Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

UV/Vacuum-UV Advanced Oxidation Process for the treatment of micropollutants from drinking water sources… Serrano Mora, Adrian 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_may_serranomora_adrian.pdf [ 2.41MB ]
Metadata
JSON: 24-1.0228628.json
JSON-LD: 24-1.0228628-ld.json
RDF/XML (Pretty): 24-1.0228628-rdf.xml
RDF/JSON: 24-1.0228628-rdf.json
Turtle: 24-1.0228628-turtle.txt
N-Triples: 24-1.0228628-rdf-ntriples.txt
Original Record: 24-1.0228628-source.json
Full Text
24-1.0228628-fulltext.txt
Citation
24-1.0228628.ris

Full Text

  UV/VACUUM-UV ADVANCED OXIDATION PROCESS FOR THE TREATMENT OF MICROPOLLUTANTS FROM DRINKING WATER SOURCES UNDER COMMON OPERATIONAL TEMPERATURES  by  Adrian Serrano Mora B.Sc., Universidad de Costa Rica, 2013  A THESIS SUBMITTED IN PARTIAL FULLFILMENT 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) April 2016  ©Adrian Serrano Mora, 2016 ii  ABSTRACT Vacuum-UV Advanced Oxidation Process (AOP) has been identified as a viable candidate to treat chemical contaminants in drinking water. In contrast to the commercially available and widely applied UV/H2O2 AOP, VUV AOP does not rely on the addition of chemicals for the generation of hydroxyl radicals. As a result, the technology is very appealing to small systems and rural communities which often lack infrastructure and qualified personnel to operate complex systems.  The primary objective of this research was to investigate the effect of temperature on the removal efficiency and energy consumption of a UV/VUV flow-through photoreactor. Additionally, the optical absorption properties at 185 nm of water and naturally occurring solutes such as Natural Organic Matter (NOM), nitrate, sulphate and chloride were studied between 3.6 and 25.0 °C. The secondary objective of this research was to investigate the effect of UV/VUV treatment on the removal of a mixture of micropollutants (e.g., pharmaceuticals, hormones, pesticides).  The results showed that temperature does not have a significant impact on process efficiency and energy consumption. Moreover, owing to the presence of naturally occurring solutes, it was found that the production of OH radicals was to some extent limited at 3.6 °C, therefore slightly lower removals were observed at low temperature. When evaluated on a mixture of micropollutants, the UV/VUV process could potentially degrade a variety of substances with feasible energy consumption. Other than for two of the 12 micropollutants tested, at a flow-rate of 1.9 L min-1, the VUV process was competitive to the UV/H2O2 process.   iii  PREFACE  My contribution to this work was to conduct the literature review, identify research gaps, develop the research proposal, design and run the experiments, analyze experimental data and write this document under the supervision of Dr. Madjid Mohseni, Professor of the Chemical and Biological Engineering Department at the University of British Columbia. Professor Mohseni largely contributed to the development of experimental plans, discussion of results and the revision of this manuscript draft. The planning of the experiments and analysis of the samples that lead to the results that are showed in Chapter 6 were performed in collaboration with Dr. Benoit Barbeaú, Professor of the Civil, Geological and Mining Engineering Department at École Polytechnique de Montreal.  iv  TABLE OF CONTENTS  ABSTRACT .............................................................................................................................................. ii PREFACE ............................................................................................................................................... iii TABLE OF CONTENTS .................................................................................................................... iv LIST OF TABLES ................................................................................................................................ vii LIST OF FIGURES ................................................................................................................................ x ACKNOWLEDGEMENTS .............................................................................................................. xiii DEDICATION .................................................................................................................................... xiv 1. INTRODUCTION .................................................................................................................... 1 1.1. Advanced Oxidation Processes (AOPs) .............................................................................. 1 1.2. UV/Vacuum-UV Advanced Oxidation Process ................................................................ 3 1.3. Limitations of the UV/VUV Advanced Oxidation Process ............................................ 4 1.4. Economic and Energetic Feasibility ..................................................................................... 5 1.5. Effect of Temperature ............................................................................................................ 5 1.6. Project Summary ..................................................................................................................... 6 2. LITERATURE REVIEW ......................................................................................................... 7 2.1. Removal of Common Micropollutants by UV/VUV ........................................................ 7 2.2. Impact of Natural Organic Matter (NOM) and Other Naturally Occurring Ions on the UV/VUV Process Efficiency ...................................................................................................... 8 2.3. Impact of Hydrodynamics on the UV/VUV Process Efficiency .................................... 9 2.4. Comparison between UV/VUV and Other Advanced Oxidation Processes .............. 10 2.5. Absorbance of Pure Water and Inorganic Ions at 185 nm ............................................. 10 2.6. Quantum Yield Dependence on Temperature ................................................................. 11 3. KNOWLEDGE GAPS AND THESIS OBJECTIVES .................................................... 12 3.1. Knowledge Gaps ................................................................................................................... 12 3.2. Hypothesis .............................................................................................................................. 12 3.3. Thesis Objectives .................................................................................................................. 12 3.4. Research Significance ............................................................................................................ 13 4. MATERIALS AND METHODS .......................................................................................... 14 TABLE OF CONTENTS v  4.1. Chemicals................................................................................................................................ 14 4.2. Raw Water Characteristics ................................................................................................... 14 4.3. Experimental Apparatus ....................................................................................................... 15 4.3.1. UV/VUV Collimated Beam ........................................................................................ 15 4.3.2. UV Collimated Beam.................................................................................................... 16 4.3.3. UV/VUV Photoreactor ............................................................................................... 17 4.3.4. Vacuum-UV Absorption Setup .................................................................................. 18 4.4. Experimental Methodology ................................................................................................. 20 4.4.1. Temperature Effect ...................................................................................................... 20 4.4.2. Energy Consumption ................................................................................................... 21 4.4.3. Removal Equivalent Dose ........................................................................................... 22 4.4.4. Micropollutant Degradation ........................................................................................ 24 4.4.5. Optical Properties at 185 nm and Their Dependence on Temperature ............... 25 4.5. Analytical Methods ........................................................................................................... 26 4.5.1. HPLC-UV ...................................................................................................................... 26 4.5.2. HPLC-APCI-MS/MS ................................................................................................... 27 4.5.3. HPSEC ........................................................................................................................... 27 4.5.4. Total Organic Carbon (TOC) ..................................................................................... 27 4.5.5. Ion Chromatography (IC) ............................................................................................ 27 4.5.6. UV Absorbance ............................................................................................................. 28 4.5.7. Other Analytical Methods............................................................................................ 28 5. EFFECT OF TEMPERATURE ............................................................................................ 29 5.1. Absorption of Radiation at 185 nm .................................................................................... 29 5.2. Removal Efficiency with Distilled Water .......................................................................... 35 5.3. Removal Efficiency with Raw Water .................................................................................. 39 5.4. Energy Requirements ........................................................................................................... 44 5.5. Summary ................................................................................................................................. 50 6. TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS .................. 51 6.1. Preliminary Kinetic Study .................................................................................................... 52 6.2. Flow-through Photoreactor ................................................................................................. 63 6.3. Estimation of the Electrical Energy per Order................................................................. 66 6.4. Summary ................................................................................................................................. 67 TABLE OF CONTENTS vi  7. CONCLUSIONS AND RECOMMENDATIONS ........................................................... 69 7.1. Overall Conclusions .............................................................................................................. 69 7.2. Research Outcome Significance .......................................................................................... 70 7.3. Recommendations for Future Work .................................................................................. 70 REFERENCES ...................................................................................................................................... 71 APPENDICES ...................................................................................................................................... 81 A. Corrections to the Absorbance Measurements ..................................................................... 81 B. Supplementary Data for the Absorption of Major Solutes at 185 nm .............................. 82 C. Supplementary Data for the Removal of Micropollutants .................................................. 86      vii  LIST OF TABLES Table 4.1 List and characteristics of the chemicals used .................................................................. 14 Table 4.2 Seymour Reservoir Raw Water Characteristics ................................................................ 15 Table 4.3 Solute concentration and pathlengths used in the absorption experiments ................ 26 Table 5.1 Average absorption coefficient at 185 nm of pure water at different temperatures and its impact on the thickness of the illuminated volume. Error bars represent the standard deviation among three replicates ......................................................................................................... 30 Table 5.2 Comparison of the absorption coefficients at 185 nm of pure water reported in literature and the one calculated in this study .................................................................................... 31 Table 5.3 Average values for the molar absorption coefficient at 185 nm of nitrate at different temperatures. Error bars represent the standard deviation among three replicates .................... 32 Table 5.4 Average values for the molar absorption coefficient at 185 nm of sulphate at different temperatures. Error bars represent the standard deviation among three replicates .... 32 Table 5.5 Average values for the molar absorption coefficient at 185 nm of chloride at different temperatures. Error bars represent the standard deviation among three replicates .... 32 Table 5.6 Comparison of the molar absorption coefficient at 185 nm of chloride and sulphate reported in literature (Weeks et al., 1963) and the one calculated in this study ............................ 33 Table 5.7 Average values for the molar absorption coefficient at 185 nm of Suwannee River NOM at different temperatures. Error bars represent the standard deviation among three replicates .................................................................................................................................................. 33 Table 5.8 Comparison between the measured and the calculated absorption coefficient for the raw water used in experimentation at different temperatures. ........................................................ 34 Table 5.9 Average air gap temperature under different operational conditions for an amalgam Hg lamp ................................................................................................................................................... 35 Table 5.10 Average absorption coefficient at 185 nm of raw water at different temperatures and its impact on the thickness of the illuminated volume. Error represents the standard deviation among three replicates ......................................................................................................... 40 Table 5.11 Fraction of 185 nm photons absorbed by each component of the aqueous matrix at different temperatures ........................................................................................................................... 42 Table 5.12 Flow-rates and number of passes used in the determination of the Electrical Energy per Order of the system ........................................................................................................................ 45 LIST OF TABLES viii  Table 5.13 Real number of passes through the photoreactor to observe a 1-log removal of atrazine at 3.6 °C .................................................................................................................................... 47 Table 5.14 Comparison between EEO values reported in literature and the one obtained in this study for the removal of atrazine ......................................................................................................... 50 Table 6.1 Photochemical parameters of the substances present in the micropollutant mixture 54 Table 6.2 UV254nm direct photolysis first order reaction rate for each micropollutant ................. 55 Table 6.3 Global pseudo-first order reaction rate constant for the removal of each micropollutant. Error represents a 95% confidence interval for the slope of the linear regression ................................................................................................................................................ 57 Table 6.4 OH radical degradation pathway pseudo-first order reaction rate constant for the removal of each micropollutant ........................................................................................................... 58 Table 6.5 Reaction rate constant, concentration and scavenging strength for each major solute in the water matrices tested .................................................................................................................. 58 Table 6.6 Formation of some chlorine based radicals from chloride ............................................ 59 Table 6.7 Removal observed for each micropollutant at a flow rate of 1.9 L min-1 with different water matrices. Error represents the uncertainty in the removal calculated for two samples .... 63 Table 6.8 Estimated absorption coefficient of each water matrix and 185 nm photon scavenging fraction of each major solute in solution ....................................................................... 64 Table 6.9 Estimated Electrical Energy per Order for each micropollutant in several water matrices at 1.9 L min-1 and 20 °C. Error represents the uncertainty associated in the EEO calculation................................................................................................................................................ 66 Table A.1 Refractive index used in the calculation of the Fresnel relation ................................... 81 Table C.1 Removal observed for each micropollutant at a flow rate of 2.7 L min-1 with different water matrices. Error represents the uncertainty in the removal ................................... 86 Table C.2 Estimated Electrical Energy per Order for each micropollutant in several water matrices at 2.7 L min-1 and 20.5 °C. Error represents the calculated uncertainty for the EEO .. 86 Table C.3 Estimated 1-log removal 254 nm based dose for each micropollutant in several water matrices at 20.5 °C. Error represents the calculated uncertainty for the 254 nm based dose .......................................................................................................................................................... 87 Table C.4 Estimated number of passes required to achieve 1-log removal for each micropollutant in several water matrices at 1.9 L min-1 and 20.5 °C. Error represents the calculated uncertainty for the 254 nm based dose ............................................................................ 87 LIST OF TABLES ix  Table C.5 Estimated number of passes required to achieve 1-log removal for each micropollutant in several water matrices at 2.7 L min-1 and 20.5 °C. Error represents the calculated uncertainty for the 254 nm based dose ............................................................................ 88    x  LIST OF FIGURES Figure 4.1 UV/VUV collimated beam apparatus with the Petri dish reacting vessel .................. 16 Figure 4.2 UV collimated beam apparatus ......................................................................................... 17 Figure 4.3 Diagram and cross section of the UV/VUV photoreactor .......................................... 18 Figure 4.4 Flow-through UV/VUV photoreactor ............................................................................ 18 Figure 4.5 Experimental setup to determine the absorption coefficient and molar absorption coefficient at 185 nm at different temperatures ................................................................................ 19 Figure 4.6 Analytical analyses performed on the samples obtained during the experiments ..... 20 Figure 4.7 Experimental setup to study the effect of temperature on the removal of model micropollutant (i.e., atrazine) ................................................................................................................ 21 Figure 4.8 Experimental setup to determine the EEO for the removal of atrazine ..................... 22 Figure 4.9 Information required to calculate the 254 nm based removal equivalent dose ......... 23 Figure 4.10 Dependence of Atrazine removal with the 254 nm based dose delivered to the aqueous media. Equation shows the curve that best fit the experimental data ............................ 24 Figure 5.1 Comparison in light attenuation intensity by a substance with a) lower absorption coefficient b) higher absorption coefficient through an identical pathlength ............................... 29 Figure 5.2 Thermocouple positioning inside the air gap of the photoreactor .............................. 35 Figure 5.3 Removal efficiency dependence on temperature at two different flow-rates for an amalgam Hg lamp. Error bars represent the calculated uncertainty for the removal of two replicates .................................................................................................................................................. 36 Figure 5.4 Removal efficiency dependence on temperature at two different flow-rates for a standard Hg lamp. Error bars represent the calculated uncertainty for the removal of two replicates .................................................................................................................................................. 37 Figure 5.5 185 nm photon absorption profile at different temperatures....................................... 38 Figure 5.6 Removal efficiency of atrazine in raw water at two different temperatures. Error bars represent the calculated uncertainty for the removal of four replicates ................................ 39 LIST OF FIGURES xi  Figure 5.7 Comparison of the 185 nm photon absorption profile in the photoreactor between water matrices at different temperatures ............................................................................................ 41 Figure 5.8 Comparison of the electrical energy consumed by the system at different temperatures with a flow-rate of 2.7 L min-1. Error bars represent the standard deviation in the energy measurements of four replicates ............................................................................................. 44 Figure 5.9 Kinetics of atrazine degradation in the flow-through photoreactor. Units of the pseudo first order constant apparent reaction rate are cm2 mJ-1 ..................................................... 46 Figure 5.10 254 nm based dose delivered per pass in the photoreactor at different flow-rates a) 0.5 L min-1; b) 1.0 L min-1; c) 1.7 L min-1 & d) 2.9 L min-1 .............................................................. 47 Figure 5.11 Electrical Energy per Order for the removal of atrazine at 3.6 °C and different flow-rates. Error bars represent the calculated uncertainty for the EEO for two replicates ....... 48 Figure 5.12 Energy consumed by the lamp and ballast at each flow-rate to achieve a 1-log removal of atrazine and its relationship with the EEO ..................................................................... 49 Figure 6.1 Names and chemical structures of the substances present in the micropollutant mixture ..................................................................................................................................................... 51 Figure 6.2 Overall average removal observed by direct photolysis. Error bars represent the standard deviation among the removal observed for each water matrix tested ........................... 56 Figure 6.3 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water ........................................................................................................ 61 Figure 6.4 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water + 50 ppm Carbonate .................................................................. 61 Figure 6.5 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water + 2 ppm SRNOM ....................................................................... 62 Figure 6.6 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water + 25 ppm Chloride ..................................................................... 62 Figure 6.7 185 nm photon absorption profile in the photoreactor with different water matrices .................................................................................................................................................................. 65 Figure B.1 Linearity in the absorbance of nitrate solutions at 3.6 °C ............................................ 82 Figure B.2 Linearity in the absorbance of nitrate solutions at 17.5 °C .......................................... 82 Figure B.3 Linearity in the absorbance of sulphate solutions at 3.6 °C ......................................... 83 LIST OF FIGURES xii  Figure B.4 Linearity in the absorbance of sulphate solutions at 17.5 °C ....................................... 83 Figure B.5 Linearity in the absorbance of chloride solutions at 3.6 °C ......................................... 84 Figure B.6 Linearity in the absorbance of chloride solutions at 17.5 °C ....................................... 84 Figure B.7 Linearity in the absorbance of SRNOM solutions at 3.6 °C ........................................ 85 Figure B.8 Linearity in the absorbance of SRNOM solutions at 17.5 °C...................................... 85    xiii  ACKNOWLEDGEMENTS First and foremost, I would like to express my gratitude and sincere thanks to my supervisor Dr. Madjid Mohseni for giving me the opportunity to work under his supervision. I appreciate his guidance, leadership, feedback and support throughout this research.  Keyvan Maleki and Heidi Backous: thanks for making working for RES’EAU WaterNET a pleasure and a wonderful experience. My sincere appreciation for the CHBE staff: Doug Yuen, Graham Liebelt, Charles Cheung, Alex Thng, Serge Millaire, Gordon Cheng, Richard Ryoo, Ivan Leversage, Marlene Chow and the administrative staff, specially to Helsa Leong. Thanks to all of you for your hard work. Special thanks to Metro Vancouver for providing me the raw water used throughout the course of the project. I would like to acknowledge École Polytechnique de Montreal, particularly to Dr. Barbeaú and Sung Vo Duy for their time and collaboration in conducting the analytical analysis of the PPCPs samples.  To all the research group and fellow summer students for creating the nice environment needed to spend long days in the lab. Zaki, Laith, Ramin, Mehdi, Adel, Ata, Pranav, Sean, Saad, Umar, John, Macarena and Clara: Thanks for all the discussions, pieces of advice and coffee breaks throughout the past couple of years.  To my mentors at Universidad de Costa Rica, Esteban Durán and Adolfo Ulate, thanks for your support and willingness to help. I look forward to joining the team in the near future. Last but not least, to my friends and family back in Costa Rica, thanks for making me feel that I never left.   xiv  DEDICATION         To mom, dad and sister, thanks for your endless love and support.   1  1. INTRODUCTION Chemical contamination of water bodies by natural and man-made substances has been identified as a key environmental problem. Many of these chemical contaminants are classified as micropollutants, since they are present in concentrations of ng L-1. Substances of mass consumption such as herbicides, pesticides, hormones, pharmaceuticals and personal care products are examples of micropollutants. Some of these substances are known to bio-accumulate (thus moving their way up in the food chain), they can also have adverse effects on the endocrine system of animals and others may promote bacterial resistance (Schwarzenbach, et al., 2006). Conventional water treatment processes (e.g., coagulation, flocculation and filtration) are generally ineffective at removing many of the micropollutants found in water sources. Historically, two methods have been applied in commercial scale to remove or degrade these contaminants: ozone in combination with granulated activated carbon (GAC) adsorption, and UV-hydrogen peroxide (H2O2) advanced oxidation process (AOP). In spite of the fact that UV/H2O2 technology has been successfully applied in large scale in several locations around the world, it suffers from the high cost of hydrogen peroxide and the added complexity of post treatment monitoring and quenching of the unreacted chemical. Moreover, given the remoteness in which small and rural communities are often located, the transportation and storing of large quantities of chemicals becomes a logistic challenge. Therefore, the implementation of this technology in small systems (i.e., up to 500 people) (Government of British Columbia), which often lack trained and specialized personnel, funds and infrastructure, becomes infeasible.  1.1. Advanced Oxidation Processes (AOPs) Over the years, advanced oxidation processes (AOPs) have been positioned as an effective way to degrade micropollutants from drinking water. AOPs refer to the group of technologies that rely on the formation of hydroxyl radicals (OH) to induce the oxidation of chemical compounds in water (Andreozzi et al., 1999). The OH radical is known to be a powerful oxidant. It has the second largest oxidation potential, with a value of 2.80 V (Beltrán F. J., 2003) and reacts rapidly and unselectively with other chemical substances with reaction INTRODUCTION 2  rate constants in the order of 108 to 1010 M-1 s-1 (Buxton et al., 1988). There are different ways in which hydroxyl radicals can be formed in aqueous solutions. For instance, the Fenton and Photo Fenton processes use iron salts (Fe2+ and Fe3+) in combination with hydrogen peroxide to generate OH radicals (Andreozzi et al., 1999; Oppenländer, 2003); photocatalysis on the other hand, relies on the excitation of a semi-conductor (i.e., TiO2) by UV radiation to reduce water and generate the radicals (Mills et al., 1993; Andreozzi et al., 1999); and finally, the decomposition of ozone (Tomiyasu et al., 1985; Hoigné, 1988) can be enhanced by UV radiation resulting in a high yield of OH radicals (Glaze et al., 1988). UV/Hydrogen Peroxide Advanced Oxidation Process The UV/H2O2 process is perhaps the most extensively studied advanced oxidation process (Stefan & Williamson, 2004; Tühkanen, 2004). It relies on the cleavage of hydrogen peroxide by the action of UV radiation, thus forming two hydroxyl radicals (Baxendale & Wilson, 1957; Andreozzi et al., 1999). However, hydrogen peroxide poorly absorbs UV radiation, reason why a relatively high concentration is needed to generate enough OH radicals in the system. Considering the cost of hydrogen peroxide at around 0.76 USD per kilogram (USP Technologies) and the concentration required in the system (minimum 5 mg L-1), the overall cost of operation may escalate quickly. Despite its high cost, UV/H2O2 process is the only advanced oxidation technology that has been commercialized and applied in the large scale. To just name a few, a proprietary technology by Trojan UV (TrojanUVPhox™) is being used in the Tucson Airport Remediation Project for the treatment of 1,4-Dioxane at a design operating flow-rate of 1317 m3 hr-1. The system is able to achieve a removal greater than 97% (1.6 log reduction) and the resulting water benefits approximately 50000 users. In Big Spring, Texas, UV/H2O2 AOP is used for the potable reuse of secondary waste water (flow rate up to 283 m3 hr-1) containing 1,4-Dioxane (0.5 log removal) and NDMA (1.2 log removal) for a population of 27000 people. In Orange County, California, the Groundwater Replenishment System is currently the largest indirect potable reuse plant with a flow rate of 11,040 m3 hr-1 with an oxidative step to degrade NDMA and 1,4-Dioxane to meet the State of California regulations to levels below 10 ng L-1 (ppt) and 3 μg L-1 (ppb), respectively. Finally in Aurora, Colorado, in the Aurora Reservoir Water Purification Facility, 7,886 m3 hr-1of water is treated for the reduction of NDMA (1.2 INTRODUCTION 3  log removal), taste and odour compounds (2-MIB and geosmin), microcystin (an algal toxin), atrazine (a pesticide) and chlorotetracycline (a pharmaceutical). 1.2. UV/Vacuum-UV Advanced Oxidation Process In recent years, UV/Vacuum-UV (VUV) has emerged as a strong candidate to fulfill the role of UV/H2O2 in small systems. Radiation in the Far or Deep-UV, as is alternatively named, is able to induce water photolysis (Barrett & Baxendale, 1960), resulting in the formation of hydroxyl radicals (•OH) without the addition of chemicals. Hence, there is no need for additional storing infrastructure and the hazards associated with chemical handling are eliminated, thus resulting in a safer process for the operator. Moreover, taking into account the remoteness in which small and rural water systems are often located, the implementation of a chemical free process, that significantly reduces the chemical supply chain and potentially the overall cost of operation, becomes appealing. Water photolysis occurs through two different photochemical reactions: water homolysis (1.1) and water ionization (1.2). The quantum yield (i.e., quantum efficiency) reported for the formation of hydroxyl radicals is 0.33 (Gettoff & Schenck, 1968) and 0.045 (Sokolov & Stein , 1966), for equations (1.1) and (1.2), respectively. 𝐻2𝑂 + ℎ𝑣185𝑛𝑚 → 𝐻𝑂• + 𝐻• (1.1) 𝐻2𝑂 + ℎ𝑣185𝑛𝑚 → 𝐻𝑂• + 𝐻+ + 𝑒𝑎𝑞−  (1.2) High energy photons (i.e., 185 nm photons) are emitted by low pressure mercury lamps, which have their main emissions at 254 nm (253.7 nm) (Oppenländer, 2003). As reported by Masschelein (2002), the energy output at 185 nm is approximately 8% of the one measured at 254 nm. The strong absorption of oxygen at wavelengths below 200 nm, requires the purging of the lamp surroundings and the path to the solution with nitrogen or to work under a vacuum (thus, the denomination Vacuum-UV) to prevent the formation of ozone from molecular oxygen and to maximize the transmission of VUV photons. In order to aid in the transmission of 185 nm photons from the lamp to the solution, the lamp and the protective sleeve must be fabricated with natural or fused-quartz (Masschelein, 2002; Oppenländer, 2003).  INTRODUCTION 4  Once in the aqueous media, photons are absorbed by water and react according to reactions 1.1 & 1.2. Given the absorption coefficient of water at 185 nm, as reported by Weeks et al. (1963), of 1.8 cm-1 it is possible to calculate that 90% of the incident photons will be absorbed in a layer of approximately 5.6 mm. As a result, any reaction vessel (e.g., from a Petri dish to an annular reactor) will present two distinct zones; the first one rich in 185 nm photons in which the photochemical homolysis of water and chemical oxidation occurs, and the other in which, only 254 nm photons are present and only direct photolysis will happen. Moreover, as a consequence of the emission of 185 and 254 nm radiation by the Hg lamp, the UV/VUV process is able to simultaneously induce disinfection and oxidation processes. 1.3. Limitations of the UV/VUV Advanced Oxidation Process As a consequence of the strong absorption of 185 nm radiation by water, it would be highly inefficient to use regular UV disinfection reactors for oxidation purposes. Therefore, UV/VUV photoreactors need to be designed in such a way that most of the liquid is irradiated by 185 nm radiation. As a result, UV/VUV photoreactors are much smaller in size and less water can be effectively treated by unit. Moreover, it has been showed by Dobrović et al., (2007) and Imoberdorf & Mohseni (2011 and 2012), that the efficiency of the process is limited by mass transfer phenomena, particularly the one associated with the micropollutant. This limitation can be overcome if the hydrodynamic conditions inside the photoreactor are optimized. In addition, natural water (i.e., surface or groundwater) contains a variety of naturally occurring solutes such as Natural Organic Matter (NOM), bicarbonate, nitrate, sulphate, chloride which are known, according to several studies (Kustchera et al., 2009; Mouamfon et al., 2011; Li et al., 2011; Imoberdorf & Mohseni, 2012), to impact the efficiency of the process by either scavenging hydroxyl radicals or acting as a filter (i.e., photon scavenging or screening). The reduction in the efficacy of the process will lead to larger energy consumption in order to be able to meet the desired extent of removal.   INTRODUCTION 5  1.4. Economic and Energetic Feasibility The figure of merit, Electrical Energy per Order (EEO) was introduced by Bolton & Cater (1994) to benchmark the relative performance of the different AOPs. It is defined as the electrical energy required to obtain 90 % reduction (i.e., 1-log reduction) of a target pollutant in a cubic meter of contaminated water. It was suggested that values below 2.5 kW h m-3 order-1 (Bolton & Cater, 1994) were indicative of an economically and energetically feasible operation. In a flow-through process, the EEO (kW h m-3 order-1) can be calculated by: 𝐸𝐸𝑂 =𝑃𝑒𝑙?̇? ∙ log (𝐶𝑜𝐶 ) (1.3) where, Pel refers to the power consumed by the UV lamp in kW; ?̇? to the operational flow rate in m3 h-1, C0 and C stand for the initial and final concentrations of the pollutant in mol L-1, respectively. 1.5. Effect of Temperature As observed by Barrett & Mansell (1960), Weeks et al. (1963), Halmann & Platzner (1966), and Kröckel & Schmidt (2014), the absorption coefficient of water at 185 nm shows a dependence on temperature. As the temperature of the aqueous solution changes, the thickness of the layer in which the photons are absorbed and where the OH radicals are generated will vary, thus affecting the mass transfer limitation of the system. The dependence of this optical property within temperatures commonly used in practical operation is, so far, unknown. As with pure water, naturally occurring solutes such as NOM, nitrate, sulphate and chloride could show temperature dependence for the absorption of 185 nm photons; therefore, affecting their role as inner filters (i.e., photon scavenging) in the system and the efficiency of the process. Although it is possible to find in the literature the molar absorption coefficient for the inorganic ions and NOM at room temperature, their dependence on temperature (particularly at lower temperature ranges) has not been reported yet.  INTRODUCTION 6  1.6. Project Summary This study aimed to quantify the effect of temperature on the efficiency and energy consumption of a UV/VUV photoreactor for drinking water treatment. In addition, the dependence on temperature of the absorption optical properties at 185 nm of pure water and commonly found solutes such as NOM, nitrate, sulphate and chloride was studied. Finally, the impact of treatment on a mixture of micropollutants was assessed under different water types (i.e. different water quality) and the Electrical Energy per Order was estimated for each micropollutant.   7  2. LITERATURE REVIEW 2.1. Removal of Common Micropollutants by UV/VUV Although there are several reports in the literature regarding the application of 172 nm to induce degradation of organics through OH radical attack (e.g., Jakob et al., 1993; Gonzalez & Braun, 1994; Oppenländer & Gliese, 2000; Oppenländer et al, 2005 and 2008) this section will focus on the results observed when seizing the emission of 185 nm photons by mercury lamps. As there are two energy sources in the system, the mechanism of removal will combine, in some cases, the action of direct photolysis by 254 nm photons and OH radical attack produced by water photolysis. The use of UV/VUV radiation (254 + 185 nm) to degrade micropollutants has been subject of study by numerous researchers. For instance, micropollutants such as p-chlorobenzoic acid (i.e., p-CBA) (Han et al., 2004; Azrague & Osterhus, 2009), 2,4-dichlorophenoxyacetic acid (i.e., 2,4-D) (Imoberdorf & Mohseni, 2012), taste and odour compounds such as geosmin and 2-methylisoborneol (i.e. 2-MIB) (Kutschera et al., 2009; Zoschke et al., 2012), perfluorodecanoic acid (Wang et al., 2010), sulfamethoxazole (Mouamfon et al., 2011), clofibric acid (Li et al., 2011), ibuprofen, ketoprofen (Szabó et al., 2011) and naproxen (Arany et al., 2013) have been used to assess the efficiency of the process under a variety of experimental conditions. In all these studies, the photoreactor used was operated in batch mode to obtain kinetic parameters and to study the extent of mineralization of each micropollutant (i.e., decrease in the TOC). In addition, these studies were carried at temperatures between 10 and 30 °C, and none explored the effect of temperature on process efficiency. Furthermore, all of these studies conducted the experiments in Milli-Q water and some generated synthetic waters by supplementing DOC, alkalinity and nitrate to Milli-Q water. A few studies, like the one by Kutschera et al., (2009), Zoschke et al., (2012), Azrague & Osterhus, (2009) and Imoberdorf & Mohseni, (2012) used natural water to investigate the effect of naturally occurring background components on the efficiency of the process. Moreover, despite the valuable information that all these studies have obtained, the lack of a standardized way to report the results, makes the information hard to transfer to other UV/VUV systems. Namely, reporting the results in terms of fluence (i.e., UV dose) as was LITERATURE REVIEW 8  done by Kutschera et al., (2009) will enable a more readily comparison of the effectiveness of the process towards the removal of a certain compound.  As observed in these studies, degradation will happen through two different mechanisms. The contribution of each pathway will depend on the characteristics of each compound. Due to the presence of 254 nm radiation, direct photolysis will occur when enough energy is absorbed by the compound to induce the breakdown of the chemical bonds. The sensitivity of a compound to direct photolysis by 254 nm photons will be dominated by the molar absorption coefficient and the photolysis quantum yield at 254 nm (Kim & Tanaka, 2009). On the other hand, the generation of OH radicals from water photolysis with 185 nm photons enables the advanced oxidation pathway. The contribution of these two mechanisms was observed clearly in the extensive study by Kim & Tanaka, (2009). They studied the degradation of a mixture of 30 pharmaceuticals and personal care products (i.e., PPCPs). For instance, they observed that diclofenac is easily degraded by direct photolysis and that the radical pathway did not enhance its removal. In contrast, propranolol showed very little change with the action of 254 nm (≈25% removal) but the presence of OH radicals increased the removal to ≈80%. The authors also found that some compounds such as ethenzamide (an analgesic) and clarithromycin (an antibiotic) were not effectively removed by either approach. On the other hand, the fate of the degradation byproducts of the micropollutants has not been extensively studied. Their sensitivity to further treatment and hazardous properties it is often unknown. Few studies, such as the one by Szabó et al., (2011) and Arany et al., (2013) found that the byproducts of ibuprofen, ketoprofen and naproxen were more resistant to UV/VUV treatment than the parent compounds.  2.2. Impact of Natural Organic Matter (NOM) and Other Naturally Occurring Ions on the UV/VUV Process Efficiency The removal efficiency of the UV/VUV process is hindered by solutes that absorb 185 nm photons (i.e., inner filter effect) and scavenge (i.e., react with) OH radicals. Attenuation of 185 nm photons depends on the molar absorption coefficient and the concentration of each compound (i.e., absorption coefficient) while the OH radical scavenging effect will depend on the reaction rate constant of each substance towards the radical species. So far, the attenuation of 185 nm radiation by naturally occurring solutes in natural waters has been barely LITERATURE REVIEW 9  investigated (see section 2.5). Weeks et al., (1963) and Fox et al., (1978) have reported values for the interaction of substances like chloride, carbonate and sulphate with 185 nm photons. On the other hand, the •OH scavenging reactions have been widely studied. For instance, the reaction rate constant between bicarbonate and OH radicals is in the range of 1.5 x106 to 8.5 x107 M-1 s-1 (Weeks & Rabani, 1966; Buxton et al., 1988) while, in contrast, several NOM standard isolates react with rates in the range of 1.21 to 10.36 x108 MC-1 s-1 as reported by Westerhoff et al., (2007) and McKay et al., (2011 & 2014). Moreover, the scavenging effect of the inorganic carbon (CO32- & HCO3-) will show some slight differences according to the solution pH value, given the equilibria between these two species and their respective reactivity towards OH radicals (Azrague & Osterhus, 2009). It has been reported that carbonate has a reaction rate with OH radicals that is from 28 to 45 times larger than the one of bicarbonate (Weeks & Rabani, 1966; Buxton et al., 1988; Glaze et al., 1995). Therefore, the OH scavenging effect due to the inorganic carbon will increase with the pH of the solution. The impact of NOM and inorganic carbon on the efficiency of the process has been studied by Kutschera et al., (2009), Azrague & Osterhus (2009), Li et al., (2011), Mouamfon et al., (2011) and Imoberdorf & Mohseni (2012). 2.3. Impact of Hydrodynamics on the UV/VUV Process Efficiency Owing to the high reactivity and localized concentration profile of hydroxyl radicals in just a few millimeters of the aqueous solution, mass transfer limitation significantly impacts the degradation process (i.e., diffusion controlled process). Several approaches have been used to improve the photoreactor hydrodynamics and overcome, to some extent, the mass transfer limitation. For instance, Dobrović et al., (2004) and Imoberdorf & Mohseni (2011) increased the velocity of the fluid inside the reactor (i.e., increase on Reynolds number) while Han et al., (2004) and Oppenländer et al., (2005) improved the mixing conditions by means of gas sparging (N2, O2 or air). Moreover, the photoreactor design can be modified in order to promote a more intimate contact between the micropollutant and the hydroxyl radicals. The study by Bagheri and Mohseni (2015) compared two different annular photoreactor configurations. They modified the annular section by adding baffles and observed a significant increase in the removal efficiency, with the baffled photoreactor. LITERATURE REVIEW 10  2.4. Comparison between UV/VUV and Other Advanced Oxidation Processes Several studies have compared the effectiveness of different AOPs while few have compared the energy requirements for each process tested. In comparison to photocatalysis (i.e., UV/TiO2), UV/VUV was shown to degrade the micropollutant (p-CBA) with faster reaction rates (Han et al., 2004). However, only one configuration for the photocatalytic process was tested (i.e., thin film placed in the reactor wall). Other configurations, such as a fluidized photoreactor may improve the efficiency of the process.  The UV/VUV process efficiency has often been compared (Li et al., (2011), Mouamfon et al., (2011), Imoberdorf & Mohseni (2011) and Zoschke et al., (2012)) to the one by the UV/H2O2 process. However, in the studies by Li et al., (2011) and Mouamfon et al., (2011) the concentration of hydrogen peroxide in the system was relatively high (50 to 100 mg L-1) in comparison to the one used in real commercial operations (maximum 20 mg L-1) (Sarathy & Mohseni, 2009). As a consequence of the larger amount of OH radicals present in the system, the pseudo-first order reaction rates observed will be dependent on the amount of H2O2 in solution. Despite this limitation, in all studies reaction rates observed for the UV/VUV process were comparable or even faster than those observed with UV/H2O2. Another approach to compare AOPs is based on the Electrical Energy per Order. This parameter offers a more direct comparison between processes. By this mean, Zoschke et al., (2012) compared UV/O3 and UV/H2O2 with UV/VUV for the removal of geosmin and 2-MIB. The authors assumed a value of 10 kWh per kg of H2O2 and 12 kWh per kg of O3 to account for the energy cost associated with these two chemicals and have a more fair comparison among processes. The EEO values calculated followed the order UV/O3 < UV/VUV < UV/H2O2, with the peculiarity that only the latter was over the 2.5 kW h m-3 limit recommended by Bolton & Cater (1994). 2.5. Absorbance of Pure Water and Inorganic Ions at 185 nm Water and commonly found organic and inorganic solutes in raw water are able to absorb 185 nm photons. The photon scavenging process hinders the formation of OH radicals from water photolysis thus, reducing the degradation efficiency of the process. Moreover, LITERATURE REVIEW 11  Barrett & Mansell (1960), Weeks et al., (1963), Halmann & Platzner (1965) and Stevenson (1968) demonstrated that the attenuation of 185 nm photons (i.e., absorption coefficient) by pure water increases with temperature. Overall, the range of temperatures covered by these studies was from 23.5 to 91.8 °C. The marked temperature dependence of the absorption coefficient observed in these studies can be attributed to the formation of hydrogen bonds. It has been shown by Barrett & Mansell (1960) and Ito (1960) that the absorption properties of an absorbing system are more sensitive to temperature when hydrogen bonds are present.  On the other hand, the study of the optical properties of solutes commonly found in raw water has been scarcely investigated. Of the few studies available, Weeks et al. (1963) reported values for the molar absorption coefficient at 185 nm and 24 °C of sulphate, chloride and carbonate of 190, 3300 and 1000 M-1 cm-1, respectively. Moreover, Fox et al., (1978) showed that the absorbance of chloride increases with temperature and that in comparison to other halides (Br- & I-), it shows the strongest temperature dependence. 2.6. Quantum Yield Dependence on Temperature The quantum yield (Φ) or quantum efficiency of a photochemical reaction is defined as the ratio between the moles produced or reacted of a particular chemical species over the total amount of photons absorbed by the media per unit time (Braslavsky, 2007). In other words, it expresses the likelihood for a particular photochemical process to happen. For the case of UV/VUV process, the quantum yield of interest is the one associated with the production of OH radicals from water photolysis. Moreover, if the absorption of 185 nm photons is hindered by the presence of organic and inorganic solutes, the production of OH radicals will be negatively impacted. In addition, one could expect a dependence on temperature of the quantum yield if the solutes absorbing properties are, as with pure water, temperature dependent. Although there is no literature available for this particular process, several reports (Zellner et al., 1990; Moortgat et al., 1983; Talukdar et al., 1988; Roehl et al., 1994; Lee & Yoon, 2004) have shown that the quantum yield for different photochemical processes is dependent on temperature. Specifically, in all these studies the quantum efficiency increased with temperature.   12  3. KNOWLEDGE GAPS AND THESIS OBJECTIVES 3.1. Knowledge Gaps 1) To best of my knowledge, there is no study that has focused on the potential effect that temperature may exert on the efficiency and power requirement of the UV/VUV AOP. 2) The absorption coefficient of pure water at 185 nm has not been reported below 23.5 °C. 3) The temperature dependence of the molar absorption coefficient of naturally occurring solutes such as NOM, chloride, sulphate and nitrate has not been reported in the literature. 3.2. Hypothesis The evidence found in the literature indicates that the absorption coefficient of pure water will increase with temperature. This means that, in a mass transfer limited system, the expansion of the layer in which OH radicals can be formed will result in greater reaction zone and thereby a more efficient process. The presence of solutes, such as the ones found in natural water, will on the other hand scavenge OH radicals and 185 nm photons thus, limiting the efficiency of the process. However, it is so far unknown how much the 185 nm photon scavenging process will be affected by temperature.  3.3. Thesis Objectives 1) Study the impact of temperature on the removal efficiency and the Electrical Energy per Order of the operation.  2) Study the 185 nm absorbance of pure water and major occurring solutes in raw water and their dependence on temperature. 3) Investigate the UV/VUV process efficacy towards the removal of a mixture of micropollutants in a variety of water matrices. KNOWLEDGE GAPS AND THESIS OBJECTIVES  13  3.4. Research Significance The UV/VUV process has shown great potential to economically remove chemical contaminants from water. Because of its chemical free nature and disinfection capabilities, it is an attractive drinking water treatment technology for rural and developing communities or countries around the world in which access to chemicals or economical resources for water treatment is limited. Considering that water temperature is one of the operational parameters in which water treatment plants have no control on, it is of great practical significance to study the impact of water temperature on the efficiency and energy consumption of the process. With a unit that resembles a full scale photoreactor, under common temperatures found throughout the year in the northern hemisphere and with the presence of a variety of micropollutants, insights regarding the process performance and energetic viability will be available to engineers and small systems that are in need for a engineered solution for their drinking water treatment plant.   14  4. MATERIALS AND METHODS 4.1. Chemicals The chemicals used in this research and their characteristics are listed in Table 4.1.  The chemicals were used as received from the suppliers. Table 4.1 List and characteristics of the chemicals used Chemical Grade Supplier Acetonitrile HPLC Fisher Scientific Atrazine Analytical Standard Sigma Aldrich Caffeine Analytical Standard Sigma Aldrich Carbamazepine Analytical Standard Sigma Aldrich Deethylatrazine Analytical Standard Sigma Aldrich Diclofenac Analytical Standard Sigma Aldrich Estradiol Analytical Standard Sigma Aldrich Fluoxetine Analytical Standard Sigma Aldrich Linuron Analytical Standard Sigma Aldrich Medroxyprogesterone Analytical Standard Sigma Aldrich Nitrogen 4.8 (99.998%) Praxair Norethindrone Analytical Standard Sigma Aldrich Phosphoric Acid HPLC Fisher Scientific Potassium Iodate ACS Fisher Scientific Potassium Iodide ACS Fisher Scientific Potassium Phthalate ACS Fisher Scientific Progesterone Analytical Standard Sigma Aldrich Sodium borate, decahydrate Reagent Fisher Scientific Sodium Carbonate ACS Fisher Scientific Sodium Chloride ACS BDH Sodium Nitrate ACS Fisher Scientific Sodium Sulphate ACS EMD Sulfamethoxazole Analytical Standard Sigma Aldrich Suwannee River NOM - IHHS 4.2. Raw Water Characteristics Raw water was collected from the Seymour Capilano Filtration Plant (Source: Seymour Reservoir) located in Vancouver, British Columbia, between September 2014 and May 2015. Water was collected in 20 L drums and was stored at 5 °C inside a temperature controlled room. A summary of the characteristics of the raw water used is shown in Table 4.2. The MATERIALS AND METHODS 15  values shown in Table 4.2 are the average value obtained from the analysis of raw water samples collected over the period of the study.  Table 4.2 Seymour Reservoir Raw Water Characteristics  Parameter Value Dissolved Organic Carbon - mg L-1 2.6543 ± 0.0594  Nitrate - mg L-1 0.6342 ± 0.0102 Sulphate - mg L-1 1.1370 ± 0.0426 Chloride - mg L-1 0.7368 ± 0.0060 Alkalinity as CaCO3 - mg L-1 2.08 ± 0.14 pH 5.90 ± 0.09 Absorbance at 254 nm 0.1152 ± 0.0011 Turbidity - NTU 0.89 ± 0.09 SUVA - m-1 mg L-1 4.3401 ± 0.1056 4.3. Experimental Apparatus 4.3.1. UV/VUV Collimated Beam An UV/VUV collimated beam apparatus (see Figure 4.1) as described by Duca et al., (2014) was used to study the decomposition kinetics of a mixture of micropollutants and to estimate the fluence (i.e., UV dose) within the flow-through photoreactor. Equipped with a low pressure amalgam Hg lamp (Light Sources, Inc.) the collimated beam set up was constantly purged with nitrogen (1.5 L min-1) to prevent the formation of ozone. To ensure a steady output from the UV/VUV lamp, a time period of at least an hour was waited prior to the start of experiments. The Petri dish (reacting vessel) used in the experiments was made of glass sides and quartz bottom which allowed the transmittance of 254 and 185 nm photons. The Petri dish had an average diameter of 4.59 ± 0.18 cm which resulted in an average exposure area of 16.57 ± 0.14 cm2. The volume of sample irradiated was 25 mL. A mixer was used to thoroughly mix the contents of the vessel and to overcome mass transfer limitations. MATERIALS AND METHODS 16   Figure 4.1 UV/VUV collimated beam apparatus with the Petri dish reacting vessel 4.3.2. UV Collimated Beam  An UV collimated beam apparatus (see Figure 4.2), equipped with a low pressure high output amalgam Hg lamp (Light Sources, Inc.) was used to study the effect of direct photolysis on the degradation of a mixture of micropollutants. The same Petri dish and batch volumes described in the previous section were used with the UV collimated beam. A magnetic stirrer bar was used to mix the solution and an optical paper was placed between the bottom of the Petri dish and the plastic surface of the magnetic stirrer to avoid scratching the quartz bottom surface.  MATERIALS AND METHODS 17   Figure 4.2 UV collimated beam apparatus 4.3.3. UV/VUV Photoreactor An UV/VUV annular photoreactor was designed and fabricated (see Figure 4.4) to study the removal efficiency of a variety of micropollutants. The photoreactor (see Figure 4.3) was equipped with a low pressure amalgam mercury lamp (GPHVA357T5L, Light Sources, Inc.) surrounded by a natural quartz dome ended sleeve (Light Sources, Inc.). The air gap between the Hg lamp and the sleeve is of 5 mm. The open end of the sleeve was capped, preventing the circulation of air thus minimizing the formation of ozone. Nevertheless, an ambient ozone sensor (EZ-1X, Eco Sensors) was used to monitor the concentration of this substance in the photoreactor surroundings. The photoreactor body was made of Plexiglas with a thickness of 6.35 mm which resulted in a water layer of 5 mm that flowed around the sleeve. The flow-rate was manipulated via an analog pump controller and monitored through a flow-meter. MATERIALS AND METHODS 18   Figure 4.3 Diagram and cross section of the UV/VUV photoreactor  Figure 4.4 Flow-through UV/VUV photoreactor 4.3.4. Vacuum-UV Absorption Setup An experimental unit capable of detecting 185 nm radiation was designed and built (see Figure 4.5) to study the temperature dependence of the molar absorption coefficient of pure water and dilute solutions of commonly found solutes (i.e., NOM, chloride, sulphate and nitrate). The unit consisted of a UV/VUV low pressure amalgam Hg lamp (Light Sources, MATERIALS AND METHODS 19  Inc.), a temperature controlled cuvette holder (T2, Quantum Northwest) coupled with a temperature controller (TC1, Quantum Northwest), a radiometer (IL1700 Research Radiometer, International Light Technologies) coupled with a 185 nm gold cathode sensor (SED185, International Light Technologies). A digital thermometer (HH801B, Omega) coupled with a type T thermocouple (0.81 mm diameter, Omega) was used to measure the temperature inside the quartz cuvettes used (2, 5 and 10 mm, Starna). Nitrogen at a flow-rate of 1.7 L min-1 was used to purge the lamp casing (i.e., to prevent ozone formation) and to avoid condensation on the quartz cuvette at lower temperatures. A thermostated circulator unit (Isotemp 250LCU, Fisher Scientific) was used to maintain the recirculating water used by the temperature controller at 5 °C.  Figure 4.5 Experimental setup to determine the absorption coefficient and molar absorption coefficient at 185 nm at different temperatures 1. UV/VUV source 2. Shutter 3. Temperature controlled cuvette holder 4. 185 nm sensor 5. Nitrogen purging   20  4.4. Experimental Methodology 4.4.1. Temperature Effect To study the effect of temperature, the flow-through photoreactor was placed and operated inside a temperature controlled room. The room was set at two temperatures of 5 and 20 °C. Distilled and raw water collected from the Seymour Reservoir was stored inside the temperature controlled room to achieve thermal equilibrium (approx. 72 hours) prior to the start of the experiment. Batches of 20 L were spiked with atrazine to achieve an initial concentration close to 250 μg L-1. The photoreactor was turned on and tap water recirculated through it for at least 1 hour to ensure that the Hg lamp output was constant. Subsequently, the tank containing the spiked aqueous solution was connected to the system and a stabilization time of 3.5 minutes was used with the purpose of flushing the system and to have a steady flow-rate through the photoreactor. Samples from the inlet tank and the outlet of the reactor were taken in triplicate for analysis (see Figure 4.6). Once the entire batch was treated, the reactor was flushed by passing tap water for at least 5 min at a flow-rate of 5.7 L min-1. The experiments carried with distilled water were conducted in duplicate while the ones with raw water were performed four times.   Figure 4.6 Analytical analyses performed on the samples obtained during the experiments Matrix usedDistilled WaterHPLC-UVRaw Water0.45 μm filtrationHPLC-UVHPSEC TOC UV Abs ICMATERIALS AND METHODS 21   1. Inlet Water Tank 2. Sampling Point #1 3. Pump 4. Flow Meter 5. Photoreactor 6. Sampling Point #2 7. Ballast  8. Power Meter 9. To waste Figure 4.7 Experimental setup to study the effect of temperature on the removal of model micropollutant (i.e., atrazine) 4.4.2. Energy Consumption The power requirement of the Hg lamp and its ballast, in addition to the one of the pump was measured by a P4400 Kill A Watt Electricity Usage Monitor (P3 International) for all experiments that were carried out in the flow-through photoreactor. The performance of the P4400 instrument was checked with an Amprobe® multimeter (ACD-31P). Additionally, in order to calculate the Electrical Energy per Order (EEO) of the process, four different flow-rates (i.e., 0.5, 1.0, 1.9 and 2.7 L min-1), one temperature (SP = 5 °C, water temperature ≈ 3.6 °C ) and raw water batches of 8 L spiked with atrazine to achieve an initial concentration close to 200 μg L-1 were used. For these sets of experiments, each batch of water was passed through the photoreactor for a number of times and samples were taken from the inlet and outlet tanks. Each experimental condition was conducted in duplicate. According to previous data, a minimum of 5 passes were needed to achieve a removal greater than 90% at the highest flow-rate used (2.7 L min-1). After each pass and collecting the samples, the contents of the outlet tank were transferred to the inlet tank and the process was repeated. In order to avoid cross contamination, the tanks were rinsed with tap water prior to the transfer and one liter of solution was purged from the photoreactor before collecting treated water in the outlet tank. MATERIALS AND METHODS 22    1. Inlet Water Tank 2. Sampling Point #1 3. Pump 4. Flow Meter 5. Photoreactor 6. Sampling Point #2 7. Ballast  8. Power Meter 9. Outlet Water Tank Figure 4.8 Experimental setup to determine the EEO for the removal of atrazine  4.4.3. Removal Equivalent Dose To estimate the dose delivered in the flow-through photoreactor, degradation experiments were carried out in the UV/VUV collimated beam (see section 4.3.1). Atrazine was used as a model micropollutant and aqueous solutions (C0 ≈ 180 μg L-1) were irradiated for a certain amount of time. Exposure time can be translated into fluence or dose if the irradiance of the source is known. The 254 nm irradiance of the UV/VUV lamp was measured by means of iodide/iodate actinometry (Rahn et al., 2006). To carry out the actinometric procedure, a 25 mL sample containing I-, IO3- and borax (Na2B4O7·10H2O) at a respective concentration of 0.60 M, 0.10 M and 0.010 M were irradiated for a period of 30 s. Subsequently, the absorbance at 352 nm of the solution was measured and the irradiance (mW cm-2) was calculated from equation 4.1: 𝐸′ = 23.407 [𝐴352 (𝑠𝑎𝑚𝑝𝑙𝑒) −  𝐴352 (𝑏𝑙𝑎𝑛𝑘)]𝐴𝑟𝑒𝑎 (𝑐𝑚2) × 𝐸𝑥𝑝𝑜𝑠𝑢𝑟𝑒 𝑡𝑖𝑚𝑒 (𝑠) (4.1) As described by Rahn et al., (2006) the constant 23.407 is the result of dividing the energy of a mole of photons at 254 nm (4.716 x 105 J Einstein-1) over the product of the quantum yield of triiodide (0.73) at 254 nm and the molar absorption coefficient of triode at MATERIALS AND METHODS 23  352 nm (27600 M-1 cm-1). An irradiance of 0.752 ± 0.187 mW cm-2 was calculated for the UV/VUV source of the collimated beam. The actinometric experiments were carried in triplicate. Batches of 25 mL of atrazine solution (C0≈180 μg L-1) were irradiated in duplicate for times ranging between 2.5 and 17.5 min which translate in fluences between 113 and 790 mJ cm-2. Removals of up to 96 % were observed which enabled to estimate the 254 nm based dose delivered in the flow-through photoreactor system (see Figure 4.9).  Between experimental runs the Petri dish was rinsed with deionized water. Subsequently, the Petri dish was rinsed at least three times with fresh solution to avoid any cross contamination.  It is important to note that the results obtained can only be applied to experiments carried out with the same water matrix (i.e., Seymour Capilano Filtration Plant raw water, DOC ≈ 2.5 mg L-1 and 2 mg L-1 alkalinity as CaCO3).  Figure 4.10 shows the data set used to obtain a best fit equation to estimate the 254 nm based dose delivered in the photoreactor.   Figure 4.9 Information required to calculate the 254 nm based removal equivalent dose Collimated beam degradation experiments254 nm actinometryFlow-through degradation experimentsRemoval Equivalent DoseMATERIALS AND METHODS 24   Figure 4.10 Dependence of Atrazine removal with the 254 nm based dose delivered to the aqueous media. Equation shows the curve that best fit the experimental data 4.4.4. Micropollutant Degradation The removal efficacy of the UV/VUV treatment of a mixture of micropollutants was assessed under a variety of water matrices by conducting experiments in the flow-through photoreactor (see section 4.3.3). A total of nine pharmaceuticals, two herbicides (plus one major metabolite) and one stimulant were part of the cocktail of micropollutants that were subject to treatment. The target initial concentration for each compound was 500 ng L-1. Two different flow-rates (i.e., 1.9 and 2.7 L min-1) were studied and the temperature controlled room was at a temperature of 22 °C which resulted in a water temperature of 20.5 °C. Four different water matrices were used to observe the impact of different constituents on the removal efficiency. Raw water collected from the Seymour Capilano Filtration Plant (DOC≈2.5 mg L-1 and 2 mg L-1 alkalinity as CaCO3) was used as a control. The other three matrices studied separately contained an additional amount of DOC (2 mg L-1 Suwannee River NOM), carbonate (50 mg L-1) and chloride (25 mg L-1). Each combination of experimental conditions was tested in duplicate. Degradation kinetics were studied by irradiating small batches (25 mL) of the four different water matrices previously described in the UV/VUV collimated beam setup (see section 4.3.1). Irradiation times between 2.5 and 15 min (fluences between 97 and 587 mJ cm-2) y = -3E-10x4 + 8E-07x3 - 0.0008x2 + 0.4064xR² = 0.99650204060801000 100 200 300 400 500 600 700 800Atrazine Removal (%)UV based dose (mJ cm-2)MATERIALS AND METHODS 25  were used to study the decay of each component of the micropollutant mixture. These experiments were not replicated. Instead, a random control was performed to corroborate the results obtained. To differentiate the effects of direct photolysis by 254 nm and OH radical degradation pathway, control samples of each water matrix tested were irradiated in a UV collimated beam (see section 4.3.2) for a time of 20 min (approx. fluence of 587 mJ cm-2). In the same manner, no replicates were conducted but a random control was performed to verify the results observed. Between experimental runs the Petri dish was rinsed with deionized water. Subsequently, the dish was rinsed at least three times with fresh solution to avoid any cross contamination. Analyses of the samples from this experimental stage were performed by means of HPLC-APCI-MS/MS (see section 4.5.2) in collaboration with École Polytechnique de Montreal. 4.4.5. Optical Properties at 185 nm and Their Dependence on Temperature Pure water obtained from an ELGA Option-Q water purification system (PURELAB) and aqueous solutions of nitrate, sulphate, chloride and Suwannee River NOM (SRNOM) at different concentrations (see Table 4.3) were used to study the dependence on temperature of the molar absorption coefficient of pure water and each solute using the experimental setup described in section 4.3.4. As with the collimated beam apparatus and flow-through photoreactor, the system was turned on and a period of at least an hour was waited before collecting any data to guarantee a steady output from the radiation source. During the warm up period and between measurements, a shutter was used to avoid sensor saturation and to prevent the degradation of rubber parts of the detector.    MATERIALS AND METHODS 26  Table 4.3 Solute concentration and pathlengths used in the absorption experiments Substance Concentration Pathlengths used (cm) Pure Water Depending on temperature from 55.5 M (3.5 °C) to 55.3 M (25 °C) 0.2, 0.5 & 1.0 Nitrate (0.0639 ± 0.0007) mM (0.1201 ± 0.0003) mM (0.2493 ± 0.0008) mM 0.2 & 0.5 Sulphate (1.1993 ± 0.0240) mM (2.1390 ± 0.0208) mM (3.1228 ± 0.0117) mM Chloride (0.0633 ± 0.0003) mM (0.1159 ± 0.0002) mM (0.2250 ± 0.0016) mM Suwannee River NOM (0.1728 ± 0.0045) mMC (0.3555 ± 0.0013) mMC (0.4851 ± 0.0047) mMC Prior to its use, each quartz cuvette was rinsed thoroughly with pure water to remove any particulate attached to the interior walls. Subsequently, the cuvette was rinsed five times with the solution being tested and was slowly filled using a Pasteur pipette to avoid the formation of air bubbles inside the cuvette. Optical paper was used to dry and remove any particulate from the surface of the cuvette. Once the cuvette was placed inside the temperature controlled holder, a type T thermocouple (0.81 mm diameter, Omega) was inserted in the cuvette to monitor the temperature of the solution. At the end of each set of measurements, the cuvettes were rinsed thoroughly with pure water, dried up and stored in their respective cases. 4.5. Analytical Methods 4.5.1. HPLC-UV A Dionex Ultimate 3000 HPLC (high performance liquid chromatography) system equipped with a 4 μm (particle diameter) C18 reversed-phase analytical column (3.9 x 150 mm, Nova-Pak®, Waters) was used to quantify the concentration of atrazine. Acidified water (pH MATERIALS AND METHODS 27  2.3, H3PO4) and acetonitrile were used as the mobile phase (60:40 %, v/v). The mobile phase was pumped at a flow-rate of 1 mL min-1. The thermostated column compartment was kept at 35 °C and quantification was done by UV detection at λ = 222 nm. Each vial containing atrazine was analyzed in triplicate. 4.5.2. HPLC-APCI-MS/MS A TSQ Quantum Ultra AM triple quadrupole mass spectrometer (Thermo Fisher Scientific) with an atmospheric pressure chemical ionization (APCI) source coupled with an HPLC system was used to quantify the micropollutants used in Chapter 6. The analysis was performed at École Polytechnique de Montreal. Details of the instrumental analysis can be found elsewhere, (Morissette et al., 2015).  4.5.3. HPSEC A Waters HPSEC (high performance size exclusion chromatography) system equipped with a 125 Å (pore size), 10 μm (particle diameter) analytical column (7.8 x 300 mm, Protein-Pak®, Waters) was used to quantify the changes in the apparent molecular weight of the natural organic matter present in raw water samples. Phosphate buffer (0.02 M, pH 6.8) with an ionic strength of 0.1 M (NaCl) was used as the mobile phase. The mobile phase was pumped at a flow-rate of 0.7 mL min-1. Detection was done by a Waters 2487 dual absorbance UV detector set at 260 nm. Each sample containing NOM was analyzed in triplicate. 4.5.4. Total Organic Carbon (TOC) A Shimadzu TOC-VCPH analyzer was used to determine the total organic carbon content of the aqueous samples. The combustion of the sample was carried at 680 °C. The result reported by the instrument software is the average of three determinations (SDmax = 0.1 & CVmax = 2%). Each sample was analyzed in triplicate. 4.5.5. Ion Chromatography (IC) A Dionex ICS-1100 ion chromatography system equipped with an IonPac® AS22-Fast analytical column coupled with an ASRS-300 (4mm) suppressor was used to determine the concentration of inorganic anions. An aqueous solution containing carbonate and bicarbonate MATERIALS AND METHODS 28  (4.5 & 1.4 mM, respectively) was used as the mobile phase. The mobile phase was pumped at a flow-rate of 1.2 mL min-1. The quantification was done by a DS6 heated conductivity cell. Each sample was analyzed in triplicate. 4.5.6. UV Absorbance An Agilent Cary 100 spectrophotometer operated in double beam mode was used to determine the absorbance at 254 nm of the samples. Quartz cuvettes (Hellma) with a pathlength of 10 mm were used in the determinations. Each sample was analyzed in triplicate. Between measurements the cuvettes were rinsed at least three times with a small portion of the solution to be tested to avoid cross contamination. 4.5.7. Other Analytical Methods Alkalinity The Standard Method 2320 B was followed to measure the alkalinity of the aqueous samples. Sulphuric acid (0.02 N) was used as the titrator of the 100 mL sample. Each sample was analyzed in triplicate. The pH was monitored with a pH handheld meter (pH 5+, Oakton) equipped with a pH and a temperature probe. Turbidity A Hach 2100Q portable turbidimeter was used to measure the turbidity units (NTU) of the water samples. Each sample was analyzed in triplicate. Between measurements the cuvettes were rinsed at least three times with a small portion of the solution to be tested to avoid cross contamination.  29  5. EFFECT OF TEMPERATURE  This chapter describes the effect of temperature on the removal efficiency and energy consumption observed in a UV/VUV photoreactor. Moreover, the absorption properties of pure water and commonly found solutes such as nitrate, chloride, sulphate and Suwannee River NOM (i.e., a surrogate for dissolved organic carbon) at 185 nm and their dependence on temperature were investigated. First, the interaction between water and the aforementioned components with 185 nm radiation is introduced to later explore its effect in the removal efficiency of a micropollutant with waters of different quality. Finally, the energy demand and feasibility of the process are assessed by determining the Electrical Energy per Order (EEO). 5.1. Absorption of Radiation at 185 nm The absorption coefficient of a substance can be extracted from the Beer-Lambert law and it represents the substance strength to attenuate the intensity of radiation as it passes through it. As depicted in Figure 5.1, if the absorption coefficient increases, the amount of light that is able to travel through the solution decreases.     Figure 5.1 Comparison in light attenuation intensity by a substance with a) lower absorption coefficient b) higher absorption coefficient through an identical pathlength The Beer-Lambert law (Eq. 5.1) states that the attenuation of radiation is dependent on pathlength, b, the concentration, c, and the molar absorption coefficient, ε, of the material. Moreover, the absorption coefficient, k, is the product of the molar absorption coefficient and a) b) EFFECT OF TEMPERATURE 30  the concentration of the substance in the medium (Eq. 5.2). Consequently, equation 5.1 can be written as equation 5.3:  𝐴 = log (𝐼0𝐼) = 𝜀 ∙ 𝑐 ∙ 𝑏 (5.1) 𝑘 =  𝜀 ∙ 𝑐 (5.2) 𝐴 = log (𝐼0𝐼) = 𝑘 ∙ 𝑏 (5.3) If the absorbance, pathlength and concentration are known, the molar absorption coefficient or the absorption coefficient can be easily obtained from equations 5.2 and 5.3, respectively (refer to Appendix A for a detailed explanation on how to manipulate the experimental data). As was previously stated in section 4.4.5, pure water and aqueous solutions of organic and inorganic solutes at different concentrations were used to determine their optical properties and their dependence on temperature. For a start, the absorption of pure water was investigated and the values for the calculated absorption coefficient are reported in Table 5.1. Table 5.1 Average absorption coefficient at 185 nm of pure water at different temperatures and its impact on the thickness of the illuminated volume. Error bars represent the standard deviation among three replicates Temperature (°C) Absorption Coefficient (cm-1) 185 nm % Transmittance Path length at which 90% Absorption of 185 nm occurs 3.6 0.79 ± 0.13 16.22 1.27 cm 17.5 1.23 ± 0.11 5.89 0.81 cm 25 1.53 ± 0.09 2.95 0.65 cm These results clearly show the dependence of the absorption coefficient of pure water on temperature. The increase in its magnitude, respectively from 0.79 to 1.23 cm-1 (statistically significant, α = 0.05, 4 degrees of freedom) between 3.6 and 17.5 °C, result in that 90% of the photons reaching the solution are absorbed in a layer of 1.27 and 0.81 cm, respectively. In addition, it is possible to see that the 185 nm transmittance decreased from 16.2 to 5.9% as the temperature increases from 3.6 to 17.5 °C, respectively. Hence, without proper hydrodynamic conditions inside the reactor, the system will be more sensitive to mass transfer limitation as temperature increases. The absorption coefficient was also measured at 25 °C to validate and EFFECT OF TEMPERATURE 31  compare the results obtained with the reported values in the literature (Barrett & Baxendale (1960), Barrett & Mansell (1960), Weeks et al., (1963), Halmann & Platzner, (1966) and Kröckel & Schmidt (2014)). As shown in Table 5.2, the values obtained in this study fall within the range of data reported in the aforementioned studies. Table 5.2 Comparison of the absorption coefficients at 185 nm of pure water reported in literature and the one calculated in this study Temperature (°C) Absorption Coefficient (cm-1) Reference 25 1.20 Barrett & Baxendale, 1960 25 1.46 Barrett & Mansell, 1960 25 1.80 Weeks et al., 1963 24 1.47 Halmann & Platzner, 1966 25.5 1.62 Kröckel & Schmidt, 2014 25 1.53 This study As pointed out by Kröckel & Schmidt (2014), the variation among different studies are mainly due to the presence of trace contaminants in water, particularly DOC and inorganic ions and the manipulation of the experimental data, namely the correction required to compensate for reflection losses in the beam pathlength. In this study, the pure water used contained less than 10 ppb DOC and had a conductivity of 18.2 MΩ. In addition, the glassware used in the experiments was baked at 520 °C for an hour to remove any organic carbon. Finally, the experimental absorbance was corrected due to reflection losses according to the equation derived by Goldring et al., (1953), and a detailed explanation is presented in Appendix A.  Subsequently, the molar absorption coefficient of chloride, nitrate and sulphate were investigated. These solutes were chosen due to their presence in the raw water used in the degradation experiments. The calculated molar absorption coefficients at different temperatures for nitrate, sulphate and chloride are respectively shown in Table 5.3, Table 5.4 and Table 5.5.    EFFECT OF TEMPERATURE 32  Table 5.3 Average values for the molar absorption coefficient at 185 nm of nitrate at different temperatures. Error bars represent the standard deviation among three replicates Temperature (°C) Molar absorption coefficient (M-1 cm-1) 3.6 4632 ± 877 17.5 4737 ±968 25 4779 ± 1115 Table 5.4 Average values for the molar absorption coefficient at 185 nm of sulphate at different temperatures. Error bars represent the standard deviation among three replicates Temperature (°C) Molar absorption coefficient (M-1 cm-1) 3.6 109 ± 41 17.5 134 ± 39 25 146 ± 41 Table 5.5 Average values for the molar absorption coefficient at 185 nm of chloride at different temperatures. Error bars represent the standard deviation among three replicates Temperature (°C) Molar absorption coefficient (M-1 cm-1) 3.6 1877 ± 767 17.5 2636 ± 683 25 3063 ± 699 It can be seen from the results that among the inorganic solutes tested, nitrate had the highest molar absorption coefficient, followed by chloride and sulphate. The latter showed negligible absorption of 185 nm photons, whereas, nitrate and chloride showed a strong absorption at 185 nm. On the other hand, absorption by chloride showed the largest change (39% difference), among the inorganic solutes tested, as temperature varied from 3.6 to 25 °C. However, for each ion, the differences between the values obtained are not statistically significant (α = 0.05, 4 degrees of freedom). As a consequence, it must be stated that there is not a significant statistical impact of temperature on the absorption properties of the ions tested. On the other hand, the results obtained are in agreement to those reported by Weeks et al., (1963) at 25 °C for chloride and sulphate ions. A comparison of the results is shown in Table 5.6.    EFFECT OF TEMPERATURE 33  Table 5.6 Comparison of the molar absorption coefficient at 185 nm of chloride and sulphate reported in literature (Weeks et al., 1963) and the one calculated in this study Ion Literature Value (M-1 cm-1) This Study (M-1 cm-1) Difference (%) Chloride 3300 3063 7.18 Sulphate 190 146 23.2 Table 5.7 Average values for the molar absorption coefficient at 185 nm of Suwannee River NOM at different temperatures. Error bars represent the standard deviation among three replicates Temperature (°C) Molar absorption coefficient (MC-1 cm-1) 3.6 1251 ± 746 17.5 1341 ± 791 25 1402 ± 866 Last but not least, the absorption of a standardized carbon source, Suwannee River NOM (RO isolate, 2R101N), was studied and the results for its molar absorption coefficients at 185 nm at different temperatures are shown in Table 5.7.  The results obtained indicate that the absorption of 185 nm photons by this particular NOM isolate has a weak dependence on temperature. Nonetheless, as with the inorganic ions, the difference in the molar absorption coefficient at different temperatures is not statistically significant (α = 0.05, 4 degrees of freedom). In comparison, the absorption of 185 nm radiation by SRNOM is stronger than the one by sulphate, but is not as strong as the one by nitrate or chloride.  Altogether, the data obtained for water and each solute can be used to estimate, as long as the water quality is known (i.e., DOC, nitrate, sulphate & chloride), the absorption coefficient at 185 nm of a source water. As mentioned before, the absorption coefficient is the product of the molar absorption coefficient and the concentration of a substance. When dealing with a mixture, the absorption coefficient can be calculated as  𝑘 = ∑ 𝜀𝑖 ∙ 𝑐𝑖𝑛𝑖=1  (5.4) This is a great tool that can be used to aid in the assessment of a given source water to be treated by UV/VUV process without the need of an instrument to measure the molar absorption coefficient of the water. For instance, given the water quality characteristics of the raw water used in this research, the estimated absorption coefficient was within 10 to 15 % error of the measured value, as shown in Table 5.8. It is important to note that it was assumed EFFECT OF TEMPERATURE 34  that the dissolved organic carbon present in water interacted with 185 nm radiation in the same manner as Suwannee River NOM, which may not be necessarily the case.  Table 5.8 Comparison between the measured and the calculated absorption coefficient for the raw water used in experimentation at different temperatures. Temperature (°C) Measured Absorption Coefficient, k (cm-1) Calculated Absorption Coefficient, k (cm-1) Error (%) 3.6 1.00 ± 0.21 1.15 ± 0.23 15.0 17.5 1.46 ± 0.16 1.63 ± 0.23 11.6 25.0 1.78 ± 0.12 1.96 ± 0.24 10.1 Moreover, owing to the additive property of absorbance (Eq. 5.1.5), it is now possible to estimate the fraction (XA) of 185 nm photons that will be absorbed by each chemical species (Eq. 5.1.6) in solution. In doing this, it will be possible to understand and separate the role of each solute in the photon scavenging process (this will be discussed in details in section 5.3).  𝐴 = 𝑏 ∙ ∑ 𝜀𝑖 ∙ 𝑐𝑖𝑛𝑖=1 (5.5) 𝑋𝐴 =𝜀𝐴 ∙ 𝑐𝐴∑ 𝜀𝑖 ∙ 𝑐𝑖𝑛𝑖=1 (5.6)  EFFECT OF TEMPERATURE 35  5.2. Removal Efficiency with Distilled Water Degradation experiments with the flow-through VUV/UV photoreactor (as described in section 4.3.3) were conducted with 40 L batches of distilled water spiked with atrazine. The aqueous solution lacked background natural organic matter or alkalinity; therefore, all the 185 nm photons were absorbed by water and the OH radicals were scavenged almost exclusively by the target micropollutant. Additionally, the temperature of the air gap (i.e., the space between the lamp and the sleeve) was monitored by two thermocouples (T type) placed 180° apart (see Figure 5.2) to estimate if the lamp was performing under adverse conditions. The average values of the measurements made are shown in Table 5.9.     Figure 5.2 Thermocouple positioning inside the air gap of the photoreactor Table 5.9 Average air gap temperature under different operational conditions for an amalgam Hg lamp Water Temperature (°C) Air gap Temperature (°C) 3.7 41.4 17.8 56.6 Difference 15.2 As can be seen, even though there is a substantial change in air gap temperature between both water temperatures tested (see Table 5.9) it is possible to infer, from the results 1. Hg lamp 2. Lamp holder 3. Thermocouple 4. Hg lamp base EFFECT OF TEMPERATURE 36  showed in Figure 5.3, that the lamp is operating at a temperature that enables a constant rate of 185 nm photon emission that is not dependent on environmental temperature. If this was not the case, having in mind that low pressure Hg lamps tend to underperform at very low environmental temperatures, the removal efficiency observed at 3.7 °C would have been smaller than the one observed at 17.8 °C, just as shown in Figure 5.4, where the results of a replicate set of degradation experiments performed with a standard Hg lamp are portrayed. As can be seen from Figure 5.4, the removal efficiency changes significantly between the temperatures tested and it increases with temperature. Given that the experimental conditions are the same, the results obtained suggest that the output of the lamp is being affected by the surroundings temperature.  These observations are supported respectively by the findings of Bagheri & Mohseni (2014), who reported a stable output of 185 nm photons for temperatures in the amalgam position in the range between 60 and 80 °C and by Duran et al., (2010), who showed that standard Hg lamps have an increase in the radiant power as the water temperature in the photoreactor annulus increases from 5 to 15 °C, where a maximum radiant power was observed, to then decrease as the temperature of the media continues to increase from 15 to 35 °C.   Figure 5.3 Removal efficiency dependence on temperature at two different flow-rates for an amalgam Hg lamp. Error bars represent the calculated uncertainty for the removal of two replicates 91.385.786.578.97075808590952.7 3.5Removal (%)Flow-rate (L min-1)3.7 °C 17.8 °CEFFECT OF TEMPERATURE 37   Figure 5.4 Removal efficiency dependence on temperature at two different flow-rates for a standard Hg lamp. Error bars represent the calculated uncertainty for the removal of two replicates When comparing Figure 5.3 and Figure 5.4, higher efficiencies were observed with the amalgam Hg lamp and with lower flow-rate. The former can be explained by the relatively higher output (3-4 times) of the amalgam lamp. An increased emission of 185 nm photons results in a higher yield of OH radicals from water photolysis, while the latter is due to a longer retention time in the photoreactor (4.4 and 3.4 s for 2.7 and 3.5 L min-1, respectively), which leads to a higher dosage of 185 nm photons into the system. Interestingly, in Figure 5.3, a slight improvement in the removal is observed when the system is operated at low temperature. Given it has been shown that the lamp power output is constant, the observed difference is attributed to a change in the optical properties of the solution. As reported in Table 5.1, the absorption coefficient of water increases with temperature from 0.78 to 1.23 cm-1 from 3.6 to 17.5 °C, respectively. Moreover, the geometry of the annular photoreactor along with to the flow-rates and temperatures used in experimentation result in Reynolds numbers between 545 and 813, indicating that the system operates under laminar flow regime and its efficiency is limited by the mass transfer of the micropollutant to the AOP layer. Figure 5.5 shows the photon absorption profile in the photoreactor annulus. It can be seen that there is a slight but significant change in the strength by which 185 nm photons are absorbed by water. Consequently, the concentration profile of OH radicals resulting from water photolysis will change and the mass transfer limitation will 49.341.866.262.230405060702.7 3.5Removal (%)Flow-rate (L min-1)3.7 °C 17.8 °CEFFECT OF TEMPERATURE 38  be, to some extent, overcome. In other words, the same amount of 185 nm photons will be absorbed but a less localized production of OH radicals will be observed as the water temperature decreases.  Figure 5.5 185 nm photon absorption profile at different temperatures 0204060801001.40 1.51 1.63 1.74 1.86 1.97 2.08 2.20 2.31 2.43 2.54Photons Absorbed (%)Reactor Radius (cm)3.6 °C 17.5 °CEFFECT OF TEMPERATURE 39  5.3. Removal Efficiency with Raw Water The presence of additional substances, such as the ones present in natural water sources, which are able to scavenge 185 nm photons (i.e., act as inner filters) creates an aqueous solution with complex optical properties. Although water, due to its concentration, will absorb majority of the 185 nm photons, the high molar absorptivities of some inorganic solutes and organic matter (i.e., NOM) could play a role in hindering the efficiency of the process, despite their relatively low concentrations. As was mentioned previously, the temperature dependence of the absorption properties of these naturally occurring solutes has not been reported. Similar to the previous section, degradation experiments were performed by spiking raw water collected from Seymour Reservoir with atrazine to an initial concentration close to 250 µg L-1. The characteristics of the raw water used are shown in Table 4.2. The experiments were carried out at a flow-rate of 2.7 L min-1 (i.e., approx. residence time of 4.4 s) and at water temperatures of 3.6 and 17.5 °C. The degradation efficiencies observed under the experimental conditions are shown in Figure 5.6.  Figure 5.6 Removal efficiency of atrazine in raw water at two different temperatures. Error bars represent the calculated uncertainty for the removal of four replicates 37.743.030354045Removal (%)Water Temperature3.6 °C 17.5 °CEFFECT OF TEMPERATURE 40  The impact of water matrix is evident when comparing the efficiencies of the experiments carried out with distilled and raw water. In average, a percent reduction of approximately 55% was observed among the water matrices tested. The marked reduction was mainly due to the scavenging effect of hydroxyl radicals by NOM and the inorganic solutes dissolved in solution, which results in the reduced availability of OH radicals required to oxidize the target contaminant. Moreover, in contrast to the results obtained with distilled water, the removal efficiency when using natural water was slightly larger (by approx. 5.3%) when the system was operated at a higher temperature.  As mentioned before, the difference in the observed removal efficiency could be attributed to the change in the optical properties of the solution when background constituents (i.e., NOM, alkalinity and inorganic ions) are present. In order to evaluate this hypothesis, the absorption of 185 nm radiation by raw water was studied at different temperatures. Given the complexity of the water matrix, the global absorption coefficient (i.e., the ratio of the calculated absorbance and the pathlength) was calculated and the results are shown in Table 5.10. Table 5.10 Average absorption coefficient at 185 nm of raw water at different temperatures and its impact on the thickness of the illuminated volume. Error represents the standard deviation among three replicates Temperature (°C) Absorption Coefficient (cm-1) 185 nm % Transmittance 90% Absorption of 185 nm occurs at 3.6 1.00 ± 0.21 10.00 1.00 cm 17.5 1.46 ± 0.16 3.47 0.68 cm 25 1.78 ± 0.12 1.66 0.56 cm As was the case for pure water, the absorption coefficient of raw water increased with temperature. In particular, the absorption coefficient increased from 1.00 to 1.46 cm-1 when the media temperature increased from 3.6 to 17.5 °C. However, due to the presence of NOM, bicarbonate and inorganic ions (i.e., nitrate, sulphate and chloride), the attenuation of radiation at 185 nm was stronger in raw water. For instance, for pure and raw water at 3.6 °C, absorption of 90% of the photons reaching the aqueous media occurred in a layer of 1.27 and 1.00 cm, respectively. Consequently, the 185 nm photon absorption profile for raw water was steeper than the one for pure water at the same temperature, as shown in Figure 5.7.  EFFECT OF TEMPERATURE 41   Figure 5.7 Comparison of the 185 nm photon absorption profile in the photoreactor between water matrices at different temperatures Furthermore, the removal efficiencies observed and the absorption profile suggest that for raw water, mass transfer limitation does not have a dominant role in affecting the performance of the system. If this was the case, smaller degradation efficiency would have been observed when operating the system at warmer temperatures, similar to the case of pure water. This behavior would have been a consequence of a reduced photoreaction volume (increase in the attenuation of 185 nm photons as temperature rises) where OH radicals can be formed to react with the micropollutant. In other words, a more localized production of OH radicals would deplete the organic contaminant present in the AOP layer rapidly, leaving a larger, untreated fraction in an OH radical free zone. At this point, any additional reaction with OH radicals will happen after the micropollutant diffuses to the photoreaction zone. However, for raw water, the efficiency is a consequence of the competition for photons by the matrix components. Using equation 5.6, it is possible to estimate the fraction of photons absorbed by each component of the matrix for which the molar absorption coefficient and the concentration is known. As an illustration, for raw water, the fraction of 185 nm photons absorbed by water is estimated by equation 5.7 as: 𝑋𝐻2𝑂 =𝜀𝐻2𝑂 ∙ 𝑐𝐻2𝑂𝜀𝐻2𝑂 ∙ 𝑐𝐻2𝑂 + 𝜀𝐷𝑂𝐶 ∙ 𝑐𝐷𝑂𝐶 + 𝜀𝑁𝑂3− ∙ 𝑐𝑁𝑂3− + 𝜀𝑆𝑂42− ∙ 𝑐𝑆𝑂42− + 𝜀𝐶𝑙− ∙ 𝑐𝐶𝑙− (5.7) 0204060801001.40 1.51 1.63 1.74 1.86 1.97 2.08 2.20 2.31 2.43 2.54Photons Absorbed (%)Reactor Radius (cm)Raw water at 17.5 °CPure water at 17.5 °CRaw water at 3.6 °CPure water at 3.6 °CEFFECT OF TEMPERATURE 42  The estimated fraction of 185 nm photons absorbed by each component is shown in Table 5.11. Above all, the respective increase in the fraction of 185 nm photons absorbed by water from 0.683 to 0.753 (approx. 10% increase) between 3.6 and 17.5 °C can be used to explain the slightly larger removal observed as the temperature increases. Recalling that approximately 33% (i.e., quantum yield of 0.33) of the 185 nm photons absorbed by water will result in the production of OH radicals (i.e., water photolysis) it is possible to expect an increase in OH radical concentration if more VUV photons are being absorbed by water. Consequently, under the experimental conditions tested, the availability of OH radicals will increase with water temperature thus resulting in an enhanced treatment of the micropollutant present in the AOP layer. With raw water then, the mass transfer limitation imposed by a steeper absorption profile (see Figure 5.7) will be overcome by a larger yield of OH radicals. Nonetheless it is important to note that, if the hydrodynamic conditions are not ideal, a portion of the micropollutant will pass through the unit without receiving any treatment. Table 5.11 Fraction of 185 nm photons absorbed by each component of the aqueous matrix at different temperatures Temperature (°C) 3.6 17.5 Component Concentration (M) Fraction of 185 nm photons absorbed Water 55.5/55.4 0.683 0.753 DOC 2.210 x10-4 0.240 0.182 Nitrate 1.023 x10-5 0.041 0.030 Sulphate 1.184 x10-5 0.001 0.001 Chloride 2.078 x10-5 0.034 0.034 Other factors that may be suggested to show dependence on temperature, and hence explain the observed results with raw water, include the reaction rate and the OH radical quantum yield from water photolysis. There is some evidence in the literature (see section 2.6) suggesting that the quantum yield of hydroxyl radicals from water photolysis may show a dependence on temperature, especially for natural waters (i.e., waters containing 185 nm absorbing solutes). Reaction rate with OH radical, on the other hand, is unlikely to have a significant impact on the system, according to the observation made by Cheney (1996) who reported that OH radical reactions show little dependence on temperature given the small activation energy required. This was also observed by Oppenländer & Xu (2008) who found EFFECT OF TEMPERATURE 43  that in the 20-50 °C range, there was no significant temperature effect on the degradation rates of rhodamine B (a dye) and methanol while using an excimer lamp (emitting 172 nm photons).  EFFECT OF TEMPERATURE 44  5.4. Energy Requirements The electrical energy consumption was monitored throughout the experimentation as described in section 4.4.2. It was found to be twice as large for the lamp & ballast combined in comparison to the one required by the pump. On average, 67% of the input energy was consumed by the lamp and the ballast, while 33% was consumed by the pump. Based on the 6 minutes required to treat the entire batch of water for each experimental run, the combined averaged electrical requirement, in kW h, at 3.6 and 17.5 °C are shown in Figure 5.8.  The total energy consumed by the setup at 3.6 and 17.5 °C was 9.77 x10-3 and 9.07 x10-3 kW h, respectively. However, it can be seen that the energy required by the lamp and ballast did not vary as much as the pumping energy, which decreased due to the difference in the viscosity of the medium at the two temperatures tested.  Figure 5.8 Comparison of the electrical energy consumed by the system at different temperatures with a flow-rate of 2.7 L min-1. Error bars represent the standard deviation in the energy measurements of four replicates Given the lower removal efficiency obtained at lower temperatures, the Electrical Energy per Order (EEO) for the system was calculated at 3.6 °C for four different flow-rates. As a result of the removal observed with a single pass through the photoreactor at 2.7 L min-1 (see section 5.3), it was calculated that a minimum of 5 passes were needed to obtain a removal larger than 90% at this particular flow-rate. The amount of passes required at each flow-rate to achieve at least 1-log removal are shown in Table 5.12. 0.630 0.6330.347 0.2740.000.200.400.600.801.001.203.6 °C 17.5 °CElectrical Energy Consumption (kW h x 102)Temperature (°C)Lamp + Ballast PumpEFFECT OF TEMPERATURE 45  Table 5.12 Flow-rates and number of passes used in the determination of the Electrical Energy per Order of the system Flow-Meter (gal min-1) Measured Flow-Rate (L min-1) Passes for at least               90 % Removal Minimum 0.5 1 0.25 1.0 2 0.5 1.9 4 0.75 2.7 5 To calculate the EEO, equation 1.3 (EEO equation) was modified in order to account for the peculiar energy input required per pass through the photoreactor as shown in equation 5.8. The first term in the right-hand side of equation 5.8 represents the EEO due to the energy requirements of the UV/VUV lamp while the second terms represent the pumping contribution to the overall energetic demand of the system. 𝐸𝐸𝑂 =𝑃𝑙𝑎𝑚𝑝 ∙ 𝑅𝑒𝑎𝑙 # 𝑜𝑓 𝑝𝑎𝑠𝑠𝑒𝑠?̇? ∙ 60+𝑃𝑝𝑢𝑚𝑝?̇? ∙ 60 (5.8) Several steps are required to obtain the real number of passes required to achieve 90% removal. To begin with, the observed removal at each pass was correlated to a 254 nm based dose by means of equation 5.9 (see Figure 4.10). Subsequently, given that atrazine decays with a pseudo-first order kinetic behavior (see Figure 5.9), it is possible to link the 254 nm based dose given to the system with the observed atrazine removal. Owing to the fact that the interest is to know the dose at which 90% removal is observed, from the linear regression equation it is possible to calculate that a dose close to 590.7 mJ cm-2 is required to achieve 1-log removal of atrazine from raw water. % 𝑅 = −3𝑥10−10 ∙ 𝐷𝑜𝑠𝑒4 + 8𝑥10−7 ∙ 𝐷𝑜𝑠𝑒3 − 8𝑥10−4 ∙ 𝐷𝑜𝑠𝑒2 + 0.4064 ∙ 𝐷𝑜𝑠𝑒 (5.9) EFFECT OF TEMPERATURE 46   Figure 5.9 Kinetics of atrazine degradation in the flow-through photoreactor. Units of the pseudo first order constant apparent reaction rate are cm2 mJ-1 As a consequence of the different flow-rates studied, the dose delivered to the system will vary with each pass through the unit. As the flow-rate increases, the residence time (i.e., contact time) inside the photoreactor decreases, resulting in a decreased dose of radiation. Figure 5.10 shows the average dose delivered per pass at each flow-rate studied. Given that the kinetic behavior of atrazine decay will not change, it is possible to extract from each one of these graphs, the actual number of passes required to deliver a dose of approximately 590.7 mJ cm-2. The results of these analyses are shown in Table 5.13. Once this is known, the EEO for the removal of atrazine can be calculated for each flow-rate studied (see Figure 5.11).   y = 0.0039xR² = 0.99500.511.522.530 136 272 408 544 680-ln(C/C0)254 nm based dose (mJ cm-2)EFFECT OF TEMPERATURE 47  Number of Passes    254 nm based dose (mJ cm-2) Number of Passes    254 nm based dose (mJ cm-2) Figure 5.10 254 nm based dose delivered per pass in the photoreactor at different flow-rates a) 0.5 L min-1; b) 1.0 L min-1; c) 1.7 L min-1 & d) 2.9 L min-1 Table 5.13 Real number of passes through the photoreactor to observe a 1-log removal of atrazine at 3.6 °C 1-log removal 254 nm dose (mJ cm-2) Flow-Rate      (L min-1) Residence Time (s) Real number of passes 590.7 0.5 23.8 0.95 1.0 11.9 1.7 1.9 6.3 3.2 2.7 4.4 4.4  y = 0.0016xR² = 0.99930.00.30.50.81.00 170 340 510 680ay = 0.0029xR² = 0.97430.00.51.01.52.00 170 340 510 680by = 0.0055xR² = 0.98150.01.02.03.04.00 170 340 510 680cy = 0.0074xR² = 0.98930.01.32.53.85.00 170 340 510 680dEFFECT OF TEMPERATURE 48   Figure 5.11 Electrical Energy per Order for the removal of atrazine at 3.6 °C and different flow-rates. Error bars represent the calculated uncertainty for the EEO for two replicates Figure 5.11 shows the total and individual contribution of each component to the Electrical Energy per Order. An average value of 63.2 W was used as the electrical energy required by the Hg lamp and ballast. As can be seen, the EEO decreases from 2.00 to 1.72 kW h m-3 order-1 as the flow-rate increases from 0.5 to 2.7 L min-1. The decrease in the EEO indicates that the dominant factor in the calculation is the flow-rate of water treated. This is, the additional energy expense (i.e., additional lamps and ballasts) required to treat a larger flow-rate of water will play a minor role in determining the economic feasibility of the process (see Figure 5.12). In other words, providing treatment to a larger volume of water will dominate the outcome of the figure-of-merit. In case of the EEO at higher temperatures, it is expected to be lower than the values shown in Figure 5.11, given the slight degradation improvement which will impact the required dose and the number of passes required to achieve a removal of 90%. 2.00 1.79 1.77 1.722.271.96 1.92 1.870.00.51.01.52.02.53.00.5 1.0 1.9 2.7EEO (kW h m-3order-1 )Flow-rate (L min-1)Lamp + Ballast PumpEFFECT OF TEMPERATURE 49   Figure 5.12 Energy consumed by the lamp and ballast at each flow-rate to achieve a 1-log removal of atrazine and its relationship with the EEO In comparison to other advanced oxidation processes (see Table 5.14), for the removal of atrazine, the UV/VUV process shows great potential to compete with the UV/H2O2 process, particularly at higher flow-rates. The EEO value reported by Muller et al., (2000) of 1.67 kW h m-3 order-1 for the UV/H2O2 process considers the stored electrical energy cost in hydrogen peroxide (i.e., 10 kW kg-1). On the other hand, Martjin et al. (2008) reported for a pilot study an EEO that ranges from 1.5 to 0.7 kW h m-3 order-1 as the concentration of H2O2 changes from 4 to 25 mg L-1. Unfortunately, it is not clear if they include or not the cost of the oxidant in their calculations. In addition to the cost of peroxide, it is important to remember that the implementation of UV/H2O2 AOP requires a system to monitor and quench unreacted H2O2, thus increasing the cost and complexity of operation. In contrast, the UV/VUV process does not require ancillary operations, which add to its attractiveness for small and rural water system applications.      1.701.751.801.851.901.952.002.050.000.050.100.150.200.250.300 0.5 1 1.5 2 2.5 3EEO(kW h m-3order-1 )1-log removal required energy  (kW)Flow-rate (L min-1)Lamp + Ballast Energy EEOEFFECT OF TEMPERATURE 50  Table 5.14 Comparison between EEO values reported in literature and the one obtained in this study for the removal of atrazine Advanced Oxidation Process Oxidant Concentration (mg L-1) EEO (kW h m-3 order-1) Reference H2O2/O3 H2O2 = 1.7 O3 = n.a. 0.102 Muller et al., (2000) UV/O3 O3 = n.a. 0.90 UV/H2O2 H2O2 = 17 1.67 UV/H2O2 H2O2 = 4-25 1.5-0.7 Martijn et al., (2008) UV/VUV  1.72-2.00 This Study 5.5. Summary The optical properties of pure water and some common matrix constituents were studied between 3.6 and 25 °C. For pure water, the absorption coefficient increases from 0.79 to 1.53 cm-1 when the temperature increases from 3.6 to 25 °C. Furthermore, for the inorganic ions investigated nitrate was the one with the highest molar absorption coefficient, followed by chloride and sulphate. Regarding their dependence on temperature, the differences in the molar absorption coefficient were found to be not statistically significant. Owing to a weaker attenuation of light at lower temperatures, the mass transfer limitation of the system was, to some extent, overcome thus a larger removal of atrazine was observed when operating the system with distilled water at a temperature of 3.6 °C. In contrast, when raw water was used, the presence of background NOM who acted as an inner filter, reduced, to a larger extent, the amount of photons available for water photolysis at 3.6 °C. Consequently, better removal efficiency was observed when operating the system at a water temperature of 17.5 °C. Finally, the Electrical Energy per Order was calculated at different flow-rates. EEO values between 1.72 and 2.00 kW h m-3 order-1 were obtained and the data suggest that further increase in the flow-rate will result in lower EEO values, hence demonstrating the potential feasibility of this AOP to be implemented in small drinking water systems.  51  6. TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS As described in section 4.4.4, four different water matrices were used to assess the impact of UV/VUV treatment on the degradation of a mixture of micropollutants. The kinetic parameters that characterize the degradation process were measured by conducting experiments in two different collimated beam apparatus, as described in sections 4.3.1 & 4.3.2. Besides, degradation experiments were carried in the flow-through photoreactor described in section 4.3.3. The names and chemical structures of the micropollutants used in these experiments are shown in Figure 6.1.     Atrazine Caffeine Carbamazepine    Deethylatrazine Diclofenac Estradiol    Fluoxetine Linuron Medroxyprogesterone  Figure 6.1 Names and chemical structures of the substances present in the micropollutant mixture  TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 52     Norethindrone Progesterone Sulfamethoxazole Figure 6.1 (Cont.) Names and chemical structures of the substances present in the micropollutant mixture 6.1. Preliminary Kinetic Study There are two mechanisms by which an organic substance can be degraded in aqueous solution when irradiated by 254 and 185 nm photons: direct photolysis and oxidation through reaction with OH radicals. Giving the relatively low concentration of micropollutants in water in comparison to the concentration of water (hence scavenging the majority of the 185 nm radiation), direct photolysis by the action of 185 nm photons can be assumed negligible for compounds present at ppb or ppt levels. In contrast, water weakly absorbs 254 nm radiation and given the relatively high output by Hg lamps at this wavelength, a significant amount of radiation is available to interact directly with the organics present in water. By way of example, considering a compound C in aqueous solution, its removal can be expressed as: −𝑑[𝐶]𝑑𝑡= 𝑘•𝑂𝐻,𝐶 ∙ [• 𝑂𝐻]𝑠𝑠 ∙ [𝐶] + 𝑘𝐷,𝐶 ∙ [𝐶] (6.1) −𝑑[𝐶]𝑑𝑡= [𝐶] ∙ (𝑘•𝑂𝐻,𝐶 ∙ [• 𝑂𝐻]𝑠𝑠 + 𝑘𝐷,𝐶) (6.2) where, k•OH,C and kD,C respectively are the second and first order reaction rate constants for the OH radical and direct photolysis pathways. [•OH]ss and [C] are the concentration of OH radicals at the steady state and the compound C, respectively. The first term in parenthesis in 6.2, can be combined, assuming a steady state concentration of OH radicals, in a pseudo-first order reaction rate term, k’. At this point, both terms can be combined in a global constant reaction rate constant, kapp. −𝑑[𝐶]𝑑𝑡= [𝐶] ∙ (𝑘′ + 𝑘𝐷,𝐶) (6.3) TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 53  −𝑑[𝐶]𝑑𝑡= [𝐶] ∙ 𝑘𝑎𝑝𝑝 (6.4) Given the different radiant (i.e., irradiance) characteristics of the variety of low pressure Hg lamps available and the geometry of the experimental setups, the dose delivered to the system can be used to modify equation 6.4, thus enabling the comparison of results among experimental conditions and apparatus, (Bolton & Stefan, 2002). The dose, D, is the product of the average fluence rate (i.e., equal to irradiance in a collimated beam), I, and time. Dividing equation 6.4 by the average fluence rate, I, results in a reduction based on the dose delivered to the system.  −𝑑[𝐶]𝑑(𝐼 ∙ 𝑡)= −𝑑[𝐶]𝑑𝐷= [𝐶] ∙ 𝑘𝑎𝑝𝑝 (6.5) − ∫𝑑[𝐶][𝐶]𝐶𝐶0= 𝑘𝑎𝑝𝑝 ∙ ∫ 𝑑𝐷𝐷0 (6.6) − ln ([𝐶][𝐶]0) = 𝑘𝑎𝑝𝑝 ∙ 𝐷 (6.7) As described by Levenspiel (1972), kapp can be obtained from the slope of the linear equation that best fits the data of a plot of the left hand side term of equation 6.7 as a function of the delivered dose.  To differentiate the effects of both degradation pathways, experiments were carried out in the presence of exclusively 254 nm radiation. Following the same logic, provided that the removal follows a first order decay, the direct photolysis constant reaction rate, kD,C can be extracted from the experimental data. Subsequently, the pseudo first order reaction rate constant for the OH radical oxidation pathway, k’, can be obtained from: 𝑘′ = 𝑘𝑎𝑝𝑝 − 𝑘𝐷,𝐶 (6.8) As noted by Kim & Tanaka (2009), compounds with high molar absorption coefficients at 254 nm and high photolysis quantum yields are more prone to undergo direct photolysis. Furthermore, as described by Lopez et al., (2003), in a system where the optical light path and delivered dose are constant, the decay due to direct photolysis is exclusively dependent on the photolysis quantum yield and molar absorption coefficient of the TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 54  compound. With this in mind, the values for these two photochemical parameters and their product are shown in Table 6.1. Table 6.1 Photochemical parameters of the substances present in the micropollutant mixture Compound Molar Absorption Coefficient  (ε254nm) at 254 nm      (M-1 cm-1) Direct Photolysis Quantum Yield (Φ) at 254 nm        (mol Einstein-1) ε·Φ                  (L Einstein-1 cm-1) First Order Reaction Rate with OH radicals1 k•OH (M-1 s-1) x109  Reference Atrazine 3860a 0.046b 178 2.30 ± 0.14 a. Hessler et al., (1993) b. Nick et al., (1992) Caffeine 8386 0.0018 ± 0.0003 15.1 6.40 ± 0.71 Rivas et al., (2011) Carbamazepine 6070 0.0006 ± 0.0001 3.64 8.02 ± 1.90 Pereira et al., (2007) Deethylatrazine 6800 0.042 286 1.20 Beltrán et al., (1996) Diclofenac 4260 ± 130 0.384 ± 0.075 1636 8.38 ± 1.24 Canonica et al., (2008) Estradiol 420 ± 20 0.067 ± 0.007 28.1 14.1 Mazellier et al., (2008) Fluoxetine 790 ± 25 0.41 ± 0.04 324 9 ± 1.8 Wols et al., (2014) Linuron 13437 0.0360 484 4.30 Benitez et al., (2009) Medroxy-progesterone N/A N/A N/A N/A  Norethindrone N/A N/A N/A N/A  Progesterone 17020 ± 350 0.022 ±0.005 374 N/A Méité et al., (2010) Sulfamethoxazole 16760 ± 380 0.046 ± 0.021 771 5.82 ± 1.99 Canonica et al., (2008) The information in Table 6.1 suggests that the most sensitive compound to 254 nm radiation is diclofenac, followed by sulfamethoxazole. On the other hand, compounds such as carbamazepine, estradiol and caffeine are not expected to be extensively degraded by direct photolysis. The remaining substances will show similar susceptibility to the action of UV254 nm radiation given the relatively close values for the product of the molar absorption coefficient and the photolysis quantum yield (i.e., between 178 and 484 L Einsteins-1 cm-1). The dose                                                  1 Reference: Wols & Hofman-Caris (2012) TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 55  based first order reaction rate constants for direct photolysis (254 nm radiation only) obtained from UV collimated beam irradiations are shown in Table 6.2. It is important to recall that the UV control runs were not replicated and that the sample was taken after a dose of approximately 590 mJ cm-2 (to match the UV dose delivered in the other UV/VUV CB apparatus) was delivered to the solution. Nonetheless, it is still possible to obtain a rough estimate of this kinetic parameter.   Table 6.2 UV254nm direct photolysis first order reaction rate for each micropollutant Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride  Compound 254 nm Direct Photolysis first order reaction rate, kD,C x103        (cm2 mJ-1) Atrazine 0.4 0.5 0. 4 0. 3 Caffeine Negligible Carbamazepine N/A 0. 2 0. 2 N/A Deethylatrazine 0. 3 0. 2 0. 3 0. 2 Diclofenac 4.8 5.2 4.3 5.0 Estradiol Negligible Fluoxetine 2.7 2.3 1.9 2.0 Linuron 2.2 2.4 2.0 2.0 Medroxyprogesterone N/A 0. 1 N/A 0. 5 Norethindrone 0. 9 0. 6 0. 6 0. 5 Progesterone Negligible Sulfamethoxazole 2.1 1.6 2.4 2.0 The average removal observed by the action of direct photolysis (after a dose of approximately 590 mJ cm-2) for each compound is shown in Figure 6.2. As predicted by the photochemical parameters, diclofenac was the compound most affected by the action of 254 nm photons. Interestingly, fluoxetine, linuron and sulfamethoxazole showed similar removal, even though the sulfamethoxazole is, in theory, slightly more sensitive to 254 nm radiation. Accordingly, caffeine, carbamazepine and estradiol were not significantly impacted by UV254 nm radiation. TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 56   Figure 6.2 Overall average removal observed by direct photolysis. Error bars represent the standard deviation among the removal observed for each water matrix tested For the purpose of obtaining the pseudo first order reaction rate constant for the OH radical oxidation pathway, k’, degradation experiments were carried under 254 and 185 nm irradiation in the UV/VUV collimated beam apparatus. As mentioned earlier, when both degradation pathways contribute, an overall apparent pseudo first order reaction rate constant, kapp, will characterize the removal of the target contaminant. Table 6.3 shows the values obtained for this overall kinetic parameter. As with the UV control runs, no replicates were done but given that several batches of the solution were irradiated at 6 different doses (from 0 to 590 mJ cm-2), it is possible to calculate (when the case allows) a 95% confidence interval for the slope of each linear regression.       19.29.013.593.470.7 72.414.631.769.40255075100Removal by direct photolysisTREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 57  Table 6.3 Global pseudo-first order reaction rate constant for the removal of each micropollutant. Error represents a 95% confidence interval for the slope of the linear regression Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride  Compound Global pseudo-first order reaction rate constant, kapp x103        (cm2 mJ-1) Atrazine 3.78 ± 0.32 1.69 ± 0.11 2.56 ± 0.19  2.41 ± 0.33 Caffeine 6.27 ± 0.96 2.87 ± 0.13 4.78 ± 0.17 4.22 ± 0.98 Carbamazepine 10.65 ± 1.89 4.57 ± 0.16 7.36 ± 0.45 6.69 ± 1.19 Diclofenac 14.65 ± 0.88 18.17 ± 1.46 11.26 ± 0.32 11.21 ± 1.26 Deethylatrazine 1.62 ± 0.34 0.48 ± 0.06 1.03 ± 0.28 1.23 ± 0.26 Estradiol 9.34 ±0.21 10.69 ± 2.50 6.43 ± 0.71 8.89 ± 3.97 Fluoxetine 18.60 ± 3.65 8.93 ± 1.05 14.37 ± 1.64 9.08 ± 1.53 Linuron 10.15 ± 1.13 5.15 ± 0.14 7.09 ± 0.18 6.95 ± 0.79 Medroxyprogesterone 9.84 ± 1.07  3.30 ± 0.29 5.55 ± 0.38 7.29 ± 0.75 Norethindrone 15.18 5.41 ± 0.28 7.71 ± 1.43 6.63 Progesterone 4.98  1.83 ± 0.34 1.12 6.63 ± 1.18 Sulfamethoxazole 8.65 ± 0.69 6.65 ± 0.90 7.69 ± 0.67 6.18 ± 0.82 Comparison of the results presented in Table 6.2 & Table 6.3 demonstrates the significant impact of the OH radical reaction pathway on the degradation of each compound tested (see Figures 6.3 to 6.6). For all the micropollutants, even those highly sensitive to 254 nm radiation, the reaction rate increased markedly. Of all the micropollutants, deethylatrazine showed the smallest improvement when advanced oxidation conditions were present in the system. Given this compound is one of three most common derivatives (i.e., a by-product) of atrazine degradation (Prosen & Zupančič-Kralj, 2005), it is hypothesized that there was a generation of this substance along the irradiation process which negatively impacted the observed degradation. Consequently, the observed kinetic parameter may not reflect the actual interaction of this compound with OH radicals.      TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 58  Table 6.4 OH radical degradation pathway pseudo-first order reaction rate constant for the removal of each micropollutant Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound Pseudo-first order reaction rate constant, k’ x103 (cm2 mJ-1) Atrazine 3.4  1.2 2.2 2.1 Caffeine 6.3 2.8 4.8 5.2 Carbamazepine 10.7 4.4 7.2 6.7 Deethylatrazine 1.3 0.3 0.7 1.0 Diclofenac 9.9 13.0 7.0 6.6 Estradiol 9.3 9.7 6.4 8.9 Fluoxetine 15.9 6.6 12.5 7.4 Linuron 8.0 2.7 5.1 4.7 Medroxyprogesterone 9.8 3.2 5.6 6.8 Norethindrone 14.3 4.8 7.1 6.1 Progesterone 5.0 1.8 1.1 6.6 Sulfamethoxazole 6.6 5.1 5.3 4.3 By means of equation 6.8, it is possible to obtain the pseudo first order reaction rate constant for the OH radical degradation pathway (see Table 6.4).  The results clearly show the impact of water matrix on the degradation rate by OH radical action. Besides the background NOM, alkalinity and chloride present in the raw water used, the other water matrices tested contained a larger concentration of each of these solutes. According to the concentrations used, the OH radical scavenging potential of each matrix will vary and can be calculated from equation 6.9 and the results are shown in Table 6.5. • 𝑂𝐻𝑆𝑐𝑎𝑣𝑒𝑛𝑔𝑖𝑛𝑔 = 𝑘•𝑂𝐻,𝑆𝑖 ∙ [𝑆𝑖] (6.9) where kOH,Si is the second order reaction rate constant between OH radicals and the scavenger, Si. Table 6.5 Reaction rate constant, concentration and scavenging strength for each major solute in the water matrices tested Solute 𝒌•𝑶𝑯,𝑺𝒊 [𝑺𝒊] 𝒌•𝑶𝑯,𝑺𝒊 ∙ [𝑺𝒊] Reference Bicarbonate 8.5x106 M-1 s-1 8.16x10-4 M 6.9x103 s-1 Buxton et al., 1988 Carbonate 3.9x108  M-1 s-1 1.00x10-5 M 3.9x103 s-1 Chloride 4.3x109 M-1 s-1 7.05x10-4 M 3.0x106 s-1 Jayson et al., 1973 Suwannee River NOM 1.4x104 L mg-1 s-1 2 mg L-1 2.8x104 s-1 Sarathy et al., 2011 TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 59  The scavenging terms shown in Table 6.5 suggest that, at the concentrations used in the experiments, chloride has the highest •OH scavenging potential, specifically two orders of magnitude larger than the one calculated for SRNOM and the bicarbonate-carbonate (i.e. alkalinity) containing matrices. However, it is possible to note that the reaction rate constants observed with 25 ppm chloride are comparable, with few exceptions (i.e., estradiol, fluoxetine and progesterone), to the ones observed with 2 ppm SRNOM. The discrepancy between the scavenging effect of chloride and the values obtained for the reaction rate constants can be explained by, as described by Jayson et al., (1973) & Liao et al., (2001), the regeneration of OH radicals from HOCl•-, an intermediate formed in the manifold of reactions resulting from the generation of chlorine based radicals as shown in Table 6.6 (equations 6.10 to 6.15). The reaction rate constant for the reverse reaction shown in equation 6.10 was reported to be 6.1 ± 0.8 x109 s-1 by Jayson et al., (1973). As can be seen from Table 6.6, elucidating the impact of chloride when present in the UV/VUV process is a complex task and is beyond the scope of this study. Nonetheless, it is possible to hypothesize that the chlorine based radicals formed will react unselectively in the medium, hence promoting to some extent the degradation of organic compounds.  Table 6.6 Formation of some chlorine based radicals from chloride2  Reaction Reaction Rate Constant  𝐶𝑙− + • 𝑂𝐻 → 𝐻𝑂𝐶𝑙 •− 4.3 x109 M-1 s-1 (6.10) 𝐻𝑂𝐶𝑙 •−+ 𝐻+ → 𝐻2𝐶𝑙𝑂 • 3.0 x1010 M-1 s-1 (6.11) 𝐻𝑂𝐶𝑙 •−+ 𝐻+ → 𝐶𝑙 • +𝐻2𝑂 2.1 x1010 M-1 s-1 (6.12) 𝐻2𝐶𝑙𝑂 • → 𝐶𝑙 • +𝐻2𝑂 5.0 x108 s-1 (6.13) 𝐶𝑙 • +𝐶𝑙− → 𝐶𝑙2− • 8.5 x109 M-1 s-1 (6.14) 𝐶𝑙2− •  + 𝑂𝐻− → 𝐻𝑂𝐶𝑙− • +𝐶𝑙− 4.0 x109 M-1 s-1 (6.15) When compared, at a chloride concentration of 25 ppm, the scavenging term is three orders of magnitude smaller than the reverse reaction rate constant, meaning that the release of OH radicals is dominant. However, it is possible to hypothesize that there is still a slight OH radical scavenging action. Moreover, according to the results presented in section 5.1, chloride                                                  2 Reference: (Jayson et al., 1973; Liao et al., 2001 & De Laat et al., 2004) TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 60  (ε185nm = 2636 M-1 cm-1 at 17.5 °C) will act as an inner filter, thus reducing the amount of 185 nm photons available in the system to photolize water. In a similar manner, SRNOM (ε185nm = 1341 MC-1 cm-1 at 17.5 °C) will also play a role, to a lesser extent, in scavenging 185 nm photons. In the case of water matrix containing additional alkalinity, the reaction rate constants obtained are, in general, the lowest among the water matrices tested. Even though a decrease was expected in the reaction rates (in comparison to those expected for only SCFP raw water), the results show that alkalinity has the greatest impact in the degradation rate, despite its relatively low scavenging term. This could be explained by the fate of the OH radical once it is scavenged. For example, it was previously shown that when an OH radical is scavenged by a chloride ion several radicals could potentially be formed thus compensating to some extent the absence of OH radicals. On the other hand, the scavenging of OH radicals by carbonate or bicarbonate ions will produce carbonate radicals (Hoigné, 1988). The carbonate radical has a lower oxidation potential (2.3 vs 2.7 V) (Medinas et al., 2007) and is more selective in its reaction with aromatic compounds (Chen et al., 1975) in comparison to the one of the hydroxyl radical. Given the mixture of micropollutants used and the complexity of their chemical structures, carbonate radicals will interact differently with each one of the aromatic micropollutants. The kinetics of these interactions are not know yet. Another factor to take into account is that the recombination of carbonate radicals occur at rates between 106-107 (Neta et al., 1988) and could play a role in limiting the efficiency of the process. However, as with chloride radicals, the radical pathway is complex and has not been extensively studied. Finally, OH radicals will be scavenged by NOM and its subsequent degradation by-products due to the presence of conjugated bonds (i.e., dense electronic zones) (Westerhoff et al., 1999; Buchanan et al., 2004). With a second order reaction rate ranging from 1.21 to 10.36 x108 MC-1 s-1 (McKay, et al., 2014) and with a concentration, in average, several times those of the micropollutants, NOM will strongly compete for OH radicals.  The contribution of each degradation mechanism to the overall pseudo first order reaction rate observed for the removal of each micropollutant is summarized in Figure 6.3, Figure 6.4, Figure 6.5 and Figure 6.6. The contribution is defined by the ratio  %•𝑂𝐻 =𝑘′𝑘𝑎𝑝𝑝 (6.16) TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 61  %𝐷𝑖𝑟𝑒𝑐𝑡 𝑃ℎ𝑜𝑡𝑜𝑙𝑦𝑠𝑖𝑠 =𝑘𝐷,𝐶𝑘𝑎𝑝𝑝 (6.17)  Figure 6.3 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water  Figure 6.4 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water + 50 ppm Carbonate 0%20%40%60%80%100%•OH Radical Attack UV (254 nm) Direct Photolysis0%20%40%60%80%100%•OH Radical Attack UV (254 nm) Direct PhotolysisTREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 62   Figure 6.5 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water + 2 ppm SRNOM  Figure 6.6 Contribution of each degradation pathway to the overall pseudo first order reaction rate observed in raw SCFP water + 25 ppm Chloride  0%20%40%60%80%100%•OH Radical Attack UV (254 nm) Direct Photolysis0%20%40%60%80%100%•OH Radical Attack UV (254 nm) Direct Photolysis 63  6.2. Flow-through Photoreactor The removal of mircopollutants was further assessed by performing degradation experiments in a flow-through photoreactor. The runs were carried out at a water temperature of 20.5 °C and two flow-rates (i.e., 1.9 and 2.7 L min-1) to observe the impact of residence time (i.e., 6.3 and 4.4 s, respectively) on the degradation efficiency of the process. Furthermore, as in the previous section, four different water matrices were used to assess the impact of different solutes in the efficacy of the process. Progesterone was excluded from the results due to problems with the quantification. Table 6.7 shows the removal observed at 1.9 L min-1 for each water matrix tested (results at 2.8 L min-1 are shown in Appendix C). Table 6.7 Removal observed for each micropollutant at a flow rate of 1.9 L min-1 with different water matrices. Error represents the uncertainty in the removal calculated for two samples Type Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride  Compound % Removal Observed Drugs Carbamazepine 81.9 ± 6.6 51.7 ± 1.4 72.2 ± 1.4 60.2 ± 3.2 Diclofenac 91.6 ± 2.8 90.2 ± 4.9 87.6 ± 5.9 83.1 ± 0.3 Fluoxetine 88.1 ± 6.6 79.7 ± 6.8 88.6 ± 9.4 68.6 ± 5.0 Sulfamethoxazole 83.2 ± 3.2 87.6 ± 0.6 78.9 ± 1.3 66.3 ± 2.4 Stimulant Caffeine 70.9 ± 3.1 35.8 ± 0.7 58.8 ± 1.1 50.7 ± 2.2 Hormones Estradiol 86.3 89.0 ± 5.5 61.2 ± 4.3 67.5 ± 3.0 Medroxyprogesterone 77.2 ± 9.5 19.0 ± 0.7 53.9 ± 4.1 51.9 ± 11.6 Norethindrone 87.7 ± 0.4 60.4 ± 1.0 76.1 ± 1.4 70.7 ± 18.2 Herbicides Atrazine 46.3 ± 2.6 28.2 ± 1.0 39.5 ± 1.0 36.2 ± 1.5 Deethylatrazine 12.3 ± 0.2 8.7 ± 0.5 18.5 ± 0.4 24.6 ± 0.3 Linuron 81.0 ± 4.4 62.0 ± 3.0 75.0 ± 4.9 68.9 ± 1.6 As expected, for most of the micropollutants, better removal efficiencies were observed for the case when only SCFP water was used. Again, the comparatively low OH radical scavenging strength of the SCFP water explains the larger removals. Moreover, seven compounds degraded the least in the presence of 50 ppm carbonate, which reaffirms that alkalinity plays a significant role hindering the efficiency of the process, despite its relatively low scavenging term, when compared to the other two synthetic water matrices tested. However, for compounds sensitive to direct photolysis with 254 nm radiation (i.e., diclofenac, sulfamethoxazole, fluoxetine & linuron), the effect of alkalinity on the removal was not as TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 64  significant (note that alkalinity scavenges OH radicals), and similar removals were observed among the matrices tested. On the other hand, for most compounds, similar removals were observed between the water matrices containing additional DOC and chloride. This supports the hypothesis that chloride, when scavenges OH radicals, is able to regenerate a significant fraction of them, thus not having the expected impact predicted from its theoretical scavenging potential. Furthermore, in contrast to the experiments conducted with the collimated beam, the performance observed in the experiments carried out in the flow-through photoreactor was influenced by the hydrodynamics and the optical properties of the solution. The Reynold numbers calculated for the two flow rates used were approximately 607 and 863; hence, the system operated under laminar flow regime and was subject to mass transfer limitation. Owing to this condition, the profile of photon absorption (see Figure 6.7) was estimated from the optical properties (i.e., molar absorption coefficient) that were measured at 17.5 °C in order to observe the impact of each water matrix on the attenuation of 185 nm photons. Additionally, the contribution of each solute to the photon scavenging was calculated (see Table 6.8) according to equation 5.6. It was assumed that all dissolved organic carbon behaved, in optical terms, as SRNOM and bicarbonate (i.e., chemical specie present at the solution pH value of 6) had a molar absorption coefficient of approximately 269 M-1 cm-1 (reported by Duca, 2015).  Table 6.8 Estimated absorption coefficient of each water matrix and 185 nm photon scavenging fraction of each major solute in solution  185 nm Photon Absorption Fraction Distribution Water Matrix 185 nm Absorption coefficient (cm-1) Water DOC Cl- NO3 SO42- HCO3- SCFP 1.63 0.75 0.18 0.03 0.03 9.75E-04 1.67E-03 SCFP + 50 ppm Carbonate 1.85 0.66 0.16 0.03 0.03 8.58E-04 1.21E-01 SCFP + 2 ppm SRNOM 1.85 0.66 0.28 0.03 0.03 8.57E-04 1.47E-03 SCFP + 25 ppm Chloride 3.49 0.35 0.09 0.55 0.01 4.55E-04 7.79E-04 TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 65  The estimated absorption coefficient for the different aqueous matrices at 185 nm indicates that 185 nm photons are attenuated the most in the 25 ppm chloride solution, followed by the water containing SRNOM, carbonate and the control matrix. The 185 nm photon scavenging contribution of nitrate, sulphate and bicarbonate can be considered negligible in all matrices tested. In addition, owing to the relatively high molar absorption coefficient of chloride, it absorbed 55% of the 185 nm photons reaching the aqueous media, leaving only 35% to interact directly with water to produce OH radicals. As indicated before, these radicals will be, to some extent, scavenged by chloride. Furthermore, according to reactions 6.10 to 6.15 and the results obtained in both collimated beam and flow-through photoreactor, it is possible to hypothesize that, apart from the OH radicals that are released back into solution, the micropollutants may interact with chloride based radicals, thus forming chlorinated by-products. Altogether, these two mechanisms may explain the degradation behavior observed for the chloride containing matrix, which is able to overcome relatively strong mass transfer limitation as shown from the photon absorption profile shown in Figure 6.7. On the other hand, the NOM and carbonate matrix had similar absorption coefficients, resulting in similar photon absorption profiles (Figure 6.7). In both cases, water absorbed similar amount of 185 nm photons (≈66%), but the removal efficiencies were lower in the carbonate matrix for 8 out of 11 compounds. This is in agreement with the global pseudo-first order reaction rates observed in the UV/VUV collimated beam apparatus (Table 6.3).   Figure 6.7 185 nm photon absorption profile in the photoreactor with different water matrices 0204060801001.40 1.63 1.86 2.08 2.31 2.54% Photons AbsorbedReactor Radius (cm)SCFP + 25 ppm ChlorideSCFP + 50 ppm CarbonateSCFP + 2 ppm SRNOMSCFP 66  6.3. Estimation of the Electrical Energy per Order The results obtained from the collimated beam and flow-through degradation experiments were used to estimate roughly the Electrical Energy per Order for each compound tested. To start with, the global reaction rate constant, kapp, was used to obtain the UV based dose required to achieve 90% removal of a particular compound. Secondly, the amount of theoretical passes needed to achieve 1-log removal was obtained from the average dose delivered to the system, which can be calculated for each compound knowing the average removal observed in the flow-through experiments. The ratio of these two quantities allowed quantifying the total energy input required to obtain 90% removal. Lastly, the EEO can then be calculated from equation 6.18. The results for the EEO calculated for each compound at 1.9 L min-1 are shown in Table 6.9 (EEO results at 2.7 L min-1 are shown in Appendix A). 𝐸𝐸𝑂 =𝑃𝑙𝑎𝑚𝑝 ∙ 𝑅𝑒𝑎𝑙 # 𝑜𝑓 𝑝𝑎𝑠𝑠𝑒𝑠?̇? ∙ 60 (6.18) Table 6.9 Estimated Electrical Energy per Order for each micropollutant in several water matrices at 1.9 L min-1 and 20 °C. Error represents the uncertainty associated in the EEO calculation Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound Estimated EEO (kW h m-3 order-1) Atrazine 2.05 ± 0.19 3.85 ± 0.31 2.54 ± 0.09 2.84 ± 0.41 Caffeine 1.03 ± 0.03 2.88 ± 0.27 1.44 ± 0.07 1.81 ± 0.10 Carbamazepine 0.75 ± 0.04 1.75 ± 0.11 1.00 ± 0.07 1.39 ± 0.06 Deethylatrazine 9.75 ± 7.30 14.09 ± 11.42 6.25 ± 1.35 4.52 ± 0.21 Diclofenac 0.52 ± 0.01 0.55 ± 0.02 0.61 ± 0.06 0.72 ± 0.02 Estradiol 0.64 0.58 ± 0.04 1.35 ± 0.19 1.14 ± 0.13 Fluoxetine 0.60 ± 0.11 0.80 ± 0.02 0.59 ± 0.17 1.10 ± 0.09 Linuron 0.77 ± 0.05 1.32 ± 0.17 0.92 ± 0.09 1.09 ± 0.06 Medroxyprogesterone 0.86 ± 0.14 6.07 ± 1.43 1.65 ± 0.26 1.75 ± 0.34 Norethindrone 0.61 ± 0.04 1.38 ± 0.17 0.89 ± 0.12 1.04 ± 0.15 Sulfamethoxazole 0.71 ± 0.03 0.61 ± 0.07 0.82 ± 0.05 1.17 ± 0.06 As in the previous section, Progesterone was excluded from the analysis due to problems in its quantification. Under the circumstances of the experiments, as mentioned before, the parameters obtained for deethylatrazine were influenced by the degradation of atrazine; hence, the estimated EEO for this particular compound will be, to some extent, lower TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 67  in a controlled system where no atrazine is present. Nonetheless, deethylatrazine is expected to have a larger EEO than atrazine, given its relatively low reaction rate with OH radicals of 1.20x109 M-1 s-1 in comparison to the one of atrazine (2.30 ± 0.14x109 M-1 s-1) (Wols & Hofman-Caris, 2012) and, as shown in Table 6.1, a lower sensitivity to direct photolysis. From the results obtained, it is possible to note that apart from deethylatrazine and atrazine, most EEO estimated are below the target of 2.5 kW h m-3 order-1, suggesting that the UV/VUV process, even with adverse scavenging matrices, is capable of treating a variety of compounds at an affordable cost. However, it is important to note that there will be an even larger number of degradation by-products that may still have adverse effects on flora and fauna. For example Choi et al., (2013) studied the toxic effects of 5 possible by-products of atrazine when degraded by UV and UV/H2O2. They found that one by product (i.e., hydroxyatrazine) had a greater toxicity that atrazine itself and three other had a lower toxicity evaluation (i.e., deisopropylatrazine, deethylatrazine and deethyldeisopropylatrazine) when using Daphnia magna as a model organism. Although there are several studies that have identified and dealt with degradation by-products resulting from the application of UV based AOPs (Lopez et al., 2003; Vogna et al., 2004a; Vogna et al., 2004b; Prosen & Zupančič-Kralj, 2005) , there is a lack of information regarding the potential toxic effects of these substances.  In comparison to other Advanced Oxidation Processes, Shu et al., (2013) reported an EEO value for the UV/H2O2 process in batch operation (in synthetic waters) of 3.5 and 3.7 kW h m-3 order-1 for the removal of diclofenac (C0 = 40 mg L-1) and an EEO value of 7.1 and 5.0 kW h m-3 order-1 for the removal of caffeine (C0 = 20 mg L-1) with peroxide concentrations of 25 and 50 mg L-1, respectively. In another study by Ijpelaar et al., (2010), EEO values for the UV/H2O2 process operated in continuous mode (in natural water with 4 mg L-1 DOC and 188 mg L-1 HCO3-) were reported at approximately 0.65, 0.4 0.2 and 0.1 kW h m-3 order-1 for atrazine, carbamazepine, diclofenac and sulfamethoxazole (initial concentration for all micropollutants between 2-4 μg L-1) for a peroxide concentration of 10 mg L-1, respectively. In comparison to this study, the values estimated for the UV/VUV process are larger in all cases. 6.4. Summary The impact of the UV/VUV AOP process in a variety of micropollutants was evaluated by conducting a preliminary kinetic study and degradation experiments in a flow-TREATMENT EFFECT ON A MIXTURE OF MICROPOLLUTANTS 68  through photoreactor. The kinetic study showed that the OH radical pathway is the dominant mechanism of removal and although some substances like diclofenac, sulfamethoxazole, fluoxetine and linuron can be easily degraded by only 254 nm radiation, the presence of OH radicals reduces significantly the energy required (dose) to observe the same removal extent.  Among the different water matrices tested, overall, the one with additional alkalinity showed the greatest impact on the efficiency of the process. On the other hand, despite its high OH radical scavenging potential, the matrix with additional chloride showed similar results to the matrix with additional DOC. In view of the relatively high molar absorption coefficient of chloride at 185 nm, the formation of chlorine based radicals is likely another degradation pathway that may explain the comparable results to the ones observed with the matrix containing additional DOC. The Electrical Energy per Order was estimated for each micropollutant in the different water matrices tested. At a flow-rate of 1.9 L min-1 only deethylatrazine and atrazine had EEO values larger than 2.5 kW h m-3 order-1. Specific cases such as the one of caffeine and medroxyprogesterone in the water matrix with additional alkalinity were above the target value as well. Nonetheless, the results show that the UV/VUV AOP can be considered as an option to economically degrade micropollutants from drinking water. Despite the promising results, it is of great importance to understand the fate of the micropollutants after treatment to assess the safety of the water produced.  69  7. CONCLUSIONS AND RECOMMENDATIONS 7.1. Overall Conclusions The effect of temperature on the efficacy and energy consumption of a continuously operated flow-through UV/VUV photoreactor was investigated. Furthermore, the impact of treatment was assessed for a variety of micropollutants and the Electrical Energy per Order was calculated for each compound under the influence of different water matrices. Finally, the influence of temperature on the optical properties of water and major solutes at 185 nm was studied. The overall conclusions of this work are as follows:  The optical absorption properties of pure water, SRNOM, nitrate, chloride and sulphate were determined between 3.5 and 25 °C increase with temperature. Moreover, the absorption coefficient of water increased from 0.79 to 1.53 cm-1 from 3.6 to 25 °C. For the solutes tested, nitrate showed the largest molar absorption coefficient, followed by chloride, SRNOM and sulphate. On the other hand, the attenuation of 185 nm by chloride was the most sensitive to temperature. However, the differences observed for each solute at the temperatures tested were not statistically significant.   The performance of the UV/VUV system operated with raw water was enhanced at higher temperatures due to the increased absorption of 185 nm photons to induce water photolysis. It was estimated that at 3.6 °C, water absorbed approximately 68% of the photons reaching the solution while at 17.5 °C, the fraction absorbed increased to 75%. In the case of pure water, given the absence of photon scavengers, the majority of 185 nm photons were absorbed by water thus, the reduction in mass transfer limitation due to an increase in the photoreaction zone at lower temperatures enhanced the performance of the system.  The Electrical Energy per Order for the removal of atrazine decreased with an increase in flow-rate. EEO values between 2.00 and 1.72 kW h m-3 order-1 were calculated for the removal of this micropollutant. These results show that, for the case of atrazine, the UV/VUV process can be considered as an alternative to the UV/H2O2 process  CONCLUSIONS AND RECOMMENDATIONS  70   The kinetic study performed for a variety of micropollutants showed that the OH radical pathway is the dominant mechanism of removal and although some substances like diclofenac, sulfamethoxazole, fluoxetine and linuron can be easily degraded by only 254 nm radiation, the presence of OH radicals reduces significantly the energy required (dose) to obtain the same removal extent.  The Electrical Energy per Order values obtained for each compound under the four water matrices tested demonstrate the potential applicability of the UV/VUV process to economically degrade a variety of micropollutants. 7.2. Research Outcome Significance The outcomes obtained from this research contribute to the scientific and practical body of knowledge. This is the first study that has applied Vacuum-UV spectroscopy to study the dependence on temperature of the 185 nm absorbance of water and some major solutes within temperatures commonly found in areas of the world with marked seasonal variations. Moreover, it was shown that, as long as the source water quality is known, it is possible to calculate within 10 to 15% error the absorption coefficient at 185 nm of an aqueous solution. This can be used as a tool to quickly establish the impact that UV/VUV treatment may have on a particular source water. Furthermore, it was shown that the effect of water temperature on the efficiency and energy consumption of the UV/VUV process was not extensive. Therefore, provided that water quality allows, this AOP can be considered as a potential drinking water treatment alternative for small systems, even if they suffer from extreme seasonal temperature variations. 7.3. Recommendations for Future Work  Extend the study of the optical properties and their dependence on temperature for other solutes such as nitrite, phosphate, bicarbonate, and NOM isolates.  Explore the effect of temperature on the 185 nm photolysis quantum yield of water.  Investigate the impact of the UV/VUV process on chlorine demand and disinfection by-product potential.  Assess the fate and toxicity of the degradation byproducts formed by the action of the UV/VUV process.  71  REFERENCES Andreozzi, R., Caprio, V., Insola, A., & Marotta, R. (1999). Advanced oxidation processes (AOP) for water purification and recovery. Catalysis Today, 53(1), 51-59. Arany, E., Szabó, R., Apáti, L., Alapi, T., Ilisz, I., Mazellier, P., . . . Gajda-Schrantz, K. (2013). Degradation of naproxen by UV, VUV photolysis and their combination. Journal of Hazardous Materials, 262, 151-157. Azrague, K., & Osterhus, S. W. (2009). Persistent organic pollutants (POPs) degradation in natural waters using a V-UV/UV/TiO2 reactor. Water Science & Technology, 9(6), 653-660. Bagheri, M., & Mohseni, M. (2014). Computational fluid dynamics (CFD) modeling of VUV/UV photoreactors for water treatment. Chemical Engineering Journal, 256, 51-60. Bagheri, M., & Mohseni, M. (2015). Impact of hydrodynamics on pollutant degradation and energy efficiency of VUV/UV and H2O2/UV oxidation processes. Journal of Environmental Engineering, 164, 114-120. Barrett, J., & Baxendale, J. H. (1960). The Photolysis of Liquid Water. Transactions of the Faraday Society, 37-43. Barrett, J., & Mansell, A. L. (1960). Ultra-violet Absorption Spectra of the Molecules H2O, HDO and D2O. Nature, 138. Baxendale, J. H., & Wilson, J. A. (1957). THE PHOTOLYSIS OF HYDROGEN PEROXIDE AT HIGH LIGHT INTENSITIES. Transactions of the Faraday Society, 344-356. Beltrán, F. J. (2003). Ozone reaction kinetics for water and wastewater systems. Florida, USA: CRC Press. Beltrán, F., González, M., Rivas, F., & Álvarez, P. (1996). Aqueous UV radiation and UV/H2O2 oxidation of atrazine first degradation products: deethylatrazine and deisopropylatrazine. Environmental Toxicology and Chemistry, 15(6), 868-872. Benitez, F. J., García, C., Acero, J. L., & Real, F. J. (2009). Removal of Phenylurea Herbicides from Water by using Chemical Oxidation Treatments. World Academy of Science, Engineering and Technology, 58, 673-680. REFERENCES  72  Bolton, J., & Cater, S. (1994). Homogeneus Photodegradation of Pollutants in Contaminated Water: An Introduction. In Surface and Aquatic Environmental Photochemistry (pp. 467-490). Florida: CRC Press. Bolton, J., & Stefan, M. (2002). Fundamental photochemical approach to the concepts of fluence (UV dose) and electrical energy efficiency in photochemical degradation reactions. Research on Chemical Intermediates, 28(7), 857-870. Braslavsky, S. E. (2007). Glossary of terms used in photochemistry. Pure and Applied Chemistry, 79(3), 293-465. Buchanan, W., Roddick, F., Porter, N., & Drikas, M. (2004). Enhanced biodegradability of UV and VUV pre-treated natural organic matter. Water Science and Technology: Water Supply, 4(4), 103-111. Buxton, G., Greenstock, C., Helman, P., & Ross, A. (1988). Critical Review of Rate Constants for Reactions of Hydrated Electrons, Hydrogen Atoms and Hydroxyl Radicals (.OH/.O-) in Aqueous Solution. Journal of Physical and Chemical Reference Data, 17(2), 513-886. Canonica, S., Meunier, L., & von Gunten, U. (2008). Phototransformation of selected pharmaceuticals during UV treatment of drinking water. Water Research, 42, 121-128. Chen, S. N., Hoffman, M., & Parsons , G. (1975). Reactivity of the carbonate radical toward aromatic compounds in aqueous solution. Journal of Physical Chemistry, 1911-1912. Cheney, V. (1996). AM1 calculations on reactive oxygen species. Part 1. Analysis of hydroxyl radical reactions. Journal of Molecular Structure, 219-237. Choi, H.-J., Kim, D., & Lee, T.-J. (2013). Photochemical degradation of atrazine in UV and UV/H2O2 process: pathways and toxic effects of products. Journal of Environmental Science and Health, Part B: Pesticides, Food, Contaminants , and Agticultural Wastes, 48(11), 927-934. Daimon, M., & Masumura, A. (2007). Measurement of the refractive index of distilled water from the near-infrared region to the ultraviolet region. Applied Optics, 46(18), 3811-3820. De Laat, J., Truong, G., & Legube, B. (2004). A comparative study of the effects of chloride, sulfate and nitrate ions on the rates of decomposition of H2O2 and organic compounds by Fe(II)/H2O2 and Fe(III)/H2O2. Chemosphere, 55, 715-723. REFERENCES  73  Dobrović, S., Juretić, H., & Ružinski, N. (2007). Photodegradation of Natural Organic Matter in Water with UV Irradiation at 185 and 254 nm Importance of Hydrodynamic Conditions on the Decomposition Rate. Separation Science and Technology, 1421-1432. Duca, C. (2015, February 24). Effect of water matrix on Vacuum UV process for the removal of organic micropollutants in surface water (Thesis/Dissertation). Electronic Theses and Dissertations (ETDs) 2008+. University of British Columbia. Retrieved November 10, 2015, from https://open.library.ubc.ca/cIRcle/collections/24/items/1.0167680 Duca, C., Imoberdorf, G., & Mohseni, M. (2014). Novel Collimated Beam Setup to Study the Kinetics of VUV-Induced Reactions. Photochemistry and Photobiology, 90(1), 238-240. Duran, E., Taghipour, F., & Mohseni, M. (2010). Irradiance modeling in annular photoreactors using the finite-volume method. Journal of Photochemistry and Photobiology A: Chemistry, 81-89. Fox, M., Barker, B., & Hayon, E. (1978). Far-ultraviolet solution spectroscopy of chloride ion. Journal of the Chemical Society, Faraday Transactions 1, 74, 1776-1785. Gettoff, N., & Schenck, M. (1968). Primary products of liquid water photolysis at 1236, 1470 and 1849 A. Photochemistry and Photobiology, 167-178. Glaze, W., Kang, J.-W., & Chapin, D. (1988). The Chemistry of Water Treatment Processes Involving Ozone, Hydrogen Peroxide and Ultraviolet Radiation. Ozone Science & Engineering, 335-352. Glaze, W., Lay, Y., & Kang, J. (1995). Advanced Oxidation Processes. A Kinetic Model for the Oxidation of 1, 2-Dibromo-3-chloropropane in Water by the Combination of Hydrogen Peroxide and UV Radiation. Industrial Engineering & Engineering Chemistry Research, 2314-2323. Goldring, L. S., Hawes, R. C., Hare, G. H., Beckman, A. O., & Stickney, M. E. (1953). Anomalies in Extinction Coefficient Measurements. Analytical Chemistry Journal, 25(6), 869-878. Gonzalez, M., & Braun, A. (1994). Vacuum-ultraviolet (VUV) photolysis of water: mineralization of atrazine. Chemosphere, 28(12), 2121-2127. Government of British Columbia. (n.d.). Small Water System Handbook. Retrieved September 26, 2015, from http://www2.gov.bc.ca/assets/gov/environment/air-land-water/small-water-system-guidebook.pdf Griesmann, U., & Burnett, J. (1999). Reactivity of nitrogen gas in the vacuum ultraviolet. Optics Letters, 24(23), 1699-1701. REFERENCES  74  Halmann, M., & Platzner, I. (1966). Temperature Dependence of Absorption of Liquid Water in the Far-Ultraviolet Region. The Journal of Physical Chemistry, 580-581. Han, W., Zhang, P., Zhu, W., Yin, J., & Li, L. (2004). Photocatalysis of p-chlorobenzoic acid in aqueous solution under irradiation of 254 nm and 185 nm UV light. Water Research, 38(19), 4197-4203. Heraeus . (n.d.). Suprasil UVL. Retrieved November 8, 2015, from http://heraeus-quarzglas.com/media/webmedia_local/downloads/broschren_mo/Suprasil_UVL_synthetic_fused_silica.pdf Heraeus Noblelight. (2015). Retrieved October 2, 2015, from http://www.heraeus-noblelight.com/en/productsandsolutions/uv/amalgamlamps.aspx Hessler, D. P., Gorenflo, V., & Frimmel, F. H. (1993). Degradation of Aqueous Atrazine and Metazachlor Solutions by UV and UV/H2O2 — Influence of pH and Herbicide Concentration. Acta hydrochimica et hydrobiologica, 21(4), 209-214. Hoigné, J. (1988). The Chemistry of Ozone in Water. In S. Stucki (Ed.), Process Technologies for Water Treatment (pp. 121-141). New York: Plenum Press. Ijpelaar, G., Harmsen, D., Beerendonk, E., van Leerdam, R., Metz, D., Knol, A., . . . Krijnen, S. (2010). Comparison of Low Pressure and Medium Pressure UV Lamps for UV/H2O2 Treatment of Natural Waters Containing Micro Pollutants. Ozone: Science & Engineering, 32, 329-337. Imoberdorf, G., & Mohseni, M. (2011). Degradation of natural organic matter in surface water using vacuum UV irradiation. Journal of Hazardous Materials, 186(1), 240-246. Imoberdorf, G., & Mohseni, M. (2012). Kinetic study and modeling of the vacuum-UV photoinduced degradation of 2,4-D. Chemical Engineering Journal, 187(1), 114-122. Ito, M. (1960). The Effect of Temperature on Ultraviolet Absorption Spectra and its Relation to Hydrogen Bonding. Journal of Molecular Spectroscopy, 4(1-6), 106-124. Jakob, L., Hashem, T., Bürki, S., Guindy, N., & Braun, A. (1993). Vacuum-ultraviolet (VUV) photolysis of water: oxidative degradation of 4-chlorophenol. Journal of Photochemistry and Photobiology A: Chemistry, 75(2), 97-103. Jayson, G. G., Parsons, B. J., & Swallow, A. J. (1973). Some Simple, Highly Reactive, Inorganic Chlorine Derivatives in Aqueous Solution. Journal of the Chemical Society, Faraday Transactions 1, 69, 1597-1607. REFERENCES  75  Kim, I., & Tanaka, H. (2009). Photodegradation characteristics of PPCPs in water with UV treatment. Environmental International, 35(5), 793-802. Kröckel, L., & Schmidt, M. (2014). Extinction properties of ultrapure water down to deep ultraviolet wavelengths. Optical Materials Express, 4(9), 1932-1942. Kustchera, K., Börnick, H., & Worch, E. (2009). Photoinitiated oxidation of geosmin and 2-methylisoborneol by irradiation with 254 nm and 185 nm UV light. Water Research, 2224-2232. Lee, C., & Yoon, J. (2004). Temperature dependence of hydroxyl radical formation in the hv/Fe3+/H2O2 and Fe3+/H2O2 systems. Chemosphere, 923-934. Levenspiel, O. (1972). Chemical Reaction Engineering. New York, United States of America: John Wiley & Sons. Li, W., Lu, S., Qiu, Z., & Lin , K. (2011). UV and VUV photolysis vs UV H2O2 and VUV H2O2 treatment for removal of clofibric acid from aqueous solution. Environmental Technology, 1063-1071. Liao, C., Kang, S., & Wu, F.-A. (2001). Hydroxyl radical scavneging role of chloride and bicarbonate ions in the H2O2/UV process. Chemosphere, 44, 1193-1200. Light Sources Inc. (2015). Retrieved October 02, 2015, from http://www.light-sources.com/solutions/germicidal-uvc-lamps/uv-germicidal-lamps/low-pressure-amalgam-lamps/ Lopez, A., Bozzi, A., Mascolo, G., & Kiwi, J. (2003). Kinetic investigation on UV and UV/H2O2 degradations of pharmaceutical intermediates in aqueous solution. Journal of Photochemistry and Photobiology A: Chemistry, 156(1-3), 121-126. Martijn, B., Kamp, P., & Kruithof, J. (2008). UV/H2O2 Treatment: An Essential Barrier in a Multi Barrier Approach For Organic Contaminant Removal. IUVA News, 10(4), 12-19. Masschelein, W. (2002). Ultraviolet Light in Water and Wastewater Sanitation. (R. Rice, Ed.) Florida: CRC Press. Mazellier, P., Méité, L., & De Laat, J. (2008). Photodegradation of the steroid hormones 17β-estradiol (E2) and 17α-ethinylestradiol (EE2) in dilute aqueous solution. Chemosphere, 73(8), 1216-1223. McKay, G., Dong , M., Kleinman, J., Mezyk, S., & Rosario-Ortiz, F. (2011). Temperature Dependence of the Reaction between the Hydroxyl Radical and Organic Matter. Environmental Science & Technology, 45(16), 6932-6937. REFERENCES  76  McKay, G., Kleinman, J., Johnston, K., Dong, M., Rosario-Ortiz, F., & Mezyk, S. (2014). Kinetics of the reaction between the hydroxyl radical and organic matter standards from the International Humic Substance Society. Journal of Soils and Sediments, 14(2), 298-304. Medinas, D., Cerchiaro, G., Trindade, D., & Augusto, O. (2007). The carbonate radical and related oxidants derived from bicarbonate buffer. IUBMB Life, 255-262. Méité, L., Szabó, R., Mazellier, P., & De Laat, J. (2010). Cinétique de phototransformation de polluants organiques émergents en solution aqueuse diluée. Revue de sciences de l'eau, 23(1), 31-36. Mills, A., Davies, R. H., & Worsley, D. (1993). Water Purification by Semiconductor Photocatalysis. Chemical Society Reviews, 22(6), 417-425. Moortgat, G. K., Seiler, W., & Warneck, P. (1983). Photodissociation of HCHO in air: CO and H2 quantum yields at 220 and 300 K. The Journal of Chemical Physics, 1185-1190. Morissette, M. F., Vo Duy, S., Arp, H. P., & Sauvé, S. (2015). Sorption and desorption of diverse contaminants of varying polarity in wastewater sludge with and without alum. Environmental Science: Processes & Impacts, 17, 674-682. Mouamfon, M., Li, W., Lu, S., Chen, N., Qiu, Z., & Lin, K. (2011). Photodegradation of Sulfamethoxazole Applying UV- and VUV-Based Processes. Water, Air and Soil Pollution, 265-274. Müller J-P, Gottschalk C, Jekel M (2000) Comparison of Advanced Oxidation Processes in Flow-Through Pilot Plants, Proceedings of the 2nd International Conference on Oxidation Technologies for Water and Wastewater Treatment, Clausthal-Zellerfeld, Germany, 28–31 May 2000, electronic release. Neta, P., Huie, R., & Ross, A. (1988). Rate Constants for Reactions of Inorganic Radicals in Aqueous Solution. Journal of Physical and Chemical Reference Data, 17(3), 1027-1284. Nick, K., Schöler, H. F., Mark, G., Söylemez, T., Akhlaq, M. S., Schuchmann, H. P., & von Sonntag, C. (1992). Degradation of some triazine herbicides by UV radiation such as used in the UV disinfection of drinking water. Journal of Water Supply Research and Technology-Aqua, 41(2), 82-87. Oppenländer, T. (2003). The Photochemical Purification of Water and Air. Weinheim: Wiley. Oppenländer, T., & Gliese, S. (2000). Mineralization of organic pollutants (homologous alcohols and phenols) in water by vacuum-UV-oxidation (H2O-VUV) with and incoherent xenon-excimer lamp at 172 nm. Chemosphere, 40(1), 15-21. REFERENCES  77  Oppenländer, T., & Xu, F. (2008). Temperature Effects on the Vacuum-UV (VUV)-Initiated Oxidation and Mineralization of Organic Compounds in Aqueous Solution Using a Xenon Excimer Flow-through Photoreactor at 172 nm. Ozone: Science and Engineering, 99-104. Oppenländer, T., Walddörfer, C., Burgbacher, J., Kiermeier, M., Lachner, K., & Weinschrott, H. (2005). Improved vacuum-UV (VUV)-initiated photomineralization of organic compounds in water with a xenon excimer flow-through photoreactor (Xe2* lamp, 172 nm) containing an axially centered ceramic oxygenator. Chemosphere, 60(3), 302-309. Pereira, V., Weinberg, H., Linden, K., & Singer, P. (2007). UV Degradation Kinetics and Modeling of Pharmaceutical Compounds in Laboratory Grade and Surface Water via Direct and Indirect Photolysis at 254 nm. Environmental Science & Technology, 41(5), 1682-1688. Prosen, H., & Zupančič-Kralj, L. (2005). Evaluation of photolysis and hydrolysis of atrazine and its first degradation products in the presence of humic acids. Environmental Pollution, 133(3), 517-529. Rahn, R., Bolton, J., & Stefan, M. (2006). The Iodide/Iodate Actinometer in UV Disinfection: Determination of the Fluence Rate Distribution in UV Reactors. Photochemistry and Photobiology, 611-615. Rivas, J., Gimeno, O., Borralho, T., & Sagasti, J. (2011). UV-C and UV-C/peroxide elimination of selected pharmaceuticals in secondary effluents. Desalination, 279(1-3), 115-120. Roehl, C., Orlando, J., Tyndall, G., Shetter, R., Vázquez, G., Cantrell, C., & Calvert , J. (1994). Temperature Dependence of the Quantum Yields for the Photolysis of NO2 Near the Dissociation Limit. Journal of Physical Chemistry, 7837-7843. Sarathy, S., & Mohseni, M. (2009). The fate of natural organic matter during UV/H2O2 advanced oxidation of drinking water. Canadian Journal of Civil Engineering, 36(1), 160-169. Sarathy, S., Bazri, M., & Mohseni, M. (2011). Modeling th Transformation of Chromophoric Natural Organic Matter during UV/H2O2 Advanced Oxidation. Journal of Environmental Engineering, 137(10), 903-912. Schwarzenbach, R., Escher, B., Fenner, K., Hofstetter, T., Johnson, A., von Gunten, U., & Wehrli, B. (2006). The Challenge of Micropollutants in Aquatic Systems. Science, 313, 1072-1077. REFERENCES  78  Shu, Z., Bolton, J., Belosevic, M., & Gamal El Din, M. (2013). Photodegradation of emerging micropollutants using the medium-pressure UV/H2O2 Advanced Oxidation Process. Water Research, 47, 2881-2889. Sokolov, U., & Stein , G. (1966). Photolysis of Liquid Water at 1470 A. The Journal of Chemical Physics, 2189-2192. Stefan, M. I., & Williamson, C. T. (2004). UV light-based applications. In S. Parsons (Ed.), Advanced Oxidation Processes for Water and Wastewater Treatment. London: International Water Association. Stevenson, D. P. (1968). On the Monomer Concentration in Liquid Water. The Journal of Physical Chemistry, 69(7), 2145-2152. Szabó, R. K., Megyeri, C., Illés, E., Gadja-Schrantz, K., Mazellier, P., & Dombi, A. (2011). Phototransformation of ibuprofen and ketoprofen aqueous solutions. Chemosphere, 84(11), 1658-1663. Talukdar, R. K., Longfellow, C. A., Gilles, M. K., & Ravishankara, A. R. (1998). Quantum yields of O(1D) in the photolysis of ozone between 289 and 329 nm as a function of temperature. Geophysical Research Letters, 25(2), 143-146. Tomiyasu, H., Fukutomi, H., & Gordon, G. (1985). Kinetics and mechanism of ozone decomposition in basic aqueous solution. Inorganic Chemistry, 24(19), 2962-2966. Trojan UV. (n.d.). Retrieved October 1, 2015, from http://www.trojanuv.com/resources//casestudies/ECT/Treatment_of_Groundwater_Contaminated_with_1_4_Dioxane___Tucson__Arizona_Case_Study___Environmental_Contaminant_Treatment.pdf Trojan UV. (n.d.). Retrieved October 1, 2015, from http://www.trojanuv.com/resources//casestudies/ECT/Indirect_Potable_Reuse___Big_Spring__Texas_Case_Study.pdf Trojan UV. (n.d.). Retrieved October 1, 2015, from http://www.trojanuv.com/resources/trojanuv/casestudies/TrojanUVPhox_TM__Case_Study___Orange_County_Water_District.pdf Trojan UV. (n.d.). Retrieved October 1, 2015, from http://www.trojanuv.com/resources//casestudies/ECT/Aurora__Colorado_ARWPF_Case_Study___Environmental_Contaminant_Treatment.pdf Tühkanen, T. A. (2004). UV/H2O2 processes. In S. Parsons (Ed.), Advanced Oxidation Processes for Water and Waste Water Treatment. London: International Water Association. REFERENCES  79  USP Technologies. (n.d.). Retrieved October 1, 2015, from http://www.h2o2.com/faqs/FaqDetail.aspx?fId=25 Vogna , D., Marotta, R., Napolitano, A., Andreozzi, R., & d'Ischia, M. (2004). Advanced oxidation of the pharmaceutical drug diclofenac with UV/H2O2 and ozone. Water Research, 414-422. Vogna, D., Marotta, R., Andreozzi, R., Napolitano, A., & d'Ischia, M. (2004). Kinetic and chemical assessment of the UV/H2O2 treatment of antiepileptic drug carbamezapine. Chemosphere, 497-505. Wang, B. B., Cao, M. H., Tan, Z. J., Wang, L. L., Yuan, S. H., & Chen, J. (2010). Photochemical decomposition of perfluorodecanoic acid in aqueous solution with VUV light irradiation. Journal of Hazardous Materials, 181(1-3), 187-192. Weeks, J., & Rabani, J. (1966). The Pulse Radiolysis of Deaerated Aqueous Carbonate Solutions I. Transient Optical Spectrum and Mechanism II.pk for OH radicals . The Journal of Physical Chemistry, 2100-2106. Weeks, J., Meaburn, G., & Gordon, S. (1963). Absorption Coefficients of Liquid Water and Aqueous Solutions in the Far Ultraviolet. Radiation Research, 559-567. 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 Research, 33(10), 2265-2276. Westerhoff, P., Mezyk, S., Cooper, W., & Minakata, D. (2007). Electron Pulse Radiolysis Determination of Hydroxyl Radical Rate Constants with Suwannee River Fulvic Acid and Other Dissolved Organic Matter Isolates. Environmental Science & Technology, 41(13), 4640-4646. Wols, B. A., & Hofman-Caris, C. H. (2012). Review of photochemical reaction constants of organic micropollutants required for UV advanced oxidation processes in water. Water Research, 46, 2815-2827. Wols, B. A., Harmsen, D. J., Beerendonk, E. F., & Hofman-Caris, C. H. (2014). Predicting pharmaceutical degradation by UV (LP)/H2O2 processes: A kinetic model. Chemical Engineering Journal, 255(1), 334-343. Zellner, R., Exner, M., & Herrmann, H. (1990). Absolute OH Quantum Yields in the Laser Photolysis of Nitrate, Nitrite, anb Dissolved H2O2 at 308 and 351 nm in the Temperature Range 278-353 K. Journal of Atmospheric Chemistry, 411-425. REFERENCES  80  Zoschke, K., Dietrich, N., Börnick, H., & Worch, E. (2012). UV-based advanced oxidation processes for the treatment of odour compounds: efficiency and by-product formation. Water Research, 46(16), 5365-5373.   81  APPENDICES  A. Corrections to the Absorbance Measurements Due to the passage of radiation through different media, reflection losses have to be taken into account before reporting the measured value of absorbance. The Fresnel relation for the fraction reflected was calculated as  𝑓 = (𝑛1 − 𝑛2𝑛1 + 𝑛2)2 (A.1) The refractive index values used are shown in Table A.1 Refractive index used in the calculation of the Fresnel relation Media Refractive Index Reference Fused-Quartz 1.57505 Daimon & Masumura, 2007 Nitrogen 1.0003515 Heraeus Water 1.458039 Griesmann & Burnett, 1999 According to Goldring et al., (1953), the true absorbance, A0, can be calculated from 𝐴0 = 𝐴′ − 0.434 ∙ (1 − 𝑇2) ∙ (𝑓1 ∙ 𝑓21 − 𝑓1)  (A.2) with,  𝐴0 = log (𝐼0𝐼)  (A.3) 𝑇 =𝐼𝐼0  (A.4)   82  B. Supplementary Data for the Absorption of Major Solutes at 185 nm The following figures demonstrate that the solutions of inorganic ions and SRNOM used in the experiments followed the Beer-Lambert Law  Figure B.1 Linearity in the absorbance of nitrate solutions at 3.6 °C  Figure B.2 Linearity in the absorbance of nitrate solutions at 17.5 °C   0.0000.2250.4500.6750.0500 0.1000 0.1500 0.2000 0.2500AbsorbanceConcentration (mM)2 mm 5 mm0.0000.2250.4500.6750.0500 0.1000 0.1500 0.2000 0.2500AbsorbanceConcentration (mM)2 mm 5 mmAPPENDIX B 83   Figure B.3 Linearity in the absorbance of sulphate solutions at 3.6 °C  Figure B.4 Linearity in the absorbance of sulphate solutions at 17.5 °C  0.0000.1000.2000.3001.1000 1.6000 2.1000 2.6000 3.1000AbsorbanceConcentration (mM)2 mm 5 mm0.0000.1000.2000.3001.1000 1.6000 2.1000 2.6000 3.1000AbsorbanceConcentration (mM)2 mm 5 mmAPPENDIX B 84   Figure B.5 Linearity in the absorbance of chloride solutions at 3.6 °C  Figure B.6 Linearity in the absorbance of chloride solutions at 17.5 °C       0.0000.1000.2000.3000.0500 0.1000 0.1500 0.2000 0.2500AbsorbanceConcentration (mM)2 mm 5 mm0.0000.1000.2000.3000.4000.0500 0.1000 0.1500 0.2000 0.2500AbsorbanceConcentration (mM)2 mm 5 mmAPPENDIX B 85    Figure B.7 Linearity in the absorbance of SRNOM solutions at 3.6 °C  Figure B.8 Linearity in the absorbance of SRNOM solutions at 17.5 °C    0.0000.1000.2000.1500 0.2000 0.2500 0.3000 0.3500 0.4000 0.4500 0.5000AbsorbanceConcentration (mM)2 mm 5 mm0.0000.0500.1000.1500.2000.2500.1500 0.2000 0.2500 0.3000 0.3500 0.4000 0.4500 0.5000AbsorbanceConcentration (mM)2 mm 5 mm 86  C. Supplementary Data for the Removal of Micropollutants Table C.1 Removal observed for each micropollutant at a flow rate of 2.7 L min-1 with different water matrices. Error represents the uncertainty in the removal Type Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound  % Removal Observed Drugs Carbamazepine 76.0 ± 6.0 42.4 ± 1.1 63.5 ± 1.2 53.1 ± 2.8 Diclofenac 86.4 ± 2.6 81.9 ± 4.4 78.7 ± 5.2 77.0 ± 0.8 Fluoxetine 87.3 ± 5.8 62.0 ± 4.6 70.8 ± 5.4 59.7 ± 5.2 Sulfamethoxazole 73.4 ± 2.8 74.6 ± 0.4 69.8 ± 1.1 63.6 ± 2.4 Stimulant Caffeine 61.8 ± 2.7 29.4 ± 0.5 48.9 ± 0.9 47.4 ± 2.0 Hormones Estradiol 78.2 ± 11.3 80.8 ± 5.0 54.1 ± 3.6 65.4 ± 2.8 Medroxyprogesterone 65.3 ± 7.1 6.1 32.3 ± 3.2 50.7 ± 11.5 Norethindrone 87.0 ± 0.4 32.3 ± 0.4 67.7 ± 1.0 63.7 ± 16.2 Herbicides Atrazine 42.5 ± 2.4 17.2 ± 0.6 27.5 ± 0.7 32.7 ± 1.4 Deethylatrazine 14.7 ± 0.2 3.0 ± 0.2 20.9 ± 0.5 18.7 ± 0.2 Linuron 76.3 ± 4.1 53.1 ± 2.5 62.3 ± 4.0 62.2 ±  Table C.2 Estimated Electrical Energy per Order for each micropollutant in several water matrices at 2.7 L min-1 and 20.5 °C. Error represents the calculated uncertainty for the EEO Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound Estimated EEO (kW h m-3 order-1) Atrazine 1.62 ± 0.14 4.78 ± 0.69 2.80 ± 0.18 2.27 ± 0.39 Caffeine 0.93 ± 0.04 2.58 ± 0.14 1.34 ± 0.05 1.40 ± 0.08 Carbamazepine 0.63 ± 0.02 1.63 ± 0.05 0.89 ± 0.04 1.19 ± 0.15 Deethylatrazine 5.64 ± 2.36 N/A 3.84 ± 1.37 4.34 ± 0.31 Diclofenac 0.45 ± 0.01 0.53 ± 0.02 0.58 ± 0.02 0.61 ± 0.05 Estradiol 0.59 ± 0.06 0.54 ± 0.04 1.15 ± 0.13 0.85 ± 0.02 Fluoxetine 0.44 ± 0.02 0.93 ± 0.05 0.73 ± 0.02 0.99 ± 0.22 Linuron 0.62 ± 0.03 1.19 ± 0.14 0.92 ± 0.05 0.92 ± 0.05 Medroxyprogesterone 0.85 ± 0.05 14.3 2.31 ± 0.91 1.27 ± 0.39 Norethindrone 0.44 ± 0.03 2.30 ± 0.23 0.80 ± 0.06 0.89 ± 0.16 Sulfamethoxazole 0.68 ± 0.03 0.66 ± 0.05 0.75 ± 0.04 0.89 ± 0.09    APPENDIX C 87  Table C.3 Estimated 1-log removal 254 nm based dose for each micropollutant in several water matrices at 20.5 °C. Error represents the calculated uncertainty for the 254 nm based dose Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound 1-log removal 254 nm based dose (mJ cm-2) Atrazine 606.1 ± 17.7 1354.7 ± 44.4 885.8 ± 30.3 959.6 ± 17.1 Caffeine 365.6 ± 5.9 822.5 ± 37.5 479.8 ± 33.7 442.9 ± 5.7 Carbamazepine 215.2 ± 3.1 500.7 ± 31.1 311.2 ± 12.6 343.7 ± 4.7 Deethylatrazine 1439.4 ± 16.4 7676.7 ± 87.9 2303 ± 20.3 2303 ± 24.9 Diclofenac 156.7 ± 8.3 126.5 ± 3.7 203.8 ± 18.6 205.6 ± 4.7 Estradiol 247.6 ± 35.3 215.2 ± 2.6 359.8 ± 9.0 258.8 ± 1.8 Fluoxetine 123.8 ±  2.0 258.8 ± 4.7 159.9 ± 3.9 253.1 ± 3.7 Linuron 225.8 ± 5.3 451.6 ± 34.7 324.4 ± 31.3 329.0 ± 7.5 Medroxyprogesterone 235.0  ± 5.5 697.9  ± 17.1 411.3 ± 15.0 315.5 ± 7.5 Norethindrone 151.5 426.5 ± 18.1 299.1 ±  5.1 348.9 Sulfamethoxazole 264.7 ± 8.2 343.7 ± 5.6 299.1 ± 8.4 371.5 ± 6.8  Table C.4 Estimated number of passes required to achieve 1-log removal for each micropollutant in several water matrices at 1.9 L min-1 and 20.5 °C. Error represents the calculated uncertainty for the 254 nm based dose Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound # of passes required to achieve 1-log removal Atrazine 3.70 ± 0.34 6.94 ± 0.56 4.59 ± 0.15 5.13 ± 0.74 Caffeine 1.86 ± 0.05 5.20 ± 0.48 2.60 ± 0.12 3.26 ± 0.17 Carbamazepine 1.35 ± 0.07 3.16 ± 0.20 1.80 ± 0.13 2.50 ± 0.12 Deethylatrazine 17.59 ± 13.16 25.42 ± 20.60 11.27 ± 2.43 8.16 ± 0.37 Diclofenac 0.93 ± 0.01 0.99 ± 0.04 1.10 ± 0.11 1.30 ± 0.04 Estradiol 1.16 1.04 ± 0.07 2.43 ± 0.33 2.05 ± 0.23 Fluoxetine 1.08 ± 0.19 1.45 ± 0.04 1.06 ± 0.30 1.99 ± 0.15 Linuron 1.38 ± 0.09 2.38 ± 0.30 1.66 ± 0.17 1.97 ± 0.10 Medroxyprogesterone 1.56 ± 0.25 11.0 ± 2.6 2.98 ± 0.47 3.15 ± 0.61 Norethindrone 1.10 ± 0.07 2.49 ± 0.30 1.61 ± 0.22 2.04 ± 0.15 Sulfamethoxazole 1.29 ± 0.05 1.10 ± 0.12 1.48 ± 0.09 2.12 ± 0.11    APPENDIX C 88  Table C.5 Estimated number of passes required to achieve 1-log removal for each micropollutant in several water matrices at 2.7 L min-1 and 20.5 °C. Error represents the calculated uncertainty for the 254 nm based dose Water Matrix SCFP SCFP + 50 ppm Carbonate SCFP + 2 ppm SRNOM SCFP + 25 ppm Chloride Compound # of passes required to achieve 1-log removal Atrazine 4.16 ± 0.36 12.24 ± 1.77 7.17 ± 0.45 5.81 ± 0.99 Caffeine 2.39 ± 0.09 6.61 ±  0.35 3.43 ± 0.13 3.58 ± 0.20 Carbamazepine 1.61 ± 0.06 4.17 ± 0.13 2.29 ± 0.09 3.04 ± 0.37 Deethylatrazine 14.45 ± 6.06 N/A 9.83 ± 3.51 11.12 ± 0.80 Diclofenac 1.16 ± 0.02 1.35 ± 0.06 1.49 ± 0.04 1.57 ± 0.13 Estradiol 1.51 ± 0.15 1.40 ± 0.10 2.96 ± 0.34 2.17 ± 0.06 Fluoxetine 1.12 ± 0.05 2.38 ± 0.12 1.87 ± 0.06 2.53 ± 0.56 Linuron 1.60 ± 0.08 3.04 ± 0.35 2.36 ± 0.13 2.37 ± 0.13 Medroxyprogesterone 2.18 ± 0.13 37 5.91 ± 2.34 3.26 ± 1.01 Norethindrone 1.13 ± 0.07 5.91 ± 0.60 2.04 ± 0.15 2.27 ± 0.40 Sulfamethoxazole 1.74 ± 0.07 1.68 ±   0.12 1.92 ± 0.10 2.28 ± 0.24    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0228628/manifest

Comment

Related Items