<%BANNER%>

Stabilized Landfill Leachate Treatment Using Physico-Chemical Treatment Processes

Permanent Link: http://ufdc.ufl.edu/UFE0041629/00001

Material Information

Title: Stabilized Landfill Leachate Treatment Using Physico-Chemical Treatment Processes Coagulation, Anion Exchange, Ozonation, Membrane Filtration, and Adsorption
Physical Description: 1 online resource (144 p.)
Language: english
Creator: SINGH,SHRAWAN KUMAR
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ADSORPTION -- COAGULATION -- LANDFILL -- LEACHATE -- MEMBRANE
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Leachate is generated by the percolation of rainwater through the layers of wastes in a landfill. A combination of several physical, chemical, and microbial processes in the landfill results in contaminants being transferred from the waste to the leachate. Several wastewater treatment processes have been studied to effectively treat contaminants in leachate. Reliability, simplicity, and cost efficiency make biological treatment the most commonly used leachate treatment method. However, these methods are not very effective in treating leachate containing less biodegradable organics drawn out from stabilized landfills. Physico-chemical treatment methods have been found more effective in treating such leachate; however, studies on leachate treatment by physico-chemical treatment methods suggest that use of a single treatment or a combination of more than one treatment process often does not produce effluent water that meets water quality standards for direct discharge to groundwater or surface water. Among the physico-chemical processes, membrane filtration technologies, especially reverse osmosis (RO), has been found effective in removing the contaminants from leachate to meet effluent discharge regulations; however, increased fouling of these membranes due to the presence of high concentrations of organic matter and salts reduces the cost effectiveness of these systems. Reducing the concentration of leachate organic matter prior to membrane treatment was hypothesized to reduce the fouling frequency of membranes. The objective of this doctoral research was to evaluate the effectiveness of coagulation, anion exchange and ozonation as a pretreatment option for treating stabilized leachate using high pressure nano-filtration (NF) and reverse osmosis (RO) membranes. Additionally, adsorption of organic matter using activated carbon (AC) has been traditionally used in water and wastewater treatment; however, the selectivity of AC based on their pore sizes (micro-porous and meso-porous) for leachate treatment has never been investigated. Part of this study therefore focused the organic matter adsorption profile of three different ACs for stabilized leachate treatment. A maximum of 70%, 34%, and 23% reduction in DOC was observed after Fe (III), MIEX, and ozone treatment, respectively. Coagulation (Fe (III)), magnetic ion exchange resin (MIEX), and ozonation effectively removed humic and fulvic-like organic matter from stabilized leachate; more than 60% removal was observed for these types of organic matter. However, these pretreatments did not improve in the permeate flux of NF and RO membranes as compared to raw leachate treatments. In the coagulation and MIEX pretreated leachate, the pH of the feed leachate increased throughout the experiment. This was suspected to have caused excessive precipitation of carbonate minerals as compared to the raw leachate treatment on the membrane surface, causing a greater decline in permeate flux. During the membrane treatment of ozone pretreated leachate, the divalent ions present in the ozonated leachate are hypothesized to have coagulated with the by-products formed during ozonation, with these coagulated solids then precipitating on the membrane surface, leading to a greater permeate flux decline as compared to the raw leachate treatment. The adsorption study showed that the micro and meso-porous AC had similar organic matter adsorption capacity (0.2 g TOC/g AC) during stabilized leachate treatment with faster organic matter diffusion in meso-porous AC. A maximum of 80% TOC was removed from both types of AC.
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 SHRAWAN KUMAR SINGH.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Townsend, Timothy G.
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 2011
System ID: UFE0041629:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041629/00001

Material Information

Title: Stabilized Landfill Leachate Treatment Using Physico-Chemical Treatment Processes Coagulation, Anion Exchange, Ozonation, Membrane Filtration, and Adsorption
Physical Description: 1 online resource (144 p.)
Language: english
Creator: SINGH,SHRAWAN KUMAR
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ADSORPTION -- COAGULATION -- LANDFILL -- LEACHATE -- MEMBRANE
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Leachate is generated by the percolation of rainwater through the layers of wastes in a landfill. A combination of several physical, chemical, and microbial processes in the landfill results in contaminants being transferred from the waste to the leachate. Several wastewater treatment processes have been studied to effectively treat contaminants in leachate. Reliability, simplicity, and cost efficiency make biological treatment the most commonly used leachate treatment method. However, these methods are not very effective in treating leachate containing less biodegradable organics drawn out from stabilized landfills. Physico-chemical treatment methods have been found more effective in treating such leachate; however, studies on leachate treatment by physico-chemical treatment methods suggest that use of a single treatment or a combination of more than one treatment process often does not produce effluent water that meets water quality standards for direct discharge to groundwater or surface water. Among the physico-chemical processes, membrane filtration technologies, especially reverse osmosis (RO), has been found effective in removing the contaminants from leachate to meet effluent discharge regulations; however, increased fouling of these membranes due to the presence of high concentrations of organic matter and salts reduces the cost effectiveness of these systems. Reducing the concentration of leachate organic matter prior to membrane treatment was hypothesized to reduce the fouling frequency of membranes. The objective of this doctoral research was to evaluate the effectiveness of coagulation, anion exchange and ozonation as a pretreatment option for treating stabilized leachate using high pressure nano-filtration (NF) and reverse osmosis (RO) membranes. Additionally, adsorption of organic matter using activated carbon (AC) has been traditionally used in water and wastewater treatment; however, the selectivity of AC based on their pore sizes (micro-porous and meso-porous) for leachate treatment has never been investigated. Part of this study therefore focused the organic matter adsorption profile of three different ACs for stabilized leachate treatment. A maximum of 70%, 34%, and 23% reduction in DOC was observed after Fe (III), MIEX, and ozone treatment, respectively. Coagulation (Fe (III)), magnetic ion exchange resin (MIEX), and ozonation effectively removed humic and fulvic-like organic matter from stabilized leachate; more than 60% removal was observed for these types of organic matter. However, these pretreatments did not improve in the permeate flux of NF and RO membranes as compared to raw leachate treatments. In the coagulation and MIEX pretreated leachate, the pH of the feed leachate increased throughout the experiment. This was suspected to have caused excessive precipitation of carbonate minerals as compared to the raw leachate treatment on the membrane surface, causing a greater decline in permeate flux. During the membrane treatment of ozone pretreated leachate, the divalent ions present in the ozonated leachate are hypothesized to have coagulated with the by-products formed during ozonation, with these coagulated solids then precipitating on the membrane surface, leading to a greater permeate flux decline as compared to the raw leachate treatment. The adsorption study showed that the micro and meso-porous AC had similar organic matter adsorption capacity (0.2 g TOC/g AC) during stabilized leachate treatment with faster organic matter diffusion in meso-porous AC. A maximum of 80% TOC was removed from both types of AC.
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 SHRAWAN KUMAR SINGH.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Townsend, Timothy G.
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 2011
System ID: UFE0041629:00001


This item has the following downloads:


Full Text

PAGE 1

1 STABILIZED LANDFILL LEACHATE TREATMENT USING PHYSICO CHEMICAL TREATMENT PROCESSES : COAGULATION, ANION EXCHANGE, OZONATION, MEMBRANE FILTRATION AND ADSORPTION By SHRAWAN KUMAR SINGH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 1

PAGE 2

2 201 1 Shrawan Kumar Singh

PAGE 3

3 To my parents who sacrificed their present for my future

PAGE 4

4 ACKNOWLEDGMENTS I cannot thank enough to my mentor and committee chairman Dr. Timothy G. Townsend for providing me a wonderful opportunity to work in one of the leading solid waste management group s in the USA. It was his continuous guidance and encouragement that kept me motivated throughout the degree program to be focused and work hard towards the ultimate goal. I am and will be indebted to Dr. Townsend for the efforts and time he spent in train ing me not only on academic and research front s but also giving me several lessons of multi tasking time management, presentation skills, professionalism, and problem assessment and their solving skills. I will be highly grateful to Dr. Paul Chadik, who has given me a lot of selfless support and provided me the lab space along with several of the instruments such as an ozone generator, lab scale membrane setup etc, which were the essential elements of my research. I am equally grateful to Dr. David Mazy ck and Dr. Tre a v o r Boyer, who were always there to help guide me during my experiment design and data analysis with their expert opinions. I am very thankful to my committee members Dr. John Sansalone and Dr. El Shall Hassan who guided me throughout my gr aduate studies and the research. I acknowledge the support of my friends and fellow graduate students, post docs who provided me a commendable support in planning research, conducting field work, and providing valuable inputs on data analysis and for bei ng there in hours of need. With a great advisor and his wonderful research group I never felt to be half a globe away from my family and friends in India. I would also like to thank the sources of my inspiration, my parents Sri Satya Narain Singh and Sm t. Krishna Devi, parents in law Dr. S.P. Sharma and Smt. Pushpa

PAGE 5

5 Sharma, and brothers Sri Pawan Singh and Sri Vikas Singh for their love, relentless support and never say die spirit without which I would have never reached this point. Words of thanks do not thank enough for the support and strength I have got from my lovely wife Dr. Bhawana Sharma for constantly motivating me to do better and better in my tough times. In all the tough and frustrating times, she was always there to prevent me going down and making me even strong er and focused. Last but not the least; I would also like to thank several of my g ator f riends who have been there for me and my wife, like a family. Their companionship helped us live for four years without meeting our other family members back in India.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 1.1 Background and Problem Statement ................................ ................................ 15 1.2 Research Objectives ................................ ................................ ......................... 21 1.3 Research Approach ................................ ................................ .......................... 23 1.4 Outline of Dissertation ................................ ................................ ....................... 24 2 EVALUATION OF COAGULATION AND ANION EXCHANGE AS A PRETREATMENT FOR STABILIZED LANDFILL LEACHATE TREATMENT USING HIGH PRESSURE MEMBRANES ................................ .............................. 27 2.1 Introduction ................................ ................................ ................................ ....... 27 2.2 Experimental Materials Methods and Analysis ................................ ................. 31 2.2.1 Materials ................................ ................................ ................................ .. 31 2.2.2 Coagulation Experiment ................................ ................................ .......... 32 2.2.3 MIEX Experiment ................................ ................................ .................... 33 2.2.4 Membrane Experiment ................................ ................................ ............ 34 2.2.5 Analytical Methods ................................ ................................ .................. 36 2.3 Results and Discussion ................................ ................................ ..................... 38 2.3.1 Leachate Characteristics ................................ ................................ ......... 38 2.3.2 Leachate Treatment Using Coagulant ................................ ..................... 39 2.3.3 MIEX Experiment ................................ ................................ .................... 42 2.3.4 Comparison of Coagulation to MIEX Treatment ................................ ...... 43 2.3.5 Leachate Treatment Using Membranes ................................ .................. 44 2.4 Conclusions ................................ ................................ ................................ ...... 47 3 EFFECT OF OZONATION AS A PRETREATMENT FOR STABILIZED LANDFILL LEACHATE TREATMENT USING HIGH PRESSURE MEMBRANES 62 3.1 Introduction ................................ ................................ ................................ ....... 62 3.2 Experimental Material Methods and Analysis ................................ ................... 65 3.2.1 Materials ................................ ................................ ................................ .. 65 3.2.2 Landfill Leachate Ozonation ................................ ................................ .... 67 3. 2.3 Landfill Leachate Treatment Using Membrane ................................ ........ 68

PAGE 7

7 3.2.4 Analytical Methods ................................ ................................ ................. 69 Model ................................ ................................ ................................ ............ 71 3.3 Results and Discussion ................................ ................................ ..................... 72 3.3.1 Leachate Characterization ................................ ................................ ....... 72 3.3.2 Ozonation ................................ ................................ ................................ 73 3.3.3 Leachate Treatment Using Membranes ................................ .................. 78 3.4 Summary and Conclusions ................................ ................................ ............... 81 4 EQUILIBRIUM AND INTRAPARTICLE DIFFUSION OF STABILIZED LANDFILL LEACHATE ONTO MICRO AND MES O POROUS ACTIVATED CARBON ........... 94 4.1 Introduction ................................ ................................ ................................ ....... 94 4.2 Experimental Material and Methods ................................ ................................ .. 97 4.2.1 Landfill Leachate and Activated Carbons ................................ ............... 97 4.2.2 Batch Experiments ................................ ................................ ................. 98 Isotherm experiment ................................ ................................ .................. 98 Diffusivity experiment ................................ ................................ ................. 98 4.2.3 Rapid Small Scale Column Tests ................................ ........................... 99 4.2.4 Analysis ................................ ................................ ................................ 100 4.3 Results and Discussion ................................ ................................ ................... 101 4.3.1 Adsorption Equilibrium Study ................................ ............................... 101 4.3.2 Intra particle Diffusivity ................................ ................................ .......... 105 4.3.3 Column Experiment ................................ ................................ .............. 109 4.4 Summary and Conclusions ................................ ................................ ............. 110 5 SUMMARY AND CONCLUSIONS ................................ ................................ ........ 121 5.1 Su mmary ................................ ................................ ................................ ........ 121 5.2 Conclusions ................................ ................................ ................................ .... 126 5.3 Future work ................................ ................................ ................................ ..... 127 APPENDIX A EFFEC T OF COAGULANT DOSE ON FILTRATE pH AND DOC ........................ 129 B REACTION MECHANISM OF OZONE AND HYDROXYL RADICALS TO ORGANIC MATTER ................................ ................................ ............................. 131 LIST OF REFERENCES ................................ ................................ ............................. 133 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 144

PAGE 8

8 LIST OF TABLES Table page 2 1 Previous studies on leachate treatment using coagulation process. .................. 50 2 2 Properties of NF 90 and BW 30 membranes used in the study ......................... 51 2 3 Physico chemical characteristics of ACL, NCL, and NRL leachate during the study period ................................ ................................ ................................ ........ 51 2 4 Percentage distribution and total volume (V t ) of DOM in Region I to Region V of ACL, NCL, and NRL leachate derived from FRI analysis ............................... 51 2 5 Percentage rejection 1 of salts 2 and DOM 3 in permeate in RO and NF experiment ................................ ................................ ................................ ......... 51 3 1 Previous studies on application of ozonation for landfill leachate treatment. ...... 84 3 2 Properties of reverse osmosis (BW 30) and nano filtration (NF 90) membranes used in the study ................................ ................................ ............. 85 3 3 Leachate characteristics of ACL, NCL, and NRL leachate during the study period ................................ ................................ ................................ ................. 85 3 4 Percentage distribution and total volume (V t ) of DOM in Region I to Region V of ACL, NCL, and NRL leachate derived from FRI analysis ............................... 85 3 5 Percentage rejection 1 of salts 2 and DOM 3 from raw and ozonated ACL, NCL, and NRL leachate treatment using RO and NF membranes .............................. 85 3 6 Power law equation and the value of r 2 of flux decline with ozonation time for raw and ozonated leachate treatment using RO and NF membranes ................ 86 3 7 The correlation coefficient (r 2 ) and the filtration constants of the Hermia's filtration model for raw and ozonated ACL, NCL, and NRL leacha te treatment using RO membranes ................................ ................................ ......................... 86 3 8 The correlation coefficient (r 2 ) and the filtration constants of the Hermia's filtration model for raw and ozonated ACL, NCL, and NRL leachate treatment using NF membranes ................................ ................................ ......................... 86 4 1 Previous studies on landfill leachate trea tment using activated carbon adsorption process ................................ ................................ ........................... 112 4 2 Physico chemical characteristics of ACL leachate during the study period ...... 113 4 3 Characteristics of Calgon F 300, Norit HD 4000, and Darco 12x40 AC ........... 113

PAGE 9

9 4 4 Design parameters of full scale and RSSCT ................................ .................... 113 4 5 Compar ative empty bed contact time (EBCT) of full scale and RSSCT columns for Calgon F 300, Norit HD 4000, and Darco 12x40 AC .................... 113 4 6 Isotherm parameters obtained using non linear method for organic matter absorption onto Calgon F 300, Norit HD 4000, and Darco 12x40 AC .............. 114 4 7 Boundary layer thickness (C) of different particle sizes of Calgon F 300, Norit HD 4000, and Darco 12x40 AC ................................ ................................ ........ 114 4 8 Effe ctive diffusion coefficient (De) of organic matter onto different particle sizes of Calgon F 300, Norit HD 4000, and Darco 12x40 AC ........................... 114 4 9 Volume distribution of DOM in region I to region V of leachate derived from FRI analysis ................................ ................................ ................................ ...... 114

PAGE 10

10 LIST OF FIGURES Figure page 1 1 (a) Reaction that hydrolysis products of coagulants follow when coagulant is added to water containing natural organic matter. (b) MIEX resin anion exchange process. (c) Model reacti on of ozonation of unsaturated organic compound ................................ ................................ ................................ ........... 2 6 2 1 Reaction that hydrolysis products of coagulants follow when coagulant is added to water cont aining natural organic matter. ................................ .............. 53 2 2 MIEX resin anion exchange process. ................................ ................................ 53 2 3 Laboratory scale Osmonics SEPA CF membrane experimental setup .............. 54 2 4 Operationally defined excitation and emission wavelength boundaries for five EEM regions ................................ ................................ ................................ ....... 54 2 5 Fluorescence EEMs for leachate samples (a) ACL leachate (b) NCL leachate (c) NRL leachate ................................ ................................ ................................ 55 2 6 Effect of coagulant dose on (a) pH, (b) DOC and (c) SUVA of ACL, NCL, and NRL leachate. ................................ ................................ ................................ ..... 56 2 7 Effect of coagulation on removal of DOM in each region derived from FRI analysis of ACL, NCL, and NRL leachate ................................ ........................... 57 2 8 Effect of MIEX dose and mixing time on UV 254 absorbance of (a) ACL, (b) NCL, and (c) NRL leachate. ................................ ................................ ................ 58 2 9 Effect of MIEX treatment on (a) DOC, (b) SUVA, and (c) pH of ACL, NCL, and NRL leachate. ................................ ................................ .............................. 59 2 10 Effect of MIEX on removal o f DOM of each region derived from FRI analysis of ACL, NCL, and NRL leachate. ................................ ................................ ........ 60 2 11 Effect on normalized permeate flux as a function of filtration time for pretreated (FeCl 3 and MIEX) and raw ACL leachate using NF 90 and BW 30 membrane ................................ ................................ ................................ .......... 61 2 12 Effect on f eed pH as a function of time for pretreated (FeCl 3 and MIEX) and raw ACL leachate using NF 90 and BW 30 membrane operation ...................... 61 3 1 Sc hematic diagram of laboratory scale ozonation setup ................................ .... 88 3 2 Experimental setup of laboratory scale Osmonics SEPA CF membrane ........... 88

PAGE 11

11 3 3 Operationally defined excitation and emission wavelength boundaries for five regions of excitation emis sion matrix ................................ ................................ .. 89 3 4 Fluorescence EEM of (a) ACL, (b) NCL, and (c) NRL leachate .......................... 89 3 5 Evolution of the off gas ozone concentration as a function of ozonation time for ACL, NCL, and NRL leachate ................................ ................................ ........ 90 3 6 Effect of ozonation on (a) pH, (b) UV 254 absorbance, and (c) DOC of ACL, NCL, and NRL leachate. ................................ ................................ ..................... 91 3 7 Effect of ozonation on removal of DOM for each EEM region derived from FRI analysis of (a) ACL, (b) NCL, and (c) NRL leachate ................................ .... 92 3 8 Effect on normalized permeate flux for treatment of raw and ozonated (a) ACL, (b) NCL, and (c) NRL leachate using BW 30 and NF 90 membrane ......... 93 4 1 Adsorption of leachate onto (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. ................................ ................................ .............................. 115 4 2 Weber and Moris intra particle diffusion for removal of TOC using (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. ................................ ......... 116 4 3 Homogeneous particle diffusion model for TOC removal in leachate using (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. ............................ 117 4 4 Adsorption kinetics of organic matter at different particle sizes (a) Calgon F 300, (b) Norit HD 4000, an d (c) Darco 12x40 AC. ................................ ............ 118 4 5 Effect of EBCT on the TOC removal of leachate using Calgon F 300, Norit HD 4000, and Darco 12x40 AC. ................................ ................................ ....... 119 4 6 Fluorescence EEM of ACL leachate and operationally defined E x E m boundaries for five EEM regions. ................................ ................................ ...... 119 4 7 Effect of EBCT on DOM of each EEM region while leachate treatment using (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. ....................... 120 A 1 Effect of coagulant dose on pH and DOC of ACL, N CL, and NRL leachate. .... 130

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy STABILIZED LANDFILL LEACHATE TREATMENT USING PHYSICO CHEMICAL TREATMENT PROCESSES : COAGULATION, ANION EXCHANGE, OZONATION, MEMBRANE FILTRATION AND ADSORPTION By Shrawan Kumar Si ngh May 201 1 Chair: Timothy G. Townsend Major: Environmental Engineering Sciences Leachate is generated by the percolation of rainwater through the layers of wastes in a landfill. A combination of several physical, chemical, and microbial processes in the landfill results in contaminants being transferred from the waste to the leachate Several wastewater treatment processes have been studied to effectively treat contaminants in leachate. Reliability, simplicity, and cost efficiency make biological treatment the most commonly used leachate treatment method. However, these methods are not very effective in treating leachate containing less biodegradable organics drawn out from stabilized landfills. P hysico chemical treatment methods have been found more effective in treating such leachate ; however, s tudies on leachate treatment by phy sico chemical treatment methods suggest that use of a single treatment or a combination of more than one treatment process often does not produce effluent water that meet s water quality standards for direct discharge to groundwater or surface water. Amon g the physico chemical processes, membrane filtration technologies especially reverse osmosis (RO) has been found effective in removing the contaminants from leachate to meet effluent discharge regulations ; however, increased

PAGE 13

13 fouling of these membranes d ue to the presence of high concentrations of organic matte r and salts reduces the cost effectiveness of these systems. R educing the concentration of leachate organic matter prior to membrane treatment was hypothesized to reduce the fouling frequency of me mbranes. The objective of this doctoral research was to evaluate the effectiveness of coagulation, anion exchange and ozonation as a pretreatment option for treating stabilized leachate using high pressure nano filtration (NF) and reverse osmosis (RO) mem branes. Additionally, adsorption of organic matter using activated carbon (AC) has been traditionally used in water and wastewater treatment; however, the selectivity of AC based on their pore sizes (micro porous and meso porous) for leachate treatment ha s never been investigated. Part of this study therefore focused the organic matter adsorption profile of three different AC s for stabilized leachate treatment. A maximum of 70%, 34%, and 23% reduction in DOC was observed after Fe (III), MIEX, and ozone treatment, respectively. Coagulation (Fe (III)) magnetic ion exchange resin (MIEX) and ozonation effectively remove d humic and fulvic like organic matter from s tabilized leachate ; m ore than 60% removal was observed for these types of organic matter. However, these pretreatment s did not improve in the permeate flux of NF and RO membranes as com pared to raw leachate treatments In the coagulation and MIEX pretrea ted leachate, the pH of the feed leachate increas ed throughout the experiment This was suspected to have caused excessive precipitation of carbonate minerals as compared to the raw leachate treatment on the membrane surface, causing a greater decline in permeate flux During the membrane treatment of ozone pretreated leachate, the divalent ions present in the ozonated leachate are hypothesized to have

PAGE 14

14 coagulated with the by products formed during ozonation with these coagulated solids then precipitat ing on the membrane surface, leading to a greater permeate flux decline as compared to the raw leachate treatment. The adsorption study showed that the micro and meso porous AC had similar organic matter adsorption capacity (0.2 g TOC/g AC) during stabilized leachate treatment with faster organic matter diffusion in meso porous AC. A maximum of 80% TOC was removed from both types of AC.

PAGE 15

15 CHAPTER 1 INTRODUCTION 1.1 Background and Problem Statement Economic advantages of landfills still make them the most accepted method for ultimate disposal of municipal and industrial solid waste (Renou et al., 2008). However, landfills face a major challenge of ma naging leachate generated by the percolation of rainwater through the layers of waste. A combination of several physical, chemical, and microbial processes in the waste results in contaminants from the waste being transferred to the leachate Improper ma nagement of leachate can potentially contaminate the surface and groundwater bodies (Christensen et al., 2001; Cossu et al., 2003; Eggen et al., 2010 ). Hence, leachate treatment is an essential part of effective landfill management process. The characte ristics of leachate are very heterogeneous and typically contain high concentrations of organic matter, salts, ammonia, and toxic trace components. A variety of parameters such as age, landfill operation method, moisture availability, waste composition, a nd climate, control the concentration of contaminants in the leachate. In particular, the composition of leachate varies greatly depending upon the age of a landfill and can be characterized as young, intermediate, and stabilized leachate (Kjeldsen et al. 2002). Generally, young leachate contains higher concentrations of organic and inorganic contaminants, which gradually decreases in the subsequent years ( Statum et al., 2004 ). As the landfill grows old, the concentration of biodegradable organic matter ( Biochemical Oxygen Demand ( BOD 5 ) ) may approach zero with leachate contain ing mostly refractory organic matter ( Chemical Oxygen

PAGE 16

16 Demand ( COD) ) ; leachate with BOD 5 Reinhart and Grosh, 1998 ). Tradi tional landfill leachate treatment methods can be classified into three major categories : (a) leachate transfer, which includes leachate recirculation into the landfill, or hauling the leachate to domestic wastewater treatment plant, (b) biological treatme nt using aerobic and anaerobic processes, and (c) physico chemical treatment methods such as chemical oxidation, coagulation flocculation, adsorption, chemical precipitation, air stripping, and membrane filtration. Each leachate treatment method has its o wn advantages and disadvantages. The controlled leachate recirculation has been largely used in the past decade and has shown advantages in terms of faster waste stabilization and decrease in COD of leachate, however, a very limited data are available ab out the quality of long term recirculated leachate (Rodriguez et al., 2004). Additionally, leachate recirculation cannot be continued throughout the life of a landfill; leachate needs to be removed from the landfill and treated, to dry the waste in the la ndfill. Leachate mixing with domestic wastewater has also been used but the presence of organic inhibitory compounds with low biodegradability and heavy metals reduces the treatment efficiency of the wastewater treatment plant (Cecen and Aktas, 2004). Reliability, simplicity, and cost efficiency make biological treatment the most commonly used leachate treatment method (Li et al., 2007). Biological treatment methods effectively remove the high biodegradable organic concentration of the leachate generat ed from young landfills when the BOD 5 /COD ratio has a high value

PAGE 17

17 (>0.5) (Renou et al., 2008). However, the major presence of refractory organic matter in th e stabilized leachate tends to limit the effectiveness of these processes. Leachate generated from stabilized landfills is typically characterized by COD in the range of 500 to 5000 mg/L, pH ranging from 7.5 to 8.5, a low biodegradability, and a significant amount of high molecular weight organic compounds (Rivas et al., 2004). Physico chemical treatment methods have been found useful for treating leachate containing refractory soluble organics (humic and fulvic like compounds) or as a polishing step for biologically treated leachate ( Alvarez Vazquez et al., 2004; Kurniawan et al 2006) These physico chemical methods have been used individually or in combination with other methods at various landfill sites (Chianese et al., 1999; Trebouet et al., 2001 ; Tatsi et al., 2003; Monze Ramirez and Velasquez, 2004; Kurniawan et al., 2006 ; Maranon et al., 2009 ). Studies on leachate treatment by physico chemical treatment methods suggest that use of a single treatment methodology does not produce effluent water that meets water quality standards for direct discharge to groundwater or surf ace water (Renou et al., 2008). Treatment of stabilized leachate with a combination of biological and physico chemical methods may provide efficient removal of contaminants but often does not meet the stringent regulations required to discharge the treate d water (Rivas et al., 2003, 2004; Monje Ramirez and Velasquez, 2004; Ntampou et al., 2006; Kurniawan et al., 2006) For an instance, Rivas et al. (2003, 2004) obtained COD removal in the range of 80 to 96% for an initial COD value of 11,000 mg/L from sta bilized leachate by using various combinations of pH shift, ozonation, coagulation flocculation, oxidation, and adsorption by activated carbon, but none of the tested processes reduced the COD

PAGE 18

18 levels enough that the effluent could be directly discharged. However, membrane filtration technologies such as nano filtration (NF) and reverse osmosis (RO) have been studied individually and in combination with other treatment methods by various researchers and have been found effective in removing the contaminants from the leachate to meet effluent discharge regulations ( Baumgarten and Seyfried, 1996; Peters, 1998; Bohdziewicz et al., 2001; Tabet et al., 2002; Palma et al., 2002; Thorneby et al., 2003). Although use of membrane filtration systems such as NF and RO ha s been found effective for treating landfill leachate, these systems may not provide a cost effective treatment ( Trebouet et al., 2001; Kurniawan et al., 2006). Heavy loading of particulate matter as well as the organic and inorganic compounds typica lly found in leachate tend to accumulate on the membrane surface and clog the pores of membranes a process known as fouling (Pearce, 2007). Most of the natural organic matter (NOM) components have been considered as a major membrane fouling agents. In particular, the compounds of low diffusion coefficients such as humic and fulvic like organic matter, which are hydrophobic in nature, have high fouling potential (Hong and Elimelech, 1997 ; Ta n g et al. 2007, Bruggen et al., 2008; Huang et al., 2009; Wang et al., 2011 ). At high divalent ion concentrations, the electrostatic repulsion between negatively charged humic and fulvic like organic matter and the membrane surface s reduce s The decreased repulsion of these compounds with the membrane surfaces lead to increased deposition of these compounds o n the membrane surface This caus es increased fouling of the membrane ( Yuan and Zydney, 2000; Tang et al. 2007). Additionally, other NOM such as organic macro molecules polysaccharide s proteins,

PAGE 19

19 and their coll oids may alter the pore structures through the pore blocking, pore narrowing, and cake layer formation, resulting in decrease in membrane permeability (Hermia et al., 1982; Huang et al., 2008). F eed water containing h igher concentrations of polysaccharide and protein like organic matter as compared to humic and fulvic like organic matter may thus exhibit more pronounced fouling ( Kwon et al. 2006; Lee and Lee, 2007; Amy, 2008). Fouling severely limits the filtration efficiency of membranes and causes flu x decline, a decrease in permeate quality, gradual membrane degradation, and increased operational costs of the treatment system. However, pretreatment of leachate that can reduce the concentrations of humic and fulvic like organic matter can be helpful i n reducing the concentration of foulant on the membrane surfaces and thereby in preventing frequent fouling and flux decline, which in turn may increase the life of the membranes (Shon et al., 2004; Xu et al., 2006; Agenson and Urase, 2007; Herzberg and El imelech, 2007). Several studies have been conducted for the leachate treatment using membrane systems; however, the studies were generally focused on improving the rejection of contaminants from the leachate by changing the membrane configuration, membra ne surface properties and membrane cleaning regime (Peters, 1998; Palma et al., 2002; Ushikoshi et al., 2002; Thorneby et al., 2003). The studies conducted with the pretreatment methods were also generally focused on improving the treatment performance in terms of contaminants rejection (Linde et al., 1995; Bohdziewicz et al., 2001; Ahn et al., 2002; Li et al., 2 007). V ery few studies were conducted on evaluating pretreatment methods for reducing the fouling of commonly used membranes while

PAGE 20

20 treating landf ill leachate (Baumgarten and Seyfried, 1996; Trebouet et al., 2001; Meier et al., 2002; Liu et al., 2008). Among the pre treatment methods studied, biological pre treatment methods were often found ineffective with the membranes; hence, this research was conducted to evaluate the performance of cost effective commonly used physico chemical methods (coagulatio n, anion exchange, and ozonation) as pretreatment options to treat stabilized leachate using RO and NF membranes. T raditionally, these processes have been found effective in removing or transforming the humic and fulvic like organic matter from the aqueo us matrix. C oagulant s remove organic matter by charge stabilization process ( Fettig et al., 1996; Amokrane et al., 1997 ; Tatsi et al., 2003 ; Ntampou et al., 2006 ). When coagulant is added to the aqueous matrix, the negatively charged organic molecules bi nd or adsorb on the positively charged hydroxides and precipitate as shown in Figure 1 1 (a) A nion exchange resin s replace their anions by preferential binding with negatively charged organic molecule as shown in Figure 1 1 (b) ( Fearing et al., 2004; All pike et al., 2005 ; Bo yer and Singer, 2005 ; Humbert et al., 2005; Boyer and Singer, 2006 ). O zone breaks the unsaturated and aromatic structure s present as part of this organic matter into smaller aliphatic organic matter as shown in Figure 1 1 (c) and reduces the concentration of humic and fulvic like organic matter ( Rivas et al., 2003; ; Bila et al., 2005; Tizaoui et al, 2007 ) Reduced concentrations of these organic matter reduces the fouling potential of membranes ( Amokrane et al., 1997 ; Bila e t al., 2005; Humbert et al., 2005; Ntampou et al., 2006; Tizaoui et al, 2007 ). Figures 1 1 (a, b, and c) are also presented individually in more detail in later chapters.

PAGE 21

21 Additionally, the literature suggest s that activated carbon (AC) is an effective method to remove high molecular weight organic matter from the aqueous solution, but very few studies can be found on the use of AC for landfill leachate treatment. In the leachate treatments, AC has been genera lly studied in combination with other treatment methods, but a detailed study that determine d the adsorption profile of organic matter removal using AC from the stabilized leachate, which helps in designing a n AC treatment system has not been reported (Fet tig et al., 1996; Kurniawan et al., 2006). These studies have always used micro porous AC (pore size < 2nm), whereas stabilized leachate contain high molecular weight and size organic matter, which may reduce the adsorption efficiency of AC due to pore bl ockage. A detailed study on the use of different types of AC based on their pore sizes for stabilized leachate treatment has not been found. 1.2 Research Objectives Stabilized landfill leachate contains higher amounts of biologically refractory organic m atter and the l iterature suggest s that the use of physico chemical treatment methods such as coagulation, anion exchange, ozonation, and adsorption by activated carbon can reduce the concentration of heavier organics in leachate, thereby should reduce the frequ ency of fouling of membranes. The objective s of this doctoral research w ere to explore the effectiveness of coagulation, anion exchange, and ozonation for reducing the fouling of NF and RO membranes while treating stabilized landfill leachate and to study the selectivity of AC based on their pore sizes for treating stabilized landfill leachate. More specifically, an objective of the first study was to evaluate the efficiency of Fe (III) salt (coagulant) and MIEX (anion exchange) as a pretreatment op tion for treating

PAGE 22

22 stabilized landfill leachate using NF and RO membrane s In the coagulation process, the hydrolysis products of the positively charged coagulant react with the negatively charged high molecular weight organic matter and precipitates. Iro n salt has been found better for performing coagula tion for leachate treatment however, there is no perfect correlation obtained between the coagulant dose and the treatment efficiency. Additionally, t he anion exchange resin in particular, magnetic ion e xchange (MIEX) has shown greater potential as an alternative to coagulation process for the removal of high molecular weight organic matter and removes a wider range of organics than coagulation (Boyer and Singer, 2006). Th is study is an important contrib ution to evaluate the efficiency of Fe(III) coagulant and MIEX for treating stabilized leachate using membranes. The o bjective of the second study was to investigate the effectiveness of ozone (oxidation process) as a pretreatment for stabilized landfill l eachate treatment using NF and RO membranes. Ozone has a high oxidation potential and high reactivity with the humic and fulvic like organic matter. It transforms these recalcitrant organic matters into more biodegradable form, or even oxidizes them t o carbon dioxide (Monze Ramirez and Velasquez, 2004). Additionally, ozone has been shown to reduce the fouling of micro filtration and ultra filtration membranes (Lee et al., 2005; Kim et al., 2008). However, very little or no information is available on u sing ozone as a pretreatment option while treating stabilized landfill leachate using NF and RO membranes. Ads orption on to AC has been reported as an effective method for the removal of high molecular weight refractory organic matter from aqueous soluti on AC has been seldom used for landfill leachate treatment, either as a single treatment step or in

PAGE 23

23 combination with other treatment options. In addition to the availability of very few studies on the use of AC for leachate treatment, more rigorous stud ies in terms of adsorption isotherm determination and diffusion of organic matter onto AC, which are the primary factors to design and optimize a full scale or pilot scale AC treatment system has been rarely found. The specific o bjective of the third stud y was to determine the effectiveness of organic matter adsorption on to three AC s selected based on their pore sizes for treating stabilized landfill leachate. 1.3 Research Approach Laboratory scale experiments were conducted to achieve the objectives des cribed in section 1.2 Objective 1. To evaluate the efficiency of Fe (III) (coagulant) and MIEX (anion exchange) as a pretreatment option for treating stabilized landfill leachate using NF and RO membrane systems. Approach. Leachate samples were colle cted from three different landfills. Leachate characteristics were determined based on their dissolved organic carbon (DOC), ultra violet 254 (UV 254) absorbance, and florescence excitation emission matrix (EEM). Batch experiments of coagulation and anio n exchange process were conducted and the treatment efficiency was evaluated for various doses. The optimum coagulant and MIEX doses were determined to run fouling experiments using one NF and one RO membranes each in unadjusted pH conditions Permeate f lux and permeate quality w ere also analyzed. Objective 2. To investigate the effectiveness of ozonation as a pretreatment for treating stabilized landfill leachate using RO and NF membranes.

PAGE 24

24 Approach. Leachate from three different landfills was collecte d. The characteristics of leachate were determined and l aboratory scale experiments were conducted in two phase s At first, leachate was treated with a fix e dose of ozone for variable durations and optimum ozonation duration was determined in terms of re moval of DOC, UV 254 absorbance, and different types of organic matter derived from EEM. In the second phase, the ozonated leachat e was tested for the time dependent permeate flux and the permeate quality. The permeate flux data was used to determine the pressure, cross flow filtration model (Hermia, 1982). Objective 3. To determine the equilibrium adsorption and intra particle diffusion of organic matter into micro and meso porous AC while treating stabilized landfill leachate. Approach. Three commercial AC; one micro porous and two meso porous were selected for treating leachate collected from a stabilized landfill. Isotherm and diffusivity experiments were performed to determine the isotherm profile, the adsorption capacity, the rate limiting adsorption process, and the rate of organic matter diffusion onto AC. Rapid small scale column tests (RSSCT) were conducted to determine the amount and the type of organic matt er removed using these three AC for leachate treatment. 1.4 Outline of Dissertation This dissertation is divided into five Chapters. Chapter 1 introduces the research, the problem statement, the research objectives and the approach followed to achieve specific objectives. Chapters 2 to 4 are classified into three stand alone manuscripts. Chapter 2 describes the efficiency of Fe (III) (coagulant) and MIEX (anion exchange) as a pretreatment option for treating stabilized landfill leachate using NF and R O

PAGE 25

25 membrane systems. Chapter 3 evaluates the use of ozone as a pretreatment option for treating stabilized landfill leachate using NF and RO membrane systems. Chapter 4 describes the effectiveness of AC for treating stabilized landfill leachate. Chapter 5 provides summary, conclusions, and recommendations for future work. The references cited are listed at the end of dissertation.

PAGE 26

26 Figure 1 1. (a) Reaction that hydrolysis products of coagulants follow when coagulant is added to water containing natural organic matter. (b) MIEX resin anion exchange process (c) Model reaction of ozonation of unsaturated organic compound C=C O 3 R 3 R 4 R 2 R 1 C=O R 2 R 1 O=C R 3 R 4 + + C=O OH R 1 (a) (b) (c)

PAGE 27

27 CHAPTER 2 EVALUATION OF COAGULATION AND ANION EXCHANGE AS A PRETREATMENT FOR STABILIZED LANDFILL LEACHATE TREATMENT USING HIGH PRESSURE MEMBRANES 2.1 I ntroduction The percolation of rainwater through the layers of degr aded waste in the landfill generates high strength contaminated wastewater referred as leachate. A combination of several physical, chemical, and microbial processes in the waste results in contaminants from the waste being transferred to the leachate, wh ich makes leachate potential contaminant source to the surface and groundwater bodies (Christensen et al., 2001). Despite the fact that the modern landfill designs minimize leachate generation by reducing the influx of moisture ; landfills still face a maj or challenge of efficient management or treatment of leachate The heterogeneous nature of leachate composition that varies with age, landfill operation method, moisture availability, waste composition, and climate, makes the selection of leachate treatment process even more difficult. In particular, composition of leachate varies greatly on the age of the landfill and can be characterized as young, intermediate, and stabilized leachate (Kjeldsen et al., 2002). Youn g leachate generally contains high concentrations of inorganic and biodegradable organic contaminants and biological methods have been found effective for its treatment. However, as the landfill grows old, the concentration of BOD 5 reduces and leachate co ntains mostly refractory organic matter making biological treatment methods less effective. When the leachate has BOD 5 /COD<0.1, it is termed as stabilized leachate; the physico chemical treatment methods such as coagulation flocculation, chemical oxidatio n, adsorption, and

PAGE 28

28 membrane systems are more effective than biological treatment for such leachate (Renou et al., 2008). Treatment of stabilized leachate with one or combination of more than one biological or physico chemical treatment methods often does not meet the stringent regulations required to discharge the treated water (Monje Ramirez and Velasquez, 2004; Kurniawan et al., 2006; Ntampou et al., 2006). However, the membrane filtrations technologies especially reverse osmosis (RO) has been found ef fective in removing the contaminants from the leachate to meet effluent discharge regulations (Peters, 1998; Palma et al., 2002; Tabet et al., 2002). A common issue with the membrane systems when treating landfill leachate is the fouling by organic matter and salts present in leachate, which limit the filtration efficiency and cause flux decline. Fouling of membranes is generally influenced by the composition of feed water, concentration of constituents (hydrophilic and hydrophobic organic matter), water chemistry (pH and divalent ion concentration), hydrodynamic conditions (permeate flux and cross flow velocity), and membrane surface properties (hydrophobicity, surface morphology, and charge) (Li and Elimelech, 2004; Xu, 2006) Several researchers have s hown that the high molecular weight humic and fulvic like organics (natural organic matter (NOM)) present in the feed water may adsorb on the hydrophobic membrane surfaces and increase the fouling (Jucker and Clark, 1994; Nilson and DiGiano, 1996). Stabil i zed leachate contains higher concentrations of these types of organic matter; hence t he ir reduction from stabilized leachate prior to membrane treatment can provide a substantial increase in membrane performance (Huang et al., 2009).

PAGE 29

29 Coagulation with in organic salts (Fe (III) and Al (III)) has been used to remove dissolved organic matter (DOM) from leachate and has shown 50 to 75% COD removal from stabilized leachate (Amokrane et al., 1997; Tatsi et al., 2003). When the inorganic coagulants are dissolve d in water they form a series of intermediate hydrolysis products and the negatively charged organic molecule s present in solution may bind or adsorb with the precipitating hydrolysis products as shown in Figure 2 1 (Letterman et al., 1999). The c oagulati on process is highly dependent on the pH of the solution and a typical profile of coagulation dose on the pH and organic matter removal efficiency is explained in Appendix A using the results of present study A considerable amount of studies have been conducted on leachate treatment using coagulant as shown in Table 2 1 Generally, iron and aluminum salts have been used as coagulant for leachate treatment and iron salt has been found a better performing coagula nt T he se studies were mainly focused on the effect of different operating variables such as coagulant dose and pH on the treatment efficiency achieved in terms of COD ; however, there is no perfect correlation obtained between coagulant dose and treatment efficiency (Amokrane et al., 1997; Yoon et al., 1998; Wang et al., 2002; Tatsi et al., 2003). For instance, Amokrane et al. (1998) obtained a maximum of 55% COD removal at a dose of 2 g/L Fe (III), whereas Wang et al. (2002) obtained 80% COD removal effi ciency at a dose of 1.5 g/L Fe (III) salt for leachate containing slightly higher COD than the leachate used by Amokrane et al. (1998). Comstock et al. (2010) studied the use of sulfate salt of Fe (III) for treatment of stabilized leachate collected from different landfills and observed a considerably wide range of DOC removal due to the specific characteristics of each leachate. They also

PAGE 30

30 concluded that the coagulation process for leachate treatment should be selected based on the specific UV absorbance of leachate, instead of ratio of BOD 5 and COD (Comstock et al., 2010). Few studies have been conducted using coagulant as a pretreatment option for leachate treatment using membranes; such as, Amokrane et al (1997) used a Fe (III) coagulant along with lim e and hydrogen peroxide oxidant and found that treated leachate had lower fouling potential than raw leachate for reverse osmosis (RO) membranes. Yoon et al. (1998) have reported a 59 to73% removal of larger organic matter (MW>500) that are mostly respons ible for organic fouling of membranes; hence, coagulation can be used as a pretreatment option for reduc ing the fouling frequency of membranes. However, a comprehensive study on the effectiveness of coagulation as a pretreatment option for treating stabil ized leachate using high pressure membrane s NF and RO have not been reported. M agnetic ion exchange resin (MIEX) has also been found to have a greater potential to remove DOC and UV a bsorbing compounds and remove wider range of molecular weight and aromaticity of organic matter than coagulants such that MIEX can effectively remove organic matter of molecular weight > 1000 Da and water with low SUVA values (<2) ( Fearing et al., 2004; Allpike et al., 2005 ) However, coagulant s selectively remove high molecular weight (>5000 Da) and aromatic rich organic matter (SUVA>2) ( Edzwald and Tobiason, 1999; Fearing et al., 2004; Allpike et al., 2005; Humbert et al., 2005; Bo yer and Singer, 2005 ; Boyer and Singer, 2006). MIEX has macro porous structure, strong base functional group, polyacrylic matrix in chloride form and is used in completely mixed reactor in contrast to traditional ion exchange resin. It

PAGE 31

31 has also been tested as a pretreatment option to reduce fouling of low pressure membranes for low DOC water (Fabris et al., 2007). However, its application for the landfill leachate treatment has been very limited. The stabilized leachate contains higher portion of negatively charged humic and fulvic like organic material ( Baker and Curry, 2004) and MIEX resin shows preferential removal of UV absorbing compounds which are represented by the humic and fulvic like organic matter as shown in Figure 2 2 ; hence, it was hypothesized that MIEX can be used as a pretreatment option to reduce organic fouling of membranes and improve the treatment performance. No studies were found on evaluation of treatment of MIEX treated leachate with high pressure NF and RO membrane systems. The objective of this research was to evaluate the effec tiveness of coagulation and anion exchange process for stabilized landfill leachate treatment and their use as a pretreatment option for treating stabilized landfill leachate using NF and RO membrane systems. Leachate samples were collected from three dif ferent landfills and were characterized based on their dissolved organic carbon (DOC), ultra violet 254 (UV 254) absorbance, specific ultra violet absorbance (SUVA), and fluorescence excitation emission matrix (EEM). These DOM properties were studied as a n indicator parameter for providing the insight of selecting and determining efficiency of coagulation and anion exchange for stabilized leachate treatment with membranes. 2.2 Experimental Materials Methods and Analysis 2.2.1 Materials Experiments were c onducted on leachate generated from three municipal solid waste (MSW) lined landfills located in Florida, USA. Leachate samples were collected from the leachate lift stations or the leachate manholes in Nalgene containers and kept

PAGE 32

32 at 4 0 C in the dark. All three landfills have been fully or partially operated as a bioreactor landfill where leachate has been injected into the waste to accelerate the rate of waste degradation. The Alachua County Southwest Landfill (ACL) has a closed lined MSW cell that has b een operated as a bioreactor since 1990. Leachate samples were collected multiple times for conducting experiments in the duration from November 2009 to February 2010. The North Central Landfill (NCL) located in Polk County, Florida has two landfill cell s; the first cell was closed in April 2000 after accepting waste for 11 years and the second cell was closed in 2007 after seven years accepting waste. The second cell was operated as a bioreactor landfill. The leachate samples were collected from the le achate storage tank, where leachate from both the cells is mixed. The New River Regional Landfill (NRL) located at Union county, Florida, has three cells of waste ranging in age from 5 to15 years. The leachate samples were collected from a common manhole where leachate from all the landfill cells is drained. Technical grade ferric chloride (FeCl 3 .6H 2 O) purchased from Fisher Scientific was used as a coagulant for the experiments. The magnetically enhanced polyacrylic resin (MIEX) was obtained from the O rica Watercare of Watkins, CO, USA in the slurry form. Dow filmtec (Minneapolis, MN) provided flat sheet NF 90 and BW 30 membranes for conducting membrane performance studies. Membrane filtration experiments were conducted using high pressure cross flow system Osmonics SEPA CF Membrane Cell as shown in the schematic diagram ( Figure 2 3 ). 2.2.2 Coagulation Experiment Preliminary jar test coagulation experiments were performed to determine the optimum coagulant dose. A bench scale jar testing apparatus (Phipps and Bird, Richmond, VA) equipped with six 2000 mL beakers was used to conduct the

PAGE 33

33 experiments. Before startin g the experiments, leachate was brought to room temperature (approximately 23 0 C). Leachate samples were thoroughly shaken for re suspension of settled solids and 1000 mL of leachate was transferred to each jar test beaker. All experiments were conducted in duplicate. Dry crystals of coagulant were weighed using an analytical balance (Mettler Toledo) and were added to leachate. Coagulant doses in the range of 1 to 10 g/L of ferric salt (FeCl 3 .6H 2 O) were used in the experiment, corresponding to 3.7 to 37 .0 mmol/L of Fe(III) salt, respectively. Immediately after adding the coagulant, samples were rapidly mixed at 100 rpm for 5 minutes followed by gentle mixing for 25 minutes at 35 rpm. Finally, the flocs were allowed to settle for another 30 minutes (Tre bouet et al., 2001; Tatsi et al., 2003), after which the samples were collected from the sampling port located approximately 2 cm above the bottom of beaker at the 500 mL mark. A control experiment was conducted without adding any coagulant. No pH adjust ment was made during the experiment. Previous studies have used COD, color, and turbidity as a parameter for selecting and determining the efficiency of coagulation as a leachate treatment option (Comstock, et al., 2010); however, in order to use coagula tion as a pretreatment option with membranes, characterization of DOM present in leachate is important because the characteristics of DOM present in leachate directly influences the membrane treatment efficiency (Bellona et al., 2004). Each sample was ana lyzed for pH and DOC. To determine the chemical complexity of DOM SUVA, and fluore scence EEM were also analyzed.

PAGE 34

34 2.2.3 MIEX Experiment Preliminary MIEX experiments were performed using a jar test procedure to determine optimal MIEX dose and mixing time. Unless otherwise indicated, the same apparatus and procedures as used for the coagulation experiment were used. The experiments were conducted at MIEX doses ranging from 2 to 10 mL/L. To prepare the MIEX doses, MIEX slurry was vigorously shaken and trans ferred to a 10 mL graduated cylinder and allowed to settle for 30 minutes. After settling, specific doses were prepared by adding or removing the resin with a glass pipette. These MIEX doses were added to the 1000 mL leachate samples using de ionized org anic free water. Immediately after adding the MIEX doses, the leachate samples were rapidly mixed at 100 rpm for 60 minutes followed by 30 minutes of settling. During the mixing period, aliquots were periodically collected (5, 10, 20, 30, 60, and 90 minu tes) using a glass pipette. The samples were analyzed for pH and UV 254 absorbing organic matter After 60 minutes of mixing followed by 30 minutes of settling, samples were analyzed for DOC SUVA, and EEM. To determine the optimum mixing time and the M IEX dose, the results of UV 254 analysis were plotted with respect to the mixing time for all different MIEX doses. 2.2.4 Membrane Experiment Experiments were conducted to evaluate the performance of raw and treated ACL leachate using NF and RO membranes The ACL leachate was found hardest to treat in terms of DOC and SUVA removal by Fe (III) salt and was the most stabilized in terms of lowest BOD 5 /COD; hence the membrane performance was evaluated following a conservative approach and using only ACL leac hate. Membrane performance was evaluated by measuring permeate flux and rejection of DOM and salt. To conduct the

PAGE 35

35 membrane experiments, approximately 8 to 10 L of ACL leachate was pretreated separately using the optimum dose of coagulant FeCl 3 and MIEX a s determined by the batch experiments. The flat sheet membranes were cut into 14.6 cm x 9.5 cm coupons and stored as dry in the dark. These coupons were soaked in MilliQ water in the dark for 24 hours before use. The coupons were placed into the membrane cell sandwiched between a low foulant feed spacer of thickness 0.86 mm (34 mil) and a permeate carrier of thickness 0.2 mm (7.8 mil). Feed solution was pumped using a variable speed Hydra Cell Industrial pump from a Nalgene feed tank. The temperature of the feed tank was maintained constant (23 1 0 C) using the recirculation water bath (RTE 5B, Neslab Instruments Inc.). The membrane cell was pressurized using a hydraulic hand pump (Enerpac, P142) and the pressure was maintained constant (20.7 bar) through out each experiment. The cross flow and the operational pressures were adjusted using the concentrate return valve and the bypass valve and were monitored by a flow meter and a pressure gauge. The flow meter and pressure gauge were connected in the conce ntrate line. The permeate flux was determined volumetrically by measuring the time taken to a predetermined volume of permeate generated. Concentrate and permeate were recycled to the feed tank except for the intermittently collected permeate samples for DOC and UV 254 absorbance measurement. The physical and the operational characteristics of NF and RO membranes are presented in Table 2 2 Both the membranes are aromatic polyamides thin film composite with zeta potential of 26.5 mV, and 5.2 mV, resp ectively. Pure water flux

PAGE 36

36 (flux after pre compaction) of NF and RO membranes was obtained as 1.2 0.3 and 2.6 0.4 m/day, respectively. Membrane runs were conducted in two steps. First, the membranes were pre compacted by filtering de ionized water for 4 8 hours at a constant pressure of 13.8 bar (200 psi) and a cross flow velocity of 20 cm/sec, until constant permeate flux (pure water flux) was achieved. After the pre compaction run, the membrane system was stopped and the de ionized water was drained ou t from the feed tank. Approximately 8 to 10 L of feed leachate was filtered using 0.7 m glass fiber filter (whatman) and added to the tank. The membrane system was restarted and operated at similar pressure and cross flow velocity as in pre compaction r un for 24 hours. Feed pressure and cross flow velocity were maintained constant throughout the experiment and permeate flux was measured intermittently. Feed and permeate samples were collected intermittently. 2.2.5 Analytical Methods Water quality par ameters pH, conductivity, and temperature was measured using a multi function Orion Research instrument. Solution pH was measured with an accuracy of 0.02. Temperature was measured in degrees celsius with a scale accuracy of 0.15 o C. Conductivity was measured in mS/cm or S/cm depending upon the concentration of sample. Salt rejection in the membrane experiments were measured by measuring conductivity of feed and permeate. Samples were filtered through pre rinsed 0.45 m nitrocellulose filter (Millipore) to measure DOC, UV 254 absorbance, and EEM. The filtered samples were stored in capped 40 mL glass vials at 4 0 C in the dark; samples were brought to room

PAGE 37

37 temperature before analysis. DOC was measured using Tekmer TOC analyzer. UV 254 absorba nce was measured using HACH DR 4000 spectrophotometer. EEM was measured using a 1 cm quartz cell on a Hitachi F 2500 fluorescence spectrophotometer. The EEM was obtained by scanning the samples at 5 nm increments over an excitation (E x ) wavelength in th e range of 200 nm to 500 nm. For each E x wavelength, the emission (E m ) wavelengths were detected at 5 nm increments in the range of 290 nm to 550 nm. To limit the second order Ra y l e i gh scattering, a limit of 290 nm cutoff was used for all the samples (Ch en et al., 2003). The data were processed in MATLAB following the procedures by Chen et al., 2003 and Cory and Mcknight, 2005. Each time the instrument was used; the EEM of de ionized water was analyzed and was subtracted from the EEM response of samples containing DOC to reduce the effect of scattering (Chen et al., 2003). The area under the Raman water peak (at E x 350 nm) was calculated for de ionized water and the intensities of the EEM response for the samples were normalized by the Raman water area; the EEM response of the samples was then normalized by the DOC concentration; and EEMs were plotted in MATLAB usin g the contour function. The amount of various types of DOM was quantified using Fluorescence Regional Integration (FRI) technique (Chen et al., 2003). The FRI was used to integrate the area under the EEM spectra as shown in equation 2 1. ( 2 1) where, V i Ex Em ) is the fluorescence intensity at a particular Ex Em wavelength pair, d Ex is the increment of Ex wavelength (5 nm) and d Em is the increment of Em wavel ength (5 nm).

PAGE 38

38 EEM were divided into five E x E m regions and concentration of DOM in each region was quantified by integrated volume in each region. The E x E m regions were divided based on the EEM peaks associated with specific type of organic compounds as shown in Figure 2 4 These regions were selected using the literature (Chen et al., 2003; Chen et al., 2003; Baker, 2004; Hudson, 2007). Shorter E x (<250 nm) and shorter E m wavelengths (<380nm) are associated with simple aromatic protein like compounds s uch as tyrosine (Regions I and Region II). The peaks at E x wavelengths in the range of 250 to 280 nm and E m wavelengths <380 nm are associated with soluble microbial byproducts type compounds and are kept in Region IV. The fulvic like compounds (Region I II) show peaks at shorter E x wavelengths (<250 nm) and longer E m wavelengths (>350 nm). Peaks at longer E x wavelengths and longer E m wavelengths are associated with humic like compounds and shown in Region V. More specifically the fulvic like compounds s how peak intensities at E x E m wavelength band of 220 250/400 450 nm and humic like compounds at 320 360/400 470 nm. 2.3 Results and Discussion 2.3.1 Leachate Characteristics Relevant physico chemical characteristics of ACL, NCL, and NRL leachate samples collected during the research period are summarized in Table 2 3 Multiple samples were collected and composition of leachate was relatively constant over the research period. All three leachate s were characterized by slightly alkaline pH, dark brown color, a low concentration of biodegradable content as represented by BOD 5 and high concentration of refractory organics represented by COD. Leachate from all three landfills ha d BOD 5 /COD in the ran ge of 0.02 to 0.14. These properties are consistent with typical stabilized or intermediate stabilized leachate (Reinhart and Grosh, 1998 ;

PAGE 39

39 Statom et al., 2004). These types of leachate are found difficult to treat using conventional biological treatment methods and in most cases require more intensive physico chemical treatment. All three leachate s had DOC in the range of 425 to 670 mg/L with UV 254 absorbance of 8.3 to 14.7 cm 1 The SUVA for three leachate were in the range of 1.6 to 2.22 m 1 /(mg/L DO C) The Fluorescence EEM of leachate from all three landfills is shown in Figure 2 5 Leachate from all three landfills showed an overlapping peak in Region III and V, representing the presence of fulvic and humic like organics in leachate. ACL and NCL leachate also showed overlapping peak in the region IV and V and peaks in Region I and II representing the presence of microbial derived organic and protein like organic matter. The distribution of DOM in each region was derived using FRI analysis as show n in Table 2 4 Among all three leachates NRL leachate had the highest total volume (V t ) representing maximum concentrations of DOM in NRL leachate. Additionally, Region V showed maximum volume among all five regions in all three leachates representing leachates contained maximum concentrations of humic like organic matter. 2.3.2 Leachate Treatment Using Coagulant Coagulation dose efficiency was assessed by measuring the change in leachate pH, DOC, SUVA and fluorescence EEM. A decreasing pH was obser ved with an increase in coagulant dose in all three leachates ( Figure 2 6 (a)) The results are shown as average of the duplicate experiments with error bars as standard deviation. The pH of leachate dropped in the range of 3 to 5.2 at different coagulan t doses. Amokrane et al. (1995) also observed a decrease in supernatant pH with an increase in coagulant dose due to the increased release of H + in solution after hydrolysis. NRL

PAGE 40

40 l eachate had a lower alkalinity (3000 mg/L as CaCO 3 ) as compared to ACL (6600 mg/L as CaCO 3 ) and NCL (5400 mg/L as CaCO 3 ) leachate, which caused a faster decrease in pH of NRL leachate than ACL and NCL leachate. Contrary to ACL leachate, the NCL and NRL leachate showed an increase in pH at higher coagulant doses, due to form ation of hydrolysis product Fe(OH) 4 that starts accepting available hydrogen ions and reducing free H + As the pH of supernatant in the coagulation process reduces, the charge densities of t he DOM also decreases (Ritchie and Perdue, 2003). Lower charge densities of DOM require lower coagulant doses to initiate charge neutralization and precipitation (Shin et al. 2003) Hence, as the coagulant dose was increased higher DOC and SUVA removal was observed as shown in Figure 2 6 (b and c). A maximum of 68%, 71%, and 45% DOC ( Figure 2 6 (b)), and 48 %, 60 % and 90 % SUVA removal was observed in ACL, NCL, and NRL leachate, respectively ( Figure 2 6 (c)) However, these maximum removals were observed at different coagulant doses The ACL leachate showed maximum removal at a coagulant dose of 37 mmol Fe(III)/L, whereas NCL, and NRL leachate showed at 29.6 and 18.5 mmol Fe(III)/L, respectively. Yoon et al. (1998) also observed 59 to 73% TOC reduction w hile treating landfill leachate using Fe (III) salt. Comstock et al. (2010) studied the use of sulfate salt of Fe (III) for treating leachate collected from ACL, NCL, and NRL landfills and observed 20%, 30%, and 70% DOC reduction, respectively for a fix d ose of 17.9 mmol Fe (III)/L. In the present study, at a coagulant dose of 17.9 mmol Fe(III)/L of chloride salt, slightly lower DOC removal was observed in all three leachates. M aximum removal of DOC and SUVA was observed at lower coagulant doses in NCL a nd NRL leachate as compared to ACL

PAGE 41

41 leachate because the coagulation efficiency was dominated by the solution pH. The pH of coagulated leachate corresponding to the maximum reduction of DOC and SUVA in all three leachate s was in the range of 4.2 to 5.2, wh ich is the pH range for highest coagulation efficiency for Fe(III) salt (Babcock and Singer, 1979). Amokrane et al (1995) also observed optimum pH of 5.5 for maximum COD removal from stabilized leachate. Effects of coagulant doses on removal of differe nt types of organics in all three leachate s as derived by FRI analysis are shown in Figure 2 7 DOM present in Region V showed the highest removal among DOM of all regions and the order of removal of in all three leachate s was found similar as Region V>Re gion III>Region IV>Region II>Region I. The h ighest removal of DOM from Region V and Region III was expected because they belong to humic and fulvic like organics that are favorable to coagulation process ( Letterman et al., 1999 ). The DOM present in Regio n I did not show any particular pattern with increase in coagulant dose due to low concentrations of DOM in this region. Low DOM concentration may show some noise during EEM measurement leading error in FRI analysis. A coagulant dose of 22 mmol Fe (III) /L was used for leachate pretreatment to conduct the membrane experiments. At this coagulant dose, NRL leachate showed the maximum DOC removal where as ACL leachate showed the least removal M embrane experiments were performed using the leachate that showed lowest DOC removal among all three leachate. A pproximately 40% of the DOC and 40 to 60% of the humic and fulvic like organic matter as determined by FRI analysis was removed from all three leachate at this coagulant dose.

PAGE 42

42 2.3.3 MIEX Experiment Leachate treatment using MIEX showed an increased removal of UV 254 absorbing organic compounds with the increase in MIEX dose in all three leachate s ( Figure 2 8 ). UV 254 absorbance reduced at high rate at the start of mixing for each MIEX dose, while it plateaued after some time. The UV 254 absorbance was observed constant after approximately 30 minutes of mixing. It was also observed that majority of UV 254 absorbing materials were removed in ini tial 20 minutes of mixing in all three leachate s for each MIEX dose, which is consistent with the results observed by Boyer and Singer (2005) for drinking water treatment. Increased DOC and SUVA removal was observed with the increase of MIEX doses ( Figur e 2 9 (a and b )). Highest reduction in DOC was observed in NRL leachate followed by ACL and NCL leachate, respectively. Higher concentrations of TDS might have caused lowest rejection of DOC and UV 254 absorbance in NCL leachate because inorganic ions pr esent in solution may occupy some spaces of MIEX resin causing reduction in DOC removal capacity (Boyer and Singer, 2006). A maximum of 30%, 20%, and 34% DOC removal was observed from ACL, NCL, and NRL leachate, respectively at a MIEX dose of 10 mL/L. A recent study on the use of MIEX for treating leachate collected from similar landfills also showed an average of 35% DOC removal for slightly higher MIEX doses (Boyer et al., 2010). The results suggest that for all three leachate s most of the DOC was remo ved using low MIEX doses and with the increase of MIEX doses, rate of DOC removal decreases. A maximum of 18 %, 19 % and 2 1 % of SUVA removal was observed from ACL, NCL, and NRL leachate, respectively. It was also observed that the pH of all three leachate s were slightly increased for each MIEX doses and remain approximately same as shown in Figure 2 8 (c).

PAGE 43

43 The FRI analysis results of the DOM removed from each region of the MIEX treated leachate are shown in Figure 2 10 The pattern for DOM rejection in each region was found similar in all three leachate s with respect to rate of removal of organics from each region. Protein like DOM present in Region I and II had the maximum rate of removal followed by fulvic and humic like DOM present in Region III and V, respectively. On comparing the amount of DOM removed from each region, the results show that the DOM present in Region III and V were removed to maximum in all three leachate s The FRI analysis results also show ed that the rate of DOM removal decrease d with the increase of MIEX dose. The DOM removal profile gets almost plateau after a MIEX dose of 4 mL/L in all three leachate s which is in agreement with the results obtained for the removal of DOC and SUVA A dose of 5 mL/L MIEX was selected for mem brane experiment because most of the DOC was removed at an applied MIEX dose of 4 to 6 mL/L in all three leachate s Also, a mixing time of 30 minute s was selected for MIEX treatment, because in all three leachate s most of the UV 254 absorbing organic mate rial was removed in initial 20 30 minutes of mixing. 2.3.4 Comparison of Coagulation to MIEX Treatment Both the coagulation and anion exchange process effectively removes humic and fulvic like organics from leachate as represented by removal of DOC, SUVA and results obtained from FRI analysis ; however, coagulation process was found more effective than MIEX treatment. Leachate treatment using MIEX removed DOC in the range of 20 to 34% for the applied doses, which is approximately 4 0% lower reduction than c oagulation process. A high removal (> 50 %) of SUVA were observed in all three leachate s using the coagulant whereas MIEX removed a n average of 20% SUVA from

PAGE 44

44 all three leachates. The coagulant showed better performance than MIEX in terms of removing high m olecular weight organics (humic and fulvic like organic matter) and a maximum of 80% these organics were removed by coagulant in all three leachates M IEX removed a maximum of 40% humic and fulvic like organic matter and showed removal of wider range of organics in terms of molecular weight as observed by maximum removal of protein like o rganic matter in all three leachates However, the amount of low molecular weight organic matter (protein like) removed by coagulation process was still h igher than MIEX treatment. Microbial derived organics (region IV) were least removed after MIEX treatment whereas coagulant showed least removal for protein like organic matter. Fearing et al., ( 2004 ) and Allpike et al., ( 2005 ) have also found similar r esults in terms of removing organic matter of different range of molecular weight using coagulant and MIEX for treatment of low DOM containing water 2.3.5 Leachate Treatment Using Membranes As expected both membranes showed high rejection for DOM and sa lts present in the feed leachates More than 99% of DOM and 93% salts were rejected from all three leachate s from both the membranes with slightly greater salt rejection from RO membrane than NF membrane. The effect on permeate flux for treatment of raw and pretreated leachate using NF and RO membranes is shown in Figure 2 1 1 An average initial flux of 0.85 m/day (8.2 mL/min) and 0.55 m/day (5.3 mL/min) was observed with the NF and RO membranes, respectively. At the start of membrane treatment, a shar p decrease in permeate flux was observed in all three feed leachate s however, in long term operation of raw and coagulated leachate the rate of permeate flux reduction slowed down whereas MIEX treated leachate showed continuous sharp reduction in permeate flux with both

PAGE 45

45 membranes. The rate of permeate flux decline was observed to be faster for the coagulant and MIEX treated leachate than that of raw leachate with both the membrane s. The maximum flux decline was observed in MIEX treated leachate. Although, the rate of permeate flux decline for the raw and coagulated leachate in RO membrane was comparable, the NF membranes showed a distinctly higher rate of flux decline for pretrea ted leachate as compared to raw leachate. The plausible reason of increased flux decline with pretreated leachate lies within the complex chemistry of feed leachate and the membrane surfaces that might have changed during the experimental run because t he feed pH of all three feed solutions in NF as well as RO operation increased during the run as shown in Figure 2. 12 The increase in feed pH may have been caused by the permeation of CO 2 of the solution through the membrane s NF and RO membranes are well known to be highly permeable to gases such as CO 2 ( Mitsoyannis and Saravacos 1977 ). In the aqueous phase, CO 2 gas is dissolved as carbonic acid (H 2 CO 3 ) ( CO 2 gas +H 2 O H 2 CO 3 ) The permeation of CO 2 gas through the membrane will tend to decrease H 2 CO 3 in the solution. As the H 2 CO 3 decreases, bicarbonate anions (HCO 3 ) will transform to H 2 CO 3 to maintain the equilibrium (HCO 3 +H + H 2 CO 3 ). This leads to an increase in solution pH. The change in feed pH might have affected the flux decline in two ways. First, an i ncrease in pH may increase the potential for scaling on the membrane surface due to precipitation of CaCO 3 ( Mitsoyannis and Saravacos, 1977 and Linde and Jonsson, 1995). Amokrane et al. (1998) studied the use of Fe (III) salt for reducing the f ouling potential of RO membranes and observed that alkaline pH conditions increase d the

PAGE 46

46 fouling potential of membranes due to alkalinity precipitation; however, they did not operate the membrane system with treated leachate. Second, as the solution pH increases, the surfaces of these membranes tend to get more negatively charged due to deprotonation of carboxyl groups ( COOH COO ) of membranes However, the rate of deprotonation of membrane surface molecules may not be the same as the H + consumed by bicarbonate and an increase in pH of feed water may still be observed. The increase in negative charge of the membrane surface molecule may experience increased repulsion causing membrane pores to shrink. Chil dress and Elimelech (2000 ) have found increased negative charge and reduction in pore sizes of NF membrane surfaces at high pH conditions. This reduction in pore size reduce s the permeate flux generation. However, the effect of feed pH on permeate flux is contradictory A f ew researchers have found no effect on water flux with the change in feed pH for the RO membranes, though the membrane surfaces bec ame more negatively charged with the increase in pH ( Wagner et al., 2009; Hoang et al., 2010). Another possible reason for increased fouling after Fe (III) salt coagulation might be the complex formation between iron and remaining NOM that may also precipitate on the membrane surface and causes fouling (Hong and Elimelech, 1997). Trebouet et al., (2001) also observed that lea chate without any pre treatment is the best way to use with NF membranes MIEX treated leachate had the highest pH of 8.1 among all three leachate s at the start of experiments and at this pH, most of the alkalinity is generally in the form of carbonate, which does not easily pass through the membrane due to increased

PAGE 47

47 repulsion between carbonates and negatively charged membrane surface s. The precipitation of carbonate alkalinity might have caused even greate r permeate flux reduction in MIEX treated leacha te than other two feed leachate ( Mitsoyannis and Saravacos, 1977 ; Qin et al., 2004). Cornelissen et al. (2010) used an anion exchange resin ( Fluidized Ion Exchange (FIX)) to reduce fouling of NF membranes but did not observe reduced fouling and concluded that after anion exchange treatment the fouling of membranes was more pronounced by other organic matter such as polysaccharides than humic like organic matter These compounds are harder to remove by the MIEX as shown in Figure 2 10 and may cause pronoun ced fouling due to their charge neutrality ( Kwon et al. 2006; Lee and Lee, 2007; Amy, 2008 ) Humber t et al. (2007) also observed increased fouling of ultrafiltration membranes while using MIEX as a pretreatment option for surface water treatment and concluded that t he microbial derived organics present in MIEX pretreated water contribute d to a larger extent in the fouling as compared to the raw water treatment. A complete understanding of increased flux decline was beyond the scope of this study. 2 .4 Conclusions This research was conducted to investigate the treatment of stabilized bioreactor landfill leachate using coagulation, anion exchange, and membrane treatment processes. Coagulation and anion exchange were also studied as a pretreatment step for leachate treatment using membranes. FRI analysis of fluorescence EEM of coagulated leachate showed that the coagulant has highest affinity to remove humic and fulvic like organic matter. Approximately 80% of humic and fulvic like organics were remo ved by coagulation. A maximum of 68%, 71%, and 45% reduction in DOC was obtained from coagulation of

PAGE 48

48 ACL, NCL, and NRL leachate, respectively. All three leachate s had low SUVA values representing smaller fraction of humic and fulvic like organics that le ad to smaller DOC removal from coagulation. A coagulant dose of 22 mmol Fe(III)/L was selected for the membrane experiment because this dose generated optimum pH condition for coagulation in all three leachate s Use of anion exchange resin MIEX showed l esser DOC removal from all three leachate s as compared to coagulation treatment. A maximum of 30%, 20%, and 34% DOC removal was observed from ACL, NCL, and NRL leachate. However, most of the DOC was removed at lower MIEX doses. MIEX removed 30 to 60% of negatively charged humic and fulvic like organics present in leachate. A MIEX dose of 5 mL/L was selected for membrane experiments for membrane experiments. The research showed that a pretreatment (coagulant (Fe III) and anion exchange resin (MIEX)) th at only reduces humic and fulvic like organic matter from stabilized leachate did not improve the permeate flux of NF and RO membranes. An increased permeate flux decline was observed for pretreated leachate as compared to raw leachate in both membrane tr eatments possibly due to a continuous high increase in pH of the pretreated leachates that might have caused increased precipitation of CaCO 3 on the membrane surface. The MIEX pretreated leachate showed even higher permeate flux decline as compared to le achate pretreated with coagulant, possibly due to pronounced fouling caused by microbial derived organics. As future work, further analysis is required to study the cause of increased flux decline of pretreated leachate as compared to raw leachate. Addit ional membrane fouling experiments need to be conducted at fixed pH conditions to determine the effect

PAGE 49

49 of pH on permeate flux. Additionally, the effect of pretreated leachate at variable coagulant and MIEX dose should be studied. A different coagulant (e .g. alum) can also be studied as a pretreatment option to avoid potential iron fouling.

PAGE 50

50 Table 2 1. Previous studies on leachate treatment using coagulation process. Reference Initial concentrations in leachate Coagulant Dose (g/L) Experimental pH condition Maximum COD/TOC* Removal (%) COD or TOC (mg/L) BOD 5 /COD pH Amokarane et al. (1998) 4 100 0.05 8.2 FeCl 3 2 4.9 55 Al 2 (SO 4 ) 3 0.9 5.5 42 Yoon et al. (1998) 282 417 FeCl 3 1.0 48* Wang et al. (2002) 5 800 0.07 7.6 FeCl 3 .6H 2 O 1.0 3 24 Tatsi et al. (2003) 5 350 0.20 7.9 FeCl 3 1.5 10 80 Al 2 (SO 4 ) 3 1.5 6.2 30 70 900 0.38 6.2 FeCl 3 5.5 10 38 Al 2 (SO 4 ) 3 3.0 6.2 40 Ntampou et al. (2006) 1 010 0.17 FeCl 3 1.1 3.5 72 Wang et al. (2009) 600 700 0.01 Fe 2 (SO 4 ) 3 0.6 66.7 Polymeric aluminum 1.0 33 Pi et al. (2009) 18 725 0.05 6.3 FeCl 3 0.7 5 38 Poly FeCl 3 0.6 5 43.6 Li et al. (2010) 2 817 0.05 8.6 FeCl 3 1.7 5.5 68 Al 2 (SO 4 ) 3 3.9 6.0 53 Comstock et al. (2010) 2 076 0.03 8.01 Fe 2 (SO 4 ) 3 0.06 28

PAGE 51

51 Table 2 2. Properties of NF 90 and BW 30 membranes used in the study Parameter NF 90 BW 30 Membrane type 1 Fully aromatic membrane Fully aromatic membrane Operating pH range 1 3 10 2 11 Maximum operating temperature ( 0 C) 1 45 45 Maximum operating pressure (bar (psi)) 1 41.4 (600) 41.4 (600) Pure water flux (m/day) 2 2.6 0.4 1.2 0.3 Salt rejection (%) 2 96.0 0.86 96.7 1.0 Virgin membrane zeta potential (pH 4.5/7) (mV) 3 17.5/ 26.5 3.8/ 5.2 1 From Dow FilmTec ; 2 Pure water flux was considered as permeate flux at the end of 48 h comp action period. Salt rejection was determined for the feed solutio n containing 10 mM NaCl at pH 7; 3 Tang et al. 2007. Table 2 3 Physico chemical characteristics of ACL, NCL, and NRL lea chate during the study period Parameter ACL NCL NRL Average Std dev Average Std dev Average Std dev pH (S.U.) 7.71 0.09 7.61 0.06 7.50 0.05 Conductivity (mS/cm) 13.14 1.4 14.44 7.97 TDS (mg/L) 6 280 65 7 163 19 4 418 22 BOD 5 (mg/L) 46 146.5 410 COD (mg/L) 2 300 46.8 2 225 35.3 2 915 190 BOD 5 /COD 0.02 0.06 0.14 DOC (mg/L) 527 56.6 425 35.9 663 NH 3 N (mg/L) 1060 63 740 42 400 28.3 Alkalinity (mg/L as CaCO 3 ) 6 600 5 400 3 000 UV 254 8.3 0.4 8.58 1.5 14.7 SUVA (m 1 /mg/L DOC) 1.6 0.1 2.01 2.22 Table 2 4. Percentage distribution and total volume (V t ) of DOM in Region I to Region V of ACL, NCL, and NRL leachate derived from FRI analysis Region ACL leachate NCL leachate NRL leachate Region 1 1 1 1 Region 2 6 4 2 Region 3 13 13 12 Region 4 18 15 9 Region 5 61 66 75 V t (AU nm 2 [mg C/L] 1 ) 6 205 5 317 6 926 Fluorescence Index 2.23 2.15 1.61 Table 2 5. Percentage rejection 1 of salts 2 and DOM 3 in permeate in RO and NF experiment Type of feed water Initial concentration RO treatment (BW 30) NF treatment (NF 90)

PAGE 52

52 Conductivity (mS/cm) UV 254 absorbance (cm 1 ) Salt rejection (%) DOM rejection (%) Salt rejection (%) DOM rejection (%) Raw leachate 13.11.4 8.30.4 93.50.3 99.40.1 93.00.3 99.80.1 FeCl 3 treatment 14.41.1 3.80.8 93.50.5 98.80.3 93.10.4 99.60.1 MIEX treatment 11.20.3 5.10.4 93.80.3 99.30.1 92.30.3 99.50.1 1 Percentage rejection of salt and DOM was calculated using average concentrations of permeate samples collected during each experiment and the initial concentration; 2 Salt concentrations was measured using conductivity; 3 DOM was measured as UV 254 absorb ance.

PAGE 53

53 Figure 2 1. Reaction that hydrolysis products of coagulants follow when coagulant is added to water containing natural organic matter. Molecular structure of NOM was adopted from Alvarez Puebla et al. (2006) Figure 2 2 MIEX resin anion exchange process Molecular structure of NOM was adopted from Alvarez Puebla et al. (2006)

PAGE 54

54 Figure 2 3 Laboratory scale Osmonics SEPA CF membrane experimental setup Figure 2 4 Operationally defined excitation and emission wavelength boundaries for five EEM regions (Chen et al. 2003).

PAGE 55

55 Figure 2 5 Fluorescence EEMs for leachate samples (a) ACL leachate (b) NCL leachate (c) NRL leachate

PAGE 56

56 Figure 2 6 Effect of coagulant dose on (a) pH, (b) DOC and (c) SUVA of ACL, NCL, and NRL leachate The data points and error bars represent average and standard deviation of duplicate experiments respectively

PAGE 57

57 Figure 2 7 Effect of coagulation on removal of DOM in each region derived from FRI analysis of ACL, NCL, and NRL leachate

PAGE 58

58 Figure 2 8 Effect of MIEX dose and mixing time on UV 254 absorbance of (a) ACL, (b) NCL, and (c) NRL leachate D ata points and error bars represent average and standard deviation of duplicate experiments respectively

PAGE 59

59 Figure 2 9 Effect of MIEX treatment on (a) DOC, (b) SUVA and (c) pH of ACL, NCL, and NRL leachate D ata points and error bars represent average and standard deviation of duplicate experiments respectively

PAGE 60

60 Figure 2 10 Effect of MIEX on removal of DOM of each region derived from FRI analysis of ACL, NCL, and NRL leachate.

PAGE 61

61 Figure 2 1 1 Effect on normalized permeate flux as a function of filtration time for pretreated (FeCl 3 and MIEX) and raw ACL leachate using NF 90 and BW 30 membrane Figure 2 1 2 Effect on feed pH as a function of time for pretreated (FeCl 3 and MIEX) and raw ACL leachate using NF 90 and BW 30 membrane operation

PAGE 62

62 CHAPTER 3 EFFECT OF OZONATION AS A PRETREATMENT FOR STABILIZED LANDFILL LEACHATE TREATMENT USING HIGH PRESSURE MEMBRANES 3.1 Introduction Though the modern landfill designs minimize the generation of leachate by reducing the influx of moisture, landfills still face a major challenge of managing leachate generated by the percolation of rainwater through the layers of waste. In the process, waste transfers its contaminants into leachate which if improperly managed may pollute surrounding surface and groundwater bodies (Christensen et al., 2001). The characteristics of leachate are complex and influenced by the factors such as age of landfill, type of waste disposed, cover material used in landfill, and management of liquid and gas produc tion at the landfill. Traditionally leachate is characterized based on the age of landfill as young, intermediate, and stabilized leachate. Young leachate contains high amount of biodegradable organic matter and effectively treated by biological treatme nt methods. As the landfills grow old, biodegradable content of leachate reduces and leachate mostly contains biologically refractory organic matter. The stabilized leachate typically contains a ratio of BOD and COD less than 0.1 making p hysico chemical treatment methods such as coagulation, chemical oxidation, adsorption, and membrane systems more effective (Wiszniowski et al., 2006; Renou et al., 2008). Several studies have been conducted to treat stabilized leachate using individual or combination o f more than one treatment methods but often treated leachate do not meet the stringent regulations required to discharge the treated water (Monje Ramirez and Velasquez, 2004; Kurniawan et al., 2006; Ntampou et al., 2006). High pressure membrane filtration s such as nano filtration (NF) and reverse osmosis (RO) have been

PAGE 63

63 e ffective ly used to treat stabilized leachate that can meet the effluent discharge standards (Peters, 1998; Palma et al., 2002; Tabet et al., 2002). However, high concentrations of organic matter and salts present in stabilized leachate tend to adsorb on the membrane surface and block the pores, also termed as fouling reduces the filtration and economic efficiency of the membrane systems. Fouling of membranes is also influenced by the pres ence of specific type of compounds such as h ydrophobic humic and fulvic like organic matter and divalent calcium, pH, and the membrane surface properties ( Hong and Elimelech, 1997 ; Li and Elimelech, 2004; Xu, 2006) Hence, a pretreatment that reduces humi c and fulvic like organic matter from leachate can provide an effective increase in the life of membranes. Ozone has high oxidation potential (E 0 =2.07V) and high reactivity and selectivity toward organic pollutants such as aromatic compounds. Ozone ruptures the C=C bonds or aromatic ring and produces aliphatic acids and aldehydes (Anderson et al., 1985 ; Westerhoff et al., 1998 Jing et al., 2008 ) A t hig h pH conditions (pH>8), ozone produces hydroxyl radicals (O 3 +H 2 2 +2 OH) that have even higher oxidation potential (E 0 =2.8V) than ozone molecule, and accelerate the removal of recalcitrant organic matter from complex wastewater matrix (Wang et al., 2003) A typical ozone reaction mechanism with the organics is presented in Appendix B. Ozone transforms recalcitrant organic matter into more biodegradable form; h ence, ozone has been normally used as a pretreatment step for treating mature landfill leachat e (Monze Ramirez and Velasquez, 2004; Wu et al., 2004; Bila et al., 2005; Ntampou et al., 2006). A summary of pertinent literature on leachate treatment using ozonation process is presented in Table 3 1. These studies were generally focused on

PAGE 64

64 improving the biodegradability of leachate or the use of ozone in combination with other treatment methods such as advance oxidation, coagulation and adsorption (Fettig et al., 1996; Rivas et al., 2003; Bila et al., 2005; Ntampou et al., 2006; Tizaoui et al, 2007 ). Rivas et al. (2003) studied the effect of ozonation on stabilized leachate treatment and observed a maximum of 30% COD reduction for an initial COD of 5,000 mg/L. Bila et al. (2005) studied the effect of ozonation on the biodegradability of mature landf ill leachate and observed an increase in BOD 5 /COD ratio from 0.05 to 0.3 that can be further treated by biological processes. Ntampou et al. (2006) used a combination of ozonation and coagulation processes for removing the COD from the leachate and observ ed 65% COD removal after 60 minutes of ozonation using 2 g/h ozone and an overall COD removal of 80% using the combined process. Though o zone can oxidize all the organic matter into its highest oxidation state the complexity of leachate composition require s high ozone doses and the respective reaction may take longer time causing this process to be economically unfavorable ( Monze Ramirez and Velasquez, 2004 ) However, ozone has high react ivity with potential membrane fouling humic and fulvic like o rganic matter ; hence pre ozonation of leachate may reduce the membrane fouling during leachate treatment. Ozone has been shown to reduce the fouling of membranes during surface and ground water treatment ( Karnik et al., 2005; Lee et al., 2005; Brown et al ., 2008; Kim et al., 2008) Karnik et al. (2005) showed a significant decrease in ceramic membrane fouling by application of ozone gas prior to filtration for lake water containing a maximum TOC of 11.6 mg/L Brown et al. (2008) applied variable doses of ozone to surface water for studying their effect on fouling characteristics of RO membranes and observed

PAGE 65

6 5 increased permeate flux at lower ozonation dose of 0.3 mg ozone/L. H owever, no information could be found on using ozone as a pretreatment option wh ile treating stabilized leachate using nanofiltration (NF) and reverse osmosis (RO) membranes. The objective of this research was to investigate the effectiveness of ozonation as a pretreatment option for treating stabilized landfill leachate using RO an d NF membranes. The experiments were conducted in two phase s wherein at first, an optimum ozone dose was determined for leachate treatment. In the second phase, the ozonated leachate was tested for the time dependent permeate flux and the permeate quali ty. The permeate flux data was used to determine the type of fouling occurred model (Hermia, 1982). 3.2 Experimental Material Methods and Analysis 3.2.1 Materials Leachate samples were collected from three municipal solid waste landfills ; the Alachua County Southwest Landfill (ACL), the North Central Landfill (NCL), and the New River Regional Landfill (NRL) all located in different counties in Florida, USA. All three land fills have been fully or partially operated as a bioreactor landfill where leachate has been added into the waste to accelerate the rate of waste degradation. The ACL has a lined MSW cell that accepted waste for 10 years before closing in 1998 and has bee n operated as a bioreactor since 1990. Leachate samples were collected from the leachate lift station where all the leachate is collected. The NCL has two landfill cells; the first cell was closed in 2000 after accepting waste for 11 years and the second cell was closed in 2007 after eight years of accepting waste. The second cell was operated as a bioreactor landfill. The leachate samples were collected from the

PAGE 66

66 leachate storage tank, where leachate from both the cells is mixed. NRL has three cells ha ving waste ranging in age from 5 15 years. The leachate samples were collected from a common aeration tank where leachate from all the landfill cells is drained and aerated. The leachate samples were collected multiple times in the duration from February 2010 to July 2010 in Nalgene containers and kept at 4 0 C in the dark until used in experiments. Landfill leachate ozonation experiments were carried out in laboratory scale semi batch bubble reactor consisting of a Plexiglas column with height of 45.7 cm a nd internal diameter of 10 cm, having a total volume of 2000 mL, as shown in schematic diagram ( Figure 3 1). Ozone was produced using high purity oxygen as a feed gas to the laboratory scale ozone generator (Pacific lab series ozone generator, Model: L20) and supplied to the ozonation column through a ceramic porous diffuser of porosity 10 to15 m, at the bottom of the reactor. The concentration of ozone in the inlet and the outlet gas stream of the column was measured using gas phase digital ozone analyzer (T API Model 452). The residual gas stream was passed through an ozone destruction unit. Dow filmtec (Minneapolis, MN) provided RO (BW 30) and NF (NF 90) membranes as flat sheet for conducting membrane performance studies. The physical and the operational characteristics of these membranes are presented in Table 3 2 The flat sheet membran es were cut into 14.6 cm x 9.5 cm coupons and stored as dry in the dark. These coupons were soaked in MilliQ water in the dark for 24 hours before use. Membrane filtration experiments were conducted using high pressure cross flow Osmonics SEPA CF Membra ne Cell as shown in the schematic diagram (Figure 3 2).

PAGE 67

67 The membrane coupons were placed into the membrane cell sandwiched between a low foulant feed spacer of thickness 0.86 mm (34 mil) and a permeate carrier of thickness 0.2 mm (7.8 mil). Feed solution was pumped using a variable speed Hydra Cell Industrial pump from a 15 L Nalgene feed tank. A recirculation water bath (RTE 5B, Neslab Instruments Inc.) was used to control the feed water temperature. A hydraulic hand pump (Enerpac, P142) was used to pr essurize membrane cells and the pressure was maintained constant (300 psi) throughout the experiment. T he cross flow velocity and the operation pressure were controlled by v alves connected in the concentrate return line and the bypass line and monitored b y a flow meter (0.13 to 3.15 lpm) and a pressure gauge (0 to 68.9 bar). 3.2.2 Landfill Leachate Ozonation The ozone generator was turned on approximately 20 minutes prior to the start of each experiment to stabilize the rate of ozone generation by the in strument. The concentration of ozone in the ozone/oxygen gas mixture was controlled by adjusting the ozone production rate from the ozone generator by changing the feed gas (oxygen) pressure into the generator and the voltage applied. Before starting th e experiments, leachate was brought to room temperature of approximately 23 0 C. Leachate samples were thoroughly shaken for re suspension of settled solids and 1000 mL of leachate was transferred to the column. To determine the optimum ozonation time for the selected ozone dose, leachate samples were ozonated for 5, 10, 15, and 30 min at an ozone dose of 1 3 8 x 10 4 mol/L ( 66.7 0.3 mg/L ) and feed gas flow rate of 3.5 L/min. At the end of each experiment, samples were withdrawn from the column and analyzed The experiments were conducted without any adjustment of pH and at the end of experiments; samples were analyzed for pH,

PAGE 68

68 dissolved organic carbon (DOC), ultra violet 254 (UV 254) absorbance, and fluorescence excitation emission matrix (EEM). All experi ments were conducted in duplicate. The amount of ozone consumed in ozonation of leachate was calculated by subtracting the ozone concentration of outlet gas stream to the feed gas. 3.2.3 Landfill Leachate Treatment Using Membrane Experiments were condu cted to evaluate the performance of raw and ozone treated leachate using RO and NF membranes. The performance was evaluated by measuring permeate flux and rejection of dissolved organic matter (DOM) and salt. To conduct the membrane experiments, approxim ately 8 to 10 L of all three leachate s was pretreated separately using optimum ozone dose and time as determined by the batch experiments. Membrane experiments were conducted in two steps; wherein membranes were first pre compacted by filtering de ionized water for 24 to 48 hours at constant pressure of 13.8 bar (200 psi), constant cross flow velocity of 20 cm/sec, and at constant temperature of 23 1 0 C, until constant permeate flux (pure water flux) was achieved. After pre compaction run, the membrane syst em was stopped and the de ionized water was drained out from the feed tank. Approximately 8 to 10 L of pre filtered feed leachate was added to the tank. The leachate was pre filtered using 0.7 m glass fiber filter (whatman). The membrane system was the n restarted and operated at similar operating conditions as in pre compaction run for 24 hours. Feed pressure, cross flow velocity, and temperature were maintained constant throughout the experiment. The permeate flux was measured intermittently by measu ring the time taken to a predetermined volume of permeate generated. Concentrate and permeate were

PAGE 69

69 recycled to the feed tank except for the intermittently collected samples for pH, conductivity, DOC and UV 254 absorbance measurement. 3.2.4 Analytical M ethods A multi function Orion Research instrument was used to analyze water quality parameters pH, conductivity, and temperature. Solution pH was measured to an accuracy of 0.02. Temperature was measured in degrees celsius with a scale accuracy of 0. 15 0 C. Conductivity was measured in mS/cm or S/cm depending upon the concentration of sample. Salt rejection in the membrane experiments were measured by measuring conductivity of feed and permeate. All samples were filtered through pre rinsed 0.45 m nitrocellulose filter (Millipore) to analyze DOC, UV 254 absorbance, and EEM. The filtered samples were stored in capped 40 mL glass vials at a 4 0 C in the dark; samples were brought to room temperature before analysis. DOC was measured using Tekmer TOC a nalyzer. UV 254 absorbance was measured using HACH DR 4000 spectrophotometer. Organic free de ionized water was used to calibrate the spectrophotometer for UV 254 absorbance measurement. EEM was measured using a 1 cm quartz cell on a Hitachi F 2500 flu orescence spectrophotometer. The EEM was obtained by scanning the samples at 5 nm increments over an excitation wavelength (E x ) in the range of 200 nm to 500 nm. For each E x the emission wavelengths (E m ) were detected at 5 nm increments in the range of 2 90 nm to 550 nm. To limit the second order Ra y le i gh scattering, a limit of 290 nm cutoff was used for all the samples (Chen et al., 2003). The data were processed in MATLAB following the procedures Chen et al., 2003 and Cory and Mcknight, 2005.

PAGE 70

70 Each tim e the instrument was used, the EEM of de ionized water was analyzed and was subtracted from the EEM response of samples containing DOC to reduce the effect of scattering (Chen et al., 2003). The area under the Raman water peak (at E x 350 nm) was calculate d for de ionized water and the intensities of the EEM response for the samples were normalized by the Raman water area; the EEM response of the samples was then normalized by the DOC concentration; and EEM were plotted in MATLAB using the contour function. The amount of various types of DOM removed by each treatment was quantified using Fluorescence Regional Integration technique (FRI) (Chen et al., 2003). The FRI was used to integrate the area under the EEM spectra as shown in equation 3 1. ( 3 1) where, V i ex em ) is the fluorescence intensity at a particular E x E m wavelength pair, d ex is the increment of E x (5 nm) and d em is the increment of E m (5 nm). EEM were divided into five E x E m regions and concentration of DOM in each region was quantified by integrated volume in each region. The E x E m regions were divided based on the EEM peaks associated with specific types of organic compounds (Figure 3 3). These regions wer e selected using the literature (Chen et al., 2003; Chen et al., 2003; Baker, 2004; Hudson, 2007). Shorter E x (<250 nm) and shorter E m (<350nm) are associated with simple aromatic protein like compounds such as tyrosine (Regions I and Region II). The pea ks at E x in the range of 250 to 280 nm and E m <380 nm are associated with soluble microbial byproducts type compounds and kept in Region IV. The fulvic like compounds (Region III) show peaks at shorter E x (<250 nm)

PAGE 71

71 and longer E m (>350 nm). Peaks at longe r E x and longer E m are associated with humic like compounds and shown in Region V. The fouling occurred during raw and ozone treated leachate treatment using RO and NF membranes Koltuniewicz et al., 1995 is a classic filtration model that uses flux decline data to determine the type and cause of fouling occurred during the filtration proce ss (Reis and Zydney, 2007). The model equations have been previously used by several researchers for various types of feed water with different types of membranes, however, no study was found that analyzes fouling mechanism due to stabilized leachate trea tment using NF and RO membranes ( Koltuniewicz et al., 1995; Keskinler et al., 2004; Brown et al, 2008; Lohwacharin and Takizawa, 2009 ). The model assumes a single pore size of the membrane and the operations are performed at constant pressure in cross flow filtration system The filtration model descri bes four fouling mechanisms during membrane filtration as standard pore blocking, cake filtration, intermediate pore blocking, and complete blocking of pores. For standard pore blocking of mem branes, it is assumed that the particle diameter is less than the pore diameter and pore diameter is decreased by deposition of particles on the pore walls can be expressed as equation (3 2) ; ( 3 2) where, J t is flux rate (L/m 2 h ) at time t ; J 0 is initial flux rate (L/m 2 h) ; K s is standard blocking constant (L 1 ) ; A is area of membrane (m 2 ) ; t is time of filtration; (h)

PAGE 72

72 The cake filtration model (equation 3 3) assumes that particles do not enter into the pores due to their large sizes and form a cake layer over the membrane surface. ( 3 3) where, K c is cake filtration constant (h/L 2 ) Intermediate pore blocking is defined as the pore blockage due to the deposition of particles on the pores as well as on the pore surface and can be expressed as equation 3 4. ( 3 4) where, K i is intermediate pore blocking constant (L 1 ). When the particles seal the membrane pores but do not accumulate on each other, the blocking is assumed as complete blocking and is modeled as equation 3 5. ( 3 5) where, K b is complete blocking constant (h 1 ) The flux data obtained from the membrane experiments were fitted in the Hermia model equations (equations 3 2 to 3 5) for each leachate to determine the type of pore blocking mechanism and to calculate the correlation coefficient (r 2 ) of each model. 3.3 Results and Discussion 3.3.1 Leachate Characterization Relevant physicochemical characteristics of ACL, NCL, and NRL leachate samples collected during the research period are summarized in Table 3 3 The

PAGE 73

73 composition of leachate was relatively constant over the resear ch period. All leachate were characterized by slightly alkaline pH, dark in color, a low concentration of biodegradable content as represented by BOD 5 and high concentration of refractory organic matter represented by COD. Leachate from all three landfi lls has BOD 5 /COD in the range of 0.02 to 0.11 and pH above 7. These properties are consistent with typical stabilized or intermediate stabilized leachate (Reinhart and Grosh, 1998 ; Statom et al., 2004). All three leachate s contained high DOC and UV 254 a bsorbing organic compounds. These types of leachate are characterized as refractory to conventional biological treatment processes and require physicochemical treatment processes for treatment. The fluorescence EEM of all three leachate s is shown in Figur e 3 4. All three leachate s showed distinct peak in the region III representing the presence of fulvic like organics. ACL and NCL landfills are older than NRL, and showed more contour lines in region V, representing higher presence of humic like organic c ompounds in ACL and NCL leachate than NRL leachate. The distribution of DOM in each region was derived from FRI analysis as shown in Table 3 4 Among all three leachates ACL leachate had the highest total volume (V t ) representing maximum concentrations o f DOM in ACL leachate. Additionally, Region V showed maximum volume among all five regions in all three leachates representing leachates contained maximum concentrations of humic like organic matter. The ACL leachate also showed the maximum DOM in region V among all three leachates. 3.3.2 Ozonation Experiments were conducted to determine the efficiency of ozone for treatment of stabilized landfill leachate. An ozone dose of 70 mg/L was passed through the bubble

PAGE 74

74 column for variable durations. The off gas ozone concentration from the column was measured with the ozonation time as shown in Figure 3 5. The data points for all three leachate s of different batch experiments were found overlapping, representing the re producibility of results and the validity of experimental procedure. The off gas ozone concentration was observed to increas e slowly at the start of experiment in all three leachate s representing faster kinetics of ozone reaction at the start of experime nt. After this period, the off gas ozone concentration was observed to increase more or less exponentially due to the decreased availability of organic compounds that can react with ozone. The ACL and NCL leachate showed a slower increase in off gas ozon e concentration than NRL leachate, right from the start of experiment, showing a relatively faster kinetics of ozonation in ACL and NCL leachate. Higher concentrations of humic and fulvic like organic compounds lead to faster ozonation in ACL and NCL leac hate. In the initial alkaline pH conditions of all three leachate s (7 to 8), ozone (O 3 ) molecules directly react with the susceptible organic compounds present in the leachate. Additionally, formation of high oxidation potential hydroxyl radicals ( OH) c ontribute in the oxidation reaction (Kurniawan et al., 2006). The off gas ozone concentration almost platued after an ozonation of 8 to 10 minutes in ACL and NCL leachate and 15 minutes in NRL leachate at lower concentration than input ozone dose, represe nting a constant rate of ozonation after initial faster kinetics of ozonation and less availability of compounds that easily reacts with ozone. It also indicates that ozone reactions are still taking place with possibly the products of initial reactions. If the experiments were carried out for much longer time, the off gas ozone concentration would level off with approximately input ozone concentrations.

PAGE 75

75 The pH of ozonated samples were measured as a function of ozonation time as presented in Figure 3 6 (a) and it was found that in the NRL leachate, pH of ozonated leachate decreased from 7.7 to 7.0 because oxidation of organic matter forms secondary by products of acidic nature causing a decrease in pH of solution. However, application of ozone may also strip carbon dioxide from the leachate a nd the decrease in carbon dioxide may lead to decrease in bicarbonate concentrations in the solution (Wu et al., 2004). This drop in bicarbonate may consume H + ions, causing an increase in pH. The ACL and the NCL l eachate contained high alkalinity, and CO 2 stripping due to ozonation might have caused a slight increase in pH. A 15 to 20% decrease in alkalinity was also observed in all three leachate s with a trend of higher decrease in alkalinity with the ozonation t ime and showed a maximum alkalinity drop after 30 minutes of ozonation. The UV 254 absorbing compounds can be attributed to the presence of unsaturated and aromatic organic compounds with C = C and C = O st ructures such as phenolic, poly aromatic hydrocarbons (PAHs), aromatic ketone s and aromatic aldehydes. A significant reduction in UV 254 absorbance was observed with the increase in ozonation time as shown in Figure 3 6 (b). After 30 minutes of ozonation, more than 65% of UV 254 absor bing compounds from the ACL leachate and 78% from NCL and NRL leachate were either removed or transformed into other forms. Few other researchers also observed approximately similar reduction in UV 254 absorbance during ozonation of similar type of leacha te (Ntampou et al., 2006; Rivas et al., 2003). Ntampou et al. (2006) observed a 70% reduction in UV 254 absorbance and a 50% reduction in COD after 30 minutes of ozonation for an initial UV 254 absorbance and

PAGE 76

76 COD of 6.76 cm 1 and 1010 mg/L, respectively. Rivas et al. (2003) observed a 60% reduction in UV 254 absorbance after 30 minutes of ozonation at initial UV 254 absorbance of 1.5 cm 1 The rate of reduction of UV 254 absorbance was almost similar in all three leachate s independent of their initial value; however, UV 254 absorbance reduction was found faster at the start of ozonation, which slowed down with the increase in ozonation time. The possible reason behind this type of ozone kinetics is the mechanism of ozone reaction, where molecular ozone quickly oxidizes the unsaturated bonds of aromatic rings and form by products such as aliphatic acids and aldehydes. These by products slow down the oxidation reaction (Westerhoff et al., 1999; Tizaoui et al., 2007). A maximum of 13 to 17 % reduction in DOC was observed in all three leachate s ; after 30 minutes of ozonation (Figure 3 6 (c )), which is comparable to the results of previous studies. Cortez et al. (2010) observed approximately 11% TOC (initial TOC 284 mg/L) reduction in landfill leachate for almost similar ozone dose. With the increase in ozonation time, the reduction of DOC increased; however, the rate of DOC reduction was not always constant in all three leachate s All three leachates showed a faster DOC removal at the start of ozonation which slowed down as the ozonation progressed. Ntampou et al. (2006) observed a similar trend for COD removal with the ozonation time. A slightly higher DOC removal rate was observed in the NRL leachate as compared to ACL and NCL leachate because mineral ization of humic and fulvic like organic matter is harder than smaller organic matter and ACL and NCL leachate had higher concentrations of humic and fulvic like organics than NRL leachate. The se

PAGE 77

77 results are consistent with the off gas ozone concentration s also, such that lower the off gas ozone concentration, higher the DOC removal and with the increase in ozonation time, the rate of DOC removal became almost constant. ACL leachate showed highest amount of DOC removal among all three leachate s at the 30 minute ozonation time due to its slightly increased pH. Increased pH conditions increase the formation of OH radicals that has higher oxidation potential than ozone leading to faster oxidation of organic matter (Kurniawan et al., 2006) The rate of DOC removal was observed lower than that of UV 254 absorbance removal, representing faster ozone reaction with the unsaturated and aromatic compounds and smaller rate of complete mineralization of DOC. Effect of ozonation on the removal of different types o f organics with the increase in ozonation time was derived from the FRI analysis. The normalized concentrations of DOM in each region are presented in Figure 3 7 All three leachate s showed almost similar pattern of DOM removal, such that, DOM present in region V and region III had a maximum reduction at each ozonation time. Ozone preferentially reacts with humic and fulvic like organic compounds that contain aromatic and unsaturated organic compounds. These humic and fulvic like compounds break into sm aller aliphatic organic compounds; confirming the slightly smaller rate of DOM removal of organics present in region I and region II. The rate of DOM removal became almost constant after 10 minutes of ozonation in all three leachate s The kinetics of UV 254 absorbance, DOC, and DOM in region III and region V removal was observed to be decreased after 10 minutes of ozonation; hence, an optimum ozonation time of 10 minutes was selected for the membrane experiments. An

PAGE 78

78 ozonation of 10 minutes correspond s t o an average of 55% UV 254 absorbance and 8% of DOC removal in all three leachate s The pH after 10 minutes of ozonation was in the range of 7 to 8 for all three leachate s which is also within the pH range of raw leachate. 3.3.3 Leachate T reatment U sin g M embranes As expected, RO and NF membrane treatment for raw and ozone treated leachate showed a high rejection efficiency of salts and DOM (Table 3 5 ). RO membrane rejected approximately 93 to 96% salts from RO membranes and 91 to 95% from the NF membra ne for all three leachate s treatment. The DOM as measured by the UV 254 absorbance was rejected to more than 99% by RO and NF membrane treatment of all three leachate s. The effect on permeate flux for treatment of raw and ozone treated leachate using RO and NF membrane is shown in Figure 3 8. A slight fluctuation in the data was observed that can be explained by the smaller variation s in feed temperature, pressure, and random sampling during the experiment At the start of experiment (for approximately 2 hours), the permeate flux decreased rapidly in all conditions, which slowed down for remainder of the experiment. The pattern of flux decline in all conditions were fitted in the power law model equation (y=ax b ) with a value of r 2 >0.87 as shown in Table 3 6 Hyuang et al. (2000) evaluated ozonation as a pretreatment on flux parameters of ultra filtration membranes for water treatment and observed similar pattern in the permeate flux decline. The derived power law equations can be used to predict t he performance of these RO and NF membranes for treating raw and ozone treated leachate in the applied conditions.

PAGE 79

79 A 10 to 14% decrease in permeate flux was observed for raw leachate treatment using RO membrane, whereas the NF membranes showed a flux decli ne of approximately 18 to 19% after 24 hours of experiment. Though, leachate pretreated with ozone had approximately 4%, 12%, and 10% less DOC and 54%, 64% and 59% less UV 254 absorbance with respect to raw ACL, NCL, and NRL leachate, respectively; leacha te pretreatment with ozone did not reduce fouling of membranes and a faster permeate flux decline was observed, for the applied experimental conditions. The RO treatment of these leachate showed a permeate flux reduction of 17 to 23% at the end of experim ent. The NF treatment showed even greater flux decline of 23% in ACL leachate and 32% in NCL and NRL leachate. Clearly, the reduction in DOC and UV 254 absorbance after ozonation did not reduce the fouling of membranes. The increased fouling can be cau sed by the factors such as constituents of feed water membrane properties and the hydrodynamic conditions (Bellona et al., 2004; Tang et al., 2007). During the experiments, the hydrodynamic conditions were kept constants; hence, the fouling was mainly c aused by different characteristics of raw and ozone treated leachate and different membrane surface properties of RO and NF membranes. The pH of each solution was in the range of 7 to 8 for all three leachates and during the membrane experiments the pH of each solution were not significantly changed. Hence, the changes in pH of feed solution during membrane experiments should not have a significant effect on permeate flux. A mong all three leachate s, ACL leachate contained slightly higher concentrations of humic and fulvic like organic compounds (Table 3 4 ) that are hydrophobic in nature

PAGE 80

80 and have tendency to adsorb on the hydrophobic membrane surfaces at high pressure operations, which might be a possible reason of slightly higher flux decline in ACL leac hate as compared to NCL and NRL leachate (Bellona et al., 2004; Tang et al., 2007). Additionally, the membrane surface roughness plays an important role in fouling such that higher the roughness, higher the fouling potential in presence of humic and fulvi c like organic matter (Hobbs et al., 2000) NF membranes have higher surface roughness than RO membranes (Table 3 2 ), which explains higher flux decline of NF membranes than RO membranes In order to determine the possible reasons of increased flux decl ine of ozone treated leachate than raw leachate, the was used. The model uses flux decline data and determines the type and cause of fouling. As presented in Table 3 7 and Table 3 8 each membrane showed closely related correlat ion coefficients (r 2 ) that vary to a small percentage (0 to10%) for standard, cake, intermediate and complete fouling mechanisms. This implies that the membranes have faced all four types of fouling by each feed waters throughout the experiment duration; however the rate of occurrence of each fouling was not the same. Among all four types of fouling constants, K b was highest in all leachate treatment experiments which shows the complete pore blocking mechanism w as the most dominant fouling processes and the particles of the feed water block the pores by accumulating on the membrane surface but not over each other Comparing the results between raw and ozonated leachate treatment showed that the ozonated leachate h ad higher fouling constant values than the raw leachate in all experiments, showing a faster flux decline in ozon e treated leachate. Ozonation may increase the particle sizes due to ozone induced

PAGE 81

81 particle destabilization (Lee et al., 2005) The destabili zation reduces the net negative charge and thus increases the possibilities of particle collision and coagulation. Ozonation increases the concentration of carboxylic acids in the solution that may cause increased association with calcium present in the l eachate, resulting in precipitation of complexes of calcium on the membrane surfaces and reduced permeate flux. Chandrakanth and Amy (1996) found that calcium in feed water act as a coagulant aid for the by products formed during ozonation of NOM along wi th the formation of increased Ca NOM complexes. The increased formation of Ca NOM complexes and coagulated NOM over calcium may precipitate on the membrane surface and increase s flux decline. Karnik et al. (2005) observed that i ncreasing the ozone concen tration beyond a threshold value had no beneficial effect on permeate flux recovery Brown et al. (2008) also observed a decrease in permeate flux of RO membranes due to increase in particle size s at high ozone doses caused by coagulation of calcium and N OM., 3.4 Summary and Conclusions Application of high pressure membrane system such as NF and RO can provide a efficient landfill leachate treatment option, however, the frequent fouling of these membranes due to the presence of higher concentrations of humic and fulvic like organics limits its operation. The l iterature suggest that the use of ozone can decrease the concentrations of DOM and COD from the landfill leachate but there is no literature available on the effectiveness of ozone as a pretreatment method for treating landfill leachate using high pressure membranes. The first objective of this research was to determine the effectiveness of ozone for treating stabilized landfill leachate and to determine optimum ozone dose for membrane treatment operations. A fixed ozone dose of 70 mg/L was used at a feed gas flow rate of 3.5 L/min for 5 to 30 minutes to

PAGE 82

82 tr eat leachate from three different landfills. A maximum of 78% drop in UV 254 absorbance and 23% drop in DOC was observed. Faster ozone kinetics was observed at the start of experiments, which platued after 8 to10 minutes of ozonation in all three leachat e s The characterization of organic matter using fluorescence EEM and FRI analysis also showed that the ozone removed most of the humic and fulvic like organics at the start of ozonation. The second objective was to evaluate the performance of membrane operations for treatment of raw and ozone treated leachate. An optimum ozonation time of 10 minutes was used to run the membrane experiment. RO and NF membranes rejected greater than 99% of DOM and 91% salts from all three leachate s in the experiment dur ation. The flux decline with the operation time was measured, which followed a power law equation; such that a faster flux decline at start of membrane operation was observed. The developed power law equations can be used to predict the performance of th ese RO and NF membranes for treatment of these types of leachate. A faster flux decline was observed for the ozonated leachate treatment than raw leachate in all conditions and NF membranes showed a faster flux decline as compared to the RO membranes. Th flow, constant pressure filtration model and the results suggest that the fouling was primarily caused by the blocking of the membrane pores by the particles that accumulate on the membrane surface but not over each other thereby not allowing the filtrate to pass through. The rates of fouling constants were found higher in NF membranes than RO membranes confirming a higher flux decline in NF membranes.

PAGE 83

83 This research showed that the use of ozonation as a pre treatment option for treating stabilized leachate using high pressure membranes did not help increase the permeate flux In the applied experimental conditions, the ozonation of leachate before membrane treatment did not reduce the fouling potential of me mbranes. The ozonation provided a significant reduction or transformation of humic and fulvic like organic matter from leachate; however, the selected ozone dose for membrane treatment might have caused additional complexation of calcium and the by produc ts of ozonation reaction, which might have reduced the permeate flux of membrane. Hence, a careful selection of ozone dose is required to consider ozonation as a pretreatment option for stabilized leachate treatment using membranes. Furthermore, membran e fouling experiments should be conducted for the pretreated leachate using variable ozone doses. Possibly, a lower ozone dose can be used as compared to the dose used in the present study to pre treat leachate. Lower doses may cause lesser coagulation c aused by the calcium ions present in leachate. Anti scalant can also be used as an additional pretreatment to reduce the possible coagulation caused by divalent ions.

PAGE 84

84 Table 3 1. Previous studies on a pplication of ozonation for landfill leachat e treatment Initial concentrations in leachate Maximum COD or DOC* Removal by ozonation (%) Ozone dose Additional treatment with ozone Reference COD/DOC* (mg/L) BOD 5 /COD pH 640/205** 8.24 Adsorption+O 3 Fettig et al. (1996) 518 8.3 66 1.7 g O 3 /g COD Biological treatment +O 3 Baig and Liechti (2001) 5 000 0.17 8 9 30 1.3 3.5 g O 3 /hr Adsorption+O 3 Rivas et al. (2003) 3 460 0.04 8.2 48/ 10 TOC 0.5 3 g O 3 /L Coagulation+O 3 and NH 3 stripping Silva et al. (2004) 6 500/4 000 TOC 0.08 8.1 15 TOC 1.2 g O 3 /L Coagulation+O 3 O 3 /H 2 O 2 and O 3 /UV Wu et al. (2004) 1 090 0.04 8.3 75 1.2 12.5 g O 3 /L O 3 and O 3 /H 2 O 2 Wang et al. (2004) 4 000 0.05 8 8.5 70/48* 0.5 9.0 g O 3 /L Coagulation+O 3 Bila et al. (2005) 1 010 0.17 8 80 1.5 2.0 g O 3 /hr Coagulation+O 3 Ntampou et al. (2006) 5 230 0.09 8.7 27 1.1 g O 3 /L O 3 and O 3 /H 2 O 2 Tizaoui et al. (2007) 743/284 TOC 0.01 3.5 10/7 TOC 8 g O 3 /hr O 3 and O 3 /H 2 O 2 Cortez et al. (2010)

PAGE 85

85 Table 3 2 Properties of reverse osmosis (BW 30 ) and nano filtration (NF 90) membranes used in the study Parameter BW 30 NF 90 Membrane type 1 Fully aromatic membrane Fully aromatic membrane Operating pH range 1 2 11 3 10 Maximum operating temperature ( 0 C) 1 45 45 Maximum operating pressure (bar (psi)) 1 41.4 (600) 41.4 (600) Pure water flux (m/day) 2 1.1 0.4 2.4 0.3 Salt rejection (%) 2 96.7 1.0 96.0 0.86 Virgin membrane zeta potential (pH 4.5/7) (mV) 3 3.8/ 5.2 17.5/ 26.5 Surface roughness (nm) 3 68.3 ( 12.5 ) 129.5 ( 23.4 ) 1 From Dow FilmTec ; 2 Pure water flux was considered as permeate flux at the end of 48 h compaction period. Salt rejection were determined for the feed solutio n containing 10 mM NaCl at pH 7; 3 Tang et al., 2007. Table 3 3 Leachate characteristics of ACL, NCL, and NRL leachate during the study period Parameter ACL NCL NRL Average Std dev Average Std dev Average Std dev pH (S.U.) 7.71 0.01 7.47 0.02 7.74 0.05 Conductivity (mS/cm) 13.79 1.4 8.36 7.90 TDS (mg/L) 6 546 30 4 221 11 4 034 24 BOD 5 (mg/L) 57 135 260 COD (mg/L) 2 286 46 2 166 3 3 2 0 25 190 BOD 5 /COD 0.02 0.06 0.1 2 DOC (mg/L) 523 10.7 413 9.9 276 2.4 Alkalinity (mg/L as CaCO 3 ) 6 800 3 200 800 UV 254 (cm 1 ) 9.0 11.9 7.1 SUVA (m 1 /mg/L DOC) 1.7 2.9 2.6 Table 3 4 Percentage distribution and total volume (V t ) of DOM in Region I to Region V of ACL, NCL, and NRL leachate derived from FRI analysis Region ACL leachate NCL leachate NRL leachate Region I 1 1 0 Region II 5 4 2 Region III 13 13 12 Region IV 17 14 10 Region V 64 68 76 V t (AU nm 2 [mg C/L] 1 ) 4 946 4 185 3 846 Fluorescence Index 2.21 2.04 1.89 Table 3 5 Percentage rejection 1 of salts 2 and DOM 3 from raw and ozonated ACL, NCL, and NRL leachate treatment using RO and NF membranes Leachate Feed Initial concentration RO treatment NF treatment

PAGE 86

86 from leachate Conductivity (mS/cm) UV 254 (cm 1 ) Salt rejection (%) DOM rejection (%) Salt rejection (%) DOM rejection (%) ACL Raw 13.20.2 8.00.4 93.50.5 99.30.1 91.50.4 99.70.01 Ozonated 12.90.2 4.00.1 92.60.6 98.80.1 91.40.3 98.80.04 NCL Raw 12.90.2 12.10.3 93.10.9 99.40.1 93.10.4 99.80.02 Ozonated 12.80.1 4.20.1 93.60.4 98.60.2 94.00.2 99.30.04 NRL Raw 6.80.1 6.40.1 96.10.2 99.70.1 95.90.1 99.70.02 Ozonated 6.70.1 2.40.1 96.40.3 99.30.1 95.60.2 99.50.05 1 Percentage rejection of salt and DOM was calculated using average concentrations of permeate samples collected during each experiment and the initial concentration ; 2 Salt concentrations was measured using conductivity ; 3 DOM was measured as UV 254 absorbance Table 3 6 Power law equation and the value of r 2 of flux decline with ozonation time for raw and ozonated leachate treatment using RO and NF membranes Leachate from Type of Feed leachate RO NF Equation r 2 Equation r 2 ACL Raw y=1.0163x 0.051 0.90 y=0.9779x 0.058 0.87 Ozonated y=0.9448x 0.041 0.96 y=0.9576x 0.063 0.92 NCL Raw y=0.9784x 0.028 0.93 y=0.9422x 0.047 0.96 Ozonated y=0.9595x 0.068 0.92 y=0.9282x 0.094 0.96 NRL Raw y=0.9722x 0.022 0.92 y=0.9465x 0.04 0.89 Ozonated y=0.9680x 0.051 0.91 y=0.8943x 0.084 0.91 y=Permeate flux (m/day) ; x= Ozonation time (hrs) Table 3 7 The correlation coefficient (r 2 ) and the filtration constants of the Hermia's filtration model for raw and ozonated ACL, NCL, and NRL leachate treatment using RO membranes Leachate from Type of Feed leachate Standard Cake Intermediate Complete r 2 K s (L 1 ) r 2 K c (hr.L 2 ) r 2 K i (L 1 ) r 2 K b (h 1 ) ACL Raw 0.98 1.97E 03 0.98 1.04E 03 0.98 2.88E 03 0.98 6.90E 03 Ozonated 0.86 2.85E 03 0.89 1.30E 03 0.87 3.60E 03 0.85 9.50E 03 NCL Raw 0.85 1.93E 03 0.86 7.80E 04 0.86 2.16E 03 0.85 6.00E 03 Ozonated 0.78 4.56E 03 0.98 1.82E 03 0.88 5.05E 03 0.87 1.51E 02 NRL Raw 0.89 7.44E 04 0.80 1.82E 04 0.89 7.21E 04 0.88 4.30E 03 Ozonated 0.88 2.28E 03 0.89 5.20E 04 0.88 2.16E 03 0.87 9.60E 03 Table 3 8 The correlation coefficient (r 2 ) and the filtration constants of the Hermia's filtration model for raw and ozonated ACL, NCL, and NRL leachate treatment using NF membranes Leachate from Type of Feed leachate Standard Cake Intermediate Complete r 2 K s (L 1 ) r 2 K c (hr.L 2 ) r 2 K i (L 1 ) r 2 K b (h 1 ) ACL Raw 0.91 4.39E 03 0.93 1.30E 03 0.92 4.33E 03 0.90 1.43E 02 Ozonated 0.90 4.29E 03 0.93 1.30E 03 0.91 4.33E 03 0.89 1.50E 02 NCL Raw 0.87 2.42E 03 0.89 7.80E 04 0.95 4.33E 03 0.86 1.01E 02

PAGE 87

87 Ozonated 0.85 4.55E 03 0.90 1.30E 03 0.87 5.05E 03 0.83 2.13E 02 NRL Raw 0.84 2.17E 03 0.86 2.60E 04 0.84 1.44E 03 0.83 1.22E 02 Ozonated 0.93 3.16E 03 0.96 7.80E 04 0.94 3.60E 03 0.91 2.04E 02

PAGE 88

88 Figure 3 1. Schematic diagram of laboratory scale ozonation setup Figure 3 2. Experimental setup of laboratory scale Osmonics SEPA CF membrane

PAGE 89

89 Figure 3 3. Operationally defined excitation and emission wavelength boundaries for five regions of excitation emission matrix (Chen et al., 2003). Figure 3 4. Fluorescence EEM of (a) ACL, (b) NCL, and (c) NRL leachate

PAGE 90

90 Figure 3 5 Evolution of the off gas ozone concentration as a function of ozonation time for ACL, NCL, and NRL leachate

PAGE 91

91 Figure 3 6. Effect of ozonation on (a) pH, (b) UV 254 absorbance, and (c) DOC of ACL, NCL, and NRL leachate The data points and error bars represent average and standard deviation of duplicate experiments respectively

PAGE 92

92 Figure 3 7. Effect of ozonation on removal of DOM for each EEM region derived from FRI analysis of (a) ACL, (b) NCL, and (c) NRL leachate

PAGE 93

93 Figure 3 8. Effect on normalized permeate flux for treatment of raw and ozonated (a) ACL, (b) NCL, and (c) NRL leachate using BW 30 and NF 90 membrane

PAGE 94

94 CHAPTER 4 EQUILIBRIUM AND INTRA PARTICLE DIFFUSION OF STABILIZED LANDFILL LEACHATE ONTO MICRO AND MESO POROUS ACTIVATED CARBON 4.1 Introduction L eachate is generated when rainwater percolates through the layers of waste in the landfill In the process, waste transfer pollutants to the percolating water, which might cause serio us problems to the surrounding surface and groundwater bodies if not properly managed (Christensen et al., 2001). Hence, leachate treatment is an essential part of effective landfill management. Several treatment methods have been reported in the literat ures that help reduce the hazardous nature of leachate; however, selection of treatment method depend upon leachate characteristics that change with age, landfill operation method, moisture availability, waste composition, and climate In particular, comp osition of leachate varies greatly on the age of the landfill and can be characterized as young, intermediate, and stabilized leachate (Kjeldsen et al., 2002). B iological methods have been found effective to reduce the contaminants from the leachate gener ated from young landfills, as it contains high concentrations of biodegradable organic matter (Timur et al., 2000 ; Borghi et al., 2003 ). However, as the landfill grows old, leachate contains mostly refractory organic matter, making biological treatment me thods less effective. When the leachate contains BOD 5 /COD<0.1, it is termed as stabilized leachate and p hysicochemical treatment methods have been found more effective for treating such old leachate Several different treatment methods s uch as coagulation and flocculation (Tatsi et al., 2003), chemical oxidation (Monze Ramirez and Velasquez, 2004), membrane based technologies (Chianese et al., 1999; Trebouet et al., 2001) and activated carbon adsorption (Kurniawan et al., 2006; Maranon et al., 2009) have

PAGE 95

95 been investigated however, high concentrations of refractory organic matter present in stabilized leachate always limit their treatment efficiency. Adsorption on to activated carbon (AC) has been reported as an effective method for the removal of high molecular weight refractory organic matter from aqueous solution ( Halim et al., 2010 ) Foo and Hameed (2009) have recently reviewed the studies on the landfill leachate treatment using AC and have found it a potentially viable leachate treatment method. V ery few studies can be found on the use of AC for landfill leachate treatment, either as a single treatment step or in combination with other treatment options as presented in Table 4 1. Generally, the AC adsorption process has been used in com bination with other treatment methods such as biological treatment methods, ozonation, and coagulation processes and has shown a wide range of treatment efficiencies for organic matter. P owdered AC (PAC) has been often found to be used with biological tre atment methods. Kargi and Pamukoglu (2003) used a PAC with biological treatment system and achieved 49% COD removal efficiency for an initial COD of 7 000 mg/L. Granular AC (GAC) has been often used with physico chemical treatment processes in leachate t reatment; however, Morawe et al. (1995) used GAC as a polishing step after biological treatment of landfill leachate and observed a 70% COD removal efficiency for an initial COD of 940 mg/L. Kurniawan et al. (2006) investigated the treatment of landfill l eachate using ozone and G AC together and achieved an 86% reduction in COD from an initial COD of 7 811 mg/L Maranon et al. (2009) used AC adsorption as a tertiary treatment option for biologically pretreated landfill leachate and achieved a maximum of 63 % reduction in COD from an initial COD of 785 mg/L The

PAGE 96

96 COD adsorption capacities were in the range of 0.06 to 0.56 g /g AC depending upon the type of leachate and the additional treatment used with AC Several other studies have also been reported in t he literature, however, all these studies have used micro porous AC (pore size < 2 nm) due to their higher adsorption capacity ( Morawe et al., 1995; Gotvajn et al., 2009 ). However, use of meso porous AC (pore size 2 to 50 nm) can be an ideal option for stabilized leachate treatment, which has not been reported in the literature. Stabilized leachate contains high molecular size organic matter that might not be able to en ter into commonly used micro porous structure of AC and may block the pore openings at the surface of granule consequently, reducing the adsorption capacity (Pignatello et al., 2006). H owever, the meso porous AC have bigger pore opening a nd should allow the large organic molecules to diffuse within the pore and not affecting the adsorption capacity. In addition to the availability of very few studies on the use of AC for leachate treatment, the studies in terms of adsorption isotherm determination and d iffusion of organic matter onto AC, which are the primary factors to design and optimize a full scale or pilot scale AC treatment system has been rarely found in the literature. Hence, in the present research, the adsorption of organic matter present in s tabilized leachate was studied onto three different AC s which were selected based upon their pore sizes One micro porous and two meso porous AC s were selected to compare their adsorption isotherm profile, the rate limiting adsorption processes, and the rate of organic matter diffusion. Rapid small scale column tests (RSSCT) were conducted to determine the amount and the type of organic matter removed using these three AC s for leachate treatment.

PAGE 97

97 4.2 Experimental Material and Methods 4.2.1 Landfill L eachate and Activated Carbons Experiments were conducted using leachate generated from a lined and capped waste disposal unit of the Alachua County Southwest Landfill (ACL) located in Florida, USA. Landfill accepted municipal solid waste in the duration o f May 1988 to December 1998 and has been operated as a bioreactor landfill since 1990 through various leachate recirculation activities. Leachate was collected from the leachate lift station, where all the leachate from the landfill is drained. Leachate samples were collected multiple times in the duration from August 2009 to December 2009 in Nalgene containers and kept at 4 0 C in the dark until used in experiments. Relevant physico chemical characteristics of leachate are presented in Table 4 2 The com position of leachate was relatively constant over the research period and was characterized by slightly alkaline pH, dark in color, with a low concentration of biodegradable content as represented by BOD 5 and high concentration of refractory organic matte r as represented by COD. These properties are consistent with typical stabilized leachate (Statom et al., 2004). Three AC used in the study were Calgon Filtrasorb F 300 (Calgon Carbon Corporation, Pittsburgh PA), Norit HD 4000 (The Netherlands), and Dar co 12x40 (Norit Americas Inc. Texas). General characteristics of AC are detailed in Table 4 3 AC s were selected with varying pore sizes. The Calgon F 300 showed an average pore size diameter of 18 with predominantly micro porous AC structure. The Norit HD 4000 and the Darco 12x40 were predominantly meso porous AC s with pore size diameters of 32 and 42 respectively.

PAGE 98

98 4.2.2 Batch Experiments B atch experiments were conducted for developing the adsorption isotherms and to determine the intra par ticle diffusivity of organic matter onto selected AC. Leachate was filtered using a glass fiber filter prior to each experiment. All experiments were conducted in duplicate for quality control purposes and at constant temperature of 23 0 C. Isotherm experi ment Isotherm experiments were conducted using predetermined adsorbent doses of 0 to 100 g /L. Activated carbon from 0 to 10 g was added in separate cleaned, pre dried amber bottles and 100 mL of filtered leachate was added in each amber bottle. Amber bo ttles were kept on a rotary shaker, which was operated at 150 rpm. Amber bottles were wrapped with aluminum foil to avoid the ir exposure to light. S amples were shaken for approximately 240 hours until the equilibrium was achieved. Rivas et al. (2006) al so achieved the equilibrium within 200 hours while studying the kinetics of adsorption of landfill leachate onto AC. After equilibration time, samples were filtered using glass fiber filter to avoid AC particles entering into the samples to be analyzed. Samples were analyzed for total organic carbon (TOC). Diffusivity experiment The intra particle diffusivity of organic matter was determined by short term kinetic experiment using AC particles of different sizes. Activated carbon were crushed and sieved through the sieve numbers 20, 35, 50, and 200. P articles retained at the top of the sieves were used in the experiment. All fractions were cleaned with deionzed water and were dried at 105 0 C for 24 hours. Each type and particle size of AC were weighed to 1 g and taken into clean dried 40 mL glass vials. Glass vials were filled with 30 mL leachate and kept on the rotary

PAGE 99

99 shaker, which was operated at 150 rpm. After every hour, one sample containing each size AC was removed from the shaker and samples w ere collected for TOC analysis by filtering leachate AC solution using glass fiber filters. 4.2.3 Rapid Small Scale Column Tests Crittenden et al. (1986, 1991) presented the methodology for predicting full scale AC column performance using RSSCT s that can generate breakthrough profile in a very short period of time as compare d to full scale or pilot scale studies. RSSCT s also require lesser volume s of water for testing performance of AC to that water (Crittenden et al., 1986; 1991). The small scale colu mns used in the study were designed depending upon the design parameters of a full scale column ( Table 4 4 ) and the scaling equations (equations 4 1 and 4 2) developed by Crittenden et al. (1991). ( 4 1) ( 4 2) where, EBCT is e mpty bed contact time (min), R is AC particle radius (cm), t is t ime (min) to complete the tes t, V is h ydraulic loading rate (m/min), and R e is R eynolds number The subscripts SC and LC represent the small and long column. The calculated design parameters of RSSCT columns for each AC are presented in Table 4 4 RSSCT columns (0.5 cm diameter x 12 cm length) were made up of quartz glass with PTFE inserts to prevent the sorption of organic matter. Syringe pump (Harvard apparatus model 33) was used to supply the constant flow rate to the columns. The flow rate of leachate to the column was monitored by the flow meter installed in the syringe pump. E xperiments were conducted at 6.6, 10, 15 30, and 60 minutes empty

PAGE 100

100 bed contact time (EBCT) of long column; corresponding EBCT for small scale columns as calculated by equation 4 1 and 4 2 for each AC are listed in Table 4 5 Samples were collected after 15 minutes of small scale column run and were analyzed for TOC to determine the amount of organic matter removed at each EBCT. Type of organic matter adsorption onto AC with respect to EBCT was analyzed by characterizing the organic matter present in the samples using fluorescence spectroscopy. The samples were filtered using 0.45 m filter before analyzing the fluorescence spectrum. 4.2.4 Analysis A multi function Orion Research instrument was used to analyze water quality parameters pH and temperature. TOC was measured using TOC V CPH/CPN an alyzer and the fluorescence excitation emission spectrum was measured using the Hitachi F 2500 Fluorescence spectrophotometer. Excitation emission matrix (EEM) was generated for the excitation wavelength (E x ) of 200 nm to 500 nm at 5 nm increments and for each E x the emission wavelength (E m ) was detected at 5 nm increments in the range of 290 to 550 nm. Data were processed in MATLAB following the procedures developed by Chen et al. (2003) and Cory and Mcknight (2005). First, the effect of scattering was reduced by subtracting EEM response of de ionized water to EEM response of DOM containing samples followed by calculating the area under the Raman water peak (at E x 350 nm) for de ionized water. Then, the intensities of the EEM response for the DOM co ntaining samples were normalized by the Raman water area. Afterwards; the calculated EEM response of the sample was normalized by the DOC concentration; and EEM were plotted in MATLAB using the contour function.

PAGE 101

101 The contour plot of EEM response was divi ded into five E x E m regions based on the EEM peaks associated with specific types of organic compounds. These regions were decided using the literature Chen et al. (2003); Chen et al. (2003); Baker (2004); Hudson (2007). The amount of DOM present in each region was quantified using Fluorescence Regional Integration (FRI) technique (Chen et al., 2003). The volume of each region under the EEM spectra was calculated using equation 4 3. ( 4 3) where, V i represents the volume unde ex em ) is the fluorescence intensity at a particular E x E m pair; d ex is the increment of E x (5 nm), and d em is the increment of E m (5 nm). 4.3 Results and Discussion 4.3.1 Adsorption Equilibrium Study To effectively design an adsorptio n system, it is essential to develop most appropriate mathematical description of adsorption isotherms. Several isotherm equations (e.g. equations 4 4, 4 5, and 4 6) have been developed for the AC and organic matter adsorption process and the parameters o f these equations express the surface properties and the affinity of adsorbent at a fixed temperature and pH. Each model varies with respect to their assumption of thermodynamic adsorption process and has their own set of advantages and disadvantages. Th e Freundlich isotherm (equation 4 4) has been widely used and can be applied for heterogeneous surfaces along with multi layer adsorption processes and the amount of adsorbate adsorbed onto adsorbent increases infinitely with increase in concentration; how ever, it does not fit data very well for the low concentration systems (Freundlich, 1906). The Langmuir adsorption

PAGE 102

102 isotherm (equation 4 5) assumes the monolayer sorption process, which limits its applicability at high concentration systems (Langmuir, 1916 ). Freundlich and Langmuir isotherms are two parameter isotherm equations with their own limitations; Redlich and Peterson (1959) developed a three parameter isotherm equation (equation 4 6) to incorporate the features of both Freundlich and Langmuir isot herms. At low concentrations, the Redlich Peterson isotherm follows monolayer sorption and at high concentrations its behavior approaches to Freundlich isotherm. Freundlich isotherm: ( 4 4) Langmuir isotherm: ( 4 5) Redlich Peterson isotherm: ( 4 6) w here, Q e is the amount of solute adsorbed at equilibrium (mg/g); C e is the concentration of adsorbate in solution at equilibrium (mg/L); K F and n are adjus table parameter in Freun dlich equation; Q m and b are adjus table parameter in Langmuir equation; K R a R and b R are adjustable parameter in Redlich Peterson equation. The most common approach to assess the adsorption capacity using above mentioned isotherm equations is to use the linear regression analysis of the experimental data and the isotherm with R 2 closest to one was assumed to be the best fit isotherm equation. However, transforming non linear equation to linear form may generate an error in analysis; hence, a non linear a nalysis approach as developed by Ho et al. (2001) was followed for each isotherm. Non linear regression error functions

PAGE 103

103 (equations 4 7 to 4 11) were used to assess most suitable isotherm models that fit to the experimental data. Sum of absolute errors: ( 4 7) Sum of error square: ( 4 8) Composite fractional error method ( 4 9) ( 4 10) Average relative error ( 4 11) where, Q e is measured equilibrium concentration of adsorbate in the solid phase, and is the calculated concentration of adsorbate in the solid ph ase. Each error function generated different set s of isotherm parameters; hence, a procedure that normalizes the parameters and combines the errors was adopted. First, one isotherm and one error function was selected and isotherm parameter values that m inimize the error function were determined using solver add in function of Microsoft E xcel. Using these isotherm parameter values, values for other error functions were determined and at the end five values were obtained. A s imilar procedure of minimizin g the error function was applied to eac h error function and a different isotherm parameter set and error function values was obtained. Then, the five values of each error function

PAGE 104

104 were normalized using the maximum of these values and the normalized values were summed up. The parameter set that provided smallest sum of normalized error was considered as the best parameter set values. The non linear isotherm plots for three different AC are presented in Figure 4 1. Among each of the three isotherms, Freu ndlich and Redlich Peterson isotherms gave a better fit to the experimental data, with the best fit in Redlich Peterson isotherm. Langmuir isotherm did not show a good fit to the experimental data, further strengthening the fact that it does not fit for h igh concentration solutions. The model equation for Langmuir isotherm also showed a sharp increase in Q e for higher C e values. Rivas et al. (2006) also observed a sharp increase in Q e for COD adsorption Langmuir isotherm at high C e values. The parameter s of the three isotherm equations for the three AC s are shown in Table 4 6 The Freundlich and Redlich Peterson isotherms predicted almost similar adsorption capacity to the organic matter for each of the three AC. The Freundlich isotherm constant (K F ) for each of the three AC s were in between organic matter and AC in each of the three AC. Freundlich isotherm predicted a rate of 0.3 g TOC adsorption/g AC for 50% TOC removal from the leachate by Norit HD 4000 AC. A slightly higher adsorption of 0.47 g TOC adsorption/g AC was observed with Calgon F 300. The Redlich Peterson Model predicted lower values than Freundlich model as 0.22, 0.18, and 0.19 g TOC adsorption/g A C with Calgon F 300, Norit HD 4000, and Darco 12x40 AC, respectively. Though, the micro porous AC has larger surface area and should have higher adsorption capacity than meso porous AC the presence of high molecular weight and

PAGE 105

105 size organic matter (humic and fulvic like) may block the pores of micro porous AC at the surface due to steric effect and reduce the adsorption capacity during stabilized leachate treatment ( Karanfil et al., 2006 ; Pignatello et al., 2006 ). The meso porous AC have pore opening in t he range of 2 to 50 nm, which are in the molecular size range of humic like organic molecule (<2.5 nm to 40 nm ) (Osterberg et al., 1992). These bigger organic molecule should not block the pores of meso porous AC particle and AC should use their effective adsorption capacity. This might be the possible reason of almost similar adsorption capacities observed in micro porous and meso porous AC ( Pignatello et al., 2006 ) Although, there is little data available on landfill leachate adsorption onto AC, Xing et al. (2008) reported 0.2 g TOC adsorption/g of coal based AC, a similar adsorption capacity as predicted by Redlich Peterson model for the set of AC s and leachate used in this study. 4.3.2 Intra particle Diffusivity The sorption of organic matter onto AC is a complex phenomenon, where properties of both adsorbate and adsorbent play an important role. The adsorption process can be controlled by following sequential steps: (1) the bulk solution transport, where the adsorbate diffuses from solution to th e boundary layer of solution surrounding the AC particles, (2) film diffusion, where adsorbate diffuses through the liquid film surrounding the AC particles, and (3) pore diffusion and adsorption, where adsorbent is transported to the pores of AC to availa ble adsorption sites. One or more than one process can be involved in the adsorption process and the slowest process controls the rate of adsorption process. Weber and Moris (1963) (equation 4 12) developed a widely accepted kinetic based model that rep resents the time dependent intra particle diffusion of components

PAGE 106

106 and showed that the sorption process is diffusion controlled if its rate is dependent upon the rate at which adsorbate and adsorbent diffuse towards one another. ( 4 12) w here, q t (mg/g) is the adsorbate loading on the solid phase at time t, k id (mg/g min 1/2 ) is the intra particle diffusion rate constant, and C (mg/g) is the constant that is proportional to the thickness of boundary layer; larger the value of C greater the boundary layer thickness (Mckay et al., 1980). The straight line plot of q t and t 1/2 represents the sorption process as diffusion controlled, however, if the data shows multiple linear plots, the sorption process is controlled by more than one processes. At first, the sharper portion of the plot represents external resistance to mass transfer followed by gradual adsorption where intra particle diffusion is the controlli ng factor. Figure 4 2 shows the relation of mass of TOC adsorbed per unit mass of different particle sizes of three different AC to the t 1/2 Each of the three AC showed two straight lines for the particle sizes greater than 0.3 mm. The first portion o f straight line represents the diffusion process controlled by external surfaces and the second portion of straight line showed the intra particle diffusion. Extrapolation of the linear portion of the plots to the y axis shows intercepts that provide the boundary layer thicknesses. An increased boundary layer thickness was observed with the decrease in AC particle sizes ( Table 4 7 ) representing slower intra particle diffusion for smaller AC particle. The deviation of straight lines from the origin repres ents the difference in diffusion rate in initial and final stages of adsorption. The diffusion rate constants (k id ) were observed to be decreasing with the decreasing AC particle sizes ( Table 4 7 ) and k id varied to approximate square root of the AC partic le sizes similar to as observed by McKay et al.

PAGE 107

107 (1980; 1987). The dependence of k id on particle diameter shows that the intra particle diffusion is the predominant rate controlling step during part of adsorption process McKay et al. (1980). The AC parti cle sizes less than 0.3 mm did not show similar intra particle diffusion pattern in micro and meso porous AC, wherein, the micro porous Calgon F 300 AC showed two straight lines but the meso porous AC Norit HD 4000 and Darco 12x40 showed only one straight line representing only micro porous diffusion as the main diffusion process in meso porous AC for these sizes of AC particles. A higher boundary layer thickness in the meso porous AC as shown in Table 4 7 also supports the observed results. The Weber Mo ris model (equation 4 12) predicted that the intra particle diffusion was the main rate limiting factor for organic matter adsorption onto each of the three AC. Additionally Boyd et al. (1947) generated a homogeneous particle diffusion model (HPDM) (equat ion 4 13) for adsorbent phase controlled diffusion (such as intra particle diffusion) of adsorbate onto spherical particles to determine the rate of diffusion process. ( 4 13) w here, F(t) is the fractional attainment of equilibrium at time t, D e the effective diffusion coefficient of adsorbate onto adsorbent (m 2 /sec), r the radius of adsorbent particle assumed to be spherical (m), and z in an integer. F(t) values can be calculated as where q t and q e are adsorbate loading on the adsorbent at time t and when equilibrium is achieved respectively.

PAGE 108

108 adsorption on spherical particles as shown in equations 4 14 and 4 15 (Ver meulen, 1953); ( 4 14) ( 4 15) where, The slope of plot ln[1 F 2 (t)] and t as shown in Figure 4 3 gives the value of effective diffusion coefficient (D e ). From the plots, it is evident that the data satisfactorily fit in the entire range of diffusion for each size AC particles. The r 2 values of each plot are presented in Table 4 8 The D e values as presented in Table 4 8 shows that the meso porous Norit HD 4000 has the highest rate of diffusion among all three AC in an order of Norit HD 4000>Darco 12x40>Calgon F 300. There is no data available to compare the results of TOC adsorption onto these AC. The type of intra particle diffusivity of TOC onto AC was determined as shown in Figure 4 4. Generally, the intra kinetic adsorption profile does not vary with the change in size of AC particles, however, if the adsorption profile varies with the size of AC particles, the diff usivity is termed as porous Calgon F 300 AC showed that as the size of AC particles was changed the kinetic adsorption profile also changed, representing the proportional diffusivity of organic matter in to AC. However, the meso porous Norit HD 4000 and Draco 12x40 AC showed constant intra particle diffusivity of organic matter onto AC. The possible reason of such behavior in meso porous AC was bigger pore diameters of AC particles that allow bigger orga nic molecules (humic and

PAGE 109

109 fulvic like) to enter into the pores and least effect of the particle sizes, whereas in the micro porous AC, the pore diameters are small, which does not allow these organic molecules to enter into the pore as easily as in meso por ous AC. 4.3.3 Column Experiment RSSCT experiments were conducted to determine the efficiency of AC columns for removal of organic matter at different EBCT. As shown in Figure 4 5, an increased TOC removal was observed with the increase in EBCT. However, EBCT greater than 30 minutes did not show any significant increase in TOC removal in each of the three AC, representing at least 30 minutes of EBCT was required for maximum TOC removal. Additionally, each of the three AC showed almost similar TOC removal efficiency, with slightly better TOC removal performance in Calgon F 300 AC. M ore than 85% of TOC was removed from leachate using any of the three AC at 60 min of EBCT. Rivas et al. (2003) obtained a 90% COD removal using micro porous AC which is comparable to the results of current st udy. The fluorescence EEM analysis of leachate and samples generated at different EBCT was conducted. The EEM shows the presence of organic matter at specific E x and E m wavelengths by peaks Shorter E x (<250 nm) and shorter E m (<350nm) are generally as sociated with the simple aromatic protein like compounds such as tyrosine (Regions I and Region II). The peaks at E x in the range of 250 to 280 nm and E m <380 nm are associated with soluble microbial byproducts type compounds and are shown in Region IV. The fulvic like compounds (Region III) show peaks at shorter E x (<250 nm) and longer E m (>350 nm). Peaks at longer E x and longer E m are associated with humic like compounds and shown in Region V. The EEM of leachate as shown in Figure 4 6 showed peaks in the Region II, Re gion III, and Region V, representing the presence of

PAGE 110

110 protein, fulvic, and humic like organic matter in the leachate. The FRI analysis was conducted using equation 4 1 and showed the maximum concentration of organic matter in the Region V ( Table 4 9 ). The fractional removal of organic matter of each region with respect to the EBCT for each of the three AC is shown in Figure 4 7. Each type of organic matter was removed and the amount of organic matter removed was found increasing with the increase in EBCT. However, among all type of organic matter, highest removal was observed for the fulvic like organic matter from each AC. Smaller organic matter of Region I showed increased adsorption for smaller EBCT, however, they desorbed at higher EBCT due to preferential adsorption of humic and fulvic like organic matter on to AC. 4.4 Summary and Conclusions Adsorption of leachate onto three different AC micro porous Calgon F 300, meso porous Norit HD 4000 and Darco 12x40 AC was studied to assess their effectiveness for organic matter removal. The adsorption capacity, the rate of organic matter diffusion, and the type of organic matter removed were studied using isotherms, diffusivity, and RSSCT experiments, respectively. Each of the three AC s howed almost similar organic matter adsorption capacity, with slightly better adsorption capacity in micro porous Calgon F 300 AC. Among the adsorption isotherm model tested, three parameter model Redlich Peterson gave the best fit to the experimental dat a, which contains the feature of Freundlich and Langmuir isotherm. The Langmuir isotherm did not give a good fit for the experimental data. The Weber Moris model equation of diffusivity showed that the overall diffusion of organic matter was controlled by the intra particle di ffusion process, which is a two step process. The diffusion process wa s initially controlled by macro pore diffusion fo llowed

PAGE 111

111 by relatively slow micro pore diffusion. The effect of AC particle sizes on the diffusion process showed that as the AC particle sizes were reduced, the overall diffusion rate constant also decreased and the boundary layer thickness across the adsorbent particles increased. Among all three AC studied, meso porous Norit HD 4000 AC showed maximum rate of diff usion of organic matter onto AC. Micro porous Calgon AC showed a proportional diffusivity of organic matter, whereas meso porous Norit HD 4000 and Darco 12x40 AC showed constant diffusivity onto AC particles. The RSSCT results showed a slightly almost sim ilar performance in terms of TOC removal with slightly better TOC removal by micro porous AC at smaller EBCT The FRI analysis of fluorescence EEM showed greatest removal for fulvic like organic compounds by each of the three AC. The protein like organic matter tend s to adsorb at smaller EBCT and then desorb s at higher EBCT. Hence, this study demonstrates that micro as well as meso porous AC provide almost similar organic matter adsorption capacity while treating stabilized leachate, however, meso porous AC provide slightly faster organic removal than micro porous AC. Even though the AC showed very encouraging results in terms of TOC removal, the process will be best suited when combined with other pretreatments. The high concentrations of organic matte r present in leachate may exhaust the AC quickly and making the treatment process economically unfavorable. Furthermore, it will be valuable to determine the effectiveness of meso porous AC for stabilized leachate treatments using field scale design param eters, because, generally all the studies been conducted for leachate treatment have used micro porous AC. The performance of RSSCTs should also be validated with the filed scale study.

PAGE 112

112 Table 4 1. Previous studies on landfill leach ate treatment using activated carbon adsorption process Initial concentrations AC type/precursor/GAC/ PAC Micro porous/Meso porous AC COD Removal (%) Maximum Adsorption capacity (g/g AC ) Additional treatment with AC Reference COD /DOC* (mg/L) BOD 5 /COD 879 940 0.004 Calgon F400 (GAC) Micro Porous 70 0.56 Biological treatment+AC Morawe et at. (1995) 205* Norit NRS EA (PAC) 0.06/0.08 AC/Ozone+AC Fettig et al. (1996) 10 750 0.6 Norit SA 4 (PAC) 38 ASP+AC Aktas and Cecen (2001) 7 000 Commercial PAC 49 Biological treatment+AC Kargi and Pamukoglu (2003) 5 000 0.17 Norit 0.8 Micro porous 90 0.2 Ozone+AC Rivas et al. (2003) 3 600 0.11 Norit 0.8 (GAC) Micro porous 75 0.9 1.4 AC Rivas et al. (2004) Chemviron AQ 40 (GAC) Micro porous 55 Picacarb 1240 (GAC) Micro porous 55 7 811 0.08 GAC (type PHO 8/35 LBD) Micro Porous 60/86 0.16 AC/ Ozone+AC Kurniawan et al. (2006) 110 TOC Coal based PAC Micro Porous 54 Coagulation+ AC Xing et al. (2008) Coal based GAC 50 Wood based GAC 33 Wood based PAC 14 2 461 0.24 Darco 20x40 (GAC) 40 AC Gotvajn et al. (2009) 785 Organosorb 10MB Micro Porous 55 0.150 0.157 AC Maranon et al. (2009) Filtracarb CC65/1240 Micro Porous 51 0.145 0.175

PAGE 113

113 Table 4 2 Physico chemical characteristics of ACL leachate during the study period Parameter Average Std dev pH (S.U.) 7.71 0.01 Conductivity (mS/cm) 13.79 1.4 TDS (mg/L) 6546 30 BOD 5 (mg/L) 57 COD (mg/L) 2300 46 BOD 5 /COD 0.02 DOC (mg/L) 523 10.7 NH 3 N (mg/L) 1060 63 Alkalinity (mg/L as CaCO 3 ) 6800 UV 254 (cm 1 ) 9.0 SUVA (m 1 /mg/L DOC) 1.7 Table 4 3 Characteristics of Calgon F 300, Norit HD 4000, and Darco 12x40 AC Parameter Calgon F 300 Norit HD 4000 Darco 12x40 Iodine number (mg/g) 900 a 500 b 625 b Surface area (m 2 /g) 816 546 588 Pore size () 18 32 42 Mesh size 8x30 a 10x30 b 12x40 b Effective particle size (mm) 0.8 1.0 a 0.6 0.8 0.65 b a Values obtained from Calgon group ; b Values obtained from Norit group Table 4 4 Design parameters of full scale and RSSCT Parameter Full scale Calgon F 300 Norit HD 4000 Darco 12x40 Grain size US mesh 8x30 35x50 35x50 35x50 EBCT (min) 5 60 1.3 12 0.9 8.4 0.6 5.2 Hydraulic loading rate (mm/min) 17 205 4.6 4.6 4.6 Time to process (hours) 28.8 4 4 4 Table 4 5 Comparative empty bed contact time (EBCT) of full scale and RSSCT columns for Calgon F 300, Norit HD 4000, and Darco 12x40 AC EBCT full scale column EBCT for RSSCT (min) (min) Calgon F 300 Norit HD 4000 Darco 12x40 6.6 1.3 0.9 0.6 10 2 1.4 0.9 15 3 2.1 1.3 30 6 4.2 2.6 60 12 8.4 5.2

PAGE 114

114 Table 4 6 Isotherm parameters obtained using non linear method for organic matter absorption onto Calgon F 300, Norit HD 4000, and Darco 12x40 AC Isotherm Constant Calgon F 300 Norit HD 4000 Darco 12x40 Freundlich K F 0.04 0.05 0.04 1/n 1.63 1.52 1.59 Langmuir Q M 16.98 5.69 7.09 b 0.001 0.01 0.01 Redlich Peterson K R 0.71 0.56 0.63 a R 578.81 917.82 647.16 b R 1.89 2.03 1.88 Table 4 7 B oundary layer thickness (C) of different particle sizes of Calgon F 300, Norit HD 4000, and Darco 12x40 AC Mean AC particle size (mm) Diffusion rate constant (k id ), (mg/g min 1/2 ) Boundary layer thickness (C), (mg/g) Calgon F 300 Norit HD 4000 Darco 12x40 Calgon F 300 Norit HD 4000 Darco 12x40 0.84 3.27 2.26 4.01 14.6 14.8 16.8 0.68 2.96 1.36 3.29 16.1 15.1 16.9 0.40 2.89 1.53 2.97 16.5 15.1 17.3 0.19 2.73 0.10 0.04 16.6 15.2 17.5 Table 4 8 Effective diffusion coefficient (De) of organic matter onto different particle sizes of Calgon F 300, Norit HD 4000, and Darco 12x40 AC AC Particle size range (mean sizes)mm Calgon F 300 Norit HD 4000 Darco 12x40 Diffusion coefficients (m 2 /s) F(t) vs. (t) R 2 Diffusion coefficient s (m 2 /s) F(t) vs. (t) R 2 Diffusion coefficients (m 2 /s) F(t) vs. (t) R 2 >0.84 (0.84) 145x10 10 0.95 369 x10 10 0.93 253 x10 10 0.99 0.50 0.85 (0.68) 109 x10 10 0.97 182 x10 10 0.93 124 x10 10 0.96 0.30 0.50 (0.40) 533 x10 11 0.97 643 x10 11 0.97 728 x10 11 0.95 0.075 0.30 (0.19) 151 x10 11 0.99 105 x10 11 0.70 809 x10 12 0.86 Table 4 9 Volume distribution of DOM in region I to region V of leachate derived from FRI analysis Region Volume (AU nm 2 /mg C/L) Region I 137.56 Region II 302.58 Region III 706.35 Region IV 996.18 Region V 3 267.10

PAGE 115

115 Figure 4 1. Adsorption of leachate onto (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. Data points and error bars are average and standard deviation of the duplicate experiments, respectively.

PAGE 116

116 Figure 4 2. Weber and Moris intra particle diffusion for removal of TOC using (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. Data points were obtained using average of duplicate experiments in the model equations.

PAGE 117

117 Figure 4 3. Homogeneous particle diffusion model for TOC removal in leachate using (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. Data points were obtained using average of duplicate experiments in the model equations.

PAGE 118

118 Figure 4 4. Adsorption kinetics of organic matter at different particle sizes (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. Data points and error bars are average and standard deviation of normalized TOC obtained from duplicate experiments, respectively.

PAGE 119

119 Figure 4 5. Effect of EBCT on the TOC removal of leachate using Calgon F 300, Norit HD 4000, and Darco 12x40 AC. Data points and error bars are average and standard deviation of percentage TOC removal obtained from duplicate experiments, respectively. Initial TOC =670 mg/L. Figure 4 6. Fluorescence EEM of ACL leachate and operationally defined E x E m boundaries for five EEM regions.

PAGE 120

120 Figure 4 7. Effect of EBCT on DOM of each EEM region while leachate treatment using (a) Calgon F 300, (b) Norit HD 4000, and (c) Darco 12x40 AC. Volumes were derived from FRI analysis.

PAGE 121

121 CHAPTER 5 SUMMARY AND CONCLUSIONS 5.1 Summary Landfills are still the most accepted method for ultimate disposal of municipal and industrial solid waste due to their economic advantages. However, landfills face a major challenge of managing the leachate generated by the percolation of rainwater through the layers of waste. Traditionally, leachate treatment has been conducted by one or the combination of more than one treatment processes such as leachate recirculation, biological, and physico chemical treatment met hods. The characteristics of leachate have been found very heterogeneous and vary with the age of landfill. The biological treatment methods are effective for the young leachate, but as the landfill grows old, major presence of refractory organic matter in the leachate limits the effectiveness of these processes. However, physico chemical treatment methods such as coagulation and flocculation, chemical oxidation, activated carbon adsorption and membrane systems have been found more effective for such old leachate. It has been observed that the u se of a single treatment methodology does not produce effluent water that will meet water quality standards for direct discharge to groundwater or surface water, however, membrane filtration technologies such as n ano filtration (NF) and reverse osmosis (RO) have been found effective in this regard. Nevertheless, the membrane systems may not provide a cost effective treatment due to excessive fouling caused be the presence of high concentrations of high molecular w eight humic and fulvic like organic matter The current research was divided into three major objectives. The first objective was to evaluate and compare the effectiveness of Fe (III) coagulation and MIEX anion

PAGE 122

122 exchange processes for reducing the foulin g of NF and RO membranes The second objective was to study the effectiveness of oxidation of organic matter using ozone as a pretreatment step to reduce the membrane fouling. The third objective of the study was to evaluate the organic matter adsorption parameters for three different AC based on their pore sizes for stabilized leachate treatment. In the first study, at first, batch coagulation and anion exchange experiments were conducted to determine their effectiveness of organics removal from stabiliz ed leachates. All three leachate showed a variable amount of maximum DOC removal using the Fe(III) coagulant. A maximum of 70% DOC removal was observed in NCL leachate, which is in the similar range as suggested in the literature. The amount of DOC remo ved was correlated to the pH and alkalinity present in the leachate. Fluorescence regional integration (FRI) method as suggested by Chen et al. (2003) was used to determine the type of organics removed by the coagulant and approximately 80% removal of hum ic and fulvic like organics were observed in all three leachate. An optimum coagulant dose of 22 mmol Fe(III)/L was selected for the membrane experiment because this dose generated optimum pH conditions for coagulation in all three leachate. A maximum o f 30% DOC was removed using MIEX; however, most of the DOC was removed at lower MIEX doses. Increase in MIEX doses did not remove additional DOC and the doses used in the study correspond to the doses typically used in drinking water treatments. There is no literature available to compare the results of leachate treatment using MIEX; hence this study is an important contribution for determining the MIEX doses for leachate treatment. MIEX removed 30 to 60% of negatively charged

PAGE 123

123 humic and fulvic like organ ics present in leachate. An optimum MIEX dose of 5 mL/L was selected for membrane experiments for membrane experiments. The coagulation and MIEX pretreatment did not show any improved performance of NF and RO membranes in terms of contaminant rejection from leachate as compared to raw leachate treatment. Additionally, an increased fouling was observed for coagulated as well as MIEX pretreated leachate as compared to raw leachate in both membrane treatments. During the membrane operation, an increase in feed pH was observed in each condition, which might have caused an increased precipitation of alkalinity over the membrane surface, leading to increased fouling. Trebouet et al., (2001) also observed that leachate without any pre treatment is the best wa y to use with NF while investigating coagulation as a pretreatment option for treating leachate with NF membranes. No literature was found to compare the results of MIEX treated leachate coupled with membranes; however, Cornelissen et al. (2010) used an a nion exchange resin (Flu i dized Ion Exchange (FIX)) to reduce fouling of NF membranes for surface water treatment but could not observe reduced fouling of NF membranes. The second objective of the study was to investigate the effectiveness of ozonation as a pretreatment option for treating stabilized landfill leachate using RO and NF membranes. A fixed ozone dose of 70 mg/L was used at a feed gas flow rate of 3.5 L/min f or 5 to 30 minutes to treat leachate from three different landfills. A maximum of 78% drop in UV 254 absorbance and 23% drop in DOC was observed. Faster ozone kinetics was observed at the start of experiments, which platued after 8 to10 minutes of ozonat ion in all three leachate. The characterization of organic matter using fluorescence EEM and FRI analysis also showed that the ozone removed most of the

PAGE 124

124 humic and fulvic like organic matter at the start of ozonation. Landfill leachate literature contains several studies on the kinetics of ozonation, however, no studies has been reported on using ozonation as a pretreatment option for leachate treatment using membrane systems. An optimum ozonation time of 10 minutes was determined from the batch experimen ts and used to run the membrane experiment. RO and NF membranes rejected greater than 99% of DOM and 91% salts from all three leachate in the experiment duration. A faster flux decline was observed for the ozonated leachate treatment than raw leachate. Between RO and NF membranes, NF membranes fouling was mainly caused by the particles blocking the membrane pores and not allowing the filtrate to pass through. Hence, in th e applied experimental conditions, ozonation of leachate before membrane treatment did not reduce the fouling potential of membranes. However, selections of ozone dose that can further reduce the concentrations of humic and fulvic like organic matter, the reby reducing the potential of complete blockage of membrane pores may reduce the fouling frequency of membranes. Hence, the pretreatment options studied in this research were found effective in removing or transforming humic and fulvic like organic matter from leachate. However, the pretreatments did not provide improved permeate flux in the applied experimental conditions as compared to raw leachate for NF as well as RO membranes. The third objective of the study was to determine the adsorption profile of stabilized landfill leachate onto three different AC micro porous Calgon F 300, meso porous Norit HD 4000 and Darco 12x40 AC which were selected based on their pore

PAGE 125

125 sizes AC has been generally studied in combination with other treatment methods for landfill leachate treatment but a detailed study that determine the adsorption profile of organic matter removal using AC from the stabilized leachate, which helps in designing a AC treatment system has not been reported. Each of the three AC showed alm ost similar organic matter adsorption capacity irrespective of their pore sizes Among the adsorption isotherm model tested, three parameter model Redlich Peterson gave the best fit to the experimental data. The Langmuir isotherm did not give a good fit for the experimental data. The Weber Moris model equation of diffusivity showed that the overall diffusion of organic matter onto AC particles was controlled by the intra particle diffusion process. The effect of AC particle sizes on the diffusion proces s showed that as the AC particle sizes were reduced, the overall diffusion rate constant also decreased and the boundary layer thickness across the adsorbent particles increased. Among all three AC, meso porous Norit HD 4000 AC showed maximum rate of diff usion of organic matter onto AC. The RSSCT results showed that 30 minutes EBCT provides the maximum TOD removal and further increase in EBCT does not improve the TOC removal Among all three AC, micro porous AC showed slightly better TOC removal than oth er two meso porous AC. The FRI analysis of fluorescence EEM showed greatest removal for fulvic like organic compounds by each of the three AC. The protein like organic matter tends to adsorb at smaller EBCT and desorb at higher EBCT. The breakthrough st udy of meso porous Norit HD 400 AC showed a rapid exhaustion of AC, making AC treatment as an expensive option for leachate treatment. This study provides an important contribution in selecting the type of AC for landfill leachate treatment.

PAGE 126

126 5.2 Conclus ions Following specific conclusions were drawn from the research: A maximum of 70% removal of DOC was observed using Fe (III) salt as a coagulant at unadjusted leachate pH. Coagulation process was most effective at the pH range of 4.2 to 5.2. Highest DO C removal was observed in the same pH range for all three leachate. Approximately 80% of humic and fulvic like organic matter was removed by coagulation. A maximum of 34% DOC removal was observed for leachate treatment using anion exchange resin MIEX for t he applied doses. Most of the DOC was removed at lower MIEX doses. Additionally, most of the ion exchange process occurred within initial 20 minutes of experiment. Most of the organic matter removed by the MIEX was humic and fulvic like compounds. A maximum of 60% humic like organic compounds were removed by MIEX. In the applied experimental conditions, an increased fouling was observed for coagulated as well as MIEX pretreated leachate as compared to raw leachate in both membrane treatments. MIEX tre ated leachate showed faster flux decline than Fe (III) salt treated leachate. A maximum of 78% reduction in UV 254 absorbance and 23% reduction in DOC was observed with the use of a fixed ozone dose of 70 mg/L at a feed gas flow rate of 3.5 L/min. Faster ozone kinetics was observed at the start of experiments, which platued after 8 to10 minutes of ozonation, representing that initially ozone reacts quickly with the available unsaturated bond organic compounds but it does not react as quickly with the bypr oducts of initial reactions. Ozon ation removed more than 80% of the humic and fulvic like organic matter in initial 10 minutes of ozonation. RO and NF membranes rejected greater than 99% of DOM and 91% salts from all three ozone treated leachate in the exp eriment duration. A faster flux decline with RO and NF membranes for the ozone treated leachate than raw leachate showed increased fouling of membranes.

PAGE 127

127 Between RO and NF membranes, NF membranes showed a faster flux decline than RO membranes while treating ozone pretreated leachate with membranes. blocking the membrane pores and not allowing the filtrate to pass through. The standard and intermediate blocking mechanism also play ed a significant role in membrane fouling. Ozonation at high doses might increase the particle size by forming metal complexes, which might precipitate on the membrane surface and decreases the flux. Ozonation at lower doses might help in reducing the f ouling frequency of membranes. Among the selected AC, all three AC showed almost similar adsorption capacity. Three parameter Redlich Peterson isotherm model gave the best fit for the equilibrium experimental data. Weber Moris model predicted that the i ntra particle diffusion is the main governing phenomenon of adsorption in all three AC. The diffusion rate constants were slightly dependent on the size of AC particles. As the size of AC particles reduces, the overall diffusion rate constant also decrea ses. Micro porous AC showed a proportional diffusivity, however, meso porous AC showed constant diffusivity of organic matter onto AC particles. The RSSCT results showed a slightly better performance of meso porous AC than micro porous AC in terms of TOC removal at high EBCT. All three AC showed a greater adsorption affinity towards fulvic like organic compounds. 5.3 Future work The results of the first study show that coagulant Fe(III) and anion exchange resin MIEX can effectively remove humic and fulvi c like organic matter from stabilized leachate. However, the selected doses of Fe (III) and MIEX for pre treating leachate did not help improve the permeate flux of NF as well as RO membranes as compared to raw leachate. Further analysis should be conduc ted to study the cause of increased flux decline. Additional membrane fouling experiments should be conducted at fixed pH

PAGE 128

128 conditions to determine the effect of pH on permeate flux. Additionally, t he effect of pretreated leachate at variable coagulant and MIEX dose can be studied. A different coagulant (e.g. alum) can also be studied as a pretreatment option to avoid potential iron fouling. Ozone helped reduce the concentration of humic and fulvic like organic matter either by removing or transforming to other forms during stabilized leachate treatment; however, the selected ozone dose as a pretreatment option did not improve the permeate flux of NF as well as RO membranes as compared to that of raw leachate. Further membrane fouling experiments can be co nducted for the pretreated leachate using variable ozone doses. Possibly, a lower ozone dose can be used as compared to the dose used in the current study to pre treat leachate. Lower doses may cause lesser coagulation due to the divalent ions present in leachate. Anti scalant can also be used as an additional pretreatment to reduce the possible coagulation caused by divalent ions. The laboratory scale adsorption study provided almost similar adsorption capacity of micro and meso porous AC with higher organic matter adsorption rate in meso porous AC. In the field scale, all the studies have been conducted using micro porous AC It will be valuable to determine the effectiveness of meso porous AC for stabilized leachate treatments using field scale de sign parameters. The performance of RSSCTs should also be validated with the filed scale study.

PAGE 129

129 APPENDIX A EFFECT OF COAGULANT DOSE ON FILTRATE p H AND DOC When the coagulant is added into a solution, the metal ion reacts with the hydroxide ion, a product of hydrolysis reaction and releases H + ion into the solution. The released H + ion reduces the pH of solution. As the coagulant dose increases, the concentration of H + also increases that further reduced supern atant pH Similar trend s of supernatant pH were observed in all three leachates with the increase in coagulant dose as shown in Figure A 1 (a) The NRL leachate showed faster decrease in pH as compared to ACL and NCL leachate with the increase in coagula nt doses due to the presence of lower alkalinity in NRL leachate. The alkalinity of NRL leachate was 3000 mg/L as CaCO 3 however ACL and NCL leachate contained alkalinity of 6600 mg/L as CaCO 3 and 5400 mg/L as CaCO 3 respectively. At very high coagulant doses the hydrolysis product Fe(OH) 4 starts accepting available hydrogen ions and reducing free H + in the solution and thereby increasing the pH of supernatant solution (Letterman et At high coagulant doses, NCL and NRL leachate showed increase in pH however, in ACL leachate, pH did not reach low enough to form Fe(OH) 4 that can accept H + ion from solution and increase the pH. With the addition of coagulant, as the solubility product of metal hydroxides exceeds its li mit, the particulate hydroxides start to precipitate. These hydroxides precipitates are stabilized by the coating of negatively charged natural organic matter (NOM). An increased reduction in DOC was observed in all three leachate s with the increase in c oagulant dose as shown in Figure A 1 (b). A t lower coagulant doses in NRL leachate, faster drop in pH also caused higher concentrations of positively charged metal hydroxides Higher concentrations of metal hydroxides lead to increased binding

PAGE 130

130 or sorption of NOM to precipitating hydroxides and higher DOC reduction was observed as compared to ACL and NCL leachate. When the coagulant dose was further increased the surface area of these metal hydroxides also increased and higher DOC removal was o bserved. At very high coagulant doses, the pH of the solution passes through the point of minimum solubility of metal hydroxides and the charges on the surface of precipitates becomes positive and this change tend to stabilize the particles and may releas e NOM into the solution as observed in NRL and NCL leachate (Letterman et al., 1999). The ACL leachate did not release NOM into the solution because pH of solution could not reach low enough at the applied coagulant doses such that the charges of precipit ate become increasingly positive and release the adsorbed NOM to the solution. Figure A 1 Effect of coagulant dose on pH and DOC of ACL NCL, and NRL leachate. The data points and error bars represent average and standard deviation of duplicate experiments respectively

PAGE 131

131 APPENDIX B REACTION MECHANISM OF OZONE AND HYDROXYL RADICALS TO ORGANIC MATTER Ozone has high oxidation potential (E 0 =2.07V) and high reactivity and selectivity toward organic pollutants such as aromatic compounds. Ozone ruptures the C=C bonds or aromatic ring and produces ketones and easters These intermediate products can be further transformed into organic acids th rough the oxidation process as shown in equation B 1. ( B 1) A t high pH conditions (pH> 8), ozone produces hydroxyl radicals (equation B 2) that have even higher oxidation potential (E 0 =2.8V) than ozone molecule, and accelerate the removal of recalcitrant organic matter from complex wastewater matrix. Hydroxyl radical mainly reacts with the organics via hydrohen abstraction or OH radical addition (equation B 3) Hydroxyl radical can r eact with aromatic C=C structure and produces phenolic and ethers via addition reactions. It can also produce alcohols by reacting with C C structures. These intermediates can be mineralized or transformed into small organic acids by ozone molecules and OH radicals during ozonation A typical pathway of OH radical reaction can be given as equation B 4 to B 9. O 3 +H 2 2 +2 OH ( B 2) RH+ OH R +H 2 O ( B 3 ) R +O 2 ROO ( B 4 ) ROO ( B 5 ) ROO + RO +C C ( B 6 ) C=C O 3 R 3 R 4 R 2 R 1 C=O R 2 R 1 O=C R 3 R 4 + + C=O OH R 1 C=C O

PAGE 132

132 RO ( B 7 ) ROOH Alcohols Ketones Organic acids ( B 8 ) ROH Ketones Organic acids ( B 9 ) Equations ( B 3 to B 9 ) are adopted from Li et al. (2008).

PAGE 133

133 LIST OF REFERENCES Agenson, K.O., Urase, T., 2007. Change in membrane performance due to organic fouling in nanofiltration (NF)/reverse osmosis (RO) applications. Separation and Purification Technology 55, 147 156. Ahn, W.Y., Kang, M.S., Yim, S.K., Choi, K.H., 2002. Advanced landfill leachate treatment using an integrated membrane process. Desalination 149, 109 114. Aktas O., Cecen, F., 2001. Addition of activated carbon to batch activated sludge reactorsin the treatment of landfill leachate and domestic wastewater J ournal of Chem ical Technol ogy and Bio technol ogy 76 793 802. Allpike, B.P., Heitz, A., Joll, C.A., Kagi, R.I., Abbt Braun, G., Frimmel, F.H., Brinkmann, T., Her, N., Amy, G., 2005. Size exclusion chromatography to characterize DOC removal in 356 drinking water treatment. Environmental S cience and Technology 39, 2334 2342. Alvarez Puebla, R.A., Valenzuela Calahorro, C., Garrido, J.J., 2006. Theoretical study on fulvic acid structure, conformation and aggregation A molecular modelling approach. Science of the Total Environment358, 243 25 4. Alvarez Vazquez, H., Jefferson, B., Judd, S.J., 2004. Membrane bioreactors vs. conventional biological treatment of landfill leachate: a brief review. Journal of Chemical Technology and Biotechnology 79, 1043 1049. Amokrane, A., Comel, C., Veron, J., 1997. Landfill leachates pretreatment by coagulation flocculation. Water Research 31, 2775 2782. Amy, G., 2008. Fundamental understanding of organic matter fouling of membranes. Desalination 231, 44 51. Anderson, J., Johnson, D., Chistman R.F., 1985. The reaction of ozone with isolated aquatic fulvic acid Organic Geochemistry 8 (1), 65 69. Babcock, D.B. Singer, P.C., 1979. Chlorination and coagulation of humic and fulvic acids. Journal of American Water Works Association 71, 149 152 Baig, S., Liechti, P.A., 2001. Ozone treatment for biorefractory COD removal. Water Science and Technology 43 (2) 197 204. Baker, A. Curry, M., 2004. Fluorescence of leachates from three contrasting landfills. Water Research 38, 2605 2613. Baumgart en, G., Seyfried, C.F., 1996. Experience and new developments in biological pretreatment and physical post treatment of landfill leachate. Water Science and Technology 34 (7 8), 445 453.

PAGE 134

134 Bellona, C., Drewes, J.E., Xu, P., Amy, G., 2004. Factors affecting the rejection of organic solutes during NF/RO treatment A literature review. Water Research 38, 2795 2809. Benson, C.H., Barlaz, M.A., Lane, D.T., Rawe, J.M., 2007. Practice review of five bioreactos/recirculation landfills. Waste Management 27, 13 29. Bila, D.M., Montalvao, A.F., Silva, A.C., Dezotti, M. 2005. Ozonation of a landfill leachate: evaluation of toxicity removal and biodegradability improvement. Journal of Hazardous Material B 117 235 242. Bohdziewicz, J., Bodzek, M., Gorska, J., 2001. Application of pressure driven membrane techniques to biological treatment of landfill leachate. Process Biochemistry 36, 641 646. Borghi, A.D., Binaghi, L., Converti, A., Borghi, M.D. 2003. Combined treatment of leachate from sanitary landfill and mu nicipal wastewater by activated s ludge. Chemical and Biochemical Engineering Quarterly 17 (4) 277 283. Boyd, G.E., Myers, L.S., Adamson, A.W., 1947. The exchange adsorption of ions from aqueous solutions by organic zeolites. III. Performance of deep ads orbent beds under non aquarium conditions. Journal of American Chemical Society 69, 2849 2859. Boyer, T., Graf, K., Comstock, S.E., Townsend, T.G., Magnetic ion exchange treatment of stabilized landfill leachate. Submitted to Chemosphere, 23 November 20 10. Boyer, T., Singer, P.C., 2005. Bench scale testing of a magnetic ion exchange resin for removal of disinfection by product precursors. Water Research 39, 1265 1276. Boyer, T., Singer, P.C., 2006. A pilot scale evaluation of magnetic ion exchange tre atment for removal of natural organic material and inorganic ions. Water Research 40, 2865 2876. Boyer, T. Singer, P.C., 200 8 Stoichiometry of removal of natural organic matter by ion exchange. Environmental Science and Technology 42, 608 613. Brown, S .L., Leonard, K.M., Messimer, S.L., 2008. Evaluation of ozone pretreatment on flux parameters of reverse osmosis for surface water treatment. Ozone: Science and Engineering 30, 152 164. Cecen, F., Aktas, O., 2004. Aerobic co treatment of landfill leacha te with domestic wastewater. Environmental Engineering Science 21 (3), 303 312. Chandrakanth, M.S., Amy, G.L., 1996. Effects of ozone on the colloidal stability and aggregation of particles coated with natural organic matter. Environmental Science and Technology 30, 431 443.

PAGE 135

135 Chen, J., LeBoeuf, E.J., Dai, S., Gu, B., 2003 Fluorescence spectroscopic studies of natural organic matter fractions. Chemosphere 50, 639 647. Chen W., Westerhoff, P., Leenheer, J.A., Booksh, K., 2003. Fluorescence excitation emission matrix regional integration to quantify spectra for dissolved organic matter. Environmental Science Technology 37, 5701 5710. Chianese, A., Ranauro, R., Verdone, N., 1999. Treatment of landfill leachate by reverse osmosis. Water Research 33, 64 7 652. Childress, A.E., Elimelech, M., 1996. Effect of solution chemistry on the surface charge of polymeric reverse osmosis and nanofiltration membranes. Journal of Membrane Science 119, 253 268. Childress, A.E., Elimelech, M., 1996. Relating nanofiltr ation membrane performance to membrane charge (electrokinetic) characteristics. Environmental Science and Technology 34, 3710 3716. Christensen, T.H., Kjeldsen, P., Bjerg, P.L., Jensen, D.L., Christensen J.B., Baun, A., Albrechtsen, H.J., Heron, G., 2001. Biogeochemistry of landfill leachate plumes. Applied Geochemistry 16, 659 718. Comstock, S.E.H., Boyer, T.H., Graf, K.C., Townsend, T.G., 2010. Effect of landfill characteristics on leachate organic matter properties and coagulation treatability. Chem osphere 81, 976 983. Cornelissen, E.R., Chasseriaud, D., Siegers, W.G., Beerendonk, E.F., Van der Kooij, D., 2010. Effect of anionic flu i dized ion exchange (FIX) pre treatment on nanofiltration (NF) membrane fouling. Water Research 44, 3283 3293. Cortez, S., Teixeira, P., Oliveira R., Mota, M., 2010. Ozonation as polishing treatment of mature landfill leachate. Journal of Hazardous Materials 182, 730 734. Cory, R.M., Mcknight, D.M., 2005. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinines in dissolved organic matter. Environmental Science and Technology 39, 8142 8149. Cossu, R., Raga, R., Rossetti, D., 2003. "The PAF model: an integrated approach for landfill sustainability." Waste Management 23, 37 44. Critt enden, J.C., Berrigan, J.K., Hand, D.W., 1986. Design of rapid small scale adsorption tests for a constant surface diffusivity. Journal of Water Pollution Control Federation 58 (4), 312 319. Crittenden, J.C., Reddy, P.S., Arora, H., Trynoski, J., Hand, D .W., Perran, D.L., Summers, R.S., 1991. Predicting GAC performance with rapid small scale column tests. Journal of American Water Works Association 83 (1), 77 87.

PAGE 136

136 Edzwald, J.K., Tobiason, J.E., 1999. Enhanced coagulation: US requirements and a broader v iew. Water Science Technology 40 (9) 63 70. Eggen, T., Moeder, M., Arukwe, A., 2010. Municipal landfill leachate: A significant source for new and emerging pollutants. Science of the Total Environment 408 (21), 5147 5157. Fabris R., Lee, E.K., Chow, C.W.K., Chen, V., Drikas, M., 2007. Pre treatments to reduce fouling of low pressure micro filtration (MF) membranes. Journal of Membrane Science 289, 231 240. Fearing, D.A., Banks, J., Guyetand, S., Eroles, C.M., Jefferson, B., W ilson, D., Hillis, P., Campbell, A.T., Parsons, S.A., 2004. Combination of ferric and MIEX (R) for the treatment of a humic rich water. Water Research 38, 2551 2558. Fettig, J., Stapel, S., Steinert, C., Gieger, M., 1996. Treatment of landfill leachate by preozonation and adsorption in activated carbon columns. Water Science and Technology 34 (9), 33 40. Foo K.Y., Hameed, B.H., 2009. An overview of landfill leachate treatment via activated carbon adsorption process. Jou rnal of Hazardous Materials 171, 54 60. Freundlich, H.M.F., 1906. Over the adsorption in solution. Journal of Physical Chemistry 57, 385 470. Gotvajn, A.Z., Tisler, T. Koncan, T. 2009. Comparison of different treatment strategies for industrial landfill leachate Journal of Hazardous Materials 162 1446 1456. Halim, A.A., Aziz, H.A., JOhari, M.A.M., Ariffin, K.S., 2010. Comparison study of ammonia and COD adsorption on zeolite, activated carbon and composite materials in landfill leachate treatment. Desalination 262, 31 35. Her, N., Amy, G., Plottu Pecheux, A., Yoon, Y., 2007. Identification of nanofiltration membrane foulants Water Research 41, 3936 3947. Hermia, J., 1982. Constant pressure blocking filtration laws Application to power law non newtonian fluids. Transaction of Institution of Chemical Engineers 60, 183 187. Herzberg, M., Elimelch, M., 2007. Biofouling of reverse osmosis membranes: Role of biofilm enhanced osmotic pressure. Journal of Membrane Science 295, 11 20. Ho, Y.S., Porter, J.F., Mckay, G., 2002. Equilibrium isotherm studies for the sorption of divalent metal ions onto peat: Copper, nickel, and lead single component system. Water, Air, and Soil Pollution 141, 1 33. Hoang, T., Stevens, G., Kentish, S., 2010. The effect of feed p H on the performance of a reverse osmosis membrane. Desalination 261, 99 103.

PAGE 137

137 Hobbs, C., Hong, S., Taylor, J., 2000. Effect of membrane properties on fouling in RO/NF membrane filtration of surficial groundwater. American Chemical Society, Washington DC August 20 24, 2000, 40 (2), 287 289. Hong, S., Elimelech, M., 1997. Chemical and physical aspects of natural organic matter (NOM) fouling of nanofiltration membranes. Journal of Membrane Science 132, 159 181. Hudson, N., Baker, A., Reynolds, D., 2007. Fluorescence analysis of dissolved organic matter in natural, waste and polluted waters A review. River Research and Applications 23, 631 649. Huang, H., Schwab, K., Jacangelo, J.G., 2009. Pretreatment for low pressure membranes in water treatment: A r eview. Environmental Science and Technology 43 (9), 3011 3019. Humbert, H., Gallard, H., Suty, H., Croue, J.P., 2005. Performance of selected anion exchange410 resins for the treatment of a high DOC content surface water. Water Research 39, 1699 1708. H umbert, H., Gallard, H., Jacquemet, V. Croue, J.P., 200 7 Combination of coagulation and ion exchange for the reduction of UF fouling properties of a high DOC content surface water Water Research 41, 3803 3811. 2008. Direct flow microfiltration of aquasols II. On the role of colloidal natural organic matter. Journal of Membrane Science 325 (2), 903 914. Hyung, H., Lee, S., Yoon, J., Lee, C.H., 2000. Effect of preozonation on flux and water quality in ozonation ultra filtration hybrid system for water treatment. Ozone: Science and Engineering 22, 637 652. Jucker C., Clark, M.M., 1994. Adsorption of aquatic h u mic substances on hydrop hobic ultrafiltration membranes. J ournal of Membrane Sci ence 97 37 52. Karanfil T., Dastgheib, S.A., Mauldin, D., 2006. Exploring molecular sieve capabilities of activated carbon fibers to reduce the impact of NOM preloading on trichloroethylene adsorption. Environmental Science and Technology 40, 1321 1327. Kargi, F., Pamukoglu, M.Y., 2003. Powdered activated carbon added biological treatment of pre treated landfill leachate in a fed batch reactor. Biotechnology Letters 25, 695 699. Karnik, B.S., Davies, S.H.R., Chen, K.C., Jaglowski, D.R., Baumann, M.J., Masten, S.J., 2005. Ef fects of ozonation on the permeate flux of nanocrystalline ceramic membranes. Water Research 39, 728 734

PAGE 138

138 Keskinler, B., Yildiz, E. Erhan, E., Dogru, M., Bayhan, Y.K., Akay, G., 2004. Crossflow Microfiltration of low Concentration Nonliving Yeast Suspensions J ournal of Membrane Sci ence 233 59 69. Kim, J., Davies, S.H.R., Baumann, M.J., Tarabara, V.V., Masten, S.J., 2008. Effect of ozone dosage and hydrodynamic conditions on the permeate flu x in a hybrid ozonation ceramic ultrafiltration system treating natural water. Journal of Membrane Science 311, 165 172. Kjeldsen, P., Barlaz, M.A., Rooker, A.P., Baun, A., Ledin, A., Christensen, T.H., 2002. Present and long term composition of MSW land fill leachate: A review. Critical Reviews in Environmental Science and Technology 32 (4), 297 336. Koltuniewicz, A.B., Field, R.W., Arnot, T.C., 1995. Cross flow and dead end microfiltration of oily water emulsion part I: Experimental study and analysis of flux decline. Journal of Membrane Science 102, 193 207. Kurniawan, T.A., Lo, W. H., Chan, G.Y.S., 2006. Radicals catalyzed oxidation reactions for degradation of recalcitrant compounds from landfill leachate. Chemical Engineering Journal 125, 35 57. Kurniawan T.A., Lo, W. H., Chan, G.Y.S., 2006. Degradation of recalcitrant compounds from stabilized landfill leachate using a combination of ozone GAC adsorption treatment. Journal of Hazardous Materials B137, 443 455. Kwon, B., Cho, J., Park, N., Pellegrino, J., 2006. Organic nanocolloid fouling in UF membranes. Journal of Membrane Science 279, 209 219. Langmuir, I., 1916. The adsorption of gases on plane surface of glass, mica, and platinum. Journal of American Che mical Society 40, 1361 1368. Lee S., Lee C. H., 2007. Effect of membrane properties and pretreatment on flux and NOM rejection in surface water nanofiltration. Seperation and Purification Technology 56 (1), 1 8. Lee, S., Lee, K., Wan, W.M., Choi, Y., 200 5. Comparison of membrane permeability and a fouling mechanism by pre ozonation followed by membrane filtration and residual ozone in membrane cells. Desalination 178, 287 294. culation: Chapter 6: Water quality and treatment: A handbook of community water. The American water works association, 5 th edition, Mc Graw Hill Book Company, 6.1 6.6. Li, F., Wichmann, K., Heine, W., 2008. Treatment of methanogenic landfill leachate wit h thin open channel reverse osmosis membrane modules. Waste Management 29 (2), 960 964

PAGE 139

139 Li, J., JiuHui Q.U., HuiJuan L., RuiPing L., Xu Z., Y iNing H., 2008. Species transformation and structure variation of fulvic acid during ozonation Science in China Series B: Chemistry 51 (4), 373 378. Li, Q., Elimelech, M. 2004. Organic fouling and chemical cleaning of nanofiltration membran es: measurements and mechanisms. Environ mental Sci ence and Technol ogy 38 (17) 4683 4693. Li, Z., Zhou, S., Qiu, J., 2007. Combined treatment of landfill leachate by biological and membrane filtration technology. Environmental Engineering Science 24 (9), 1245 1256. Li, W., Hua, T., Zhou, Q., Zhang, S., Li, F., 2010. Treatment of stabilized landfill leachate by the com bined process of coagulation/ flocculation and powder activated carbon adsorption Desalination 264, 56 62. Linde, K., Jonsson, A.S., 1995. Nanofiltration of salt solution and landfill leachate. Desalination 103, 223 232. Liu Y., Li, X., Wang, B., Liu, S ., 2008. Performance of landfill leachate treatment system with disc tube reverse osmosis units. Frontiers of Environmental Science and Engineering in China 2 (1), 24 31. Lohwacharin, J., Takizawa, S., 2009. Effects of nanoparticles on the ultrafiltrati on of surface water Journal of Membrane Science 326, 354 362. Maranon, E., Castrillon, L., Fernandez Nava, Y., Fernandez Mendez, A., 2009. Tertiary treatment of landfill leachate by adsorption. Waste Management Research 27,527 533. Mckay, G, Otterburn M.S., Sweeney, A.G., 1980. The removal of color form effluent using various adsorbents III silica: rate processes. Water Research 14, 15 20. Mckay, G, Otterburn, M.S., Aga, J.A., 1987. Intraparticle diffusion process occurring during adsorption of dye stuffs. Water, Air and Soil Pollution 26, 381 390. McKnight, D.M., Boyer, E.W., Westerhoff, P.K., Doran, P.T., Kulbe, T., Anderson, D.T., 2001. Spectrofluorometric characterization of dissolved organic matter for indication of precursor organic material and aromaticity. Limnology and Oceanography 46 (1), 38 48. Meier, J., Melin, T., Eilers, L.H., 2002. Nanofiltration and adsorption on powdered adsorbent as process combination for the treatment of severely contaminated wastewater. Desalination 146, 361 366. Mitsoyannis E., Saravacos G.D., 1977. Precipitation of calcium carbonate on reverse osmosis membranes Desalination 21 (3), 235 240.

PAGE 140

140 Monje Ramirez, I., Orta de Velasquez, M., 2004. Removal and transformation of recalcitrant organic matter from stabilized landfill leachates by coagulation ozonation coupling processes. Water Research 38, 2359 2367. Morawe, B., Ramteke, D.S., Vogelpohl, A. 1995. Activated ca rbon column performance studies of biologically treated landfill leachate Chemical Engineering and Processing 34 (3) 299 303. Nilson J.A. DiGiano, E.A., 1996. Influence of NOM compos ition on nanofiltration, J ournal of American Water Works Association 88 (5) 53 66. Ntampou, X., Zouboulis, A.I., Samaras, P., 2006. Appropriate combination of physico chemical methods (coagulation/flocculation and ozonation) for the efficient treatment o f landfill leachates. Chemosphere 62, 722 730. coagulation. Water Science and Technology 40 (9), 47 54. ropean Biophysics Journal 21, 163 167. Palma, L.D., Farrantelli, P., Merli, C., Petrucci, E., 2002. Treatment of industrial landfill leachate by means of evaporation and reverse osmosis. Waste Management 22, 951 955. Pearce, G., 2007. Introduction to membranes: Filtration for water and wastewater treatment. Filtration and Separation 44 (2), 24 27. Peters, T.A., 1998. Purification of landfill leachate with reverse osmosis and nanofiltration. Desalination 119, 289 293. Pi, K.W., Li, Z., Wan, D.J., Gao L.X., 2009. Pretreatment of municipal landfill leachate by a combined process Process Safety and Environmental protection 87, 191 196. Pignatello, J.J., Kwon, S., Lu, Y., 2006. Effect of natural organic substances on the surface and a dsorptive proper ties of environmental black carbon (char): attenuation of surface activity by humic and fulvic acids. Environmental Science and Technology 40 7757 7763. Qin, J.J., Oo, M.H., Lee, H., Coniglio, B., 2004. Effect of feed pH on permeate pH and ion rejection under acidic conditions in NF process. Journal of Membrane Science 232, 153 159. Redlich, O., Peterson, D.L., 1959. A useful adsorption isotherm. Journal of Physical Chemistry 63, 1024.

PAGE 141

141 Reinhart, D.R., Grosh, C.J., 1998. Analysis of Florida MSW landfi ll leachate quality. Gainesville, FL. Florida center for solid and hazardous waste management 31 53. Reis, R., Zydney, A., 2007. Bioprocess membrane technology. Journal of Membrane Science 297, 16 50. Renou, S., Givaudan, J.G., Poulain, S., Dirassouyan F., Moulin, P., 2008. Landfill leachate treatment: Review and opportunity. Journal of Hazardous Materials 150, 468 493. Ritchie J.D., Perdue, E.M., 2003. Proton binding study of standard and reference fulvic acids, humic acids, and natural organic mat ter Geochimica et Cosmochimica Acta 67, 85 96. Rivas, F.J., Beltran, F., Gimeno, O., Acedo, B., Carvalho, F., 2003. Stabilized leachates: ozone activated carbon treatment and kinetics. Water Research 37, 4823 4834. Rivas F. J., Beltran, F., Carvalho, F., Acedo, B., Gimeno, O., 2004. Stabilized leachate: sequential coagulation flocculation+ chemical oxidation process. Journal of Hazardous Materials B116, 95 102. Rivas, F. J., Beltran, F., Gimeno, O., Frades, J., Carvalho, F., 2006. Adsorption of land fill leachates onto activated carbon equilibrium and kinetics. Journal of Hazardous Materials B 131, 170:178. Rodriguez, J., Castrillon, L., Maranon, E., Sastre, H., Fernandez, E., 2004. Removal of non biodegradable organic matter from landfill leachates by adsorption. Water Reseach 38, 3297 3203 Stoichiometry of coagulation revisited. Environ mental Science and Technology 42, 2582 2589 Shon, H.K., Vigneswaran S., Kim, I.S., Cho, J., Ngo, H.H., 2004. Effect of pretreatment on the fouling of membranes: Applicationin biologically treated sewage effluent. Journal of Membrane Science 234, 111 120. Silva, A.C., Dezotti, M., Sant' Anna Jr., G.L., 2004. Treatment and detoxification o f a sanitary landfill leachate. Chemosphere 55, 207 214. Singer, P.C., Bilyk, K., 2002. Enhanced coagulation using a magnetic ion exchange resin. Water Research 36, 4009 4022. Statom, R.A., Thyne, G.D., Mccray, J.E., 2004. Temporal changes in leachate chemistry of a municipal solid waste landfill cell in Florida, USA. Environmental Geology 45, 982 991.

PAGE 142

142 Tabet, K., Moulin, P., Vilomet, J.D., Amberto, A., Charbit, F., 2002. Purification of landfill leachate with membrane processes: Pr eliminary studies for an industrial plant. Separation and Technology 37 (5), 1041 1063. Tang, C.Y., Kwon, Y.N., Leckie, J.O., 2007. Fouling of reverse osmosis and nano filtration membranes by humic acid Effects of solution composition and hydrodynamic c onditions. Journal of Membrane Science 290, 86 94. Tatsi, A.A., Zouboulis, A.I., Matis, K.A., Samaras, P., 2003. Coagulation flocculation pre treatment of sanitary landfill leachates. Chemosphere 53, 737 744. Thorneby, L., Hogland, W., Stenis, J., Mathi asson, L., Somogyi, P., 2003. Design of a reverse osmosis plant for leachate treatment aiming for safe disposal. Waste Management Research 21, 424 435. Timur, H. Ozturk, I. 1997. Anaerobic Treatment of Leachate Using Sequencing Batch Reactor and Hybri d Bed Filter. Water Science and Technology 36 (6 7) 501 508. Tizaoui, C., Bouselmi, L., Mansouri, L., Ghrabi, A., 2007. Landfill leachate treatment with ozone and ozone/hydrogen peroxide systems. Journal of Hazardous Material 140, 316 324. Trebouet, D. Schlumpf, J.P., Jaouen, P., Quemeneur, F., 2001. Stabilized landfill leachate treatment by combined physicochemical nanofiltration processes. Water Research 35 (12), 2935 2942. Urase, T., Salequzzaman, M., Kobayashi S., Matsuo, T., Yamamoto, K., Suzuk i, N., 1997. Effect of high concentration of organic and inorganic matters in landfill leachate on the treatment of heavy metals in very low concentration level. Water Science and Technology 36 (12), 349 356. Ushikoshi, K., Kobayashi, T., Uematsu, K., Toji, A., Kojima, D., Matsumoto, K., 2002. Leachate treatment by the reverse osmosis system. Desalination 150, 121 129. Van der Bruggen B., Manttari, M., Nystrom,, M, 2008. Drawbacks of applying nanofiltration and how to avoid them: a review, Separation and Purification Technology 63, 251 263. Van Wagner, E.M.V., Sagle, A.C., Sharma, M.M., Freeman, B.D., 2009. Effect of crossflow testing conditions, including feed pH and continuous feed filtration, on commercial reverse osmosis membrane performance. Jo urnal of Membrane Science 345, 97 109. Vermeulen, T., 1953. Theory of irreversible and constant pattern solid diffusion. Industrial Engineering Chemistry 45, 1664 1670.

PAGE 143

143 Wang, F., Gamal El Din, M., Smith, D.W., 2003. Oxidation of aged raw landfill leacha te with O 3 only and O 3 /H 2 O 2 : Treatment efficiency and molecular size distribution analysis. Ozone Science and Engineering 26, 287 298. Wang, F., Smith, D.W., Gamal El Din, M., 2003. Application of advanced oxidation methods for landfill leachate treatmen t A review. Journal of Environmental Engineering Science 2, 413 427. Wang, S., Liu, C., Li, Q., 2011. Fouling of microfiltration membranes by organic polymer coagulants and flocculants: Controlling factors and mechanisms Water Research 45, 357 365. Wa ng, X., Chen, S., Gu, X., Wang, K., 2009. Pilot study on the advanced treatment of landfill leachate using a combined coagulation, fenton oxidation and biological aerated filter process Waste Management 29, 1354 1358. Wang, Z., Zhang, Z., Lin, Y., Dend N., Tao, T., Zhuo, K., 2002. Landfill leachate treatment by a coagulation photooxidation process. Journal of Hazardous Materials B95, 153 159. Weber, W.J., Moris, J.C., 1963. Kinetics of adsorption onto carbon from solutions. Journal of Sanitary Engi neering Division ASCE 89, 31 60. Westerhoff P., Aiken, G., Amy,G., Debroux,J., 1999. Relationship between the structure of organic matter and its reactivity towards molecular ozone and hydroxyl radicals. Water Research 33 (10). 2265 2276. Wiszniowski, J ., Robert, D., Suurmacz Gorska, J., Miksch, K., Weber, J.V., 2006. Landfill leachate treatment methods: A review. Environmental Chemistry Letter 4, 51 61. Wu, J.J., Wu, C.C., Ma, H.W., Chang, C.C., 2004. Treatment of landfill leachate by ozone based adv anced oxidation processes. Chemosphere 54, 997 1003. Xing, W., Ngo, H.H., Guo, W.S., Hagare, P., 2008. Physico chemical processes for landfill leachate treatment: Experiments and mathematical modeling. Separation Science and Technology 43, 347 361. Xu, P ., Drewes, J.E., Kim, T.U., Bellona. C., Amy, G. 2006. Effect of membrane fouling on transport of organic contaminants in NF/RO membrane applications. Journal of Membrane Science 279, 165 175. Yoon, J., Cho, S., Cho, Y., kim, S., 1998. The characteristics of coagulation of fenton reaction in the removal of landfill leachate organics. Water Science and Technology 38 (2). 209 214. Yuan, W., Zydney, A.L., 2000. Humic acid fouling during ultrafiltration. Environmental Science and Technolo gy 34, 5043 5050.

PAGE 144

144 BIOGRAPHICAL SKETCH Shrawan Kumar Singh was born in Raebareli, India, to Krishna Devi and Sri Satya Narain Singh. He enrolled in the Govind Ballabh Pant University of Agriculture and Technology, Pantnagar, India in August 1998, and graduated with a Bachelor of Technology in Agricultural Engineering in July, 2002. He did his Master of Technology i n Environmental Engineering and Management from the In dian Institute of Technology Kanpur India in July 2004 before enrolling in graduate program at Department of Environmental Engineering Sciences at the University of Florida in January 2006, to study solid and hazardous waste management.