate,
NH4NO3, 1.6g/L; strontium nitrate, Sr(NO3)2, 1.21g/L; and agar, 15g/L). The manure
sample (8 mL) was centrifuged at 2400 rpm for 80 min, and the supernatant obtained was
filtered twice by using 0.45 pm and 0.22 pm syringe filters. One mL of filtered sample
was then serially diluted into 9 mL SM buffer. Three mL of the filtered and diluted
sample containing bacteriophage were added along with 66 uL isopropyl-B-D-
thiogalactoside (IPTG) (200 mg/mL) and 360 pL 5-bromo-4-chloro-3-indolyl-B-D-
galactopyranoside (X-gal) (10 mg/mL) dissolved in dimethylforrnamide. The solution
was swirled to mix the components and evenly distributed onto two 100x15mm Petri
dishes. After the agar had hardened, the plates were inverted and incubated at 37 0C.
Pale blue plaques appeared after 4 hrs and the color intensified as the plaques developed
(after 7 to 8 hrs). Each plaque produced was assumed to have initiated from a single

baceriophage. The plaque numbers were enumerated after 16 hrs of incubation.

29

2.3 Results and Discussion

The storage of the swine manure slurry was followed over a three-month period
from June 12 through September 4, 1995. The experimental studies were divided into
two parts: those pertaining to storage and those on the ozonation of the manure slurry.
2.3.1 Storage Experiment
Gas phase analysis: The release of ammonia and hydrogen sulfide from the surface of the
stored manure is summarized in Figure 2.5. The concentration of ammonia reached its
maximum concentration in the 4th week then dropped off to a lower and more constant
level for the remaining test period. The concentration of hydrogen sulfide waserratic
during the first seven weeks, but then increased steadily during the last four weeks of the
experiment. The final discharge achieved was 63 mg. The total mass of ammonia
released during 12-week storage (ca. 30 mg) was much less the total mass of hydrogen

sulfide emitted (ca. 360 mg).

Liquid phase analysis: The pH of the fresh manure was initially 8.8. After 2 weeks of
storage the pH of the slurry decreased to about 7.2 and then remained relatively constant
for the remaining 10 weeks (Figure 2.6). There was little variation in the pH of the
manure slurry in each of the four zones during the entire storage period. The production
and accumulation of acidic intermediates as a result of the decomposition of organic
compounds is probably the major reason for the initial decrease in pH. This is consistent
with our observation that the concentrations of volatile fatty acids increase during storage.

Sufficient buffering capacity may have prevented any further decrease in pH. The

30

buffering capacity may, in part, originate from alkaline metabolites produced from the
metabolism of proteins.

The concentration of ammonia nitrogen in the swine manure steadily increased in all
of the four zones during the first seven weeks of storage (Figure 2.7). The initial
concentration of ammonia nitrogen ranged from 3300-3400 mg/L (week 0) to 3600-3800
mg/L (week 7). The ammonia concentration was consistently higher in zone 4 than in the
other zones. This increase is presumed to be the result of the production of ammonia
nitrogen from the microbial degradation of nitrogenous organic compounds. After week
7, the ammonia concentrations in all 4 zones remained essentially constant at
approximately 3600-3700 mg/L.

The concentration ranges of BOD5, COD and total solids of the stored manure slurry
were respectively 8,000-26,000 mg/L, 18,000-58,000 mg/L, and 7,000-35,000 mg/L
(Figures 2.8-2. 10). The solids contribute most of the organic material to the slurries.
Therefore, the concentrations of organic material (BOD5 and COD) were normalized
using the solid concentrations. The ratio of COD to total solids was essentially
unchanged over 12 weeks of storage and within the range of 2 :t: 0.5. Evans et al. (1979)
reported a COD/T S ratio for untreated slurry of 1.3. The ratio of BOD5 to COD observed
in the upper three zones increased during the first four weeks of storage from 0.45 to 0.6
(Figure 2.11), substantiating the conclusion that during this initial period there is a
conversion of organic material to a more biodegradable form. This observation is also
substantiated by the work of Williams and Evans (1981), which showed that an increase

on BOD5 was due to the production of more biodegradable soluble intermediates from

31

the degradation of higher molecular weight components such as fiber and protein. The
values of BOD5 to COD observed in zone 4 appeared to be almost random.

Manure slurry is commonly recycled as fertilizer because of its high nutrient content.
These nutrients include phosphate and nitrogen. The concentrations of nitrate could not
be accurately measured due to the presence of interfering substances in the manure slurry.
The effect of storage on the phosphate concentration was quantified. The concentration
of soluble phosphate in the upper three zones remained essentially constant over the
course of the experiment, suggesting that phosphate was not being released during the
storage process (Figure 2.12). The concentration of soluble phosphate in the bottom
(zone 4) increased from 520 mg/L to 1,100 mg/L during the first six weeks, then
subsequently decreased to 590 mg/L. The initial increase is thought to be due to the
release of soluble phosphate from the solid material. The latter decrease may be due to
further settling of the solid material below the sampling port and insufficient mixing of
any phosphate released from those solids. Alternatively, it could be due to a decrease in
metabolic activity or a change in the rate at which phosphate was released from the solids
(possibly due to the early release of the more easily metabolized fraction, with what
remained being much more difficult to metabolize).

The concentration of sulfate was also monitored in the storage tank. It was found
that the concentration of sulfate remained relatively constant at approximately 500 mg/L
for the first 8 weeks, then decreased to < 100 mg/L by the 12th week (Figure 2.13).
Concurrently, in the last 4 weeks, there appeared to be an increase in the ratio of sulfide

concentration to the total solid concentration (Figure 2.14). This is thought to be due to

32

the utilization and transformation of sulfide by the sulfur-reducing bacteria that likely
began to proliferate after the 8th week.

Volatile fatty acid (VFA) concentrations were determined from a composite sample
taken from the four zones. Acetate was found to be the predominant VFA, followed in
decreasing order by propionate, butyrate, iso-butyrate, iso—valerate, and valerate (Table
2.2). This is consistent with the observation of Cooper and Comforth (1978). Total VFA
concentrations peaked around the 10‘'1 week (Figure 2.15). Based upon the work of
Spoelstra (1979), VFA accumulation is likely to be due to the production of VFAs during
the anaerobic degradation of fiber (mainly cellulose and hemicellulose) and protein. The
subsequent decrease is thought to be due to the consumption of VFAs by methanogenic
bacteria. This result was verified by the observation of lots of gaseous bubbles
accumulated at the upper zone of the storage tank.

Phenols (phenol, p-cresol, and p-ethylphenol) and indoles (indole and skatole) are
regarded as important malodorous components in swine waste slurry (Schaefer, 1977;
Spoelstra, 1977; Williams and Evans, 1981; Williams, 1983). The concentrations of
phenolic and indolic metabolites were analyzed from composite samples from the 4 zones
of the fermentation reactor. In the raw manure, p-cresol predominated, followed in
decreasing order by phenol, p-ethylphenol, and skatole (Figure 2.16). The concentration
of p-cresol peaked around the 6th week of storage. The concentration of phenol did not
change during storage. The concentration of p-ethylphenol increased rapidly during the
first week storage then remained constant. The concentration of indole was below
detection limits. The concentration of skatole rose until the fourth week, then gradually

reduced to non-detectable levels after 10 weeks storage.

33

Table 2.2 The Concentrations of Each Volatile Fatty Acid during 12 Weeks of Storage

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pefiod

Time Acetate Propionate Butyrate i-Butyrate Valerate i-Valerate

(weekS) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)
0 10060 2170 1070 240 290 300
1 10410 2310 1320 260 240 290
2 10380 2240 1320 270 240 300
3 10500 2250 1350 280 240 300
4 10600 2240 1360 310 260 310
5 10270 2300 1340 320 250 320
6 10280 2290 1310 330 240 320
7 10310 2320 1280 340 240 330
8 10730 2310 1120 340 240 330
9 12000 2510 390 360 240 350
10 11990 2520 300 370 250 360
11 10600 2480 260 380 250 370
12 10820 1350 230 360 230 360

 

 

 

 

 

 

 

 

 

As shown in Figure 2.17, the odor intensity of the swine manure tended to increase
with storage time (although the variations in intensity for any specific time were large).
The intensity of the odor was also found to increase with tank depth (over the entire
course of the experiment). As such, it is apparent that anaerobic storage results in an
increase in odor intensity, most likely due to the formation of odorous chemical
compounds resulting from microbiological reactions.

Numberous attempts have been made to relate the odor intensity to the specific
chemical characteristics of manure slurries. Barth et al. (1974), Williams (1984), Carney
and Dod (1989), and Misselbrook et al. (1993) related the concentrations of specific
malodorous compounds (along with other chemical compponents) to the odor intensity by

using category scaling and ratio scaling:

34

where k1, k2, a, and b are empirically determinant constants. Although the concentration
of ammonia nitrogen in the stored swine manure slurry increased with the storage time,
the odor intensity determined by Olfactometry test was found to have a low correlation
with ammonia nitrogen for stored swine manure slurry (Figure 2.18). The regression
coefficients, R, for category and ratio scalings were found to be 0.34 and 0.41 ,
respectively. However, the ratio of sulfide to total solids concentration seemed to be a
better indicator of odor intensity than ammonia nitrogen (Figure 2.19), with regression
coefficients for category and ratio scaling of 0.64 and 0.61, respectively. The
mathematical expressions for these relationships can be written as:

Odor intensity (ppm of l-butanol)=262.82 Ln(Sulfides/T S) +2909

Odor intensity (ppm of 1-butanol)=7129.5(Sulfidesl'I‘S)°‘28

2.3.2 Ozonation Experiment
For the ozonation experiments, equal volumes of manure slurry were taken from

each of the four zones of the fermentation reactor and mixed. The composite sample was
then ozonated at a dosage of 1 g ozone per liter slurry. Following ozonation, the original
(control) and ozonation samples were analyzed to evaluate the efficacy of ozone for the
reduction of odor along with evaluating the changes in the physical, chemical, and
microbiological properties of the slurry.

Ozonation resulted in an increase in the pH of the swine manure slurry (Figure 2.20).

Similar results were obtained by Watkins et a1. (1997). After bubbling oxygen into the

35

swine manure (pH=6.5), Watkins et a1. (1997) found that the pH increased to 7.1. In our
study, sparging with helium resulted in an increase in the pH of the manure slurry from
7.6 to 8.7. Ozonation resulted in an increase from 7.6 to 8.1. The increase in pH upon
sparging with oxygen or helium is thought to be due to the volatilization of VFAs.
However, during ozonation, both the VFA volatilization and VFA generation by
oxidization of its precursors are likely to occur. Apparently, the formation of VFAs is
significant as the concentration of total VFAs was observed to increase upon ozonation
(Figure 2.21) and the pH that is observed after ozonation is lower than that observed
when sparging with helium was used (8.1 vs. 8.7). Additionally, when the slurry that had
previously been sparged with helium (pH=8.7) was ozonated the pH was 8.4 (rather than
8.1 without sparging as a pretreatment method).

The concentration of sulfide decreased as a result of ozonation (Figure 2.22),
whereas the concentration of sulfate increased (Figure 2.23). This phenomenon indicates
that the sulfides are easily oxidized by ozone and transformed into sulfate. A mass
balance approach was used to try to relate the conversion of sulfide to sulfate. However,
the increase in mass of sulfate exceeded, in most cases, the mass of sulfides removed
from the system by ozonation. Thus, there appears to be an additional source of sulfate
beyond the oxidation of sulfides by ozone. The additional sulfate formed may be due to
the oxidation of sulfur-containing metabolites.

A comparison of the odor intensity of the untreated and ozonated samples
demonstrated the ability of ozone to significantly reduce the odor of the swine manure
slurry (Figure 2.24). The odor intensities of the slurries were all reduced to

approximately the same level, irrespective of the initial odor intensity of the untreated

36

samples. There was also less variation in the measurements of odor intensity of the
ozonated samples (as compared to the raw samples), suggesting that the perception of
reduced intensity was much more consistent among the six panelists. Based upon these
results, the efficacy of ozonation is independent of the initial odor intensity of the stored
manure. The reoccurrence of odor was also assessed by storing the ozonated samples,
which were open to the atmosphere, in a hood at room temperature. Although the odor
intensity of these samples was found to increase gradually with time (Figure 2.25), the
odor of these stored ozonated samples was much less offensive than that of the original
raw manure slurry.

Ozonation of the swine manure slurry did not significantly modify the BOD5, COD,
solids, ammonia, and phosphate concentrations of the slurry (Figures 2.26-2.30). These
results indicate that ozonation preserves both the organic concentration and the nutrient
content of the manure slurry, the nutrient content is preserved. Ozonation did not
significantly change the concentrations of the individual VFAs (Table 2.3).

The phenolic and indolic metabolites in the stored slurry samples were completely
degraded by ozone a dosage of 1 g/L. It is thought that the change in the odor upon
ozonation is likely to be due to the rapid oxidation of these simple aromatic compounds.

The effect of ozonation on the microbiological characteristics (e.g., aerobic and
anaerobic bacterial numbers, total coliforms, total E. coli and total coliphage) of the
slurry were evaluated. The numbers of total aerobes in the raw slurry (zero week)
decreased by almost 50% after ozonation, while the numbers of total anaerobes decreased
by about 25% (Figure 2.31). In the last 5 weeks of storage, ozone appeared to be less

effective at reducing the numbers of total aerobes than in the early weeks of storage. The

Table 2.‘.

Storage
H

 

/7/r//f7/T//F

37

Table 2.3 The Concentrations of Each Volatile Fatty Acid for Stored and Ozonated Swine

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Manure Slurry
Storage Weeks (mg/L) C2 C3 C4 i-C4 C5 i—C5
Raw 10,060 2,170 1 ,070 240 290 300
O Ozonated 10,500 2,260 1,190 230 230 280
1 Raw 10,410 2,310 1,320 260 240 290
Ozonated 1 1,290 2,370 1,280 250 210 270
2 Raw 10,380 2,240 1 ,320 270 240 300
Ozonated 10,920 2,230 1,260 270 210 260
3 Raw 10,500 2,250 1,350 280 240 300
Ozonated 1 1,510 2,350 1,310 280 220 280
4 Raw 10,600 2,240 1,360 310 260 310
Ozonated 10,740 2,240 1 ,270 280 220 280
Raw 10,270 2,300 1,340 320 250 320
5 Ozonated 10,890 2,290 1,240 300 210 290
Raw 10,280 2,290 1,310 330 240 320
6 Ozonated 1 1,360 2,410 1,290 330 210 310
7 Raw 10,310 2,320 1,280 340 240 330
Ozonated 10,920 2,320 1,210 320 210 300
8 Raw 10,730 2,310 1,120 340 240 330
Ozonated 1 1,150 2,350 1,070 340 210 310
9 Raw 12,000 2,510 390 360 240 350
Ozonated 12,690 2,400 390 350 210 320
10 Raw 1 1 ,990 2,520 300 370 250 360
Ozonated 12,650 2,520 280 350 210 320
1 1 Raw 10,600 2,480 260 380 250 370
Ozonated 9,850 2,540 260 380 210 350
12 Raw 10,820 1,350 230 360 230 360
Ozonated 9,810 1,610 240 380 210 350

 

 

 

 

 

 

 

 

 

38

total aerobes and aerobes in the raw, untreated swine manure slurry gradually decreased
with time of storage, indicating unfavorable conditions for the growth of total aerobes and
anaerobes in the storage system. Nevertheless, it is thought that some species of
anaerobes such as sulfur-reducing bacteria proliferate during storage. After 7 weeks of
storage, ozone became ineffective at reducing the number of total anaerobes. One
explanation might be that unfavorable conditions in the reactor resulted in the sporulation
of vegetative cells. These spores would be expected to be more resistant to ozone than
the vegetative cells. The number of coliforms decreased with ozonation of swine manure
slurry by a factor of 10—1000 (Figure 2.32). The number of E. coli decreased in the fresh
manure substantially upon ozonation (from 50,000 to 0). However, the number of E. coli
in untreated swine manure slurry declined to zero after 3 weeks of storage, so that by
week 3 there was no difference in the numbers of E. coli in the untreated and ozonated
slurries (Figure 2.33). Coliphage plaque numbers, which indicate of the presence of

viruses, decreased significantly after ozonation (Figure 2.34).

2.4 Conclusions
In general, the production of malodors from swine waste manure is the combined
effect of odorous compounds that are present in the fresh manure and those that

accumulate due to the degradation of organic matter by anaerobic microorganisms during

storage.

1. The storage of manure slurry resulted in (a) an increase in the concentration of

sulfides per unit of total solids due to sulfate reduction by sulfur-reducing bacteria and

39

(b) an increase in the concentrations of ammonia nitrogen due to the mineralization of
organic nitrogen.

The storage of manure slurry resulted in the accumulation of total volatile fatty acids
during the first ten weeks of storage. The subsequent reduction in the concentration
of VFAs during the final two weeks of storage was ascribed to the decomposition of
acetate into methane and carbon dioxide by methanogenic bacteria.

Microbial populations of aerobes, anaerobes, and E. coli were found to decrease
during storage, suggesting an unfavorable environment for these bacteria compared
with that of the host. E. coli was undectable after 4 weeks of storage.

Ozonation (at a dosage of 1 g/L) is effective in reducing the concentrations of
sulfides, phenol, p-presol, p-ethylphenol, and skatole in swine manure.

The soluble phosphate concentration did not change through ozonation, indicating the
preservation of nutrient content.

. The ozonation at a dosage of 1 g/L of fresh or stored swine manure reduced the odors
to acceptable levels.

As the concentrations of total VFAs increased and ammonia remained unchanged
upon ozonation, while the malodor was significantly reduced, it appears that neither
VFAs nor ammonia contribute extensively to the malodor of the manure.

Ozonation is able to kill some of the indicator microorganisms such as coliforms, E.
coli and coliphage. However, it does not provide a significant benefit in reducing the

organic loading (as measured by BOD5, COD, and VFAs).

40

2.5 References

l. ASTM E-544-75, (1975(r6approved 1981)), Standard Practices for Referencing
Suprathreshold Odor Intensity, Am. Soc. Testing. Materials, Vol. 15.07:32-44 (1984).

2. CL. Barth; L.B. Polkowski, “Identifying odorous components of stored dairy manure”,
Transactions of the ASAE, 17:737—741 (1974).

3. PG. Carney; V.A. Dodd, “The measurement of agricultural malodours ”, J. Agric.
Engng. Res., 43: 197-209 (1989).

4. P. Cooper; I.S. Comforth, “Volatile fatty acids in stored animal slurry”, J. Sci. F d
Agric., 29:19-27 (1978).

5. MR. Evans; E.A. Deans; M.P.W. Smith; D.F. Ellarn; S. Baines; J.L. Woods, “Effect of
micro-organism residence time on aerobic treatment of piggery waste”, Agricultural
Wastes, 1:67-85 (1979).

6. DC. Hardick, Agricultural Problems Related to Odour Prevention and Control, In:
Odour Prevention and Control of Organic Sludge and Livestock farming Waste,
Nielsen, V.C., Voorburg, J .H.& L'Hermite, P., editors, Elsevier Applied Science
Publishers, London, 21-26 (1986).

7. RE. Hungate, “The anaerobic mesophilic cellulolytic bacteria”, Bacteriol. Rev., 14:1-
49 (1950).

8. RE. Hungate AND D.W. Fletcher, Laboratory Manual of General Bacteriology, 2nd
Ed., Scholar's Library, New York, 44 (1957).

9. M.M. Ijzerrnan; C. Hagedom, “Improved method for coliphage detection based on B-

galactosidase induction”, Elsevier Science Publishers B.V., Amsterdam, Netherland

(1992).

41

10. AM. Jirka; M.J. Carter, “Chemical oxygen demand, reactor digestion method’, Anal.
Chem, 47(8): 1397 (1975).

11. TH. Misselbrook; C.R. Clarkson; B.F. Pain, “Relationship between concentration
and intensity of odours for pig slurry and broiler houses”, J. Agric. Engng. Res,
55:163-169 (1993).

12. L.H. Nowell; J. Hoigné, “Interaction of iron (II) and other transition metals with
aqueous ozone”, In Proc. Of the 8th Ozone World Congress, 2:E80-E95 (1988).

13. W.F. Ritter, “Odour control of livestock wastes: state-of -the-art in North America”,
J. Agric. Engng. Res., 42:51-62 (1989).

14. J. Schaefer, “Sampling, characterization and analysis of malodours”, Agriculture and
Environment, 3: 121-128 (1977).

15. Standard Methods for the Examination of Water and Wastewater, American Public
Health Association, Inc., 18th ed., Washington, DC. (1992).

16. SF. Spoelstra, “Volatile fatty acids in anaerobically stored piggery wastes”,
Netherlands Journal of Agricultural Science, 27:60-66 (1979).

17. 8D. Watkins; S.M. Hengemuehle; H.L. Person; M.T. Yokoyama; S.J. Masten,
“Ozonation of swine manure wastes to control odors and reduce the concentrations of
pathogens and toxic fermentation metabolites”, Ozone Science & Engineering,
19:425-437 (1997).

18. A.G. Williams, “Indicators of piggery slurry odour offensiveness”, Agricultural
Wastes, 10: 15-36 (1984).

19. A.G. Williams; M.R. Evans, “Storage of piggery slurry”, Agricultural Wastes, 3: 311-

321(1981).

42

20. M.T. Yokoyama; C. Tabori; E.R. Miller; M.G. Hogberg, “The effects of antibiotics in
the weaning pig diet on growth and the excretion of volatile phenolic and aromatic

bacterial metabolites”, Amer. J. Clin. Nutr., 35:1417-1424 (1982).

43

 

 

     
  

\
I
Ventilation

Trap Solutions

/

Storage Tank

\

       
   

Flow Meter Sampling Ports

  

Air Cylinder

  

 

Figure 2.1 System Setup for the Storage Experiment

 

Figure 2.2 The photo of the Fermentation Tank for Storage Experiment

 

 

l:|

r— DDI—.1

 

 

 

Cooling System

 

 

 

 

 

 

45

 

 

 

 

 

 

 

Flow Controller

UV Spectrophotometer

 

 

 

 

 

 

 

 

 

[l O
O
i— f
e)
O C)
Ozone Generator
Oxygen Cyli‘nder A
V F
Trap Solution (KI)

 

Reactor (Gas-Washing Bottle)

Figure 2.3 System Setup for the Ozonation Experiment

 

46

 

Figure 2.4 The Photo of the Dynamic Dilution Binary Scale Olfactometer

47

 

   
 

50 ‘ -o- Ammonia
-D— Hydrogen Sulfides

 

 

Mass released during each week (mg)

 

 

0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.5 Effect of Time on the Release of Ammonia and Hydrogen Sulfide
from Stored Swine Manure

 

 

 

 

10.00 '
9.00
8.00 -
7.00 . W
6.00 ~
“:1. 5.00 -
4.00 J
3.00 - +Zonel +Zone 2
2.00 - +Zone 3 -x-Zone 4
1.00 -
0.00 . . . . . . . . T . .
0 1 2 3 4 5 6 7 8 9 10 11 12

Storage Time (weeks)

Figure 2.6 Effect of Time on the pH in Stored Swine Manure Slurry

48

4000.00
" 3900.00 -
3 3800.00 -
g, 3700.00 -

— -A\
_- ‘\
E3600.00- / " "1

§ - '9

g 3500.00 ;_ i ‘
3400.00 f, a /
.30.... /

 

 

 

 

3mom ' l I T 1 T I T I U I F
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.7 Effect of Time on the Concentrations of Ammonia Nitrogen in

 

 

 

Stored Swine Manure Slurry
30000
T +Zone 1
25000 -
'l' +Zone 2
A 20000 - +Zone 3
§ 1 T -x-Zone 4
1 15W) "‘ A «L
Q e
at: 10000 /, , ‘ ‘ ‘ ’ e
5000
0 ' T I I I I r

 

 

0 1 2 3 4 5 6 7 8 9 10 11 12
StorageTime(weeks)

Figure 2.8 Effect of Time on the Concentrations of BODS in Stored Swine
Manure Slurry

49

70000.00

 

+Zone 1

 

 

 

 

0 1 2 3 4 5 6 7 8 9 10 11 12
StorageTime(weeks)

Figure 2.9 Effect of Time on the Concentrations of COD in Stored Swine

 

  
  
     
  

 

 

 

Manure Slurry
40000.00
35000.00 - T +20!“
+ 2
A 30000.00 - 2°":
E +Zone3
v 25000.00 - _x_ Zone 4
g 20000.00 -
3 15000.00 -”
O
i" 10000.00 /"\x
5000.00
00m I I f I I I I I I I I I fl
0123456789101112

Storage Time (weeks)

Figure 2.10 Effect of Time on the Concentrations of Total Solids in Stored
Swine Manure Slun'y

50

 

 

 

 

 

 

0.8
0.6 « l5 .1
.. ., ,,4\{),,‘,-..4p\#
8 x , I. F" I, . 1:
3 11’ ‘ ' .1.
0.4 :
+Zone 1 +Zone 2
+Zone 3 -X— Zone 4
002 I I l T T I I T I I T
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.11 Effect of Time on the Ratios of BODS to COD in Stored Swine
Manure Slurry

1200.00

 

1000.00 -

   

(mg/L)

 

 

a

§ 800.00 -

§ 600.00 -

3

%

a, 400.00

“e

8 200.00 ~ +Zone1 +Zon62

‘3 +Zone 3 +Zone 4
0.00

 

 

 

0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.12 Effect of Time on the Concentrations of Soluble Phosphate in
Stored Swine Manure Slurry

51

800.00

 

A 700.00
V600.00 _
‘3 500 00
E .
U]

3 400.00 -
,3 300.00 -
6-

O
3 200.00 ~

 

(3 100.00 -

 

 

 

00m I I I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.13 Effect of Time on the Concentrations of Soluble Sulfate in

 

   
  
 
 
 
 

Stored Swine Manure Slurry

4.00E-03
8 3'50E'03 l +Zone 1 +Zone 2
2100303 “ +Zone 3 +Zone 4
.3 2.50E-03 -
“52.00303 -
m 1 50B 03
g .

 

; 10013-03 ‘

 

O OOE'I'm T r r 'I' 1 r I T T fi fi
0 1 2 3 4 5 6 7 8 9 10 l 1 12
Storage Time (weeks)

Figure 2.14 Effect of Time on the Ratios of Total Sulfides to Total Solids in
Stored Swine Manure Slurry

52

 

 

 

 

 

Mom I I j I T I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.15 Effect of Time on the Concentrations of Total VFAs in Stored
Swine Manure Slurry

 

 

 

Cone. of Volatile Metabolite (mg/L)

 

 

0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)
+Phenol +p-Cresol +p—Ethylphenol +Skatole

Figure 2.16 Effect of Time on the Concentrations of Phenolic and Indolic
Metabolites in Stored Swine Manure Slurry

53

 

 

 

 

1800.00
1600.00 -
E 1400.00 - .
e.
2 1200.00 - v‘ 1"
1000.00 -
i . N‘ gaui'
- 800.00 - \
fl up
“a y/ \\§/
. 600.00 / VC/
8 /
8 400.00
200.00 +Zone 1 +Zone 2
00m I I 7 I j I I @m 3I I I zoncT4
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.17 Effect of Time on the Odor Intensity in Stored Swine Manure
Slurry

54

 

°8°

1600 - I
y=4770x-38000 ' '

2 I
R =0.34 ' I

§§

§§§§

l

l

 

Odor Intensity (ppm)

 

 

 

I

8.09 8.14 8.19 8.24 8.29
Ln (Concentration of Ammonia)

 

 

Ln (Odor Intensity)

 

 

 

5 . 1 .
8.09 8.14 8.19 8.24 8.29

Ln (Concentration of Ammonia)

Figure 2.18 Relationships between Odor Intensity and Concentrations of
Ammonia Nitrogen in Stored Swine Manure Slurry

55

 

1800

g

y = 260x + 2900 'l'
R2 = 0.64

§

   
  

1200 1

i

 

Odor Intensity (ppm)
I

§§§

 

 

-9.2 -8.2 -7.2 -6.2 -5.2
Ln (Salads/Total Solids)

 

7.5 n y = 028x + 8.9

 

Ln (Odor Intemlty)

 

 

 

5 I I I I

-9.2 -8.2 -7.2 -6.2 -5.2
Ln (Sulfldwfl‘otal Solids)

Figure 2.19 Relationships between Odor Intensity and Concentrations of
Sulfides/Total Solids in Stored Swine Manure Slurry

56

 

pH
LII

3 J + Untreated -D— Ozonated

 

 

O l 1 I I T I I I I I U
0 1 2 3 4 5 6 7 8 9 10 11 12

Storage Time (weeks)

 

Figure 2.20 Effect of Time and Ozonation on the pHs in Stored Swine
Manure Slurry

 

 

 

 

 

 

’E‘
5
8
E
‘5
2 44000 ~ +Untreated -D—Ozonated
e
0 420.00 .
4m.m I 1' I I 1 I T T I I l l
0 l 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.2] Effect of Time and Ozonation on the VFA Concentrations in
Stored Swine Manure Slurry

Cone. of Total Sulfide: (mg/L)

57

 

+ Untreated

 

4
2 I I
0

-D— Ozonated

 

 

r

0 1 2 3 4 5 6 7 8

Storage Time (weeks)

Figure 2.22 Effect of Time and Ozonation on the Concentrations of Total

Cone. of Soluble Sulfate (mg/L)

Sulfides in Stored Swine Manure Slurry

 

 

+ Untreated

 

—D— Ozonated

 

 

0 1 2 3 4 5 6 7
Storage Tlme (weeks)

Figure 2.23 Effect of Time and Ozonation on the Sulfate Concentrations in

Stored Swine Manure Slurry

58

 

1800 -
1600 -
1400 -
1200 4

+ Untreated

 

Cone. of l-Butanol (mg/L)

-D- Ozonated

 

 

5 6 7 8 9 10 ll 12

Storage Time (weeks)

Figure 2.24 Effect of Time and Ozonation on the Odor Intensity of Stored

Swine Manure Slurry

 

Odor Intensity (ppm)

 

+ 12-Jun
-I- 19-Jun
-I- 26-Jun
-a- 3-Jul
+ IO-Jul
+ 17-Jul
-I— 24—Jul
+ 31-Jul
—I- 7-Aug
-I— 14-Aug
+21-Aug
+28-Aug

 

+4-Sep
12 14

r I l

8 10

Weeks

Figure 2.25 Effect of Subsequent Storage on the Odor Intensity of
Ozonated Swine Manure Slurry

59

 

18000

-O—Untreated -D—Ozonated

 

 

BOD; (mg/L)

 

 

 

 

0 1 2 3 4 5 6 7 8 9 10 ll 12
Storage Time (weeks)

Figure 2.26 Effect of Time and Ozonation on the BODS Concentrations in

 

 

 

 

Stored Swine Manure Slurry
30000 *
27000 . -K
§ 24000 - ..'~ // \
a _‘ I
c: 21000 \
U 'I " }I_.‘
18000 ’ '\
' + Untreated —D— Ozonated '
15“ I 1 I' I I I I I— —I I I I
0 1 2 3 4 5 6 7 8 9 10 11 12

Storage Time (weeks)

Figure 2.27 Effect of Time and Ozonation on the COD Concentrations in
Stored Swine Manure Slurry

60

 

18000

+ Untreated -D—Ozonated

 

 

 

 

i
c8
:3
‘53
S
e
0
6“) I I T I I I I I I T I I
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.28 Effect of Time and Ozonation 0n the Concentrations of Total
Solids in Stored Swine Manure Slurry

 

 

Cone. of Ammonia-Nitrogen (mg/L)

 

 

 

2500 ‘ -+— Untreated —n— Ozonated
2% I I I I I I I I I I I
0 1 2 3 4 5 6 7 8 9 10 11 12
Storage'lime (weeks)

Figure 2.29 Effect of Time and Ozonation on the Concentrations of
Ammonia Nitrogen in Stored Swine Manure Slurry

61

 

 

 

 

 

3
.3
a.
3
E
.9.
.e
:1
'3
E
g 200 ‘ + Untreated -D— Ozonated
8
O I I I I r I I I I I I
0 1 2 3 4 5 6 7 8 9 10 1 1 12
Storage Time (weeks)

Figure 2.30 Effect of Time and Ozonation on the Concentrations of Soluble
Phosphate in Stored Swine Manure Slurry

 

 

 

 

z + Untreated Aerobes
1.00E+04 ~ + Ozonated Aerobes
+ Untreated Anaerobes
—a— Ozonated Anaer
1 .mE'I'O3 I I f IomS I I I I I j I I

 

0 1 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.31 Effect of Time and Ozonation on Aerobes and Anaerobes in
Stored Swine Manure Slurry

62

 

1 .00E+08
1 .00E+O7
1 .00E+06

   
 

£10013+03 «
.2

 

 

 

ELOOE+02 ~
groomm - +Untreated -D—Ozonated
1.(X)E+m I I I I I I I I I j I

0 1 2 3 4 5 6 7 8 9 10 11 12

Storage Time (weeks)

Figure 2.32 Effect of Time and Ozonation on Total Coliforms in Stored
Swine Manure Slurry

+Untreated -D-Ozonated

Most Probable Number (per mL)

 

 

 

0 l 2 3 4 5 6 7 8 9 10 11 12
Storage Time (weeks)

Figure 2.33 Effect of Time and Ozonation on E. Coli in Stored Swine
Manure Slurry

63

 

+ Untreated -0- Ozonated

 

Number (per mL)

 

 

 

O I I r I I I I 1 I I I

0 l 2 3 4 5 6 7 8 9 10 1 1
Storage Time (weeks)

Figure 2.34 Effect of Time and Ozonation on Coliphage Plaques Produced
on E. Coli in Stored Swine Manure Slurry

Chapter 3

THE USE OF OZONE TO REDUCE THE CONCENTRATION OF
MALODOROUS METABOLITES IN SWINE MANURE SLURRY BY USING A

PILOT-SCALE SYSTEM

3.1 Introduction

Environmental pollution by livestock waste is a major nation-wide concern of the
animal agriculture industry. An increasing number of civil litigations, more restrictive
township ordinances, demands for the protection of public health and new Federal and
State regulations governing surface run-off and ground water contamination necessitate
that the livestock industry take action to control the environmental problems associated
with livestock production.

Malodors emanating from livestock production are the major source of complaints
against livestock farms. As livestock farms have moved to larger and more intensive
production units, odor complaints from nearby residents have correspondingly increased
(Ritter, 1989). Innovative methods to control, reduce or eliminate malodors associated
with livestock farms must be developed and employed to ensure the future viability of the

industry.

65

More than 160 odorous compounds have been identified in livestock wastes (O’Neill
and Phillips, 1992). Among these compounds, the volatile phenolic and indolic
metabolites are among the most malodorous of the compounds found in swine manure
slurry (Spoelstra, 1977). These compounds are formed from the bacterial degradation of
tyrosine and tryptophan in the intestinal tract and during storage of swine waste
(Yokoyama et al., 1982). Other volatile components that have been identified in swine
are the C2 to C5 carboxylic acids (Schaefer, 1977), aromatic carboxylic acids, alcohols,
carbonyl compounds, and sulfur-containing compounds (Yasuhara et al., 1984).
Williams and Evans (1981) indicated that during storage the odor offensiveness
increased, as did the concentration of malodorous compounds including phenol, p-cresol,
skatole and volatile fatty acids (VFAs).

Ozonation is recognized as an environmentally safe and effective process for the
treatment of industrial effluents, drinking water and sewage (Masten and Davies, 1994).
Ozonation has the potential to reduce the malodorous metabolites produced during
anaerobic storage of livestock waste by chemically oxidizing these compounds. Ozone is
a very strong oxidant that attacks the cell wall and the cytoplasmic membrane of the cell
(Block, 1982), thereby inactivating the organism. As such, ozone has the potential to
reduce the number of micro-organisms present in waste, and thereby control the rate of
production of these malodorous metabolites. Previous researches (Wu et al., 1998;
Watkins et al., 1997) indicate that ozone preferentially oxidizes odor producing
metabolites, reduces odor intensity and inactivates micro-organisms.

Scale-up of ozone reactors typically occurs after bench-scale and pilot-scale

experiments have been conducted (Finch and Smith, 1990). In the previous chapter, it

66

was reported that an ozone dosage of 1 g/L was found to be effective in controlling the
malodor of fresh and stored swine manure slurry using a lab-scale reactor. However, it is
also imperative to test the ozonation process using a pilot-scale reactor prior to the
installation of a full-scale ozonation system. This will allow us to determine the effect of
the operational parameters on the removal of malodors in the swine manure. As the
reaction rate of those major malodorants including phenolic and indolic compounds were
found to be very fast (Hoigné and Bader, 1983), the efficiency of mass transfer becomes
the main controlling factor in ozonating swine manure. Therefore, an attempt should be
made to enhance the mass transfer of ozone during the treatment. Based upon this, the
pilot-scale system will be designed to optimize the mixing of the liquid and gaseous
ozone. A cost analysis will also be done to evaluate the capital and operational costs of

the pilot-scale system. These costs will be related to the full-scale system.

3.2 Materials and Methods
3.2.] Manure Collection and Storage

Fresh swine manure was collected from the Michigan State University Swine
Research and Teaching Farm (Figures 3.1-3.2). Four 115 kg finishing pigs were placed in
a metabolism crate and fed a standard com-soy diet. The crate had a slotted floor, which
allowed for the collection of the liquid fraction of the manure slurry (hereafter, referred to
as the manure slurry). Each day’s manure slurry was stored at 4 0C to minimize changes
that might have occurred in its physicochemical and biological properties. After
collection over a 5-day period, the manure slurry was decanted into a 150 liter storage

tank. During three weeks of storage the manure was maintained at 18 0C. It was stirred

67

and sampled once per week to determine the concentrations of COD, phosphates,

sulfides, sulfates, phenolic and indolic metabolites, and volatile fatty acids.

3.2.2 Ozone and Ozonation Reactor

A semi-continuous pilot-scale ozonation system (Figures 3.3-3.4) was constructed
and used for treating the manure slurry. The 3.5-gallon reactor was equipped with four
baffles, each with 1/ 10 of the width of the reactor diameter, located on the wall of the
reactor and a multiple-stage mixer consisting of Rushton disc turbines with the ratio of
1/3 of the turbine size to the reactor size, which was demonstrated to be the optimum
ratio in maintaining the effective mixing with the minimum power needed (Oldshue,
1983). Using this design, the formation of central vortex or swirling would be avoided
(McCabe et al., 1985; Tatterson, 1994) and complete mixing of liquid along the
horizontal and vertical paths would be produced in the tall reactor (Oldshue, 1983). To
maintain the desired temperature during treatment (14 °C, 18 °C, 25 °C), a water bath
was used to circulate water around the jacket of the reactor.

Ozone was generated from pure oxygen by corona discharge using Ozonia® Model
No. GTC-O.5C ozone generator (Ozonia Corp., Lodi, NJ). A Venturi in-line injector
(Mazzei Injector Corp., Bakersfield, CA) was installed at the inlet of the ozone stream.
A circulator pump (Grundfos Pumps Corp., Clovis, CA) was used to circulate the liquid
through the venturi. This system resulted in the generation of very small bubbles (< 1
mm diameter) of ozone-enriched oxygen. The inlet stream of ozone flowed at 1.8 Umin
and had a concentration of 153 mg of ozone per liter of gas. Detention times of 10, 20,

30, and 40 minutes provided ozone dosages of 0.25, 0.5, 0.75, and 1 g/L, respectively. To

68

reduce foaming caused by the bubbling of ozone into the manure slurry, 10 mL of an anti-
foaming agent (Anti-foam 289, Sigma Co., St. Louis, M0), were added to the slurry just
prior to ozonation.

At weekly intervals during the three weeks of storage, 3-gallon aliquots including
fresh and stored manure slurry were ozonated. In addition, the last aliquot of ozonated
manure slurry was stored for an additional four weeks to determine if the malodors

reoccurred.

3.2.3 Odor Evaluation

A odor panel, consisting of five untrained graduate students in the Department of
Civil and Environmental Engineering at Michigan State University, was asked to evaluate
the odor of the raw and ozonated manure slurry samples. Each individual in the panel
ranked the samples in order of worst to best odor and classified the odor of each sample
as acceptable, unacceptable, or extremely unacceptable. The classifications of acceptable,
unacceptable, and extremely unacceptable were ranked as l, 2, and 3 respectively. A
repeat measure analysis of variance was used to explain the subjectively determined
measure of odor at weeks 0, 1, 2, and 3 (The SAS System for Information Delivery,
Version 6.12., SAS Institute, Inc., Cary, NC). Independent variables were the panelists
and ozone dosages. The orthogonal test for sphericity was P=0.99, indicating that use of
the major assumptions for this procedure were not violated and the univariate tests were

valid. Specific contrasts were performed between ozone dosage levels.

69

3.2.4 Chemical Analysis

The aqueous ozone concentration was quantified using an ozone monitor (Exidyne
Instrumentation Technologies Inc., Exton, PA) through which a small amount of liquid
was passed across a detection probe. The gaseous ozone concentrations in the inlet and
outlet streams were determined spectrophotometrically. The absorbance of the ozone
solution was measured in a 2-mm flow-through UV cell at a wavelength of 258 nm. An
extinction coefficient of 3000 M4ch was used to convert absorbance to concentration
(Nowell and Hoigné, 1988).

Chemical oxygen demand (COD) was determined using pre-packaged Hach vials (0-
1,500 mg/L) (Hach Company, Loveland, CO). Phosphate and sulfate were determined
using ion chromatography (Method 4110, Standard Methods for the Examination of
Water and Wastewater). Ammonia was measured with an ammonium electrode
connected to a pH meter and quantification was accomplished using the method of known
addition (Method 4500-NH3, Standard Methods for the Examination of Water and
Wastewater 1995). Sulfides were measured using a purge and trap method (Barth and
Polkowski , 1974), employing nitrogen gas to purge hydrogen sulfide from the solution.
All sulfide species were first converted to hydrogen sulfide by the addition of 10 mL of
concentrated sulfuric acid to lower the pH of the samples to 2. The trap solution was
prepared by adding 15 mL of zinc acetate (2N) to 135 mL deionized water. After
trapping, the sulfides were determined according to Method 4500-82' (Standard Methods
for the Examination of Water and Wastewater, 1995). Aldehydes were determined by the
formation of 2,4-dinitrophenylhydrazones (DNPH) using solid-phase extraction and high

performance liquid chromatography (HPLC). 15 mL of DNP reagent (0.25 g 2,4-

70

dinitrophenylhydrazine were dissolved in 100 mL of 6 M hydrochloric acid) were added
into 100 mL of aqueous sample. The mixture was stirred for 5 minutes and then passed
through a C18 bonded cartridge that has been previously conditioned with methanol. The
cartridge, which contains the DNPH derivative, was washed with DI water and eluted
with two 500 uL aliquots of methylene chloride directly into a 1 mL volumetric flask.
The concentrated extract was injected into a HPLC equipped with C18 column and an
ultraviolet (UV) absorbance detector whose wavelength was set to 365 nm. The mobile
phase was used a binary mixture consisting of 80% acetonitrile and 20% DDI. The flow
rate was operated at 1 mIJmin.

The concentrations of volatile fatty acids (including acetate, propionate, butyrate,
isobutyrate, valerate, isovalerate) were determined using a HP 5890 gas-liquid
chromatograph (Hewlett-Packard Inc., Avondale, PA) with a flame ionization detector
(FID). One mL of concentrated hydrochloric acid was added to 10 mL of manure slurry
sample. After the mixture was agitated in a vortex mixer, it was centrifuged at 12,000 G
for 10 minutes to remove particulate matter. The supernatant (3 11L) was injected directly
into a 900 mm x 6 mm glass column packed with Supelco GP 10% SP-2100 on 100/ 120
Chromosorb W AW (Supelco Inc., Bellefonte, PA). The carrier gas was nitrogen flowing
at 25 mIJmin. The column temperature was held at 110 0C for the first three minutes
after injection; the temperature was ramped at 5 0C/min to 170 0C and held for fifteen
minutes. Injection port and FID temperatures were maintained at 150 and 200°C,
respectively.

The concentrations of phenol, p-cresol, p-ethylphenol, indole, and skatole were

quantitatively determined in the swine manure slurry by gas-liquid chromatography

71

(Yokoyama et al., 1982). The pH of each sample (10 mL) was adjusted to 4.0 with
concentrated sulfuric acid. The analytes were extracted from the acidified samples twice
with 30 mL of methylene chloride. The extract was evaporated down to a volume of
approximately 1 mL using a rotary evaporator, transferred to a screw cap vial, and made
up to 4.0 mL with methylene chloride. Samples (2 11L) were analysed directly by GC-FID
(HP 5890), using a stainless steel column (2 m x 2 mm) packed with 4% Carbowax
80/120 Carbopack (Supelco, Bellefonte, PA). The oven temperature was programmed to
increase from 100 0C to 175 0C at a ramping rate of 8 0C/rnin. Injection port and FID
temperatures were maintained at 200 0C. The efficiency of recovery of the volatile

phenolic and indolic metabolites was determined to be greater than 95% (Yokoyama et

al., 1982).

3.3 Results and Discussion
3.3.1 Effect of Storage and Ozonation

The stored manure slurry was analyzed weekly for its chemical characteristics and
odor. After three weeks of storage, the pH of the swine manure slurry decreased from 8.9
to 7.9. During storage, the VFA concentration increased from 62 to 240 mM. The
concentrations of other chemical analytes (e.g., ammonia-nitrogen, sulfides, phenolic
metabolites, phosphate and sulfate) also increased (Table 3.1). The chemical oxygen
demand (COD) of the manure remained constant, around 14,000 mg/L, during the storage
period. The concentrations of all aldehydes, including formaldehyde, acetaldehyde,
propionaldehyde, butyraldehyde, valeraldehyde, and hexaldehyde, in the all fresh and

stored manure samples were below the detection limits (1 ppb). Acetate was the

72

predominant VFA, followed in decreasing order of concentration by propionate, butyrate,
isovalerate, isobutyrate, and valerate. This is consistent with the observation of Cooper
and Comforth (1978). The concentrations of all VFAs were observed to increase during
storage. The odors (Table 3.2) of the fresh and one-week-storage samples were classified
as unacceptable and the odors of the samples held for longer storage times were deemed
as extremely unacceptable.

The ozonation of the fresh and stored manure slurries was evaluated at different
dosages of ozone to determine the minimal dose for effective odor reduction (Table 3.2).
The program used in SAS is provided in the Appendix. The effect of overall treatment
was highly significant at P=0.0001, while the effect of the panelists was not significant at
P=0.34. The effect of storage time was significant at P=0.0047. At all four weeks, the
negative control treatment (0 g/L ozone dosage) differed at P< 0.05 from all dosages
greater than 0.25 g/L; the 0.25 g/L dosage differed from all higher dosages at P<0.05, and
the 0.5 g/L dosages did not differ from higher dosages at P>0. 13. These results indicate
that no significant additional beneficial effect was realized at ozone dosages greater than
0.5 g/L. Therefore, the results from the odor panel indicateded that ozone is effective in
reducing malodor to an acceptable level at a dosage of 0.5 g/L.

Sulfides were observed to be removed rapidly from either fresh or stored manure
slurries using ozone dosages greater than 0.25 g/L. The concentration of phenol (Figure

3.5) in the manure slurry that had been stored for more than 2 weeks was observed to

73

Table 3.1 The Characteristics of Fresh and Stored Swine Manure

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fresh 1St Week 2IE Week 3rd Week
Manure Storage Storage Storage

H 8.9 8.7 8.1 7.9
NH3-N (mg/L) 2960 3400 4000 4930
Sulfides (mg/L) 1.6 8.0 12.0 8.0
COD (mg/L) 13500 13500 13500 13500
Phenol (mg/L) 15.8 13.6 52.0 61.4
p-Cresol (mg/L) 2.0 22.5 133.8 167.5
p—Ethylphenol (mg/L) 0.6 4.8 3.6 5.4
Indole (mg/L) ND ND ND ND
Skatole (mg/L) ND 4.4 2.5 2.0
Dissolved phosphate (mgL) ND 200 1550 2150
Dissolved sulfate (mg/L) 1660 4120 6030 4000
Acetate (mM) 22.1 29. 1 48.0 93. l
Propionate (mM) 2.8 2.5 3.9 8.2
Butyrate (mM) 1.2 1.3 2.9 5.1
Isobutyrate (mM) 0.1 0.2 0.5 0.9
Valerate (mM) 0.2 0.2 0.2 0.3
Isovalerate (mM) 0.3 0.4 0.6 1.1
Total VFAs (mM as carbon) 60.7 74.8 125.3 241.3

 

ND = Not Detectable

 

Table 3.2 Evaluation of the Odor of Fresh and Stored Manure before and after Ozonation

74

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Panelist Fresh Manure 0.25 g/L 0.5 g/L 0.75 g/L 1 g/L
Ozone ozone Ozone Ozone

V 2 2 1 1 1

W 2 3 1 1 1

X 1 1 1 1 1

Y 2 2 2 1 1

Z 2 l 1 1 1
Panelist 1St Week 0.25 g/L 0.5 g/L 0.75 g/L 1 g/L
Storage Ozone Ozone Ozone Ozone

V 2 2 2 2 1

W 2 2 1 1 1

X 3 2 l 1 1

Y 3 3 2 1 1

Z 2 2 1 1 l
Panelist 2'”ka 0.25 g/L 0.5 g/L 0.75 g/L 1 g/L
Storage Ozone Ozone Ozone Ozone

V 3 3 2 l 1

W 3 2 l 1 1

X 2 3 2 1 1

Y 3 3 1 l 1

Z 3 2 1 l 1

Panelist 3rd Week 0.25 g/L 0.5 g/L 0.75 g/L 1

Storage Ozone Ozone Ozone Ozone

V 3 2 1 2 l

W 2 3 2 1 1

X 3 3 2 1 1

Y 3 2 1 1 1

Z 3 2 1 1 2

 

 

 

 

 

 

l: acceptable; 2: unacceptable; 3: extremely unacceptable

 

75

decrease with increasing ozone dosage. In the fresh manure slurry and the manure slurry
that had been stored for only one week, the concentration of phenol did not follow a
consistent pattern, although a substantially lower phenol concentration was measured
after ozonation at a dose of 1 g/L (Figure 3.5). Similar trends for the decrease in the
concentration of p-cresol (as compared to phenol) were observed except that the
concentration of p-cresol in the fresh slurry did not decrease until 1 g/L of ozone had been
applied (Figure 3.6). The concentrations of p-ethylphenol (Figure 3.7) and skatole (Figure
3.8) were also observed to decrease significantly. The concentrations of other analytes,
such as ammonia nitrogen, phosphate, sulfate, COD, and VFAs, were not affected by the
ozonation treatment.

Phenol, p—cresol, p-ethylphenol, indole, skatole, hydrogen sulfide, and VFAs are
thought to be the major malodorous compounds contributing to the odor problem of
swine manure slurry (Spoelstra, 1977, 1979, 1980; Williams and Evans, 1981; Williams’
1984). Storage of manure results in the formation and accumulation of these malodorous
compounds through anaerobic microbial activity and fermentation (Welsh et al., 1977).
The results of this study confirm the finding of Welsh et a1. (1977) as the rapid
accumulation of malodorous compounds during the three weeks of storage correlated
with an increase in odor intensity. Among the malodorous compounds, the concentration
of p-cresol was observed to increase the most significantly. It is thought that p-cresol
contributes most directly to the odor of swine manure sluny. Ozone dosages greater than
0.5 g/L resulted in a substantial decrease in the concentration of this compound. This

result did not depend on the length of storage.

76

The concentrations of VFAs did not decrease even after ozone treatment at a dosage
of 1 g/L, although a marked improvement in odor was evident. Therefore, at the pH of
the manure slurry (7 to 9), VFAs do not appear to be a suitable indicator of odor intensity.
VFA apparently play a minor role in contributing to the odor characteristics of the manure
slurry (as compared with the phenolic and indolic metabolites).

The microbiological data, including numbers of aerobes, anaerobes and coliphage,
were also analyzed. The number of total aerobes decreased significantly during 3-week
storage and decreased to a lesser extent upon ozonation (Figure 3.9). The number of total
anaerobes decreased slightly during 3-week storage and upon ozonation (Figure 3.10). It
is thought that the conditions under which the liquid manure slurry is stored are
unfavored to the growth of these microorganisms, especially to the aerobes. Ozone dosed
from 0.25 to 1 g/L was found to deactivate and kill some of the aerobes and anaerobes.
This result is consistent with that presented in Chapter 2. The number of coliphage was
found to increase after ozonation at lower dosages, whereas, its numbers decreased at the
higher ozone dosage (Figure 3.11). These results indicated that ozone can induce the
release of coliphage while killing its host, E.coli. When the ozone dosage was increased,
ozonation resulted in a decrease in the number of coliphage. In earlier work (see Chapter
2), only an ozone dosage (1 g/L) was used. As such, the numbers of coliphage were
observed to decrease.

The utilization of ozone in the pilot-scale system was evaluated using three different

parameters: ozone utilized, transfer efficiency, and transferred mass. These are defined as

(Chrostowski et al., 198; Langlais et al., 1990):

77

 

. . (Cin -C°m)xV8
0U (Ozone Ut1lized, mg/L) = V
l
C. - C
TE (Transfer Efficiency, percent) = _( mC out )

in

TM (Transferred Mass, mg/min) = TE x Cin x Q

where, Cm = ozone concentration in the feed gas (mg/L); Com = ozone concentration in the

['65

x100

effluent gas (mg/L); Vg = volume of applied ozone (L); V1: volume of liquid treated (L);

Cm = aqueous ozone residual in the reactor (mg/L); Q = applied flowrate of ozone

(Umin). The experimental results are summarized in Table 3.3. The applied ozone was

fully consumed by the manure slurry using this pilot-scale ozonation system. Therefore,

almost no wasted ozone can be seen in the off-gas when the pilot-scale system was used

to treat the fresh and stored swine manure.

Table 3.3 The Systematic Parameters of Ozonating Stored Manure

 

 

 

 

 

 

 

Time Applied Dosage C1,. Com Cres OU TE TM
(min) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (%) (mg/min)
0 0 0 0 0 0 0 0
10 250 153 0 0 250 100 275
20 500 153 1 0 498 99.4 273
30 750 153 2.5 0 744 98.4 271
40 1000 153 4 0 988 97.4 268

 

 

 

 

 

 

 

 

Ci“ = Ozone concentration in the feed gas (mg/L)
Com = Ozone concentration in the effluent gas (mg/L)
Cm = Aqueous ozone residual in the reactor (mg/L)
OU = Ozone Utilized (mg/L)

TE = Transfer Efficiency (percent)

TM = Transferred Mass (mg/min)

 

78

3.3.2 Comparison of Stripping, Oxygenation, and Ozonation

To compare the effects of stripping with nitrogen, oxidation by oxygen, and
oxidation by ozone, we treated the manure slurry with nitrogen, oxygen, and ozone-
enriched oxygen (Figures 3.12 to 3.15). A flowrate of 1.8 11min and treatment time of 40
minutes were used in all experiments. The percentage removals reported are calculated
for the individual processes. For example, if the concentration of a compound were to be
reduced by 10% during stripping, 25% during oxygenation and 100% during ozonation,
the percent removals for volatilization (stripping), aerobic degradation and reaction with
ozone are reported as 10, 15 and 75%. Our experiments showed that less than 20% of the
phenol was removed by nitrogen and oxygen. In contrast, approximately 80% of the
phenol was removed by its reaction with ozone at a dosage of 1 g/L. The stripping of p-
cresol by nitrogen appears to be insignificant (less than 10%) compared with that
removed by the oxidation of oxygen (40%) and ozone (50%). Approximately 15% of the
p-ethylphenol was removed by stripping, whereas 30% was removed by oxygenation and
55% by ozone. Approximately 30% of skatole was removed by stripping, whereas only
15% was removed by oxygenation and 55% by ozonation. The use of nitrogen and
oxygen to compare the efficiency of ozone in removing the odorous compounds
demonstrates that the reaction of these compounds with ozone is the predominant
mechanism resulting in the reduction in the odor of the manure slurry. Stripping had a
greater effect on the removal of skatole and p-ethylphenol as compared to p-cresol and
phenol, most likely because the first two compounds are more hydrophobic than the latter
two. About 50% of the p-cresol was removed during oxygenation (for 40 minutes),

which would have resulted both in the stripping of p-cresol and in an enhancement of

79

activity of aerobic microorganisms. It appears that p—cresol is biodegraded more easily
than the other phenolic and indolic compounds, since the enhancement in the efficiency at
which p-cresol was removed during treatment with oxygen (as compared to that with

nitrogen) significantly greater than that observed for the other compounds.

3.3.3 Comparison of Varying Temperature and Using Hydrogen Peroxide

To determine the effect of temperature (14-25 °C) on the efficiency of the treatment
of the swine manure slurry, samples which had been stored for one month were ozonated
at different temperatures using an ozone dosage of 1 g/L. Temperature did not have a
significant effect (at the 95% confidence interval) on the removal of either the phenolic or
the indolic compounds (Figures 3.16-3.19). The aqueous solubility of ozone decreases
with increasing temperature, whereas the reaction rate of ozone with organic compounds
increases with increasing temperature. In the swine manure slurry, which is a very
complex mixture, these two factors appear to counter-balance each other over the
temperature range studied (14-25 °C). This result would indicate that, for this
temperature range, the design of an ozonation process will not be affected by seasonal
variations in the temperature of the manure slurry.

The effect of ozone combined with hydrogen peroxide on the removal of the
phenolic and indolic compounds was also evaluated. Hydrogen peroxide to ozone ratios
of 0, 0.51, 0.84, 1.4 g/g were used. No significant benefit (at the 95% confidence
interval) from the addition of hydrogen peroxide was obtained (Figures 3.20-3.23).
Although advanced oxidation processes (AOPs) have been used to improve the efficiency

of ozonation of water and wastewater (Brunet et al., 1985; Glaze et al., 1987; Preis et al.,

199

the!
0201
suni
had

rnay
deco
(2) [l
react
utco

Odorc

80

1995; Roche and Prados, 1986), it has been reported that AOPs are not advantageous for
the oxidation of specific organic compounds which are highly reactive with molecular
ozone (Masten and Butler, 1986; Trapido et al., 1994; Beltran et al., 1996). Our studies,
similarly, indicate that the addition of hydrogen peroxide (at the ratios from 0 - 1.4 g/g)
had no beneficial effect for the oxidation of the phenolic and indolic compounds. This
may have been due to: (1) the rapid consumption of the free radicals generated during the
decomposition of ozone by other non-target compounds in such a complicated matrix, or
(2) the design of the reactor in which hydrogen peroxide was added at the top of the
reactor and ozone at the bottom. Had all of the ozone reacted before it could have come
in contact with the hydrogen peroxide, free radicals could not have formed. As such, the

odorous metabolites in the manure slurry could be oxidized only by ozone.

3. 3.4 Effect of Subsequent Storage of Manure Slurry after Ozone Treatment

To determine the extent to which the malodor would reoccur after ozone treatment,
the ozonated manure slurry was stored at room temperature (18 °C) and analyzed each
week for one month. Results indicate that the concentrations of the phenolic and indolic
compounds decreased to a level that was less than that observed in the manure slurry
immediately after ozonation (Table 3.4). The odor intensity remained at the same level as

that observed in the freshly ozonated manure sluny. In addition, the concentrations of the

81

Table 3.4 Analysis of Stored Ozonated Manure (1 g/L) over a Period of Four Weeks

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Storage Time (week) 0 I 2 3 4

Phenol (mg/L) 15.1 23.3 6.4 1.5 1.7
p-Cresol (mg/L) ND 5.2 ND ND ND
p-Ethylphenol (mg/L) 2.6 7.2 5.9 1 .8 1.4
Indole (mg/L) ND ND ND ND ND
Skatole (mg/L) ND 2.0 2.2 ND ND
Acetate (mM) 92.1 77.2 29.2 5.0 4.6
Propionate(mM) 7.7 7. 1 3.0 0.9 1.0
Butyrate (mM) 2.0 1.9 1.5 0.6 0.4
Isobutyrate (mM) 4.2 4.1 3.6 2.3 1.0
Valerate (mM) 1.0 1.0 0.6 0.3 0.3
Isovalerate (mM) 2.4 2.2 2.3 1.8 0.7
VFAs (mM as Carbon) 249 216 102 34 23

 

 

 

 

0: Freshly Ozonated Manure Slurry

I : One Week of Subsequent Storage after Ozonation

2: Two Weeks of Subsequent Storage after Ozonation
3: Three Weeks of Subsequent Storage after Ozonation
4: Four Weeks of Subsequent Storage after Ozonation
ND: Not Detectable

VFAs were observed to decrease significantly (from 250 mM to 20 mM) during one-
month of storage. The major malodorous metabolites (phenolic and indolic compounds)
were eliminated during ozonation and were not regenerated during subsequent storage.

It is possible that these metabolites were not regenerated during storage because (1) ozone
degraded the tryptophan and tyrosine, the precursors of the phenolic and indolic
compounds, or (2) ozone inactivated the class of malodor-causing bacteria that degrade
tryptophan and tyrosine into the phenolic and indolic compounds. On the other hand, the
concentrations of VFAs were found to decrease during storage (post-ozonation). Since a

wide class of bacteria would be able to use the VFAs as a growth substrate, it is thought

that 02
to imp
accutr
gener
proce
four

for a

0P

of

ti

82

that ozonation is ineffective at reducing the number of these microorganisms responsible
to impede methanogenesis. This phenomenum was verified by observing the rising and
accumulating of gaseous bubbles on the surface of storage tank probably due to the
generation of carbon dioxide and methane. Because it was observed that the ozonation
process effectively removes odorous metabolites and inhibits their formation for at least
four weeks following treatment, the ozonated manure slurry could be potentially stored
for at least four weeks prior to land application without encountering the problems

associated with malodors.

3.3.5 Economic Analysis of Pilot-scale Ozonation System

An economic analysis of the pilot-scale system, including construction and
operational costs, is summarized in Tables 3.5 and 3.6. The capital costs include the cost
of materials, equipment, and construction. The operational cost includes the supply of
oxygen gas and electric consumption when operating the ozonation system. Considering
the current electricity rate in Michigan ($0.07/kW-hr) and 40 minutes ozone treatment (at
1 g/L ozone dosage), the operational cost was estimated to be approximately $2.00 for
treating 100 gallons of swine manure. When using a detention time of 20 minutes,
corresponding to an ozone dosage of 0.5 g/L, the operational cost would be reduced to
half ($1.00/ 100 gallons). At this dosage, a maximum capacity of 144 gallons of liquid
manure per day can be obtained using a treatment cycle of 30 minutes including 5
minutes for filling, 20 minutes of treatment, and 5 minutes for flushing. Thus, the
operational cost can be estimated to be about $500 per year. The unit cost including

capital and operational costs to treat the liquid manure is $0.024 dollar/gallon manure as

calcula
liquid <
cost w
from?
of S]

$3.8C

A.

#17 T‘ITY‘IT Fl7

 

83

calculated in Table 3.7. If this rate is applicable to the full-scale system used to treat the
liquid discharge of MSU Swine Farm (about 1,600 gallons per day), the total operational
cost will be $38 per day. On average, 10 pigs (at 220 lbs) are sent to the market per day
from MSU Swine Farm (Tengman, 1995). Thus, the profits on each pig (within the range
of $11.00 — 19.00 for 1993-1996 reported by Iowa State University) will be reduced by

$3.80 if the ozonation system is used to treat the manure waste.

Table 3.5 The Capital Cost for the Pilot-scale Ozonation System

Items Cost S dollars
Ozone 5 000
Pilot-scale reactor 1 000
Venturi in' 100
Circulation 200

Ozone monitor 3 000
'on facili 000
-control 1 000
Flowmeter l
Tubin and fittin 500

Total

 

84

Table 3.6 The Operational Cost for the Pilot-scale System (Based on the calculation at 0.5

 

 

 

 

 

 

 

 

g/L ozone dosage)
Items Power Consumption Cost ($)/ 100 Gallons Treated
(watt)

Ozone generator 115 0.18
Circulation pump 85 0.13
Agitator Facility 50 0.08
Temperature-control System 60 0.10
Oxygen supply 0.50
Total 0.99

 

 

 

 

Table 3.7 The Calculation of Total Cost for the Pilot-scale System

Initial cost: $ 14,000

Initial annual operating cost: $ 500, and increasing cost per year (3%): $ 15
Estimated salvage value: $ 500

Life expectancy: 15 years

Interest factor: 8 %

 

The total cost (present worth): Initial cost + Annual operating cost + Increasing operating
cost - Salvage value

15 I4
=14,000+500[(1+0'08) ’1} 15 FINDS) ‘1-14][-———1 ]—500(1+0.08)"5

0.08(1+0.08)" 0.08 0.08 (1+0.08)“

 

 

=14,000+4300+650-160

=1 8790

 

The total volume of swine manure during 15 years: 144 (gal/day) x 365 (day/year) x 15
(year)=788,400 (gallon)

Therefore, 18790 (dollar)/7 88400 (gallon): $ 0.024 dollar/gallon

85

3.4 Conclusions

Although ozonation has been extensively used in water and wastewater treatment, the
application of ozone has not been used in the field to remove malodors from swine
manure slurry. Based on the pilot-scale study, ozone treatment is a quick and effective
method for removing the malodorous components from swine manure slurry. The design
of an ozone facility to treat swine manure slurry will need to consider neither temperature
variations nor the length of storage (up to 4 weeks) prior to treatment. A minimum
dosage of ozone, 0.5 g/L, has been established to reduce the odor intensity of fresh and
stored manure slurry to an acceptable level. The results of this study have encouraged us
to install a full-scale ozonation system on a 2000-swine operation to investigate the
feasibility of odor-control in a full-scale environment. To protect groundwater, the
quality of which can be degraded by the migration of contaminants from the land
discharged slurry, the treatment of the manure slurry using a lagoon or other biological
treatment after ozone treatment is strongly recommended in order to diminish such
components as ammonia-nitrogen, VFAs, and COD. Direct land application of ozone-
treated swine manure slurry should been evaluated to determine the impact of high
organic loading on the environment and ecology of soils and plants, even though the
odorous substances have been substantially removed.

In the economic analysis, the operational cost when using 0.5 g/L ozone dosage was

about $1.00 perlOO gallons of liquid swine manure. The capital costs were $14,000.
When amortizing the capital and operational costs over 15 years, it was determined that

the profits from the swine operation would be reduced by $3.80 per pig marketed. This

cost '1

{Q

86

cost is excessive and modifications will be necessary if ozonation is to be implemented

on a large scale by swine producers. Future work will focus on these modifications.

3.5 References

1. A.G. AWilliams; M.R. Evans, “Storage of Piggery Slurry”, Agricultural Wastes,
3:311-321 (1981).

2. CL. Barth; L.B. Polkowski, “Identifying Odorous Components of Stored Dairy
Manure”, Transactions of the ASAE, 17:737-741, 747 (1974).

3. E]. Beltran; G. Ovejero; J. Rivas, “Oxidation of Polynuclear Aromatic Hydrocarbons
in Water. 4. Ozone Combined with Hydrogen Peroxide ”, Journal of Industrial
Engineering Chemistry Research, 35:891-898 (1996).

4. J .C. Block, In: Ozonation Manual for Water and Wastewater Treatment (Masschelein,
W.J., Ed), John Wiley and Sons, New York, 69-72 (1982).

5. R. Brunet; M.M. Bourbigot; M. Doré, “Oxidation of Organic Compounds through the
Combination Ozone-Hydrogen Peroxide”, Ozone Science & Engineering, 6: 163-183
(1985).

6. P. Cooper; I.S. Comforth, “Volatile Fatty Acids in Stored Animal Slurry”, Journal of
the Science of Food and Agriculture, 29:19-27 (1978).

7. RC. Chrostowski; A.M. Dietrich; I. H. Suffet, “Laboratory Testing of Ozonation
Systems Prior to Pilot-plant Operations”, Journal of AWWA, 74:38-43 (1982).

8. AD. Eaton; L.S. Clesceri; A.E. Greenberg (Eds), Standard Methods for the
Examination of Water and Wastewater; American Public Health Association, Inc.,

19th ed., Washington, DC, 1995.

9.0

\0.

ll

87

9. GR. Finch; D.W. Smith, “Pilot-scale Evaluation of the Effects of Mixing on Ozone

10.

ll.

12.

13.

14.

15.

16.

17.

Disinfection of Escherichia Coli in a Semi-batch Tank Reactor”, Ozone Science &

Engineering, 13:593-605 (1991).

W.H. Glaze; J.W. Kang; D.H. Chapin, “The Chemistry of Water Treatment Processes
Involving Ozone, Hydrogen Peroxide and Ultraviolet Radiation”, Ozone Science &
Engineering, 9:335-352 (1987).

B. Langlais; D.A. Reckhow; D.R. Brink (Eds.), “Ozone in Water Treatment:
Application and Engineering”, Lewis Publishers, Chelsea, MI, 339-340 (1990).

SJ. Masten; J .N. Butler, “Ultraviolet-Enhanced Ozonation of Organic Compounds: 1,
2-Dichloroethane and Trichloroethylene as Model Substrates”, Ozone Science &
Engineering, 8:339-353 (1986).

SJ. Masten; S.H. Davies, “The Use of Ozonation to Degrade Organic Contaminants
in Wastewater”, Environmental Science & Technology, 28: 181A-185A (1994).

W.L. McCabe; J .C. Smith; P. Harriott, “Unit Operations of Chemical Engineering”,
McGraw-Hill Publications Co., New York, NY, 208-252 (1985).

L.H. Nowell; J. Hoigné, “Interaction of Iron (11) and Other Transition Metals with
Aqueous Ozone”, In Proceedings of the 8‘h Ozone World Congress, 2:E80-E95
(1988).

J.Y. Oldshue, “Fluid Mixing Technology”, McGraw-Hill Publications Co., New
York, NY, 155-215 (1983).

D.H. O’Neill; V.R. Phillips, “A Review of the Control of Odour Nuisance from

Livestock Buildings: Part 3, Properties of the Odorous Substances which have been

ix)

88

Identified in Livestock Wastes or in the Air around them”, Journal of Agricultural
Engineering Research, 53:23-50 (1992).

18. S. Preis; S. Kamenev; J. Kallas; R. Munter, “Advanced Oxidation Processes against
Phenolic Compounds in Wastewater Treatment”, Ozone Science & Engineering,
17:399-418 (1995).

19. W.F. Ritter, “Odour Control of Livestock Wastes: State-of -the-Art in North
America”, Journal of Agricultural Engineering Research, 42: 51-62 (1989).

20. P. Roche; M. Prados, “Removal of Pesticides by Use of Ozone or hydrogen
Peroxide/Ozone”, Ozone Science & Engineering, 17:657-672 (1995).

21. J. Schaefer, “Sampling, Characterization and Analysis of Malodours, Agriculture and
Environment”, 3: 121-128 (1977).

22. SF. Spoelstra, “Simple Phenols and Indoles in Anaerobically Stored Piggery
Wastes”, Journal of the Science of Food and Agriculture, 28:415-423 (1977).

23. SF. Spoelstra, “Volatile Fatty Acids in Anaerobically Stored Piggery Wastes”,
Netherlands Journal of Agricultural Science, 27:60-66 ( 1979).

24. SF. Spoelstra, “Origin of Objectionable Odorous Components in Piggery Wasted and
the Possibility of Applying Indicator Components for Studying Odour Development”,
Agriculture and Environment, 5:241-260 (1980).

25. GB. Tatterson, “Scaleup and Design of Industrial Mixing Processes”, McGraw-Hill
Publications Co., New York, NY, 53-105 (1994).

26. CL. Tengman, “Gravitational Liquid-solid Separation Immediately below Slats
Concentrating Swine Manure Phosphorus in Solid”, Master Thesis, Department of

Agricultural Engineering, Michigan State University (1995).

27. .N

27.

28.

29.

30.

31.

32.

33.

89

M. Trapido; J. Veressinina; R. Munter, “Ozonation and AOP Treatment of
Phenanthrene in Aqueous Solutions”, Ozone Science & Engineering, 16:475-485
(1994).

B.D. Watkins; S.M. Hengemuehle; H.L. Person; M.T. Yokoyama; S.J. Masten,
“Ozonation of Swine Manure Wastes to Control Odors and Reduce the
Concentrations of Pathogens and Toxic Fermentation Metabolites”, Ozone Science &
Engineering, 19:425-437 (1997).

F.W. Welsh; D.D. Schulte; E.J. Kroeker; H.M. Lapp, “The Effect of Anaerobic
Digestion upon Swine Manure Odours”, Canadian Agricultural Engineering, 19:122-
126 (1977).

A.G. Williams, “Indicators of Piggery Slurry Odour Offensiveness”, Agricultural
Wastes, 10:15-36 (1984).

J .J . Wu; S. Park; S.M. Hengemuehle; M.T. Yokoyama; H.L. Person; SJ. Masten,
“The Effect of Storage and Ozonation on the Physical, Chemical, and Biological
Characteristics of Swine Manure Slurries”, Ozone Science & Engineering, 20:35-50
(1998)..

A. Yasuhara; K. Fuwa; M. J imbu, “Identification of Odorous Compounds in Fresh
and Rotten Swine Manure”, Agricultural and Biological Chemistry, 48:3001-3010
(1984).

M.T. Yokoyama; C. Tabori; E.R. Miller; M.G. Hogberg, “The Effects of Antibiotics
in the Weaning Pig Diet on Growth and the Excretion of Volatile Phenolic and

Aromatic Bacterial Metabolites The American Journal of Clinical Nutrition,

35:1417-1424 (1982).

90

Fresh Discharge l-week Storage 2-week Storage 3—week Storage

 

 

 

 

 

 

 

 

 

 

 

 

 

r
if

 

l l l

       

         

 

Storage Tank

 

I Ozone Treatment System

 

 

 

Figure 3.1 The

Collection System for the Swine Manure

91

 

Figure 3.2 The Photo of the Manure Collection System

 

 

 

 

 

 

 

 

 

 

 

1: Oxygen Cylinder

2: Ozone Generator

3: UV Spectrophotometer
4: Three Way Valve

5: K1 Solution

6: Pressure Gauge

7: Flow Meter

8: Flow Rate Controller
9: Venturi Injector

10: Circulation Pump

11: Ozonation Contactor
12: Sampling Ports

13: Dissolved Ozone Monitor
14: Agitator Controller

Figure 3.3 The Configuration of Pilot-Scale System for Ozone Treatment

93

. acteirir

 

 

Figure 3.4 The Photo of Pilot-Scale System for Ozone Treatment

Phenol (mg/L)

94

 

 

 

  
  

Storage Time
(week)

 
   
 

0.25
0.5

Ozone Dosage (g/L)

Figure 3.5 Changes in Phenol Concentration during the Storage and
Ozonation of Swine Waste Slurry

p -Cresol (myL)

  
  

95

y.“ .1». men (sun-z. ;

a“ «.-

Storage Time
(week)

 
   
 

0.25
0.5

Ozone Dosage (g/L) I

Figure 3.6 Changes in p -Cresol Concentration during the Storage
and Ozonation of Swine Waste Slurry

p -Ethylphenol (mg/L)

96

 

 

 

 

 

 

   

 

Storage Time
(week)

 

0.25

0.5
Ozone Dosage (g/L)

Figure 3.7 Changes in p—Ethylphenol Concentration during the
Storage and Ozonation of Swine Waste Slurry

Skatole (mg/L)

97

Ozone Dosage (gJL)

Figure 3.8 Changes in Skatole (3-methyl-indole) Concentration
during the Storage and Ozonation of Swine Waste Slurry

   

Storage Time
(week)

98

   

Storage Time (week)

Figure 3.9 Changes in Number of Total Aerobes during the Storage and Ozonation of
Swine Manure Slurry

 

 

 

OzoneDosage(g/L) 1

Figure 3.10 Changes in Number of Total Anaerobes during the Storage and Ozonation of
Swine Manure Slurry

 

 

 

 

 

Omne Dosage (g/L)

Figure 3.11 Changes in Number of Coliphage during the Storage and Ozonation of
Swine Manure Slurry

101

 

8

U!
C

Concentration (mg/L)
‘5

    

 

 

 

 

 

 

30 —0— Nitrogenation
20 ‘ +Oxygenation
10 ‘ +Ozonation
0 r n x
0 10 20 3O 40

Time (min)

Figure 3.12 Effect of Different Treatment Processes on the Removal of

Phenol (The flowrate of ozone was 25 mg/L-min)

 

8

1

U1
0

Concentration (mg/L)
8

    
  

 

 

 

 

 

 

 

     

30 ~
20 _ + Nitrogenation
10 +Oxygenation
+ Ozonation
0 I I I
O 10 20 30 40

Time (min)

Figure 3.13 Effect of Different Treatment Processes on the Removal of

p -Cresol (The flowrate of ozone was 25 mg/L-min)

=.
p: 23:33:39: ..v
Ax:-

F1;

F1

102

 

 

U»)
C
1

N
M
1

Concentration (mg/L)
5: 8

+ Nitrogenation

p—r
o
1

+ Oxygenation

M
1

 

+ Ozonation

 

O

 

I r I I

10 20 30 40
Time (min)

C

Figure 3.14 Effect of Different Treatment Processes on the Removal of
p -Ethylphenol (The flowrate of ozone was 25 mg/L-min)

 

 

 

 

      

Concentration (mg/L)
Ox

 

 

 

   

+ Ninogenation
2 ‘ +Oxygenation
+Ozonation
0 I r r
0 10 20 30 40

Time (min)

Figure 3.15 Effect of Different Treatment Processes on the Removal of
Skatole (The flowrate of ozone was 25 mg/L-min)

103

 

 
 
    

80 +Temp=12 C
70 +Temp=18 C
+Temp=25 C

  
 

8 S
l

‘5

Concentration (mg/L)

 

 

 

30 ~
20 a
10 a
O I I I
0 0.25 0.5 0.75 1
Ozone Dosage (g/L)

Figure 3.16 Effect of Temperature on the Removal of Phenol

 

 

 

 

 

90

80

70
A +Temp=12 C
i 60 - +Temp=18 C
‘5 50 1 +Temp=25 C
5
E 40 .
5
§ 30 -
U

20 -

10 -

0 I I
0 0.25 0.75 1

0.5
Ozone Dosage (g/L)
Figure 3.17 Effect of Temperature on the Removal of p -Cresol

73E» ====.=-.u.==.. v

Fig

104

 

 

60
50 1
l +Temp=12 C

i 40 +Temp=18 C
v +Temp=25 C
r:
8
g
8

 

 

 

 

0 0.25 0.5 0.75 1
Ozone Dosage (g/L)

Figure 3.18 Effect of Temperature on the Removal of p -Ethylphenol

 

12 -
10
+Temp=12 C
8 . +Temp=l8 C
+Temp=25 C

Concentration (mg/L)
ax

 

 

 

 

4 _
2 s
O I F I

0 0.25 0.5 0.75 1

Ozone Dosage (g/L)
Figure 3.19 Effect of Temperature on the Removal of Skatole

A. {aw—-~ =aout-ro-‘FI-1‘Iauv‘

F1

 

105

 

 
 
   
   

 

 

 

 

90 '

80 +H202/03=0

70 +I'1202/03=0.51
g 60 - +H202/03=0.84
V —X— H202/O3=1.36
: 50 -l
i

40 .4
i
a 30 d
U

20 q

10 1

O I I I

0 0.25 0.75 1

0.5
Ozone Dosage (g/L)

Figure 3.20 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of Phenol

 

 
 
 
  
     

 

 

 

90 »

80

+H202/03=0

70 .? +H202/O3=0.51
§ 60 - +H202/O3=O.84
v +H202/o3=1.36
= 50 -
8
r .0.
§ 30 -
U

20 4

10 -

0

0 0.25 0.5 0.75 1
Ozone Dosage (g/L)

Figure 3.21 Effect of the Ratio of Hydrogen Peroxide t0 Ozone on the
Removal of p -Cresol

106

 

 
 
 
    

40 + H202IO3=0

35 -I- H202/O3=0.51
A + H202/O3=0.84
§ 30 ‘ + H202/O3=1.36

 

 

0 0.25 0.5 0.75 1
Ozone Dosage (g/L)

Figure 3.22 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of p -Ethylphenol

 

14 >

12 + H202/O3=0
-I- H202103=0.51

10 , + H202/O3=0.84

—K— H202/O3=1.36

Concentration (mg/L)

 

 

 

 

0 0.25 0.5 0.75 1
Ozone Dosage (g/L)

Figure 3.23 Effect of the Ratio of Hydrogen Peroxide to Ozone on the
Removal of Skatole

4.1

wa'
rap
19'
Tr
US

[ht

Chapter 4

MASS TRANSFER OF OZONE IN A SEMI-BATCH STIRRED REACTOR

4.1 Introduction

The increasing demand for high quality water has accelerated the use of ozone in
water and wastewater treatment. With its powerful oxidation potential, ozone reacts
rapidly with many organic and inorganic compounds (Hoigné and Bader, 1983‘, 1983b,
1985). Thus, the reaction rate for many ozonation processes is limited by mass-transfer.
Traditionally, the ozonation of the most environmental contaminants is accomplished
using a semi-batch or continuous system in which the gaseous ozone is passed through
the liquid phase. Before ozone can react with any substance in the liquid phase, it must
pass across an interface between gaseous ozone and liquid. As ozone is a sparingly
soluble gas, its transfer efficiency is controlled by such physical parameters as
temperature, gaseous flowrate, mixing speed, type of diffusers, and reactor geometry, and
such chemical parameters as pH, ionic strength, and composition of the aqueous solution.
Since the transfer efficiency of mass transfer of ozone into the aqueous phase can be
described by two variables, mass-transfer and partition coefficients, the effect of these

systematic parameters on the mass-transfer and partition coefficients will be the focus of

107

thi

C01

1.!

fit

130

th.

Bl

th

th

Sl

ir.

108

this study. A mass-transfer model will be deve10ped and used to predict the ozone

concentration profile during the period of mass-transfer.

4.2 Materials and Methods

A semi-batch stirred reactor (Figures 4.1-4.2) made of pyrex glass with a capacity of
1.5 L was utilized to investigate the mass transfer of ozone. To prevent any interference
from impurities, deionized distilled water (DDI) was used. Ozone was generated from
pure oxygen by corona discharge using an ozone generator (Model T-408, Polymetrics
Inc., San Jose, CA), which produced an ozone concentration of ca. 1.3 % (by volume) in
the oxygen gas stream. The oxygen stream was dried using a molecular sieve trap (SIP
Brand #G5301-21) prior to ozone generation. In this reactor, a fritted glass diffuser that
is able to generate fine bubbles (bubble diameter < 1 mm) was located near the bottom of
the reactor and a magnetic stirrer bar was placed under the diffuser to achieve complete
mixing throughout the liquid. In addition, a water jacket around the reactor maintained
the desired temperature with water circulated from a water bath.

The concentrations of gaseous ozone were monitored using an UV
spectrophotometer (UV 1201 , Shimadzu, Japan) by passing ozone gas, from both the
inlet and outlet, through two 2-mm flow cells. The concentration of aqueous ozone was
also determined spectrophotometrically. In this case, the liquid was pumped through a 1-
cm flow cell. The absorbance of the ozone was monitored at the wavelength 258 nm. An
extinction coefficient of 3000 M'lcm’l was used to convert absorbances to concentration

units (Nowell and Hoigné, 1988).

 

0Z0

C011

the

Du

Slrt

5pc“

Th

C0

sy

109

Before initiating each experiment, a three-way valve was used to bypass the inlet
ozone into a potassium iodine solution (trap) until it was observed that the inlet
concentration was constant. When a desired concentration of inlet ozone was obtained,
the run was started by rotating this valve and admitting ozone in oxygen into the reactor.
During the run, the ozone absorbances in the inlet and outlet gas streams and in the liquid
stream were monitored at five-minute intervals by using a kinetic diskette in the UV
spectrophotometer. When all absorbances became stable, the experiment was stopped.
The rate of ozone decomposition was determined by monitoring the decrease of
concentration with time after the introduction of gaseous ozone was ceased. The

systematic parameters that were varied are summarized below:

0 Temperature: 14, 22, 30 °C

0 Inlet concentration: 5.4, 22.4, 44.3, 62.4 mg/L

0 Inlet flowrate: 100, 200, 300, 400, 500 mL/min

0 pH: 6.0-6.5 (unbuffered distilled deionized water), 2.1 (acidified by adding 50 mL of
a solution which was made by the addition of 8 mL concentrated HZSO4 and 120.7 g
NaZSO4 to 0.5 liter DDI water), 6.7 (buffered by adding 50 mL of phosphate solution
which was made by the addition of 92.84 g Nazi-IP04 and 141.13 g KH2P04 to 1 liter

DDI water)

0 Ionic Strength: 0.1, 0.2, 0.3, 0.4, 0.5 (M)

4.3

liq
co
co:
rat
in‘

E5

ef:
Ne

all

(16

de

he
CC

Cc

110

4.3 Model Development

A mathematical model of the mass transport of ozone from the gas into the bulk
liquid in a semi-batch reactor needs to be developed to determine the mass-transfer
coefficient, which is relatively important when the reaction of ozone and target
compounds is controlled by the rate of mass transfer instead of by the chemical reaction
rate. Approaches to developing the mass-transfer model have been proposed by many
investigators (Kuo etal., 1976, Yocum, 1979, Roth and Sullivan, 1981, Shaffer and
Esterson, 1982, Anselmi etal., 1984, Roustan etal., 1987, Sotelo etal., 1989). However,
among these models, only the model proposed by Anselmi et a1. (1984) considered the
effect of head space above the liquid on the concentration of ozone in the gas phase.
Nevertheless, this model still neglects the effect of ozone self-decomposition in the
aqueous and gas phases. As such, Anselmi’s model is inappropriate for ozone solutions
having a pH>6 and for low flowrate systems, in which the self-decomposition of
moisturized ozone gas may be significant due to longer detention times in the system. To
develop a general model which can be used in any situation, the effect of ozone self-
decomposition was taken into consideration although this resulted in more complicated
mathematical computations. Fortunately, Laplace operators can be employed as a
mathematical tool to determine the analytical solutions. In this first model, the overall
mass balance between the aqueous phase and gas phase, including the gas bubble and
headspace have been examined thoroughly. It is also assumed that the gas flow is
completely mixed. As there is some dispute whether the flow pattern of gas is
completely mixed or plug flow (Nishikawa etal., 1981), a second model was developed

in which it was assumed that the flow of the bubbles in the system could be modeled as

111

plug flow. In order to simulate a plug flow system, the reactor is divided into 10 sections,
each completely mixed. As a result, in the second model there are eleven ordinary

differential equations that could be solved only by using numerical methods.

4.3.1 Mixed-F low Model

The following assumptions were made:

(1) complete mixing of all liquid and gas phases;

(2) constant temperature and pressure;

(3) constant liquid volume;

(4) first order self-decomposition of ozone in both gas and liquid phases;

(5) two-film model to describe the liquid side resistance (neglecting the gas phase

resistance) since ozone is sparingly soluble in water.

I Outlet Ozone

Headspace

 

Gas Phase Liquid Phase

 

 

Inlet Ozone

it
he

(lit

(2)
l

(3')

112

The mass balance of ozone in all liquid and gas phases can be expressed as:

 

 

 

dC ,
L _ * _ _ = _ _
Vx dt —KLaxVx(CL CL) kaxCL KLaxVx(an CL) kaxCL (4.1)
deg . , ,
H = x Cm—C —K xVx aC —C 4.2
0X dt Q ( g g ) La ( g L) ( )
VX d: =Qx(Cg -C;)—k’xV’xC; (4.3)

where the initial conditions are:

(1) t=0, CL(O) = o
(2) t=0, Cg'(0) = 0

(3) t=0, C;(0) = 0
and the terms are defined as:
V : volume of the liquid phase reactor (L)
K M : mass transfer coefficient in the liquid phase (1/min)
k : first order decomposition rate constant for dissolved ozone in the liquid phase
(l/min)
k’ : first order decomposition rate constant for moisturized gaseous ozone (llmin)
CL : dissolved ozone concentration in the liquid phase (mg/L)
CZ: saturated dissolved ozone concentration in the liquid phase which is equilibrium
with the gas phase (mg/L)
(1 : partition coefficient for ozone

H : gas hold up (total volume of bubbles) in the reactor (L)

V’ : volume of the reactor headspace (L)

II

If

lh

113

I

Cg : ozone concentration leaving the reactor or inlet ozone concentration to the

headspacc(mg/L)

C2" : inlet ozone concentration to the gas phase (bubbles) (mg/L)

C 8 : outlet ozone concentration leavrng the headspace (mg/L)

Q : ozone flow rate (Umin)

Taking the Laplace transformation of each the mass balance equations yields:

 

v x (56, (S) — CL (0)) = KuV(aC; (S) - CL (S)) - kvc“, (S) (4.4)
.. , c‘" .. .. .

Ho x (SC, (S) — C, (0)) = Q53 - QC, (S) - Kai/(aC, (S) — C. (S)) (4.5)

V’x (so; (S) — c; (0)) = Q(C‘, (S) — (E; (S)) — k’V’C‘; (S) (4.6)

According to the initial conditions:

aKuV

6‘ s =
‘( ) (VS+KUV+kV

 

"0 aK "1
C S = ——i"——- .
) g( ) (S+Ku+k)C‘(S) (47)

After substitution, we can write that:

 

 

M %XQXC:
Cg(S)= 2 (4.8)
aKLa V
HOS+Q+aKLaV————-
S+Km+k
(S +Ku +k)QC;" (4 9)

_ S[H032 +(H0KM +H0k+Q+aKmV)S+(QKLa +Qk+akKLaV)]

A, =H0Ku +Hok+Q+aKLaV

If lt
we 6 AzzQKu+Qk+akKLaV

then,

I._..-4 ..

114

.. (s +K,,, +k)Qc;';'

 

 

 

 

 

 

 

 

 

 

 

C 5 _ (4.10)
8” S(HOSZ+2,S+22)
= (s + Km + 1:)ch (4.11)
H 5(5—4" +fl2—4H02, )(S_—2, —,[2,2—4H,2,)
° 2n0 2H0
1,2 — 4110/12 = (HOKM + Hok — Q — aKLaV)2 + 4HoaKL02V > 0
C; (S)=14-+ B 2 + C 2 (4.12)
S S_-4+(/4 -0101, 53’1"”: 411042
2H0 21110

(K La + k)QC§"
A = 4.
QC;"(—2., + J23 - 4H0A2 + 2H0Ku + 2H0k)
2: - 4742 — moi. — 411.4.
C = ch'(—A, — J33 — 4110/1, + 2HOKLa + 2H0k)
2.2 + 4742 -4110, -4110),

 

 

 

 

where, B =

 

 

 

Taking the reverse Laplace transformation yields:

 

 

 

 

2 2
4‘44}; 4’1”)” t)+CxExp( 3‘ JA‘ 4H°32 t) (4.13)

C t=A+BxEx
8 0 p( 110 2H0

In the same way, the ozone concentration in liquid phase can be found as:

 

 

2 2
—2, +JA, 4H0V/12 t) + FX Exp( A, [71, 4H0V2

C t =D+ExEx
‘0 p( 2H0V 2H0V

’ t) (4.14)

 

 

115

aKuQCg'

D = ——
4.
zanoxuocg"

E = 2 2

’11 ’4H042 - 41E ’4H032
F _ zanoxuocg'

2.2 - 411,1, + 4742 -4110,

 

 

Then, the outlet ozone concentration from the second gas phase can also be obtained as:

 

-1, +JA2—4HOA, t)

 

C;(t) =W+XxExp[(- Q —k’)t]+YxExp(

 

 

 

 

 

 

 

 

 

 

 

 

W 2H0
—2, —,/2,2 —4H,2,
+ Z X Exp( 2H0 t) (4.15)
K k Ci"
W = ( La + )3] 8
1 k’ —
M + ( Q ))
g,(--Q-, — k’ + KL, + k)QCf,"
X = Q V V Q Q
(-7 - k')[Ho(-‘—/7 -' k')2 + M—V -' k') + 42]
Q(—21 + Jlllz —4H0/l2 + 2H0KLa + 2kHo)(%)C;"
Y =
— + 2 -4H
("An/412 '4H0/b +212 ‘4H032X A JZAH 032 +%+k’)
0
Q(—A1 — ([2,: —4H0/l2 + 2H0Ku + 2kH0)(%)C§"
z =
— — 2 —
(4W 411.4. +2.2 ~4Hotx " ‘b‘ “7°" +—‘%+k')
2H0 v
4.3.2 Plug-Flow Model

The following assumptions were made:

fi—d

116

(1) the liquid in the reactor and the gas in headspace are both completely mixed. The flux
of gas in the reactor can be described as plug flow;

(2) constant temperature and pressure;

(3) constant liquid volume;

(4) first order self-decomposition of ozone at both gas and liquid phase;

(5) two-film model to describe the liquid side resistance (neglecting the gas phase

resistance) as ozone is sparingly in water.

4

Outlet Ozone

 

 

//
00f Headspace

 

Gas Phase
(bubbles)

Liquid Phase

 

 

 

Inlet Ozone

 

 

The plug flow region in portion of the reactor containing the gas phase (bubbles) can be

divided into 10 small sections, each of which is assumed to be completely mixed.

Mass balance equations of ozone at all liquid and gas phases:

10x dX(t)
dt

 

= K“ (ax(Y, (t)+Y (t)+1’3(t)+Y4 (t)+YS (t)+Y6 (t)+Y7 (t)+Y8 (t)+Y9(t)
+Y,o(t))-10XX(t))—10xk, xX(r) (4.16)

H0 xd—g;(—€-)-=10xQx(C;" —Y,(t))—k2 xH0 xY,(t)—KLa xVx(axY,(t)—X(t)) (4.17)

117

dY2 (t)

 

 

 

 

 

 

 

 

Hox =10xQx(Y,(t)—Y2(t))—k2xHoxY,(t)—K,AxVx(axY2(t)—X(t)) (4.18)
(113(1)

Hox =10xQx(Y,(t)—Y3(t))—k2xHoxY,(t)—KmxV><(axY3(t)—X(t)) (4.19)
dY(t)

Hox “ =leQ><(Y3(t)—Y,(t))-k2xHoxY,(t)—KmxV><(axY,(t)—X(t)) (4.20)
dY(t)

Hox 5 =10xQ><(Y,(t)—Y5(t))—k2xHoxY5(t)—KuxVx(axY5(t)-X(t)) (4.21)
dY.(t)

Hox =10xQx(Y5(t)—Y6(t))—k2xHoxY6(t)—KuxVx(axY6(t)—X(t)) (4.22)
dY7(t)

Hox r =10xQx(Y,(t)—Y7(t))—k2xHoxY,(t)—KmxVx(axY,(t)—X(t)) (4.23)
dYsU)

H0 x = 10xQ><(Y7 (1)—Y, (t))—k2 xH0 xY,(t)—K,a xVx(axY,(t)—X(t)) (4.24)
dY9(t)

Hox d =1OxQx(Y,(r)-Y9(t))—k2xHoxY9(t)—K,axVx(axY9(t)—X(t)) (4.25)
dY10(t)

HOXT=lonx(x.)(t)-Yro(t))-k2XHoinoa)-KiaXVX(aXY10(t)—X(t)) (4-26)

Vl x? = Qx(Yw(r) —Z(t))—k2 xv; XZ(t) (4.27)

The initial conditions are:

X(O) = 0, Y1 (0) = 0, Y, (0) = 0, Y3 (0) = 0, Y, (0) = 0, Y, (0) = 0, Y, (0) = 0, Y, (0) = 0, Y, (0) = 0,
Y, (0) = 0, Y,, (0) = 0, 2(0) = 0

The terms used in the above equations are defined as:
X(t): dissolved ozone concentration in the liquid phase;

Yj(t): gaseous ozone concentration in the jth section of the bubble phase, j is the number

of the section;

Z(t): gaseous ozone concentration in the headspace.

sofiv

44d

4:1

obs
unh
tha
rea
an
an(
on
du
dn
efi
Us

an

CC

ta

31

118

This system of equations was solved numerically by using the mathematical computer

software, MAPLE V (Release 4.0, Waterloo Maple Inc., Waterloo, Ontario)

4.4 Results and Discussion
4.4.] Self-decomposition of Ozone in Water and Headspace

The order and rate of the decomposition of dissolved ozone were determined by
observing the reduction in the ozone concentration under different temperatures in
unbuffered DDI water and buffered DDI water at pH of 6.7. Results (Table 4.1) indicate
that the rate of ozone decomposition in unbuffered DDI water occurs via a first order
reaction (Figure 4.3), while in buffered DDI water (pH=6.7) ozone decomposition
follows second order reaction (Figure 4.4). According to the results of many researchers
and summarized by Sugimitsu et a1. (1989), the decomposition of ozone has a reaction
order of 1, 1.5, or 2. Gurol and Singer (1982) concluded that the major differences are
due to (a) the use of different analytical techniques to measure the concentration of
dissolved ozone, (b) uncertainties in the data analysis and data interpretation, (c) the
effect of solution composition, ((1) the possible presence of impurities in the reagents
used. Therefore, in our studies, the discrepancy in the reaction order for the unbuffered
and buffered DDI water can be explained as the effect of solution composition.

The rate of ozone decomposition was also determined at pH 2.1. At this pH, the
concentration of dissolved ozone decreased quite slowly. As such, the decomposition
rate can be said to be negligible compared with that obtained under higher pH conditions.
This is consistent with the observation of other researchers (Gurol and Singer, 1982;

Sugimitsu et al., 1989). Increasing the temperature is also expected to increase the rate of

119

self-decomposition of dissolved ozone in water. The enhancement of self-decomposition
of ozone in water can be demonstrated by the relationship derived from the Arrhenius

equation (Kuo et al., 1976, Ghaieb and Bes, 1989):

k, = k0[0H' 1'" exp(—E/RT) (4.28)

where kd is the decomposition rate constant, k0 is the frequency factor, [OH'] is the
concentration of hydroxide ion (mole/L), m is the reaction order, E is the activation
energy (kcal/mole), R is the gas constant (kcal/K-mole), and T is temperature (K).
The effect of temperature (over the range from 14 °C to 30 °C) on the decomposition rate
of ozone was more significant for the buffered DDI water. The decomposition rate
constant increased by ca. 140%. On the other hand, over a temperature ranged from 14
°C to 30 °C the ozone decomposition rate constant in unbuffered DDI water increased by
only ca. 10%. Using equation 4.25, the activated energies were calculated to be 1.4 and
9.6 Kcal/mole, respectively, for the unbuffered and buffered DDI water. This result
implies that the decomposition rate constant of ozone in buffered DDI water is much
more sensitive to the change in temperatures than it is in unbuffered DDI water.

The rate of decomposition of ozone in the moisturized gas in the headspace of the
reactor was estimated (Figure 4.5). The concentrations of ozone at the entrance and exit
of the headspace were unchanged at different flowrates. Therefore, the self-

decomposition of moisturized gaseous ozone can be neglected.

120

Table 4.1 Self-decomposition of Dissolved Ozone in DDI Water under Different
Temperature

 

 

 

 

 

 

 

Mathematical Expression of Ozone Decomposition : — d[O3 ] / dt = k d [03 ]"'
DDI without buffer DDI buffered at pH=6.7
Temperature (°C) m kd m Rd
14 1 0.0201 2 0.00284
22 1 0.0215 2 0.00402
30 1 0.0229 2 0.00691

 

 

 

 

 

 

4.4.2 Transfer of Ozone in DDI Water (without pH buffer)

The mass transfer of ozone in DDI water was determined by changing the various
concentrations of inlet ozone, system temperature, and ozone flow rate. To measure the
mass-transfer coefficient, Ka and the partition coefficient, 01, the experimentally
determined concentrations of outlet ozone and dissolved ozone were used in the two
models. In the first model, it was assumed that the gas phase was completely mixed,
while in the second phase it was assumed that the movement of gas bubbles in the gas
phase followed plug-flow conditions. The logarithmic concentration of ozone in the
bubbles (Beltran et al., 1997) was used as the average driving force in plug-flow model.
To validate these models, the concentration of inlet ozone was varied under the
operational conditions of 14 °C and 400 mIJmin flowrate. This data was used to
calculate the mass-transfer and partition coefficients for ozone (Table 4.2). These results
indicate that, by using these two parameters, the transfer of ozone can be modeled (Figure
4.6). For most experiments, the plug-flow model better simulated the data than did the
mixed-flow model. This result supports the conclusion proposed by Nishikawa et al.

(1981) that the movement of ozone bubbles in a semi-batch reactor is best represented by

121

a plug-flow model. Although the concentrations of inlet ozone differed, by an order of
magnitude, there is little variation in the fitted values of the mass transfer and the

partition coefficient.

Table 4.2 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Inlet Ozone Concentrations

 

 

 

 

 

 

 

 

 

14°C, Q=400 mL/min Mixed-flow Model Plug-flow Model
Cone. of Inlet Ozone 01 KLa R7 0t KLa RT
(mg/L) (min") (min")
5.41 0.332 0.2245 0.961 0.339 0.1779 0.932
22.37 0.343 0.2249 0.969 0.344 0. 1763 0.979
44.29 0.338 0.2236 0.975 0.342 0.1728 0.988
62.37 0.348 0.2273 0.962 0.353 0.1774 0.990

 

 

 

 

 

 

The effect of temperature on mass-transfer and partition coefficients of ozone
indicates that as the temperature increases, the mass-transfer coefficient increases and the
partition coefficient decreases (Table 4.3). Adams et al. (1981) indicated that
temperature is one of the most significant factors to affect the mass transfer coefficient of
oxygen, with relationship as below:

9 (T-20)

KLam = KLam.) (4.29)

The temperature factor, 0, for oxygen has been reported to vary from 1.016 to 1.037
(Adams et al., 1981). Our experimental results for ozone mass-transfer also show that a
similar relationship may exist. Using the mass-transfer coefficient for ozone that was
derived from the plug-flow model, the 0 value can be estimated to be 1.031 (R2=0.96).

As such, the effect of temperature on the mass-transfer coefficient of ozone is similar to

122

that for oxygen. The partition coefficient for ozone was reported to be 0.34, 0.24, and
0.17 for temperatures of 12 °C, 21 °C, and 31 °C, respectively (Caprio et al., 1981). Thus,

the partition coefficients we obtained agree satisfactorily with those of Caprio et al.

(1981).

Table 4.3 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Temperatures in DDI Water

 

 

 

 

 

 

 

 

 

C=46mg/L, Q=400mL/min Mixed-flow Model Plug-flow Model
Temperature (°C) or KLa R2 0t KLa R2
(min'l) (min!)
14 0.355 0.2152 0.957 0.353 0.1723 0.989
22 0.253 0.2851 0.964 0.249 0.2397 0.970
30 0.178 0.3551 0.971 0.176 0.2800 0.971

 

 

 

 

 

The effect of flowrate on the mass-transfer coefficient in a semi-batch stirred reactor
was also investigated (Table 4.4). The mass-transfer coefficient of ozone was found
increase with increasing the gas flowrate. Yocum (1980) indicated that the effect of an
increase in the gas feed rate was found to be similar to that of increasing the turbine
speed. This was thought to be due to an increase in the absorption rate of ozone, because
of the larger interfacial area created at the higher gas feed rates. The mathematical
relationship between mass-transfer coefficient and superficial velocity, Ug (flowrate
divided by section area of reactor), of oxygen (Figueiredo and Calserbank, 1979 and
Nishikawa et al., 1981) and ozone (Yocum, 1980, Anselmi et al., 1984, Roustan et al.,

1987, Stankovic, 1988, and Laplanche et al., 1991) is a power law (KLa= aUgb). From the

data we obtained at 14 °C, the constants, a and b, were determined by plotting the

 

relat

scali

can

wit

me

 

fifi—I‘fi 7— T

123

relationship between mass-transfer coefficient and superficial velocity on a logarithmic
scale. As shown in Figure 4.7, the equation:

KLa = 0.091 118“” (4.30)

can be used to describe this relationship. The value of b obtained for our system falls
within the values of 0.24 to 0.82 that have been determined by those investigators
mentioned above. The variations in the values for b predominately depend on the
geometry of the reactor, the range of superficial velocity covered, the experimental
method used to measure the concentration of dissolved ozone, and the composition of

solution used.

Table 4.4 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Flowrates in DDI Water

 

 

 

 

 

 

 

 

T=14°C, C=46 mg/L Mixed-flow Model Plug-flow Model
Inlet Flowrate 0t KLa R2 (1 Km RT
(mL/min) (min'l) (min")
100 0.358 0.1717 0.909 0.367 0.0947 0.977
200 0.340 0.1647 0.939 0.326 0.1358 0.990
300 0.355 0.1727 0.982 0.349 0.1472 0.989
400 0.355 0.2152 0.957 0.353 0.1723 0.989
500 0.353 0.2582 0.963 0.369 0.1861 0.988

 

 

 

 

 

 

 

4.4.3 Transfer of Ozone in DDI Water (pH =2 acidified by sulfuric acid)

To prevent the self-decomposition of ozone, the experiments to determine the effect

of ionic strength on mass transfer and partition coefficients were conducted at a pH of

2.1. This solution was prepared by adding sulfuric acid to adjust the pH. An appropriate

amount of sodium sulfate was added to obtain the desired ionic strength. As shown in

 

124

Table 4.5, the partition coefficients were not affected by the ionic strength of the water.
On the contrary, ionic strength was found to have a significant effect on the mass-transfer
coefficient. As the ionic strength increased, the size of the bubbles emitted from the
diffuser was observed to decrease significantly. This decrease in size may be due to the
increase in surface tension as the ionic strength of the solution is increased. Therefore,

ionic strength is an important criterion in any ozonation system design.

Table 4.5 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Ionic Strength in DDI Water (Acidified to pH=2)

 

 

 

 

 

 

22 °C, Q=400 mIJmin Mixed-flow Model Plug-flow Model
Ionic Strength (mole/L) (3; Km (min'l) R2 a KLa (min'l) R2
0.1 0.208 0.4020 0.968 0.21 1 0.3033 0.994
0.2 0.214 1.0430 0.979 0.211 0.5701 0.997
0.3 0.210 2.2353 0.936 0.208 0.8910 0.983
0.4 0.21 1 2.2620 0.660 0.206 0.9805 0.955

 

 

 

 

 

 

 

 

 

The effect of temperature and flowrate on the mass-transfer and partition
coefficients was studied using the acidified water (Table 4.6). Our results show that the
plug-flow model yields a higher correlation for the fitted coefficients than did the mixed-
flow model. The conclusion is consistent with that found in unbuffered DDI water.
Within a temperature range of 14 - 30 °C, the 0 value ranged from 1.020 to 1.025 for all
five flowrates. Therefore, the effect of temperature on the mass-transfer coefficients is
identical to that found in DDI water at the flowrate range of 100 — 500 mIJmin. The
power law was again used to examine the effect of flowrate on the mass-transfer

coefficient under different temperatures. The relationships determined are given below:

125

T =14°C, KLa =0.115(U, )"-“83
T = 22 °C, KLa = 0.147 (U, W28 (4.31)
T = 30 °C, KLa = 0.170 (U, )°-‘”

The values of b in pH 2.1 water fall within the range of 0.42 to 0.49. As a comparison,

the value obtained in unbuffered DDI water was 0.41.

4.4.4 Transfer of Ozone in DDI Water (pH =6. 7 buffered by phosphate salt)

The transport of ozone in water (at circumneutral pH) was investigated. At
circumneutral pH, the effect of self-decomposition of ozone is more significant than that
observed at lower pH values. By using the mass-transfer model developed previously,
the effect of temperature and flowrate on the mass-transfer and partition coefficients of
ozone is shown in Table 4.7. The ionic strength of the solution was fixed at 0.1 M. The
results show that the mass-transfer model of ozone remains applicable for this system and
the plug-flow model is better than mixed-flow model for fitting the system parameters,
although the correlation factors were lower than those obtained at pH 2.1. This may due
to the errors associated with measuring dissolved ozone at higher pH as a result of the
rapid decomposition of dissolved ozone. This also helps explain the greater variation in
the partition coefficients. The temperature factor, 0, which ranged from 1.026 to
1.036, for different flowrates was also found to vary more at high pH than observed at

lower pH. The power laws found for the different temperatures are given below:

T = 14 °C, KLa = 0.103 (U, )°-‘2°
T = 22 °C, KLa = 0.139 (U, )°-5" (4.32)
T = 30°C, KLa =0.161(U1g )“55

 

Emir-~_

LENers

\LN\\\\

 

126

The values of b, apparently, are larger than those obtained at pH 2.1. It also implies that
at circumneutral pH, the effect of superficial velocity on mass-transfer coefficients is

greater than observed in acidic pH water.

Table 4.6 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Temperatures and Flowrates in DDI Water (Acidified to pH=2)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T=14 6C Mixed-flow Model Plug-flow Model

Inlet Flowrate (ml/min) 0t KLa (min'l) R2 a KLa(min'1) R2
100 0.278 0.2386 0.822 0.278 0.1241 0.954
200 0.274 0.2506 0.966 0.266 0.1721 0.997
300 0.272 0.2949 0.967 0.271 0.2090 0.985
400 0.284 0.3239 0.968 0.282 0.2424 0.995
500 0.293 0.3589 0.951 0.291 0.2698 0.992

T=22 °C Mixed-flow Model Plug-flow Model

Inlet Flowrate (ml/min) a KLa (min’l) R7 a KLa (min'l) R2
100 0.224 0.2864 0.933 0.219 0.1574 0.975
200 0.215 0.3142 0.977 0.209 0.2122 0.962
300 0.224 0.3379 0.971 0.222 0.2357 0.979
400 0.232 0.3873 0.963 0.230 0.2843 0.992
500 0.238 0.4352 0.974 0.238 0.3188 0.996

T=30 °C Mixed-flow Model Plug-flow Model

Inlet Flowrate (ml/min) 0t KLa (min'l) R2 at KLa(min'1) RT
100 0.160 0.3554 0.818 0.158 0.1853 0.928
200 0.156 0.3722 0.985 0.156 0.2456 0.983
300 0.159 0.4102 0.961 0.159 0.2876 0.990
400 0.164 0.5073 0.979 0.164 0.3525 0.983
500 0.166 0.5080 0.978 0.164 0.3862 0.986

 

 

 

Tab

 

mi: : m::: REE:

 

127

Table 4.7 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Temperatures and Flowrates in DDI Water (pH=6.7 Buffered by

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Phosphate System)
T=14 6C Mixed-flow Model Plug-flow Model
Inlet Flowrate (mIJmin) 0t KLa (min'T) R2 at KLa(min'1) R2
100 0.305 0.2385 0.935 0.326 0.1 125 0.986
200 0.292 0.3094 0.948 0.310 0.1735 0.988
300 0.304 0.3515 0.922 0.313 0.2268 0.986
400 0.303 0.4027 0.921 0.307 0.2771 0.971
500 0.332 0.4257 0.876 0.332 0.3027 0.967
T=22 °C Mixed-flow Model Plug-flow Model
Inlet Flowrate (mIJmin) 0t KLa (min'I) R2 at KLa (min'I) R2
100 0.225 0.3029 0.743 0.232 0.1479 0.927
200 0.209 0.3899 0.980 0.210 0.2405 0.981
300 0.220 0.3673 0.958 0.222 0.2719 0.971
400 0.237 0.41 16 0.884 0.233 0.3328 0.904
500 0.238 0.5618 0.967 0.240 0.3866 0.986
T=30 °C Mixed-flow Model Plug-flow Model
Inlet Flowrate (mIJmin) a KLa (min'l) R7 a KLa(n1in") R2
100 0.168 0.3090 0.574 0.166 0.1709 0.841
200 0.158 0.4681 0.971 0.159 0.2808 0.965
300 0.158 0.6301 0.957 0.158 0.3982 0.956
400 0. 172 0.5890 0.942 0.170 0.4318 0.966
500 0.173 0.7200 0.981 0.171 0.4798 0.987

 

 

 

 

 

 

 

liql

dc

fa.

128

The enhancement factor, E , which is due to a chemical reaction taking place in the
liquid phase (Danckwerts, 1970), is defined as KL(with reaction)/KL(without reaction).
In the absence of other reactive species at circumneutral or alkaline pH, the self-
decomposition of ozone is the major reaction involving ozone. As such, the enhancement
factor can be determined by dividing the mass-transfer coefficient found in water
buffered at a pH of 6.7 by the mass-transfer coefficient in water having a pH of 2.1. The
enhancement factors determined for the same ionic strength and temperature of the water
are tabulated in Table 4.8. Except for the enhancement factors obtained for a flowrate of
100 mIJmin, all enhancement factors were found to be slightly greater than 1. It is
thought that the mass-transfer coefficients obtained at this lower flowrate are less reliable
as the correlation factors obtained in fitting the mass-transfer model were poorer than
those obtained for higher flowrate. These results indicate that the enhancement of mass-
transfer coefficient by self-decomposition of ozone over this pH range is insignificant as
compared with the enhancement obtained for the reactions of some organic compounds
(Augugliaro and Lizzuti, 1978). Augugliaro and Lizzuti (1978) found that the

enhancement factor was greater than 3 when ozonating wastewater containing high

concentrations of phenol.

Table 4.8 Enhancement Factors for the DDI Water (pH=6.7) Buffered with Phosphate
Solution

 

Flowrate (mIJmin)

 

 

 

Temperature 100 200 300 400 500
14°C 0.91 1.01 1.09 1.14 1.12
22°C 0.94 1.13 1.15 1.17 1.21

 

 

30°C 0.92 1.14 1.38 1.22 1.24

 

 

 

 

 

 

 

 

4.4.5

1171118

mixc

and

111.111

and

mo

flo

s11

00

129

4.4.5 Comparison of Experimental Data and Model Prediction

In this section, the simulations obtained using mixed-flow and plug-flow mass-
transfer models are compared with the experimental data. It should be noted that the
mixed-flow model can be solved by either analytical or numerical methods. Figures 4.8
and 4.9 show the consistency of the solutions obtained using both the analytical and
numerical methods. The concentration of ozone in the headspace (outlet concentration)
and the experimentally determined dissolved ozone concentration are compared to the
model prediction. To minimize the space requirements, this comparison is made only for
flowrate and temperature of 400 mIJmin and 22 °C (Figures 4.10 - 4.15), although
similar results are obtained for the other flowrates and temperatures. The results indicate
that both the concentrations of outlet and dissolved ozone predicted by either mixed-flow
model or plug-flow model are slightly greater than the values observed experimentally.
These variations may be due to the experimental errors, or some optimistic assumptions
such as complete mixing in water and flow pattern of gas bubbles. The prediction
obtained using the plug-flow model was close to the experimental data. Gas holdup,
measured by monitoring the increase in height of water level while gas was sparged into
water, was employed in computing the concentration of ozone in the mass-transfer
models. Gas holdup was estimated to be 0.01425 L when the flowrate was 400 mIJmin.
The difference in the behavior of mixed-flow or plug-flow pattern for the gas bubbles

might be insignificant due to such a small volume of gas bubbles in the reactor.

130

4.4.6 Sensitivity Analysis of Mass-Transfer Model of Ozone

Sensitivity analysis is usually used to determine the uncertainty of each parameter.
The sensitivity of the mass-transfer model to each parameter was tested by changing each
parameter over a range (i10 — 20%) of its value while the other parameters were held
constant at their base values. The effects of mass-transfer coefficient, partition
coefficient, self-decomposition coefficient, and gas holdup were determined on the
profiles of dissolved ozone and outlet ozone concentrations (Figures 4.16-4.19). It is
found that, of all the parameters tested, the partition coefficient (0t) had the greatest effect
on the dissolved ozone concentration. Following this, was the mass-transfer coefficient.
However, the profile of the concentration of outlet ozone gas appeared to be insensitive
to all parameters tested. Any change in the self-decomposition coefficient or the gas
holdup did not have any effect on either concentration. Therefore, it is very important to
accurately determine the value of the partition coefficient when using the ozone mass-

transfer model to predict the concentrations of outlet ozone gas and dissolved ozone.

4.4. 7Application of Mass-transfer Model in a Pilot-scale System

Based on the successful use of mass-transfer model (plug-flow model) in a lab-scale
reactor, this model was used to demonstrate its applicability for a pilot-scale reactor,
which had been operated to treat the stored swine manure (see Chapter 3). The effect of

temperature and agitation speed on the partition and mass-transfer coefficients for ozone

in DDI water was investigated (see Table 4.9):

131

Table 4.9 Fitting Parameters Obtained from the Mass-transfer Equations. Effect of
Changes in Temperatures and Agitation Speeds in DDI Water in a Pilot-scale

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Reactor

Q=443 mL/min, 200 rpm 0t KLa R2
T=15 °c 0.356 0.3671 0.971
=20 °c 0.268 0.4327 0.980
T=30 cc 0.213 0.4764 0.957

Q=443 mIJmin, T=20 °c a Kta R2
0 rpm 0.257 0.3590 0.988
200 rpm 0.268 0.4269 0.981
400 rpm 0.268 0.4327 0.980
600 rpm 0.278 0.4795 0.938

 

Similar effects of temperature on the mass-transfer and partition coefficient were
observed in both the lab-scale and pilot-scale reactors (Table 4.9). An increase in
temperature resulted in a decrease in the partition coefficient and an increase in the mass-
transfer coefficient. Increases in agitation speed, while keeping other conditions
constant, effectively enhanced the mass-transfer coefficient. It is thought that the larger
the agitation intensity the bigger the contact area between gas bubbles and liquid. This
would also increase the detention time of the gas bubbles in the reactor. In figures 4.20
and 4.21, the concentrations of dissolved ozone and outlet ozone obtained from
experimental data and model predictions are compared. It appears that the plug-flow

model can also be used to describe ozone mass transfer in the pilot-scale system.

4.5 Conclusions

From the results presented in this chapter, several conclusions can be drawn.

132

The self-decomposition of ozone in unbuffered DDI water follows a first order kinetic
model. However, ozone decomposes by a second-order reaction in DDI water
buffered with phosphate at pH of 6.7. Increasing the temperature of the solution
accelerated the self-decomposition of aqueous ozone. However, the rate constant for
the self-decomposition of moisturized gaseous ozone was found to be zero.

A mass-transfer model using plug-flow in the bubble phase was successfully used in
the semi-batch laboratory scale reactor as well as in the pilot-scale reactor. Although
other investigators have used the mixed-flow model more often than plug-flow model
for the analysis of mass-transfer of ozone, our data show that use of plug-flow model
yields higher correlation factors for mass-transfer and partition coefficients of ozone.
The mass-transfer coefficient increased with increasing flowrate, temperature, and
ionic strength. The effect of flowrate on the mass-transfer coefficient can be
described by using a power law. The effect of temperature can be corrected by a
temperature coefficient (0), which was estimated to fall within the same range as that
observed for oxygen. In addition, it was noted that an increase in ionic strength
resulted in smaller ozone bubbles. Therefore, while maximizing the mass-transfer
efficiency of an ozone reactor, the chemical composition of liquid should be taken
into consideration in the design of the reactor.

Of the factors tested, partition coefficient was found to be the only parameter that
varied both with temperature and aqueous composition. However, temperature is still
the major factor to affect the partition coefficient.

Any chemical reaction in the ozonation system including ozone self-decomposition

and its reaction with organic and inorganic compounds will enhance the mass-transfer

133

coefficient of ozone. The enhancement factor for the ozone self-decomposition was
observed to fall within the range of 0.9-1.4 (pH of 6.7 and flowrate ranging from 100
to 500 mIJmin. Therefore, the enhancement factor caused by ozone self-
decomposition at this pH tested is insignificant.

Using a sensitivity analysis, the partition coefficient (a) was found to be the most
sensitive factor that affected the concentration profile of dissolved ozone. The mass-
transfer coefficient had a minor effect on the dissolved ozone. The concentration of
dissolved ozone was unaffected by the self-decomposition rate constant of ozone and

gas-holdup. For the concentration profile of outlet ozone, all of these four parameters

are insensitive.

4.6 References

1.

CE. Adams, Jr.; D.L. Ford; W.W. Eckenfelder, Jr., “Development of Design and
Operational Criteria for Wastewater Treatment”, CB1 Publishing Co., Boston, MA
(1981).

G. Anselmi; P.G. Lignola; C. Raitano; G. Volpicelli, “Ozone Mass Transfer in Stirred
Vessel”, Ozone: Sci. & Eng., 6: 17-28 (1984).

V. Augugliaro; L. Rizzuti. “The pH Dependence of the Ozone Absorption Kinetics in
Aqueous Phenol Solutions”, Chem. Eng. Sci, 33: 1441-1447 (1978).

F]. Beltran; J .F. Garcia-Araya; J .M. Encinar, “Henry and Mass Transfer Coefficients
in the Ozonation of Wastewater”, Ozone: Sci. & Eng., 19:281-296 (1997).

V. Caprio; A. Insola; P.G. Lignola; G. Volpicelli, “A New Attempt for the Evaluation

of the Absorption Constant of Ozone in Water”, Chem. Eng. Sci, 37: 122-124 (1981)

134

6. RV. Danckwerts, “Gas-Liquid Reactions”, McGraw-Hill Book Company, New York
(1972).

7. J. Ghaieb; R.S. Bes, “Self Decomposition and Ozone Mass Transfer in a Laboratory
stirred Reactor”, in Proceedings of Ninth Ozone World Congress, New York, Vol.
2:504-512 (1989).

8. MD. Gurol; P.C. Singer, “Kinetics of Ozone Decomposition: A Dynamic Approach”,
Environ. Sci. Technol., 16:377-383 (1982).

9. J. Hoigné; H. Bader, “Rate Constants of Reactions of Ozone with Organic and
Inorganic Compounds in Water-I (Non-dissociating Organic Compounds)”, Wat.
Res., 17:173-183 (1983).

10. J. Hoigné; H. Bader, “Rate Constants of Reactions of Ozone with Organic and

Inorganic Compounds in Water-II (Dissociating Organic Compounds)”, Wat. Res.,

17:185-194 (1983).

11. J. Hoigné; H. Bader; W.R. Haag; J. Staehelin, “Rate Constants of Reactions of Ozone
with Organic and Inorganic Compounds in Water-HI (Inorganic Compounds and
Radicals)”, Wat. Res., 19:993-1004 (1985).

12. CH. Kuo; K.Y. Li; C.P. Wen; J .L. Weeks, Jr., “Absorption and Decomposition of
Ozone in Aqueous Solutions”, AIChE Symp., 73:230-241 (1977).

13. A. Laplanche; N. Le Sauze; G. Martin; B. Langlais, “Simulation of Ozone Transfer in
Water. Comparison with a Pilot Unit”, Ozone: Sci. & Eng., 13:535-558 (1991)

14. M.M. Lopes De Figueiredo; P.H. Calderbank, “The Scale-up of Aerated Mixing

Vessels for Specified Oxygen Dissolution Rates”, Chem. Eng. Sci, 34: 1333-1338
(1979).

135

15. M. Nishikawa; M, Nakamura; H. Yagi; K. Hashimoto, “Gas Absorption in Aerated
Mixing Vessels”, Journal of Chemical Engineering of Japan, 14:219-226 (1981).

16. L.H. Nowell; J. Hoigné, “Interaction of Iron (H) and Other Transition Metals with
Aqueous Ozone”, in Proceedings of 8th Ozone World Congress, New York, Vol. 2:
E80-E95 (1988).

17. J .A. Roth; D.E. Sullivan, “Solubility of Ozone in Water”, Ind. Eng. Chem. F undam,
20:137-140 (1981).

18. M. Roustan; J.P. Duguet; B. Brette; E. Brodard; J. Mallevialle, “Mass Balance
Analysis of Ozone in Conventional Bubble Contactors”, Ozone: Sci. & Eng., 9:289-
297 (1987).

19. M. Roustan; R.T. Wang; D. Wolbert, “Modeling Hydrodynamics and Mass Transfer
Parameters in 3 Continuous Ozone Bubble Column”, Ozone: Sci. & Eng., 18:99-115
(1996).

20. S. Sheffer; G.L. Esterson, “Mass Transfer and Reaction Kinetics in the Ozone Tap
Water System”, Wat. Res., 16:383-389 (1982)

21. J .L. Sotelo; F.J. Beltran; F.J. Benitez; J. Beltran-Heredia, “Ozone Decomposition in
Water: Kinetic Study”, Ind. Eng. Chem. Res., 26:39-43 (1987).

22. J.L. Sotelo; F.J. Beltran; F.J. Benitez; J. Beltran-Heredia, “Henry’s Law Constant for
the Ozone-Water System”, Wat. Res., 23: 1239-1246 (1989).

23. I. Stankovic, “Comparison of Ozone and Oxygen Mass Transfer in a Laboratory and

Pilot Plant Operation”, Ozone: Sci. & Eng., 10:321-338 (1988).

136

24. H. Sugimitsu; S. Okazaki; T. Moriwaki, “Decomposition of Ozone in Aqueous

Solutions”, in Proceedings of Ninth Ozone World Congress, New York, Vol. 1:97-
105 (1989).

25. F. H. Yocum, “Ozone Mass Transfer in Stirred Vessel”, AIChE Symp. Ser., 76: 135-
141 (1979).

137

 

[EU]
0
O

A
I.

 

 

 

 

 

 

(I

 

   

 

 

1: Oxygen Cylinder

2: Ozone Generator

3: Cooling System

4: Flow Controller

5: Three-way Valve

6: UV Spectrophotometer
7: Three-way Valve

8: Semi-batch Reactor

9: Trap Solution (KI)

Figure 4.1 The System Setup for the Semi-batch Ozonation Process

138

 

Figure 4.2 The Photo of the System Setup for the Semi-batch Ozonation Process

139

First Order Decomposition of Dissolved Ozone in DDI Water (14 C)

 

  

 

 

 

 

 
    
 

 

 

 

 

20 .
g 15 y = 13067600201"
2 ..

:5; 10 R — 0.9856
5 5

0 T F I I l I I *I‘

0 20 40 60 80 100 120 140
Time (min)
First Order Decomposition of Dissolved Ozone in DDI Water (22C)
12

A 10 3
§, 3 i y = 9.9199e°°~‘5"
E 6 - R2 = 0.9993
6
s 4 -
U 2 a

0 l 4 l t i t

0 20 40 60 80 100 120
Time (min)
First Order Decomposition of Dissolved Ozone in DDI Water (30 C)
10 4

y = 7.935e°°°”"
R2 = 0.9814

    

Conc. (mg/L)
as

 

 

 

4 1

2 a

0 l U I T I
0 20 40 6O 80 100

Time (min)

Figure 4.3 First-order Decomposition of Dissolved Ozone in DDI Water

 

140

Second Order Decomposition of Ozone in Liquid
(pH=6.63, I=0.1M, T=l4 C))

 

 
    

 

 

 

 

 
 
    

 

 

 

0.3
0.25 - j
y = 0.002836x _

8 0.2 - 2
x R = 0.977214
5 0.15
2 0.1

0.05

0 I I
Time (min)
Second Order Decomposition of Ozone in Liquid

0 5 (DI-1:663, I=0.1M, T=22 C)

0.4 a y = 0.004015x
5, 0.3 - R2 = 0.997032
1.!
9 0.2

0.1

0 I T I
0 2° 4° 60 80 100
Time (min)

Second Order Decomposition of Ozone in Liquid
(pH=6.63, I=0.1M, T=30 C)

 

y = 0.006910x
R2 = 0.999783

 
   

1/C-1/C0
O
.a.

 

 

 

Time (min)

Figure 4.4 Second-order Decomposition of Dissolved Ozone in DDI Water
(pH=6.7)

141

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1: Oxygen Cylinder

2: Ozone Generator

3: Cooling System

4: Flow Controller

5: Three-way Valve

6: Gas Washing Bottle (To moisture the gaseous ozone before entering into headspace)
7: UV Spectrophotometer

8: Cone Beaker (2.0 L to simulate the headspace in a semi-batch reactor)

9: Trap Solution (KI)

Figure 4.5 System Setup for Determining the Self-decomposition rate of Moisturized
Gaseous Ozone

142

 

1.8 ~
1.6 a
1.4 3
1.2 ~

0.8 a
0.6 a
0.4 .4
0.2 e

0 r r l

Logarithmic Driving Force

 

 

 

 

0 3 6 9
(dlosl/tlt)+ktl[osr"

Figure 4.6 Method to Fit the Mass-transfer and Partition Coefficients by
Linear Regression

 

-1.5

-1.6 -
-1_7 s y = 0.411x - 2.3947

2
=0. 8 7
-1.8n R 93

-1.9 ~

Ln(KLa)

 

 

'2.4 I F I

 

 

0 0.5 1 1.5
ang)

Figure 4.7 Relationship between Mass-transfer Coefficient of Ozone and
Superficial Velocity

143

 

 

 

 

 

12
5‘, 10 —
5
0
=1 3 ..
S
o
E 6—
O
.2
5‘5 4 * +Analytical Solution
8’
5 2 a +Numerica1 Solution
0 T I l l r
0 5 10 15 20 25 30

Time (min)

Figure 4.8 Comparison of Dissolved Ozone Concentration Calculated by
Analytical and Numerical Solutions in Mixed-flow Model for DDI
Water at Flowrate 400 mL/min and Temperature 22 C

 

45l

 

20 _‘ +Analytlcal SOlution

15 a + Numerical Solution

 

 

 

0 T l f I I

0 5 10 15 20 25 30

Cone. of Gaseous Ozone in Headspace (mg/L)

Time (min)

Figure 4.9 Comparison of Exit Ozone Concentration Calculated by Analytical

and Numerical Solutions in Mixed-flow Model for DDI Water
at Flowrate 400 mL/min and Temperature 22 C

144

 

12.00

10.00 ~

8.00

1

6.00 4

  

+Prediction by Mixed-flow Model

   

Cone. of Dissolved Ozone (mg/L)

 

 

 

4.00
-I—Prediction by Plug-flow Model
2.00 J +Observation by Experiment
0.00 “ I I l I
O 5 10 15 20 25 30
Time (min)

Figure 4.10 Comparison of Dissolved ozone Concentration in DDI Water
Obtained Model Prediction with that Experimentally Determined
at Flowrate 400 mL/min and Temperature 22 C

 

 

+ Prediction by Mixed-flow Model

 

Conc. of Gaseous Ozone in Headspace (mg/L)

 

 

15.00 -
10 00 _ +Prediction by Plug-flow Model
5 00 _ +Observation by Experiment
0.00 . , , 1 , I
0 5 10 15 20 25 30
Time (min)

Figure 4.11 Comparison of Exit Ozone Concentration in DDI Water Obtained
Model Prediction with that Experimentally Determined at Flowrate
400 mL/min and Temperature 22 C

145

 

 

 

 

 

14.00
g 12.00 —
E
2 10.00 -
8
G 8.00 -
'3
2.
g 6.00 -
.3 +Prediction by Mixed-flow Model
“5 4.00 P - -
6 + redlctlon by Plug-flow Model
a
5 2.00 z , +Observationby Experiment
0.00 . r r a t r ,
0 5 10 15 20 25 30

Time (min)

Figure 4.12 Comparison of Dissolved Ozone Concentration in DDI Water
(pH 2.1) Obtained Model Prediction with that Experimentally
Determined at Flowrate 400 mL/min and Temperature 22 C

 

 

 

 

 

g 60.00
5
§ 50.00 -
a.
M
'3

.00 ~
:95 40
.E
g 30.00 -
O +Prediction by Mixed-flow Model
3 20.00 . .
2 +Predlctlon by Plug-flow Model
0 10.00 +Observation by Experiment
h-
O
6
'5' 0.00 . r . . . r
U

0 5 10 15 20 25 30

Time (min)

Figure 4.13 Comparison of Exit Ozone Concentration in DDI Water (pH=2. l)
Obtained ModelPrediction with that Experimentally Determined
at Flowrate 400 mL/min and Temperature 22 C

146

 

  
 
 
 
  
 
 
 
 

 

 

 

1 2 .00
g, 10.00 -
E
g 8 ()0 —
3 .
O
E 6.00 l
g +prediction by Mixed-flow Model
9 4.00 4
"5 + Prediction by Plug-flow Model
6
5 2.00 1 +Observation by Experiment
0.00 r i r t
O 5 10 15 20 25 30
Time (min)

Figure 4.14 Comparison of Dissolved Ozone Concentration in DDI Water
(pH 6.7) Obtained Model Prediction with that Experimentally
Determined at Flowrate 400 mL/min and Temperature 22 C

60.00

 

50.00 ~

40.00 ~

  

0.00 '1
+ Prediction by Mixed-flow Model

   

Cone. of Gaseous Ozone in Headspace
(mg/L)

 

 

 

20'00 “ + Prediction by Plug-flow Model
10.00 - + Observation by Experiment
0.00 . , , r
0 5 10 15 20 25 30
Time (min)

Figure 4.15 Comparison of Exit Ozone Concentration in DDI Water (pH=6.7)
Obtained ModelPrediction with that Experimentally Determined
at Flowrate 400 mL/min and Temperature 22 C

147

 

u—u
N

1...:
O

 

 

 

Cone. of Dissolved Ozone (mg/L)

 

0 l l I T I
0 5 10 15 20 25 30

Time (min)

+OriginalKLa ++10%KLa +-10%KLa
++20% KLa +40% KLa

 

Ur
O

Conc. of Outlet Ozone (mg/L)
t—I ... N N W W vb uh

Lit O U! O U! C UI O VI
1 1 1 1 l 1 1 1 I

 

 

 

 

O I l T If I

0 5 10 15 20 25 30

Time (min)

-O- +Orig/inal KLa -I- +10% KLa
+-/10 ++20% KLa
+40% KLa

Figure 4.16 Sensitivity Profile of Dissolved and Exit Ozone Concentrations
by Changing the Mass-transfer Coefficient in DDI Water

148

 

Cone. of Dissolved ozone (mg/L)

 

 

 

 

O l I I l I
0 5 10 15 20 25 30

Time (min)

+Origina1 Alpha ++10% Alpha + -10% Alpha
++20% Alpha + -20% Alpha

 

Cone. of Outlet Ozone (mg/L)

 

 

 

0 l I I l l
0 5 10 15 20 25 30

 

Time (min)

+Original Alpha ++10% Alpha + -10% Alpha
++20% Alpha + -20% Alpha

Figure 4.17 Sensitivity Profile of Dissolved and Exit Ozone Concentrations
by Changing the Partition Coefiicient Coefficient in DDI Water

149

 

 

 

 

 

E
I)
l:
S
O
1:
0
.2.
’n
"5
ii
6
U
0 I 1 I fl I
0 5 10 15 20 25 30
Time (min)

+Original k1 ++10% kl +-10%k1++20% kl +-20%k1

 

 

 

 

 

E
8
O
5
E
8
G
2
5
e
O I I 1 I 7
0 5 10 15 20 25 30
Time (min)

+Original k1 ++10% k1+-10% k1++20% kl +-20% kl

Figure 4.18 Sensitivity Profile of Dissolved and Exit Ozone Concentrations
by Changing the Self-decomposition Coefficient Coefficient in
DDI Water

150

 

 

 

 

 

12
'5‘. 0-
E 1
0
§ 8‘
o
u
E 6 ‘
8
'5 4i
“5
2 2 -
Q
U
0 I l I I T
0 5 10 15 20 25 30
Time (min)
+Origina1 H0 ++10% H0 +40% H0
++20% H0 +40% H0

 

 

 

 

 

E.
0
a
O
N
O
8
7.:
:1
O
‘8
t5
=
G
U
0 I I l I I
0 5 10 15 20 25 30
Time (min)
+Original H0 —l—+10%H0 +40% H0
+ +20% H0 + 40% H0

Figure 4.19 Sensitivity Profile of Dissolved and Exit Ozone Concentrations
by Changing the Gas Hold-up in DDI Water

151

 

 

 

 

 

6 5
'5‘. 5 ~
31
8 4 J
S
C
e 3 “
g + Experimental Measurement
2
.2
a + Predicted by Plug-flow Model
1
0 ’ I r I I I
0 5 10 15 20 25 30

Time (min)

Figure 4.20 Comparison of Dissolved Ozone Concentration in DDI Water
in Pilot-scale System Obtained by Model Prediction and

Experimental Data at the Flowrate 443 mL/min and Temperature
22 C

 

 
  
 
   

 

 

 

25
20 ~
ii
5 15 4
0
=
S
2 10 «
a -0— Experimental Measurment
5 ‘ +Predicted by Plug-flow Model
0 ~ , . f T
0 5 10 15 20 25 30
Time (min)

Figure 4.21 Comparison of Exit Ozone Concentration in DDI Water in Pilot-
scale System Obtained by Model Prediction and Experimental
Data at the Flowrate 443 mL/min and Temperature 22 C

Chapter 5

OXIDATION KINETICS OF PHENOLIC AND INDOLIC COMPOUNDS IN

AQUEOUS ENVIRONMENT BY OZONE

5.1 Introduction

Phenolic (phenol, p-cresol, and p-ethylphenol) and indolic (indole and skatole)
compounds are regarded as among the most malodorous compounds found in swine
manure slurry (Spoelstra, 1977). Although many researchers have investigated the
kinetics of the reaction of ozone with phenol and p—cresol (Gould and Weber, 1976;
Augugliaro and Rizzuti, 1978; Hoigné and Bader, 1983; Singer and Gurol, 1983; Zheng
et al., 1993), no effort has been made to investigate the kinetics of the reaction of these
compounds in a complex matrix such as wastewater. Although, in full-scale contactors
ozone gas is bubbled into the liquid being treated, few researchers have determined the
kinetics of ozone in such heterogeneous systems. In a few cases, empirical kinetic
models of ozonation have been proposed for heterogeneous systems in order to determine
an apparent rate constant for the overall reaction as a function of both flow and chemical

parameters (Roth et al., 1982), but these constants are only limited to the specific reactor

type used.

152

153

The objectives of this study were to describe the kinetics of the oxidation of phenolic
and indolic compounds by ozone in a semi-batch reactor, to determine the effect of such
system parameters, such as flowrate, temperature, pH, and solution composition on the
removal of these malodorous compounds, and to develop a mathematical model, which
combines the mass transfer of ozone and ozonation kinetics, to predict the degradation of

those compounds in synthetic and real swine manure.

5.2 Materials and Methods
5.2.] Preparation of Bufi‘er and Chemical Solutions

Stock buffer solutions for pH control were prepared according to Table 5.1. The
ionic strength of stock buffer solutions was fixed at 3 M by adding a specific amount of
sodium sulfate into each of the solutions. Malodorous chemicals such as phenol, p-
cresol, p-ethylphenol, indole, and skatole, respectively, were weighted and dissolved in
DDI water. The phenolic and indolic compounds were placed in light-sensitive bottles
and stored in the refrigerator for use. All chemicals were used within a week of
preparation. Distilled deionized water (DDI water) was used to prepare all solutions.
Aqueous ozone was prepared by continuously passing gaseous ozone from a fritted
diffuser through DDI water to which had been added with 50 mL of a pH buffer to
achieve the desired pH and ionic strength. The concentration of aqueous ozone achieved

a stable level (ca. 16 mg/L at pH 2 and 11 mg/L at pH 6.7).

154

Table 5.1 The Composition of Stock Buffer Solution

 

 

 

 

 

pH Reagents

2.1 8 mL concentrated H2804 + 120.7 g Nast4 in 500 mL DDI water
6.7 92.84 g NazHPO4 + 141.13 g KHzPO4 in 1000 mL DDI water

9.9 74.2 g H3BO3 + 40 g NaOH + 95.14 g Na2804 in 1000 mL DDI water

 

 

 

5.2.2 Stoichiometric Factor and Reaction Rate Constant

This experiment was conducted in a series of 30 mL glass tubes into which were
placed mini-stir bars to keep the solution completely mixed. The stoichiometric factor of
the reactions was obtained by mixing aqueous target compound (phenol, p-cresol, p-
ethylphenol, indole, or skatole) and ozone solutions of known concentrations. In order to
avoid, as much as possible, the interference of other reactions (i.e., between ozone and
byproducts), in all cases, the initial concentrations of target compounds were kept much
higher than that of ozone so that the ozone could be consumed mainly by its reaction with
parent compounds. Two concentrations of target compounds were used: 100 and 200
mg/L. An appropriate volume of the ozone solution was added to the tube to react with
the target compounds. After the aqueous ozone was consumed, chemical analyses were
done to determine the consumption of target compound. The stoichiometric factor was
then computed as n=A[03]/A[M]. All experiments were conducted in triplicate.

The concentrations of target compounds were determined using high-performance

liquid chromatography (HPLC) (Perkin Elmer Corporation, Norwalk, CT). A platinum
guard and analytical columns (EPS C18, Alltech Corporation, Deerfield, IL) were used.

The HPLC was operated at the flowrate of l mIJmin and the ratio of mobile phase 43:57

155

(acetonitrilezwater). A wavelength in the UV diode array detector (Perkin Elmer
Corporation, Norwalk, CT) of 255 nm was used.

The rate constants for the reaction of the target compounds and ozone were
estimated using the method of competition kinetics (Xiong and Graham, 1992) and
relative reaction-rate constants (Hoigné and Bader, 1983“). Considering a pair of organic

compounds (M1 and M2) in aqueous solution, they will compete for ozone, as described

 

 

 

 

 

 

below:
d Ml
_[dt ]=”k1,03[M1103] (5.1)
d M
[ (”21:—km, [M2103] (5.2)
Thus,
d[Ml]= k1,o3[M1] (5 3)
dlel 19.0.1le
f M \
.. m.
[(2.03 =k1.03 >[M2] < (54)
Log 1 '
([MJ.)

p-Cresol was chosen as the reference compound as the reaction rate constant of ozone
with p-cresol had been determined (Zheng et al., 1993). From this reaction rate constant,

the reaction rate constant of the other target compounds could be calculated.

5.2.3 Factors Afiecting the Removal of the Odorous Target Compounds by Ozone
A semi-batch reactor having a capacity of 1.5 L was used to observe the reduction of
phenolic and indolic compounds by ozone in aqueous solution. In this phase, two

aqueous solutions containing phenolic and indolic compounds, respectively, were utilized

156

to determine the effect of pH, flowrate, temperature, ammonia, COD, and VFA on the
oxidation of the target compounds by ozone. The ozonation system was configured as
shown in Figure 5.1 (the photo can be seen in Figure 4.2). In each experiment, pH, water
temperature, ionic strength, and the concentrations of the target compounds were
maintained at the desired levels. In addition, the gaseous concentration of inlet ozone
was maintained at the desired value by adjusting the voltage valve on the panel of the
ozone generator (Model T-408, Polymetrics Inc., San Jose, CA). When the experiment
began, the kinetic program (Kinetic-2) in the UV spectrophotometer (UV 1201,
Shimadzu Incorporation, Japan) was initiated to monitor and record the gaseous ozone
concentration at both inlet and outlet lines at 5 minute intervals. In the meantime, the
liquid samples were withdrawn from the reactor at regular intervals to measure the
concentrations of target compounds by HPLC and dissolved ozone by the indigo method
(Bader and Hoigné, 1982). The total liquid volume withdrawn for sampling purposes

from the reactor was less than 7% of the initial volume in the reactor.

5.2.4 The Ozonation of Synthetic and Real Swine Manure in a Semi-batch Reactor
In order to better understand the reaction kinetics between ozone and the target

malodorous components, a formulation of synthetic swine manure was at the

concentrations of major malodorous composition reported by Wu et al. (1998) and

Yasuhara (1980). The concentration of each compound is listed in Table 5 .2. In these

experiments, temperature (14, 22, 30 °C) and flowrate (100, 300, 500 mIJmin) were

varied to investigate their effect on the rate of ozonolysis of the target malodorous

157

Table 5.2 Concentrations of Malodors in the Synthetic Manure

 

 

 

 

 

 

 

 

Phenolic and Indolic Concentration Volatile Fatty Concentration
Compounds (mg/L) Acids (mg/L)
Phenol 28. 1 Acetate 3000
p-Cresol 210 Propionate 170
p—Ethylphenol 3.5 iso-Butyrate 34. l
Indole 5. l Butyrate 322
Skatole 12.8 iso-Valerate 128
Valerate 83.2

 

 

 

 

 

compounds. The pH value of the synthetic manure was controlled using a phosphate
buffer system (pH=6.7). The ionic strength of the synthetic manure was 0.1 M. A control
experiment, in which the manure slurry was sparged with oxygen using the same
flowrates as that used in the ozone experiments, was conducted to investigate the effect of
stripping and oxygenation. Experimental procedures were the same as those documented
in Section 5.2.3 except for the analytical methods used for the phenolic and indolic
metabolites. Standards and samples of phenolic and indolic compounds were extracted
by methylene chloride (1:1). Gas chromatography (HP 5890, Hewlett-Packard Inc.,
Avondale, PA) was employed to separate and quantify these compounds. Each sample (3
uL) was analyzed directly by GC-FID, using a stainless steel column (2 mx2 mm) packed

with 21% Carbowax 4000 and WAW-DMCS (60/80 mesh). The column temperature
was held at 100 0C for the first 1 min after injection; the temperature was ramped at 10

°C/min to 125 oC and held for 8 min, then ramped at 2 °C/rnin to 155 0C. The total run

158

for each sample was 26.5 min. Injection port and FID temperatures were maintained at
200 and 250°C, respectively.

The concentrations of volatile fatty acids (including acetate, propionate, butyrate,
iso-butyrate, valerate, iso-valerate) were also determined by GC/FID. 0.5 mL of
concentrated hydrochloric acid was added to 10 mL of manure slurry sample. After the
mixture was agitated in a vortex mixer, it was centrifuged at 12,000 G for 10 min to
remove particulate matter. An aliquot (3 uL) of the supernatant was injected directly into
a 900 mm x 6 mm glass column packed with Supelco GP 10% SP-2100 on 100/ 120

Chromosorb W AW (Supelco Inc., Bellefonte, PA). The carrier gas was nitrogen flowing
at 25 mIJmin. The column temperature was held at 125 0C for the first 5 min after
injection; the temperature was ramped at 4 oC/min to 175 0C and held for 25 min, then

ramped at 5 oC/min to 200 0C and held for 15 min. The total run for each sample was
62.5 min. Injection port and FID temperatures were maintained at 150 and 200°C,
respectively.

Real swine manure was collected from the storage pit of the one of the swine houses
at MSU swine farm. After collection the manure was placed in a four-liter glass bottle for

two weeks to allow fermentation to occur. Oxygenation and ozonation experiments were

conducted at a temperature of 22 0C and a flowrate of 300 mIJmin. Samples were taken
at ozone dosages from O to 2 g/L at increments of 0.25 g/L. At the ozone dosage used,
19.3 min was needed for an incremental increase of 0.25 g/L ozone. The analytical

methods for phenolic and indolic compounds were mentioned in the previous paragraph.

159

5.3 Model Development

A model combining mass transfer and chemical oxidation kinetics was developed to

predict the accumulation of dissolved ozone concentration and the decrease in the

concentrations of malodorous compounds in the synthetic and real swine manure during

lab-scale ozonation. The mass-transfer model for ozone in DDI water was employed.

Incorporated into this model were the concepts of enhancement factor and chemical

kinetics. The following basic assumptions were made:

1.

2.

Liquid phase is completely mixed;

The flow pattern of gaseous ozone through the system is plug-flow;

Constant temperature and pressure are maintained during the experimental period.
Ozone is consumed mainly by malodorous parent compounds (phenol, p-cresol, p-
ethylphenol, indole, and skatole); and

The enhancement factor is constant during the ozonation process.

Mass transfer and the ozonation reactions can be described mathematically by the

following equations:

dl03]

 

= ExKu(a g; -[o,1)—2n. <k.)[0,1[M1.. -k, [0.1—kilos?“

d M .
__[_d?]_L_—_k5[M],. +(k,)[03l[Ml.~

160

where:

[03] : Dissolved ozone concentration (mole/L)
E : Enhancement factor by chemical reaction
K M : Mass - transfer coefficient of ozone in solution in the absence of ozone - reactive

compounds (l/sec)
a : Partition Coefficient of ozone

C'" : Gaseous concentration of inlet ozone (mole/L)

ni :Stoichiometric factor for malodorous compound, i

k. : Re action rate constant for malodorous compound i with dissolved ozone
(L/sec - mole)

: Concentration of malodorous compound (mole/L)

k f :Recation Coefficient by the byproducts or other compounds which also

consume available ozone (l/sec)
k d : Self - decomposition rate constant of dissolved ozone
(1/sec for m equal to 1; L/sec - mole for m equal to 2)
m : Self - decomposition order of dissolved ozone
kI : Coefficient for the removal of malodorous compound, i, by stripping and oxygenation

The number of system equations depended on the number of initial malodorous
compounds in the ozonation reactor, e. g., if five malodorous compounds were selected,
there would be six simultaneous equations in the model. Since the equations are non-
linear, it is impossible to find an analytical solution for this model. Instead, numerical
methods can be used to solve these non-linear simultaneous equations. Mathematica 3.0
(Wolfram Research Inc., Champaign, IL) provides a numerical package to obtain the
solutions in the reaction system as long as we can estimate the values of some parameters.
To distinguish the effect of the reaction between ozone and the intermediates formed

from the oxidation of parent compounds, two models have been used. Model I assumes

that

the

Oil

COT

5.4

5.4

SIC

re

CC

al

b.

1‘1

H-

161

that the reaction coefficient is zero. In Model I], the reaction coefficient is evaluated to fit
the experimental data. It is especially important to determine the effects of side reactions
on the degradation of target parent compound by ozone as the byproducts are also able to

compete with their parent compounds for the available ozone.

5.4 Results and Discussion
5.4.1 Stoichiometric Factors and Rate Constants for Ozonation Reactions

The stoichiometric factors and rate constants for the reaction of ozone with phenolic
and indolic compounds are summarized in Table 5.3. Results show that the
stoichiometric factors for phenol are 1.81i0.16 and 1.86i0. 16 at pH 2.1 and pH 6.7,
respectively. Hoigné and Bader (1983) indicated that for the ozonation of phenol-like
compounds, 2.5 moles of ozone are needed to consume 1 mole of phenol; Li et al. (1979)
also reported that the stoichiometric factor is 2. However, Eisenhauer (1978), Gould and
Weber (1976), and Roth et al. (1982) found that for the total destruction of phenol,
between 4-6 moles of ozone are required to remove 1 mole of phenol. These values are
higher than that found in this work because they did not distinguish between the amount
of ozone that reacts with phenol and that which reacts with the ozonation byproducts.
Since reaction conditions were such to favor the reaction of ozone with the target
chemical over any other reaction, the stoichiometric factors found in this work seem to be

reasonable. For p-cresol, the stoichiometric factors were found to be 1.00 :l: 0.07 and 1.38

i 0.11 at pH 2.1 and pH 6.7, respectively. The value reported by Zheng et al. (1993) for

Tabl

E

2.3 L E

:E:

 

162

Table 5.3 The Stoichiometric Factor and Rate Constants for the Reaction of Ozone with
Phenolic and Indolic Compounds under Different pH Conditions
(C.L.= Confidence Limits)

 

pH=2.l, I=0.1, T=22 °c

 

 

 

 

 

 

Stoichiometric factor 95% CL. Reaction rate constant 95% CL.
(M'lsec'l)
phenol 1.81 0. 16 460 10
p-cresol 1 .00 0.07 5,200 NRa
p-ethylphenol 1 . 13 0.05 1 ,650 280
indole 0.68 0.07 45,000 2,900
skatole 0.90 0.03 10,200 2,200

 

 

 

 

 

pH=6.7, I=0.1, T=22 °c

 

Stoichiometric factor 95% CL. Reaction rate constant 95% CL.

 

 

 

 

 

 

 

 

 

 

 

(M"sec")
phenol 1.86 0.16 750,000 33,000
p-cresol 1 .38 0. l 1 850,000 NR”|
p-ethylphenol 1.29 0.1 1 990,000 1 10,000
indole 0.98 0.07 5,000,000 3,600,000
skatole 0.89 0.04 4,500,000 780,000

 

“ Value was determined by Zheng et al. (1993) and CL. were not reported

cresol isomers was 3; the value documented by Beltran et al. (1990) for o-cresol was 2.
Using the stoichiometric factor for p-cresol determined in this work, it appears that
approximately 1 mole of ozone is consumed by 1 mole of p-cresol. Due to the lack of
references relating to the ozonation kinetics of p-ethylphenol, indole, and skatole, no
comparison could be made for the stoichiometric factors of these compounds. In our
work, the stoichiometric factor of p—ethylphenol was found about 1.2 for the pH range we
used. For indolic compounds, however, the stoichiometric factor at pH 6.7 was less than
1 (but close to 1). The only case is where the stoichiometric factor was less than one,

occurred with indole. Its stoichiometric factor was 0.68 :l: 0.07 at pH 2.1. One

ex;

att

163

explanation for this result is that one of the byproducts, which formed after ozone initially
attacked indole, reacted with indole to further consume the parent compound.

The reaction rate constants for the phenolic and indolic compounds were found to
increase significantly with increasing pH. This increase for phenolic compounds is
thought to be due to greater reactivity of the phenolate ion as compared that of phenol.
Augugliaro and Rizzuti (1978) and Hoigné and Bader (1983) both showed that the
phenolate ion has a much larger rate constant towards ozone than that of phenol
(109 M’ls'l for phenolate ion and 500 M'ls'l for phenol). Hoigné and Bader (1983) also
demonstrated that the total reaction rate constants for phenolic compounds increase over a
wide range of pH values by a factor of 10 per pH unit, corresponding to the incremental
increase in the degree of dissociation. Although another explanation is that hydroxyl
radicals cause the increase of the reaction rate with pH, Beltran et al. (1992) negated this
hypothesis when they found that the rate constants for the ozonation of o-cresol in
aqueous solution with and without tert—butanol, a radical scavenger, were nearly identical.
In our experiment, we used the phosphate buffer system to control the pH at 6.7. Masten
et al. (1996) demonstrated that, at this pH, hydrogen phosphate (HPO42') and dihydrogen
phosphate (H2PO4’) are the predominant OH radical scavengers in oxidizing
trichlorobenzene by advanced oxidation process. Thus, the direct reactions between
ozone and the target compounds predominate because any free radicals generated from
the decomposition of ozone are likely to be scavenged by the phosphate species present in
the buffer system we used.

The reaction rate constants determined at pH 2.1 were ranked, indole > skatole > p-

cresol > p-ethylphenol > phenol. The electrophilic reaction is though to be the major

mec
aror
pos:

l'CSl

CH
mc
eth

Th

is
G.
el
of

fa

164

mechanism by which aromatic compounds react with ozone. This is especially true for
aromatic compounds that have electron donating groups substituted at the ortho and para
positions because of the high electron densities that result from this substitution. As a
result of the high electron density, these compounds are very reactive with ozone.
(Langlais et al., 1991). As p-cresol contains two electron donating groups, -OH and -
CH3 (Geissman, 1968), the reactivity of p-cresol by the initial attack of the ozone
molecule is expected to be greater than that of phenol (with only a -OH group). The
ethyl group (-CH2CH3) on p—ethylphenol is also electron donating group (Hart, 1991).
The reactivity of p-ethylphenol is, therefore, expected to be greater than that of phenol.
p-Cresol has a greater reaction rate constant than p-ethylphenol because the methyl group
is a better electron donating group than is the ethyl group. In addition, it was reported by
Geissman (1968) that indole and skatole are very reactive nucleophiles, and undergo
electrophilic substitution reactions with ease. Nevertheless, no explanation can be
offered at this point for why the reaction rate constants of indolic compounds are much
faster than those of phenolic compounds because of structural differences in the two
groups of compounds. A pH 6.7, the reaction rate constants for the three phenolic
compounds are almost the same magnitude (close to 1x106 M'ls'l) and the rate constants

of indole and skatole are 5x106 M"s'1.

5.4.2 Factors Afiecting the Oxidation of Phenolic and Indolic Compounds by Ozone
The effect of design parameters of ozonation system and physical and chemical
characteristics of the swine manure on the reduction of the concentrations of phenolic and

indolic compounds was determined. The physical parameters studied were flowrate and

temp
swir

mU

con
red‘
cor
(Fi
c01
col

02

in
ta

b«

165

temperature; the chemical parameters were pH and other constitutional components in
swine manure, such as VFAs, COD, ammonia, etc. By varying the flowrate (100-500
mIJmin) the concentration of ozone in the inlet gas stream was maintained at 64.58
mg/L, it could be shown that the reduction in the concentrations of phenolic and indolic
compounds increased as the flowrate also increased (Figures 5.2-5.6). If we consider the
reduction in concentration as a function of ozone dosage rather than time, the reduction in
concentrations appear to depend solely on ozone dosage and be independent of flowrate
(Figures 5.7-5.11). As all of the ozone bubbled into the reactor was utilized and the
concentration of dissolved ozone during the experiment was zero, all ozone dosed was
consumed. Therefore, ozone dosage is thought to a valuable parameter in designing the
ozonation system to treat the wastewater containing phenolic and indolic compounds.
pH has a significant effect on the reduction in the concentrations of phenolic and
indolic compounds (Figures 5.12-5. 16). At pH of 2.1, in a solution containing the three
target phenolic compounds, p-cresol concentration was observed to rapidly decrease to
below detection limits within 30 minutes. On the contrary, the reduction of phenol
concentration was retarded during the initial twenty minutes, then began to accelerate
after p-cresol was consumed. The p-ethylphenol concentration decreased to just between
the concentrations of phenol and p-cresol. As mentioned in Section 5.4.1, the largest
value for the reaction rates at pH 2.1 was that for p—cresol (5,200 M'ls'l), followed by p-
ethylphenol (1650 M'ls'l), and then by phenol (460 M"s"). Based upon these rate
constants, one would expect that, initially, the available ozone would be consumed
mainly by pocresol, and that phenol would not be oxidized until most of the p—cresol had

disappeared. The results are consistent with the hypothesis. In studies conducted at pH

 

6.7
cor
cor
Cir
5211
co
dc
pl
re

fc

166

6.7 and 9.9, the rates of degradation of the three phenolic compounds, when present in a
complex mixture, were not affected by increases in pH over this range. The reaction rate
constants for these phenolic compounds are almost identical (at the same order) at the
circumneutral and alkaline pH. Thus, the three components decreased according to the
same pattern. It is worthy to note that in the presence of phenol and p-ethylphenol the
concentration of p-cresol was significantly reduced at pH of 2.1 as the majority of ozone
dosed into the system reacted with even though the reaction rate constant for p-cresol at
pH 2.1 is much less than that in the solution having higher pHs. The effect of pH on the
reduction in the concentrations of indolic compounds in a complex mixture solution was
found to be insignificant, within the pH range of 2.1 to 9.9.

Temperature did not affect the rates of degradation of phenolic or indolic
compounds when present in the complex mixture (Figures 5.17-5.21). This result is
consistent with that found for pilot-scale studies and is thought to be due to the decrease
in the solubility of ozone and the increase in reaction rate constant with increasing
temperature. Over this temperature range (14 — 30 °C), these two factors appeared to
counter-balance each other.

Dextran (M.W. -:-:- 75,000), a type of starch, was used to simulate the complex
organic constituents found in the swine manure. The COD in the phenolic or indolic
solution containing dextran was 15,000 mg/L, which is the approximate concentration of
the liquid part of swine manure. The results (Figures 5.22-5.24) indicate that dextran
slightly retards the rate of oxidation of the phenolic compounds. This is thought to be
because the dextran molecule is surface active and results in a resistance at the interface

where available ozone can oxidize these chemicals. However, the effect of dextran on the

 

rauao
rau:c
possi
such

thetz

and
500
the
ex;

tha

US

TC

167

rate of oxidation of indolic compounds is insignificant, apparently because the reaction
rate constants of indolic compounds towards ozone are large enough to overcome any
possible interference that might exist at the interface (Figures 5.25-5.26). Other factors,
such as the presence of ammonia and volatile fatty acids, did not affect for the removal of
the target odorous compounds (Figures 5.27-5.36).

To compare the effects of stripping and oxygenation on the removal of the phenolic
and indolic compounds, oxygen was used to purge the solution. The maximum flowrate,
500 mIJmin, was employed. Almost no change occurred in the concentration of any of
the target compounds during a period of 80 minutes. This result differs from our previous
experiment in which “real” swine manure was aerated using the pilot-scale system. In
that study, the concentrations of the target compounds were decreased approximately by
15%, 50%, 40% and 45% for phenol, p—cresol, p—ethylphenol, and skatole, respectively,
using oxygen. A flowrate of 1,800 mIJmin and a treatment period of 40 minutes,
respectively, were used. Therefore, it is hypothesized that microbiological metabolism,
activated by oxygen, is the real mechanism by which the phenolic and indolic compounds

disappear during aeration of the real swine manure.

5. 4. 3. Comparison between Experimental Result and Model Prediction

A model combining mass transfer and oxidation kinetics was employed to predict
the concentration profiles of phenolic and indolic compounds in water under different pH
values and flowates. Although, in previous experiments, the mass-transfer coefficients,
partition coefficients, stoichiometric factors, and reaction rate constants have been

determined, the enhancement factor and reaction factor (kf) are still unknown. Once

168

those parameters are determined, the oxidation model can be used to simulate the
ozonation process.
The enhancement factor is defined as the ratio between the actual and maximum

physical absorption rates (Danckwerts, 1970; Charpentier, 1981):

 

N
=—"—’—.— (5.5)
KLa[O3]
m—m
N03 = 'V 0 (5.6)

where N03 is determined from the difference between the molar flux of ozone at the
reactor inlet and outlet, m, and mo, KLa is the liquid phase volumetric mass-transfer
coefficient, [03'] is the dissolved ozone concentration at the water-gas interface, and V is
the liquid volume in reactor. Until the phenolic or indolic compounds were removed
from the reactor, the gaseous ozone concentration in the stream was not detectable. Thus,
the enhancement factor is a constant during the reaction period. The E values at

temperature of 22 °C were shown in Table 5.4 .

Table 5.4 The Enhancement Factor Calculated under Different Flowrates and pH Values

2.1 =6.7
100 mIJmin 1.9 2.0

300 mIJmin 3.8 3.2
mIJmin 4.4 3.8

 

 

 

Anoth

reactions ir

Ozonation

 

 

These com
1976; Sin:
to carry 0L
To overco
COElficlen
impurities

“Wilme-

CXpe.
The -

inl6r

169

Another important feature to be considered is the occurrence of competitive
reactions involving ozone with the intermediates formed during the ozonation.
Ozonation of phenolic compounds yields dihydroxybenzenes as the initial byproducts.
These compounds are as reactive with ozone as the parent phenols (Gould and Weber,
1976; Singer and Gurol, 1983). These circumstances make kinetic studies more difficult
to carry out especially for an initial reaction system with more than one parent compound.
To overcome the difficulty in using the mathematical model, an overall reaction
coefficient, kf, was used to describe the consumption of ozone by all intermediates or
impurities. The overall reaction coefficient could be estimated by fitting our

experimental data to find a minimum residual, which is defined as below:

 

N
Z (X j — X Experiment )2

R: "' 5.7
N ( )

 

where Xj is the predicted concentration for the parent compounds in reactor, XExpcdmcm is
the observed concentration during the experiment, and N is the number of total
observations.

Phenolic mixtures (phenol, p—cresol, p—ethylphenol) and indolic mixtures (indole and
skatole) were ozonated respectively under the following conditions: pH 2.1, temperature
of 22 °C, flowrate of 300 mIJmin, and ionic strength of 0.1 M. In Figures 5.37-5.41, the
experimental data and the predicted values determined using Models I and II are shown.
The values for the minimum residuals used in Model 11 are 0.8 and 1.0 for the phenolic
compounds and indolic compounds, respectively. When the reactions between

intermediates and ozone were ignored, the predicted concentration profiles of those

 

compc
the ox
effect
comp
fitted

betw

Tal

170

compounds decreased much faster than those determined experimentally. Therefore, for
the oxidation of phenolic and indolic compounds, it is very important to consider the
effect of side reactions during ozonation. At pH 6.7, the reaction coefficients were also
compared. The reaction coefficients (kf) are much larger than those found at pH 2.1. The
fitted values for reaction coefficient are summarized in Table 5.5 and a comparison

between experimental data and model predictions can be seen in Figures 5.42-5.56.

Table 5.5 The Reaction Coefficients for Phenols and Indoles under Different Flowrates
at pH 6.7

Phenols s' Indoles s'
100 mUmin 350 450

300 mIJmin 400 400
500 mIJmin 350 400

 

5.4.4 Oxidation of Synthetic Manure and Stored Swine manure by Ozone

Synthetic manure and stored swine manure were ozonated to investigate the
oxidation of phenolic and indolic compounds in complex matrices. The synthetic manure
was ozonated at different temperatures and flowrates. Changing the temperature from 14
— 30 °C had no effect on the removal of phenolic and indolic compounds (Figures 5.57-
5.61). This is in agreement with our previous experiments in which individual phenolic
or indolic compounds were ozonated. The effect of flowrate (100, 300, 500 mIJmin) is
shown in Figures 5.62-5.66. As observed previously, when plotting the concentration of

compound as a function of ozone dosage, rather than time, the concentration profiles of

171

phenolic and indolic compounds appeared to be essentially independent of flowrate
(Figures 5.67-5.71). Based upon the results presented in these figures, when the ozone
dosage is greater than 0.5 g/L, phenol and p—cresol would be removed completely; when
the ozone dosage is greater than 0.1 g/L, p—ethylphenol, indole, and skatole would also be
removed. Therefore, an ozone dosage of 0.5 g/L can be established as that which will
remove all of the malodorous components in the synthetic manure. It can be predicted
that higher ozone dosage should be needed to treat the real swine manure when having
the same original concentrations of phenolic and indolic compounds. In our pilot-scale
study, although the ozone dosage of 0.5 g/L was established for the odor acceptability, 1
g/L ozone dosage was still needed to remove all of the phenolic and indolic compounds.
For the engineering practical, however, the use of ozone dosage in designing the reaction
system should adopt the one for the improvement of malodors to an acceptable level,
instead of the complete removal of the malodorous substances.

The concentrations of VFAs were reduced by less than 10% during ozonation for 80
minutes (Figure 5.72). It is consistent with the finding in the chapter 2 and chapter 3 that
the VFAs were not oxidized by ozone in the real manure. This can be explained by the
very low rate constants of ozone with VFAs (Hoigné and Bader, 1983), which means that
VFAs are not able to compete with the more readily oxidable substrates for ozone.

In the synthetic manure, stripping and oxygenation were ineffective at removing any
of the phenolic or indolic compounds at the same hydrodynamic conditions as that used
with ozone gas. As such, the losses of phenolic and indolic compounds from the
synthetic manure appear to be due entirely to ozonation reactions. The oxidation model

was used to predict and compare the decrease of those target compounds. Results show

172

that the prediction using Model I results in a much faster reduction than that observed
experimentally. Thus, the reaction coefficient (kf) should be used to take into account
competing reactions. The reaction coefficients are summarized in Table 5.6 for the
various flowrates. The fitted values of reaction coefficients are in the same range as that
obtained when the individual phenolic or indolic compounds were ozonated. The

comparisons of experimental observations and model predictions can be seen in Figures

5.73-5.87.

Table 5.6 The Reaction Coefficients for Synthetic Manure Determined at Different
Flowrates

Flowrates Reaction Coefficient s'
100 mIJmin 300

300 mIJmin 400
mIJmin 600

 

The initial characteristics of the stored swine manure are given in Table 5.7. The
manure was oxygenated and ozonated using a lab-scale reactor at a flowrate of 300
mIJmin and temperature of 22 °C. Phenol, p-cresol, p-ethylphenol and skatole, were
found to be reduced by about 10%, 30%, 20% and 25% after two and half hours of
oxygenation (Figure 5.88). Although some attempts had been tried to fit the
concentration curves, no satisfactory mathematical relationship could be obtained.
Microbiological metabolism of the odorous compounds is thought to be the major
removal mechanism when oxygen was provided, resulting in an increase in microbial

activity. However, as the degradation rate of phenolic and indolic compounds in swine

173

Table 5.7 Chemical Characteristics of Real Swine Manure Taken from the Storage Pit

7.5 Acetate 93.87
COD 000 'onate 14.38
Phenol 75.2 iso-B 2.70

1 186.6 B 9.66

Eth l m 20.9 iso-Valerate 2.17
Indole m ND Valerate 1.58

Skatole 10.0 VFAs as Carbon 299.07

 

manure by microbial metabolism is slow, this removal mechanism can be ignored when
compared with the removals obtained by ozone. During ozonation, all of these
malodorous compounds were oxidized after 2 hours, i.e., at an ozone dosage of 1.5 g/L
(Figure 5.89). The VFA concentrations did not change with either oxygenation or
ozonation. The pH values increased from 7.5 to 8.4 after oxygenation and from 7.5 to 8.2
after ozonation for the period of two and half hours. As the initial pH was 7.5, the
reaction rate constants were assumed to be 10 times greater than those obtained for a pH
of 6.7. Thus, the rate constants used in the oxidation model are 107 M'ls'l for the
phenolic compounds and 5x107 M"s'l for skatole. A reaction coefficient of 40,000 3'1
was found to yield the minimum residual in fitting the experimental data (Figures 5.90-
5.93). This value, which is much greater than that obtained for synthetic manure, is likely
to be large because of impurities in real swine manure that can react with ozone.

The use of a model to predict the removals of odorous compounds in swine manure
would be easier if the reaction coefficient could be determined from a “gross parameter”

such as COD. As such, the relationship between kf and COD was considered. We

174

assume that the mathematical relationship between COD and kf can be expressed as

krzkf’ (COD), the intrinsic reaction coefficient, kf’, is calculated to be 1.33 for the real
manure we tested. Similarly, kf’ is calculated as 0.125 for the synthetic manure, in which
the COD is 3,200 mg/L. It should be noted that the pH of real manure is about a unit
larger than that of synthetic manure. Therefore, the concentration of hydroxide ion in the
real manure would be 10 times that in the synthetic manure. If the reaction rates of ozone
with by-products or impurities would also increase for ten times by increasing the pH as
suggested by Hoigné and Bader (1983), the mathematical expression in using COD would

be valid for the two manures.

5.5 Conclusions

The combination of a mass transfer and kinetic study has led to the development of a
model to predict the degradation of malodorous, phenolic and indolic compounds in a
semi-batch ozonation reactor. For this study, several important conclusions can be

drawn:

1. Reaction rate constants of the phenolic and indolic compounds increase with
increasing pH.

2. No dissolved ozone and outlet gaseous ozone can be observed prior to the complete
removal of the phenolic and indolic compounds. One hundred percent utilization of
ozone was obtained under our experimental conditions. Therefore, ozone dosage
becomes a very important and reliable parameter for the design of an ozone reactor to

remove malodorous substances from the liquid manure slurry.

175

3. The effect of pH on the oxidation of phenolic and indolic compounds is
predominately determined by their reaction rate constants and competition kinetics
with ozone. For the solution containing phenolic compounds at lower pH values, the
degradation of phenol is seriously retarded as its reaction rate is the smallest of the
target compounds; the degradation of p-cresol is dramatically accelerated as it has the
greatest reaction rate. As these reaction rates of phenolic compounds are identical at
pH of 6.7, the degradation of these compounds follows similar patterns.

4. The reduction of phenolic and indolic compounds is unaffected by temperature and by
the presence of other compounds, such VFAs and ammonia. Although COD,
simulated by dextran, might cause a slight decrease in the rate of oxidation of these
malodorous compounds by ozone, the effect is still small due to their fast reaction
rates.

5. The removal of phenolic and indolic compounds by stripping and oxygenation was
insignificant at the gas flowrates used. Therefore, the slight reduction of
concentrations of these compounds in real swine manure is attributed to microbial
degradation when ozone-enriched oxygen was used. The difference in the results
obtained in pure water and synthetic manure is likely due to the large numbers of
microorganisms in the real manure.

6. A model combining mass transfer and oxidation kinetics of ozone should include the
effect of side reactions involving ozone and byproducts. Using Model I, which
ignores the effect of side reactions, the predicted rates of the oxidation of these
malodorous compounds are much greater than those observed experimentally. The

reaction coefficients for artificial and real swine manure were estimated as 400

176

(5") and 40,000 (5"), respectively. It is apparent that presence of ozonated reactive
impurities results in a significant increase in the ozone dosage required to oxidize the

target compounds.

5.6 References
1. V. Augugliaro; L. Rizzuti, “The pH Dependence of the Ozone Absorption Kinetics in

Aqueous Phenol Solutions”, Chem. Eng. Sci., 33: 1441-1447 (1978).

N

. H. Bader; J. Hoigné, “Determination of Ozone in Water by the Indigo Method: A

Submitted Standard Method”, Ozone Sci. & Eng., 4: 169-176 (1982).

U.)

. F.J. Beltran; J.M. Encinar; J.F. Garcia-Araya, “Ozonation of o-Cresol in Aqueous
Solutions ”, Wat. Res. , 24: 1309- 13 16 (1990).
4. F]. Beltran; J.M. Encinar; J.F. Garcia-Araya, “Absorption Kinetics of Ozone in
Aqueous o-Cresol Solutions ”, Can. J. Chem. Eng., 70: 141-147 (1992).

5. J.C. Charpentier, “Mass Transfer Rates in Gas-liquid Absorbers and Reactors”, In
Advances in Chemical Engineering, Academic Press, New York, 11:3-133 (1981).

6. RV. Danckwerts, “Gas-Liquid Reactions”, McGraw-Hill Book Company, New York,
96-151 (1972).

7. HR. Eisenhauer, “The Ozonation of Phenolic Wastes”, Journal of WPCF, 40: 1887-
1899 (1968).

8. T.A. Geissman, “Principles of Organic Chemistry”, 3rd Edition, W. H. Freeman and
Company, San Francisco, 547-548, 720-721 (1968).

9. MD. Gurol; P.C. Singer, “Dynamics of the Ozonation of Phenol-H Mathematical

Simulation”, Wat. Res., 17:1173-1181 (1983).

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

177

JP. Gould; WJ. Weber, “Oxidation of Phenols by Ozone”, Journal of WPCF, 48:47-
60 (1976).

H. Hart, “Organic Chemistry- A Short Course”, 8th Edition, Houghton Mifflin
Company, Boston, 107-134 (1991).

J. Hoigné; H. Bader, “Rate Constants of Reactions of Ozone with Organic and
Inorganic Compounds in Water-1(Non-dissociating Organic Compounds)”, Wat. Res.,
17:173-183 (1983).

J. Hoigné; H. Bader, “Rate Constants of Reactions of Ozone with Organic and
Inorganic Compounds in Water-11(Dissociating Organic Compounds)”, Wat. Res.,
17:185-194 (1983).

B. Langlais; D.A. Reckhow; D.R. Brink (Eds), “Ozone in Water Treatment:
Application and Engineering”, Lewis Publishers, Chelsea, MI, 12-13 (1991).

KY. Li; C.H. Kuo; J.L. Weeks, Jr., “A Kinetic Study of Ozone-Phenol Reaction in
Aqueous Solutions”, AIChE Journal, 25 :583—591 (1979).

SJ. Masten; M.J. Galbraith; S.H. Davies, “Oxidation of Trichlorobenzene Using
Advanced Oxidation Processes”, Ozone Sci. & Eng., 18:535-548 (1996).

J.A. Roth; W.L. Moench; K.A. Debalak, “Kinetic Modeling of the Ozonation of
Phenol in Water", Journal of WPCF, 54:135-139 (1982).

PC. Singer; M.D. Gurol, “Dynamics of the Ozonation of Phenol-I Experimental
Observations”, Wat. Res., 17:1163-1171 (1983).

SF. Spoelstra, “Simple Phenols and Indoles in Anaerobically Stored Piggery

Wastes”, Journal of the Science of Food and Agriculture, 28:415—423 (1977).

178

20. S. Wolfram, “Mathematica 3.0”, 3rd Edition, Wolfram Research Inc., Champaign, IL,
1997.

21. JJ. Wu; S. Park; S.M. Hengemuehle; M.T. Yokoyama; H.L. Person; J.B. Gerrish; S.J.
Masten, “The Use of Ozone to Reduce the Concentration of Malodorous Metabolites
in Swine Manure Slurry”, Accepted for Publication by Journal of Agricultural
Engineering Research, 1998.

22. F. Xiong; NJ. Graham. “Rate Constants for Herbicide Degradation by Ozone
Ozone Sci. & Eng., 14:283-301 (1992).

23. A.Yasuhara, “Relation between Odor and Odorous Components in Solid Swine
Manure”, Chemosphere, 9:587-592 (1980).

24. Y. Zheng, D.O. Hill; C.H. Kuo, “Rates of Ozonation of Cresol Isomers in Aqueous

Solutions”, Ozone Sci. & Eng., 15:267-278 (1993).

179

 

1: Oxygen Cylinder

: Ozone Generator

: Cooling System

: Flow Controller

: Three-way Valve

: UV Spectrophotometer
: Three-way Valve

: Semi-batch Reactor

: Trap Solution (KI)

\OmflOMAWN

Figure 5.1 The System Setup for the Semi-batch Ozonation Process

180

 

200.00
180.00
160.00 ~
140.00 a
120.00 4
100.00 1
80.00 ~
60.00 «
40.00 ~
20.00 a

0.00

Cone. of Phenol (mg/L)

 

+ 100 mL/min
+200 mL/min
+ 300 mL/min
+400 mL/min
+ 500 mL/min

   
  
  

 

 

20 40 60 80 100 120
Time (min)

Figure 5.2 The Effect of Flowrate on the Oxidation of Phenol in the Mixture
of Phenolic Compounds at pH 6.7 and Temperature 22 C

200.00

 

180.00

160.00 d
140.00 -
120.00 ~
100.00 -
80.00 —
60.00 d
40.00 —
20.00 a

Conc. of p -Cresol (mg/L)

 

0.00

+ 100 mL/min
-I— 200 mL/min
+ 300 mL/min
+400 mL/min
+ 500 mL/min

    
  
  

 

 

20 40 60 80 100 120

Time (min)

Figure 5.3 The Effect of Flowrate on the Oxidation of p -Cresol in the
Mixture of Phenolic Compounds at pH 6.7 and Temperature

22C

181

 

100.00

    
  
  

 

 

 

90.00 -0— 100 mL/min
g 80.00 - ‘1 +200 mL/min
E 7000 ~ +300 mL/min
9
g 60.00 4 +400 mL/min
2: 50.00 4 +500 mL/min
“.1 40.00 ~
fit
"5 30.00 —
t:
8 20.00 ~
0

10.00 ~

0.00 —

0 100 120

Time (min)

Figure 5.4 The Effect of Flowrate on the Oxidation of p -Ethylphenol in the
Mixture of Phenolic Compounds at pH 6.7 and Temperature

 

 

 

 

22 C

60.00

50"” —o— 100 mL/min
:2: +500 mL/min
'6 30.00 ~
.5
Q-
Q
d 20.00 4
8
o

1000 ~

0.00 -

0 10 20 30 40 50 60 70 80

Time (min)

Figure 5.5 The Effect of Flowrate on the Oxidation of Indole in the Mixture
of Indolic Compounds at pH 6.7 and Temperature 22 C

182

 

 

 

 

60.00
A 50.00
§ -0- 100 mL/min
5;; 40'00 ‘ +300 mL/min
3 30.00 _ +500 mL/min
m
h!
3 20.00 J
:
¢
U

l0.00 e

0.00 J

O 10 20 30 4O 50 60 70 80
Time (min)

Figure 5.6 The Effect of Flowrate on the Oxidation of Skatole in the

Mixture of Indolic Compounds at pH 6.7 and Temperature

 

22 C

200.00

180.00

160.00 _ +Q=100 mL/min
"‘ -I- —200 '
'5‘. 140.00 ~ Q' mum“
g +Q=300 mL/min
g +Q=400 mL/min
g 100 00 — ,
3,. +Q=500 mL/nun
"5
d
t:
O
U

 

 

 

40.00 ~
20.00 «
0.00
0.00 0.50 1.00 1.50 2.00 2.50
Ozone Dosage (g/L)

 

3.00

Figure 5.7 The Effect of Ozone Dosage on the Oxidation of Phenol in the

Mixture of Phenolic Compounds by Different Flowrates at pH

6.7 and Temperature 22 C

183

 

 

 

 

 

200.00
A +Q=100 mL/min
é, 150.00 4 .
E +Q=200 mL/min
E +Q=300 mL/min
3 100.00 - +Q=400 mL/min
A, +Q=500 mL/min
"5
2
5 50.00 ~

0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00

Ozone Dosage (g/L)

Figure 5.8 The Effect of Ozone Dosage on the Oxidation of p -Cresol in the
Mixture of Phenolic Compounds by Different Flowrates at pH

 

 

 

 

 

6.7 and Temperature 22 C

100.00
g) 80.00 +Q=100 mL/rrun
é -I—Q=200 mL/min
‘3' + 300 mL/min
E 60.00 ~ Q:
t_.~.. +Q=400 mL/min
h
5 +Q=500 mL/min
“1‘ 40.00 «
fi.
G...
O
2
5 20.00 T
L)

0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00

Ozone Dosage (g/L)

Figure 5.9 The Effect of Ozone Dosage on the Oxidation of p -Ethylphenol
in the Mixture of Phenolic Compounds by Different Flowrates
at pH 6.7 and Temperature 22 C

184

 

 

 

 

60.00
+Q=100 mL/min

E“ 40.00 fi +Q=300 mL/mln
E +Q=500 mL/min
8
.5
"5
2 20.00 ~
O
L)

0.00 -

0.00 0.20 0.40 0.60 0.80 1.00
Ozone Dosage (g/L)

Figure 5.10 The Effect of Ozone Dosage on the Oxidation of Indole in the
Mixture of Indolic Compounds by Different Flowrates at pH
6.7 and Temperature 22 C

 

 

 

 

60.00
A +Q=100 mL/min
t-J
E 40.00 ~ +Q=300 mL/min
35’ +Q=500 mL/min
‘5
..‘:
U)
“5
g 20.00 4
O
u
0.00 — »
0.00 0.20 0.40 0.60 0.80 1.00

Ozone Dosage (g/L)

Figure 5.11 The Effect of Ozone Dosage on the Oxidation of Skatole in the
Mixture of Indolic Compounds by Different Flowrates at pH 6.7
and Temperature 22 C

185

 

200.00
180.00
160.00 -
140.00 ~
120.00 ~
100.00 ~
80.00 i
60.00 —
40.00 -
20.00 i

0.00 t
0 20 40 60 80 100 120

Cone. of Phenol (mg/L)

 

 

 

 

Time (min)

Figure 5.12 Efi‘ect of pH on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at Flowrate 300 mL/min and Temperature
22 C

 

200.00
180.00
160.00 1
140.00 —
120.00 -
100.00 1
80.00 4

60.00 ~

40.00 ~

20.00 4

0.00

Cone. of p -Cresol (mg/L)

 

 

 

 

0 20 4O 60 80 100 120

Time (min)

Figure 5.13 Efi‘ect of pH on the Oxidation of p -Cresol in the Mixture of
Phenolic Compounds at Flowrate 300 mL/min and Temperature
22 C

186

 

100.00
90.00
80.00 J +pH=2.1
70.00 —
60.00 4
50.00 4
40.00 4
30.00 ~
20.00 ~
10.00 «

0.00

Cone. of p -Ethylphenol (mg/L)

 

 

 

 

0 20 40 60 80 100 120
Time (min)
Figure 5.14 Effect of pH on the Oxidation of p -Ethylphenol in the Mixture

of Phenolic Compounds at Flowrate 300 mL/min and
Temperature 22 C

 

Conc. of Indole (mg/L)

 

 

 

 

 

0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.15 Effect of pH on the Oxidation of Indole in the Mixture of Indolic
Compounds at Flowrate 300 mL/min and Temperature 22 C

187

60.00

 

 

 

 

 

 

50.00 T
g 40.00 4 +pH=6.7
0
E J +pH=9.9
g 30.00
m
“5
a 20.00 4
fl
6
D
10.00 —«
0.00 . i
0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.16 Effect of pH on the Oxidation of Skatole in the Mixture of
Indolic Compounds at Flowrate 300 mL/min and
Temperature 22 C

 

200.00
180.00
160.00 +T=15 C
140.00 4 +T=22 C
120.00 ~
100.00 5
80.00 ~
60.00 ~
40.00 ~
20.00 4

0.00 fl
0 20 40 60 80 100 120

Conc. of Phenol (mg/L)

 

 

 

 

Time (min)

Figure 5.17 Effect of Temperature on the Oxidation of Phenol in the
Mixture of Phenolic Compounds at pH 6.7 and Flowrate 300
mL/min

188

200.00
180.00
160.00

 

140.00
120.00
100.00
80.00
60.00
40.00
20.00

0.00 t t I
0 20 40 60 80 100 120

Conc. of p -Cresol (mg/L)

 

 

 

 

Time (min)

Figure 5.18 Effect of Temperature on the Oxidation of p -Cresol in the
Mixture of Phenolic Compounds at pH 6.7 and Flowrate
300 mL/min

100.00

 

90.00
30.00
70.00 J
60.00 ~
50.00 J
40.00 -t
30.00 ~
20.00 «
10.00 ~

Conc. of p -Ethylphenol (mg/L)

 

 

 

0.00

 

0 20 40 60 80 100 120
Time (min)

Figure 5.19 Effect of Temperature on the Oxidation of p -Ethylphenol in the

Mixture of Phenolic Compounds at pH 6.7 and Flowrate
300 mL/min

189

 

Conc. of Indole (mg/L)

 

 

 

. A

0 5 10 15 20 25 30 35 40

 

 

Time (min)

Figure 5.20 Effect of Temperature on the Oxidation of Indole in the Mixture
of Indolic Compounds at pH 6.7 and Flowrate 300 mL/min

60.00

 

50.00 —
40.00 -
30.00 1

20.00 ~

Cone. of Skatole (mg/L)

10.00 a

 

 

 

0.00

 

 

. k
0 5 10 15 20 25 30 35 40

Time (min)
Figure 5.21 Effect of Temperature on the Oxidation of Skatole in the

Mixture of Indolic Compounds at pH 6.7 and Flowrate 300
mL/min

190

200.00
180.00
160.00 4
140.00 1
120.00 ~
100.00 ,
80.00 4
60.00 4
40.00 ~
20.00 «

0.00

 

 
 
     

+No Dextran

+ With Dextran
(COD=15000)

Conc. of Phenol (mg/L)

 

 

 

0 20 40 60 80 100 120

Time (min)

Figure 5.22 Effect of Dextran on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

200.00 ~
180.00
160.00 -'
140.00 1
120.00 -
100.00 -
80.00 -+
60.00 ~
40.00 4
20.00 «

0.00

 

  

+ No Dextran

+ With Dextran
(COD=15000)

Conc. of p -Cresol (mg/L)

 

 

 

0 20 40 60 80 100 120

Time (min)

Figure 5 .23 Effect of Dextran on the Oxidation of p -Cresol in the Mixture
of Phenolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

191

100.00

90.00
8000 + NO Dextran

 

700° 1 + With Dextran
60.00 1 (COD=15000)

50.00 «
40.00 a
30.00 5
20.00 a
10.00 e
0.00

Conc. of p -Ethylphenol (mg/L)

 

 

 

 

0 20 40 60 80 100 120
Time (min)
Figure 5.24 Effect of Dextran on the Oxidation of p -Ethylphenol in the

Mixture of Phenolic Compounds at pH 6.7, F lowrate 300
mL/min, and Temperature 22 C

 

 

 

 

 

 

60.00
50-00 “ +No Dextran
g 4000 ‘ +With Dextran
2 (COD=15000)
G
"= 30.00 i
.5
"5
d 20.00 .
8
L)
10.00 «
0.00 t A
0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.25 Effect of Dextran on the Oxidation of Indole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

192

 

 

 

 

 

 

60.00
50.00 ~ +No Dextran
5 40-00 1 +With Dextran
% (COD=15000)
3 30.00 -
m
G.-
c
a 20.00 «
E
O
0
10.00 1
0.00 . i
0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.26 Effect of Dextran on the Oxidation of Skatole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

200.00
180.00
160.00 —
140.00 a

 

+ No Ammonia

120.00 3 +With Ammonia (5000
100.00 - ' my”

80.00 ~
60.00 a
40.00 e
20.00 -

0.00 t
0 20 40 60 80 100 120

Time (min)

Conc. of Phenol (mg/L)

 

 

 

 

Figure 5.27 Effect of Ammonia on the Oxidation of Phenol in the Mixture of
Phenolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

Cone. of p -Cresol (mg/L)

200.00

193

 

180.00

160.00 2
140.00 5
120.00 «
100.00 ‘4
80.00 5
60.00 3
40.00 «
20.00 3

—O— No Anunonia

+ With Ammonia (5000
mg/L)

 

 

0.00

I

20 40 60 80 100

Time (min)

 

 

120

Figure 5.28 Effect of Ammonia on the Oxidation of p -Cresol in the Mixture

Cone. of p -Ethylphenol (mg/L)

of Phenolic Compounds at pH 6.7, Flowrate 300 mL/min, and

Temperature 22 C

 

100.00 .

90.00
3000 ~
70.00 -
60.00 —
50.00 ~
40.00 —
30.00 J
20.00 4
10.00 -

-O- No Ammonia

+ With Ammonia (5000
mg/L)

 

 

0.00

 

20 40 60 80 100

Time (min)

 

120

Figure 5.29 Effect of Ammonia on the Oxidation of p -Ethylphenol in the

Mixture of Phenolic Compounds at pH 6.7, Flowrate 300

mL/min, and Temperature 22 C

194

60.00

 

50.00

+No Ammonia
40.00 4

+With Ammonia (5000 mg/L)
30.00 4

20.00 4

Cone. of Indole (mg/L)

10.00 4

 

 

 

 

0.00

 

0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.30 Effect of Ammonia on the Oxidation of Indole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

 

 

 

 

 

 

60.00
50.00 ~
E0 4000 - +No Ammonia
.2
e . .
E 30.00 g +W1th Ammon1a(5000
m mg/L)
“5
2 20.00 ~
0
D
10.00 4
0.00 . U
0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.31 Effect of Ammonia on the Oxidation of Skatole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

200.00

195

 

180.00 ‘

160.00 4
140.00 4
120.00 4
100.00 4
80.00 4
60.00 -
40.00 4
20.00 -

Conc. of Phenol (mg/L)

 

0.00

+ No VFAs

+With VFAs (240 mM as
Crbon)

 

 

20 40 60 80 100

Time (min)

 

120

Figure 5.32 Effect of VFAs on the Oxidation of Phenol in the Mixture of

Phenolic Compounds at pH 6.7, Flowrate 300 mL/min, and

Temperature 22 C

200.00

 

180.00
160.00 «
140.00 ~
120.00 4
100.00 e
80.00 ~
60.00 ~
40.00 —
20.00 ~

Cone. of p -Cresol (mg/L)

+ No VFAs

+With VFAs (240 mM as Carbon)

 

 

0.00

 

20 40 60 80 100

Time (min)

 

120

Figure 5.33 Effect of VFAs on the Oxidation of p -Cresol in the Mixture of

Phenolic Compounds at pH 6.7, Flowrate 300 mL/min, and

Temperature 22 C

196

 

100.00

90.00
80.00 4 +N° VFAS

70.00 4 +With VFAs (240 mM as
60.00 4 cam“)

50.00 4
40.00 4
30.00 4
20.00 4
10.00 4
0.00

Conc. of p -Ethylphenol (mg/L)

 

 

 

 

0 20 40 60 80 100 120
Time (min)
Figure 5.34 Effect of VFAs on the Oxidation of p -Ethylphenol in the

Mixture of Phenolic Compounds at pH 6.7, Flowrate 300 mL/min,
and Temperature 22 C

 

 

 

 

 

 

60.00
50.00 -
A +No VFAs
E 40.00 4
v +With VFAs (240 mM as
g Carbon)
30.00 4
.5
“5
d 20.00 4
8
L)
10.00 -
0.00 4 4 U
0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.35 Effect of VFAs on the Oxidation of Indole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

197

 

 

 

 

 

 

60.00
50.00 ~
g +No VFAs
E 40.00 —
E +With VFAs (240 mM as
Q
E 30.00 — cam“)
(I)
"5
g 20.00 4
Q
D
10.00 4
0.00 . a
0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.36 Effect of VFAs on the Oxidation of Skatole in the Mixture of
Indolic Compounds at pH 6.7, Flowrate 300 mL/min, and
Temperature 22 C

 

 
  
  
   

 

 

 

200.00
180.00 ~
+ '
A 160.00 q Experimental Data
E4 140.00 — -I—Prediction by Model]
'5 12000 F +PredictionbyModel II (kf=0.8)
1:
0 100.00 4
at":
.5 80.00 ~
g 60.00 —
D 40.00 ~
20.00 4
0.00 t
0 20 40 60 80 100 120
Time (min)

Figure 5.37 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction
at pH 2.1, Flowrate 300 mL/min, and Temperature 22 C

198

 

 

 

 

 

200.00
+Expen'mental Data
,3 150.00 4
E + Prediction by Model I
E +Prediction by Model 11 (kf=0.8)
9 100.00 4
Q.
0—
Q
o
5 50.00 4
u
0.00
0 20 40 60 80 100 120

Time (min)

Figure 5.38 Comparison of p -Cresol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 2.1, Flowrate 300 mL/min, and Temperature

 

 

 

 

 

22 C
100.00
3 +Experimental Data
3,, 80.00 4
E -l— Prediction by Model I
g 60.00 - ‘. +Prediction by Model 11 (kf=0.8)
'2.
'3.
5
:4: 40.00 4
..‘3'
Q
g 20.00 4
Q
u
0.00
0 20 40 60 80 100 120

Time (min)

Figure 5.39 Comparison of p -Ethylphenol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 2.1, Flowrate 300 mL/min, and Temperature
22 C

199

 

60.00

+ Experimental Data
45.00 4 . .

-I— Prediction by Model I

+ Prediction by Model 11
30.00 4 (”:1”)

Cone. of Indole (mg/L)

15.00 -

 

 

 

 

0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.40 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 2.1, Flowrate 300 mL/min, and Temperature 22 C

 

 

 

 

 

60.00
+ Experimental Observance

g “'00 - +Theoretical Model Prediction
‘5’ + Modified Model Prediction
g (kf=1.0)
g 30.00 4
m
"5
d
:1
Q
D 15.00 4

0.00 T s A . . A . ,. Y t

0 5 10 15 20 25 30 35 40

Time (min)

Figure 5.4] Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 2.1, Flowrate 300 mL/min, and Temperature 22 C

200

 

 
 
 
  
 
 
   

 

 

 

200.00

180.00 +Experimental Data
A 160.00 +Prediction by ModelI
3', 140.00 4 +Predictioon by Model 11 (kf=350)
‘: 120.00 4
S
3 100.00 4
a.
“a 80.00 4
a
g 60.00 4
L)

40.00 4

20.00 4

0.()() I T I I I D
0 20 40 60 so 100 120
Time (min)

Figure 5.42 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, Flowrate 100 mL/min, and Temperature 22 C

 

 

 

 

 

200.00
+Experimental Data
g 150.00 4 +Prediction by Model I
g +Prediction by Model 11 (kf=3 50)
25°:
{3 100.00 4
a.
hut
Q
d
5 50 00 4
U o
0.00 I I I I I l
0 20 40 60 80 100 120
Time (min)

Figure 5.43 Comparison of p -Cresol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 6.7, Flowrate 100 mL/min, and Temperature
22 C

201

 

100.00

 
 
 
 

+ Experimental Data
-I- Prediction by Model I
+Prediction by Model H (kf=350)

75.00 4

 

 

Cone. of p -Ethylphenol (mg/L)

 

 

50.00 4
25.00 4
0.00 I I I I I l
0 20 40 60 80 100 120

Time (min)

Figure 5.44 Comparison of p -Ethylphenol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 6.7, Flowrate 100 mL/min, and Temperature

 

 

 

 

 

22 C

200.00
A +Experimental Data
4 150.00 4
E“ + Prediction by Model I
'g + Prediction by Model II (kf=400)
3 100.00 -
n.
h-
a
d
=
5 50.00 4

0.00
0 2O 40 60 80 100 120

Time (min)

Figure 5.45 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, Flowrate 300 mL/min, and Temperature 22 C

202

 

 

200.00
+ Experimental Data
3 150.00 ~
g -I— Prediction by Model I
792 + prediction by Model 11 (kf=400)
5 100.00 ~
.5
G
8
5 50.00 «
0.00

 

 

 

O 20 40 60 80 100 120

Time (min)

Figure 5.46 Comparison of p -Cresol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 6.7, Flowrate 300 mL/min, and Temperature

 

 

 

 

 

22 C
100.00
3 +Experimental Data
h
a 75.00 ~
E + Prediction by Model 1
a
é, +Prediction by Model I] (kf=400)
'3'. 50.00 a
'5
“3
a.
b-
Q
g 25.00 «
O
D
0.00 .
0 20 4O 60 80 100 120

Time (min)

Figure 5.47 Comparison of p -Ethylphenol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 6.7, Flowrate 300 mL/min, and Temperature
22 C

200.00

203

 

150.00 o

100.00 ~

Cone. of Phenol (mg/L)

50.00 1

 

0.00

-O- Experimental Data
-l— Prediction by Model I

+ Prediction by Model 11
(kf=350)

 

20 40 60 80 100

Time (min)

 

 

120

Figure 5.48 Comparison of Phenol Concentration in the Mixture of Phenolic
Compounds between Experimental Data and Model Prediction

at pH 6.7, Flowrate 500 mL/min, and Temperature 22 C

200.00

 

150.00 +

100.00 ‘1

Cone. of p -Cresol (mg/L)

50.00 E

-0— Experimental Data

-I— Prediction by Model I

+ Prediction by Model II
(kf=350)

 

 

0.00

Figure 5.49 Comparison of p -Cresol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 6.7, Flowrate 500 mL/min, and Temperature

22C

20 40 60 80 100

Time (min)

 

 

120

204

100 . 00

 

 

 

 

 

3 +Experimental Data
3 75.00 ~ +Prediction by Model I
g + Prediction by Model 11
g (kf=350)
E» 50.00 1
“3
a.
“a
d 25.00 ~
=
Q
D

0.00

O 20 40 60 80 100 120

Time (min)

Figure 5.50 Comparison of p -Ethylphenol Concentration in the Mixture of
Phenolic Compounds between Experimental Data and Model
Prediction at pH 6.7, Flowrate 500 mL/min, and Temperature

 

 

 

 

 

22 C
60.00
+Experimental Data
a 45 ‘00 A +Prediction by Model I
E
j + Prediction by Model [I
'3 30.00 e , (“=4”)
.5
“a
d
:1
O
D 15.00 c
0.00 - . -
O 20 40 60 80

Time (min)

Figure 5.51 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, Flowrate 100 mL/min, and Temperature 22 C

205

 

 

 

 

60.00
-0- Experimental Data
a 45.00 4
E +Prediction by Model 1
f3 +Prediction by Model 11
g 30.00 « (kf=450)
m
"5
‘s’
L) 15.00 i
0.00
0

 

Time (min)

Figure 5.52 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, Flowrate 100 mL/min, and Temperature 22 C

 

 

 

 

60.00
+Experimenta lData

A 45.00 4
g +Prediction by Model I
E
% +Prediction by Model 11
E 30.00 i (kf=400)
“a
d
'5‘
L) 15.00 ~

0.00 —

0 20 40 6O 80

Time (min)

Figure 5.53 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, Flowrate 300 mL/min, and Temperature 22 C

206

 

 

 

 

60.00
+Experimental Data

A 45.00 o
é +Prediction by Model I
% +Prediction Model II (kf=400)
g 30.00 ~
m
"5
d
8
D 15.00 -

0.00 1

O 20 4O 6O 80
Time (min)

Figure 5.54 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, F Iowrate 300 mL/min, and Temperature 22 C

 

 

 

 

60.00 ;
+Experimental Data
A 45.00 a , ,
g —I— Prediction by Model I
E
g + Prediction by Model II
E 30.00 — (kf=400)
“a
d
a
O
D 15.00 e
0.00 -
0 20 40 6O 80
Time (min)

Figure 5.55 Comparison of Indole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, Flowrate 500 mL/min, and Temperature 22 C

207

 

 

 

 

60.00 L
. +E rimental Data

g 45.00 — x”
E —I-— Prediction by Model I
g + Prediction by Model 11 (kf=400)
g 30.00 i
m
“a
2
5 15.00 4

0.00 —

0 20 4O 60 80

Time (min)

Figure 5.56 Comparison of Skatole Concentration in the Mixture of Indolic
Compounds between Experimental Data and Model Prediction
at pH 6.7, F lowrate 500 mL/min, and Temperature 22 C

 

 

 

 

 

 

30.00
25.00
A +T=15 C
E” 20.00 +T=22 C
2.; +T=30 C
E
g 15.00
H:
“a
d 10.00
E
a
o
5.00
0.00 I I a
o 10 20 3o 40 50
Time (min)

Figure 5.57 Effect of Temperature on the Oxidation of Phenol in Synthetic
Manure by Ozone at pH 6.7 and Flowrate 300 mL/min

 

 

 

208

250.00

 

200.00 +T=15 C

 
 
    

+T=22 C

5—0
KI!
p
O
C

4

+T=30 C

100.00 c

Cone. of p -Cresol (mg/L)

(II
p
O
O
;

 

 

0.00

 

 

I

0 10 20 30 40 50

Time (min)

Figure 5.58 Effect of Temperature on the Oxidation of p -Cresol in Synthetic
Manure by Ozone at pH 6.7 and Flowrate 300 mL/min

 

 

 

 

 

 

E: -o—T=15 C
Q
5} —I—T=22 C
=-
5 +T=3O C
Fa.”
..'}
Q
d
E
Q
o
a
o 10 20 3o 40 50

Time (min)

Figure 5.59 Effect of Temperature on the Oxidation of p -Ethylphenol in
Synthetic Manure by Ozone at pH 6.7 and Flowrate 300 mL/min

209

 

 

 

 

 

 

 

5.00
4.50
a 4.00 +T=15 C
a 3'50 ‘ +T=22 C
‘3 3.00 ~
'8 +T=30 C
'6 2.50 «
.5
.5 2.00
'a’ 1.50
O
0 1.00
0.50
0.00 1%
0 10 20 30 40 50

Time (min)

Figure 5.60 Effect of Temperature on the Oxidation of Indole in Synthetic
Manure by Ozone at pH 6.7 and Flowrate 300 mL/rnin

 

14.00 '

12.00

10.00

8.00 c

6.00 ~

4.00 -

Cone. of Skatole (mg/L)

2.00 ~

 

 

 

0.00

 

0 10 20 30 40 50

Time (min)

Figure 5.61 Effect of Temperature on the Oxidation of Skatole in Synthetic
Manure by Ozone at pH 6.7 and Flowrate 300 mL/min

210

 

  
  
  

 

 

 

 

30.00
A 25.00 ‘ +Q=100 mL/min
5 20.00 . +Q=300 mL/min
E +Q=500 mL/min
g 15.00 e
E
5

10.00 .
‘e’
:3

5.00 -

0.00 .

0 20 40 60 so 100

Time (min)

Figure 5.62 Effect of Flowrate on the Oxidation of Phenol in Synthetic
Manure at pH 6.7 and Temperature 22 C

 

  
  
  

 

 

 

 

250.00
+Q=100 mL/min
a 200 00
E +Q=3OO mL/min
'3 150.00 - +Q=500 mL/min
“.9
“3‘ 100.00 4
d
8
Q 50.00 -
0.00 .
0 20 40 60 80 100
Time (min)

Figure 5.63 Effect of Flowrate on the Oxidation of p -Cresol in Synthetic
Manure at pH 6.7 and Temperature 22 C

211

 

+Q=100 mL/min
+Q=300 mL/min
+Q=500 mL/min

Cone. of p -Ethylphenol (mg/L)

 

 

 

 

E %
0 20 40 60 80 100

Time (min)

Figure 5.64 Effect of F lowrate on the Oxidation of p -Ethylphenol in Synthetic
Manure at pH 6.7 and Temperature 22 C

 

 

 

 

 

a +Q=100 mL/min
a +Q=300 mL/min
g +Q=500 mL/min
.5
"5
d
'5'
u
n a
0 20 40 6O 80 100

Time (min)

Figure 5.65 Effect of Flowrate on the Oxidation of Indole in Synthetic
Manure at pH 6.7 and Temperature 22 C

 

 

212

 

 

 

 

 

14.00 ,
12.00
% +Q=100 mL/min
10.00 .
g +Q=300 mL/min
2
g 8'00 " +Q=500 mL/min
m 6.00 «
'0-
°- l
g 4.00
c
u
2.00 -
0.00 - a It
0 20 40 60 80 100

Time (min)

Figure 5.66 Effect of Flowrate on the Oxidation of Skatole in Synthetic
Manure at pH 6.7 and Temperature 22 C

 

 

 

 

 

30.00 ~
A 25.00 +Q=100 mL/min
EL +Q=300 mL/min
E 20.00 ~ .
: +Q=500 ern
G
E
g 15.00J
A:
“a 10 00
a - -
5
o

5.00 o

0.00 _ I t

0.00 0.50 1.00 1.50 2.00

Ozone Dosage (g/L)

Figure 5.67 Effect of Ozone Dosage on the Oxidation of Phenol in Synthetic
Manure by Different Flowrates at pH 6.7 and Temperature 22 C

213

 

 

 

 

 

250.00 ,
:5 200.00 +Q=100 mL/min
E +Q=300 mL/min
E 150.00 4 +Q=500 mL/min
2
9
€- 100.00 ~
E
O 50.00 —

0.00 i , or
0.00 0.50 1.00 1.50 2.00

Ozone Dosage (g/L)

Figure 5.68 Effect of Ozone Dosage on the Oxidation of p -Cresol in
Synthetic Manure by Different Flowrates at pH 6.7 and
Temperature 22 C

 

+ Q= 100 mL/min
—I— Q=300 mL/min
+ Q=500 mL/min

Cone. of p-Ethylphenol (mg/L)
N
O
O

 

 

 

 

0.00 0.50 1.00 1.50 2.00
Ozone Dosage (g/L)

Figure 5.69 Effect of Ozone Dosage on the Oxidation of p -Ethylphenol in
Synthetic Manure by Different Flowrates at pH 6.7 and
Temperature 22 C

214

 

5.00
4.50
4-00 +Q=100 mL/min
3.50
+Q=300 mL/min

+Q=500 mL/min

Cone. of Indole (mg/L)
N
U!
O

 

 

 

 

0.00 0.50 1.00 1.50 2.00
Ozone Dosage (g/L)

Figure 5.70 Effect of Ozone Dosage on the Oxidation of Indole in Synthetic
Manure by Different Flowrates at pH 6.7 and Temperature 22 C

 

14.00

”'00 -o—Q=100 mL/min

”-00 +Q=300 mL/min
8.00 +Q=500 mL/min

Cone. of Skatole (mg/L)

 

 

 

 

0.00 0.50 1.00 1.50 2.00
Ozone Dosage (g/L)

Figure 5.71 Effect of Ozone Dosage on the Oxidation of Skatole in Synthetic
Manure by Different Flowrates at pH 6.7 and Temperature 22 C

215

 

 

 

 

 

E 120 4 ~10:
U
a 90 —
E
5
g 60 I
O
D
<
g 30 J
O I f fl
0 10 20 30 40
Time (min)

Figure 5.72 Change of VFAs Concentration in Synthetic Manure by
Ozonation at pH 6.7, Flowrate 300 mL/min, and Temperature 22 C

 

  
  
  

 

 

 

 

 

30.00
+Experimental Data
25.00
a +Prediction by Model I
E 20'0” ‘ +PredictionbyModel 11 (kf=300)
'5
9 15.00 4
a
“5
d 10.00 4
H
O
u
5.00 4
0.00 r 1
0 20 40 60 80 100

Time (min)

Figure 5.73 Comparison of Phenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 100 mL/min, and Temperature 22 C

-—b

 

 

250.00

Cone. of p -Cresol (mg/L)

200.00 ‘

150.00

100.00

50.00

0.00

216

 

+ Experimental Data

  
  
  

-I— Prediction by Model I

I

+Prediction by Model II ch=3 00)

l

 

 

 

f

0 20 40 60 80

 

   

Time (min)

100

Figure 5.74 Comparison of p -Cresol Concentration in the Synthetic Manure

Cone. of p -Ethylphenol (mg/L)

3.50

between Experimental Data and Model Prediction at pH 6.7,
Flowrate 100 mL/min, and Temperature 22 C

 

     

 

 

 

 

 

3-00 ‘ +Experimental Data
2.50 4 +Prediction by Model I
2.00 4 +Prediction by Model 11 (kf=300)
1.50 4
1.00 e
0.50 ~
0.00 4|:
0 20 40 6O 80

Time (min)

Figure 5.75 Comparison of p -Ethylphenol Concentration in the Synthetic
Manure between Experimental Data and Model Prediction at pH

6.7, Flowrate 100 mL/min, and Temperature 22 C

100

 

 

 

217

 

 

 

 

 

 

5.00
+E rimental Data
A 4.00 m
fi -I— Prediction by Model I
5
:5» 3-00 +Prediction by Model 11 (kf=300)
a
.5
“a 2.00
2
6
1.00
0.00 I E
0 20 40 60 80 100
Time (min)

Figure 5.76 Comparison of Indole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
F lowrate 100 mL/min, and Temperature 22 C

 

 

 

 

 

 

 

12.00
10.00 +Experimental Data
5 8.00 +Predrct1on by Model I
j; +Prediction by Model 11 (kf=300)
f: 6.00
(I)
"5
d 4.00
:1
G
D
2.00 —
0.00 fl i
0 20 4O 60 80 100
Time (min)

Figure 5.77 Comparison of Skatole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 100 mL/min, and Temperature 22 C

218

 

 

 

 

 

 

 

30.00
25.00 + Experimental Data
El: 2000 -I- Prediction by Model I
‘g +Prediction by Model 11 (kf=400)
g 15.00
a.
“5
g 10.00
0
D
5.00
0.00 I E
0 20 40 60 80 100
Time (min)

Figure 5.78 Comparison of Phenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

 

 

 

 

 

 

250.00

+E rimental Data

a 200.00 xpe

E -I— Prediction by Model I

I; 15000 ‘ +Prediction by Model II (kf=400)

‘1’

a? 100.00 4

d

8

D 50.00 e

0.00 I I
0 20 40 60 80 100

Time (min)

Figure 5.79 Comparison of p -Cresol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

219

 

 

 

 

 

3.50
i 3'00 —O—Experimental Data
5 2 50 4
"g ' -I— Prediction by Modell
0
if: 2'00 ‘ +PredictionbyMode1 II (kf=400)
g 1.504
Q.
"5 1.00J
d
‘5
u 0.50 5

0.00 4 . . . ~ r I

0 20 40 60 80 100

Time (min)

Figure 5.80 Comparison of p -Ethylphenol Concentration in the Synthetic
Manure between Experimental Data and Model Prediction at pH
6.7, Flowrate 300 mL/min, and Temperature 22 C

 

 

 

 

 

5.00
a 4'00 + Experimental Data
5 +Prediction by Model I
o 3.00
g +Prediction by Model II (kf=400)
‘8 2.00
d
a
6

1.00

0.00 fl E

0 20 4O 60 80 100

Time (min)

Figure 5.81 Comparison of Indole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

220

 

 

 

 

 

12.00
A 1000 +Experimental Data
g 8.00 +Prediction by Model I
0
g + Prediction by Model II (kfi300)
.5 6.00
m
.8 4 00
a .
8
D

2.00

0.00 J I

0 20 40 60 80 100
Time (min)

Figure 5.82 Comparison of Skatole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

 

 

 

 

 

 

30.00
”'00 +Experimental Data
§ 20.00 +Prediction by Model I
'2‘ +Prediction by Model II (kf=600)
g 15.00
a.
"5
d 10.00
i:
G
D
5.00
0.00 F I
0 20 40 60 80 100
Time (min)

Figure 5.83 Comparison of Phenol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 500 mL/min, and Temperature 22 C

221

 

 

 

 

 

250.00
a 200.00 . +Experimental Data
5 -I- Prediction by Model I
'§ 150.00 ~ . .
2 . +Predlcuon by Model II (kf=600)
“.9
.5 100.00
Q
g
L) 50.00
0.00 ‘ ‘ I I l I
0 10 20 30 4O 50 60 70 80 90

Time (min)

Figure 5.84 Comparison of p -Cresol Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
F lowrate 500 mL/min, and Temperature 22 C

 

 

 

 

 

 

3.50
"‘ 3.00 .
é +Expenmental Data
5’ 2.50 ~ . .
S +Predlctlon by Model I
i 2.00 4 . .
'5'. +Predlctlon by Model II (kf=600)
.:
as 1.50 4
.5
c 1.00 o
d
8
Q 0.50 5

0.00 4 .— E

O 20 40 60 80 100
Time (min)

Figure 5.85 Comparison of p -Ethylphenol Concentration in the Synthetic
Manure between Experimental Data and Model Prediction at
pH 6.7, Flowrate 500 mL/min, and Temperature 22 C

222

 

 

 

 

 

5.00 l
A 4_00 - +Experimental Data
E. +Prediction by Model I
'5'; 3'00 4 +Prediction by Model II (11%600)
e
.5
"5 2.00 4
d
a
5

1.00 4

0.00 ~ i I

0 20 40 60 80 100

Time (min)

Figure 5.86 Comparison of Indole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 500 mL/min, and Temperature 22 C

 

 

 

 

 

 

12.00
10.00
g +Experimental Data
5 3-00 ‘ -I-Predictionby Model 1
0
O
3 6.00 _ +Prediction by Model II (kf=600)
2
g 4.00 -
a
Q
D
2.00 ~
0.00 4 l a
0 20 4O 60 80 100
Time (min)

Figure 5.87 Comparison of Skatole Concentration in the Synthetic Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 500 mL/rnin, and Temperature 22 C

223

200.00
180.00
160.00 ~
140.00 -
120.00 4
100.00 ~
80.00
60.00

 

 

Cone. (mg/L)

 

 

 

Time (min)

+ Phenol + p-Cresol + p-Ethylphenol —)(— Skatole

Figure 5.88 Effect of Oxygenation on the Concentration of Phenolic and
Indolic Compounds in Real Swine Manure at Flowrate 300

 

 

 

 

 

mL/min and Temperature 22 C
200.00
150.00 ~
‘3’ 100.00 o
8'
D
50.00 4
~:~.
0.00 r . —"
0 20 40 6O 80 100 120 140 160

Time (min)
+Phenol +p-Cresol +p-Ethylphenol -)(- Skatole

Figure 5.89 Effect of Ozonation on the Concentration of Phenolic and
Indolic Compounds in Real Swine Manure at Flowrate 300
mL/min and Temperature 22 C

 

 

 

224

 

80.00

 
  
  
 

 

 

 

 

-O—Experimental Data
A 60.00 «
é +Prediction by Model I
‘2' +Prediction by Model H %40,000)
5.3 40.00 4
Q:
"5
d
:l
O
D 20.00 4
0.00
0 50 100 150

Time (min)

Figure 5.90 Comparison of Phenol Concentration in the Real Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

 

200

 

 

200.00
+Experimental Data
A 150.00 ‘
é -l—Prediction by Model 1
E: + Prediction by Model 11 04:“,000)
9 100.00 4
5:
Q
:3
r:
5 50.00 4
0.00

 

0 20 40 60 80 100 120 140 160
Time (min)
Figure 5.91 Comparison of p -Cresol Concentration in the Real Manure

between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/rnin, and Temperature 22 C

 

 

225

 

 

 

 

 

 

30.00
g + Experimental Data
5
:5 20.00 ‘ +Prediction by Model I
s:
i +Prediction by Model II (kf=40,000)
'3'.
5
“J
42' 10.00 4
d
r:
O
L)

0.00
0 50 100 150 200

Time (min)

Figure 5.92 Comparison of p -Ethylphenol Concentration in the Real Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

 

 

 

 

 

 

15.00
-O-Experimental Data
23‘ . .
E 10.00 _ +Predlctlon by Model I
A; +Prediction by Model II (kf=40,000)
E
(D
“a
g 5.00 4
O
L)
0.00 - ~
0 50 100 150 200
Time (min)

Figure 5.93 Comparison of Skatole Concentration in the Real Manure
between Experimental Data and Model Prediction at pH 6.7,
Flowrate 300 mL/min, and Temperature 22 C

Chapter 6

CONCLUSIONS AND RECOMMENDATIONS

6.1 Conclusions

The comprehensive objective of this research work was to investigate and
understand the formation of malodorous compounds in the swine manure during storage
and to explain the process by which ozone controls the malodors of swine manure. The
major purposes of this study were to find the minimum ozone dosage that was effective at
reducing the concentrations of malodorous metabolites in the swine manure, to determine
the major physical and chemical parameters that affected effectiveness of the ozonation
process, to establish design guidelines for operation of the pilot-scale system, and to
develop the oxidation model that could be used to predict the concentrations of phenolic
and indolic compounds from the synthetic and real swine manure.

The malodorous compounds in swine manure are mainly formed by anaerobic
decomposition of their precursors, such as proteins and fibers. Malodorants, including
phenolic and indolic compounds, sulfides, ammonia, and volatile fatty acids, were
accumulated during a storage period of 3 months. Ozone was found to reduce the odor
intensity to an acceptable level using the dosage of 1g/L in the lab-scale reactor.
Concurrent with the decrease in the odor intensity during ozonation process, is the

decrease in the concentrations of phenolic and indolic metabolites and sulfides. The

226

 

 

 

227

concentrations of other odorous compounds, however, such as ammonia and volatile fatty
acids, remain unchanged. These results suggest that ammonia and VFAs play a minor
role in contributing to the odors released from the swine manure.

E. Coli and total coliform were destroyed by ozonation. The aerobic and anaerobic
bacteria were also reduced by about 1-2 orders of magnitude at an ozone dosage of 1 g/L.
These results indicate that ozone has the potential to kill or inhibit pathogenic
microorganisms. Most importantly, the malodors will not reoccur after subsequent
storage of ozonated swine manure. This also implies that the land spreading of ozonated
swine manure will not result in major odor releases even after several weeks of post-
treatment storage. As the nutrient content is unchanged by ozonation, the manure still
maintains its nutrient value.

When the ozonation process was enlarged to pilot-scale, the effective ozone dosage
of 0.5 g/L was sufficient to bring the odor of the swine manure to an acceptable level.
This lower value was made possible by the new reactor design which enhanced ozone
mass-transfer in the liquid manure by using a venturi injector and an agitation system.
Temperature variations (12-25 °C) did not affect the efficiency of the ozonation process.
This implies that the effect of seasonal change will be insignificant. The addition of
hydrogen peroxide to ozone did not provide additional oxidation of the malodorous
substances. This is because the manure probably contained radical scavengers that
consumed the free radicals generated by the reaction of ozone and hydrogen peroxide.
Therefore, the advanced oxidation processes may not apply to the treatment of swine

manure in reducing the malodors.

228

An economic analysis was conducted for the pilot-scale system. It appears that
$3.80 will be deducted from the profit of selling a 220 lb pig into the market. Therefore,
any attempt to reduce the cost of treating the manure will be appreciated by the swine
producers. The first alternative is to use the PSA (pressure swing adsorption) for oxygen
generation to replace the use of liquid oxygen. Approximately one half of the operational
costs are for oxygen cylinder rental. Although the capital cost will increase by using PSA
system, the operational cost may be decreased since PSA only requires a small amount of
electricity. The second alternative is to install the ozone diffuser directly into the lagoon.
By this idea, the capital cost will be reduced significantly since no reactor is required.
However, we have to evaluate the efficiency of using in-situ ozonation in the lagoon to
control the malodors and the effect of ozone on the ecology of bacteria and
microorganisms in the lagoon.

In the mass-transfer study, a model that considered the effect of liquid-gas
interaction and headspace on the mass-transfer of ozone was developed to evaluate the
mass-tansfer and partition coefficients in both lab-scale and pilot-scale reactors.
Temperature is found to be the key factor in influencing the partition coefficient, the most
sensitive factor in affecting the concentration profiles of dissolved ozone and gaseous
ozone. The factors affecting the mass-transfer coefficient include flowrate, agitation
speed, ionic strength, pH, and temperature.

In kinetic studies, the reaction rate constants for phenolic and indolic compounds
were pH dependent. At high pH, the reaction rates of these compounds with ozone
increased significantly. The stoichiometric factors for phenol is 2 and for p-cresol, p-

ethylphenol, indole, and skatole is 1. In the study of synthetic manure, it was found that

229

the decrease in the concentrations of phenolic and indolic compounds only depends on
the ozone dosage because the concentrations of the dissolved ozone in the solution and
gaseous ozone in the outlet stream could not accumulate until those compounds were
removed. Finally, a dynamic model incorporating the mass transfer of ozone with the
oxidation kinetics was developed to predict the degradation of phenolic and indolic
compounds in the synthetic and real swine manure. It was noted that the reactions
between ozone and byproducts or impurities should be considered since the
concentrations of the target compounds predicted by the model were reduced much faster
than those observed in the experiments. Therefore, a reaction coefficient was defined and
included in the model to simulate the effect of side reactions on the concentrations of
phenolic and indolic compounds. The reaction coefficients found in synthetic and real
manure could be related to an easily determined gross parameter (COD). This is
especially important if the mathematical relationship can be established, then we can

predict the concentrations of malodorous compounds using ozone treatment.

6.2 Recommendations
This study has demonstrated the ability of ozone to reduce the malodors of swine
manure using lab-scale and pilot-scale systems. Some recommendations are encouraged
to do for the future research.
1. Although ozone has the ability to oxidize the malodorous compounds in the swine
manure, high organic loading is still a problem for the protection of groundwater.
Therefore, a complete treatment process (for example: preozonation and biological

treatment) should be developed to solve the problem.

 

 

230

2. In-situ ozone treatment in a lagoon appears to be a promising technology to control
the formation of malodors and to avoid the expense of installing an ozonation reactor.
It is worthwhile to evaluate the efficiency of ozonation on lagoon odors and
management.

3. Based on the successful use of the oxidation model in predicting the degradation of
phenolic and indolic compounds in swine manure, one may modify the reaction
coefficient for use with a gross parameter, such as COD, to obtain more precise
coefficient for general use.

4. The sulfur-containing compounds contribute to the malodorous characteristics of the
swine manure. Therefore, a GC detector (FPD) should be purchased for evaluating

the reduction of the sulfur-containing metabolites by ozone.

APPENDIX

231

Program Used for the Repeated Measures Analysis of Variance in SAS System

data ozone;
input person trtmt smellO smelll smell2 smell3;
cards;

112233
123332
131221
141212
151111

212232
223223
231112
241111
251111

311323
321233
331122
341111
351111

412333
422332
432211
441111
451111

512233
521222
531111
541111
551112

e
9

*Proc glm;

*class person trtmt;

*model smellO - smell3 = person trtmt;

*repeated time 4 (O 1 2 3) contrast (1)/summary printh printe;

 

 

232

Proc glm;

class person trtmt;

model smellO - smell3 = person trtmt;

repeated week 4 (O l 2 3)/summary printh printe;

contrast ‘0 vs .25 g/L ’ trtmt l —1 O O 0;
contrast ‘0 vs .5 g/L ’ trtmt 1 O —1 O O;
contrast ‘0 vs .75 g/L ’ trtmt l O O -l O;
contrast ‘Ovs lg/L ’trtmtl O O O -1;

contrast ‘0.25 g/L vs 0.5 g/L ’ trtmt O 1 -1 O 0;
contrast ‘0.25 g/L vs 0.75 g/L ’trtmt 0 l 0 -1 O;
contrast ‘0.25 g/L vs 1 g/L ’ trtmt O 1 0 0 -1;

contrast ‘O.50 g/L vs 0.75 g/L ’ trtmt O O l -1 O;
contrast ‘O.50g/Lvs lglL ’trtmtO O 1 O -l;