<%BANNER%>

Evaluaton of Seasonal Variations of Dissolved Organic Matter and Halide Concentrations in the St. Johns River, FL, USA

Permanent Link: http://ufdc.ufl.edu/UFE0042582/00001

Material Information

Title: Evaluaton of Seasonal Variations of Dissolved Organic Matter and Halide Concentrations in the St. Johns River, FL, USA Relationship to Ion Exchange Treatment Potential
Physical Description: 1 online resource (67 p.)
Language: english
Creator: Walker, Krystal
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: bromide, chloride, dissolved, exchange, ion, johns, magnetic, matter, organic, river, st
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: EVALUATON OF SEASONAL VARIATIONS OF DISSOLVED ORGANIC MATTER AND HALIDE CONCENTRATIONS IN THE ST. JOHNS RIVER, FL, USA: RELATIONSHIP TO ION EXCHANGE TREATMENT POTENTIAL This work has two overall objectives; the first to evaluate the seasonal variation in dissolved organic matter and halide (chloride and bromide) chemistry and concentrations, the second to evaluate the treatment potential of magnetic ion exchange resins. Both were accomplished through the ten month sampling of the St. Johns River, near Astor, FL. The concentrations of dissolved organic carbon varied from 12 to 26 mg/L, while the fluorescence index and specific ultraviolet absorbance both indicate the organic matter was of terrestrial origin. Halide chemistry also varied greatly through the year and are comparable to other large scale studies of bromide and chloride from different parts of the world. The data collected in this study have showed that the St. Johns River has high concentrations of dissolved organic matter, chloride and bromide throughout the year and could prove difficult to treat to drinking water standards. To evaluate a treatment potential for the St. Johns River, the anion exchange resin, MIEX was used in two different forms. The first form was the traditional chloride resin (MIEX-Cl), utilizing chloride as the mobile counter ion, and the other form, MIEX-HCO3 was regenerated to have bicarbonate as the mobile counter ion. For this work the removal of dissolved organic carbon, ultraviolet absorbance at 254 nm, bromide and sulfate removals were compared for both resins through the entire year. Both resins showed similar removal potentials for all but bromide and sulfate. By the final regeneration sulfate removal differences between the two resins was 33%, with the MIEX-Cl showing the best removal. MIEX-Cl resin differed greatly from MIEX-HCO3 in that the resin released bromide into the receiving waters. This phenomenon was further investigated with jar tests simulating halides in the St. Johns River. All resins were also imaged with scanning electron microscopy to describe both resins after fourteen regenerations and compare those to virgin resin. Overall, the work revealed MIEX-HCO3 has the potential to be used in water treatment, even under the variable conditions of a river like the St. Johns River.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Krystal Walker.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Boyer, Treavor H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042582:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042582/00001

Material Information

Title: Evaluaton of Seasonal Variations of Dissolved Organic Matter and Halide Concentrations in the St. Johns River, FL, USA Relationship to Ion Exchange Treatment Potential
Physical Description: 1 online resource (67 p.)
Language: english
Creator: Walker, Krystal
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: bromide, chloride, dissolved, exchange, ion, johns, magnetic, matter, organic, river, st
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: EVALUATON OF SEASONAL VARIATIONS OF DISSOLVED ORGANIC MATTER AND HALIDE CONCENTRATIONS IN THE ST. JOHNS RIVER, FL, USA: RELATIONSHIP TO ION EXCHANGE TREATMENT POTENTIAL This work has two overall objectives; the first to evaluate the seasonal variation in dissolved organic matter and halide (chloride and bromide) chemistry and concentrations, the second to evaluate the treatment potential of magnetic ion exchange resins. Both were accomplished through the ten month sampling of the St. Johns River, near Astor, FL. The concentrations of dissolved organic carbon varied from 12 to 26 mg/L, while the fluorescence index and specific ultraviolet absorbance both indicate the organic matter was of terrestrial origin. Halide chemistry also varied greatly through the year and are comparable to other large scale studies of bromide and chloride from different parts of the world. The data collected in this study have showed that the St. Johns River has high concentrations of dissolved organic matter, chloride and bromide throughout the year and could prove difficult to treat to drinking water standards. To evaluate a treatment potential for the St. Johns River, the anion exchange resin, MIEX was used in two different forms. The first form was the traditional chloride resin (MIEX-Cl), utilizing chloride as the mobile counter ion, and the other form, MIEX-HCO3 was regenerated to have bicarbonate as the mobile counter ion. For this work the removal of dissolved organic carbon, ultraviolet absorbance at 254 nm, bromide and sulfate removals were compared for both resins through the entire year. Both resins showed similar removal potentials for all but bromide and sulfate. By the final regeneration sulfate removal differences between the two resins was 33%, with the MIEX-Cl showing the best removal. MIEX-Cl resin differed greatly from MIEX-HCO3 in that the resin released bromide into the receiving waters. This phenomenon was further investigated with jar tests simulating halides in the St. Johns River. All resins were also imaged with scanning electron microscopy to describe both resins after fourteen regenerations and compare those to virgin resin. Overall, the work revealed MIEX-HCO3 has the potential to be used in water treatment, even under the variable conditions of a river like the St. Johns River.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Krystal Walker.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: Boyer, Treavor H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042582:00001


This item has the following downloads:


Full Text

PAGE 1

1 EVALUATON OF SEASONAL VARIATIONS OF DISSOLVED ORGANIC MATTER AND HALIDE CONCENTRATIONS IN THE ST. JOHNS RIVER, FL, USA: RELATIONSHIP TO ION EXCHANGE TREATMENT POTENTIAL By KRYSTAL WALKER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Krystal Walker The views expressed in this article are those o f the author and do not reflect the official policy or position of the United States Air Force, Department of Defense, or the U.S. Government

PAGE 3

3 To my daughter, Amber, thank you for always being there

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor, Dr. Treavor Boyer, for his guidance and oversight of my work I also wish to thank my committee, Dr. Joseph Delfino and Dr. Jean Claude Bonzongo, for their time and he lpful comments on this thesis. Also, f or their time and advice on the s canning electron microscope images, I would like to thank Dr. Kevin Powers and Dr. Gary Schi e ffele. I want to thank all of my colleagues in Dr. Boyer s g roup for all their help and support along with the St. Johns River Water Management District for the sample collection This work was supported in part by the UF Water Institute (Project: Scenario Based Analysis of Surface Water Withdrawals from the St. Johns River Basin). I also would like to thank the Unite d States Air Force for giving me the opportunity to pursue my goals Finally, I would like to thank most my daughter Amber for her constant support and understanding.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................7 LIST OF FIGURES .........................................................................................................................8 LIST OF ABBREVIATIONS ........................................................................................................10 ABSTRACT ...................................................................................................................................11 CHAPTER 1 INTRODUCTION ..................................................................................................................13 2 SEASONAL VARIATIONS IN THE CHEMISTRY OF HALIDES AND DISSOLVED ORGANIC MATTER IN THE ST. JOHNS RIVER, FL, USA .............................................14 2.1 St. Johns River Overview ..............................................................................................14 2.2 Materials and Methods ..................................................................................................16 2.2.1 Sampling Location ............................................................................................16 2.2.2 Sample Collection .............................................................................................16 2.2.3 Analytical Methods ...........................................................................................17 2.3 Results and Discussion ..................................................................................................18 2.3.1 Hydrology and General Water Quality .............................................................18 2.3.2 Chlorid e and Bromide .......................................................................................19 2.3.4 DOM .................................................................................................................21 2.4 St. Johns River Summary ..............................................................................................23 2.5 Supplementary Material ................................................................................................23 3 ROBUSTNESS OF MIEX TREATMENT: EFFECT OF SEASONAL CHANGES IN WATER QUALITY AND REGENERATION EFFICIENCY ..............................................30 3.1 Magnetic Ion Exchange Overview ................................................................................30 3.2 Materials and Methods ..................................................................................................32 3.2.1 Sampling Location ............................................................................................32 3.2.2 Materials ............................................................................................................32 3.2.3 SJR Jar Tests .....................................................................................................33 3.2.4 Resin Regeneration ...........................................................................................33 3.2.5 Bromide Release Jar Tests ................................................................................34 3.2.6 Analytical Methods ...........................................................................................34 3.3 Results and Discussion ..................................................................................................36 3.3.1 DOM Removal ..................................................................................................36 3.3.2 Inorganic Anion Removal .................................................................................37 3.3.3 MIEX Resin Characteristics ..............................................................................38

PAGE 6

6 3.3.4 Bromide Release Jar Test ..................................................................................40 3.4 Magnetic Ion Exchange Summary ................................................................................41 3.5 Supplementary Material ................................................................................................42 4 CONCLUSIONS AND RECOMMENDATIONS .................................................................49 APPENDIX A NITROGEN, SULFATE AND IODIDE ................................................................................51 A.1 Materials and Methods ..................................................................................................51 A.2 Results and Discussion ..................................................................................................52 B ADDITIONAL SEM IMAGES ..............................................................................................57 LIST OF REFERENCES ...............................................................................................................62 BIOGRAPHICAL SKETCH .........................................................................................................67

PAGE 7

7 LIST OF TABLES Table page 31 St. Johns River jar test regeneration numbers and corresponding dates that samples were collected from the river. ............................................................................................43 32 Synthetic waters used in bromide release jar tests ............................................................44

PAGE 8

8 LIST OF FIGURES Figure page 21 St. Johns River hydrology for As tor, FL site from October 2009 to August 2010. ............................................................................................................................................24 22 St. Johns River water quality for Astor, FL site collected from October 2009 to August 2010.. .....................................................................................................................25 23 Absolute concentrations in water of (a) chloride, reported in mg/L, and (b) bromide, reported in g/L, from St. Johns River Astor, FL site. ......................................................26 24 Ratio o f Cl-/Br(mol/mol) for St. Johns River at Astor, FL site. .......................................27 25 Box and whisker plot comparison of Cl-/Br(mol/mol) data across different spatial and temporal scales ............................................................................................................28 26 Absolute concentrations in St. Johns River waters of (a) dissolved organic carbon (DOC), (b) SUVA254 and (c) fluorescence index (FI), ......................................................29 31 Com parison of St. Johns River data for raw waters, MIEX Cl and MIEX HCO3 treated waters. Figures are (a) absolute DOC concentration, (b) percent DOC removal, (c) absolute UV254 and (d) percent UV254 removal.. ...........................................45 32 Comparison of St. Johns River data for raw waters, MIEX Cl and MIEX HCO3 treated waters. Figures are (a) absolute bromide concentration, (b) percent bromide removal, (c) absolute sulfate concentration and (d) percent sulfate removal. ...................46 33 Field emission scanning electron microscopy images of (a) 40 magnification of regenerated MIEX Cl resin beads (b) 250 magnification of surface of MIEX Cl beads (c) 1,000 magnificati on cross section of regenerated MIEX Cl bead ...................47 34 Concentration of (a) chloride, (b) bromide, and (c) sulfate from bromide release jar tests. ...................................................................................................................................48 A 1 Raw absolute concentrations in St. Johns River of (a) nitrate, measured in mg/L and (b) total dissolved nitrogen, measured in mg/L for Astor, FL site. .................................54 A 2 Percent o f total dissolved nitrogen present as nitrate in the St. Johns River over sampling period. .................................................................................................................55 A 3 Absolute concentrations of (a) sulfate and (b) iodide in the SJR for the sampling period. ................................................................................................................................56 B 1 Field emission scanning electron microscopy images of MIEX HCO3 and virgin MIEX Cl resin beads ..........................................................................................................58

PAGE 9

9 B 2 Field emission scanning electron microscopy images of regenerated MIEX Cl resin surfaces in greater detail. ...................................................................................................59 B 3 Field emission scanning electron microscopy images of regenerated MIEX HCO3 resin surfaces in greate r detail. ...........................................................................................60 B 4 Field emission scanning electron microscopy images of virgin MIEX Cl resin surfaces in greater detail.. ..................................................................................................61

PAGE 10

10 LIST OF ABBREVIATIONS BrBromide ClChloride DI Deionized DOC Dissolved organic carbon DOM Dissolved organic matter EEM Excitation emission matrix FI Fluorescence index FL Florida IIodide MIEX Magnetic ion exchange MIEX Cl Magnetic io n exchange in chloride form MIEX HCO3 Mag netic ion exchange in bicarbonate form NO3 Nitrate SEM Scanning electron microscope SJR St. Johns River SJRWMD St. Johns River Water Management District SO4 2 Sulfate SUVA254 Specific ultraviolet absorbance at 254 nm TDN T otal dissolved nitrogen TDS Total dissolved solids USGS United States Geological Survey UV254 Ultraviolet absorbance at 254 nm

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for the Degree of Master of Science EVALUATON OF SEASONAL VARIATIONS OF DISSOLVED ORGANIC MATTER AND HALIDE CONCENTRATIONS IN THE ST. JOHNS RIVER, FL, USA: RELATIONSHIP TO ION EXCHANGE TREATMENT POTENTIAL By Krystal Walker De cember 2010 Chair: Treavor Boyer Major: Environmental Engineering Sciences This work has two overall objectives; the first to evaluate the seasonal variation in dissolved organic matter and halide (chloride and bromide) chemistry and concentrations, the second to evaluate the treatment potential of magnetic ion exchange resins. Both were accomplished through the ten month sampling of the St Johns River, near Astor, FL The concentrations of dissolved organic carbon varied from 12 to 26 mg/L, while the fl uorescence index and specific ultraviolet absorbance both indicate the organic matter was of terrestrial origin. Halide chemistry also varied greatly through the year and are comparable to other large scale studies of bromide and chloride from different parts of the world. The data co llected in this study have showed that the St. Johns River has high concentrations of dissolved organic matter, chloride and bromide throughout the year and could prove difficult to treat to drinking water standards. To evaluate a treatment potential for the St Johns River, the anion exchange resin, MIEX was used in two different forms. The first form was the traditional chloride resin (MIEX Cl), utilizing chloride as the mobile counter ion, and the other form MIEX HC O3 was re generated to have bicarbonate as the mobile counter ion. For this work the removal of dissolved organic

PAGE 12

12 carbon, ultraviolet absorbance at 254 nm, bromide and sulfate removals were compared for both resin s through the entire year. Both resins showed similar removal potentials for all but bromide and sulfate. By the final regeneration sulfate removal differences between the two resins was 33%, with the MIEX Cl showing the best removal. MIEX Cl resin differed greatly from MIEX HCO3 in that the res in released bromide into the receiving waters. T his phenomenon was further investigated with jar tests simulating halides in the St. Johns River All resins were also imaged with scanning electron microscopy to describe both resins after fourteen regener ations and compare those to virgin resin. Overall, the work revealed MIEX HCO3 has the potential to be used in water treatment, even under the variable conditions of a river like the St. Johns River

PAGE 13

13 CHAPTER 1 INTRODUCTION The St. Johns River Water Management District (SJRWMD) is one of five water management districts in the state of Florida and encompasses 12,300 square miles in northeast to east central Florida ( SJRWMD 2006) The district is responsible for meeting the current and future ground and surface water demands (S JRWMD, 2010b) These demands have historically been met by groundwater sources from the surficial, intermediate and Floridan Aquifer s (SJRWMD, 2006) The SJRWMD is expecting t he population to increase to 5.9 million by 2020, which will mean an increase in water demand (SJRWMD, 2006) To meet these demands many alternatives have been explored, e.g., water reclamation water conservation, desal ination, and surface water withdrawals (SJRWMD, 2006) Five projects al ong the St. Johns River (SJR ) are being evaluated for drinking water supplies by the SJRWMD and two sites are currently removing water from the SJR (SJRWMD, 2006) It is currently estimated that a total of 262 million gallons per day could be withdrawn from the SJR to provide drinking water to the district (SJRWMD, 2006, 2010a) The SJR is unique, it originates in the east central Florida near Indian River County, FL and flows northward to Jacksonville, FL where it flow s into the Atlantic Ocean (SJRWMD, 2006) It is also a very wide and slow moving river, the width averaging 2 miles across and the elevation change from the headwaters to the mouth is ~30 ft (Thoburn, 2002) This slow movement and low gradient means that at periods of low flow at the northern end of the river can be influenced by tides from the ocean as far upstream as Lake George (SJRWMD, 2006) Although studies have been conducted on the SJR to understand the effects of withdrawals on the environment, little has been done to evaluat e the water quality parameters as they pertain to drinking water treatment.

PAGE 14

14 CHAPTER 2 SEASONAL VARIATIONS IN THE CHEMISTRY OF HALIDES AND DISSOLVE D ORGANIC MATTER IN THE ST. JOHNS RIVER, F L, USA 2.1. St. Johns River Overview Although there are previous s tudies conducted on the SJR, FL, USA to evaluate the effects of surface water withdrawals on ecological and biogeochemical outcomes (SJRWMD, 2010a) there is a gap in knowl edge pertaining to the chemistry of halides and dissolved organic matter (DOM). Moreover, new research findings on halides and DOM would complement existing studies on the river. In particular, it is not known how the concentration of bromide varies with respect to chloride. This is important because understanding the speciation of bromide relative to chloride is expected to provide new insights into the hydrology and geochemistry of the river (Magazinovic et al., 2004; Alcala and Custodio, 2008; D'Alessandro et al., 2008) In addition, bromide is a precursor to harmful disinfection byproducts that can be formed during water treatment (Hua et al., 2006) Although the concentration of DOM in the SJR is known, there is very little information on the chemistry of DOM. Understanding the chemistry of DOM is fundamental to understanding the numerous reactions that DOM takes part in, such as oxidation reduction, photochemistry and light attenuation, contaminant binding, and impact on engineered water treatment processes. Bromide and chloride are natural constituents found in surface waters at various levels (Magazinovic et al., 2004) According to Stumm and Morgan (1996) seawater ratios of chloride to bromide are approximately 300:1 (w/w) Several studies, at various locations around the world, have been conducted looking at the spatial spread of bromide and chloride levels in surface water s. These studies have determined a wide range of chloride to bromide (Cl-/Br-) ratios, depe nding upon the location (Magazinovic et al., 2004; Alcala and Custodio, 2008; D'Alessandro et al., 2008) The Cl-/Brratio has been used previously as a tracer to determine the

PAGE 15

15 origin and evolution of groundwater, surface water, and brines in oil and gas exploration (Alcala and Custodio, 2008) It can also be very important for understanding the ecology of a given water and revealing anthropogenic influences on that water. For example previous research has shown that waters from areas where Br based pesticides are used in agriculture have Cl-/Brratios of 275 718 (mol/mol), w hile leaching of septic wastes into water ha s a ratio of ~463 (mol/mol) ( Alcala and Custodio, 2008) Studying the changes in the DOM in the SJR is important to understanding the ecosystem of the river. This understanding is important because plant and soil derived DOM represents a significant part of terrestrial carbon budge ts and is a dominant linkage between terrestrial and aquatic ecosystems (Jaffe et al., 2008) Also, DOM is the largest pool of detrital organic carbon, in fact dissolved organic carbon (DOC) is the most important i ntermediate in the global carbon cycle (Battin et al., 2008) DOC concentrations in a body of water are not only a reflection of the nature of the watershed but are also influenced by seasonal variation (Leenheer and Croue, 2003) To date, no information is available on the correlation between bromide and chloride in the SJR. To add to the lack of knowledge regarding halide chemistry, iodide (presented in Appendix A) is rarely studied in natural waters and it is not known if there is a correlation between iodide and chloride. Many studi es have looked at the temporal and spatial characteristics of DOM t o understand the ecology of surface water(s) (Jaffe et al., 2008) Few studies have been dedicated to one location through an entire course of seas ons to understand the impact of seasonal changes on DOM. This study is aimed at filling some of these gaps in knowledge. The overall goal of this work is to study the concentration and speciation of halides and DOM in the SJR The specific objectives of this work are: (1) to evaluate the temporal trends in

PAGE 16

16 the ratios of Cl-/Br-; (2) to evaluate the temporal trends in the specific ultraviolet absorbance (SUVA) and fluorescence excitation emission matrix (EEM) spectra of DOM; and (3) to discuss the implicat ions of the chemistry of halides and DOM on natural and engineered processes. 2.2. Materials and Methods 2.2.1. Sampling Location All samples were collected from the SJR at Astor, FL, which is located 204 km upstream from the mouth of the river (US G eological S urvey (USGS) 2009) Sources of recharge for the river are groundwater, stormwater runoff, and rainfall. The sampling location was chosen because it is near proposed locations of surface water withdrawal (SJRWMD, 2010a) and has a USGS gage station (02236125) that monitors gage height, discharge, water temperature, dissolved oxygen, turbidity, and specific conductance. The location of the gage station is latitude 2910 The drainage area for the gage station is 8,352 km2 (SJRWMD, 2009) There are no wastewater treatment plants within 8 km radius of the sample location (SJRWMD, 2009) The land use for the drainage area is wetland (32%) agriculture (24%), forest (16%), urban (13%), and other (15%) (SJRWMD, 2009) The SJR is characterized by high salt content (median chloride is 235 mg/L) and contains a high organic matter content ( median total organic carbon is ~17 mg C/L ) (SJRWMD, 2009) 2.2.2. Sample Collection Samples were collected twice a month by the SJRWMD or author for a 10 month period starting in October 2009. Samples col lected by the author were taken from either the east or west bank, from a dock ~12 ft from the shore. Samples collected by the SJRWMD were collected from the side of a boat roughly in the middle of the river, by a kalawasi tube ~7 ft long and blended in a plasti c container before being dispensed into sample bottle s Samples were

PAGE 17

17 delivered to the Department of Environmental Engineering Sciences, University of Florida within 24 hr of collection and stored at 4C. Samples were analyzed within 2 weeks of receipt. 2.2.3. Analytical Methods All standards were prepared from ACS grade chemicals and deionized (DI) water. Dry materials were weighed on a Mettler AE 160 analytical balance. The pH of all samples was measured with an Accumet AP71 pH meter with a n Ag/AgCl pH/ ATC probe that was calibrated using pH 4, 7, and 10 buffer solutions. All subsequent analyses were performed on samples that were filtered through a 0.45 m nylon membrane filter (Millipore). All filters were pre rinsed with 500 mL of DI water and ~20 mL of sample. DOC was measured on a Shimadzu TOC VCPH total organic carbon analyzer utilizing combustion at 680 C for analysis, equipped with an ASI V autosampler. The stock standard for DOC was made from potassium hydrogen phth alate. Chloride and bromide were measured on a Dionex ICS 3000 ion chromatograph equipped wi th IonPac AG22 guard column, AS22 analytical column and ARS 4mm suppressor running 4.8 mM Na2CO3/1.0 mM NaHCO3 eluent with 100 L sample loop and a flow rate of 1.5 ml/min, following manufac turers recommendations. Temperature of the column and detector compartments was set to 35 C and the suppressor current was set at 40 mA Stock standards for chloride were prepared from sodium chloride and stock standards for bromide were prepared from s odium bro mide All DOC, chloride, and bromide measurements were made in duplicate and averaged. Each run on the TOC VCPH and I CS 3000 were monitored using calibration check standards (4, 40 and 400 mg/L for chloride; 0.025, 0.25 and 2.5 mg/L for bromide; and 1, 5, 10, 15 and 20 mg/L for DOC) to ensure that the measured concentration fell within 10% of the known value. Ultraviolet absorbance at 254 nm (UV254) was measured a Hitachi U 2900 spectrophotometer in a 1 cm quartz cuvette. Specific UV254 absorbanc e (SUVA254) was calculated by dividing UV254

PAGE 18

18 by DOC. Fluorescence excitation emission ( EEM ) spectra were collected on a Hitachi F 2500 fluorescence spectrophotometer using a 1 cm quartz cuvette. Samples were scanned at 5 nm increments over an excitation of 220400 nm and 2 nm increments over an emission of 250500 nm. For the measurement excitation slit was set at 5.0 nm, emission slit was set at 2.5 nm and scan speed was set at 1500 nm/min. The raw fluorescence EEMs were processed in MATLAB using an in house program (Comstock et al. 2010; Appel and Boyer, 2010) The fluorescence index (FI) is defined as the ratio of fluorescence intensity at emissions of 470 nm and 520 nm at an excitation of 370 nm (Cory and McKnight, 2005) and was calculated using the inhouse MATLAB program. All data presented in this thesis will be the average for the entire sampling peri od the standard deviation for the entire sampling period. 2.3. Results and Discussion 2.3.1. Hydrology and General Water Quality The hydrology and general water quality parameters of the river for the entire sampling period are shown in Figures 21a c an d 22 a d. Solid black lines represent the data collected by the USGS gage station (2009) except for pH which was measured in the laboratory, and the red line, when present, represents the 13 year median data reported by the SJRWMD (2009) The SJRWMD (2009) rates the Astor sit e as having good water quality. Over the course of the study period the gage height of the river has been variable, the highest point registering 2 ft and lowest point at 0.25 ft (Figure 2 1a). The river is also charac terized by a variable discharge rate, at times even flowing in reverse (Figure 21b). The specific conductance for the river has the lowest reading at 600 S/cm to the highest at ~1,200 S/cm, nearly double the lowest point (Figure 2 1c). In terms of tot al dissolved solids (TDS), this represents a range of 400800 mg/L using the estimator that TDS is 0.75 the specific conductance (Fishman and Friedman, 1989) for comparison brackish water has a TDS of ~1000 mg/L and seawater a TDS of ~3,500 mg/L

PAGE 19

19 (Hem, 1985) Interestingly, the period of highest conductance does not correlate with the periods of negative discharge. As stated in the Introduction, the SJR flows from south to north, where it meets the Atlantic Ocean. The negative discharge would suggest that water is flowing in reverse, and one would think this corresponds to a surge of water from the north (i.e., the Atlantic Ocean). The TDS for these periods of negative discharge do not support this theory, the TDS measures ~600 mg/L, which is not near brackish water, and so no saltwater is present in the surge. Figures 22 a d provide further insight int o the quality of the water at the sampling site. Dissolved oxygen (DO) (Figure 2 2a) content follows an inverse of winter/summer trend as temperature, with DO highest in the winter season and lowest in the summ er season. The pH of the water ( as measured in the laboratory) was on average 7.3 0.4, with the lowest reading at 6.3 and the highest at 8 (Figure 22b). The turbidity (Figure 2 2c) of the water ranged from 1 FNU to 8 FNU, which is below the level of biological impact threshold of 25 FNUs (Klein, 2008) Figure 22d shows th e sampling periods seasonal variation in the temperature of the water, wi th the coldest period starting in December 2009 through March 2010 and the warmest period starting in June 2010 and continuing through September 2010. This data follows the typical trend in winter/summer seasons for this portion of Florida. 2.3.2. Chlo ride and Bromide Figure 23a b represents the concentrations of chloride and bromide in the river for the entire sampling period. The xaxis is in months and spans the entire sampling period. The yaxis is measured in mg/L for chloride (Figure 2 3a) and g/L for bromide (Figure 2 3b). All data were generated from analysis of collected samples. Figure 23a shows the concentration of chloride in the river has changed greatly (nearly doubled from low to peak concentration) through the course of the year, generally increasing from October 2009 through March 2010 and

PAGE 20

20 then began to decrease until May 2010. The average chloride concentration was 195 35 mg/L. The specific conductance also increased during this time period (Figure 2 1c) and recall the corresp onding TDS range was 400800 mg/L. The bromide concentration has varied through the study period, again with the maximum concentration (>1,000 g/L) more than double the minimum concentration (~500 g/L) (Figure 23b) with an average of 768 175g/L. A s a note, von Gunten (2003) reports a typical bromide concentrations range of 2700 g/L in U.S. and European waters, while research by Kroening (2004) shows a historical range in the SJR of 190 17,000 g/L. It should also be noted that bromide concentrations over 100 g/L will form problematic concentrations of disinfection byproducts (DBPs) (von Gunten, 2003) Chlor ide and bromide peaks, while not exactly aligning, do seem to follow a similar trend. Other research in the SJR found disagreement with chloride and bromide concentrations (Kroening, 2004) and was able to better link bromide concentrations with negative discharge periods. Bromide concentrations are influenced by multiple factors including geology and anthropogenic influences (Magazinovic et al. 2004; Alcala and Custodio, 2008; D'Alessandro et al., 2008) Recent research has focused on understanding the relationships between chloride and bromide levels and relating those to their potentia l sources (Magazinovic et al. 2004; Alcala and Custodio, 2008; D'Alessandro et al., 2008) Data from this study was plotted and compared with this recent research to understand the influences at the sample site. T he Cl-/Brratio is used by many researchers to predict the concentration of bromide in a water given only the chloride and also as an indication of upstream activities in the river that will influence the biogeochemistry of that river (Alcala and Custodio, 2008) Figure 24 shows the Cl-/Brratio trend, on a molar basis, for this study averages 582 84. The ratio increase s from

PAGE 21

21 October 2009 to February 2010, which agrees with the increase in chloride concentration for the same time period (Figure 2 3a). Figure 25 compares studies from three different countries, with the results from this study. This study is less variab le than the previous work, owning mostly to this works focus on a single location, in contrast to the other work focused across a country/region (Magazinovic et al., 2004; Alcala and Custodio, 2008; D'Alessandro et al., 2008) It can also be seen that the results from the current study align more closely with the study in Spain (Alcala and Custodio, 2008) and is only slightly similar to the lower results from the other two countries (Magazinovic et al., 2004; D'Alessandro et al., 2008) Alt hough these studies were conducted over a period of 1 4 years depending on location, the data discussed were representative of that entire time period (Magazinovic et al., 2004; Alcala and Custodio, 2008; D'Alessandro et al., 2008) So, it is likely the difference seen when comparing data, are due to the differences in geology in these countries. 2.3.4. DOM Figure 26 a c represents the DOC, SUVA254, and FI for the entire year of the sampling period. The xaxis of both figures is represented in months, t he y axis is presented in different units; part (a) is measured in mg/L, part (b) in L/mg m and part (c) is unitless. All data were generated from collected samples analyzed in the laboratory. Total organic carbon (TOC) measurements are used to quantify organic matt er present in a body of water, and TOC can be split into two fractions : particulate organic carbon and DOC (Leenheer and Croue, 2003) Because DOC represents > 90% of TOC (Leenheer and Croue, 2003) it is often used as a surrogate for the concentration of organic matter (Croue et al., 2000) For the SJR the concentration of DOC (Figure 26a) changes through the sampling period, starting out at its highest point in October 2009 to reaching its lowest point in August 2010.

PAGE 22

22 Jaffe et al. (2008) collected data from a 10 different locations in North America and found a DOC range of 1 30 mg/L, with most surface waters being less than 6 mg/L. The SJR has an average concentration of 18.9 3.4 mg/L, well above the 6 mg/L but within the range of the Jaffe et. al (2008) study. To further the understanding of the of DOM SUVA is used to determine the quality of the DOM (Owen et al., 1993) For the sampling period the SUVA254 remained constant, at ~4 0.3 L/mg m (Figure 2 6b), which indicates a terrestrially derived organic matter (Croue et al., 2000; Leenheer and Croue, 2003) Fluorescence is also used to determine the quality of organic matter (Croue et al., 2000) A high FI indicates microbial organic matter, while a low FI indicates terrestrial organic matter (Croue et al., 2000; Cory and McKnight, 2005; Hur et al., 2007) The fluorescence method utilized in the laboratory is uncorrected, unlike those often reported in literature. To better understand the data presented, the FI was determined for known DOM (Suwannee River Natural Organic Matter Reverse Osmosis isolate from International Humic Substances Society and a natural spring water from the Ichentuckee Springs river run with DOM of microbial origin), and data presented in F igure 26c was scaled between these two points. For the study period FI (Figure 2 6c) was fairly constant at 1.90 0.03, which falls within the range of terrestrial organic as determined by the lab comparison study. The data generated from the SUVA254 a nd FI are in agreement, the organic matter present in the SJR through the course of the sampling period is of terrestrial origin. When considering a source for drinking water it is important to understand the concentration of organic matter present and al so the origin of the organic matter. This is important because terrestrial organic matter will form DBPs during water treatment (Hua and Reckhow, 2007; Sani et al., 2008)

PAGE 23

23 2.4. St. Johns River Summary The SJR is a dynamic system undergoing various changes in the chemistry of the river throughout the year. Chloride and bromide concentrations have both doubled in the year, going from 130270 mg/L and 5001,000 g/L respectively. The concentration of bromide in the river is high. Comparing the Cl-/Brratio coincides well with other reported data and provides an important understanding of comparing temporal and spatial variations. The DOC of the river has also undergone extreme changes, again more than doubling from 12 27 mg/L and is much higher than the typical range for surface waters. Finally, the two parameters used to determine the quality of the organic matter, SUVA254 and FI, have remained fairly constant through the year and in agreement that the organic ma tter is of terrestrial origin. Looking at all of this data together provides a detailed picture of the water treatment hurdles present in the SJR. The concentrations of DOC and bromide are both high and fluctuate considerably during the year. The organic matter is of terrestrial origin, as SUVA and FI indicate, and as such will form DBPs during water treatment. Finally, knowing that high bromide will lead to problematic DBPs, it is obvious that treatment of these waters to meet drinking water standar ds will be a challenge. 2.5. Supplementary Material Appendix A presents additional details into the chemistry of the SJR. The concentrations of nitrogen, sulfate, and iodide in the SJR during the sampling period are reported. All measurement techniques and data are included and discussions of each are presented in this section.

PAGE 24

24 Figure 21. St. Johns River hydrology for Astor, FL s ite from October 2009 to August 2010. Figures presented are (a) gage height, (b) discharge, and (c) specif ic conductance. Dashed line, when present, represents the 13 year me di an reported by the S t J ohns R iver W ater M anagement D istrict (2009) All data were collected from U S G eological S urvey online database (2009)

PAGE 25

25 Figure 22. S t J ohns R iver water quality for Astor, FL site collec ted from October 2009 to August 2010. Figures represent: (a) di ssolved oxygen (DO) (b) pH, (c) temperature, and (d) turbidity. Dashed line, when present, represents 13 year medi an (SJRWMD, 2009) All data, except for pH, were collected from U.S. Geo logical Survey online database (2009)

PAGE 26

26 Figure 23. Absolute concentrations in water of (a) chloride, reported in mg/L, and (b) bromide, reported in g/L, from S t J ohns R iver Astor, FL site. Error bars represent standard deviation for each sample. Data in parenthesis represents the average concentration and standard deviation for the entire year.

PAGE 27

27 Figure 24. Ratio of Cl-/Br(mol/mol) for S t J ohns R iver at Astor, FL site. The y axis is measured in (mol/mol) and the months of the study are on the on xaxis. Error bars represent the standard deviation for each sample. Data in parenthesis represents the average and standard deviation

PAGE 28

28 Figure 25. Box and whisker plot comparison of Cl-/Br(mol/mol) data across different s patial and temporal s cales (Magazinovic et al., 2004 (Australia); Alcala and Custodio, 2008 ( Spain ) ; D'Alessandro et al., 2008 ( Sicily ) ) The vertical line s represent the range of the data, horizontal lines represent the median of the data. The boxes represent the median of the upper and lower quartiles of the data.

PAGE 29

29 Figure 26. Absolute concentrations in S t J ohns R iver waters of (a) dissolved organic carbon (DOC) measured in mg/L, (b) SUVA254, presented in L/mg m and (c) fluorescence index (FI) unitless ratio of intensi ty at emission of 370 nm over inten s i ty at emission of 520 nm both at excitation of 370 nm Error bars not included for fluore scence index, represent standard deviation for each sample. Data in parenthesis represents the average and standard deviation for the entire year. SUVA254 is scaled between microbial organic matter (1) a nd terrestrial organic matter (5 ) and fluorescence index is scaled between terrestrial (1.8) and microbial (2.1) to better understand the data.

PAGE 30

30 CHAPTER 3 ROBUSTNESS OF MIEX TREATMENT: EFFECT OF SEASONAL CHANGES IN WATER QUALITY AND REGENERATION EFFICIENCY 3.1. Magnetic Ion Exchange Overview The SJR lo cated in FL, USA, has very high concentrations of DOM and bromide. The water quality in the SJR is important because the river is being considered as an alternative drinking water source, and is estimated to supply 262 million gallons per day by 2020 (SJRWMD, 2010a) It is well known that DOM affects water treatment processes and finished water quality. For example, DOM adds color, taste and odor to raw drinking water (Leenheer and Croue, 2003) ; forms DBPs during water treatment (Hua and Reckhow, 2007; Sani et al., 2008) ; exerts a demand on oxidation and coagulation chemicals (Sharp et al., 2006; Sani et al., 2008) ; fouls membranes (Lozier et al., 2008) ; and contributes to biological growth in distribution systems (Clement et al., 2003) Also during water treatment, bromide is oxidized to hypobromous acid by free chlorine and other oxidants, and then reacts with DOM to form brominated DBPs (Cowman and Singer, 1996; Liang and Singer, 2003; von Gunten, 2003; Hua and Reckhow, 2007; Richardson et al., 2008; Krasner, 2009; Hsu and Singer, 2010) Thus, advanced processes, such as ion exchange, nanofiltration, and reverse osmosis, will be required to effectively treat water from the SJR to produce high quality drinking water. Magnetic ion exchange (MIEX), using chlor ide as the mobile counter ion, has been shown in previous research to be an effective process for removing DOM and to a limited extent bromide (Tan et al., 2005; KabschKorbutowicz et al., 2008; Zhang et al., 2008; H su and Singer, 2010) and has been shown to be as effective as or more effective than activated carbon adsorption and enhanced coagulation (Singer and Bilyk, 2002; Humbert et al., 2008; Jarvis et al., 2008; Drikas e t al., 2009; Bond et al., 2010) A few studies have evaluated MIEX treatment as a function of source water quality and regeneration efficiency using pilot plant studies (Singer et

PAGE 31

31 al., 2007; Drikas et al., 2009) a nd through multiple regenerations in bench scale experiments (Mergen et al., 2008) Other studies have lo oked at spatially diverse waters and compared the MIEX removal capabilities (Singer and Bilyk, 2002; Kitis et al., 2007; Boyer and Singer, 2008) A major limitation of previous MIEX research is that its long term p erformance due to changing water quality and regeneration efficiency has not been systematically studied. A disadvantage of MIEX treatment, and anion exchange in general, is the production waste regenerant brine that is difficult to dispose One potentia l solution is to use a different, more benign mobile counter ion such as bicarbonate. Previous work by Holl and Kiehling (1981), Jelinek et al. (2004) and Matosic et al. (2000) have demonstrated the use of bicarbonate as a mobile counter ion in anion exchange reactions. Recent work by Rokicki and Boyer (2010) demonstrated that MIEX resin with the bicarbonate mobile counter ion (MIEX HCO3) performed very similar to traditional MIEX resin with chloride as th e mobile counter ion (MIEX Cl). No work has been done to evaluate the longterm performance of MIEX HCO3 resin due to changing water quality or to compare its long term performance with MIEX Cl resin. The goal of this work is to evaluate the robustness of MIEX treatment with respect to seasonal changes in water quality for a river high in DOM and bromide. The specific objectives of this work are: (1) to quantify the differences between MIEX Cl and MIEX HCO3 resins for DOM removal; (2) to quantify the dif ferences between MIEX Cl and MIEX HCO3 resins for bromide removal; (3) to assess the viability of regenerating MIEX Cl and MIEX HCO3 resins multiple times; (4) to examine the physical structure of MIEX resin; and (5) to study the performance of MIEX resin under conditions of rapidly changing water quality. This work is also expected to provide insights into water quality and treatment as a result of climate change.

PAGE 32

32 3.2. Materials and Methods 3.2.1. Sampling L ocation All samples were collected from the SJR at Astor, FL, which is located 204 km upstream from the mouth of the river (USGS, 2009) Sources of recharge for the river are groundwater, stormwater runoff, and rainfall. The sampling location was chosen because it is near proposed locations of surface water withdrawal (SJRWMD, 2010a) and has a USGS gage station (02236125) that monitors gage height, discharge, water temperature, dissolved oxygen, turbidity and specific conductance. The location of the gage station is latitude 2910 8131 The drainage area for the gage station is 8,352 km2 (SJRWMD, 2009) There are no wastewater treatment plants within 8 km radius of the sample location (SJRWMD, 2009) Land use for the chosen sites drain age area is wetland (32%), agriculture (24%), forest (16%), urban (13%), and other (15%) (SJRWMD, 2009) The SJRWMD (2009) identifies the river as having a high salt content (median chloride concentration of 235 mg/L) and a high DOM content (median total organic carbon concentration of 17 mg C/L). Samples were collected twice a month by the SJRWMD or the author for a 12 month period starting in Oct ober 2009. Samples collected by the author were taken from either the east or west bank, from a dock ~12ft from the shore line. Samples collected by the SJRWMD were collected from the side of a boat at roughly the center of the river, by a kalawasi tube ~7ft long and blended in a plastic container before being dispensed into sample bottles. Samples were delivered to the Department of Environmental Engineering Sciences, University of Florida within 24 hr of collection and stored at 4C. Samples were analyzed with in 2 weeks of receipt. 3.2.2. Materials MIEX resin from Orica Watercare was used in this work. Two mobile counter ions: bicarbonate (HCO3 -) and chloride (Cl-) were evaluated based on previous research by Rokicki

PAGE 33

33 and Boyer (2010) MIEX HCO3 resin was originally produced from virgin MIEX Cl resin, and both MIEX Cl and MIEX HCO3 resins were regenerated after every experiment, as described in Section 3.2.4. 3.2.3. SJR Jar T ests All experiments were conducted at ambient laboratory temperature and open to the atmos phere. Jar tests were performed using a Phipps & Bird PB 700 jar tester with 2 L jars. One liter of sample water was used in all jar tests. A MIEX resin dose of 10 mL was measured as the volume of wet settled resin in a graduated cylinder. The jar tests were conducted at 100 rpm for 20 min and the resin was allowed to settle for at least 5 min before samples were decanted from the jar (Boyer and Singer, 2005) MIEX resin was then r insed two times with DI water as outlined below, covered and stored until regeneration. All jar tests were performed in duplicate and the data shown are the average of duplicate samples. 3.2.4. Res in R egeneration MIEX resin was regenerated to alter the mobile counter ion available for exchange on the resin. All resins were regenerated with 1 L of salt solution (sodium chloride or sodium bicarbonate), which contained the desired mobile counter ion. The regeneration concentration was based upon the equivalent capacity of MIEX Cl resin, which was previously determined to be 0.52 meq/mL (Boyer and Singer, 2008) Previous experiments also showed that a regeneration solution with 10 times the equivalent capacity of the resin dose would be sufficient to f ully regenerate the resin (Rokicki and Boyer, 2010) The MIEX resin was mixed in the regeneration solution using a Phipps & Bird PB 700 jar tester for 20 min to allow for exchange of ions. The resin was allowed to settle for 5 min before t he supernatant liquid was decanted. To ensure that there was no remaining regeneration solution, the resin was washed with DI water by adding 1 L of DI water, mixing for 10 min, and settling for 5 min before decanting the

PAGE 34

34 supernatant liquid. This washing was repeated a second time before the resins were covered and stored in DI water. The SJR jar test, resin regeneration and washing of the resin was repeated for every sample collection from October 2009 through August 2010, leading to 17 regenerations of MIEX Cl and MIEX HCO3 resin. Table 31 lists the dates that SJR water was collected and the corresponding regeneration number. 3.2.5. Bromide R elea se Jar T ests A set of jar tests were conducted to determine if MIEX resin could be retaining a removed ani on and then releasing it later in the presence of a more preferable anion. To begin, a large batch of virgin MIEX Cl resin was converted to MIEX HCO3 resin. One jar was used for the entire experiment, which contained 10 mL of resin. For each cycle, 1 L of synthetic water was added containing a pre determined concentration of anions, as shown in Table 32. The composition of the synthetic water was based on halide and sulfate levels in the SJR. After each jar test, 1 mL of partially exhausted MIEX HCO3 resin was removed and 1 mL of fresh MIEX HCO3 resin was added. This was to simulate how a full scale MIEX treatment plant is operated, with a small portion of the resin being removed and replaced with freshly regenerated resin (Kitis et al., 2007; Hsu and Singer, 2009; Neale and Schafer, 2009) The concentration of anions in the synthetic water was altered to simulate periods of low and high concentrations of sulfate. 3.2.6. Analytical M ethods All standards w ere prepared from ACS grade chemicals and DI water. Dry materials were weighed on a Mettler AE 160 analytical balance. All samples were vacuum filtered with a 0.45 m nylon membrane (Millipore) to remove any MIEX resin particles and to ensure only dissol ved species were analyzed. Filtered samples were analyzed for DOC, ultraviolent absorbance at 254 nm (UV254) chloride, bromide, and sulfate. DOC was measured on a

PAGE 35

35 Shimadzu TOC VCPH total organic carbon analyzer utilizing combustion at 680C for analysis equipped with an ASI V autosampler. The stock standard for DOC was made from potassium hydrogen phthalate Chloride, sulfate and bromide were measured on a Dionex ICS 3000 ion chromatograph equipped wi th IonPac AG22 guard column, AS22 analytical colum n and ARS 4mm suppressor running 4.8 mM Na2CO3/1.0 mM NaHCO3 eluent with 100 L sample loop and a flow rate of 1.5 ml/min, following manufacturers recommendations. Temperature of the column and detector compartments was set to 35 C and the suppressor cur rent was set at 40 mA. Chloride stock standards were made from sodium chloride, sulfate stock standards were made from potassium sulfate, and bromide stock standards were made from sodium bromide. All DOC, chloride, sulfate and bromide measurements w ere made in duplicate and averaged. Each run on the TOC VCPH and ICS 3000 were monitored using calibration check standards to ensure that the measured concentration fell within 10% of the known value. UV254 was measured a Hitachi U 2900 spectrophotometer in a 1 cm quartz cuvette. Data presented in this paper represents the average for the entire sampling period the standard deviation for the entire sampling period. Virgin MIEX Cl, regenerated MIEX Cl, and regenerated MEX HCO3 resins were characterize d using a field emission scanning electron microscope (SEM) (JEOL JSM 6335F). The regenerated MIEX Cl and MIEX HCO3 resins were sampled after 14 regeneration cycles. Each MIEX resin was prepared for microscopy by lyophilizing ~0. 1 mL of wet resin to prevent any distortion of the resin during drying. The resin samples were then mounted to the pedestal with carbon tape and analyzed. Resin surface morphologies were characterized without coating using an accelerating voltage of 5.0 kV and a ~15 mm working distance in order to minimize

PAGE 36

36 charging, enhance surface features, and observe the resin at magnifications ranging from 40 to 5,000. 3.3. Results and Discussion 3.3.1. DOM R emoval Figures 31a and c show the DOC concentration and UV254 absorbance, respectiv ely, for raw and MIEX treated (i.e., MIEX Cl and MIEX HCO3) water. The DOC concentration of the SJR was at its maximum value (26 mg/L) at the beginning of the study (October 2009), and varied between 15 21 mg/L for a majority of the study. The corresponding DOC concentrations for MIEX Cl and MIEX HCO3 treated waters maintained a concentration of 4.5 0.7 mg/L and 5.1 0.7 mg/L, respectively, for the duration of the study, despite the changes in the DOC concentration of the SJR. When evaluating the over all removal of DOC (see Figure 3 1b), both MIEX Cl and MIEX HCO3 resins begin with ~90% removal and continue to have similar removal efficiencies with an average efficiency of 75% and 72%, respectively. Mergen et al. (2008) compared multiple MIEX Cl resin single use removal studies (i.e., regeneration 0 in this study), and reported a DOC removal range of 4093%. One difference is that the DOC range of the raw waters in the Mergen et al. (2008) work was 1.65.1 mg/L, well below the concentration of this study. The UV254 absorbance of the SJR water and MIEX treated waters (Figure 3 1c and d) followed a nearly identical pattern as DOC (Figure 3 1a and b). The UV254 concentration is reduced to ~0.1 0.2 1/cm by both resins, which equates to a ~85% removal. Other studies have reported a UV254 removal range of 80 95% (Hum bert et al., 2007; Kitis et al., 2007) For both MIEX Cl and MIEX HCO3 treated waters, the DOC concentrations and UV254 absorbance behaved similarly. This illustrates that bicarbonate form anion exchange has the same DOC and UV254 removal efficiency as chloride form anion exchange, and regeneration with sodium bicarbonate is able to achieve the same regeneration efficiency as regeneration with sodium

PAGE 37

37 chloride. Also, Figure 31 illustrates the robustness of the MIEX process in terms of fluctuating DOC an d UV254 (Figure 3 1a and c). Both resins show great efficiency at removing the aromatic organic carbon content, as seen by the reduction of UV254 absorbance, but for DOC there still remains 5 mg/L. This remaining DOC is likely a combination of the nonar omatic DOC, which is not favorably removed by MIEX resin (Ates et al., 2007) and/or competition by other anions for exchange sites (Kitis et al., 2007; Neale and Schafer, 2009; Hsu and Singer, 2010) 3.3.2. Inorganic Anion R emoval The bromide and sulfate concentrations in the SJR vary greatly though the year (see Figure 32a and c). The bromide concentration starts at ~600 g/L and increases through the sampling period with occasional fluctuations (April May 2010), with an average concentration of 768 175 g/L. The bromide concentration re aches its highest value in August 2010. Sulfate also starts the sampling period at its lowest concentration, ~20 mg/L, and increases to nearly 120 mg/L in April 2010, then begins to decline and has an average of 55 24 mg/L. Figures 32b and d, show that both MIEX Cl and MIEX HCO3 resins have the highest removals of bromide and sulfate at regeneration 0. Bromide removals for MIEX Cl and MIEX HCO3 range from 10 50% and 25 60%, respectively, with an average removal of 30 28% and 37 11%, respectively, f or each resin Other researchers evaluating bromide removal by MIEX Cl resin reported 10 90% bromide removal (DOC range 025 mg/L and initial concentration bromide range 10300 g/L) depending upon the MIEX resin dose and additional anions in the water (Singer and Bilyk, 2002; Neale and Schafer, 2009; Hsu and Singer, 2010) Results from this work, for both virgin and regenerated MIEX Cl and MIEX HCO3 resins, fall within the published data. Note that MIEX Cl resin experiences periods of bromide release (see Figure 3 2b), which are shown by the negative removal periods. This is likely due to the resin not being fully regenerated and releasing bromide in the presence of more favorable anions, such as sulfate

PAGE 38

38 or DOM (Kitis et al., 2007; Neale and Schafer, 2009) This idea is further explored in the bromide release experiments discussed in Section 3.3.4. Sulfate is always removed by both forms of MIEX resin, between 6096% for M IEX Cl and 4095% for MIEX HCO3 depending on the number of regenerations. Kitis et al. (2007) reported 924% sulfate removal, with a similar sulfate concentration to this study and low DOC (2.64.3 mg/L) This work showed a much higher removal, with both MIEX Cl and MIEX HCO3, with the average removal for each being77 11% and 65 17%, respectively. Unlike DOC and UV254 performance, the MIEX HCO3 appears to not be as effective as MIEX Cl at removing sulfate over numerous regenerations, with a 33% difference between the two resins at the end of the sampling term. This phenomenon could explain why the MIEX Cl resin releases bromide (Figure 3 2b), i.e., the MIEX HCO3 resin does not have as high a preference for sulfate as the MIEX Cl resin. Another possibility is that the MIEX Cl is performing better at sulfate removal because it is releasing excess bromide to take on sulfate, the preferred ion (Neale and Schafer, 2009) Comparing DOC and UV254 removal (Figure 31b and d) with the removal efficiency of bromide and sulfate (Figure 3 2b and d), ther e is an obvious disparity, the resin efficiency decreases over time for both inorganic anions while it remains constant for DOM. This could be caused by the resin retaining DOM, sulfate, and/or bromide during regeneration, thus limiting the number of avai lable removal sites for subsequent jar tests. It could also be a preference for DOM over the inorganic anions when the resins encounter high concentrations of each. 3.3.3. MIEX Resin C haracteristics MIEX resin is a polyacrylate macroporous polymer with strong base quaternary ammonium functional groups for ion exchange and magnetic iron oxide incorporated into the resin structure to aid in settling. There is very little published information on the physical characteristics of MIEX resin. For example, N eale et al. (2010) shows an SEM image of MIEX

PAGE 39

39 resin, however, the history of the resin or how it was prepared are not clear. Furthermore, over numerous regeneration cycles in this study, it was noted that the regenerated MIEX Cl and MIEX HCO3 resins used in the SJR jar tests began to behave differently. Regenerated MIEX HCO3 resins began to agglomerate and clump to the bottom of the jar during mixing and the efficiency of sulfate removal by regenerated MIEX HCO3 resin began to decrease. Therefore, SEM im ages of regenerated MIEX Cl and MIEX HCO3 resins were compared with a sample of virgin MIEX Cl resin to provide new insights into the physical structure of MIEX resin (see Figure 33 and Figure B 1a e in appendix B). Figure 3 3a shows multiple regenerated MIEX Cl resin beads in the field of view and illustrates that even with numerous regenerations there does not appear to be any mechanical attrition of the resin beads. Comparing all three resins at 40 magnification, (Figure 3 3a, Figure B 1a and c) all appear to have the same range of bead diameters, <10 m to over 200 m, but the virgin MIEX Cl resin (Figure B 1c) has a larger portion of fine particles than the other two. Figure 33b is a magnified image of regenerated MIEX Cl resin. Comparing Figure 33b with Figures B 1b and d in Appendix B shows that the regenerated MIEX HCO3 resin is slightly different from the virgin and regenerated MIEX Cl resins. In general, the surfaces of the regenerated MIEX HCO3 resin have both a greater number and more var iability in surface features than do the regenerated and virgin MIEX Cl resins. The regenerated and virgin MIEX Cl resins are more similar in surface appearance, with a much smoother look. Figure 3 3c shows a crosssection of a regenerated MIEX Cl resin bead and a crosssection of the virgin MIEX Cl resin is also shown in Figure B 1e in Appendix B. Looking at both images it is obvious that the internal structures of the resin beads are very complex and allow for interactions not only on the surface but w ithin the resin. The darker features are assumed to be pores, as the structure is known to be porous, but could be a result of how the bead

PAGE 40

40 was sheared. Porebetween the two resin s. 3.3.4. Bromide Release Jar T est MIEX Cl resin released bromide at different points during the SJR jar tests (see discussion in Section 3.3.2). Ishii and Boyer (2010) saw similar results in a full scale MIEX Cl water treatment plant, where sulfate was released into treated waters when the raw water supply was switched from a well with a high sulfate concentration to a well with a low sulfate concentration A series of jar tests were conducted to investigate the possible cause for bromide release as illustrated in Figures 3 4a c. During regenerations 110, MIEX HCO3 resin removed chloride and bromide from the synthetic water, and removal plateaued at 12% a nd 20%, respectively. To understand this consider that MIEX resin has a limited potential for removal (i.e. 0.52 meq/ m L of charge sites ; Boyer and Singer, 2008) Takin g this and considering only chloride, for simplicity 10 mL of MIEX HCO3 has the potential to remove 182 mg/L of chloride. T he synthetic water is dosed with 200 mg/L of chloride, so at first the resin is capable of removing nearly all the chloride. In each cycles after the initial, only 1 mL of exhausted resin is removed and replace d by 1 mL fresh resin (capable of removing ~18 mg/L of chloride). As the experiment continues the resin becomes saturated with chloride more easily and is unable to take on any more chloride. In the actual experiment synthetic waters also contained other competing anions (bromide and sulfate) During regenerations 11 15, 50 mg/L of sulfate was added to the synthetic water and the removal of bromide and chloride decreased as sulfate is the more preferred anion (Neale and Schafer, 2009) In fact, the resin released chloride (regeneration 11) and bromide (regenerations 12 15). The bromide is the least favorable anion, as is seen in Figu re 3 4b, and resin releases nearly 1.5 times the initial concentration at regeneration 14. This provides strong evidence for the potential of MIEX resin to sequester

PAGE 41

41 anions and then exchange the anions into receiving waters when a more preferred anion is present. To further prove this point, when the synthetic water was returned to its initial condition the residual sulfate is released through the last five regenerations (Figure 3 4c). At the same time, chloride exchange (Figure 34a) reaches a plateau a t the same level as seen in regenerations 510. The bromide concentration for the final five regenerations, (Figure 34b) has a low removal and is even released into the treated waters again at regeneration 19, possibly due to bromide being held on the re sin and being released again in the presence of sulfate. This shows that the performance of MIEX resin depends not only on the concentrations of anions but more importantly which anions are present, as MIEX has a greater affinity for some anions (sulfate) over others (bromide). 3.4. Magnetic Ion Exchange Summary MIEX Cl and MIEX HCO3 resins have similar removal efficiency for DOC and UV254, but removal is affected by different types of organic matter and competition by inorganic anions. Th is study has shown that both MIEX Cl and MIEX HCO3 resins are capable of removing ~70% of the DOC in the SJR over variable concentrations, but consistently 30% of the DOC remains. Inorganic anions behave differently from DOM when bromide and sulfate removals are considered. In the case of bromide, overall the removal rates were similar ~30%, but the MIEX Cl resin occasionally released bromide into the water, while the MIEX HCO3 resin consistently removed it. W hen sulfate removal is considered, MIEX Cl re sin removed more sulfate than MIEX HCO3 resin, with a 33% difference between the two resins at the end of the experiment. Overall the MIEX HCO3 and MIEX Cl resins have a similar removal performance, which proves MIEX HCO3 resin could be a viable candidate to replace traditional MIEX Cl resin for anion removal, but considerations would have to be made depending on the makeup of the water. Analysis of the regenerated MIEX Cl and MIEX HCO3 resins, along with virgin MIEX Cl resin,

PAGE 42

42 show differences in the surf aces of the resins. These differences could be related to the regeneration efficiency of the resins, but further work would be required to prove this. Bromide release jar tests performed with MIEX HCO3 resin prove the resins preference for certain anions, such as sulfate, and also prove the resin is capable of sequestering certain anions and then releasing them later into receiving waters, as was evident with the b romide release. Future work on MIEX HCO3 resins should be focused on furthering the knowl edge of the resin performance, what different anions MIEX HCO3 is capable of removing, and if there is a breaking point where the resin can no longer be effectively regenerated. Comparisons of many different types and sources of waters should be analyzed for removal efficiency of MIEX HCO3. Alternative pathways for regenerating MIEX HCO3 resin, such as CO2, should be evaluated in combination with resin performance studies. Further work is also needed on understanding the structure of MIEX resin A study should be undertaken to visualize the surface changes of the resin as it goes through multiple regenerations, focus should be on characterizing the surface features. The porosity of MIEX resin should also be analyzed to determine its pore size pore volume and total surface area. This information would help better explain how the resin is capable of sequestering anions during regeneration and could be applicable to all MIEX resins. 3.5. Supplementary Material Appendix B contains additional SEM images o f virgin MIEX Cl and regenerated MIEX HCO3 resins. Images were produced under the same parameters as presented in section 3.2.6 and further details are provided in the figure captions.

PAGE 43

43 Table 3 1. St. Johns River jar test regeneration numbers and corresponding dates that samples were collected from the river. Regeneration 0 represents fresh resin and the start of the sampling period. Regeneration number Date water collected 0 10/21/09 1 11/4/09 2 11/20/09 3 12/16/09 4 1/7/10 5 2/17/10 6 2/24/10 7 3/17/10 8 4/4/10 9 4/27/10 10 5/10/10 11 6/1/10 12 6/14/10 13 6/28/10 14 7/8/10 15 7/21/10 16 8/6/10 17 8/22/10

PAGE 44

44 Table 3 2. Synthetic waters used in bromide release jar tests. For regenerations 1 10 and 1520 only chloride (200 mg/L ) and bromide (1000 g/L) were added to synthetic waters. Sulfate (50 mg/L) was added for regenerations 1115 only. Regeneration numbers Cl (mg/L) Br (g/L) SO 4 2 (mg/L) 1 10 200 1000 0 11 15 200 1000 50 16 20 200 1000 0

PAGE 45

45 Fig ure 31. Comparison of St. Johns River data for raw waters, MIEX Cl and MIEX HCO3 treated waters at resin dose of 10 mL/ L and contact time of 20 min. Figures are (a) absolute DOC concentration, (b) percent DOC removal, (c) absolute UV254 and (d) pe rcent UV254 removal. Data presented in parenthesis are average, one standard deviation.

PAGE 46

46 Figure 32. Comparison of St. Johns River data for raw waters, MIEX Cl and MIEX HCO3 treated waters at resin dose of 10 mL/L and contact ti me of 20 min. Fig ures are (a) absolute bromide concentration, (b) percent bromide removal, (c) absolute sulfate concentration and (d) percent sulfate removal. Data presented in parenthesis are average, one standard deviation.

PAGE 47

47 Fig ure 33. Field emission scan ning electron microscopy images of (a) 40 magnification of regenerated MIEX Cl resin beads (b) 250 magnification of surface of MIEX Cl beads (c) 1,000 magnification cross section of regenerated MIEX Cl bead collected after 14th regeneration. Samples we re lyophilized and characterized without coating at an accelerating voltage of 5.0kV and ~15mm working distance. a b c

PAGE 48

48 Figure 34. Concentration of (a) chloride (b) bromide and (c) sulfate f r om bromide release jar tests. Resin dose was 10 mL/ L and contact time of 20 min. For each regeneration 1 mL of used MIEX HCO3 resin was removed and replaced with 1 mL of fresh MIEX HCO3 resin. Stripped bar represents the concentration added into each batch of synthetic water. Black bars represent regeneration s with only chloride (200 mg/L) and bromide (1000 g/L). White bars represent the regenerations that contain sulfate (50 mg/L) in addition to chloride (200 mg/L) and bromide (1000 g/L).

PAGE 49

49 CHAPTER 4 CONCLUSIONS AND RECOMMENDATIONS With the dwindling ground water supplies and an ever increasing population, many water management districts in Florida are looking to other sources for drinking water. One district, the SJRWMD, has proposed the SJR as a source to supply their future needs. This study has shown th at the SJR will be a complex water for treatment, with its high concentration of terrestrial DOM and bromide. The fact that these two factors together will lead to DBP problems means that special efforts should be made early in the process to remove both. This study also demonstrated the potential of using anion exchange (MIEX Cl or MIEX HCO3 resin) to remove bromide and DOM from the SJR. Future work for this research would be a longer term study that better explains the sources of DOM, chloride and bro mide in the river. The ability to track these origins would allow water treatment plant operators to better predict when treatment methods need adjustment, rather than reacting to the changes. The Br-/Clratio for the river should be further developed a nd studied with data from multiple sites along the river. This information would provide better predictability and identification of the sources of bromide and chloride in the SJR. Long term studies should also be combined with MIEX HCO3 anion exchange j ar tests and pilot studies. Research evaluating MIEX HCO3 resin needs to focus on understanding the failure point of the resin, when does it no longer regenerate effectively, and also study the use of alternative regenerants such as CO2, which could be a cost savings. More focus is needed on characterizing the surface features of the MIEX resin. Analysis should be conducted on the resin from virgin resin through multiple regenerations to understand the changes occurring on the surface. Additional work, such as porosimetry, should be done to understand pore volume, size and total

PAGE 50

50 surface area. All this combined could be used to explain the anion sequestering discussed in this paper.

PAGE 51

51 APPENDIX A NITROGEN SULFATE AND IODIDE A.1 Materials and M e thods T otal dissolved nitrogen ( T D N) was measured on a Shimadzu TOC VCPH total organic carbon analyzer equipped with a TNM 1 Total Nitrogen Module and ASI V autosampler. Nitrogen stock standards were prepared from potassium nitrate. Nitrate and sulfate w ere measured on a Dionex ICS 3000 ion chromatograph under the same conditions mentioned in previous sections (Chapters 2 & 3) Nitrate stock standards were made from sodium nitrate and sulfate standards were made from potassium sulfate. T D N, nitrate and sulfate measurements were made in duplicate and averaged. Each run on the TOC VCPH and I CS 3000 were monitored using calibration check standards (0.1, 0.5, 1, 1.5 and 2 mg N /L for TDN, 0.05, 0.5 and 5 mg N /L for nitrate, 0.5, 5 and 50 mg/L for sulfate) to ensure that the measured concentration fell within 10% of the known value. Due to iodide concentrations below the detection limit of the ion chromatography protocol, Standard Method 4500IC. Catalytic Reduction Method was used to determine iodide lev els ( Clesceri et al., 1989 ) The exact details of the analysis are outlined below Regents for iodide measurements were prepared every three months using DI water, except for ferrous ammonium sulfate which was prepared daily. All standards were prepared according to Standard Method 4500IC except for arsenious acid. For arsenious acid, a step wise addition of concentrated sodium hydroxide was required to dissolve the solid and then concentrated sulfuric acid was added to neutralize the base. 11 N sulfuric acid was used for all additions of acid. The Standard Methods 1030 E. (Clesceri et al., 1989) method detection limit was performed with DI and nanopure wate r to determine a detection limit of 2 g/L. A calibration curve and check points were created using the appropriate amount of standard iodide

PAGE 52

52 solution and DI water to make up 10 mL at given points (2, 5, 10, and 12 g/L). 10 mL of sample, calibration cur ve or calibration check solutions were placed in individual glass test tu bes. Initially 1 mL sodium chloride 0.5 mL arsenious acid and 0.5 mL sulfuric acid were added to each tube. Samples were then capped and placed in a water bath set to 30 C 0.5C, along with the ceric ammonium sulfate. Both were allowed to equilibrate for 1015 min. After equilibration, caps were removed and 1 mL ceric ammonium sulfate solution was added to each tube, inverted at least one time timed for 15 min and returned to t he water bath. At the 15 min point ( 0.1 min) 1 mL of ferrous ammonium sulfate was added and shaken. Immediately 1 mL of potassium thiocyanate was added to the tubes and inverted at least one time Tubes were returned to the water bath. When all tubes were completed samples were analyzed on a Hitachi U 2900 spectrophotometer for ultraviolet absorbance at 525 nm and 510 nm Analysis was completed within 1 hr as directed by Standard Method 4500IC (Clesceri et al., 1989) A.2 Results and D iscussion To better understand the ecology of the SJR TDN and nitrate concentrations were tracked through the study period (Figure A 1 a and b). The xaxis spans the entire year of the study, the yaxis is in mg N/L. All data were generated from analysis of collected samples. Total nitrate concentration for the year averages 0.15 0.09 mg/L, with a minimum below the detection limit of 0.05 mg/L and a maximum > 0.28 mg/L (Figure A 1a), while TDN averages 1.0 0.2 mg/L, with a minimum of 0.7 mg/L and a maximum of 1.4 mg/L (Figure A 1b) T he peak nitrate concentrations occur in November 2009 an d May 2010, which does fall within typical planting periods in the area downstream of the sample site (Stephens et al., 1991) To further understand the nitrogen present in the site waters, the nitrate/ TDN was plotted f or the sampling period (Figure A 2). The percentage of nitrate varies through the year, ranging

PAGE 53

53 from less than 5% to over 20%. On average nitrate represents 14 6% of the TDN concentration. Iodide and sulfate are other inorganic anions present in waters. Their concentrations are plotted in Figure A 3a and b. The xaxis on Figure A 3a is measured in mg/L and in Figure A 3b it is measured in g/L. The y axis for both figures represents the months of the study period. Sulfate concentrations were at the lowest at the start of the study, October 2009, at 24 mg/L (Figure A 3a) and the highest peaks are in February 2010 and March 2010, at >100mg/L. The average sulfate concentration in the SJR was 54.7 23.6 mg/L. Comparing the trend of concentrations for sulfate (Figure A 3a) with those for chloride (Figure 2 3a) they are very similar, with peak concentrations at the same time period. Iodide concent rations were much more variable ranging f rom 4 15 g/L, with an average of 8.8 3.2g/L. As with chloride and bromide concentrations, iodide and sulfate maximum concentrations are at least twice the minimum concentration (Figures A 3 and 2 3). Overall data shows concentrations ranges for iodide and sulfate fluctuate throughout the sampling period.

PAGE 54

54 Figure A 1. Raw absolute concentrations in S t. Johns R iver of (a) nitrate measured in mg/L and (b) total dissolved nitrogen measured in mg/L for Ast or, FL site Data in parenthesis represent average and standard deviation respectively. Error bars represent standard deviation for each sample. Nitrate measurements at 0.05 mg/L were set to the lower detection limit for experiment, actual concentratio n was measured at lower concentration.

PAGE 55

55 Figure A 2. Percent of total dissolved nitrogen present as nitrate in the S t. J ohns R iver over sampling period. Data in parenthesis are the average and standard deviation, respectively for the entire sampling period. Error bars represent standard deviation for each sample.

PAGE 56

56 Figure A 3. Absolute concentrations of (a) sulfate and (b) iodide in the SJR for the sampli ng period. Data in parenthesis represents the average and mean concentrations for the year respectively. Error bars represent the standard deviation for each sample

PAGE 57

57 APPENDIX B ADDITIONAL SEM IMAGE S SEM images were taken of MIEX Cl and MIEX HCO3 resins to better understand the physical changes noted in section 3.3.3 Images were captured under the same conditions mentioned i n section 3.3.3. Supplemental f igure B 1a e show the images captured and discussed in section 3.3.3. Also included in this supplemental section are additional images (supplemental figures B 24) taken of the regenerat ed MIEX Cl, regenerated MIEX HCO3 and virgin MIEX Cl resins. These images were focused more on the surfaces of the various bead and their exact descriptions are included in the figure captions.

PAGE 58

58 Figure B 1. Field emission scanning el ectron microscopy images of (a) 40 magnification of regenerated MIEX HCO3 resin beads (b) 250 magnification surface view of regenerated MIEX HCO3 beads collected after 14th regeneration. Also included are the following images of virgin MIEX Cl resin for comparison; (c) 40 magnification of virgin MIEX Cl resin beads (d) 250 magnification surface view of virgin MIEX Cl beads and (e) 1,000 magnification of cross section of virgin MIEX Cl bead c a b d e

PAGE 59

59 Figure B 2. Field emission scanning electron m icroscopy images of regenerated MIEX Cl resin surfaces in greater detail. (a) 1,000 magnification of regenerated MIEX Cl resin surface features (b) 5,000 magnification of regenerated MIEX Cl resin surface features and (c) 15,000 magnification of regene rated MIEX Cl resin surface features. a b c

PAGE 60

60 Figure B 3. Field emission scanning electron microscopy images of regenerated MIEX HCO3 resin surfaces in greater detail. (a) 40 magnification of regenerated MIEX HCO3 resin beads (b) 500 magnification of regenerated MIEX HCO3 resin bead (c) 1,000 magnification of regenerated MIEX HCO3 resin surface features (d) 5,000 magnification of regenerated MIEX HCO3 resin surface features and (e) 20,000 magnification of regenerated MIEX HCO3 resi n pore structure d c e b a

PAGE 61

61 Figure B 4. Field emission scanning electron microscopy images of virgin MIEX Cl resin surfaces in greater detail. (a) 1,000 magnification of virgin MIEX Cl resin surface features (b) 5,000 magnification of virgin MIEX Cl resin surface features and (c) 20,000 magnification of virgin MIEX Cl resin pore structure a b c

PAGE 62

62 LIST OF REFERENCES Alcala, F.J., Custodio, E., 2008. Atmospheric chloride deposition in continental Spain. Hydrological P rocesses 22, 3636 3650. Apell, J .N. B oyer,T .H ., 2010. Combined ion exchange treatment for removal of dissolved organic matter and hardness. Water R esearch 44, 24192430. Ates, N ., K itis, M ., Y etis, U ., 2007. Formation of chlorination byproducts. In waters with low SUVAcorrelations with S UVA and differential UV spectroscopy. Water R esearch 41, 41394148. Battin, T .J ., K aplan, L .A ., F indlay, S ., H opkinson, C .S ., M arti, E ., P ackman, A .I ., N ewbold, J .D ., S abater, F ., 2008. Biophysical controls on organic carbon fluxes in fluvial networks. Nat ure G eoscience 1, 95 100. Bond, T ., G oslan, E .H ., P arsons, S A ., J efferson, B ., 2010. Disinfection byproduct formation of natural organic matter surrogates and treatment by coagulation, MIEX ( R ) and nanofiltration. Water R esearch 44, 16451653. Boyer, T .H ., S inger, P .C ., 2005. Benchscale testing of a magnetic ion exchange resin. f or removal of disinfection byproduct precursors. Water R esearch 39, 12651276. Boyer, T .H ., S inger, P .C ., 2008. Stoichiometry of removal of natural organic matter by ion exchange. Environmental S cience & T echnology 42, 608 613. Clement, J., S pencer, C., C apuzzi, A .J ., C amper, A ., V an A ndel, K ., S andvig, A ., 2003. Influence of D istribution S ystem I nfrastructure on B acterial R egrowth: H igh Q uality W ater. A WWA R esearch F oundation and A merican W ater W orks A ssociation, D enver, CO. Clesceri, L S ., G reenberg, A .E ., T russell, R .R .., 1989. Standard M ethods for the E xamination of W ater and Wa stewater, 17th E dition. American P ublic H ealth Association, W ashington, D .C Comstock, S .E .H ., B oyer T .H ., G raf, K .C ., T ownsend, T .G., 2010. E ffect of landfill characteristics on leachate organic matter properties and coagulation treatability. Chemosphere 81, 976983. Cory, R .M ., M c K night, D .M ., 2005. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environmental S cience & T echnology 39, 8142 8149. Cowman, G A ., S inger, P .C ., 1996. Effect of bromide ion on haloacetic acid speciation resulting from chlorination and chloramination of aquat ic humic substances. Environmental S cience & T echnology 30, 1624.

PAGE 63

63 Croue, J P ., K orshin, G .V ., B enjamin, M ., 2000. Characterization of natural oranic m atter in drinking w ater: M onitoring and a nalysis, AWWA R esearch F oundation and A merican W ater W orks A sso ciation, D enver, CO. D'alessandro, W ., B ellomo, S ., P arello, F ., B rusca, L ., L ongo, M ., 2008. Survey on fluoride, bromide and chloride contents in public drinking water supplies in S icily (I taly). Environmental M onitoring and A ssessment 145, 303313. Drikas M ., D ixon, M ., M orran, J ., 2009. Removal of MIB and geosmin using granular activated carbon with and without MIEX pre treatment. Water R esearch 43, 51515159. Fishman, M J ., F riedman, L .C ., 1989. Methods for D etermination of I norganic S ubstances in W ater and F luvial S ediments. Techniques of W ater R esources I nvestigations of the U nited S tates G eological S urvey. U.S Geological S urvey, D enver, CO. Hem, J .D ., 1985. Study and I nterpretation of the C hemical C haracteristics of N atural W ater: W ater S upply P aper 2254. U.S Geological S urvey, D enver, CO. Hsu, S ., S inger, P .C ., 2009. Application of anion exchange to control nom interference on lime softening. Journal A merican W ater W orks A ssociation 101, 8594. Hsu, S ., S inger, P .C ., 2010. Removal of bromide and nat ural organic matter by anion exchange. Water R esearch 44, 2133 2140. Hua, G .H ., R eckhow, D .A ., 2007. Comparison of disinfection byproduct formation from chlorine and alternative disinfectants. Water R esearch 41, 16671678. Hua, G .H .,R reckhow, D .A ., K im, J ., 2006. Effect of bromide and iodide ions on the formation and speciation of disinfection byproducts during chlorination. Environmental S cience & T echnology 40, 3050 3056. Humbert, H .,G allard, H ., J acquemet, V ., C roue, J .P ., 2007. Combination of coagulation and ion exchange for the reduction of UF fouling properties of a high DOC content surface water. Water R esearch 41, 3803 3811. Humbert, H ., G allard, H ., S uty, H ., C roue, J .P ., 2008. Natural organic matter ( NOM ) and pesticides removal using a combination of ion exchange resin and powdered activated carbon ( PAC ). Water R esearch 42, 16351643. Hur, J ., J ung, N .C ., S hin, J .K ., 2007. Spectroscopic distribution of dissolved organic matter in a dam reservoir impacted by turbid storm runoff. Environmental M onitor ing and A ssessment 133, 5367. Ishii, S .K .L ., B oyer, T .H ., 2010. Evaluating the secondary effects of magnetic ion exchange: F ocus on corrosion potential in the distribution system. Desalination U nder R eview. Jaffe, R ., M c K night, D ., M aie, N ., C ory, R ., M c D owell, W H ., C ampbell, J .L ., 2008. Spatial and temporal variations in DOM composition in ecosystems: T he importance of longterm

PAGE 64

64 monitoring of optical properties. Journal of G eophysical R esearch B iogeosciences 113, 115. Jarvis, P ., J efferson, B ., D ixon, D ., P arsons, S .A ., 2008. Treatment options and their effect on NOM coagulant floc structures. Journal A merican W ater W orks A ssociation 100, 6473. Jelinek, L ., P arschova, H ., M atejka, Z ., P aidar, M ., B ouzek, K ., 2004. A combination of ion exchange and elect rochemical reduction for nitrate removal from drinking water P art I : N itrate removal using a selective anion exchanger in the bicarbonate form with reuse of t he regenerant solution. Water E nvironment R esearch 76, 26862690. Kabsch K orbutowicz, M ., M ajewska N owak, K ., W innicki, T ., 2008. Water treatment using MIEX ( R ) DOC/ultrafiltration process. Desalination 221, 338344. Kitis, M .,H arman, B .I ., Y igit, N .O ., B eyhan,M ., N guyen, H ., A dams, B ., 2007. The removal of natural organic matter from selected T urkis h source waters using magnetic ion exchange resin ( MIEX ( R )). Reactive & F unctional P olymers 67, 14951504. Klein, R .D ., 2008. Timber harvest and t urbidity in N orth c oastal C alifornia w atersheds. The T hird I nteragency C onference on R esearch in the W atersheds, E stes P ark, CO, pp. 201212. Krasner, S W ., 2009. The formation and control of emerging disinfection byproducts of health concern. Philosophical T ransactions of the R oyal S ociety A M athematical P hysical and E ngineering S ciences 367, 40774095. Kroening, S .E ., 2004. Streamflow and W ater Q uality C haracteristics at S elected S ites of the S t. Johns R iver in C entral F lorida, 19332002. In: I nterior, U .S .D .O .T ., U .S G .S. (eds.). U. S Geological S urvey, D enver, CO. Leenheer, J ., C roue, J ., 2003. Characterizing aquatic dissolved organic matter. Environmental S cience & T echnology, 18a 26a. Liang, L ., S inger, P .C ., 2003. Factors influencing the formation and relative distribution of haloacetic acids and trihalomethanes in drinking water. Environmental S cience & T e chnology 37, 2920 2928. Lozier, J ., C appucci, L ., A my, G ., L ee, N ., J acangelo, J ., H uang, H ., Y oung, T ., M ysore, C ., E meraux, C ., C louet, J ., C roue, J. P ., H eijmann, B ., 2008. Natural O ranic M atter F ouling of L ow P ressure M embrane S ystems: H igh Q uality W at er. A WWA R esearch F oundation and A merican W ater W orks A ssociation, D enver, CO. Magazinovic, R .S ., N icholson, B .C ., M ulcahy, D E ., D avey, D .E ., 2004. Bromide levels in natural waters: I ts relationship to levels of both chloride and total dissolved solids an d the implications for water treatment. Chemosphere 57, 329335.

PAGE 65

65 Matosic, M ., M ijatovic, I ., H odzic, E ., 2000. Nitrate removal from drinking water using ion exchange comparison of chloride and bicarbonate form of the resins. Chemical and B iochemical E ngi neering Q uarterly 14, 141 146. Mergen, M R .D ., J efferson, B ., P arsons, S .A ., J arvis, P ., 2008. Magnetic ionexchange resin treatment: I mpact of water type and resin use. Water R esearch 42, 19771988. Neale, P .A ., S chafer, A .I ., 2009. Magnetic ion exchange: I s there potential for international development? Desalination 248, 160168. Owen, D .M ., A my, G L ., C howdhury, Z.K ., 1993. Characterization of natural organic matter and it's relationship to treatability: M onitoring and analysis. A WWA R esearch F oundation and A merican W ater W orks A ssociation, D enver, CO. Richardson, S .D ., F asano, F ., E llington, J .J ., C rumley, F .G ., B uettner, K .M ., E vans, J .J ., B lount, B .C ., S ilva, L .K ., W aite, T .J ., L uther, G W ., M c K ague, A .B ., M iltner, R J ., W agner, E .D ., P lewa, M .J ., 2008. Occurrence and mammalian cell toxicity of iodinated disinfection byproducts in drinking water. Environmental S cience & T echnology 42, 83308338. Rokicki, C .A ., B oyer, T .H ., 2010. Bicarbonate form anion exchange: A ffinity, regeneration, and stoichiometry. Water R esearch, doi:10.1016/j.waters.2010.10018. Sani, B ., B asile, E ., R ossi, L ., L ubello, C ., 2008. Effects of pre treatment with magnetic ion exchange resins on coagulation/flocculation process. Water S cience and T echnology 57, 5764. Sharp, E .L ., P arso n, S .A ., J efferson, B ., 2006. Coagulation of NOM : L inking character to treatment. Water S cience and T echnology 53, 6776. Singer, P .C ., B ilyk, K ., 2002. Enhanced coagulation using a magnetic ion exchange resin. Water R esearch 36, 40094022. Singer, P .C ., S chneider, M ., E dwards B randt, J ., B udd, G .C ., 2007. M IEX for removal of DBP precursors: P ilotplant findings. Journal A merican W ater W orks A ssociation 99, 128139. S JRWMD 2006. St. Johns R iver W ater M anagement D istrict: D istrict W ater S upply P lan 2005, sj 20062. St. Johns R iver W ater M anagement D istrict, P alatka, FL. SJRWMD 2009. Online tools, GIS, data: 20010002 ( S t. Johns R iver). St. Johns R iver W ater M anagement D istrict. Available from < http://www.sjrwmd.com/watershedfacts/factPages/20010002.html >. S JRWMD 2010a. Water S upply: F requently A sked Q uestions ( FAQ). St. Johns R iver W ater M anagement District. Available from < http://www.sjrwmd.com/surfacewaterwithdrawals/FAQs.html >.

PAGE 66

66 S JRWMD 2010b. Water S upply: T he S t. Johns R iver as a drinking water source. St. Johns R iver W ater M anagement D istrict. Available from < http://www.sjrwmd.com/surfacewaterwithdrawals/ >. Stephens, J M ., B rown, S .P ., T readwell, D ., W ebb, S ., G evens, A ., D unn, R .A ., K idder, G ., S hort, D ., S imone, G W ., 1991. Florida V egetable G ardening G uide. Stumm, W ., M organ, J .J ., 1996. Aquat ic C hemistry, 3rd E dition. John W iley & S ons, I nc., N ew Y ork, NY. U S G eological S urvey, 2009. U SGS S urfaceW ater D aily D ata for F lorida: USGS 02236125 S t. Johns R iver at A stor, FL. U. S. Geological S urvey. Available from < http://waterdata.usgs.gov/nwis/dv?referred_module=sw&site_no=02236125 > Tan, Y .R ., K ilduff, J .E ., K itis, M ., K aranfil, T ., 2005. Dissolved organic matter removal and disinfection byproduct formation contr ol using ion exchange. Desalination 176, 189200. Thoburn, B ., 2002. St. Johns A merican H eritage R iver: S tate of the R iver. Available from < http://water.epa.gov/type/watersheds/named/heritage/upload/2003_07_07_heritage_sor_ sor_st_johns.pdf > von G unten,U u., 2003. Ozonation of drinking water: part II Disinfection and byproduct formation in presence of bromide, iodide or chlorine. Water R esearch 37, 14691487. Zhang, R ., V igneswaran, S ., N go, H ., N guyen, H ., 2008. Fluidized bed magnetic ion exchange ( MIEX ( R )) as pre treatment process for a submerged membrane reactor in wastewater treatment and reuse. Desalination 227, 8593.

PAGE 67

67 BIOGRAPHICAL SKETCH Krystal Walker graduated from Eastern New Mexico Univers ity, Portales NM in July 2000, with a BS in biology and c hemistry. In April 2002, she joined the United States Air Force, commissioning as a Second Lieutenant in June 2002. Since then she has served the United States Air Force as a chemist in a variety of locations and job positions In 2009, she was selected by the Air Force Institute of Technology to pursue her masters degree at the University of Florida.