"Applied Science, Faculty of"@en . "Chemical and Biological Engineering, Department of"@en . "DSpace"@en . "UBCV"@en . "Dixit, Fuhar"@en . "2017-10-02T16:00:30Z"@en . "2017"@en . "Master of Applied Science - MASc"@en . "University of British Columbia"@en . "Harmful algal blooms have markedly increased in frequency over the past two decades, due to rising global temperatures and chemical runoff produced by modern farming practices. Cyanobacteria, also known as the blue green algae, are an essential part of the aquatic food chain. Cyanobacterial toxins, which are released from the algal biomass, can contaminate drinking, recreational and agricultural water reserves. Such environmental and health hazards pose a serious threat to communities and many aquatic ecosystems. In particular, small, shallow lakes that are sources of drinking water for rural communities have been particularly affected by elevated levels of algal toxins. Conventional water treatment systems are inefficient in removing these toxins, and may produce hazardous by-products. Additionally, the effectiveness of chemical oxidants, like chlorine and ozone, is hindered by the presence of natural organic matter (NOM) and inorganic ions in the contaminated water. Such challenges reveal the need for alternative treatment technologies capable of removing algal toxins. \r\nAnionic ion exchange (IX) resins offer a promising and cost-effective treatment alternative for natural waters affected by high algal content and high dissolved organic carbon (DOC) levels. This research investigates the efficiency of strongly basic ion exchange resins for the removal of microcystin-LR (MCLR), the most common toxin released from the algal biomass, inorganic ions, and NOM from natural water sources. We focused specifically on optimizing resin dosage, resin regeneration, and scaling up such procedures for particular source waters. Our results showed that the resin exhibited an excellent adsorption capacity of 3850 \u00B5g/L, removing more than 80% of the MCLR within 10 minutes at examined resin dosages (10 to 1000 mg/L; 1mL of resin = 221 mg). Additionally, we examined the influence of operating parameters such as pH and adsorption capacity of the resin, as well as the impact of NOM fractions on the uptake of MCLR. The charge density and molecular weight of the source water NOM were found to play a major competitive role in the uptake of Microcystin-LR. The results of this study bring the promising potential of IX resins for the removal of Microcystins and NOM from surface waters to light."@en . "https://circle.library.ubc.ca/rest/handle/2429/63181?expand=metadata"@en . "ANION EXCHANGE RESINS FOR THE REMOVAL OF MICROCYSTINS FROM SURFACE WATER by Fuhar Dixit B. Tech., Indian Institute of Technology (B.H.U), Varanasi, 2014 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemical and Biological Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) September 2017 \u00C2\u00A9 Fuhar Dixit, 2017 ii Abstract Harmful algal blooms have markedly increased in frequency over the past two decades, due to rising global temperatures and chemical runoff produced by modern farming practices. Cyanobacteria, also known as the blue green algae, are an essential part of the aquatic food chain. Cyanobacterial toxins, which are released from the algal biomass, can contaminate drinking, recreational and agricultural water reserves. Such environmental and health hazards pose a serious threat to communities and many aquatic ecosystems. In particular, small, shallow lakes that are sources of drinking water for rural communities have been particularly affected by elevated levels of algal toxins. Conventional water treatment systems are inefficient in removing these toxins, and may produce hazardous by-products. Additionally, the effectiveness of chemical oxidants, like chlorine and ozone, is hindered by the presence of natural organic matter (NOM) and inorganic ions in the contaminated water. Such challenges reveal the need for alternative treatment technologies capable of removing algal toxins. Anionic ion exchange (IX) resins offer a promising and cost-effective treatment alternative for natural waters affected by high algal content and high dissolved organic carbon (DOC) levels. This research investigates the efficiency of strongly basic ion exchange resins for the removal of microcystin-LR (MCLR), the most common toxin released from the algal biomass, inorganic ions, and NOM from natural water sources. We focused specifically on optimizing resin dosage, resin regeneration, and scaling up such procedures for particular source waters. Our results showed that the resin exhibited an excellent adsorption capacity of 3850 \u00C2\u00B5g/L, removing more than 80% of the MCLR within 10 minutes at examined resin dosages (10 to 1000 mg/L; 1mL of resin = 221 mg). Additionally, we examined the influence of operating parameters such as pH and adsorption capacity of the resin, as well as the impact of NOM fractions on the uptake of MCLR. The charge iii density and molecular weight of the source water NOM were found to play a major competitive role in the uptake of Microcystin-LR. The results of this study bring the promising potential of IX resins for the removal of Microcystins and NOM from surface waters to light. iv Lay Summary Cyanobacterial toxins, such as microcystins, which are released from algal biomass have become a serious environmental and health hazard in regards to their potential to contaminate drinking, recreational and agricultural water sources. The challenges with conventional water treatment process present the need for alternative treatment technologies capable of removing algal toxins. Anionic ion exchange resins offer a promising and cost-effective alternative for natural waters affected by high cyanotoxin content and high dissolved organic carbon (DOC) levels. This research has investigated the efficiency of strongly basic ion exchange resins for the removal of microcystins, inorganic ions, as well as natural organic matter (NOM) from natural water sources, with a specific focus on optimizing the resin dosage, regeneration, and scale up potentials for particular source waters. v Preface This dissertation is submitted for the degree of Master of Applied Science at The University of British Columbia. I was in charge of conducting the literature survey, identifying the research gaps and developing the research proposal, conducting the experiments, performing and analyzing the data (unless otherwise stated in below), writing and preparing the manuscripts, conference presentations and posters throughout the course of this research. The research described herein was conducted under the supervision of Professor Madjid Mohseni in Chemical and Biological Engineering department of The University of British Columbia. The following is the list of manuscripts that have been communicated in academic journals. Journals: \u00EF\u0082\u00B7 Dixit, F., Barbeau, B., Mohseni, M. \u00E2\u0080\u009CStoichiometry of Microcystin-LR removal in the presence of mineral anions via ion exchange\u00E2\u0080\u009D. (Under Review) \u00EF\u0082\u00B7 Dixit, F., Barbeau, B., Mohseni, M. \u00E2\u0080\u009CCharacteristics of competitive uptake between Microcystin-LR and Natural Organic Matter (NOM) fractions by strongly basic anion exchange resins\u00E2\u0080\u009D. (Under Review). \u00EF\u0082\u00B7 Dixit, F., Barbeau, B., Mohseni, M. Simultaneous uptake of NOM and Microcystin-LR by anion exchange resins: Effect of inorganic ions and resin regeneration. (Under Review). vi Table of Contents Abstract .......................................................................................................................................... ii Lay Summary ............................................................................................................................... iv Preface .............................................................................................................................................v Table of Contents ......................................................................................................................... vi List of Tables ................................................................................................................................ xi List of Figures .............................................................................................................................. xii List of Abbreviations ................................................................................................................. xiv Acknowledgements ................................................................................................................... xvii Dedication ................................................................................................................................... xix Chapter 1: Background .................................................................................................................1 1.1 Harmful Algal Blooms (HABs) ...................................................................................... 1 1.2 Natural Organic Matter ................................................................................................... 2 1.3 Treatment of Cyanotoxins............................................................................................... 2 1.4 Anion Exchange Process................................................................................................. 3 1.5 Research Rationale.......................................................................................................... 3 1.6 Thesis Layout .................................................................................................................. 4 Chapter 2: Literature Review .......................................................................................................6 2.1 Harmful Algal Blooms (HABs) ...................................................................................... 6 2.1.1 Factors Governing Cyanobacterial Growth and Development of Blooms ................. 7 2.2 Cyanobacterial Toxins .................................................................................................... 9 2.2.1 Neurotoxins ................................................................................................................. 9 vii 2.2.2 Hepatotoxins ............................................................................................................. 11 2.2.3 Lipopolysaccharides ................................................................................................. 13 2.2.4 Dermatotoxins ........................................................................................................... 13 2.2.5 Toxic Amino Acids ................................................................................................... 15 2.3 Cyanotoxin Detection and Quantification .................................................................... 16 2.3.1 Physicochemical Methods ........................................................................................ 17 2.3.2 Biochemical Assays .................................................................................................. 17 2.3.3 Field Test Kits ........................................................................................................... 18 2.3.4 Quality Assurance and Control ................................................................................. 19 2.4 Microcystins .................................................................................................................. 21 2.4.1 Guidelines for Water Usage ...................................................................................... 21 2.4.2 Geographical Distribution in Canada........................................................................ 22 2.4.3 Environmental Stability ............................................................................................ 23 2.5 Organic Matter in Surface Water .................................................................................. 24 2.5.1 Natural Organic Matter (NOM) ................................................................................ 24 2.5.2 Organic Matter from Cyanobacterial Cells ............................................................... 25 2.6 Treatment of Cyanotoxins............................................................................................. 27 2.7 Anion Exchange Process............................................................................................... 30 2.7.1 Anion Exchange for Organic Contaminants ............................................................. 30 2.7.2 Regeneration of IX Resins ........................................................................................ 33 2.8 Knowledge Gaps ........................................................................................................... 33 2.9 Research Objectives ...................................................................................................... 36 2.10 Engineering Significance .............................................................................................. 37 viii Chapter 3: Experimental Methodology .....................................................................................38 3.1 Experimental Procedures .............................................................................................. 38 3.1.1 Glassware Preparation .............................................................................................. 38 3.1.2 Chemicals and Reagents ........................................................................................... 38 3.1.3 Water and Resin Preparation .................................................................................... 39 3.1.4 Isotherm and Kinetic Studies .................................................................................... 40 3.1.5 Resin Regeneration and Treatment of Spent Brine ................................................... 40 3.2 Analytical Methods ....................................................................................................... 41 3.3 Statistical Analysis ........................................................................................................ 42 Chapter 4: Kinetics of Microcystin-LR Uptake by IX Resins .................................................43 4.1 Introduction ................................................................................................................... 43 4.2 MCLR Removal Mechanism ........................................................................................ 44 4.3 Isotherm Models ........................................................................................................... 45 4.3.1 Langmuir Msotherm model ...................................................................................... 45 4.3.2 Freundlich Isotherm Model....................................................................................... 46 4.4 Kinetic Models .............................................................................................................. 46 4.4.1 Pseudo-first-order Kinetic Model ............................................................................. 47 4.4.2 Pseudo-second-order Kinetic Model......................................................................... 47 4.4.3 Film and Intra-particle Diffusion Model................................................................... 47 4.5 MCLR Uptake Capacity on IX Resin ........................................................................... 49 4.6 Kinetic Results .............................................................................................................. 52 4.7 Conclusions ................................................................................................................... 56 Chapter 5: Effect of Background NOM Concentration on MCLR Uptake ...........................57 ix 5.1 Introduction ................................................................................................................... 57 5.2 Effect of Resin Dosage ................................................................................................. 58 5.3 MCLR and NOM Uptake Capacities on IX Resin........................................................ 59 5.4 MCLR Removal Kinetics: Effect of Background NOM Concentration ....................... 62 5.5 Conclusion .................................................................................................................... 65 Chapter 6: Impact of NOM Characterisitcs on MCLR Uptake ..............................................67 6.1 Introduction ................................................................................................................... 67 6.2 Competitive Uptake of NOM and MCLR .................................................................... 68 6.3 Isotherm Modelling ....................................................................................................... 72 6.4 Effect of NOM Characteristics on Kinetics of MCLR Uptake ..................................... 77 6.5 Effect of pH................................................................................................................... 82 6.6 Comparative Studies on Surface Water ........................................................................ 85 6.7 Conclusion .................................................................................................................... 87 Chapter 7: Simulated Environmental Conditions: Effect of Inorganic Ions and Multiple Loading Tests ...............................................................................................................................89 7.1 Introduction ................................................................................................................... 89 7.2 Impact of Background NOM Fractions and Sulphate Ions on MCLR Uptake ............. 90 7.3 Simulated Environmental Conditions: Effect of Inorganic Ions................................... 93 7.4 Impact of Inorganic Ions on MCLR Uptake in Presence of Low-Molecular Weight Organics .................................................................................................................................... 96 7.5 Impact of Inorganic Ions on MCLR Uptake in Presence if Higher-Molecular Weight Organics .................................................................................................................................... 99 7.6 Multiple Loading Tests ............................................................................................... 102 x 7.7 MCLR Degradation in Brine ...................................................................................... 105 7.8 Conclusion .................................................................................................................. 105 Chapter 8: Conclusions and Future Work ..............................................................................107 8.1 Conclusions ................................................................................................................. 107 8.2 Recommendations for Future Work............................................................................ 108 Bibliography ...............................................................................................................................110 Appendices ..................................................................................................................................138 Appendix A VUV-IX Combined Treatment........................................................................... 138 Appendix B Determination of Rate Limiting Step ................................................................. 140 B.1 Pseudo Second Order Kinetics................................................................................ 140 B.2 Pseudo First Order Kinetics .................................................................................... 140 B.3 Biot Number Estimation ......................................................................................... 140 Appendix C Charge Density and Resin Porosity Estimation.................................................. 143 C.1 Charge Density Estimation ..................................................................................... 143 C.2 Resin Porosity Estimation ....................................................................................... 143 Appendix D Conference Presentations ................................................................................... 144 xi List of Tables Table 2.1 Methods to Detect Three of the Most Common Cyanotoxins in Drinking Water ........ 20 Table 3.1 Characteristics of Standard NOM Isolates.................................................................... 39 Table 4.1 Isotherm Constants and Thermodynamic Parameters .................................................. 51 Table 4.2 Kinetic Parameters of MCLR Uptake by Strongly Basic IX resins ............................. 55 Table 5.1 Adsorption Isotherm Constants and Thermodynamic Parameters for MCLR and Suwannee River NOM .................................................................................................................. 61 Table 5. 2 Kinetic Parameters of MCLR Uptake in Presence of Different NOM Fractions ........ 64 Table 6.1 Adsorption Isotherm Constants and Thermodynamic Parameters for MCLR and NOM Fractions ........................................................................................................................................ 73 Table 6.2 Kinetic Parameters of MCLR Uptake in Presence of Different NOM Fractions ......... 82 Table 6.3 Kinetic Parameters of MCLR Uptake in Presence of Surface Water ........................... 87 Table 7.1 IX Site Occupancy Under Different Water Matrix ....................................................... 92 Table 7.2 Influence of Various Concentrations of Anions on Uptake of MCLR and DOC at Resin Dosage of 200 mg/L...................................................................................................................... 95 Table 7.3 Influence of Various Concentrations of Anions on Uptake of MCLR in Presence of PLFA 3 mg C/L at Resin Dosage of 200 mg/L ............................................................................ 98 Table 7.4 Influence of Various Concentrations of Anions on Uptake of MCLR in Presence of PLFA 3 mg C/L at Resin Dosage of 200 mg/L .......................................................................... 100 Table C.1 Resin Porosimetry Results for IX Resin (Purolite A860) .......................................... 143 xii List of Figures Figure 2.1 The Cyanobacterial Neurotoxins, (A) Anatoxin-a, (B) Anatoxin-a (s), (C) Saxitoxin 11 Figure 2.2 Structure of icrocystins (A) General Structure, (B) Cylindrospermopsin, (C) Lipopolysaccharide Endotoxin, (D) Microcystin-LR ................................................................... 15 Figure 2.3 Toxic Amino Acids: (A) \u00CE\u00B2-N-methylamino-L-alanine (BMAA), and (B) L-2,4- diaminobutyric Acid ..................................................................................................................... 16 Figure 4.1 Stoichiometry of MCLR Removal by IX .................................................................... 44 Figure 4.2 Effect of Resin Dosage on MCLR Uptake at pH 7; Lines Indicate a Fit to the Freundlich Isotherm. ..................................................................................................................... 51 Figure 4.3 MCLR Removal Kinetics Under Different Initial MCLR Concentrations; Lines Indicate a Fit to the Pseudo-second-order Kinetic Model, Studies Performed at IX Dosage of 200 mg/L~ 1 mL/L at pH=7. ................................................................................................................ 52 Figure 4.4 Biot Number as a Function of MCLR Concentration at Resin Dosage of 200 mg/L (~1 mL/L) ...................................................................................................................................... 54 Figure 5.1 (a) Removal of MCLR (Co = 25 \u00C2\u00B5g/L) by A860 Resin After 24 h for Waters with Increasing DOC Concentrations and Resin Dosages (b) DOC Removal by A860 Resin After 24 h (c) Adsorption Isotherms of MCLR on A860 Resin at pH 7.0. .................................................... 61 Figure 5.2 (a) MCLR Removal Kinetics and NOM Uptake Under Different Background NOM Concentrations (b) DOC, UV254 Absorbance and MCLR Reduction after 20 Minutes of Contact Time at Dosage of 200mg/L (~ 1mL/L) ....................................................................................... 65 Figure 6. 1 Effect of Resin Dosage on (a) MCLR Removal at pH 7 and (b) DOC Removal (Co = 100 \u00C2\u00B5g MCLR/L; 3 mg C/L; 24 Hours of Contact Time) ............................................................ 72 xiii Figure 6.2 (a) Adsorption Isotherms of MCLR on A860 Resin at pH 7.0 (*Lines Indicate the Fit to a Freundlich isotherm), (b) LCOCD Results for Suwanee River NOM (SRNOM) and (c) Suwanee River Humic Acid (SRHA) at Different IX Dosages (200 mg/L and 500 mg/L .......... 74 Figure 6.3 (a) MCLR Uptake as a Function of Contact Time *(Co= 25 \u00C2\u00B5g MCLR/L, 200 mg/L Resin and DOC of 3mg C/L), (b) TOC and UVA Reduction for 60 Minutes of Operation......... 81 Figure 6.4 Effect of pH on (a) Uptake of MCLR (b) Uptake of DOC and Equivalent Chloride Release Studies at pH = 7 ............................................................................................................. 84 Figure 7. 1 Kinetics of MCLR Uptake in Presence of Different NOM Fractions (a) In Presence of 5 mg/L Sulphate Ions MCLR (b) Reduction in DOC and Sulphate Ion Concentration ........... 91 Figure 7. 2 MCLR Kinetics in Presence of SRNOM and Inorganic Ions .................................... 94 Figure 7.3 MCLR Uptake Kinetics in Presence of PLFA and Inorganic Ions ............................. 97 Figure 7.4 MCLR Uptake Kinetics in Presence of SRHA and Inorganic Ions........................... 101 Figure 7.5 MCLR, DOC and UV254 Absorbance Reduction for (a) Continuous Operation for 4000 BV and 1000 BV after Regeneration (b) 2000 BV of Operation (c) MCLR Degradation in Brine as a Function of Time........................................................................................................ 104 Figure A.1 MCLR Degradation by VUV at Different Dosages ................................................. 138 Figure A.2 MCLR Degradation by VUV (IX Pretreatment) ...................................................... 139 xiv List of Abbreviations AOM Algogenic Organic Matter Biot Biot Number BV Bed Volumes Ct Concentration at time t Co Concentration at time 0 Ce Concentration at equilibrium Da Dalton (g/mol) DBP Disinfection By-Products DIC Dissolved Inorganic Carbon DOC Dissolved Organic Carbon DON Dissolved Organic Nitrogen EL Eagle Lake eq Equivalents FA Fulvic Acid HA Humic Acid HPLC High performance Liquid Chromatography HPSEC High Performance Size Exclusion Chromatography hr Hour IC Ion Chromatography IX Ion Exchange k1 First Order Rate Constant xv k2 Second Order Rate Constant kIPD Intraparticle Diffusion Rate Constant Kf Freundlich Constant LC-OCD Liquid Chromatography with Organic Carbon Detection MCLR Microcystin-LR MIEX Magnetic Ion Exchange MW Molecular Weight NOM Natural Organic Matter n Freundlich parameter PL Pony Lake PLFA Pony Lake Fulvic Acid q Sorption Capacity qm Maximum Sorption Capacity qe Sorption Capacity at Equilibrium SPE Solid Phase Extraction SR Suwannee River SRHA Suwannee River Humic Acid SRNOM Suwannee River Natural Organic Matter SUVA Specific UV Absorbance TOC Total Organic Carbon t Time UV Ultraviolet xvi UV254 UV Absorbance at 254 nm %UVT Percentage UV Transmittance VAID Van Anda Improvement District WHO World Health Organization xvii Acknowledgements I would like to first thank my advisor, Prof. Madjid Mohseni, for his expert guidance and endless enthusiasm. Prof. Mohseni\u00E2\u0080\u0099s mentorship helped me to learn, explore and develop as a researcher and I cannot thank him enough for giving me the support and privilege to work independently. I also thank my committee members Prof. Benoit Barbeau and Prof. Pierre Berube. Prof. Barbeau provided valuable guidance and suggestions for my research work, inspiring me to seek greater perspectives in my research work and Prof. Berube has the distinction of having taught the best water treatment course I\u00E2\u0080\u0099ve ever taken. Special thanks are owed to my parents, Iti Dixit and Prasanna Kumar Dixit, who have supported me throughout my years of education. I also owe particular thanks to my aunt, Ms. Dominga Valerio and my uncle Hari Singh, for their endless love and care. It was with their support that I was able to learn freely, make new friends and cherish my Vancouver memories. Major thanks are due to my fellow labmates over the years, including Siddharth Bhartia, Pranav Chintalapati, Adrian Serrano, Sean McBeath, Ataollah Kheryendish, Dr. Reza Rezaei, John Bergese, Joerg Winter, Adel Hajimalyeri, Dr. Laith Furatian, Dr. Syed Zaki, Mike Cheung, Maggie Han, Kai Song, Dr. Mohammad Umar, Kiara Allen, Emily Forsee, Dr. Sue Satyro, Riley Witter and Francois St. Pierre. Special thanks to my friend Dr. Mehdi Bazri, for all his input, assistance and timely guidance for my career, current and future research work. I would also acknowledge Prof. Naoko Ellis for her mentorship and support for the environmental sustainability projects. xviii Outside of the lab, I owe thanks to my friends and fellow members of the Indian Graduate Student Association (aka fun) Committee, including Gaurav Goyal, Shubham Jain, Raghav Raghav Grover, Varun Rangu, Charu Datt, Lakshana Sreenivasan, Abhinav Ajaykumar, Aarya VC, Manish Vashishta, Manav Bharti, Payel Ganguly, Mohammed Farooq, Abhishek Gautam, Abhishek Agarwal, Akhil Kumar, the cricket and badminton team members, including David Lim, Prashanth Krishnamoorthy, Sudipta Mitra, Manu Modugil, Gurvinder Gill, Kevin Yeo, Somesh Daga and WIL India colleagues in Vancouver, including Olivia Allen, Rajat Jain, Lina Azeez, and Allison Matfin. I am also thankful to Ila Dwivedi for her moral support and friendship throughout this work. I would also like to sincerely acknowledge the IC-IMPACTS team who have assisted me in my time here, including Prof. Nemy Banthia, Helena Fehr, Sue Roppel, Ashish Mohan and Angela Reid. Special thanks to Keyvan Maleki for his guidance and support throughout the program. I would also acknowledge the efforts of other members of the RES\u00E2\u0080\u0099EAU-WaterNET Network, including Candace Cook, Megan Wood, Heidi Backous and Adrian On for their support during the experimental work. I should also acknowledge funding, primarily from RES\u00E2\u0080\u0099EAU-WaterNET and IC-IMPACTS NCE. I should also thank my grandparents Mr. Ramnik Kumar Dixit and Uma Dixit, and my family members, including Pahuni, Divyanshi, Poorva and Sandeep Diwan. I would also like to sincerely acknowledge my uncle Sarvanja Dwivedi for his constant support and valuable guidance. xix Dedication To my parents I dedicate this thesis. 1 Chapter 1: Background 1.1 Harmful Algal Blooms (HABs) Cyanobacteria, also known as blue-green algae, are an essential part of the aquatic food chain. Under certain favorable conditions, rapid multiplication of cyanobacteria can lead to harmful algal blooms (HAB). Eutrophic waters, containing chemical factors like phosphorous and nitrogen, are important in controlling the growth of cyanobacteria, and raw and treated sewage, agricultural run-offs, pesticides and animal wastes are important sources of such nutrients (Boyer, 2007; Carmichael et al., 2001; Carmichael and Boyer, 2016; Chorus and Bartram, 1999; El-Shehawy et al., 2012; Health Canada, 2016; Metcalf and Codd, 2014; Svrcek and Smith, 2004). Cyanobacterial blooms are particularly harmful due to the release of specific toxins known as cyanotoxins. Cyanobacterial toxins, which are released from algal biomass have become a serious environmental and health hazard due to their potential to contaminate drinking, recreational and agricultural water reserves. Consumption of water with elevated levels of cyanotoxins may result in several health complications such as gastroenteritis, and external exposure to such water may lead to skin irritation or an allergic response. This poses serious threat to many aquatic ecosystems throughout the world and across North America. Microcystin-LR (MCLR) is the most commonly encountered toxic microcystin variant worldwide. Though regarded primarily as hepatotoxins, which cause liver damage and dysfunction, these toxins can also cause widespread damage to the kidney, lung, heart, adrenal glands, and thymus. Heath Canada guidelines recommend a safe limit of 1.5 \u00C2\u00B5g/L in drinking water and a concentration of 20 \u00C2\u00B5g MCLR/L for safe recreational usage (Carmichael et al., 2001; Falconer, 1999; Falconer and Humpage, 2006, 1996; Graham et al., 2010; Health Canada, 2016). 2 1.2 Natural Organic Matter Natural Organic Matter (NOM) is a complex mixture of organic substances originating from the degradation of plants and animals, and is found in both natural surface and ground water. Bloom events also may elevate the organic content of water bodies by release of algogenic organic matter (AOM) after cell death (Henderson et al., 2010; Mergen et al., 2008; Sathya et al., 2015). NOM may pose a number of challenges to the drinking water supply and distribution. It is a precursor for disinfection by-products, it impacts the taste and odor of potable water, and it deteriorates the biological stability of water by increasing bacterial regrowth and biofilm formation within distribution systems. Additionally, it can reduce the efficiency of water treatment technologies, e.g., it causes membrane fouling, attenuates ultraviolet (UV) irradiation, and scavenges OH radicals (\u00E2\u0080\u00A2OH) during the advanced oxidation processes (AOPs) (Bazri et al., 2016b, 2012; He et al., 2011; Kabsch-Korbutowicz et al., 2008; Matilainen et al., 2002). 1.3 Treatment of Cyanotoxins Removal of cyanotoxins from water is a major challenge. Conventional water treatment methods such as coagulation, flocculation, sedimentation and filtration are only partly effective at addressing the challenge of cyanobacterial toxins due to their limited capability of removing extracellular toxins. Moreover, the management of toxic sludge increases the operational and maintenance costs for these systems. Treatment processes such as UV-based advanced oxidation, ozonation and chlorination can provide significant barriers against dissolved toxins. However, environmental factors, such as presence of NOM, inorganic ions, pH and the risk of incomplete oxidation of released intracellular toxins, make oxidation-based process challenging to operate during a toxic bloom event. Adsorption based processes have been shown to be promising 3 treatment options for the uptake of cyanotoxins from bloom laden waters. However, the challenges of designing an adsorption contactor for toxin removals in source waters with elevated NOM concentration remains a challenge. Therefore, processes that are effective for simultaneous removal of cyanotoxins and NOM are desirable (Dixon et al., 2011; Lee and Walker, 2011; Merel et al., 2009; Sathishkumar et al., 2010; Sharma et al., 2012; Wang et al., 2007). 1.4 Anion Exchange Process Ion exchange process is the reversible exchange of negatively charged counter ions (such as, Cl-, OH-, HCO3-), from the surface of polymeric resin to surrounding aqueous matrix. Anionic ion exchange (IX) resins are being increasingly used as cost-effective treatment for the removal of dissolved organic carbon (DOC). Application of IX process has received considerable attention because of its effectiveness, ease of operation, scale up or down capabilities, small footprint and relatively low cost. Under pH relevant to water treatment, microcystins are negatively charged. For example, the pKa value for MCLR has been estimated as 2.3 (Walker, 2014), which result from the presence of two negatively charged carboxylic groups and one positively charged amine group. Therefore, IX can potentially be used for bloom laden waters for simultaneous NOM and cyanotoxin removal. 1.5 Research Rationale Despite many studies that evaluated IX resins for NOM removal, little research has been conducted on the effectiveness of IX for cyanotoxin uptake from bloom laden waters. Considering that algae bloom waters commonly result in the co-occurrence of elevated microcystin and NOM concentrations, it would be of interest to understand the extent to which IX resins can be used to 4 remove microcystins resulting from a bloom. Further, the NOM characteristics (e.g., charge density, molecular weight distribution) and the background water matrix (e.g., presence of NO3-, SO42-) can influence the treated water quality and also the long-term performance of the treatment process. Therefore, an understanding of the microcystin removal capability and kinetics by IX, as well as regeneration efficiency of the IX resins will be very valuable for commercial applications. Assuming that the IX process is efficient for the sorption of microcystins, it is expected that the process would produce a toxic spent brine that shall be treated prior to its disposal in the environment. To date, no strategies have been proposed in the scientific literature to address this challenge as well. 1.6 Thesis Layout This thesis compiles the results achieved throughout the course of this research, and discusses them in the form of six chapters following introduction and experimental methods chapters. The objectives of this research were achieved through a series of well controlled experimental work carried out at the University of British Columbia (UBC) and Ecole Polytechnique de Montreal. The information gathered is presented as follows: Chapter 1: Presents a general background on research topic and concerning issues. Chapter 2: Provides a comprehensive literature review around cyanotoxins and the application of IX for simultaneous NOM and cyanotoxin removal. Research objectives and rationale are articulated and the significance of this research is addressed in this chapter. Chapter 3: Provides detailed description of the experimental plan and research methodology as well as all the assays and analytical techniques employed in this study. 5 Chapter 4: Contributes primarily towards the main objective by providing an understanding of the mechanism and kinetics of microcystin-LR by a strongly basic anion exchange resin. Chapter 5: Provides an in depth look into the simultaneous MCLR and NOM uptake by IX process and involves comparative studies with different natural water sources. Chapter 6: Studies the key NOM characteristics that influence the MCLR uptake by IX process and evaluates the rate-limiting steps during IX process by applying various mathematical and physical models. Chapter 7: Demonstrates the impact of background water matrix on IX performance by evaluating four natural water sources under various IX treatment conditions. Moreover, the impact of multiple loading cycles and resin regeneration was also investigated in this study. Chapter 8: Presents overall conclusions, summarizes the key findings of the work, and provides recommendations for future research. 6 Chapter 2: Literature Review 2.1 Harmful Algal Blooms (HABs) Algae are naturally-occurring components of fresh, marine and brackish waters worldwide (Elliott, 2012; Walker, 2014). These microscopic plants are an essential part of the aquatic food chain, and are largely benign. However, cyanobacteria, also known as blue-green algae, in large quantities can be quite harmful (Carmichael and Boyer, 2016; Chatziefthimiou et al., 2016; EPA, 2012; Falconer, 1999). These bacteria, which grow in nutrient rich waters, are not always visible at low concentrations. However, under certain favorable conditions, rapid multiplication of cyanobacteria can lead to harmful algal blooms (HAB) (Falconer and Humpage, 2005; He et al., 2016). Several factors, such as excess nutrients or unusually high water temperatures, can contribute to HABs in a freshwater ecosystem. Specifically, by-products of livestock waste, fertilizer runoffs and wastewater treatment plant discharges containing nitrogen and phosphorous, provide the excess nutrients required to result in a bloom (Health Canada, 2016; Reichwaldt and Ghadouani, 2012; Svrcek and Smith, 2004). Harmful algal blooms can have a significant impact on the environment and human health. Blooms produce offensive odors and result in reduced water clarity, contributing to a decrease in the recreational value of impacted reservoirs (Falconer, 1999). Cyanobacterial blooms are particularly harmful due to the release of specific toxins known as cyanotoxins. Consumption of water with elevated levels of cyanotoxins may result in several health complications such as gastroenteritis, and external exposure to such water may lead to skin irritation or an allergic response (Carmichael et al., 2001; Carmichael and Boyer, 2016; EPA, 2012). Furthermore, chronic exposure to 7 cyanobacterial toxins, either by consumption or by topical exposure, may result in liver or nervous system damage, or cancer (Falconer, 1991; Falconer and Humpage, 2006, 2001, 1996). In addition to affecting human health, the onset, and eventual collapse and decay of cyanobacterial blooms can have a significant impact on wildlife. Hypoxic, or low oxygen, conditions created by increased ammonia concentrations in water can increase fish kills (Carmichael et al., 2001; Carmichael and Boyer, 2016; EPA, 2012). 2.1.1 Factors Governing Cyanobacterial Growth and Development of Blooms The formation of cyanobacterial blooms occurs when specific conditions permit the rapid proliferation of bacterial cells. Several physical, chemical and biological factors govern these conditions, and the bloom occurrence can be influenced by both, natural and human activities (Catherine et al., 2013; Izaguirre et al., 2007). Eutrophic waters, containing chemical factors like phosphorous and nitrogen, are important in controlling the growth of cyanobacteria, and raw and treated sewage, agricultural run-offs, pesticides and animal wastes are an important source of such nutrients (Downing et al., 2001; Elliott, 2012; Merel et al., 2013; Reichwaldt and Ghadouani, 2012). Iron, due to its direct involvement in photosynthesis and nitrogen fixation, has also been found to be an essential micronutrient for regulating cyanobacterial bloom development (He et al., 2016; Molot et al., 2014; Sevilla et al., 2010). Alkalinity and pH play a key role in determining the speciation of inorganic carbon. Together with oxygen, carbonate, bicarbonate and carbon dioxide affect the mobilization of nutrients (Sinha et al., 2012). Consequently, cyanobacteria are found more commonly in alkaline waters, because of their tendency to use bicarbonate as a source of inorganic carbon (De Maagd et al., 1999; Kotak and Zurawell, 2007). 8 In addition to the chemical composition, the hydrodynamic properties of water are also critical to the development of algal blooms (Dupuis and Hann, 2009; Salmaso and Cerasino, 2012). Slow moving or still waters are more favorable for bloom development and stability, in contrast to waters with faster mixing and greater flow (Bullerjahn et al., 2016; Falconer, 1999; Krienitz et al., 2013; Reichwaldt and Ghadouani, 2012; Svrcek and Smith, 2004). Furthermore, climatic conditions like temperature and rainfall pattern can have a great impact on cyanobacterial growth rates (Carmichael and Boyer, 2016; Krienitz et al., 2013). Rising global temperatures may be contributing to an increase in the frequency of toxic blooms, along with their expansion into more temperate ecosystems. Strong winds can also influence bloom development by promoting the accumulation of cyanobacterial populations near shorelines accumulation (Carmichael and Boyer, 2016; Chorus and Bartram, 1999). The biological environment of cyanobacteria can also impact their proliferation, and therefore bloom frequency. Aquatic plants compete with the phytoplankton cyanobacteria for nutrients and light, affecting bloom occurrences. Zooplanktons, which primarily feed on phytoplankton, are not able to easily digest cyanobacteria, thereby impacting cyanobacterial population size (de Bernardi and Giussani, 1990; Kotak and Zurawell, 2007; Svrcek and Smith, 2004). Many cyanobacteria also possess photo-adaptive characteristics and can outcompete other phytoplanktons for light in deeper and more turbid waters (de Bernardi and Giussani, 1990). Zebra mussels also actively graze on phytoplankton and selectively reject toxic cyanobacteria, providing toxic cyanobacteria with a competitive advantage, and resulting in an increased population of toxic variants (EPA, 2012; Fishman et al., 2010; Health Canada, 2016; Ho and Michalak, 2015; Kotak and Zurawell, 2007; Paerl and Paul, 2012; Svrcek and Smith, 2004). 9 2.2 Cyanobacterial Toxins Cyanobacteria have the potential to produce a wide variety of compounds that are toxic to mammalian systems (Health Canada, 2016; Kotak and Zurawell, 2007). Cyanobacterial toxins, which are released as the cells die, can remain in the water for up to three weeks after the algal bloom disappears. Although many such toxic compounds have been identified, an understanding of the complete range of cyanotoxins, of their prevalence, and of their impact on human health, is still incomplete (Carmichael and Boyer, 2016; EPA, 2012; Meriluoto and Codd, 2005; Metcalf and Codd, 2014). In general, cyanotoxins are divided into different classes based on their primary mode of action in mammalian systems. 2.2.1 Neurotoxins Cyanobacterial neurotoxins disrupt the normal propagation of nerve impulses from neurons to muscles, thereby causing paralysis and eventually death due to respiratory failure following ingestion (Banack et al., 2014; A. M\u00C3\u00A9jean et al., 2014). Of these neurotoxins, one of the most acutely toxic sub-classes is that of anatoxins (Funari and Testai, 2008). Anatoxin-a, an alkaloid secondary amine, binds to acetylcholine receptors at the junctions between nerves and muscle tissues, called synapses. Its toxicity arises from its ability to mimic acetylcholine, the neurotransmitter responsible for the neuromuscular stimulation. Additionally, Anatoxin-a cannot be broken down by acetylcholinesterase, the enzyme responsible for breaking down acetylcholine in the neuromuscular junction. This results in the over-stimulation of muscles regulated by cholinergic input, including those responsible for breathing. Drastic consequences of anatoxin-a\u00E2\u0080\u0099s evasion of enzymatic degradation include fatigue, and ultimately death via respiratory failure (Annick M\u00C3\u00A9jean et al., 2014; Osswald et al., 2013; Parsons et al., 2000). 10 Anatoxin-a(s) is a natural organophosphate compound, which, in contrast anatoxin-a, binds to and inhibits the acetylcholinesterase enzyme. The inhibition of acetylcholinesterase also results in the over-stimulation of muscle tissue by acetylcholine, and consequent respiratory failure. Both anatoxin-a and \u00E2\u0080\u0093a(s) are only produced by cyanobacteria in freshwater environments (D\u00C3\u00B6rr et al., 2010; Annick M\u00C3\u00A9jean et al., 2014). The paralytic shellfish poisons (PSPs), saxitoxin and neo-saxition are common components of the red tides in marine environments, but are produced only by cyanobacteria in freshwater. The saxitoxins are a group of twenty structurally related compounds, which are responsible for the periodic halts in shellfish harvesting worldwide and human fatalities following contaminated shellfish consumption (Etheridge, 2010; Wang et al., 2016). Saxitoxins act by blocking sodium channels in nerves, suppressing the propagation of nerve impulses (Codd et al., 1999; Wiese et al., 2010). In contrast to anatoxin-a and anatoxin-a(s), saitoxin activity ultimately leads to paralysis and death by preventing muscle stimulation altogether (Bragg et al., 2015; O\u00E2\u0080\u0099Neill et al., 2016). The toxic effects of anatoxin-a, anatoxin-a(s) and saxitoxin, expressed as LD50 in mice, manifest at very low concentrations 200, 20 and 10 \u00C2\u00B5g/kg, respectively, indicating their extremely high potency. Few studies have evaluated the frequency of cyanobacterial neurotoxins in Canadian freshwaters and indicate that neurotoxins may be rare in Canadian freshwaters (Downing et al., 2001; Health Canada, 2016; O\u00E2\u0080\u0099Neill et al., 2016; Parsons et al., 2000). 11 . Figure 2.1 The Cyanobacterial Neurotoxins, (A) Anatoxin-a, (B) Anatoxin-a (s), (C) Saxitoxin 2.2.2 Hepatotoxins The most commonly reported cyanobacterial hepatotoxins in freshwater are microcystins (MCs), nodularin, and cylindrospermopsin (Chorus and Bartram, 1999; Funari and Testai, 2008; Kotak and Zurawell, 2007). Though regarded primarily as hepatotoxins, which cause liver damage and dysfunction, these toxins can also cause widespread damage to the kidney, lung, heart, adrenal glands, and thymus (in the case of cylindrospermopsin). Cylindrospermopsin is a toxic alkaloid characterized by a tricyclic ring structure. It is a hepatotoxin that also demonstrates cytotoxic effects in the thymus, heart and gastrointestinal tract. Cylindrospermopsin is more commonly encountered in the tropical and sub-tropical climates of Australia and in Florida (Machado et al., 2017; Moreira et al., 2017; Rzymski and Poniedzia\u00C5\u0082ek, A B C 12 2014). No Canadian cyanobacterial strains have yet been found to possess the ability to produce cylindrospermopsin (Health Canada, 2016). As a result, cylindrospermopsin has not received much attention in Canada (Health Canada, 2016; Kotak and Zurawell, 2007; Rzymski and Poniedzia\u00C5\u0082ek, 2014). Microcystins are the most commonly occurring hapatotoxins, and have been extensively studied across the globe. They are cyclic, seven amino acid-containing peptides, with over 70 variants having been identified worldwide (Figure 2.2) (Health Canada, 2016; Kondo et al., 1992; Pavagadhi et al., 2013). The variations in structure arise through amino acid substitutions at positions two and four of the cyclic peptide ring, which heavily influence the toxicity of these microcystin analogues (the LD50 values can vary from 50 to 1200 \u00C2\u00B5g/kg in mice) (Chen et al., 2016; Graham et al., 2010; Health Canada, 2016). One of the most toxic microcystin variants is microcystin LR (MCLR), which contains leucine (L) and arginine (R) at positions two and four of the peptide structure (Chen et al., 2016; Health Canada, 2016; Kotak and Zurawell, 2007; Woolbright et al., 2017). Death and tissue damage due to microcystins occurs in a dose dependent manner. At higher doses, microcystins cause irreversible damage to mammalian liver by inhibiting protein phosphatase 1 and protein phosphatase 2A, which causes numerous ultrastructural changes that ultimately result in cell death (Boyer, 2007; Carmichael, 1992). As cells die, blood pools in the liver, eventually doubling organ weight. Microcystins can also cause gastrointestinal and kidney damage, and chronic exposure to these toxins eventually results in hepatomas (Chen et al., 2016; Graham et al., 2010; Health Canada, 2016). In addition, while dermal or eye exposure causes irritation, accidental ingestion of water through activities like bathing, swimming, water sports and diving can be even more detrimental under high toxin concentrations (Chen et al., 2016; Metcalf and Codd, 2014; Nagata 13 et al., 1995; Pereira et al., 2004). Microcystins can also cause gastrointestinal and kidney damage, and chronic exposure to these toxins eventually results in hepatomas (Chen et al., 2016; Graham et al., 2010; Health Canada, 2016). 2.2.3 Lipopolysaccharides Lipopolysaccharide (LPS) toxin production is a standard characteristic of Gram negative bacteria, including coliforms and cyanobacteria. A cyanobacterial bloom may result in high LPS concentrations in raw water, increasing the risk associated with recreational activities in these waters (Carmichael, 1992; Codd et al., 1999; Downing et al., 2001; Funari and Testai, 2008). LPS exposure and ingestion cause the body to release pyrogenic compounds, which result in fever, diarrhea, vomiting, and hypotension (Mayer et al., 1985). 2.2.4 Dermatotoxins Aplysiatoxins and lyngbyatoxins, which are produced by certain marine and freshwater cyanobacteria cause dermatitis among people who come in contact with the cyanobacterial filaments (Lajeunesse et al., 2012). These toxins are addressed in the Guidelines for Canadian Recreational Water Quality due to their nature as human allergens (Chorus and Bartram, 1999; Health Canada, 2016). However, dermatotoxins are generally not considered harmful when ingested with drinking water. 14 A B 5 4 3 2 1 7 C 6 15 Figure 2.2 Structure of icrocystins (A) General Structure, (B) Cylindrospermopsin, (C) Lipopolysaccharide Endotoxin, (D) Microcystin-LR * In figure 2.2 (A): X= Leucine (L), arginine (R), tyrosine (Y); Z= arginine (R), alanine (A), methionine (M); Side chain nomenclature- 1: (D-Ala-), 2: L-X-, 3: D-erythro-p-methyl-Asp-, 4: L-Z, 5: ADDA-, 6: D-Glu, 7: N-Methyldehydro-Ala. 2.2.5 Toxic Amino Acids Like plants, algae, and fungi, cyanobacteria produce a wide range of amino acids. \u00CE\u00B2-N-methylamino-L-alanine (BMAA) is an amino acid produced by free-living cyanobacteria, which is acutely toxic to a range of aquatic animals including protozoa, brine shrimp and fish larvae (Figure 2.3) (Brand et al., 2010; Cox et al., 2003). In humans, BMMA is associated with the neurological disease amyotrophic lateral sclerosis (ALS) (Nunn and Ponnusamy, 2009). BMMA is often produced in conjunction with 2-,4-diamminobutyric acid (DAB) (Figure 2.3). Due to its D 16 elemental similarities with BMMA, DAB is often associated with neurotoxicity (Health Canada, 2016). Figure 2.3 Toxic Amino Acids: (A) \u00CE\u00B2-N-methylamino-L-alanine (BMAA), and (B) L-2,4- diaminobutyric Acid 2.3 Cyanotoxin Detection and Quantification Cyanobacterial blooms are mixtures of heterogeneous species, each of which may or may not produce toxins. The biological complexity of blooms makes identifying and quantifying specific cyanotoxins in drinking water very difficult (Merel et al., 2013; Nicholson and Burch, 2001). Moreover, there is no single method available to assess different toxin variants in a sample of bloom water (Westrick, 2010). Cell counts, biomass measurements and chlorophyll-a measurements have traditionally been utilized to monitor cyanobacterial cells (Chen et al., 2011; Oehrle et al., 2010). However, no direct relationship has been observed between the number of bacterial cells and cyanotoxins levels in surface water studies (He et al., 2016). A B 17 2.3.1 Physicochemical Methods Chromatographic separation is one of the most widely used techniques when it comes to analyzing microcystins, nodularins, anatoxins, cylindrospermopsins and saxitoxins in commercial and research laboratories (Bragg et al., 2015; Oehrle et al., 2010). While most cyanotoxins can be analyzed with high performance liquid chromatography (HPLC), run times of about one hour have been reported for some compounds (Spoof et al., 2010). The introduction of ultra-high performance liquid chromatography (UPLC) in mid-2000s, has allowed the same compounds to be analyzed in approximately 5-10 minutes (Boua\u00C4\u00B1\u00CC\u0088cha et al., 2002; Fayad et al., 2015; McElhiney and Lawton, 2005). Generally, these LC systems, when calibrated against known standard cyanotoxins, exhibit good UV-absorbance characteristics, providing an excellent avenue for cyanotoxin assessment (ISO, 2005). However, compounds like saxitoxins and BMAA cannot be observed by conventional UV detectors, and must be chemically modified by fluorescent tags for detection through LC (D\u00C3\u00B6rr et al., 2010; He et al., 2016; Oehrle et al., 2010). . Another highly sensitive physicochemical method for cyanotoxin detection is mass spectrometry (MS). With the exception of LPS, all known cyanotoxins can be identified and detected through MS (Zheng et al., 2016). 2.3.2 Biochemical Assays Several biochemical assays have been developed for the identification of microcystins in water. The two most commonly used assays are the enzyme linked immunosorbent assay (ELISA) and protein phosphatase inhibition assay (PPIA) (An and Carmichael, 1994; Kim et al., 2003). ELISAs have proven to be highly sensitive for microcystin detection, but have shown variable reactivity towards different microcystin analogues. Consequently, this assay is often unable to distinguish 18 between individual variants, and obtained concentration values are reported as MCLR equivalents (An and Carmichael, 1994; Aranda-Rodriguez et al., 2003; Kim et al., 2003; Kondo et al., 2000; Liu et al., 2014). Furthermore, low pH and naturally occurring water contaminants like natural organic matter (NOM), metals and salts interfere with the results obtained through ELISA. However, the potential for interference varies depending upon the type of ELISA used (McElhiney and Lawton, 2005; Nagata et al., 1995; Weller, 2013). PPIAs measure the phosphate released from phosphorylated proteins in the presence of a phosphatase enzyme preparation and inhibitors like microcystins. Due to the commercial availability of enzymes, PPIAs are rapid, sensitive and readily accessible (Heresztyn, 2001). However, like ELISA, they do not exhibit the same sensitivity for all microcystins (An and Carmichael, 1994). The detection limit of total microcystins, reported as MCLR equivalents by PPIA, is approximately 0.1 \u00C2\u00B5g/L or less. However, PPIAs are significantly affected by a variety of matrix interferences from iron, manganese and other protein phosphatase inhibitors (McElhiney and Lawton, 2005). 2.3.3 Field Test Kits Field test kits allow the rapid determination of microcystin presence or absence in a sample. These kits are often used as tools to assess the toxicity of a bloom and therefore, to optimize treatment plant operations. Commercially available test kits can generally measure between 1 to 5 \u00C2\u00B5g/L of microcystins and use a variety of methods including immunochromatography, ELISA and protein phosphatase inhibition for detection (Izydorczyk et al., 2009; McElhiney and Lawton, 2005). Intracellular toxins must be released with an extra lysing agent in order to determine the level of total microcystins (Aranda-Rodriguez et al., 2015). Unfortunately, these lysing agents may 19 interfere with the test method and provide incorrect readings (Fastner et al., 1998; Nicholson and Burch, 2001; Zhao et al., 2013). 2.3.4 Quality Assurance and Control There are only two standardized methods for the quantitative analysis of microcystins in drinking water: ISO method 20179, which utilizes SPE-HPLC-UV; and EPA Method 544, which uses SPE LC-MS/MS (ISO, 2005; Shoemaker, 2015). These methods are not widely used in commercial laboratories in Canada, making it difficult for responsible authorities to have confidence in the resulting data. In order to address this issue, quality assurance (QA) and quality control (QC) protocols and standard operating procedures have been developed for various methods of microcystin analysis, such as HPLC-PDA and ELISA (Health Canada, 2016). 20 Table 2.1 Methods to Detect Three of the Most Common Cyanotoxins in Drinking Water Methods Anatoxin Cylindrospemopsin Microcystin-LR Biological Assays Mouse Yes Yes Yes Enzyme Linked Immunosorbent Assays (ELISA) No No No Protein Phosphatase Inhibition Assays (PPIA) No Yes No Gas Chromatography Gas Chromatography with Mass Spectrometry (GC-MS) Yes No No Liquid Chromatography Liquid Chromatography/ Ultraviolet-Visible (LC/UV or HPLC) Yes Yes Yes Liquid Chromatography/ Fluorescence (LC/FL) Yes No No Liquid Chromatography Triple Quadrupole Mass Spectrometry (LC/MS/MS) Yes Yes Yes 21 2.4 Microcystins Microcystins are cyclic peptide, hepatotoxic cyanobacterial metabolites, produced primarily by the microcystis genus in North America. Their resilience to chemical and biological breakdown, prevalence, and potential to reach high concentrations in blooms and scums, makes them arguably the most important of all freshwater cyanotoxins (Chen et al., 2016; Meriluoto and Codd, 2005). As inhibitors of protein phosphatase activity in the liver, microcystins progressively cause cell collapse and separation, tissue damage, organ failure, and death. Microcystin-LR (MCLR) is the most commonly encountered toxic microcystin variant worldwide. Other variants, including MCLA, MCRR and MCYR, occasionally dominate or co-occur with MCLR during bloom events (Meriluoto and Codd, 2005; Svrcek and Smith, 2004). 2.4.1 Guidelines for Water Usage Different jurisdictions have set various, yet stringent, limits and guidelines for the concentration of microcystins in drinking water. Heath Canada guidelines recommend a safe limit of 1.5 \u00C2\u00B5g/L in drinking water while the Australian National Health and Medical Research Council has established a guideline in 2001 for total microcystin concentration of 1.3 \u00C2\u00B5g/L expressed as MCLR toxicity equivalents (Health Canada, 2016; Brenton C Nicholson and Burch, 2001). WHO established a provisional guideline value of 1 \u00C2\u00B5g/L for total MCLR in drinking water. Based on WHO risk categories for recreational contact, microcystin concentrations <10 \u00C2\u00B5g/L represent a low probability of adverse health effects, while 10 \u00C2\u00B5g/L to 20 \u00C2\u00B5g/L indicate a moderate risk, and >20 \u00C2\u00B5g/L represents a high risk (Chorus and Bartram, 1999; EPA, 2012; Health Canada, 2016). The Health Canada Guidelines recommend a concentration of 20 \u00C2\u00B5g MCLR/L for safe recreational usage; whereas, 20 \u00C2\u00B5g MCLR/L is considered under \u00E2\u0080\u0098Danger: Tier II\u00E2\u0080\u0099 in the state of California and 22 a \u00E2\u0080\u0098Level 3: Serious Risk\u00E2\u0080\u0099 with health alert in the state of Indiana and Nebraska (EPA, 2012; Health Canada, 2012). 2.4.2 Geographical Distribution in Canada Studies on occurrence of microcystins in Canada began in the late 1950s with the successful isolation and culture of Microcystis aeruginosa from blooms samples from the Little Rideau Canal in Ontario. The isolated compound, composed of a stable cyclic peptide group composed of seven amino acids, was originally called Fast Death Factor (FDF). This was later included in a group called \u00E2\u0080\u009Cmicrocystins\u00E2\u0080\u009D following the discovery of similar structures in the early 1970s (Health Canada, 2016; Kotak and Zurawell, 2007). Since the discovery of microcystins, studies have shown that these compounds are highly prevalent in Canada. In particular, small and shallow lakes that are the source of drinking water for many small and rural communities have shown incidences of elevated levels of algal toxins (Bullerjahn et al., 2016; Carmichael and Boyer, 2016; Kotak and Zurawell, 2007). The lakes in the interior region of British Columbia and water bodies around the coastal areas of the province have also indicated high toxin concentrations (Andersen et al., 1993). In 2011-2012, an algal bloom on an inland lake in British Columbia was reported to have an MCYR concentration of 25 \u00C2\u00B5g/L and an MCLR concentration of 6 \u00C2\u00B5g/L, affecting two municipal drinking water uptakes. The bloom continued into 2012, with intake concentrations rising to 62 \u00C2\u00B5g/L for MCYR and 7 \u00C2\u00B5g/L for MCLR. However, the total amount of microcystins in the treated water was below the B.C. drinking water guideline of 1.5 \u00C2\u00B5g/L (BC Ministry of Health, 2012; Health Canada, 2016). In 2005, Alberta initiated a routine microcystin monitoring program for lakes and reservoirs. Consequently, high concentrations (> 25 \u00C2\u00B5g MCLR/L) are reported in more than 100 reservoirs 23 across the province of Alberta (Alberta Environment, 2012; Kotak and Zurawell, 2007). Similarly, more than twenty-five euphoric bodies in Manitoba, including slow flowing prairie rivers, have also shown elevated toxin concentrations (Manitoba Water Stewardship, 2012; Orihel et al., 2012). In contrast, microcystins in the drinking water reserves of Saskatchewan have not been extensively monitored or well reported (Saskatchewan Environment, 2012). However, microcystins have been reported in bloom samples affecting recreational lakes in the province (Andersen et al., 1993). For instance, recurrent cyanobacterial blooms have been observed on a lake sourcing drinking water treatment plants in Saskatchewan. MCLR, specifically, has been detected in Lake Ontario and Lake of the Woods, Ontario, since the mid-2000s. However, a vast majority of the sites reported toxin concentrations below 20 \u00C2\u00B5g/L (Ontario Ministry of Environment, 2014, 2012; Winter et al., 2011). In Quebec, cyanobacterial toxin concentrations have been documented by water treatment plants since the early 2000s. Cyanobacterial cell counts in raw water samples often reported concentrations above 20000 cells/mL, with MCLR concentrations of 6 \u00C2\u00B5g/L. In addition to microcystins, presence of anatoxins (3 \u00C2\u00B5g/L) were also reported in water sources. Studies have indicated that approximately 60 % of the Quebec water treatment plants could cope with high MCLR concentrations (60 g/L); however, increased anatoxin-a concentrations would be difficult to deal with due to its high resistance towards conventional treatment (Carri\u00C3\u00A8re et al., 2010; Minist\u00C3\u00A8re du D\u00C3\u00A9veloppement durable de l\u00E2\u0080\u0099Environnement et des Parcs du Qu\u00C3\u00A9bec, 2012; Zamyadi et al., 2012). 2.4.3 Environmental Stability Microcystins are environmentally stable, their toxicity persisting between a few days to a few weeks after bloom disappearance. Because their chemical structure is stable in water, microcystins 24 are resistant to chemical hydrolysis and oxidation, and can remain for more than three weeks following bloom treatment with copper sulphate algaecide (Jones, 1994). Furthermore, these toxins cannot be destroyed by boiling (Smith et al., 2008). There are some methods of eliminating microcystins from water. Approximately 77% of these toxins can be removed from the reservoir surface via photo-degradation (Tsuji et al., 1995). However, this procedure requires approximately 22 days. Moreover, this technique is primarily useful in shallow water systems, and increasing the depth of coverage to 1m and 4 m reduces toxin degradation to 24% and 9%, respectively (Health Canada, 2016). Temperature also impacts the removal of these toxins from water. The degradation of MCLR and MCRR is highest at 30\u00C2\u00B0C, and occurs fairly rapidly between 10\u00C2\u00B0C and 30\u00C2\u00B0C. Within four days, the toxin concentrations can decline from 20 mg/L to below detection limits (< 0.1 \u00C2\u00B5g/L), in the presence of microcystin degrading bacteria (Genus: Sphingomonas) (Park et al., 2001). At temperatures below 10\u00C2\u00B0C, degradation is slow, and only about 30% of the toxin is degraded at 5\u00C2\u00B0C after a seven days (Falconer, 1999; Kenefick et al., 1993). This stability of microcystins in aqueous matrices necessitates adopting appropriate water treatment technologies in order to provide safer and cleaner drinking water. 2.5 Organic Matter in Surface Water 2.5.1 Natural Organic Matter (NOM) Natural Organic Matter (NOM), is a complex mixture of various organic substances found in surface and ground water sources. NOM is chemically heterogeneous, comprised of hydrophobic, hydrophilic and neutral fractions with a wide range of molecular weights and chemical characteristics, and it majorly originates from dead and decaying plant and animal bodies (Bazri et al., 2016a; Graf et al., 2014; Kennedy et al., 2005; Westerhoff et al., 1999). The presence of 25 NOM brings many potential issues with respect to water quality and treatment processes. As a precursor of disinfection by-products, it significantly impacts the taste and odor of potable water. Moreover, it deteriorates the biological stability of water by increasing bacterial regrowth and biofilm formation within distribution systems. Additionally, it can reduce the efficiency of advanced oxidation processes (AOPs) for disinfection and micro-pollutant degradation. Evidence of membrane fouling and scavenging of hydroxyl radicals (\u00E2\u0080\u00A2OH) during UV-based oxidation processes have also been documented (Bolto et al., 2004; Swietlik et al., 2004). Several methods of characterization using UV spectroscopy, size exclusion chromatography (SEC), chemical fractionation, and excited emission matrix (EEM) fluorescent spectroscopy have been developed to study NOM (Bazri and Mohseni, 2016; Cornelissen et al., 2008). However, because these methods monitor individual properties of NOM, they cannot be considered comprehensive. Major parameters used to quantify the amount of organic carbon (dissolved in water) associated with NOM are total and dissolved organic carbon (TOC, DOC) and UV absorbance at 254 nm (UV254), which measures the amount of conjugated carbon-carbon double bonds within the structure (Allpike et al., 2007, 2005; Bazri et al., 2012; Huber et al., 2011; Pelekani et al., 1999; Sarathy et al., 2011). 2.5.2 Organic Matter from Cyanobacterial Cells Bloom events often elevate the total organic carbon (TOC) levels of water bodies. Algogenic organic matter (AOM), which consists of intracellular organic matter (IOM) released from algal cell lysis and extracellular organic matter (EOM), and natural organic matter (NOM) together cause a number of issues in potable water treatment (Henderson et al., 2010; Wert and Rosario-26 Ortiz, 2013). They adversely affect the aesthetic quality of water and the performance of water treatment processes (Cornelissen et al., 2008; Newcombe , Morrison, 2002). AOM shows a higher nitrogen content than NOM of terrigenous origin due to its proteinaceous origin. The Total Organic Nitrogen (TON) to Total Organic Carbon (TON/TOC) ratios follow the order of algal cells = IOM>EOM>NOM. AOM, which consists of more hydrophilic and less aromatic carbons, and low molecular weight compounds, is more biodegradable than NOM compounds (Health Canada, 2016; Henderson et al., 2010; Wert and Rosario-Ortiz, 2013). AOM characteristics also change as the algal bloom passes from the growth phase to the death phase. EOM is primarily composed of compounds of lower molecular weight, such as glycolic and amino acids, and is released during the exponential growth phase. IOM is mostly composed of higher molecular weight compounds like polysaccharides and is mostly released during the death phase (Wert and Rosario-Ortiz, 2013). Bloom events often necessitate the removal of NOM, AOM, IOM and EOM, since these compounds can react readily with chlorine during water disinfection processes and form suspected carcinogenic halogenated by-products (Ritchie and Perdue, 2003). Various chemical and physical treatment processes have been proposed to address the issue of high DOC in aqueous matrices. Processes such as ozonation and UV based oxidation processes are known to be effective, but these are often very expensive and lead to the partial removal of organics and formation of biodegradable intermediates. These issues, hence, necessitate the need for a pre-treatment process to improve water treatment operations (Coral et al., 2013; Sharma et al., 2012; Westerhoff et al., 1999; Zamyadi et al., 2015). Physical processes such as filtration, coagulation/flocculation and ion exchange (IX) have gained attention in this regard. Of these processes, IX has demonstrated great potential as a pre-treatment process for high DOC waters and has documented a significant reduction in the disinfection by-product formation potential and 27 membrane fouling rates when applied prior to coagulation or activated carbon (Bazri and Mohseni, 2016; Bolto et al., 2004; Coral et al., 2013). 2.6 Treatment of Cyanotoxins A variety of chemical, physical, and biological treatment methods can be employed to remove the intracellular (cell-bound) and extracellular (dissolved) cyanotoxins in water treatment plants. A brief discussion of the methods that may be effective in this regard are discussed below (He et al., 2016; Health Canada, 2016; Svrcek and Smith, 2004). A multi-barrier approach is required to remove both the intracellular and extracellular microcystins from drinking water (Lawton et al., 1998). Conventional filtration with coagulation, clarification and granular filtration can be effective in removing cyanobacterial cells and intracellular toxins (Camacho et al., 2015; Dixon et al., 2011; Henderson et al., 2010). However, accumulation of cyanobacterial cells in clarifiers can lead to microcystin breakthrough into filtered waters (Park et al., 2001; Zamyadi et al., 2015, 2013, 2012). The efficiency of conventional filtration processes for toxin removal varies from 60% to 99.9% and depends upon several factors, such as cyanobacterial species and cell density, coagulant dosage, pH, flocculation time, filter backwash and clarified sludge removal. Further, peroxidation can cause cell lysis and increase concentrations of dissolved microcystins in the treatment plant. Similarly, cell lysis can also be observed in the clarifier sludge (Carri\u00C3\u00A8re et al., 2010; Coral et al., 2013). Consequently, conventional filtration is generally considered to have limited effectiveness for the removal of extracellular microcystins and additional processes such as chemical oxidation, reverse osmosis and adsorption are required (Chorus and Bartram, 1999; Coral et al., 2013; Sathya et al., 2015; Zamyadi et al., 2015). 28 A variety of chemical oxidants including chlorine, ozone and permanganates can be effective for the destruction of dissolved microcystins. The effectiveness of chemical oxidation depends upon several factors such as oxidant type and dosage, solution pH, oxidant contact time, and presence of NOM (Acero et al., 2008; Rodr\u00C4\u00B1, 2007; Sordo et al., 2007). The general trend of effectiveness for toxin degradation is as follows: ozone > permanganate > chlorine and chlorine based oxidants (Acero et al., 2008, 2005; Zamyadi et al., 2015). Although chemical oxidation can be effective at removing microcystins, the concentration of dissolved microcystins can increase following oxidation processes due to cell lysis, resulting in an increased concentration of dissolved toxins in drinking water. This can be addressed by applying processes such as conventional filtration to remove cyanobacterial cells prior to chemical oxidation (Al Momani and Jarrah, 2010; Rodr\u00C3\u00ADguez et al., 2007). UV irradiation/photolysis may be effective for the breakdown of microcystins, however, the efficiency of UV photolysis depends upon the exposure time, UV dose, absorption coefficient of the target microcystin, and surface water characteristics (He et al., 2011; Health Canada, 2016; Lee et al., 2004a, 2004b; Svrcek and Smith, 2004). Membrane based processes are another alternative to remove toxins from water. Microfiltration and ultrafiltration membranes are effective for the removal of cyanobacterial cells and intracellular toxins, with some reports suggesting removals of more than 98% for the cell bound toxins (Sorlini et al., 2013). However, these are less efficient at removing dissolved extracellular toxins. Nanofiltration (NF) and Reverse Osmosis (RO) can achieve high removal rates for both intracellular and extracellular toxins, but require cyanobacterial cells to be removed first to prevent membrane fouling. For all membrane based processes, size exclusion is the dominant rejection mechanism, and factors such as hydrophobicity, initial microcystin concentration, flux, recoveries, 29 molecular weight distribution of NOM and resulting membrane fouling impact the removal efficiency of these processes (Camacho et al., 2015; Kim and Dempsey, 2010; Meriluoto and Codd, 2005; Svrcek and Smith, 2004) Adsorption based processes have been shown to be promising treatment options for the uptake of cyanotoxins from bloom laden waters. Adsorbents such as iron oxide nanoparticles and carbon based adsorbents have been extensively investigated for adsorptive removal of MCLR and other related toxins (Huang et al., 2007; Lee and Walker, 2011; Teng et al., 2013). However, challenges of designing an adsorption contactor for toxin removals in source waters with elevated NOM concentration remains an issue for the adoption of this technology. NOM in aquatic environments can simultaneously compete with cyanotoxins for uptake sites on the surface of the adsorbent, thereby reducing the uptake efficiency. NOM concentration (mg/L) is several orders of magnitude higher than the concentration of cyanotoxins (\u00C2\u00B5g/L) and other micro-pollutants (ng/L). Therefore, NOM can dramatically reduce the uptake capacity of cyanotoxins; however, the latter is not expected to influence NOM uptake (Li et al., 2002; Pelekani and Snoeyink, 1999). Factors like microcystin variant, pH, and contact time affect the removal efficiency of activated carbon. Moreover, the raw water quality, specifically the presence of NOM and inorganic ions, and limited regeneration and reusability issues, limit the usage of activated carbon for commercial applications (Pelekani and Snoeyink, 1999; Sun et al., 2012; Verdickt et al., 2011). Therefore, processes that are effective for simultaneous removal of cyanotoxin and NOM with good regeneration efficiencies are desirable. Physical processes like anion exchange (IX) are being increasingly used as cost-effective treatment for the removal of dissolved organic carbon (DOC) (Boyer et al., 2008; Boyer and Singer, 2006, 2005; Treavor H. Boyer and Singer, 2008; Cornelissen et al., 2010; Grefte et al., 2013; Huang et 30 al., 2015; Humbert et al., 2005; Johnson and Singer, 2004; Kabsch-Korbutowicz et al., 2008) and can potentially be used for bloom laden waters for simultaneous NOM and cyanotoxin removal. Application of anionic ion exchange (IX) process has received considerable attention because of its effectiveness, ease of operation, scale up or down capabilities, small footprint and relatively low cost. 2.7 Anion Exchange Process The reversible exchange of negatively charged counter ions (such as, Cl-, OH-, HCO3- ), from the surface of polymeric resin to surrounding aqueous matrix is called anion exchange (IX) process. IX resins are being increasingly used as cost-effective treatment for the removal of dissolved organic carbon (DOC) (Huang et al., 2015; Kabsch-Korbutowicz et al., 2008). Under pH relevant to water treatment, microcystins are negatively charged and could potentially be removed via IX. For example, the pKa value for MCLR has been estimated as 2.2 (Walker, 2014), which results from the presence of two negatively charged carboxylic groups and one positively charged amine group. Therefore, depending upon the characteristics of microcystins and the background water matrix undergoing the IX treatment, microcystins can be removed effectively using IX. To the best of the author\u00E2\u0080\u0099s knowledge, no studies have been performed to evaluate the efficacy of IX process for the removal of microcystins from natural waters. 2.7.1 Anion Exchange for Organic Contaminants It has been well demonstrated that IX is capable of removing a broad spectrum of NOM components and DBP precursors (Cornelissen et al., 2008; Graf et al., 2014; Tan and Kilduff, 2007a; Verdickt et al., 2011). Properties such as polymeric matrix and degree of cross linking 31 (Acrylic or styrene), functional group properties, such as strongly basic (Quaternary ammonium Type I or II), or weakly basic (tertiary ammonium), ionic form (Cl-, OH-), ion exchange capacity, water content, and particle size play a key role in determining the applicability of IX resins for various purposes (Bazri et al., 2016b; Hu et al., 2014; Ishii and Boyer, 2011; Tan and Kilduff, 2007). The quaternary ammonium strongly basic resins (SBR) with acrylic backbone and smaller bead size are reported to show superior performance at NOM removal compared to other resins (Bazri et al., 2016a; Bolto et al., 2002b; Greluk and Hubicki, 2011). Literature studies have reported that Purolite A860, a strongly basic macroporous resin is highly effective in removing a major fraction of NOM components from surface water matrices in comparison to other commercially available resins (Bazri et al., 2016a). In terms of polymeric backbone, it has been reported that polystyrene resins are more selective towards smaller organics, <1 kilo Dalton (kDa), and aromatic moieties than acrylic species (Bazri and Mohseni, 2016; Bolto et al., 2002a). This behavior is governed by their electrostatic interactions and hydrophobic bonding. Macroporous resins with higher water content and more open structure have been found to increase the resin performance for smaller molecular weight NOM removal (Boari et al., 1974; Bolto et al., 2004; Huang et al., 2015; Lawton et al., 1998; Sun et al., 2015; Tan and Kilduff, 2007; Zhang et al., 2014). NOM properties such as molecular weight (MW), polarity (hydrophobicity), and charge density have been reported to significantly influence the treatment efficacy during the IX treatment. IX process has shown to be more effective towards removing low to medium molecular weight organics and is also reported to be a cheaper alternative to granular activated carbon (GAC) (Bazri et al., 2016b; Cornelissen et al., 2008; Huang et al., 2015; Kabsch-Korbutowicz et al., 2008) 32 However, higher molecular weight hydrophobic fractions are not removed by IX resins due to size exclusion. Removal of organics by IX resins can take place via two major mechanisms, namely, electrostatic exchange of counter ions (e.g., Cl-, OH-, HCO3-) and/or adsorption of ions on resin surface. However, studies indicate that uptake mechanism for DOC removal is dominated by ion exchange with only a small percentage of the DOC in the <1 kDa and >10 kDa fractions being removed by adsorption (Bazri et al., 2016b; Bazri and Mohseni, 2016). Further, IX resins poorly remove very large organic molecules (hydrophobic humic compounds), due to size exclusion. A similar reduction in uptake is observed for small molecular weight neutral fractions, and therefore, the molecular size distribution and charge density of the organic matter can play a significant role in competing with microcystins for uptake suites (Huber et al., 2011; Zhang et al., 2014). Regardless of the removal mechanism, kinetic studies on the uptake of organic molecules indicate pore diffusion to be the rate-limiting step at low resin concentrations. At low solute concentrations or under high resin dosage, film diffusion can also become significant. However, no studies have yet been performed to investigate the uptake mechanism of microcystins via IX resins. In addition to the presence of NOM, the composition of the aqueous matrices, specifically anionic species, are expected to impact the uptake and the sorption kinetics of the adsorbent (Bazri and Mohseni, 2016; Li and Sengupta, 2000). Based on their affinities, these anionic species would compete for a limited number of active sites on the resin and thereby, reduce the uptake of cyanotoxins under these conditions. Anion exchange resins are reported to have higher affinity towards sulfate compared to organic components and hence, the presence of high sulfate ion concentration adversely impacts the toxin removal by IX (Verdickt et al., 2011; Willison and Boyer, 2012). Nitrates are not strong competitors for organics, even though it has been reported 33 that at high concentrations they influence the removal of organic matter (Ates and Incetan, 2013; Kim et al., 2012). Literature findings also suggest that bicarbonates and chlorides have equal affinity for anion exchange resins and lower preference as compared to organic molecules (Bolto et al., 2002a; Hsu and Singer, 2010). Considering that algae bloom waters commonly result in the co-occurrence of elevated microcystin, inorganic ions and NOM concentrations, the knowledge about the impacts of NOM properties and inorganic species on the uptake of microcystins via IX is very limited. 2.7.2 Regeneration of IX Resins To maintain the treated water quality, resin regeneration is frequently required. This is done by restoring the depleted/occupied resin pores and sites which get saturated by organic contaminants over the course of treatment. Regeneration with brine (NaCl ~10 % W/V, a condition equivalent to an application of 0.1 kg NaCl/L of resin), is generally used to elute the ions adsorbed from the resins and is the most efficient way of restoring the capacity of anionic exchange resins for further cycles (Bazri and Mohseni, 2016; Bolto et al., 2004). Brine management after resin regeneration is one of the greatest challenges with the ion exchange process (Cornelissen et al., 2008; Verdickt et al., 2011). Presence of cyanotoxins along with the high salt and organic contaminants adds to the treatment and handling costs of the toxic brine. However, no studies have been performed to address the challenge of high microcystin concentrations in the regenerated brine solutions. 2.8 Knowledge Gaps IX resins are being increasingly used as cost-effective treatment for the removal of dissolved organic carbon (DOC) from surface waters. Considering that algae bloom waters commonly result in the co-occurrence of elevated microcystin and NOM concentrations, it is would be of interest 34 to understand to what degree IX resins can be used to remove microcystin during a toxic bloom. At this time, no strategies have yet been proposed in the scientific literature to address this challenge. Assuming that the IX process is efficient for the uptake of microcystins, it is expected that the process would produce a toxic spent brine that shall be treated prior to its disposal in the environment. Therefore, a set of research questions were formulated to answer these knowledge gaps: 1. Are strongly basic IX resins capable of removing microcystin-LR from aqueous matrices? If yes, what is the uptake mechanism (Ion Exchange or Adsorption)? Under pH relevant to water treatment, microcystins are negatively charged. For example, the pKa values of microcystin-LR (MCLR) has been estimated as 2.3, which result from the presence of two negatively charged carboxylic groups and one positively charged amine group. Therefore, the primary hypothesis for this question is that microcystin-LR can be removed using IX under different water matrices. The uptake of toxins can be either through exchange of electrostatic ions or adsorption, as observed in case of other organic contaminants. It would be very valuable to know which of these mechanisms dominate in case of microcystins. 2. How would the performance and robustness of IX process for toxin uptake change under different process parameters (e.g., resin dose, contact time, long-term operation)? The primary hypothesis for this question is that the MCLR removal by the IX resin is influenced by the change in resin dosage and contact time. Kinetic experiments investigated on multiple loading cycles along with different regeneration scenarios can enable us to compare and select the best operating IX conditions for a given application. 35 3. What are the factors that govern the kinetics of microcystin-LR removal under IX operation? It is hypothesized that film diffusion and intraparticle diffusion can impact the kinetics of microcystin-LR uptake under IX application. For organic contaminants such as NOM, the kinetics is majorly influenced by intraparticle diffusion; however, at lower solute concentrations film diffusion also plays a key role in the uptake process. During bloom events, the microcystin-LR concentrations are significantly lower (100-1000 fold) than the NOM. However, studies that investigate the effect of film and intraparticle diffusion of microcystin-LR by IX process are rare. 4. What is the effect of source water characteristics, specifically pH, characteristics of background NOM (charge density and molecular weight distribution), inorganic ions on the uptake of microcystin-LR? The characteristics of NOM, mainly charge density, molecular weight, and hydrophobicity, play a significant role in the efficacy of the IX process for DOC removal. Depending on the intensity of each of these parameters one may dominate and determine the fate of microcystin-LR removal during the treatment. That said, it is hypothesized that increasing the concentration of NOM fractions and inorganic ions results in a lower MCLR uptake by the IX resins. 5. Is IX regeneration effective for microcystin-LR recovery? If yes, how to treat the toxic spent brine prior to its disposal in the environment? Assuming that the IX process is efficient for the sorption of microcystins, it is expected that the process would produce a toxic spent brine that shall be treated prior to its disposal in the environment. At this time, no strategies have yet been proposed in the scientific literature to address this challenge. 36 2.9 Research Objectives Given the concern over the presence of microcystins and organic contaminants in surface waters, the main objective of this study was to investigate the efficiency of a strongly basic anion exchange resin for the co-removal of MCLR and NOM. To the best of our knowledge, no studies have been conducted on this specific topic. The overall goal of this research was achieved through a systematic and detailed experimental work focusing on the following sub-objectives: 1. Study the influence of operational parameters such as resin dosage and contact time along with, environmental factors such as NOM characteristics (concentration, charge density and molecular weight distribution) and presence of other anionic species on the uptake capacity of IX resin. a. Different operating conditions (i.e. resin dose, pH and contact time). b. Different NOM sources c. Different inorganic ions (Sulphates, Nitrates, etc.) d. Isotherm and Kinetic studies 2. Investigate the competitive uptake mechanism between MCLR and NOM using the resin dosages and NOM fractions selected from sub-objective 1. a. Ideal Adsorption Solution Theory (IAST) b. Equivalent Background Concentration (EBC) Model 3. Study the kinetics of uptake and analyze the rate controlling step. a. Film and Pore Diffusion Modelling (FDM and PDM). b. Kinetic Models (Intraparticle diffusion, pseudo-first order knietic model, etc.) 4. Investigate the impact of multiple loading cycles and resin regeneration. 37 a. Batch and consecutive loading b. Effect of operating conditions (NaCl concentration and contact time) 5. Optimize the regeneration process and evaluate the possibility of degrading the toxins in the spent brine. a. Toxin degradation with caustic b. Effect of storage conditions (Contact time, temperature, etc.) 2.10 Engineering Significance This study was intended to lay out a detailed investigation of the key factors affecting the efficacy of suspended bed IX process for MCLR removal. The overall significance of this work was to provide knowledge about the efficiency and effectiveness of IX resins for microcystin uptake under different operational parameters. This knowledge would then be useful in optimizing different treatment conditions (resin dosage and contact time) under different water matrices and will help gauge the robustness of the process for various water sources. The obtained knowledge can be further extended to study and understand the efficiency of combining IX with other processes (e.g., advanced oxidation processes) for cyanotoxin removal within a treatment train. Kinetic data obtained could be used to simulate, design, and fabricate efficient reactor set-up configuration in order to maximize microcystin and NOM removal as well as regeneration efficiency. This will also be useful in understanding the kinetics and limitations of the removal process and provide comparison of lab water studies with IX performance on actual bloom and surface water. 38 Chapter 3: Experimental Methodology This chapter describes the common experimental methodology and procedures followed to fulfill the stated objectives of this research. Methods and experiments specific to each chapter are presented as part of that chapter (in chapters 5-7). 3.1 Experimental Procedures 3.1.1 Glassware Preparation All glassware and tools (flasks, beakers, pipette tips, funnels) were thoroughly washed and triple rinsed with Milli-Q water. This was followed by oven drying at 80\u00C2\u00B0C for 1 hour and baking at 550\u00C2\u00B0C for 5 hours. Pipette tips were autoclaved and stored in sealed plastic bags to avoid potential contamination. Filter papers used for water filtration and resin separation were triple rinsed with Milli-Q water prior to use (Bazri and Mohseni, 2016). 3.1.2 Chemicals and Reagents Dry microcystin-LR (1 mg powder) was purchased from Enzo Lifesciences (Farmingdale, NY, USA) and was used as received. Suwannee River Natural Organic Matter (SRNOM), Suwannee River Humic Acid (SRHA) and Pony Lake Fulvic Acid (PLFA) were obtained from the International Humic Substances Society (IHSS, St. Paul, MN, USA). All solutions were prepared in Milli-Q water (resistivity 18.2 M.\u00CE\u00A9.cm). 39 Table 3.1 Characteristics of Standard NOM Isolates NOM fractionsa Charge Density (meq/g C)b Average Apparent Molecular Weight (Da)c Pony Lake Fulvic Acid (PLFA) 6.8 760 Suwannee River Natural Organic Matter (SRNOM) 10.2 1030 Suwannee River Humic Acid (SRHA) 8.9 1540 aAll waters contained 40 mg/L NaHCO3. Data obtained were estimates and were used for comparison only (Bazri and Mohseni, 2016). bEstimated charge density (Bazri et al., 2016b; Ritchie and Michael Perdue, 2003). cEstimated molecular weights of NOM fractions using LC-OCD technique (Bazri et al., 2016b; Bazri and Mohseni, 2016). 3.1.3 Water and Resin Preparation All experiments involving synthetic water were performed by preparing a stock solution of 700 mg/L IHSS isolate, filtered through 0.45 \u00C2\u00B5m pre-rinsed membrane filters and stored in dark at 4\u00C2\u00B0C for up to two weeks. The final pH was adjusted to 7.0 \u00C2\u00B1 0.1 with NaOH (0.1 N) and NaHCO3 (40 mg/L). MCLR stock solution (0.01 mM), was prepared by mixing 1 mg of dry MCLR in 1 L of Milli-Q water and stored at 4\u00C2\u00B0C for up to two weeks. Purolite\u00C2\u00AE A860 (Purolite, Bala Cynwyd, PA, USA), a strongly basic anionic macroporous resin, was used for all the experiments. Mercury porosimetry tests were performed to determine the resin pore volume distribution (Giesche, 2006). Resin properties are illustrated in Table C.1 (Appendix C). Prior to use, resins were initially regenerated by mixing 10 Bed Volumes (BV) of 10% (W/V) NaCl solution, a condition equivalent to an application of 0.1 kg NaCl/L of resin (or 6.2 lb/ft3). 40 Surface waters were collected from the water treatment plant of the city of West Vancouver (referred hereafter as Eagle Lake), Van Anda Improvement District (Vancouver Island, BC), Shawnigan Lake (Vancouver Island, BC) and Middle River First Nation community (BC). The water samples were pre-filtered with 0.45 \u00C2\u00B5m pre-rinsed membrane filters stored in dark at 4\u00C2\u00B0C for up to four weeks and filtered before use with 0.22 \u00C2\u00B5m pre-rinsed membrane filters. 3.1.4 Isotherm and Kinetic Studies For isotherm studies, 10-1000 mg of resins (221 mg dry weight of resin = 1 mL of wet resin) were mixed for 24 hours (equilibrium) with a source water having 0.1 mg MCLR/L and 2, 3, 5 and 8 mg C/L initial concentration with three reference NOM fractions, Suwannee River NOM, Suwannee River Humic Acids and Pony Lake Fulvic Acids. Control tests without resin were always included in the study. After 24 h, the resins were filtered from the treated water using a 0.45 \u00C2\u00B5m pre-rinsed syringe filters (Fisher Scientific, ON, Canada) prior to determining the final DOC and MCLR concentrations. For kinetic studies, a dry resin dosage of 200 mg (equivalent to approx. 0.9 mL) was mixed with 1 L of water in circular 1 L beakers for contact times varying from 2 minutes to 24 hours in a Phipps & Bird 9900 Jar tester (Richmond, VA, USA) operated at 150 rpm. Triplicate experiments were performed for every experimental condition. A control sample that included contaminant solution without resin was included and all the samples were analyzed in triplicates. 3.1.5 Resin Regeneration and Treatment of Spent Brine Consecutive batch treatments of raw water using 1 mL/L (corresponding to water/resin: 1000 mL/1 mL = 1000 bed volumes, BV) of resins and 60 min contact time were carried out to simulate the 41 performance of resin under commercial suspended bed operations. These conditions were selected on the basis of literature findings by other researchers where resin dosages of 5-20 mL/L and residence times of 10-60 min have been reported for commercial applications (Bazri and Mohseni, 2016; Boyer and Singer, 2006; Slunjski et al., 2000). Resin regeneration was performed by conservatively mixing 10 Bed Volumes (BV) of 10% (W/V) NaCl solution, a condition equivalent to an application of 0.1 kg NaCl/L of resin. Spent brine treatment was investigated for MCLR degradation by raising pH at 9-13 using NaOH (1N) and storing in dark at 4\u00C2\u00B0C for periods up to two weeks. 3.2 Analytical Methods The concentration of Microcystin-LR was determined by solid phase extraction (Waters Sep \u00E2\u0080\u0093Pak tC18 cartridges, Mississauga, Canada) followed by high performance liquid chromatography (Dionex Ultimate 3000, USA) with a UV detector at 238 nm and a C18 analytical column (Waters, Mississauga, Canada). HPLC-grade methanol and phosphoric acid (Fisher Scientific, Ottawa, ON, Canada) were used as eluents for MCLR detection in the HPLC. The HPLC was operated under isocratic conditions using a mobile phase consisting of 50% methanol and 50% phosphate buffer adjusted to a pH of 2.8 at a flow rate of 1 mL/min with 100 \u00C2\u00B5L injections. The retention times for NOM and MCLR were 3 minutes and 17.5 minutes, respectively, indicating a good peak separation which avoided NOM interference. TOC, DOC and Dissolved Inorganic Carbon (DIC) were measured using a TOC analyzer (GE Sievers M5310 C, Boulder, CO, USA). UV254 analyses were performed using a UV visible spectrophotometer (Cary 100 UV-Vis Spectrophotometer, Agilent Technologies, USA) with a path length of 1 cm. Chlorides, sulphates and nitrates were measured 42 using ion chromatography (Dionex ICS- 1100, USA) equipped with an electrical suppressor and AS22 Fast column, according to the USEPA 300.0 reference method (John D. Pfaff, 1993). 3.3 Statistical Analysis The reproducibility of all the triplicate isotherm and kinetic experiments were determined in terms of standard deviation i.e., one of the most commonly adopted quantitative data quality objective (DQOs) (du Prel et al., 2009; Yasutaka et al., 2017). The standard deviation is a measure that is used to quantify the amount of variation or dispersion of a set of data values. The experiments with coefficient of variation (COV; i.e. 100 times the ratio of standard deviation to mean) of 10% or higher were re-performed multiple times (n>3), until the obtained COV was 10% or below. Consequently, the 95% confidence interval (two-tails) was also calculated for all the studies (COV<10%) based on the equation 3.1 (du Prel et al., 2009; Yasutaka et al., 2017). \u00F0\u009D\u0091\u008B \u00C2\u00B1 \u00F0\u009D\u0091\u00A1 \u00C3\u0097\u00F0\u009D\u0091\u00A0\u00E2\u0088\u009A\u00F0\u009D\u0091\u009B 3.1 where X is the mean; n is the number of samples, s is the standard deviation and the t value for 95% confidence interval is estimated based on a two-tailed t-distribution (du Prel et al., 2009; Yasutaka et al., 2017). The calculated 95% confidence interval values indicate that if the sample population is sampled on numerous occasions and interval estimates are made on each occasion, the resulting confidence interval covers the true value in 95 of the 100 studies performed. 43 Chapter 4: Kinetics of Microcystin-LR Uptake by IX Resins 4.1 Introduction Microcystins are amongst the most common family of cyanobacterial toxins detected in surface waters (Chen et al., 2016; Enzo Life Sciences, 2010; Falconer and Humpage, 2005; Kotak and Zurawell, 2007; Meriluoto and Codd, 2005). Under pH relevant to water treatment (pH=5 to pH=8), microcystins are negatively charged (De Maagd et al., 1999). For example, the pKa value of MCLR has been estimated to be 2.2, resulting from the presence of two negatively charged carboxylic groups and one positively charged amine group. Therefore, depending upon the characteristics of the microcystins and the background water matrix undergoing the IX treatment, the toxins can be removed effectively using IX. The uptake of microcystins can occur either through the exchange of electrostatic ions or via adsorption, as observed in the case of other organic contaminants such as chlorophenols, pentachlorophenate and benzene carboxylates (Arias-paic et al., 2016; Bolto et al., 2004, 2002b; Croue and Humbert, 2005; Li and Sengupta, 2001). We wanted to find out which of these mechanisms dominate the removal of microcystins and under what conditions. In particular, this chapter is focused on investigating the microcystin-LR (MCLR) uptake mechanism by strongly basic IX resins. The influence of operating parameters such as resin dosage and contact time on the uptake capacity of the resin was also investigated in this study. The results of these studies were fitted with various kinetic models including pseudo-first order, pseudo-second order and intraparticle diffusion model, to comprehensively determine the rate determining step. 44 4.2 MCLR Removal Mechanism Removal of organics by IX process may be controlled by two major mechanisms, namely ion exchange and adsorption (Graf et al., 2014; Cornelissen et al., 2008; Bazri and Mohseni, 2016). A similar behavior could be expected for MCLR uptake as well. MCLR has two ionizable carboxyl groups, indicating a possibility of reversible electrostatic exchange of resin counter ions (Cl-) with the MCLR molecule. Adsorption may also contribute towards the toxin uptake and is the major mechanism when investigated with activated carbon, peat and iron oxide nanoparticles (Pavagadhi et al., 2013b; Lee and Walker, 2011; Sathishkumar et al., 2010). To investigate the uptake, chloride release studies were performed with 200 mg/L (`0.9 mL/L) of resin on Milli-Q water spiked with MCLR at concentrations from 4 mg/L to 0.1 mg/L. The released chloride meq\u00E2\u0080\u0099s were plotted versus the MCLR meq\u00E2\u0080\u0099s and the results are illustrated in Figure 4.1. A linear plot passing through origin with a regression coefficient (R2) > 0.99 and a linear slope of 1.01 \u00C2\u00B1 0.01 for A860, indicates that ion exchange is the prevalent mechanism for MCLR uptake. Figure 4.1 Stoichiometry of MCLR Removal by IX * Error bars represent standard deviations of experimental replicates (n = 3) 00.00050.0010.00150.0020.00250 0.0005 0.001 0.0015 0.002 0.0025Chloride meqsMCLR meqsMCLR-DI WaterX=Y45 4.3 Isotherm Models The adsorption isotherm indicates the distribution of adsorbent between the liquid and the solid phases at equilibrium state. The isotherms render certain constants whose value articulates surface properties and affinity of adsorbent. In this study, the equilibrium experimental data for the MCLR uptake onto IX resin were analyzed using the two most well-known isotherm models, i.e. Langmuir and Freundlich models (Dada et al., 2012; Ding et al., 2012; Lian et al., 2014). The isotherm constants for the two models were obtained by linear regression methods and are presented in Table 4.1. 4.3.1 Langmuir Msotherm model The Langmuir model assumes monomolecular layer adsorption without any interaction between the adsorbed molecules (Dada et al., 2012; Sahetya et al., 2013). The principal assumption of the Langmuir theory is that the uptake occurs on a homogeneous surface by monolayer adsorption without any interaction between the adsorbed ions. Therefore, the model is applicable to homogeneous adsorption where the adsorption of each molecule has equal sorption activation energy. The Langmuir model can be represented as: \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 =\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00C3\u0097\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0090\u00BF\u00C3\u0097\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0090\u00BF1+(\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0090\u00BF\u00C3\u0097\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092) (4.1) where Ce and qe are the equilibrium concentration (\u00C2\u00B5g/L) and amount of adsorbed compounds (\u00C2\u00B5g/g), qm is qe for a complete monolayer uptake (\u00C2\u00B5g/g) and KL is a constant related to the affinity of the binding sites (L/\u00C2\u00B5g). The linearized form of the Langmuir equation can be represented as: \u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092=1\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00C3\u0097\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0090\u00BF+\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A (4.2) 46 The Langmuir constants qm and KL were obtained by linear regression method. 4.3.2 Freundlich Isotherm Model Freundlich isotherm model assumes that the uptake process occurs on heterogeneous surfaces (Dada et al., 2012; Ebie et al., 2001; Sun et al., 2012). The isotherm equation states that heat of adsorption decreases in magnitude with increase in the extent of adsorption. This isotherm model is defined by the equation: \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 = \u00F0\u009D\u0090\u00BE\u00F0\u009D\u0090\u00B9 \u00C3\u0097 \u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u00921/\u00F0\u009D\u0091\u009B (4.3) where KF and 1/n are the Freundlich constants related to uptake capacity and intensity, respectively. The linearized form of this equation is written as: \u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 =1\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u009B\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092 + \u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u009B\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0091\u0093 (4.4) The linear plot of ln qe vs. ln Ce yields slope of value 1/n and an intercept ln KF. Freundlich constant (1/n) values between 0.1 and 1 indicate favorable uptake. Low values of 1/n (< 1) indicates a weak interaction between adsorbent and adsorbate, while 1/n equal to 1 indicates linear uptake with identical adsorption energies for all the sites. 4.4 Kinetic Models Adsorption kinetics was thoroughly studied to investigate the mechanism of adsorption and the rate determining step. Several kinetic models such as pseudo-first-order, pseudo-second-order and intraparticle diffusion models have been examined for defining the rate determining step and mechanism of adsorption. 47 4.4.1 Pseudo-first-order Kinetic Model The pseudo-first-order rate equation is given as (See appendix B.2 for details): log(\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 \u00E2\u0088\u0092 \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u00A1) = \u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u0094 \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 +\u00F0\u009D\u0091\u009812.303\u00F0\u009D\u0091\u00A1 (4.5) where qe and qt are the amounts of adsorbed MCLR on the adsorbent at equilibrium and at time t respectively (\u00C2\u00B5g/g), and k1 is the first-order adsorption rate constant (min\u00E2\u0088\u00921) (Sahetya et al., 2013; Sathishkumar et al., 2010). 4.4.2 Pseudo-second-order Kinetic Model The pseudo-second-order equation is given as (See Appendix B.1 for details): \u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u00A1=1k2.\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092+\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 (4.6) where k2 (g mg\u00E2\u0088\u00921 min\u00E2\u0088\u00921 ) is the pseudo-second-order rate constant (Pavagadhi et al., 2013; Sahetya et al., 2013). 4.4.3 Film and Intra-particle Diffusion Model A solid\u00E2\u0080\u0093liquid uptake process is usually characterized by external mass transfer (film diffusion) or intraparticle diffusion, or both. The intraparticle diffusion model was used to examine the impact of external diffusion and intra-particle diffusion mechanism described on the basis of Eq. (4.7) (Weber and Morris, 1963): qt= kpt1/2 (4.7) where qt is the time-dependent adsorbate concentration on the resin (\u00C2\u00B5g/g), kp is the intraparticle diffusion rate constant (\u00C2\u00B5g/g min), and t is the contact time. 48 The plot of qt vs. t1/2 at different initial solution concentrations (yields kp) should be linear if intraparticle diffusion is involved in the process. The plotted line must pass through origin if intraparticle diffusion is the sole rate controlling step. If the plot presents multi-linearity or does not pass through the origin, two or more steps occurring in the uptake process (Li and Sengupta, 2000; Matsui et al., 2009). Kinetic data from Figure 4.3 were further analyzed using the Pore diffusion model (PDM) and Film diffusion model (FDM). Parameters were estimated by fitting the equations into the kinetic data using nonlinear optimization schemes as described elsewhere (Bazri and Mohseni, 2016). The analytical solution for intraparticle diffusion model (IDM) for single sized resin beads in a completely stirred tank reactor are described by the following equations: \u00F0\u009D\u0091\u0088\u00F0\u009D\u0091\u00A1 = \u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009C\u00E2\u0088\u0092\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009C\u00E2\u0088\u0092\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092=\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092= 1 \u00E2\u0088\u0092 \u00E2\u0088\u0091 \u00F0\u009D\u009C\u0094\u00E2\u0088\u009E\u00F0\u009D\u0091\u009B=16(\u00F0\u009D\u009C\u0094+1)9+9\u00F0\u009D\u009C\u0094+ \u00F0\u009D\u009B\u00BD\u00F0\u009D\u0091\u009B2\u00F0\u009D\u009C\u00942\u00C3\u0097 exp (\u00E2\u0088\u0092 \u00F0\u009D\u0090\u00B7\u00F0\u009D\u0091\u008E,1\u00F0\u009D\u009B\u00BD\u00F0\u009D\u0091\u009B2\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u0085\u00F0\u009D\u0091\u009D2 ) (4.8) \u00F0\u009D\u009C\u0094 \u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u00A0 \u00F0\u009D\u0091\u0090\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u0090\u00F0\u009D\u0091\u00A2\u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u0091 \u00F0\u009D\u0091\u0093\u00F0\u009D\u0091\u009F\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u009A: \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u0089\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009C=11+ \u00F0\u009D\u009C\u0094 (4.9) \u00F0\u009D\u009B\u00BD\u00F0\u009D\u0091\u009B \u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009F\u00F0\u009D\u0091\u0092 \u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u009B \u00E2\u0088\u0092 \u00F0\u009D\u0091\u00A7\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009F\u00F0\u009D\u0091\u009C \u00F0\u009D\u0091\u009F\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u00A0 \u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u0093 \u00F0\u009D\u0091\u00A1\u00E2\u0084\u008E\u00F0\u009D\u0091\u0092 \u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u00A2\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u009B: tan \u00F0\u009D\u009B\u00BD\u00F0\u009D\u0091\u009B =3\u00F0\u009D\u009B\u00BD\u00F0\u009D\u0091\u009B1+ \u00F0\u009D\u009C\u0094\u00F0\u009D\u009B\u00BD\u00F0\u009D\u0091\u009B2 (4.10) where U(t), is the fractional attainment of equilibrium and Co. Ct and Ce are concentrations of solute (mg/L) at time t= 0, at time t, and at equilibrium, respectively. Rp is the radius of the resin (assuming spherical, 0.375 cm) and Da,1 is the apparent diffusivity (cm2/s). For film diffusion controlled removal, the following equation represents the changes in the MCLR concentration of the solution: \u00F0\u009D\u0091\u0088(\u00F0\u009D\u0091\u00A1) = 1 \u00E2\u0088\u0092 exp (\u00E2\u0088\u00923.\u00F0\u009D\u0090\u00B7\u00F0\u009D\u0091\u0093.(\u00F0\u009D\u0091\u0089\u00E2\u0080\u00B2\u00F0\u009D\u0090\u00B6\u00E2\u0080\u00B2+\u00F0\u009D\u0091\u0089\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009C)\u00F0\u009D\u0091\u0085\u00F0\u009D\u0091\u009D.\u00F0\u009D\u009B\u00BF.\u00F0\u009D\u0090\u00B6\u00E2\u0080\u00B2\u00F0\u009D\u0091\u0089) (4.11) 49 where C\u00E2\u0080\u0099 is the resin exchange capacity (meq/L), \u00CE\u00B4 is the film thickness (10-3 cm), Co is the initial solute concentration (meq/L), V is the solution volume (L), V\u00E2\u0080\u0099 is the resin volume (L), Df is the film diffusion coefficient (cm2/s). All other parameters (Df and Da,1) are assumed to be constant and are estimated based on nonlinear optimization schemes. The curve fitted for Df agreed well with the experimental data (0.95 < R2 < 0.98), while the quality of fit for the pore diffusion model (0.96 < R2 < 0.99), was well fit under the dilute condition assumption as described elsewhere (Bazri and Mohseni, 2016). The rate-controlling step was further investigated using the dimensionless Biot number (Bi, Appendix B.3) which is the ratio of internal mass transfer (i.e., pore diffusion) to external mass transfer (i.e., film diffusion) resistances (Ko et al., 2001) \u00F0\u009D\u0090\u00B5\u00F0\u009D\u0091\u0096 =\u00F0\u009D\u0091\u0098\u00F0\u009D\u0091\u0093.\u00F0\u009D\u0091\u0085\u00F0\u009D\u0091\u009D\u00F0\u009D\u0090\u00B7\u00F0\u009D\u0091\u009D\u00F0\u009D\u0091\u0092 (4.12) where kf (cm/s) is the external mass transfer coefficient (kf = Df/\u00CE\u00B4) and Df (cm2/s), is the film diffusion coefficient and Dpe (cm2/s) is effective pore diffusion coefficient (\u00CE\u00B4 is film thickness \u00E2\u0089\u0088 10-3 cm according to (Helfferich, 1965). The Bi << 1 indicates film diffusion as the rate-limiting step where Bi >>1 shows pore diffusion to be the rate limiting step. 4.5 MCLR Uptake Capacity on IX Resin The uptake capacity of Purolite A860 resin for MCLR was initially measured at pH 7, with 100 \u00C2\u00B5g/L of MCLR at different resin dosages (10 -1000 mg/L). Figure 4.2 illustrates the uptake of MCLR plotted as percent removal vs. resin dose. It can be seen that the toxin uptake improved with increasing resin dosage from 10 mg/L to 1000 mg/L for all investigated water matrices. This is in agreement with previous studies where the availability of higher surface area and more active 50 sites at higher resin concentrations resulted in greater contaminant uptake (Chen et al., 2015; Bolto et al., 2004; Kabsch-Korbutowicz et al., 2008; Lian et al., 2014). The difference in the uptake of MCLR by the resin was not statistically significant for the IX dosages above 50 mg/L. This can be explained by the fact that at dosages above 50 mg/L, the ratio of exchangeable resin sites to the total contaminants (meq/meq) is approximately 100-folds higher. Thus, there are excessively more sites available for the MCLR uptake and the competitive behavior is less significant. For all the investigated conditions, the residual MCLR concentrations were below the USEPA standard of 0.3 \u00C2\u00B5g/L (for young children (EPA, 2012)) at the resin dosage of 200 mg IX/L (~ 1mL/L) and above. Moreover, when the initial resin dosage was 10 mg/L, the toxin uptake capacity could reach as high as 3850 \u00C2\u00B5g /g, indicating that A860 has a great potential for the removal of MCLR. Noteworthy, this study tested very low resin concentrations to investigate the competitive effects. Actual IX dosages applied in the industry are typically in the range of 1000 - 5000 mg/L (5-20 mL/L, five to ten times higher than investigated dosages). Therefore, a complete MCLR elimination is expected from source water at these commercial dosages and it can be observed that Purolite A860 exhibits a great potential for the practical applications in removing algal toxins from toxin laden waters. 51 Figure 4.2 Effect of Resin Dosage on MCLR Uptake at pH 7; Lines Indicate a Fit to the Freundlich Isotherm. * Error bars represent standard deviations of experimental replicates (n = 3) Table 4.1 Isotherm Constants and Thermodynamic Parameters Parameters MCLR in DI Water Kf (mg/g/(mg/L)1/n) 27.94 \u00C2\u00B1 1.60 1/n 0.63 \u00C2\u00B1 0.07 R2 0.97 Data reported with 95th confidence intervals for the respective thermodynamic parameters In order to obtain the theoretical adsorption capacity and investigate the distribution of MCLR molecules on the resin, adsorption equilibrium isotherms were fitted to the Langmuir and Freundlich (qe=Kf.Ce1/n) models (section 4.3; parameters are provided in Table 4.1). The latter was highly suitable (R2 > 0.97) for modelling the isotherms of MCLR sorption on A860 resin which 0204060801000 200 400 600 800 1000 1200% MCLR UptakeResin Dosage (mg/L)MCLR-DI WaterFreundlich Isotherm52 suggests heterogeneous adsorption of MCLR (Figure 4.2). Note that the obtained 1/n values were positive and lower than 1, indicating a weak interaction between the resin and MCLR, and therefore, unfavorable adsorption (Dada et al., 2012; Sahetya et al., 2013; Sun et al., 2012). Table 4.1 displays the regression data for the performed studies. 4.6 Kinetic Results The effect of initial MCLR concentration on the uptake rate was investigated at a fixed resin dosage (200 mg IX/L ~ 1 mL/L) with different initial concentrations of MCLR (10 \u00E2\u0080\u0093 100 \u00C2\u00B5g/L) at pH 7 and 22\u00C2\u00B0C. As depicted in Figure 4.3, a rapid uptake with around 80% removal was achieved within 10 minutes with the initial concentration of 100 \u00C2\u00B5g/L. Figure 4.3 MCLR Removal Kinetics Under Different Initial MCLR Concentrations; Lines Indicate a Fit to the Pseudo-second-order Kinetic Model, Studies Performed at IX Dosage of 200 mg/L~ 1 mL/L at pH=7. * Error bars represent standard deviations of experimental replicates (n = 3) 0501001502002503003504004505000 10 20 30 40 50 60MCLR uptake (\u00C2\u00B5g/g)Time (minutes)100 \u00C2\u00B5g/L50 \u00C2\u00B5g/L25 \u00C2\u00B5g/L10 \u00C2\u00B5g/L53 To investigate further, the experimental data were fitted to a pseudo-first-order kinetics, a pseudo-second order kinetics, and the intra-particle diffusion model (see details, section 4.4). The kinetic parameters calculated by the three models are listed in Table 4.2. The pseudo-second-order kinetics model was found to provide the best fit to the kinetic data (refer to Figure 4.3). The pseudo first order kinetic model under-predicted the equilibrium uptake values for all the conditions examined. The highest k1 and k2 values reported were with 100 \u00C2\u00B5g/L MCLR (Table 4.2). Increasing the initial MCLR concentration from 10-100 \u00C2\u00B5g/L resulted in a three fold increase in the pseudo second order rate constant values, while the qe values increased by around 10 folds (45 \u00C2\u00B5g/g to 450 \u00C2\u00B5g/g), indicating a faster uptake at higher MCLR concentrations. The intraparticle diffusion model was used to examine the impact of external diffusion and intra-particle diffusion mechanism (Weber and Morris, 1963) described on the basis of the model equation (qt= kpt1/2; section 4.4). The calculated values of kp were 4.78, 13.12, 24.31 and 46.26 \u00C2\u00B5g/g min, for 10, 25, 50 and 100 \u00C2\u00B5g/L MCLR, respectively. According to the model, the plot of qt vs t1/2 should be linear if intra-particle diffusion is involved in the process (Zhang et al., 2014; Chen et al., 2015). The R2 values of the intra-particle diffusion model for all MCLR-DI matrix studies listed in Table 4.2 are low (<0.9), indicating that intra-particle diffusion is not the only rate controlling step and the process involves more than one kinetic stage (Guo et al., 2003). The rate-controlling step was further investigated using the dimensionless Biot number which is the ratio of internal mass transfer (i.e., pore diffusion) to external mass transfer (i.e., film diffusion) resistances (Ko et al., 2001). As explained earlier (Section 4.4), Bi << 1 indicates the film resistance as the rate limiting step; whereas, Bi >> 1 indicates the pore diffusion to be the rate-controlling step. Parameters were estimated by fitting the equations into the kinetic data using nonlinear optimization schemes as described elsewhere (Bazri and Mohseni, 2016) (Appendix 54 B.3). The Biot numbers obtained for 10, 25, 50 and 100 \u00C2\u00B5g/L MCLR concentration in DI water were 0.72, 0.90, 3.32 and 6.58, indicating that film diffusion is the rate determining step at lower MCLR concentrations and pore diffusion at higher MCLR concentrations (above 25 \u00C2\u00B5g/L) at the dosage of 200 mg IX/L (1 mL/L). Therefore, it was inferred that the overall uptake process was controlled by film diffusion in DI water for commonly encountered MCLR concentrations in bloom samples that are usually below 25 \u00C2\u00B5g MCLR/L (at toxin to resin ratio of 4X10-8 and below (meq/meq)) as illustrated in Figure 4.4). Figure 4.4 Biot Number as a Function of MCLR Concentration at Resin Dosage of 200 mg/L (~1 mL/L) 012345670 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16BiotResin Loading X 10-6 (Meq/Meq)MCLRFilm DiffusionPore Diffusion55 Table 4.2 Kinetic Parameters of MCLR Uptake by Strongly Basic IX resins Model Parameters MCLR-10 \u00C2\u00B5g/L MCLR 25 \u00C2\u00B5g/L MCLR 50 \u00C2\u00B5g/L MCLR-100 \u00C2\u00B5g/L Pseudo-first Order qe 20 \u00C2\u00B1 1 97 \u00C2\u00B1 2 192 \u00C2\u00B1 5 372 \u00C2\u00B1 7 kl 0.052 \u00C2\u00B1 0.002 0.060 \u00C2\u00B1 0.002 0.064 \u00C2\u00B1 0.002 0.070 \u00C2\u00B1 0.003 R2 0.90 0.98 0.96 0.94 Pseudo-second Order k2 0.24 \u00C2\u00B1 0.01 0.32 \u00C2\u00B1 0.01 0.56 \u00C2\u00B1 0.02 0.74 \u00C2\u00B1 0.01 qe 47 \u00C2\u00B1 3 127 \u00C2\u00B1 5 238 \u00C2\u00B1 4 454 \u00C2\u00B1 8 R2 0.98 0.99 0.99 0.99 Intra-particle diffusion kp 4.78 \u00C2\u00B1 0.15 13.12 \u00C2\u00B1 0.30 29.31 \u00C2\u00B1 0.41 46.26 \u00C2\u00B1 0.74 C 82.90 13.25 15.22 14.75 R2 0.80 0.97 0.82 0.87 Biot Number Bi 0.72 0.90 3.32 6.58 *Data reported with 95th confidence interval for the respective kinetic parameters 56 4.7 Conclusions In the present study, a strongly basic macroporous resin (Purolite A860) was employed to remove Microcystin-LR (MCLR) from aqueous matrices. The objective of this work was to identify the mechanism of MCLR uptake (i.e. electrostatic ion exchange or adsorption) and to investigate the impact of different influencing factors, such as MCLR and resin concentration, on uptake. The A860 resin exhibited an excellent adsorption capacity of 3850 \u00C2\u00B5g/g. The removal of MCLR by this resin was primarily attributed to electrostatic interactions. Kinetic experiments further revealed that more than 70% of the toxin was removed within 10 minutes at a resin dosage of 200 mg/L, with the rate of uptake rising proportionally to MCLR/resin (ratio) concentration. Both external diffusion and intra-particle diffusion play a key role in the uptake kinetics of MCLR. In DI water matrices, film diffusion is the limiting factor at MCLR to resin ratios of 4X10-8 and below (meq/meq). Pore diffusion serves as the rate limiting step for MCLR to resin ratios above 4X10-8(meq/meq). These results reveal that anion exchange resins, given their efficiency and low cost, display promising potential for the removal of microcystins. However, microcystins do not occur alone in algal bloom waters. Because these toxins commonly co-occur with elevated NOM concentrations, we found it pertinent to investigate the impact of NOM characteristics and inorganic ions on the uptake of microcystins by IX resins. The following chapters will evaluate the efficacy of the IX process for the simultaneous removal of MCLR, NOM, and inorganic ions. 57 Chapter 5: Effect of Background NOM Concentration on MCLR Uptake 5.1 Introduction In the previous chapter we demonstrated that IX are capable of removing microcystins from spiked laboratory (Deionized (DI) Water) water. However, bloom events often elevate the total organic carbon (TOC) levels of water bodies. The presence of natural organic matter (NOM) along with algogenic organic matter (AOM) gives rise to many challenges associated with water quality and treatment processes. As a precursor of disinfection by-products, organic matter significantly impacts the taste and odor of potable water. Moreover, it deteriorates the biological stability of water by increasing bacterial regrowth and biofilm formation within distribution systems. Additionally, it can reduce the efficiency of advanced oxidation processes (AOPs) for disinfection and micro-pollutant degradation (Acero et al., 2008; Coral et al., 2013; Dixon et al., 2011; Lee et al., 2004; Rodr\u00C3\u00ADguez et al., 2007; Sordo et al., 2007). Therefore, processes that can effectively and simultaneously remove cyanotoxins and NOM are necessary. To address these challenges, anionic ion exchange (IX) resins are being increasingly used as cost-effective treatments for the removal of dissolved organic carbon (DOC) (Huang et al., 2015; Kabsch-Korbutowicz et al., 2008). Considering the IX process\u00E2\u0080\u0099s effectiveness for the uptake of microcystins, the main objective of this study was to investigate the efficiency of a strongly basic anion exchange resin for the co-removal of MCLR and NOM. We also investigated the influence of environmental factors such as NOM concentration and other anionic species on resin uptake capacity. 58 5.2 Effect of Resin Dosage The effect of resin dosage on the uptake of MCLR in the presence of different NOM concentrations was studied to provide valuable understanding of the capacity and efficacy of the resin A860 for MCLR removal in high DOC waters. Figure 5.1(a) illustrates the uptake of MCLR plotted as percent removal vs. resin dose. It was observed that more than 80% removal (initial MCLR = 100 \u00C2\u00B5g/L) was achieved in DI water for all the resin dosages tested (10, 50, 200, 500 and 1000 mg/L). Moreover, when the initial resin dosage was 10 mg/L, the toxin uptake capacity could reach as high as approximately 3800 \u00C2\u00B5g/g, indicating that A860 has a great potential for the removal of MCLR. In the experiments evaluating MCLR removal in NOM-containing waters, a significant reduction in the uptake of MCLR was observed at the lower resin dosages. For instance, at the resin dosage of 10 mg/L with a DOC of 2 mg/L, toxin uptake was reduced twofold: from 80% in DI water to 42%. Source water characteristics, specifically the concentration and character of NOM can impact the uptake of cyanotoxins through competitive uptake mechanisms (Cook et al., 2001; Matsui et al., 2012). In particular, compounds of similar size and shape are believed to be the most competitive, and the adsorbent pore size and volume distributions play a key role in the competitive uptake (Dixon et al., 2011; Matsui et al., 2012; Pelekani and Snoeyink, 1999). Further, more than 70% of the DOC reduction (Fig 5.1(b)) was observed for all the NOM matrices indicating a strong competition by NOM and saturation of uptake sites available for MCLR leading to a decrease in the toxin uptake. This is also evident by observing the results at higher DOC concentrations, where increasing the NOM concentration resulted in a further decrease in the MCLR uptake from 80% in DI water to 34% and 28% at DOC of 5 mg/L and 8 mg/L, respectively. For all the investigated conditions, increasing the resin dosage improved the toxin removal. The highest tested IX dosage of 1000 mg/L (~ 4.5 mL/L) achieved more than 99% MCLR removal, 59 even in the source water of 8 mg C/L. Therefore, IX treatment provided a final MCLR concentration of <0.2 \u00C2\u00B5g/L, a value below WHO guideline of 1 \u00C2\u00B5g/L. Noteworthy, this study tested very low resin concentrations to investigate the competitive effects. Actual IX dosages applied in the industry are typically in the range of 1000 - 5000 mg/L (Kabsch-Korbutowicz et al., 2008). Thus, IX appears as a promising process for effective removal of MCLR from aqueous matrices. 5.3 MCLR and NOM Uptake Capacities on IX Resin The uptake amount of MCLR and NOM at equilibrium on A860 at different resin dosages is illustrated in Figures 5.1 (a) and 5.1 (b), respectively. In order to obtain the theoretical adsorption capacity and investigate the distribution of MCLR molecules on the resin, adsorption equilibrium isotherms were fitted to the Freundlich model (qe=Kf.Ce1/n ; parameters are provided in Table 5.1). The model was highly suitable (R2 > 0.97) for modelling the isotherms of MCLR sorption on A860 resin which suggests heterogeneous adsorption of MCLR (Figure 5.1 (c)). The synergistic adsorption mechanisms of hydrophobic effects, cation-\u00CF\u0080 bonding, and \u00CF\u0080-\u00CF\u0080 electron donor-acceptor (EDA) interactions between the organic contaminants influence the uptake of MCLR onto the resin and incline the system towards heterogeneous adsorption (Zhang et al., 2014). Noteworthy, the 1/n values for all the water matrices were positive and lower than 1, indicating a weak interaction between the resin and MCLR, and, therefore, unfavorable adsorption (Dada et al., 2012; Sahetya et al., 2013). As the NOM concentration increased in source waters, the values of 1/n slightly decreased from 0.63 to 0.49. The most important effect was observed on the values of Kf which declined from 28 ((mg/g)(L/mg)1/0.63) in buffered MQ water to 1.2 60 ((mg/g)(L/mg)1/0..59)) in the presence 2 mg C/L. Increasing Suwannee river NOM from 2 to 8 mg C/L (a 4-fold increase) further decreased Kf from 1.2 to 0.26 (approximately a 5 fold decrease). 02040608010010 50 200 500 1000MCLR Removal (%)Resin Dosage (mg/L); *221 mg dry resin= 1 mLDI WaterDOC 2 mg/LDOC 3 mg/LDOC 5 mg/LDOC 8 mg/L02040608010010 50 200 500 1000DOC Removal (%)Resin Dosage (mg/L)DOC 2 mg/LDOC 3 mg/LDOC 5 mg/LDOC 8 mg/L(a) (b) 61 Figure 5.1 (a) Removal of MCLR (Co = 25 \u00C2\u00B5g/L) by A860 Resin After 24 h for Waters with Increasing DOC Concentrations and Resin Dosages (b) DOC Removal by A860 Resin After 24 h (c) Adsorption Isotherms of MCLR on A860 Resin at pH 7.0. * 1 mL resin= 221 mg dry weight of resin (24 hours in desiccator); Error bars represent standard deviations of experimental replicates (n = 3) Table 5.1 Adsorption Isotherm Constants and Thermodynamic Parameters for MCLR and Suwannee River NOM Parameters * MCLR isotherms Suwannee NOM Isotherm with 5 mg C/L MCLR-DI Water With 2 mg C/L With 3 mg C/L With 5 mg C/L With 8 mg C/L Kf (mg/g/(mg/L)1/n) 27.94 \u00C2\u00B1 1.60 1.19 \u00C2\u00B1 0.09 0.60 \u00C2\u00B1 0.03 0.42 \u00C2\u00B1 0.02 0.26 \u00C2\u00B1 0.01 88.32 \u00C2\u00B1 4.76 1/n 0.63 \u00C2\u00B1 0.07 0.54 \u00C2\u00B1 0.03 0.51 \u00C2\u00B1 0.03 0.50 \u00C2\u00B1 0.04 0.49 \u00C2\u00B1 0.05 0.46 \u00C2\u00B1 0.10 R2 0.97 0.99 0.99 0.98 0.98 0.99 *Fitted with the Freundlich equation: (qe=Kf.(Ce)n ; data reported with 95th confidence intervals for the Kf and n values 0500100015002000250030003500400045000 20 40 60 80Uptake (\u00C2\u00B5g/g)Equilibrium Concentration (\u00C2\u00B5g/L)DI WaterDOC 2mg/LDOC 3 mg/LDOC 5 mg/LDOC 8 mg/L(c)62 5.4 MCLR Removal Kinetics: Effect of Background NOM Concentration Figure 5.2 (a) depicts the removal kinetics of MCLR under different NOM concentrations for a constant dosage of 200 mg IX/L (~ 1mL/L). The experimental data were fitted to a pseudo-first-order kinetic, a pseudo-second order kinetic and the intra-particle diffusion model (See section 4.4 for details). The kinetic parameters calculated by the three models are listed in Table 5.2. The pseudo-second-order kinetics model was found to provide the best fit to the kinetic data. The results from previous studies with different initial MCLR concentrations were also used to evaluate the effect of background NOM concentration on MCLR removal (Table 5.2). Increasing the background NOM concentration from 2 to 8 mg C/L (four times) resulted in a decrease (nearly half) of the kinetic constants. The NOM kinetic rate constants (k2) at 8 mg C/L are higher (double) than that of MCLR indicating a faster uptake of NOM. Increasing the background DOC concentration also resulted in a decrease in the MCLR equilibrium loading (pseudo-second order parameter qe at 60 min) from 130 \u00C2\u00B5g/g in DI water to approximately 105, 100, 90 and 80 \u00C2\u00B5g/g, at NOM concentrations of 2, 3, 5 and 8 mg C/L, respectively. The removal efficiency was also reduced from more than 90% in DI water to approximately 60% in 8 mg C/L waters, indicating a strong competition for limited ion exchange sites. As depicted in Figure 5.2 (a) and (b), more than 30 % of DOC was removed during the first 20 minutes of operation for all dosages. The qe values obtained for NOM at 60 min were magnitudes higher than the values for MCLR indicating a large coverage of the resin sites. In comparison to the pseudo-second order qe values for MCLR of approximately 130 \u00C2\u00B5g/g, NOM qe value was approximately 9000 \u00C2\u00B5g/g (a 70-fold higher loading). Since NOM concentrations are in mg/L, and therefore, are much higher than MCLR concentrations, a strong competition for the limited active sites can be observed leading to an overall decrease in MCLR uptake . 63 The intraparticle diffusion model (IPD) was used to examine the impact of external diffusion and intra-particle diffusion mechanism described in section 4.4 (Weber and Morris, 1963). The calculated values of kid were 4.78, 13.1, 24.3 and 46.3 \u00C2\u00B5g/g min1/2, for 10, 25, 50 and 100 \u00C2\u00B5g/L MCLR, respectively. In the presence of background NOM, the value of kid was reduced further to 5.62 \u00C2\u00B5g/g min1/2 with 8 mg C/L. According to the model, the plot of qt vs t1/2 should be linear if intra-particle diffusion is involved in the process (Zhang et al., 2014; Chen et al., 2015). The R2 values of the intra-particle diffusion model for all MCLR-DI matrix studies listed in Table 5.2 are low (<0.9), indicating that intra-particle diffusion is not the only rate controlling step and the process involves more than one kinetic stage (Guo et al., 2003). The rate-controlling step was further investigated using the dimensionless Biot number which is the ratio of internal mass transfer (i.e., pore diffusion) to external mass transfer (i.e., film diffusion) resistances (Ko et al., 2001). Parameters were estimated by fitting the equations into the kinetic data using nonlinear optimization schemes (see details in Appendix B.3) (Bazri and Mohseni, 2016). The uptake kinetic is significantly influenced by the presence of background NOM where a shift in the rate controlling step, from film diffusion (Bi=0.9) to pore diffusion (Bi=58) is observed as the DOC concentration is increased from 0 to 2 mg/L. This observation is in agreement with the IPD model where R2 values > 0.90 were obtained for all MCLR-NOM matrices. Therefore, it is inferred that the overall uptake process is controlled by intra-particle diffusion in presence of background NOM and is controlled either by external diffusion or intraparticle diffusion in DI water (depending on the MCLR concentration as discussed in section 4.4). 64 Table 5. 2 Kinetic Parameters of MCLR Uptake in Presence of Different NOM Fractions Model Parameters MCLR-10 \u00C2\u00B5g/L MCLR 25 \u00C2\u00B5g/L MCLR 50 \u00C2\u00B5g/L MCLR-100 \u00C2\u00B5g/L 25 \u00C2\u00B5g/L MCLR with 2 mg C/L 25 \u00C2\u00B5g/L MCLR with 3 mg C /L 25 \u00C2\u00B5g/L MCLR with 5 mg C/L 25 \u00C2\u00B5g/L MCLR with 8 mg C/L NOM 8 mg C/L with 25 \u00C2\u00B5g/ MCLR Pseudo-first Order qe 20 \u00C2\u00B1 1 97 \u00C2\u00B1 2 192 \u00C2\u00B1 6 372 \u00C2\u00B1 8 53 \u00C2\u00B1 1 52 \u00C2\u00B1 1 51 \u00C2\u00B1 1 50 \u00C2\u00B1 1 6514 \u00C2\u00B1 50 kl 0.052 \u00C2\u00B1 0.002 0.060 \u00C2\u00B1 0.002 0.064 \u00C2\u00B1 0.002 0.070 \u00C2\u00B1 0.003 0.057 \u00C2\u00B1 0.003 0.055 \u00C2\u00B1 0.003 0.042 \u00C2\u00B1 0.003 0.032 \u00C2\u00B1 0.003 0.053 \u00C2\u00B1 0.001 R2 0.90 0.98 0.96 0.94 0.88 0.89 0.91 0.91 0.98 Pseudo-second Order k2 0.24 \u00C2\u00B1 0.01 0.32 \u00C2\u00B1 0.01 0.56 \u00C2\u00B1 0.02 0.74 \u00C2\u00B1 0.01 0.27 \u00C2\u00B1 0.03 0.25 \u00C2\u00B1 0.02 0.20 \u00C2\u00B1 0.02 0.16 \u00C2\u00B1 0.01 0.34 \u00C2\u00B1 0.01 qe 47 \u00C2\u00B1 3 127 \u00C2\u00B1 5 238 \u00C2\u00B1 4 454 \u00C2\u00B1 9 105 \u00C2\u00B1 4 101 \u00C2\u00B1 4 90 \u00C2\u00B1 4 81 \u00C2\u00B1 4 9887 \u00C2\u00B1 63 R2 0.98 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.97 Intra-particle diffusion kid 4.78 \u00C2\u00B1 0.17 13.12 \u00C2\u00B1 0.34 29.31 \u00C2\u00B1 0.46 46.26 \u00C2\u00B1 0.84 8.78 \u00C2\u00B1 0.37 7.27 \u00C2\u00B1 0.11 6.81 \u00C2\u00B1 0.25 5.62 \u00C2\u00B1 0.27 1045 \u00C2\u00B1 98 C 82.9 \u00C2\u00B1 5 13.3 \u00C2\u00B1 1 15.2 \u00C2\u00B1 2 14.8 \u00C2\u00B1 3 42.3 \u00C2\u00B1 5 42.9 \u00C2\u00B1 4 44.4 \u00C2\u00B1 5 48.7 \u00C2\u00B1 5 85.7 \u00C2\u00B1 4 R2 0.80 0.97 0.82 0.87 0.93 0.98 0.89 0.90 0.93 Biot Number Bi 0.72 0.90 3.32 6.58 58 61 67 71 2071 Data reported with 95th confidence interval for the respective kinetic parameters 0204060801001200204060801001200 20 40 60DOC remaining (%)MCLR uptake (\u00C2\u00B5g/g)Time (minutes)DI WaterDOC 2mg/LDOC 3mg/LDOC 5 mg/L(a) 65 Figure 5.2 (a) MCLR Removal Kinetics and NOM Uptake Under Different Background NOM Concentrations (b) DOC, UV254 Absorbance and MCLR Reduction after 20 Minutes of Contact Time at Dosage of 200mg/L (~ 1mL/L) *Lines indicate a fit to pseudo-second order kinetic model; initial MCLR concentration=25 \u00C2\u00B5g/L; DOC= 2-8 mg C/L; Error bars represent standard deviations of experimental replicates (n=3) 5.5 Conclusion Through extensive kinetic experimets, ion exchange was identified as a potential process for the simultaneous removal of MCLR and DOC from aqueous matrices. Increasing resin dosage improved toxin uptake. A resin dosage of 1000 mg/L (~ 4.5 mL/L) was sufficient to reduce the toxin concentration of 100 g/L to below the WHO guideline of 1 \u00C2\u00B5g/L in all investigated water matrices. 0204060801002 mg/L 3 mg/L 5 mg/L 8 mg/LRemoval (%)DOC Concentration (mg/L)DOCUV254MCLR(b) 66 However, a significant reduction in MCLR uptake was observed in the presence of background NOM compounds (SRNOM). Intra-particle diffusion plays a key role in governing the uptake kinetics of MCLR in presence of organic contaminants. NOM compounds with an average charge density considerably higher than MCLR are preferably removed, resulting in a decline in the kinetic rate of MCLR removal. Overall, this study indicates that anionic IX is a prmosing treatment alternative for removing MCLR from natural waters, even under elevated NOM concentrations. We next conducted further studies to assess how specific NOM characteristics and inorganic ions impact the efficacy of the IX process for MCLR removal. 67 Chapter 6: Impact of NOM Characterisitcs on MCLR Uptake 6.1 Introduction As shown in earlier chapters and verified in other studies, the use of strongly basic ion exchange resins (IX) is an effective tool for the simultaneous removal of natural organic matter (NOM) (Croue and Humbert, 2005; Dixit and Mohseni, 2016; Hu et al., 2014; Humbert et al., 2005) and microcystins from surface waters. A solid\u00E2\u0080\u0093liquid uptake process is usually characterized by external mass transfer (film diffusion) or intraparticle diffusion, or both (Weber and Morris, 1963). Kinetic studies investigating the removal of MCLR by IX resins have concluded that pore diffusion is the rate-limiting step in this process. However, film diffusion can also become significant at low solute concentration. It has been well documented that NOM properties such as molecular weight (MW), polarity (hydrophobicity), and charge density significantly influence IX treatment efficacy (Bazri and Mohseni, 2016; Bolto et al., 2004, 2002b; Huang et al., 2015; Zhang et al., 2014). The IX process has been shown to be more effective at removing low to medium molecular weight organics. (Bazri et al., 2016b; Cornelissen et al., 2008; Croue and Humbert, 2005; Huang et al., 2015; Kabsch-Korbutowicz et al., 2008; Li and Sengupta, 2001; Singer and Bilyk, 2002; Sun et al., 2015). However, to the best of the author\u00E2\u0080\u0099s knowledge, no studies elucidating the impact of NOM properties on Microcystin uptake by IX have been performed. Extensive kinetic experiments have been carried out on three standard NOM isolates with different MWs and charge densities: Suwannee River NOM (SRNOM), Suwannee River Humic Acid (SRHA), and Pony Lake Fulvic Acid (PLFA). The characteristics of these NOMs are represented in Table 3.1 (Bazri and Mohseni, 2016). We obtained these commercially available NOM fractions from the \u00E2\u0080\u0098International Humic 68 Substances Society\u00E2\u0080\u0099 and utilized them to evaluate the effect of NOM charge density and molecular weight distribution on the removal of MCLR via the IX process. Furthermore, we employed mathematical and physical models such as the film diffusion and pore diffusion models, to investigate the impact of these properties on the rate-limiting step for MCLR uptake. Finally, we compared the uptake of MCLR in synthetic water with that in surface water under equivalent conditions. 6.2 Competitive Uptake of NOM and MCLR The effect of resin dosage on the uptake of MCLR in the presence of different NOM fractions was investigated to provide an understanding of the capacity and efficacy of the resin A860 for MCLR removal from different water matrices. Further, studies were performed at lower resin concentrations in order to investigate competitive behavior and therefore, the selected dosages were well below the commercial operational dosages which are as high as 4500 mg/L (~20 mL/L) (Kabsch-Korbutowicz et al., 2008). The DOC removal followed the order SRNOM > SRHA> PLFA, i.e., in the order of the increasing charge density of these respective fractions (Figure 6.1(a), for 1000 mg/L ~ 4.5 mL/L resin dosage). Previous studies have indicated that charge density and molecular weight distribution of NOM play key roles in the uptake process by IX resins (Bazri et al., 2016a; Bazri and Mohseni, 2016). However, for lower resin dosages (<200 mg/L ~ 1 mL/L and below), the lower charge density fraction, PLFA (700 Da, 6 meq/g C), shows higher DOC removal in comparison to SRHA, (1500 Da, 8 meq/g C), which might be due to the blockage of resin pores and channels by higher molecular weight SRHA fractions (Bazri and Mohseni, 2016). The removal of MCLR in the presence and absence of NOM fractions after 24 hours of contact time is depicted in Figure 6.1 (b). It can be seen that the toxin uptake improved with increasing 69 resin dosage from 10 mg/L to 1000 mg/L (0.05 mL/L to 4.5 mL/L) for all investigated water matrices. This is in agreement with previous studies where the availability of higher surface area and more active sites at higher resin concentrations resulted in greater contaminant uptake (Bolto et al., 2002b; Kabsch-Korbutowicz et al., 2008). To evaluate the competitive NOM uptake, resin dosages of 10, 50, 200, 500 and 1000 mg/L (corresponding to an approximate dose ratio of 1:1, 5:1, 20:1, 50:1 and 100:1 meq/meq (resin/contaminants)) were tested. It can be observed that MCLR uptake is significantly affected by the presence of NOM fractions in the water matrix, especially at lower resin dosages (i.e., 200 mg/L (~1 meq/mL) and below). For instance, at the resin dosage of 10 mg/L with an initial SRNOM of 3 mg C/L, toxin uptake was reduced from 80% in DI water to approximately 40%. In particular, NOM fractions with higher charge densities can impact the uptake of cyanotoxins and micropollutants through competitive uptake mechanisms. Higher MCLR removal was observed in the presence of a lower charge density fraction (PLFA), as compared to other NOM fractions with the same initial DOC concentration (3 mg C/L) for all the investigated resin dosages below 200 mg/L. Presence of highly charged SRNOM decreased the toxin removal from >80% in DI water to approximately 40, 50 and 70% at resin dosages of 10 mg/L, 50 mg/L and 200 mg/L (0.05 mL/L, 0.23 mL/L and 0.91 mL/L), respectively. However, as illustrated in Figures 6.1 (a) and 6.1 (b), MCLR and DOC removals were not statistically significant (p > 0.05) at dosages of 500 mg/L and 1000 mg/L (2.3 mL/L and 4.5 mL/L), where the ratio of exchangeable resin sites to the total contaminant milli-equivalents (meq/meq) was 10:1 and above (resin: 0.8 meq/mL). Thus, it can be concluded that increasing the charge density of organic fraction leads to a decrease in toxin uptake at lower resin dosages. 70 Evidence of pore restrictions and pore blockage was observed in the presence of higher molecular weight fraction organics (i.e., SRHA), where the toxin removal decreased by eight-folds: from 80% in DI water to approximately 10% at a dosage of 10 mg/L. This is also evident from the DOC reduction where dosages >500 mg/L were required to obtain > 99% MCLR and >80% DOC reduction. Even though the sites available are approximately 50 times more abundant than the total contaminant concentration (meq), the pore restriction and blockage by higher molecular weight fractions leads to a significant reduction in the actual number of available sites (Mergen et al., 2008; Zhang et al., 2014). Further evidence of pore blocking is also observed in presence of SRHA, as depicted in Figure 6.2 (c). Here the LCOCD reveal that the removal of organic compounds from SRHA is significantly lower than SRNOM at similar resin dosages. The mercury porosimetery tests on the resin (Table C.1, Appendix C) reveals that micropores (<2 nm) occupy approximately 25% of the resin pore volume, mesopores (2-50 nm) occupy 55% and the remaining 20% of the pore volume is macropores (>50 nm). The median pore diameter of the resin is approximately 6 nm with an average pore diameter of 7.2 nm, indicating a high presence of primary mesopores (2- 8 nm) connected to micropores (32%) through inner pores and channels. It has been well documented by previous studies that the uptake of organic molecules involves diffusion through inner pores and channels of the resin and a similar behavior can be expected for MCLR (Bazri et al., 2016a, 2016b). At any given angle, the possible length of MCLR molecule is in the range of 2.94 nm to 1.4 nm (Sathishkumar et al., 2010). Therefore, the MCLR molecule has the capacity of accessing some of the resin micropores and the majority of resin mesopores (median: 6nm). The hydrodynamic diameter of Suwannee River Fulvic Acid (SRFA), a compound with an apparent average molecular weight of 1070 Da was reported in the range of 0.8 nm-5.8 nm, with the majority of fractions < 4 nm and an average diameter of 2 nm, determined through ultraviolet (UV)/ 71 flouroscence (FL) detection (Wells, 2015). Therefore, certain SRFA fractions block the resin micropores but the majority of the resin mesopores (6 nm pores) will not be blocked by SRFA since its maximum dimension is 5.8 nm. SRNOM, with an average molecular weight of 1030 Da, is expected to exhibit a similar hydrodynamic diameter. However, it can be assumed that the higher molecular weight fractions of SRHA (> 2000 Da, > 40% of initial DOC (LCOCD)) are capable of blocking the resin micropores and most of the resin mesopores, thereby resulting in the lower observed removal of MCLR and DOC. Thus, the competitive uptake of MCLR is not only dependent upon the charge density of the background NOM fraction, but also, and more importantly, on the NOM molecular weight distribution. In experiments evaluating MCLR removal in NOM-free waters, more than 80% removal (Co= 100 \u00C2\u00B5g MCLR/L) is observed in DI water for all the resin dosages tested (10, 50, 200, 500 and 1000 mg/L). Moreover, the toxin uptake capacity could reach as high as approximately 3800 \u00C2\u00B5g/g (resin dosage= 10 mg/L). For all the investigated conditions, the highest tested dosage of 1000 mg/L achieved more than 99% MCLR removal and more than 80% DOC reduction. IX treatment provided a final MCLR concentration of < 0.2 \u00C2\u00B5g/L, a value well below Health Canada guideline of 1.5 \u00C2\u00B5g/L and the EPA standard of 0.3 \u00C2\u00B5g/L (Health Canada, 2016; USEPA, 2015). Therefore, a complete toxin elimination is expected from source water at commercial dosages which are two to five times higher than the investigated dosages. Thus, IX process exhibits a great potential for practical applications targeting the removal of algal toxins and NOM from bloom laden waters. 72 Figure 6. 1 Effect of Resin Dosage on (a) MCLR Removal at pH 7 and (b) DOC Removal (Co = 100 \u00C2\u00B5g MCLR/L; 3 mg C/L; 24 Hours of Contact Time) *NOM Acronyms: SRNOM = Suwanee River Natural Organic Matte; SRHA = Suwanee River Humic Acid; PLFA = Pony Lake Fulvic Acid (International Humic Substances Society); 1 mL resin= 221mg dry weight of resin (24 hours in desiccator) 6.3 Isotherm Modelling In order to investigate the distribution of MCLR molecules on the resin and obtain the theoretical adsorption capacity, adsorption isotherm results were fitted to the Freundlich model (qe=Kf.Ce1/n; parameters are provided in Table 6.1. The high values of regression coefficients obtained for all the matrices indicate that equilibrium isotherms were well described by the Freundlich model (Figure 6.2 (a)). As the charge density of background NOM increased in the source waters, the values of 1/n decreased slightly from 0.63 to 0.51. The most important effect was observed on the values of Kf which declined from 28 ((mg/g)(L/mg)1/n) in buffered MQ water to 2.8 ((mg/g)(L/mg)1/n)) in the presence of PLFA (3 mg C/L). Increasing the charge density by testing SRNOM further decreased Kf from 2.8 to 0.6 (approximately a 5-fold decrease). Similarly, using 00.0050.010.0150.020.0250.030 1 2 3 4 5DOC Removal (meq/L)Resin Dosage (meq/L)SRNOM (3mg C/L)PLFA (3mg C/L)SRHA 3 mg C/L0204060801000 250 500 750 1000MCLR Removal (%)Resin Dosage (mg/L)MCLR-DI WaterPLFA 3mg/LSRNOM 3 mg/LSRHA 3mg/LCorresponding dose in mL/L1 2 3 4 5(b) (a) 73 SRHA, a higher molecular weight NOM fraction, further decreased the Kf from 2.8 to 0.5 (approximately a 6-fold decrease). Table 6.1 Adsorption Isotherm Constants and Thermodynamic Parameters for MCLR and NOM Fractions Parameters * MCLR isotherms SRNOM Isotherm with 3 mg C/L PLFA Isotherm with 3 mg C/L SRHA Isotherm with 3 mg C/L With DI Water With PLFA With SRNOM With SRHA *Kf (mg/g/(mg/L)1/n) 27.9 \u00C2\u00B1 1.60 2.83 \u00C2\u00B1 0.09 0.60 \u00C2\u00B1 0.03 0.47 \u00C2\u00B1 0.04 88.3 \u00C2\u00B1 4.76 80.5 \u00C2\u00B1 5.26 50.4 \u00C2\u00B1 8.76 *1/n 0.63 \u00C2\u00B1 0.07 0.59 \u00C2\u00B1 0.03 0.51 \u00C2\u00B10.03 0.77 \u00C2\u00B1 0.14 0.46 \u00C2\u00B1 0.10 0.71 \u00C2\u00B1 0.03 0.51 \u00C2\u00B1 0.11 *R2 0.97 0.96 0.99 0.98 0.99 0.97 0.98 a CoEBC (\u00C2\u00B5mol/L) - 0.23 0.16 0.09 *Fitted with the Freundlich equation: (qe=Kf.(Ce)1/n; data reported with 95th confidence intervals for the Kf and n values a Equivalent background concentration (EBC), estimated using the IAST model (Newcombe et al., 2002) 74 Figure 6.2 (a) Adsorption Isotherms of MCLR on A860 Resin at pH 7.0 (*Lines Indicate the Fit to a Freundlich isotherm), (b) LCOCD Results for Suwanee River NOM (SRNOM) and (c) Suwanee River Humic Acid (SRHA) at Different IX Dosages (200 mg/L and 500 mg/L * Error bars represent standard deviations of experimental replicates (n = 3) 0500100015002000250030003500400045000 20 40 60 80Uptake (\u00C2\u00B5g/g)Equilibrium Concentration (\u00C2\u00B5g/L)DI WaterPLFASRNOMSRHAIncreasing Charge DensityIncreasing Molecular Weight(a)-1.00E-031.00E-033.00E-035.00E-037.00E-039.00E-030 1000 2000Relative Intensity (I/Io)Apparent Molecular Weight (Da)SRNOMRaw WaterIX 200mg/LIX 500mg/LDI Water-1.00E-031.00E-033.00E-035.00E-037.00E-039.00E-031.10E-020 2000 4000Relative Intensity (I/Io)Apparent Molecular Weight (Da)SRHARawWaterIX 200mg/LIX 500mg/L(c)75 To further clarify the NOM effect, the equivalent background compound (EBC) method was used and the EBC parameters (KEBC, 1/nEBC and CoEBC) were determined as previously described (Newcombe et al., 2002). The EBC model has been used in the past to predict the adsorption isotherms of competing compounds under known isotherms of the target compound in presence and absence of NOM. (Ebie et al., 2001; Matsui et al., 2012). The multicomponent interactions are predicted using the ideal adsorbed solution theory (IAST), described below. Where, KEBC and 1/nEBC are the Freundlich single solute isotherm parameters for EBC and Km and 1/nm are single solute isotherm parameters for the micropollutant. CoEBC was determined using the following equations: \u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u0092 =\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u0092+\u00F0\u009D\u0091\u009E\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092(\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u0092+\u00F0\u009D\u0091\u009B\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009E\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009A\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0091\u009A) \u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009A (6.1) \u00F0\u009D\u0090\u00B6\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092 =\u00F0\u009D\u0091\u009E\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u0092+\u00F0\u009D\u0091\u009E\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092(\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u009A\u00F0\u009D\u0091\u0092+\u00F0\u009D\u0091\u009B\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u009E\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u009A\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0091\u009A)\u00F0\u009D\u0091\u009B\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00B5\u00F0\u009D\u0090\u00B6 (6.2) The 1/nEBC was held constant at 0.60 and the obtained EBC concentrations of respective compounds CoEBC in \u00C2\u00B5mol/L, are summarized in Table 6.1. The values of CoEBC were found to be in the order: SRNOM > PLFA> SRHA. Approximately 0.23 \u00C2\u00B5mol/L of SRNOM, a higher charge density fraction (~10 meq/ g C), competes with MCLR for the active sites, while it is 0.16 \u00C2\u00B5mol/L for PLFA (6.8 meq/g C), a 1.5-fold decrease in the concentration of competing compounds. Therefore, it can be observed that the concentration of competing compounds increases with increasing charge density. In the present study, the authors have presented the CEBC values in \u00C2\u00B5mol/L, while studies by other researchers with activated carbon have often reported CEBC values in mg/L with a hypothesis that the compounds with similar shape and size are the most competitive for active sites (Ebie et al., 2001). The hypothesis is true for studies with activated carbon since it is very well established that 76 micropollutants are sorbed in pores of similar sizes on activated carbon (Ebie et al., 2001; Matsui et al., 2012). However, sorption on IX resins is majorly based on electrostatic sorption (rather than hydrophobic), therefore, there is a high possibility that small NOM molecules (say 0.5 nm) can get sorbed into larger micropores (2 nm), mesopores (20 nm) or even marcopores (50 nm), since the resin surface is highly charged (0.8 meq/ mL). Consequently, using 1000 Da (~MCLR: 995 Da) as the definition of competition might be incorrect for IX resins. Nevertheless, for comparison, the CEBC values for SRNOM, PLFA and SRHA were determined based on the assumption of 1000 Da as the molecular weight of competing compounds. The calculated values for SRNOM, PLFA and SRHA were 0.12 mg/L, 0.08 mg/L and 0.05 mg/L, respectively. Noteworthy, the CEBC concentrations are < 5% of the initial NOM concentration (3 mg C/L) of these respective compounds. Therefore, only a small fraction of the total organic compounds present in the solution compete with MCLR for active sites. The EBC modelling results for SRNOM and PLFA were in good agreement with literature findings where competitive concentrations of < 10% of initial DOC have been reported for Geosmin, MIB and Atrazine (Cook et al., 2001; Pelekani and Snoeyink, 1999). However, there were few discrepancies with the higher molecular weight fraction, SRHA. Increasing molecular weight of NOM fractions resulted in a decrease in the concentration of competing compounds, where only 0.09 \u00C2\u00B5mol/L SRHA, the compound with the highest average molecular weight distribution (1520 Da), competed with MCLR for active sites (a 2.5-fold decrease compared to SRNOM (0.23 \u00C2\u00B5mol/L; 1030 Da). This is in contradiction to the observed experimental results depicted in Figure 6.1 (b), where lower toxin removal is observed in presence of SRHA. This discrepancy originates from the lack of knowledge about the particular NOM component which accounts for the majority of competitive effect and its adsorption characteristics. The crucial components of the NOM might 77 not show the same behavior as the bulk of the NOM and the use of bulk parameters such as DOC might be inappropriate under these circumstances (Newcombe et al., 2002). The NOM from PLFA (data not shown; 500 Da-1200 Da) and SRNOM were less diverse (majorly 600-2000 Da, Figure 6.2 (b)) than SRHA (500 Da-4000 Da, Figure 6.2(c)). Selection of bulk DOC as the isotherm of competing compound in a diverse NOM background for SRHA might have resulted in the discrepancy for the aforementioned reasons. Further reductions in the removal of MCLR in presence of SRHA can be attributed to pore blocking, which has been discussed in details in earlier sections, majorly since the EBC modelling does not incorporate pore blocking as a competitive mechanism (Newcombe et al., 2002). Based on observed results, it can be concluded that the EBC model does not provide a comprehensive answer to the competitive uptake on IX resins majorly in presence of diverse and higher molecular weight NOM fractions. Nevertheless, it provides a good estimate of competition under certain NOM matrices. Therefore, it can be concluded that the competition arises mostly from the differences in the charge densities of the respective NOM fractions, as determined by the EBC modelling, which is in agreement with literature findings on NOM uptake in absence of pore blocking phenomenon (Arias-paic et al., 2016; Bazri and Mohseni, 2016). 6.4 Effect of NOM Characteristics on Kinetics of MCLR Uptake As depicted in Figure 6.3 (a), a rapid uptake of around 60% removal was achieved within 5 minutes with the initial toxin concentration of 25 \u00C2\u00B5g/L in DI water, whereas 60% of MCLR was removed after 60 minutes in the presence of different NOM fractions. Rapid diffusion into the resin pores and availability of large number of vacant sites might have resulted in this initial rapid uptake in 78 NOM-free water (Bazri and Mohseni, 2016). In general, the 60 minutes toxin uptake followed the order MCLR-DI> MCLR-PLFA > MCLR- SRNOM > MCLR-SRHA. The experimental data were fitted to a pseudo-first order kinetics, a pseudo-second order kinetics and the intraparticle diffusion models to further investigate the MCLR removal process (Table 6.2). The highest k1 and k2 values were observed in the absence of NOM for all the investigated conditions. Overall, the pseudo-second order kinetics model was found to provide the best fit to the data (refer to Figures 6.3 (a) and 6.3 (b)). As observed earlier, the charge density and molecular weight distribution of NOM also play vital roles in the uptake of contaminants by the IX process. The pseudo second order kinetics rate constant, i.e., k2 PLFA, was nearly 1.5-fold greater than k2 MCLR-PLFA. This observation can be attributed to the faster diffusion of smaller molecular weight PLFA molecules (700 Da vs 995 Da) into the inner pores and channels of the resin resulting in a competitive uptake (Zhang et al., 2014). Increasing the charge density by using SRNOM (1.4 fold increase) resulted in a further decrease of the pseudo-second order rate constant value k2 MCLR-SRNOM (1.3 fold decrease). NOM characteristics also influence the MCLR equilibrium loading (pseudo-second order qe values). The observed qe values for MCLR decreased from approximately 130 \u00C2\u00B5g/g in DI water to approximately, 100 \u00C2\u00B5g/g in the presence of SRNOM (Table 6.2). As depicted in Figure 6.3 (a) and 6.3 (b), more than 40% of DOC was reduced for both PLFA and SRNOM water matrix during the first 30 minutes of operation. The qe values obtained for NOM fractions were orders of magnitude higher than the values for MCLR indicating a large coverage of resin sites. For instance, the qe value for PLFA surface concentration was approximately 6500 \u00C2\u00B5g/g (1.8 \u00C3\u009710-2 meq/meq) in comparison to that for the MCLR concentration of approximately 100 \u00C2\u00B5g/g (3\u00C3\u009710-5 meq/meq) (Table 6.2). A NOM fraction with higher molecular weight fractions (SRHA, 1.5 fold increase in 79 average molecular weight), reduced the qe values to approximately 50 \u00C2\u00B5g/g (1.5\u00C3\u009710-5 meq/meq) (a two-fold decrease). Noteworthy, as observed in Table 6.2, here the uptake of competing NOM compounds was approximately 3800 \u00C2\u00B5g/g (1.4 \u00C3\u009710-2 meq/meq) which was approximately 1.5-fold lower than the DOC uptake capacities observed in the presence of lower molecular weight fractions (1400 Da for SRHA vs 1030 Da for SRNOM and 700 Da for PLFA), supporting the previous arguments made on evidences of pore blockage. In the presence of larger molecules most of the resin macropores and primary mesopores are blocked by the uptake of charged bulky organic groups resulting in underutilization of smaller sized mesopore volume and the associated surface area of the resin (Ebie et al., 2001; Li et al., 2003; Pelekani and Snoeyink, 1999). The intraparticle diffusion model (IPD; Section 4.4) was used to examine the impact of film diffusion and intra-particle diffusion mechanism (Weber and Morris, 1963). The calculated values of kid were 13.1, 12.8 and 7.3 \u00C2\u00B5g/g min1/2, for DI water, PLFA and SRNOM with 25 \u00C2\u00B5g/L initial MCLR concentration, respectively. In the presence of background SRHA, the value of kid was reduced further to 4.1 \u00C2\u00B5g/g min1/2. According to the model assumptions, the plot of qt vs t1/2 should be linear if intra-particle diffusion is involved in the process (Hu et al., 2015). The R2 values of the intra-particle diffusion model for all conditions tested listed in Table 6.2 are high (>0.9), indicating that intra-particle diffusion is involved in the uptake process. The rate-controlling step was further investigated using the dimensionless Biot number (Bi) which is the ratio of internal mass transfer (i.e., pore diffusion) to external mass transfer (i.e. film diffusion) resistances (Ko et al., 2001). Parameters were estimated by fitting the pore diffusion (PDM) and film diffusion modelling (FDM) equations to the kinetic data using nonlinear optimization schemes as reported elsewhere (Bazri and Mohseni, 2016). Resin beads were 80 assumed to be spherical with an average radius (Rp) of 375 \u00CE\u00BCm (resin size range reported by the manufacturer, Rp : 150 \u00E2\u0080\u0093 600 \u00CE\u00BCm). The Biot numbers for MCLR in DI water are <1 for initial MCLR concentrations of 25 \u00C2\u00B5g/L and below at a resin dosage of 200 mg/L and below (Table 6.2, See details chapter 4, section 4.6, meq/meq of 4\u00C3\u009710-8 and below). The Biot numbers obtained for MCLR in the presence of PLFA, SRHA and SRNOM were 59, 61 and 70, respectively, indicating the uptake kinetics is controlled primarily by intraparticle diffusion in the presence of NOM fractions. The uptake of NOM fractions was also controlled by intraparticle diffusion (Bi > 500) for all tested conditions with an initial resin dosage of 200 mg/L (~ 0.9 mL/L). Note that in NOM-free waters, both intraparticle diffusion and film diffusion govern the MCLR removal process (Bi ~ 1). In the presence of PLFA and SRNOM, about 50% reduction in toxin removal was observed after twenty minutes of contact time with 30% reduction in the DOC and UV254 absorbing compounds. Increasing the contact time to 60 minutes further resulted in 70% reduction in MCLR with approximately 55% DOC reduction. Therefore, it can be concluded that in the presence of smaller molecular weight and highly charged compounds, increasing the contact time or increasing the resin dosage, can improve the toxin uptake. However, for SRHA increasing the contact time did not improve the toxin and DOC removal to a significant extent due to excessive pore blockage produced by higher molecular weight fractions. Therefore, for matrices with higher molecular weight NOM fractions, an increase in resin dosage would be a better option as opposed to increasing the contact time. It should be noted that a significant portion of the DOC and UV254 absorbing compounds is also removed by the resin which can favor the performance of downstream oxidation processes. Hence, IX could simultaneously remove MCLR as well as serve as pretreatment for downstream oxidative processes. 81 Figure 6.3 (a) MCLR Uptake as a Function of Contact Time *(Co= 25 \u00C2\u00B5g MCLR/L, 200 mg/L Resin and DOC of 3mg C/L), (b) TOC and UVA Reduction for 60 Minutes of Operation * Error bars represent standard deviations of experimental replicates (n = 3) 01020304050607080901000 20 40 60Percent MCLR Removal (%)Time (minutes)MCLR-DIPLFA 3mg/LSRNOM 3 mg/LSRHA 3 mg/L(a)0102030405060708000.20.40.60.811.20 10 20 30 40 50 60UVA Removal (%)TOC Concentration (C/Co)Time (Minutes)MCLR-PLFAMCLR-SRNOMMCLR-SRHA(b) 82 Table 6.2 Kinetic Parameters of MCLR Uptake in Presence of Different NOM Fractions Model Parameters* MCLR Removal NOM Removal DI SRNOM PLFA SRHA PLFA SRNOM SRHA Pseudo-first Order qe 97 \u00C2\u00B1 2 52 \u00C2\u00B1 1 94 \u00C2\u00B1 2 22 \u00C2\u00B1 3 4214 \u00C2\u00B1 50 3789 \u00C2\u00B1 29 2296 \u00C2\u00B1 79 kl 0.060 \u00C2\u00B1 0.002 0.055 \u00C2\u00B1 0.003 0.058 \u00C2\u00B1 0.003 0.029 \u00C2\u00B1 0.010 0.045 \u00C2\u00B1 0.001 0.048 \u00C2\u00B1 0.001 0.053 \u00C2\u00B1 0.001 R2 0.98 0.89 0.98 0.97 0.96 0.98 0.97 Pseudo-second Order k2 0.32 \u00C2\u00B1 0.01 0.25 \u00C2\u00B1 0.01 0.29 \u00C2\u00B1 0.02 0.23 \u00C2\u00B1 0.01 0.37 \u00C2\u00B1 0.03 0.34 \u00C2\u00B1 0.02 0.29 \u00C2\u00B1 0.03 qe 127 \u00C2\u00B1 5 101 \u00C2\u00B1 4 115 \u00C2\u00B1 7 49 \u00C2\u00B1 4 7873 \u00C2\u00B1 87 9887 \u00C2\u00B1 63 4785 \u00C2\u00B1 73 R2 0.99 0.99 0.99 0.99 0.98 0.97 0.96 Intra-particle diffusion kid 13.1 7.27 12.78 4.14 845 \u00C2\u00B1 34 713 \u00C2\u00B1 58 648 \u00C2\u00B1 88 C 13.3 42.9 15.9 18.6 86.2 \u00C2\u00B1 7 85.7 \u00C2\u00B1 4 93 \u00C2\u00B1 9 R2 0.97 0.98 0.97 0.90 0.94 0.93 0.91 Biot Number Bi 0.90 59 61 70 670 752 872 *Data reported with 95th confidence interval for the respective kinetic parameters 6.5 Effect of pH Studies were performed at different pH values ranging from 4 to 10 at a resin dosage of 200 mg/L (~ 0.9 mL/L) to understand the effect of pH on the removal of Microcystin-LR (Figure 6.4 (a)). In general, the uptake followed the trend: pH 4 ~ pH 5 ~ pH 6 ~ pH 7 > pH 8 > pH 9 > pH 10 for all the water matrices (p<0.01). MCLR molecule contains various pH-dependent charged moieties. The pKa of MCLR is 2.09 and 2.19, indicating that the cationic species [(COOH)2(NH2+)] are produced at pH < 2.09 and anionic species [(COO-)2(NH2+)] are produced at pH > 2.19 (Pavagadhi 83 et al., 2013). Therefore, above pH = 2.19, MCLR molecule is negatively charged. On the other hand, the resin carries cationic charges below pH = 7.5, since the pHzpc (point of zero charge) of the resin is approximately of 7.5 (data not shown), which is similar to the reported values of other strongly basic resins (Graf et al., 2014; Moreno-Castilla et al., 2000). Consequently, for aqueous matrices with pH < 7.5, electrostatic interactions may play a major role in the MCLR removal process. The obtained slopes for the released chloride ions (meqs), as observed at pH= 7 (Figure 6.4 (a)), is 1.04 \u00C2\u00B1 0.02 (p >0.05), and very close to the theoretical value of 1 (i.e. X=Y). A similar behavior for the uptake of NOM fractions is observed in Figure 6.4 (b). The obtained slopes for the released chloride ions (meqs) for PLFA, SRNOM and SRHA are 1.02 \u00C2\u00B1 0.01, 1.04 \u00C2\u00B1 0.03 and 1.01 \u00C2\u00B1 0.01, respectively. Therefore, it can be concluded that the uptake mechanism of MCLR via IX process is primarily through electrostatic exchange of ions with minimal influence of hydrophobic adsorption. As depicted in Figure 6.4 (a), a stable MCLR uptake is observed for pH = 4 to pH = 7, while, a significant reduction in the MCLR uptake capacity is observed for pH= 8 to pH = 10. This behavior can be explained by the fact that at pH> pHzpc (pHzpc~ 7.5), hydrophobic interactions impact the MCLR removal process (Li and SenGupta, 2000). The n-octanol/water distribution ratio log Dow for MCLR remains positive in the acidic conditions and decreases from 2.2 at pH = 1 to -1.7 at pH= 10 (significantly decreasing beyond pH >7.5), indicating a considerable change in the hydrophobicity of the MCLR molecule under basic pH conditions (De Maagd et al., 1999). Consequently, at pH > 7.5, reduction in the electronegativity of the resin surface combined with change in the MCLR hydrophobicity result in a significant decrease in the uptake process. Note that most of the surface waters have pH in the range of 6 -7.5 and, therefore, IX exhibits a great potential for simultaneous DOC and MCLR from these water matrices under these conditions. 84 Figure 6.4 Effect of pH on (a) Uptake of MCLR (b) Uptake of DOC and Equivalent Chloride Release Studies at pH = 7 * Error bars represent standard deviations of experimental replicates (n = 4) 050100150200MCLR uptake (\u00C2\u00B5g/g)pHMCLR-DI WaterMCLR-PLFAMCLR-SRNOMMCLR-SRHA(a)4 5 6 7 8 9 10024681012141618DOC uptake (mg/g)pHPLFASRNOMSRHA4 5 6 7 8 9 10(b)00.00050.0010.00150.0020.00250 0.001 0.002Chloride meqsMCLR meqsMCLR-DIWaterX=Y 00.020.040.060.080 0.02 0.04 0.06 0.08Chloride meqsTotal meqs (NOM)PLFANOMSRHA85 6.6 Comparative Studies on Surface Water To assess the effectiveness of the resin with natural waters, Eagle Lake (EL) raw water and treated water (processed through 50 \u00C2\u00B5m screening followed by coagulation (Poly aluminum chloride (low dose < 20 mg/L) and sodium hydroxide (25%)), flocculation and ultra-filtration (UF) membrane treatment) was collected from the water treatment plant of West Vancouver city. The surface water had a DOC of 3.2 mg/L and a pH = 7.0 \u00C2\u00B1 0.3, with 5 mg/L sulphate ions in the raw water. The rationale for the selection of this specific water source was that it was an actual drinking water source and had similar NOM characteristics with the organic fractions that were used previously in the laboratory studies (Size Exclusion Chromatography data not shown). SRNOM and the EL Raw Water NOM had similar molecular weight distributions with averages of 1030 Da and 1050 Da, respectively (data not shown; p > 0.05). Similarly, PLFA and EL-UF treated water NOM had average molecular weights of 680 Da and 700 Da, respectively (data not shown, p > 0.05). The pseudo second-order kinetic parameters and removals measured after 60 minutes of contact time at a dosage of 200 mg/L are presented in Table 6.3. Comparative results were obtained for the laboratory and surface water samples. For instance, the pseudo-second order qe values for MCLR uptake from EL-UF and PLFA were approximately 112 and 117 \u00C2\u00B5g/g, respectively. The kinetic rate constants for these aqueous matrices were also not statistically different (p> 0.05). In general, the k-values decreased with increasing background DOC concentration. For instance, the k2 values for Eagle Lake filtered water were 1.5 fold higher than the Eagle Lake raw water (DOC ~3 fold higher). A similar trend of decreasing k values with increasing DOC was observed for surface waters from Shawnigan Lake, Middle River and Van Anda Improvement District (VAID). The EL raw water had 5 mg/L of sulphate which was reduced to 1.2 mg/L after sixty minutes of operation (a 75% decrease). Sulphates are well known for their competition with NOM for active 86 sites (Ates and Incetan, 2013). Consequently, only 63 % MCLR and 38 % DOC removals were obtained in the EL water (a 1.2-fold decrease compared to SRNOM with same initial DOC). As a reference, the pseudo-second order kinetic rate constant values for MCLR decreased 4-folds in DI water spiked with 5 mg/L sulphate ions without NOM (data not shown), while it decreased only by 1.5-fold with 5 mg C/L background SRNOM in absence of sulphates (data not shown). At low sulphate concentrations (<1 mg/L), as in the case of EL-UF filtered water, a high correlation between the laboratory and surface water results were observed. These results show that the resins are effective for the removal of MCLR in the presence of NOM and inorganic compounds from surface waters. Also, the results from laboratory scale experiments are comparable to those from the surface water. 87 Table 6.3 Kinetic Parameters of MCLR Uptake in Presence of Surface Water Water matrix Initial Concentrations MCLR DOC (\u00C2\u00B5g/L) (mg/L) Initial pH Pseudo Second Order Parameters for MCLR Biot Number MCLR Removal (%) DOC Removal (%) k2 (\u00C2\u00B5g/g min) qe (\u00C2\u00B5g/g) R2 DI water 25.0 \u00C2\u00B1 1.5 < 0.2 7.0 \u00C2\u00B1 0.2 0.32 \u00C2\u00B1 0.01 127 \u00C2\u00B1 5 0.99 0.90 92 - PLFA 25.0 \u00C2\u00B1 1.5 1.5 \u00C2\u00B1 0.1 7.0 \u00C2\u00B1 0.2 0.29 \u00C2\u00B1 0.02 117 \u00C2\u00B1 6 0.99 61 79 62 SRNOM 25.0 \u00C2\u00B1 1.5 3.0 \u00C2\u00B1 0.1 7.0 \u00C2\u00B1 0.2 0.25 \u00C2\u00B1 0.01 101 \u00C2\u00B1 4 0.99 59 72 45 Eagle Lake Raw Water 25.0 \u00C2\u00B1 1.5 3.2 \u00C2\u00B1 0.1 7.0 \u00C2\u00B1 0.3 0.22 \u00C2\u00B1 0.01 97 \u00C2\u00B1 3 0.98 68 63 38 Eagle Lake Filtered Water 25.0 \u00C2\u00B1 1.5 1.3 \u00C2\u00B1 0.2 6.9 \u00C2\u00B1 0.3 0.31 \u00C2\u00B1 0.01 112 \u00C2\u00B1 5 0.98 58 77 63 Shawnigan Lake Water 25.0 \u00C2\u00B1 1.5 4.3 \u00C2\u00B1 0.3 7.0 \u00C2\u00B1 0.3 0.19 \u00C2\u00B1 0.01 91 \u00C2\u00B1 4 0.98 - 54 43 Middle River Water 25.0 \u00C2\u00B1 1.5 5.3 \u00C2\u00B1 0.2 7.2 \u00C2\u00B1 0.2 0.18 \u00C2\u00B1 0.02 87 \u00C2\u00B1 3 0.99 68 52 51 Van Anda Water 25.0 \u00C2\u00B1 1.5 5.8 \u00C2\u00B1 0.6 7.6 \u00C2\u00B1 0.3 0.15 \u00C2\u00B1 0.01 77 \u00C2\u00B1 4 0.98 - 44 37 6.7 Conclusion In the preceding studies, IX effectively removed MCLR and DOC from aqueous matrices. A dosage of 1000 mg/L (4 mL/L) was sufficient to reduce the initial toxin concentration of 100 \u00C2\u00B5g/L to below the WHO Guideline of 1 \u00C2\u00B5g/L for all investigated resin dosages. However, loading NOM onto the IX resin reduced the MCLR uptake capacity from approximately 3800 \u00C2\u00B5g/g to 800 \u00C2\u00B5g/g. We found that, in the presence of NOM fractions, toxin removal kinetics are controlled by internal pore diffusion. Charge density and molecular weight distribution of the NOM matrix also impact 88 resin performance, with the latter being the strongest influencing factor at lower resin dosages (200 mg/L ~ 1 mL/L and below). The equivalent background concentration (EBC) model effectively demonstrated the competition between the NOM fractions and MCLR on the IX resins. The lower molecular weight charged NOM fractions (<1000 Da) compete with MCLR molecules for active sites in the interior of the resin while the larger molecular weight compounds reduce the available active sites by pore blocakge. However, the EBC model also had certain limitations in estimating the competitive NOM concentrations when it came to the pore blockage mechanism. We also found that MCLR and DOC removal performance remain consistent in the pH=4 to pH=7 range, and the addition of inorganic ions like sulphates reduce toxin uptake at any given resin dosage. Finally, we obtained comparable results with synthetic laboratory water and surface water as long as the NOM chracteristics and sulphate concentrations remained consistent. We next sought to assess resin performance over multiple treatment cycles, toxic brine management and the potential adeverse impacts of algae cells on treatment performance. Considering that inorganic anions co-exist with NOM in natural waters the next chapter also evaluates the impact of the most commonly occuring inorganic ions (sulphates, nitrates and bicarbonates) on the uptake of MCLR and NOM via IX process. . 89 Chapter 7: Simulated Environmental Conditions: Effect of Inorganic Ions and Multiple Loading Tests 7.1 Introduction The composition of aqueous matrices, specifically the presence anionic species, is expected to impact the uptake and the sorption kinetics of the adsorbent (Bazri and Mohseni, 2016; Bolto et al., 2004; Boyer and Singer, 2006; Cornelissen et al., 2010; Li and Sengupta, 2000). Based on their affinities, these anionic species compete for a limited number of active sites on the resin and reduce the uptake of cyanotoxins. Anion exchange resins are reported to have a higher affinity for sulfates compared to organic components. Hence, high sulfate ion concentrations adversely impact toxin removal by IX (Ates and Incetan, 2013; Hu et al., 2014; Verdickt et al., 2011; Willison and Boyer, 2012). Likewise, nitrates do not strongly compete with organics, but can influence the removal of organic matter at high concentrations (Ates and Incetan, 2013; Kim et al., 2012; Li and Sengupta, 2001; Pavagadhi et al., 2013). Moreover, inorganic ions such as bicarbonates and chlorides have lower preference as compared to organic molecules (Bolto et al., 2002a; Croue and Humbert, 2005; Ding et al., 2012; Hsu and Singer, 2010; Humbert et al., 2005; Li and SenGupta, 2000). Considering that inorganic anions co-exist with NOM in natural waters it would be valuable to understand the extent of their influence on the simultaneous removal of NOM and MCLR by IX resins. Consequently, the main objective of this study was to investigate the efficiency IX resins for the removal of MCLR in presence of different inorganic species (sulphates, nitrates and bicarbonates). We looked at the kinetics to understand the resins\u00E2\u0080\u0099 uptake process and to determine rate controlling step. Additionally, we also examined the impact of multiple loading cycles and 90 resin regeneration. Since the IX process has proven efficient for microcystin uptake, it is expected that the regeneration process would produce a toxic spent brine that shall be treated prior to its disposal in the environment. To date, no strategies have been proposed to address this challenge. Consequently, studies were performed to optimize the regeneration process and to evaluate if it is possible to degrade toxins in the spent brine. 7.2 Impact of Background NOM Fractions and Sulphate Ions on MCLR Uptake It has been well established that anion exchange resins have a higher affinity for sulfates compared to organic components (Ates and Incetan, 2013; Hu et al., 2014; Verdickt et al., 2011; Willison and Boyer, 2012). Hence, studies were performed to investigate the effect of sulphate ions on the uptake kinetics of MCLR. Figure 7.1 (a) depicts the uptake kinetics of MCLR in presence of different NOM fractions at an initial sulphate ion concentration of 5 mg/L (Sulphate ion concentration in Eagle Lake raw water). MCLR uptake (pseudo second order qe) by IX resins decreased from 127 \u00C2\u00B5g/g in DI water to 96 \u00C2\u00B5g/g in presence of 5 mg/L sulphate ion concentration. This reduction can be attributed to the higher charge density of sulphate ions that resulted in a reduction of the available exchange sites on the resin surface for MCLR uptake. Addition of different NOM fractions resulted in a further decrease and the MCLR uptake can be inversely related to the average molecular weight distribution of the background water matrix. The uptake in presence of PLFA (760 Da) was approximately 76 \u00C2\u00B5g/g (pseudo second order qe) and was nearly double to that observed in case of a higher molecular weight fraction, SRHA (1450 Da), which was approximately 32 \u00C2\u00B5g/g. In presence of SRNOM, the uptake was 64 \u00C2\u00B5g/g and was nearly half of that observed in NOM free water. Therefore, presence of NOM and Sulphates in the water matrix results in a further decrease in the MCLR uptake. 91 Figure 7. 1 Kinetics of MCLR Uptake in Presence of Different NOM Fractions (a) In Presence of 5 mg/L Sulphate Ions MCLR (b) Reduction in DOC and Sulphate Ion Concentration (*Lines indicate a fir to pseudo-second order kinetics, initial DOC 3 mg C/L and sulphate 5 mg/L; Error bars represent standard deviations of experimental replicates (n = 3)) More than 50% reduction in sulphate ion concentrations was observed in presence of DI water after a contact time of 60 minutes (Figure 7.1). Addition of PLFA further reduced the sulphate 0204060801001200 10 20 30 40 50 60MCLR Uptake (ug/g)Time (minutes)DI WaterPLFA 3mg C/LSRNOM 3 mg C/LSRHA 3 mg C/L010203040506070809010001020304050600 10 20 30 40 50 60DOC Remaining (%)Sulphate Uptake (%)Time (minutes)DI WaterPLFA 3 mg C/LSRNOM 3mg C/LSRHA 3mg C/L(b) (a) 92 uptake to 45%, with approximately 50% reduction in DOC. Therefore, it can be inferred that the smaller molecular weight fractions were able to diffuse into the inner pores and channels of the resins and competed with sulphates and MCLR molecules for uptake sites. However, as observed, competition with higher charge density NOM fraction (SRNOM), was more significant and sulphate ions uptake reduced to < 40% with approximately 40% reduction in DOC. This resulted in 90% site occupancy (meq/meq, Table 7.1) in contrast to 82% site occupancy in presence of PLFA. Thereby, a further decrease in the MCLR uptake resulting from a stronger competition. In presence of larger molecular weight humic acid fraction, SRHA (1450 Da), only 20% reduction in DOC and 25 % reduction in sulphate ion concentrations was observed. The total site occupancy was approximately 33% of the total number of available sites with an equilibrium attainment in 40 minutes for all the components. Therefore, providing a further evidence of pore blockage and size exclusion phenomenon in presence of humic acid fractions. Table 7.1 IX Site Occupancy Under Different Water Matrix Parameter DI water with 25 \u00C2\u00B5g MCLR/L In presence of 5 mg/L Sulphate Ions and 25 \u00C2\u00B5g MCLR/L DI water PLFA 3 mg C/L SRNOM 3 mg C/L SRHA 3 mg C/L Percentage Site Occupancy (meq/meq) 0.1 71.6 82.3 84.9 33.1 (*Initial toxin concentration of 25 \u00C2\u00B5g/L, DOC of 3 mg C/L and resin dosage of 200 mg/L (0.9 mL/L)) 93 7.3 Simulated Environmental Conditions: Effect of Inorganic Ions Previous studies demonstrated that sulphate ions impacted the uptake of MCLR and NOM fractions via IX process. Consequently, the effect of other inorganic ions (nitrates and bicarbonates) on the uptake of MCLR and SRNOM was also studied and the observed responses in presence and absence of SRNOM after 60 minutes of contact time are shown in Table 7.2. The pseudo second order rate constant values for MCLR declined in the order: kDI water > kAlkalinity > kNitrates >kSulphates Anion exchange resins are reported to have higher affinity towards sulfate compared to organic components and hence the presence of high sulfate ion concentration adversely impacts the toxin removal by IX (Hsu and Singer, 2010; Ishii and Boyer, 2011; Mergen et al., 2008; Tan and \u00C3\u0083, 2007). This was evident from the decline in the values of the kinetic rate constants as the initial sulfate ion concentration was increased from 0 to 100 mg/L. Further addition of 3 mg/L NOM in presence of 100 mg/L sulphates resulted in a significant decrease in the toxin removal and approximately 70% reduction in uptake was observed (Figure 7.2). Nitrates are not strong competitors for organics, however, it has been reported that at high concentrations they influence the removal of organic matter (Boyer and Singer, 2008). A similar behavior was observed when the nitrate ion concentration was 5 and 30 mg/L. Here, the toxin uptake was reduced to 70% in comparison to more than 90% in DI water. Further increase in the nitrate ion concentration significantly decreased the rate constant (nearly half) and only 50 % of toxin removal was observed after 60 minutes of contact time. Moreover, alkalinity did not impact the toxin uptake and MCLR uptake for the different alkalinity conditions were nearly constant. A similar trend was observed for the DOC removal as well where the DOC removal percentage under different alkalinity 94 conditions was not statistically significant. This is in agreement with literature findings that suggest that bicarbonates and chlorides have equal affinity for anion exchangers and lower preference as compared to organic molecules (Boyer and Singer, 2008; Walker and Boyer, 2011). Figure 7. 2 MCLR Kinetics in Presence of SRNOM and Inorganic Ions (*Initial DOC 3 mg C/L and 25 \u00C2\u00B5g MCLR/L; Error bars represent standard deviations of the respective parameters (n=3)) 01020304050607080901000 10 20 30 40 50 60MCLR Uptake (ug/g)Time (minutes)DI WaterAlkalinity 30 mg/LNitrate 30 mg/LSulphate 30 mg/L95 Table 7.2 Influence of Various Concentrations of Anions on Uptake of MCLR and DOC at Resin Dosage of 200 mg/L Ions Concentration of ions (mg/L) MCLR Removal (%) Initial DOC (0 mg/L) MCLR Removal (%) Initial DOC (3 mg/L) DOC Removal (%) Initial DOC (3 mg/L) Uptake amount (%) Pseudo second order rate constants Uptake amount (%) Pseudo second order rate constants Uptake amount (%) k2 qe R2 k2 qe R2 None 0 90 0.32 \u00C2\u00B1 0.01 127 \u00C2\u00B1 5 0.99 71 0.25 \u00C2\u00B1 0.02 101 \u00C2\u00B1 4 0.99 56 Sulphate 5 66 0.09 \u00C2\u00B1 0.01 96 \u00C2\u00B1 5 0.93 49 0.09\u00C2\u00B1 0.01 72 \u00C2\u00B1 5 0.94 44 30 49 0.08 \u00C2\u00B1 0.01 72 \u00C2\u00B1 4 0.93 37 0.08\u00C2\u00B1 0.01 55 \u00C2\u00B1 3 0.92 32 100 40 0.08 \u00C2\u00B1 0.01 59 \u00C2\u00B1 3 0.94 21 0.07\u00C2\u00B1 0.01 30 \u00C2\u00B1 2 0.92 29 Nitrate 5 76 0.21 \u00C2\u00B1 0.01 108 \u00C2\u00B1 5 0.95 55 0.09\u00C2\u00B1 0.01 79 \u00C2\u00B1 4 0.96 42 30 71 0.17\u00C2\u00B1 0.01 102 \u00C2\u00B15 0.96 46 0.08\u00C2\u00B1 0.01 70 \u00C2\u00B1 4 0.96 38 100 56 0.09\u00C2\u00B1 0.01 80 \u00C2\u00B15 0.96 32 0.08\u00C2\u00B1 0.01 51 \u00C2\u00B1 2 0.95 32 Alkalinity 5 84 0.25\u00C2\u00B1 0.01 116 \u00C2\u00B15 0.97 66 0.09\u00C2\u00B1 0.01 96 \u00C2\u00B14 0.95 56 30 83 0.21\u00C2\u00B1 0.01 114 \u00C2\u00B15 0.96 64 0.09\u00C2\u00B1 0.01 95 \u00C2\u00B1 5 0.96 55 100 82 0.19\u00C2\u00B1 0.01 112 \u00C2\u00B16 0.97 62 0.09\u00C2\u00B1 0.01 95 \u00C2\u00B1 5 0.95 54 The pseudo second order qe values for MCLR also declined in presence of high sulphates and nitrate concentrations where the values decreased form 127 \u00C2\u00B5g/g in DI water to 59 and 80 \u00C2\u00B5g/g in 96 presence of 100 mg/L of nitrates and sulphates, respectively. Further addition on NOM to these matrices reduced the values to 51 and 30 \u00C2\u00B5g/g, respectively, in presence of 3 mg C/L. Moreover, alkalinity did not impact the toxin uptake and MCLR uptake for the different alkalinity conditions were nearly constant in DI water. A similar trend was observed for the DOC removal as well where the DOC removal percentage under different alkalinity conditions was not statistically significant (p>0.5). This shows that sulphates and nitrates might reduce the uptake capacity of MCLR, while in the presence of alkalinity the capacity would remain more or less unaffected. This is in agreement to previous research studies that have indicated that sulphates and nitrates affect the adsorbent performance for NOM and micropollutant uptake by competing for the limited available active sites (Ates and Incetan, 2013; Croue and Humbert, 2005). Thus, for source waters with high sulphate and nitrate concentrations, a higher resin dosage would be sufficient for simultaneous MCLR and NOM removal. 7.4 Impact of Inorganic Ions on MCLR Uptake in Presence of Low-Molecular Weight Organics The impact of different inorganic ions on the uptake of MCLR is depicted in Figure 7.3. In general, at an inorganic ion concentration of 5 mg/L with 3 mg C/L of PLFA, the toxin uptake followed the order Sulphates < Nitrates < Alkalinity. Indicating that, the sulphate ions were the most competitive, followed by nitrates and alkalinity, a behavior similar to that observed with NOM. This was also evident from the pseudo second order rate constant values which followed the order: k alkalinity > k NO3- > k SO42-(Table 7.3). However, in contrast to the uptake capacity of approximately 70 \u00C2\u00B5g/g, observed in presence of NOM, approximately 85 \u00C2\u00B5g/g of MCLR was removed by the IX resin in presence of background PLFA with the same 3 mg C/L concentration. Increasing the 97 concentration of the inorganic species from 0 mg/L to 100 mg/L resulted in a further decrease in the rate constants and resin uptake capacities for MCLR. For instance, the k NO3- value at 5 mg/L nitrate ion concentration was nearly half of the value in inorganic free water. A further increase in the nitrate ion concentration to 100 mg/L resulted in a threefold decrease in the value of the kinetic rate constants. A similar trend can be observed for alkalinity and sulphate ions as well. Approximately40 % reduction in the inorganic ion concentration and >35% DOC reduction was observed for all water matrices containing sulphates and nitrates. This high uptake of background matrix ions resulted in a significant decrease in the number of active exchange sites on the IX resin surface for MCLR removal. Consequently, MCLR uptake decreased from approximately 80% in inorganic free water to less than 35% in presence of sulphates and nitrates. A discrepancy from this trend can be observed in the presence of alkalinity where a high DOC reduction (> 60%) is accompanied with a high reduction in alkalinity (> 40%) and MCLR (> 70%). This can be attributed to the phenomeon of \u00E2\u0080\u0098salting-out\u00E2\u0080\u0099, where organic molecules shrink and reorganize under high chloride and bicarbonate ion concentrations, easing the diffusion into the inner pores and channels of the resin. Figure 7.3 MCLR Uptake Kinetics in Presence of PLFA and Inorganic Ions 0204060801001200 10 20 30 40 50 60MCLR Uptake (ug/g)Time (minutes)DI WaterAlkalinity 30 mg/LNitrate 30 mg/LSulphate 30 mg/L98 (*Initial DOC 3 mg C/L and MCLR 25 \u00C2\u00B5g/L; Error bars represent standard deviations of the respective parameters (n=3)) Table 7.3 Influence of Various Concentrations of Anions on Uptake of MCLR in Presence of PLFA 3 mg C/L at Resin Dosage of 200 mg/L Ions Concentration of ions (mg/L) MCLR Removal (%) Initial DOC (3 mg/L) DOC Removal (%) Inorganic Ion Removal (%) Uptake amount (%) Pseudo second order rate constants Uptake (%) Uptake (%) K2 qe R2 None 79 0.31 \u00C2\u00B1 0.1 124 \u00C2\u00B1 6 0.99 55 - Sulphate 5 64 0.14 \u00C2\u00B1 0.1 86 \u00C2\u00B1 4 0.96 53 42 30 44 0.12 \u00C2\u00B1 0.1 67 \u00C2\u00B1 2 0.94 42 37 100 29 0.09 \u00C2\u00B1 0.1 41 \u00C2\u00B1 2 0.95 36 34 Nitrate 5 72 0.15 \u00C2\u00B1 0.1 94 \u00C2\u00B1 4 0.97 55 39 30 51 0.12 \u00C2\u00B1 0.1 74 \u00C2\u00B1 2 0.96 45 37 100 35 0.10 \u00C2\u00B1 0.1 52 \u00C2\u00B1 2 0.94 39 35 Alkalinity 5 74 0.16 \u00C2\u00B1 0.1 116 \u00C2\u00B1 7 0.93 64 44 30 72 0.13 \u00C2\u00B1 0.1 113 \u00C2\u00B1 7 0.94 63 42 100 71 0.11 \u00C2\u00B1 0.1 109 \u00C2\u00B1 6 0.94 61 41 *Data reported with 95% confidence intervals of the respective compounds 99 7.5 Impact of Inorganic Ions on MCLR Uptake in Presence if Higher-Molecular Weight Organics The toxin uptake behavior followed a similar pattern as observed earlier in the presence of SRHA, with 3 mg C/L (Figure 7.4). Sulphates were the most competitive followed by nitrates and alkalinity. However, the toxin uptake was reduced up to 14% in presence of inorganic ions, a nearly 6 fold decrease from the value observed in organic-free water (Table 7.4). Further, the pseudo second order rate constant values were nearly half of the observed values for PLFA at similar concentrations. The uptake capacity of the resins also decreased from 85 \u00C2\u00B5g/g in PLFA to 51 \u00C2\u00B5g/g in SRHA, and further decreased to approximately 15 \u00C2\u00B5g/g and 16 \u00C2\u00B5g/g on addition of 10 mg/L of sulphates and nitrate ions, respectively. A strong competition by the highly charged sulphate and nitrate ions can thus be observed here. 100 Table 7.4 Influence of Various Concentrations of Anions on Uptake of MCLR in Presence of PLFA 3 mg C/L at Resin Dosage of 200 mg/L Ions Concentration of ions (mg/L) MCLR Removal (%) Initial DOC (3 mg/L) DOC Removal (%) Inorganic Ion Removal (%) Uptake amount (%) Pseudo second order rate constants Uptake (%) Uptake (%) K2 qe R2 None 52 0.23 \u00C2\u00B1 0.1 51 \u00C2\u00B1 3 0.99 15 - Sulphate 5 24 0.07\u00C2\u00B1 0.01 32 \u00C2\u00B1 2 0.97 12 22 30 16 0.06\u00C2\u00B1 0.01 22 \u00C2\u00B1 1 0.97 10 17 100 14 0.05\u00C2\u00B1 0.01 15 \u00C2\u00B1 2 0.96 8 13 Nitrate 5 28 0.07\u00C2\u00B1 0.01 34 \u00C2\u00B1 4 0.93 14 19 30 21 0.06\u00C2\u00B1 0.01 25 \u00C2\u00B1 4 0.94 12 16 100 15 0.06\u00C2\u00B1 0.01 17 \u00C2\u00B1 2 0.92 9 15 Alkalinity 5 29 0.07\u00C2\u00B1 0.01 37 \u00C2\u00B14 0.89 15 22 30 21 0.07\u00C2\u00B1 0.01 34 \u00C2\u00B1 5 0.91 15 21 100 15 0.07\u00C2\u00B1 0.01 31 \u00C2\u00B1 5 0.90 14 19 101 Figure 7.4 MCLR Uptake Kinetics in Presence of SRHA and Inorganic Ions (*Initial DOC 3 mg C/L and 25 \u00C2\u00B5g MCLR/L; Error bars represent standard deviation of the respective parameters (n=3)) The uptake of DOC and all the inorganic species was nearly half of the value observed for other organic matrices. For example, the DOC reduction for SRNOM, PLFA and SRHA in presence of 30 mg/L sulphate ions was approximately 32 %, 42 % and 10 %, respectively. Similarly, the nitrates at an initial concentration were reduced by 31%, 35% and 9%, in presence of SRNOM, PLFA and SRHA, respectively. Based on previous discussions, it can be inferred that size exclusion and pore blockage by the higher molecular weight compounds of humic acid fraction is responsible for this behavior. Therefore, the larger molecular weight organic fractions blocked the resin pores resulting in a lower uptake of all the components of the water matrix including inorganics, organic fractions and MCLR. 010203040506070800 10 20 30 40 50 60MCLR Uptake (ug/g)Time (minutes)DI WaterAlkalinity 30 mg/LNitrate 30 mg/LSulphate 30 mg/L102 7.6 Multiple Loading Tests Multiple loading tests were performed to simulate a stirred suspended ion exchange process with side-stream purge and regeneration. Brine recovery studies with a concentration of 2, 6 and 8 mg MCLR/L (with 250, 750 and 1000 mg C/L) were tested with 5%, 8 %, 10 % and 15% NaCl solution and the MCLR recoveries were calculated. Based on the studies (data not shown), 10% and 15% NaCl resulted in the highest recovery % (statistically similar) of MCLR from the resin. In addition, two different regeneration times (60 and 120 minutes) were also investigated to optimize the MCLR recovery and resin performance over multiple cycles. It was observed that for a cycle of 4000 Bed Volumes (BV), the MCLR uptake was reduced from 59% for the first 1000 BV to only 9% after 4000 BV of operation without any regeneration (Figure 7.5 (a)). However, the DOC and UVA254 removals were still in the 40-50% range. This observation confirms that Suwannee river NOM is more readily exchanged than MCLR. Performing a regeneration with 10% NaCl for a contact period of one hour allowed to recover 78% of the MCLR and approximately 50% of the DOC. After regeneration, the resin removed 33%, 50% and 52% of MCLR, DOC and UVA254 in the following 1000 BV of operation. This was a significant difference compared to the first 1000 BV with fresh virgin resin, where MCLR removal of 60% had been measured. Therefore, MCLR performance was more sensitive than DOC to the regeneration performance. This observed recovery after regeneration is not efficient considering the long term operation and therefore further investigations were performed to regenerate after 1000 BV, 2000 BV and 3000 BV of operation to optimize the IX process and maintain a constant NOM and MCLR removal efficiency. Figure 7.5 (b) depicts the performance of A860 resin regenerated after 2000 BV. The figure compares the results of a fresh virgin resin (DI rinsed) with a resin regenerated after 2000 BV of operation, with two different regeneration contact times (1 hour and 2 hours) during the first 1000 103 BV and 2000 BV of operation. As observed, the toxin uptake was nearly stable when the resin was regenerated after 2000 BV of operation. Increasing the resin regeneration time also improved the toxin and DOC uptake over the next subsequent cycles. For example, 54% toxin removal was achieved after one hour of contact time which increased to 59% with two hours of contact time and remained more than 50% after the next 1000 BV. Therefore it was inferred that an optimized regeneration after 2000 BV was effective in maintining a steady rate of MCLR reduction along with simultaneous uptake of dissolved organics and UV254 absorbing compounds. 010203040506070Percentage Removal (%)1000 BV 2000 BV 3000 BV 4000 BV 1000 BVMCLR RemovalDOC RemovalUV254 Removal(a)Regeneration104 Figure 7.5 MCLR, DOC and UV254 Absorbance Reduction for (a) Continuous Operation for 4000 BV and 1000 BV after Regeneration (b) 2000 BV of Operation (c) MCLR Degradation in Brine as a Function of Time. (* Above conditions which are statistically different at p =0.05 using a t-test for independent samples; Error bars represent standard deviation of respective parameters (n=3)) 01020304050607080Percentage Removal (%)MCLR RemovalDOC RemovalUV254 Removal0100200300400500600700800900MCLR Concentration (ppb)BrineBrine pH 12Initial 5 hours 48 hours 120 hours 240 hours(c) 1000 BV 2000 BV Virgin 60 min 120 min Reg. Virgin 60 min 120 min Reg. 105 7.7 MCLR Degradation in Brine Brine management after resin regeneration is one of the greatest challenges with ion exchange resins (Sun et al., 2015; Xiao et al., 2016). Presence of cyanotoxins along with the high salt and organic contaminants adds to the treatment and handling costs of the toxic brine (Arias-paic et al., 2016; Cornelissen et al., 2008; Mergen et al., 2008). For most water treatment plants operating with ion exchange, the regenerated brine is often sent to the wastewater treatment plants. In order to avoid returning to the toxic eluate to the environment the possibility of using caustic soda to raise the pH at 12, a condition favorable to organic matter hydrolysis, was investigated. As shown in Figure 7.5 (c), the brine solution at pH 12 had a faster rate of MCLR degradation as compared to the untreated spent brine. A contact time of 14 days was needed to achieve 70% toxin degradation at pH 12. Considering the cost of caustic, the long contact time needed, the risk of handling caustic and the acid (needed to neutralize caustic before discharge), the use of high pH to treat spent brine does not appear as an interesting option. 7.8 Conclusion Three different inorganic species were selected to study their impact on MCLR uptake under the influence of different organic matter fractions. Sulphates were the most competitive, followed by nitrates and alkalinity. All three NOM fractions exhibited a similar behavior (reduction in removal) under the influence of these inorganic species. However, a significant reduction in toxin uptake capacity and kinetic rates was observed upon increasing the concentration of nitrate and sulphate ions. Evidence of pore restrictions was observed in presence of humic acid. In this case, even after 40 minutes of contact time, contaminant uptake did not increase and there was only 33% site 106 occupancy. Comparative results were obtained with surface water matrices obtained from the drinking water reservoirs of three different communities in British Columbia, indicating that the IX process offers great potential as a pretreatment process for the simultaneous removal of MCLR, DOC and inorganic species from varied water sources. 107 Chapter 8: Conclusions and Future Work 8.1 Conclusions Extensive kinetic experimets were conducted and ion exchange process was identified as a potentially effective technology for the removal of MCLR from aqueous matrices. The main conclusions obtained from this study are listed as follows: \u00EF\u0082\u00B7 The results showed that Purolite A860, a strongly basic anionic exchange resin exhibited excellent adsorption capacities of 3850 \u00C2\u00B5g/g for MCLR, and the removal of MCLR was primarily attributed to electrostatic interactions. \u00EF\u0082\u00B7 Increasing the resin dosage improved the MCLR uptake and a resin dosage of 1000 mg/L (~4.5 mL/L) was sufficient to bring the initial MCLR concentration of 100 \u00C2\u00B5g/L to below the Health Canada guideline of 1.5 \u00C2\u00B5g/L for all investigated water matrices. \u00EF\u0082\u00B7 The kinetic experiments revealed that more than 70 % of the MCLR removal was achieved within 20 minutes for a resin dosage of 200 mg/L with an increase in the uptake rate as the ratio of MCLR/resin concentration increased. \u00EF\u0082\u00B7 The overall removal of MCLR is influenced by both external diffusion and intra-particle diffusion. A significant reduction (>90%) in the MCLR concentration (C0 = 25 \u00C2\u00B5g/L) was observed in the pH 4 - pH 8 range, indicating minimal influence of source water pH under practical operating conditions of pH 6 - pH 7.5. \u00EF\u0082\u00B7 A significant reduction in the uptake of MCLR was observed in the presence of background NOM compounds and it was inferred that intra-particle diffusion plays a key role in governing the uptake kinetics of MCLR in presence of organc contaminats. 108 \u00EF\u0082\u00B7 NOM compounds with an average charge density considerably higher than that of MCLR are preferably removed, thereby resulting in a decline in the MCLR uptake and the kinetic rates. \u00EF\u0082\u00B7 The smaller molecular weight charged NOM fractions compete with the MCLR molecules for active sites in the iner pores of the resin, while the larger molecular weight compounds reduce the available active sites by pore blocakge. \u00EF\u0082\u00B7 IX is also effective for MCLR uptake under the presence of anioic pollutants encountered in the real aquatic systems such as bicarbonates, nitrates and sulphates. \u00EF\u0082\u00B7 Sulphates are the most competitive for exchange sites and result in a lower MCLR and DOC uptake, followed by nitrates and bicarbonates which exhibit competitive behaviours at higher concentrations. \u00EF\u0082\u00B7 The resin can be used for multiple cycles without compromising the DOC reduction and the toxin removal is highly sensitive to regeneration conditions than NOM. \u00EF\u0082\u00B7 Regeneration after 2000 BV was effective in maintining a steady rate of MCLR reduction along with simultaneous uptake of dissolved organics and UV254 absorbing compounds. \u00EF\u0082\u00B7 Comparative results were obtained with surface water matrices obtained from the drinking water reservoirs of three different communities in British Columbia, indicating that IX process offers a great potential as a pretreatment process for the simultaneous removal of MCLR, DOC and inorganic species from water sources. 8.2 Recommendations for Future Work This study was intended to lay out a detailed investigation of the key factors affecting the efficacy of suspended bed IX process for MCLR removal and can be further used to assess the feasibility 109 of IX process to be used in combinaion with downstream treatment processes such as advanced oxidation for treatment of bloom laden waters. Appendix A, illustrates the results from a preliminary study conducted by using IX as a pre-treatment to an advanced oxidation process (Vaccuum UV). The results infer that when combined with further downstream processes, IX can be beneficial and can degrade >99% of the toxin. Moreover, studies involving the investigation of the waste brine management can be useful for commercial IX operations. This can involve investigation of temperature on brine degradation. Overall, this data could contribute to build upon the existing understating and create a wide database of the performance of IX process versus different water qualities. Conclusions made regarding the potential cost savings by using IX as a pre-treatment for downstream treatment processes involving total life cycle cost of IX-VUV is also recommended. 110 Bibliography Acero, J.L., Rodriguez, E., Meriluoto, J., 2005. Kinetics of reactions between chlorine and the cyanobacterial toxins microcystins. Water Res. 39, 1628\u00E2\u0080\u009338. doi:10.1016/j.watres.2005.01.022 Acero, J.L., Spoof, L., Meriluoto, J., Rodr\u00C4\u00B1, E.M., 2008. Oxidation of MC-LR and -RR with chlorine and potassium permanganate\u00E2\u0080\u00AF: Toxicity of the reaction products. Water Res. 42, 1744\u00E2\u0080\u00931752. doi:10.1016/j.watres.2007.10.039 Al Momani, F.A., Jarrah, N., 2010. Treatment and kinetic study of cyanobacterial toxin by ozone. J. Environ. Sci. Heal. Part A 45, 719\u00E2\u0080\u0093731. doi:10.1080/10934521003648933 Alberta Environment, 2012. Personal communication with D. Reid, Regional Integration Branch, Environmental Operations Division, Alberta Environment and Water. Allpike, B.P., Heitz, A., Joll, C.A., Kagi, R.I., 2007. A new organic carbon detector for size exclusion chromatography. J. Chromatogr. A 1157, 472\u00E2\u0080\u0093476. doi:10.1016/j.chroma.2007.05.073 Allpike, B.P., Heitz, A., Joll, C.A., Kagi, R.I., Abbt-Braun, G., Frimmel, F.H., Brinkmann, T., Her, N., Amy, G., 2005. Size Exclusion Chromatography To Characterize DOC Removal in Drinking Water Treatment. Environ. Sci. Technol. 39, 2334\u00E2\u0080\u00932342. doi:10.1021/es0496468 An, J., Carmichael, W.W., 1994. Use of a colorimetric protein phosphatase inhibition assay and enzyme linked immunosorbent assay for the study of microcystins and nodularins. Toxicon 32, 1495\u00E2\u0080\u00931507. doi:10.1016/0041-0101(94)90308-5 Andersen, R.J., Luu, H.A., Chen, D.Z.X., Holmes, C.F.B., Kent, M.L., Le Blanc, M., Taylor, F.J.R. \u00E2\u0080\u0098Max., Williams, D.E., 1993. Chemical and biological evidence links microcystins to salmon \u00E2\u0080\u009Cnetpen liver disease.\u00E2\u0080\u009D Toxicon 31, 1315\u00E2\u0080\u00931323. doi:10.1016/0041-0101(93)90404-7 111 Aranda-Rodriguez, R., Jin, Z., Harvie, J., Cabecinha, A., 2015. Evaluation of three field test kits to detect microcystins from a public health perspective. Harmful Algae 42, 34\u00E2\u0080\u009342. doi:10.1016/j.hal.2015.01.001 Aranda-Rodriguez, R., Kubwabo, C., Benoit, F.M., 2003. Extraction of 15 microcystins and nodularin using immunoaffinity columns. Toxicon 42, 587\u00E2\u0080\u0093599. doi:10.1016/j.toxicon.2003.08.001 Arias-paic, M., Cawley, K.M., Byg, S., Rosario-ortiz, F.L., 2016. Enhanced DOC removal using anion and cation ion exchange resins. Water Res. 88, 981\u00E2\u0080\u0093989. doi:10.1016/j.watres.2015.11.019 Ates, N., Incetan, F.B., 2013. Competition impact of sulfate on NOM removal by anion-exchange resins in high-sulfate and low-SUVA waters. Ind. Eng. Chem. Res. 52, 14261\u00E2\u0080\u009314269. doi:10.1021/ie401814v Banack, S.A., Metcalf, J.S., Bradley, W.G., Cox, P.A., 2014. Detection of cyanobacterial neurotoxin \u00CE\u00B2-N-methylamino-l-alanine within shellfish in the diet of an ALS patient in Florida. Toxicon 90, 167\u00E2\u0080\u0093173. doi:10.1016/j.toxicon.2014.07.018 Bazri, M.M., Barbeau, B., Mohseni, M., 2016a. Evaluation of Weak and Strong Basic Anion Exchange Resins for NOM Removal. J. Environ. Eng. 142, 4016044. doi:10.1061/(ASCE)EE.1943-7870.0001111 Bazri, M.M., Barbeau, B., Mohseni, M., 2012. Impact of UV/H2O2 advanced oxidation treatment on molecular weight distribution of NOM and biostability of water. Water Res. 46, 5297\u00E2\u0080\u00935304. doi:10.1016/j.watres.2012.07.017 Bazri, M.M., Martijn, B., Kroesbergen, J., Mohseni, M., 2016b. Impact of anionic ion exchange resins on NOM fractions: Effect on N-DBPs and C-DBPs precursors. Chemosphere 144, 112 1988\u00E2\u0080\u00931995. doi:10.1016/j.chemosphere.2015.10.086 Bazri, M.M., Mohseni, M., 2016. Impact of natural organic matter properties on the kinetics of suspended ion exchange process. Water Res. 91, 147\u00E2\u0080\u0093155. doi:10.1016/j.watres.2015.12.036 BC Ministry of Health, 2012. Office of the Provincial Health Officer, British Columbia Ministry of Health. Personal communication with B. Boettger. Boari, G., Liberti, L., Merli, C., Passino, R., 1974. Exchange equilibria on anion resins. Desalination 15, 145\u00E2\u0080\u0093166. doi:10.1016/S0011-9164(00)82079-2 Bolto, B., Dixon, D., Eldridge, R., 2004. Ion exchange for the removal of natural organic matter. React. Funct. Polym. 60, 171\u00E2\u0080\u0093182. doi:10.1016/j.reactfunctpolym.2004.02.021 Bolto, B., Dixon, D., Eldridge, R., King, S., 2002a. Removal of THM precursors by coagulation or ion exchange. Water Res. 36, 5066\u00E2\u0080\u00935073. doi:10.1016/S0043-1354(02)00232-4 Bolto, B., Dixon, D., Eldridge, R., King, S., Linge, K., 2002b. Removal of natural organic matter by ion exchange. Water Res. 36, 5057\u00E2\u0080\u00935065. doi:10.1016/S0043-1354(02)00231-2 Boua\u00C4\u00B1\u00CC\u0088cha, N., Maatouk, I., Vincent, G., Levi, Y., 2002. A colorimetric and fluorometric microplate assay for the detection of microcystin-LR in drinking water without preconcentration. Food Chem. Toxicol. 40, 1677\u00E2\u0080\u00931683. doi:10.1016/S0278-6915(02)00103-5 Boyer, G.L., 2007. The occurrence of cyanobacterial toxins in New York lakes: Lessons from the MERHAB-Lower Great Lakes program. Lake Reserv. Manag. 23, 153\u00E2\u0080\u0093160. doi:10.1080/07438140709353918 Boyer, T.H., Singer, P.C., 2008. Stoichiometry of Removal of Natural Organic Matter by Ion Exchange. Environ. Sci. Technol. 42, 608\u00E2\u0080\u0093613. doi:10.1021/es071940n 113 Boyer, T.H., Singer, P.C., 2008. Removal of natural organic material by a magnetic ion exchange resin. Water Sci. Technol. Water Supply 8, 167. doi:10.2166/ws.2008.052 Boyer, T.H., Singer, P.C., 2006. A pilot-scale evaluation of magnetic ion exchange treatment for removal of natural organic material and inorganic anions. Water Res. 40, 2865\u00E2\u0080\u00932876. doi:10.1016/j.watres.2006.05.022 Boyer, T.H., Singer, P.C., 2005. Bench-scale testing of a magnetic ion exchange resin for removal of disinfection by-product precursors. Water Res. 39, 1265\u00E2\u0080\u00931276. doi:10.1016/j.watres.2005.01.002 Boyer, T.H., Singer, P.C., Aiken, G.R., 2008. Removal of Dissolved Organic Matter by Anion Exchange: Effect of Dissolved Organic Matter Properties. Environ. Sci. Technol. 42, 7431\u00E2\u0080\u00937437. doi:10.1021/es800714d Bragg, W.A., Lemire, S.W., Coleman, R.M., Hamelin, E.I., Johnson, R.C., 2015. Detection of human exposure to saxitoxin and neosaxitoxin in urine by online-solid phase extraction-liquid chromatography\u00E2\u0080\u0093tandem mass spectrometry. Toxicon 99, 118\u00E2\u0080\u0093124. doi:10.1016/j.toxicon.2015.03.017 Brand, L.E., Pablo, J., Compton, A., Hammerschlag, N., Mash, D.C., 2010. Cyanobacterial blooms and the occurrence of the neurotoxin, beta-N-methylamino-l-alanine (BMAA), in South Florida aquatic food webs. Harmful Algae 9, 620\u00E2\u0080\u0093635. doi:10.1016/j.hal.2010.05.002 Bullerjahn, G.S., McKay, R.M., Davis, T.W., Baker, D.B., Boyer, G.L., D\u00E2\u0080\u0099Anglada, L. V., Doucette, G.J., Ho, J.C., Irwin, E.G., Kling, C.L., Kudela, R.M., Kurmayer, R., Michalak, A.M., Ortiz, J.D., Otten, T.G., Paerl, H.W., Qin, B., Sohngen, B.L., Stumpf, R.P., Visser, P.M., Wilhelm, S.W., 2016. Global solutions to regional problems: Collecting global expertise to address the problem of harmful cyanobacterial blooms. A Lake Erie case study. 114 Harmful Algae 54, 223\u00E2\u0080\u0093238. doi:10.1016/j.hal.2016.01.003 Camacho, F.P., Bongiovani, M.C., Silva, M.O., Coldebella, P.F., Sousa Pessoa Amorim, M.T., Bergamasco, R., 2015. Coagulation/Flocculation/Flotation/Nanofiltration Processes Using Moringa Oleifera as Coagulant of Eutrophized River. Icheap12 12Th Int. Conf. Chem. Process Eng. 43, 1123\u00E2\u0080\u00931128. doi:10.3303/CET1543188 Carmichael, W.W., 1992. Cyanobacteria secondary metabolites-the cyanotoxins. J. Appl. Bacteriol. 72, 445\u00E2\u0080\u0093459. doi:10.1111/j.1365-2672.1992.tb01858.x Carmichael, W.W., Azevedo, S.M.F.O., An, J.S., Molica, R.J.R., Jochimsen, E.M., Lau, S., Rinehart, K.L., Shaw, G.R., Eaglesham, G.K., 2001. Human fatalities form cyanobacteria: Chemical and biological evidence for cyanotoxins. Environ. Health Perspect. 109, 663\u00E2\u0080\u0093668. doi:10.1289/ehp.01109663 Carmichael, W.W., Boyer, G.L., 2016. Health impacts from cyanobacteria harmful algae blooms: Implications for the North American Great Lakes. Harmful Algae 54, 194\u00E2\u0080\u0093212. doi:10.1016/j.hal.2016.02.002 Carri\u00C3\u00A8re, A., Pr\u00C3\u00A9vost, M., Zamyadi, A., Chevalier, P., Barbeau, B., 2010. Vulnerability of Quebec drinking-water treatment plants to cyanotoxins in a climate change context. J. Water Health 8, 455. doi:10.2166/wh.2009.207 Catherine, Q., Susanna, W., Isidora, E.-S., Mark, H., Aur\u00C3\u00A9lie, V., Jean-Fran\u00C3\u00A7ois, H., 2013. A review of current knowledge on toxic benthic freshwater cyanobacteria \u00E2\u0080\u0093 Ecology, toxin production and risk management. Water Res. 47, 5464\u00E2\u0080\u00935479. doi:10.1016/j.watres.2013.06.042 Chatziefthimiou, A.D., Metcalf, J.S., Glover, W.B., Banack, S.A., Dargham, S.R., Richer, R.A., 2016. Cyanobacteria and cyanotoxins are present in drinking water impoundments and 115 groundwater wells in desert environments. Toxicon 114, 75\u00E2\u0080\u009384. doi:10.1016/j.toxicon.2016.02.016 Chen, H., Burke, J.M., Prepas, E.E., 2011. Cyanobacterial Toxins in Fresh Waters, in: Encyclopedia of Environmental Health. Elsevier, pp. 860\u00E2\u0080\u0093871. doi:10.1016/B978-0-444-52272-6.00333-0 Chen, L., Chen, J., Zhang, X., Xie, P., 2016. A review of reproductive toxicity of microcystins. J. Hazard. Mater. 301, 381\u00E2\u0080\u0093399. doi:10.1016/j.jhazmat.2015.08.041 Chen, Y., Zhang, X., Liu, Q., Wang, X., Xu, L., Zhang, Z., 2015. Facile and economical synthesis of porous activated semi-cokes for highly efficient and fast removal of microcystin-LR. J. Hazard. Mater. 299, 325\u00E2\u0080\u0093332. doi:10.1016/j.jhazmat.2015.06.049 Chorus, I., Bartram, J., 1999. Toxic Cyanobacteria in Water: A guide to public health consequences, monitoring and management. E FN Spon, NY, World Heal. Organ. Codd, G., Bell, S., Kaya, K., Ward, C., Beattie, K., Metcalf, J., 1999. Cyanobacterial toxins, exposure routes and human health. Eur. J. Phycol. 34, 405\u00E2\u0080\u0093415. doi:10.1080/09670269910001736462 Cook, D., Newcombe, G., Sztajnbok, P., 2001. The application of powdered activated carbon for MIB and geosmin removal: Predicting PAC doses in four raw waters. Water Res. 35, 1325\u00E2\u0080\u00931333. doi:10.1016/S0043-1354(00)00363-8 Coral, L.A., Zamyadi, A., Barbeau, B., Bassetti, F.J., Lapolli, F.R., Pr\u00C3\u00A9vost, M., 2013. Oxidation of Microcystis aeruginosa and Anabaena flos-aquae by ozone: Impacts on cell integrity and chlorination by-product formation. Water Res. 47, 2983\u00E2\u0080\u00932994. doi:10.1016/j.watres.2013.03.012 Cornelissen, E.R., Chasseriaud, D., Siegers, W.G., Beerendonk, E.F., van der Kooij, D., 2010. 116 Effect of anionic fluidized ion exchange (FIX) pre-treatment on nanofiltration (NF) membrane fouling. Water Res. 44, 3283\u00E2\u0080\u00933293. doi:10.1016/j.watres.2010.03.007 Cornelissen, E.R., Moreau, N., Siegers, W.G., Abrahamse, A.J., Rietveld, L.C., Grefte, A., Dignum, M., Amy, G., Wessels, L.P., 2008. Selection of anionic exchange resins for removal of natural organic matter ( NOM ) fractions. Water Res. 42, 413\u00E2\u0080\u0093423. doi:10.1016/j.watres.2007.07.033 Cox, P.A., Banack, S.A., Murch, S.J., 2003. Biomagnification of cyanobacterial neurotoxins and neurodegenerative disease among the Chamorro people of Guam. Proc. Natl. Acad. Sci. 100, 13380\u00E2\u0080\u009313383. doi:10.1073/pnas.2235808100 Croue, J., Humbert, H., 2005. Performance of selected anion exchange resins for the treatment of a high DOC content surface water. Water Res. 39, 1699\u00E2\u0080\u00931708. doi:10.1016/j.watres.2005.02.008 Dada, A.., Olalekan, A.., Olatunya, A.., Dada, O., 2012. Langmuir , Freundlich , Temkin and Dubinin \u00E2\u0080\u0093 Radushkevich Isotherms Studies of Equilibrium Sorption of Zn2+ Unto Phosphoric Acid Modified Rice Husk. IOSR J. Appl. Chem. 3, 38\u00E2\u0080\u009345. doi:10.9790/5736-0313845 de Bernardi, R., Giussani, G., 1990. Are blue-green algae a suitable food for zooplankton? An overview. Hydrobiologia 200\u00E2\u0080\u0093201, 29\u00E2\u0080\u009341. doi:10.1007/BF02530326 De Maagd, P.G.J., Hendriks, a. J., Seinen, W., Sijm, D.T.H.M., 1999. pH-dependent hydrophobicity of the cyanobacteria toxin microcystin-LR. Water Res. 33, 677\u00E2\u0080\u0093680. doi:10.1016/S0043-1354(98)00258-9 Ding, L., Deng, H., Wu, C., Han, X., 2012. Affecting factors, equilibrium, kinetics and thermodynamics of bromide removal from aqueous solutions by MIEX resin. Chem. Eng. J. 117 181\u00E2\u0080\u0093182, 360\u00E2\u0080\u0093370. doi:10.1016/j.cej.2011.11.096 Dixit, F., Mohseni, M., 2016. Anion exchange resins for Removal of cyanobacterial toxins from surface water. Accepted for presentation at the American Water Works Association \u00E2\u0080\u0093 Water Quality Technology Conference (AWWA-WQTC), Indianapolis, IN. Dixon, M.B., Richard, Y., Ho, L., Chow, C.W.K., Neill, B.K.O., Newcombe, G., 2011. A coagulation \u00E2\u0080\u0093 powdered activated carbon \u00E2\u0080\u0093 ultrafiltration \u00E2\u0080\u0093 Multiple barrier approach for removing toxins from two Australian cyanobacterial blooms. J. Hazard. Mater. 186, 1553\u00E2\u0080\u00931559. doi:10.1016/j.jhazmat.2010.12.049 D\u00C3\u00B6rr, F.A., Rodr\u00C3\u00ADguez, V., Molica, R., Henriksen, P., Krock, B., Pinto, E., 2010. Methods for detection of anatoxin-a(s) by liquid chromatography coupled to electrospray ionization-tandem mass spectrometry. Toxicon 55, 92\u00E2\u0080\u009399. doi:10.1016/j.toxicon.2009.07.017 Downing, J.A., Watson, S.B., McCauley, E., 2001. Predicting Cyanobacteria dominance in lakes. Can. J. Fish. Aquat. Sci. 58, 1905\u00E2\u0080\u00931908. doi:10.1139/cjfas-58-10-1905 du Prel, J.-B., Hommel, G., R\u00C3\u00B6hrig, B., Blettner, M., 2009. Confidence interval or p-value?: part 4 of a series on evaluation of scientific publications. Dtsch. Arztebl. Int. doi:10.3238/arztebl.2009.0335 Dupuis, A.P., Hann, B.J., 2009. Warm spring and summer water temperatures in small eutrophic lakes of the Canadian prairies: potential implications for phytoplankton and zooplankton. J. Plankton Res. 31, 489\u00E2\u0080\u0093502. doi:10.1093/plankt/fbp001 Ebie, K., Li, F., Azuma, Y., Yuasa, A., Hagishita, T., 2001. Pore distribution effect of activated carbon in adsorbing organic micropollutants from natural water. Water Res. 35, 167\u00E2\u0080\u0093179. doi:10.1016/S0043-1354(00)00257-8 El-Shehawy, R., Gorokhova, E., Fern\u00C3\u00A1ndez-Pi\u00C3\u00B1as, F., del Campo, F.F., 2012. Global warming 118 and hepatotoxin production by cyanobacteria: What can we learn from experiments? Water Res. 46, 1420\u00E2\u0080\u00931429. doi:10.1016/j.watres.2011.11.021 Elliott, J.A., 2012. Is the future blue-green? A review of the current model predictions of how climate change could affect pelagic freshwater cyanobacteria. Water Res. 46, 1364\u00E2\u0080\u00931371. doi:10.1016/j.watres.2011.12.018 Enzo Life Sciences, 2010. Microcystins & other cyanobacterial toxins. EPA, 2012. Cyanobacteria and Cyanotoxins\u00E2\u0080\u00AF: Information for Drinking Water Systems. United States Environ. Prot. Agency 1\u00E2\u0080\u00939. Etheridge, S.M., 2010. Paralytic shellfish poisoning: Seafood safety and human health perspectives. Toxicon 56, 108\u00E2\u0080\u0093122. doi:10.1016/j.toxicon.2009.12.013 Falconer, I.R., 1999. An overview of problems caused by toxic blue-green algae (cyanobacteria) in drinking and recreational water. Environ. Toxicol. 14, 5\u00E2\u0080\u009312. doi:10.1002/(SICI)1522-7278(199902)14:1<5::AID-TOX3>3.0.CO;2-0 Falconer, I.R., 1991. Tumor promotion and liver injury caused by oral consumption of cyanobacteria. Environ. Toxicol. Water Qual. 6, 177\u00E2\u0080\u0093184. doi:10.1002/tox.2530060207 Falconer, I.R., Humpage, A.R., 2006. Cyanobacterial (blue-green algal) toxins in water supplies: Cylindrospermopsins. Environ. Toxicol. 21, 299\u00E2\u0080\u0093304. doi:10.1002/tox.20194 Falconer, I.R., Humpage, A.R., 2005. Health risk assessment of cyanobacterial (blue-green algal) toxins in drinking water. Int. J. Environ. Res. Public Health 2, 43\u00E2\u0080\u009350. doi:10.3390/ijerph2005010043 Falconer, I.R., Humpage, A.R., 2001. Preliminary evidence for in vivo tumour initiation by oral administration of extracts of the blue-green algaCylindrospermopsis raciborskii containing the toxin cylindrospermopsin. Environ. Toxicol. 16, 192\u00E2\u0080\u0093195. doi:10.1002/tox.1024 119 Falconer, I.R., Humpage, A.R., 1996. Tumour promotion by cyanobacterial toxins. Phycologia 35, 74\u00E2\u0080\u009379. doi:10.2216/i0031-8884-35-6S-74.1 Fastner, J., Flieger, I., Neumann, U., 1998. Optimised extraction of microcystins from field samples - A comparison of different solvents and procedures. Water Res. 32, 3177\u00E2\u0080\u00933181. doi:10.1016/S0043-1354(98)00073-6 Fayad, P.B., Roy-Lachapelle, A., Duy, S.V., Pr\u00C3\u00A9vost, M., Sauv\u00C3\u00A9, S., 2015. On-line solid-phase extraction coupled to liquid chromatography tandem mass spectrometry for the analysis of cyanotoxins in algal blooms. Toxicon 108, 167\u00E2\u0080\u0093175. doi:10.1016/j.toxicon.2015.10.010 Fishman, D.B., Adlerstein, S.A., Vanderploeg, H.A., Fahnenstiel, G.L., Scavia, D., 2010. Phytoplankton community composition of Saginaw Bay, Lake Huron, during the zebra mussel (Dreissena polymorpha) invasion: A multivariate analysis. J. Great Lakes Res. 36, 9\u00E2\u0080\u009319. doi:10.1016/j.jglr.2009.10.004 Funari, E., Testai, E., 2008. Human Health Risk Assessment Related to Cyanotoxins Exposure. Crit. Rev. Toxicol. 38, 97\u00E2\u0080\u0093125. doi:10.1080/10408440701749454 G. Newcombe, J. Morrison, C.H., 2002. Simultaneous adsorption of MIB and NOM onto activated carbon . 1 . Characterisation of the system and NOM adsorption. Carbon N. Y. 40, 2135\u00E2\u0080\u00932146. Giesche, H., 2006. Mercury Porosimetry: A General (Practical) Overview. Part. Part. Syst. Charact. 23, 9\u00E2\u0080\u009319. doi:10.1002/ppsc.200601009 Graf, K.C., Cornwell, D. a., Boyer, T.H., 2014. Removal of dissolved organic carbon from surface water by anion exchange and adsorption: Bench-scale testing to simulate a two-stage countercurrent process. Sep. Purif. Technol. 122, 523\u00E2\u0080\u0093532. doi:10.1016/j.seppur.2013.12.012 120 Graham, J.L., Loftin, K.A., Meyer, M.T., Ziegler, A.C., 2010. Cyanotoxin Mixtures and Taste-and-Odor Compounds in Cyanobacterial Blooms from the Midwestern United States. Environ. Sci. Technol. 44, 7361\u00E2\u0080\u00937368. doi:10.1021/es1008938 Grefte, A., Dignum, M., Cornelissen, E.R., Rietveld, L.C., 2013. Natural organic matter removal by ion exchange at different positions in the drinking water treatment lane. Drink. Water Eng. Sci. 6, 1\u00E2\u0080\u009310. doi:10.5194/dwes-6-1-2013 Greluk, M., Hubicki, Z., 2011. Efficient removal of Acid Orange 7 dye from water using the strongly basic anion exchange resin Amberlite IRA-958. Desalination 278, 219\u00E2\u0080\u0093226. doi:10.1016/j.desal.2011.05.024 Guo, B., Hong, L., Jiang, H.-X., 2003. Macroporous Poly(calcium acrylate\u00E2\u0088\u0092divinylbenzene) Bead A Selective Orthophosphite Sorbent. Ind. Eng. Chem. Res. 42, 5559\u00E2\u0080\u00935567. doi:10.1021/ie030156i He, X., Liu, Y.L., Conklin, A., Westrick, J., Weavers, L.K., Dionysiou, D.D., Lenhart, J.J., Mouser, P.J., Szlag, D., Walker, H.W., 2016. Toxic cyanobacteria and drinking water: Impacts, detection, and treatment. Harmful Algae 54, 174\u00E2\u0080\u0093193. doi:10.1016/j.hal.2016.01.001 He, X., Pelaez, M., Westrick, J.A., Shea, K.E.O., Hiskia, A., Triantis, T., Kaloudis, T., Stefan, M.I., De, A.A., Dionysiou, D.D., 2011. Efficient removal of microcystin-LR by UV-C / H 2 O 2 in synthetic and natural water samples. Water Res. 46, 1501\u00E2\u0080\u00931510. doi:10.1016/j.watres.2011.11.009 Health Canada, 2016. Cyanobacterial Toxins in Drinking Water, Guidelines for Canadaian Drinking Water Quality. Health Canada, 2012. Guidelines for Canadian Recreational Water Quality \u00E2\u0080\u0093 Third Edition. 121 Helfferich, F., 1965. Ion-Exchange Kinetics. V. Ion Exchange Accompanied by Reactions. J. Phys. Chem. 69, 1178\u00E2\u0080\u00931187. doi:10.1021/j100888a015 Henderson, R.K., Parsons, S.A., Jefferson, B., 2010. The impact of differing cell and algogenic organic matter (AOM) characteristics on the coagulation and flotation of algae. Water Res. 44, 3617\u00E2\u0080\u00933624. doi:10.1016/j.watres.2010.04.016 Heresztyn, T., 2001. Determination of cyanobacterial hepatotoxins directly in water using a protein phosphatase inhibition assay. Water Res. 35, 3049\u00E2\u0080\u00933056. doi:10.1016/S0043-1354(01)00018-5 Ho, J.C., Michalak, A.M., 2015. Challenges in tracking harmful algal blooms: A synthesis of evidence from Lake Erie. J. Great Lakes Res. 41, 317\u00E2\u0080\u0093325. doi:10.1016/j.jglr.2015.01.001 Hsu, S., Singer, P.C., 2010. Removal of bromide and natural organic matter by anion exchange. Water Res. 44, 2133\u00E2\u0080\u00932140. doi:10.1016/j.watres.2009.12.027 Hu, J., Chen, Y., Guo, L., Chen, X., 2015. Chemical-free ion exchange and its application for desalination. Desalination 365, 144\u00E2\u0080\u0093150. doi:10.1016/j.desal.2015.02.033 Hu, J., Martin, A., Shang, R., Siegers, W., Cornelissen, E., Heijman, B., Rietveld, L., 2014. Anionic exchange for NOM removal and the effects on micropollutant adsorption competition on activated carbon. Sep. Purif. Technol. 129, 25\u00E2\u0080\u009331. doi:10.1016/j.seppur.2014.03.019 Huang, W., Cheng, B., Cheng, Y., 2007. Adsorption of microcystin- LR by three types of activated carbon. J. Hazard. Mater. 141, 115\u00E2\u0080\u0093122. doi:10.1016/j.jhazmat.2006.06.122 Huang, W., He, H., Dong, B., Chu, H., Xu, G., Yan, Z., 2015. Effects of macro-porous anion exchange and coagulation treatment on organic removal and membrane fouling reduction in water treatment. Desalination 355, 204\u00E2\u0080\u0093216. doi:10.1016/j.desal.2014.10.045 122 Huber, S. a., Balz, A., Abert, M., Pronk, W., 2011. Characterisation of aquatic humic and non-humic matter with size-exclusion chromatography \u00E2\u0080\u0093 organic carbon detection \u00E2\u0080\u0093 organic nitrogen detection (LC-OCD-OND). Water Res. 45, 879\u00E2\u0080\u0093885. doi:10.1016/j.watres.2010.09.023 Humbert, H., Gallard, H., Suty, H., Crou\u00C3\u00A9, J.-P., 2005. Performance of selected anion exchange resins for the treatment of a high DOC content surface water. Water Res. 39, 1699\u00E2\u0080\u00931708. doi:10.1016/j.watres.2005.02.008 Ishii, S.K.L., Boyer, T.H., 2011. Evaluating the secondary effects of magnetic ion exchange: Focus on corrosion potential in the distribution system. Desalination 274, 31\u00E2\u0080\u009338. doi:10.1016/j.desal.2011.01.061 ISO, 2005. Water quality\u00E2\u0080\u0094Determination of microcystins\u00E2\u0080\u0094Method using solid phase extraction (SPE) and high performance liquid chromatography (HPLC) and ultraviolet (UV) detection: ISO 20179. Geneva, Switzerland. Izaguirre, G., Jungblut, A.-D., Neilan, B.A., 2007. Benthic cyanobacteria (Oscillatoriaceae) that produce microcystin-LR, isolated from four reservoirs in southern California. Water Res. 41, 492\u00E2\u0080\u0093498. doi:10.1016/j.watres.2006.10.012 Izydorczyk, K., Carpentier, C., Mr\u00C3\u00B3wczy\u00C5\u0084ski, J., Wagenvoort, A., Jurczak, T., Tarczy\u00C5\u0084ska, M., 2009. Establishment of an Alert Level Framework for cyanobacteria in drinking water resources by using the Algae Online Analyser for monitoring cyanobacterial chlorophyll a. Water Res. 43, 989\u00E2\u0080\u0093996. doi:10.1016/j.watres.2008.11.048 John D. Pfaff, 1993. Method 300.1 \u00E2\u0080\u009CDetermination of inorganic anions in drinking water by ion chromatography\u00E2\u0080\u009D rev. 1.0. United States Environ. Prot. Agency. Johnson, C.J., Singer, P.C., 2004. Impact of a magnetic ion exchange resin on ozone demand and 123 bromate formation during drinking water treatment. Water Res. 38, 3738\u00E2\u0080\u00933750. doi:10.1016/j.watres.2004.06.021 Jones, G., 1994. Release and degradation of microcystin following algicide treatment of a Microcystis aeruginosa bloom in a recreational lake, as determined by HPLC and protein phosphatase inhibition assay. Water Res. 28, 871\u00E2\u0080\u0093876. doi:10.1016/0043-1354(94)90093-0 Kabsch-Korbutowicz, M., Majewska-Nowak, K., Winnicki, T., 2008. Water treatment using MIEX\u00C2\u00AE DOC/ultrafiltration process. Desalination 221, 338\u00E2\u0080\u0093344. doi:10.1016/j.desal.2007.01.092 Kenefick, S.L., Hrudey, S.E., Peterson, H.G., Prepas, E.E., 1993. Toxin Release from Microcystis Aeruginosa after Chemical Treatment. Water Sci. Technol. 27, 433\u00E2\u0080\u0093440. Kennedy, M.D., Chun, H.K., Yangalp, V.A.Q., Heijman, B.G.J., Schippers, J.C., 2005. Natural organic matter ( NOM ) fouling of ultrafiltration membranes\u00E2\u0080\u00AF: fractionation of NOM in surface water and characterisation by LC-OCD. Desalination 178, 73\u00E2\u0080\u009383. doi:10.1016/j.desal.2005.02.004 Kim, H.C., Dempsey, B.A., 2010. Removal of organic acids from EfOM using anion exchange resins and consequent reduction of fouling in UF and MF. J. Memb. Sci. 364, 325\u00E2\u0080\u0093330. doi:10.1016/j.memsci.2010.08.032 Kim, Y.M., Oh, S.W., Jeong, S.Y., Pyo, D.J., Choi, E.Y., 2003. Development of an Ultrarapid One-Step Fluorescence Immunochromatographic Assay System for the Quantification of Microcystins. Environ. Sci. Technol. 37, 1899\u00E2\u0080\u00931904. doi:10.1021/es026191i Kim, Y.S., Lee, Y.H., An, B., Choi, S.A., Park, J.H., Jurng, J.S., Lee, S.H., Choi, J.W., 2012. Simultaneous Removal of Phosphate and Nitrate in Wastewater Using High-Capacity Anion-Exchange Resin. Water Air Soil Pollut. 223, 5959\u00E2\u0080\u00935966. doi:10.1007/s11270-012-124 1331-1 Ko, D.C.K., Porter, J.F., McKay, G., 2001. Film-pore diffusion model for the fixed-bed sorption of copper and cadmium ions onto bone char. Water Res. 35, 3876\u00E2\u0080\u00933886. doi:10.1016/S0043-1354(01)00114-2 Kondo, F., Ikai, Y., Oka, H., Okumura, M., Ishikawa, N., Harada, K., Matsuura, K., Murata, H., Suzuki, M., 1992. Formation, characterization, and toxicity of the glutathione and cysteine conjugates of toxic heptapeptide microcystins. Chem. Res. Toxicol. 5, 591\u00E2\u0080\u0093596. doi:10.1021/tx00029a002 Kondo, F., Matsumoto, H., Yamada, S., Tsuji, K., Ueno, Y., Harada, K., 2000. Immunoaffinity purification method for detection and quantification of microcystins in lake water. Toxicon 38, 813\u00E2\u0080\u0093823. doi:10.1016/S0041-0101(99)00194-4 Kotak, B.G., Zurawell, R.W., 2007. Cyanobacterial toxins in Canadian freshwaters: A review. Lake Reserv. Manag. 23, 109\u00E2\u0080\u0093122. doi:10.1080/07438140709353915 Krienitz, L., Dadheech, P.K., Fastner, J., Kotut, K., 2013. The rise of potentially toxin producing cyanobacteria in Lake Naivasha, Great African Rift Valley, Kenya. Harmful Algae 27, 42\u00E2\u0080\u009351. doi:10.1016/j.hal.2013.04.005 Lajeunesse, A., Segura, P.A., G\u00C3\u00A9linas, M., Hudon, C., Thomas, K., Quilliam, M.A., Gagnon, C., 2012. Detection and confirmation of saxitoxin analogues in freshwater benthic Lyngbya wollei algae collected in the St. Lawrence River (Canada) by liquid chromatography\u00E2\u0080\u0093tandem mass spectrometry. J. Chromatogr. A 1219, 93\u00E2\u0080\u0093103. doi:10.1016/j.chroma.2011.10.092 Lawton, L. a., Cornish, B.J.P. a., MacDonald, A.W.R., 1998. Removal of cyanobacterial toxins (microcystins) and cyanobacterial cells from drinking water using domestic water filters. 125 Water Res. 32, 633\u00E2\u0080\u0093638. doi:10.1016/S0043-1354(97)00267-4 Lee, D., Kim, S., Cho, I., Kim, S., Kim, S., 2004a. Photocatalytic oxidation of microcystin-LR in a fluidized bed reactor having TiO 2 -coated activated carbon. Seperation Purif. Technol. 34, 59\u00E2\u0080\u009366. doi:10.1016/S1383-5866(03)00175-8 Lee, D., Kim, S., Kim, S., Chung, I., Kim, S., 2004b. Photocatalytic oxidation of microcystin-LR with TiO 2 -coated activated carbon 102, 93\u00E2\u0080\u009398. doi:10.1016/j.cej.2004.01.027 Lee, J., Walker, H.W., 2011. Physicochemical and Engineering Aspects Adsorption of microcystin-Lr onto iron oxide nanoparticles. Colloids Surfaces A Physicochem. Eng. Asp. 373, 94\u00E2\u0080\u0093100. doi:10.1016/j.colsurfa.2010.10.032 Li, L., Quinlivan, P.A., Knappe, D.R.U., 2002. Effects of activated carbon surface chemistry and pore structure on the adsorption of organic contaminants from aqueous solution. Carbon N. Y. 40, 2085\u00E2\u0080\u00932100. Li, P., Sengupta, A.K., 2001. Entropy-driven selective ion exchange for aromatic ions and the role of cosolvents. Colloids Surfaces A Physicochem. Eng. Asp. 191, 123\u00E2\u0080\u0093132. Li, P., Sengupta, A.K., 2000. Intraparticle diffusion during selective ion exchange with a macroporous exchanger. React. Funct. Polym. 44, 273\u00E2\u0080\u0093287. Li, P., SenGupta, A.K., 2000. Intraparticle Diffusion during Selective Sorption of Trace Contaminants: The Effect of Gel versus Macroporous Morphology. Environ. Sci. Technol. 34, 5193\u00E2\u0080\u00935200. doi:10.1021/es001299o Li, Q., Snoeyink, V.L., Mari\u00C3\u00B1as, B.J., Campos, C., 2003. Pore blockage effect of NOM on atrazine adsorption kinetics of PAC: The roles of PAC pore size distribution and NOM molecular weight. Water Res. 37, 4863\u00E2\u0080\u00934872. doi:10.1016/j.watres.2003.08.018 Lian, L., Cao, X., Wu, Y., Sun, D., Lou, D., 2014. A green synthesis of magnetic bentonite 126 material and its application for removal of microcystin-LR in water. Appl. Surf. Sci. 289, 245\u00E2\u0080\u0093251. doi:10.1016/j.apsusc.2013.10.144 Liu, L., Xing, C., Yan, H., Kuang, H., Xu, C., 2014. Development of an ELISA and Immunochromatographic Strip for Highly Sensitive Detection of Microcystin-LR. Sensors 14, 14672\u00E2\u0080\u009314685. doi:10.3390/s140814672 Machado, J., Campos, A., Vasconcelos, V., Freitas, M., 2017. Effects of microcystin-LR and cylindrospermopsin on plant-soil systems: A review of their relevance for agricultural plant quality and public health. Environ. Res. 153, 191\u00E2\u0080\u0093204. doi:10.1016/j.envres.2016.09.015 Manitoba Water Stewardship, 2012. Personal communication with K. Phillip, Office of Drinking Water, Manitoba Water Stewardship. Matilainen, A., Lindqvist, N., Korhonen, S., Tuhkanen, T., 2002. Removal of NOM in the different stages of the water treatment process. Environ. Int. 28, 457\u00E2\u0080\u0093465. doi:10.1016/S0160-4120(02)00071-5 Matsui, Y., Ando, N., Sasaki, H., Matsushita, T., Ohno, K., 2009. Branched pore kinetic model analysis of geosmin adsorption on super-powdered activated carbon. Water Res. 43, 3095\u00E2\u0080\u00933103. doi:10.1016/j.watres.2009.04.014 Matsui, Y., Yoshida, T., Nakao, S., Knappe, D.R.U., Matsushita, T., 2012. Characteristics of competitive adsorption between 2-methylisoborneol and natural organic matter on superfine and conventionally sized powdered activated carbons. Water Res. 46, 4741\u00E2\u0080\u00934749. doi:10.1016/j.watres.2012.06.002 Mayer, H., Tharanathan, R.N., Weckesser, J., 1985. 6 Analysis of Lipopolysaccharides of Gram-Negative Bacteria. pp. 157\u00E2\u0080\u0093207. doi:10.1016/S0580-9517(08)70475-6 McElhiney, J., Lawton, L.A., 2005. Detection of the cyanobacterial hepatotoxins microcystins. 127 Toxicol. Appl. Pharmacol. 203, 219\u00E2\u0080\u0093230. doi:10.1016/j.taap.2004.06.002 M\u00C3\u00A9jean, A., Mazmouz, R., Paci, G., Moncoq, K., Regad, L., Combes, A., Pichon, V., Ploux, O., 2014. 6. Biosynthesis of the cyanobacterial neurotoxins anatoxin-a and homoanatoxin-a: From the genome to the metabolites. Toxicon 91, 166\u00E2\u0080\u0093167. doi:10.1016/j.toxicon.2014.08.014 M\u00C3\u00A9jean, A., Paci, G., Gautier, V., Ploux, O., 2014. Biosynthesis of anatoxin-a and analogues (anatoxins) in cyanobacteria. Toxicon 91, 15\u00E2\u0080\u009322. doi:10.1016/j.toxicon.2014.07.016 Merel, S., Lebot, B., Cl\u00C3\u00A9ment, M., Seux, R., Thomas, O., 2009. Ms identification of microcystin-LR chlorination by-products. Chemosphere 74, 832\u00E2\u0080\u0093839. doi:10.1016/j.chemosphere.2008.10.024 Merel, S., Walker, D., Chicana, R., Snyder, S., Baur\u00C3\u00A8s, E., Thomas, O., 2013. State of knowledge and concerns on cyanobacterial blooms and cyanotoxins. Environ. Int. 59, 303\u00E2\u0080\u0093327. doi:10.1016/j.envint.2013.06.013 Mergen, M.R.D., Jefferson, B., Parsons, S.A., Jarvis, P., 2008. Magnetic ion-exchange resin treatment: Impact of water type and resin use. Water Res. 42, 1977\u00E2\u0080\u00931988. doi:10.1016/j.watres.2007.11.032 Meriluoto, J., Codd, 2005. Microcystins and nodularins, Cyanobacterial Monitoring and Cyanotoxin Analysis. Metcalf, J.S., Codd, G. a., 2014. Cyanobacterial Toxins (Cyanotoxins) in Water: A Review of Current Knowledge. Found. Water Res. 44, 47. Minist\u00C3\u00A8re du D\u00C3\u00A9veloppement durable de l\u00E2\u0080\u0099Environnement et des Parcs du Qu\u00C3\u00A9bec, 2012. Personal communication with C. Robert, Minist\u00C3\u00A8re du D\u00C3\u00A9veloppement durable, de l\u00E2\u0080\u0099Environnement et des Parcs du Qu\u00C3\u00A9bec. 128 Molot, L.A., Watson, S.B., Creed, I.F., Trick, C.G., McCabe, S.K., Verschoor, M.J., Sorichetti, R.J., Powe, C., Venkiteswaran, J.J., Schiff, S.L., 2014. A novel model for cyanobacteria bloom formation: the critical role of anoxia and ferrous iron. Freshw. Biol. 59, 1323\u00E2\u0080\u00931340. doi:10.1111/fwb.12334 Moreira, C., Mendes, R., Azevedo, J., Vasconcelos, V., Antunes, A., 2017. First occurrence of cylindrospermopsin in Portugal: a contribution to its continuous global dispersal. Toxicon 130, 87\u00E2\u0080\u009390. doi:10.1016/j.toxicon.2017.02.024 Moreno-Castilla, C., L\u00C3\u00B3pez-Ram\u00C3\u00B3n, M.., Carrasco-Mar\u00C4\u00B1\u00CC\u0081n, F., Lopez-Ramon, M. V, Carrasco-Marin, F., 2000. Changes in surface chemistry of activated carbons by wet oxidation. Carbon N. Y. 38, 1995\u00E2\u0080\u00932001. doi:Doi 10.1016/S0008-6223(00)00048-8 Nagata, S., Soutome, H., Tsutsumi, T., Hasegawa, A., Sekijima, M., Sugamata, M., Harada, K.-I., Suganuma, M., Ueno, Y., 1995. Novel monoclonal antibodies against microcystin and their protective activity for hepatotoxicity. Nat. Toxins 3, 78\u00E2\u0080\u009386. doi:10.1002/nt.2620030204 Newcombe, G., Morrison, J., Hepplewhite, C., 2002. Simultaneous adsorption of MIB and NOM onto activated carbon. I. Characterisation of the system and NOM adsorption. Carbon N. Y. 40, 2135\u00E2\u0080\u00932146. doi:10.1016/S0008-6223(02)00097-0 Nicholson, B.C., Burch, M.D., 2001. Evaluation of analytical methods for detection and quantification of cyanotoxins in relation to Australian drinking water guidelines. Cooperative Research Centre for Water Quality and Treatment. Nicholson, B.C., Burch, M.D., 2001. Evaluation of Analytical Methods for Detection and Quantification of Cyanotoxins in Relation to Australian Drinking Water Guidelines. Nunn, P.B., Ponnusamy, M., 2009. \u00CE\u00B2-N-Methylaminoalanine (BMAA): Metabolism and 129 metabolic effects in model systems and in neural and other tissues of the rat in vitro. Toxicon 54, 85\u00E2\u0080\u009394. doi:10.1016/j.toxicon.2009.03.008 O\u00E2\u0080\u0099Neill, K., Musgrave, I.F., Humpage, A., 2016. Low dose extended exposure to saxitoxin and its potential neurodevelopmental effects: A review. Environ. Toxicol. Pharmacol. 48, 7\u00E2\u0080\u009316. doi:10.1016/j.etap.2016.09.020 Oehrle, S.A., Southwell, B., Westrick, J., 2010. Detection of various freshwater cyanobacterial toxins using ultra-performance liquid chromatography tandem mass spectrometry. Toxicon 55, 965\u00E2\u0080\u0093972. doi:10.1016/j.toxicon.2009.10.001 Ontario Ministry of Environment, 2014. Personal communication with S. Deshpande, Standards Development Branch, Ontario Ministry of the Environment. Ontario Ministry of Environment, 2012. Personal communication with S. Deshpande, Standards Development Branch, Ontario Ministry of the Environment. Orihel, D.M., Bird, D.F., Brylinsky, M., Chen, H., Donald, D.B., Huang, D.Y., Giani, A., Kinniburgh, D., Kling, H., Kotak, B.G., Leavitt, P.R., Nielsen, C.C., Reedyk, S., Rooney, R.C., Watson, S.B., Zurawell, R.W., Vinebrooke, R.D., Smith, R.E.H., 2012. High microcystin concentrations occur only at low nitrogen-to-phosphorus ratios in nutrient-rich Canadian lakes. Can. J. Fish. Aquat. Sci. 69, 1457\u00E2\u0080\u00931462. doi:10.1139/f2012-088 Osswald, J., Carvalho, A.P., Guimar\u00C3\u00A3es, L., Guilhermino, L., 2013. Toxic effects of pure anatoxin-a on biomarkers of rainbow trout, Oncorhynchus mykiss. Toxicon 70, 162\u00E2\u0080\u0093169. doi:10.1016/j.toxicon.2013.04.014 Paerl, H.W., Paul, V.J., 2012. Climate change: Links to global expansion of harmful cyanobacteria. Water Res. 46, 1349\u00E2\u0080\u00931363. doi:10.1016/j.watres.2011.08.002 Park, H.-D., Sasaki, Y., Maruyama, T., Yanagisawa, E., Hiraishi, A., Kato, K., 2001. 130 Degradation of the cyanobacterial hepatotoxin microcystin by a new bacterium isolated from a hypertrophic lake. Environ. Toxicol. 16, 337\u00E2\u0080\u0093343. doi:10.1002/tox.1041 Parsons, P.J., Camp, N.P., Edwards, N., Ravi Sumoreeah, L., 2000. Synthesis of (\u00C2\u00B1)-Anatoxin-a and Analogues. Tetrahedron 56, 309\u00E2\u0080\u0093315. doi:10.1016/S0040-4020(99)00909-6 Pavagadhi, S., Tang, A.L.L., Sathishkumar, M., Loh, K.P., Balasubramanian, R., 2013. Removal of microcystin-LR and microcystin-RR by graphene oxide: Adsorption and kinetic experiments. Water Res. 47, 4621\u00E2\u0080\u00934629. doi:10.1016/j.watres.2013.04.033 Pelekani, C., Newcombe, G., Snoeyink, V.L., Hepplewhite, C., Assemi, S., Beckett, R., 1999. Characterization of Natural Organic Matter Using High Performance Size Exclusion Chromatography. Environ. Sci. Technol. 33, 2807\u00E2\u0080\u00932813. doi:10.1021/es9901314 Pelekani, C., Snoeyink, V.L., 1999. Competitive adsorption in natural water: Role of activated carbon pore size. Water Res. 33, 1209\u00E2\u0080\u00931219. doi:10.1016/S0043-1354(98)00329-7 Pereira, P., Dias, E., Franca, S., Pereira, E., Carolino, M., Vasconcelos, V., 2004. Accumulation and depuration of cyanobacterial paralytic shellfish toxins by the freshwater mussel Anodonta cygnea. Aquat. Toxicol. 68, 339\u00E2\u0080\u0093350. doi:10.1016/j.aquatox.2004.04.001 Quinlivan, P.A., Li, L., \u00C3\u0083, D.R.U.K., 2005. Effects of activated carbon characteristics on the simultaneous adsorption of aqueous organic micropollutants and natural organic matter. Water Res. 39, 1663\u00E2\u0080\u00931673. doi:10.1016/j.watres.2005.01.029 Reichwaldt, E.S., Ghadouani, A., 2012. Effects of rainfall patterns on toxic cyanobacterial blooms in a changing climate: Between simplistic scenarios and complex dynamics. Water Res. 46, 1372\u00E2\u0080\u00931393. doi:10.1016/j.watres.2011.11.052 Ritchie, J.D., Michael Perdue, E., 2003. Proton-binding study of standard and reference fulvic acids, humic acids, and natural organic matter. Geochim. Cosmochim. Acta 67, 85\u00E2\u0080\u009393. 131 doi:10.1016/S0016-7037(02)01044-X Ritchie, J.D., Perdue, E.M., 2003. Proton-binding study of standard and reference fulvic acids, humic acids, and natural organic matter. Geochim. Cosmochim. Acta 67, 85\u00E2\u0080\u009396. doi:10.1016/S0016-7037(02)01044-X Rodr\u00C3\u00ADguez, E., Onstad, G.D., Kull, T.P.J., Metcalf, J.S., Acero, J.L., von Gunten, U., 2007. Oxidative elimination of cyanotoxins: Comparison of ozone, chlorine, chlorine dioxide and permanganate. Water Res. 41, 3381\u00E2\u0080\u00933393. doi:10.1016/j.watres.2007.03.033 Rodr\u00C4\u00B1, E., 2007. Oxidation of microcystins by permanganate\u00E2\u0080\u00AF: Reaction kinetics and implications for water treatment. Water Res. 41, 102\u00E2\u0080\u0093110. doi:10.1016/j.watres.2006.10.004 Rzymski, P., Poniedzia\u00C5\u0082ek, B., 2014. In search of environmental role of cylindrospermopsin: A review on global distribution and ecology of its producers. Water Res. 66, 320\u00E2\u0080\u0093337. doi:10.1016/j.watres.2014.08.029 Sahetya, T.J., Dixit, F., Balasubramanian, K., 2013. Waste citrus fruit peels for removal of Hg(II) ions. Desalin. Water Treat. 3994, 1\u00E2\u0080\u009313. doi:10.1080/19443994.2013.852483 Salmaso, N., Cerasino, L., 2012. Long-term trends and fine year-to-year tuning of phytoplankton in large lakes are ruled by eutrophication and atmospheric modes of variability. Hydrobiologia 698, 17\u00E2\u0080\u009328. doi:10.1007/s10750-012-1068-2 Sarathy, S.R., Bazri, M.M., Mohseni, M., 2011. Modeling the Transformation of Chromophoric Natural Organic Matter during UV/H2O2 Advanced Oxidation. J. Environ. Eng. 137, 903\u00E2\u0080\u0093912. doi:10.1061/(ASCE)EE.1943-7870.0000390 Saskatchewan Environment, 2012. Personal communication with S. Ferris, Municipal Branch, Saskatchewan Ministry of Environment. Sathishkumar, M., Pavagadhi, S., Vijayaraghavan, K., Balasubramanian, R., Ong, S.L., 2010. 132 Experimental studies on removal of microcystin-LR by peat. J. Hazard. Mater. 184, 417\u00E2\u0080\u0093424. doi:10.1016/j.jhazmat.2010.08.051 Sathya, K., Saranya, P., Swarnalatha, S., Mandal, A.B., Sekaran, G., 2015. Removal of microcystin-RR , a membrane foulant using exocellular polymer from Enterobacter ludwigii\u00E2\u0080\u00AF: Kinetic and isotherm studies. Desalination 369, 175\u00E2\u0080\u0093187. doi:10.1016/j.desal.2015.05.008 Sevilla, E., Martin-Luna, B., Vela, L., Teresa Bes, M., Luisa Peleato, M., Fillat, M.F., 2010. Microcystin-LR synthesis as response to nitrogen: transcriptional analysis of the mcyD gene in Microcystis aeruginosa PCC7806. Ecotoxicology 19, 1167\u00E2\u0080\u00931173. doi:10.1007/s10646-010-0500-5 Sharma, V.K., Triantis, T.M., Antoniou, M.G., He, X., Pelaez, M., Han, C., Song, W., Shea, K.E.O., De, A.A., Kaloudis, T., Hiskia, A., Dionysiou, D.D., 2012. Destruction of microcystins by conventional and advanced oxidation processes\u00E2\u0080\u00AF: A review. Sep. Purif. Technol. 91, 3\u00E2\u0080\u009317. doi:10.1016/j.seppur.2012.02.018 Shoemaker, J.., 2015. US EPA Method 544: Determination of microcystins and nodularin in drinking water by solid phase extraction and liquid chromatography/tandem mass spectrometry (LC/MS/MS). EPA/600/R-14/474. National Exposure Research Laboratory, Office of Research an Development, U.S. Environmental Protection Agency, Cicinnati, OH. Singer, P.C., Bilyk, K., 2002. Enhanced coagulation using a magnetic ion exchange resin. Water Res. 36, 4009\u00E2\u0080\u00934022. doi:10.1016/S0043-1354(02)00115-X Sinha, R., Pearson, L.A., Davis, T.W., Burford, M.A., Orr, P.T., Neilan, B.A., 2012. Increased incidence of Cylindrospermopsis raciborskii in temperate zones \u00E2\u0080\u0093 Is climate change responsible? Water Res. 46, 1408\u00E2\u0080\u00931419. doi:10.1016/j.watres.2011.12.019 133 Slunjski, M., Cadee, K., Tattersall, J., 2000. MIEX\u00C2\u00AE resin water treatment process. Proc. Aquatech 26\u00E2\u0080\u009329. Smith, M.J., Shaw, G.R., Eaglesham, G.K., Ho, L., Brookes, J.D., 2008. Elucidating the factors influencing the biodegradation of cylindrospermopsin in drinking water sources. Environ. Toxicol. 23, 413\u00E2\u0080\u0093421. doi:10.1002/tox.20356 Sordo, A., Metcalf, J.S., Acero, J.L., Rodr\u00C4\u00B1, E., 2007. Kinetics of the oxidation of cylindrospermopsin and anatoxin-a with chlorine , monochloramine and permanganate. Water Res. 41, 2048\u00E2\u0080\u00932056. doi:10.1016/j.watres.2007.01.033 Sorlini, S., Gialdini, F., Collivignarelli, C., 2013. Removal of cyanobacterial cells and Microcystin-LR from drinking water using a hollow fiber micro filtration pilot plant. Desalination 309, 106\u00E2\u0080\u0093112. doi:10.1016/j.desal.2012.09.028 Spoof, L., Neffling, M.-R., Meriluoto, J., 2010. Fast separation of microcystins and nodularins on narrow-bore reversed-phase columns coupled to a conventional HPLC system. Toxicon 55, 954\u00E2\u0080\u0093964. doi:10.1016/j.toxicon.2009.06.015 Sun, J., Li, X., Quan, Y., Yin, Y., Zheng, S., 2015. Effect of long-term organic removal on ion exchange properties and performance during sewage tertiary treatment by conventional anion exchange resins. Chemosphere 136, 181\u00E2\u0080\u0093189. doi:10.1016/j.chemosphere.2015.05.002 Sun, W.L., Xia, J., Li, S., Sun, F., 2012. Effect of natural organic matter (NOM) on Cu(II) adsorption by multi-walled carbon nanotubes: Relationship with NOM properties. Chem. Eng. J. 200\u00E2\u0080\u0093202, 627\u00E2\u0080\u0093636. doi:10.1016/j.cej.2012.06.118 Svrcek, C., Smith, D.W., 2004. Cyanobacteria toxins and the current state of knowledge on water treatment options: a review. J. Environ. Eng. Sci. 3, 155\u00E2\u0080\u0093185. doi:10.1139/s04-010 134 Swietlik, J., D, A., Raczyk-stanis, U., Nawrocki, J., 2004. Reactivity of natural organic matter fractions with chlorine dioxide and ozone 38, 547\u00E2\u0080\u0093558. doi:10.1016/j.watres.2003.10.034 Tan, Y., Kilduff, J.E., 2007. Factors affecting selectivity during dissolved organic matter removal by anion-exchange resins. Water Res. 41, 4211\u00E2\u0080\u00934221. doi:10.1016/j.watres.2007.05.050 Teng, W., Wu, Z., Fan, J., Chen, H., Feng, D., Lv, Y., Wang, J., Asiri, A.M., Zhao, D., 2013. Ordered mesoporous carbons and their corresponding column for highly efficient removal of microcystin-LR. Energy Environ. Sci. 6, 2765. doi:10.1039/c3ee41775a Tsuji, K., Watanuki, T., Kondo, F., Watanabe, M.F., Suzuki, S., Nakazawa, H., Suzuki, M., Uchida, H., Harada, K.-I., 1995. Stability of microcystins from cyanobacteria\u00E2\u0080\u0094II. Effect of UV light on decomposition and isomerization. Toxicon 33, 1619\u00E2\u0080\u00931631. doi:10.1016/0041-0101(95)00101-8 USEPA, 2015. Health Effects Support Document for the Cyanobacterial Toxin Microcystins. Washington, DC. Verdickt, L., Closset, W., Haeseleer, D. V, Cromphout, J., 2011. Applicability of ion exchange for NOM removal from a sulfate rich surface water incorporating full reuse of the brine, in: IWA Specility Conference on Natural Organic Matter. Costa Mesa, CA, USA. Walker, H.W., 2014. Harmful Algae Blooms in Drinking Water: Removal of Cyanobacterial Cells and Toxins. Advances in Water and Wastewater Transport and Treatment CRC Press. Walker, K.M., Boyer, T.H., 2011. Long-term performance of bicarbonate-form anion exchange: Removal of dissolved organic matter and bromide from the St. Johns River, FL, USA. Water Res. 45, 2875\u00E2\u0080\u00932886. doi:10.1016/j.watres.2011.03.004 Wang, D.-Z., Zhang, S.-F., Zhang, Y., Lin, L., 2016. Paralytic shellfish toxin biosynthesis in 135 cyanobacteria and dinoflagellates: A molecular overview. J. Proteomics 135, 132\u00E2\u0080\u0093140. doi:10.1016/j.jprot.2015.08.008 Wang, H., Ho, L., Lewis, D.M., Brookes, J.D., Newcombe, G., 2007. Discriminating and assessing adsorption and biodegradation removal mechanisms during granular activated carbon filtration of microcystin toxins. Water Res. 41, 4262\u00E2\u0080\u00934270. doi:10.1016/j.watres.2007.05.057 Weber, W.J., Morris, J.C., 1963. Kinetics of adsorption on carbon solution. J. Sanit. Engeering Div. Proceedings.American Soc. Civ. Eng. 89, 31\u00E2\u0080\u009360. Weller, M., 2013. Immunoassays and Biosensors for the Detection of Cyanobacterial Toxins in Water. Sensors 13, 15085\u00E2\u0080\u009315112. doi:10.3390/s131115085 Wells, M., 2015. Conductivity-Dependent Flow Field-Flow Fractionation of Fulvic and Humic Acid Aggregates. Chromatography 2, 580\u00E2\u0080\u0093593. doi:10.3390/chromatography2030580 Wert, E.C., Rosario-Ortiz, F.L., 2013. Intracellular Organic Matter from Cyanobacteria as a Precursor for Carbonaceous and Nitrogenous Disinfection Byproducts. Environ. Sci. Technol. 47, 6332\u00E2\u0080\u00936340. doi:10.1021/es400834k Westerhoff, P., Aiken, G., Amy, G., Debroux, J., 1999. Relationships between the structure of natural organic matter and its reactivity towards molecular ozone and hydroxyl radicals. Water Res. 33, 2265\u00E2\u0080\u00932276. doi:10.1016/S0043-1354(98)00447-3 Westrick, J.A., 2010. Chapter 3: Detection of cyanotoxins during potable water treatment. In: Manual of water supply practices M57: Algae\u00E2\u0080\u0094Source to treatment. American Water Works Association, Denver, CO. Wiese, M., D\u00E2\u0080\u0099Agostino, P.M., Mihali, T.K., Moffitt, M.C., Neilan, B.A., 2010. Neurotoxic Alkaloids: Saxitoxin and Its Analogs. Mar. Drugs 8, 2185\u00E2\u0080\u00932211. doi:10.3390/md8072185 136 Willison, H., Boyer, T.H., 2012. Secondary effects of anion exchange on chloride, sulfate, and lead release: Systems approach to corrosion control. Water Res. 46, 2385\u00E2\u0080\u00932394. doi:10.1016/j.watres.2012.02.010 Winter, J.G., DeSellas, A.M., Fletcher, R., Heintsch, L., Morley, A., Nakamoto, L., Utsumi, K., 2011. Algal blooms in Ontario, Canada: Increases in reports since 1994. Lake Reserv. Manag. 27, 107\u00E2\u0080\u0093114. doi:10.1080/07438141.2011.557765 Woolbright, B.L., Williams, C.D., Ni, H., Kumer, S.C., Schmitt, T., Kane, B., Jaeschke, H., 2017. Microcystin-LR induced liver injury in mice and in primary human hepatocytes is caused by oncotic necrosis. Toxicon 125, 99\u00E2\u0080\u0093109. doi:10.1016/j.toxicon.2016.11.254 Xiao, K., Xu, F., Jiang, L., Dan, Z., Duan, N., 2016. The oxidative degradation of polystyrene resins on the removal of Cr(VI) from wastewater by anion exchange. Chemosphere 156, 326\u00E2\u0080\u0093333. doi:10.1016/j.chemosphere.2016.04.116 Yasutaka, T., Naka, A., Sakanakura, H., Kurosawa, A., Inui, T., Takeo, M., Inoba, S., Watanabe, Y., Fujikawa, T., Miura, T., Miyaguchi, S., Nakajou, K., Sumikura, M., Ito, K., Tamoto, S., Tatsuhara, T., Chida, T., Hirata, K., Ohori, K., Someya, M., Katoh, M., Umino, M., Negishi, M., Ito, K., Kojima, J., Ogawa, S., 2017. Reproducibility of up-flow column percolation tests for contaminated soils. PLoS One 12, e0178979. doi:10.1371/journal.pone.0178979 Zamyadi, A., Coral, L.A., Barbeau, B., Dorner, S., Lapolli, F.R., Pr\u00C3\u00A9vost, M., 2015. Fate of toxic cyanobacterial genera from natural bloom events during ozonation. Water Res. 73, 204\u00E2\u0080\u0093215. doi:10.1016/j.watres.2015.01.029 Zamyadi, A., Dorner, S., Sauv\u00C3\u00A9, S., Ellis, D., Bolduc, A., Bastien, C., Pr\u00C3\u00A9vost, M., 2013. Species-dependence of cyanobacteria removal efficiency by different drinking water 137 treatment processes. Water Res. 47, 2689\u00E2\u0080\u00932700. doi:10.1016/j.watres.2013.02.040 Zamyadi, A., MacLeod, S.L., Fan, Y., McQuaid, N., Dorner, S., Sauv\u00C3\u00A9, S., Pr\u00C3\u00A9vost, M., 2012. Toxic cyanobacterial breakthrough and accumulation in a drinking water plant: A monitoring and treatment challenge. Water Res. 46, 1511\u00E2\u0080\u00931523. doi:10.1016/j.watres.2011.11.012 Zhang, M., Li, A., Zhou, Q., Shuang, C., Zhou, W., Wang, M., 2014. Effect of pore size distribution on tetracycline adsorption using magnetic hypercrosslinked resins. Microporous Mesoporous Mater. 184, 105\u00E2\u0080\u0093111. doi:10.1016/j.micromeso.2013.10.010 Zhao, H., Qiu, J., Fan, H., Li, A., 2013. Mechanism and application of solid phase adsorption toxin tracking for monitoring microcystins. J. Chromatogr. A 1300, 159\u00E2\u0080\u0093164. doi:10.1016/j.chroma.2013.02.048 Zheng, B., Fu, H., Berry, J.P., McCord, B., 2016. A rapid method for separation and identification of microcystins using capillary electrophoresis and time-of-flight mass spectrometry. J. Chromatogr. A 1431, 205\u00E2\u0080\u0093214. doi:10.1016/j.chroma.2015.11.034 138 Appendices Appendix A VUV-IX Combined Treatment Figure A.1 MCLR Degradation by VUV at Different Dosages 0102030405060708090100MCLR Removal (%)VUV MCLR Degradation200 mJ/cm2 120 mJ/cm2 90 mJ/cm2 60 mJ/cm2 30 mJ/cm2139 Figure A.2 MCLR Degradation by VUV (IX Pretreatment) 0102030405060708090100Combined VUV-IX for MCLR Removal0 mJ/cm230 mJ/cm260 mJ/cm290 mJ/cm2120 mJ/cm2200 mJ/cm21000 mg/L 500 mg/L 200 mg/L 50 mg/L 10 mg/L 140 Appendix B Determination of Rate Limiting Step B.1 Pseudo Second Order Kinetics The pseudo-second-order reaction considers that the rate is directly proportional to the number of active surface sites present on the adsorbent surface. The pseudo-second-order equation is given as: \u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u00A1=1k2.\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092+\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 (1) where k2 (g mg\u00E2\u0088\u00921 min\u00E2\u0088\u00921 ) is the pseudo-second-order rate constant (Pavagadhi et al., 2013; Sahetya et al., 2013). B.2 Pseudo First Order Kinetics Lagergren demonstrated that the rate of adsorption of solute on the adsorbent depends on the adsorption capacity and follows the pseudo-first-order equation used to estimate the kad, the mass transfer coefficient. The pseudo-first-order rate equation is given as: log(\u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 \u00E2\u0088\u0092 \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u00A1) = \u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u009C\u00F0\u009D\u0091\u0094 \u00F0\u009D\u0091\u009E\u00F0\u009D\u0091\u0092 +\u00F0\u009D\u0091\u009812.303\u00F0\u009D\u0091\u00A1 (2) where qe and qt are the amounts of adsorbed Hg(II) ions on the adsorbent at equilibrium and at time t respectively (mg/g), and k1 is the first-order adsorption rate constant (min\u00E2\u0088\u00921 ) (Sahetya et al., 2013; Sathishkumar et al., 2010). B.3 Biot Number Estimation By using the dimensionless Biot number representing the ratio of internal mass transfer 141 (pore diffusion) to external mass transfer (film diffusion) impedances, the rate-controlling step was determined using the following equation (Bazri and Mohseni, 2016) \u00F0\u009D\u0090\u00B5\u00F0\u009D\u0091\u0096 =\u00F0\u009D\u0090\u00BE\u00F0\u009D\u0091\u0093.\u00F0\u009D\u0091\u0085\u00F0\u009D\u0091\u009D\u00F0\u009D\u0090\u00B7\u00F0\u009D\u0091\u009D\u00F0\u009D\u0091\u0092 (3) where kf is the external mass transfer coefficient (kf :Df/\u00CE\u00B4) and Df, is film diffusion coefficient and Dp,e is effective pore diffusion coefficient (\u00CE\u00B4 is film thickness \u00E2\u0089\u0088 10-3 cm) and were estimated from experimental data using non linear optimization schemes (Bazri and Mohseni, 2016). Resin beads were assumed to be spherical with an average radius (Rp) of 375 \u00CE\u00BCm (resin size range reported by the manufacturer, Rp : 150 \u00E2\u0080\u0093 600 \u00CE\u00BCm). The Bi << 1 indicates film diffusion as the rate-limiting step where Bi >>1 shows pore diffusion to be the rate limiting step. Da accounts for free liquid diffusion (Dl) and sorption to resins resistances, and tortuous diffusion pathway through inside the resins and is correlated to effective pore diffusivity (Dp,e) as follows (Weber Jr and DiGiano, 1996): Da=\u00F0\u009D\u0090\u00B71.\u00CE\u00B5/\u00CF\u0084[(1\u00E2\u0088\u0092\u00CE\u00B5) \u00CF\u0081s KD + \u00CE\u00B5 ]=\u00F0\u009D\u0090\u00B7\u00F0\u009D\u0091\u009D\u00F0\u009D\u0091\u0092.\u00CE\u00B5 [(1\u00E2\u0088\u0092\u00CE\u00B5) \u00CF\u0081s KD + \u00CE\u00B5 ] (4) \u00E2\u0081\u0084 Where KD is the linear equilibrium partition coefficient, \u00CE\u00B5 is the resin porosity ~ 0.46 (Bazri and Mohseni, 2016), \u00CF\u0084 is the tortuosity of the resin and is estimated to be ~ 3 (Bazri and Mohseni, 2016; Li and Sengupta, 2000; Li and SenGupta, 2000). \u00CF\u0081s is solid phase density of 1.08 g/cm3 (manufacturer), and Dp,e is effective pore diffusion coefficient (cm2/s). 142 \u00CE\u00B5/\u00CF\u0084 accounts for reduction in Dl because of the tortuosity of the diffusion path and the term [(1-\u00CE\u00B5) \u00CF\u0081s KD + \u00CE\u00B5 ] is referred to as retardation factor, by which the liquid diffusivity is reduced due to local microscale partitioning. Assuming a linear distribution of toxin between the solid and liquid phases was plausible because of the low concentrations of the solute (i.e, ~ 1-100 \u00C2\u00B5g MCLR/L). The R2 values obtained for the linear correlation were between 0.92-0.98. 143 Appendix C Charge Density and Resin Porosity Estimation C.1 Charge Density Estimation Charge densities for all NOM fractions were estimated using the Henderson-Hasselbalch equation as described elsewhere (Bazri and Mohseni, 2016; Ritchie and Michael Perdue, 2003). The estimated charge densities of all the NOM fractions are illustrated in Table A.1. The resin exchange capacity was 0.8 meq/mL (Manufacturer) and 1 mL of wet resin approximated 221 mg after drying in desiccator (Bazri and Mohseni, 2016). C.2 Resin Porosity Estimation Mercury porosimetry tests were performed to determine the resin pore volume distribution as described elsewhere (Giesche, 2006). The resin porosity data for A860 is illustrated in Table C.1. Table C.1 Resin Porosimetry Results for IX Resin (Purolite A860) Pore Size Pore Volume (mL/g) Pore Volume (%) Cumulative Pore Area (m2/g) Pore Area (%) Mesopores (<2nm) 0.00906 56 1.56 74.6 Micropores (2-50 nm) 0.00435 26 0.53 25.3 Macropores (>50 nm) 0.00311 18 0.001 < 0.1 144 Appendix D Conference Presentations \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Anion exchange resins for the removal of cyanobacterial toxins from surface water. Water Quality and Technology Conference. Oral Presentation. Indianapolis, IN. November 2016. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Removal of Algal toxins from surface water by anion exchange resins. 17th Canadian National Drinking Water Conference. Oral Presentation. Ottawa, ON. October 2016. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Algal toxin removal using strongly basic anion exchange resins. IC-IMPACTS Summer Institute on Nanotechnologies. Oral Presentation. Edmonton, AB. June 2016. \u00EF\u0082\u00B7 Dixit, F., Chintalapati, P., Mohseni, M. Algal Toxins- What are the viable treatment options for Small Systems. British Columbia Water and Waste Association (BCWWA) Annual Conference and Tradeshow. Oral Presentation. Whistler, BC. April 2016. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Cyanobacterial toxin removal using ion exchange resins. RES\u00E2\u0080\u0099EAU WaterNET Annual General Meeting. Poster Presentation. Whistler, BC. April 2016. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Ion exchange resins for the removal of Microcystin-LR from surface water. IC-IMPACTS Annual General Meeting 2016. Oral Presentation. Vancouver, BC. March 2016. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Algal toxin removal using anion exchange resins. IC-IMPACTS Summer Institute on Optical Sensing. Oral Presentation. Toronto, ON. July 2015. 145 \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Algal toxin removal using strongly basic ion exchange resins. Water and Engineering Student Talks (WEST) Conference. Poster Presentation. Vancouver, BC. June 2015. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Anion exchange resins for removal of algal toxins from surface water. RES\u00E2\u0080\u0099EAU WaterNET Annual General Meeting. Poster Presentation. Kelowna, BC. April 2015. \u00EF\u0082\u00B7 Dixit, F., Mohseni, M. Anion exchange resins for the removal of cyanobacterial toxins from surface water. IC-IMPACTS Annual General Meeting 2015. Poster Presentation. Vancouver, BC. March 2015. "@en . "Thesis/Dissertation"@en . "2017-11"@en . "10.14288/1.0355879"@en . "eng"@en . "Chemical and Biological Engineering"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "Attribution-NonCommercial-NoDerivatives 4.0 International"@* . "http://creativecommons.org/licenses/by-nc-nd/4.0/"@* . "Graduate"@en . "Anion exchange resins for the removal of microcystins from surface water"@en . "Text"@en . "http://hdl.handle.net/2429/63181"@en .