1 FATE OF MERCURY IN C ONTAMINATED SOILS TR EATED WITH ALUMINUM BASED DRINKING WA TER TREATMENT RESIDU ALS (AL WTR ) By KATHERINE Y. DELIZ QUI ONES 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 2013
2 2013 Katherine Y. Deliz Qui ones
3 To my son Edrian Javier
4 ACKNOWLEDGMENTS I thank my research committee Dr. Lena M a, Dr. Mark Brenner and Dr. William Wise for their time, their expert advice and both technical and moral support throughout the development and completion of this project. I especially thank my committee chair and mentor Dr. Jean Claude Bonzongo, for his g u idance, his support and his motivation to care for both my emotional and professional wellbeing He has led me with his knowledge and has helped me grow as a person and a researcher. I also want to give very special thanks to Dr. Doug Levy and Dr. Anne Donnelly for their financial s upport through the NSF sponsored SPICE and SEAGEP scholarships for their guidance and for their time knowledge sharing, and professional exper iences I a m in debt to both and cannot thank them enough for the positive impact they had on my professional and personal development I a m also grateful for the support of my family in Puerto Rico and for the support of my Church in Gainesville The latter has been my family away from home and I would like to thank them for their unc onditional love and support during difficult times and for providing guidance when it was most needed I want to thank my dear son Edrian for keeping my feet on the ground and being a constant reminder of the things that are really important in life. And l ast but not least I want to thank all my friends back home and in Gainesville for making my PhD journey a memorable one.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 M ERCURY CONTAMINATION IN SOILS: VALUE ADDED WASTE AS A POTENTIAL COST EFFECTIVE APPROACH FOR REMEDIATION ..................... 14 2 METALS IN CONTAMINATED SOILS: OVERVIEW OF REMEDIATION TECHNIQUES AND RATIONALE FOR U SING WATER TREATMENT RESIDUALS AS A SORBENT FOR MERCURY ................................ .................... 21 2.1. Metals Sorption in Soils ................................ ................................ ................... 22 2.2. Influence of Soil Fractions on Me tal Speciation and Sorption .......................... 24 2.2.1. Clay minerals ................................ ................................ .......................... 24 2.2.2. Oxides and hydrated metal oxides ................................ ......................... 25 2.2.3. Organic matter ................................ ................................ ........................ 26 2.3. Mercury in Soils ................................ ................................ ............................... 28 2.4. Remediation Techniques for Hg Contaminated S oils ................................ ...... 30 2.4.1. Ex situ techniques ................................ ................................ .................. 30 2.4.2. In situ techniques ................................ ................................ ................... 32 2 .5. Rationale for Using WTRs in Remediation of Hg Contaminated Soils ............. 35 2.5.1. WTRs: Overview of current knowledge ................................ .................. 36 2.5.2. Physi cochemical composition of Al WTRs ................................ ............. 38 2.5.3. Al WTR as a soil supplement/substitute ................................ ................. 39 2.5.4. Toxicity of WTRs ................................ ................................ .................... 39 2.5.5. Reactivity and potential for use in Hg remediation ................................ 40 3 IMMOBILIZATION OF MERCURY IN CONTAMINATED SOILS BY DRINKING WATER TREATMENT RESIDUAL S: ROLE OF MERCURY SPECIATION AND DISSOLVED ORGANIC CARBON ................................ ................................ ......... 43 3.1 Introduction ................................ ................................ ................................ ....... 43 3.2. Research Motivation ................................ ................................ ........................ 44 3.3 Materials and Methods ................................ ................................ ...................... 45 3.3.1 Collection and characterization of aluminum based drinking water treatment residuals (Al WTRs) ................................ ................................ ...... 45
6 3.3.2 Collection and characterization of soils ................................ ................... 45 3.3.3 Preparation of mercury spiked ORS soils ................................ ................ 46 3.3.4 Selective sequential extraction procedure (SSE) ................................ .... 47 3.3.5 Elemental analysis ................................ ................................ ................... 49 3.3.6 Column leaching experime nts ................................ ................................ 49 3.4 Results and Discussion ................................ ................................ ..................... 52 3.4.1 Characterization of Al WTR ................................ ................................ ..... 52 3.4.2 Physico chemical analysis of ORS soil ................................ .................... 52 3.4.3 Mercury leached from ORS loamy soil ................................ .................... 53 3.4.4 Mercury leaching from Hg s piked ORS soils ................................ ........... 57 3.5 Conclusions on Column Leaching Experiments ................................ ................ 58 4 MECHANISMS OF MERCURY IMMOBILIZATION BY ALUMINUM BASED DRIN KING WATER TREATMENT RESIDUALS: IMPLICATIONS FOR SOIL REMEDIATION ................................ ................................ ................................ ....... 76 4.1 Introduction ................................ ................................ ................................ ....... 76 4.2. Materials and Methods ................................ ................................ ..................... 77 4.2.1 Collection and characterization of Al WTR sample ................................ .. 77 4.2.2 Preparation of fresh and aged mercury spiked Al WTRs ........................ 78 4.2.3 Sample analysis by selective sequential extraction (SSE) procedure ..... 79 4.2.4 Total Hg analysis ................................ ................................ ..................... 82 4.2.5 Physical analysis of Al WTR samples ................................ ..................... 83 4.3 Results and Discussion ................................ ................................ ..................... 84 4.4. Summary of the Main Findings ................................ ................................ ........ 91 4.5 Future Research Avenues ................................ ................................ ................ 93 5 GENERAL CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK ................................ ................................ ................................ ................... 107 LIST OF REFERENCES ................................ ................................ ............................. 110 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 120
7 LIST OF TABLES Table page 3 1 Physicochemical characterization of Al WTR sample collected from the Bradenton Drinking Water Treatment Facility (Florida, USA) and used in this study. Adapted from Hovsepyan and Bonzongo (2009). ................................ .... 60 3 2 Elemental analysis of the Al WTR sample used in this study. Adapted from Hovsepyan and Bonzongo (2009). ................................ ................................ ..... 60 3 3 Physico chemical variables for Hg contaminated soil colle cted from Oak Ridge Site (ORS), Tennessee, USA. ................................ ................................ .. 61 3 4 Elemental analysis of Hg contaminated soil collected from Oak Ridge Site (ORS), Tennessee, USA. ................................ ................................ ................... 61 3 5 Concentrations (mg/kg) and distribution (%) of Hg, in the different chemical fractions of ORS (NSS), and Hg spiked ORS (HSS) soils; Oak Ridge, Tennessee, USA. ................................ ................................ ............................... 62 3 6 Mass of mercury (in either g or mg) in each chemical fraction of ORS soils (NNS) and Hg spiked ORS soils (HSS) used in column leaching experiments. ................................ ................................ ................................ ....... 63 3 7 Mass balance of Hg in column leaching studies using Oak Ridge Si te (ORS) soil. The initial total Hg mass in each column was ~0.54 mg, correspon ding to a THg concentration of 54.71 m g of Hg/kg soil. Soil remaining in columns was analyzed after leaching with 82 pore volumes of SPLP solution. ................ 68 3 8 Mass balance of Hg in Column leaching studies using Oak Ridge soil. The initial total Hg (THg) mass in each column was 0.54 mg, corresponding to a concentration of 54.71 mg of Hg/kg soil. Soil remaining in columns w as analyzed after leaching with 82 pore volumes using SRW (DOC= 53.3 mgC/L) ................................ ................................ ................................ ................ 70 3 9 Mass balance of Hg in Column leaching studies using Oak Ridge Site (ORS) soil. The initial total Hg (THg) mass in each column was 23 m g, corresponding to a concentration of 2567.4 mg of Hg/kg soil. Soil remaining in columns was analyzed after leaching with 40 pore volumes using SPLP solution (pH=4.22). ................................ ................................ ............................. 72 3 10 Mass balance of Hg in column leaching studies using Oak Ridge Site (ORS) soil. The initial total Hg mass in each column was 23 m g, corresponding to a concentration of 2567.4 mg of Hg/kg soil. Soil remaining in columns was analyzed after leaching w ith 52 pore volumes using Suwannee River water (SRW) with a DOC concentration of 53.3 mg/L and a pH of 4.2. ........................ 74
8 3 11 Percent of mobile fractions of Hg retained in Hg spiked ORS soil (HSS) amende d with Al WTR at 2% and 5% application rates and using two different incorporation schemes (uniformly mixed and bottom layer/liner) when leached with water containing increasing DOC concentration added as Suwannee River water (SRW) diluted in DI water ................................ .............. 75 4 1 Concentration and distribution (%) of Hg, Al and Si in specific chemical fractions of Hg spiked Al WTRs using the fractionation method adapted from Tessier et al.1979 (adapted from Hovsepyan and Bonzongo 2009) ................... 94 4 2 Concentration and distribution (%) of Hg, in specific chemical fractions of Hg spiked Al WTR samples determined by the SSE method adapted from Bloom et al. (2003) ................................ ................................ ................................ ........ 95 4 3 Abundance (%) of Hf 4f Al 2p S 2p O 1s C 1s N 1s and Cl 2p3 on the different Hg spiked Al WTR samples, as determined by XPS analysis ................................ 105
9 LIST OF FIGURES Fi gure page 2 1 Influence of solution pH on Al WTR efficiency and behavior. ............................ 42 3 1 Efficiency of Al WTR in immobilizing H g in a lab contaminated sandy soil leached with synthetic precipitation leaching procedure (SPLP) solution (Hovsepyan, 2008) ................................ ................................ ............................. 64 3 2 Experimental design for the first set of column leaching expe riments using Oak Ridge Site (ORS) soils with a starting THg concentration of 54.71 4.3 mg Hg/kg soil.. ................................ ................................ ................................ .... 65 3 3 Experimental design for the second set of column leaching experiments using Hg spi ked Oak Ridge Site (ORS) soil spiked to increase THg concentration from 54.71 mg Hg/ kg soil to 2567.4 mg Hg/kg soil ...................... 66 3 4 Hg leached from Oak Ridge Site (ORS) soil with an initial THg concent ration of 54.71 m g of Hg/kg soil. Soil columns were leached with synthetic precipitation leaching proc edure (SPLP) solution (pH=4.22) .............................. 67 3 5 Hg leached from Oak Ridge Site (ORS) soil with init ial THg concentration of 54.71 m g of Hg/kg soil using Suwannee River water with a DOC concentration of 53.3 mg C/L and pH=4.2 ................................ ......................... 69 3 6 Hg leached from Hg spiked Oak Ridge Site (ORS) soil with in itial THg concentration of 2567.40 of Hg/kg soil (corresponding to a total mass of ~23 mg Hg per column) using synthetic precipitation leaching procedure (SPLP) solution, pH=4.2 ................................ ................................ ................................ 71 3 7 Hg leach ed from Hg spiked Oak Ridge Site (ORS) soil with an initial THg concentration of 2567.40 mg Hg/kg of soil, corresponding to a mass of ~23 mg Hg/column. Columns were leached sequentially with solutions of different concentrations of dissolved organic carbon prepared using the Suwannee River water diluted in DI water ................................ ................................ ............ 73 4 1 SEM micrograph of Al WTR collected from Bradenton Drinking Water Treatment facility (Florida USA). ................................ ................................ ......... 96 4 2 EDS spectra of Al WTR samples ................................ ................................ ........ 97 4 3 Elemental mapping of aged Hg spiked Al WTR samples using TEAM EDS technology ................................ ................................ ................................ .......... 98 4 4 XRD spectra of 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg Al WTR .. 99 4 5 XPS wide scan survey spectrum. ................................ ................................ ..... 100
10 4 6 XPS spectra of Hg 4f ................................ ................................ .......................... 101 4 7 XPS spectra of O 1s ................................ ................................ ........................... 102 4 8 XPS spectra of Al 2p ................................ ................................ .......................... 103 4 9 XPS spectra of Si 2p ................................ ................................ .......................... 104 4 10 alumina surface, with reduced Hg(I) Hg(I) binuclear species ................................ ...................... 106
11 LIST OF ABBREVIATIONS A l WTR Aluminum based water treatment residuals CVAFS Cold vapor atomic fluorescence spectroscopy DOC Dissolve d organic carbon eCEC Effective cation exchange capacity H g Mercury SEM EDS Scanning electron spectroscopy Energy dispersive spectroscopy SPLP Synthetic leaching procedure SSE Selective sequential extraction THg Total mercury TOC Total organic carbon XPS X ray photoelectron spectroscopy XRD X ray diffraction spectroscopy
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 FATE OF MERCURY IN C ONTAMINATED SOILS TR EATED WITH ALUMINUM BASED DRINKING WA TER TREATMENT RESIDU ALS (AL WTR ) By Katherine Y. Deliz Quiones August 2013 Chair: Jean Claude J. Bonzongo Major: Environmental Engineering Science s Research in remediation of mercury (Hg) contaminated soil s remains challenging as it requires the deve lopment of cost effective and efficient remedial approaches that remove or immobilize contaminants while avoiding advers e effects on treated ecosystems S everal techniques have been developed in the past few decades. However, the in situ implementation of most of these techniques remains hampered by their excessive cost and limited environmental benefits to treated systems. In this study, the potential of an abundant and readily available waste material, the aluminum based d rinking water treatment residuals (Al WTRs) is evaluated, by emphasizing the concept added to contaminated soils. Column leaching studies were conducted to highlight the effect of acidic pH and dissolved organic carbon on the efficiency of Al WTR to immob ilize the leachable Hg fraction of a contaminated loamy soil Results showed (1) t he addition of Al WTR reduces Hg leaching potential of contaminated soils, and this effect is more pronounced i n soils with a high percentage of the soil Hg bound in both the water extractable and easily exchangeable fractions (2) Dissolved organic carbon concentration (DOC) levels > 26 mg C/L tend to reduce the efficiency of Al WTR to immobilize Hg; and (3) the
13 addition of Al WTR as a bottom liner is advantageous for fine tex tured soils with high clay and silt content. In addition a combination of analytical techniques including selective sequential extraction (SSE), scanning electron microscopy combined with X ray energy dispersive spectrometry (SEM EDS) X ray diffraction ( XRD) and X ray photoelectron spectroscopy (XPS) were used to determine Hg speciation and binding to Al WTR. These analyses confirmed Hg immobilization by Al WTR, occurs largely as a consequence of sorption onto amorphous Al 2 O 3 or other oxides such as SiO x prese nt in the residual fraction. In addition solid phase analysis suggest the initial incorporation of Hg to Al WTR occurs through a combination of outer sphere and inner sphere complexation driven by electrostatic attraction (on negatively charged partic les) and covalent bonding (on O and S contained in the di fferent fractions of Al WTRs). Further solid phase analyses are suggested to better understand how Hg binds to Al WTR and therefore gain insight on the long term stability of Hg Al WTR complexes.
14 CH APTER 1 MERCURY CONTAMINATION IN SOILS : VALUE ADDED WASTE AS A POTENTIAL COST EFFECTIVE APPROACH FOR REMEDIATION Heavy metals occur naturally in the environment as a consequence of processes such as pedogenesis and lithogenesis at levels that are regarded as trace and rarely toxic However, a nthropogenic activities and subsequent disturbances of natural systems have increase d since the ons et of industrialization leading to an exponential release of these metals to air, soil, and water systems. This contam ination has therefore chang ed the distribution patterns and impacts of metals on living organisms H eavy metal pollution is now a widespread environmental problem with negative implications at the global scale. An inventory conducted in the year 2000 show ed that in the USA alone about 80% of superfund sites are contaminated with heavy metals and more than 50,000 metal contaminated sites a wait remediation (Ensley, 2000) The current US for pollutants of serious environmental concern contains the following metals: silver ( Ag ) arsen ic ( As ) beryllium ( Be ) cadmium ( Cd ) chromium ( Cr ) copper ( Cu ) mercury ( Hg ) nickel ( Ni ) lead ( Pb ) antimony ( Sb ) selenium ( Se ) thallium ( Tl ) and zinc ( Zn ) (http://water.epa.gov/scitech/methods/cwa/pollutants.cfm). Unlike organic pollutants which are oxidized to carbon di oxide by microbial action, metals can be bio transformed leading to their environmental persistence, potential transfer to food chains and adverse effects on ecological functions and human health (Steinnes and Friedland, 2006) Past and current research in soil and sediment remediation ha s facilitat ed the identification and development of sorbents with high sorption capacity for toxic metals. Achieving efficient pollutant removal has been one of the key drivers in remediation
15 research. As a consequence a host of sorbents used in remediation of metal contaminated soils/sediments is available on the market or can be tailored to help attain specific sorption affinities for targeted classes of environmental pollutants (Leun and SenGupta, 2000) H igh adsorption capacity makes sorbent materials very effective in scavenging targeted contaminants from the complex biogeochemical composition of soil/sediment matrices. Unfortunately, the properties of various soil s /sediment s differ significantly, making the efficiency of most available technologies matrix dependent (Mull igan et al., 2001b) Although the biogeochemistry of metals in soils and sediments is well understood, research on the development of cost effective and envi ronment ally friendly remediation techniques remains challenging. In fact, the remediation of metal contaminated soils remains one of the most intractable problems of environmental restoration with relevance at both national and international levels (Hovsepyan and Bonzongo, 2009) The d evelopment and subsequent approval and application of a remediation technique for m etal contaminated soils is not a straightforward process It requires risk assessment studies that provide understanding of source s of contamination potential dispersion patterns environmental and health effects and the cost of applying such a technique Accordingly, most remediation techniques described in the literature are not implemented and do not go further than the experimental stage. Current methods for remediation of metal contaminated soil s and sediment s include physical separation, thermal pr ocesses, biological decontamination, electro kinetics, washing, stabilization, and solidification techniques. O nly a few of these techniques have been tested commercially, and their widespread use remains limited
16 largely a consequence of their prohibitive costs (Hovsepyan and Bonzongo, 2009) Therefore, a cost effective and efficient remedial approach that remove s or immobilizes metals in soils while avoiding adverse effects on treated systems is still needed A s mentioned earlier, a mongst metals of serious environmental and human health concern is mercury (Hg) In fact, Hg is probably one of the most toxic and perva sive metals in natural environments (Serrano et al., 2012) This stems primarily from its demons trated strong ability to bio accumulate and bio magnify in food chains and its harmful effects on living organisms. Wh ereas some heavy metals are essential for living organisms (animals and plants) in small amounts, Hg serve s no known biological function and can be toxic at very low concentrations causing serious damage to the nervous, cardiovascular and immune system s Because of these toxic characteristics, Hg is also part of the Substance Priority List of the Agency for Toxic Substances and Disease Reg istry (ATSDR), ranking third based on its frequency, toxicity, mobility, and long residence time (ATSDR, 2011) Major pathways involved in the deposition of Hg onto soil e nvironments include natural processes such as weathering of rocks, volcanic events and geothermal activity, anthropogenic sources like chlor alkali plants and mine wastes, and several diffuse sources such as coal fired power plants (Bose 2 009) and deposition of re emitted but previously deposited Hg Although most Hg cell chlor alkali plants in the US have been closed or converted to Hg free processes, a few facilities are stil l in operation. In fact, all closed chlor alkali plants in the US are now either superfund sites or Corrective Action Sites under the Resource Conservation and Recovery Act (RCRA) and require appropriate clean up actions (US EPA, 2013) Other
17 contributors to soil Hg pollution are both current (primarily in the developing world) and historic gold mining activities. Th e latter has left a legacy of Hg contaminated sites in the U.S. (Bonzongo et al., 1996a; Bonzongo et al., 1996b; Gustin et al., 1994) wh ereas the former continues to be an important source of Hg contamination in gold rich countries in South America, Africa, and Asia (Bonzongo et al., 2002; Donkor et al., 2006; Donkor et al., 2005; US EPA, 2000) Unfortunately, Hg has a long retention time in soils, and Hg accumulated in soils can bl eed into the surrounding environments for a long time, possibly hundreds of years (US EPA, 1997) Within soils, Hg can also be redistributed and transformed into different chemical forms, increasing the risk for biological exposure and toxic contaminated soils. Therefore, the determination of total Hg concen trations in soils is insufficient to accurately predict the environmental risks associated with Hg pollution. This is because Hg behavior, toxicity and impacts are highly dependent on its chemical speciation (Adriano, 2001; Ma and Rao, 1997; Snchez Polo and Rivera Utrilla, 2002) The latter controls the degree of toxicity and mobility of Hg, and therefore, its bioavailability and leaching potential from porous media (Bernaus et al., 2006; Bloom et al., 2003; Fernndez Martnez et al., 2005) Thus, besides the determination of total Hg content in contaminated soil matrices, performing studies that can assess the different Hg binding forms has been shown to be essential for evaluating environmental risks (Biester and Nehrke, 1997) Understanding the factors that influence metal retention and mobility in soils is fundamental to control and prevent its mobilization. Concerns regarding the accelerated increase of Hg introduction in to the environment and the resulting implications, along with the recent strengthening of most
18 environmental reg ulations at both state and federal levels have stimulated research on the development of new technologies for the treatment of Hg contaminated gas and water effluents and for the remediation of Hg contaminated soils and other toxic metals such as arsenic The developed decontamination techniques are categorized under biological, physical, and chemical methods (Hamby, 1996) and can be applied either ex situ (off site) or in situ (on site). They include convention al engineering based methods such as excavation, electro kinetics, soil washing/flushing, volatilization, thermal desorption and adsorption, and emerging technologies such as phytoremediation (Cox et al., 1996; Cunningham et al., 1995; Moreno et al., 2005a; Moreno et al., 2005b; Suer and Allard, 2003; Suer and L ifvergren, 2003; Wang and Greger, 2006) T he cleanup of Hg contaminated soils however, represents a significant challenge and the different remediation technologie s which are currently available, have several limitations. Recently, aluminum based drinking water treatment residuals (Al WTRs) w ere proposed for in situ immobilization of Hg in contaminated soils (Hovsepyan, 2008; Hovsepyan and Bonzongo, 2009) First, in the US alone more than 2 million metric tons of WTRs are produced daily (Agyin Biri korang and O'Connor, 2009) Second, this generated waste material is disposed of in landfills, stored in onsite lagoons, or sometime s discharged into river systems (Novak and Watts, 2004) The disposal of this waste product can be expensive ( ~ $50 / ton) and increases the overall cost of the water purification process (Novak and Watts, 2004) Also, h andling of WTRs may account for 50 % of the total operating budget Accordingly, t he use of WTRs for soil remediation represents a way of adding value to this waste material with potential significant benefits associated with the recovery of currently polluted systems and elimination of
19 diffuse sources related to the leaching of Hg contaminated soils Because current remediation methods of Hg polluted soils are rather expensive and have several other disadvantages, the use of Al WTRs as soil amendments to immobilize Hg c ould significantly lower the cost of potable water, eliminate the need for disposing of these waste materials in land fills and provide environmental benefits as di scussed elsewhere (Hovsepyan, 2008; Hovsepyan and Bonzongo, 2009) The expectation for high performance of Al WTRs in the immobilization of Hg in contaminated soils stems from the results of preliminary experiments c onducted in our research group (Hovsepyan, 2008; Hovsepyan and Bonzongo, 2009) Sorption isotherms indicated a strong affinity of Hg for Al WTRs and a relatively high maximum sorp tion capacity of ~79 mg Hg/g Al WTRs (Hovsepyan and Bonzongo, 2009) C olumn leaching studies using Hg contamin a ted soils amended with Al showed that Al WTRs c ould efficiently contr ol Hg mobility by significantly reducing or eliminating Hg leachability from soils (Hovsepyan, 2008) There are, however limited data on the long term stability of Hg compounds formed after soil amendments by sorbent additions (Biester and Zimmer, 1998) and a limited understanding of the mechanisms of Hg fixation onto Al WTRs and how changes in key environmental variables affect the ability of Al WTRs to sorb and immobilize Hg from soil solutions. T his research builds up on the findings of Hovsepyan (2008) The first chapter of this dissertation outlines the research questions or problem statement and establishes the need for cost effective and environmental friendly solutions to control/reduce the mobilization and bioavailabili ty of Hg in contaminated solid matrices. The second chapter is a review of relevant literature focusing on
20 background information on sorption of metals in soils, current remediation techniques, in soil remediation Ch apter 3 presents the results of laboratory experiments designed to assess the potential of Al WTR to reduce Hg leachate from both naturally contaminated soils and Hg spiked soils. Chapter 4 focuses on the determination of the different mechanism(s) involve d in the mobilization of Hg onto Al WTRs Here, physicochemical information are sought to gain insight s into the potential long term stability of formed Hg WTR complexes, if used in in situ soil remediation. Finally C hapter 5 summarizes the main findings o f this dissertation.
21 CHAPTER 2 METALS IN CONTAMINATED SOILS: OVERVIEW OF REMEDIATION TECHNIQUES AND RATIONALE FOR USING WATER TREATMENT RESIDUALS AS A SORBENT FOR MERCURY Decades of research in soil remediation have significantly improved our knowledge in this field. Unfortunately, the identification of efficient remediation methods remains challenging, as it requires the development of cost effective and efficient remedial approaches that remove metals and other contaminants, while avoiding adverse effe cts on treated systems. It is now quite well established that the complexity of metal contaminated soils varies with site location, the source and type of metal contamination, and the history and age of metal contamination. Anthropogenic sources of metals that result in soil pollution levels which are of concern to humans include mining activities, nuclear material processing, wastewater sludge, metal plating, and the industrial manufacture of batteries, metal alloys, munitions, electrical components, paint s, preservatives and insecticides (Roane and Kellogg, 1996) The focus on the location of the contaminated site allows the different remediation techniques to be classified into two categories only. First, the most common approach to cleaning metal contaminated soil sites is physical removal and landfilling. This is not only an expensive option, but it merely moves the contamination from one location to another. The second approach takes place onsite and in this in situ treatment, metals are either extracted from contamina ted soils, or simply stabilized so that they can no longer move off site. However, for this second option it is important that the remediation process be as noninvasive and environmentally benign as possible, especially if the end product is intended to b e a healthy and productive ecosystem
22 2.1. Metals Sorption in Soils For metals to be adsorbed to soil surfaces they first need to be attracted by charges generated o n the surface of soils and soil constituents. The two main types of charges that can develo p on soil surfaces are permanent and pH dependent (Essington, 2004) Permanent charges are developed by isomorphic substitution when the mineral is formed, thus, it is a property of the mineral and cannot be affected by changes in the surrounding environment. On the other ha nd, pH dependent charges are influenced by mineral surfaces and environmental variables, and fluctuate as the result of protonation at low pH (formation of positive charges) and deprotonation at high pH (formation of negative charges) (Essington, 2004) The relative affiniti es of metal ions for adsorption sites on charged surfaces determine the stability of the metal ligand bond, and are influenced by properties of the metal such as ionic radius, ionization potential, electronegativity, hydration and polarizability (Zhou and Haynes, 2010) In view of these conditions, the selective adsorption and hysteresis of metals can be largely explained in terms of the Lewis hard soft acid base (HSAB) principle. Hard acids are those with a high electronegativ ity, low polarizability and small ionic radius, wh ereas soft acids are the opposite of these. Accordingly, hard acids such as Na + K + Mg +2 Ca +2 Mn +3 Fe +3 and Cr +3 show a preference for hard bases like water and Fe oxides; wh ereas soft acids such as Hg + 2 and Cd +2 prefer to complex with soft bases like clay minerals. Once metals have been attracted to soil surfaces different physical and chemical forces control the adsorption reactions driving the complexation between metals and the different soil const ituents such as organic matter, clay minerals, carbonates, alumino silicates and iron, manganese and aluminum oxides and hydroxides. The main mechanisms for metal adsorption into these soil constituents
23 include inner sphere complexation, outer sphere compl exation and diffuse ion association (Lestan et al., 2003; Strawn and Sparks, 1999) Inner sphere complexation occur s when the ligand displaces a water of hydration from the coordination sphere (Essington, 2004) This process results in the formation of stable metal soil complexes, which are only weakly affected by the ionic strength of the soil solution (Sparks, 1995) T hese initial reactions of metal attachment to the surface of soil particles are usually followed by diffusion of metal cat ions into the macro and micro pores of soil particles, hindering subsequent desorption of these metals. The m ajor soil constituents involved in the specific adsorption of metals include organic matter and amorphous and crystalline hydrous oxides of Fe, Al and Mn (Zhou and Haynes, 2010) As explained previously, several characteristics of metals ions can in fluence their binding forces and thus the sorption mechanisms. For instance the high electrical charge of the atomic nucleus, small ionic size and high polarizability of heavy metals makes them suitable for specific adsorpti on reactions in soil Trace anio nic metals are also preferentially adsorbed by inner sphere complexation over major anions. However, if specific adsorption sites become saturated these trace elements can also be accumulated in non specific adsorption sites. In outer sphere complexation metals are bound to the soil particles by electrostatic attraction. This process is also known as non specific adsorption because it occurs when at least one solvent (water) molecule interposes between the surface functional group and the bound metal ion (Zhou and Haynes, 2010) Outer sphere complexation reactions are reversible, diffusion controlled, stoichiometric process es in which the adsorbents show differences in the selectivity for metals, and can be
24 influenced by pH, valence of the metal, surface charg e and the concentration of ions. According to the principle of electroneutrality, i n order to stay in equilibrium the non specific adsorption of metals should be followed by desorption of stoichiometric quantities of cou nter ions. This process is also known as ion exchange because the addition of other cations into the system, in sufficient concentration, will result in the replacement of original cations at the soil surface. Outer sphere complexation of metals can occur via physical adsorption (though van der Waals forces), cation exchange and hydrogen bonding (Kabata Pendias, 2001) Just as in outer sphere complexation reactions, in diffuse ion associations metal ions are surrounded by water of hydration. Hence, metals accumulate at the interface of charged surfa ces because of electrostatic interactions, but are not directly bound to the soil surface (McLean and Bledsoe, 1992) These reactions occur rapidly and can be reversed easily by changes in environmental conditions such as pH and concentration of ions in solution. 2.2. Influence of Soil Fractions on Metal Speciation and Sorption The affinity of meta l ions for adsorption sites will depend on the characteristics of the sorbent (soil) and associated aqueous phases (Zhou and Haynes, 2010) Adsorption capacity is different for different types of soil (mineral vs organic). Ac cordingly, the abundance of organic matter, clay minerals and hydrous oxides of Al, Fe and Mn in soil will tend to dictate the type of metals that will be adsorbed by either inner sphere or outer sphere complexation and to what extent such sorption can occ ur 2.2.1. Clay minerals The texture of soils and their composition is extremely important in sorption processes. In general all clay minerals are characterized by having a high surface area,
25 which provides them with numerous negative adsorption sites (h igh CEC) for metals cations to bind. Hence, much work has focused on the ability of clay minerals (e.g. kaolinite, allophone, vermiculite, montmorillonite) to adsorb heavy metals (Abollino et al., 2003; Abollino et al., 2008; Malandrino et al., 2006; Sarkar et al., 2000) Permanently charged sites on clay mineral surfaces interact with metal ions by means of non specific electrostatic forces (Adriano, 2001) T here is also however evidence that suggest s metals can be held in a non exchangeable form by chemical interactions between heavy metals and SiOH and AlOH groups o n the edges of clay minerals (Abollino et al., 2003; Petruzzelli, 19 97) Both of these mechanisms are pH dependent. Accordingly a study by Abollino et al. (2003) shows that metal complexes wit h silanol (Si O) and aluminol (Al O) groups at the clay particles surface is unsupported under low pH (2.5 3.5) conditions. Also, a study by Lippold and Lippmann Pipke (2009) showed that affinity of metals for clay mineral sorption sites is also affected at high concentrations of humic matter throughout the acidic range. Sorption capacities of some of these minerals will also differ when saturated with different cations (Kabata Pendias, 2001) 2.2.2. Oxides and hydrated metal o xides Metal oxides and hydroxides are secondary minerals formed as a result of mechanical (heat, water ice, pressure) and chemical/biological (Fe and Mn oxidizing bacteria such as Thiobacillium sp and Metallogenium sp respectively) weathering. Oxides a nd hydrated metal oxides that exist as poorly crystalline, colloidal size particles, with a high reactive surface, can form strong complexes with numerous trace metals (Essington, 2004) However, when these accessory minerals associate with primary minerals and other seconda ry minerals their surface properties are masked and the
26 access of solutes to interlayer adsorption sites gets blocked. For example, in natural soils clays are usually coated by oxides which make the clay surface un available for metal sorption directly f rom the soil solution (Yin et al., 1997) I n this case sorption depend s on the affinity of metals for the oxide instead of the clay. According to Kabata Pendias (2001) isomorphic substitution, cation exchange and oxidation are the major mechanisms involved in the sorption of trace metals by metal (hydr) oxides. However the stability of these secondary minerals is subject to redox state, pH, organic matter content and microb es in the soil. For example, decreasing pH conditions limit the adsorption of metal cations (by increasing the sorption capacity for anions), whereas at high pH the presence of Fe and Mn oxides generally increases the adsorption of metals in soils (Adriano, 2001) Accordingly soil minerals with low pH zpc have greater ability to attract and retain cations over a broad soil pH range because their surface charge is mainly negative at pH values common in soils On the other hand, minerals with higher pH zpc are better at retaining anions (Essington, 2004) In addition adsorption of s ome metals by (hydr)oxides increase s significantly in the presence of humic substances (Adriano, 2001) 2.2.3. Organic matter In soil, organic matter (OM) ca n be found as organic residues that are partially decomposed organic matter, or as humus/soil organic matter (SOM), that consists of non humic and humic substances. The nature of SOM is a function of the soil type, vegetation and microbial fauna and plays an important role in the environment, affecting among other things the mobility of metal ions in soil. Polyvalent metal ions can form bonds with non humic compounds such as organic acids, amino acids, proteins and polysaccharides, through a range of mecha nisms including ion exchange, covalent
27 bonds and chelation (Kabata Pendias, 2001; Zhou and Haynes, 2010) Metal ions can also form strong bonds with humic substances, which are high molecular weight molecules that exist as heterogeneous, complex, three dimensional amorphous structures, because of their high number of O containing fu nctional groups such as phenolic, enolic, carboxyl ( COOH) and alcoholic OH and C=O groups (Zhou and Haynes, 2010) Although the heterogeneous nature of humic substances provides a large number of binding sites that range fr om weak forces of attraction to stable coordinate inner sphere complexion, the high degree of selectivity of OM indicates that most heavy metals preferentially coordinate with functional groups forming inner sphere complexes instead of just being adsorbed by simple exchange reactions (Adriano, 2001; Zhou and Haynes, 2010) Specificity of metals toward the different functional groups is determined by the metal properties and the different soil characteristics The mult iple interactions between SOM and metal ions can lead to either water soluble or insoluble complexes. For example, when metals bind to relatively stable humic compounds in the solid phase, these metals are taken out of solution, which decreases their mobil ity and bioavailability in soils (Adriano, 2001; Zhou and Haynes, 2010) However, when metal ions bind to OM in the aqueous phase, soluble organic matter can act as a carrier for metals increasing their mobility to o ther environmental compartments (Adriano, 2001; Zhou and Haynes, 2010) Different factors affect the complex ion ability of specific metals to organic matter. Some of these relate to properties of the metal itself, bec ause small ionic radius and high electronegativity favors the formation of stronger complexes. Other factors include the nature of the organic material and the quantity, type and reactivity of the different functional groups
28 (McLean and Bledso e, 1992; Zhou and Wong, 2003; Zhou and Haynes, 2010) Furthermore, adsorption of metals by SOM and the mobility of metal OM complexes are strongly related to soil pH. According to Yin et al. (2002) DOM increases at high pH, facilitating the mobility of metals through the formation of soluble metal OM complexes. In addition, in soils with high concen trations of OM, high pH tend s to decrease proton assist ance of formation of soluble metal OM complexes. 2.3. Mercury in Soils Mercury (Hg) is a naturally occurring element that cycles through the atmosphere, the hydrosphere, the biosphere and the solid ge osphere (Johannessen et al., 2005) In soils w h ere it tends to concentrate, it ca n be found as free ions in both mono and di valent forms (i.e. Hg 2 +2 and Hg +2 ), bound to inorganic and organic ligands, or present as solid minerals including HgO and HgS (Schuster, 1991) Factors controlling the transformations of Hg from any of the above forms to another include pH, redox potential, ionic strength and concentrations of organic matter (OM) (Babiarz et al., 1998; Schuster, 1991) Hg is a chalcophilic metal, which means it has a high affinity for sulfides and low affinity for Lewis bases such as carboxyl groups. Some of the most abundant forms of Hg in the environment include HgS Hg(OH) 2 and HgCl 2 D epending on the physical and chemical parameters of the soil these compounds can be further compl exed with organic ligands (Gabriel and Williamson, 2004) It is important to understand the factors affecting Hg speciation because the nature of sorption/desorption reactions taking place in the soil strongly influences Hg toxicity. Accordingly several studies have focused on the effects of p H, OM, redox conditions and competitive ions on the sorption and desorption reactions of Hg in soil/sediment (Aiken et al., 2003; He et al., 2007; Jing et
29 al., 2007; Kim et al., 2004a; Kim et al., 2004b; McLean and Bledsoe, 1992; Sarkar et al., 1999; Skyllberg et al., 2006; Skyllberg and Drott, 2010; Wang et al., 2009; Wang and Greger, 2006; Yang et al., 2 006; Yang et al., 2008) Among the factors controlling the adsorption of Hg ions by soil constituents, pH is one of the most i mportant (Yin et al., 1996) Solubility of Hg compounds is related to their oxidation states and var ies from negligible (e.g. HgS, Hg 0 ) to very soluble (e.g. HgCl 2 Hg(OH) 2 Hg 2 SO 4 ) (US EPA, 1997) According to Adriano (2001) at low pH cations (e.g. HgCl + ) and neutral (HgCl 2 ) Hg species are formed, wh ereas oxy/hydroxyl Hg species (Hg(OH) 2 ) predominate at higher pH. Yin et al. (1996) show ed that as the pH increases from 3 to 5 the proportion of hydr oxide Hg species increases. A similar study by Andersson (1979) show ed that in acidic soils (pH<4.5 5) Hg +2 is mainly bo und to organic ligands, whereas a higher affinity is shown for iron oxides and clay minerals in soils with circumneutral to neutral pH Although adsorption of metal ions is usually gr e ater at higher pH, studies have shown a higher adsorption of Hg +2 in aci dic media (Hovsepyan and Bonzongo, 2009; Yin et al., 1996) T he reduction/oxidation potential (Eh) is a nother important parameter used to determine/predict H g mobility in the environment Under oxidizing condit ions H g chloride compounds (e.g. HgCl 2 Hg 2 Cl 2 ) and hydroxyl Hg species ( e.g., Hg(OH) 2 HgO) predomina te In contrast, under reducing conditions most Hg is bound to S and amorphous FeS. Soil texture also has an effect on Hg mobility because metals are more firm ly retained in clay soils than sandy soils. Microorganisms can also assist in the transformation of Hg to the organometallic methyl Hg compounds, a group of very highly mobile and toxic forms of Hg (Camps Arbestain et al., 2009)
30 Another factor that affects the adsorption of metal ions is SOM. It is well known that Hg can form strong bonds with organic matter in soil (Grigal, 2003; Yin et al., 1997) Column studies by Miretzky et al (2005) showed an increase sorption capacity for Hg at higher organic matter content. Whereas the opposite effect was observed in soils with a high content of dissolve organic matter (DOM) (Yin et al., 1997) Although Hg occurs naturally in the environment, anthropogenic disturbances have led to its accelerated release and accumulation in ecosystems. For instance, the States was estimated to be about 0.11 mg/kg (Adriano, 2001) T otal Hg (THg) concentration near gold mining sites were reported to contain concentrations ranging from1 32 mg/kg to 635 mg/kg (Kim et al., 2003) and those areas surrounding chlor alkali plants had THg concentrations ranging from 4.3 mg/kg to 1150 mg/kg (Bernaus et al., 2006) The remova l of Hg (II) from soils and sediments, therefore, assumes significance. 2.4. Remediation Techniques for Hg Contaminated Soils 2.4.1. Ex situ t echniques Excavation has been implemented at full scale and is the most commonly used method for cleaning metal co ntaminated sites (Hinton and Veiga, 2001) It involves the remov al of contaminated soil and its subsequent transport to hazardous landfills or off site treatment facilities. Disposal in landfills is expensive and represent s an additional problem because the majority of the soil mass is not composed of the actual pollut ant, which reduces the space to discard other hazardous materials. In addition excavation of large contaminated areas and the need to replace the soil that has been removed increase costs substantially (Mulligan et al., 2001a; Mulligan et al., 2001b)
31 Furthermore, disturbing the soil increases the risk of spreading Hg contamination (Bower et al., 2008) Soil w ashing removes heavy metals from the soil through a combination of phy sical separation and chemical extraction procedures. The physical separation of Hg from the soil involves sepa ration of larger particles (e.g., by gravity) from finer particles, the latter of which will subsequently be homogenized and chemically treated. C hemical extraction removes heavy metals from the fine particles by adding a washing solution of water enhanced with chemical additives such as leaching agents, surfactants, bio surfactants, inorganic and organic acids, chelating agents or a mixture of the above (Dahrazma and Mulligan, 2007; US EPA, 2007) Physical separation reduces the volume of soil that will require chemical treatment as well as the costs, however the applicability and effectiveness of these techniq ue may be limited for complex waste mixtures (such as metals mixed with organic compounds), soils with more than 40% silt or clay content, certa in types metal speciation, and high pH and moisture content (US EPA, 2007) Volatilization involves the use of high temperatures and low pressure to volatilize contaminants in soils. After volatilization Hg can be recovered through condensation, to collect l iquid elemental mercury (US EPA, 2007) or immobilization (Mulligan et al., 2001b) Some factors affecting the performance and the cost of this technique include soil type and texture, organic and moisture content and Hg co ncentration in the soil (US EPA, 2007)
32 Amalgamation is the dissolution of mercury in other metals like copper, nickel, zinc and tin, to form a semi solid alloy. This process is sometimes followed by encapsulation to prevent volatili zation from the amalgam (US EPA, 2007) 2.4.2. In situ t echniques Soil flushing techniques such as soil washing remove contaminants from the soil and concentrate them in a liquid phase that can be collected and treated as was te water In such cases, chemicals used as extract ing solutions are applied to the contaminated soil by surface flooding, sprinklers or soil injection (NRC, 1999) Overall, the leachate needs to be collected and treated as hazardous waste water. Electro kinetic remediation uses a low electric current to move water and metal ions (by electro osmosis and electro migration respectively) between a cathode and an anode inserted vertically in to the contaminated zone (Mulligan et al., 2001b; Suer and Lifvergren, 2003) The effectiveness of this technique depends on both solubility and transport of the contaminant through the soil matrix. Electro remediation of Hg contaminated soils is challenging because of the low solubility of some Hg species (HgS, Hg (l), Hg 2 Cl 2 ) at pH values common to most natural soils (Cox et al., 1996) In s itu t hermal d esorption is a technology that applies heat and vacuum to the surface of contaminated soils to modify and extrac t volatile contaminants (US EPA, 2007) Bioremediation is a process in which metal contaminants are modified as a result of the activity of organisms such as bacteria (Lovley and Coates, 1997; Rathinasabapathi et al., 2006) or plants (Cao et al., 2009) For instance, organic acids produced by both bacteria and f ung i can induce leaching and metal removal in contaminated soils (Mulligan et al., 2001b) I nvestigations on the use of plants in soil
33 clean up show promise for certain metals Bioremedi ation has also been proposed as a second stage to treat wastewaters generated from soil treatment techniques like soil washing and soil flushing. Microbial processes being studied for metal remediation include biotransformation, bio precipitatio n, and bio sorption (Chen and Wilson, 1997) Examples of these processes include the use of Hg reductase, produced by some Pseudomonas strains, to convert mercuric ions to Hg 0 under aerobic conditions. However, with time, p roduced Hg 0 will accumulate within the microbial biomass and will have to be extracted. In addition, Hg contaminated wastewater effluents can be exposed to microbial populations that conv ert soluble and bioavailabl e Hg to less soluble forms (e.g. HgS) followed by s edimentation/filtration (US EPA, 2007) Although these microbial driven approaches can be cost effective, their performances can be negatively affected by high Hg concentrations and other changing water quality variables (pH, nutrients, temperature, etc ) (US EPA, 2007) A biotechnology prog ram was developed recently at the National Research Institute for Biotechnology in Germany. The approach in this program is based on the use of proprietary microbes to reduce ionic Hg species to Hg 0 and the major advantage of this techn ology is that the microorganisms used are able to grow regardless of Hg concentrations (even in the presence of Hg concentrations > 100 mg/L), resulting in Hg reduction efficiencies up to 99% (Hazardous Waste Consultant, 1996) Phyto extraction and phyto volatilization are emerging technologies that use plants to extract heavy metals from the soil and either accumulate them in aboveground shoots or volatilize them (Garbisu and Alkorta, 2001) These techniques are les s expensive than most technologies that are currently available. However, effectiveness is
34 dependent upon several plant characteristics. Hyper accumulators of heavy metals in harvestable parts that tolerate high metal concentrations and have a rapid mass g rowth are the b est option (Garbisu and Alkorta, 2001) Nevertheless, metals need to be present in soluble forms for plant uptake. Chemicals for increasing the availability of metals in soils include acidifying agents chelating materials and fertilizer salts (Moreno et al., 2005b) Plants can also be genetically engineered to enhance their ability to extract and transform or volatilize pollutants such as Hg (Hussein et al., 2007; Moreno et al., 2005a) Studies by Hussein et al. (2007) showed that chloroplast transgenic plants are able to accumulate various forms of Hg at concentrations higher than the soil. M ore research is needed to enhance plant ability to extract metals and to understand how bioavailability correlates with metal uptake from soils (Mulligan et al., 2001b) To date there are no reports of naturally occurring plants that can be classified as hyper accumulators of Hg. Immobilization by solidification and vitrification is a set of techniques that includes s olidificati on as an approach to reduce the mobility, and subsequently the toxicity of metals in soils through physical binding or encapsulation of the contaminant with cement ing agents (Hinton and Veiga, 2001) or use of heat. For instance, vitrification is one example of a solidification process that requires thermal energy to form a glassy soli d that is chemically durable and leach resistant. This process can be performed in situ or ex situ For in situ vitrification electrodes are inserted vertically in the contaminated site, and then an electric current with a heat of about 1,600 2,000C is p assed though, melting the adjacent soil and solidifying it as it cools (Mulligan et al., 2001b; US EPA, 2007) The heat spreads outwardly and downwardly (about 20 feet),
35 and each electrode has a maximum treatment capa city of 1,000 tons (US EPA, 2007) The effectiveness of this process can be affected by high concentration s of clay, debris and moisture. Some disadvantages include restriction of soil u se and high implementation costs. For e x situ vitrification contaminated material is heated in a s melter or furnace. Additionally, metals in contaminated soils can be immobilized by use of environmentally benign sorbents. This process requires the additio n of amendments with high sorption affinity for the target metal(s), and can be relatively inexpensive depending on the type of sorbent used However, to be effective the metal has to be irreversibly bound to the sorbent in a non labile and non biological ly available form under natural soil conditions, and fairly stable under changing environmental parameters ( e.g., pH, oxidation and reduction). Although the metal will not be available for bio uptake, immobilization techniques do not change the total conta minant level in the soil and the lack of data regarding stability of long term metal sorbent complexes continue to limit the general acceptance of in situ immobilization by regul atory agencies (Zhou and Haynes, 2010) With re gard to Hg, different Hg binding compounds have been tested to remove Hg from solutions (clays, resins, silica gels, zeolites, and activated carbon), but with only a few methods proposed for Hg remediation of soil (Chen et al., 2004; Meng et al., 1998) The focus in this area has been on a readily available waste material, the aluminum drinking water treatment residuals (WTRs). 2.5. Rationale for Using WTRs in Remediation of Hg Contaminated Soils Across the nation, wa ste residuals that are produced in abundance during drinking water treatmen t processes are either discarded in landfills, stored in onsite
36 lagoons, or simply discharged into river s (Novak and Watts, 2004) D isposal of this waste product can be expensive and increases the overall cost of water purificatio n (Novak and Watts, 2004) One way to add value to this abundant and readily available waste material is to take advantage of its physic chemical reactivity with and high sorption capacity for certain inorganic pollutants to remediat e contaminated environments. So far, ongoing research has focused primarily on the potential use of WTRs as cost effective material for the remediation of environmental compartments contaminated with oxyanions such as PO 4 3 AsO 4 3 and SeO 4 2 (Ippolito et al., 2011) However, the findings of Hovsepyan and Bonzongo (2009) and results from other res earch groups (Brown et al., 2005; Chiang et al., 2012) point to a high potential for metal cation immobilization by WTRs. 2.5.1. WTRs: Overview of current knowledge Water treatment processes used worldwide to produce safe drinking water generate a wide variety of residual products (Ippolito et al., 2011) In the conventional coagulation filtration treatment process, suspended solids and natural organic matter are removed from the raw wate r supply by addition of coagulants such as Al 2 (SO 4 ) 3 3 or Fe 2 (SO 4 ) 3 The process results in the production of water treatment residuals or WTRs. Accordingly, aluminum (Al) and iron (Fe) based WTRs (referred to thereafter as Al WTRs and Fe WTRs respectively) are the most common by products of drinking water treatment facilities. In addition to the above coagulants, the chemical composition of WTRs is also impacted by the mineral and elemental signature of the treated raw water (Elliott et al., 2002) With regard to soil remediation with WTRs, the use of Fe WTRs in metal contaminated soils is less attractive because of the highly redox sensitive nature of iron, which could result in iron oxy/hydro xide
37 dissolution under anoxic conditions and release of previously sorbed metal ions. Therefore, this study focuses solely on aluminum based WTRs which hereafter ar e referred to as Al WTRs. Extensive research has now shown that, as a result of their chemi cal composition and high reactivity, WTRs can potentially be used as soil amendments to increase phosphorus (P) sorption capacity and reduce the impacts of runoff and infiltration on water quality (Dayton et al., 2003; Novak and Watts, 2004; Silveira et al., 2006) L aboratory studies suggest that the mechanism of P sorption onto WTR matrices tends to be biphasic, with an initial fast sorption reaction on the surface attributed to electrostatic interactions between P and WTRs, followed by a slower sorption step associated with intra particle diffusion of P into micropores (Makris et al., 2005) In a study by Novak and Watts (Novak and Watts, 2004) P sorption data on WTRs were best fit to a first order reaction model, wh ereas P kinetic data by Makris et al. (2005) fit a second order reaction model best This d iscrepancy can be attributed to factors such as the lack of homogeneity of WTRs used in these two studies. T hese authors nevertheless concur on the strong ability of WTRs to sorb P. More recent studies have explored the ability of WTRs to sorb positively c harged metal ions. Hovsepyan and Bonzongo (2009) investigated the ability of Al WTR to sorb mercury from aqueous solutions, and using Fe WTR, Chiang et al. (2012) studied the removal of cadmium (Cd), lead (Pb), and zinc (Zn) from aqueous solutions. Research by Brown et al. (2005) found that WTR amendments in soil reduced the NH 4 NO 3 extractable Cd, Pb, and Zn fractions, suggesting a strong ability for immobilization of these metals as well. Although data on the mechanisms of metal sorption on WTRs are
38 still l acking, these preliminary findings point to the potential of WTRs as sorbent s for metal cations, and further research is necessary to find out how WTRs can be used to improve environmental quality. 2.5.2. Physicochemical composition of Al WTRs A s stated earlier, Al WTRs are formed as a result of addition of aluminum salts (e. g Al 2 (SO 4 ) 3 ) and other coagulants (such as polymers) to raw water to remove colloids, silt and clay size particles, and color (Dayton and Basta, 2001; Elliott et al., 2002; Makris and Harris, 2006) They consist of particles that settle as a result of coagulation and flocculation (Makris et al., 2005) The amorphous structure of Al WTRs is caused by the formation of am orphous Al oxides and hydroxides (Makris and Harris, 2006; Novak and Watts, 2004) According to the American Society of Civil Engineers, amorphous aluminum oxides/hydroxides account for 50 to 150 g/kg of WTRs (Dayton and Basta, 2001) and depending on the treatment process used, Al WTRs can also contain small quantities of activated carbon, organic matter, and sediments (Makris et al., 2004) In our preliminary study, analysis of Al WTR mater ial collected from a dry disposal site in Bradenton Florida, showed that Al WTR can have rather low initial metal content compared to federal regulatory limits for soils, while exhibiting a relatively high specific surface area or SSA. In fact, the N 2 BET and CO 2 BET analyses revealed SSA of 48 m 2 /g and 120 m 2 /g, respectively, suggesti ng a high er density of sorption sites within micropores. Total Hg and other analyzed metal concentrations were found to be lower than the regulatory limits for land applicatio n of sewage sludge (US EPA 40 CFR Part 503).
39 2.5.3. Al WTR as a soil supplement/s ubstitute Dayton and Basta (2001) investigated the prospect of using seventeen WTRs (fourteen of which were Al WTRs) as soil substitutes by comparing physic c hemical properties and nutrient composition of the WTRs with typical soils. They found that most of the studied Al WTRs had soil like qualities with the potential to be used as soil substitutes for land reclamation purposes. The physic chemical properties of WTRs were found adequate for plant growth. If used solely for crop growth, the drawback of WTR application would actually be the reduction of P bioavailability, which could lead to P deficiency in plants if applied at excessively high rates. However, th is shortcoming can be easily corrected by applying a P fertilizer to the treated soil whenever necessary (Hyde and Morris, 2000) 2.5.4. Toxicity of WTRs Dayton and Basta (2001) also evaluated the potential toxicity of Al WTRs using the toxicity characteristic leaching procedure (TCLP), a US EPA standard procedure used to profile waste materials as hazardous or non hazardous for the purpose of landfil ling. The metal content of all studied Al WTRs was significantly below the regulatory levels for TCLP and therefore categorized as non hazardous waste. Indeed, metal concentrations of WTRs are not regulated by the US EPA 503 st atute for sewage sludge (US EPA, 1997) Studies by Hyde and Morris (2000) have also shown that metal concentrations of studied WTRs were lower than the US EPA regulatory limits for metals of environmental concern, and the dissolved Al concentration s in extracts of Al WTRs did not produce toxicity symptoms in soyb eans and corn (Dayton and Basta, 2001) Studies by Sotero Santos et al. (2005) demonstrate d that Al WTRs did not cause acute toxicity to the aquatic invertebrate Daphnia similis in a 48 hour toxicity assay.
40 Also, no evidence of aluminum toxicity was found when alum based sludge was used as growth media (Babatunde and Zhao, 2006) Our own preliminary results obtained using MetPlate, a bi oassay based on bacterial galactosidase inhibition by toxic metals (Gao et al. 2009), showed a total elimination of soil leachate toxicity when treated with Al WTR in column studies (Hovsepyan, 2008) 2.5.5. Reactivity and potential for use in Hg r emediation The expectation of a high performance by Al WTRs in the immobilization of Hg in con taminated soils stems from laboratory experiments, which assessed the potential of Al WTRs to adsorb and immobilize Hg. Overall, results show ed that Al WTRs have a strong potential for Hg immobilization in soils (Hovsepyan and Bonzongo, 2009) F or instance, results of Hg adsorption by Al WTR from aqueous solutions ( Figure 2 1 A ) show the highest Hg removal at a very acidic pH of 3, wh ereas the lowest Hg removal was obse rved at pH 5. Hg sorption decreased from pH 3 to 5 and then increased gradually with increasing pH, from 5 to 8. Measurements of the zeta potential ( ) of tested Al WTRs as a function of pH showed that as the pH decreases from 7 to 3, the surface charge of Al WTRs becomes more negative ( Figure 2 1 B ). This unusual trend could explain, at least partly, the high Hg sorption at pH near 3, but not the lowest Hg removal observed at pH 5. Because there is no direct correlation between the observed Hg sorption beh avior and pH on one hand, and between the measured zeta potential and pH on the other, it is likely that electrostatic attraction do es not constitute the dominant force governing Hg sorption on WTRs. These data suggest that additional forces/mechanisms are involved, and studies are needed to (i) elucidate the mechanism(s) of Hg sorption onto Al WTRs, and (ii) determine the associated
41 implications for the long term stability of formed Hg WTR complexes versus changes in key environmental parameters, if used f or soil remediation. In conclusion initial research has established the ability of WTRs to immobilize anions and toxic metals and eliminate metal toxicity in soils (Chiang et al., 2012; Dayton et al., 2003; Hovsepyan, 2008; Hovsepyan and Bonzongo, 2009; Makris and Harris, 2006; Sujana et al., 1998) Whereas the effect of pH may contribute to the efficiency of WTR sorption of dissolved ions by controlling the net charge at the surface of particles, electrostatic forces co uld constitute just one of the several mechanisms for contaminant sorption by WTRs. The long term goal of the study initiated here is to develop an efficient and cost effective remediation technique for metal contaminated soils. This can be achieved throug h a 3 step approach that includes : (i) studies of interactions between metal cations and sorption sites of the complex and highly heterogeneous drinking WTRs, (ii) field studies to lay the ground work for field application s ; and (iii) field remediation stu dies of contaminated soils. This dissertation focuses only on step one
42 Figure 2 1. Influence of solution pH on Al WTR efficiency and behavior. A) Hg sorption by Al WTR initial solution concentration of 40 mg/L. B) fluctuation on Al WTR zeta potenti al. A B
43 CHAPTER 3 IMMOBILIZATION OF MERCURY IN CONTAMINATED SOILS BY DRINKING WATER TREATMENT RESIDUALS: ROLE OF MERCURY SPECIATION AND DISSOLVED ORGANIC CARBON 3.1 Introduction Mercury (Hg) occurs naturally in the environment but its a ccelerated use and r elease since the onset of the industrial revolution has created a legacy of Hg contaminated sites worldwide For instance, the average background concentration of total Hg (THg) was estimated to be about 0.11 mg/ kg (Adriano, 2001) However, THg concentration s in soil samples collected near gold mining sites and areas surrounding chlor alkali plants range from 132 to 635 m g Hg /kg of soil (Kim et al., 2004b) and from 4.3 to 1150 mg Hg /kg of soil (Bernaus et al., 2006) respectively In the US, p ast releases of Hg from nuclear weapon production during the cold war have contaminated soil, ground and surface waters, sediments, and biota at many DOE sites. For instance, Hg co ntamination is of special concern at the Oak Ridge Site (ORS) in Tennessee where ~330 Mg of Hg were released to the headwaters of the East Fork Poplar Creek from 1950 to 1963. As a result, nearly 324 hectares around the Y 12 National Security Facility ar e heavily contaminated with Hg (Liu et al., 2006) Unfortunately, no remediation technique has so far been proved to be both cost effective and environment ally friendly for large Hg contaminated areas such as the ORS mentioned above, which awaits remed iation Adsorption and desorption processes are of special importance in Hg dynamics in soils They tend to control the fate, transport and bioavailability of metals in general (Sarkar et al., 2000; Yin et al., 1996) One of the main concerns with regard to Hg contaminated soils is the potential for Hg mobilization and contamination of surface and
44 ground waters, with implications for food chain contamination and human exposure. The US Environmental Protection Agency ( EPA) has established a maximum level of 2 Hg /L (or 2 ppb) for drinking water. To protect aquatic ecosystems, regulatory limits have also been established in terms of Hg levels allowed in effluents discharged to the environment. R ehabilitation of contami nated soils impacted by historic anthropogenic activities however, remains challenging for several reasons (Ciccu et al., 2003) as discussed in Chapter 2. For cost saving amendme nt of Hg contaminated soils with suitable sorbents that render Hg harmless is generally accepted, if accompanied by a comprehensive environmental impact study. Although i mmobilization techniques do not remove Hg from the soils, they reduce Hg mobility and bioavailability through adsorption processes. In this study, the ability of aluminum based drinking water treatment residuals ( Al WTRs ) to sorb and immobilize the mobile fractions of Hg from contaminated soils was assessed using column leaching studies 3.2. Research Motivation Preliminary studies conducted by our research group relied on laboratory contaminated sandy soils to emphasize the low retention capacity of Hg by sand particles, and therefore, the ability of Al WTRs to trap water leachable H g f rom soil solution (Hovsepyan, 2008; Hovsepyan and Bonzongo, 2009) T hese studies use d different Al WTR application rates and mixing schemes ( mix ed vs. liner [i.e. introduced as a bottom layer] ) and showed a significan t reduction of Hg leached effluent compared to the non treated control samples (Figure 3.1). In addition to the reduction of Hg leachability, bioassays using treated and non treated soils showed that both soil toxicity and Hg bioavailability to methylating microorga n isms were eliminated or significantly reduced with the incorporation Al WTRs into Hg contaminated soils. Although these
45 results for sandy soils are encouraging, soil composition and speciation of Hg in soils contaminated by historic anthropogeni c activities could affect the efficiency of Al WTRs. For instance, high levels of dissolved organic matter in soil water could outcompete Al WTR in Hg binding efficiency leading to poor Hg retention K nowledge of the association of Hg with different inorg anic and organic soil fractions may help determine if Al WTR application is appropriate Accordingly, in this study, soils with higher fractions of silt and clay than sand (loamy soils) were used without (i.e. as collected) and with Hg addition to signific antly increase the Hg load of the soil. S ampling and experimental methods follow. 3.3 Materials and Methods 3.3.1 Collection and c haracterization of aluminum based drinking water treatment residuals ( Al WTR s) The Al WTR sample used in this study was collec ted from the Manatee County Drinking Water Treatment Pl ant in Bradenton, Florida. Details on the collection and characterization of this Al WTR sample were described by Hovsepyan and Bonzongo (2009) Briefly, the stabilized Al WTR material was collected from an open air disposal site and transported to our laboratory at the Univer sity of Florida Gainesville Here Al WTRs were air dried, sieved to obtain a relatively homogen ous matrix, and then characterized for pH, electrical conductivity ( EC ) effective cation exchange capacity ( CECe ) organic carbon content, total met al concentration and specific surface area (SSA) 3.3.2 Collection and c haracterization of s oils Soils co ntaminated by historic anthropogenic activities were collected from the Oak Ridge Site (ORS) mentioned earlier. Soil samples were obtained from Big Turtle
46 Creek Park (sample site coordinates: N 35 59.44, W 084 19.031 and N 35 59.463, W 084 18.983) Onc e in the laboratory the collected soil was air dried and then ground and sieved through a 2 mm mesh to remove coarse organic and inorganic debris. After sieving the soil was mixed thoroughly and subsamples were used for determination of physic chemical v ariables The s oil particle size distribution reflects the p ercent of clay, silt and sand in a given soil and was determined in this study using the USDA Soil Survey Lab Method (USDA, 1992) The organic carbon content was measured according to (Walkley and Black, 1934) This procedure is based on the oxidation of organic carbon with potassium dichromate followed by titrati on with Fe 2+ to a burgundy endpoint (Walkley and Black, 1934) The effective cation exchange capacity (CECe), which is a measure of the quantity of sites on soil surfaces that can retain cations by electrostatic forces, was determined according to Sumner and Miller (1996) Briefly, CECe was determined using the concentrations of base cations (Ca 2+ Mg 2+ K + and Na + ) and aluminum (Al 3+ ) extracted by NH 4 Cl and measured by Inductively Coupled Plas ma Atomic Emission Spectrometry (ICP AES). The pH and the electrical conductivity (EC) of Hg contaminated soils were measured in 1:1 (mass/volume (m/v)) and 1:2 (m/v) Nanopure water suspensions respectively, using a pH meter (model 240, Corning) and EC m eter (model 1054, Markson). 3.3.3 Preparation of mercury spiked ORS soils To gain insight into the behavior of Hg in historically and freshly contaminated soils, a portion of collected ORS soil was spiked with Hg in our laboratory. The intent was to signi ficantly increase the THg concentration of the soil, with a likely increase in the water extractable and easily exchangeable fractions of soil Hg. To accomplish this,
47 a subsample of ORS soil was spiked using a flooding approach by bringing dry, sieved soi ls into co ntact with a n HgCl 2 solution in a 1:4 ratio (m/v). Produced slurries were then equilibrated for 7 days in closed plastic containers in a fume hood. Next, the mixtures were allowed to dry at room temperature inside the fume hood by opening the l id s of the containers Following this initial step, wet and dry cycles were simula ted by first saturating the Hg soil mixture with deionized water and then letting it air dry at room temperature. This experimental approach was designed to help incorporate Hg into the different soil chemical and organic fractions. After a total of 4 wet dry cycles over a period of about 2 months, dry and well homogenize d subsamples of the Hg spiked ORS soil were used to determine THg concentrations by cold vapor atomic fluores cence spectrophotometry (CV AFS) Hg distribution among the different Al WTR mineral and organic fractions was determined by the selective sequential extraction (SSE) procedure described below 3.3.4 Selective sequential extraction procedure (SSE) This tec hnique enables determination of the different fractions of Hg based on their affinity for specific groups of chemical compounds in soils. A selective sequential extraction was performed on both ORS and Hg spiked ORS soils. For these analyses, s ub samples ( ~1 g) of air dried samples were extracted sequentially using a procedure adopted from Bloom et al. (2003) A ll extractions were carried out in triplicate according the procedure described below Fraction 1 (F1) targets water soluble Hg O ne gram of soil was extracted at room temperature with 10 m L of Nanopure water for 18 hours with continuous agitation at a rate of 200 rpm. Following the agitation step, the mixture wa s centrifuged at 5,000 rpm for 30 min (Beckman J2 HS, Tritech Field Eng. Inc.) and the supernatant removed with
48 a pipette. Next, the residue was rinsed with 10 m L of Nanopure water for 5 minutes and the mixture centrifuged again at 5,000 rpm for 15 min. T he supernatant was carefully withdrawn using a 5 m L pipette and added to the first supernatant fraction. The combined supernatant was used for the analysis of THg and t he solid residue was used in the next extraction step. Besides the extraction step, the centrifugation, rinsing, and filtration steps were ident ical for all fractions (F1 to F4), except for F5 Fraction 2 (F2) targets HgO and HgSO 4 species T he residue from F1 was treated with 10 mL of a mixture of CH 3 COO H and 0.01 M HCl (pH =2). The mixture w as agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrifugation, and rinsing as described in F1. Fraction 3 (F3) targets Hg bound to organic compounds, primarily humic acids T he residue from F2 was extracted wit h 10 mL of a 1M KOH solution. The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrifugation and rinsing Fraction 4 (F4) targets amorphous Hg, organo sulfur, Hg amalgams and Hg associated with crystalline Fe Mn oxides The residue from F3 was extracted with 10 mL of 12M HNO 3 The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrif ugation and rinsing Fraction 5 (F5) targets residual species such as Hg bound t o sulfides (e.g. HgS) The residue from F4 was extracted with 10mL of aqua regia solution (HCl and HNO 3 mixture in a 3:1 ratio respectively) The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrif ugati on and rinsing
49 3.3.5 Elemental a nalysis ORS soil and Hg spiked ORS soil sample s were analyzed for total metal concentration s after hot acid digestion. For this analysis, about 1 g of dry of soil sample was digested at 110C with 30 m L of HNO 3 /H 2 SO 4 mixtur e (7:3, v/v) in a closed Teflon vessel. The mixture was allowed to cool and afterwards diluted to 50 m L with Nanopure water. The digested and diluted solution s were analyzed for T Hg concentration s by cold vapor atomic fluorescence spectrometry (CV AFS). O ther metals such as Pb, C d, Cr, Zn, Cu, Al, Fe and Ca were analyzed by i nductively coupled plasma atomic emission spectroscopy ( ICP AES ) On the aqueous fractions obtained from each step of the SSE procedure, T Hg was analyzed following overnight treatmen t with bromine mono chloride an oxidizing agent made of the mixture of KBr and BrO 3 dissolved in concent rated HCl. At the end of this room temperature oxidation process, samples were pre reduced by addition of 30% NH 2 OH HCl ) prior to SnCl 2 reduction of io nic Hg to Hg 0 and detection by CV AFS according to US EPA method 1631 3.3.6 Column l eaching e xperiments Pore volume To determine the porosity of tested soils, about 4.89 g of air dried soils had Nanopure water added until saturation was reached. The vo lume of dry soil used was determined using a Picnometer. Total porosity was calculated to be 3.3 cm 3 from Equation 1 (Wasay et al., 2001) (3 1 ) where, V (mL) is the volume of water added to attain soil saturation (V=2 cm 3 ), and V s is volume of soil used ( V s = 1 .9373 cm 3 ).
50 To determine the pore volume of the soil, a clear PVC column with 1.3 cm internal diameter wa s packed with ~9 grams of soil to achieve a soil height of 5.15 cm. Pore volume was calculated from Equation 2, (3 2 ) where, V p (mL or cm 3 ) is the pore volume of the packed soil inside the column (V p =3.3 cm 3 ), r is the radius of the column (r=0.635 cm), h is the height occupied by soil in the column ( h= 5.15 cm) and f is the total porosity of the soil ( f= 0.508). Synthetic precipitation leaching procedure (SPLP) solution A SPLP mother solution was prepared by mixi ng 1% (v/v) concentrated nitric acid metal grade/Nanopure water solution with 1% (v/v) concentrated sulfuric acid trace metal grade/Nanopure water solution in a 2:3 (v/v) ratio respectively. The SPLP extracting solution (EPA method 1312) was prepared by diluting 0.4 ml of the mother solution in a 2.0 mL volumetric flask with Nanopure water to reach a final pH of 4.2 0.02 (Townsend et al., 2006) The mother solution was refrigerated and used to prepare SPLP solutions used during all column leaching experiments. Experimental design Soil columns were custom made out of clear PVC ( 20 cm long and 1.3 cm wide) equipped with a drainage hole at the base and covered with wool netting to prevent soil loss and to minimize dead end volume. Two different sets of column leaching experiments were performed for this study. For the first experi ment the columns were packed with 9 g of soil (controls), or 9 g of soil plus Al WTRs added at application rat es of 2% and 20% (mass Al WTRs/mass soil for treated soils). These application rates of 2 and 20% were selected as low and high end members base d on preliminary results, although practically, an application rate of 20% might not be
51 considered cost effective in in situ soil remediation. In columns with 2% and 20 % treatments, Al WTRs was uniformly mixed with the soil and then added to the column. On ce packed, s oil columns were held vertical on a wo oden rack placed on the lab bench. Control (no WTR added) and Al WTR treated soil columns were then leached by gravity with either a SPLP solution (pH=4.2) or a dissolved organic matter solution prepared fr om Suwannee River water, following the experimental design shown in Figure 3 1. All treatments were run in triplicate for a total of 18 columns and 6 sets of experiments. For the second experiment, columns were packed with 9 g of soil (controls) or 9 g of soil plus Al WTRs added at application rat es of 2% and 5% (mass Al WTRs/mass soil). Columns with 2% and 5% treatment included two incorporation schemes: (i) uniform mix ing of Al WTRs with the soil (2% Al WTR and 5% Al WTR) and (ii) Al WTRs added as a bo ttom l ayer at the base of the column (5% Al WTR liner). Once packed, s oil columns were held vertical on a wo oden rack placed on the lab bench. Control and Al WTR treated soil columns were leached by gravity with either a SPLP solution (pH= 4.2) or a soluti on with increasing concentrations of a dissolved organic matter (0%, 10%, 25%, 50%, 75%), following the experimental design shown in Figure 3 2. All treatments were run in triplicate for a total of 24 columns and 8 experimental variations. Leaching procedu re SPLP solution was used to simulate the effect of soil leaching that might occur because of acid rain (pH~4.3), whereas Suwannee River water was used for its rich organic content to simulate the effect that organic rich soil water might have on the stab ility of Hg WTR complexes formed in treated soils. Columns were
52 leached daily with 2 pore volumes (6.6 mL). Leachate samples were collected in 25 mL acid washed scintillation vials, preserved HCl and refrigerate d at 4C until digestion and analysis for THg as described earlier in section Differences between Hg concentrations in leachates from different treatments were evaluated using the Independent Sample t test with SPSS statistical so ftware (vs. 11.0). Tests were performed at a confidence level of 95%. 3.4 Results and Discussion 3.4.1 Characterization of Al WTR Data on physico chemical characterization and elemental composition of the Al WTR s used are presented in Tables 3 1 and 3 2 Q ualitatively, the Al WTR used had an elemental composition similar to those reported by several previous studies and summarized in a recent review by Ippolito et al. (2011). T his same Al WTR was used in previous studies in our laboratory, and a thorough de scription/discussion of its characteristics can be found in Hovsepyan and Bonzongo (2009) 3.4.2 Physico chemical analysis of ORS soil Data obtained from the physic chemica l analysis of ORS soil are presented in Table 3 3. Using the United States Department of Agriculture definition it can be with a ratio of sand/silt/clay fractions being about 40/40/20% (Schwedt, 2001) Fine particles in the soil (silt and clay) make loamy soils efficient at retaining nutrients but pollutants as well. In such soils, the different particle sizes allow for the formation of pockets that increase water holding capacity and air penetration. ORS soil porosity was 50.81.3 which is characteristic of soils with medium to fine texture (40 60%). On the acidity scale, this soil is considered neutral with an average
53 measured pH of 7.6 0.03. The high value of CEC e of this soil (24.060.38 cmol/kg) is consistent with the measured organic carbon content and neutral pH. The ORS soil showed concentrations of Hg, Al, Fe and Mn well ab ove the US EPA soil clean up target levels for residential and industrial areas (Table 3 4). In addition, the distribution of Hg in the different soil fractions in the native ORS (NSS = non spiked soil) and laboratory Hg spiked ORS (HSS=Hg Spiked Soil) is presented in Table 3 5. Overall, NSS had about 99% of the Hg bound to the mineral lattice (oxide fraction) and residual fraction (Hg bound to sulfides), whereas HSS had a higher T Hg, with about 70% of the total concentration found in the water soluble (F1 ) and exchangeable (F2) fractions. Therefore, HSS is expected to easily leach out Hg in column experiments. 3.4.3 Mercury leached from ORS loamy soil ORS soils amended with Al WTR at different application rates showed consistently lower concentrations of Hg in SPLP leachates compared to control soil samples with no Al WTR additions (Figure 3 4). The sum of water soluble, exchangeable and Hg humic fractions comprised <1% of THg concentration, corresponding to a mass of about ~3 to 4 g (Table 3 5). These observations point to the high affinity of Hg for this ORS loamy soil, in which Hg is strongly associated with soil fractions other than F1, F2, and F3. The total mass of Hg extracted by fractions F1, F2, and F3 was released after lea ching the equivalent of 4 pore volumes with SPLP solution in both the control soil and soil treated with 2% (m/m) Al WTR. In contrast, ORS soil mixed with Al WTR at the application rate of 20% required a total of 12 pore volumes to extract and release the same amount of Hg. Hg bound to Fe / Mn (hydr) oxides was only partially extracted in WTR amended soils.
54 Al WTR has been reported to have a very high sorption capacity for Hg, reaching up to 70 mg Hg/g of Al WTR (Hovsepyan and Bonzongo, 2009) Theoretical calculations suggest that Al WTR at 2% and 20% application rates should be able to sorb all the Hg present in the soils used It is also known that at pH values at or bel ow that of the SPLP solution used (pH 4.22), most metals are soluble because of H + competition at the exchange sites occupied by metal cations (Dong et al., 2009) This competition decreases metal sorption at acidic pH values, and increases metal mobilization through soils. Furthermore, because 80 % of Hg in ORS soil was bound to the Fe/Mn oxides fraction (Table 3 5 and Table 3 6) an acid and redox sensitive fraction a decrease in soil pH on one hand and the development of anoxic conditions on the other would release Hg bound to this fraction. Thu s, increased and continuous Hg leaching after reaching the plateau observed in these experiments can be attributed to both mobilization of pH impacted Hg complexes and dissolution of (hydr)oxide minerals On the other hand, preliminary studies focusing on Hg sorption by Al WTR have shown high sorption capacities even under acidic conditions (Hovsepyan and Bonzongo, 2009) Therefore, it is likely that the biochemical re duction of oxide minerals and possibly soil compaction and channeling formed over time during soil leaching experiments do contribute to the observed delayed/ reduced efficiency of Al WTR sorption capacity This can be explained by the slow and temporal rel ease of Hg initially bound to oxide minerals as the latter undergo reduction and subsequent dissolution, releasing Hg to solution. Additionally, soil compaction and channeling would limit the contact of the formed soluble Hg in Al WTR in amended soils res ulting in extended Hg bleeding Accordingly, although changes in pH of the soil system could have somehow
55 influenced the efficiency of the Al WTR treatments, it is likely that redox reactions and the lack of physical contact between soluble Hg species and Al WTR particles played a n important role with regard to the observed Hg leaching trends This is especially true after leaching with 26 pore volumes by which time the soils become highly compacted inside the columns, resulting in reduced porosity develo pment of preferential flow paths and anoxic conditions This conclusion is supported by the SSE data (Tables 3 5 and 3 6), which show that prior to leaching, ~80% of the THg concentration in tested soil is bound to Fe/Mn (hydr)oxides, and after leaching ex periments, a near total depletion of this specific Hg fraction was observed. Although leachates collected throughout these experiments, from both control and amended soils, showed Hg concentrations that are lower than the Florida leachate criteria for grou ndwater (2.1 ppm), concentrations of Hg leached from control columns were significantly higher (p<0.5) than the concentration leached from 2% and 20% Al WTR treatments. At the end of the leaching events (total of 82 pore volumes), the amount of Hg retained in non amended soils corresponded to 63% of initial total Hg mass, as compared to 73% retention by soils amended with 2% Al WTR and 85% retention in soils amended with 20% Al WTR (Table 3 6). However, calculated Hg retention potential after leaching with the first 26 pore volumes, before soil compaction and channeling became significant, was different. The 2% Al WTR treatment yielded a calculated retention of 87% of total Hg in soil and the 20% Al WTR treatment yielded a calculated retention of 99% of tota l Hg compared to only 80% retention of Hg in non treated soils. Although the ORS soil has an initial high Hg retention capacity, treatment with Al WTR improved Hg retention by about 10 to 20%.
56 The efficiency of Al WTR amended soils to retain Hg when leache d with Suwannee River water (TOC= 53.3 mg/L and pH 4.2) was significantly reduced, compared to Hg leached with SPLP solution. All samples showed release of Hg from pore volumes 10 16, and subsequent increase d every 10 pore volumes (Figure 3 5). At the end of the leaching procedure with Suwannee River water (82 pore volumes), the amount of Hg retained in non amended soils corresponded to 53% of initial total Hg mass in the column in contrast to 58% and 59% retention by soils amended with 2% Al WTR and 20% Al WTR, respectively (Table 3 7). The difference between control and Al WTR treated samples was not significant (p>0.05). Interactions of Hg with dissolved organic matter (DOM) can lead to the formation of Hg DOM complexes, which can be highly mobile (Adriano, 2001; Allard and Arsenie, 1991; Skyllberg et al., 2006; Zhou and Haynes, 2010) Suwannee River water (SRW) has a very high DOC concentration, composed of a wide variety of organic compounds and containing a significa nt fraction as humic and fulvic acids. In addition, SRW has an acidic pH. Therefore, an increase in Hg released from soils when leached with SRW compared to SPLP solution could be explained simply by the well established high Hg DOM affinity. A study by Yang et al (2008) showed that the addition of different types of dissolved organic compounds to soils decrease s Hg sorption and depending on soil type, sorption decrease could vary fro m 23 to 43% of the initial sorbed mass. This study showed the low capacity of Al WTR to retain Hg when exposed to DOC rich water with low pH. T his is however, a rather extreme example as DOC concentrations in pore waters rarely exceed 50 mg C/L as test ed here. At the same time, the results suggest that the efficiency of Hg immobilization in Al WTR treated soils could be negatively impacted under flood
57 conditions with DOC rich water. Because the effect of pH on leaching can be negated by blending treatme nt techniques e.g. combining Al WTR with lime, studies should be carried out on how to counter the action of dissolve d organic matter. 3.4.4 Mercury leaching from Hg spiked ORS soils Mercury spiked ORS soil saw its total Hg concentration increase to an a verage of 2567.4 78.5 mg of Hg/kg of soil. Selective sequential extraction analysis showed the following distribution of the main chemical fractions: 42% in the water soluble fraction, 29% in the exchangeable fraction, 5% in the organo complexed fraction 6% bound to the Fe/Mn oxide fraction and 2% in the residual fraction (Table 3 5). Spiking of ORS soil with Hg increased concentration of leachable Hg. I n this manipulated sample, >70% of Hg was extracted from the highly mobile and relatively mobile Hg fr actions. And based on the maximum retention capacity of Al WTR for Hg as determined by Hovsepyan and Bonzongo (2009) soil amendment with 2% (0.18 g Al WTR / 8.82 g soil ) and 5% (0.45 Al WTR g / 8.55 g soil ) Al WTR should theoretically provide retention of up to 14.2 mg and 35.5 mg of Hg, respectively. Accordingly, the 2% application rate should be able to control Hg leaching from the water soluble fraction (F1) of this soil (11. 55 mg), whereas the 5% application rate should stop all leachable Hg. At the end of the leaching experiments (40 pore volumes), control Hg spiked ORS soils leached with SPLP consistently released Hg with retention of 42% of the mobile fraction. In Al WTR treated soils, 75% of the mobile fraction was retained in soils with a 2% Al WTR application rate. For soils treated with 5% Al WTR, about 85% of Hg in the mobile fraction was retained in the uniformly mixed incorporation scheme versus 90% retention in co lumns containing 5% Al WTR used as bottom liner (Table 3 9).
58 Concentrations of Hg leached from control columns were significantly greater (p<0.5) than Hg concentration leached from all Al WTR treated soils. For mercury spiked soils leached with increasing concentrations of SRW (Table 3 11), controls released a higher concentration of Hg, with Hg retention of 37% of the mobile Hg fraction after 54 pore volumes. In contrast, 71% retention of the same fraction was retained in soils amended with 2% Al WTR. S oi ls containing 5% Al WTR showed retention of about 78%, regardless of the incorporation scheme (Table 3 10). Concentrations of Hg leached from control columns were significantly higher (p<0.5) than Hg concentration leached from 2% and 5% Al WTR treated soil s, but no significant difference (p>0.5) was observed between soils treated with 5% Al WTR uniformly mixe d with the soil and 5% Al WTR used as a liner. Finally, although Al WTR applied as liner might play an important role by eliminating the negative effe cts of compaction and channeling and increasing chances for physical contact between soluble Hg and Al WTR, differences in Hg retention rates leached with solutions containing > 26 mg C/L might not be worthwhile if the cost involved in excavation to apply t he Al WTR liner is considered 3.5 Conclusions on Column Leaching Experiments Among all treatments, non treated soils (0% Al WTR) consistently displayed higher Hg release from soils leached with both synthetic precipitation (pH=4.22 0.05) and increasing concentrations of Suwannee River water (SRW) with TOC=53.3 mg/L and pH=4.20 0.05. Increased Hg retention was observed for ORS soils treated with Al WTR in the order NTS < 2% Al WTR < 20% Al WTR for soils leached with SPLP solution. Although ORS soils tr eated with Al WTR showed a slight improvement in Hg retention compared to the control, when leached with Suwannee River water, there was no significant difference between 2% Al WTR and 20% Al WTR treatments.
59 Mercury retention was improved in ORS soils as w ell as Hg spiked ORS soils by application of Al WTR treatments However, Al WTR treatments seem to be more relevant for soils with a high concentration of Hg in mobile and bioavailable fractions. Because of the high affinity of Hg for dissolved organic ma tter, the different Al WTR treatments are more efficient at sorbing Hg from soil when leached with an acid solution (SPLP) than when leached with a solution with high concentration of dissolved organic matter (TOC >13 mg/L). Development of anoxic condition s and both c ompaction and channel formation in used soils led to the co occurrence of increased Hg release and reduce d contact of Hg with Al WTR resulting in decreased Al WTR sorption efficiency. Application of Al WTR as a liner is suggested for fine tex ture soils with limited water flow, to increase Hg WTR contact and thus sorption. In general the le a chability of Hg in Al WTR treated Hg contaminated soils is significantly lower than that of non treated soils However, this significance difference disapp ears when soils are leached with solutions containing high DOC concentrations.
60 Table 3 1. Physicochemical characterization of Al WTR sample collected from the Bradenton Drinking Water Treatment Facility (Florida, USA) and used in this study. Adapted from Hovsepyan and Bonzongo (2009) Parameter Mean Value a b Average values from previous Al WTR studies Units pH c 5.6 0.01 6.5 0.3 e EC d 0.36 0.01 1.6 0.9 e dS/m e CEC e 45.8 0.09 cmol/kg SSA N 2 48 0.3 36 e m 2 /g SSA CO 2 120 0.3 104.9 e m 2 /g Organic Carbon 12.7 0.3 % a All the values are mean of triplicates; b values represent standard errors of the mean; c at soil water ratio of 1:1; d electric conductivity at a soil water ratio of 1:2 (m/v); e effective cation exchange capacity ; e from Ippolito et al. (2011) Table 3 2. Elemental analysis of the Al WTR sample used in this study. Adapted from Hovsepyan and Bonzongo (2009) Element Mean Value b, c Average values from previous Al WTR studies d Units Regulatory Limit f (mg/kg) Al 73.8 3.2 119 24 g/kg nd Fe 3.7 0.1 37 20 g/kg nd Ca 2.26 0.5 10 4 g/kg nd As 8.01 1.1 Nd e mg/kg 41 Cu 141 3.4 624 581 mg/kg 1500 Cr 81.1 1.3 20 7 mg/kg 1200 Hg 0.02 0.003 0.46 mg/kg 17 Pb 1.99 0.4 22 12 mg/kg 300 Zn 14.37 1.3 98 31 mg/kg 2800 Al OE a 66 9 g/kg Fe OE 14 8 g/kg a OE stands for ammonium oxalate extractable; b all the val ues are mean of triplicates; c values represent standard errors of the mean; d from Ippolito et al. (2011) ; e Not defined ; f US EPA 40 CFR Part 503, pollutant limits for meeting land exception al quality criteria
61 Table 3 3. Physico chemical variables for Hg contaminated soil collected from Oak Ridge Site (ORS), Tennessee, USA. Parameter Mean Value Units Sand 43 0.5 d % Silt 37 1 d % Clay 20 0.5 d % Density 2.52 0.2 e g/cm3 Porosity 50.8 1.3 e % pH a 7.60 0.03 e TOC b 3.66 0.76 e % eCEC c 24.06 0.38 d Cmolc/kg a At a soil water ratio of 1:1 (m/v); b Total organic carbon; c Effective cation exchange capacity; d Mean S.D. (n=2), e Mean S.D. (n=3) Table 3 4. Elemental analysis of Hg contaminated soil collected from Oak Ridge Site (ORS), Tennessee, USA. Elements Mean concentration in mg/kg SCTL/Direct Exposure c (mg/kg) Residential Industrial Mercury (Hg) 54.71 4.3 3 17 Lead (Pb)
62 Table 3 5. Concentrations (mg/kg ) and distribution (%) of Hg, in the different chemical fractions of ORS (NSS), and Hg spiked ORS (HSS) soils; Oak Ridge, Tennessee, USA. Sample Fraction Concentration mg/kg a % of total ORS soil (NSS) F1: Water soluble 0.10 0.02 0.21 F2: Exchangeab le 0.06 0.00 0.11 F3: Organo complexed 0.18 0.02 0.37 F4: Fe/Mn oxides 39.21 1.70 79.61 F5: Residual 9.71 0.65 19.71 Recovery from SSE 49.25 1.07 90.01 Total Hg in soil 54.71 4.31 F1: Water soluble 1085.4 76.45 42.28 F2: Exchangeable 763.55 82.22 28.69 Hg spiked ORS soil (HSS) F3: Organo complexed 129.75 33.18 5.05 F4: Fe/Mn oxides 161.25 43.34 6.28 F5: Residual 57.98 11.30 2.26 Recovery from SSE 2170.93 37.72 84.56 Total Hg in soil 2567.40 78.5 a M ean 1SD ( n=3 )
63 Table 3 6. Mass of mercury (in either g or mg) in each chemical fraction of ORS soils (NNS) and Hg spiked ORS soils (HSS) used in column leaching experiments. Sample Fraction/Description Typical compounds Average Hg mass per 9 g of soil a Sequential addition of Hg mass Mass units ORS soil (NSS) F1: Water soluble HgCl 2 1.11 g F2: Exchangeable HgO and HgSO 4 0.62 1.73 g F3: Organo complexed Hg humic acids 1.98 3.71 g F4: Fe/Mn oxides Mineral lattice, Hg 0 429.88 433.59 g F5: Res idual HgS, m HgS, HgAu 106.41 540 g Hg spiked ORS soil (HSS) F1: Water soluble HgCl 2 11.55 m g F2: Exchangeable HgO and HgSO 4 7.84 19.39 m g F3: Organo complexed Hg humic acids 1.38 20.77 m g F4: Fe/Mn oxides Mineral lattice, Hg 0 1.72 22.48 m g F5: Residual HgS, m HgS, HgAu 0.62 23.10 m g a Percent of each chemical fraction in Table 3 3 was used to calculate corresponding Hg mass in each fraction of soil used in the differ ent column leaching experiments
64 Figure 3 1. Efficiency of Al WTR in immobilizing Hg in a lab contaminated sandy soil leached with synthetic precipitation leaching procedure (SPLP) solution (Hovse pyan, 2008)
65 Figure 3 2 Experimental design for the first set of column leaching experiments using Oak Ridge Site (ORS) soils with a starting THg concentration of 54.71 4.3 mg Hg/kg soil. The experimental design included 6 columns packed with non treated soils (i.e., control without Al WTR addition), 6 columns pa cked with 2% Al WTR treated soils and 6 columns packed with 20% Al WTR treated soil. Soils + Al WTR in each column added to a total mass of 9 g, with a THg mass of 23.1 mg. For each treatment 3 columns were leached with the SPLP solution (pH=4.2) and the o ther 3 with Suwannee River water containing a DOC concentration of 53.3 mg C/L). A total of 18 columns were prepared and used in leaching experiments.
66 Figure 3 3 Experimental design for the second set of column leaching experiments using Hg spiked Oa k Ridge Site (ORS) soil spiked to increase THg concentration from 54.71 mg Hg/kg soil to 2567.4 mg Hg/kg soil. The study included 6 columns packed with non treated soils (control, no WTR added), 6 columns packed with 2% Al WTR treated soils, 6 columns pack ed with 5% Al WTR treated soils, and 6 columns packed with 5% Al WTR added as bottom liner. Soils + Al WTR were added in each column to a final total mass of 9g corresponding to a total mass of 23.1 mg Hg. For each treatment, 3 columns were leached with SP LP solution and another 3 with organic rich solutions of increasing concentrations prepared for Suwannee River water with an initial concentration of 53.3 mg C/L.
67 Figure 3 4 Hg leached from Oak Ridge Site (ORS) soil with an initial THg concentratio n of 54.71 m g of Hg/kg soil. Soil columns were leached with synthetic precipitation leaching p rocedure (SPLP) solution (pH=4.22), NTS: non t reated or control soil; 2% and 20% WTR represent ORS soil treated Al WTRs at 2 % and 20% application rates.
68 Table 3 7. Mass balance of Hg in column leaching studies using Oak Ridge Site (ORS) soil. The initial total Hg mass in each column was ~0.54 mg, correspon ding to a THg concentration of 54.71 m g of Hg/kg soil. Soil remaining in columns was analyzed after leaching with 82 pore volumes of SPLP solution. Control Soil + 0% Al WTR Soil + 2% Al WTR Soil + 20% Al WTR THg in leachate ( g) 202.3 147.9 82.7 THg left in column ( g) 337.7 392.1 457.3 Retention of Mobile fraction (in %) 63.0 73.0 85.0
69 Figure 3 5 Hg leached from Oak Ridge Site (ORS) soil with initial THg concentration of 54.71 m g of Hg/kg soil using Suwa nnee River water with a DOC concentration of 53.3 mg C/L and pH=4.2. NTS: ORS soil without Al WTR; and 2% and 20% Al WTR SRW are ORS soils treated with Al WTR and leached with Suwannee River water.
70 Table 3 8. Mass balance of Hg in Column leaching studies using Oak Ridge soil. The initial total Hg (THg) mass in each column was 0.54 mg, correspo nding to a concentration of 54.71 mg of Hg/kg soil. Soil remaining in columns was analyzed after leaching with 82 pore volumes using SRW (DOC= 53.3 mg C /L ) Control (0% Al WTR) Soil + 2% Al WTR Soil + 20% Al WTR THg in leachate ( g) 252.4 228.3 221.3 THg left in column ( g) 287.6 311.7 318.7 Percent of retention (%) 53 58 59
71 Figure 3 6 Hg leached from Hg spiked Oak Ridge Site (ORS) soil with initial THg concentration of 2567.40 of Hg/kg soil (corresponding to a total mass of ~23 mg Hg per column) using synthet ic precipitation leaching procedure (SPL P) solution pH=4.2 when treated with different application rates and schemes of Al WTR. NTS: non t reated soil; WTR: soil treated with aluminum based water t reatmen t residual; SPLP: synthetic precipitation leaching p rocedure solution.
72 Table 3 9. Mass ba lance of Hg in Column leaching studies using Oak Ridge Site (ORS) soil. The initial total Hg (THg) mass in each column was 23 m g, corresponding to a concentration of 2567.4 mg of Hg/kg soil. Soil remaining in columns was analyzed after leaching with 40 por e volumes using SPLP solution (pH=4.22). Control 0% Al WTR Soil + 2% Al WTR Soil + 5% Al WTR Soil + 5% Al WTR liner THg in leachate ( m g) 14.7 8.5 6.6 5.6 THg left in column ( m g) 8.3 14.5 16.4 17.4 Retention of mobile fraction (%) a 42 75 85 90 a Mobile fraction = water soluble fraction (F1) + easily exchangeable fraction (F2)
73 Figure 3 7. Hg leached from Hg spiked Oak Ridge Site (ORS) soil with an initial THg concentration of 2567.40 mg Hg/kg of soil, corresponding to a mass of ~23 mg Hg/column. Besides the controls (i.e. soil with no Al WTR addition), soils were treated with Al WTR at 2% and 5% application rates. The 5% app lication rate was used either as a bottom layer (liner approach) or thoroughly mixed with soil. Columns were leached sequentially with solutions of different concentrations of dissolved organic carbon prepared using the Suwannee River water diluted in DI w ater (0%= p ure DI water; 10% = DI solution containing 5.33 mg C/L; 25% = 13.33 mg C/L; 50% = 26.65 mg C/L; and 75% = 39.98 mg C/L) 5.33 mg/L 0 mg/L 13.33 mg/L 26.65 mg/L 39.98 mg/L
74 Table 3 10. Mass balance of Hg in column leaching studies using Oak Ridge Site (ORS) soil. The initial total Hg mass in ea ch column was 23 m g, corresponding to a concentration of 2567.4 mg of Hg/kg soil. Soil remaining in columns was analyzed after leaching with 52 pore volumes using Suwannee River water (SRW) with a DOC concentration of 53.3 mg/L and a pH of 4.2. Control 0% Al WTR Soil + 2% Al WTR mixed Soil + 5% Al WTR mixed Soil + 5% Al WTR as bottom liner THg in leachate ( m g) 16.01 9.4 7.98 7.84 THg left in column ( m g) 7.09 14.8 15.12 15.26 Retention of Mobile fraction (%) a 37 71 78 79 a Mobile fraction = water solubl e fraction (F1) + easily exchangeable fraction (F2)
75 Table 3 11. Percent of mobile fractions of Hg r etained in Hg spiked ORS soil (HSS) amended with Al WTR at 2 % and 5% application rates and using two di fferent incorporation schemes ( uniformly mixed and bottom layer/liner) when leached with water containing increasing DOC concentration added as Suwannee River water (SRW) diluted in DI water # of PV a DOC b (mg/L) Soil with no Al WTR c Soil+ 2% Al WTR c Soil+ 5% Al WTR c Soil+ 5% Al WTR as liner c 20 0.00 59.43 79.69 87.58 88.04 30 5.33 57.47 78.14 86.70 87.11 40 13.33 55.46 76.75 86.08 86.44 46 26.65 48.35 74 .74 83.51 83.66 52 39.98 36.55 70.62 77.94 78.66 a PV=pore volume, b DOC=dissolved organic carbon c Percent Hg retained
76 CHAPTER 4 MECHANISMS OF MERCURY IMMOBILIZATION BY A LUMINUM BASED DRI NKING WATER TREATMENT RESIDUALS: IMPLICATIONS FOR SOIL REMEDIA TION 4.1 Introduction effects on the environment and human health have placed this trace metal in the top tier of the most toxic contaminants. Different technologies are currentl y being used for the removal of Hg from contaminated solid matrices and include soil washing, volatilization, excavation and phytoremediation, as discussed in Chapter 2. Yet, adsorption technologies have recently received heightened attention because of e as e of application and relatively low cost (Kim et al., 2003; Wang and Mulligan, 2009) M aterials that are efficient with regard to Hg binding generally carry sulfur nitrogen and oxygen containing functional g roups (Kim et al., 2003; Wang and Mulligan, 2009) F ine particle materials with large surface area s, such as oxides, oxyhydroxides and layer silicates are among the principal sorbents for metals such as Hg Understanding t he mechanisms of Hg binding to different sites in soils is key to predicting Hg mobility and long term environmental fate P revious studies show aluminum and iron (hydr)oxides are efficient sinks for many contaminants including Hg (Axe and Trivedi, 2002; Behra et al., 2001; Bonnissel Gissinger et al., 1999; Ehrhardt et al., 2000; Jeong et al., 2007; Kim et al., 2004a; Kim et al., 2004b; Sarkar et al., 1999) For instance, laboratory s tudies investigating the interactions be tween Hg and pure minerals such as Fe and Al (hydr)oxides were carried out by Kim et al. (2004a; Kim et al., 2004b) Their findings demonstrated that the presence of sul f ate in solution significantly increase d Hg(II ) sorption, whereas the occurrence of chloride resulted in reduced Hg(II) sorption from solution
77 C hemical characterization of Al WTRs revealed the presence of O, S and N among the prevalent elements in this waste material (Hovsepyan and Bonzongo, 200 9; Makris et al., 2004) Aluminum oxides in these waste materials appear to present primarily in amorphous form s which are not detectable by XRD spectroscopy (Hovsepyan and Bonzongo, 2009; Makris et al., 2004) This study investigates the sorption mechanisms involved in the immobilization of Hg when Al WTR is used as sorbent. A combination of analytical techniques including c hemical fractionation and spectroscopic methods (e.g ., X ra y photoelectron spectroscopy (XPS), and X ray diffraction (XRD)), were used to gain insight into the binding mechanisms of Hg onto the Al WTR surface. 4.2. Materials and Methods 4.2.1 Collection and characterization of Al WTR sample The Al WTR sample used was collected from the Manatee County Drinking Water Treatment Plant in Bradenton, Florida, USA. In this water treatment plant, Al WTRs are generated by addition of alum and copolymers of sodium acrylate and acrylamide (Makris et al., 2005) Sampled Al WTR was collected from an open air disposal site and transported to the University of Florida in Gainesville, w h ere it was further air dried and sieved to obtain relatively homogenous material. Al WTRs were then characterized by d etermining key variables such as pH, electrical conductivity (EC) cation exchange capacity (eCEC ), organic carbon content, total metal concentrations, and specific surface area (SSA) measured using both nitrogen (SSA N 2 at 77 K) and carbon dioxide (SSA CO 2 at 273 K) sorption methods (Hovsepyan and B onzongo, 2009)
78 4.2.2 Preparation of fresh and aged mercury spiked Al WTRs Hg spiked Al WTRs were prepared by using a fl ooding approach conducted by bringing dry Al WTR into contact with a HgCl 2 solution in a 1:4 ratio (mass/volume). Samples were prepared in triplicate. Produced slurries were equilibrated for 7 days in closed containers in a fume hood. Next, the mixtures w ere allowed to dry at room temperature by opening the l i d s of the containers for another 7 days. Following this initial step of Hg incorporation into Al WTRs, a series of wet and dry cycles was started by first saturating the Hg (Al WTRs) mixtures with dei onized water for 5 days and then air drying for another 5 days at room temperature. This experimental approach was designed with the intention of forc ing Hg into the micro pores of the Al WTRs as surface sites became progressively saturated over time. Afte r a total of 12 wet dry cycles (~4 months), aliquots of the Hg spiked Al WTRs materials were used to determine THg concentration by ICP AES and Hg distribution among the different Al WTRs mineral and organic fractions by use of a selective sequential extra ction (SSE) procedure. The remaining Hg spiked Al WTR samples were left to age untouched in covered containers for a period of four years. Fresh Hg Al WTR spiked samples were prepared in a similar way to the aged samples H owever, these samples were only e xposed to 4 wet dry cycles in a period of about 2 months. For these experiments and analyses, control samples run in parallel were represented by raw Al WTRs and no Hg addition. After the Hg incorporation period, samples were analyzed for THg concentration s and Hg chemical fractionation. Hg spiked Al WTR samples were analyzed The solid phase analyses of the samples were performed using by Scanning Electron Microscopy Energy Dispersive X Ray Spectroscopy (SEM EDS), X ray Photoelectron Spectroscopy (XPS), a nd chemical fractionation
79 4.2.3 S ample analysis by s elective sequential extraction (SSE) procedure Two approaches were used for the selective sequential extraction of Hg in Al WTR samples. For aged Hg (Al WTR) samples a full SSE was performed (Hovsepyan 2008) One gram of air dried Al WTRs sample (e.g. Hg spiked Al WTRs and control Al WTRs with no Hg added) was extracted sequentially using a method adapted from the technique proposed by Tessier et al. (1979) The f ollowing fractions were targeted and all extractions were carried out in triplicate. Fraction 1 (F1) targets water soluble Hg. One gram of soil was extracted at room temperature with 8 ml of Nanopure water for 3 hours with continuous agitation at a rate o f 200 rpm. Following the agitation step, the mixture was centrifuged at 10,000 rpm for 30 min (Beckman J2 HS, Tritech Field Eng. Inc.) and the supernatant was removed with a pipette. Next, the residue was rinsed with 8 ml of Nanopure water for 5 minutes a nd the mixture was centrifuged again at 10,000 rpm for 30 min. The supernatant was carefully withdrawn and added to the first supernatant fraction. The combined supernatant was then filtered (0.45 Hg. The solid residu e was then used in the next extraction step. Besides the extraction step, the centrifugation, rinsing, and filtration steps were identical for all fractions (F1 to F5), except for F6. Fraction 2 (F2) targets exchangeable Hg. The residu e from F1 was treated with 8 ml of magnesium chloride 1M (MgCl 2 pH 7). The mixture was agitated at 200 rpm at room temperature for 1 hour, prior to centrifugation, rinsing, and filtration as described in F1. Fraction 3 (F3) targets Hg bound to carbonates The residue from F2 was extracted with 8 m l of 1M sodium acetate (NaOAc) solution with a pH adjusted to 5.0
80 with acetic acid (HOAc). The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 5 hours, prior to centrifugation, rinsing, and filtration. Fra ction 4 (F4) targets Hg bound to Fe Mn oxides The residue from F3 was extracted with 20 ml of 0.04 M hydroxylamine hydrochloride (NH 2 OH.HCl) in 25% (v/v) HOAc in containers placed in a water bath for 6 hours at 96 3C, with occasional agitation by hand. Upon cooling, the mixture was centrifuged, rinsed, and the supernatant filtered as described in F1. Fraction 5 (F5) targets Hg bound to organic matter The residue from F4 was extracted with 3 m l of 0.02 M nitric acid (HNO 3 ) and 5 m l of 30% hydrogen perox ide (H 2 O 2 ) that had pH adjusted to 2 with HNO 3 The mixture was heated to 85 2C in a water bath for 3 hours with intermittent agitation by hand. An additional 3 m l of 30% H 2 O 2 (pH adjusted to 2 with HNO 3 ) was added and the mixture heated for an addition al 3 hours at 85 2C. Upon cooling, and to prevent the adsorption of extracted Hg onto oxidized Al WTR components, 5 m l of 3.2 M ammonium acetate (NH 4 OAc) in 20% (v/v) HNO 3 was added and the sample was diluted to 20 m l and agitated continuously for 30 mi n utes on an orbital shaker at 200 rpm. These steps were then followed by centrifugation and filtration as described in F1. Fraction 6 (F6) targets residual species The residual fraction was quantified by subtracting the sum of Hg obtained in fractions F1 5 from the amount of Hg obtained in the total Hg analysis. In addition to the above SSE method developed initially for the fractionation of solid phase me tals in general (Tessier et al., 1979) more recent research has produced SSE techniques that are specific to Hg, and the method proposed by Bloom et al
81 (2003) is widely used because it emphasizes Hg fractions prone to methylation. A ccordingly, s ub samples (~1 g) of air dried Al WTR were also analyzed following a modification of Bloom et al. (2003) Fraction 1 (F1) targets water soluble Hg One gra m of Al WTR was extracted at room temperature with 10 ml of Nanopure water for 18 hours with continuous agitation at a rate of 200 rpm. Following the agitation step, the mixture was centrifuged at 5,000 rpm for 30 min utes (Beckman J2 HS, Tritech Field Eng Inc.) and the supernatant removed with a pipette. Next, the residue was rinsed with 10 ml of Nanopure water for 5 minutes and the mixture was centrifuged again at 5,000 rpm for 15 min. The supernatant was carefully withdrawn using a 5 ml pipette and add ed to the first supernatant fraction. The combined supernatant was then filtered (0.45 for the analysis of total Hg. The solid residue was then used in the next extraction step. Besides the extraction step, the centrifugation, rinsing, and fil tration steps were ident ical for fractions F1, F2, F3, and F4. Fraction 2 (F2) targets Hg present in the sample as HgO and HgSO 4 species The residue from F1 was treated with 10 m l of CH 3 COO + 0.01 HCl (0.1M, pH =2). The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrifugation, rinsing, and filtration as described in F1. Fraction 3 (F3) targets Hg that is bound to humic acids The residue from F2 was extracted wit h 10 m l of a 1M KOH solution. The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrifugation, rinsing, and filtration.
82 Fraction 4 (F4) targets Hg in amorphous minerals organo sulfur, Hg amalgams and Hg associated with crystalline Fe Mn oxides The residue from step F3 was extracted with 10 m l of 12M HNO 3 The mixture was agitated on an orbital shaker at 200 rpm at room temperature for 18 hours, prior to centrifugation, rinsing, and filtration. Fraction 5 (F5) target s residual Hg species, mainly Hg sulfide minerals (e.g. HgS). This residual fraction was not chemically extracted. After the F4 step, the solid phase was rinsed with DI water and air dried. The dry samples were than analyzed by XPS to determine the type of association between residual Hg and sorbent. Details on sample preparation and analysis are described later 4.2.4 Total Hg analysis For total Hg analysis on solid Al WTR samples 5 ml of aqua regia (mixture of concentrated HNO 3 and HCl) and 5 ml of hydrofl uoric acid (HF) were added to ~0.5 to 1 g of dry Al WTR samples. Digestion was performed overnight in capped Teflon vessel s at 110C. Upon cooling, digested samples were diluted to 50 ml with a saturated solution of boric acid to dissolve any formed fluorides For the aqueous fr actions obtained through SSE concentrations of Hg in supernatants were determined following overnight digestion with bromine mo nochloride (mixture of KBr and BrO 3 dissolved in concentrated HCl) at room temperature. This step allows for the oxidation of so luble organic matter and dissolution of colloidal Hg. After the oxidation step, 10 L of 30% hydroxylamine chloride (NH 2 OH HCl) was added to digested samples. The samples were swirled and allowed to react for at least 5 minutes. Following sample digestion, total Hg concentrations on both solid and aqueous samples
83 were determined by cold vapor atomic fluorescence spectrophotometry (CV AFS), using US EPA method 1631. 4.2.5 Physical analysis of Al WTR samples Representative dry samples of control and Hg spiked Al WTR samples were pulverized using a ball mill grinder The pulverized samples were analyzed by the analytical techniques described below. Morphology and elemental composition of a ged samples were determined via scanning electron microscopy (SEM), carried out in the JSM 6330F field emission scanning electron microscope unit equipped with an X ray energy dispersive spectrometer (SEM EDS) The characterization capabilities of EDS analysis are based on the fundamental principle that each element has a unique atomic structure and therefore a unique set of peaks on its X ray spectrum. This analysis was performed at the Ma j or Analytical Instrumentation Center (MAIC) at the University of Florida X ray diffraction (XRD) was used to seek the products of bulk phase sorption, and to obtain information on crystallinity and the mineralogical and geochemical environment in which Hg is found within the Al WTR matrix. Al WTR s amples were prepared for x ray diffraction (XRD) analysis by grinding in a ball mill to powder form and loading into cavity mounts (Harris and White, 2008) X ray diffraction analyses were conducted using a computer controlled x ray diffractometer (Ultima IV X Ray Diffractometer, Rigaku Corporation, Japan) e quipped with stepping motor and graphite crystal monochromator. Samples we 60 degrees 2 Near surface analysis of aged Hg spiked Al WTR and residual fractions of non spiked and fresh and aged Hg spiked Al WTR samples was performed by X ray
84 photoelectron spectrosco py (XPS). This was perfor med using a Perkin Elmer 5100 XPS System with AugerScan software at the Ma j or Analytical Instrumentation Center (MAIC) The Al WTR sample was loaded into the sample chamber on double sided copper tape (3M) and pumped down to a vacuum of 10 8 Torr. Sa mples were then radiated using photons of Mg with energy of 1253.6 eV. The survey scans were recorded using a fix ed pass energy of 89.45 eV, and narrow scan spectra of the Hg 4f, Al 2p, O 1s Cl 2p3 and C1s levels were recorded using a fixed pass energy of 35.75 eV and a range of 10 30 swaps. The recorded lines were fitted using a curve fitting program with Gaussian Lorentzian peak shape. Once the peaks were fitted each spectrum was calibrated to the binding energy of C 1s photoelectrons at 284.6 eV (calibration varied between samples from 2 2.75 eV) Binding energy (BE eV) of each fitted and calibrated peak was compared to the NIST X ray Photoelectron Spectroscopy database 20, version 3.5 ( http://srdata.nist.gov/xps/ ) for full characterization and spe ciation of surface bound elements. 4 .3 Result s and Discussion Data on physicochemical characterization and elemental composition of Al WTR samples are discussed in Chapter 3. In general, pH and EC of samples were lower than values reported in previous studies, whereas eCEC was higher than values rep o rted for soils (Essington, 2004) In addition, Al WTRs showed a macro element composition similar to those of previous Al WTR studies (Agyin Birikorang and Oonnor, 2007; Makris et al., 2004) Soon after spiking with HgCl 2 c oncentratio n and distribution of Hg, Al and Si in specific chemical fractions of Hg spiked Al WTR, showed Hg being incorporated into the residual fraction of Al WTR samples, along with Si, while Al wa s found primarily in the Fe/Mn oxide fraction.
85 Total mercury concen tration and its distribution among the different chemical fractions of aged Hg spiked Al WTR, fresh Hg spiked Al WR and non spiked Al WTR is presented in Table 4 4. THg concentration in non spiked Al WTR samples averaged 30 g Hg/kg Al WTR or ppb THg con centration determined on 4 year old Hg spiked Al WTR (8201 435 mg Hg/kg Al WTR or ppm) was considerably lower than the initial THg concentration measured soon after sample spiking (i.e. an average 24 050 mg Hg/kg Al WTR or ppm). This large difference in r ecovered THg mass is likely a consequence of : (i) Hg incorporated deeper into the micropores of Al WTR becoming inaccessible to the digestion procedure used to extract Hg, and (ii) po ssible losses attributable to Hg ability to distribute between solid and gas phase. Previous studies that induc ed Hg volatilization in soils reported Hg loss as high as 100% for sandy soils and 20% for clay soils, when initially spiked with either HgCl 2 or HgNO 3 2 (Rogers and McFarlane, 1979) However, the high Hg sorption capacity of Al WTR and the establish ed role of its micropores in retaining sorbed conta minants (Hovsepyan and Bonzongo, 2009; Makris et al., 2005; Yang et al. 2006) suggest that Hg could be strongly locked into the internal channels and micropores and n ot easily leachable. If true, this could be an advantage with regard to the use of Al WTR in soil remediation by in situ immobilization. However, this should be verified analytically, and an alternative extraction procedure such as bomb digestion would be more appropriate. In general, SSE of the different samples showed that 4 year old Hg spiked Al WTR had the highest percentage of Hg in the residual fraction (F5), wh ereas both fresh Hg spiked Al WTR and non spiked Al WTR had a higher percent of Hg in f r act ion F4, which corresponds to Hg bound to crystalline oxide minerals. These results
86 demonstrate that Hg incorporation into the micropores and becoming part of the operationally defined residual fraction is likely a rate limiting step in Hg immobilization by Al WTRs. Accordingly, WTR contact time with the contaminant is of primary concern (Ippolito et al., 2011) Al WTRs were spiked with a high concentration of HgCl 2 solution, but nevertheless, both aged and fresh Hg spiked Al WTR had low Hg mass recovered in the water soluble fraction (F1), only 0.32% and 0.12% of THg concentrations for fresh and aged samples respectively. The HgO/HgSO 4 or easily exchangeable fraction (F2) represented only 0.67% an d 3.88% of the total Hg for aged and fresh Hg spiked Al WTR samples, respectively. Contaminants in the water soluble and exchangeable fractions are the most mobile and thus present the highest environmental risk. Still, these results show that even highly soluble Hg species are rapidly incorporated in to the humic acid fraction (F3) and crystalline oxide fraction (F4). These results clearly demonstrate the potential of Al WTR to reduce Hg mobility if used in remediation of Hg contaminated soils. Scanning e lectron secondary images revealed that Al WTR samples are heterogenous mixtures of particles with variable sizes (Figure 4 1A ) A closer view of the sample showed particles with irregular surfaces (Figure 4 1B), which could potentially provide these sorben ts with larger reactive areas for sorption of Hg or other metals. The elemental composition analysis revealed spectra dominated by Al, O, Si, P, S, Cl, Ca, a nd Fe (Figure 4 2A and 4 2B) as previously reported by others (Hovsepyan and Bonzongo, 2009; Makris et al., 2004) The EDS spectrum of Hg spiked samples showed a Hg peak in addition to the above elements confirming its sorption onto Al WTR particles (Figure 4 2C). However, the Hg peak/shoulder in the EDS scan from the
87 aged Hg spiked Al WTR sample (Figure 4 3B) was smaller than the peak obtained from the same sample when it was freshly spiked four years ago The EDS signal is totally dependent on concentration and depth. If the elements are not within surface (because they are concentrated inside the mesopores and micropores of the particle), or in a local concentration above ~1 3 % then they are not likely to be detected. Previous studies using the Weber Morris intra particle diffusion mo del showed that Hg is incorporated into the micropores of Al WTR likely involving a two step process starting with (i) relatively fast electrostatic attraction to the surface of the particle, followed by (ii) slow diffusion into Al WTR micropores (Hovsepyan and Bonzongo, 2009) The latter step would be responsible for the deeper sequestration of Hg within particles resulting in a non labile fraction isolated from the bulk of what might be leachable (Axe and Trivedi, 2002) In addition, the linearity of model ( R 2 = 0.9948, p < 0.005) confirms that intra particle diffusion is in effect a rate limiting step for Hg sorption on Al WTRs (Hovsepyan an d Bonzongo, 2009) The mechan isms of intra particle diffusion could also be supported by SSA N 2 and CO 2 N 2 analysis (Table 4 1), which revealed Al WTR particles have a large internal network of mesopores and micropores (Hovsepyan and Bonzongo, 2009) Accordingly the reduced Hg peak shown on the EDS spectra of aged Hg spiked Al WTR is most likely a result of the above slow incorporation of Hg over time making it un detectable by EDS analysis. Elemen tal mapping by SEM EDS was used to determine spatial distribution of elements within the 4 year old Hg spiked Al WTR sample. This technique provides a 2D scan of the elements, which can assist in the identification of the different chemical
88 phases (Figure 4 4). Results placed Hg in the vicinity of major elements in the order: S, Cl, Si, Al, O. Although spatial distribution of elements in the sample might suggest Hg is mainly forming bonds with S and Cl ions, this technique cannot be used to determi ne bondin g between elements X ray diffraction analysis showed a high abundance of quartz (SiO 2 ) and calomel (HgCl 2 ) with no apparent crystal line phase for aluminum (Figure 4 5), suggesting that amorphous Al (oxy) hydroxides dominate the Al WTRs (Makris et al., 2004) O verall, SEM EDS mapping, XRD analysis, and Hg fractions obtained from SSE of 4 year old Hg spiked Al WTRs suggest Hg is forming bonds with amorphous aluminum oxide minerals in Al WTR particles. However, further microscopic studies are neces sary to determine mechanisms of Hg sorption into Al WTR. Comparative XPS analyses of (i) aged Hg spiked Al WTR, and the residual fraction of (ii) fresh Hg spiked Al WTR and (ii) aged Hg spiked Al WTR were performed to gain insight into Hg speciation and b inding to specif ic chemical functional groups as a function of time. X ray s can penetrate up to 1 m into the sample, however useful e signals are only obtained from the first 1 30 monolayers (0.3 10 nm). The depth of the analysis will be determined by th e nature of the matrix under study. Thus, regardless of the element abundance, if it is not found within the first 10 nm of the surface it will not provide a useful signal for the analysis but will instead undergo inelastic collisions while traveling to the surface, giving rise to a stepped background, as shown in the wide scan survey (Figure 4 6). X ray photoelectron spectroscopy wide scan surveys revealed the surface of the aged Hg spiked Al ereas the surfa ce of the residual fraction of fresh and aged Hg spiked Al WTR is primarily composed of C,
89 O, Si and N. The absence of silicon in the surface of aged Hg spiked Al WTR suggests that Si is found more than 10 nm from the surface, or simply deeper than what ca n be detected with this technique. On the other hand, the absence of an aluminum peak in the residual fraction of fresh and aged Hg spiked Al WTR could be explained by (i) the removal of the Al oxide fraction during the selective sequential extraction, (ii ) amorphous Al oxides being deeper within the pores of the sample, or (iii) shadowing from other elements that are bound to Al and perhaps covering the surface of Al oxides. In general, apparent differences in the elemental composition of the surface could be a consequence of lower sensitivity and limited penetration of the technique to this specific matrix, changes in the composition of the surface after selective sequential extraction, and the chemical interactions between elements in the sample. In addit ion, the presence of Hg4f peaks in all Hg spiked samples clearly confirmed the sorption and incorporation of Hg into Al WTR particles. Characterization of the Hg4f peaks for the residual fraction of aged Hg spiked Al WTR (Figure 4 7A) showed two major peak s at 100.80 eV (60.2%) and 105.12 eV (39.8%), both corresponding to HgO species. After deconvolution, one peak for HgCl 2 at 101.26 eV (37.1%) and another for HgO at 105.27 eV (62.9%) were identified from the Hg4f peak obtained with the aged Hg spiked Al W TR. The Hg4f peak of fresh Hg spiked Al WTRs was also de convoluted into two major species, Hg(H 2 NC(O)NHC(O)NH 2 ) 2 Cl 2 at 101.31 eV (57.2%) and HgO at 105.04 eV (42.8%). The presence of HgO in all samples and a higher concentration of HgO in the residual fra ction of aged Hg spiked Al WTRs compared to the other samples confirmed that Hg added as HgCl 2 is slowly incorporat ed into the residual fraction of the Al WTRs and over time Hg oxides are formed. A higher concentration of
90 HgCl 2 as compared to HgO in aged Hg spiked Al WTR was expected because of the fact that Hg spiked samples were prepared using a HgCl 2 solution. Matching species for Hg(H 2 NC(O)NHC(O)NH 2 ) 2 Cl 2 in the C 1s peak of fresh Hg spiked Al WTRs and for HgCl 2 in the Cl 2p3 peak of aged Hg spiked Al WT Rs were also found during high resolution scans (data not show n ). High resolution scans of O 1s were rather wide, giving some indi cation of a possible multiplicity of adsorption sites The O 1s peak for the residual fraction of aged Hg spiked Al WTR was dec onvoluted in two components at 532.5 eV, and 530.86 eV corresponding to oxygen atoms in SiO 2 and HgO species (Figure 4 8A). Fresh Hg spiked Al WTRs only showed peaks for SiO 2 (Figure 4 8B). A non matching HgO peak in the deconvoluted spectr um and fitting of this last sample could be a consequence of an energy shift from 529.91 eV, which is the reported binding energy (BE) for HgO, giving rise to a BE which is not characteristic of the compound. Energy shifts or changes in the BE of specific compounds can b e explained by (i) changes in the oxidation state of the sample and (ii) collisions from e in the inner layers of the samples. The O 1s peak of aged Hg spiked Al WTR was deconvoluted in 5 components corresponding to oxygen atoms in Al 2 O 3 SiO 2 and HgO spec ies (Figure 4 8C). Further analyses of the samples were made to determine matching peaks for Si 2p and Al 2p (Figures 4 9 and 4 10). Because the residual fraction of Hg spiked Al WTRs have a higher concentration of silica and mercury than aluminum (Table 4 3), and XPS analysis revealed Hg is forming HgO in the residual fraction, we could assume the Hg is forming oxides mainly or exclusively with SiO x species. However, the XPS survey of the residual fraction of non
91 spiked Al WTR shows a peak for aluminum (Fig ure 4 6 B). The decrease/disappearance of the Al 2p peak and appearance of Hg 4f peak in the Hg spiked samples seem to be related, and suggest aluminum oxides are present in the residual fraction of Al WTRs and bound to Hg. Previous EXAFS studies by Kim et a l (2004a) on modes of Hg (II) sorption to synthetic well characterized Al 2 O 3 ) indicat ed that both outer sphere and inner sphere monodentate surface complexation might be involved in Hg alumina sorption (Figure 4 11). 4.4. Summary of the Main Findings Secondary images produced through SEM analysis showed that the Al WTR particles are characterized by irreg ular surfaces, which could result in large reactive surface areas for sorption of Hg. Sample digestion and EDS spectra revealed an elemental composition similar to that reported i n previous Al WTR studies. X ray diffraction analysis showed the presence of SiO 2 (quartz) and Hg 2 Cl 2 (Calomel) The latter being the result of spiking Al WTR samples with a HgCl 2 solution. N o crystalline phase for aluminum was observed, suggesting that aluminum (hydr)oxides present in WTR sample s were primarily in amorphous state Selective sequential extraction of 4 year old Hg spiked Al WTR soon after spiking with HgCl 2 revealed Hg is incorporated into the residual fraction (82%) of Al WTR samples, along with Si (99%), wh ereas Al is found primarily in the Fe/Mn oxide fraction ( 50%) followed by the residual fraction (30%). Selective sequential extraction of non spiked Al WTR, 4 year old Hg spiked Al WTR and, fresh Hg spiked Al WTR, showed 4 year old samples had the highest percentage of Hg in the residual fraction (F5), wh ereas both fresh Hg spiked Al WTR and non spiked Al WTR had a higher percent of Hg in the Fe/Mn oxide fraction (F4). These results demonstrate that Hg entry into the micropores and incorporation into the operationally defined residual fraction is likely a rate l imiting step in Hg immobilization by Al WTRs. Accordingly, WTR contact time with the contaminant could be a significant factor with regard to sorption and immobilization efficiencies (Ippolito et al. 2011) In addition, the rapid incorporation of Hg into t he humic acid fraction (F3) and oxide fraction (F4) of Al WTRs points to a potential weakness as these fractions are prone to redox transformations and subsequent mobilization.
92 THg concentration determined on 4 year old Hg spiked Al WTR was considerably l ower than the initial THg concentration measured soon after sample spiking with HgCl 2 Although the observed difference in concentration values can not be explained with data at hand, this rather large difference can potentially be attributed to : (i) deep er incorporation of Hg into the micropores of the Al WTR, preventing the extraction of Hg by means of the acid digestion procedure used here; and (ii) Hg loss to the gaseous phase over time A high Hg sorption capacity of Al WTR and the established role o f its micropores in retaining sorbed contaminants (Ho vsepyan and Bonzongo, 2009; Makris et al., 2005; Yang et al., 2006) suggest that Hg is likely concealed within the internal micropores. This can be verified analyti cally by use of hot acid bomb digestion. A n d if confirmed analytically this could then be an advantage with regard to the use of Al WTR in soil remediation. X ray photoelectron spectroscopy wide scan surveys revealed differences in the surface composition of Al WTR samples. The absence of an aluminum peak in the residual fraction of both fresh and aged Hg spiked Al WTR is most likely the result of shadowing from other elements, which are bound to Al and perhaps covering the surface of Al oxides. Wh ereas th e absence of silicon in the surface of aged Hg spiked Al WTR suggests that this element is found in deeper layer s of the Al WTR and cant be easily detect ed with this technique. Elemental mapping of 4 year old Al WTR samples placed Hg in the vicinity of maj or elements in the order: S, Cl, Si, Al, O. Although spatial distribution of elements in the sample might suggest Hg is mainly forming bonds with S and Cl ions, this technique cannot be used to determi ne bonding between elements Further analysis of the sa mple using X ray photoelectron spectroscopy, revealed Hg is binding to Al and Si oxides. Accordingly, what seems to be a reduced relation between Hg and Al, O, could be the result of shadowing from Hg binding at the surface of Al oxide. XPS analysis showe d the surface of Hg spiked Al WTR is composed of HgO, HgCl 2 and Hg(H 2 NC(O)NHC(O)NH 2 ) 2Cl 2 species. Mercury oxide being the most abundant and persistent in the samples. Vanishing of the Al peak in Hg spiked Al WTR compared to non spiked Al WTR is most likel y a consequence of Hg binding to Al 2 O 3 and the inability of XPS to pick up a use ful e signal. High abundance of SiOx in the residual fraction of 4 year old Al WTR samples most likely indicates complexation of Hg O Si. Although the specific mechanism of so rption could not be elucidated through SSE, SEM EDS, XRD and XPS analysis, we speculate that Hg is binding to the surface of Al and Si oxides by a combination of mechanism such as electrostatic attraction at the surface of the mineral lattice and/or by fo rming covalent bonds with oxygen in Al and Si oxides. In addition, preliminary studies on mechanisms of Hg sorption into Al 2 O 3 ) suggest outer sphere and inner sphere monodentate are most probable
93 4.5 Future Research A venues To identif y the binding sites of Hg within the Al WTR matrix, a combination of more than three physicochemical techniques is needed. In addition to Transmission Electron Microscopy (TEM), Electron Backscatter Diffraction (EBSD) and Thermo desorption coupled with atomic fluorescence spectrometry extended X ray absorption fine structure (EXAFS) and X ray absorption near edge structure (XANES) methods would be appropriate EXAFS and XANES can provide insight into molecular scale information that can help define the chemica l speciation of Hg at a molecular level and distinguish between different Hg species. EXAFS with its high kinetic energy range (150 2000 eV) provides elemental specificity and allows extraction of the signal from a surface monolayer or a single buried la yer in the presence of a large background signal. XANES with its low kinetic energy range (5 150 eV) provides crucial information such as formal valence, coordination environment, and subtle geometrical distortions.
94 Table 4 1. Concentration and distrib ution (%) of Hg, Al and Si in specific chemical fractions of Hg spiked Al WTRs using the fractionation method adapted from Tessier et al. (1979) (adapted from Hovsep yan and Bonzongo 2009) Element Fraction Concentration mg/kg a % of total Hg Water soluble 146 6 0.60 Exchangeable 766 32 3.18 Carbonate 230 23.1 0.95 Fe/Mn oxides 889.5 150 3.69 Organic 2162.8 306 8.99 Residual 19855 172 82.3 Total 24050 842 Water soluble
95 Table 4 2. Concentration and distribution (%) of Hg, in specific chemical fractions of Hg spiked Al WTR samples determined by the SSE met hod adapted from Bloom et al. (2003) Al WTR sample Fraction Concentration mg/kg a % of total Aged Hg spiked Water soluble 26.1 3.1 0.32 HgO, HgSO 4 10.0 1.0 0.12 Bound to humic acids 1650.6 148.0 20.13 Crystalline oxides 1445.7 94.1 17.63 Residual/HgS 5369.2 65.47 Total 8201.6 435.7 Water soluble 21.4 0.8 0.67 Fresh Hg spiked HgO, HgSO 4 124.6 10.5 3.88 Bound to humic acids 720.4 68.9 22.43 Crys talline oxides 1833.4 200.9 57.08 Residual/HgS 512.3 15.95 Total 3212.1 78.2 Water soluble 9.5 x10 4 3.5x10 5 3.17 Non spiked HgO, HgSO 4 3.9 x10 4 8.5x10 5 1.30 Bound to humic acids 0.01 3.7x10 4 33.33 Crystalline oxides 0.02 1.8x10 3 66. 67 Residual/HgS 2.2x10 3 7.33 Total 0.03 6.4x10 3 a Mean standard error
96 Figure 4 1. SEM micrograph of Al WTR collected from Bradenton Drinking Water Treatmen t facility (Florida USA) showing A) heterogene ous morphology of material and B) s urface area of a single Al WR particle. B A
97 Figure 4 2 EDS spectra of Al WTR samples A) non spiked Al W TR B) Hg spiked Al WTR, THg=24050 842 mg Hg/kg Al WTR C) 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg Al WTR C O Al S i P Hg S C l Ca Fe A B C
98 Figure 4 3. Elemental mapping of aged Hg spiked Al WTR samples using TEAM EDS technology. A) Aluminum B) Silica C) Oxygen D) Mercury E) Sulfur and F) Chloride A E B E C F
99 Figure 4 4 XRD spectra of 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg Al WTR
100 A B Figure 4 5. XP S wide scan survey spectru m of A) residual fraction of 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg Al WTR and B) non spiked Al WTR.
101 A B C Figure 4 6. XPS spectra of Hg 4f in, A) the residual fraction of 4 year old Hg spiked Al WTR, THg =5369.17 mg/kg B) the residual fraction of fresh mercury spiked Al WTR, THg =512.33 mg/kg and C) 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg.
102 A B C Figure 4 7. XPS spectra of O 1s in, A) the residual fraction of 4 year old Hg spiked Al WTR, THg=5369.17 mg/kg B) the residual fraction of f resh mercury spiked Al WTR, THg=512.33 mg/kg and C) 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg.
103 Figure 4 8. XPS spectra of Al 2p in 4 year old Hg spiked Al WTR, THg=8,202 mg Hg/kg.
104 A B Figure 4 9. XPS spectra of Si 2p in A) the residual fraction of 4 year old Hg sp iked Al WTR, THg=5369.17 mg/kg B) the residual fraction of fresh mercury spiked Al WTR, THg=512.33 mg/kg.
105 Table 4 3. Abundance (%) of Hf 4f Al 2p S 2p O 1s C 1s N 1s and Cl 2p3 on the different Hg s piked Al WTR sample s, as determined b y XPS analysis Al WTR sample Element Abundance (%) 4 year old Hg spiked Al WTR Hg 4f 0.9 Al 2p 10.5 Si 2p nd a O 1s 46.7 C 1s 39.1 N 1s 1.6 Cl 2p3 1.2 Residual fraction of non spiked Al WTR Hg 4f nd Al 2p 10.1 Si 2p nd O 1s 47.7 C 1s 4 0.9 N 1s 1.3 Cl 2p3 nd Residual fraction of 4 year old Hg spiked Al WTR Hg 4f 1.2 Al 2p nd Si 2p 26.6 O 1s 40.7 C 1s 29.2 N 1s 2.3 Cl 2p3 nd Residual fraction of Fresh Hg spiked Al WTR Hg 4f 0.7 Al 2p nd Si 2p 14.1 O 1s 42.0 C 1s 40.4 N 1s 2.8 Cl 2p3 nd a Not detected
106 Figure 4 10. alumina surface, with reduced Hg(I) Hg(I) binuclear species sorbing as (1) a monodentate (mononuclear) complex on a singly coordinated ox ygen (upper right); (2) a bidentate corner sharing (binuclear) complex on two singly coordinated oxygens (upper left); and (3) a monodentate (mononu clear) complex on a singly coordinated oxygen site of a bridging hydrated Al octahedron (bottom center). N ot shown is outer sphere sorption of the Hg 2 (OH) 2 aqueous species. The Hg(I) Hg(I) distance is 2.54 in all cases. Figure was obtained from Kim et al. (2004a)
107 CHAPTER 5 GENERAL CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK Waste material formed by coagu lation during d rinking water treatment offers a tremendous opportunity for development of soil remediation technologies based on the value added to waste In the US alone, more than 2 million metric tons of water treatment residuals (WTRs) are pr oduced daily (Agyin Birikorang et al., 2009) but simply discarded. Unfortunately, the d isposal of WTRs can be expensive and it increases the overall cost of the water purification process (Novak and Watts, 2004) Accordingly, using WTRs in the remediation of contaminated soils would prov ide a way of adding value to this abundant and readily available waste material. The first comprehensive study using WTRs for land application was performed in the early 19 90s ( e.i Elliot et al. 1990) and focused on the immobilization of negatively charg ed ions such as phosphate Since then, progress has been made and current scientific knowledge of the composition environmental behavior, and potential uses of WTRs has improved significantly (Ippolito et al., 2011) However, a lthough most studies conducted to date have focused primarily on the potential use of WTRs as a cost effective approach to remediate soils contaminated with oxyanions the potential for remediating metal contaminated soils is emerging Given the previously established ability of Al WTR to adsorb mercury (Hg) from aqueous solutions (Hovsepyan, 2008; Hovsepyan and Bonzongo, 2009) this research focused on : (1 ) assessing the efficiency of Al WTR to immobilize Hg in contaminated soils (long term versus freshly contaminated soils) as a function of leaching solution type, with emphasis on pH and DOC content and (2) investigation of potential
108 mechanisms involved in the binding of Hg within the Al WTR matrix to help predict the long term stability of sorbed Hg. The main findings of this research can be summarized as follows: In general, the addition of Al WTR reduces the Hg leaching potential of Hg contaminated soils. However, in soils with high content of silt and clay, the high retention capacity of metal cations such as Hg n + naturally limits the release of Hg, even when soils are leached with a low pH solution such as SPLP. The benefits of Al WTR treatments are more pronounced i n soils with hi gher percentages of Hg bound in the water extractable and easily exchangeable soil fractions. Soil leaching with organic rich water collected from the Suwannee River increased Hg release compared to SPLP, but only for DOC concentrations in excess of ~15 m g C/L. The reduced efficiency of Al WTR to immobilization under these conditions is probably a consequence of the high affinity of Hg organic ligands. Accordingly, DOC appears to be a key variable and needs to be taken into account as the use of Al WTR in soil remediation is contemplated. In addition to DOC, knowledge of Hg speciation is necessary to justify the use of in situ Hg immobilization by soil amendment with Al WTRs. Only soils with high Hg levels in the easily extractable fractions are worth trea ting with Al WTR With regard to application rates and schemes, numbers between 2 and 5% should be adequate, and the use of Al WTR as a liner would be advantageous only for fine textured soils with high clay and silt contents. Results of solid phase anal ysis of Al WTR spiked with Hg using X ray diffraction identified Hg 2 Cl 2 (Calomel) as the only Hg based crystalline mineral. No crystalline aluminum phases were identified by XRD suggesting that if present, amorphous aluminum (hydr) oxides dominate in the Al WTR sample and/or crystalline phases occur at levels below the detection of the technique used In contrast, XPS analysis showed the presence of Al 2 O 3 SiO x and HgO. Although specific mechanisms of sorption could not be clearly established through the use of SSE, SEM EDS, XRD and XPS a few indications suggest that Hg is likely found at the surface of Al and Si oxides through outer sphere and inner sphere monodentate complexation and a combination of mechanism s such as electrostatic attraction and cova lent bonding could be at play. Future research avenues should focus on: Further elucidation of Hg binding sites in the Al WTRs by use of a combination of more efficient techniques such as Thermo desorption coupled with atomic fluorescence spectrometry E XAFS and XANES
109 D etermination of the bioavailability of Hg sorbed onto Al WTR. P roduction of methyl Hg is one of the primary concerns with regard to the fate of Hg in the environment. Therefore, reduction and/or elimination of Hg methylation by soil microo rganisms would be a tremendous advantage with regard to the in situ use of the technique. Pilot studies conducted in situ for the determination of long term efficiency and monitoring of potential adverse implications
110 LIST OF REFERENCES Abollino, O., Aceto, M., Malandrino, M., Sarzanini, C., Mentasti, E., 2003. Adsorption of heavy metals on Na montmorillonite. Effec t of pH and organic substances. Water Research 37, 1619 1627. Abollino, O., Giacomino, A., Malandrino, M., Mentasti, E., 2008. Interaction of metal ions with montmorillonite and vermiculite. Applied Clay Science 38, 227 236. Adriano, D., 2001. Trace Elemen ts in Terrestrial Environments: Biogeochemistry, Bioavailability, and Risks of Metals. Springer Verlag, New York. Agyin Birikorang, S., O'Connor, G.A., 2009. Aging effects on reactivity of an aluminum based drinking water treatment residual as a soil amend ment. Science of The Total Environment 407, 826 834. Agyin Birikorang, S., Oonnor, G., 2007. Liability of drinking water treatment residuals (WTR) immobilized phosphorus: aging and pH effects. Journal of Environmental Quality, pp. 1076 1085. Agyin Birikor Based Drinking Water Treatment Residuals Safe for Land Application? University of Florida, IFAS Extension, pp. 1 7. Aiken, G., Haitzer, M., Ryan, J.N., Nagy, K., 2003. Interactions between dissolved organic matter and mercury in the Florida Everglades. J.Phys.IV France, pp. 29 32. Allard, B., Arsenie, I., 1991. Abiotic reduction of mercury by humic substances in aquatic system an important process for the mercury cycle. Water Air & Soil Pollution 56, 457 4 64. Andersson, A., 1979. Mercury in Soil: The biogeochemistry of mercury in the environment. Elsevier, North Holland Biomedical Press, Amsterdam, the Netherlands, pp. 79 112. ATSDR, 2011. Priority list of hazardous substances. http://www.atsdr.cdc.gov/spl/ Axe, L., Trivedi, P., 2002. Intraparticle Surface Diffusion of Metal Contaminants and their Attenuation in Microporous Amorphous Al, Fe, and Mn Oxides. Journal of Colloid and Interface Science 247, 259 265. Babatunde, A.O., Zhao, Y.Q., 2006. Constructive Approaches Toward Water Treatment Works Sludge Management: An International Review of Beneficial Reuses. Critical Reviews in Environmental Science and Technology 37, 129 164.
111 Babiarz, C., Benoit, J., Shafer, M., Andren, A., Hurley, J., Webb, D., 1998. Seasonal influences on partitioning and transport of total and methylmercury in rivers from contrasting watersheds. Biogeochemistry 41, 237 257. Behra, P., Bonnissel Gissinger, P. Alnot, M., Revel, R., Ehrhardt, J.J., 2001. XPS and XAS Study of the Sorption of Hg(II) onto Pyrite. Langmuir 17, 3970 3979. Bernaus, A., Gaona, X., Esbr, J.M., Higueras, P., Falkenberg, G., Valiente, M., 2006. Microprobe Techniques for Speciation Analy sis and Geochemical Characterization Science & Technology 40, 4090 4095. Biester, H., Nehrke, G., 1997. Quantification of mercury in soils and sediments acid digestion versus pyrolysis. Fresenius' Journal of Analytical Chemistry 358, 446 452. Biester, H., Zimmer, H., 1998. Solubility and changes of mercury binding forms in contaminated soils after immobilization treatment. Environmental Science & Technology 32, 2755 2762. Bloom N.S., Preus, E., Katon, J., Hiltner, M., 2003. Selective extractions to assess the biogeochemically relevant fractionation of inorganic mercury in sediments and soils. Analytica Chimica Acta 479, 233 248. Bonnissel Gissinger, P., Alnot, M., Lickes, J. P. Ehrhardt, J. J., Behra, P., 1999. FeOOH (Goethite) and Amorphous Silica. Journal of Colloid and Interface Science 215, 313 322. Bonzongo, J.C., Heim, K.J., Warwick, J.J., Lyons, W.B., 1996a. Me rcury levels in surface waters of the Carson River Lahontan Reservoir system, Nevada: Influence of historic mining activities. Environmental Pollution 92, 193 201. Bonzongo, J.C., Lyons, W.B., Hines, M.E., Warwick, J.J., 2002. Mercury in surface waters of three mine dominated aquatic systems: Idrija River, Slovenia; Carson River, Nevada USA; and Madeira River, Brazil. Geochemical Exploration and Environmental Analysis 2, 111 120. Bonzongo, J.C.J., Heim, K.J., Chen, Y.A., Lyons, W.B., Warwick, J.J., Miller, G.C., Lechler, P.J., 1996b. Mercury pathways in the Carson River Lahontan reservoir system, Nevada, USA. Environmental Toxicology and Chemistry 15, 677 683. Bose G., 2008. Mercury as a serious health hazard for children in gold mining areas. Environmental Research 107, 89 97. Bower, J., Savage, K.S., Weinman, B., Barnett, M.O., Hamilton, W.P., Harper, W.F., 2008. Immobilization of mercury by pyrite (FeS2). Environmental Pol lution 156, 504 514.
112 Brown, S., Christensen, B., Lombi, E., McLaughlin, M., McGrath, S., Colpaert, J., Vangronsveld, J., 2005. An inter laboratory study to test the ability of amendments to reduce the availability of Cd, Pb, and Zn in situ. Environmental P ollution 138, 34 45. Camps Arbestain, M., Rodrguez Lado, L., M. Bao, M., Macas, F., 2009. Assessment ofMercury Polluted Soils Adjacent to an OldMercury Fulminate Production Plant. Hindawi Publishing Corporation, Applied and Environmental Soil Science. Ca o, X., Ma, L., Shiralipour, A., Harris, W., 2009. Biomass reduction and arsenic transformation during composting of arsenic rich hyperaccumulator Pteris vittata L. Environ Sci Pollut Res. Chen, C.C., McKimmy, E.J., Pinnavaia, T.J., Hayes, K.F., 2004. XAS s tudy of mercury(II) ions trapped in mercaptan Functionalized mesostructured silicate with a wormhole framework structure. Environmental Science & Technology 38, 4758 4762. Chen, S.L., Wilson, D.B., 1997. Genetic engineering of bacteria and their potentia l for Hg2+ bioremediation. Biodegradation 8, 97 103. Chiang, Y.W., Ghyselbrecht, K., Santos, R.M., Martens, J.A., Swennen, R., Cappuyns, V., Meesschaert, B., 2012. Adsorption of multi heavy metals onto water treatment residuals: Sorption capacities and app lications. Chemical Engineering Journal 200, 405 415. Ciccu, R., Serci, A., Fadda, S., Peretti, R., Zucca, A., 2003. Heavy metal immobilization in the mining contaminated soils using various industrial wastes. Minerals Eng., pp. 187 192. Cox, C.D., Shoesmi th, M.A., Ghosh, M.M., 1996. Electrokinetic remediation of mercury contaminated soils using iodine/iodide lixiviant. Environmental Science & Technology 30, 1933 1938. Cunningham, S.D., Berti, W.R., Huang, J.W.W., 1995. Phytoremediation of contaminated soil s. Trends in Biotechnology 13, 393 397. Dahrazma, B., Mulligan, C.N., 2007. Investigation of the removal of heavy metals from sediments using rhamnolipid in a continuous flow configuration. Chemosphere 69, 705 711. Dayton, E.A., Basta, N.T., 2001. Characte rization of Drinking Water Treatment Residuals for Use as a Soil Substitute. Water Environment Research 73, 52 57. Dayton, E.A., Basta, N.T., Jakober, C.A., Hattey, J.A., 2003. Using treatment residuals to reduce phosphorus in agricultural runoff. Journal American Water Works Association 95, 151 158.
113 Dong, D., Zhao, X., Hua, X., Liu, J., Gao, M., 2009. Investigation of the potential mobility of Pb, Cd and Cr(VI) from moderately contaminated farmland soil to groundwater in Northeast, China. Journal of Hazard ous Materials 162, 1261 1268. Donkor, A.K., Bonzongo, J.C., Nartey, V.K., Adotey, D.K., 2006. Mercury in different environmental compartments of the Pra River Basin, Ghana. Science of The Total Environment 368, 164 176. Donkor, A.K., Bonzongo, J.C.J., Nart ey, V.K., Adotey, D.K., 2005. Heavy metals in Sediments of the gold mining impacted Pra River Basin, Ghana, West Africa. Soil & Sediment Contamination 14, 479 503. Ehrhardt, J.J., Behra, P., Bonnissel Gissinger, P., Alnot, M., 2000. XPS study of the sorpti on of Hg(II) onto pyrite FeS2. Surface and Interface Analysis 30, 269 272. Elliott, H.A., O'Connor, G.A., Lu, P., Brinton, S., 2002. Influence of water treatment residuals on phosphorus solubility and leaching. Journal of Environmental Quality 31, 1362 136 9. Ensley, B.D., 2000. Rationale for use of phytoremediation. In: Phytoremediation of Toxic Metals Using Plants to Clean up the Environment. Raskin, I. and Ensley, B.D., Eds., Wiley, New York., 1 12 pp. Essington, M., 2004. The soil chemical environment:an overview, Soil and water chemistry: an integrated approach. CRC Press LLC, Boca Raton, FL, pp. 1 34. Fernndez Martnez, R., Loredo, J., Ordez, A., Rucandio, M.I., 2005. Distribution and mobility of m ercury in soils from an old mining area in Mieres, Asturias (Spain). Science of The Total Environment 346, 200 212. Gabriel, M.C., Williamson, D.G., 2004. Principal biogeochemical factors affecting the speciation and transport of mercury through the terres trial environment. Environmental Geochemistry and Health 26, 421 434. Garbisu, C., Alkorta, I., 2001. Phytoextraction: a cost effective plant based technology for the removal of metals from the environment. Bioresource Technology 77, 229 236. Grigal, D.F., 2003. Mercury sequestration in forests and peatlands: A review. Journal of Environmental Quality 32, 393 405. Gustin, M.S., Taylor, G.E., Leonard, T.L., 1994. High levels of mercury contamination in multiple media of the Carson River drainage basing of Ne vada implications for risk assessment. Environmental Health Perspectives 102, 772 778. Hamby, D.M., 1996. Site remediation techniques supporting environmental restoration activities A review. Science of the Total Environment 191, 203 224.
114 Harris, W., Wh ite, G., 2008. X ray diffraction techniques for soil mineral identification. p81 115. In, Methods for soil analysis:Part5 mineralogical methods. Soil Science Society of America, Ulery and R. Drees (eds.), Madison, WI, pp. 81 115. Hazardous Waste Consultant 1996. Remediating Soil and Sediment Contaminated with Heavy Metals Elsevier Science, Netherlands. He, Z.Q., Traina, S.J., Weavers, L.K., 2007. Sonolytic desorption of mercury from aluminum oxide: Effects of pH, chloride, and organic matter. Environmental Science & Technology 41, 779 784. Hinton, J., Veiga, M., 2001. Mercury Contaminated Sites: A Review of Remedial Solutions. Proc. National Institute for Minamata Disease, Forum 2001. March 19 20, Minamata, Japan. Hovsepyan, A., 2008. Immobilization of merc ury in contaminated soils using aluminum drinking water treatment residuals. UFL dissertation, 135 pp. Hovsepyan, A., Bonzongo, J. C.J., 2009. Aluminum drinking water treatment residuals (Al WTRs) as sorbent for mercury: Implications for soil remediation. Journal of Hazardous Materials 164, 73 80. Hussein, S., Ruiz, O.N., Terry, N., Daniell, H., 2007. Phytoremediation of mercury and organomercurials in chloroplast transgenic plants: Enhanced root uptake, translocation to shoots, and volatilization. Environm ental Science & Technology 41, 8439 8446. Hyde, J.E., Morris, T.F., 2000. Phosphorus availability in soils amended with dewatered water treatment residual and metal concentrations with time in residual. Journal of Environmental Quality 29, 1896 1904. Ippol ito, J.A., Barbarick, K.A., Elliott, H.A., 2011. Drinking Water Treatment Residuals: A Review of Recent Uses. Journal of Environmental Quality 40, 1 12. Jeong, Y., Fan, M., Singh, S., Chuang, C.L., Saha, B., van Leeuwen, H., 2007. Evaluation of iron oxide and aluminum oxide as potential arsenic(V) adsorbents. Chemical Engineering and Processing 46, 1030 1039. Jing, Y.D., He, Z.L., Yang, X.E., 2007. Effects of pH, organic acids, and competitive cations on mercury desorption in soils. Chemosphere 69, 1662 166 9. Johannessen, S.C., Macdonald, R.W., Eek, K.M., 2005. Historical Trends in Mercury Sedimentation and Mixing in the Strait of Georgia, Canada. Environmental Science & Technology 39, 4361 4368. Kabata Pendias, A., 2001. Trace elements in soils and plants. 3rd ed. CRC Press, Boca Raton, ????
115 Kim, C.S., Bloom, N.S., Rytuba, J.J., Brown, G.E., 2003. Mercury Speciation by X ray Comparison of Speciation Methods. Environmental Science & Technology 37, 5102 5108. Kim, C.S., Rytuba, J., Brown, G.E., 2004a. EXAFS study of mercury(II) sorption to Fe and Al (hydr)oxides II. Effects of chloride and sulfate. Journal of Colloid and Interface Science 270, 9 20. Kim, C.S., Rytuba, J.J., Brown G.E., 2004b. EXAFS study of mercury(II) sorption to Fe and Al (hydr)oxides I. Effects of pH. Journal of Colloid and Interface Science 271, 1 15. Lestan, D., Grcman, H., Zupan, M., Bacac, N., 2003. Relationship of soil properties to fractionation of Pb a nd Zn in soil and their uptake into Plantago lanceolata. Soil & Sediment Contamination 12, 507 522. Leun, D., SenGupta, A.K., 2000. Preparation and characterization of magnetically active polymeric particles (MAPPs) for complex environmental separations. E nvironmental Science and Technology 34, 3276 3282. Lippold, H., Lippmann Pipke, J., 2009. Effect of humic matter on metal adsorption onto clay materials: Testing the linear additive model. Journal of Contaminant Hydrology 109, 40 48. Liu, G., Cabrera, J., Allen, M., Cai, Y., 2006. Mercury characterization in a soil sample collected nearby the DOE Oak Ridge Reservation utilizing sequential extraction and thermal desorption method. Science Total Environment 369, 384 392. Lovley, D.R., Coates, J.D., 1997. Bio remediation of metal contamination. Current Opinion in Biotechnology 8, 285 289. Ma, L.Q., Rao, G.N., 1997. Effects of phosphate rock on sequential chemical extraction of lead in contaminated soils. Journal of Environmental Quality 26, 788 794. Makris, K.C., Harris, W.G., 2006. Time dependency and irreversibility of water desorption by drinking water treatment residuals: Implications for sorption mechanisms. Journal of Colloid and Interface Science 294, 151 154. Makris, K.C., Harris, W.G., O'Conn o, G.A., Obreza, T.A., 2004. Phosphorus Immobilization in Micropores of Drinking for Long Term Stability. Environmental Science & Technology 38, 6590 6596. Makris, K.C., Harris, W.G., O'Connor, G.A., Obreza, T.A., E lliott, H.A., 2005. Physicochemical Properties Related to Long Term Phosphorus Retention by Drinking Water Treatment Residuals. Environmental Science & Technology 39, 4280 4289.
116 Malandrino, M., Abollino, O., Giacomino, A., Aceto, M., Mentasti, E., 2006. Ad sorption of heavy metals on vermiculite: Influence of pH and organic ligands. Journal of Colloid and Interface Science 299, 537 546. McLean, J.E., Bledsoe, B.E., 1992. Behavior of Metals in Soils: Ground Water Issue. U.S.EPA. EPA/540/S 92/018. Meng, X., Hu a, Z., Dermatas, D., Wang, W., Hsiu Yu, K., 1998. Immobilization of mercury(II) in contaminated soil with used tire rubber. Journal of Hazardous Materials 57, 231 241. Miretzky, P., Bisinoti, M.C., Jardim, W.F., 2005. Sorption of mercury (II) in Amazon soi ls from column studies. Chemosphere 60, 7 7. Moreno, F.N., Anderson, C.W.N., Stewart, R.B., Robinson, B.H., 2005a. Mercury volatilisation and phytoextraction from base metal mine tailings. Environmental Pollution 136, 341 352. Moreno, F.N., Anderson, C.W.N ., Stewart, R.B., Robinson, B.H., Ghomshei, M., Meech, J.A., 2005b. Induced plant uptake and transport of mercury in the presence of sulphur containing ligands and humic acid. New Phytologist 166, 445 454. Mulligan, C.N., Yong, R.N., Gibbs, B.F., 2001a. An evaluation of technologies for the heavy metal remediation of dredged sediments. Journal of Hazardous Materials 85, 145 163. Mulligan, C.N., Yong, R.N., Gibbs, B.F., 2001b. Remediation technologies for metal contaminated soils and groundwater: an evaluati on. Elsevier Science Bv, pp. 193 207. Novak, J.M., Watts, D.W., 2004. Increasing the phosphorus sorption capacity of southeastern Coastal Plain soils using water treatment residuals. Soil Science 169, 206 214. NRC, 1999. In Groundwater and soil cleanup: im proving management of persistent contaminants. National Academy Press: Washington, D.C., 113 174 pp. Petruzzelli, G., 1997. Soil sorption of heavy metals. In:Ecoloical issues and environmental impact assessment; edited by Cheremisinoff, P.N. Golf Publishi ng Company, Huston Texas, 145 174. Pirrone, N., Cinnirella, S., Feng, X., Finkelman, R., Friedli, H., Leaner, J., Mason, R., Mukherjee, A., Stracher, G., Streets, D., Telmer, K., 2009. Global Mercury Emissions to the Atmosphere from Natural and Anthropogen ic Sources, in: Mason, R., Pirrone, N. (Eds.), Mercury Fate and Transport in the Global Atmosphere. Springer US, pp. 1 47.
117 Rathinasabapathi, B., Babu Raman, S., Kertulis, G., Ma, L., 2006. Arsenic resistant proteobacterium from the phyllosphere of arsenic hyperaccumulating fern (Pteris vittata L.) reduces arsenate to arsenite. Can J. Microbiol 52, 695 700. Roane, T.M., Kellogg, S.T., 1996. Characterization of bacterial communities in heavy metal contaminated soils. Canadian Journal of Microbiology 42, 593 6 03. Rogers, R., McFarlane, J., 1979. Factors influencing the volatilization of mercury from soil. Journal of Environmental Quality. Sarkar, D., Essington, M.E., Misra, K.C., 2000. Adsorption of mercury(II) by kaolinite. Soil Science Society of America Jour nal 64, 1968 1975. Sarkar, D., O'Connor, G.A., Ruple, G.J., Sartain, J.B., 1999. Reuse of Carlton reject water: II. Fate and transport of Ra 226. Soil and Crop Science Society of Florida Proceedings 58, 38 44. Schuster, E., 1991. The behavior of mercury in the soil with special emphasis on complexation and adsorption processes a review of the literature. Water Air and Soil Pollution 56, 667 680. Schwedt, G., 2001. The Essential Guide to Environmental Chemistry. John Wiley and Sons,LTD, Chichester,UK. Serra Hg(II) by Coprecipitation in Sulfate Cement Systems. Environmental Science & Technology 46, 6767 6775. Silveira, M.L., Miyittah, M.K., O'Connor, G.A., 2006. Phosphorus release fr om a manure impacted spodosol: Effects of a water treatment residual. Journal of Environmental Quality 35, 529 541. Skyllberg, U., Bloom, P.R., Qian, J., Lin, C.M., Bleam, W.F., 2006. Complexation of mercury(II) in soil organic matter: EXAFS evidence for l inear two coordination with reduced sulfur groups. Environmental Science & Technology 40, 4174 4180. Skyllberg, U., Drott, A., 2010. Competition between Disordered Iron Sulfide and Natural Organic Matter Associated Thiols for Mercury(II) An EXAFS Study. En vironmental Science & Technology 44, 1254 1259. Sotero Santos, R.B., Rocha, O., Povinelli, J., 2005. Evaluation of water treatment sludges toxicity using the Daphnia bioassay. Water Research 39, 3909 3917. Sparks, D., 1995. Environmental soil chemistry. Ac ademic Press Inc, San Diego, p. 267. Steinnes, E., Friedland, A.J., 2006. Metal contamination of natural surface soils from long range atmospheric transport: Existing and missing knowledge. Environmental Reviews 14, 169 186.
118 Strawn, D.G., Sparks, D.L., 1999. Sorption kinetics of trace elements in soils and soil materials. In:Fate and transport of heavy metals in vadose zone; edited by Selim, H.M and Iskandar, I.K. CRC Press LLC, 1 28. Suer, P., Allard, B., 2003. Mercury trans port and speciation during electrokinetic soil remediation. Water Air and Soil Pollution 143, 99 109. Suer, P., Lifvergren, T., 2003. Mercury contaminated soil remediation by iodide and electroreclamation. Journal of Environmental Engineering Asce 129, 441 446. Sujana, M.G., Thakur, R.S., Rao, S.B., 1998. Removal of fluoride from aqueous solution by using alum sludge. Journal of Colloid and Interface Science 206, 94 101. Sumner, M., Miller, W., 1996. Cation exchange capacity and exchange coefficients. Metho ds of Soil Analysis. Part 3: Chemical Methods., D. L. Sparks, ed., SSSA Book Series, Madison, WI, 1201 1229. Snchez Polo, M., Rivera Adsorption of Cd(II) and Hg(II) on Ozonized Activated Carbons. Environmental Science & Technology 36, 3850 3854. Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential Extraction Procedure for the Speciation of Particulate Trace Metals. Analytical Chemistry, pp. 844 851. Townsend, T., Brajesh, D., Thabet, T., 200 6. Interpretation of Synthetic Precipitation Leaching Procedure (SPLP) Results for Assessing Risk to Groundwater from Land Applied Granular Waste. Environmental Engineering Science, pp. 239 251. US EPA, 1997. Best Management Practices (BMPs) for Soil Treat ment Technologies, US EPA Office of Solid Waste, Washington, DC. US EPA, 2007. Treatment Technologies For Mercury in Soil,Waste, and Water. U.S. Environmental Protection Agency. Office of Superfund Remediation and Technology Innovation, Washington, DC. US EPA, 2013. Final national priority list (NPL) sites by state. http://www.epa.gov/superfund/sites/query/queryhtm/nplfin.htm US EPA, S., 2000. Mercury transport and fate in watersheds. National Center for Environmental Research, pp. 1 8. USDA, 1992. Soil Su rvey Laboratoty Methods Manual. U.S. Department of Agriculture,Washington DC. Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soi l Science 37, 29 38.
119 Wang, J., Deng, B., Chen, H., Wang, X., Zheng, J., 2009. Removal of Aqueous Hg(II) by Polyaniline: Sorption Characteristics and Mechanisms. Environmental Science & Technology 43, 5223 5228. Wang, S., Mulligan, C.N., 2009. Enhanced mobi lization of arsenic and heavy metals from mine tailings by humic acid. Chemosphere 74, 274 279. Wang, Y.D., Greger, M., 2006. Use of iodide to enhance the phytoextraction of mercury contaminated soil. Elsevier Science Bv, pp. 30 39. Wasay, S.A., Barrington S., Tokunaga, S., 2001. Organic Acids for the In Situ Remediation of Soils Polluted by Heavy Metals: Soil Flushing in Columns. Water, Air, and Soil Pollution 127, 301 314. Yang, J.Y., Yang, X.E., He, Z.L., Li, T.Q., Shentu, J.L., Stoffella, P.J., 2006. E ffects of pH, organic acids, and inorganic ions on lead desorption from soils. Environmental Pollution 143, 9 15. Yang, Y.K., Liang, L., Wang, D.Y., 2008. Effect of dissolved organic matter on adsorption and desorption of mercury by soils. Journal of Envir onmental Sciences China 20, 1097 1102. Yin, Y., Allen, H., Huang, C., 1997. Kinetics of Mercury(II) Adsorption and Desorption on Soil Environ. Sci. Technol., pp. 496 503. Yin, Y., Impellitteri, C.A., You, S. J., Allen, H.E., 2002. The importance of organic matter distribution and extract soil:solution ratio on the desorption of heavy metals from soils. Science of The Total Environment 287, 107 119. Yin, Y.J., Allen, H.E., Li, Y.M., Huang, C.P., Sanders, P.F., 1996. Adsorption of mercury(II) by soil: Effects of pH, chloride, and organic matter. Journal of Environmental Quality 25, 837 844. Zhou, L.X., Wong, J.W.C., 2003. Behavior of heavy metals in soil:effect of dissolved organic matter. In: Geochemical and hydrological reactivity of heavy metals in soil; ed ited by Selim, H.M. and Kingery, W.L. CRC Press LLC, 245 269. Zhou, Y. F., Haynes, R.J., 2010. Sorption of Heavy Metals by Inorganic and Organic Components of Solid Wastes: Significance to Use of Wastes as Low Cost Adsorbents and Immobilizing Agents. Crit ical Reviews in Environmental Science and Technology 40, 909 977.
120 BIOGRAPHICAL SKETCH Katherine Y. Deliz Quiones was born in Puerto R ico in 1978. After completing her elementary to high school education at Isabela Puerto Rico she was accepted to the D epartment of Biology at the University of Puerto Rico in Mayaguez (UPRM) Upon completing her Bachelor of Science degree in b io logy she was accepted in to the master program at the same institution. Once s he received her Master of Science from the U niversity of Puerto Rico she started working as a Research Scientist 2 at the Experimental Station of the University of Puerto Rico in Rio Piedras In addition she worked as a Research Scientist for the US Fish and Wildlife Service at Boqueron, Puerto Rico. I n 2006 she was admitted to the D epartment of Environmental Engineering Science s at the University of Florida, Florida, USA, (UFL) wh ere she completed her PhD.