<%BANNER%>

Sustainable Phosphorus Removal from Surface Water

Permanent Link: http://ufdc.ufl.edu/UFE0041717/00001

Material Information

Title: Sustainable Phosphorus Removal from Surface Water
Physical Description: 1 online resource (88 p.)
Language: english
Creator: Persaud, Amar
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adsorption, algae, alum, area, bed, bench, byproduct, chemical, column, composition, continuous, eutrophic, eutrophication, exchange, flow, fluidized, industrial, ion, jar, jesup, lake, layne, metal, orthophosphate, phosphate, phosphorus, phosx, physical, pilot, precipitate, precipitation, prevention, property, regeneration, resin, scale, sludge, sulfate, surface, sustainable, test, total, treatment, tributary, water
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphorus (P) is often a limiting nutrient in plant productivity. An excess of its bioavailable form, orthophosphate (OP), will cause eutrophication. Natural surface water systems may become overloaded with OP, therefore, it is pertinent to develop a strategy of remediation. Adsorption through upward flowing column has proven to be an effective method of removing OP from surface water. In this research it was found that the use of industrial by-products, also known as low-cost materials (LCMs), as compared to commercially available materials, such as polymeric adsorbents, in the treatment method can become feasible but performance will vary due to changing composition and prepared particle sizes. LCMs tested were, alum sludge, ferric sludge, steel slag, iron slag, recycled concrete, fly ash, and limestone of which alum sludge performed best. It was found that the advantages of using commercially available materials are ease of implementation, consistency of performance, and reusability. Commercially available materials tested were PhosX, MIEX, and Dowex22 of which PhosX performed best. Both alum sludge and PhosX are effective at removing OP from water by adsorption under continuous flow. Implementing fluidized beds to replicate bench-scale column treatment on a larger scale which utilize P-selective materials has proven to remove OP effectively. Bench-scale studies have helped to successfully predict the removal of OP by the pilot-scale fluidized bed with typically > 70% OP removal by both. Materials such as PhosX, a polymeric adsorbent resin, can be regenerated and was found to work as effectively as its virgin version. However, when using a reusable commercially available material, such as PhosX, the by-product of regeneration is a contaminated brine solution. The P in solution can be removed by precipitation and the cleaned brine solution can be used to regenerate another batch of exhausted resin. The technique of considering all aspects of a treatment process, while keeping sustainability in mind, has proven to be useful in developing a highly effective P treatment system.
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 Amar Persaud.
Thesis: Thesis (M.E.)--University of Florida, 2010.
Local: Adviser: Boyer, Treavor H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041717:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041717/00001

Material Information

Title: Sustainable Phosphorus Removal from Surface Water
Physical Description: 1 online resource (88 p.)
Language: english
Creator: Persaud, Amar
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adsorption, algae, alum, area, bed, bench, byproduct, chemical, column, composition, continuous, eutrophic, eutrophication, exchange, flow, fluidized, industrial, ion, jar, jesup, lake, layne, metal, orthophosphate, phosphate, phosphorus, phosx, physical, pilot, precipitate, precipitation, prevention, property, regeneration, resin, scale, sludge, sulfate, surface, sustainable, test, total, treatment, tributary, water
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, M.E.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphorus (P) is often a limiting nutrient in plant productivity. An excess of its bioavailable form, orthophosphate (OP), will cause eutrophication. Natural surface water systems may become overloaded with OP, therefore, it is pertinent to develop a strategy of remediation. Adsorption through upward flowing column has proven to be an effective method of removing OP from surface water. In this research it was found that the use of industrial by-products, also known as low-cost materials (LCMs), as compared to commercially available materials, such as polymeric adsorbents, in the treatment method can become feasible but performance will vary due to changing composition and prepared particle sizes. LCMs tested were, alum sludge, ferric sludge, steel slag, iron slag, recycled concrete, fly ash, and limestone of which alum sludge performed best. It was found that the advantages of using commercially available materials are ease of implementation, consistency of performance, and reusability. Commercially available materials tested were PhosX, MIEX, and Dowex22 of which PhosX performed best. Both alum sludge and PhosX are effective at removing OP from water by adsorption under continuous flow. Implementing fluidized beds to replicate bench-scale column treatment on a larger scale which utilize P-selective materials has proven to remove OP effectively. Bench-scale studies have helped to successfully predict the removal of OP by the pilot-scale fluidized bed with typically > 70% OP removal by both. Materials such as PhosX, a polymeric adsorbent resin, can be regenerated and was found to work as effectively as its virgin version. However, when using a reusable commercially available material, such as PhosX, the by-product of regeneration is a contaminated brine solution. The P in solution can be removed by precipitation and the cleaned brine solution can be used to regenerate another batch of exhausted resin. The technique of considering all aspects of a treatment process, while keeping sustainability in mind, has proven to be useful in developing a highly effective P treatment system.
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 Amar Persaud.
Thesis: Thesis (M.E.)--University of Florida, 2010.
Local: Adviser: Boyer, Treavor H.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041717:00001


This item has the following downloads:


Full Text

PAGE 1

1 SUSTAINABLE PHOSPHORUS REMOVAL FROM SURFACE WATER By AMAR PREM PERSAUD A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ENGINEER ING UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Amar Prem Persaud

PAGE 3

3 To the world

PAGE 4

4 ACKNOWLEDGMENTS First and foremost I thank God for blessing me with the ca pability to achieve great things. I thank my family for ensuring that I live an honest life, and the love of my life, Anasuya Bulkan, for her encouraging words and reminding me of visions of success and a blissful life Also, during my stay at the University of Florida I crossed paths with a great friend, German Calvo who proved to me that there is still hope for the new generation of people and hence a sound future for the world. I express my gratitude to Dr. Sherry Brant Williams for providing funding through SJRWMD Contract # 25104, Lake Jesup Total Phosphorus Removal Treatment Technologies Floating Island Pilot Project. Also, I thank Dr. Mark Brown and his research group, Sam Arden and Hugo Sindelar IV, for working alongside of me to launch a pilot scale study. Mo st importantly I thank Dr. Treavor Boyer for entrusting me with such a large and significant project and his mentorship and the assistance of a promising student, Pedro Palomino which all in the end, contributed to overcoming all obstacles

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 LIST OF ABBREVIATIONS ........................................................................................... 11 ABSTRACT ................................................................................................................... 13 CHAPTER 1 INTRODUCTION .................................................................................................... 15 2 LITERATURE REVIEW .......................................................................................... 20 3 MATERIALS AND METHODS ................................................................................ 24 Treatment Materials ................................................................................................ 24 Surface Water ......................................................................................................... 25 Jar Testing .............................................................................................................. 26 Continuous Flow Studies ........................................................................................ 27 Bench Scale Experiments ................................................................................ 27 Pilot Scale Experiments ................................................................................... 29 Sustainability Studies .............................................................................................. 30 Regeneration Procedure for PhosX .................................................................. 30 Precipitation Procedure .................................................................................... 30 Reuse of Dirty Wash Solution ........................................................................... 31 Analytical Methods .................................................................................................. 32 4 RESULTS AND DISCUSSION ............................................................................... 37 Low Cost Materials ................................................................................................. 37 Bulk Screening for P Selective Materials ................................................................ 38 Surface Waters ................................................................................................. 38 Jar Testing ........................................................................................................ 40 Performance of P Selective Materials ..................................................................... 43 Surface Waters ................................................................................................. 43 Continuously Flowing Treatment ...................................................................... 45 Cyclically Flowing Treatment ............................................................................ 46 Pilot Scale Studies ........................................................................................... 47 Effect of Algal Blooms ...................................................................................... 48 Sustainability Studies .............................................................................................. 49

PAGE 6

6 Regeneration .................................................................................................... 50 Precipitati on ...................................................................................................... 50 Reuse of Regeneration Solution ....................................................................... 51 5 CONCLUSION ........................................................................................................ 69 Evaluati on of Materials ............................................................................................ 69 Low Cost Materials ........................................................................................... 69 Commercially Available Resins ........................................................................ 70 Pilot Application ...................................................................................................... 70 Sustainability of Process ......................................................................................... 71 Recommendations for Further Research ................................................................ 72 APPENDIX A LOW COST MATERIAL PROPERTIES .................................................................. 73 B COLUMN TESTING RESULTS .............................................................................. 78 C FLOATING ISLAND TREATMENT SYSTEM RESULTS ........................................ 80 LIST OF REFERENCES ............................................................................................... 84 BIOGRAPHICAL SKETCH ............................................................................................ 87

PAGE 7

7 LIST OF TABLES Table page 3 1 Description of Low Cost Materials Investigated for Phosphorus Removal ......... 34 3 2 Description of IEX Resins Investig ated for Phosphorus Removal ...................... 34 4 2 Average Metal Composition (mg/kg) of Low Cost Materials rounded to 3 significant figures ................................................................................................ 53 4 3 Water Quality Characteristics for Lake Jesup and Sanford Avenue Canal ......... 53 4 4 Water Quality Characteristics for Sanford Avenue Canal and Lake Alice ........... 53 4 5 Removal of P by PX using Sanford Avenue Canal April water in Bench Scale Continuous Flow Study ....................................................................................... 54 4 6 Removal of P by AS Using Sanford Avenue Canal April Water .......................... 54 4 7 Comparison of AS1 and AS2 .............................................................................. 55 A 1 Metal Composition (mg/kg) of Low Cost Materials ............................................. 73

PAGE 8

8 LIST OF FIGURES Figure page 3 1 Sampling locations for A) Lake Jesup and Sanford Avenue Canal, and B) Lake Alice. .......................................................................................................... 35 3 2 Fl oating Island Treatment System showing A) the fluidized bed, and B) crosssectional view of fluidized bed system. ..................................................... 36 4 1 Composition of OP of A) Sanford Avenue Canal, and B) Lake Jesup using b atches of water from April. Initial Sanford Avenue Canal April TP = 289 ..................................................... 55 4 2 Rate of OP removal by A) LCMs (4 g/L), and B) IEX resins (1 and 4 mL/ L) in Sanford Avenue Canal water. C/C0 = 2.45 for FA at 60 min. Initial OP = 88.9 to 159.4 g P/L. .................................................................................................. 56 4 3 P remaining after 60 min of treatment of Sanford Avenue Canal water for A) LCMs, and B ) IEX resins. Peak of TPAS = 2.1; TPFS = 1.5; TPFA = 5.9; OPFA = 2.4. Initial TP = 125.3 to 286.6 g P/L; initial OP = 88.9 to 159.4 g P/L. ........... 57 4 4 P species removed after 60 minutes treatment of Canal Water for IEX resins. Where Pcomparison = (TPr OPr) / TPo, TPr = TP removed, TPo = Initial TP and OPr = OP removed. ............................................................................................ 58 4 5 Change in pH following 60 min of treatment of Sanford Avenue Canal water for A) LCMs, and B) IEX resins. .......................................................................... 58 4 6 Change in chloride and sulfate following 60 min of treatment of Sanford Avenue Canal water for A) LCMs, and B) IEX resins. Initial chlori de = 112 to 191 mg/L; initial sulfate = 22.5 to 40.6 mg/L. ...................................................... 59 4 7 OP remaining during bench scale continuous flow treatment by A) PX and AS, B) PX, AS1, MX and FS, and C) PX and AS, in differe nt water batches. Initial OP concentrations of Sanford Avenue Canal (SAC) April = 215.23 g P/L, and SAC June = 106.83 to 107.63 g P/L, LA July = 387.12 g P/L. .......... 60 4 8 Sulfate remaining duri ng bench scale continuous flow treatment by PX and AS. Initial sulfate concentrations Lake Alice (LA) July = 18.3 mg/L. ................... 61 4 9 Cyclic Flow, 12 hours treatment and 12 hours resting period, comp ared to continuous treatment. Lake Alice water was used from July and August for bench scale continuous and 12/12 experiments respectively. Initial OP concentration of Lake Alice (LA) July = 387.12 g P/L, and LA August = 396.15 g P/L. .................................................................................................... 61

PAGE 9

9 4 10 TP removal by FITSthrough mesocosm and fluidized beds between 10/09/2009 to 10/19/2009 during pilot study. ...................................................... 62 4 11 OP Removal by fluidized beds from mesocosm (influent) water and fluidized beds between A) 10/9/2009 to 10/19/2009, and B) 12/01/2009 to 12/12/2009 during pilot study. ............................................................................................... 62 4 12 Turbidity in Lake Alice (LA) during algae bloom and removal by mesocosm respectively between 10/02/2009 to 10/09/2009 during pilot study. ................... 63 4 13 Change of pH during algae bloom in Lake Alice (LA) pH between 10/02/2009 to 10/09/2009 during pilot study. ......................................................................... 64 4 14 TP removal from Lake Alice by FITS between 10/02/2009 to 10/09/2009 during pilot study. ............................................................................................... 64 4 15 OP r emoval from Lake Alice by FITS between 10/02/2009 to 10/09/2009 during pilot study. ............................................................................................... 65 4 16 Removal of OP by Regenerated PX from the benchscale continuous flow studies. Initial OP for Sanford Avenue Canal (SAC) June = 107.10 g P/L and 90.49 g P/L for PX and Regenerated PX respectively. ..................................... 65 4 17 OP removal from mesocosm (influent) water by batches of PX regenerated A) 1x (11/02/2009 to 11/13/2009), B) 2x (12/15/2009 to 12/21/2009), and C) 3x (01/12/2010 to 01/20/2009) during pilot study. ............................................... 66 4 18 Volume of 0.1M CaCl added to precipitate A) OP, and B) NOM from DWSs from v arious starting concentrations. Initial OP, Jar 1 = 42mg P/L, Jar 2 = 21 mg P/L, Column = 49 mg P/L. Initial UV254 Jar 1 = 1.9 cm1, Jar 2 = 2.6 cm1. .... 67 4 19 OP removal from mesocosm (influent) water by PX regenerated by cleaned DWS between 11/18/2009 and 11/30/2009. Resin used on FITS. ..................... 68 A 1 Elemental spectrum and dot map of Steel Slag done by scanning electron microscopy (SEM) and energy dispersive x ray fluorescence elemental microanalysis. This helped to identify calcium silicate as a dominant mineral. ... 73 A 2 X ray diffraction peak graphs of A) alum sludge, B) fly ash, C) ferric sludge, D) iron slag, E) limestone, F) recycled concrete, and G) steel slag showing mineral compositions .......................................................................................... 74 B 1 Change in A) Chloride, and B) Sulfate during column testing by P hosX (PX) and Alum Sludge (AS). Initial Chloride = 193.84 mg/L, Sulfate = 29.7 mg/L. ..... 78 B 2 Change in pH during column testing by PhosX (PX) and Alum Sludge (AS). ..... 79

PAGE 10

10 B 3 Change in pH during column testing by PhosX (PX), MIEX (MX), Alum Sludge (AS), and Ferric Sludge (FS). ................................................................. 79 B 4 Change in Total Phosphorus (TP) during c olumn testing by PhosX (PX), MIEX (MX), Alum Sludge (AS), and Ferric Sludge (FS). Initial TP = 125.99 ug P/L to 152.87 g P/L. .......................................................................................... 79 C 1 Phosphorus speciation throughout operation of Floating Isl and Treatment System showing treatment by mesocosm and PhosX packed fluidized bed. ..... 81 C 2 Total Phosphorus Removal by Floating Island Treatment System through Mesocosm and Fluidized beds. .......................................................................... 81 C 3 Orthophosphate Removal by Floating Island Treatment System through virgin PhosX and Regenerated PhosX Fluidized beds. ...................................... 82 C 4 Change in pH in Lake Alice throughout operation of the Floating Island Treatment System. ............................................................................................. 82 C 5 Change in turbidity in Lake Alice and removal by mesocosm throughout operation of the Floating I sland Treatment System. ........................................... 83 C 6 OP removal by fluidized beds from mesocosm (influent) water comparing PhosX (PX) dosed at 0.25 gals flowing at 0.5 gpm for 12 hrs per day until total exhaustion of PX between 03/23/2010 to 04/10/2010 during pilot study. ... 83

PAGE 11

11 LIST OF ABBREVIATION S 12/12 12 hours on then 12 hours off AS2 A second batch of Alum Sludge PLRG Pollution Load Reduction Goals SWIM Surface Water Improv ement and Management Act of 1987 (3x) Something done three times, if 1x means it was done one time, etc. AS Alum Sludge BV Bed Volume cm Centimeter DI Deionized DWS Dirty Wash Solution DX Dowex22 DX (1) Dowex22 dosed at 1mL/L DX (4) Dowex22 dosed at 4 mL/L FA Fly Ash FDEP Florida Department of Environmental Protection FITS Floating Island Treatment System FS Ferric Sludge gpm Gallon per minute IEX ion exchange IS Granulated Blast Furnace Iron Slag L Liter LA Lake Alice LCM Low Cost Mate rials, a.k.a. industrial by products

PAGE 12

12 LJ Lake Jesup LS Limestone N Normality M Molar mL Milliliter MX MIEX MX (1) MIEX Dosed at 1mL/L MX (4) MIEX dosed at 4mL/L OP Orthophosphate, bioavailable phosphorus or PO4 3 P Phosphorus PAC Powder Activated C arbon PX PhosX PX (1) PhosX dosed at 1 mL/L PX (4) PhosX dosed at 4 mL/L PZC Point of Zero Charge RC Recycled Concrete rpm Revolutions per minute SAC Sanford Avenue Canal SJRWMD St. Johns River Water Management District SS Basic Oxygen Furnace Steel Slag TN Total Nitrogen TP Total Phosphorus UV254 Ultraviolet absorbance at 254 nm XRD X ray diffraction

PAGE 13

13 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Engineering SUSTAINABLE PHOSPHORUS REMOVAL FROM SURFACE WATER By Amar Prem Persaud May 2010 Chair: Treavor H. Boyer Major: Environmental Engineering Sciences Phosphorus (P) is often a limiting nutrient in plant productivity. An excess of its bioavailable form, orthophosphate (OP) will cause eutrophication. Natural surface water systems may become overloaded with OP therefore, it is pertinent to develop a strategy of remediation. Adsorption through upward flowing column has proven to be an effective method of removing OP from surface water. In this research it was found that the use of industrial by products also known as low cost materials (LCMs), as compared to commercially available materials, such as polymeric adsorbents in the treatment method can become feasible but performance will vary due to changing composition and prepared particle sizes. LCMs tested were, alum sludge, ferric sludge, steel slag, iron slag, recycled concrete, fly ash, and limestone of which alum sl udge performed best. It was found that the advantages of using commercially available materials are ease of implementation, consistency of performance and reusability. Commercially available materials tested were PhosX, MIEX, and Dowex22 of which PhosX performed best. Both alum sludge and PhosX are effective at removing OP from water by adsorption under continuous flow Implementing fluidized beds to replicate benchscale column treatment on a larger scale which utilize P selective materials has

PAGE 14

14 proven to remove OP effectively. B ench scale studies have helped to successfully predict the removal of OP by the pilot scale fluidized bed with typically >70% OP removal by both Materials such as P hos X a polymeric adsorbent resin, can be regenerated and was found to work as effectively as its virgin version. However, when using a reusable commercially available material such as P hos X the by product of regeneration is a contaminated brine solution. The P in solution can be removed by precipitation and the cleaned brine solution can be used to regenerate another batch of exhausted resin. The technique of considering all aspects of a treatment process, whil e keeping sustainability in mind, has proven to be useful in developing a highly effective P treatment system.

PAGE 15

15 CHAPTER 1 INTRODUCTION Surface waters continue to be overloaded with phosphorus (P) from point and nonpoint sources, which exist in sediment and surface water runoff. An excessive amount of P and nitrogen in surface waters can cause eutrophication. In t his case, P is the limiting nutrient in plant productivity. Therefore, emphasis must be placed on P removal to eliminate eutrophication (Vollenweider et al., 1980). P mainly occurs in sediment and is released into the water ( dissolution) by biogeochemica l processes (nutrient cycle). The main form released is PO4 3 -, also known as orthophosphate (OP) which is immediately available to algae (bioavailable). Diffusion is caused by factors such as temperature, pH, redox and advective forces from flowing water enhancing the exchange of P from sediment to OP in water. These are naturally occ urring and cannot be controlled; the act of altering the environment to such a great extent is difficult. Therefore, the bioavailable P, also known as OP, existing should then be the target for removal if eutrophication is to be controlled, as it is impractical to dredge all the P in sediment from waterways or drastically change the water quality characteristics. Eutrophication can increase the rate of algal production, which consumes dissolved oxygen and blocks the sunlight in the water thus, making it difficult for native plants and marine life to survive. In addition, when algae die it results in a high chemical oxygen demand where an accelerated production causes a larger a mount of dead organisms. It was reported by the Florida Department of Environmental Protection (FDEP) that cyanobacterial algae blooms have become a problem in Lake Jesup (FDEP, 2008) These algae blooms can release toxins and increase pH that are harmful

PAGE 16

16 to humans and animal life that are in contact with the water (Jones et al., 2006) Alga e blooms are aesthetic issues as well causing foul aromas and discoloration of lakes. Sources of P pollution include agricultural drainage, municipal and industrial eff luents, and urban stormwater runoff. In a specific case urban stormwater drainage has been cited as the cause of increased levels of total phosphorus (TP) in soil (McCormick and Newman, 2009) This results in the loss of the abundant native macrophyte species and a shift to periphyton communit ies which have greater impacts on the dissolved oxygen and trophic levels in the Everglades (McCormick and Newman, 2009) In fact, P removal techniques have been developed for agricultural drainage (Dayton and Basta, 2003) and municipal and industrial wastewater effluents (Blaney and Cinar, 2007) to attempt source treatment. Tributaries often contain O P not only from dissolution but from surface water runoff when located in populated or agricultural zones. F or example, Lake Jesup in central Florida receives much of its OP from surface water runoff. During wet years, the a verage annual watershed loading of TP is 18.7 metric ton s P/yr in Lake Jesup (Jia, 2007) This is coming mainly from its tributaries which results in a lgae blooms. The goal set by the St. Johns River Management District ( SJRWMD ) was to remove 9 metric tons P/yr from the watershed (SJRWMD, 2008). Upon entering the lake, most biological activity occurs because the majority of the surface area of a large l ake is directly exposed to sunlight which is the driving force for photosynthesis. If one were to treat a lake directly, P in sediment and organic P have to be removed. Tributaries are shaded quite often, so less biological activity is likely to happen an d a high OP concentration exists, where OP should be the target. O ther than the existence of natural organic

PAGE 17

17 matter ( NOM ) from terrestrial sources, not much plant activity could occur in an area blocked from the sun. The SJRWMD had embarked on adopting Pol lution Load Reduction Goals (PLRG) as reflected in the Florida Administrative Code 6240 (FAC 62 40) and has instilled the Surface Water Improvement and Management (SWIM) Act of 1987, which involves m ainly the achievement of PLRGs (Jia, 2007) A plan like this helps to improve lakes because it focuses on reducing the pollution in storm water runoff by reducing excess nutrients and other pollutants which affect water quality (SFWMD, 2009) Beyond policy, researchers have thought of various techniques to c apture P by physical chemical or biological processes at point sources and nonpoint sources. In order to keep the environment in its pristine condition, all water quality characteristics must be considered while removing the P from water sources. By exami ning real surface water from the Lake Jesup watershed, it is quickly realized that it is important to not only track the removal of P but the water chemistry that will affect its removal and how the removal will in turn affect the water chemistry. Hence a removal process can beco me a detriment to the environment if a thorough investigation is not performed. Other constituents in the water can also compete to be removed if they have a similar affinity for removal mechanisms as OP does for example, NOM, which is a dissolved constituent in the water. NOM can take up space on the surface of an adsorbent that could have been used to remove more OP (Weng and Van Riemsdijk, 2008) Hence, OP can be outcompeted by NOM and other anions in the water. Due to the fact that sustainability of the process is very important, water treatment processes should have an ambition to adapt a synergestic approach by considering

PAGE 18

18 source of materials, ease of use, and total environmental effects. For example, the use of industrial by products would be part of a conscious effort to improve the total environment. Industrial processes have waste materials from the production of their specific products. Some of these materials can be used to remove OP by methods such as adsorption or prec ipitation. The waste products will be considered as low cost materials (LCMs) from here on, because the only cost asso ciated with the material is shipping. By adapting this approach, their use s are part of a sustainable effort because they will be fully ut ilized before inevitabl y dispos ing of them Comparing LCMs to commercially available resins that are designed for water treatment ( ion exchange (IEX) resin s), it is important to establish a benchmark to evaluat e the efficacy of using LCMs and the feasibili ty of using IEX resins. Sustainability, for this project, is therefore defined as a treatment process that would include beneficial reuse of industrial by products, recycling of materials, handling of by products of material use, ease of implementation, and a renewable energy source. The IEX resins have an advantage whereby they are designed for specific contaminants, and can be regenerated and reused. However, in the process, it produces a brine solution containing P. It is important to consider whether th e P can be retrieved and marketed, and if the brine solution can be utilized again, to reduce the harmful discharges from facilities equipped to embark on this process. All environmental factors must be carefully considered in designing an effective treatm ent process. All of the above considerations describe a variety of synergies that are important in order to design a safe and effective treatment process. In the case of this project, they are applied to the removal process of P from surface water bodies. Hence, a pilot scale

PAGE 19

19 treatment facility was constructed due to the fact that surface waters are large and nonuniform in water quality. It embodied a floating, solar powered and sustainable method to remove P with an aim of ensuring that a safe aquatic environment is maintained This facility will be designated as the Floating Island Treatment System (FITS) upon which a physical chemical treatment process was designed. The overall goal of this project is to improve surface water quality by removing P, whic h will control and prevent eutrophication. The specific objectives are (1) to discuss the properties of LCMs and its P removal mechanisms, (2) to evaluate the effectiveness of LCMs and IEX resins for P removal while screening for P selective materials, (3) to evaluate O P removal by P selective materials by comparing laboratory scale and pilot scale continuous flow treatment, and (4) to evaluate disposal and regeneration options for P selective materials and their byproducts to ensure the process is sustainable.

PAGE 20

20 CHAPTER 2 LITERATURE REVIEW Three common mechanisms of removing dissolved constituents from water are ion exchange (IEX), adsorption, and precipitation. Seeking materials that exhibit those mechanisms would be beneficial in order to remove the bioavailable phosphorus or orthophosphate (OP) in water. There are multitudes of materials available on the market for sale that are designed especially for water treatment and the removal of dissolved matter, such as IEX resins (Boyer and Singer, 2008; Blaney and Cinar, 2007) However in an effort to consider sustainability factors of a treatment process, by reducing its footprint, then it is beneficial to consider the reuse of industrial byproducts. These products are low cost materials (LCMs) and the process should allow LCMs to be used once more before an inevitable disposal. Beneficial reuse of LCMs like steel slag, fly ash and water treatment residuals such as alum sludge and ferric sludge have been investigated for removal capabilities of OP from wastewa ter (Johansson, 1999; Mortula and Gagnon, 2007a; Ugurlu and Salman, 1998) E lements such as iron, aluminum, calcium and magnesium have an affinity for OP and it can be determined which materials are worth investigating based on their composition, such as limestone (Baker et al., 1998) and iron slag. Commercially available materials such as IEX resins and other polymeric adsorbents have also been used for phosphorus (P) removal (Blaney and Cinar, 2007) For example, MIEX which can be applied to water high i n natural organic matter (NOM) (Boyer and Singer, 2008) Dowex22, and PhosX (Blaney and Cinar, 2007) P was removed in the past by using various materials and adopting a number of techniques of experimentation. Jar testing is a popular method to carry out batch

PAGE 21

21 experiments, whereby a large amount of materials can be quickly tested and selected for removal of P (Mortula and Gagnon, 2007a) Another method is the column experiment, where a continuous supply of water is pumped through a media to test for its t hreshold (Mortula and Gagnon, 2007b) It was determined that adsorption through a column was an effective way of removing P (Baker et al., 1998) especially if a treatment system has to handle an infinite flow of water rather than a fixed volume. Column exp eriments also provided an ideal environment that allowed adsorption onto metal hydroxide surfaces. It was implied that a continuously flowing system can be applied to receive wastewater from septic tank s from which P could seep into a surface water body (B aker et al., 1998) Pilot studies are important to in order to take laboratory studies to the next level and determine if a hypothesis will work in the real environment. In previous batch studies materials such as alum sludge, blast furnace slag and granulated activated carbon performed well, with over 70% removal of P at high initial concentrations of 2.5 mg P/L and more from deionized water. M aterials were screened and tested for their capacity through adsorption column experiments (Mortula and Gagnon, 2007b; Baker et al., 1998) Previous column experiments resulted in fly ash yielding removal of 99% of P at initial concentrations ra nging from 20 mg P/L to 50 mg P /L (Ugurlu and Salman, 1998) Alum sludge and granulated activated carbon performed well and w ere relatively inert to water quality characteristics such as pH but can be affected by a change in pH (Mortula and Gagnon, 2007b) However, blast furnace slag yielded the greatest removal of P but raised the pH (Mortula and Gagnon, 2007a) Particle size was deemed an important physical characteristic of materials and drastically reduce the surface area available for adsorption as it gets larger (Mortula and

PAGE 22

22 Gagnon, 2007b) ; this is also important in the same way for IEX (Boyer and Singer, 2008) Materials that have available calcium usually precipitate P from solutions they are in contact with (Baker et al., 1998) LCMs were good candidates for OP removal but it must be confirmed if they are formidable opponents to the usage of commercially available resins and if they can be applied to a real treatment process. On numerous occasions, there were more gaps in the studies For instance, some column studies only used deionized water spiked with OP (Mortula and Gagnon, 2007b; Ugurlu and Salman, 1998) which in fact did not pose a conclusion on how it will work in a real scenario. Current P treatment methods require high concentrations, but when dealing with large natural surface water bodies, the P concentrations are much lower. This must be addressed to contribut e meaningfully to the research involved in recovery of the natural environment. Past studies were also limited whereby essential water quality characteristics were not fully investigated showing the possibility that the chemistry affected a treatment process and in turn if the treatment process was affected. For example, the usage of alum sludge could release sulfates in the water, which may methylate mercury where it exists forming methylmercury [CH3Hg]+ (Benoit, 1997) which results in bioaccumulation and food web magnification. A comparative study of LCMs and IEX resins is essential to test their P removing capabilit ies and their effect s on organic rich surface water T herefore, water from a real source will be beneficial so that all constituents can be observed. It was determined by researchers that c ompetition exist ed between NOM and P for removal by materials using IEX or adsorption as a mechanism of removal (Guan and Shang, 2006) It is

PAGE 23

23 necessary to prove that these materials are effective in the real environment and can be applied to a large scale treatment system. A comparative study between benchscale and pilot scale studies will give insight on this. Commercially available IEX resins and adsorbents can be regenerated by removing the P from the s urface sites by specific reg en eration processes and reapplying them to the treatment system (Blaney and Cinar, 2007; Boyer and Singer, 2008) because they are useful when regenerated with the possibility of multiple regenerations (Apell, 2009) With P diss olved in the dirty wash solution for these materials, it is possible to remove it by precipitation (Sibrell and Montgomery, 2009) so that P is not being released into the environment. Hence, i t is crucial to compare all factors involved in a treatment proc ess to evaluate the best approach. This will enable the use of an economical and eco friendly process for removal of P from natural surface waters.

PAGE 24

24 CHAPTER 3 MATERIALS AND METHOD S Treatment Materials Two classes of materials were evaluated in this work: low cost materials (LCMs) and ion exchange (IEX) resins All LCMs were analyzed for their composition and physical characteristics. The LCMs evaluated in this work consisted of waste byproducts and natural materials, as described in Table 3 1, and the only associated cost was shipping. The LCMs alum sludge (AS), steel slag (SS), and iron slag (IS) were crushed with a mortar and pestle and sieved through U.S. Standard sieves 30 and 40, to give a particle size range of 420 (FS) limestone (LS), and recycled concrete (RC) were dried under ambient laboratory conditions before being crushed as above. A batch of Class F Fly ash (FA), which was received in a powder ed form w as used in its original stat e. These materials were chosen on the basis of studies which show that Fe and Al oxides have faster adsorption k inetics than those of Ca and Mg (Guan and Shang, 2006) The LCMs were weighed as dry material s ranging from 1 to 16 g for jar tests. The resin s that were investigated are commercial IEX resins and polymeric adsorbents, which are designed for municipal and industrial water treatment. All resins will be considered as IEX resins and are described in Table 3 2 The IEX resins, MIEX (MX), PhosX (PX) and Dowex 22 (DX), were used in its original state and stored in deionized (DI) water. They were measured using a graduated cyl inder with doses ranging from 1 to 8 mL of wet settled material. The densities of the IEX resins were determined as the mass of dr y material per volume of wet settled material. All materials

PAGE 25

25 were evaluated following standard jar tests procedures, as described in the next section. Materials used for bench scale co ntinuous flow column studies were crushed, washed and measured as a wet volume in a graduated cylinder, which r epresents the bed volume (BV) of water treated. Materials used for pilot scale studies were prepared in a similar manner. PX was prepared as described in column studies. AS was crushed and sieved to grain sizes of between 0.5 inches and 0.75 inches, washed, and soaked. Surface W ater Bench scale experiments were conducted with surface water obtained from Lake Jesup and one of its tributaries, the Sanford Avenue Canal based on a project focus directed to the Lake Jesup watershed. Lake Jesup is located near the city of Sanford in central Florida and is part of the St. Johns River system Lake Jesup is impaired by high concentrations of total phosphorus (TP), total nitrogen (TN), unionized ammonia, and low concentrations of dissolved oxygen (FDEP, 2003) Lake Jesup and its tributaries are also rich in organic material. Figure 31A shows a map of Lake Jesup and a detail of the sampling locations where Lake represents Lake Jesup and Canal represents the Sanford Avenue Canal All samples were col lected by the St. Johns River Water Management District (SJRWMD) and delivered to the Department of Environmental Engineering Sciences at the University of Florida. Samples were stored at 4C upon receipt and were collected on four different dates in 2009. Seasonal differences in water quality and differences in water quality for the two sample locations will be discussed in a subsequent section.

PAGE 26

26 It was later determined that it was more feasible to launch a pilot scale study in Lake A lice on the University of Florida campus due to its close proximity. Therefore, experiments were also performed on water samples collected from Lake Alice and subsequently, the pilot scale facility was launched there Lake Alice located in Gainesville, FL is an open water/marsh system that receives surface water runoff from t he University of Florida campus in addition to a wastewater treatment plant situated to the east of it. I t is in the center of the University of Florida campus traversing the western stretch of M useum R oad. It is purported to have high levels of TP, orthophosphate ( OP ), and organic matter making it similar to the conditions found at Lake Jesup. Thus, Lake Alice provided an excellent replacement for the conditions experienced at Lake Jesup The s ample location for bench scale continuous flow studies was at an outflow canal near Baughman Center. F igure 3 1B exhibits the sampling location for column experiments and location of the Floating Island Treatment System (FITS) All graphs and fig ures refer to Lake Jessup, Sanford Avenue Canal, and Lake Alice as LJ, SAC, and LA, respectively, and in the case of bench scale experiments, followed by the month of procurement; for example Sanford Avenue Canal water obtained in April will be presented as SAC April in tables and figures. All water used for bench scale experiments were obtained during the year of 2009. Jar Testing Standard jar tests were conducted using a Phipps and Bird PB 700 jar tester at ambient laboratory temperature (20 22C) to investigate the materials P removing abilities. Two liters of Lake Jesup or Sanford Avenue Canal water were added to each jar. Various doses of LCMs or IEX resins were measured and added to each jar.

PAGE 27

27 The following constant mixing speeds were used: 100 rpm for AS, FA, and IEX resins and 200 rpm for RC, LS, IS, SS and FS because these materials were denser. All jar tests were conducted for 60 minutes, with samples collected after 5 minutes of mixing (no settling), 30 minutes mixing (no settling), and 60 minu tes mixing (with 30 min utes of undisturbed settling). The material doses were tested in duplicate. All results are average values of duplicate samples with error bars showing one standard deviation, unless otherwise noted. Raw water and treated water sampl es were measured for pH, turbidity, OP, and TP. Samples were also analyzed for inorganic (Boyer and Singer, 2008) All glass sample containers were soaked overnight in a 6% nitric acid bath, rinsed three times with DI water, and air dried. Jar test paddles and jar s were washed with laboratory detergent and DI water, rinsed six times with DI water and air dried. Plastic sample containers used for sampling were not reused. Continuous F low S tudies Bench Scale Experiments Following the concept of small scale laboratory based experiments (Mortula and Gagnon, 2007b) a fixed bed column was setup. The column used had 0.7854 cm inner diameter and a height of 2 cm with 25 pore size polyethylene frits on each end. The treatment material was prepared soaked overnight and washed of its suspended fine grains The column was then filled with 1 mL of wet material after which it was topped off with DI water to eliminate bubbl es. Tubing was connected to allow for an upflow direction. If pumped downward, further compaction of the material will occur and other particles will build up, hindering the free flow of water. An upward flow allows the grains to remain somewhat suspended causing an opening of pores between grains, hence

PAGE 28

28 providing additional surface area that were originally covered by grains in contact with each other. The flow rate used was 2 mL/min, preset by using DI water as source water and measuring the outflow in a graduated c ylinder. The concept of bed volumes (BV) was applied and in this case, 1 mL flow was equivalent to 1 BV because 1 mL of material was used. Therefore, the treatment flow rate was 2 BV per minute. BV will now represent the volume of any solution used in continuous flow studies, which is derived from the volume of material used. Before an experiment started, the system was flushed by pumping 120 BV of DI water. Water to be tr eated was filtered, through a Whattmann GF/A filter (1.6 size) to control clogging of frits. Two studies were performed: continuous flow and cyclic 12 hour s on and 12 hour s off (12/12) where 1 hour samples were taken in 125 m L Erlenmeyer flasks every 3 hours and left to flo w without sampling for 12 hour s or to rest for 12 hours during the night, respectively. A predetermined breakthrough OP concentration (Mortula and Gagnon, 2007b) was determined as 50% removal after which one extra sample was taken at the next sample point to confirm the end of an exper iment. Immediate trends in OP changes were observed by measuring with a Hach 850 Colorimeter. The water quality of raw water before and after filtration and all following samples were analyzed for OP, TP, anions, and pH. Each time a new batch of water was received, a continuous flow column test was performed using the most promising materials PX and AS, in order to facilitate a reference for comparison. The column setup was cleaned by scrubbing the interior of columns, and then pumping 50 BV 6% nitric acid and 200 BV of DI water through the entire system.

PAGE 29

29 Pilot S cale Experiments Based on the physical chemical research presented in this project fluidized beds were installed on a combined treatment system which included biological treatment and deployed into Lake Alice T he biological treatment proved to be useful to replicate the filt r ation of the raw water as done in the benchscale column studies reducing turbidity, and some amount of OP. The fluidized beds replicated on a larger scale the bench scale colu mn setup keeping the treatment of 2 BV/min as a constant factor. T he incorporation of two fluidized beds w ere implemented in order to use two small chambers rather than one large chamber t o reduce the pressure required to pump water through it and head los s. This was convenient because it allowed two materials to be compared at the same time. As shown in Figure 32A, the fluidized beds were mounted in a position to receive water from the mesocosm, Figure 32B illustrates a cross section of the fluidized bed. Each fluidized bed was 6 inches in interior diameter and 24 inches high where sediment shields covered both ends to prevent treatment materials from leaving. They were each packed with 946 mL (0.25 gallons) of either AS or PX. Water from the mesocosm was pumped upward at a rate of 2 BV/min (0.5 gpm) through each fluidized bed with a bilge pump for 10 to 12 hours per day and rested through the night. Grab sampling was done every day at consistent times before and after treatment. Therefore each marker in g raphs represents a different day. A predetermined breakthrough of OP removal was set at 50%; thereafter two fresh prepacked fluidized beds replaced the exhausted fluidized beds to continue treatment and experiments. All graphs presented that refer to the pilot study show absolute values rather than C/C0 values on the y axis because significant changes in water quality were observed and it was easier to

PAGE 30

30 visualize the quantity of OP that was being removed by the use of absolute values. All samples were anal yzed for OP, TP, Ultraviolet absorption at 254 nm (UV254) anions, turbidity, and pH. Sustainability Studies Regeneration P rocedure for PhosX PX resins were regenerated by a cleaning and neutralization process. Cleaning solution consisted of 2% NaCl and 4% NaOH as recommended by the manufacturer whereby 1 g dried chemical in 100 mL DI water represented 1% A column procedure used in previous studies for cleaning PX was not followed exactly (Blaney and Cinar, 2007) but inspired a modified procedure, in lieu 12 equivalent BV of cleaning solution with used PX were shaken on an Innova 2000 platform shaker at 220 rpm for 30 minutes. The dirty wash solution (DWS) was decanted and stored in a glass vial for further analysis. 20 mL of DI was added to the PX and s haken at 200 rpm for 10 minutes to remove any remaining contaminants. 100 mL of water was added to the cleaned PhosX which was titrated with a 0.1N HCl solution at 25 L increments until a pH of 7 was sustained for 10 mi nutes whilst being shaken at 200 rpm It was ensured that the pH did not go below 5 as the Fe will be stripped from the PX. The solution was decanted and the PX stored in DI. For larger scale regeneration related to the pilot study, the shaker plate was set at 220 rpm and 8 BV equivalent DI was added after DWS was decanted. 0.1N HCl was added in 1 m L increments. Precipitation Procedure The DWS from r egenerating PX contain ed all pollutants that the r esin had adsorbed. Therefore, this was an effort to separate the P from the solution. The DWS

PAGE 31

31 may then be reused for regeneration as described later 1 mL was drawn from the DWS and diluted by 100 (i.e. diluted by 100x) to remain within the range that a Hach 850 Colorimeter can read. This sample was filtered through a 0.45 d f or OP measured by the Hach meter (Hach OP), UV254, and pH. 10 BV of DWS was titrated with 0.1M CaCl2 solution. Titration involved continuously adding 2 m L of 0.1M CaCl2 and allowing mixing for 2 to 3 minutes; at each increment, 0.25 m L DWS was withdrawn and diluted to 100x, filtered and analyzed for Hach OP, UV254, and pH When a sign ificant drop (<25%) in OP was observed 0. 1 m L increments CaCl2 were then added following the above titration procedure until the Hach Colorimeter showed <0.05 mg P/L. Results were normalized based on addition and extraction of fluid to and from the DWS : (Normalized value = (Vol. Remaining + Vol Ca Solution Added Vol. Withdrawn) x Actual Reading / Starting Vol.). Then, results were multiplied by 100 to show concentrations bef ore dilution. For a larger scale p ilot study the same procedure was followed with a few exceptions due to the larger scale. 1 M CaCl2 solution was made and was added in increments of 200 m L then 100 m L when a significant drop in OP was noticed. AS should be disposed of after being used according to EPA guidelines for disposal of alum sludge with arsenic content (MacPhee et al., 2001) Reuse of Dirty Wash Solution After a DWS is generated, it should be considered reus able before final disposal Therefore, after a DWS was generated in pilot studies P was precipitate d using the above procedure filtered through a 0.45 according to the regene ration procedure after all OP was removed. However, it became difficult to remove the precipitate during the final 3 BV of solution; therefore, fresh wash

PAGE 32

32 solution was made to replace th e loss The solution was analyzed for Hach OP pH, and UV254 absorbance to ensure all OP was removed and the behavior of NOM was tracked. Ana ly tical Methods ACS specified grade or better chemicals were used to prepare all standard aqueous chemicals. Aqueous samples were analyzed as follows. An Accumet AP71 pH meter with a pH/ATC probe was used to measure pH. The pH meter was calibrated before each use with pH 4, 7, and 10 buffer solutions. Turbidity was measured on a Lamotte 2020 instrument that was calibrated daily wi th a 1 NTU standard. Chloride, sulfate and nitrate were measured on a Dionex ICS 3000 ion chromatograph equipped with IonPac AG22 guard column and AS22 analytical column. All inorganic anions were measured in duplicate with average values reported. The relative difference between duplicate samples was <1%, and the relative difference between calibration check standards and known concentration was <20% percent. OP was measured using a Hach DR850 colorimeter using an ascorbic acid method equival ent to U.S. EPA Method 365.2 and was checked periodically for its accuracy by measuring 0.05, 0.1, and 0.2 mg P/L standards. Samples were also sent to the UF/IFAS Analytical Services Laboratory for OP and TP in 20 mL scintillation vials and followed U.S. E PA Method 365.1. The TP samples were acidified to pH <2 with sulfuric acid for preservation after which autoclave digestion using ammonium persulfate and sulfuric acid was performed. The agreement between OP measurements for the Hach DR 850 and Analytical Services Laboratory were approximately 30 to 60% The Hach DR 850 was used to show immediate trends in P removal. All OP data presented in tables and figures were reported from the Analytical Services Laboratory data, unless otherwise specified

PAGE 33

33 UV254, TOC and TN were measured according to proc e dures outlined by Banerjee ( 2010) Solid samples were analyzed as follows. The elemental composition of the LCMs was determined by two different methods. Samples were sent to the Analytical Services Laboratory, whic h follows U.S. EPA Method 200.7 to analyze for plant available iron, aluminum, and calcium using the Mehlich 3 extraction technique for metals. The LCMs were also digested following US EPA method 3050b using h eat plus n itric a cid followed by hyd rogen peroxi de oxidation, and metals were determined by Thermo Jarrell Ash Trace ICP AES. They were analyzed in triplicate doses following the USEPA 3050B method. All results were compared to data from the digestion of standard reference 2709 soil (NIST, 2002) All da ta retrieved and reported in this project as greater than 0 mg/kg were above the Thermo Jarrell Ash Trace ICP AES minimum detection limit which varied per element Point of z ero charge (PZC) was determined by typically varying the percentage by weight of sample in the water and duplicate doses were performed. S amples w ere dried overnight measured, placed in beaker s of nanopur e water and w ere bubbled with N2 gas for at least 20 min utes while covering with parafilm. 5 mL of bubbled water was then injected into a narrow neck vial with 0.5 g of material The sample s w ere agitated to ensure mixing. Head space was then filled with N2 gas for about 30 seconds after which the sample vial s w ere sealed promptly. Samples were shaken for 24 hrs before pH levels were analyzed. Pore size, pore volume and surface area were analyzed following procedures used by Bach and Mazyck (2007)

PAGE 34

34 XRD m ethod was as follow s. Samples were crushed to less than 75 then mounted on plexiglas cavity sample holders (Harris and White, 2008) Each sample was scanned from 2 (Harris and White, 2008) The instrument used was a computer controlled x ray powder diffractometer equipped with stepping motor and graphite crystal monochromator (Harris and White, 2008) Results were reported in graphical form whereby, the dominant existing minerals were estimated. Table 31 Description of Low Cost Materials Investigated for Phosphorus Removal Material Source Description Alum sludge (AS) (Mortula and Gagnon, 2007a) Peace River Manasota Regional Water Supply Authority, Arcadia, FL Surface water treatment plant that uses aluminum sulfate to treat water from the Peace River a Ferric sludg e (FS) David L. Tippin Water Treatment Facility, Tampa, FL Surface water treatment plant that uses ferric sulfate to treat water from the Hillsborough River. Granulated blast furnace iron slag (IS) Civil & Marine Inc., Cape Canaveral, FL Nonmetallic byp roduct from iron production. Basic oxygen furnace steel slag (SS) (Johansson, 1999) Levy Enterprises, Valparaiso, IN Byproduct of manufacturing steel from pig iron. Class F fly ash (Pidou and Avery, 2008) Boral Materials Technologies, Tampa, FL Combusti on of ground or powdered coal. Recycled concrete (Pidou and Avery, 2008) Florida Concrete Recycling Inc, Gainesville, FL Concrete aggregate collected from demolition sites. Limestone Florida Rock Industries Inc., Gainesville, FL Natural rock mined from v arious locations. a Contain powdered activated carbon, which is used prior to coagulation process. Table 32 Description of IEX Resins Investigated for Phosphorus Removal Material Manufacturer Structure and Density, (g/L) Application PhosX (PX) SolmeteX Macroporous polymer resin impregnated with iron oxide particles ; = 0.343 Developed specifically for phosphate removal (Blaney and Cinar, 2007) MIEX (MX) Orica Watercare Macroporous, polyacrylic anion exchange res in with strongbase, type II functional groups ; = 0.299 Used in water treatment for removal of organic material (Boyer and Singer, 2007) Dowex22 (DX) Dow Chemical Macroporous, polystyrene anion exchange resin with strongbase, type II functional groups = 0.415 Conventional anion exchange resin; simila r to IRA910

PAGE 35

35 A B Figure 3 1. Sampling locations for A ) Lake Jesup and Sanford Avenue Canal and B ) Lake Alice.

PAGE 36

36 A B Figure 32. Floating Island Treatment System showing A) the fluidized bed, and B) crosssectional view of fluidized bed system.

PAGE 37

37 CHAPTER 4 RESULTS AND DISCUSSI ON Low Cost Materials Tables 4 1 and 42 list the physical and chemical properties of the low cost materials (LCMs), which have been used in the bench scale studies of this project Results in Table 42 were adjusted to 3 significant figures. The composition of materials such as alum sludge (AS) and ferric sludge (FS) varied due to different source treatment methods and the others differed by industrial processes Each was examined using several parameters in order to explain their removal mechanisms. Some LCMs exhibit metal hydroxide surfaces favorable to adsorption and others with high calcium contents, precipitation occurs (Baker et al., 1998) Aluminum ( Al ) calcium ( Ca ) and iron ( Fe ) have strong affinities for p hosphorus (P) and the composition of the tested materials should reflect its efficacy to remove P. It was observed previously that materials higher in Al and Fe hydroxides have faster adsorption kinetics (Sibrell a nd Montgomery, 2009) AS and FS had high proportions of Al and Fe (see Table 42) respectively, and are the precipitates of coagulation processes forming aluminum hydroxide and ferric hydroxide, respectively. Therefore, AS and FS will experience adsorption because of their metal hydroxide surfaces, but recycled concrete (RC) and limestone (LS) will treat by precipitation due to their high calcium content. Further studies were done to determine the surface area of the materials as it helps to compare which may have more adsorption sites. Surface area changes with the pore size distribution if grain size is held constant, because if a material is porous then it has a greater surface area F or example, activated carbon is porous and has a large surface area ( Tennant and Mazyck, 2007) The AS obtained was said by the water treatment

PAGE 38

38 plant, as cited in Table 31, to have had a large quantity of powder activated carbon and it was observed that the surface area was large compared to other materials. Moreover intra particle diffusion of P within the pore structure plays a large role in the materials adsorption rate and capacity (Sibrell and Montgomery, 2009) Point of zero charge (PZC) was examined as well When the pH is below the PZC the surface is positively cha rged and vice versa for above. The performance of removal will depend on the conditions of the water, such as pH which enhances precipitation when high but hinders adsorption when above its PZC AS, FS, iron slag (IS), and steel slag (SS) are expected to r emove P by adsorption, while RC, LS, and fly ash (FA) are expected to remove P by precipitation. X ray diffraction (XRD) analysis was performed to help determine the dominant minerals by analyzing the crystal structures. M aterials such as SS, RC and LS hav e many minerals present because they all consist of raw minable materials, such as limestone (contained in RC and LS), and hematite (SS is a by product from its use in steel alloys). AS and FS did not respond significantly to XRD and this may primarily be due to their formation, which is independent of geologic processes. Table 41 shows the specific mineral contents. Bulk Screening for P Selective Materials Surface Waters The two raw waters investigated here showed very different physicochemical properties While the raw waters were of similar pH values, the greater total organic carbon (T OC ) content of Sanford Avenue Canal reflects on the greater ultraviolet absorbance at 254 nm ( UV254) of the highly colored water. The water quality characteristics for Lak e Jesup and Sanford Avenue Canal are summarized in Table 4 3 Both Lake Jesup and Sanford Avenue Canal waters have turbidity on account of high

PAGE 39

39 concentration of algae; more so with Lake Jesup because it is directly exposed to sunlight which drives photosynthesis. Figure 41 presents a particular phenomenon occurring related to the nutrient cycle transitioning from the tributary to lake. The orthophosphate (OP) is high in the Sanford Avenue Canal compared to other P species and as the water drains into Lake Jesup the OP is converted to other forms of P which, with turbidity readings in Table 43, represents great biological activity. This is due to the fact that tributaries are mainly shaded from the sunlight and the hydraulic flushing rates are high not allo wing algae growth, in this case Sanford Avenue Canal has an average flow rate of 11,237 acre ft/year (Jia, 2007), whereas lakes are open, slow flowing and exposed bodies of water where the sunlight drives photosynthesis. Overall, the Sanford Avenue Canal c ontributes 2.2 metric ton TP/year (Jia, 2007). P speciation was of particular interest in these water bodies P speciation was measured by the ratio of OP to total phosphorus ( TP ) (i.e., OP:TP), where TP includes OP condensed polyphosphates, organic phosphorus, and phosphorus in biomass. When comparing the OP to TP and natural organic matter (NOM) to TOC contents it can be understood the biological activity occurring in the surface water. For P treatment technologies removal of OP relative to TP was an important aspect as will be discussed in subsequent sections Lake Jesup was characterized by a low concentration of OP, low OP: TP, high concentration TOC, and moderate specific ultraviolet absorbance ( SUVA ), where SUVA L/mg C m = 100 x (UV254 cm1 / TOC mg/L ) The major seasonal trends for Lake Jesup were an increase in turbidity, decrease in SUVA, inc rease in TP, and decrease in OP: TP. The seasonal trends suggest increased biological activity within Lake Jesup as

PAGE 40

40 winter progressed to summer. For example, the increase in turbidity reflect ed increased biomass, the decrease in SUVA was a result of microbial byproducts contributing to the organic matt er pool, and the decrease in OP: TP was a result of OP incorporation into biomass. Historic w eat h er data for the O r lando Sanford area (KSFB) showed an increase in temperature progressing through the sampling times and low rainfall (WUnderground, 2009) This may explain the increased concentration of P in the Sanford Avenue Canal via dissolution and evaporation and greater biological act ivity through the seasons in Lake Jesup Sanford Avenue Canal was characterized by hi gh concentration of OP, high OP: TP, high TOC concentration, and high SUVA. Overall the tributary water draining into the lake is the main source of nutri ent overloading a nd is rich in TOC and OP. Jar Testing Phosphorus r emoval. OP removals of more than 50% were observed for most LCM and ion exchange (IEX) resin treatment experiments on Lake Jesup and Sanford Avenue Canal waters. Sanford Avenue Canal water became the focus of jar test experiments because it had a higher concentration of OP. The mechanisms of IEX and adsorption removes dissolved constituents in the water and the materials achieved the removal of P mainly in the form of OP. Figure 4 2 represe nts the rate of OP removal in Sanford Avenue Canal water. It can be seen that the LCMs can be grouped into general categories of removal : good, moderate, and undesirable or no effect. FA, which added P to the water was an example of undesired effect on the treatment goal. The FA seemed to be a very contaminated material. Alt hough there w ere slight effect s by IS and LS, this can be categorized as having no effect and does not serve as a useful material here, removing

PAGE 41

41 about 515% OP. Moderate removal was ach ieved by SS and FS at about 40% removal. The downward gradient suggests that if given more time they may have the ability to remove more OP AS and RC had the best performance removing greater than 70% OP. In Figure 4 2 AS showed a steep increase in OP rem oval (70% removal) in the first 5 minutes to maintain a plateau region up to 60 minutes. In other words, AS reached its peak removal within 5 minutes and had no effect afterwards. RC removed about 20% OP at 5 minutes and then gradually removed OP providing a final removal of about 80% at 60 minutes. All other LCMs exhibit a low OP removal rates from observing the gradients shown in Figure 42 Though we can see good performance from AS and moderate performance from FS, kinetically, it ended in the addition of TP; refer to F igure 4 3 An observation during jar testing showed that the grains of AS and FS were breaking down and may have caused release of TP. The LCMs were sieved to the same particle size, with exception to FA so not only did the kinetic data suggest that AS had more adsorption sites for P than the other materials, this was confirmed by the surface area studies shown in Table 4 1 previously In jar testing IEX resins were represented by their acronyms followed by its dosage in brackets. For example, PhosX dosed at 1 mL/L was represented as PX (1). From Figure 41, MIEX ( MX ) and Dowex 22 ( DX ) resins dosed at both 1 mL/L and 4 mL/L seemed to have a quick removal rate with DX outperforming MX after 60 minutes. PhosX ( PX ) resin dosed at both 1/mL/L and 4mL/L performed a bit differently in that its initial removal was slow but in the end its capacity proved to be greater exhibiting good removal rates whereas MX and DX achieved their peak performance in the beginning and leveled off towards the end. PX (4) followed a steady removal rate of OP, of about

PAGE 42

42 30% at 5 minutes to 60% at 30 minutes and 80% at 60 minutes. Note that PX(4) was applied to Lake Jesup water from January spiked with OP providing an initial concentration of 127.6 g P/L. MX (4) quickly removed more than 45% at 5 minutes and then show ed a slow removal rate over time. Although low dose s of resin exhibit ed lower OP removal, PX(1) removed more than 50% OP at 60 m inutes and its removal rate was between DX(4) and MX(4) at 60 minutes. Figure 44 helps to deduce that when the bar for Pcomparison is >0 then OP was not the only constituent removed. It can be seen that the IEX resins are not only able to remove OP but other P species can be targeted. This is a major advantage that can supplement the good OP removal. Overall, t he IEX resins removal of TP and OP were achieved with PX having the greatest effect on OP and comparable removals by MX and DX. Furthermore, the 4 g/L of LCMs used is comparable in OP remov al performance to 4 mL/L of the IEX res ins. Inorganic w ater q uality. Abundant growth of algae made the water from Lake Jesup and Sanford Avenue Canal highly turbid which was increased by either the breaking apart of the LCMs during jar testing or precipitation reactions AS, FS, IS, SS, FA, RC, and LS caused turbidities to increase by approximately 11, 25, 8, 15, 124, 33, and 26 NTU respectively at a dose of 4 g/L. The IEX resins resulted in a negligible change in turbidity. Changes in pH were observed and are represented in F igure 4 5 AS, F S, IS, LS and the IEX resins maintained the pH at approximately constant conditions Even t hough SS showed moderate performance in OP removal it raised the pH to 10. RC was a competitor with AS with respect to OP removal, but RC substantially raised the pH to

PAGE 43

43 almost 10 in the Sanford Avenue Canal water. Also, the calcium in RC formed a precipitate with P which was undesirable due to the need of an additional handling process (Baker et al., 1998) The anions, c hlori de and sulfate, are shown in F igure 4 6 T he LCMs did not change the chloride concentration; therefore there is no concern of total dissolved solids added to the water. AS and FS increased the level of sulfate as shown in Figure 46. This was due to the fact that these were the byproducts of water treatment with aluminum sulfate and ferric sulfate salts and hence had sulfate content With respect to the IEX resins, MX and DX proved to be true IEX resins, whereby they added chloride to the water and removed sulfate in a stoichiometric ratio (Blaney and Cinar, 2007; Bolto and Dixon, 2002) However, the PX resin behaved more like an adsorbent, which removed many anionic species Nitrate, in addition to OP is a problem as it is a nutrient and can contribute to eutrophication. However, because there is a natural nitrogen cycle by which nitrogen gets fixed into plants, nitrogen has not been a focus for this project. Overall, MX, PX, AS, and FS showed the most promise because these materials substantially removed OP and had minimal sideeffects on the surf ace waters treated. These materials were defined as P selective materials, and were subject to additional testing as described in subsequent sections. Performance of P Selective Materials Surface Waters Two raw waters were used in this study from different locations in Florida. One came from the Sanford Avenue Canal and the other was from Lake Alice Gainesville, Florida. A local source of water was more economical and time efficient to conduct studies due to accessibility. Alt hough the TOC content of Sanfo rd Avenue Canal is

PAGE 44

44 high er than Lake Alice, the OP:TP and SUVA were comparable. Table 44 shows the water quality of the sample locations. Because of the highly turbid samples, it was necessary to filter the water through a Whattman GF/A filter to remove any particles that ma y clog the column setup. Table 44 shows a substantial decrease in TP and a slight decrease in OP after filtering. It has been previously determined that the materials were more effective with dissolved constituents in the water; therefore, it is not crucial to keep the particulate P suspended in the water. Waters used for jar testing allowed an understanding that the major seasonal trends for Sanford Avenue Canal were increases in OP and TP with a net increase in OP: TP, decrease in TOC, and decrease in SUVA as the year progressed through a dry season Here water batches were obtained from contrasting seasons in the Sanford Avenue Canal, i.e., from dry to wet, but temperatures remained moderately warm (WUnderground, 2009) This resulted i n a significant decrease in OP and TP and an incre ase in TOC from the dry season to the wet season in Florida. This signified dilution of OP whereas the higher concentration of TOC and SUVA suggested stormwater runoff as its source. At the time of collecti ng samples from Lake Alice it ha d close resemblance to a batch of water collected from Sanford Avenue Canal in April whereby SUVA and OP: TP were comparable. Based on these dissolved constituents the use of Lake Alice became valid as it was comparable to Sanford Avenue Canal Lake Alice samples showed increases in TOC and O P during a wet period. Lake Alice is designed to collect stormwater runoff from the University of Florida where rain resulted in more contaminants entering the wetland

PAGE 45

45 Continuously Flowi ng Treatment Targeted P s pecie s OP has been determined as the major targeted P species by the materials used in this project from jar testing experiments. The continuous fl ow column results shown in the Tables 45 and 46 further emphasized that OP was th e targeted P specie because TP versus OP removed is similar throughout treatment. To further emphasize this point, O P r emaining was subtracted from TP remaining and the result remained constant throughout treatment sho wing that other P species were left b ehind while OP was targeted. OP r emoval. From the jar testing an idea of which materials work best was derived. The LCMs, AS and FS, and the IEX resins, MX and PX were tested for their capability to endure continuous treatment. Further screening was done here. Figure 4 7B showed that AS, MX, and PX achieved 80% OP removal in the initial 120 BV of continuous flow However, MXs capacity greatly degraded shortly after 120 BV indicating that all of its ion exchange sites became utilized quickly MX had great removal but it seem ed to work best for the treatment of fixed volume s of water as observed in jar tests FS removed 40% OP at the beginning and continued to treat at 30% which was not sufficient for the large volume of water that needed to be treated and t he treatment goals set out for the project. Figure 4 7 collectively shows that PX and AS achieved cons istent removal of over 50% for larger BV s even under the high ly concentrated OP conditions of water used in Figure 47A and 4 7C Overall, the porous P se le ctive materials, AS and PX, have greater capacities for removing OP from surface water under continuous flow. Sulfate e ffect. Figure 48 shows an important water quality characteristic if mercury (Benoit, 1997) were to be a concern. Sulfate concentration was affected in the

PAGE 46

46 beginning of treatment within the first 1000 BV by both PX and AS. AS increased the sulfate concentration to 231% whereas PX removed 83% of the sulfate in solution. This was reflected in Figure 46 as well which predicted such changes in sulfate for the two materials. After 1000 BV the effect diminished indicating that AS had some free sulfate which needed to be washed or released before application. This is important to note so that it does not affect the water quality of surface water bodies. Cyclically Flowing Treatment Motivation. It w as determined that PX and AS were both porous materials and that porous materials exhibit a greater removal capacity if allowed to rest (Blaney and Cinar, 2007) This is due to internal diffusion, whereby the OP molecules on the surface of the material get time to migrate to the inner pore walls freeing surface area immediately in contact with the flowing water. Another factor that could govern this is desorption of other anions bound to the surface of t he materials freeing adsorption sites. The eventual pilot scale system was envisioned to be powered by solar energy so it was more sensible to have it in operation during the day and off during the night. Therefore, a cyclic 12 hours on and 12 hours off (1 2/12) was examined. OP r emoval. It was found that PX and AS yielded over 50% removal consistently as in the continuous flow studies, but after resting overnight an increase in OP removal was observed at the beginning of the next day as compared to the end of the previous day. Figure 49 shows that AS and PX were capable of enduring twice as much BVs before exhaustion as compared to PX that was subjected to continuous flow without a break. Lake Alice water was used in this experiment because it had been deci ded that it was more feasible to launch the FITS in Lake Alice.

PAGE 47

47 Pilot Scale Studies P r emoval. The construction of a pilot scale facility was found necessary to ensure that the benchscale experiments can be applied in the real environment Based on the fi ndings of the 12/12 experiments it has been determined that giving the materials a rest extends their us eful life. T he FITS was run on solar power conveniently during the day and left to rest overnight, so the material will benefit from this. The results i n F igure 4 10 show that the overal l treatment of TP by the FITS was effective yielding greater than 60% by combini ng the mesocosm and fluidized bed filled with PX for a duration of 9 to 10 days until a 50% breakthrough point was observed. The PX fluidized bed achieved over 70% removal of OP as shown in Figure 4 11 whe n the mesocosm water was considered as the influent The PX seems to have 810 days of good removal with the flow rate of 2 BV/min in Lake Alice. Subtracting the integral of the removal curves from the integral of the influent curves and dividing by the amount of material used in Figures 411A and 411B, the PX can remove between 2.0 to 3.5 mg P per mL PX resin depending on the water chemistry flowing at a rate of 2 BV/min. This removal will al ways depend on the flow rate and competing constituents in the water, such as NOM and sulfate. The difference between the behavior of OP in Figure 411A and 411B was representative of this. Many environmental conditions occurring day by day will affect the system. AS did not remove P as well as predicted by the column studies. This may be due to a change in composition. A second batch of AS was received from the same source as the first but at a different time of year Water passing through a treatment f acility is not uniform so the aluminum sulfate to activated carbon proportions used for treatment and sludge content will differ throughout the year. From here on the second batch of AS

PAGE 48

48 was used for the fluidized beds. Table 4 7 gives a comparison of the f irst batch of AS (AS1 ) and second AS (AS2). The composition changed slightly. Another physical factor that can play a role in the reduction of removal was that the grain sizes were much larger after preparation, than in the benchscale column studies H ence reducing the surface area significantly results in less sites available for adsorption (Mortula and Gagnon, 2007a) It was observed in Figure 411 that the AS did achieve some removal of OP and if the grain sizes were smaller it will yield better resul ts. However, to prepare a large batch of AS at a smaller, effective grain size became time consuming and resulted in waste of material Turbidity and pH. As a note, in the following section, Figure 412 and Figure C 5 show that the mesocosm removed a significant amount of turbidity, which was favorable because clogging of the fluidized bed could be an issue with high turbidities. The lake remained between pH 7 to 9 with little changes to other inorganic water quality by the fluidized beds. Much of the TP i n Lake Alice was contributed by the level of turbidity in the water, which in this case was directly from to biological activity related to algae blooms. Overall, the bench scale column studies can be scaledup to a larger fluidized bed and applied to the real environment. The entire project presented the possibility to accurately predict the behavior of the fluidized beds depending on the material applied and water conditions. The use of water from real surface water systems also helped with understanding how changes in water quality affect performance. Effect of Algal B looms The effects of algal blooms are apparent in a lake. It becomes an aesthetic issue and aquatic life starts to die due to reduced oxygen supply. During the times of algae

PAGE 49

49 blooms the turbidity was higher than prior to algae blooms as shown in Figure C 5. The pH was increased to around 10 as a result of the biological activity, see Figure 413. In comparing Figures 411 and 412 the pH levels track closely with the rising and falling of t urbidity. This is a rather common observation as algae blooms impact the pH and turbidity (Foott et al., 2009) However, the adsorption mechanism is affected by changes in pH whereby water with a high pH has a higher concentration of negatively charged OHions that will then surround the surface sites and compete with OP which is also negatively charged (Zhu and Jyo, 2005) Therefore, a rise in pH will have an adverse effect on the removal capability of the treatment materials used because PX and AS act as adsorbents. This effect was shown in Figures 414 and 415 where there was little to no removal of TP and OP, respectively, by AS and some removal of TP and OP by PX from the beginning of an algae bloom season. Some P was released at the end as the lake w ater spiked above pH 10.5. Overall, the AS seemed to be more greatly affected by the change in pH than PX. Sustainability Studies The attempts to remove P made throughout this project always kept sustainability within arms reach. The sustainable value of considering a LCM as a media of removal is significant, whereby an extra life could be given to the waste byproduct before its inevitable disposal. But, it has been determined that although LCMs can perform well in smallscale batch or continuous treatment the existence of commercially available IEX resins can prove to be easier to implement and provide consistently good efficacy for OP removal.

PAGE 50

50 IEX resins are expensive and it would be a shame if after the resin is exhausted it is disposed of. However, I EX resin producers boast the reusability of these materials (Sylvester and Moller, 2004) In order to reuse them they must be regenerated and as PX was the best candidate for implementing in a fluidized bed for the removal of OP, then the regeneration, reapplication, and disposal of byproducts had been experimented with. Regeneration Before starting the large pilot scale regeneration, it must be determined if regeneration is a possible option. Figure 416 shows a comparison between the fresh (virgin) PX and once regenerated PX in duplicate, which was done in a bench scale column experiment. It was found that the materials tracked each other very closely proving that the PX can be regenerated and effectively reapplied to a system. The largest disagreement was a brief 20% difference in removal. Due to the success of the bench scale regeneration, it was decided to apply PX regeneration to the FITS. From here it was deemed practical to investigate multiple regeneration cycles. Figure 417 proved that PX can be regenerated multiple times. Mesocosm effluent represented influent water here. Regenerations were done three times (3x) and all were in very close agreement with the virgin PX used simultaneously. Over 60% removal was observed at most times until exhaustion in all regeneration cases. Again, having two fluidized beds furnished the convenience of comparing two materials under the same conditions at the same time. Precipitation The re generation process of PX resulted in a dirty caustic brine solution, also known as dirty wash solution ( DWS ), which was very concentrated in all contaminants

PAGE 51

51 that the PX had originally removed and a pH of 14. For example, the duplicate regenerated PX used in F igure 416 resulted in a DWS that contained 49 mg P/L and 14.1 cm1 of UV254 absorbing NOM. It is unusual for an environmental engineer to think that such a solution can be disposed of without reducing the hazardous status of the DWS This defeats the purpose of removing OP and would result in P being rereleased to the environment. To prevent this OP should be removed from th e DWS. In this case, the DWS was titrated with a predetermined molari t y of calcium solution which resulted in a precipitate. This precipitate from past research was most likely hydroxyapatite and/or, in smal ler quantities, CaPO4 specie s because of the high pH conditions that exist (Bowden et al., 2009; Sibrell and Montgomery, 2009) Figure 4 18 shows a variety of DWSs that were treated with the calcium solution significantly reducing the OP in solution; but i f too much was used then OP seemed to become released. Jar 1 and Jar 2 existed from two jar tests done on Lake Alice water for the purpose of attaining two batches of used PX resins to produce two extra batches of DWSs to experiment with. Data for the c olu mn line was derived from precipitating a DWS of a PX resin used in a column study on Sanford Avenue Canal water from April. UV254 was measured throughout the precipitation. Initial UV254 readings varied but were all reduced by 7% to 2 0 % and pH remained around 14. The NOM was not greatly affected but there was some removal maybe due to complexation of the calcium precipitate by the NOM. Reuse of Regeneration Solution The cleaned DWS had a long history starting with the preparation of the solution to cleaning it of contaminants. With such labor involved and the hazard it may cause if poured down the drain, it was thought that maybe the cleaned solution can be used again to regenerate the PX resin. After the OP in the DWS was precipitated, the

PAGE 52

52 precipitate was filtered out of solution. Three quarters of the solution was recovered from this process; filtering the remaining quarter became difficult with the resources available. This quarter was replaced with fresh solution. It was found that the cleaned DWS worked once again to clean the PX of its contaminants. After the regeneration process, this PX batch was reapplied to a fluidized bed of the FITS. The DWS regenerated PX performed as well as the virgin PX that it was simultaneously compared to, yielding over 60% removal until exhaustion; see Figure 419. Therefore, it is possible to reuse the DWS once, after which a decision can be made to precipitate OP, neutralize, remove NOM and dispose of the safer solution. Table 41 Physical Properties of LCMs Parameter A S 1 FS IS SS FA RC LS Al a (mg/kg) 2461 9.5 3084 65.4 1271 1054 2.8 Caa (mg/kg) 1142 10810 12460 36290 6910 28080 34240 Fea (mg/kg) 63.6 801.0 78.0 674.0 503.0 527.0 27.1 Pore sizeb () 76.58 69.06 278.98 69.27 160.9 Surface areab (m2/g) Pore Volu meb (cm3/g) 227.40 0.44 0 6.67 0.01 0 2.05 0.014 2.68 0.004 3.53 0.014 Point of zero charge (pH) 6.18 6.87 9.28 12.46 11.60 11.97 9.47 X ray diffraction crystal structure c Q None C C, Ag, L Q D, C, Q D, C a Plant available quantity b Data v erified for accuracy by measuring a standard powdered activated carbon of known characteristics c Abbreviations: Q = quartz, C = calcite, D = dolomite, Ag = aragonite, L = larnite

PAGE 53

53 Table 4 2 Average Metal Composition (mg/kg) of Low Cost Materials round ed to 3 significant figures Al Ca Fe As Pb Cu Mg Mn Zn FS 2470 26000 149000 18.4 12.0 8.42 863 243 626 SS 13900 9990 0 12400 0. 140 7. 65 27.7 26600 6720 167 AS 93500 2470 2920 122 7.71 276 650 107 107 IS 56900 177000 2210 1. 38 0. 556 4.21 21400 1790 39. 2 RC 4330 139000 3920 5. 49 6. 08 1 5.0 16500 64.7 70. 4 LS 722 155000 880 2. 67 1. 89 5. 26 80500 45.7 656 STD REF 2709a 24200 17100 28100 13. 8 13.0 29. 4 12400 424 1030 a All elements in STD REF 2709 were within NIST 2002 specified ranges (NIST, 2002; USEPA 1996) Table 4 3 Water Quality Characteristics for Lake Jesup and Sanford Avenue Canal Month pH Turbidity TOC SUVA ClSO42NO3 TN OP TP OP: TP 2009 NTU m g C/L L/mg Cm mg/L mg/L mg N/L mg N/L g P/L g P/L Lake Jesup Jan. 7.02 2.86 16.1 3.8 109 23.9 < 0.5 1.0 17.2 111 0.15 Mar. 7.33 15.8 17.3 2.9 175 35.6 < 0.5 1.3 7.3 206 0.04 Apr. 7.40 22.3 17.8 3.2 269 57.1 < 0.5 0.7 5.6 274 0.02 Sanford Avenue Canal Jan. 7.14 2.47 23.2 4.8 122 22.6 0.6 1.2 86.2 137 0.63 Mar. 7.67 3.70 18.3 4.4 18 8 40.6 0.6 0.9 159 194 0.82 Apr. 7.63 4.18 14.8 4.3 197 35.3 0.7 0.1 232 289 0.80 Table 4 4. Water Qualit y Characteristics for Sanford Avenue Canal and Lake Alice Date pH Turb. TOC SUVA ClSO42 NO3 TN OP TP OP : TP 2009 NTU mg C/L L/mg Cm mg/L mg/L mg N/L mg N/L g P/L g P/L Sanford Avenue Canal Apr. Raw 7.59 4.18 14.44 4.44 185.15 33.02 1.28 221.78 329.07 0.67 Filt a 7.74 14.58 4.07 185.15 33.02 1.28 1.22 215.23 250.30 0.86 Jun. Raw 6.93 33.56 4.86 54.51 6.30 9.38 0.84 107.10 1 95.11 0.55 Filt a 7.57 31.48 5.22 54.51 6.30 9.38 0.84 107.10 142.95 0.75 Lake Alice Jul Raw 7.70 7.16 3.19 13.67 18.70 0.49 421.26 481.69 0.87 Filt a 7.70 7.08 3.09 13.6 18.3 0.44 387.12 423.43 0.91 Aug. Raw 7.55 8.26 2.72 15.86 20.76 1.95 506.90 528.85 0.96 Filt a 7.50 8.06 2.83 0.59 372.37 327.27 0.83 a Sample filtered through a Whattman GF/A Filter

PAGE 54

54 Table 4 5. Removal of P by PX using Sanford Avenue Canal April water in BenchScale Continuous Flow Study BV T P Removed, g P/L OP Removed, g P/L TP Remaining, g P/L OP Remaining, g P/L P Remaining, g P/L 0 0.00 0.00 246.90 216.90 30.00 120 183.22 177.64 63.68 39.26 24.42 600 158.59 149.57 88.30 67.33 20.97 1080 144.34 144.61 102.56 72.29 30.27 1560 141.29 137.12 105.61 79.78 25.83 2880 123.98 124.01 122.92 92.89 30.03 3360 107.82 107.50 139.08 109.40 29.68 3840 108.32 109.10 138.58 107.80 30.78 4320 103.06 102.60 143.84 114.30 29.54 5760 97.17 95.40 149.73 121.50 28.23 6240 94.51 97.40 152.39 119.50 32.89 Table 4 6. Removal of P by AS U sing Sanford Avenue Canal April W ater BV TP Removed, g P/L OP Removed, g P/L TP Remaining, g P/L OP Remaining, g P/L P Remaining, g P/L 0 0.00 0.00 250.70 213.50 37.20 120 184.75 187.14 65.95 26.36 39.59 600 175.38 165.73 75.32 47.77 27.55 1080 167.84 157.97 82.86 55.53 27.33 1560 155.64 148.81 95.06 64.69 30.37 2880 129.04 125.80 121.66 87.70 33.96 3360 125.58 124.06 125.12 89.44 35.68 3840 128.37 118.87 122.33 94.63 27.70 4200 133.70 116.78 117.00 96.7 2 20.28 4920 109.92 103.20 140.78 110.30 30.48 5760 104.81 99.40 145.89 114.10 31.79

PAGE 55

55 Table 47. Comparison of AS1 and AS2 Parameter AS 1 AS 2 Al a (mg/kg) 2461.00 3385.00 Ca a (mg/kg) 1142.00 554.00 Fe a (mg/kg) 63.60 46.39 Organic Matter b OM 44.10 % 52.23% a Plant available quantity b Loss on Ignition A B Figure 4 1. Composition of OP of A) Sanford Avenue Canal, and B) Lake Jesup using and Lake Jesup April TP = 274 g P/L.

PAGE 56

56 A B Figure 4 2 Rate of OP removal by A) LCMs (4 g/L) and B) IEX resins (1 and 4 mL/L) in Sanford Avenue Canal water. C/C0 = 2.45 for FA at 60 min. Initial OP = 88.9 to 159.4 g P /L.

PAGE 57

57 A B Figure 4 3 P remaining after 60 min of treatment of Sanford Avenue Canal water for A ) LCMs, and B ) IEX resins. Peak of TPAS = 2.1; TPFS = 1.5; TPFA = 5.9; OPFA = 2.4 Initial TP = 125.3 to 286.6 g P /L; initial OP = 88.9 to 159.4 g P /L.

PAGE 58

58 Figure 4 4 P species removed after 60 minutes tr eatment of Canal Water for IEX resins. Where Pcomparison = (TPr OPr) / TPo, TPr = TP removed, TPo = Initial TP and OPr = OP removed. A B Figure 4 5 Change in pH following 60 min of treatment of Sanford Avenue Canal water for A) LCMs, and B ) IEX resi ns.

PAGE 59

59 A B Figure 4 6 Change in chloride and sulfate following 60 min of treatment of Sanford Avenue Canal water for A ) LCMs, and B ) IEX resins. Initial chloride = 112 to 191 mg/L; initial sulfate = 22.5 to 40.6 mg/L.

PAGE 60

60 A B C Figure 4 7. O P rem aining during benchscale continuous flow treatment by A ) PX and AS, B ) PX, AS1, MX and FS and C ) PX and AS, in different water batches. Initial OP concentrations of Sanford Avenue Canal (SAC) April = 215.23 g P /L and SAC June = 106.83 to 107.63 g P /L LA July = 387.12 g P /L

PAGE 61

61 Figure 48. Sulfate remaining during bench scale continuous flow treatment by PX and AS. Initial sulfate concentrations Lake Alice (LA) July = 18.3 mg/L. Figure 4 9. Cyclic Flow, 12 hours treatment and 12 hours resting pe riod, compared to continuous treatment. Lake Alice water was used from July and August for bench scale continuous and 12/12 experiments respectively. Initial OP concentration of Lake Alice (LA) July = 387.12 g P /L and LA August = 396.15 g P /L

PAGE 62

62 Figur e 410. TP removal by FITSthrough mesocosm and fluidized beds between 10/09/2009 to 10/19/2009 during pilot study. A Figure 411. OP Removal by fluidized beds from mesocosm (influent) water and fluidized beds between A) 10/9/2009 to 10/19/2009, and B) 12/01/2009 to 12/12/2009 during pilot study.

PAGE 63

63 B Figure 411. Continued. Figure 412.Turbidity in Lake Alice (LA) during algae bloom and removal by mesocosm respectively between 10/02/2009 to 10/09/2009 during pilot study.

PAGE 64

64 Figure 413. Change o f pH during algae bloom in Lake Alice (LA) pH between 10/02/2009 to 10/09/2009 during pilot study. Figure 414. TP removal from Lake Alice by FITS between 10/02/2009 to 10/09/2009 during pilot study.

PAGE 65

65 Figure 415. OP removal from Lake Alice by FITS between 10/02/2009 to 10/09/2009 during pilot study. Figure 4 16. Removal of OP by Regenerated PX from the benchscale continuous flow studies. Initial OP for Sanford Avenue Canal (SAC) June = 107.10 g P/L and 90.49 g P/L for PX and Regenerated PX respectively.

PAGE 66

66 A B C Figure 4 17. OP removal from mesocosm (influent) water by batches of PX regenerated A) 1x (11/02/2009 to 11/13/2009), B) 2x (12/15/2009 to 12/21/2009), and C) 3x (01/12/2010 to 01/20/2009) during pilot study.

PAGE 67

6 7 A B Figure 4 18. Volume of 0.1M CaCl added to precipitate A) OP and B) NOM from DWSs from various starti ng concentrations. Initial OP, Jar 1 = 42mg P/L, Jar 2 = 21 mg P/L, Column = 49 mg P/L. Initial UV254 Jar 1 = 1.9 cm1, Jar 2 = 2.6 cm1.

PAGE 68

68 Figure 4 19. OP remo val from mesocosm (influent) water by PX regenerated by cleaned DWS between 11/18/2009 and 11/30/2009. Resin used on FITS.

PAGE 69

69 CHAPTER 5 CONCLUSION Evaluation of Materials Low Cost Materials The low cost materials (LCMs) performed similar to the results depicted in past research. Materials such as recycled concrete (RC) and steel slag (SS) made great improvements in reducing orthophosphate (OP) levels but they significantly changed the pH and resulted in calcium precipitate. As expected, alum sludge (AS) and ferric sludge (FS) are good prospects for the removal of OP because under a fixed volume of water, they both performed well. However, under continuous flowing conditions, AS was the best LCM because of its affinity for OP and the fact that it is very porous due to the activated carbon content, which individually removes P. AS was fairly inert to other water quality characteristics like pH, if carefully prepared for usage. The major disadvantage of using an LCM is that it is infeasible to obtain grain sizes that provide a large enough surface area to be effective. Furthermore, its composition tended to vary in different batches in time and was more sensitive to water quality changes, as was observed in the case of AS. This in itself can introduce a random er ror in any predictive study that is endeavored. In relation to a past experiment done with fly ash (FA), this material had provided 99% P removal (Ugurlu and Salman, 1998) however the batch obtained for this project greatly contaminated the water. In summ ary, LCMs are cheap and their beneficial reuse should be a priority. Therefore, if a good batch is received and tested, it can be mixed into a treatment process if it encompasses an effective grain size because it is difficult to prepare in large quantities.

PAGE 70

70 Commercially Available Resins Commercially available ion exchange (IEX) resins performed in a consistent manner and can engender the possibility of a predictable treatment system. PhosX (PX) resin was proven to be a robust material at removing OP from the surface water, despite competing anions. It had consistently achieved over 60 percent removal of OP when applied to the Floating Island Treatment System (FITS). MIEX (MX) is a great material for treating fixed volumes of water as it immediately exchang es its surface area sites for OP and natural organic matter (NOM) ions in the water. In contrast, this proves to be a disadvantage because when it is applied to a continuously flowing treatment system, it becomes exhausted early and is hence, useless in tr eatment of continuously flowing water. Pilot Application Fluidized beds were built to simulate the small scale column studies. Their inclusions in the treatment process of the FITS have proven to be greatly beneficial. PX and AS passed the screening process, and were deemed the best options of P selective materials, based on OP threshold and effects on other water quality. However, AS did not perform as well as it did in the bench scale experiments mainly, because of the two major disadvantages of LCMs, as described above. PX resin treated the surface waters effectively and consistently. However, the presence of algae blooms increases the pH and can adversely affect the adsorption process of PX. As compared to the column studies, the results were very simil ar because real surface water was used in the screening process. Also, the BV/min were kept constant which proved to be a major controlling factor for replicating the performance observed in bench scale column

PAGE 71

71 studies. Overall, the bench scale column exper iments were successfully scaled up and the concept applied to the pilot study closely resembled the bench scale experiments. Sustainability of Process In addition to the fact that the FITS is solar powered, and safe for the environment, the PX resin can al so be successfully regenerated multiple times and performs just as well as a fresh batch. This is an advantage, with inclusion of great performance, which can counteract its disadvantage of being costly. However, with regeneration comes a dirty brine solut ion, also referred to as dirty wash solution (DWS), contaminated with the P and other constituents that were removed in the treatment process. The P in the solution can be precipitated with calcium, which can then be separated from the solution by filtration and reused in another industry for fertilizer. Afterward, this cleaned solution can be reused to wash another batch of exhausted PX. This PX regenerated with the reused DWS performs well in comparison to fresh PX. Upon final disposal of the DWS, the env ironmental hazards must be considered. The DWS contains not only P but has a high pH, concentrated NOM, and possibly other contaminants. Firstly, P will be precipitated and in this instance, to further treat the DWS, NOM can be completely removed by neutralizing the solution after the precipitate is removed and then possibly treating it with MX. This clean, neutral solution can then be tested for other hazardous contaminants and then disposed of. MX from previous jar testing results and past research presen ted itself as an ideal material for NOM remov al in a fixed volume of water (Banerjee, 2010; Boyer and Singer, 2008) Overall, the design of the treatment process and selection of materials were successful and proved to be sustainable and beneficial to the environment.

PAGE 72

72 Recommendations for Further Research Based on the results of this project it was found that further research must be done to explore and exploit the possibilities that it can be of benefit. Recommendations for future research are outlined bel ow: The fact that OP is the main cause of eutrophication indicates that it must be targeted. Therefore, the oxidation of organic P to convert it to OP may be useful for complete remediation. Processes such as solar i rradiation (Frost and Xenopoulos, 2002) and a c hemical UV combination oxidation will be highly beneficial (Ridal and Moore, 1990) The concept of the FITS is successful but can be optimized to work more effectively. Applying an oxidation process before the fluidized beds, and investigating into different contact times with materials or flow rates (BV/min) are good options. Contact times will help to determine whether there is a relationship between the BV/min and either removal percentage, or duration of treatment, or both. Number of times commer cially available P selective resins can be regenerated and reused. A method to completely clean the DWS before disposal should be developed. A model based on all dissolved constituents present and a materials affinity for them can be developed. This could be used to accurately predict the performance of a treatment system and capacity of a material selected for targeted contaminants. The relationship between NOM and P during treatment may prove to be useful here. This study opens a new perspective on water treatment. Investigations into applying this process for pretreating surface water before entering a drinking water facility may prove to be useful; for example, to remove arsenic from water (Moller and Sylvester, 2009) In Florida, most of the drinking water supply comes from ground water, which is becoming depleted and plans to withdraw water from surface water bodies (WSTB et al., 2009) are currently being considered by municipalities. Conversion from a ground water to a surface water facility or building new facilities may be costly. Therefore, the pretreatment of surface water before entering treatment facilities, such as previously designed ground water plants, may be feasible.

PAGE 73

73 APPENDIX A LOW COST MATERIAL PR OPERTIES Table A 1. Metal Composition (m g/kg) of Low Cost Materials Ag Al As B Ba Be Ca Cd Co Cr Cu Fe Material AVG FS 0.4 2470.7 18.4 5.1 87.0 0.0 26013.3 6.5 18.6 9.0 8.4 148529.9 AVG SS 0.4 13869.0 0.1 31.1 58.7 0.0 99906.3 4.3 3.1 637.8 27.7 123475.8 AVG AS 0 93468.9 121.9 2 0.8 17.6 0.0 2468.1 0.0 0.0 80.2 276.1 2921.9 AVG IS 0 56868.8 1.4 75.7 322.4 6.5 177231.3 0.0 3.3 22.2 4.2 2207.9 AVG RC 0 4326.9 5.5 24.3 31.6 0.4 139020.4 0.2 1.9 14.3 15.0 3923.3 AVG LS 0 722.3 2.7 14.0 1.6 0.0 154883.9 0.0 0.0 10.0 5.3 880.2 STDRE F 2709 a 0.6 24159.7 13.8 32.9 342.2 0.4 17066.0 0.9 11.0 71.1 29.3 28108.1 K Mg Mn Mo Na Ni Pb Sb Sn Sr V Zn Material AVG FS 292.3 863.2 242.7 109.8 0.0 13.3 12.0 7.6 4.5 92.4 479.1 626.1 AVG SS 576.2 26563.2 6722.9 10.0 0.0 6.4 7.6 2.5 0 .9 134.4 759.3 166.9 AVG AS 495.5 650.2 107.3 56.4 6467.7 9.2 7.7 0.0 22.0 315.3 29.0 106.6 AVG IS 3365.1 21396.2 1790.5 0.0 1706.2 1.4 0.6 0.0 0.0 281.6 24.8 39.2 AVG RC 478.2 16541.8 64.7 2.8 2005.1 7.2 6.1 0.4 2.9 214.3 15.2 70.4 AVG LS 309.8 80445. 0 45.7 1.1 2563.5 3.7 1.9 0.0 1.3 156.1 9.7 655.7 STDREF 2709 a 4162.6 12425.3 424.4 0.0 783.1 65.3 13.0 0.6 0.0 91.6 63.9 1026.8 a All elements in STD REF 2709 were within NIST 2002 specified ranges (NIST, 2002; USEPA, 1996) Figure A 1. Elemental s pectrum and dot map of Steel Slag done by scanning electron microscopy (SEM) and energy dispersive x ray fluorescence elemental microanalysis. This helped to identify calcium silicate as a dominant mineral.

PAGE 74

74 A B Figure A 2. X ray diffraction peak graphs o f A) alum sludge, B) fly ash, C) ferric sludge, D) iron slag, E) limestone, F) recycled concrete, and G) steel slag showing mineral compositions

PAGE 75

75 C D Figure C 2. Continued.

PAGE 76

76 E F Figure C 2. Continued.

PAGE 77

77 G Figure C 2. Continued.

PAGE 78

78 APPENDIX B COLUMN TE STING RESULTS A B Figure B 1. Change in A) Chloride, and B) Sulfate during column testing by PhosX (PX) and Alum Sludge (AS). Initial Chloride = 193.84 mg/L, Sulfate = 29.7 mg/L.

PAGE 79

79 Figure B 2. Change in pH during column testing by PhosX (PX) and Alum Sludge (AS). Figure B 3. Change in pH during column testing by PhosX (PX), MIEX (MX), Alum Sludge (AS), and Ferric Sludge (FS). Figure B 4. Change in Total Phosphorus (TP) during column testing by PhosX (PX), MIEX (MX), Alum Sludge (AS), and Ferric Sludg e (FS). Initial TP = 125.99 ug P/L to 152.87 g P /L

PAGE 80

80 APPENDIX C FLOATING ISLAND TREA TMENT SYSTEM RESULTS

PAGE 81

81 Figure C 1. Phosphorus speciation throughout operation of Floating Island Treatment System showing treatment by mesocosm and PhosX packed fluidi zed bed. Figure C 2. Total Phosphorus Removal by Floating Island Treatment System through Mesocosm and Fluidized beds.

PAGE 82

82 Figure C 3. Orthophosphate Removal by Floating Island Treatment System through virgin PhosX and Regenerated PhosX Fluidized beds. Figure C 4. Change in pH in Lake Alice throughout operation of the Floating Island Treatment System.

PAGE 83

83 Figure C 5. Change in turbidity in Lake Alice and removal by mesocosm throughout operation of the Floating Island Treatment System. Figure C 6. OP r emoval by fluidized beds from mesocosm (influent) water comparing PhosX (PX) dosed at 0.25 gals flowing at 0.5 gpm for 12 hrs per day until total exhaustion of PX between 03/23/2010 to 04/10/2010 during pilot study.

PAGE 84

84 LIST OF REFERENCES Apell, J.N., 2009. Combined Ion Exchange for the Simultaneous Removal of Dissolved Organic Matter and Hardness, Environmental Engineering Sciences. University of Florida, Gainesville, p. 73. Bach, M.T., Mazyck, D.W., 2007. Strategies for overcoming pH excursions for reactivated granular activated carbon: Air and carbon dioxide treatments. Environmental Engineering Science 24, 1266 1272. Baker, M.J., Blowes, D.W., Ptacek, C.J., 1998. Laboratory Development of Permeable Reactive Mixtures for the Removal of Phosphorus from Onsite Wastewater Disposal Systems. 32, 23082316. Banerjee, P., 2010. Characterization of NOM and tracking the changes in its composition while removing phosphorus from surface water, Environmental Engineering Sciences. University of Florida, Gainesville. Benoi t, J.M., 1997. The Effect of Sulfate and Sulfide on Mercury Methylation in Florida Everglades. EPA Grant number: U915152. Blaney, L.M., Cinar, S., 2007. Hybrid anion exchanger for trace phosphate removal from water and wastewater. Water Research 41, 16031 613. Bolto, B., Dixon, D., 2002. Removal of natural organic matter by ion exchange. Water Research 36, 50575065. Bowden, L.I., A.P., J., Younger, P.L., Johnson, K.L., 2009. Phosphorus Removal from Waste Waters Using Basic Oxygen Steel Slag. Environmental Science & Technology 43, 24762481. Boyer, T.H., Singer, P.C., 2008. Stoichiometry of Removal of Natural Organic Matter by Ion Exchange. Environmental Science & Technology 42, 608613. Dayton, E.A., Basta, N.T., 2003. Using treatment residuals to reduce phosphorus in agricultural runoff. American Water Works Association 95, 151158. FDEP, 2003. Middle St. Johns Basin. Group 2 Basin Status Report. FDEP, 2008. 2008 Integrated Water Quality Assessment for Florida, 2008_Integrated_Report.pdf. Foott, J.S., Stone R., Fogerty, R., 2009. Effects of simulated algal bloom pH on juvenile lost river sucker energetics and growth. U.S. Fish and Wildlife Service, Anderson. Frost, P., Xenopoulos, M., 2002. Ambient solar ultraviolet radiation and its effects on phosphorus f lux into boreal lake phytoplankton communities. Can. J. Fish. Aquat. Sci. 59, 10901095.

PAGE 85

85 Guan, X., Shang, C., 2006. Competitive adsorption of organic matter with phosphate on aluminum hydroxide. Colloid and Interface Science 296, 5158. Harris, W.G., White G.N., 2008. X ray diffraction techniques for soil mineral identification. Soil Sci. Soc. Am. Jia, Y., 2007. Hydrologic and Water Quality Modeling of the Lake Jesup Watershed Using Hydrological Simulation Program Fortran. Johansson, L., 1999. Industrial by products and natural substrata as phosphorus sorbents. Environmental Technology 20, 309316. Jones, M., Eilers, J., Kann, J. 2006. Water Quality Effects of BlueGreen Algal Blooms in Diamond Lake, Oregon. Advancing the Fundamental Sciences 1, 102110. MacPhee, M., Charles, G., Cornwell, D., 2001. Sludge Disposal Options. EPA, pp. 4350. McCormick, P., Newman, S., 2009. Landscape responses to wetland eutrophication: loss of slough habitat in the Florida Everglades, USA. Hydrobiologia 621, 105114. Mergen M.R., B., J., 2008. Magnetic ion exchange resin treatment: Impact of water type and resin use. Water Research 42, 19771988. Mortula, M., Gagnon, G.A., 2007a. Phosphorus adsorption by naturally occurring materials and industrial by products. Environmental Engineering and Science 6, 157164. Mortula, M.M., Gagnon, G.A., 2007b. Phosphorus adsorption and oven dried alum residual solids in fixed bed column experiments. Environ. Eng. Sci. 6, 623628. NIST, 2002. Certificate of Analysis, Standard Reference Material 2709. Pidou, M., Avery, L., 2008. Chemical solutions for greywater recycling. Chemosphere 71, 147155. Ridal, J.J., Moore, R.M., 1990. A re examination of the measurement of dissolved organic phosphorus in seawater. Marine Chemistry 29, 1931. SFWMD, 2009. Surface Water Improvement and Management Program (SWIM), Watershed Management. Southwest Florida Water Management District. SJRWMD. (2008). Lake Jesup total phosphorus removal treatment technologies floating island pilot project. Presentation. Gaines vlle, FL.

PAGE 86

86 Sibrell, P.L., Montgomery, G.A., 2009. Removal of phosphorus from agricultural wastewaters using adsorption media prepared from acid mine drainage sludge. Water Research 43, 22402250. Sylvester, P., Moller, T., 2004. Phosphate removal from wastewater using a regenerable adsorption media. IWA Publishing, London. Tennant, M.F., Mazyck, D.W., 2007. The role of surface acidity and pore size distribution in the adsorption of 2methylisoborneol via powdered activated carbon. Carbon 45, 858864. Ugurlu, A., Salman, B., 1998. Phosphorus Removal by Fly Ash. Environment International 24, 911918. USEPA, 1996. Method 3050B, Acid Digestion of Sediments, Sludges and Soils, Rev. 2 ed. Vollenweider, R. A., Rast, W., Kerekes, J., 1980. The Phosphorus Loading Concept and Great Lakes Eutrophication, Phosphorus Management Strategies for Lakes pp. 207234. Weng, L., Van Riemsdijk, W.H., 2008. Humic Nanoparticles at the Oxide Water Interface: Interactions with Phosphate Ion Adsorption. Environmental Science & Technology 42, 87478752. WSTB, DELS, NRC, 2009. Review of the St. Johns River Water Supply Impact Study. The National Academies Press, Washington. WUnderground, 2009. Season Weather Averages for Orlando S anford (KSFB), Weather Underground. Zhu, X., Jyo, A., 2005. Columnmode phosphate removal by a novel highly selective adsorbent. Water Research 39, 23012308.

PAGE 87

87 BIOGRAPHICAL SKETCH Amar Prem Persaud was born in 1984 in London, England. He attended elementar y school in New Jersey, USA and then The Bishops' High School in Guyana, South America. He attained a Diploma in Civil Engineering from the University of Guyana in 2003 and later graduated with a B.S. in Civil Engineering at Florida Atlantic University in 2006 at which time was awarded Outstanding Undergraduate Leadership. He then ventured off into broadening his horizons by working in the engineering consulting industry for two years. This started in Florida where he did structural and geotechnical inspect ions with GFA International until the end of 2006. The opportunity then arose to gain project management experience in Guyana with Vikab Engineering Consultants, which he could not help but take advantage of. The duties for this job became relaxed after the third quarter of 2007, so with his ambition to gain progressive experience he relocated to California to work with Lei ghton and Associates conducting underground explorations, slope stability analys es and other geotechnical engineering work specific to t he Californian terrain. The worlds practices were leaning toward a sustainable environment; therefore, in 2008 he relocated to Florida and committed himself to attaining a M.E. in environmental engineering with a focus on the h ydrological sciences ecolog y, and policy at the University of Florida. Finances bec a me tight as savings dwindled when at that moment he met his new advisor, Dr. Treavor Boyer, in short pants at the Einstein Bros. Bagel on the UF campus and a light at the end of the tunnel became vis ible With both of them st arting a new career the initial research goal s to manage the physical chemical experimentation phase and help develop a sustainable phosphorus removal system for surface waters w ere presented by the St. John's River Water Managem ent District. "With Amar's experience, he is the man for the

PAGE 88

88 job," said Dr. Boyer. This research later expanded into a Masters Thesis focusing on phosphorus removal from organic rich surface water bodies. His main career goal is to get involved in constru ction and environmental planning by adopting a synergistic approach to the development of civilization with the security of natural resources.