Oxidation of Sulfolane in Aqueous Systems by Chemical and Photochemical Processes

University of Calgary PRISM: University of Calgary's Digital Repository

Graduate Studies The Vault: Electronic Theses and Dissertations

2019-04-08 Oxidation of Sulfolane in Aqueous Systems by Chemical and Photochemical Processes

Izadifard, Maryam

Izadifard, M. (2019). Oxidation of sulfolane in aqueous systems by chemical and photochemical processes (Unpublished doctoral thesis). University of Calgary, Calgary, AB. http://hdl.handle.net/1880/110148 doctoral thesis

University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca

UNIVERSITY OF CALGARY

Oxidation of Sulfolane in Aqueous Systems by Chemical and Photochemical Processes

Maryam Izadifard

A THESIS

SUBMITTED TO THE FACULTY OF GRADUATE STUDIES

IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSPHY

GRADUATE PROGRAM IN CIVIL ENGINEERING

CALGARY, ALBERTA

APRIL, 2019

ABSTRACT

In this research degradation of sulfolane in spiked water, contaminated ground water and soil wash water was investigated by several oxidative methods. Sulfolane is an organosulfur compound, which is commonly used for liquid- liquid aromatics extraction from mixtures containing aliphatic hydrocarbons and in Sulfinol® process for liquid natural gas treatment. Due to its large production, a significant amount of waste containing sulfolane is annually produced. Beyond this, over many years of operation, there have been some unpredicted or accidental spills, leaks from extraction units during processing as well as leachates from disposal areas from producing wells and unlined storage ponds, which have caused contamination of soil, ground water and wetland ecosystem around gas processing plants. The natural attenuation processes are not effective in sulfolane removal as they are quite slow. Based on the toxicological studies, sulfolane is considered as an emerging industrial contaminant, which should be removed from the environment.

In this study, three different oxidative methods were evaluated for sulfolane degradation and possible adaptation for field application. This study builds on previous investigations conducted on application of oxidative methods and wherever relevant these have been referenced. The first oxidative method evaluated in this study was oxidation of sulfolane using ammonium persulfate

(APS) along with ultraviolet light (UVC) and/ or bubbling ozone in spiked water with sulfolane as well as in ground water samples. To the best of our knowledge the synergistic effect of O3 and UV irradiation on activation of persulfate has not been investigated for degradation of any organic compounds prior to this study.

This study demonstrated that persulfate along with UVC and UVC/O3 can efficiently degrade sulfolane in water. More than 90% removal was achieved after 35 min and 10 min

-1 respectively. Presence of 5 mg L O3 in solution not only increased the rate of sulfolane removal

(by up to three times) but also decreased at least 25% of the required dosage of persulfate. In general, at higher pHs than 6.9 the reactions were slower, and the quenching effect of carbonate seemed to be significant. Chloride at concentrations lower then 100 mg L-1 had no effect on reaction rate. The application of these methods was also tested for ground water samples collected from a sulfolane impacted site. For 90% sulfolane degradation in groundwater 60 and 22 min

-1 irradiation was required in presence of 3 g L of APS for APS/UVC and APS/UVC/O3 systems respectively.

The second oxidative method that was evaluated was the application of CaO2/O3 and

CaO/O3 for mineralization of sulfolane in aqueous systems. This study demonstrated that the application of calcium peroxide or lime along with O3 is a viable and effective method for

-1 treatment of sulfolane in water and groundwater. If 1.6 g L of oxidants (CaO2 and CaO) were

-1 used along with 0.5 L min of O3 in a batch reactor, sulfolane and TOC were totally removed in less than 40 and 150 min, respectively. Once these conditions were established and optimised in the lab, field experiments were designed and evaluated to treat contaminated ground water samples. The field tests were successful in treating sulfolane with TOC removal within 150 min and after 4 h, respectively.

Reduced treatment time compared to UV/O3 system, applicability of lime, which is readily available, negligible matrix effect and the potential for complete mineralization of sulfolane, make

CaO/O3 or CaO2/O3 a practical method for sulfolane treatment. Among the methods tested for sulfolane remediation, this method is the only AOP method, which can be used in situ for treatment of groundwater contaminated with sulfolane.

A mechanistic study was also performed for CaO2/O3 system compared to that of

NaOH/O3. The proposed degradation pathways were quite similar. The more efficient TOC

iii

2+ removal in case of CaO2/O3 was related to complexation of Ca with oxidized sulfolane by- products. Involvement of CaCO3 as a solid formed during degradation of sulfolane, in a catalytic ozonation process, was not supported by the experimental results. Not only CaCO3 but also several other solids such as MgO, silica, zeolite and different types of activated carbons were inefficient in degradation of sulfolane along with O3. Only two types of carbon showed positive results, but the overall results were not consistent.

Both these two oxidative methods (APS/UV/O3 and CaO/O3) were effective in treating sulfolane in soil washwater. Therefore, they can be combined with a soil washing/flushing process as a treatment method for contaminated soils.

The third oxidative method tested in this study was based on application of aqueous

− − chlorine (Cl2, OCl , Cl ) along with UVC/UVB. Only hypochlorus acid (if it was added stepwise) along with UVC was effective in sulfolane degradation in water samples. This prevented the quenching effect of HOCl on reactive oxidative species. There is a possibility of formation of chlorinated by-products in this case and the presence of natural organic matter (NOM) might further complicate the application of this treatment method. Degradation of sulfolane under UVA and visible light irradiation was negligible.

While, aqueous chlorine in combination with longer wavelength ultraviolet light/visible light was not effective for sulfolane removal, it has been considered an Advanced Oxidation method for treatment of water and wastewater in the litearure as a result of production of hydroxyl radicals. Proposed hydroxyl radical production requires efficient long wavelength UV/visible light absorption of chlorine species in presence of high concentration of coloured Natural Organic

Matter (NOM); which creates some ambiguity on the reaction mechanism. Therefore, a mechanistic study was performed, and it was found that the chlorine species may be exploited to

iv degrade many organic contaminants using UVA and visible light excitation. Hydroxyl radicals are not necessary for degradation of contaminants in presence of coloured compounds in water.

Involvement of a photosensitization process for sulfolane degradation was not also evident as reactive oxygen species were not produced efficiently.

ACKNOWLEDGEMENTS

The research presented in this thesis could not have been successfully completed without the guidance and support of many individuals during my PhD study.

First, I express my sincere gratitude to my supervisor, Dr. Gopal Achari, Professor in Civil

Engineering Department and Associate Dean of Graduate Studies and Research in Schulich School of Engineering. He has given his heartful support, invaluable guidance and encouragement to me during this time. His patience and understanding gave me the opportunity to finish my study along with all my commitments not only to my family but also to my teaching career, which was very valuable to me.

I am deeply grateful to my co-supervisor Dr. Langford, Professor in Chemistry Department, who unexpectedly passed away last year. His immense knowledge and willingness to share his expertise as well as his great personality were of incredible value to me.

Special thanks go to Dr. Tay and Dr. Habibi for agreeing to be in my committee and their helpful comments and suggestions during my candidacy exam.

My special appreciation goes to Mr. Daniel Larson, our lab technologist, whose technical support, guidance and patience throughout the research contributed to its successful completion. I would also like to acknowledge and thank Mr. White at the Department of Chemistry for running MS samples.

I would also like to thank our group members Mitra Mehrabani, Nahid Hassanvand Gandaie, Elena

Vialykh, Mohammad Khan, Jordan Hollman and Linlong Yu for their cooperation in sharing knowledge and laboratory facilities. I am grateful to know them, and I have been always benefitted from their help.

I would also like to thank my friends in Calgary for their warm support and my family back home in Iran. My special gratitude goes to my parents for always believing in me and, my sisters for their emotional supports and for taking over my responsibilities. Last but not least I would like to deeply thank my husband for supporting me during all these years, without him I couldn’t complete this work and, mostly my girls Niloofar and Niki for accepting me as a busy mom. I will make up to you, I promise!

vii

To my beloved family Afshin, Niloofar & Niki

viii

Table of Content

Abstract……………………………………………………………………………………………ii Acknowledgements…………………………………………………………………………….... vi Dedication……………………………………………………………………………………… viii Table of contents………………………………………………………………………………….ix List of Tables………………………...…………………………………………………………. List of Figures and illustrations………………………………………………………………....xvi List of Symbols, Abbreviations and Nomenclature………….…………...………………….....xxii

CHAPTER ONE: INTRODUCTION 1.1 Background……………………………………………………………………………………1 1.2 Research Goals and Objectives………………………………………………………………..2

1.3 Thesis Overview…………………………………………………………………………….....4

CHAPTER TWO: LITERATURE REVIEW 2.1 Sulfolane………………………………………………………………………………………7 2.1.1 Physiochemical Properties of Sulfolane…………………………………………………..7 2.1.2 Synthesis and Manufacturing……………………………………………………………8 2.1.3 Storage of Sulfolane……………………………………………………………………...10 2.1.4 Applications……………………………………………………………………………11 2.1.5 Toxicity…………………………………………………………………………………..13 2.1.6 Environmental Problems of Sulfolane…………………………………………………...15 2.1.7 Natural Attenuation of Sulfolane………………………………………………………...17 2.1.8 Remediation……………………………………………………………………………...18 2.1.9 Sulfolane Analysis………………………………………………………………………..20

2.2 Hydroxyl Radical and Sulfate Radical Based Advanced Oxidation Processes……………….20

2.2.1 Contaminated Water/Wastewater Treatment Using AOPs……………………………….23 2.2.1.1 Effect of Interferences on AOP…………………………………………………….25

2.2.1.2 Electrical Energy Per Order (EEO)…………………………………………………26

2.3 Ozonation ……………………………………………………………………………………27

2.3.1 Mechanism of Ozonation………………………………………………………………...28 2.3.2 Production of Ozone……………………………………………………………………..30

2.3.3 Disadvantages of Ozone………………………………………………………...……….31 2.3.4 Increase of Ozonation Efficiency………………………………………………………...32

2.3.4.1 Ozonation at High pHs…………………………………………………………….32

2.3.4.2 Ozonation in Presence of Oxidants Such as Hydrogen Peroxide or Persulfate...…32

2.3.4.3 Ozonation Aided by UV Light ………………………………………………...….34

2.3.4.4 Catalytic Ozonation……………………………………………………………….35

2.3.4.4.1 Mechanism of catalytic ozonation…………………………………...…..35 2.3.4.4.1.1 Homogeneous Catalytic Ozonation…………………………...36 2.3.4.4.1.2 Heterogeneous Catalysis………………………………………41 2.3.4.4.2 Advanced oxidation Processes (AOPs) and Catalytic Ozonation…...…….47 2.3.4.4.3 Limitations of Catalytic Ozonation………………………………………..48 2.4 Persulfate……………………………………………………………………………………..49 2.4.1 Persulfate activation …………………………………………………………………….51 2.4.1.1 Heat Activation……………………………………………………………………51 2.4.1.2 UV Light Activation……………………………………………………………...52 2.4.1.3 Metal Activation using Transition Metals or Chelated Metals…………………...52 2.4.1.4 Alkali Activation……………………………………………………………….....53 2.4.1.5 Other Oxidant Activation…………………………………………………………55 2.4.1.6 Carbon Activated Persulfate……………………………………………………...56 2.4.2 Overall Comparison of Activation Methods……………………………………………57 •- 2.4.3 Quenching of SO4 ……………………………………………………………………..57 2.5 AOP using UV/Photoactive Chlorine Species……………………………………………….58 2.5.1 Comparison of Chlorine/UV Systems With Other AOPs………………………………..61 2.6 Photosensitization……………………………………………………………………………62 2.7 UV Light Sources……………………………………………………………………………63

2.7.1 Low Pressure Mercury Arc Lamp ………………………………………………………65

2.7.2 LED Light Sources……………………………………………………………………...67

2.7.3 The Ferrioxalate Actinometer ………………………………………………………….68

CHAPTER THREE: DEGRADATION OF SULFOLANE USING ACTIVATED PERSULFATE WITH UV AND UV- OZONE 3.1 Introduction…………………………………………………………………………………..70 3.2 Experimental Methods and Materials………………………………………………………..73 3.3 Results and Discussion.……………………………………………………………………...76 3.3.1 Degradation of Sulfolane using Ammonium Persulfate Activated with UVC…….…...76 3.3.1.1 Effects of Persulfate Dose and UVC Light Intensity on Sulfolane Degradation...76 3.3.1.2 Effect of Initial pH on Sulfolane Degradation with Persulfate And UV…...…....79 3.3.1.3 Effect of Carbonate on Degradation of Sulfolane using APS/UVC……………..81 3.3.1.4 Effect of Carbonate-Bicarbonate on Degradation of Sulfolane Using APS/UVC… …………………………………………………………………………………………...83 3.3.1.5 Effect of Chloride on Degradation of Sulfolane using APS/UVC……………….84

3.3.2 Degradation of Sulfolane using Ammonium Persulfate/O3/UVC……………………...85 3.3.2.1 Effect of Persulfate Dosage on Sulfolane Degradation and Control Experiments...87

3.3.2.2 Effect of Initial pH on Sulfolane Degradation with Persulfate/UVC/O3……...... 92

3.3.2.3 Effect of Carbonate on Degradation of Sulfolane Using APS/O3/UVC…………...94

3.3.2.4 Effect of Chloride on Degradation of Sulfolane Using UV/APS/O3………………95 3.3.3 Degradation of Sulfolane using Persulfate and UVA Lamps…………………………..95

3.3.4 Sulfolane Treatment in Well Water Samples using APS/UVC and APS/O3/UVC…….97 3.4 Conclusions…………………………………………………………………………………98

CHAPTER FOUR: MINERALIZATION OF SULFOLANE IN AQUEOUS SOLUTIONS BY

OZONE/CaO2 AND CAO/OZONE WITH POTENTIAL FOR FIELD APPLICATION 4.1 Introduction………………………………………………………………………..……….100 4.2 Experimental Materials and Methods…………………………………………………...….103 4.3 Results and Discussion……………………………………………………………..……....105

4.3.1 Degradation of Sulfolane Using CaO2/O3 - Different Loadings of CaO2, Effect Of UV Irradiation and Blank Experiments…………………………………………………….105

4.3.2 Degradation of Sulfolane using Lime/O3……………………………………………...113 4.3.3 Degradation of Sulfolane in Contaminated Groundwater Samples (Laboratory Tests).115 4.3.4 Degradation of Sulfolane In Contaminated Groundwater Samples (Field Tests)……..116 4.4 Conclusions…………………………………………………………………………………119

CHAPTER FIVE: MECHANISM OF SULFOLANE DERADATION USING CAO2/UV AND CATALYTIC OZONATION

5.1 Mechanism of Sulfolane Degradation using CaO2/UV……………………………………..120 5.1.1 Conclusions…………………………………………………………………………….124 5.2 Catalytic Ozonation for Sulfolane Degradation…………………………………………….124

5.2.1 Conclusions…………………………………………………………………………….130

CHAPTER SIX: TREATMENT OF SULFOLANE IN SOIL BY COMBINATION OF SOIL

WASHING/FLUSHING AND THREE AOP PROCESSES: (APS/UV, CaO2/UV AND

H2O2/UV). AN EXTENSION TO CHAPTER FIVE AND SIX

6.1 Introduction…………………………………………………………………………………131

6.2 Experimental Procedure, Results and Discussions…………………………………………….132

6.3 Conclusions…………………………………………………………………………………...137

CHAPTER SEVEN: APPLICATION OF THE UV/CHLORINE ADVANCED OXIDATION PROCESS FOR SULFOLANE TREATMENT 7.1 Degradation of Sulfolane using Bleach (NaOCl/HOCl) along with UV Light Irradiation…..138

7.2 Conclusions…………………………………………………………………………………142

CHAPTER EIGHT: ALTERNATIVE VIEW OF CHLORINE OXIDATION STIMULATED BY LONGER WAVELENGTH LIGHT

xii

8.1 Introduction…………………………………………………………………………………143

8.2 Materials and Methods……………………………………………………………………...145 8.2.1 Materials……………………………………………………………………………….145 8.2.2 Sample Preparation, Irradiation, and Analysis………………………………………...146 8.3 Photodegradation of SRFA, Coumarin Dye, and BDW …………………………………..147 8.4 Conclusions………………………………………………………………………………..156

CHAPTER NINE: CONCLUSIONS & RECOMMENDATIONS FOR FUTURE RESEARCH 9.1 Conclusions………………………………………………………………………..………..157

9.1.1 APS/UV and APS/UV/O3 systems……………………………………………………..157

9.1.2 CaO/CaO2 system………………………………………………………………………158 9.1.3 HOCl/UV system………………………………………………………………………159 9.2 Recommendations for Future Research………………………………………………..…...160

9.2.1 APS/UV and APS/UV/O3 systems…………………………………………………….160 9.2.2 Sulfolane Removal in Soil Wash Water……………………………………………….161 9.2.3 HOCl/UV System……………………………………………………………………...161 9.2.4 Catalytic Ozonation…………………………………………………………………....162

REFERENCES...... 163

xiii

List of Tables

Table 2.1 Chemical and physical properties of sulfolane………………………………...………8 Table 2.2 Thermal stability of sulfolane at different temperatures …...……………………….....9 Table 2.3 The effects of sulfolane inhalation on laboratory animals……………………………14

Table 2.4 LD50 values for sulfolane administered to animals by ingestion, subcutaneous

injection or dermal exposure …………………………………………….………..15

Table 2.5 Homogeneous catalytic ozonation….………………………………………………...36 Table 2.6 Heterogeneous catalytic ozonation metal oxides.………………………….…………43

Table 2.7 The mechanisms of catalytic ozonation proposed for MnO2…….……………………..44 Table 2.8 Mechanisms of catalytic ozonation suggested for Fe hydroxyoxides ...…..………….45 Table 2.9 Heterogeneous catalytic ozonation – metals on supports …………..………………..45 Table 2.10 Potential reactive species and their redox potentail in an activated persulfate system.51 Table 2.11 Ultraviolet energy type…………………………………….………………………..64

Table 2.12 Commonly used definitions of quantum yields……………………..………………69

Table 3.1 Pseudo-first-order degradation rate constants for (200 mg L-1-1.83 mM) sulfolane and varying concentrations of persulfate under UVC irradiation (10 lamps)……...……..77

Table 3.2 Pseudo-first-order degradation rate constants of 1.8 mM sulfolane and persulfate

(13.10 mM- 3 g L-1) under three different light intensities…………….………………79

Table 3.3 Effect of initial pH on degradation of sulfolane (220 mg L-1) in presence of 3 g L-1 of

APS and UVC 10-254 nm lamps……………………………………………………..80

Table 3.4 Pseudo-first-order degradation rate constants of 1.83 mM sulfolane and persulfate with -1 different concentrations in presence of O3 (5 mg L ) under UVC irradiation………..88

Table 3.5 Characteristics of groundwater samples………………………………………………97 -1 Table 4.1 Effect of CaO2 loading on degradation rate of 220 mg L of sulfolane bubbling O3

(0.5 L min-1)……………….………………………………………………...………106

Table 5.1 List of 10 different AC used along with ozonation for sulfolane degradation……....127

xiv

Table 6.1 Characterization of the soil samples………………………………………………...133 Table 6.2 Experiment A-F descriptions and the corresponding reaction rate constants, Wash-

Water: WW; UVC light: UVCL…………………………………………………….134

List of Figures and Illustrations

Fig. 2.1 Molecular structure of sulfolane……………………………………………………….7 Fig. 2.2 Absorption spectrum of HOCl and OCl-……………………….….……………………60 Fig. 2.3 Dependence of the ratio of HOCl/OCl- on pH………………………………………….61 Fig. 2.4 Classification of electromagnetic radiation in the wavelength range below 1200 nm and the interaction of each class with molecule ……………………………………………64 Fig. 2.5 Output spectrum of low pressure mercury lamp………………………………………..66

Fig. 2.6 Typical output spectrum of a black lamp…………………………………………….…66

Fig. 2.7 The inner works of an LED ….………………………………………………..………..67

Fig. 3.1 Degradation of sulfolane (220 mg L-1-1.83 mM) using 1 g L-1 - 4 g L-1 (4.38-17.40 mM)

ammonium persulfate under UV irradiation (10 lamps) …………………….………….77

Fig. 3.2 Changes in concentration of persulfate (13.10 mM- 3g L-1) over times in combination with

UV or UV/O3…………………………………………………………………………….78 Fig. 3.3 Effect of initial pH on TOC removal (sulfolane -220 mg L-1) in presence of (13.10 mM-3

g L-1 of APS and UVC……………………………………………………...…………...80

Fig. 3.4 Effect of carbonate concentration (100-400 ppm), pH:10.6 -10.9, on degradation of

sulfolane (220 mg L-1 - 1.83 mM) using 1 g L-1 (4.38 mM) of APS under UV irradiation

………………………………………….……………………………………………….82

Fig. 3.5 Effect of carbonate (100-400 ppm), pH:10.6 -10.9 on degradation of sulfolane (220 mg

L-1- 1.83 mM) and 3 g L-1 (13.10 mM) of APS under UV irradiation………………….83

Fig. 3.6 Effect of initial pH on degradation of sulfolane (220 mg L-1) in presence of 3 g L-1 of APS

and 400 ppm carbonate/bicarbonate under UV irradiation………………………………84

Fig. 3.7 Effect of chloride concentration (100 ppm) on degradation of sulfolane (220 mg L-1--1.83

mM) using of ammonium persulfate (3 g L-1 -13.10 mM)) under UV irradiation…..……85

xvi

Fig. 3.8 Degradation of sulfolane (220 mg L-1- 1.83 mM) using APS (4.38-13.10 mM), under

bubbling O3 and 254 nm irradiation………………………………………..……………88

Fig. 3.9 Degradation of sulfolane (220 mg L-1- 1.83 mM) using 1 g L-1 (4.38 mM) APS under UV

irradiation in the presence or absence of bubbling ozone (0.5 L min-1)………………..89

-1 Fig. 3.10 TOC removal over time for three APS loadings for 220 mg L of sulfolane and O3

flow rate of 0.5 L min-1 under 254 nm irradiation…………………………..…………90

Fig. 3.11 Degradation of sulfolane (220 mg L-1) using APS (1 g L-1) and bubbling ozone (0.5 L

min-1)…………………………………………………………………………………..91

Fig. 3.12 Effect of changing initial concentration of sulfolane on induction period of O3 (0.5

L min-1) -UVC system……………………………………………………………...... 92

Fig. 3.13 Effect of initial solution pH on degradation of sulfolane (220 mg L-1- 1.83 mM) in

-1 -1 presence of 3 g L (13.10 mM) of APS and O3 flow rate of 0.5 L min under 254 nm

irradiation……………………………………………………………………………...93

Fig. 3.14 Effect of initial solution pH on TOC removal, sulfolane (220 mg L-1 - 1.83 mM) in

-1 -1 presence of 3 g L (13.10 mM) of APS and O3 flow rate of 0.5 L min under 254 nm

irradiation……………………………………………………………………………...93

-1 Fig. 3.15 Effect of carbonate on sulfolane (220 mg L - 1.83 mM) degradation using O3 (0.5

L min-1) and persulfate (4.38 mM) under 254 nm irradiation. The pH for all solution

were above 10.5……………………………………………………………………....94

-1 -1 Fig. 3.16 Effect of chloride on sulfolane (220 mg L ) degradation using O3 (0.5 L min ) and

persulfate (1 g L-1) under 254 nm irradiation…………………………………………..95

Fig. 3.17 Degradation of sulfolane (220 mg L-1- 1.83 mM) using persulfate (3 g L-1- 13.10 mM) under 350 nm irradiation (I= 4.70×1017 photons.s-1)…………………………………...96

xvii

Fig. 3.18 Degradation of sulfolane in a ground water sample ………………………….……….98

Fig. 4.1 Degradation of 220 mg L-1 (1.83 mM) sulfolane in presence of different loadings of

-1 CaO2 and 0.5 L min of O3. The CaO2 loadings are per 100 ml of sulfolane solutions...105

-1 -1 Fig. 4.2 Degradation of 220 mg L (1.83 mM) sulfolane using 1.6 g L (22.20 mM) CaO2/O3 -1 compared and UV/O3, with the O3 flow rate of 0.5 L min ……………………………………107 -1 -1 Fig. 4.3 Degradation of sulfolane (220 mg L - 1.83 mM) using O3 (0.5 L min ) and different

concentrations of NaOH. The loadings are for 100 ml of sulfolane solutions………..109

-1 -1 Fig. 4.4 TOC removal by ozonation (0.5 L min ) in presence of CaO2 (3.00 g L – 41.62 mM) and

NaOH (5.10 g L-1)…………………………………………………………………..….109

Fig. 4.5 Degradation of sulfolane in presence of CaO2 (filtered solution after 15 min stirring of

-1 -1 1.6 g L – 22.20 mM of CaO2) under bubbling O3 (0.5 L min )……………………….111

-1 Fig. 4.6 Degradation of sulfolane (220 mg L – 1.83 mM) using 50mM CaO2, CaO, Ca(OH)2 and

-1 O3 (0.5 L min )……………………………………………..…………………………113

-2 Fig. 4.7 TOC removal using 5.0×10 M CaO2 (pH=11.0) or CaO (pH=11.85) under O3 bubbling

with a flow rate of 0.5 L min-1……………………………………………………...…114

Fig. 4.8 Degradation of sulfolane using different loadings of CaO (0.8, 1.6, 3.0 g L-1- 14.26,

-1 28.53, 53.50 mM) under bubbling O3 (0.5 L min )…………………...……………….114

-1 Fig. 4.9 Comparing different percentage of O3 with flow rate of 0.5 L min , in presence of

3 g L-1 (53.50 mM) of CaO…………………………………………………………….115

Fig. 4.10 Experimental set up for the field tests……………………………………………….117

Fig. 4.11 Degradation of sulfolane in groundwater sample spiked with sulfolane using CaO2 and

-1 CaO (50 mM) under bubbling O3 0.5 L min ……………………………...…………117 Fig. 4.12 TOC removal in 60 L of tap water in presence of 1.6 g L-1 CaO (28.53 mM) and 5 L min-

xviii

1 of O3………………………………………………………………………………...118

Fig. 4.13 Degradation of sulfolane (100 ppm) in 60 L of well water in presence of 1.6 g L-1 CaO

-1 (28.53 mM) and 5 L min of O3. The y axis is the PA measured by GC…………….118

Fig. 4.14 TOC removal in 60 L of well water in presence of 1.6 g L-1 CaO (28.53 mM) and

-1 5 L min of O3………………………………………………………………..……….119

-1 -1 Fig. 5.1 TOC removal by ozonation (0.5 L min ) in presence of CaO2 (3.00 g L - 41.62 mM) and

NaOH (5.10 g L-1 – 127.50 mM)…………………………………………...…………120

Fig. 5.2 Changes in concentration of butenoic acid (BA) over time during sulfolane (220 mg L-1 - -1 1.83 mM) degradation by ozonation (0.5 L min ) in presence of CaO2 and NaOH; pH= 12.02..124

Fig. 5.3 Degradation of sulfolane (220 mg L-1 – 1.83 mM) using silica gel (SG), zeolite and

graphene along with ozonation (0.5 L min-1) and 254 nm irradiation compare to that

without catalyst…………………………………………………………………..……126

Fig. 5.4 Degradation of sulfolane using ozonation (0.5 L min-1) along with activated carbon-A..128

Fig. 5.5 Degradation of sulfolane using ozonation (0.5 L min-1) along with activated carbon-E..128

Fig. 5.6 Adsorption of sulfolane on different loadings of AC-A and ozonated AC-A. PA: GC peak area of sulfolane…………………………………………………………………….….129 Fig. 5.7 Adsorption of sulfolane on different loadings of AC-E and ozonated AC-E. PA: GC peak area of sulfolane…………………………………………………………………….…130 Fig. 6.1 Soil washing set up………………………………………………………….………….132

-1 Fig. 6.2 Degradation of sulfolane (220 mg L – 1.83 mM) in soil washwater using APS or CaO2 along with 254 nm irradiation and ozonation (0.5 L min-1). PA: GC peak area of sulfolane…..134

Fig. 6.3 Soil flushing set up……………………………………………………………..………136

Fig. 6.4 Degradation of sulfolane in two times diluted soil wash samples using H2O2 (13 mM) and

-1 10-UV lamps and bubbling O3 (0.5 L min )…………………………………………..136

xix

Fig. 6.5 Changes in fluorescence spectra (λexci=350 nm, PMT=high) of water sample during

degradation of sulfolane using H2O2 (13 mM) and 10-UV lamps and bubbling O3

(0.5 L min-1)……..……………………………………………………………………136

Fig. 6.6 Washwater samples before and after H2O2, UV, O3 treatment………………………….137

Fig. 7.1 Degradation of sulfolane (20 ml- 220 mg L-1- 1.83 mM) using 80.60 mM NaOCl

(pH=10.5) under 8-254 nm irradiation…………………………………………..……138

Fig. 7.2 Absorption spectra of HOCl and OCl- in water………………………………………..139 Fig. 7.3 Degradation of sulfolane (220 mg L-1- 1.83 mM) using 114.00 mM of HOCl (pH=3.4) and 254 nm irradiation…………………………………………………...……………140 Fig. 7.4 Degradation of sulfolane (220 mg L-1- 1.83 mM) with sequential addition of HOCl under 254 nm irradiation…………………………………………………………………….141 Fig. 8.1 Absorption spectrum of sodium hypochlorite solution………………………………..145

Fig. 8.2 (A) Changes in absorbance of SRFA (100 ppm) in presence of (2.4 × 10−2 M) NaOCl (pH = 9.66) in dark and under 440 nm irradiation; (B) changes in absorbance at 400 nm during 62 min of dark reaction versus under irradiation plus irradiation results in absence of NaOCl; absorbance values are referenced to pure water……………….148

Fig. 8.3 Changes in the characteristic humic fluorescence (λex = 350 nm) of (100 ppm) SRFA in

presence of NaOCl (2.4×10-2 M) with the initial pH of 9.4 in dark and under 440 nm

irradiation (A) and changes in Fluorescence intensities at 446 nm during 30 min of dark

or irradiation (B). Fluorescence values are in arbitrary units under the condition specified

in the experimental section…………………………………………………………….149

Fig. 8.4 (A) Changes in absorption spectra of coumarin dye 28 μM (not shown here) in presence of (8.4 × 10−3 M) NaOCl (pH = 9.78) in dark (solid lines) and under 440 nm irradiation (dotted lines); (B) changes in absorbance of coumarin dye at λ = 400 nm; absorption of the stock coumarin dye at 400 nm is 0.10………………………………………..151 Fig. 8.5 Comparison of absorption spectra of a BDW sample in presence of NaOCl (7.7×10-2M)

after 1.5h dark and irradiation at pH=9.9 (A); Control experiment under irradiation without

NaOCl (B) and changes in absorption spectra (at 400 nm) of a BDW sample in presence

of NaOCl (7.7 ×10-2M) in dark at 23, and 35oC and under 440 nm irradiation (C)……...153

Fig. 8.6 BDW sample in presence of NaOCl (7.7 ×10-2 M) after 1.5 h in dark (23 and 35 oC) and under 440 nm irradiation………………………………………………………………154 Scheme 1.1 Summary of the oxidative methods used for degradation of sulfolane………………..3 Scheme 2.1 Chemical synthesis of sulfolane…………………………………………………..10

Scheme 2.2 Oxidation of substrate (S) with O3 ………………………………………………..…29

Scheme 2.3 Proposed mechanism for oxalic acid catalytic ozonation by Co(II)/O3 system ……………………………………………………………………………………………………39 Scheme 2.4 Three possible cases of heterogeneous catalysis for degradation of Organic Materials (OM)……………………………………………………………………………………………..42 Scheme 2.5 Two main possibilities of catalytic ozonation pathways: (up) adsorption on catalyst and oxidation by ozone or •OH radical of adsorbed organic (down) •OH or other radical species generation by reaction of ozone with reduced metal catalyst, and oxidation of organic by oxidized metal and/or in homogeneous solution…………..46 Scheme 3.1 Chemistry of persulfate…………………………………………………………..….71 Scheme 3.2 Schematic of an experimental setup with both ozonation and UV……………….….75

Scheme 5.1 Degradation pathway of sulfolane using NaOH/O3 or CaO2/O3. ……….……..123

xxi

List of Symbols, Abbreviations and Nomenclature

Symbol Definition C Concetntration of a substrate, mg L-1or mol L-1

C0 Concetntration of a substrate

ευ Relative permittivity K Rate costant, s-1 or mol L-1 s-1

Koc Carbon-water partition coefficient

KOW n-Octanol-Water partition coefficient -1 Kp Dermal pressure, mgh Λ Wavelength (nm)

λmax Maximum Wavelength (nm) Μ Dipole moment, Debye pKa -log Ka; Ka: Dissociation Constant Δ Hildebrand solubility parameter, MPa1/2 T time, min or s

Abbreviations Definition AC Activated Carbon AOPs Advanced Oxidation Processes APS Ammonium persulfate BDW Blowdown Water CCME Canadian Council of Ministers of the Environment CECs Contaminants of Emerging Concern CNS Central nervous system CNTs Carbon nanotubes COD Chemical Oxygen Demand DCM dichloromethane DBPs disinfection by products DMA dimethylacetamide DMF dimethyformamide DOC Dissolved Organic Carbon DIPA Diisopropanolamine DMSO dimethyl sulfide EEO Electrical Energy per Order EI Electron Impact ESI Electrospray Ionization FID Flame Ionization Detector

xxii

GC Gas Chromatography HPLC High Performance Liquid Chromatography ISCO In Situ Chemical Oxidation LED Light Emitting Diode LMCT Ligand-to-Metal Charge Transfer MS Mass Spectrometry NMP N-Methyl-2-pyrrolidone NOAEL Non Observable Adverse Effect Level NOM Natrural Organic Matter OA oxalic acid OSPW oil sands tailing pond water PPCPs Pharmeceutical and Personal Care Products PMT Photomultiplier PS Photosensitizer SAGD Steam Assisted Gravity Drainage SRFA Swannee River Fulvic Acids TOC Total Organic Carbon UV Ultraviolet UVA Ultraviolet, sub-type A UVB Ultraviolet, sub-type B UVC Ultraviolet, sub-type C UVCL UVC Light UV-Vis Ultraviolet – Visible VUV Vacuum Ultraviolet WW WashWater

xxiii

Chapter one: INTRODUCTION

1.1 Background

Sulfolane, is a man made commonly used industrial liquid because of its unique physiochemical properties. Annually 18,000–36,000 tons of sulfolane is produced worldwide. Due to such a large-scale industrial production, a significant amount of waste containing sulfolane is produced. Improper disposal of this waste and landfill practices as well as leakages from reservoirs, are the main sources of sulfolane contamination in the environment. Sulfolane contamination causes adverse effects to human health and the environment through impacts to soil, ground and surface waters. The contaminated sites are normally managed by first assessing the level of contamination; and if found to be above the current guidelines, remediation is necessary.

Guidelines for sulfolane in environment have existed in Alberta and British Columbia in

Canada and in California, Delaware, Indiana, Texas and in Puerto Rico in US, indicating its use and release to the environment in these jurisdictions (Yu et al., 2016). Alberta classifies sulfolane as a ‘Tier 1’ contaminant through the framework for the management of contaminated sites. For fine grained soils and also agricultural use, the cleanup level is 0.18 milligrams per kilogram

(mg/kg), and for coarse grained soil the cleanup level is 0.21 mg.kg-1. The cleanup level for groundwater is 0.09 mg L-1 for all uses (Alberta Environment 2009).

Since 1980s several active treatment methods have been suggested in literature. Adsorption on biologically activated carbon from water (McLeod et al., 1992) and biological treatment under aerobic conditions, using microbes residing in an activated sludge in wastewater treatment system or in contaminated aquifer sediments, are among the remediation methods that were investigated in the past (Greene et al., 2000; Greene et al., 1998; Chou and Swatloski, 1983; Chou and Swatloski, 1983; Fedorak and Coy, 1996).

Another popular treatment method for sulfolane is application of UV- based AOPs in an aqueous medium (spiked water and sulfolane-containing groundwater). Scala and Colon in

1979 proposed vacuum UV photolysis of sulfolane; sulfur dioxide, cyclobutane, 1-butene and ethylene were reported as the reaction products. However preliminary experiments on sulfolane degradation using H2O2/UVA in the contaminated groundwaters were reported in 2005 by

Agatonovic and Vaisman (2005). Mehrabani et al. (2016) and Yu et al. (2016a), reported on application of several oxidative methods such as application of UVA and UVC irradiation along with photoactive oxidants, including TiO2, O3, H2O2 and their combination such as H2O2/O3 for degradation of sulfolane. While these experiments were performed in a batch reactor, Mehrabani et al. (2016), investigated the mineralization of sulfolane using UV/O3/H2O2 in a flow through tubular reactor. Application of Fenton reagent is also reported in literature for sulfolane oxidation

(Yu et al., 2016b; Omar et al., 2010).

1.2 Research Goals and Objectives

The focus of this research is to build on current research and investigate the application of some of the other oxidative chemical and photochemical methods for remediation of sulfolane (present at concentrations exceeding guidelines) in water, and soil. With this goal in mind the following sub objectives are identified:

I- Investigation of sulfolane degradation in water/groundwater using persulfate/UV, O3-

persulfate/UV

II- Investigation of sulfolane degradation in water/groundwater using CaO2/O3 and CaO/O3;

laboratory and field tests

III- Study of mechanism of sulfolane degradation using CaO2/O3

IV- Study of degradation of sulfolane using hypochlorous acid and UVC light

V- Mechanistic study of natural organic matter (NOM) degradation using sodium

hypochlorite/440 nm blue LED and its application to sulfolane degradation in water

(degradation by reactive oxygen species or by photosensitization)

VI- Investigation of application of the studied (identified above) treatment methods for

degradation of sulfolane in soil washwater

Scheme 1.1 provides a summary of this study’s selected sulfate and hydroxyl based advanced oxidation methods for degradation of sulfolane.

UVC+ Ammonium persulfate (APS) CaO + O UVC+ HOCl UVA + NOM 2 3 - Mechanistic study CaO + O UVA+ Ammonium persulfate (APS) 3 Mechanism UVC+ Ammonium persulfate + O Catalytic ozonation 3

Remediation of sulfolane in soil washwater

•– • All involve SO4 and OH

Sulfate or hydroxyl radical base Advanced oxidation processes (AOPs)

Scheme 1.1 Summary of the oxidative methods used for degradation of sulfolane

1.3 Thesis Overview

This dissertation is divided into nine chapters in a paper-based format. A brief description on the content of each chapter is outlined here.

Chapter 1, the current chapter, presents a general introduction on sulfolane as an emerging industrial contaminant in groundwater and soil and the motivation to perform this research with consideration to past research. Research scope and objectives are also provided here.

Chapter 2, presents a literature review on:

▪ sulfolanes’s physiochemical properties, sources, natural attenuation processes and

environmental problems

▪ principles of sulfate and hydroxyl based advanced oxidation processes as potential methods

for sulfolane degradation and their applications for contaminated water/wastewater

▪ sulfate and hydroxyl radical production by persulfate activation

▪ ozonation

▪ catalytic ozonation

▪ photosensitization and

▪ light sources

This chapter also includes advantages and disadvantages of each treatment method, as well as role and effect of interferences on the degradation.

1 In Chapter 3 , degradation of sulfolane in spiked water by (NH4)2S2O8/UVC,

(NH4)2S2O8/O3/UVC and (NH4)2S2O8/O3/UVA methods is presented. Also, effects of O3, persulfate dosage, UV light intensity, pH and interferences such as carbonate/bicarbonate and

1 Izadifard, M, Achari G. and Langford, C.H. (2017). Degradation of sulfolane using activated persulfate with UV and UV-Ozone. Water Research, 125, 325 – 331.

4 chloride on degradation rate of sulfolane are investigated. Applicability of (NH4)2S2O8/UVC method is also evaluated for degradation of sulfolane in groundwater.

2 Chapter 4 , describes the mineralization of sulfolane in aqueous systems by CaO2/O3 and CaO/O3.

Effect of O3, CaO2 and CaO dosage, UVC irradiation and pH on reaction rates and on TOC removal is reported. A number of field pilot experiments are also designed and conducted both for water spiked with sulfolane and for groundwater contaminated with sulfolane.

Chapter 5, advances on the findings of chapter 4. Using the same experimental set up described in chapter 4, further experiments are performed to find the pathways and reaction mechanisms of sulfolane degradation using CaO2/UV. A short study is performed on possibility of using catalytic ozonation for sulfolane degradation.

Chapter 6, presents an evaluation of the AOP methods described in Chapters 4 and 5 namely

H2O2/UV, APS/UV, APS/UV/O3, CaO2/O3, CaO/O3, for remediation of sulfolane extracted from soil by a soil washing/flushing technique.

Chapter 7, reports on a short study on assessment of UV/chlorine as an advanced oxidation process for degradation of sulfolane in spiked water. The challenges of this method for practical application are discussed.

3 − − Chapter 8 , investigates a mechanistic study on aqueous chlorine (Cl2, OCl , Cl ) in combination with longer wavelength ultraviolet irradiation and visible light for water and wastewater treatment.

2 Izadifard, M.; Achari, G.; Langford, C.H. (2018). Mineralization of sulfolane in aqueous solutions by Ozone/CaO2 and Ozone/CaO with potential for field application. Chemosphere 197, 535-540. 3 Izadifard, M.; Langford, C. H., Achari, G. (2016) Alternative view of chlorine oxidation stimulated by longer wavelength light. J. Environ. Eng. 142(10), 04016048.

The cases include bleaching of organic matter (NOM) and blowdown water under longer wavelengths irradiation.

Chapter 9, presents the major conclusions drawn from this dissertation. Recommendations for future research are also presented in this chapter.

Chapter two: LITERATURE REVIEW

2.1 Sulfolane

Sulfolane, C4H8SO2, (Fig. 2.1), also known by other chemical names, such as: bondelane

A, cyclotetramethylenesulfone, dihydrobutadiene sulfone, dioxo-2,3,4,5- tetrahydrothiophene,

1,1-dioxothiolane, 1-1 -sulfalonel sulfolan, sulfoxaline, sulpolane, tetrahydrothiophene-1 ,l- dioxide, 2,3,4,5,4-tramethylene sulfone, and thiocyclopentane1 ,l-dioxide, is a cyclic sulfone, and is a well known polar solvent (Fedorak and Coy, 1966). Sulfolane contains a four-membered carbon ring and sulfonyl functional group (R2SO2) with a sulfur atom double-bonded to two oxygen atoms.

Fig. 2.1 Molecular structure of sulfolane

2.1.1 Physiochemical Properties of Sulfolane

Sulfolane has no colour or odor; it is not volatile and not highly viscous. It is heavier than water (the liquid density of sulfolane is 1.2606 g cm-3) and is easily soluble in water due to its highly polar sulfur–oxygen double bonds. The solubility of sulfolane is reported to be 1,266 g L-1 at 20°C as per Canadian Environmental Quality Guidelines (CCME, 2006). Its melting point is

7 between 27.4 and 27.8°C and boiling point is between 280 and 285 °C. It has a vapor density of

4.2 g L-1 relative to air (Stewart and Minnear, 2010; CCME 2006).

Sulfolane has a high dipole moment (μ = 4.7 Debye), elevated relative permittivity (ευ

=43.4), and a high Hildebrand solubility parameter (δ = 27.2 [MPa]1/2). This means it has high solvency power for reactions containing polarizable intermediates. Also, sulfolane is capable of strong solvation of cations by the oxygen atoms present in the sulfone group, which increases the nucleophilicity of the corresponding less solvated anions (Tilstam, 2012).

Table 2.1 Chemical and physical properties of sulfolane (Modified from Bak et al. 2018)

Property Value Units

Molecular weight 120.17 g×mol-1

Freezing point 27 oC

Melting point 28.5 oC

Flash point 165-178 oC

Density at 15oC 1.276 g cm-3

Viscosity at 30oC 0.00987 P

Vapour pressure at 25oC 0.0062 mmHg

n-Octanol-Water partition coefficient (KOW) -0.77 log

Henry's law constant=P/C 4.6×10-10 atm×m3×mol-1

Solubility in water at 25oC 379, miscible g L-1

Sorption by aquifer material <1 L Kg-1

Hildbrand solubility parameter 27.2 (MPa)1/2

pKa 12.9 -log Ka

Heat capacity at 30oC 181.5 J×mol-1×K-1

Sulfolane is thermally stable until about 220°C, when it starts to break down slowly with the formation of sulfur dioxide and a polymeric compound (Brown, 1966). Sulfolane is also chemically stable against many chemical substances including strong acids and bases. The only exceptions are sulfur and aluminum chloride (Kirk-Othmer, 1999).

A complete list of physical and chemical properties and of thermal stability of sulfolane are summarized in Table 2.1 and Table 2.2, respectively.

Table 2.2 Thermal stability of sulfolane at different temperatures (Modified after Kirk- Othmer, 1999)

o Temperature, C amount of SO2 released from 250 mL sulfolane, mg/h 180 0.6 200 2.8 220 3.3 240 24.1

Since sulfolane is a polar solvent, it selectively dissolves other polar compounds such as water, acetone, glycerol, toluene and other aromatic solvents. Aliphatic hydrocarbons are nonpolar, whereas aromatic hydrocarbons (e.g. toluene) exhibit polarity due to the ᴫ-cloud of the aromatic ring. High stability of sulfolane against strong acids and bases and its thermal stability makes it a solvent of choice for different acid-catalyzed reactions at elevated temperatures

(Tilstam, 2012). Sulfolane can be used as a solvent for a wide range of reaction types/conditions including formation of various fluoroaromatic compounds (halo exchange reactions), oxidations, phosphonylations and condensation reactions (Back et al., 2018).

Common chemical reactions involving sulfolane are: UV assisted halogenation, ring cleavage by alkali metals, ring addition catalyzed by alkali metals, Grignard reactions and complex formation by Lewis Acids (Kirk-Othmer, 1999).

2.1.2 Synthesis and Manufacturing

Sulfolane is a non-reactive, water-soluble, dipolar industrial solvent, which was first engineered and patented by Shell Oil Company in the early 1940s (Zaretskii et al., 2013). Its synthesis is based on the hydrogenation of 3-sulfolene (3) obtained by the reaction of sulfur dioxide (2) and butadiene (1) (Scheme 2.1).

Scheme 2.1 Chemical synthesis of sulfolane

Sulfolane is commercially available for purchase in anhydrous or aqueous form. Aqueous sulfolane typically has 3%–5% moisture content and is used as a freezing point depressant to ensure that it remains liquid during transport and storage (Bak et al., 2018). The worldwide production of sulfolane is approximately 18,000–36,000 tons per year.

2.1.3 Storage of Sulfolane

Typically, in the refining process, the breakdown of sulfolane occurs in the presence of oxygen or chlorides. Sulfolane reacts with oxygen to produce sulfuric and organic acids as well as aldehydes and ketones. Oxygen-degraded sulfolane has a lower pH (leading to enhanced corrosivity), a darker color than pure sulfolane and a lower extractive power for aromatics in the aromatic extraction process (Stewart and Minnear, 2010). Thus, sulfolane is stored under a nitrogen blanket where it is not in contact with atmospheric oxygen.

2.1.4 Applications

Sulfolane has a variety of applications. The most common use of sulfolane is as a solvent in liquid- liquid aromatics extraction from mixtures containing aliphatic hydrocarbons. Sulfolane dissolves the aromatics - benzene, toluene, and xylenes from a hydrocarbon stream. Sulfolane is the preferred solvent for this process due to its high selectivity for aromatics of interest and its capacity to dissolve large quantities of aromatics. After the extraction process, most of the solvent is recovered and recycled (Stewart and Minnear, 2010).

Sulfolane is also commonly used in the Sulfinol® process for liquid natural gas treatment.

The Sulfinol® process is used in a variety of industries for removal of acidic components such as

H2S, CO2, COS, CS2, mercaptans and organic sulfides from sour gas streams (Kirk-Othmer,1999).

In the Sulfinol® process, the gas stream is contacted with a mixture of sulfolane, an alkanolamine (usually diisopropanolamine), and water (Kirk-Othmer,1999). The acid components in the gases are absorbed chemically by the amine and absorbed physically by sulfolane.

Other gas-treating processes involving sulfolane are (1) hydrogen selenide removal from gasification of coal, shale, or oil sands; (2) olefin removal from alkanes; (3) nitrogen, helium, and argon removal from natural gas; (4) atmospheric CO2 removal in nuclear submarines; (5) ammonia and H2S removal from waste streams; (6) H2S, HCl, N2O, and CO2 removal from various streams and (7) H2S and SO2 removal from gas mixtures (Kirk-Othmer,-1999).

Apart from these traditional applications, sulfolane has also been used in a wide range of engineering and biomedical applications (Bak et al., 2018; Stewart and Minnear, 2010; Kirk

Othmer, 1999; Kopple et al., 1992)

These include:

• fractionation of wood tars (separating out components of wood tars for commercial use)

• curing agent for epoxy resins

• production of electronics; because of its high dielectric constant, low volatility, excellent

solubilizing characteristics, and aprotic nature, sulfolane has been used as a solvent in batteries

(particularly lithium batteries), coil-insulating component, solvent in electronic display

devices, as capacitor impregnates, and as a solvent in electroplating baths.

• production of insecticides, herbicides, and fungicides; micro-emulsions containing sulfolane

together with a cationic surfactant such as cetyltrimethylammonium bromide are useful for the

detoxification of pesticides

• process solvent in pharmaceutical manufacturing

• wood delignification for removing lignin from wood chips and freeing the cellulose fibers

• solvent in analytical processes such as fast atom bombardment-mass spectrometry, and nuclear

magnetic resonance

• polymer plasticizer; nylon, cellulose, and cellulose esters can be plasticized using sulfolane to

improve flexibility and to increase elongation of the polymer. Sulfolane is subsequently

extracted from the fiber with water to give a permeable, plasticizer-free, hollow fiber.

• polymerization solvent in polymer production; for polyureas, polysulfones, polysiloxanes,

polyether polyols, polybenzimidazoles, polyphenylene ethers, poly(1,4-benzamide)

(poly(imino-1,4-phenylenecarbonyl)), silylated poly(amides), poly(arylene ether ketones),

polythioamides, and poly(vinylnaphthalene/fumaronitrile) initiated by laser. Increased

polymerization rate, ease of polymer purification, better solubilizing characteristics, and

improved thermal stability are advantages of using sulfolane in this process.

• textile applications for improved usability (e.g., enhanced surface texture and increased

durability); sulfolane is used as solvent in spinning and casting of synthetic fibers and fabrics

• component of ink for jet printing to increase storage ability, reduce clogging of printer nozzles,

improve fixation on substrates, and water resistance.

2.1.5 Toxicity

Sulfolane is a highly stable and attractive alternative to common dipolar aprotic solvents such as dimethyl sulfoxide (DMSO), dimethyformamide (DMF) and dimethylacetamide (DMA), and N-Methyl-2-pyrrolidone (NMP). This is because sulfolane has the highest boiling point and the highest freezing point, also its viscosity at 30 °C is also much higher than that for the others

(Tilstman, 2012). According to Tilstamn the acute toxicity of sulfolane is higher than that of the other solvents if swallowed. For sulfolane the level is 1.9 g kg-1 for rat, and for DMSO, it is 14 g kg-1. The skin permeability is highest for DMSO. Sulfolane has a skin permeability of only

0.2mg/(m2/h) which is much lower than that of the others. DMSO is not toxic as a pure compound, but it can carry dissolved toxic compounds through the skin when used as a solvent.

People get exposed to sulfolane through water, since it mixes easily with water. Other exposure routes such as breathing vapors or direct skin contact (occupational exposure) are unlikely to pose a risk because sulfolane has low volatility and is not absorbed through the skin.

Unfortunately, the human health impact of sulfolane is unknown. But the toxicity effects of sulfolane has been extensively examined in animals (Bak et al. 2018; Gordon et al., 1984;

Thomson et al., 2013). In fact, animal data suggest that reproductive and developmental toxicity is only likely to occur at high concentrations. For example, sulfolane can be acutely toxic via inhalation at doses exceeding 200 mg/kg leading to convulsions, vomiting, leukopenia and even death in exposed guinea pigs, squirrel monkeys and dogs (Zhu et al., 1987). Table 2.3 summarises some of the findings. Table 2.4 lists LD50 values for sulfolane administered to animals by ingestion,

subcutaneous injection or dermal exposure (Greene and Fedorak, 1999). Similarly, the LD50 of

sulfolane is relatively high in the laboratory animals tested. But the central nervous system effects

happen at lower doses. At lethal doses, animals convulse and demonstrate difficulty in breathing

before death. In general, the effects of sulfolane is reported to be highest in monkeys, and to lesser

degree in dogs, rabbits, guinea pigs, rats and mice (Andersen et al., 1977). This evidence suggests

that sulfolane would affect humans to a greater extent than some of the smaller test animals such

as rodents (Greene, 1999). CCME (2006) reports Total daily Intake (TDI) of 0.0097 mg kg-1 body

weight for human based on animal studies (Non Observable Advese Effect Level- NOAEL- for

female rat: 2.9 mg.kg-1 bw-day-1; Uncertainty Factor: 300).

It should be noted that there are ongoing toxicity studies to find sulfolane adverse effects on human

health with higher certainty as well as on plants and on aquatic life (Shah, 2018).

Table 2.3 The effects of sulfolane inhalation on laboratory animals (Modified after Andersen et al., 1977; Greene and Fedorak, 1999)

Animal Sulfolane (mg/m3) Exposure Time Effect on Animals convulsions, decrease in white blood cells, Rat 3600 17.5 h pulmonary hemorrhage chronic lung and river inflammation, decrease in 495 8 h×d-1, 5d×wk-1 white blood cells 2.8 continuous, 90d ------Guinea pig 495 8 h×d-1, 5d×wk-1 chronic lung inflammation chronic lung inflammation, fatty liver 200 continuous, 90 d metamorphosis 2.8 continuous, 90 d ------intermittent convulsions, extreme aggression, Dog (beagle) 200 continuous, 90 d vomiting, chronic lung hemorrhage 2.8 continuous, 90 d ------Squirrel vomiting, convolutions, decrease in white blood monkey 4850 18.5 h cells, pulmonary hemorrhage death, fatty metamorphosis of liver, lung 495 8 h×d-1, 5d×wk-1 inflammation, decrease in white blood cells 2.8 continuous, 90 d ------

Table 2.4 LD50 values for sulfolane administered to animals by ingestion, subcutaneous injection or dermal exposure (Modified after Greene and Fedorak, 1999)

-1 Animal Exposure routes LD50 (g×Kg ) Source of information Mouse Oral 1.9 to 2.5 Brown et al. 1966 Injection 1.1 Alexander et al. 1959 Injection 0.6 to 1.4 Anderson et al. 1976 Rat Oral 2.1 Brown et al. 1966 Oral 1.8 Anderson et al. 1976 Injection 1.1 to 1.6 Anderson et al. 1976 Dermal >3.8 Brown et al. 1966 Guinea pig Oral 1.8 Anderson et al. 1976 Injection 1.3 Anderson et al. 1976 Rabbit Injection 0.6 to 3.5 Anderson et al. 1976

In order to to maintain, improve or protect environmental quality also to prevent the adverse effects of sulfolane on human health as well as to protect aquatic life and agricultural water uses, threre exist several guidelines for sulfolane in the environment (section 1.1).

2.1.6 Environmental Problems of Sulfolane

Sulfolane has been detected, mostly as an anthropogenic chemical, in contaminated groundwater samples and identified in edible plant tissues, taken from nearby gardens surrounding natural gas or petroleum refining sites (Bak et al., 2018). The only report of sulfolane occurring naturally in the environment was in a sample of a sponge (Batzella) and tunicate

(Lissoclinum) composite, found on Point Impossible, Victoria, Australia, which contained approximately 50 mg·kg (dry·weight)-1 sulfolane (Barrow and Capon, 1992; CCME, 2006).

Sulfolane comprised about 0.005% of the dry weight of the specimens.

In Canada, reports on the presence of anthropogenic sulfolane in the environment are limited to data collected in the vicinity of sour gas processing facilities in Western Canada (CAPP, 1997;

Wrubleski and Drury, 1997). The maximum measured sulfolane concentrations in groundwater

15 were 800 mg·L-1 in shallow till and 88 mg L-1 in bedrock. Over many years of operation, there have been some unpredicted or accidental spills, accidental leaks from extraction units of sulfolane as well as leachates from disposal areas from producing wells and unlined storage ponds, which has caused contamination of soil, ground water and wetland ecosystem around gas processing plants (Stewart and Minnear, 2010). Leakages are mostly because of sulfolane induced corrosion.

In Alberta, the south Rosevear gas plant in 24 Yellowhead County located 32 kilometers northeast of Edson, has caused contamination of sulfolane in 5 out of 122 investigated water wells close to the gas plant (Alberta Health Services, 2014).

Under standard operating conditions pure sulfolane is stable and non-aggressive to steel, but if sulfolane is contaminated by traces of oxygen, at typical conditions of 170–180°C, sulfolane decompose and corrosive products are formed. The corrosion of steel can be quite rapid, causing severe damage to industrial installations. Decomposition of sulfolane may be further enhanced by presence of water, used commonly as a sulfolane diluent, as well as chlorides (Bak et al., 2018; Mingy et al., 2005)

The migration of a contaminant from a given pollutant site into an aquifer via groundwater is mainly governed by the water solubility of that compound, its density and interactions with soil as well as its attenuation in soil or aquifer sediments. Compounds that are highly water-soluble travel much more readily in groundwater (Morgan and Watkinson, 1989). The density of contaminants dissolved in groundwater also affects their migration through an aquifer; higher density will increase the downward flow of the contaminant plume (Christensen et al., 1994).

Increases in interactions between soil and contaminant compounds will decrease the rate of migration through an aquifer.

High miscibility of sulfolane, combined with the low octanol-water partition coefficient

(Kow), carbon-water partition coefficient (Koc) and pKa values have resulted in substantial offsite migration from contaminated sites, therefore the environmental fate of sulfolane is being extensively studied (Blystone, 2011). Natural attenuation procceses of sulfolane is discussed in the following section.

2.1.7 Natural Attenuation of Sulfolane

The natural attenuation processes include a variety of physical, chemical, or biological processes that, under favorable conditions, act without human intervention to reduce the mass, toxicity, mobility, volume, or concentration of contaminants in soil or groundwater (Stewart and Minnear, 2010). Processes such as volatilization, adsorption or degradation could be involved.

Sulfolane can be considered non-volatile, as evidenced by its low vapor pressure and Henry’s Law

Constant of 8.95 × 10-10 (atm-m3 mol-1) at 25°C (CCME, Lyman et al., 1982).

Sulfolane does not effectively get adsorbed on soil, as shown by the log octanol/water partition coefficient (log Kow) of -0.77. The organic carbon partition coefficient (Koc), estimated at

0.07, shows that the compound is highly mobile in soil. Though, the adsorption depends on the soil composition. Luther et al. (1998) indicated that sulfolane had higher adsorption to clay than organic matter, but in either case the interaction is not significant and sulfolane mobility in soil is very high.

Another attenuation pathway of sulfolane is its degradation by natural processes. Sulfolane is thermally and chemically stable in the presence of many chemical substances (Kirk-Othmer

1999), therefore chemical processes are not effective in degradation of sulfolane either.

The primary attenuation mechanism appears to be biodegradation in an aerobic environment. Like other organic compounds, biodegradation of sulfolane in an aquifer depends on the microbial population and availability of nutrients such as N, P and terminal electron acceptors

(Green and Fedorak, 2001).

2.1.8 Remediation

As discussed earlier, natural processes are not effective in sulfolane removal or they are very slow. Therefore, several active treatment methods have been suggested in literature for sulfolane removal from water and soil. Wide distribution of sulfolane and low cleanup levels add a level of complexity to a practical remediation; the Alberta Tier 1 groundwater guidelines is expected to be lowered to 0.04 mg L-1 and soil guidelines will be adjusted accordingly (“What is

Sulfolane,” 2017).

In general, there are three procedures of removing sulfolane-containing contaminants from industrial/environmental areas: physical, chemical or biochemical. Physical methods comprise transfer of sulfolane from water onto a solid phase such as activated carbon, whereas the other two are related to the decomposition/oxidation of sulfolane by thermal/chemical or photochemical methods, or degradation by microorganism to harmless products (Bak et al.,

2018). The complete oxidation of sulfolane can be described as follows (Eq. 2.1):

+ 2− C4H8O2S + 6.5O2 → 4CO2 + 3H2O + 2H + SO4 Eq. 2.1

Adsorption on biologically activated carbon from water (McLeod et al., 1992) and

biological treatment under aerobic conditions, using microbes residing in an activated sludge

in wastewater treatment system or in contaminated aquifer sediments, are among the methods

evaluated (Greene et al., 2000; Chou and Swatloski, 1983; Greene et al. 1998; Fedorak and

Coy, 1996). This is due to low investment and operating expenses, high treatment efficiency

and environmental acceptability. However, aerobic microbial bioremediation requires suitable

environmental conditions such as temperature, pH, nutrients and oxygen. Also, presence of

hydrocarbons co-contaminant lower the rate of biodegradation.

Another popular treatment method for sulfolane is application of UV- based AOPs in an aqueous medium (spiked water and sulfolane-containing groundwater). Scala and Colon in

Agatonovic and Vaisman (2005). Mehrabani et al. (2016) and Yu et al. (2016a), reported on application of several oxidative methods such as application of UVA and UVC irradiation along with photoactive oxidants, including TiO2, O3, H2O2 and their combination such as H2O2/O3 for degradation of sulfolane. Mehrabani et al. (2016), further investigated the mineralization of sulfolane using UV/O3/H2O2 in a flow through tubular reactor. Application of Fenton reagent is also reported in literature for sulfolane oxidation (Yu et al., 2016b; Omar et al., 2010). The main challenge in implementing AOPs is the high cost but normally provide the highest contaminant degradation efficiencies.

2.1.9 Sulfolane Analysis

Analytical methods usually used for sulfolane are gas chromatography combined with flame ionization (FID), mass-selective detection (MS), or electrolytic conductivity detectors.

Direct sample injection into GC-FID is another technique for analysis of sulfolane in water.

Sulfolane is extremely soluble in water, therefore sample preparation techniques may include extraction with organic solvents (Headley et al., 2002).

2.2 Hydroxyl Radical and Sulfate Radical Based Advanced Oxidation Processes

Advanced Oxidation Processes (AOPs) have been extensively studied for the destruction of a wide range of organic pollutants such as pesticides, surfactants and many other industrial chemicals from water and wastewater (Legrini et al., 1993; Reynolds et al., 1989; Alvares et al.,

2001; Zhou and Smith, 2001; Oppenla¨nder, 2003; Ikehata and Gamal El-Din, 2004; Ikehata and

Gamal El-Din, 2005; Ikehata and Gamal El-Din, 2006; Andreozzi et al., 1999; Alvares et al., 2001;

Legrini et al., 1993). These processes have several advantages over conventional chemical oxidation processes using potassium permanganate or chlorine, including higher oxidation potential, no formation of potentially carcinogenic chlorinated by-products, and no persistence of the toxic oxidant (Ikehata and Gamal El-Din, 2004; Alvares, 2001; Andreozzi et al., 1999; Camel and Bermond, 1998; Schwarzer, 1995; Langlais, 1991).

AOPs are oxidation processes mediated by hydroxyl radicals (HO•) and/ or sulfate radicals

•– (SO4 ), which are strong oxidants with high redox potentials (2.8 V and 2.6 V respectively); leading to complete mineralization of the organic pollutants into end-products such as CO2 and inorganic ions (Zhou and Smith, 2001).

Hydroxyl radical-mediated AOPs, involve combinations of chemical agents such as ozone, hydrogen peroxide, transition metals, and metal oxides and in most cases an auxiliary energy sources such as ultraviolet or visible radiation, heat, electronic current and ultrasound. Examples

2+ of AOPs include O3/H2O2, O3/UV, O3/H2O2/UV, H2O2/UV, Fenton (Fe /H2O2), photo- and electro-Fenton, chelating agent assisted Fenton, photocatalysis using titanium dioxide and sonolysis. Ozonation at pH>8 is also considered as an AOP because of the enhanced generation of hydroxyl radicals under such conditions (Beltra´ n, 2003). It should be taken into consideration that hydroxyl radicals (•OH) are the primary oxidant in these AOP processes; other radical and

•- • active oxygen species such as superoxide radical anions (O2 ); hydroperoxyl radicals (HO2 );

3 • triplet oxygen ( O2) and organic peroxyl radicals (R-O-O ) are also involved in the oxidation processes (Ikehata and Gamal El-Din, 2004; Oppenla¨nder, 2003; wang et al., 2003; ; Andreozzi,

1999; Camel and Bermond, 1998; Schwarzer, 1995; Langlais et al., 1991). Numerous studies have demonstrated that HO• reacts with pollutants in water through three competing pathways: addition, hydrogen abstraction, and electron abstraction (Lian et al., 2017; Minakata; 2009; Ikehata, 2006).

Sulfate radical-mediated AOPs have recently gained a substantial amount of scientific attention.

A series of methods have been used to generate sulfate radicals through the activation of persulfate and peroxymonosulfate with UV irradiation, heat, transition-metals or carbonaceous materials.

•− Hydroxyl radical based reactions are well-studied but SO4 based AOPs have come forth recently and received considerable attention for destroying recalcitrant pollutants such as pesticides, perfluorocarboxylic acids, and cyanotoxin (Lian et al., 2017; Lutze et al., 2015; Antoniou et al.,

•− 2010; Hori et al, 2005). Sulfate radical (SO4 ) is a strong oxidant with a high redox potential and reacts with many organic compounds at nearly diffusion-controlled rates, which are comparable to •OH (Lian et al., 2017; Neta et al., 1988).

•- • The reaction mechanism for the oxidation of organic molecules by SO4 is similar to HO (Xiao et al., 2018; Luo et al., 2017, George, 2001). It might be done by:

- hydrogen-atom abstraction from alkanes, alcohols, ethers and esters

•- - SO4 addition reaction with compounds containing unsaturated bonds such as alkenes and alkynes or

- electron transfer for organic compounds that contain electron donating substituents such as amino, amine, and hydroxyl groups.

•- Electron transfer from organic compounds to SO4 is more intrinsic to sulfate radical based AOPs.

•− Hydrogen-atom abstraction and addition processes have minor contributions to SO4 oxidation

(Lian et al, 2017; Lutze et al., 2015). For hydroxyl radicals, hydrogen-atom abstraction and OH radical addition are more common (Liang and Su, 2009). Therefore, they both can complement each other with diverse reactivities, product patterns, and energy efficiencies (Deng et al., 2011;

Khan et al., 2014).

In the meanwhile, in situ remediation with activated persulfate oxidation may be preferred over peroxide based HO radical oxidation processes, as the persulfate anion is more stable and more selective, has longer life time (less aggressive) and has application over a wider pH range

(Yan et al., 2013; Petri et al., 2011).Therefore for in situ remediation it may be transported further in the subsurface.

Overall, the performance of AOP is affected by the presence of other constituents of water, such as natural organic matter, dissolved and suspended solids, water pH and temperature

(Oppenla¨nder, 2003). For example, suspended solids and color can hinder photochemical reactions by light scattering and absorption and may hinder the performance of photochemical

AOPs. Carbonate, bicarbonate and chloride ions, as well as some natural organic compounds are known to act as radical scavengers. These compounds compete with target pollutants for hydroxyl radicals; therefore, their presence increases oxidant demands and lowers the treatment efficiency

(This will be more discussed in section 2.2.1.1.). In addition, the costs of materials and equipment, as well as energy requirements and efficiency must be considered, when assessing the overall performance of AOPs (Oppenla¨nder, 2003; Wang et al., 2003; Legrini et al., 1993).

2.2.1 Contaminated Water/Wastewater Treatment Using AOPs

AOP have been sparingly applied in municipal and industrial wastewater treatment plants to degrade a number of contaminants of emerging concern (Gilmour, 2012; Abdelmelek et al.,

2011). According to Sauvé and Desrosiers (2014) contaminants of emerging concern (CEC) are naturally occurring, manufactured or manmade chemicals or materials, which are present in various environmental compartments and whose toxicity or persistence impact the metabolism of living beings. CEC include many types of contaminants such as pesticides, pharmaceuticals, personal care products, plasticizers, flame retardants, nanoparticles, perfluoroalkyl compounds, chlorinated paraffins, siloxanes, algal toxins, various trace elements including rare earths and radionuclides, etc. (Sauvé and Desrosiers, 2014).

Unfortunately, despite the progress that has been achieved over three decades of intense research in understanding and adoption of AOPs, application of this technology in wastewater treatment is limited (Vogelpohl, 2007). High cost of AOP for complete mineralization, by-product formation and inadequate understanding about the quality of the treated water are some of the reasons. In some cases, partial oxidation of contaminants to intermediates, which are less harmful or readily biodegradable in the environment is a viable option. Occasionally partial oxidation of organic contaminants can result in the formation of intermediates, which are more toxic than the parent compounds (Rizzo, 2011). The nature and number of the degradation products depends on the AOP process, reaction time, and water quality metrics. In fact, the effectiveness of AOPs is largely determined by the background water quality matrix of the contaminated water. For example, the presence of high bromide concentrations or NOM can result in the formation of regulated oxidation by-products that may lead to the deterioration of the water quality to even lower than its initial quality. Bromide is oxidized to form bromine, which then reacts with naturally

23 occurring organic matter and produce DBPs; brominated DBPs have higher health risks than chlorinated DBPs (Hua et al., 2006). Production of bromate along with other DBPs is also possible

(Gunten, 1994). Similarly, the presence of nitrates and carbonates can interfere with the destruction of the target contaminants and ultimately reduce the effectiveness of the selected AOP. In general, most of the technical difficulties associated with AOPs is because of non-selective reactions of sulfate or hydroxyl radicals. Part of the high cost of AOP is the need for determining degradation pathways as well as quantifying the cumulative effects of the resulting mixture of compounds on living systems.

According to Krishnan et al. (2017), major factors affecting the cost analysis are the influent and effluent contaminant concentrations, the quantity of oxidizing agents and catalysts, light intensity, irradiation time and the nature of the wastewater (pH, presence of solids and ions).

It is always necessary to conduct experimental studies to develop an AOP method suitable for a specific wastewater by determining chemical dosages, desired flow rate, reactor contact time and reactor configuration. This also to ensures no harmful degradation products are produced, residual reagents are limited in the effluent and, to estimate the capital and operating costs (Krishnan et al.,

2017).

With the advent of higher efficiency UV lamps, visible light catalysts, and improved reactor design as well as computational modeling, AOPs have great potential for large-scale applications in near future (Chong et al., 2010). It is worth noting that UV and O3 are currently used in water and waste water treatment plants for disinfection purposes and they are both listed among the AOPs. To make it more cost effective, AOPs can also be used in combination with other methods. For example, they can be used as pre- and/or post-treatment of biological systems. The purpose for a pretreatment case is to improve biological treatability of wastewaters. Post treatment removes those

24 contaminants not completely degraded during biological treatment. To ensure that the cost basis of the combined process is low, it is necessary to limit the duration of the advanced oxidation processes (Cesaro et al., 2013).

2.2.1.1 Effect of Interferences on AOPs

The most common constituents of water matrices, which interfere with AOP processes are nitrate, sulfate, chloride, alkalinity and NOM. According to Duca et al. (2017) theses chemicals can be involved in the following processes. They can:

• absorb the incident radiation, making the process less efficient because fewer photons will

be available for the photolysis of water

• scavenge HO• radicals, consequently reducing the concentration of HO• radicals available

for reactions with target contaminants and

• produce radicals when photolyzed, which might increase the degradation efficiency

- Overall, inorganic ions can decrease or increase the reaction efficiencies. As an example, NO3

- and NO2 are both known to absorb in the UV region. Nitrous acid and nitric acid absorb UV light at 371 nm (ɛ=2900 M-1cm-1) and 201 nm (ɛ=9900 M-1cm-1) respectively, which decreases the

- reaction efficiency. On the other hand, NO3 exhibit little reactivity towards hydroxyl radicals (k <

5 −1 - 1.0 × 10 s ) but hydroxyl radical scavenging by NO2 needs to be considered (Eq. 2.2-2.4) (Duca et al., 2017).

- • -• NO3 + hυ → NO2 + O Eq. 2.2

-• • - O + H2O → OH + OH Eq. 2.3

- • • - 10 -1 -1 NO2 + OH → NO2 + OH k=1×10 M s Eq. 2.4

Carbonate and bicarbonate ions also act as hydroxyl radical scavengers (Eq. 2.5 & 2.6) and decrease the degradation rate of AOP processes (Hoigne and Bader, 1976). During these processes carbonate radical anion is produced, which react with organic compounds, but to a lower level

•- compared to hydroxyl radicals. It should be noted that CO3 can also be produced as a result of photosensitization with excited dissolved Natural Organic Matter (3NOM) (Eq. 2.7). At the same

•- • time, CO3 is scavenged to a lesser extent than OH by NOM in surface water. For this reason,

•- • CO3 can reach a higher steady-state concentration than HO , which may compensate for its lower reactivity (Duca et al., 2017).

• 2- - •- 8 -1 -1 HO + CO3 → OH + CO3 (k= 3.9 × 10 M s ) Eq. 2.5

• 3- •- 6 -1 -1 HO + HCO → H2O + CO3 (k= 8.5 × 10 M s ) Eq. 2.6

3 2- -• •- 5 -1 -1 NOM + CO3 → NOM + CO3 (k= 1.0 × 10 M s ) Eq. 2.7

2.2.1.2 Electrical Energy Per Order (EEO)

According to Bolton et al. (2001), the Electrical Energy per Order (EEO) is a useful concept for comparing the performance of ultraviolet based AOPs for the degradation of organic contaminants. EEO is the number of kiloWatt hours necessary to degrade a given pollutant by one order of magnitude in one cubic meter of water. The EEO parameter depends on a variety of factors such as the concentration and the identity of the target contaminant, amount of the oxidants used and the characteristics of the reactor. Bolton et al. (Chemistry International, 2016) emphasized on standardization of EEO in order to be useful for comparison between reactors across studies. EEO parameter is also useful for prediction of practicality of AOP methods.

2.3 Ozonation

Ozonation is a water treatment process that destroys microorganisms and degrades organic pollutants through oxidation by ozone. Ozone is a colourless gas, which is 1.5 times denser than oxygen. Also, the aqueous solubility of ozone is ten times greater than that for oxygen under normal conditions of temperature, pressure and pH (Lawrence and Cappelli, 1977). The chemical properties of ozone depend on the structure of the molecule. Ozone comprises of three oxygen atoms (O3), which decomposes rapidly in water into molecular oxygen and oxygen atom.

Ozone is mostly used in the industry for disinfection purposes as its redox potential is higher than all other common oxidants against bacteria and viruses (Ikehata et al., 2006; Leob,

2002; Paraskeva and Graham, 2002). It has a redox potential of +2.07 V compared with +1.49 V for hypochlorous acid, +1.36 V for chlorine, +0.75 V for chloramine or +1.25 V for chlorine dioxide at 25°C (Lawrence and Cappelli, 1977). The strong electrophilic nature of this molecule is due to the presence of the third oxygen atom.

In addition, ozone can reduce the concentration of iron, manganese and sulfur by producing the corresponding oxide and elemental sulfur, which are insoluble and can be removed by filtration.

It can also reduce or eliminate taste and odor problems and selectively degrade a number of recalcitrant organic pollutants in water and wastewater (Ikehata and Gamal El-Din, 2004 and 2005,

Alvares et al., 2001; Masten and Davis, 1994). Ozone is a highly toxic gas with a pungent, irritating odour. Fortunately, it can be detected by smell at a concentration 10-20 times lower than that required to cause harmful effects even after prolonged exposure (Lawrence and Cappelli, 1977).

2.3.1 Mechanism of Ozonation

Ozonation of organic pollutants involves two types of oxidation reactions, either molecular ozone reactions (ozonolysis) or hydroxyl radical reactions, depending on the reaction conditions. pH is a key factor in ozonation (Ikehata and Gamal El-Din, 2004). At low pH, molecular ozone reactions

(direct pathway- selective oxidation) are predominant, where organic compounds are subjected to the electrophilic attack of ozone molecules.

+ − 0 O3 + 2H + 2e → O2 + H2O E = 2.07 V Eq. 2.8

The molecular ozone reactions are selective to the organic molecules having nucleophilic moieties such as carbon-carbon double bonds, OH, CH3, OCH3, and other nitrogen, oxygen, phosphorus, and sulfur bearing functional groups (Alvares et al., 2001). It is also known that aromatic compounds selectively decompose through ozone treatment. These organic molecules can be destroyed in various ways, including: i) breakage of double bond and formation of aldehydes and ketones, ii) addition of an oxygen atom to benzene rings and iii) reaction with alcohols to form organic acids (Shahamat et al., 2014).

On the other hand, at high pH (>8), ozone molecules are decomposed into hydroxyl radicals

(Ikehata and Gamal El-Din, 2004; Hordern et al., 2003), which non-specifically react with a wide range of compounds (indirect pathway). There is some direct reaction of ozone with water involved in this process as well. The following equations shows the steps involved in hydroxyl radical production:

• −4 −1 −1 O3 + H2O → 2HO + O2 k2 = 1.1 × 10 M s Eq. 2.9

− •− • −1 −1 O3 + OH → O2 +HO2 k2 = 70 M s Eq. 2.10

• • •− + O3 + HO → O2 + HO2 ↔ O2 + H Eq. 2.11

• • 9 −1 −1 O3 + HO2 → 2O2 + HO k2 = 1.6 × 10 M s Eq. 2.12

• 2HO2 → O2 + H2O2 Eq. 2.13

Therefore, there is possibility of enhancing ozonation by the application of different acidic

(for the substances reactive with molecular ozone) and basic (for the remaining organic compounds refractory to ozone) cycles during the ozonation processes (Martins and Quinta-

-Ferreira, 2014). For degradation of a mixture of pollutants, it is suggested that the advantages of both mechanisms should be exploited to have better utilization of ozone.

All mechanisms involved in ozonation are shown in Scheme 2.2.

Direct oxidation of S Degradation with O3 products

O3 + S + OH-

• Reaction of Production of OH , - - - O3 and OH O2 , HO2 , H2O2

Oxidation/Reduction of S by reactive oxygen species

Scheme 2.2 Degradation of substrate (S) with O3 (Modified after Hoigne and Bader, 1976)

According to Hordern et al. (2003), in water, there are several compounds that are capable of initiation, promotion or inhibition of the radical chain reaction process. The initiators (OH−,

− 2+ •− H2O2/HO2 , Fe , formate, humic substances) can induce the formation of superoxide ion O2 from an ozone molecule. The promoters (R2–CH–OH, aryl–R, formate, humic substances, O3) are

•− responsible for the regeneration of the O2 ion from the hydroxyl radicals. The inhibitors (CH3–

− − 2− − 2− COO , alkyl–(R), HCO3 /CO3 , H2PO4 /HPO4 and humic substances) are compounds capable of consuming hydroxyl radicals without the regeneration of the superoxide anion; Equations 2.14-

2.18. The redox potential of the radicals generated are lower than those of hydroxyl radicals.

• • 9 −1 −1 HO + O3 → O2 + HO2 k2 = 3.0 × 10 M s Eq. 2.14

• − − • 7 −1 −1 HO + HCO3 → OH + HCO3 k2 = 1.5 × 10 M s Eq. 2.15

• 2− − •− 8 −1 −1 HO + CO3 → OH + CO3 k2 = 4.2 × 10 M s Eq. 2.16

• − − • 5 −1 −1 HO + H2PO4 → OH + H2PO4 k2 < 10 M s Eq. 2.17

2− − •− 7 −1 −1 HO• + HPO4 → OH + H2PO4 k2 < 10 M s Eq. 2.18

2.3.2 Production of Ozone

Methods of ozone synthesis include: (a) electric discharge of air or oxygen; (b) electrolysis of water; and (c) ultraviolet radiation of air or oxygen. Ozone production by electric discharge of air is the most efficient method amongst others and is widely used. Air is first filtered, then dried by desiccation (alumina or silica gel). This is to prevent production of nitric acid and/or oxides of nitrogen in the generator, which accelerate decomposition of the generated ozone (Lawrence and

Capelli, 1997). Clean dry air at low pressure is passed between large area electrodes separated by an air gap with a dielectric barrier discharge across driven by a pulsed power (Diaper and Evans,

1972). The efficiency of ozone production is dependent upon the rate of gas flow, applied voltage

30 and temperature of the cooling systems (for both power supply unit as well as the ozone generator vessel).

2.3.3 Disadvantages of Ozone

There has been some criticism of ozone usage as well (Mehrjouei et al., 2015; Martins and M.

Quinta-Ferreira, 2014; Nawrocki and Hordern, 2010; Pirkanniemi and Sillanp, 2002; Hordern et al., 2003). These include:

• Equipment and operational costs are high

• There is no residual effect during disinfection

• By-products of ozonation such as brominated by-products, aldehydes, ketones, and carboxylic

acids are carcinogenic. A post-filtration by activated carbon may be necessary

• Ozone reacts slowly with certain organic substances such as inactivated aromatics or saturated

carboxylic acids

• The solubility and stability of ozone in water is low, therefore, an important technological issue

is related to ozone mass transfer. A special mixing technique is required to provide proper

contact between ozone and the contaminants

• Complete mineralization of recalcitrant organics upon ozonation is low, due to low reaction

kinetics and limited mass transfer. Ozone reacts with organic matter to form aldehydes and

carboxylic acids, both of which then do not react with ozone

• For hard water, a pre-treatment step for hardness reduction is required to prevent scale

formation

• There is potential of fire hazards and health risks associated with ozone generation

• It is applicable within a limited pH range

2.3.4 Increase of Ozonation Efficiency

The above-mentioned disadvantages make the application of ozone alone for treatment of polluted water economically unattractive and is perhaps the reason why ozonation is sometimes modified by the addition of catalysts and/or irradiation. More details on catalytic ozonation are provided later.

2.3.4.1 Ozonation at High pHs

Ozone in water disintegrates to radicals such as the hydroxyl radical (HO•) and superoxide

•- radical (O2 ) through a complex series of reactions. Hydroxyl radical production is better at higher pH levels (Ikehata and Gamal El-Din, 2004; Hordern et al., 2003).

– + 3O3 + OH + H → 2HO• + 4O2 Eq. 2.19

2.3.4.2 Ozonation in Presence of Oxidants Such as Hydrogen Peroxide or Persulfate

The decomposition cycle of ozone can also be enhanced by using ozonation aided by oxidants such hydrogen peroxide or persulfate.

The reaction of ozone with H2O2 generates HO radicals (Equations 2.20- 2.26). Dissolved H2O2 in water dissociates partially into the hydroperoxide ion (HO2- ), which rapidly reacts with ozone to initiate a radical chain mechanism that generates hydroxyl radicals (Staehelin and Hoigne (1982);

Glaze et al., 1987).

- + H2O2↔HO2 + H Eq. 2.20

- • • - HO2 + O3→HO 2+ O 3 Eq. 2.21

∙ - + ∙ O 3 + H ↔ HO 3 Eq. 2.22

• • HO 3→ HO + O2 Eq. 2.23

• • - + HO 2 → O 2 + H Eq. 2.24

• - ∙ - O 2 + O3 → O2 + O 3 Eq. 2.25

In summary, the combination of reactions leads to the following:

• 2O3 + H2O2 → 2HO + 3O2 Eq. 2.26

In this case also a two-step reaction is beneficial. An ozonation step first, to degrade highly reactive substances with ozone and introducing a dose of hydrogen peroxide in the second step.

Compared to other AOPs, ozonation by H2O2/O3 systems seem to be the most established AOP in remediation for water pollution and there is a history of implementation of H2O2/O3 systems.

Persulfate is another oxidants used to enhance ozone decomposition. Persulfate is a strong water-soluble oxidant (E = 2.1 V), with the sulfate moieties substituted for hydrogens in H2O2, which significantly increases its stability (Chen and Huang, 2015). Enhancement by persulfate is common when working at pH values above 8.0 (Chiang et al., 2006). This is attributed to the ability of O3 to initiate hydroxyl radical formation. The persulfate can be activated to initiate sulfate radical under the effect of hydroxyl radicals (Equations 2.27- 2.33). The cooperative effect of hydroxyl and sulfate radicals in removing COD is much higher than ozonation (Abu Amr et al.,

2013).

- - O3 + OH → HO2 + O2 Eq. 2.27

- . - O3+ HO2 →HO2 + O3 Eq. 2.28

- • - O3 + H2O →HO + O2+ OH Eq. 2.29

• 2- •- - HO + S2O8 → SO4 + HSO4 + ½ O2 Eq. 2.30

•- 2- • + SO4 + H2O → SO4 + HO + H Eq. 2.31

•- - 2- • SO4 + OH → SO4 + HO Eq. 2.32

Overall reaction:

2− • 2− •− + S2O8 + OH → SO4 + SO4 + 1/2O2+H Eq. 2.33

•- It is suggested that under acidic condition, a few HO• are generated, but more SO4 could be formed via the asymmetric break of peroxide bond (Yang et al., 2016a). Neutral conditions favor direct reactions with O3 and the target pollutant with a decrease in the radical species generation.

2.3.4.3 Ozonation Aided by UV Light

The ozone molecule in presence of UV light in range of 200-300 nm is decomposed into

1 O2 and the atomic singlet oxygen ( D), which is highly reactive and rapidly reacts with the water molecules to produce H2O2. Then, the hydrogen peroxide reacts with O3 to produce hydroxyl radicals (Sonntag and Gunten, 2012). A simplified reaction sequence is shown below:

1 O3 + hν → O( D) + O2 Eq. 2.34

1 O( D) + H2O → H2O2 Eq. 2.35

• H2O2→ 2 OH Eq. 2.36

Overall reaction:

• 2O3 + H2O2 → 2 OH + 3 O2 Eq. 2.37

2.3.4.4 Catalytic ozonation

Catalytic ozonation has received increasing attention during past decades as a viable alternative to single ozonation. Catalytic ozonation is a promising technology for the effective removal of water and wastewater contaminants that are refractory to conventional oxidation treatments.

The main advantages of the catalytic processes with respect to traditional non-catalytic ozonation listed in literature are (Shahamat et al., 2014; Nawrocki and Hordern, 2010; Horden et al., 2003; Pirkanniemi and Sillanp, 2002):

• better ozone utilization

• increased contaminant removal efficiency

• enhanced carbon mineralization of different toxic and bio recalcitrant organic compounds

• reduced selectivity

• controlled decomposition of O3

• optimized economic efficiency

• effective formation of hydroxyl radicals even at low pH (not always true)

Also, reactions of radicals with organic and inorganic molecules are, as opposed to direct molecular ozone reactions, fast and non-selective

2.3.4.4.1 Mechanism of catalytic ozonation

Catalytic ozonation can be divided into two main processes: homogeneous catalytic ozonation, which is catalyzed by transition metal ions present in aqueous solution, and heterogeneous catalytic ozonation which occur in the presence of solid catalysts.

It should be noted that the starting point for catalytic ozonation processes has been observation of an improved efficiency in presence of a so called catalyst and then the mechanisms have been proposed. In most cases because the process is called an AOP, production of hydroxyl radicals has been tested by hydroxyl radical scavengers such as tertiary butyl alcohol (Nawrocki and Horden, 2010).

2.3.4.4.1.1 Homogeneous Catalytic Ozonation

Some transition metal cations have been shown to improve ozonation efficiency. Table 2.5 lists some of theses catalysts for degradation of a variety of organic compounds (Nawrocki and

Horden, 2010).

Table 2.5 Homogeneous catalytic ozonation. Modified after Nawrocki and Hordern (2010) Transition Metals (Catalyst) Organic Compounds Mn Carboxylic acids (oxalic, pyruvic, glyoxalic, propionic, sulfosalicylic), N-methyl-p-aminophenol, atrazine, dinitrotoluene, dichlorophenol Mn, Fe, Cr, Ag, Cu, Zn, Co, Cd humic substances Mn, Fe, Fe chlorobenzene Mn, Fe Simazine Pb, Cu, Zn, Fe, Ti, Mn 2-cholorophenol Fe, Fe, Mn, Co Lignin sulfonate Ni, Fe, Mn, Zn, Sr, Cr, Cd, Hg, Cu 1,3,6-naphthalenetrisulfonic acid Co oxalic acid, pyruvic acid Cu oxalic acid, pyruvic acid Mn, Fe, Fe, Zn, CO, Ni azo dyes Ce phenol Fe oxalic acid Fe, Cu, Co, Mn, Ni, Zn, Cr carboxylic acids Fe, Cu, Ni, V, Mo dinitrobenzene Mn, Fe, MO, Cu, Ni benzoic acid Mn propionic acid

A variety of different mechanisms has been proposed in the literature to explain the observed effects in presence of th e transition metals. But there is no specific proposed mechanistic pathway that is widely accepted. It will be seen later that catalytic ozonation depends on several parameters, which could be different in different labs.

However, based on the common features in the proposed mechanisms, two major mechanisms of homogeneous catalytic ozonation are presented here (Nawrocki and Hordern, 2010):

1. Decomposition of ozone by metal cations followed by generation of hydroxyl radicals (Sauleda and Brillas, 2001; Piera et al., 2000; Gracia et al., 1996). Fe(II) is chosen here as an example

2+ (Equations 2.38-2.40). The mechanism of the Fe /O3 systems was proposed by Piera et al. (2000)

2+ and can be expressed using the following equations. The Fe /O3 system involves the direct reaction of Fe2+ with ozone resulting in the production of HO•:

2+ 2+ Fe + O3 → FeO + O2 Eq. 2.38

2+ 3+ • − FeO + H2O → Fe + HO + OH Eq. 2.39

• • RH+ HO → R + H2O Eq. 2.40

+ 2+ 3+ FeO2 is also able to oxidise Fe to Fe (at a slower rate) with the termination of the chain reaction:

2+ 2+ + 3+ FeO + Fe + 2H → 2Fe + H2O Eq. 2.41

In most references, the proposed mechanism doesn’t include regeneration of Fe2+. But two possibilities are mentioned by Barceló and Petrovic (2008) after proposing a second mechanism

37 for production of hydroxyl radicals, which is production of H2O2 as a result of O3 and water interaction and shared reactions with the classical Fenton homogeneous process:

2+ 3+ • - Fe + H2O2→Fe + HO + HO Eq. 2.42

Fe2++ HO•→Fe3++ HO− Eq. 2.43

• • RH +HO →H2O+ R Eq. 2.44

R•+ Fe3+→ R++Fe2+ Eq. 2.45

3+ + + Fe + H2O2↔Fe-OOH2 + H Eq. 2.46

+ • 2+ Fe-OOH2 → HO2 +Fe Eq. 2.47

3+ • 2+ + Fe + HO2 → Fe +O2+H Eq. 2.48

• • OH + H2O2 →H2O + HO2 Eq. 2.49

They suggested reduction of Fe3+ with organic radicals (Eq. 2.45) or with the intervention of

• hydroproxy radicals, HO2 . (Equations 2.46-2.49)

2+ Fe /O3/UV system is also discussed in literature under the topic of homogeneous catalytic ozonation (Nawrocki and Hordern, 2010; Hordern et al. 2003). Irradiation with UV light also causes generation of free hydroxyl radicals:

O3 + H2O + hν → H2O2 + O2 Eq. 2.50

• H2O2 + hν → 2HO Eq. 2.51

2+ The efficiency of the O3/Fe /UV is based on the enhanced generation of hydroxyl radicals as a result of the photo-Fenton reaction:

2+ 3+ • − Fe + H2O2 → Fe + HO + OH Eq. 2.52

3+ 2+ • + Fe + H2O + hν → Fe + HO + H Eq. 2.53

2+ In this case Fe is reacting with H2O2 generated by UV light and O3 to enhance hydroxyl radical production (Fenton reactions) and Fe2+ is generated using UV light.

Except for ozone decomposition and hydroxyl radical generation, the homogeneous catalyst can also form complexes with organic molecules such as carboxylic acids.

2. Complex formation between organic molecule and the catalyst and subsequent oxidation of the complex (Nawrocki and Hordern, 2010; Hordern et al., 2003).

As proposed by Pines and Reckhow in 2002, the process of oxalic acid ozonation with the

Co(II)/O3 system (at pH=6) follows a two step reaction. In the first step, a Co(II)–oxalate complex is formed, which is subsequently oxidised by ozone to form Co(III)–oxalate (Scheme 2.3). The metal centre is suspected to be the site of attack. The partial donation of electron density from oxalate ion to Co(II) may increase the reactivity of Co(II)–oxalate when compared to free Co(II).

Subsequently, the decomposition of Co(III)–complex occurs with the formation of an oxalate radical and Co(II).

2+ 2- Co + C2O4 → CoC2O4

+ - CoC2O4 + O3→ CoC2O4 + O3

+ 2+ • - CoC2O4 → Co + C2O4

• - • C2O4 + OH , O3, O2 → 2CO2

Scheme 2.3 Proposed mechanism for oxalic acid catalytic ozonation by Co(II)/O3 system.

Modified after Hordern et al. (2003)

The rate of both oxalate removal and ozone decomposition increases with decreasing pH from 6.7 to 5.3. The phenomenon of enhanced ozone decomposition with decreasing pH is contradictory to the typical relationship between ozone decomposition and pH. This indicates that the principal reaction pathway in not the auto decomposition of ozone initiated by hydroxide ions (Hordern et al., 2003; Pines and Reckhow, 2002).

Consistent with the previous example, Andreozzi et al. (1992) also found that Mn(II) accelerates the oxidation of oxalic acid under acidic conditions without production of OH radicals. The authors proposed that Mn(II)-catalysed oxidation was proceeding through complexing between oxalic acid and Mn(III), forming an intermediate product which might be easier oxidized by ozone. The following equations represent Mn (II)-catalysed ozonation of oxalic acid (OA) in aqueous solution

(Legube and Leitner; 1999):

2+ + 4+ Mn + O3+ 2H →Mn + O2+ H2O Eq. 2.54

Mn4++Mn2+ →2Mn3+ Eq. 2.55

3+ 2- 3+ 2- Mn +nAO →Mn (OA )n Eq. 2.56

3+ 2- 2+ •- 2- Mn (OA )n→Mn + OA + (n-1)OA Eq. 2.57

•- + • OA + O3+H → 2CO2+ O2+ HO Eq. 2.58

HO•+ OA 2-→…… Eq. 2.59

Based on these mechanisms the ozonation of oxalate proceeds without formation of hydroxyl radicals, even though hydroxyl radicals are formed as secondary by-products (Nawrocki, 2013).

As mentioned earlier, the mechanisms reported in literature for homogeneous catalysis are contradictory and not all the processes occurring are understood. It should be emphasised here that several parameters such as: pH of solution and concentration of the transition metal ion can influence both the efficiency of the process and its mechanism.

From a practical poit of view, since metal ions can be toxic, they have to be removed from treated water which makes homogeneous catalytic ozonation not as attractive as heterogeneous catalytic ozonation. The oxidation efficiency and selectivity of the latter can be improved by proper selection or modification of the catalysts. It can also be easily recovered from the aqueous medium

(Nawroscki and Hordern, 2010).

At least in theory finding the mechanism of homogeneous catalysis could provide an explanation for improved ozonation efficiency for wastewater containing the above-mentioned transition metals.

2.3.3.4.1.2 Heterogeneous Catalysis

In heterogeneous catalytic ozonation, the catalyst is in solid form while the reaction proceeds in bulk water or on the surface of the catalyst. In many cases the formation of hydroxyl radicals is expected to be responsible for the catalytic effects. According to Nawrocki (2013) the catalytic effect is possible when at least one of the following three conditions is fulfilled (Scheme

2.4):

- ozone is adsorbed on the surface of the catalyst

- organic molecule is adsorbed on the surface of the catalyst

- or both, ozone and organic molecule are adsorbed on the catalyst surface.

The catalytic effect cannot be expected when none of the reactants is adsorbed on the catalyst surface.

Scheme 2.4 Three possible cases of heterogeneous catalysis for degradation of Organic

Materials (OM). Modified after Nawrocki (2013)

The most widely used catalysts in heterogeneous catalytic ozonation are:

• Metal oxides (e.g. MnO2, TiO2, Al2O3, FeOOH and CeO2)

• Metals on support/ Metal Oxides

• Activated carbon

Metal Oxides

Table 2.6 lists some of the heterogeneous catalysts for degradation of a variety of organic compounds. In order to understand the mechanism of heterogeneous catalytic ozonation, main physical and chemical properties of the metal oxides should be known. According to Hordern et al. (2003) the main physical variables are surface area, density, pore volume, porosity, pore size

42 distribution as well as mechanical strength, purity and commercial availability. The most important chemical properties are: chemical stability and especially the presence of active surface sites such as Lewis acid sites, which are responsible for catalytic reactions. The main parameter which determines the catalytic properties of metal oxides, is acidity and basicity. Hydroxyl groups are present on all metal oxides surfaces. However, the amount and the properties of the hydroxyls depend on the metal oxide. The hydroxyl groups formed at metal oxide surface behave as Brönsted acid sites. Lewis acids and Lewis bases are sites located on the metal cation and coordinatively unsaturated oxygen, respectively (Nawrocki et al., 1993). According to Nawrocki et al. both

Brönsted and Lewis acid sites are thought to be the catalytic centres of metal oxide. The hydroxyls are also main ion-exchange sites. The ion-exchange properties depend on pH of the surrounding environment. Thus, the oxides may exchange cations at pH higher than those of oxides’ pHpzc and anions at pHs lower than their pHpzc.

Table 2.6 Heterogeneous catalytic ozonation−metal oxides (Modified after Nawrocki and Hrdern, 2010) Catalyst Organic compounds MnO2 Carboxylic acids (oxalic, pyruvic, sulfosalicylic, propionic, glyoxalic) N-methyl-p-aminophenol, atrazine, phenol Al2O3 Carboxylic acids (oxalic, acetic, salicylic, succinic), 2-chlorophenol, chloroethanol, NOM, dimethylphthalate TiO2 Oxalic acid, carbamazepine, naproxen, nitrobenzene FeOOH p-Chlorobenzoic acid, NOM ZnO p-Chlorobenzoic acid TiO2/AC Methylene blue CuO/Al2O3 Oxalic acid Co(OH)2 p-Chloronitrobenzene MgO Dye

In general, adsorption of organic compounds on metal oxide surfaces depends on its polarity. The strongest adsorption can be expected for organic ions, however their adsorption depends on pH. Polar compounds may get adsorbed on the surfaces, while nonpolar organics are

43 not adsorbed on the surfaces unless some hydrophobic sites are present there (e.g. in high silica zeolites). Some oxides, particularly high silica zeolites, may contain hydrophobic surface that enhances adsorption of organic compounds.

On the other hand, decomposition of ozone happens on Lewis acid centres of metal oxides.

Lewis acid sites are usually found on dry metal oxides such as alumina, zirconia, titania. These sites are extremely reactive and they dissociate water molecules. Unfortunately, because of this competition, there is no direct evidence of ozone adsorption on metal oxides in the presence of water (Nawroski and Hordern, 2010). Roscoe and Abbatt (2005) reported that there is competition between ozone and water for adsorption sites on alumina in the gaseous phase and that water molecules reveal much higher affinity towards alumina’s surface sites than ozone, which results in the inability of ozone molecules to adsorb. But still there are some reports in literature on ability of oxides to catalyze the decomposition of O3 and oxidation of contaminants by hydroxyl radicals.

The mechanisms suggested for MnO2 and FeOOH catalysts, based on well documented published papers are listed in Table 2.7 & 2.8 respectively. These examples show contradictory reports on catalytic activity and reaction mechanism of oxides.

Table 2.7 The mechanisms of catalytic ozonation proposed for MnO2 (Modified from Nawrocki 2013) Catalytic system Reaction responsible for catalytic effect MnO2 + oxalic acid + O3 Formation of surface MnO2–oxalic acid complex, ozone reacts with the complex MnO2 + sulfosalicylic acid (SSal) + O3 MnO2–SSal surface complex + O3 Catalyst decomposes ozone, TBA does not influence the MnO2/GAC + nitrobenzene + O3 catalytic ozonation Formation of surface MnO2–oxalic acid complex, ozone MnO2/GAC + oxalic acid + O3 reacts with the complex * GAC: Granular Activated Carbon

Table 2.8 Mechanisms of catalytic ozonation suggested for Fe hydroxyoxides (Nawrocki, 2013) Catalytic system Reaction responsible for catalytic effect • FeOOH + O3 & FeOOH + O3 + pCBA OH in interphase Ozone reaction with surface Complex of oxalic acid FeOOH + O3 + OA with Fe2O3

FeOOH + O3 + FeOOH + O3 + nitrobenezene Decomposition of ozone on non-dissociated hydroxyls FeOOH + O3 + OA Adsorption of ozone on Lewis sites *pCBA: para- chlorobenzoic acid; OA: oxalic acid

Metals on metal oxides

Table 2.9 lists some of the heterogeneous metal on support/metal oxide catalysts for degradation of a variety of organic compounds.

Table 2.9 Heterogeneous catalytic ozonation – metals on supports (Nawrocki and Horden, 2010) Catalyst Organic compound

Pt, Pb, Pd, Ag, Co, Ru, Ir, Rh, Re/Al2O3, SiO2, activated carbon Formic acid Humic substances, salicylic Cu(10 wt%)/Al2O3, Cu(5 wt%)/TiO2, Cu(5 wt%)/clay acid, peptides V-O/TiO2, V-O/SiO2 Sulfosalicylic acid

Ru(2%)/CeO2 Succinic acid

Ru(2%)/CeO2-TiO2 Chloroacetic and succinic acid

Ru(5%)/Al2O3, Pt(5%)/Al2O3, Pt(5%)/AC p-Chlorobenzoic acid

Ru(0.1%)/Al2O3 Dimethyl phthalate

Cu-ZrO2/Al2O3, Ru-CeO2/TiO2 Pyruvic acid, succinic acid

Rh-CeO2 Pyruvic acid

The exact mechanism is not known but two possible routes have been proposed by Legube and

Leitner (1999); Scheme 2.5.

Scheme 2.5 Two main possibilities of catalytic ozonation pathways: (up) adsorption on catalyst and oxidation by ozone or •OH radical of adsorbed organic (down) •OH or other radical species generation by reaction of ozone with reduced metal catalyst, and oxidation of organic by oxidized metal and/or in homogeneous solution. Adapted from Legube and Leitner (1999)

According to these authors, in the first mechanism, organics adsorb on the surface of the catalyst and get oxidized by ozone or hydroxyl radicals and then they desorb into the solution happens. In the second mechanism, both ozone and organic molecules adsorb on the surface of the catalyst.

Hydroxyl radical is generated by the oxidation of reduced form of metal by ozone, then adsorbed organic molecule is oxidized, and the metal is reduced again. Both routes were proposed and there is still no clarity on what the dominant mechanism is.

Activated Carbon

Activated carbon (AC) acts not only as the adsorbent but also as a catalyst in promoting ozone oxidation (Hordern et al., 2003). Combined ozonation and adsorption offer strong synergetic effects on the removal of many contaminants. Two routes of ozone decomposition are proposed:

(i) AC acts as an initiator of O3 decomposition into hydroxyl radicals and (ii) ozone reacts with surface groups, generates adsorbed H2O2, which reacts with ozone in bulk solution yielding OH radicals (Nawrocki, 2013; Liu et al., 2009). It should be noted that the amount of AC applied ranges from 0.1g L-1 to 40 g L-1. Application of large amount of catalyst is not practical, even if it is AC.

2.3.4.4.2. Advanced oxidation Processes (AOPs) and Catalytic Ozonation

The term advanced oxidation process is defined as the oxidation process, which generate hydroxyl radicals in enough quantity to affect water treatment. These processes generally use a combination of oxidation agents (ozone, hydrogen peroxide), irradiation (UV, ultrasound) and catalysts as a means of generating hydroxyl radicals (Hordern et al., 2003; Huang et al., 1993).

Hydroxyl radical is one of the most reactive free radicals and one of the strongest oxidants:

• + − 0 HO + H + e → H2O E = 2.8V Eq. 2.60

The rate at which hydroxyl radicals react with organic molecules is usually in the order of 106–

109M−1 s−1 and can be expressed as follows:

• −d[M]/dt= kOH[M][HO ] Eq. 2.61

The radicals that are formed after HO• reacts with organic molecules disproportionate or combine with each other, forming different labile intermediates which react further to produce peroxides, aldehydes, acids, hydrogen peroxide, etc. (Hoigné and Stucki, 1988). Hydroxyl radicals, due to their high reactivity, can react with almost all types of organics (ethylenic, lipid, aromatic, aliphatic) and inorganics (anions and cations), therefore they are not selective.

Since catalytic ozonation provides fast degradation of organic pollutants and also effective mineralization of both micropollutants and natural organic matter, sometimes, it is classified as one of the AOPs. Based on the discussed mechanism in the previous section, it appears that catalytic activity in catalytic ozonation does not necessarily mean generation of hydroxyl radicals and there are many examples of such activity in relevant literature without generation of radicals.

Therefore, catalytic ozonation cannot be categorized as one of the AOPs unless the mechanism involves production of hydroxyl radicals.

2.3.4.4.3 Limitations of Catalytic Ozonation

Unfortunately, despite several papers published in the field of catalytic ozonation and introducing several effective catalysts, the mechanisms of catalytic processes are still largely unknown. Even, the same catalysts, studied by different research groups, lead to different, sometimes contradictory results (Nawrocki, 2013). As it was discussed in the previous section, when proposing a mechanism, some authors suggested radical pathways involving ozone decomposition and hydroxyl radical formation, others introduced pathways, which do not involve hydroxyl radical formation. The reason could be that different experiments were performed under different conditions and sometimes the pH was uncontrolled; the physical adsorption of contaminant on catalyst system was ignored; adsorption of byproducts on the catalyst surface led

48 to false TOC or COD removal; running the reaction once as opposed to several runs (Nawrocki,

2013); or not considering deactivation of catalysts.

Even if all the above parameters are controlled, the verification of mechanisms governing catalytic ozonation seems to be particularly problematic, as the usage of catalysts in aqueous solutions will lead to competition between water, ozone and organic compounds for catalytic

(adsorptive) active sites (Nawrocki and Hordern, 2010). Understanding adsorption, desorption is also critical. Yet another challenge of catalytic ozonation processes is sustaining the decomposition of ozone to assure initiation, promotion and maintenance of radical chain reactions (Yavas and

Ince, 2017).

Other limitations of catalytic ozonation are the complexity of the catalysts (Table 2.9), high catalyst loading requirement, high cost of synthesis and the leaching of materials into the liquid system causing high consumption of catalyst and being regarded as a new pollutant in a treated wastewater (Shahamat et al., 2014).

2.4 Persulfate

2- 0 Peroxydisulfate or persulfate anion (S2O8 ) is a strong water-soluble oxidant (E = 2.1 V).

It is a hydrogen peroxide with the sulfate moieties substituted for hydrogens, which significantly increases its stability (Chen and Huang, 2015). Most common persulfate compounds are ammonium, potassium and sodium persulfate (Waclawek et al., 2017; Tsitonaki et al., 2010).

Ammonium persulfate has the lowest molecular weight and the highest solubility (85 g/100 g of

o H2O at 25 C) in water amongst the others. Sodium persulfate is mostly favored (solubility as high as 73/100 g) in literature though.

Once in water, persulfate can produce reactive species such as hydroxyl radicals and sulfate radical anions. However, different reactive species dominate at different pHs. In acidic pHs, persulfate hydrolyzes into hydrogen peroxide (E= 1.77 V) or peroxymonopersulfate anions with the oxidation potential of 1.44 V (Eq. 2.62- 2.63) (Kolthoff & Miller, 1951). Under alkaline activation conditions through the addition of hydrogen radical, persulfate generates both sulfate radicals and superoxide (Eq. 2.64) (Furman et al., 2010). Under highly alkaline conditions sulfate radical can react with hydroxide radicals to form hydroxyl radicals (Eq. 2.65) (Watts and Teel,

2006):

2- - S2O8 + 2H2O → H2O2 + 2HSO4 Eq. 2.62

2- - - S2O8 + H2O → HSO5 + HSO4 Eq. 2.63

2- 2- -• -• + 2S2O8 + 2H2O → 3SO4 + SO4 +O2 + 4H Eq. 2.64

-• - 2- • SO4 + OH → SO4 +OH Eq. 2.65

Overall, it is evident that persulfate solutions may contain several different oxidants and radical species, but the dominant species is determined by the pH of the solutions. This increases the probability of reducing the target contaminant’s concentration, which is because mixtures of oxidizing species may cause multiple pathways for degradation of the contaminants. However, such diversity of oxidant species makes the assessment of the stoichiometric amount of persulfate needed to decompose the contaminants problematic, and thus it is common practise to revert to the basic, two electron transfer associated with the persulfate anion to determine the stoichiometric persulfate demand (Tsitonaki et al., 2010). The oxidation potential of reactive species potentially present in activated persulfate systems are described in Table 2.10.

Table 2.10 Potential reactive species and their redox potentail in an activated persulfate system (Tsitonaki et al., 2010)

Species Redox potential (V) Hydroxyl radical (OH•) +2.8 •- sulfate radical (SO4 ) +2.6 2- persulfate anion (S2O8 ) +2.1

hydrogen peroxide (H2O2) +1.8

Ozone (O3) +2.07 - Peroxymonopersulfate (HSO5 ) +1.4 •- Superoxide (O2 ) -0.2

2.4.1. Persulfate activation

It was mentioned in the previous section that by shifting the reaction pH to higher levels, reactive species (hydroxyl and sulfate radicals) can be produced in persulfate system. This is called activation of persulfate. Without activation, persulfate may react with some organic compounds with a lower process efficacy than the activated persulfate.

There are different ways to activate persulfate, creation of radicals under alkaline pHs is one of them. Besides that, activation can also be with iron, minerals, heat, UV, O3, activated carbon and electrochemical methods (Petri et al., 2011; He et al., 2014; Zhang et al., 2015; Kolthoff and Miller,

1951; Waldemer et al., 2007).

The following section provides a short summery of these methods:

2.4.1.1 Heat Activation

Persulfate forms two sulfate radicals through break down of peroxide bond by providing sufficient thermal energy (Kolthoff & Miller, 1951). This activation energy depends on pH and it is reported to be the lowest at neutral and acidic pHs. The activation energies are 119-129, 134-

139 and 100-116 kJ mol-1, for neutral, basic and acidic pHs respectively (House, 1962). Further

51 formation of HO• will occur at basic pHs. Thermal activation of persulfate ranges between 40 to

70°C (Zhang et al., 2015; Ghauch et al., 2012).

2- -• S2O8 + heat → 2SO4 Eq. 2.66

Thermal activation of persulfate can be implemented via conventional or microwave heating.

Overall, because of the high energy consumption this method is not considered cost effective.

2.4.1.2 UV Light Activation

UV activation of persulfate is typically done at wavelength of 254 nm to generate a pair of sulfate radicals by cleavage of the persulfate’s peroxide bond (similar to heat activation energy)

(Dogliotti & Hayon, 1967). For activation of persulfate, application of UV at range of 248-351 nm is reported in literature (Herman, 2007). Application of persulfate along with UVA has been reported in literature by Perez-Sicairos et al. (2016) for degradation of nitrobenzene. These results are useful as low energy sources such as sunlight or LED light can be used. It should however be noted that persulfate activation with longer wavelength than 254 nm needs extended exposure periods or additional persulfate activators (Matzek and Carter, 2016).

2.4.1.3 Metal Activation Using Transition Metals or Chelated Metals

Transition metals can initiate the formation of the sulfate radicals (Tsitonaki, et al., 2010) by an electron transfer step from the metal ions (Zhang er al., 2015). The last equation shows (2.69/

2+ •- 2.71) that M is also an intrinsic scavenger of sulfate radicals, therefore the generation SO4 is limited.

2- n+ •- 2- n+1 S2O8 + M → SO4 + SO4 + M Eq. 2.67

-• n+ 2- n+1 2SO4 + M → SO4 + M Eq. 2.68

2- 2+ -• 2- 3+ S2O8 + Fe → SO4 + SO4 + Fe Eq. 2.69

•- 2+ 2- 3+ 2SO4 + Fe → SO4 + Fe Eq. 2.70

Several transition metals such as Fe(II), Mn(II), Ce(II), Co (II), etc. have been studied as metal activators (Waclawek et al., 2017; Devi et al., 2016; Zhang et al., 2015; Anipsitakis and

Dionysiou, 2004;). The non-toxic nature and availability of iron makes it the most applied

•- transition metal to produce SO4 . The reduced efficacy of the organic compounds degradation as

•- a result of the fast conversion of Fe (II) into Fe (III), which acts as a SO4 scavenger (Buxton et al., 1997), can be overcome by choosing a suitable Fe (II) dose for persulfate activation. Addition of Fe(II) or thiosulfate in a stepwise mode is also helpful (Buxton et al., 1997).

Application of the chelated transition metals also decreases the iron required for persulfate activation (Kattel E., 2018; Liang et al., 2004b). It also helps to keep the iron in solution under neutral pH. Ethylenedinitrilotetraacetic acid (EDTA) and citric acid are the preferred chelating agents as EDTA forms strong Fe (II) complexes and citric acid is favored for its biodegradability and overall feasibility (Liang et al., 2004: Rastogi et al., 2009; Tsitonaki et al., 2010; Han et al.,

2015). It should be taken into consideration that some organic chelating agents, e.g. ethylenediamine-N, N’-disuccinic acid, might interfere with the pollutant degradation as a secondary contaminant (Yan and Lo, 2013).

2.4.1.4 Alkali Activation

As mentioned earlier, persulfate in highly alkaline conditions (pH> 11) generates sulfate as well as hydroxyl radicals. According to Furman et al. (2010) the initial step of the proposed

2- mechanism is the base-catalyzed hydrolysis of persulfate to peroxomonosulfate (SO5 ) and sulfate

2- (SO4 ). Persulfate likely forms an activated complex with hydroxide that weakens the S-O bond and as a result, the S-O bond dissociates:

_ - - OH - - 2- + 3OS-O-O-SO3 + H2O [ 3OS-O-O ] + SO4 + 2H Eq. 2.71

A similar dissociation of the remaining S-O bond in peroxomonosulfate results in the formation

- of sulfate and hydroperoxide (HO2 ), which the conjugate base of hydrogen peroxide:

- - - OH - 2- + [ 3OS-O-O ] +H2O HO2 +SO4 + H Eq. 2.72

Peroxomonosulfate rapidly decomposes to hydroperoxide and sulfate at basic pH. The following net reaction results:

- - OH- - 2- + 3OS-O-O-SO3 + 2H2O HO2 + 2SO4 + 3H Eq. 2.73

The hydroperoxide formed from the hydrolysis of one persulfate molecule then reduces another

•- persulfate molecule, generating sulfate radical (SO4 ) and sulfate anion, while hydroperoxide is

•- oxidized to superoxide (O2 )

- - - - OH •- 2- + •- HO2 + 3OS-O-O-SO3 SO4 +SO4 + H + O2 Eq. 2.74

These two reactions yield the following for persulfate activation under basic conditions:

2- 2- •- •- + 2S2O8 + 2H2O →3SO4 + SO4 +O2 + 4H Eq. 2.75

Furthermore, in highly alkaline conditions, sulfate radical reacts with hydroxide to form hydroxyl radical (OH•):

•- - 2- • SO4 + OH → SO4 + OH Eq. 2.76

Since sulfuric acid is produced in persulfate reactions, additional alkalinity might be required in this case. The disadvantage of this activation method is its required high alkalinity, which might affect the natural processes of water or soil.

2.4.1.5 Other Oxidant Activation

Persulfate can also be used in combination with other oxidants such as hydrogen peroxide or O3 in field applications (Tsitonaki et al., 2010). Hydrogen peroxide activation of persulfate has been applied for soil remediation (Tsitonaki et al., 2010) and landfill leachate treatment (Hilles et al., 2016). The knowledge about the interaction between these two oxidants is still unknown, but

• •- it is proposed that H2O2 is decomposed into HO , which then activates persulfate to generate SO4

(Lominchar et al., 2018). Another suggestion is that the exothermic reactions of H2O2 propagate

•- •- SO4 formation by heat (Tsao and Wilmarth, 1960). In turn, SO4 can increase the formation of

HO•, which results in a multi-radical system (Lominchar et al., 2018):

• 2- •- − HO + S2O8 → SO4 + HSO4 + 1/2O2 Eq. 2.77

•- 2− • + SO4 + H2O → SO4 + HO + H Eq. 2.78

•- - 2− • SO4 + OH → SO4 + HO Eq. 2.79

A similar mechanism is involved when persulfate is indirectly activated by ozone (Yang et al.,

2016a):

- − O3 + OH → HO2 + O2 Eq. 2.80

- • - O3 + HO2 → HO2 + O3 Eq. 2.81

- • - O3 + H2O → HO + O2 + OH Eq. 2.82

•- • − SO4 + HO → HSO4 + 1/2O2 Eq. 2.83

2- • 2− + S2O8 + H2O2 → HO2 + 2SO4 + 2H Eq. 2.84

The indirect activation by ozone is common when pH is above 8.0 (Chiang et al., 2006). It is

•- suggested that under acidic conditions, a few HO• are generated, but more SO4 could be formed via the asymmetric break of peroxide bond when persulfate is activated by acid (Yang et al.,

2016a). Neutral conditions favor O3 direct reactions with the target pollutant with a decrease in the radical species generation.

2.4.1.6 Carbon Activated Persulfate

Recently, carbonaceous materials have been applied to activate persulfate (Sun et al. 2014).

These could be classified as a type of heterogeneous activators showing a certain advantage in being non-metallic. If a non-metal activator is used, the treated water will be free from metals and secondary contamination.

According to Duan et al. (2015), carbon materials like nano-diamonds, graphene and carbon nanotubes (CNTs) have high chemical and thermal stability, ultrahigh pore volume and large specific surface area believed to overcome the issues related to the use of toxic metal activators. For example, activated carbon (AC), pristine or surface nitrogen-modified carbon nanotubes (CNT; more concentrated pore size distribution than AC) (Yao et al. 2010) and nitrogen modified reduced graphene oxide could effectively activate persulfate by generating sulfate radicals. Contaminants such as phenols (Sun et al. 2014), AO7, methyl tert-butyl ether (MTBE), and PFOA (White et al. 2011; Huling et al. 2012; Lee et al. 2013) have been degraded successfully by this method.

There are several controversies whether the persulfate activation mechanism on

•- • carbonaceous materials is based on radical (SO4 , OH ) or non-radical active species. Some of the mechanisms proposed for production of radical species are: considering oxygen functional groups

56 on AC surface as the potential active sites for persulfate activation; CNTs surface acting as an excellent electron bridge in activation of persulfate to oxidize adsorbed water, electron transfer from CNTs to persulfate (Lee et al., 2015; Liang et al., 2009; Matzek and Carter, 2016).

2.4.2 Overall Comparison of Activation Methods

Although UV irradiation and heating are common activation techniques for persulfate, high cost and energy input limit their wide applications (Liu et al. 2014). Owing to their low cost and natural abundance, transition metal activation of persulfate is the most extensively used method to date (Li et al. 2016). However, a large amount of chemicals or chelate reagents is required due to

•- the intensive scavenging of SO4 by metal ions (Chen et al., 2016; Fang et al. 2015). In addition, the potential adverse effect of some metal ions such as Co (II) on human health also needs to be considered in the homogenous activation processes (Chen et al., 2016; Feng et al. 2015). Even though the loading of metal ions on an appropriate support can partially overcome the drawbacks of the homogeneous activation, the metal leaching problem cannot be completely resolved (Cai et al. 2015).

Overall, persulfate activated by Fe(II) is preferred over the other methods due to environmental safety considerations. Also carbon based materials are non-metallic species free from metal leaching problems, and if successful the treatment technology is considered as a green technology. But when aiming for higher efficacies, radiation-based activation is used.

•- 2.4.3 Quenching of SO4

•- Due to the complexity of the aquifer environment, the yield of SO4 - and the efficiency of

2- S2O8 -based reactions can be impacted by the chemical characteristics of groundwater such as pH,

Cl- content and alkalinity. A few chemical processes can occur, which may impact the results.

- - •- First, Cl , HCO3 and pH can scavenge SO4 to generate secondary radical species including chlorine atom, carbonate radical and hydroxyl radical (Li, 2017):

•- - 2− • SO4 + Cl → SO4 + Cl Eq. 2.85

•- - 2− •- + SO4 + HCO3 → SO4 + CO3 +H Eq. 2.86

•- - 2− • SO4 + OH → SO4 + HO Eq. 2.87

According to Li (2017) the treatment efficiency will therefore depend on the extent of the scavenging effects and the reactivity of secondary radical species with a contaminant (Li, 2017).

•- • • SO4 , HO and Cl have similar reaction rates with aromatic compounds. However, the reactivity

•- of CO3 is usually a few orders of magnitude lower than the others towards aromatic compounds

- - (Chen et al, 1975). In addition, the presence of Cl and HCO3 in groundwater can also form different chloro- and carbonated- surface complexes with Fe (III)- and Mn (IV)-oxides (Laat et al.,

2004; Valentine and Wang, 1998). These surface complexes can change the redox reactivities of

2- the surfaces of metal oxides, thus potentially impacting the rate of S2O8 activation and the redox cycle of Fe and Mn-oxides (Liu et al., 2008). Thus, it is important to understand the mechanism

2- and consequence of S2O8 interaction with aquifer chemical constituents.

2.5 AOP using UV/Photoactive Chlorine Species

Ultraviolet (UV)-driven processes with photoactive chlorine species (hypochlorous acid,

HOCl, and hypochlorite ion, OCl–) have been widely investigated for water and wastewater treatment purposes targeting organics such as pharmaceutical and personal care products (PPCPs), trichloroethylene, and benzoic acids (Wang et al. 2012; Jin et al. 2011; Feng et al., 2010; Boal et al. 2011; Buxton and Subhani, 1972a, 1972b, and 1972c). This process is considered a potential

58 advanced oxidation process (AOP), because of the production of hydroxyl radical equivalents as a result of the decomposition of excited chlorine species during exposure to light with wavelengths in the UVC (100–280 nm) region. However, OCl− absorbance has a tail out into the region, where sunlight provides a UVA (315–400 nm) source at the Earth’s surface.

The following equations (Eq. 2.88- Eq. 2.92) list reactions, which occur in the UV/chlorine process based on the homolytic cleavage of HOCl/OCl−; HOCl is produced as a result of reaction of chlorine with water.

Cl2+ H2O→ HOCl+ HCl Eq. 2.88

HOCl↔ H++ OCl- Eq. 2.89

HOCl + hν (UV photons) →HO• + Cl• Eq. 2.90

OCl-+ hν (UV photons) → •O-+ Cl• Eq. 2.91

• - • - O +H2O → HO + OH Eq. 2.92

The production of hydroxyl radicals at 254 nm has been demonstrated by Jin et al. (2010), where a yield factor (i.e. the fraction of hydroxyl radicals generated as a result of the photolysis) was used to assess the UV/Chlorine process as an AOP.

Although both HOCl and OCl– photolysis can produce •OH, the concentrations of •OH generated by HOCl and OCl– for equimolar concentrations are expected to be dissimilar because of different:

• molar absorption coefficients (Watts and Linden 2007): absorption spectrum peaked at 236

and 292 nm for HOCl and OCl– respectively (Fig. 2.2)

• quantum yields of •OH formation, and

• HO• scavenging efficiencies of HOCl and OCl– (Feng et al., 2007; Watts and Linden, 2007)

Fig. 2.2 Absorption spectrum of HOCl and OCl-

While HOCl and OCl– photolysis can produce •OH, they are also •OH scavengers with reaction rate constants of 8.46 × 104 (Watts and Linden, 2007) and 9.0 × 109 M–1 s–1 (Buxton and

Subhani, 1972), respectively.

• • OH + HOCl → H2O + ClO Eq. 2.93

•OH + OCl– → ClO• + OH– Eq. 2.94

It been also shown by Watts et al. (2007) that the concentration of •OH produced by UV/OCl– is lower than UV/HOCl. Therefore, UV/HOCl is probably a more promising AOP than UV/OCl–

(Watts and Linden, 2007). As a result, the effectiveness of the UV/chlorine AOP is sensitive to pH because the components of aqueous active chlorine are pH dependent (Fig. 2.3). According to

Deborde and von Gunten (2008), the pKa for HOCl at 25°C is 7.54.

Fig. 2.3 Dependence of the ratio of HOCl/OCl- on pH (adapted from Feng et al. 2007)

The photolysis of each chlorine species is also influenced by other factors such as irradiation wavelength and differences in water matrices. For example, the OCl– quantum yield increases with decreasing of the wavelength and a higher production ratio of •OH generated by

HOCl at 254 nm was observed when compared to sunlight irradiated at higher wavelengths

(Nowell and Hoigné, 1992b). On the other hand, the different compounds present in the water might play various roles in the promotion or inhibition of chlorine chain reactions, and/or in the photon absorption efficiencies of these compounds (Watts and Linden, 2007; Kobayashi and

Okuda, 1972).

2.5.1 Comparison of Chlorine/UV Systems with Other AOPs

Recent studies have indicated that UV/Chlorine system is potentially an alternative to the

- UV/H2O2 -AOP. This is because HOCl/OCl absorbs UV photons more efficiently than H2O2 when using typical low pressure or medium pressure UV lamps, and produces •OH relatively efficiently

61 under some conditions (Watts and Linden, 2007; Watts et al., 2007; Feng et al., 2007; Chan et al.,

2012; Nowell and Hoigné, 1992b).

However, AOPs only rely on HO•, while in the UV/Chlorine system various reactive

• • • − species of such as HO , Cl , and Cl2 exist in the reaction mixture. As a result, a series of chain reactions with the formation of chlorinated intermediates might be associated with these reactive species. Compared to other commonly used AOPs, the UV/chlorine AOP is still novel and not fully explored.

2.6 Photosensitization (Juris et al., 1988)

In certain cases where it is not possible to generate the excited state by direct absorption of a photon from an incident light source, it is possible to access it by energy transfer from a sensitizer.

Energy transfer is not the only mechanism based on which sensitizers work. Electron transfer is also a known mechanism in sensitization which may involve oxidation or reduction of the excited state of the molecules (M).

PS*+ M→ PS+ M* energy transfer

PS*+M → PS•++ M•- oxidative electron transfer

PS*+ M → PS•- + M•+ reductive electron transfer

The ability of a photosensitizer (PS) to undergo energy transfer is related to its zero-zero spectroscopic energy; the excited state energy of M should be lying below that of the sensitizer.

For electron transfer it is related to oxidation and reduction potentials of the PS+ /PS*and PS*/PS- couples. Photoinduced electron transfer mentioned in this study refers to the oxidative electron transfer process mentioned in chapter eight.

2.7 UV Light Sources

Effectiveness of photochemical studies is quite dependent on the appropriate selection of the light source. Technical and economic considerations have to be taken into account. First, the emission spectrum of the lamp must match the absorption spectrum of the reactants. Therefore, the choice of the light source is dictated by the absorption spectrum of the reactant. Secondly, from an economic point of view the power efficiency of the light source is an important factor

(Oppenlander, 2003).

In this study, persulfate, ozone, hypochlorous acid, Suwannee River Fulvic Acids (SRFA) and cumarine dye have been used along with UVA, UVC or 440 nm LED (as a visible light source) light sources, depending on the maximum absorbance of each of the listed reactants.

It should be noted that UVA, UVB and visible are some of the classifications for the electromagnetic spectrum. Figure 2.4 and Table 2.11 show the classification of electromagnetic radiation in the wavelength range below 1200 nm, and the corresponding photon energy range, which is the electromagnetic spectral range of interest in photochemistry, and the corresponding interactions with molecule M. The photochemically active region of the electromagnetic spectrum is divided into five sub bands: Vacuum-UV (VUV), UVC, UVB, UVA and visible.

Fig. 2.4 classification of electromagnetic radiation in the wavelength range below 1200 nm and the intraction of each class with molecule M (Oppenlander, 2003)

Table 2.11 Ultraviolet energy type (Modified after Tamuri et al., 2014)

Name Abbreviation Wavelength range (nm)

Ultraviolet A UVA 400-315

Ultraviolet B UVB 315-280

Ultraviolet C UVC 280-100

The following section provides an overview on low pressure mercury lamps as a conventional light source and on LEDs (Light Emitting Diodes) as a newer, more efficient light source.

2.7.1 Low Pressure Mercury Arc Lamp

The most commonly used mercury arc lamps are the low pressure type (two other types are medium and high pressure lamps) which provide almost monochromatic UV radiation at wavelength of 253.7 nm (this is usually rounded to 254 nm in the literature) mercury line. With addition of an appropriate phosphorescent coating, low pressure mercury lamps can be modified to produce UVA, UVB or visible light (Oppenlander, 2003). Low pressure mercury arc lamps are made of two metal electrodes encased in a sealed quartz body filled with mercury vapour of defined pressure (0.1 Pa) (Parson, 2004). The electrodes are in direct contact with the vapour phase mercury. Hg atoms in the gas phase are electronically excited by an electrical discharge between two electrodes. The electronically excited Hg atoms deactivate to their ground state by emission of radiation according to the energy level diagram, thus generating an intense radiating arc within the quartz envelope (Oppenlander, 2003). In addition to mercury, an inert gas is added to the lamp to initiate and maintain the discharge and to amplify the excitation of Hg atoms. The radiative deactivation of these states to the ground state result in two bands centered at 253.7 and 184.9 nm.

Typical emission spectrum of low pressure mercury lamp is presented in Fig. 2.5.

Fluorescent lamps used in this study, so-called black lamps (350 nm), are examples of light sources with a low pressure mercury lamp at heart and appropriate phosphor coating applied to the interior of the tube. The ultraviolet photons are now absorbed by the interior phosphor coating, which in turn emits photons with lower energy than the one that caused it; the difference in energy between the absorbed ultraviolet photon and the emitted visible light photon goes toward heating up the phosphor coating. In this case, an ordinary glass tube is used instead of quartz tube. The output spectrum of 350 nm lamps is presented in Fig. 2.6.

Fig. 2.5 Output spectrum of low pressure mercury lamp. Adapted from Parson (2004)

Fig. 2.6 Typical ouput of a black lamp (Tamuri, 2014)

2.7.2 LED Light Sources

Light emitting diodes (LEDs), a new generation of light sources, are potential replacements for conventional gas light sources. Efficient electrical to light energy conversion, long lifetime, directed output, availability of specific wavelengths to efficiently match chromophores, DC operation for remote locations and instant illumination are attractive characteristics of LEDs

(Schubert, 2006). An LED, which is usually small in area (less than 1 mm2) consists of a semiconducting material doped with impurities to create a p-n junction (Fig. 2.7).

Fig. 2.7 The inner works of an LED (adapted from Wikipedia)

When a forward bias is applied by connecting the n-type semiconductor to the anode and the p- type to the cathode, electrons and holes recombine and emit light. The wavelength of the emitted

67 light depends on the band gap energy of the material forming the p-n junction (different color

LEDs require different forward voltages to operate). It should be noted that LEDs are not perfectly monochromatic, but rather produce wavelengths over a narrow region of the spectrum.

2.7.3 The Ferrioxalate Actinometer

Photodecomposition of potassium ferrioxalate was developed into an actinometer by

Parker and Hatchard (Hatchard and Parker, 1956). It is one of the most accurate and widely used actinometers which covers a wavelength range between 250 nm to 577 nm. Irradiation of ferrioxalate solution results in the photogeneration of Fe2+ ions through photoinduced LMCT

(Ligand-to-Metal Charge Transfer) and a subsequent reductive reaction via CO2 radical anions,

∙- 2+ CO2 . The concentration of Fe ions has been commonly determined after complexation with

1,10- phenanthroline and the subsequent absorbance measurement at 510 nm.

III 3- II 2- •- Fe (C2O4)3 + hν→ Fe + 2C2O4 + C2O4 Eq. 2.95

•- •- C2O4 → CO2 + CO2 Eq. 2.96

III 3- •- II 2- Fe (C2O4)3 + CO2 → Fe + CO2+ 3C2O4 Eq. 2.97

The quantum yield for Fe2+ formation is nearly constant over the UV wavelength range and shows small variation with temperature, solution composition and light intensity. The recommended actinometer solution for wavelength up to 400 nm contains 0.006 M K3Fe(C2O4)3 in 0.05 M

H2SO4. For longer wavelengths, a 0.15 M solution is better suited. Quantum yields vary between

1.25 (254 -365 nm) to 1.1 at longer wavelength.

It should be noted that quantum yield is one of the most important parameters in evaluating the efficiency of a photochemical or photophysical process. Commonly used definitions of Φs are

68 presented in Table 2.12. These definitions describe quantum yields of photophysical events and of photochemical reactions with regards to the reactant decrease or to the formation of photoproduct; different quantum yield terms such as primary or product quantum yield, quantum yield of fluorescence, quantum yield of rearrangement etc. are used in photochemistry.

Table 2.12 Commonly used definitions of quantum yields (Oppenlander, 2003)

Mathematical Definition

expression

number of events per unit time by the number of photons absorbed 푑푛(푒푣푒푛푡) ⁄푑푡 Φ휆 = 퐼푎 per unit time (Ia)

number of reactant M consumed per unit time divided by the 푑푛(푀) ⁄푑푡 Φ휆 = 퐼푎 number of photons absorbed per unit time (Ia)

number of photoproduct molecules P formed per unit time divided 푑푛(푃) ⁄푑푡 Φ휆 = 퐼푎 by the number of photons absorbed per unit time (Ia)

Chapter three: DEGRADATION OF SULFOLANE USING ACTIVATED

PERSULFATE WITH UV AND UV- OZONE4

3.1 Introduction

Sulfolane (C4H8O2S- tetrahydrothiophene1,1edioxide) is an organo-sulfur compound widely used in the Sulfinol® process for processing sour gas. It is a component of a solvent solution used to remove carbon dioxide (CO2), hydrogen sulfide (H2S), carbonyl sulfide, mercaptans, and organic sulfides from natural gas. It is also used for extracting aromatics from hydrocarbon mixtures, in low boiling point alcohol separation, and fractional distillation of wood tar (Stewart and Minnear,

2010). Sulfolane's high polarity and high chemical and thermal stability, make it a favourable solvent for polar or polarizable organic compounds. The different industrial applications of sulfolane produce a significant amount of sulfolane containing waste. Spills and leaks from unlined waste storage ponds have contaminated soil and ground waters with sulfolane. The high aqueous solubility of sulfolane (1266 g L-1 at 20 oC- CCME, 2006), combined with low soil water partition coefficient (Kd) and low octanol water partition (Kow) have resulted in substantial offsite migration from contaminated sites (Greene and Fedorak, 1998; Luther et al., 1998).

Sulfolane is known to cause central nervous system (CNS) stimulation or depression (dependent on dose) in mammals. Acute toxicity testing of sulfolane on mammals yielded LD50 varying between 632 and 2504 mg kg-1 (CCME, 2006).

4 The research presented in this chapter was published in: Water Research, 125, 325 – 331. The supplementary materials are combined with the original published paper.

Bioremediation studies have shown that sulfolane is biodegradable in nutrient-enriched aerobic microcosms taken from a variety of sulfolane-contaminated environmental samples (Greene and

Fedorak, 1998; Witzaney and Fedorak, 1996; Greene et al., 2000; Chou and Swatloski, 1983).

Adsorption of sulfolane on biologically activated carbon has been tested by McLeod et al. (1992).

Agatonovic and Vaisman (2005) reported application of advanced oxidative processes (AOPs) in preliminary experiments for degradation of sulfolane using H2O2 and UVA. Recently Yu et al.

(2016) reported on application of several oxidative methods such as UVA and UVC irradiation along with photoactive oxidants, such as ozone, H2O2, and photocatalysis using TiO2 to degrade sulfolane in aqueous media. Mehrabani-Zeinabad et al., 2016 investigated the mineralization of sulfolane using UV/O3/H2O2 in a tubular reactor. The application of Fenton reagent to oxidize sulfolane has also been reported (Omar et al., 2010). One oxidant that has not been studied or reported on is persulfate. Persulfate is a strong water-soluble oxidant (E = 2.1 V), with the sulfate moieties substituted for hydrogens, which significantly increases its stability (Chen and Huang,

2015). Persulfate is stable at room temperature, but various initiators transform persulfate into radical (activated) with redox potential of 2.6 V (see Scheme 3.1) (Zhang et al., 2015; He et al.,

2014; Petri et al., 2011; Waldemer et al., 2007; Kolthoff and Miller, 1951).

❑ Oxidation

2- 2- S2O8 + 2e → 2SO4

❑ Sulfate radical formation

2- •2- S2O8 + activator → 2SO4

•2- • 2- + SO4 + H2O → HO + SO4 + H

Activator: heat, UV, transition metal, etc.

Scheme 3.1 Chemistry of persulfate

Both persulfate as well as its activated form have been applied for remediation of contaminated sites (Chen and Huang, 2015; Fang et al., 2013; Zou et al., 2014; Watts and Teel, 2006; Matzek and Carter, 2016). The persulfate anion being a weaker oxidant, without activation, will react with organics to a lesser degree. In an activated state, the radicals react with organic chemicals leading to either partial or complete mineralization (Perez-Sicairos et al., 2016; Chen et al., 2014; Wang and Liang, 2014; Zao et al., 2014). In situ remediation with activated persulfate oxidation may be preferred over peroxide based HO• oxidation processes, as the persulfate anion is more stable and may be transported further in the subsurface before being activated for contaminant degradation

(Yan et al., 2013; Petri et al., 2011).

There are different ways to activate persulfate: activation with iron, base (pH>12), minerals, heat, UV, O3 as well as ultrasonication, activated carbon and electrochemical methods.

Activation of persulfate results in highly reactive sulfate radicals (2014; Zhang et al., 2015; Abu

Amr et al., 2014; Zhao et al., 2014; Petri et al., 2011; Waldemer et al., 2007; He et al., Kolthoff and Miller, 1951;) which react with organic molecules via hydrogen abstraction, addition on double bond, and electron transfer (Neta et al., 1977). It has been reported (Liang and Su, 2009) that, sulfate radicals react with water at all pHs forming hydroxyl radicals, which are the primary reactive species under basic conditions. At pH < 7 sulfate radicals are the dominant reactive species; however, hydroxyl and sulfate radicals participate equally in reactions at neutral pHs.

This paper reports on the application of persulfate along with UV and UV/O3, which is part of a comprehensive study on application of advanced oxidation processes, for degradation of sulfolane in groundwater.

UV light can activate the persulfate by breaking the O-O bond of persulfate (33.5 kcal/mol). A quantum yield of one has been reported for acidic, basic, and neutral conditions

(Dogliott and Hayon, 1967). Both UVA and UVC light sources have been utilized as the irradiation source for persulfate activation, though UVC is known to be more efficient for degradation of contaminants with shorter reaction times (Lin and Wu, 2014; Zhang et al., 2014). The first part of this study reports on efficacy of this method for degradation of sulfolane.

Even though, persulfate activation with UV or O3 for decomposition of organic compounds

(though not for sulfolane) is reported in literature (He et al., 2014; Abu Amr et al., 2014), to the best of our knowledge the synergistic effect of O3 and UV irradiation on activation of persulfate has not been investigated for any of the organic compounds prior to this study. Section 3.3.2 provides a justification on combination of UV and O3 for activation of persulfate.

Once these methods are established, the effect of groundwater components such as carbonates and chlorides, are studied. These anions are known to affect the activity of persulfate (

Chen et al., 2015; Ahmad et al., 2013; Boni and Sbaffoni, 2012; Sra et al., 2014). To this end, well waters with known properties were collected and tested, along with spiked Milli Q waters. Finally, application of ammonium persulfate (APS) along with UVA light was evaluated.

3.2 Experimental Methods and Materials

Sulfolane with 99% purity and ammonium persulfate were purchased from Sigma Aldrich,

Canada. High purity oxygen was obtained from Praxair. Milli-Q water was used in all experiments, except for the sulfolane contaminated groundwater experiments.

All experiments were conducted in batch mode in a 150 mL quartz reactor (D = 59.0 mm), with 100 mL solution. The initial concentration of sulfolane for most of the experiments was 220 mg L-1 (1.83 mM) unless the dependence on initial concentration was being tested. All solutions were prepared by diluting an appropriate volume of a 92 mM sulfolane stock solution to 100 ml.

Ammonium persulfate was chosen for this study because it has the highest reported solubility in water (85 g/100 mL of water) and the lowest molecular weight among the most common persulfates (Lin et al., 2011; Chan et al., 2010; Lau et al., 2007). In addition, ammonium group can act as a nitrogen source for bioremediation, if bioremediation is to be coupled with

Advanced Oxidation Processes (AOP). A known amount of APS was dissolved in sulfolane solution prior to irradiation. Only in one case persulfate was added in three intervals, which was to study the quenching effect of persulfate. Persulfate concentration was measured by adding 1 ml of the irradiated samples to 10 ml water solution containing sodium bicarbonate and potassium iodide; Liang et al. (2008). After a period of 15 min, the produced iodide was titrated with sodium thiosulfate (0.02 M).

A batch photoreactor (LZC-ORG, Luzchem Research Inc.), which can be fitted with varying number of UV lamps, was used in this study. Germicidal lamps with a strong line at 254 nm were used to provide UVC irradiation. To study the possibility of using UVA for degradation of sulfolane, an experiment was also performed using black lamps (section 3.3.3). Black lamps with irradiation centered at 350 nm were used as an alternate light source for certain experiments.

Ferrioxalate actinometry (Calvert and Pitts, 1966) was used to measure light intensities.

For all experiments 10- UV lamps were used unless the effect of light intensity was studied.

For some experiments, in addition to UV, an O3 generator (A2Z 3-G LAB, A2Z O3 systems

Inc.) equipped with a glass diffuser was used. The sulfolane and O3 were homogeneously distributed across the radiation field using a magnetic stirrer. The experimental set up is shown in

Scheme 3.2.

Scheme 3.2 Schematic of an experimental setup with both ozonation and UV (Adapted from Mehrabani et al., 2016)

One mL samples were collected at pre-set times and was extracted with 2 mL of dichloromethane (DCM); the extraction efficiency was determined to be 80%. Samples in DCM were analyzed using an Agilent 6890 GC equipped with auto-sampler and Flame Ionization

Detector (FID). Data acquisition and analyses were performed using Chemstation software. The

DCM samples were injected to GC and chromatographic separation was made on a fused silica capillary column (ZB 5MSI, Phenomenex). High purity helium was used as the carrier gas with a head pressure of 250 kPa. The temperature of the injection port was set to 165o C and the injection was set on splitless mode with 2.00 mL injection volume. The initial temperature was set to 90o C, which was ramped up to 175o C at a rate of 10o C/min where it was held constant for 3 min. The

FID detector temperature was set to 250oC. GC peak areas were determined from duplicate measurements on each of the samples in a ± 0.03 mg L-1 range. External calibration was used to quantify the sulfolane in the sample by preparing various concentrations of sulfolane as standards. The detection limit for sulfolane analysis was 1 mg L-1.

3.3 Results and Discussion

3. 3.1 Degradation of Sulfolane using Ammonium Persulfate Activated with UVC

Blank experiments (which was done to correct the results for the individual effects of APS or UVC irradiation on sulfolane degradaiton) revealed that persulfate cannot degrade sulfolane and at most 3% of sulfolane was degraded after 40 min of UVC irradiation, which was because of the poor absorbance of sulfolane at 254 nm.

3.3.1.1 Effects of Persulfate Dose and UVC Light Intensity on Sulfolane Degradation

Effects of persulfate concentration on sulfolane degradation rate are shown in Fig. 3.1 and

Table 3.1. The pseudo-first-order reaction rate constants increased linearly as the persulfate concentration increased from 4.38 to 17.40 mM (1, 2, 3, 4 g L-1). Persulfate concentrations in 9.1-

17.4 mM range resulted in nearly complete sulfolane degradations within 40 min of reaction.

However, at 1 g L-1 the reagent was exhausted, and the reaction stopped after 25 min, which resulted in only 60% degradation of sulfolane. The increase in rate of reaction by increasing persulfate concentration has been also reported in literature for other organic contaminants (Zhang et al., 2014; Zhao et al., 2015; Ji et al., 2015) and can be explained by an increase in the steady- state concentration of reactive radical species (sulfate and hydroxyl radicals).

1.2 1 1g/L 2g/L 0.8 3g/L

0.6 4g/L C/C0 0.4 0.2 0 0 10 20 30 40 50 Time (min)

Fig. 3.1 Degradation of sulfolane (220 mg L-1-1.83 mM) using 1 g L-1 – 4 g L-1 (4.38-17.40 mM) ammonium persulfate under UV irradiation (10 lamps)

Table 3.1 Pseudo-first-order degradation rate constants for (200 mg L-1-1.83 mM) sulfolane and varying concentrations of persulfate under UVC irradiation (10 lamps) Different loadings of APS Pseudo first order rate constant (s-1) R2

4.38 mM- 1 g L-1 0.45×10-3 0.923

9.11 mM - 2 g L-1 1.21×10-3 0.993

13.10 mM - 3 g L-1 1.52×10-3 0.920

17.40 mM - 4 g L-1 1.90×10-3 0.972

Even though 4 g L-1 (17.40 mM) of persulfate had the highest rate, 3 g L-1 (13.10 mM) was chosen to perform the rest of the experiments as this loading was the best for persulfate/O3/UVC combination (discussed later). Moreover, we did not wish to introduce more sulfate into the system than necessary as there is a maximum acceptable concentration of sulfate in treated waters. We have to be cognizant that sulfate is also produced as a result of mineralization of sulfolane. It has been reported in literature that high concentrations of persulfate (around 100 mM) reduces the

2- •- efficiency of the reaction as excess sulfate act as scavengers of S2O8 and SO4 (Wang and Liang,

2014). In this study the quenching effect was tested for the optimized persulfate loading, which was 3 g L-1 (13.10 mM). A comparison was made by adding the 3 g L-1 (13.10 mM) of persulfate to a 100 ml solution containing the same concentration of sulfolane (1.83 mM), all at the beginning or in three steps during the same irradiation time (3×0.1 g every 10 min). Since the degradation

•- efficiency was the same in both cases, it was concluded that the quenching of produced SO4 was minimal if 3 g L-1 (13.10 mM) of persulfate was used. It should be noted that after 40- and 180- min irradiation, for sulfolane and TOC removal, the persulfate concentration reduced to 7.0 and

1.0 mM respectively (Fig. 3.2).

16 APS + O3+ UV 14 12 APS+ UV

] mM ] 10

- 2

8 8

O 2

[S 6 4 2 0 -2 0 20 40 60 80 100 120 140 Time (min)

Fig. 3.2 Changes in concentration of persulfate (13.10 mM- 3 g L-1) over times in combination with UV or UV/O3

A 13.10 mM (3 g L-1) initial persulfate concentration was selected for the experiments studying the effect of UVC light intensity. As can be seen in Table 3.2, when the UVC light intensity increased from 2.23×1017 to 3.13×1017 photons/sec, the degradation rate of sulfolane

78 increased from 0.71×103 to 1.32×103 s-1 but no significant difference was observed if the light intensity was further increased to 5.20 ×1021 photons/sec. Overall an increase in reaction rate was observed with an increase in light intensity, which is consistent with literature (Salari et al., 2009).

Increasing light intensity provides more radicals available for sulfolane and photoproduct degradation. But at higher light intensities than 3.13×1017 (provided quenching did not have an effect) probably enough oxidizing radicals are produced for photodegradation of this specific concentration of sulfolane, therefore there is no significant competition for radicals among the photodegradable species present in the solution. It has also been reported in literature that dosage of persulfate plays a more important role than that of light intensity on degradation and mineralization of dyes such as acid blue azo dye (Shu et al., 2015).

Table 3.2 Pseudo-first-order degradation rate constants of 1.83 mM sulfolane and persulfate (13.10 mM- 3 g L-1) under three different light intensities

Number of lamps Light intensities (photos.sec-1) Pseudo first order rate constant (s-1), R2

4 lamps 2.23×1017±0.05 0.71×10-3, 0.9798

6 lamps 3.13×1017±0.10 1.32×10-3, 0.9420

10 lamps 5.21×1021±0.30 1.52×10-3, 0.9899

3.3.1.2 Effect of Initial pH on Sulfolane Degradation with Persulfate and UV

An investigation on the effect of varying pH on degradation of sulfolane by UVC/APS was conducted at pH 3.9, 7.0, and 9.1. Table 3.3 and Fig. 3.3 show the degradation rate of sulfolane and TOC removal by the UVC/APS process respectively. The efficiency of sulfolane and TOC removal decreases with increasing pH from 3.9 to 9.6.

According to Liang and Su (2009), hydroxyl radicals are the primary reactive species under basic conditions, while both sulfate and hydroxyl radicals serve as free radicals in solution at neutral and acidic pH (Liang and Su, 2009). The reduced efficiency at pH 9.6 could be related to the competition of the photoproducts with sulfolane for hydroxyl radicals or because of the quenching of hydroxyl radicals at pH 9.6. Since the trend for TOC removal is consistent with degradation of sulfolane, the reduced efficiency of the reaction could be because of the quenching of the hydroxyl radicals. This matter needs more detailed investigation.

Table 3.3 Effect of initial pH on degradation of sulfolane (220 mg L-1) in presence of 3 g L-1 of APS and UVC 10-254 nm lamps pH Pseudo first order rate constant (s-1) R2

3.9 1.0×10-3 0.9892

7.1 1.00×10-3 0.9968

9.6 0.22×10-3 0.9564

1.2 1 pH=3.9

0.8 pH=7.1 0 0.6 pH=9.6 0.4

TOC/TOC 0.2 0 0 100 200 300 Time (min)

Fig. 3.3 Effect of initial pH on TOC removal (sulfolane -220 mg L-1) in presence of (13.10 mM- 3 g L-1 of APS and UVC

•- • As aforementioned, SO4 and OH radicals can oxidize a variety of organic compounds, but the oxidation mechanisms of hydroxyl and sulfate radicals can significantly differ from each other. Sulfate radicals normally undergo electron transfer reactions while HO• may also react via hydrogen-atom abstraction along with an electron-transfer process which is however less prominent in their case (Liang and Su, 2009). These results show that presence of both hydroxyl and sulfate radicals is helpful in increasing the rate of the sulfolane and TOC removal as it increases the range of the compounds to be oxidized.

3.3.1.3 Effect of Carbonate on Degradation of Sulfolane using APS/UVC

According to Tsitonaki et al. (2010) the sulfate radicals can react with several inorganic

- 6 -1 -1 3- ions, such as carbonate, or bicarbonate ions very rapidly (HCO3 : k = 1.6-9.1 × 10 M s , CO2 : k = 6.1×106 – 4.1 × 108 M-1s-1) (Shu et al., 2015). These anions can be abundant in soil and groundwater systems, and may compete with the target contaminants for the sulfate radical

(scavenging reactions), which could decrease the efficiency of the oxidative treatment.

Furthermore, these anions are often products of contaminant oxidation; thus, they may accumulate over the course of reaction further reducing contaminant treatment efficiencies. Carbonates are also known to be effective hydroxyl radical scavengers (Weeks and Rabani, 1966). Since this method is going to be applied to groundwater samples, the effect of carbonate (100-400 mg L-1) on degradation of sulfolane (220 mg L-1-1.83 mM)) was studied for 1 and 3 g L-1 (4.38 and 13.10 mM)of APS (Figs. 3.4 and 3.5 respectively).

The results show that the hydroxyl (primary reactive species in this case) and sulfate radicals are efficiently quenched by carbonate at concentration above 100 mg L-1 if 1 g L-1 (4.38 mM) of APS was used; sulfolane degradation decreased to 20% compared to 60% in the absence

81 of carbonate after 30 min of reaction. The quenching effect was less if the concentration of APS was increased to 3 g L-1 (13.10 mM). It should be noted that the pH for all three cases was above

10.0, which is in the pH range for persulfate reactions with lower reaction rates (Fig. 3.3).

Therefore, the reduced efficiency cannot completely be attributed to quenching effect of carbonate.

In the following section, effect of carbonate/bicarbonate on reaction rates has been studied by lowering the pH into the acidic region.

1.1 without carbonate 1 100 ppm carbonate 0.9

0.8 200 ppm carbonate 0 0.7 C/C 400 ppm carbonate 0.6 0.5 0.4 0.3 0 10 20 30 40 Time (min)

Fig. 3.4 Effect of carbonate concentration (100-400 mg L-1), pH:10.6 -10.9, on degradation of sulfolane (220 mg L-1 – 1.83 mM) using 1 g L-1 (4.38 mM) of APS under UV irradiation

1.2 without carbonate 1 0.3 g APS & 100 ppm carbonate 0.3 g APS & 200 ppm carbonate 0.8 0.3 g APS & 400 ppm carbonate

0 0.6

C/C 0.4

0.2

0 0 10 20 30 40 50 60 70 -0.2 Time (min)

Fig. 3.5 Effect of carbonate (100-400 mg L-1), pH:10.6 -10.9 on degradation of sulfolane (220 mg L-1- 1.83 mM) and 3 g L-1 (13.10 mM) of APS under UV irradiation

3.3.1.4 Effect of Carbonate-Bicarbonate on Degradation of Sulfolane using APS/UVC

In all of the above cases the pH of the solution was above 10, which is not common in natural water systems, where the pH is lower and both carbonate and bicarbonate are present. In the sample containing 400 mg L-1 of sodium carbonate the initial pH was lowered to a pH range between 5.7 & 7.5, where both carbonate and bicarbonate are present in the system. The results

•- (Fig. 3.6) show that at initial pH of 6.9 and 7.5, some SO4 are quenched by carbonate/bicarbonate slowing the reaction but no effect was noted at pH 5.7 as about 80% carbonate is present as H2CO3, or had escaped out of the solution as CO2. At pH=6.9 and 7.5, 80% and 95% carbonate are present as bicarbonate (along with 20% and 5% of carbonate). Both sulfate and hydroxyl radicals are present in the solution at theses pHs, which could be quenched with carbonate/bicarbonate. During these processes carbonate radical anions are produced, which also react with sulfolane, but to a lower level compared to sulfate/hydroxyl radicals. As a result, lower reaction efficiency is

83 observed at higher pHs as a result of higher carbonate/bicarbonate concentration present in solution.

1.2 pH= 5.7 1 pH=6.9

0.8 0

0.6 pH=7.5 C/C 0.4 pH=7.0 W/O carbonate- bicarbonate 0.2 0 0 20 40 60 80 100 Time (min)

Fig. 3.6 Effect of initial pH on degradation of sulfolane (220 mg L-1) in presence of 3 g L-1 of APS and 400 mg L-1 carbonate/bicarbonate under UV irradiation

3.3.1.5 Effect of Chloride on Degradation of Sulfolane using APS/UVC

The effect of Cl- on the oxidative degradation of sulfolane was investigated in UVC/APS system at neutral pH to simulate the presence of chloride in groundwater. The results show no quenching effect for chloride up to 100 mg L-1-3 mM (Fig. 3.7).

1.2 1 without NaCl

0.8 100 ppm NaCl 0

0.6 C/C 0.4 0.2 0 0 10 20 30 40 50 60 70 Time (min)

Fig. 3.7 Effect of chloride concentration (100 mg L-1) on degradation of sulfolane (220 mg L-1 - 1.83 mM) using of ammonium persulfate (3 g L-1 -13.10 mM) under UV irradiation

These results are consistent with Wang et al. (2011) results, where they studied the quenching effect of chloride ion on decolorization of Acid Orange 7 by UV/persulfate in concentration range between 1 and 10 mM and they found no significant inhibition effect on the decolorization rate.

However, their results also indicated that the inhibition tendency became clearer at higher

-1 - ●- concentrations (0.1 mol L ). They justified the results as due to reaction between Cl and 2SO4

●- ●- and the generation of Cl2 or HClO (Tsitonaki et al., 2010); which are not as efficient as hydroxyl and sulfate radicals for sulfolane degradation. Both kinetics and thermodynamics conditions could be responsible for the lower sulfolane degradation effeicy in presence of higher concentration of chloride in the reaction mixture.

3.3.2 Degradation of Sulfolane Using Ammonium Persulfate/O3/UVC

2- In the presence of UV irradiation, S2O8 can promptly transform into two sulfate anion radicals:

2- ●- S2O8 + hν → 2SO4 Eq. 3.1

●- Subsequently SO4 can oxidize the contaminant, as well as react with water to produce hydroxyl

●- radicals. SO4 reacts with water at all pHs but the ratio of sulfate and hydroxyl radicals depend on the pH of the solution. In case of O3/UV systems, ozonation can proceed via two routes: direct molecular ozone reactions (which are relatively slow and selective) and/or indirect pathway leading to ozone decomposition and the generation of hydroxyl radicals (OH•). According to

Sarathy et al. (2006) and Peyton et al. (1982); O3/UV is initiated by the photolysis of O3 by UV to

• form H2O2 and O2; OH production is carried out by the reaction between O3 and H2O2 (Equation

3.2-3.4):

O3 + H2O + hv → H2O2 + O2 Eq. 3.2

• 2O3 + H2O2 → 2 OH + 3O2 Eq. 3.3

• H2O2 + hv → 2 OH Eq. 3.4

-1 -1 -1 -1 If both persulfate (ԑ254 nm= 24.1 M cm ) and O3 (ԑ254 nm =3300 M cm ) are present in the solution under UVC (254 nm) irradiation, light absorption by O3 should be dominant but O3 has a limited solubility in water. As a result, the O3/UV process is less energetically efficient than persulfate/UV for generating large quantities of hydroxyl radicals due to the low solubility of O3 in water compared to persulfate. Considering these extinction coefficients and concentrations of 5 mg L-1

-1 for O3 and 3 g L (13.10 mM) for persulfate, the absorption of the two solution for the same path length is very similar (0.3b and 0.33b respectively). It has been reported in literature that increasing persulfate might cause self quenching. Application of O3 along with low concentrations of persulfate could be a more efficient way as in case of O3, hydroxyl radicals are produced gradually.

Since water saturates quickly with O3, the flow rate of O3 could be minimal to reduce wasting of

86 generated O3 in the process. In the following sections application of persulfate/O3 under UVC irradiation will be investigated for degradation of sulfolane in water.

3.3.2.1 Effect of Persulfate Dosage on Sulfolane Degradation and Control Experiments

Prior to the investigation of effects of persulfate dosage on degradation of sulfolane, a series of blank experiments were performed to correct the results for the individual effect of different parameters involved (considering all possible reactions by breaking down all the components of reaction reagent (APS/O3/UV), which could be responsible for sulfolane degradation). The results showed that there was no reaction if only ozone or persulfate or the combination of the two was utilized without UVC irradiation. Direct photolysis of sulfolane was not efficient under UVC irradiation, as was expected from the UV-Vis absorption spectrum of sulfolane, which reveals no significant absorbance at 254 nm (~2-3% degradation of sulfolane was observed after 30 min of irradiation). Another control experiment was to study the degradation efficiency of sulfolane using O3/UVC (these results are included in Fig. 3.11, which will be discussed later). Later the preliminary tests were conducted to determine the effect of concentration of persulfate (1, 2 and 3 g L-1- 4.38, 9.11 and 13.10 mM) on degradation of sulfolane

-1 in presence of 5 mg L O3 under UVC irradiation. The results obtained are shown in Fig. 3.8. The pseudo first order rate constants are listed in Table 3.4 in comparison with similar cases in absence of O3 as a control experiment, also to show the added advantage of O3.

1.2

1 4.38 mM

9.11 mM 0.8 13.10 mM

0.6 C/C0

0.4

0.2

0 0 5 10 15 20 25 Time (min)

Fig. 3.8 Degradation of sulfolane (220 mg L-1- 1.83 mM) using APS (4.38-13.10 mM), under bubbling O3 and 254 nm irradiation

Table 3.4 Pseudo-first-order degradation rate constants of 1.83 mM sulfolane and persulfate with -1 different concentrations in presence of O3 (5 mg L ) under UVC irradiation Pseudo First order rate constant Pseudo First order rate constant (s-1)a

-1 2 2 [APS] (mM) (s ), R for APS-UVC and R for APS-O3-UVC

4.38- 1g L-1 0.45 × 10-3, 0.923 1.83 × 10-3, 0.989

9.11- 2g L-1 1.21 × 10-3, 0.993 2.50 × 10-3, 0.992

13.1- 3g L-1 1.52 × 10-3, 0.920 2.00 × 10-3, 0.989 a Only data points up to 10 min were used in case of APS/UVC/O3 to calculate the rate constants.

Comparing the rate constants with those obtained in the absence of ozone shows that most improvement was observed for the APS concentration of 4.38 mM (Fig. 3.9).

4.38 mM APS + UV 1.2

0.8

0.6 C/C0

0.4

0.2

0 0 10 20 30 40 Time (min)

Fig. 3.9 Degradation of sulfolane (220 mg L-1- 1.83 mM) using 1 g L-1 (4.38 mM) APS under UV irradiation in the presence or absence of bubbling ozone (0.5 L min-1)

The sulfolane removal rates, with the same rate of ozone introduction, are very similar for different loadings of APS (Fig. 3.8). Since these results were not conclusive, TOC removal over time was monitored instead to find the optimum concentration of APS. The results in Fig. 3.10, show that complete TOC removal can be achieved in about 40 min in presence of 3 g L-1 of APS. In the absence of ozone for the same APS, only 10-12% decrease in TOC was observed after 40 min of reaction. It should be noted that after 20- and 40-min irradiation, for sulfolane and TOC removal the persulfate concentration reduced to 11 and 7 mM respectively (Fig. 3.10).

1.2 2 g/L 1 3 g/L 4 g/L

0 0.8

0.6

TOC/TOC 0.4

0.2

0 0 10 20 30 40 50 60 70 -0.2 Time (min)

-1 Fig. 3.10 TOC removal over time for three APS loadings for 220 mg L of sulfolane and O3 flow rate of 0.5 L min-1 under 254 nm irradiation

To study the effect of different parameters on the reaction rates, a loading of 1 g L-1 (4.38 mM)

APS was chosen for most of the experiments described hereafter. The overall results are shown in

Fig. 3.11.

Persulfate/UVC/O3 was the most efficient method for degradation of sulfolane. It should be noted that an induction period was observed if ozone and UVC were used in the absence of APS (for this reason the reaction rated could not get compared). A detailed study was performed to find the parameters affecting the induction period of O3/UVC system. It was found that the induction period was in direct relation with the light intensities; higher light intensities corresponded to shorter induction periods. A small induction period remained even with the highest light intensity that was studied. It was also found that the rate of O3 introduction had no effect on the length of the induction period (Fig. 3.11), due to quick saturation of water with O3.

1.2

APS 4.38 mM + UV 1 APS (4.38 mM) + O3 0.5L/min + UV

0.8 O3 0.5L/min + UV 0

C/C O3 0.5L/min + UV 0.6 O3 0.5L/min (slower stirrer) + UV

0.4 APS 4.38 mM

0.2

0 0 10 20 30 40 50 Time (min)

Fig. 3.11 Degradation of sulfolane (220 mg L-1) using APS (1 g L-1) and bubbling ozone (0.5 L min-1) Finally, it was noticed that for a specific light intensity and flow rate of ozone the induction period is dependent on initial concentration of sulfolane (Fig. 3.12). No induction period was observed if the initial concentration of sulfolane was reduced to 50 mg L-1 (0.42 mM). Since the solubility of

O3 in water is limited, at higher initial concentration of sulfolane, the produced radicals under irradiation could be the limiting reagent. That causes a delay in observing measurable changes in sulfolane concentration.

1.2 0.42 mM 1 0.83 mM 0.8

1.83 mM 0

0.6 C/C 0.4

0.2

0 0 5 10 15 20 25 30 35 Time (min)

Fig. 3.12 Effect of changing initial concentration of sulfolane on induction period of O3 (0.5 L min-1) -UVC system

3.3.2.2 Effect of Initial pH on Sulfolane Degradation with Persulfate/UVC/O3

Further experiments to investigate the influence of pH (3.9, 7.1, 9.5) on degradation rate of sulfolane were conducted at APS loading of 3 g L-1. As shown in Fig. 3.13, the reaction rate decreases with increasing the initial pH of the sulfolane solution (with k = 2.30 × 10-3 s-1/R2 =

0.989, 1.50 × 10-3 s-1/R2 = 0.996 and 5.00 × 10-4 s-1/R2 = 0.956 for pH = 3.9, 7.1 and 9.5 respectively). This is consistent with the TOC removal results (Fig. 3.14). Again, the lower reaction rate at higher pHs could be because of quenching of hydroxyl radicals considering the mechanism proposed by Liang and Su (2009). More hydroxyl radicals are expected to be produced in case of UVC/APS/O3.

1.2 pH=3.9 1 pH=7.1 pH=9.5 0.8

0 0.6

C/C 0.4

0.2

0 0 5 10 15 20 25 Time (min)

Fig. 3.13 Effect of initial solution pH on degradation of sulfolane (220 mg L-1- 1.83 mM) in -1 -1 presence of 3 g L (13.10 mM) of APS and O3 flow rate of 0.5 L min under 254 nm irradiation

1.2 pH=3.9, APS + O3 + UV 1

0.8 pH= 7.1, APS + O3 + UV 0

0.6 pH=9.6

0.4 TOC/TOC

0.2

0 0 20 40 60 80 Time (min)

Fig. 3.14 Effect of initial solution pH on TOC removal, sulfolane (220 mg L-1 - 1.83 mM) in -1 -1 presence of 3 g L (13.10 mM) of APS and O3 flow rate of 0.5 L min under 254 nm irradiation

3.3.2.3 Effect of Carbonate on Degradation of Sulfolane using APS/O3/UVC

Quenching effect of carbonate (100- 400 mg L-1, strong alkaline media), on sulfolane (220

-1 -1 mg L ) degradation for the UVC/APS (4.38 mM)/O3 (0.5 L min ) was also investigated. Similar to the APS/UVC system (section 3.1.3), the produced radicals in UVC/APS/O3 system are efficiently quenched by carbonate. Sulfolane degradation decreased to 35% compared to 90% in the absence of carbonate in presence of 100 mg L-1 carbonate (Fig. 3.15) after 20 min of reaction.

The reaction ceased if 400 mg L-1 of carbonate was used. As it was mentioned earlier, hydroxyl radicals are reported to be the primary reactive species in alkaline solutions, when persulfate degrades under UV irradiation. Hydroxyl radicals are also produced because of UVC and O3 interaction. The lower efficiency in this case also could be related to the inhibitive effect of pH as well.

1.2

1 APS+ O3

0.8 APS+ O3+ 100 ppm carbonate

0.6 APS+ O3+ 200 ppm carbonate C/C0

0.4 APS+ O3+ 400 ppm carbonate 0.2 APS+ O3+ 800 ppm carbonate 0 0 5 10 15 20 25 Time (min)

-1 Fig. 3.15 Effect of carbonate on sulfolane (220 mg L - 1.83 mM) degradation using O3 (0.5 L min-1) and persulfate (4.38 mM) under 254 nm irradiation. The pH for all solution were above 10.5

3.3.2.4 Effect of Chloride on Degradation of Sulfolane using APS/O3/UV

Like APS/UVC case no quenching effect for chloride was observed at neutral pH for the two concentrations of chloride tested (Fig. 3.16). Overall the results in case of APS/UVC/O3 were consistent with the results obtained in the absence of O3 but adding O3 to the system increased the degradation rate of sulfolane in all cases. An economic investigation is required to determine if reduced reaction time in presence of O3 is beneficial as O3 production requires a substantial amount of energy.

1.2

1 without NaCl

0.8 45 ppm NaCl 0

0.6 116 ppm NaCl C/C 0.4

0.2

0 0 10 20 30 40 50 Time (min)

-1 -1 Fig. 3.16 Effect of chloride on sulfolane (220 mg L ) degradation using O3 (0.5 L min ) and persulfate (1 g L-1) under 254 nm irradiation

3.3.3 Degradation of Sulfolane using Persulfate and UVA Lamps

In previous studies, the persulfate degradation (and production of sulfate radicals) has been performed by photolysis at wavelengths from 248 to 351 nm (Herrmann, 2007). As an example, application of persulfate along with UVA light has been reported in literature by Perez-Sicairos et al. (2016) for degradation of nitrobenzene. Theses results could be useful if artificial sunlight or

LED light sources are planned to be used. The application of longer wavelength light for degradation of sulfolane was also verified in this study using black lamps (maximum wavelength

-1 at 350 nm). With a loading of 3 g L (13.10 mM) of ammonium persulfate (A350nm = 0.003;

-1 -1 ε351nm=0.25 M cm ) 30% of sulfolane was degraded after 50 min of reaction (Fig. 3.17). Sulfolane was not detectable after 40 min of irradiation under 254 nm irradiation. This is reasonable

-1 -1 considering the quantum yield of sulfate formation at 254 nm (ε254 = 24.1 M cm ), which is about

3 times more than that at 350 nm (Herrmann, 2007). Sulfolane removal decreases significantly beyond this period as the persulfate concentration is not high enough (11 mM) to absorb light considering the very low absorptivity at 350 nm. More detailed experiments are required to optimize the reaction conditions with regards to light intensity, best persulfate loading and the pH of the solution.

1.2 1 0.8

0 0.6

C/C 0.4 0.2 0 0 20 40 60 80 Time (min)

Fig. 3.17 Degradation of sulfolane (220 mg L-1- 1.83 mM) using persulfate (3 g L-1- 13.10 mM) under 350 nm irradiation (I= 4.70×1017 photons/sec)

3.3.4 Sulfolane Treatment in Well Water Samples using APS/UVC and APS/O3/UVC

In order to evaluate the application of UVC/APS and UVC/APS/O3, two tests were performed on a ground water sample under the optimized condition for each case (Table 3.5 presents characteristics of the well water sample used in this study).

Table 3.5 Characteristics of groundwater samples

Parameters Well 1

Total dissolved solids (mg L-1) 726

pH 7.58

Dissolved organic carbon (mg L-1) 87.5

Anions

-1 Alkalinity (Total as CaCO3) (mg L ) 641

-1 Bicarbonate (HCO3) (mg L ) 782

-1 Sulfate (SO4) (mg L ) 38.6

Chloride (Cl) (mg L-1) 32.5

-1 Nitrate (NO3) (mg L ) 0.3

Since the sulfolane concentration was low in the provided well water samples, it was spiked with sulfolane to start with 1.83 mM initial concentration of sulfolane. Sulfolane was degraded successfully in both cases but with a lower rate compared to the experiments performed in MilliQ water. For 90% sulfolane degradation 60- and 22-min irradiation was required in presence of 3

-1 g L (13.10 mM) of APS for APS/UVC and APS/UVC/O3 system respectively. This corresponds to additional 35- and 10-min irradiation compared to experiments performed using MilliQwater.

These results show presence of O3 decreases the irradiation time. It was also observed that sufolane

97 degradation follows first order kinetics with a rate constant of 1.4 ×10-3 s-1 (R2 = 0.985), but in the absence of ozone the initial degradation of sulfolane was slower than the later stage (Fig. 3.18).

1.2

1 APS+ O3+ UV 0.8 APS+UV

0.6 C/C0 0.4

0.2

0 0 20 40 60 80 Time (min)

Fig. 3.18 Degradation of sulfolane in a ground water sample (water characteristics are shown in Table 3.5) The slower reaction rate for well water samples can be explained by the fact that the pH of the water sample was 7.6, bicarbonate concentration was around 800 mg L-1 and due to the presence of the co-contaminants such as humic substances and amines in groundwater samples, all of which compete with sulfolane for reactive radicals.

3.4 Conclusions

This study has demonstrated that persulfate along with UVC and UVC/O3 can efficiently degrade

-1 sulfolane in water. Presence of 5 mg L O3 in solution, provided by bubbling O3 into the solution, not only increases the rate of sulfolane removal (up to three times) but also decreases at least 25% of the required dosage of persulfate. In general, at higher pHs than 6.9 the reactions are slower and the quenching effect of carbonate seem to be significant. Chloride at concentrations lower then

100 mg L-1 has no effect on reaction rate.

Overall, based on the results reported in this study, by considering the sulfolane and sulfate concentration in well water samples, an acceptable concentration of persulfate can be chosen, for which the best light intensity and other parameters can be defined.

Chapter four: MINERALIZATION OF SULFOLANE IN AQUEOUS SOLUTION BY

5 OZONE/CaO2 AND CaO/OZONE WITH POTENTIAL FOR FIELD APPLICATION

4.1 Introduction

Sulfolane (C4H8O2S- tetrahydrothiophene1,1-dioxide), is a chemically inert organosulfur solvent, with high polarity and high chemical and thermal stability. It has a wide spectrum of applications which include extracting aromatics from petroleum, or removal of carbon dioxide, hydrogen sulfide, carbonyl sulfide, mercaptans, and organic sulfides from natural gas. During these processes not only large amounts of waste containing sulfolane are produced, but also at many sour-gas processing plants spills and leakage from unlined surface storage ponds have contaminated soil and ground waters with sulfolane (Stewart and Minnear, 2010). High aqueous miscibility of sulfolane, combined with the low organic-water partition (Koc) have resulted in substantial offsite migration from contaminated sites (Luther et al., 1998). Sulfolane is a persistent environmental contaminant known to cause central nervous system (CNS) stimulation or depression (dependent on dose) in mammals (CCME, 2006). Since natural processes are not effective in sulfolane removal, several treatment methods have been suggested. Adsorption on biologically activated carbon (Ying et al., 1994; McLeod et al., 1992) and biological treatment under aerobic conditions for both water and soil are among the initial remediation methods suggested (Greene et al., 2000; Chou and Swatloski, 1983). In addition to theses biological processes, Doucette et al. (2005) studied uptake of sulfolane and diisopropanolamine (DIPA) by

5 The research presented in this chapter was published in: Chemosphere 197, 535-540. The supplementary materials are combined with the original published paper.

100 cattails (Typha latifolia) in wetland and found that wetland plants could play an important role in the natural attenuation of sulfolane.

Application of AOPs for degradation of sulfolane was first suggested by Agatonovic and

Vaisman (2005) in preliminary experiments, where they investigated degradation of sulfolane in contaminated groundwaters using H2O2 and UVA. Recently Mehrabani et al. (2016), Yu et al.

(2016) and Izadifard et al. (2017), reported on application of several oxidative methods such as application of UVA and UVC irradiation along with photoactive oxidants, including TiO2, O3,

2- H2O2, persulfate or their combination such as H2O2/O3 or S2O8 /O3 for degradation of sulfolane.

While these experiments were performed in a batch reactor set up, Mehrabani et al. (2016), investigated the mineralization of sulfolane using UV/O3/H2O2 in a tubular reactor. Finally, application of Fenton reagent is also reported in literature for sulfolane oxidation (Omar et al.,

2010).

The focus of this chapter is furthering the evaluation of advanced oxidative processes by studying the application of calcium peroxide- CaO2 and calcium oxide-CaO along with O3 for degradation of sulfolane in water. Ozonation is a very commonly studied process used in treatment of wastewater produced in pulp and paper production, shale oil processing, production and usage of pesticides, dye manufacture, pharmaceutical production and many others (Hoigne, 1988;

Masten and Davies, 1994; Rice, 1996; Silva et al., 2006). O3 is a strong oxidant (redox potential

2.07 ev), which breaks down recalcitrant and toxic organic compounds into smaller molecules.

Ozone by itself is an environmentally friendly oxidant, as it decomposes to O2 without producing self-derived by products in the oxidation reactions (Guo et al., 2012).

101

Despite several advantages of using ozone, it has a few disadvantages, which limit its application in water treatment technology (Nawrocki and Hordern, 2010; Pirkanniemi and Sillanp,

2002; Hordern et al., 2003). These include:

• relatively low solubility and stability in water

• high cost of production

• insufficiency of the process to mineralize recalcitrant organics

• Reactions of ozone with organic matter usually lead to the formation of aldehydes and

carboxylic acids, both of which do not react with ozone

• may have low reaction kinetics and limited mass transfer

• pH limitation

• high selectivity of ozone

In treatment of industrial wastewaters, many researchers have improved the efficiency of

- ozonation. Advancements such as ozonation in alkaline solutions (O3/OH ), photolysis of ozone

-1 -1 (O3/UVC, εO3 at 254 nm = 3300 mol cm ) (Munter, 2001), peroxone (O3/H2O2) and catalytic ozonation are the principal advanced oxidation processes (AOPs) that are most promising for industrial wastewater treatment (Rice, 1996).

In this study, application of CaO2 and CaO along with O3 is investigated for sulfolane degradation. To the best of our knowledge this is the first time this study has been conducted. CaO2 is a relatively inexpensive solid peroxide, which has bleaching, disinfecting and deodorizing properties. It has a long history of application to site remediation and sewage treatment (Lu et al.,

2017). Depending on the pH, CaO2 can be used as a stable oxygen releasing source or as a solid

102 source for H2O2. Calcium peroxide reacts with water to produce H2O2 and Ca(OH)2; a maximum of 0.47 g H2O2/g CaO2 is released into the solution. According to Northup and Cassidy (2008), the advantage of using CaO2 is gradual release of H2O2, which minimizes the quenching effect of high concentration of H2O2 on hydroxyl radicals. Mehrabani et al. (2016) also reported that gradual addition of H2O2 increased the efficacy of sulfone degradation in H2O2/UV processes.

An analysis of CaO2/O3 results led us to investigate replacing CaO2 with lime (CaO). In both cases, the experiments were initially performed in a batch reactor with water spiked with sulfolane. This was followed by tests on well water samples collected from a contaminated site.

The success of the experiment was finally evaluated using a large scale field batch reactor.

4.2 Experimental Materials and Methods

Sulfolane with 99% purity, sodium hydroxide (97% purity), calcium oxide and CaO2

(reagent grade) were purchased from Sigma Aldrich, Canada. High purity oxygen was obtained from Praxair. Milli-Q water was used in all experiments, except for the sulfolane contaminated groundwater experiments.

The laboratory experiments were conducted in a batch reactor at room temperature. One hundred mL of sulfolane spiked solutions were introduced into the reactor followed by addition of

CaO2 or CaO at a specific dose. Gaseous ozone was produced by a A2Z 3-G lab O3 generator (A2Z

O3 systems Inc.) using pure oxygen. During the experiments, while the water sample was magnetically stirred, O3 was bubbled into the reactor through a plastic diffuser. The agitation speed was fixed at 20 rpm to ensure good gas-inducing condition and a homogeneous mixture. The ozone concentration was controlled by adjusting the electric current of the ozone generator, but for most

-1 -1 of the experiments 100% O3 with a flow rate of 0.5 L min (1.9 g h ) was used. Samples were

103 withdrawn at different intervals for GC and TOC analysis. One mL samples for GC analysis were collected at pre-set times and was extracted with 2 mL of dichloromethane (DCM); the extraction efficiency was determined to be about 80%.

For the field tests a Guardian model PB 30S1-C Plasma Block® ozone system, which could operate at 23,000 Hz (3200 V) and run up to 100 psi, was used. Ozone with a flow rate of 5 L min-

1 was bubbled into 60 L of water samples spiked with sulfolane inside a 100 L reactor.

An Agilent 6890 GC equipped with auto-sampler and Flame Ionization Detector (FID) was used in this study. Data acquisition and analyses were performed using Chemstation software.

Chromatographic separation was made on a fused silica capillary column (ZB 5MSI,

Phenomenex). High purity helium was used as the carrier gas with a head pressure of 250 kPa.

The temperature of the injection port was set to 165 oC and the injection was set on split less mode with 2.0 mL injection volume. The initial temperature was set to 90 oC, which was ramped up to

175 oC at a rate of 10 oC /min where it was held constant for 3 min. The FID detector temperature was set to 250 oC. External calibration was used to quantify the sulfolane in the sample by preparing various concentrations of sulfolane in DCM as standards. GC peak areas were determined from duplicate measurements on each of the samples in a ± 0.03 mg L-1 range. A linear calibration curve was prepared using seven different concentrations of standard solutions. The detection limit was 1 mg L-1. An Apollo 9000 combustion TOC analyzer equipped with an auto- sampler was used for TOC analysis. The detection limit was 0.1 mg L-1.

104

4.3 Results and Discussion

4.3.1 Degradation of Sulfolane Using CaO2/O3 - Different Loadings of CaO2, Effect of UV

Irradiation and Blank Experiments

In order to test the efficacy of CaO2 on sulfolane degradation, a series of ozonation experiments were carried out with different loadings of CaO2. Fig. 4.1 shows the evolution of the residual concentration of sulfolane in ozonation experiments and, Table 4.1 lists its corresponding degradation rates. It was found that the degradation rate of sulfolane increases with increasing the

-1 loading of CaO2 up to 1.6 g L (22.20 mM); but higher loadings of CaO2 didn't have any added advantage in increasing the reaction rate.

1.2 0.04 g CaO2 + O3 1 0.08 g CaO2 + O3

0 0.8 0.16 g CaO2 + O3 C/C 0.6 0.30 g CaO2 + O3 0.16 g CaO2 0.4 0.16 g CaO2 + UV 0.2

0 0 10 20 30 40 50 Time (min)

Fig. 4.1 Degradation of 220 mg L-1 (1.83 mM) sulfolane in presence of different loadings of -1 CaO2 and 0.5 L min of O3. The CaO2 loadings are per 100 ml of sulfolane solutions

105

-1 Table 4.1 Effect of CaO2 loading on degradation rate of 220 mg L of sulfolane bubbling O3 (0.5 L min-1)

-1 2 CaO2 Loading Pseudo first order rate constant (s ) R

0.4 g L-1- 5.55 mM 7.0×10-4 0.9223

0.8 g L-1- 11.10 mM 8.0×10-4 0.9978

1.6 g L-1- 22.20 mM 10.0×10-4 0.9981

3.0 g L-1- 41.62 mM 9.0×10-4 0.9988

Ozone and H2O2 (CaO2 is an effective source of H2O2 for in situ chemical oxidation (ISCO) processes) are two compounds, which have been used in previous studies along with UV to achieve higher degradation efficiencies for organic pollutants (Mehrabani et al., 2016; Yu et al., 2016,

2017). Therefore, the effect of UV irradiation on the reaction efficiency was also investigated by irradiating solutions containing different CaO2 loadings, while bubbling O3, with ten 254 nm UV lamps. In all cases, the reaction rates remained the same. This is because the pH of the solution was about 12 and at this pH, no H2O2 is produced (Northup and Cassidy, 2008). For the same reason, O3 is mostly decomposed and is not available for direct light absorption (this will be discussed later in the manuscript). It is consistent with Payton et al. results (1982), where they found comparable rate constants for degradation of tetrachloroethylene by ozonation and, photolytic ozonation at high pH values.

It should be noted that there was no reaction in the absence of ozone in either CaO2 or

CaO2/UV cases. No reaction was also observed for the given reaction time if only ozone was utilized. Heat could be generated, but temperature change was not observed plus or minus 2o. And, reaction with a commercially available CaO2 (CHEMCO) gave a very similar results (k = 12.0 ×

10-4 s-1for 1.6 g L-1 (22.20 mM)).

106

It has been previously reported that application of O3 along with UV is an efficient method for degradation of sulfolane (Mehrabani et al., 2016; Yu et al., 2016; Izadifard et al., 2017). This reaction was chosen as the reference degradation method for comparison with CaO2/O3 system.

The results show (Fig. 4.2) that the reaction in the latter case is more efficient in removing sulfolane from the solution during a 40 min reaction time; only 60% of sulfolane was removed at about the same time if O3 along with UV was used. The reaction rates in case of O3/UV were not determined as there is an induction period at the beginning. It has been found previously that the duration of the induction period is dependant on the initial concentration of sulfolane (Izadifard et al., 2017).

1.2 1 O3 + UV 0.8 0 O3 + 0.16 g CaO2

C/C 0.6 0.4 0.2 0 0 20 40 60 80

Time (min)

-1 -1 Fig. 4.2 Degradation of 220 mg L (1.83 mM) sulfolane using 1.6 g L (22.20 mM) CaO2/O3 -1 compared and UV/O3, with the O3 flow rate of 0.5 L min

The rate of sulfolane degradation is significantly lower if H2O2/O3 was used instead of

CaO2 along with O3. Mehrabani et al. (2016) reported ~90% degradation after 2 h of reaction if

H2O2 was used along with O3 (Mehrabani et al., 2016). The higher rate of sulfolane degradation could be because of the effect of pH on O3 decomposition reactions and/or effect of calcium on

107

-1 reaction rate (Imamura et al., 1982). The pH of solution using 3 g L (41.62 mM) of CaO2 was

11.0. Before discussing the pH effect, it should be noted that there are two types of oxidative reactions facilitated by O3 in aqueous solutions (Hordern et al., 2003; Langlais et al.,1991): direct oxidation by O3 or oxidation by hydroxyl radicals produced in water. In direct ozonation, organic molecules can be destroyed in various ways, including: i) breaking the double bond and formation of aldehydes and ketones, ii) addition of an oxygen atom to the benzene rings and iii) the reaction with alcohols to form organic acids (Shahamat et al., 2014).

In indirect oxidation, free hydroxyl radicals are produced via a chain reaction mechanism, which can cause a significant rise in pollutant removal efficiency (Camel and Bermond, 1998;

Hordern et al., 2003). According to Hordern et al. (2003), the pH value of the solution significantly influences ozone decomposition in water. Basic pHs (pH 8-12) promotes the decomposition of ozone, producing more hydroxyl radicals for the effective oxidization or degradation of organic compounds. At pH < 3 hydroxyl radicals do not influence the decomposition of ozone.

In order to validate if the only role for CaO2 is providing high pH for the sulfolane solutions, similar experiments were conducted by adding different concentrations of NaOH while the same flow rate of O3 was used. The results presented in Fig. 4.3 show that the sulfolane degradation rate increases with increasing NaOH concentration from 0.1 to 1.0 g L-1 (2.50- 25.00 mM). But TOC analysis (Fig. 4.4) shows that the TOC removal is not as effective as sulfolane

-1 removal if NaOH is used along with bubbling O3. It should be noted that 0.1 g L of NaOH

-1 produces the same pH as 3 g L (41.62 mM) of CaO2, which is 11, and that reaction corresponds to the slowest reaction with regards to sulfolane removal presented in Fig. 4.3.

108

1.5 NaOH (0.01 g)

NaOH (0.1 g)

1 0

NaOH (0.51 g) C/C 0.5

0 0 20 40 60 80 Time (min)

-1 -1 Fig. 4.3 Degradation sulfolane (220 mg L - 1.83 mM) using O3 (0.5 L min ) and different concentrations of NaOH. The loadings are for 100 ml of sulfolane solutions

120 CaO2 3.00 g/L 100

NaOH 5.10 g/L

) 1 - 80

40 TOC (mg L (mg TOC 20

0 0 20 40 60 80 100 120 140 160 Time (min)

-1 -1 Fig. 4.4 TOC removal by ozonation (0.5 L min ) in presence of CaO2 (3.00 g L – 41.62 mM) and NaOH (5.10 g L-1)

The next step of this study was to determine the effect of calcium on the reaction rate. Muroyama

-1 et al. (2011) observed that the addition of CaCO3 at a very low concentration of 0.1 g L -1 mM

(low concentration was used because of the limited solubility of CaCO3) enhanced the degradation

109 rate of phenol by O3 by a factor of about 2.3. They also reported a less efficient reaction if H2O2/O3 was used for degradation of phenol. The pH of the reaction was maintained at 7.0. They related the improved efficiency to the effect of pH, which caused a preferable condition for continuous generation of hydroxyl radicals, HO·, which maintained a strong oxidation ability to all the organic compounds as well as phenol and their subsidiary products; no effect for calcium was reported in this study. In another study CaO was used to study the effect of calcium on COD removal for the ozonation of phenol solutions (Hsu et al., 2007). The results show that the addition of Ca2+ can effectively enhance the COD removal rate. This was reported as a direct effect of calcium ion, which can bind with some of the intermediate products of phenol, including high molecular weight products, maleic acid and oxalic acid, to form insoluble precipitates. These calcium binding effects are reported to be responsible for the enhancement of COD removal. It should be noted that during the ozonation, the solutionpH was not controlled. Since CaO is used in this study, the pH could have a significant effect on the reaction efficiency.

To study the effect of calcium in our study, CaCl2 was used along with ozonation. No sulfolane degradation was observed if the pH was not increased. Therefore, high pH along with presence of calcium/calcium carbonate is necessary for degradation of sulfolane and consequent

TOC removal. This is consistent with literature, where a remarkable promoting effect of added

CaCl2 and Ca(OH)2 was studied on the degradation efficiencies of sodium acetate with ozone

(Yang et al., 2012). The COD removal rates of 65.73% and 83.46% were obtained, respectively, with the additions of CaCl2 and Ca(OH)2 after a reaction time of 30 min. These results were

29.37% and 47.10% higher than the results of oxidation only with ozone. They suggested that the

2+ 2- - added Ca has effectively shielded the production of CO3 and HCO3 produced by the complete oxidation of organic pollutants, and also that it promoted the progress of the ozonation reaction

110 and increased both the utilization ratio of ozone and the removal rate of organic pollutants. At the same time, they reported that the shielding efficiency of Ca(OH)2 was better than that of CaCl2, and the COD removal rate with Ca(OH)2 was 17.73% higher than that with CaCl2. The later statement could again be an evidence for enhanced effect of alkalinity on the reaction.

In order to find more evidence regarding the reaction mechanism, an experiment was performed using a very low concentration of CaO2 in water along with ozonation. This was done

-1 by stirring the reaction mixture with 1.6 g L (22.20 mM) of CaO2 for 15 min followed by filtration of the remaining solid CaO2. The results are shown in Fig. 4.5.

1.2

0.8

0 0.6

C/C 0.4

0.2

0 0 5 10 15 20 25 30 35 40 45 Time (min)

Fig. 4.5 Degradation of sulfolane in presence of CaO2 (filtered solution after 15 min stirring of -1 -1 1.6 g L – 22.20 mM of CaO2) under bubbling O3 (0.5 L min )

About 25% of sulfolane degraded after 5 min of reaction, no change in concentration of sulfolane was observed after this time. The original clear solution turned milky because of formation of white CaCO3 precipitates. The dissolved amount is only enough to carry about 25% of the reaction, therefore presence of the solid CaO2 in the solution and its gradual dissolution was required for

111 the complete mineralization of sulfolane (degradation sulfolane as well as its intermediate products).

Since high pH and presence of CaO2 seem to complement each other for effective mineralization of sulfolane, degradation of sulfolane using CaO2/O3 (or CaO/O3- section 4.3.2) could be considered as one of the catalytic ozonation processes. Catalytic ozonation has received increasing attention during past decades as a viable alternative to improve efficiency of ozonation. In most cases, catalytic ozonation processes involve hydroxyl radical generation (Zhai et al., 2009;

Nawrocki and Hordern, 2010). In catalytic ozonation, high pH plays a big role initiating the reactions and later surface chemistry is reported to be involved. Yunpeng et al. (2015) reported the

Ca(OH)2/O3 process as one of the catalytic ozonation processes, which was used for treatment of leachate from a municipal solid waste incineration power plant in China. Calcium hydroxide was introduced as the catalyst (it is not clearly stated why) in the process and the proposed mechanism was based on production of hydroxyl radicals in alkaline medium and precipitation of carbonate as calcium carbonate. CaCO3 formation is the result of complete mineralization of the organic compounds (sulfolane in our study) during the ozonation process. CaCO3 is reported by Li et al.

(2006), as a catalyst for heterogeneous oxidation of sulfur dioxide by ozone in gas phase. They proposed a mechanism with two steps: adsorption of SO2 on CaCO3 surface followed by O3 oxidation; adsorption of SO2 on CaCO3 surface was recognized to be the rate-determining step.

This experimental result indicates that CaCO3 can adsorb both O3 and SO2 on its surface. In another study adsorption of O3 on CaOwas validated (Bulanin et al., 1997). But as Nawrocki and Hordern

(2010) suggested, caution must be taken in case of catalytic ozonation in aqueous solutions, as adsorption of water on the active sites of the catalyst cannot be underestimated. Water molecules will always compete with ozone for catalytic sites such as Lewis centres or hydroxyls. Water seems

112 to be a stronger Lewis base than ozone molecule and thus the displacement of H2O by O3 at Lewis site is unlikely. More detailed mechanistic studies are required.

4.3.2 Degradation of Sulfolane using Lime/O3

Based on the results obtained using CaO2 for sulfolane degradation, effect of alkalinity and presence of calcium on reaction efficiency was evident. It was reasonable to test lime (CaO or

Ca(OH)2 as a more economical option for complete mineralization of sulfolane. The results for equivalent amounts of these two compounds are shown in Fig. 4.6. The TOC removal efficiency for CaO and CaO2 are compared in Fig. 4.7.

1.2 CaO + O3 1 CaO2 + O3 0.8 Ca(OH)2 + O3 0 0.6

C/C 0.4 0.2 0 0 10 20 30 40 50

Time (min)

-1 Fig. 4.6 Degradation of sulfolane (220 mg L -1.83 mM) using 50 mM CaO2, CaO, Ca(OH)2 and -1 O3 (0.5 L min )

113

120

100 CaO2

) 1 - 80 CaO 60

40 TOC (mg L (mg TOC 20 0 0 50 100 150 Time (min)

-2 Fig. 4.7 TOC removal using 5.0×10 M CaO2 (pH=11.0) or CaO (pH=11.85) under O3 bubbling with a flow rate of 0.5 L min-1

-4 -1 2 -4 -1 The reaction rate in case of CaO and Ca(OH)2 was 13.0 × 10 s -R = 0.9823 and 21.0 × 10 s -

2 R = 0.9519 respectively. Once dissolved in water, CaO is not distinguishable from Ca(OH)2, therefore the difference in reaction rate needs more detailed studies. Similar to CaO2 a loading of

1.6 g L-1 (28.53 mM) of CaO corresponds to a higher efficiency for sulfolane degradation (Fig.

4.8).

0.08g CaO + O3 Time (s) 0 0 500 1000 1500 2000 0.16g CaO + O3 -0.5 y = -0.0008x

-1 R² = 0.9952 0 -1.5 C/C y = -0.0012x -2 y = -0.0013x R² = 0.9943 -2.5 R² = 0.9888 -3

Fig. 4.8 Degradation of sulfolane using different loadings of CaO (0.8, 1.6, 3.0 g L-1- 14.26, -1 28.53, 53.50 mM) under bubbling O3 (0.5 L min ). The CaO loadings are per 100 mL of solutions

114

It should be noted that the percentage of O3 in the gas inlet could be decreased to 30% without any significant change in the degradation rate. This is due to the limited solubility of O3 in water and quick saturation of water sample with ozone (Fig. 4.9).

1.2 10% O3-670 1 30% O3-538

0.8 100% O3-491 0

0.6 C/C 0.4 0.2 0 0 10 20 30 40 50 Time (min)

-1 Fig. 4.9 Comparing different percentage of O3 with flow rate of 0.5 L min , in presence of 3 g L-1 of CaO

4.3.3 Degradation of Sulfolane in Contaminated Groundwater Samples (Laboratory Tests)

Application of CaO2 or CaO along with O3 for treatment of ground water contaminated with sulfolane was investigated using an actual groundwater sample spiked with sulfolane (Table

4.2 presents the characteristics of the well water sample used in this study). Sulfolane was degraded

-4 -1 -4 -1 successfully in both cases (k = 16.0 × 10 s compared to k = 10.0 × 10 s in DI water for CaO2; and k = 22.0 × 10-4 s-1 compared to k = 16.0 × 10 -4 s-1 for CaO in DI water). An unidentified component of the groundwater sample can play a positive role.

115

Table 4.2 Characteristics of groundwater samples

Parameters Well 1

Total dissolved solids (mg L-1) 726

pH 7.58

Dissolved organic carbon (mg L-1) 87.5

Anions

-1 Alkalinity (Total as CaCO3) (mg L ) 641

-1 Bicarbonate (HCO3) (mg L ) 782

-1 Sulfate (SO4) (mg L ) 38.6

Chloride (Cl) (mg L-1) 32.5

-1 Nitrate (NO3) (mg L ) 0.3

4.3.4 Degradation of Sulfolane in Contaminated Groundwater Samples (Field Tests)

An experiment was performed using 60 L of tap water spiked with sulfolane in presence

-1 -1 of 1.6 g L (28.53 mM) of CaO and O3 bubbling at 5 L min . The experimental set up, sulfolane and TOC removal results are shown in Fig.4.10-12, respectively. Sulfolane was not detectable after 150 min and 90% TOC removal was achieved after 250 min of reaction. It should be noted that the initial high pH of 11.9 -12.0 could be lowered to 6.5 by diffusing CO2 into the reactor (5 psi) for 10 min at the end the reactions instead of bubbling O3.

116

Fig. 4.10 Experimental set up for the field tests

Time (s) 0 -0.5 0 1000 2000 3000 -1 -1.5

0 y = -0.0016x -2 R² = 0.9817 -2.5 Ln C/C Ln -3 -3.5 y = -0.0022x CaO -4 R² = 0.965 -4.5 CaO2 -5

Fig. 4.11 Degradation of sulfolane in groundwater sample spiked with sulfolane using CaO2 and -1 CaO (50 mM) under bubbling O3 0.5 L min

117

45 40

35 )

1 30 - 25 20 15

TOC (mg L (mg TOC 10 5 0 0 50 100 150 200 250 300

Time (min)

Fig. 4.12 TOC removal in 60 L of tap water in presence of 1.6 g L-1 CaO (28.53 mM) and -1 5 L min of O3

Very similar results were obtained when the experiment was performed using 60 L of well water samples, (Fig. 4.13 and 4.14); no sulfolane was left after 2 h of reaction and 90% TOC removal was observed after 3 h of reaction.

160 140 120 100 80 60

40 GC PA (a.u.) GC 20 0 0 20 40 60 80 100 120 140 Time (min)

Fig. 4.13 Degradation of sulfolane (100 mg L-1- 0.83 mM) in 60 L of well water in presence of -1 -1 1.6 g L CaO (28.53 mM) and 5 L min of O3. The y axis is the PA measured by GC

118

) 1 - 30 25 20

TOC (mg L (mg TOC 15 10 5 0 0 50 100 150 200 250 300

Time (min)

Fig. 4.14 TOC removal in 60 L of well water in presence of 1.6 g L-1 CaO (28.53 mM) and

-1 5 L min of O3

4.4 Conclusions

This study demonstrates that application of calcium peroxide or lime along with O3 is a viable and effective method for treatment of water and groundwater contaminated with sulfolane.

Reduced treatment time compared to UV/O3 system, applicability of lime, which is readily available, and negligible matrix effect make this a practical method for sulfolane treatment. To the best of our knowledge this method is the only method, which can be used in situ for treatment of groundwater contaminated with sulfolane. High pH and presence of CaO/CaO2 seem to complement each other for effective mineralization of sulfolane. The effect of high pH on ozonation is already established in the literature but more mechanistic studies are required to clarify the role of solid calcium carbonate, which is produced during the reaction, on reaction rates.

119

CHAPTER FIVE: MECHANISM OF SULFOLANE DEGRADATION USING

CaO2/UV AND CATALYTIC OZONATION

5.1 Mechanism of Sulfolane Degradation using CaO2/UV

This chapter complements chapter five on application of CaO2/CaO along with ozone for degradation of sulfolane. The main goal of this study was to explain efficient TOC removal for sulfolane treatment using CaO2/O3 compared to that using NaOH/O3.

Fig. 5.1 comapres the TOC changes during sulfolane degradation using NaOH/O3 as well as CaO2/O3. Eventhough the rate of sulfolane removal from the solutions in both cases were comparable (chapter four), the TOC removal was much more efficient if CaO2 was used along with O3.

1.2

1 NaOH 0 0.8 CaO2 0.6

0.4 TOC/TOC 0.2 0 0 50 100 150 Time (min)

-1 -1 Fig. 5.1 TOC removal by ozonation (0.5 L min ) in presence of CaO2 (3.00 g L - 41.62 mM) and NaOH (5.10 g L-1 -127.50 mM)

The improved efficiency of TOC removal in case of using CaO2, must be related to the

2+ presence of Ca in the case of using CaO2, as the pH of the above mentioned reactions was set to be the same.

120

One possibility is formation of CaCO3, as a result of mineralization of sulfolane in presence of calcium. Calcium carbonate might act as a catalyst for catalytic ozonation. CaCO3 is reported by Li et al. (2006), as a catalyst for heterogeneous oxidation of sulfur dioxide by ozone in gas phase. They proposed a two-step mechanism: adsorption of SO2 on CaCO3 surface followed by O3 oxidation; with adsorption of SO2 on CaCO3 surface being recognized as the rate-determining step.

This experimental result indicates that CaCO3 can adsorb both O3 and SO2 on its surface. In another study adsorption of O3 on CaO was validated (Bulanin et al., 1997). But as Nawrocki and Hordern

(2010) suggested, caution must be exercised in case of catalytic ozonation in aqueous solutions, as adsorption of water on the active sites of the catalyst cannot be underestimated. Water molecules will always compete with ozone for catalytic sites such as Lewis centres or hydroxyls. Water seems to be a stronger Lewis base than ozone molecule and thus the displacement of H2O by O3 at Lewis site is unlikely.

In order to verify if CaCO3 act as a surface to catalyze degradation of sulfolane, a series of ozonation experiments were performed using CaCO3 along with O3. The experimental results showed that sulfolane was not adsorbed on CaCO3 surface, also degradation of sulfolane in

-1 presence of O3 bubbling for the loadings of 1.6 and 3.0 g L (16.00 and 30.00 mM) was negligible.

Calcium carbonate at low concentration along with O3 has been used for degradation of phenol by

-1 Muroyama et al. (2011). They chose 0.1 g L (1.00 mM) loading of CaCO3 because of the limited

solubility of CaCO3 in water. Addition of this amount of CaCO3 in the aqueous phase, helped maintain a rather high pH, which resulted in significant enhancement in phenol degradation rate instead of giving rise to negative scavenger effects.

Overall, no evidence of catalytic ozonation mechanism was found for this system. This was predictable as sulfolane is not effectively adsorbed on CaCO3. In addition, most likely water

121 molecules are adsorbed on the active sites of CaCO3 providing competition to O3 molecules in this aqueous system. A successful catalytic ozonation system requires adsorption of at least sulfolane or O3 on the catalyst surface.

On the other hand, Yang et al. (2012) reported a remarkable promoting effect of CaCl2 and

2+ Ca(OH)2 on degradation efficacy of sodium acetate with ozone. They suggested that Ca had

2- - effectively shielded the production of CO3 and HCO3 caused by the complete oxidation of organic molecules, which promoted the progress of the ozonation reaction and removal rate of organics. Calcium hydroxide effect was more significant than CaCl2, due to higher pH of the reaction and effect of alkalinity of ozonation reaction. Therefore, the next step of this study was to determine the effect of calcium on degradation rate of sulfolane and on TOC removal. To study the effect of calcium, CaCl2 was used along with ozonation. In this experiment, no sulfolane degradation was observed as the pH of solution was not in an alkaline range. Promoting the effect of CaCl2 on CaO2/CaO system is not informative as the system contains calcium, but it could be verified for NaOH/O3 system.

In another study CaO was used to study the effect of calcium on COD removal during the ozonation of phenol in water (Hsu et al., 2007). Their results showed that the addition of Ca2+ can effectively enhanced the COD removal rate. This was reported as a direct effect of the calcium ion, which can bind with some of the intermediate products of phenol, including maleic acid and oxalic acid, to form insoluble precipitates and as a result to enhance COD removal from the solutions. Sadrnourmohamadi and Gorczyca (2015) also investigated the effect of ozone as a stand-alone coagulation aid on the removal of dissolved organic carbon from the water with a high

-1 -1 level of DOC (13.8 mg L ) and calcium hardness (270 mg L as of CaCO3). Their results indicated that ozonation was able to lower DOC up to 27% and UV254 up to 86%. The cause was attributed to the high tendency of calcium to form strong complexes with functional groups containing

122 oxygen. Ozonation of organic compounds in water can produce oxygen-rich compounds, such as carboxylic groups, leading to increased complexation of calcium with both aqueous NOM and particle sorbed NOM.

These latter cases could be viable for sulfolane degradation in presence of calcium introduced to the solution by adding CaO2 or CaO. This hypothesis was verified for sulfolane first by finding the degradation pathway of sulfolane in case of using NaOH or CaO2 along with ozone, and then by comparing the concentrations of degradation products. These experiments were performed under similar conditions reported in chapter four, but the samples were analyzed using EI and ESI mass spectroscopy. The proposed degradation pathways, which was similar in both cases, is shown in Scheme 5.1. Both electron transfer and hydrogen transfer mechanisms were involved in degradation of sulfolane.

Scheme 5.1 Degradation pathway of sulfolane using NaOH/O3 or CaO2/O3. All degradation products listed in Scheme 5.1 were detected using MS spectra except dihydroxysulfolane

123

Considering the detected degradation products, 3-butenoic acid was a good candidate for further investigation.

The results shown in Fig. 5.2 show that the overall concentration of 3-butenoic acid over time is higher in

2+ NaOH/O3 system compared to that of CaO2/O3. This could be attributed to the presence of Ca in calcium peroxide/ O3 system and explains its enhanced TOC removal rate.

0.06

0.05

0.04

0.03 NaOH CaO2

[BA]/[Sulfolane] 0.02

0.01

0 0 20 40 60 80 100 Time (min)

Fig. 5.2 Changes in concentration of butenoic acid (BA) over time during sulfolane (220 mg L-1 - -1 1.83 mM) degradation by ozonation (0.5 L min ) in presence of CaO2 and NaOH; pH= 12.02

5.1.1 Conclusions

It is concluded that during the ozonation process, calcium ion can bind with some of the intermediate of sulfolane degradation to form insoluble precipitates. These calcium binding effects are responsible for the enhancement of TOC removal.

5.2 Catalytic Ozonation For Sulfolane Degradation

Catalytic ozonation has received increasing attention during past decades as a viable alternative to simple ozonation. The intention of using catalytic ozonation in water or wastewater

124 treatment is to improve the efficacy of ozonation. A short study was also performed to determine if a suitable catalyst can be found for degradation of sulfolane.

Magnesium oxide, silica gel, graphene, zeolite (Si/Al=280) were tested along with ozonation for sulfolane degradation.

Magnesium oxide was chosen because Mg is in the same group as Ca in periodic table and it is also reported as a catalyst for ozonation in the literature (chapter two). Our results indicate that sulfolane wasn’t degraded in this system. This could be because of more electropositive property of calcium than magnesium and more efficient production of Ca(OH)2 than Mg(OH)2 as a result of the reaction with water; which has significantly lower solubility than the latter (solubility

-12 -6 constant for Mg(OH)2 and Ca(OH)2 are 5.61×10 and 5.5×10 respectively). Overall the alkaline environment for efficient ozonation is not provided by MgO and it does not also act as a catalyst for this reaction.

There was also no reaction in case of silica gel (SG), zeolite or graphene either. Instead of performing experiment just with ozone, similar reactions were performed along with UV light to verify if presence of these solids improves the degradation reaction rate of sulfolane (the rate

-2 constants for these reactions were very close to that of O3/UV application, which was 2.6 ×10 min-1). The results are shown in Fig. 5.3-A, B & C.

125

1.2 A- silical gel 1 0.01 g SG + UV + O3 0.8 0.01 g SG + UV + 0 O3

0.6 0.06 g SG + UV + C/C O3 0.4 0.14 g SG + UV + 0.2 O3 0 0 20 40 60 80 Time (min)

1.2 B- zeolite 1 UV + O3 0.15 g zeolite + UV + O3 0.8 0.6

C/C0 0.4 0.2 0 0 20 40 60 80 Time (min)

C- graphene 1.2 UV + O3 1 0.8

0 0.12 g graphene + UV + 0.6 O3 C/C 0.4 0.2 0 0 20 40 60 80 Time (min)

Fig. 5.3 Degradation of sulfolane (220 mg L-1 -1.83 mM) using silica gel (SG), zeolite and graphene along with ozonation (0.5 L min-1) and 254 nm irradiation compare to that without catalyst

126

Ten different types of activated carbon (Table 5.1) were also tested along with ozonation.

No reaction was observed in all cases except when activated carbon type A and E were used. Even

in the latter two cases not only the dark adsorption but also the successful degradation of sulfolane

was not repeated. Some experiments showed consistent degradation of sulfolane over time, some

showed no degradation and some showed a sudden change in concentration at the beginning of

irradiation and no further change thereafter. Figures 5.4 and 5.5 show some of these inconsistent

results.

Table 5.1 List of 10 different AC used along with ozonation for sulfolane degradation

BET Average Iodine Pore Activated carbon Particle Surface Pore Precursor material number Volume Type sample size (μm) Area Width (mg/g) (cm³/g) (m²/g) (Å)

A OLC AW Coconut Shell 1700*425 1050 917 0.483 21.09

B Filtrasorb 300 Coal 2360*600 900 660 0.407 24.5

C Aquasorb CX Coconut Shell 2360*600 1100 1010 0.535 21.2

D Aquasorb CB1-MW Coconut Shell Powder 950 1001 0.698 27.42

E Haycarb Coconut Shell 4750*2000 1100 870 0.467 21.44

F Norit GCN 830 Coconut Shell 2360*600 1000 777 0.414 21.23

H Norit GAC 300 Coal 2360*600 900 797 0.527 26.24

L Aquacarb 830C Coconut Shell 2360*600 1100 890 0.483 21.56

M Aquacarb 1230C Coconut Shell 1700*600 1100 928 0.5 21.38

127

1.2 Test 1: 0.10 g AC A 1 Test 2: 0.10 g AC A Test 3: 0.16 g AC A 0.8

0 Test 4: 0.20 g AC A

0.6 Test 5: 0.30 g AC A C/C 0.4 Test 6: 0.50 g AC A 0.2 0 0 50 100 150 200 Time (min)

Fig. 5.4 Degradation of sulfolane using ozonation (0.5 L min-1) along with activated carbon-A

1.2 Teat 1: 0.08 g AC E 1 Test 2: 0.10 g AC E Teat 3: 0.30 g AC E 0.8 Test 4: 0.30 g AC E

0.6 C/C0 0.4

0.2

0 0 20 40 60 80 100 Time (min)

Fig. 5.5 Degradation of sulfolane using ozonation (0.5 L min-1) along with activated carbon E

These anomalous results could be due to the non-homogeneity of the different activated carbons, their differences in surface areas and magnification of effect in low loadings. Application of high loading of activated carbon in commercial applications might minimize the inconsistency

128 of sulfolane adsorption on the surface. Combined use of ozone and activated carbon is starting to be used for the treatment of toxic effluents (Liu et al., 2009).

Further it could be due to changes in the surface properties of active carbon upon ozonation and more importantly inefficient ozone mass transfer under the applied experimental conditions.

It has been shown that ozone also attacks activated carbon and changes its surface physical and chemical properties such as the surface area, micropore volume and active functional groups

(Valdes et al., 2002). Effect of ozonation on adsorption of sulfolane on activated carbon A and E was tested in this study. The results (Fig. 5.6 & 5.7) show reduced adsorption efficiency of sulfolane on activated carbon after 30 min ozonation of activated carbon.

300

250

200 Ozonated AC A 150

AC A PA (a.u.) PA 100

0 0 0.1 0.2 0.3 0.4 AC A Weight (g)

Fig. 5.6 Adsorption of sulfolane on different loadings of AC-A and ozonated AC-A. PA: GC peak area of sulfolane

129

1000 900 800 700 600 Ozonated E 500 AC E 400 PA (a. u.) (a. PA 300 200 100 0 0 0.1 0.2 0.3 0.4 AC E Weight (g)

Fig. 5.7 Adsorption of sulfolane on different loadings of AC-E and ozonated AC-E. PA: GC peak area of sulfolane 5.2.1 Conclusions

The result of this short study on application of heterogeneous caltalytic ozonation for sulfolane degradation was not conclusive. Most probably in most of these cases no hydroxyl radicals were produced. It is worth to continue research on application of catalytic ozonation for degradation of sulfolne, as it significantly decreases the amount of the required oxidants, also it can be done at lower pHs.

130

Chapter six: TREATMENT OF SULFOLANE IN SOIL BY COMBINING SOIL

WASHING/FLUSHING AND THREE AOP PROCESSES: (APS/UV, CaO2/UV AND

H2O2/UV)

6.1 Introduction

Soil washing/flushing systems provide a viable technique for remediating soils contaminated with a wide variety of organic and inorganic contaminants such as petroleum and fuel residues, radionuclides, heavy metals, polychlorinated biphenyls (PCBs), pentachlorophenol

(PCP), pesticides, cyanides, creosote, semi volatiles and volatile compounds (Pavel and

Gavrilescu, 2008). These techniques are widely used in Northern Europe and America for the treatment of contaminated soils (Karthika et al., 2016).

Soil washing is mostly an ex-situ remediation technique that removes contaminants from soil by washing the soil with an appropriate solvent and then separating the clean soils from the contaminated solvent (US EPA 1993, 1996). Since sulfolane is a polar organic solvent and is not strongly adsorbed onto the soil particles, water is an appropriate solvent for this process. A soil washing process transfers the contaminants from coarser soil fractions to finer fractions and to the wash water (a media transfer technology). The wash water is then treated with different remediation techniques and the washed coarser fraction may be as backfill at the site.

An evaluation of AOPs for remediation of sulfolane extracted from soil by a soil washing/flushing technique using water was conducted in this study. The following AOP processes were tested: H2O2/UV, ammonium persulfate APS/UV, APS/UV/O3, CaO2/O3, CaO/O3.

131

6.2 Experimental Procedure, Results and Discussions

Three hundred grams of air-dried ground soil sample was added to 500 ml of tap water and the mixture was stirred for 24 h in a plastic jar (see Fig. 6.1).

Fig. 6.1 Soil washing set up

The contaminated soil was provided by our sponsors and its characteristics are listed in Table 6.1.

The soil got classified as silty-clay soil in texture as it contained 8.3% of sand, 43% of silt and

48.6% of clay.

132

Table 6.1 Characterization of the soil samples

Parameters UNITS Amount Sand by hydrometer % 8.3 Silt by hydrometer % 43.0 Clay content % 48.6 Texture - Silty Clay Soluble Chloride (Cl) mg L-1 310 Soluble Conductivity dS/m 3.57 Soluble pH - 7.80 Sodium Adsorption Ratio (SAR) - 1.81 Soluble Calcium (Ca) mg L-1 93.4 Soluble Magnesium (Mg) mg L-1 22.8 Soluble Sodium (Na) mg L-1 75.0 Soluble Potassium (K) mg L-1 12.5 Saturation % % 48.0 Soluble Sulphate (SO4) mg L-1 260 Anion Sum meq L-1 14.2 Cation Sum meq L-1 10.1 Cation/EC Ratio - 2.84 Ion Balance - 0.713

The soil did not contain any sulfolane, therefore the wash water was spiked with sulfolane (to make a 220 mg L-1 -1.83 mM solution). Ammonium persulfate (APS), calcium peroxide and hydrogen peroxide along with 254 nm UV irradiation and ozone were used for degradation of sulfolane. In two specific cases, APS and CaO2 were added to the soil and then the soil was washed.

The goal was to get rid of the contaminants degradable by the oxidant before irradiation. That would possibly reduce the sulfolane degradation time. The preliminary results are shown in Fig. 6.2.

133

1200

1000

800 A B 600

C PA (a.u.) PA 400 D E 200 F 0 0 10 20 30 40 50 Time (min)

-1 Fig. 6.2 Degradation of sulfolane (220 mg L -1.83 mM) in soil washwater using APS or CaO2 along with 254 nm irradiation and ozonation (0.5 L min-1). PA: GC peak area of sulfolane

Table 6.2 Experiment A-F descriptions and the corresponding reaction rate constants, Wash water: WW; UVC light: UVCL Experiment Reaction condition k (M min-1) R2

A 100 ml WW+1.86 g (8.15 mM) APS+ 10UVCL+ O3 56.7 0.9723

B 100 ml WW+0.4 g APS (1.75 mM) +10 UVCL + O3 49.3 0.9641

-1 C 100 ml WW +100 mg L (2.93 mM) H2O2 +10 UVCL+O3 39.8 0.9700

2g (27.74 mM) CaO2+ 300 g soil, washed after + 10

D UVCL + O3 16.48 0.9700

2g (8.76 mM) APS + 300 g soil, washed after+ 10

E UVCL+ O3 36.4 0.9700

F tap water + 0.4 g APS (1.75 mM) + 10UVCL + O3 56.3 0.9526

These results only show the possibility of using these methods for treatment of sulfolane contaminated soil. The reaction rates when APS was used are comparable with those when H2O2 was applied.

134

In a further experiment, a soil flushing set up (shown in Fig. 6.3) was used to perform a limited number of tests. In this case, 150 g of dried ground soil was sonicated with 100 ml of water with or without oxidants. A filter paper, a wire mesh and another filter paper were set at the bottom of the column. Above this was the soil sample followed by a porous sand stone. The flush water was provided by tubes connected to 50 ml burettes. The hydraulic conductivity of soil samples was between 0.4-0.6 mL min-1.

Fig. 6.3 Soil flushing set up

A degradation experiment on the flush water was performed using H2O2 and O3 under 254 nm UV irradiation. The treated flush water was analyzed by GC/FID. The results are presented in

Fig. 6.4. Changes in fluorescence spectra of the samples were also recorded to monitor the degradation efficiency of the Dissolved Organic Matter (DOM) in washwater (Fig. 6.5). Figure

135

6.6 shows the appearance of the diluted flush water, before and after irradiation in presence of

H2O2/O3. These results indicate that this treatment method is also effective for DOM removal.

1.2

1 H2O2 + UV 0.8 H2O2 + O3+ UV 0.6

C/C0 0.4

0.2

0 0 10 20 30 40 50 60 Time (min)

Fig. 6.4 Degradation of sulfolane in two times diluted soil wash samples using H2O2 (13 mM) -1 and 10-UV lamps and bubbling O3 (0.5 L min )

350

300 stock 2 4 6 250 8 16 200 20 25 150 If (a.u.) If 30 56 100

0 350 400 450 500 550 600 650

Wavelength (nm)

Fig. 6.5 Changes in fluorescence spectra (λexci=350 nm, PMT=high) of water sample during -1 degradation of sulfolane using H2O2 (13 mM) and 10-UV lamps and bubbling O3 (0.5 L min )

136

Before After

Fig. 6.6 Washwater samples before and after H2O2, UV, O3 treatment

6.3 Conclusions

The following methods could successfully be used for degradation of sulfolane in soil washwater:

H2O2/UV, ammonium persulfate (APS)/UV, APS/UV/O3, CaO2/O3, CaO/O3. Performing detailed experiments and a cost analysis study are required to determine which method is the most efficient method for sulfolane removal from soil washwater.

137

Chapter seven: APPLICATION OF THE UV/CHLORINE ADVANCED OXIDATION

PROCESS FOR SULFOLANE TREATMENT

This chapter reports on a short study on assessment of the UV/Chlorine process as an advanced oxidation process for degradation of sulfolane. A detailed literature review for this chapter in provided in chapter two, section five. A similar set up to what used in chapter three and four is used for this study as well.

7.1 Degradation of Sulfolane using Bleach (NaOCl/HOCl) along with UV Light Irradiation

The first experiment was performed using a NaOCl solution under irradiation with 254 nm lamps

(Fig. 7.1). No degradation of sulfolane was observed.

1.4

1.2

0 0.8

C/C 0.6

0.4

0.2

0 0 5 10 15 20 25 Time (min)

Fig. 7.1 Degradation of sulfolane (20 ml- 220 mg L-1- 1.83 mM) using 80.60 mM NaOCl (pH=10.5) under 8-254 nm irradiation

138

Since the maximum absorption of OCl- occurs at 290 nm, an experiment was performed using 8-

300 nm lamps. It was found that the degradation of sulfolane was negligible for the same irradiation period. This could be attributed to the strong •OH scavenging action of OCl-. Most probably lower concentrations of sodium hypochlorite could result in degradation of sulfolane.

Since chlorine radicals has been observed to play a negligible role in the oxidation of organics in alkaline solutions (Nowell and Hoigné, 1992; Watts and Linden, 2007), these experiments are worth following up as there is less probability of production of disinfection by products.

On the other hand, as UV/HOCl is reported to be a more promising AOP than UV/OCl-

(Watts and Linden, 2007); the rest of the experiments were performed using HOCl/UV.

Fig. 7.2 Absorption spectra of HOCl and OCl- in water. Adapted from Feng et al., 2007

Preliminary results using HOCl under 254 nm irradiation showed about 40% reduction in concentration of sulfolane after about 15 min of irradiation (Fig. 7.3); no further degradation of sulfolane was observed even after continued irradiation. This could also be due to the scavenging effect of hypochlorous acid.

139

1.2

0.8 0

0.6 C/C 0.4

0.2

0 0 10 20 30 40 50 60 70 Time (min)

Fig. 7.3 Degradation of sulfolane (220 mg L-1- 1.83 mM) using 114.00 mM of HOCl (pH=3.4) and 254 nm irradiation

Gradual addition of HOCl instead of adding it in a single step at the beginning of the experiment, was also tested. Three different time intervals for addition of HOCl were evaluated. As can be observed from Fig.7.4, addition of HOCl every ten minutes (after the first 20min) provided the most amount of hydroxyl radicals leading to higher degradation rate of sulfolane under 254 nm irradiation (k~0.0133 min-1, R2=0.9752).

140

1.2

HOCl in Solution 1

HOCl add at 0 and 20 min 0.8

0 HOCl added at 0, 20, 30, 40 & 50 min

0.6 C/C HOCl added every 5 min 0.4

0.2

0 0 20 40 60 80 Time (min)

Fig. 7.4 Degradation of sulfolane (220 mg L-1- 1.83 mM) with sequential addition of HOCl under 254 nm irradiation

Irradiation of HOCl with 254 nm light produces both hydroxyl and chlorine radicals

(Nowell and Hoigné, 1992a, 1992b; Feng et al., 2007; Watts and Linden, 2007), and they both are involved in the following photochemical chain reactions:

(1) HO• chain reactions:

• • HO + RH→ R + H2O Eq. 7.1

R• + HOCl → RCl+ HO• Eq. 7.2

(2) •Cl chain reactions:

•Cl+ RH→ R• + HCl Eq. 7.3

R• + HOCl → ROH + HO• Eq. 7.4

Thus, in acidic solutions undesirable chlorinated products may be produced. Production of

– – – – inorganic species such as ClO4 , ClO3 , ClO2 , and BrO3 is also possible (Pisarenko et al., 2013;

Deng et al., 2014; von Gunten and Hoigné, 1994; Kang et al., 2006). To date, only a limited number of studies have investigated formation of chlorinated compounds by the UV/chlorine AOP and the

141 role of •Cl is not well established. According to Wang et al. (2015), in some studies reaction of •Cl with organic compounds is considered to be negligible (Watts and Linden, 2007; Nowell and

Hoigné, 1992b). In contrast, Fang et al. (2014) reported evidence that •Cl reacts with benzoic acid, suggesting that if NOM were to contain structurally similar components, chlorinated by-products could possibly form. In another study the reseachers reported that •Cl is rather selective, in that it has strong reactivity with electron-rich organics, such as phenol, dimethylaniline, and toluene, but less reactivity with electron-poor organics, such as nitrobenzene (Fang at al. 2018; Gebel et al.,

2010; Fang et al., 2014; Chuang, 2017). In general, HO• primarily contributes to the degradation of organic compounds compared to chlorine radicals. It is difficult for Cl• to induce an oxidation or substitution reaction. Detailed studies are required to investigate formation of chlorinated by product by chlorine/UV method.

7.2 Conclusions

Sequential addition of hypochlorous acid along with UVC irradiation is an effective method for sulfolane removal from contaminated water. Unless production of chlorinated by products is proved to be negligible, this method is not a practical method for sulfolane treatment in water.

142

Chapter eight: ALTERNATIVE VIEW OF CHLORINE OXIDATION STIMULATED

BY LONGER WAVELENGTH LIGHT6

8.1 Introduction

Ultraviolet (UV)-driven processes with photoactive chlorine species have been widely investigated for water and wastewater treatment purposes targeting organics (Buxton and Subhani 1972; Feng et al., 2007 & 2010; Jin et al., 2011; Boal et al., 2011; Wang et al., 2012). This process is considered an advanced oxidation process (AOP) because of the production of hydroxyl radical equivalents as a result of the decomposition of excited chlorine species during exposure to light with wavelengths in the UVC (100–280 nm) region. However, OCl- absorbance has a tail out into the region, where sunlight provides a UVA (315–400 nm) source at the Earth’s surface. The following equations list reactions, which occur in the UV/chlorine process based on the homolytic cleavage of HOCl/OCl-:

HOCl + hʋ (UV photons) →●OH+ ●Cl Eq. 8.1

OCl- + hʋ (UV photons) → ●O- + ●Cl Eq. 8.2

● - ● - O + H2O→ OH+ OH Eq. 8.3

This sequence is well documented in the UVC region (Buxton and Subhani, 1972; Feng et al.,

2007 & 2010; Jin et al. 2011; Boal et al., 2011; Wang et al., 2012). However, several recent studies have included irradiation of aqueous chlorine with wavelength well beyond (longer wavelength)

6 The research presented in this chapter was published in: J. Environ. Eng. 142(10), 04016048. The supplementary materials are combined with the original published paper.

143 the absorption maximum of aqueous chlorine oxidants (Shu et al. 2014; Nowell and Hoigné 1992a

& b; Chan et al., 2012; Forsyth et al., 2013). For example, Shu et al. (2014), describe a solar driven

UV/chlorine system, where light beyond 303 nm is reported to be responsible for breaking the O-

Cl bond with production of OH radicals, which then bleach oil sands process affected water

(OSPW). Forsyth et al. (2013) also report an enhanced inactivation of bacillus spores during solar photolysis of free chlorine, mentioned as an advanced disinfection process. A heterolytic cleavage of OCl− is also suggested in this case in addition to the homolytic pathway, in which chloride ion and atomic oxygen are produced. This mechanism is expected to lead to production of ozone and

− ClO2 , which are useful for disinfection purposes. Heterolytic cleavage is reported to have lower quantum yield than the homolytic one at λ = 254 and 313 nm but higher yield at λ = 365 nm. All photochemistry in these cases is assigned to the weak absorbance of the tail of the OCl− spectrum.

Yet, an alternate light absorber, the mixture constituting dissolved organic matter (DOM) or bacilli in water has much higher absorbance in the UVA and visible region than OCl−. So far, the role of the organic excited states appears to have been neglected.

This report provides evidence relevant to major natural water contamination for an alternative pathway, which makes allowance for a role of excited states of organic molecules under visible

(as well as UVA) light. In this study, a blue light-emitting diode (LED) light source with a narrow band at 440 nm is used. This illumination wavelength cannot excite chlorine oxidants (maximum absorption of OCl− is at 290 nm) in competition with DOM chromophores; therefore, significant breaking of O-Cl bonds is highly unlikely (Fig. 8.1).

144

300

1 - 250

200

) 1

- 150 cm 100 440 nm 50

0 MolarAbsorption coefficient (M 220 260 300 340 380 420 460 500 λ (nm)

Fig. 8.1 Absorption spectrum of sodium hypochlorite solution

This suggests that visible region organic chromophores may readily account for photochemistry in the presence of chlorine, where the increased oxidation susceptibility of organic excited states facilitates oxidation by chlorine oxidants. This straightforward process needs to be considered in case of long wavelength irradiation in the presence of chlorine species as oxidants when, as is the case with reactions of waters contaminated with colored contaminants, it may dominate.

8.2 Materials and Methods

8.2.1 Materials

Coumarin dye (C343), fresh sodium hypochlorite solution (reagent grade with available chlorine 10–15%; for calculating the chlorine concentration, the solution was assumed to be 12%), sodium sulfite and NaN3 (≥ 99.0%), potassium persulfate (≥ 99.0% purity), H2O2 (30% by weight in H2O) were purchased from Sigma Aldrich, and Suwannee River fulvic acid (SRFA) (a reference sample widely studied) was purchased from the International Humic Substance Society (IHSS)

(St. Paul, Minnesota). Blowdown water (BDW) obtained from a steam-assisted gravity drainage

145

(SAGD) oil sand operation was passed through a 0.45-μm filter and diluted 40 times and used.

Milli Q water was used as required in the experiments.

8.2.2 Sample Preparation, Irradiation, and Analysis

To the prepared SRFAs (100 mg L-1) and coumarin dye (2.8 ×10−5 M) stock solutions,

NaOCl solution was added immediately before starting the experiments (either dark or irradiation).

It should be noted that the concentration of NaOCl used in this study is quite high as this study is a mechanistic study to emphasize the importance of the excited state of the DOM. Even with high chlorine concentration (about three orders of magnitude higher than chlorine required for disinfection purposes), the spectroscopic absorbance of chlorine species at 440 nm is negligible compared to DOM absorption (Fig. 8.1). The exact concentration of NaOCl is mentioned in each specific case in the Results and Discussion section, as it is varied in different cases. NaN3 (1.0 ×

10−1 M) was used as a quencher to test for the possible involvement of singlet oxygen in the reaction. Hydrogen peroxide and potassium persulfate (6.9 × 10−2 M) were also tested as alternative primary oxidants to hypochlorite.

The samples (in 20 mL vials) were irradiated in a compact circular bench scale reactor, with inside diameter of 9 cm and depth of 7 cm, equipped with six 440 nm LED lamps evenly distributed in two rows. The ASMT-AL31-NPQ00 lamps with output of 3 W were purchased from

Avago Technology. For some cases of fast photoreaction, only two lamps were used for irradiation.

The number of photons absorbed by the solutions was determined by chemical actinometry using potassium ferrioxalate (Hatchard and Parker 1956). Incident photon flux, measured under exactly the same conditions as in the experiments, was evaluated to be 3.4 ± 0.2 × 1016 and 5.8 ± 0.2 ×

1015 photons.s-1 for six and two blue LED lamps respectively.

146

Changes in concentration of SRFA, coumarin dye, and BDW were monitored immediately at convenient time intervals by measuring their characteristic absorbance after initiating photoirradiation or thermal dark reaction using a UV-Vis spectrophotometer (HP 8452-A diode array spectrophotometer). SRFA samples were also analyzed by studying changes in fluorescence spectra using a characteristic excitation wavelength of 350 nm with emission in the 450–470 nm range. A Varian Cary Eclipse fluorescence spectrophotometer was used. The slit-widths were adjusted at 5 nm for both excitation and emission, and the photomultiplier was set to operate at

1,000 V.

Total organic carbon (TOC) was measured for SRFA and BDW for both dark thermal reaction and photoreaction. For TOC samples after a specific dark or irradiation time, 10% by wt. sodium sulfite was added to the solution to quench residual chlorine. An Apollo 9,000 combustion

TOC analyzer equipped with an autosampler enabled TOC analysis.

8.3 Photodegradation of SRFA, Coumarin Dye, and BDW

The effect of visible light irradiation on degradation of SRFA using the chlorine system was studied by following absorbance for dark reaction in comparison with those under 440 nm irradiation shown in Fig. 8.1; the absorption spectrum of NaOCl at its initial concentration is also included for comparison in Fig. 8.2(a). As well, changes in the characteristic humic fluorescence

(λex = 350 nm) of SRFA in presence of NaOCl in dark and under 440 nm irradiation and changes in fluorescence intensities at 446 nm during 30 min of dark or irradiation are shown in Figs. 8.3 (a and b) respectively. It should be noted that upon addition of chlorine to the SRFA solution, an immediate decrease of about 58% in the intensity of fluorescence is observed. That is why the

147 initial fluorescence intensity of SRFA is eliminated from these figures to differentiate changes in fluorescence intensity of SRFA upon irradiation and in the dark reactions.

0.25 A 0.2 0 min 17 min Dark 0.15 62 min Dark 17 min B-LED 0.1

62 min B-LED Absorbance 0.05 NaOCl

0 400 450 500 550 600 λ (nm)

B 1.2 Chlorine in 1 Dark 0.8 B-LED+

0 Chlorine

0.6 A/A 0.4 0.2 0 0 20 40 60 80 Time (min)

148

200 A 180 160 140 120 40 min Dark 100

(a.u.) 100 min Dark f I 80 40 min LED 60 110 min LED 40 20 0 300 400 500 600 700 λ (nm)

0.25 B

0.2

0.15

/I f I 0.1

0.05 B-LED- Chlorine Dark- Chlorine

0 10 30 50 70 90 110 130 Time (min)

Fig. 8.3 Changes in the characteristic humic fluorescence (λex = 350 nm) of (100 ppm) SRFA in presence of NaOCl (2.4×10-2 M) with the initial pH of 9.4 in dark and under 440 nm irradiation (A) and changes in Fluorescence intensities at 446 nm during 30 min of dark or irradiation (B). Fluorescence values are in arbitrary units under the condition specified in the experimental section

149

The hypochlorite bleaching of SRFA appears to be substantially enhanced by 440 nm irradiation. Interestingly, fluorescence data suggest that fluorophores are not bleached as extensively as other chromophores. This was also seen in SRFA-sensitized photocatalytic degradation on TiO2 by Chamoli et al. (2015). This is the case when SRFA is irradiated with just visible light, while efficient degradation/treatment of fluorophore organic compounds in OSPW as a result of hydroxyl radical production because of the UV/chlorine processes under sunlight (280–

400 nm) was recently reported by Shu et al. (2014). The present experiments were performed at

440 nm where none of the chlorine species absorb light, but SRFA does. Therefore, there is a possibility that the excited state of SRFA is oxidized by chlorine species.

To collect comparative evidence on dye excited state reactivity with OCl−, a coumarin dye

(C343 with maximum absorption of 436 nm), which was used in the first example of dye-sensitized oxidation on TiO2 with blue LED light (Ghosh et al., 2008), was chosen. Degradation of the dye over the same time period was monitored both in dark and under irradiation (no change was observed in the absence of chlorine under 440 nm irradiation). Because of a high rate of photodegradation of coumarin dye only two blue LED lamps were used in this experiment (Fig.

8.4). The improved efficiency of degradation under irradiation is clearly observed. It should be noted that the spectrum of the stock coumarin dye is not shown in Fig. 8.2 due to its high absorbance (A at 400 nm is 0.10).

150

0.05 A 45 sec Dark Chlorine

0.04 80 sec Dark- Chlorine

120 sec Dark- Chlorine 0.03 180 sec Dark- Chlorine

A (a.u.) A 0.02 200 sec Dark- Chlorine 45 sec B-LED- Chlorine 0.01 80 sec B-LED- Chlorine

0 120 sec B-LED- Chlorine 370 420 470 λ (nm)

0.40 B Dark+ Chlorine B-LED+ Chlorine

0.30

A/A0 0.20

0.10

0.00 40 90 140 190 240 Time (Sec)

151

Finally, a diluted SAGD blowdown water with similar composition to the process water reaching tailings ponds was used to verify the relevance of this pathway for industrial water samples. To parallel conditions reported by Chan et al. (2012), this sample was diluted by a factor of 40 before irradiation in the presence of NaOCl using the 440 nm LEDs.

These results also show enhanced bleaching of BDW under illumination (Fig. 8.5(A)). A control experiment was also performed for 1.5 h duration in the absence of NaOCl and under irradiation; no change was observed (Fig. 8.5(B)). However, due to the dark color of the BDW sample, there is a significant increase (up to 10° in 1.5 h) in temperature that is difficult to control. To study the effect of increasing temperature on the reaction rate, a dark experiment was performed at a controlled constant temperature of 35°C. The results show a significant effect of increasing temperature on reaction rate (Fig. 8.5(C).

152

1 A NaOCl 0.8 Dark-BDW 0.6 Irradiation-BDW

0.4 A (a.u.) A

0.2

0 370 420 470 520 570 λ (nm)

B 3 2.5 2 BDW stock 1.5 BDW B-LED 90 min

A (a. u.) A 1 0.5 0 370 420 470 520 570 620 670 λ (nm)

1.2 C 1 Dark, 23 degree

0.8 Dark, 35 degree Blue LED

0.6 C/C0 0.4

0.2

0 0 20 40 60 80 100 120 Time (min)

Fig. 8.5 Comparison of absorption spectra of a BDW sample in presence of NaOCl (7.7×10-2M) after 1.5h dark and irradiation at pH=9.9 (A); Control experiment under irradiation without NaOCl (B) and changes in absorption spectra (at 400 nm) of a BDW sample in presence of NaOCl (7.7 ×10-2M) in dark at 23, and 35oC and under 440 nm irradiation (C)

153

It is clear that enhancement of reactivity by blue LEDs remains substantial even if it was assumed the solution did reach a temperature of 35°C. The color of the solutions after the experiments is shown in Fig. 8.6 to provide visual evidence of the reaction in each case.

Dark, 23 oC Blue LED Dark, 35 oC

Fig. 8.6 BDW sample in presence of NaOCl (7.7 ×10-2 M) after 1.5 h in dark (23 and 35 oC) and under 440 nm irradiation

Overall, monitoring changes in absorption and fluorescence of SRFA, coumarin dye, and

BDW in the presence of chlorine species upon irradiation compared to dark reaction revealed that the degradation in all cases (bleaching) is stimulated by visible light, and it is thus faster than that in dark. These experiments were performed using 440 nm irradiation, where none of the chlorine species absorbs light (Maximum absorption for HOCl and OCl− are at 230 and 290 nm, respectively).

But the colored component of fulvic acids and BDW as well as coumarin dye absorb the

440 nm light. As it is common in photochemistry, the excited states of the colored components are more readily oxidized by oxidants (chlorine species in this case) than the ground state (Jones and

Fox 1994). The literature that stimulated this study is confirmed in its claims for a powerful role of chlorine based oxidants (Buxton and Subhani, 1972; Feng et al., 2007& 2010; Jin et al., 2011;

154

Boal et al., 2011; Wang et al., 2012). However, in this study, the photoprocesses are dominated by substrate excitation. In the fulvic acid case, a known long-lived triplet excited state is a plausible active species that may be a contributor to the observed photodegradation (Bruccoleri et al., 1993;

Chaikovskaya et al. 2004).

These experiments were performed using chlorine oxidants. Other oxidants might behave similarly to chlorine but there is little literature suggesting long wavelength acceleration by peroxides. To verify this, two other oxidants with no absorption in the visible region (H2O2 and potassium persulfate) were also tested for degradation SRFA. No significant difference between

−2 dark and under irradiation in the presence of 6.9 × 10 M of H2O2 or potassium persulfate was observed. It should be noted that even though the oxidation reaction with H2O2 and potassium persulfate is thermodynamically possible, it is very slow kinetically.

On the other hand, the observed effect could be because of the oxidation by singlet oxygen.

It is well known that organic dyes can sensitize singlet oxygen upon visible light irradiation, which can contribute in the oxidation process (Paul et al., 2004; Wilkinson and Brummer, 1981; Zepp et al., 1981). To determine if singlet oxygen is involved in the reported photodegradation, a comparative experiment was performed for SRFA under 440 nm irradiation in presence and absence of NaN3 as a well-known quencher of singlet oxygen. The results show that NaN3 does not suppress the photodegradation rate of SRFA. It should be noted that since pH of the solution was above nine; oxidation of sodium azide by hypochlorite would not affect scavenging of singlet oxygen (Betterton et al., 2010).

Therefore, there is strong evidence that the improved reaction efficiency under 440 nm irradiation is because of the light absorption by organic chromophores of SRFA and BDW. To measure effectiveness of the oxidation reactions, the TOC was also measured

155 in both SRFA and BDW cases in dark and under irradiation. Little change in TOC indicated that chromophore loss can be accompanied by accumulation of colorless intermediates that are not rapidly degraded. In the case of SRFA, one factor could be related to the redox properties of quinone functional groups (often considered to be important fluophore of SRFA—up to 50%), which are photosensitive and display multiple redox states (Cory and McKnight, 2005; Garg et al.,

2011). The fluorescence data, on the other hand, may suggest that some of the fluorophores are not quinone like.

8.4 Conclusions

The lesson for water treatment from these experiments is that chlorine species, already widely in use for water disinfection, may be exploited to degrade many organic contaminants using UVA and visible excitation (sunlight or visible lamps) of components of the contaminants. For example, given that high-resolution Fourier transform ion cyclotron resonance mass spectroscopy (FTICR-

MS) provides evidence that SRFA are complex mixtures of thousands of individual molecular species (Hertkorn et al., 2008), it is impossible to know what specifically the chromophores are.

Some may well be labile aggregates assembled from several of the molecules identified by mass spectroscopy (MS). This implies that each colored contaminant for which chlorine oxidants may be useful will have to be tested and characterized independently

156

Chapter nine: CONCLUSIONS & RECOMMENDATIONS FOR FUTURE RESEARCH

9.1 Conclusions

In this research, several oxidation processes for sulfolane degradation in aqueous systems were studied. These include: APS/UVC & UVA; APS/UVC/O3; CaO/O3; CaO2/O3; HOCl/UV.

Two mechanistic studies were also performed: one on sulfolane degradation mechanism using

CaO2/O3 and the other on chlorine oxidation stimulated by longer wavelength light. The conclusions drawn from this research are described in the following sections:

9.1.1 APS/UV and APS/UV/O3 systems

• Ammonium persulfate/UVC process is a sulfate/hydroxyl radical based advanced oxidation

process, which can efficiently degrade sulfolane in water (220 mg L-1-1.83 mM) within 40

min, in a batch reactor, if 3 g L-1 (13.10 mM) of persulfate was applied under UVC irradiation

with the light intensity of 5.21× 1021 ± 0.30 photons.s-1 (optimized conditions)

-1 • Bubbling O3 with the flow rate of 0.5 L min significantly decreased the reaction time (to 15

min) as well as the dosage of persulfate by 25% under the same UVC light intensity

In both APS/UVC and APS/UVC/O3 processes:

• degradation of sulfolane followed a pseudo-first order kinetic model. For the best reaction

conditions, the pseudo first order rate constants were determined to be 1.0×10-3 s-1 and 2.0×10-

3 -1 s for APS/UV and APS/UVC/O3 systems, respectively

• reaction rates were affected by persulfate dosages, initial pH and concentration of

carbonate/bicarbonate present in the reaction mixture. If 3 g L-1 of persulfate was used and the

157

pH was increased from 3.9 to 9.6, the reation rate decreased from 1.0×10-3 s-1 to 0.22×10-3 s-1

-3 -1 -3 -1 and from 2.3×10 s to 0.5×10 s for APS/UVC and APS/UVC/O3 cases, respectively. On

the other hand, in the presence of 100 mg L-1 of carbonate (after 20 min of reaction), sulfolane

removal decreased to 55% compared to 80% (in the absence of carbonate) and, 35% compared

to 90% (in the absence of carbonate) for APS/UV and APS/UVC/ O3, respectively

• low concentration of chloride (less then100 mg L-1) had no effect on the reaction rates

• It was possible also to degrade sulfone using UVA irradiation, but higher concentrations of

persulfate (> 3g L-1) were required. Optimized conditions are not reported in this study

• Sulfolane was degraded successfully in well water samples by using both methods but with a

lower rate compared to the experiments performed in MilliQ water. For 90% sulfolane

removal, 60- and 22-min irradiation (5.21× 1021 ± 0.30 photons. s-1) was required in presence

-1 of 3 g L of persulfate for APS/UVC and APS/UVC/O3 systems, respectively. The slower

reaction rate for well water samples was because of the pH greater than 6.9, higher bicarbonate

concentration and, the presence of the co-contaminants such as humic substances and amines

in groundwater samples, all of which compete with sulfolane for reactive radicals

• APS/UVC/O3 was also applicable for sulfolane degradation from soil washwater with a

reaction rate comparable with H2O2/UV/O3 system

9.1.2 CaO/CaO2 system

• Application of calcium peroxide or lime along with O3 was found to be a viable and effective

method for treatment of water and groundwater contaminated with sulfolane

• High pH and presence of CaO/CaO2 complement each other for effective mineralization of

sulfolane

158

• In a batch reactor, optimal CaO2/CaO dosage and the suitable O3 flow rate for complete

mineralization of sulfolane were found as 1.6 g L-1 and 0.5 L min-1, respectively. Both sulfolane

and TOC were totally removed in less than 40 and 150 min, respectively

• Degradation of sulfolane followed a pseudo-first order kinetics model (k=10.0 × 10-4 s-1) if 50

mM CaO2 was used. For similar applied conectration of CaO and Ca(OH)2, the rate constants

were: 13.0 × 10-4 s-1 and 21.0 × 10-4 s-1 , repectively

• Field tests using a 60 L reactor were successful in mineralization of sulfolane in both water

and groundwater under optimal conditions developed in lab. Sulfolane was not detectable after

150 min and, TOC was removed up to 90% after 250 min of reaction, if 50 mM of CaO along

-1 with 5 L min of O3 bubbling were used. The pH of the water samples could be lowered from

pH> 12 to near neutral (pH = 6.5) by bubbling CO2 after the treatment process

• The degradation pathways for sulfolane removal was found to be similar to that of NaOH/O3

system but the latter process is not efficient in TOC removal in comparable time frame. This

is because of the formation of Ca2+ complexes with the degradation products in case of using

CaO2/CaO along with O3

• A preliminary study showed that H2O2/UV, ammonium persulfate (APS)/UV, APS/UV/O3,

CaO2/O3, CaO/O3 could successfully be used for degradation of sulfolane in soil washwater.

CaO2/O3 was found to be have the slowest reaction rate, which could be because of the pH

effects on ozonation efficiency in case of using CaO2.

9.1.3 HOCl/UV system

• HOCl along with UVC successfully degraded sulfolane

159

• Stepwise addition of HOCl was necessary to maximize the available hydroxyl radicals in the

solution. If 1ml - 114.00 mM HOCl (pH=3.4) was added to a 1.83 mM sulfolane solution every

ten minutes, 80% of sulfolane was removed after 60 min under UVC irradiation (5.21× 1021 ±

0.30 photons. s-1). The optimised conditions with regards to HOCl dosage and light intensity

are yet to be found

• Degradation of sulfolane by OCl- was not successful in neither UVC nor UVA irradiation

region. This was because of the low extinction coefficient of OCl- under UVC irradiation and

high hydroxyl radical scavenging efficiency of OCl- under UVA irradiation

• Degradation of contaminants such as NOM in water in presence of chlorine species under

visible light irradiation is mainly because of the oxidation of the excited states of these

molecules by active chlorine species

9.2 Recommendations for Future Research

9.2.1 APS/UV and APS/UV/O3 systems

Since degradation of sulfolane was successfully achieved with APS/UV; APS/UV/O3 in a batch photo reactor, these processes can be adopted to pilot-scale by application of dimensionless groups of variables and parameters, and similarity concepts. The hydraulics, mass transfer of oxidants and distribution of O3, and radiation field need to be studied.

Finding degradation pathways and sulfate production analysis is also recommended. This is to ensure toxicity reduction, biodegradibility improvement and to adjust persulfate dosage for the treatment system. Also, in order to eliminate the need for UV light, application of persulfate in

160 an alkali medium (pH>11) is recommened. Hydroxyl radicals will be responsible for degradation of sulfolane, if that happens.

9.2.2 Sulfolane Removal in Soil Washwater

More detailed experiments for application of APS/UV; APS/UV/O3 and CaO/O3 for treatment of sulfolane in soil wash water is recommended. This will optimize the reaction conditions for each individual treatment process and can be used to compare their efficiencies to that of H2O2/UV system. Addition of oxidants to water during the soil washing step could decrease the amount of energy required in the treatment step.

Effect of co-contaminates should also be studied carefully as they compete for radicals in these treatment systems; the reactivity of co-contaminants might not be the same for all of these processes. Pilot scale application of these treatment methods along with a soil washing/flushing step is recommended.

9.2.3 HOCl/UV System

For this system the possibility of formation of chlorinated by products should be validated to make sure they are not produced during the treatment process, as they are regulated and shown to be carcinogenic. This problem could be magnified in actual water samples as a result of presence of

NOM in the system

161

9.2.4 Catalytic Ozonation

More detailed studies on possibility of using of catalytic ozonation for sulfolane removal from contaminated water is beneficial as compared to the other used in this research, the amount of oxidant required for sulfaolane degradation will be substantially decreased. Catalytic ozonation can also be studied at lowers pH ranges. Activated carbon is an attractive candidate, but its surface properties change upon ozonation causing decreased sulfolane adsorption. Application of higher loadings of activated carbon (>3 g L-1) and higher water volume is recommended as it overcomes the non homogeneity problem of AC as well as decreased adsorption of sulfolane on the surface as a result of ozonation.

162

REFERENCES Abu Amr, S.S., Abdul Aziz, H., Adlan, M.N., Bashir, M.J.K., 2013. Pretreatment of stabilized leachate using ozone/persulfate oxidation process. Chem. Eng. J. 221, 492-499.

Abu Amr, S.S., Abdul Aziz, H., Adlan, M.N., Alkasseh, J.M.A., 2014. Effect of ozone and ozone/persulfate processes on biodegradable and soluble characteristics of semiaerobic stabilized leachate. Environ. Prog. Sustain. Energy 33, 184-191.

Ahmad, M., Teel, A.L., Watts, R.J., 2013. Mechanism of persulfate activation by phenols. Environ. Sci. Technol. 47, 5864-5871.

AHS (Alberta Health Services), 2014. Potential sulfolane contamination; alternate water source continues to be advised. Edson. [Online]. Available: http://www.albertahealthservices.ca/ assets/news/advisories/ne-pha-2014-09-12-sulfolane-edson.pdf.

Agatonovic, V., Vaisman, E., 2005. Sulfolane impacted soil and groundwater treatability study. EBA engineering consultants Ltd., Calgary, AB. Available: http://www.esaa.org/wp-content/ uploads/2015/06/05-Agatonovic.pdf.

Alexander, W.M., Abreu B.E., Weaver L.C., Faith, H.E, Newberne, J.W., 1959. Pharmacology of some monohalogenated derivatives of sulfolane (tetrahydrothiophene-1,1-dioxide). Arch. Int. Pharmacodyn Ther. 119(3-4), 423-434.

Alvares, A.B.C., Diaper, C., Parsons, S.A., 2001. Partial oxidation by ozone to remove recalcitrance from wastewaters- A review. Environ. Technol. 22(4) 409-427.

Andersen, M.E., Jones, R.A., Mehl, R.G., Hill, T.A., Kurlansik, L., Jenkins, L.J. Jr.,1977. The inhalation toxicity of sulfolane (tetrahydrothiophene-1,l-dioxide). Toxicol. Appl. Pharmacol. 40, 463-472.

163

Andreozzi, R., Caprio, V., Insola, A., Marotta, R., 1999. AdvancedOxidation Processes (AOP) for water purification and recovery. Catal. Today 53(1) 51–59.

Anipsitakis, G.P., Dionysiou, D.D. Radical generation by the interaction of transition metals with common oxidants. Environ. Sci. Technol. 38, 3705–3712.

Antoniou, M.G., de la Cruz, A.A., Dionysiou, D.D., 2010. Degradation of microcystin-LR using sulfate radicals generated through photolysis, thermolysis and e- transfer mechanisms. Appl. Catal. B 96(3−4) 290−298.

Bak, A., Kozik, V., Dybal, P., Kus, S., Swietlicka, A., Jampilek, J., 2018. Sulfolane: Magic extractor or bad actor? Pilot-scale study on solvent corrosion potential. Sustainability 10, 3677.

Baker, D.C. 1987. U.S. Pat. 4,792,405. Retrieved from: https://patents.google.com/patent/ US479245

Barceló, D., Petrovic, M., 2008. Emerging contaminants from industrial and municipal waste occurrence, analysis and effects. Springer.

Barrow, RA. and Capon, R.J., 1992. Sulfolane as a natural product from the sponge/tunicate composite, Batzella sp. Lissoclinom sp. J. Nat. Prod. 55(9) 1330-1331.

Beltra´ n, F.J., 2003. Ozone reaction kinetics for water and wastewater systems. Lewis Publishers.

Betterton, E.A., Lowry, J., Ingamells, R., Venner, B., 2010. Kinetics and mechanism of the reaction of sodium azide with hypochlorite in aqueous solution. J. Hazard. Mater. 182(1–3) 716- 722.

Blystone, C., 2001. National Institute of Environmental Health Sciences, sulfolane presentation to National Toxicology Programme Board of Scientific Counselors Meeting. Retrieved from: http://ntp.niehs.nih.gov/ntp/about_ntp/bsc/2011/december/presentations/5_blystone_sulfolane.pdf.

164

Boal, A.K., Olson, B.E., Bajszar, G., Snyder, S.A., Stanford, B.D., Rivera, S.B., 2011. Synergistic inactivation of Bacillus globigii spores using combined aqueous chlorine and ultraviolet light. 242nd ACS National Meeting. Denver.

Bolton, J.R., Bircher, K.G., Tumas, W. Tolman, C.A., 2001. Figures-of-Merit for the technical development and application of advanced oxidation technologies for both electric- and solar- driven systems. Pure Appl. Chem. 73(4) 627-637.

Boni, M.R., Sbaffoni, S., 2012. Chemical oxidation by sodium persulphate for the treatment of contaminated groundwater. Laboratory tests. Chem. Eng. Trans. 28, 157-162.

Brown, V.K.H., Ferrigan, L.W., Stevenson, D.E., 1966. Acute toxicity and skin irritant properties of sulfolane. Brit. J. Ind. Med. 23, 302-304.

Bruccoleri, A., Pant, B.C., Sharma, D.K., Langford, C.H., 1993. Evaluation of primary photoproduct quantum yields in fulvic acid. Environ. Sci. Technol. 27(5) 889–894.

Bulanin, K.M., Lavalley, J.C., Tsyganenko, A.A., 1997. Infrared study of ozone adsorption on CaO. J. Phys. Chem. B 101, 2917-2922.

Buxton, G.V., Subhani, M.S., 1972a. Radiation chemistry and photochemistry of oxychlorine ions. Part 1. Radiolysis of aqueous solutions of hypochlorite and chlorite ions. J. Chem. Soc. Faraday Trans. 1, 68, 947-957.

Buxton, G.V., Subhani, M.S., 1972b. Radiation chemistry and photochemistry of oxychlorine ions. Part 2. Photodecomposition of aqueous solutions of hypochlorite ions. J. Chem. Soc. Faraday Trans. 1, 68, 958–969.

Buxton, G.V., Subhani, M.S., 1972c. Radiation chemistry and photochemistry of oxychlorine ions. Part 3. Photodecomposition of aqueous solutions of chlorite ions. J. Chem. Soc. Faraday Trans. 1, 68, 970–977.

165

Camel, V., Bermond, A., 1998. The use of ozone and associated oxidation processes in drinking water treatment. Review paper. Wat. Res. 32(11) 3208-3222.

Chamoli, U., Achari, G., and Langford, C. H., 2015. Self-sensitized degradation of natural organic matter (NOM) using TiO2 photocatalysis with visible light. Group report, University of Calgary. Department of Chemistry.

Cai, C., Zhang, Z., Zhang, H., 2016. Electro-assisted heterogeneous activation of persulfate by Fe/SBA-15 for the degradation of Orange II. J. Hazard. Mater. 313, 209-218.

CAPP (Canadian Association of Petroleum Producers)., 1997. Evaluation of the fate of sulfolane and DIPA in the subsurface at sour gas processing plant sites. Report prepared by the 132 Departments of Biological Sciences and Renewable Resources, CAPP Pub#1997-0004, University of Alberta, Canada.

CCME (Canadian Council of Minister of the Environment), 2006. Canadian Environmental Quality Guidelines for Sulfolane: Water and Soil. Scientific supporting document. [Online].

Cesaro, A., Naddeo, V., Belgiorno, V., 2013. Wastewater treatment by combination of Advanced Oxidation Processes and conventional biological systems. J. Bioremed. Biodeg. 4(8) 1000208.

Chaikovskaya, O.N., Levin, P.P., Suĺtimova, N.B., Sokolova, I.V., and Kuźmin, A.V., 2004. Triplet states of humic acids studied by laser flash photolysis using different excitation wavelengths. Russ. Chem. B Int., 53(2) 313–317.

Chan, T.W., Graham, N.J., Chu, W., 2010. Degradation of iopromide by combined UV irradiation and peroxydisulfate. J. Hazard. Mater. 181(1-3) 508-13.

Chan, P. Y., Gamal El-Din, M., and Bolton, J. R., 2012. A solar-driven UV/chlorine advanced oxidation process. Water Res. 46(17) 5672–5682.

166

Chuang, Y. H.; Chen, S.; Chinn, C.J.; Mitch, W.A., 2017. Comparing the UV/monochloramine and UV/free chlorine Advanced Oxidation Processes (AOPs) to the UV/hydrogen peroxide AOP under scenarios relevant to potable reuse. Environ. Sci. Technol. 51, 13859–13868.

CHEMISTRY International, 2016. Standardization of Electrical Energy Per Order (EEO)

Reporting for UV/ H2O2 Reactors. 38 (6).

Chen, W.S., Jhou, Y.C., Huang, C.P., 2014. Mineralization of dinitrotoluenes in industrial wastewater by electro-activated persulfate oxidation. Chem. Eng. J. 252, 166-172.

Chen, W.S., Huang, C.P., 2015. Mineralization of aniline in aqueous solution by electrochemical activation of persulfate. Chemosphere 125, 175-181.

Chen, J., Qian, Y., Liu, H., Huang, T., 2015. Oxidative destruction of diclofenac by thermally activated persulfate: implication for ISCO. Environ. Sci. Pollut. Res. 23(4) 3824-3833.

Chen, S. N., Hoffman, M. Z., Parsons Jr. G.H., 1975. Reactivity of the carbonate radical toward aromatic compounds in aqueous solution. J. Phys. Chem. 79(18) 1911-1912.

Chen, W.S. , Huang, C.P., 2015.Mineralization of aniline in aqueous solution by electrochemical activation of persulfate. Chemosphere 125, 175-81.

Chen, J., Hong, W., Huang, T., Zhang, L., Li, W.,Wang, Y., 2016. Activated carbon fiber for heterogeneous activation of persulfate:implication for the decolorization of azo dye. Environ. Sci. Pollut. Res. 23, 18564-18574.

Chong, M.N., Jin, B., Chow, C.W.K., Saint, C., 2010. Recent developments in photocatalytic water treatment technology: A review. Water Research, 44(10) 2997-3027.

167

Chou, C.C., R.A. Swatloski., 1983. Biodegradation of sulfolane in refinery wastewater. Proc. 37th Purdue Industrial Waste Conference. Ann Arbor Science Publishers: Ann Arbor, MI. 559-566.

Chiang, Y.P., Liang, Y.Y., Chang, C.N., Chao, A.C., 2006. Differentiating ozone direct and indirect reactions on decomposition of humic substances. Chemosphere 65, 2395–2400.

Christensen, T.H., Kjeldsen, P., Albrechtsen, H.J., Heron, G., Nielsen, P.H., Bjerg, P.L., Holm, P.E., 1994. Attenuation of landfill leachate pollutants in aquifer. Crit. Rev. Environ. Sci. Tech. 24(2), 119-202.

Cory, R.M., McKnight, D.M., 2005. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ. Sci. Technol. 39(21) 8142- 8149.

Deng, L., Huang, C.H., Wang, Y.L., 2014. Effects of combined UV and chlorine treatment on the formation of trichloronitromethane from amine precursors. Environ. Sci. Technol. 48(5) 2697– 2705.

Devi, P., Das, U., Dalai, A.K., 2016. In-situ chemical oxidation: Principle and applications of peroxide and persulfate treatments in wastewater systems. – Sci. Total Environ. 571, 643–657.

Diaper, E.W.J., Evans, F. L., 1972. Ozone m Water and Wastewater Treatment. Ann Arbor Science Publishers.

Dogliotti, L., Hayon, E., 1967. Flash photolysis of peroxydisulfate ions in aqueous solutions. The sulfate and ozonide radical anions. J. Phys. Chem. 71(8) 2511–2516.

Doucette, W.J., Chard, J.K., Moore, B.J., Staudt, W.J., Headley, J.V., 2005. Uptake of sulfolane and diisopropanolamine (DIPA) by cattails (Typha latifolia). Microchem. J. 81, 41 – 49.

168

Duan, X.; Ao, Z.; Sun, H.; Zhou, L.; Wang, G.; Wang, S., 2015. Insights into N-doping in single- walled carbon nanotubes for enhanced activation of superoxides: A mechanistic study. Chem. Commun. 2015, 51 (83) 15249−15252.

Duca, C., Imoberdorf, G., Mohseni, M., 2017. Effects of inorganics on the degradation of micropollutants with vacuum UV (VUV) advanced oxidation. J. Environ. Sci. Health A. 52, 6, 524–532.

Fang, G.D., Gao, J., Dionysiou, D.D., Liu, C., Zhou, D.M., 2013. Activation of persulfate by quinones: free radical reactions and implication for the degradation of PCBs. Environ. Sci. Technol. 47, 4605-4611.

Fang, J., Fu, Y., Shang, C., 2014. The roles of reactive species in micropollutant degradation in the UV/free chlorine system. Environ. Sci. Technol. 48, 1859–1868.

Fang, J., Liu, J., Shang, C., Fan, C., 2018. Degradation investigation of selected taste and odor compounds by a UV/chlorine Advanced Oxidation Process. Int. J. Environ. Res. Public Health 15(2) 284.

Fedorak, P. M., and Coy, D. L., 1996. Biodegradation of Sulfolane in Soil and Groundwater Samples from a Sour Gas Plant Site. Environ. Technol. 17(10) 1093-1102.

Feng, Y., Smith, D.W., Bolton, J.R., 2007. Photolysis of aqueous free chlorine species (HOCl and OCl-) with 254 nm ultraviolet light. J. Environ. Eng. Sci. 6(3) 277–284.

Feng, Y., Smith, D.W., Bolton, J.R., 2010. A potential new method for determination of the fluence (UV dose) delivered in UV reactors involving the photodegradation of free chlorine.” Water Environ. Res. 82(4) 328-334.

Feng, Y., Liu J., Wu, D., Zhou, Z., Deng Y., Zhang T., Shih, K., 2015. Efficient

169 degradation of sulfamethazine with CuCo2O4 spinel nanocatalysts for peroxymonosulfate activation. Chem. Eng. J. 280, 514-524.

Forsyth, J.E., Zhou, P., Mao, Q., Asato, S.S., Meschke, J.S., Dodd, M.C., 2013. Enhanced inactivation of Bacillus subtilis spores during solar photolysis of free available chlorine. Environ. Sci. Technol. 47(22) 12976-12984.

Furman, O.S., Teel, A.L., Watts, R.J., 2010. Mechanism of base activation of persulfate. Environ. Sci. Technol. 44, 6423-6428.

Garg, S., Rose, A., Waite, T.D., 2011. Photochemical production of superoxide and hydrogen peroxide from natural organic matter. Geochim. Cosmochim. Acta, 75(15) 4310–4320.

George, C., Rassy, H.E., Chovelon, J.M., 2001. Reactivity of selected volatile organic compounds (VOCs) toward the sulfate radical (SO4.-). Int. J. Chem. Kinet. 33, 539-547.

Ghauch, A., Tuqan, A.M., Kibbi, N. Ibuprofen removal by heated persulfate in aqueous solution: a kinetics study. Chem. Eng. J., 2012, 197, 483-492.

Ghosh, J.P., Langford, C.H., Achari, G., 2008. Characterization of an LED based photoreactor to degrade 4-chlorophenol in an aqueous medium using coumarin (C-343) sensitized TiO2. J. Phys. Chem. A, 112(41) 10310–10314. Giri, R.R., Ozaki, H., Takayanagi, Y., Taniguchi, S. Takanami, R., 2011. Efficacy of ultraviolet radiation and hydrogen peroxide oxidation to eliminate large number of pharmaceutical compounds in mixed solution Int. J. Environ. Sci. Tech. 8 (1) 19-30.

Glaze, W.H., Kang, J.W. and Chapin, D.H., 1987. The chemistry of water treatment processes involving ozone, hydrogen peroxide and ultraviolet radiation. Ozone- Sci. Eng. 9 (4) 335-352.

170

Gordon, C.J., Long, M.D., Dyer, R.S., 1984. Effect of ambient temperature on the hypometabolic and hypothermic effects of sulfolane in rats. Arch. Toxicol. 56, 123–127.

Gracia, R., Aragües, J.L., Ovelleiro, J.L., 1996.Study of the catalytic ozonation of humic substances in water and their ozonation byproducts. Ozone Sci. Eng. 18 (3) 195-208.

Grebel, J.E., Pignatello, J.J., Mitch, W.A., 2010. Effect of halide ions and carbonates on organic contaminant degradation by hydroxyl radical-based advanced oxidation processes in saline waters. Environ. Sci. Technol. 44, 6822-6828.

Greene, E.A., 1999. Microbial sulfolane degradation by environmental organisms isolated from contaminated sour gas plane sediment (Doctoral dissertation). Retrieved from https://ucalgary-primo.hosted.exlibrisgroup.com.

Greene E.A., Beatty P.H., Fedorak P.M., 2000. Sulfolane degradation by mixed cultures and a bacterial isolate identified as a Variovorax sp. Arch Microbiol. 174(1-2) 111-119.

Greene, E.A. and Fedorak, P.M., 1998. A differential medium for the isolation and enumeration of sulfolane-degrading bacteria. J. Microbiol. Methods 33(3), 255-262.

Greene, E.A. and Fedorak, P.M., 2001. Nutrient stimulation of sulfolane biodegradation in a contaminated soil from a sour natural gas plant and in a pristine soil. Environ. Technol. 22(6) 619- 629. von Gunten, U., Hoigné, J., 1994. Bromate formation during ozonation of bromide-containing waters: interaction of ozone and hydroxyl radical reactions. Environ. Sci. Technol. 28(7), 1234- 1242.

Guo, Y., Yang, L., Cheng, X., Wang, X., 2012. The application and reaction mechanism of catalytic ozonation in water treatment. J. Environ. Anal. Toxicol. 2 (7) 1000150.

171

Hatchard, C. and Parker, C.A., 1956. A new sensitive chemical actinometer. II. Potassium ferrioxalate as a standard chemical actinometer. Proc. R Soc. Lond. A 235, 518-536.

He, X., Mezyk, S.P., Michael, I., Fatta-Kassinos, D., Dionysiou, D.D., 2014. Degradation kinetics and mechanism of beta-lactam antibiotics by the activation of H2O2 and Na2S2O8 under UV-254 nm irradiation. J. Hazard. Mater. 279, 375-383.

Headley J., Fedorak, P., Dickson, L., 2002. A review of analytical methods for the determination of sulfolane and alkanolamines in environmental studies. J AOAC Int. 85, 1, 154-162.

Herrmann, H., 2007. On the photolysis of simple anions and neutral molecules as sources of O- - /OH, SOx and Cl in aqueous solution. Phys. Chem. Chem. Phys. 9, 3935-3964.

Hertkorn, N., Frommberger, M., Witt, M., Koch, B. P., Schmitt-Kopplin, P.H., Perdue, E.M., 2008. Natural organic matter and the event horizon of mass spectrometry. Anal. Chem., 80(23) 8908– 8919.

Hilles, A.H., Abu Amr, S.S., Hussein, R.A., El-Sebaie, O.D., Arafa, A.I., 2016. Performance of combined sodium persulfate/H2O2 based advanced oxidation process in stabilized landfill leachate treatment. J. Environ. Manage. 166, 493–498.

Hoigne, J. and Bader, H., 1976. The role of hydroxyl radical reactions in ozonation processes in aqueous solutions. Water Res. 10(5) 377-386.

Hoigné, J. 1988. The chemistry of ozone in process technologies for water treatment. Plenum- Press: New York.

Hordern, B.K., Ziółek, M., Nawrocki, J., 2003. Catalytic ozonation and methods of enhancing molecular ozone reactions in water treatment. A review. Appl. Catal. B 46, 639- 669.

172

Hori, H., Yamamoto, A., Hayakawa, E., Taniyasu, S., Yamashita, N., Kutsuna, S., Kiatagawa, H., Arakawa, R., 2005. Efficient decomposition of environmentally persistent perfluorocarboxylic acids by use of persulfate as a photochemical oxidant. Environ. Sci. Technol. 3(7) 2383-2388.

House, D.A., 1962. Kinetics and mechanism of oxidations by peroxydisulphate. Chem. Rev. 62(3)185-203.

Hsu, Y.C., Chen, J.H., Yang, H.C., 2007. Calcium enhanced COD removal for the ozonation of phenol solution. Water Res. 41, 71-78.

Hua, G., Reckhow, D.A., Junsung K., 2006. Effect of bromide and iodide ions on the formation and speciation of disinfection by products during chlorination. Environ. Sci. Technol. 40 (9) 3050- 3056.

Huang, C. P, Dong, C., Tang, Z., 1993. Advanced chemical oxidation: its present role and potential future in hazardous waste treatment. Waste Manage.13, 361-377.

Ikehata, K., Gamal El-Din, M., 2004. Degradation of recalcitrant surfactants in wastewater by ozonation and Advanced Oxidation Processes: A review. Ozone: Sci. Eng. 26(4) 327- 343.

Ikehata, K., Gamal El-Din, M., 2005. Aqueous pesticide degradation by ozonation and ozone- based Advanced Oxidation Processes: A review (Part I). Ozone: Sci. Eng. 27(2) 83-114.

Ikehata, K. and Gamal El-Din, M., 2006. Aqueous pesticide degradation by hydrogen peroxide/ultraviolet irradiation and Fenton-Type Advanced Oxidation Processes: A review. J. Environ. Eng. Sci. 5(2) 81–135.

Ikehata, K.; Naghashkar, N. J.; Gamal El-Din, M., 2006. Degradation of aqueous pharmaceuticals by ozonation and advanced oxidation processes: A review. Ozone: Sci. Eng. 28, 353−414.

173

Huling, S.G., Kanb, E., Caldwell, C., Park, S., 2012. Fenton-driven chemical regeneration of MTBE-spent granular activated carbon -A pilot study. J. Hazard. Mater. 205-206, 55-62.

Imamura, S., Umena, H., Kawabata, N., Teramoto, M., 1982. Ozonation of organic compounds in alkaline aqueous media. Can. J. Chem. Eng. 60 (6) 853–858.

Izadifard, M., Achari, G., Langford, C.H., 2017. Degradation of sulfolane using activated persulfate with UV and UV-Ozone. Water Res. 125, 325-331.

Ji, Y.F., Dong, C.X., Kong, D.Y., Lu, J.H., Zhou, Q.S., 2015. Heat-activated persulfate oxidation of atrazine: Implications for remediation of groundwater contaminated by herbicides. Chem. Eng. J. 263, 45-54.

Jin, J., Gamal El-Din, M., and Bolton, J.R., 2011. Assessment of the UV/chlorine process as an advanced oxidation process. Water Res., 45(4), 1890–1896.

Jones, W.E., and Fox, M. A., 1994. Determination of excited-state redox potentials by phase- modulated voltammetry. J. Phys. Chem. 98(19) 5095–5099.

Juris, A., Balzani, V.,1988. Ru(II) polypyridine complexes: photophysics, photochemistry, electrochemistry and chemiluminescence. Coord. Chem. Rev. 84, 85-277.

Kang, N., Anderson, T.A., Jackson, W.A., 2006. Photochemical formation of perchlorate from aqueous oxychlorine anions. Anal. Chim. Acta. 567(1) 48–56.

Karthika, N.; Jananee, K., Murugaiyan, V., 2016. Remediation of contaminated soil using soil washing-a review Int. J. Eng. Res. Appl. 6 (1) 13-18.

Kattel, E., 2018. Application of Activated Persulfate Processes for the Treatment of Water and High-Strength Wastewater. Doctoral thesis. Tallinn University of Technology.

174

Khan, J.A., He, X., Shah, N.S., Khan, H.M., Hapeshi, E., Fatta Kassinos, D. Dionysiou, D.D., 2014. Kinetic and mechanism investigation on the photochemical degradation of atrazine with 2− − activated H2O2, S2O8 and HSO5 . Chem. Eng. J. 2014, 252, 393−403.

Kirk-Othmer.,1999. Sulfolane and sulfones. Encyclopedia of Chemical Technology. Wiley, New York, USA.

Kolthoff, I.M., Miller, I.K., 1951. The chemistry of persulfate. I. The kinetics and mechanism of the decomposition of the persulfate ion in aqueous medium. J. Am. Chem. Soc. 73, 3055-3059.

Kopple, K.D., Baures, P.W., Bean, J.W., D'Ambrosio, C.A., Hughes, J.L., Peishoff, C.E. and Eggleston, D.S., 1992. Conformations of Arg-Gly-Asp containing heterodetic cyclic peptides: solution and crystal studies. J. Am. Chem. Soc. 1 l4 (24) 9615- 9623.

Krishnan, S., Rawindran, H., Sinnathambi, C.M., Lim, J.W., 2017. Comparison of various advanced oxidation processes used in remediation of industrial wastewater laden with recalcitrant pollutants. IOP Conf. Series: Materials Science and Engineering 206. Conference one.

De Laat, J., Truong Le, G. Legube, B., 2004. A comparative study of the effects of chloride, sulfate and nitrate ions on the rates of decomposition of H2O2 and organic compounds by Fe (II)/H2O2 and Fe (III)/ H2O2. Chemosphere 55 (5) 715-723.

Langlais, B., Reckhow, D.A., Brink, D.R.,1991. Ozone in water treatment: application and engineering. Lewis Publishers. Chelsea, Michigan, USA.

Lau, T.K., Chu, W., Graham, N.J.D., 2007. The aqueous degradation of butylated hydroxyanisole 2- by UV/S2O8 : Study of reaction mechanisms via dimerization and mineralization. Environ. Sci. Technol. 41 (2), 613–619.

175

Lawrence, J., Cappelli, F., 1977. Ozone in drinking water treatment: a review. Sci. Total Environ. 7, 99-108.

Lee, H., Lee, H.J., Jeong, J., Lee, J., Park, N.-B., Lee, C., 2015. Activation of persulfates by carbon nanotubes: oxidation of organic compounds by nonradical mechanism. Chem. Eng. J., 266, 28– 33.

Legrini, O., Oliveros, E., Braun, A.M., 1993.Photochemical processes for water treatment. Chem. Rev. 93(2) 671–698.

Legube, B., Leitner, N. K.V., 1999. Catalytic ozonation: a promising advanced oxidation technology for water treatment. Catal. Today 53, 61-72.

Loeb, B.L., 2002, Ozone: science & engineering. The first 23 years. Ozone Sci. Eng. 24(6) 399- 412.

Li, W., 2017. Sulfate radical-based advanced oxidation treatment for groundwater water treatment and potable water reuse. (Doctoral dissertation). Retrieved from: https://escholarship.org/uc/item/4qc8f723.

Li, L., Chen, Z. M., Zhang, Y. H., Zhu, T., Li, J. L. Ding, J., 2006. Kinetics and mechanism of heterogeneous oxidation of sulfur dioxide by ozone on surface of calcium carbonate. Atmos. Chem. Phys. 6, 2453–2464.

Li, Z., Yang, Q., Zhong, Y., Li, X., Zhou, L., Li, X., Zeng, G., 2016. Granular activated carbon supported iron as a heterogeneous persulfate catalyst for the pretreatment of mature landfill leachate. RSC Advances 6, 987–994.

Lian, L., Yao, B., Hou, S., Fang, J., Yan, S., Song, W., 2017. Kinetic study of hydroxyl and sulfate radical-mediated oxidation of pharmaceuticals in wastewater effluents. Environ. Sci. Technol. 51, 2954−2962.

176

Liang, C., Huang, C.F., Mohanty, N., Kurakalva, R.M., 2008. A rapid spectrophotometric determination of persulfate anion in ISCO. Chemosphere 73, 1540–1543.

Liang, C.J., Su, H.W., 2009. Identification of sulfate and hydroxyl radicals in thermally activated persulfate. Ind. Eng. Chem. Res. 48, 5558-5562.

Liang, C., Bruell, C.J., Marley, M.C., Sperry, K.L., 2004. Persulfate oxidation for in situ remediation of TCE. II. Activated by chelated ferrous ion. Chemosphere 55, 1225–1233.

Lin Y.T., Liang C., Chen J.H., 2011. Feasibility study of ultraviolet activated persulfate oxidation of phenol. Chemosphere 82(8) 1168-72.

2- Lin, C.C., Wu, M.S., 2014. UV/S2O8 process for degrading polyvinyl alcohol in aqueous solutions. Chem. Eng. Process 85, 209-215.

Liu, H., Korshin, G.V., Ferguson, J.F., 2008. Investigation of the kinetics and mechanisms of the oxidation of cerussite and hydrocerussite by chlorine. Environ. Sci. Technol. 42 (9) 3241-3247.

Liu, Z.Q., Ma, J., Cui, Y.H., Zhang, B.P., 2009. Effect of ozonation pretreatment on the surface properties and catalytic activity of multi-walled carbon nanotube. Appl. Catal. B Environ. 92, 301- 306.

Liu, J.N., Chen, Z., Wu, Q.Y., Li, A., Hu, H.Y.,Yang, C., 2016. Ozone/graphene oxide catalytic oxidation: a novel method to degrade emerging organic contaminant N, N-diethyl-mtoluamide (DEET). Sci. Rep. 6, 31405.

Liu, H., Bruton, T.A., Doyle, F.M., Sedlak, D.L., 2014. In situ chemical oxidation of contaminated groundwater by persulfate: decomposition by Fe(III)- and Mn(IV)-containing oxides and aquifer materials. Environ. Sci. Technol. 48, 10330-10336.

177

Loeb, B.L., Thompson, C.M., Drago, J., Takahara, H., Baig, S., 2012. Worldwide ozone capacity for treatment of drinking water and wastewater: A review. Ozone Sci. Eng. 34, 64-77.

Lominchar, M.A., Rodríguez, S., Lorenzo, D., Santos, N., Romero, A., Santos, A., 2018. Phenol abatement using persulfate activated by nZVI, H2O2 and NaOH and development of a kinetic model for alkaline activation. Environ. Technol. 39, 35–43.

Lu, S., Zhang, X., Xue, Y. 2017. Application of calcium peroxide in water and soil treatment: A review. J. Hazard. Mater. 337, 163-177.

Luo, S., Wei, Z., Dionysiou, D.D., Spinney, R., Hu, W.P., Chai. L., Yang. Z., Ye. T, .Xiao. R., 2017. Mechanistic insight into reactivity of sulfate radical with aromatic contaminants through single-electron transfer pathway. Chem. Eng. J. 327, 1056-1065.

Luther, S.M., Dudas, M.J., Fedorak, P.M., 1998. Sorption of sulfolane and diisopropanolamine by soils, clays and aquifer materials. J. Contam. Hydro. 32,159-176.

Lutze, H.V., Bircher, S., Rapp, I., Kerlin, N., Bakkour, R.; Geisler, M., von Sonntag, C., Schmidt, T. C. Degradation of chlorotriazine pesticides by sulfate radicals and the influence of organic matter. Environ. Sci. Technol. 2015, 49 (3) 1673−1680.

Lyman, W.J., Reehl, W.F. Rosenblatt, D.H., 1982. Handbook of chemical properties estimation methods. McGraw-Hill.

Martins, R.C., Quinta-Ferreira, R.M., 2014. A review on the applications of ozonation for the treatment of real agro industrial wastewaters. Ozone Sci. Eng. 36, 3-35.

Masten, S.J., Davies, S.H.R., 1994. The use of ozonation to degrade organic contaminants in wastewaters. Environ. Sci. Technol. 28(4), 180-185.

Matzek, L.W., Carter, K.E., 2016. Activated persulfate for organic chemical degradation: A review. Chemosphere 15, 178-188.

178

McLeod, D.W., Lin, C.Y., Ying, W.C. and Tucker, M.E., 1992. Biological activated carbon for removing sulfolane from groundwater. Proceedings 46th Industrial Waste Conference, Purdue University. Lewis Publishers: Ann Arabor, MI. 99-111.

Mehrabani-Zeinabad, M., Yu, L., Achari, G., Langford, C.H., 2016. Mineralisation of sulfolane by UV/O3/H2O2 in a tubular reactor.J. Environ. Eng. Sci. Paper 16.00018

Mehrjouei, M., Müller, S., Möller, D., 2015. A review on photocatalytic ozonation used for the treatment of water and wastewater. Chem. Eng. J. 263, 209-219

Minakata, D., Li, K., Westerhoff, P., Crittenden, J., 2009. Development of a group contribution method to predict aqueous phase hydroxyl radical (HO•) reaction rate constants. Environ. Sci. Technol. 43 (16) 6220-6227.

Mingy, L., Zhong, J., Xujiang, S., 2005. Cause of equipment corrosion and counter measures in the sulfolane recycling system of aromatics extraction unit. Pet. Process. Petrochem. 36, 30-33.

Morgan, P., Watkinson, R.J., 1989. Microbiological methods for the cleanup of soil and ground water contaminated with halogenated organic compounds. FEMS Microbiol. Rev. 63, 277-300.

Munter, R., 2001. Advanced Oxidation Processes – Current status and prospects. Proc. Estonian Acad. Sci. Chem., 50, 2, 59–80.

Muroyama, K., Suwa, S., Kawabata, A., Takami, Y., Hayashi, J., 2011. Effects of addition of hydrogen peroxide and/or calcium carbonate on ozone-decomposition of phenol sparingly dissolved in water ozone. Sci. Eng. 33, 143-149.

Nawrocki, J., Hordern, B.K., 2010.The efficiency and mechanisms of catalytic ozonation. Appl. Catal. B 99, 27-42.

179

Nawrocki, J., 2013. Catalytic ozonation in water: Controversies and questions. Discussion paper. Appl. Catal. B Environ. 142- 143, 465– 471.

Nawrocki, J., Rigney, M.P., McCormick, A., Carr, P.W., 1993.Chemistry of zirconia and its use in chromatography. Review. J Chromatogr A. 657(2) 229-82.

Neta, P., Madhavan, V., Zemel, H., Fessenden, R.W., 1977. Rate constants and mechanism of ●- reaction of SO4 with aromatic-compounds. J. Am. Chem. Soc. 99, 163-164.

Neta, P., Huie, R.E., Ross, A.B., 1988. Rate constants for reactions of inorganic radicals in aqueous solution. J. Phys. Chem. 17 (3) 1027-1284.

Northup, A., Cassidy, D., 2008. Calcium peroxide (CaO2) for use in modified Fenton chemistry. J. Hazard. Mater. 152, 1164-1170.

Nowell, L.H., Hoigné, J., 1992a. Photolysis of aqueous chlorine at sunlight and ultraviolet wavelengths–II. Hydroxyl radical production. Water Research, 26(5), 599–605.

Nowell, L.H., Hoigné, J., 1992b. Photolysis of aqueous chlorine at sunlight and ultraviolet wavelengths–I. Degradation rates. Water Research, 26(5), 593–598.

Oliver, B.G., Carey, J.H., 1977. Photochemical production of chlorinated organics in aqueous solutions containing chlorine. Environ. Sci. Technol. 11, 893-895.

Omar, A.A., Ramli, R.M., Faizura, P.N., Khamaruddin, M., 2010. Fenton oxidation of natural gas plant Wastewater. Can. J. Chem. Eng. 1, 1-6.

Oppenlander, T., 2003. Photochemical purification of water and air. WILEY-VCH.

180

Paraskeva, P., Graham, N.J.D., 2002. Ozonation of municipal wastewater effluents. Water Environ. Res. 74(6) 569–581

Parson, S., 2004. Advanced oxidation processes for water and wastewater treatment. IWA.

Paul, A., Hackbarth, S., Vogt, R.D., Röder, B., Urnison, B.K., Steinberg, C.E.W., 2004. Photogeneration of singlet oxygen by humic substances: Comparison of humic substances of aquatic and terrestrial origin. Photochem. Photobiol. Sci. 3(3), 273–280.

Pavel, L.V. Gavrilescu, M., 2008. Overview of ex situ decontamination techniques for soil cleanup. Environ. Eng. Manag. J. 7(6) 815-834.

Pérez-Sicairos, S., Corrales-López, K.A., Hernández-Calderón, Ó.M., Salazar-Gastélum, 2- M.I., Félix-Navarro, R.M., 2016. Photochemical degradation of nitrobenzene by S2O8 ions and UV radiation. Rev. Int. Contam. Ambie. 32 (2), 227-236.

Petri, B.G., Watts, R.J., Tsitonaki, A., Crimi, M., Thomson, N.R., Teel, A.L., 2011. Fundamentals of ISCO using persulfate. Serdp Estcp Remediat. 147-191.

Peyton, G.R., Huang, F.Y., Burleson, J.L., Glaze, W.H., 1982. Destruction of pollutants in water with ozone in combination with UV radiation. Environ. Sci. Technol. 16, 448–453.

Piera, E., Calpe, J.C., Brillas, E., Domènech, X., 2000. Peral, 2,4-Dichlorophenoxyacetic acid degradation by catalyzed ozonation: TiO2/UVA/O3 and Fe(II)/UVA/O3 systems, Appl. Catal. B Environ. 27, 169-177.

Pines, D.S., Reckhow, D.A, 2002. Effects of dissolved cobalt(II) on the ozonation of oxalic acid, Environ. Sci. Technol. 36(19) 4046–4051.

Pirkanniemi, K., Sillanp, M., 2002. Heterogeneous water phase catalysis as an environmental application: a review. Chemosphere 48, 1047-1060.

181

Pisarenko, A.N., Stanford, B.D., Snyder, S.A., Rivera, S.B., Boal, A.K., 2013. Investigation of the use of chlorine based advanced oxidation in surface water: oxidation of natural organic matter and formation of disinfection by products. J. Adv. Oxi. Technol. 16 (1) 137-150.

Rastogi, A., Al-Abed, S.R., Dionysiou, D.D., 2009. Effect of inorganic, synthetic and naturally occurring chelating agents on Fe(II) mediated advanced oxidation of chlorophenols.Water Res. 43, 684-694.

Reynolds, G., N. Graham, R. Perry, and R.G. Rice, 1989. Aqueous ozonation of pesticides- A review. Ozone Sci. Eng. 11(4) 339-382.

Rice, R.G., 1996. Applications of ozone for industrial wastewater treatment- A review. Ozone Sci. Eng. 18, 477-515.

Rizzo, L., 2011. Bioassays as a tool for evaluating advanced oxidation processes in water and wastewater treatment. Water Res. 45(15) 4311-4340.

Roscoe, J.M., Abbatt. J.P.D., 2005. Diffuse reflectance FTIR study of the interaction of alumina surfaces with ozone and water vapor, J. Phys. Chem. A 109, 9028-9034.

Sadrnourmohsmadi, M., Gorczyca, B. 2015., Effects of ozone as a stand-alone and coagulation- aid treatment on the reduction of trihalomethanes precursors from high DOC and hardness water. Water Res. 73, 171-180.

Salari D., Niaei, A., Aber, S., Rasoulifard, M.H., 2009. The photooxidative destruction of C.I. 2- 2009. Basic Yellow 2 using UV/S2O8 process in a rectangular continuous photoreactor, J. Hazard. Mater. 166, 61–66.

Sarathy, S.R., Mohseni, M., 2006. An overview of UV-based Advanced Oxidation Processes for drinking water treatment. IUVA News, 8(2) 16-27.

182

Sauleda, R., Brillas, E., 2001. Mineralization of aniline and 4-chlorophenol in acidic solution by ozonation catalyzed with Fe2+ and UVA light. Appl. Catal. B 29(2) 135-145.

Sauvé, S; Desrosiers, M., 2014. A review of what is an emerging contaminant. Chem Cent J. 8(1)15.

Scala, A.A., Colon, I., 1979. Vacuum ultraviolet photochemistry of tetrahydrothiophene and Sulfolane. Worcester Polytechnic Inst.: Worcester, MA, USA.

Schubert, E., F., 2006. Light Emitting Diodes.2nd edition. Cambridge publisher.

Schwarzer, H., 1995. Purification of raw potable water using Advanced Oxidation Processes. Chim. Oggi-Chem. Today 13(6)17-19.

Shah, S.M 2018. Health impact assessment of sulfolane on embryonic development of zebrafish. Master’s Thesis. University of Calgary.

Shahamat, Y.D., Farzadkia, M., Nasseri, S., Mahvi, A.H., Gholami, M., Esrafili, A., 2014. Magnetic heterogeneous catalytic ozonation: a new removal method for phenol in industrial wastewater. J. Environ. Health Sci. Eng. 12(1) 12- 50.

Shu, Z., Li, C., Belosevic, M., Bolton, J.R., and Gamal El-Din, M., 2014. Application of a solar UV/chlorine advanced oxidation process to oil sands process-affected water remediation. Environ. Sci. Technol., 48(16) 9692–9701.

Shu, H.Y., Chang, M.C., Huang, S.W., 2015. UV irradiation catalyzed persulfate advanced oxidation process for decolorization of Acid Blue 113 wastewater. Desalin. Water Treat. 54,1013- 1021.

Silva, L.M., Jardim, W.F., 2006. Trends and strategies of ozone application in environmental problems. Quim. Nova, 29(2) 310-317.

183

Shen, T., Wang, Q., Tong, S., 2017. Solid base MgO/ceramic honeycomb catalytic ozonation of acetic acid in water. Ind. Eng. Chem. Res.56 (39)10965–10971. Sra, K.S., Thomson, N.R., Barker, J.F. 2014. Stability of activated persulfate in the presence of aquifer solids. Soil Sediment. Contam. 23, 820-837.

Staehelin, J. and Hoigne, J., 1982. Decomposition of ozone in water: rate of initiation by hydroxide ions and hydrogen peroxide. Environ. Sci. Technol. 16(10) 676-681.

Stewart, O., Minnear, L., 2010. Sulfolane technical assistance and evaluation report. Alaska Department of Environmental Conservation. Retrieved at: http://dec.alaska.gov/spar /csp/sites/north-pole-refinery/docs/sulfolanereportfinal.pdf.

Sun, H., Kwan, C., Suvorova, A., Ang, H.M., Tadé, M.O., Wang, S., 2014. Catalytic oxidation of organic pollutants on pristine and surface nitrogen-modified carbon nanotubes with sulfate radicals. Appl. Catal. B 154-155, 134-141.

Tamuri, A.R., Sahar, M.A., N.S.A. Abu Bakar, Lani, M.N., Kundel, M.E., Daud, Y.M., 2014. Ultravoilet (UV) light spectrum of flourescent lamps. Conference paper. The Proceedings of 8th SEATUC Symposium.

Thompson, C.M., Gaylor, D.W., Tachovsky, J.A., Perry, C., Caracostas, M.C., Haws, L.C., 2013. Development of a chronic noncancer oral references dose and drinking water screening level for sulfolane using benchmark dose modelling. J. Appl. Toxicol. 33, 1395-1406.

Tilstam, U., 2012. Sulfolane: a versatile dipolar aprotic solvent. Org. Process Res. Dev.16 (7) 1273-1278.

Tsitonaki, A., Petri, B., Crimi, M., Mosabaek, H., Siegrist, R.L., Bjerg, P.L., 2010. In situ chemical oxidation of contaminated soil and groundwater using persulfate: A review. Crit. Rev. Environ. Sci. Technol. 40 (1) 55–91.

184

United States Environmental Protection Agency USEPA.,1993. Innovative site remediation technology: soil washing/soil flushing. EPA 542-B-93-012.

United States Environmental Protection Agency (USEPA)., 1996. A citizen's guide to soil washing. EPA 542-F-96-002.

Valentine, R. L., Wang, H.C.A., 1998. Iron oxide surface catalyzed oxidation of quinoline by hydrogen peroxide. J. Environ. Eng. 124 (1), 31-38. Valdes, H., Sánchez-Polo, M., Rivera-Utrilla, J., Zaror, C.A., 2002. Effect of ozone treatment on surface properties of activated carbon. Langmuir 18(6) 2111-2116.

Vogelpohl, A., 2007. Applications of AOPs in wastewater treatment: Water Sci. Technol. 55(12) 207–211. Q IWA Publishing.

Waclawek, S., Lutze, H.V., Grübel, K., Padil, V.V.T., Černík, M., Dionysiou, D.D., 2017. Chemistry of persulfates in water and wastewater treatment: A review. Chem. Eng. J. 330, 44-62.

Waldemer, R.H., Tratnyek, P.G., Johnson, R.L., Nurmi, J.T., 2007. Oxidation of chlorinated ethenes by heat-activated persulfate: kinetics and products. Environ. Sci. Technol. 41, 1010-1015.

Wang, P., Yang, S., Shan, L., Niu, R., Shao, X., 2011. Involvements of chloride ion in decolorization of Acid Orange 7 by activated peroxydisulfate or peroxymonosulfate oxidation. J. Environ. Sci. 23,1799-1807.

Wang, C.W., Liang, C., 2014. Oxidative degradation of TMAH solution with UV persulfate activation. Chem. Eng. J. 254, 472–478.

Wang, F., Smith, D.W., Gamal El-Din, M., 2003. Application of advanced oxidation methods for landfill leachate treatment-A review. J. Environ. Eng. Sci. 2(12) 413-427.

185

Wang, D. Bolton, J.R., Andrews, S.A., Hofmann, R., 2015. UV/chlorine control of drinking water taste and odour at pilot and full-scale. Chemosphere 136, 239-244.

Wang, D., Bolton, J.R., Hofmann, R., 2012. Medium pressure UV combined with chlorine advanced oxidation for trichloroethylene destruction in a model water. Water Res. 46, 4677-4686.

Watts, M.J., Rosenfeldt, E.J., Linden, K.G., 2007. Comparative OH radical oxidation using UV-

Cl2 and UV-H2O2 processes. J. Water Supply Res. Technol. AQUA 56, 469-477.

Watts, R. and Teel, A., 2006. Treatment of contaminated soils and groundwater using ISCO. Pract. Period. Hazard. Toxic Radioact. Waste Manage. 10, 2-9. Watts, M.J., Linden, K.G., 2007. Chlorine photolysis and subsequent OH radical production during UV treatment of chlorinated water. Water Res. 41(13) 2871–2878.

Watts, M.J., Hofmann, R., Rosenfeldt, E.J., 2012. Low-Pressure UV/Cl2 for Advanced Oxidation of Taste-and-Odor. J. Am. Water Works Assoc. 104, E58–E65.

Watts, R.J., Teel, A.L., Finn, D.D., Schmidt, J.T., Cutler, L.M., 2007. Rates of trace mineral catalyzed decomposition of hydrogen peroxide. J. Environ. Eng. 133(8) 853-858.

Weeks, J.L, Rabani, J., 1966. The Pulse radiolysis of deaerated aqueous carbonate solutions: I. transient optical spectrum and mechanism. II. pK for OH radicals1. J. Phys. Chem. 70, 2100-2106.

What is Sulfolane and why is it now a Contaminant of Concern. 2017. Retrieved from http://ridgelinecanada.com/steven-finch/

White, S.S, Fenton, S.E., Hines, E.P., 2011. Endocrine disrupting properties of perfluorooctanoic acid. J. Steroid. Biochem. Mol. Biol. 127(1-2) 16-26.

Wilkinson, F., and Brummer, J.G., 1981. Rate constants for the decay reactions of the lowest electronically excited singlet state of molecular oxygen in solution. J. Phys. Chem. Ref. Data, 10(4), 809–999.

186

Witzaney, A.M., Fedorak, P.M., 1996. A review of the characteristics, analyses and biodegradability of sulfolane and alkanolamines used in sour gas processing. Report prepared for and available from Shell Canada Limited.

Wrubleski, R.M. and Drury, C.R.,1997. Chemical contamination of groundwater at gas processing plants the problem. 23rd Annual Aquatic Toxicity Workshop, Calgary, AB, Canada.

Xiang, Y. Fang, J. Shang, C., 2016. Kinetics and pathways of ibuprofen degradation by the UV/chlorine advanced oxidation process. Water Res. 90, 301-308.

Xiao, R., Luo, Z., Wei, Z., Luo, S., Spinney, R., Yang, W., Dionysiou, D. D., 2018. Activation of peroxymonosulfate/persulfate by nanomaterials for sulfate radical-based advanced oxidation technologies. Cur. Opin. Chem. Eng.19, 51–58.

Yan, N., Liu, F., Huang, W.Y., 2013. Interaction of oxidants in siderite catalyzed hydrogen peroxide and persulfate system using trichloroethylene as a target contaminant. Chem. Eng. J. 219, 149-154.

Yan, D.Y., Lo, I.M., 2013. Removal effectiveness and mechanisms of naphthalene and heavy metals from artificially contaminated soil by iron chelate-activated persulfate. Environ. Pollut. 178, 15–22.

Yan, N., Liu, F., Huang, W.Y., 2013. Interaction of oxidants in siderite catalyzed hydrogen peroxide and persulfate system using trichloroethylene as a target contaminant. Chem. Eng. J. 219, 149-154.

Yang, Y., Guo, H., Zhang, Y., Deng, Q., Zhang, J., 2016a. Degradation of bisphenol A using ozone/persulfate process: kinetics and mechanism. Water, Air & Soil Pollut. 227, 1–12.

187

Yang, X., Sun, J., Fu, W., Shang C., Li Y., Gan W., Fang J., 2016b. PPCP degradation by UV/chlorine treatment and its impact on DBP formation potential in real water. Water Res. 98, 309-318.

Yang, D., Wang, B., Ren, H.Y., Yuan, J.M., 2012. Effects and mechanism of ozonation for degradation of sodium acetate in aqueous solution. Water Sci. Eng. 5(2) 155-163.

Yavas, A.Z., Ince, N.H., 2017. Catalytic ozonation of paracetamol using commercial and Pt- supported nanocomposites of Al2O3: The impact of ultrasound. Ultrason. Sonochem. 40, 175-182.

Ying, W.C., Hannan, S.C, Tucker, M.E, Friedrich, E.A., Lin, Cynthia, Y., 1994. Successful application of biological activated carbon process for removing sulfolane from groundwater. Proceedings of the Industrial Waste Conference.

Yu, L., Mehrabani, M., Achari, G., Langford, C.H., 2016a. Application of UV based advanced oxidation to treat sulfolane in an aqueous media, Chemosphere 160, 155-161.

Yu, L., Achari, G., Langford, C.H., 2016b. A feasibility study on sulfolane degradation in groundwater using neutral Fenton catalysts, CSCE Conference, London, Ontario, Canada.

Yu, L., Iranmanesh, S., Achari, G., Langford, C.H., Keir, I., 2017. Field evaluation of sulfolane degradation in groundwater using UV/H2O2, CSCE Conference, Vancouver, BC, Canada.

Yunpeng, Y., Xuejun, Q., Zhiliang, C., Fei, X., Facheng, Q., 2015. Catalytic ozonation of biologically treated leachate from municipal solid waste incineration power plant. Chin. J. Environ. Eng. 9, 219-224.

Zaretskii, M.I., Rusak, V.V., Chartov, E.M., 2013. Sulfolane and dimethyl sulfoxide as extractants. Coke Chem. 56, 266–268.

188

Zepp, R. G., Baughman, G. L., and Schlotzhauer, P. F., 1981. Comparison of the photochemical behavior of various humic substances in water-I. Sunlight induced reactions of aquatic pollutants photosensitized by humic substances. Chemosphere 10(1) 109-117.

Zhang, B.T., Zhang, Y., Teng, Y.H., Fan, M.H., 2015. Sulfate radical and its application in decontamination technologies. Crit. Rev. Env. Sci. Tec. 45, 1756-1800.

Zhang, Y.Q., Xie, X.F., Huang, S.B., Liang, H.Y., 2014. Effect of chelating agent on oxidation rate of aniline in ferrous ion activated persulfate system at neutral pH. J. Cent. South Univ. 21, 1441-1447.

Zhai, X.D., Wang, Q., Ma, J., Zhang, H., Deng, J., 2009. Detection of hydroxyl radicals in catalytic ozonation by using ESR spectroscopy. International Conference on Environmental Science and Information Application Technology.

Zhao, L., Hou, H., Fujii, A., Hosomi, M., Li, F.S., 2014. Degradation of 1,4-dioxane in water with heat- and Fe2þ-activated persulfate oxidation. Environ. Sci. Pollut. R. 21, 7457-7465.

Zhao, Y.S., Sun, C., Sun, J.Q., Zhou, R., 2015. Kinetic modeling and efficiency of sulfate radical- based oxidation to remove p-nitroaniline from wastewater by persulfate/Fe3O4 nanoparticles process. Sep. Purif. Technol. 142, 182-188.

Zou, X.L., Zhou, T., Mao, J., Wu, X.H., 2014. Synergistic degradation of antibiotic sulfadiazine in a heterogeneous ultrasound-enhanced Fe-0/persulfate Fenton like system. Chem. Eng. J. 257, 36-44. von Gunten, U., Hoigné, J., 1994. Bromate formation during ozonation of bromide-containing waters: interaction of ozone and hydroxyl radical reactions. Environ. Sci. Technol. 28(7) 1234- 1242. von Sonntag, C., von Gunten, U., 2012. Chemistry of ozone in water and wastewater treatment: from basic principles to applications. IWA Publishing

189

Zhou, H., Smith, D.W., 2001. Advanced technologies in water and wastewater treatment. Can. J. Civ. Eng., 28, 49–66.

Zhu, Z.H., Sun, M.L.; Li, Z.S.; Yang, Z.C., Zhang, T.B., Heng, Z.C. Xiao, B.L. Li, Q.Y. Peng, Q.Y. Dong, Y.H., 1987. An investigation of maximum allowable concentration of sulfolane in surface water. J. West China Univ. Med. Soc. 18, 376-380.

190