Sorption of U(VI) Onto Manganese Oxides in the Presence of Siderophore Desferrioxamine-B

University of Central Florida STARS

Electronic Theses and Dissertations, 2020-

2020

Sorption of U(VI) onto Manganese Oxides in the Presence of Siderophore Desferrioxamine-B

Morgan Snyder University of Central Florida

Part of the Chemistry Commons Find similar works at: https://stars.library.ucf.edu/etd2020 University of Central Florida Libraries http://library.ucf.edu

This Masters Thesis (Open Access) is brought to you for free and open access by STARS. It has been accepted for inclusion in Electronic Theses and Dissertations, 2020- by an authorized administrator of STARS. For more information, please contact [email protected].

STARS Citation Snyder, Morgan, "Sorption of U(VI) onto Manganese Oxides in the Presence of Siderophore Desferrioxamine-B" (2020). Electronic Theses and Dissertations, 2020-. 815. https://stars.library.ucf.edu/etd2020/815

SORPTION OF U(VI) ON MANGANESE OXIDES IN THE PRESENCE OF SIDEROPHORE DESFERRIOXAMINE-B

MORGAN LEIGH SNYDER B.S. Clemson University, 2016

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in the Department of Chemistry in the College of Science at the University of Central Florida Orlando, Florida

Fall Term 2020

ABSTRACT

Anthropogenic activities, such as uranium mining and the production of nuclear energy and nuclear weapons, have led to significant uranium contamination of the environment, leaving populations vulnerable to negative health effects. In order to effectively remediate uranium contamination, the mobility of uranium in the environment and its natural attenuation through sorption with active mineral phases, needs to be studied in depth first. This study aims to determine how U(VI) mobility is affected by manganese oxides in the presence and absence of siderophore Desferrioxamine-B (DFOB). Experiments focused on two common manganese minerals, pyrolusite (β-MnO2) and manganite (γ-MnOOH).

Adsorption experiments with U(VI)-DFOB were performed with pyrolusite at pH 3.5, 6,

8. Control adsorption experiments were also performed with U(VI) and DFOB in order to elucidate the role each component plays in the sorption process. The concentrations ranged from

6 µM to 78µM. For manganite, adsorption experiments were performed at pH 6 and 8 with

U(VI)-DFOB and with controls of U(VI) and DFOB and a comparison was drawn with the corresponding values from pyrolusite.

For pyrolusite, the presence of DFOB largely increased the U(VI) adsorption for all pH values. U(VI) adsorption increased as pH increased due to DFOB electrostatically adsorbing to pyrolusite while complexed to U(VI). In addition, DFOB also adsorbed separately from U(VI) leading to hydrolysis and degradation of the molecule that promoted dissolution of the mineral.

For manganite, U(VI) showed high adsorption levels in the control experiment, and these rates were unchanged when DFOB was added, leading to the conclusion that DFOB does not affect

U(VI) interaction with that substrate. Also, DFOB showed lower adsorption and smaller amounts of hydrolysis with manganite when compared to pyrolusite.

iii

ACKNOWLEDGMENTS

I would like to thank Dr. Vasileios Anagnostopoulos for giving me the opportunity to conduct research in his laboratory, for serving as the Chair of my thesis committee, and for offering encouragement and guidance as I worked on this project. I would like to thank Dr.

Melanie Beazley and Dr. Michael Hampton for serving on my thesis committee. Thank you to

Jordan Stanberry, Ilana Szlamkowicz, and Travis Hager for their assistance with various experiments and analysis related to this project. Lastly, I would like to thank my family and all of my fellow lab members for their support and for making this experience memorable.

TABLE OF CONTENTS

LIST OF FIGURES ...... vii

LIST OF TABLES ...... xi

LIST OF ACRONYMS (or) ABBREVIATIONS ...... xii

CHAPTER ONE: INTRODUCTION ...... 1

1.1. Uranium ...... 1

1.2. Siderophores ...... 5

1.3. Manganese Minerals ...... 8

1.4. Adsorption onto Mineral Surfaces ...... 10

1.5. Research Objectives ...... 12

CHAPTER TWO: METHODOLOGY ...... 14

2.1. Mineral Synthesis ...... 14

2.2. Uranium and Siderophore Solutions Preparation ...... 16

2.3. Batch Sorption Experiments...... 18

2.4. Analysis Methods ...... 20

2.4.1. Determination of Mn(II) and U(VI) ...... 20

2.4.2. U(VI)-DFOB Complex Analysis ...... 21

2.4.3. Alternative Methods for Quantifying DFOB...... 23

2.4.4. Determination of Acetate and Succinate ...... 26

CHAPTER THREE: RESULTS ...... 30

3.1. Uranium Sorption on Pyrolusite in the Presence of DFOB ...... 30

3.1.1. Effect of pH ...... 30

3.1.2. Mechanisms for DFOB and U(VI) Sorption ...... 35

3.1.2.1. Reported Pathways for DFOB Adsorption ...... 35

3.1.2.2. Adsorption Using Hydroxamate Groups ...... 37

3.1.2.3. Adsorption Using Terminal Amine Group ...... 44

3.1.3 Isotherms ...... 50

3.2 Uranium Sorption on Manganite in the Presence of DFOB...... 54

3.2.1. Effect of pH ...... 54

3.2.2. Mechanisms for U(VI) and DFOB Adsorption on Manganite ...... 57

CHAPTER FOUR: CONCLUSION ...... 62

4.1. Closing Remarks ...... 62

4.2. Future Work ...... 64

LIST OF REFERENCES ...... 66

LIST OF FIGURES

Figure 1: Distributions of solution phase U(VI) species in equilibrium with atmospheric CO2 as a

2+ 2- function of pH. UO2 concentration set at 10 µM and CO3 concentration set at 0.01 M.

Ionic strength set to zero...... 4

Figure 2: Desferrioxamine B in the form of the mesylate salt37...... 5

Figure 3: Crystal structure of pyrolusite. Large red spheres are oxygen atoms and small purple

68 spheres are manganese (IV) atoms that lie inside of the MnO6 octahedra ...... 9

Figure 4: Cross section of metal-oxide mineral surface. a) unhydrated surface, b) H2O molecules

coordinate with surface metal ions, c) Dissociation of protons from adsorbed H2O

molecules creates a hydroxylated surface, d) Hydroxylated surface can attract other water

molecules or cationic species74...... 11

Figure 5: IR spectrum of synthesized manganite ...... 15

Figure 6: XRD of synthesized manganite ...... 15

Figure 7: UV-Vis spectrum for the U(VI)-DFOB complex in the range of 350 – 900 nm ...... 17

Figure 8: Speciation of U(VI)-DFOB complex in aqueous solution. Modeled using Visual

MINTEQ...... 17

Figure 9: Calibration Curve for Mn(II) ...... 20

Figure 10: Calibration Curve for U(VI) ...... 21

Figure 11: UV-Vis calibration curve for Fe(III)-DFOB at 428 nm ...... 22

Figure 12: UV-Vis Spectrum of Fe(III)-DFOB at pH 6 ...... 23

Figure 13: UV-Vis Spectrum of free DFOB at pH 6 ...... 24

Figure 14: UV-Vis Calibration curve for free DFOB at 195 nm ...... 25

vii

Figure 15: IC calibration curve for acetate using a 5 mM NaOH eluent, elution time: 6.125 min

...... 27

Figure 16: Calibration curve for succinate using a 35 mM NaOH eluent, elution time: 4.900 min

...... 27

Figure 17: Chromatogram of succinate calibration standard (5 PPM) with 35 mM NaOH eluent,

elution time: 4.900 min...... 28

Figure 18: Chromatogram of acetate calibration standard (5 PPM) with 5 mM NaOH eluent,

elution time: 6.123 min...... 29

Figure 19: DFOB sequestration a function of pH by pyrolusite at differing pH values. 37 uM

initial concentration of U(VI) or U-DFOB complex. Error bars represent relative standard

deviation from triplicate samples...... 31

Figure 20: Uranium sorption vs pH: The percentages of Uranium that is no longer present in the

aqueous phase after introduction with pyrolusite at differing pH values. Error bars

represent relative standard deviation from triplicate samples...... 32

Figure 21: Active groups for binding of DFOB...... 36

Figure 22: Illustration of proposed mechanisms for DFOB adsorption. Proposed mechanism #1

orients the hydroxamate groups towards metal centers to participate in inner-sphere

complexation. No sequestration of U(VI) can occur through this mechanism. Proposed

mechanism #2 involves orientation of protonated terminal amine groups towards the

surface, leaving hydroxamate groups open to complexation with U(VI)...... 37

Figure 23: Hydrolysis/oxidative products of DFOB as proposed by Simanova et. al. (2010)55... 39

Figure 24: DFOB degradation pathway proposed by Winkelmann (1999)95...... 40

viii

Figure 25: Acetate and succinate released into solution. Error bars represent relative standard

deviation from triplicate samples...... 42

Figure 26: Concentration of Mn released from the inner-sphere adsorption of DFOB at pH 6 as a

function of hydrolyzed DFOB. Hydrolyzed DFOB concentration was set equal to

concentration of acetate in samples. Linear fit was applied to obtain the slope. Error bars

represent relative standard deviation from triplicate samples...... 44

Figure 27: DFOB and U(VI) adsorption onto pyrolusite with differing ionic strengths. Ionic

strength is plotted as the -log of [NaClO4]. Error bars represent relative standard

deviation from triplicate samples...... 48

Figure 28: Sorption of DFOB at pH 6 plotted as % of initial DFOB concentration (µM) adsorbed

vs. initial DFOB concentration (µM). Error bars represent relative standard deviation

from triplicate samples ...... 51

Figure 29: Sorption of U(VI) at pH 6 plotted as % of initial concentration (PPM) sorbed vs.

initial U(VI) concentration (PPM). Error bars represent relative standard deviation from

triplicate samples ...... 51

Figure 30: Isotherm of DFOB sorption excluding hydrolyzed DFOB at pH 6. Q is mg of DFOB

adsorbed per gram of pyrolusite. Error bars represent relative standard deviation from

triplicate samples...... 53

Figure 31: Isotherm of U(VI) sorption enhanced by DFOB at pH 6. Q is mg of U(VI) adsorbed

per gram of pyrolusite. Error bars represent relative standard deviation from triplicate

samples...... 54

Figure 32: U(VI) adsorption as a function of pH in the presence and absence of DFOB. Error

bars represent relative standard deviation from triplicate samples...... 55

Figure 33: DFOB sequestration a function of pH by manganite at different pH values. 37 uM

initial concentration of U(VI) or U-DFOB complex. Error bars represent relative standard

deviation from triplicate samples...... 56

Figure 34: Possible adsorption mechanisms of metal-ligand complexes. Type I involves

adsorption through the metal, Type II involves adsorption only through ligand active

groups, and Type III involves the interaction of both metal and ligand active groups to the

surface. “S” represents surface functional sites...... 58

+1/2 Figure 35: Secondary amine group in DFOB hydrogen bonding with OH2 surface groups on

manganite...... 60

Figure 36: Derivatization of DFOB to DFOD1 by replacement of terminal amine proton with

acetyl group110...... 65

LIST OF TABLES

Table 1: Summary of uranium remediation methods ...... 3

Table 2: Selected Metal-DFOB complex stability constants ...... 6

Table 3: Logarithm of stability constants for potential aqueous species...... 18

Table 4: Definition of terms for each portion of DFOB involved in separate proposed

mechanisms ...... 45

Table 5: Mass balance of DFOB after sorption onto pyrolusite at pH 6...... 45

Table 6: Mass balance of DFOB at pH 3.5 and 8...... 46

Table 7: U(VI) extracted from surface of pyrolusite with deionized water and 0.01 M HCl ...... 49

Table 8: Concentration of acetate detected in manganite suspensions. DFOB initial concentration

was 37 µM added to 10 mg of manganite. Experiments were performed in triplicate. .... 57

LIST OF ACRONYMS (or) ABBREVIATIONS

CAS – Chrome Azurol S

DFOB - Desferrioxamine-B

IC – Ion Chromatography

ICP-MS - Inductively Coupled Plasma-Mass Spectrometry

MCF - Materials Characterization Facility

PPM – Parts Per Million

PPB – Parts Per Billion

UV-Vis - Ultraviolet-Visible

WHO – World Health Organization

XRD – X-Ray Diffraction

xii

CHAPTER ONE: INTRODUCTION

1.1. Uranium

Uranium is a naturally occurring radioactive element, first discovered in 1789. The radioactive nature of uranium was discovered by Henry Becquerel1 in 1896. Uranium exists naturally as one of three isotopes; 238U is the most abundant isotope (99.2739-99.2752%) , 235U is the second most abundant isotope (0.7198-0.7202%), and trace amounts of uranium exist as

234U. Uranium exhibits a weak radioactivity and has a half-life of 4.46*109 years(238U)2.

Since its discovery, uranium has been used for many applications. Most notable of which is in the commercial production of energy. Uranium nuclear reactors account for 14% of the world’s electricity, with 20% of U.S. electricity is supplied by a total of 100 nuclear reactors3, 4.

To meet this demand for uranium, there is a reliance on mining methods in order to acquire uranium from natural ores. The nuclear fuel cycle requires multiple steps, each of which contribute to contamination in the soil or ground water. The first step is mining, which can be done either on the surface, underground, or through solution mining. The process continues through milling (physical extraction of uranium from the ore), enrichment (to reach 3.5-5% of

235U), fuel fabrication, power generation, and storage and disposal of used fuel5.

Significant contamination has occurred as a result of the mining of uranium as well as its use in various industries. Among the largest sources of this contamination is leaching from mine tailings, natural deposits, uranium combustion products, emissions from the nuclear industry, and corrosion of phosphate fertilizers containing uranium6-8. One notable contamination incident happened on July 16, 1979. The Church Rock uranium spill resulted in 1100 tons of solid waste and 93 million gallons of radioactive mill waste entering the Puerco River raising the

1 concentration of uranium to 7000 times the acceptable limit9. The World Health Organization

(WHO) lists the maximum acceptable concentration in water as 50 µg/ L10 and the tolerable daily intake of soluble uranium at 0.6 µg/kg of body weight11. In addition to the intake of soluble uranium, inhalation of respirable air-borne uranium can be a major pathway into the human body12. Inhaled uranium (1-10 µm in size) can be retained in the lungs leading to lung cancer12.

Uranium has a high binding affinity for biomolecules and is a known mutagen and an ephrotoxin12. Inhaled uranium becomes deposited in the lungs and can be absorbed into the blood stream. Uranium in the blood stream ultimately reaches the kidneys resulting in renal injury and in severe cases of kidney failure and death13.

The many observed negative health impacts of uranium contamination support the need for effective remediation methods. Adsorption removes contaminants through their adhesion to a sorbate. Membrane filtration works by using membranes that can selectively remove metals based on either charge or molecular weight. Biological methods rely on the use of bacteria that can either absorb U(VI) or can reduce it into the less mobile U(IV) state. Phytoremediation is a method that can be used to treat large areas of contamination due to some plant’s ability to adsorb and accumulate large amounts of chemical elements. Table 1 summarizes the advantages of each technique and provides examples of the versatility of each method.

Table 1: Summary of uranium remediation methods

Method Advantages Examples

Adsorption Reusable, cost effective, easy Possible sorbates: Agricultural operation14 wastes15-17, Activated carbon18, Hydrothermal carbon19, Carbon nanotubes20, Graphene oxide21 Membrane High efficiency, easy operation, Nanofiltration22, Ultrafiltration23, Filtration space saving12 Reverse Osmosis12 Biological Inexpensive and readily available Pseudomonas24, Denitrification, Methods biological materials12 ferric iron-reducing, sulphate- reducing bacteria25 Phytoremediation Economically viable, effective, no Sunflower, Indian mustard26, secondary pollution, small Willow moss, celery27 disturbance to environment12.

Adsorption is a very important technique for remediation as well as natural attenuation reasons. Nevertheless, the fate of uranium and its retention by natural substrates depend on its oxidation state and speciation. U(IV) and U(VI) are the two most important oxidation states28 with hexavalent uranium compounds being much more soluble than tetravalent compounds.

2+ 5 Under oxidizing conditions, U(VI) exists as the uranyl ion (UO2) , which is soluble in water .

The enhanced solubility of the hexavalent species poses a more significant health risk29 than its tetravalent counterpart. Speciation of uranyl in aqueous solution is dependent on pH and the presence of ligands. The soluble uranium species differ whether in acidic or alkaline

2+ + environments (Figure 1). Below pH 6.5, the uranyl ion (UO2 ) and uranyl hydroxide (UO2OH ,

UO2(OH)2) species are dominant while above pH 6.5, uranyl-carbonato species dominate

0 2- 4- - (UO2CO3 , UO2(CO3)2 , UO2(CO3)3 , and (UO2)2CO3(OH)3 ).

Figure 1: Distributions of solution phase U(VI) species in equilibrium with atmospheric CO2 as a 2+ 2- function of pH. UO2 concentration set at 10 µM and CO3 concentration set at 0.01 M. Ionic strength set to zero.

Among potential substrates for use in adsorption remediation are naturally occurring minerals. Iron oxides have shown to be effective at decreasing dissolved U(VI) concentrations30.

Other minerals like quartz, kaolinite, montmorillonite, and smectite also demonstrate high capacity for uranium. It is well documented that complexation with organic ligands can affect the sorption of metals onto mineral surfaces. Uranium has a strong affinity for organic matter, with soluble organic matter being able to complex with U(VI) and affect solubility and the ability to adsorb onto minerals31, 32. Studies performed using quartz, albite, muscovite, and phyllite noted an increase in the sorption of U(VI) when humic acid was present33. Another study using hematite (Fe2O3) as a model for metal oxides, reported the enhancement of uranium sorption at low pH values in the presence of humic acid34. The proposed mechanism for this enhancement

4 was through complexation of uranium to the organic ligand and then complexation with the surface in the order of substrate/ligand/metal34. The known ability of U(VI) to adsorb onto minerals surfaces and the effects of organic ligands is the basis of this study. One organic ligand of interest, siderophore Desferrioxamine B (DFOB), will be introduced with U(VI) to investigate the role it plays on the mobility of U(VI) in the presence of manganese minerals

1.2. Siderophores

DFOB belongs to a class of molecules called siderophores that have a low-molecular weight and exhibit a strong binding affinity for iron35. Siderophores are secreted by soil microbes and plants in order to sequester iron that is otherwise insoluble in soil conditions. Iron plays a major role in many important biological processes; however, its existence as immobile ferric complexes in oxidizing environments renders it unavailable for use. In response, siderophores are released to interact with the ferric complexes and mobilize iron for uptake into microbe and plant cells36. Siderophores contain metal binding groups, among which are catecholates and hydroxamates, and most contain 6 coordinating atoms and coordinate with iron in a pseudo octahedral geometry35.

Figure 2: Desferrioxamine B in the form of the mesylate salt37.

38 The speciation of DFOB in solution (uncomplexed) is based on the following pKa values :

+ + H4DFOB ↔ H + H3DFOB pKa = 8.39

+ - H3DFOB ↔ H + H2DFOB pKa= 9.03

- + 2- H2DFOB ↔ H + HDFOB pKa= 9.70

2- + 3- HDFOB ↔ H + DFOB pKa> 11

DFOB has an exceptionally high affinity for Fe(III) and it coordinates with iron using three hydroxamate groups39. In addition to Fe(III), DFOB has exhibited a high capacity to form complexes with many other metals. Some of these metal-DFOB complexes are listed in Table 2.

Table 2: Selected Metal-DFOB complex stability constants

Metal – DFOB Complex Stability Constant (log K) Co(III) 37.540 Al(III) 24.541 Ga(III) 28.741 In(III) 21.441 Mn(III) 29.942 Fe(III) 32.0243 Cu(II) 13.7344 Zn(II) 9.5544 Cd(II) 1044 U(VI) 17.137 Pu(IV) 30.845

When DFOB is complexed with metals, it can affect their mobility by enhancing or hindering their adsorption to minerals. In the presence of DFOB, Cd and Zn have an enhanced sorption onto montmorillonite, likely due to the positive Cd(II)-DFOB and Zn(II)-DFOB complexes interacting with the negative mineral surface46. With Pb and Cd, DFOB decreased the

6 adsorption on kaolinite47. For Cu(II), there was an observed enhancement of sorption onto montmorillonite and a hindrance of sorption with kaolinite48. Zn(II) and Cd(II) were also studied with muscovite and phlogopite and had a diminished adsorption in the presence of DFOB49. The effect of DFOB on metal mobility is pertinent due to the co-existence of ligand and metal ions in the natural environment.

DFOB has also been reported to adsorb to the surface of many minerals promoting dissolution. This has been observed using montmorillonite, hematite, palygorskite, kaolinite, and sepiolite50-54. One well studied group of minerals are manganese oxides and hydroxides.

Hausmanite (Mn(II,III)2O4), birnessite ((K,Na,Ca)Mn(III,IV)2O4), and manganite

(Mn(III)O(OH)) are well studied with DFOB and mechanisms have been proposed for the sorption and subsequent release of Mn from the mineral phase55-57. These proposed mechanisms involve DFOB’s hydroxamate groups that can inner-spherically complex with metal centers on the mineral surface50. In addition, there are reported to be contributions to sorption from DFOB’s protonated amine group and secondary amide groups50. These previously proposed mechanisms will be exploited in our study in an attempt to describe the mechanism for adsorption onto pyrolusite and manganite.

The shown ability of DFOB to form complexes with many metals as well as influence their mobility makes it a good choice for this study. DFOB is present in the natural environment and it is important to understand the role it plays in potential remediation of U(VI). While the

U(VI)-DFOB complex has been studied in terms of stability, there is more information to be gained by studying how this complex behaves with various minerals.

1.3. Manganese Minerals

Manganese is the 10th most abundant element in the Earth’s crust and the second most abundant heavy metal58. Manganese can exist naturally as Mn(II), Mn(III), and Mn(IV). This ability to exist in multiple oxidation states combined with high mobility of Mn(II) and many different environmental conditions leads to significant transformations resulting in over 100 known manganese containing minerals59. These include manganese oxides and hydroxides and one unique characteristic of these minerals is their sorption capacity, which is much higher in proportion to its concentration60. They have a much higher capacity for sorption compared to other naturally occurring minerals, and this gives them the ability to sequester and control heavy metals with great efficiency61. An example of this characteristic is that adsorption by Mn oxides is considered to be one of the most important mechanisms responsible for controlling heavy metal concentration in the world’s oceans62. Mn oxide minerals are the most abundant in today’s ore deposits, and recently they have been used in applications like catalysis, batteries, plant fertilizers and steel63. Manganese oxide minerals are a major source of manganese in the aquatic environment and are most commonly in the +2 and +4 oxidation states64. Manganese oxides and hydroxides in the +3 oxidation state are generally insoluble in aqueous environments65. Any

Mn(III) present in aqueous solution will disproportionate to Mn(II) and Mn(IV) and

66 subsequently to MnO2(s) according to the following reactions :

2Mn(III) Mn(II) + Mn(IV)

Mn(IV) + 2H2O MnO2(s)

Mn(II) is the dominant species of manganese in solution (around 91%), however manganese in other oxidation forms can be present in solution when complexed to other species like carbonates, sulphates, and organic ligands64.

Of the different Mn oxide/hydroxide structures, the two major structural classifications are tunneled and layered; both types having the MnO6 octahedron as the building block of the structure. The tunneled structure contains single, double, or triple chains of edge sharing octahedra forming tunnels that can accommodate cation species67. Layered structures contain sheets of the edge sharing octahedra with the interlayer able to accommodate cation species60.

Tunneled structures are used exclusively in the research presented in this thesis, specifically β-

MnO2 (pyrolusite) and γ-MnOOH (manganite). Pyrolusite is the most abundant and most stable polymorph of MnO2 and consists of single chains of Mn(IV)O2 octahedra that form tunnels that are 1 octahedron by 1 octahedron in cross section67. This pyrolusite structure is pictured in

Figure 3.

Figure 3: Crystal structure of pyrolusite. Large red spheres are oxygen atoms and small purple 68 spheres are manganese (IV) atoms that lie inside of the MnO6 octahedra .

Manganite is similar in structure (tunneled structure with single chains of octahedra) with the exception that manganese is the in +3 oxidation state and hydroxyl anions replace half of the oxygen atoms. In addition there is significant distortion of the octahedra due to Jahn-Teller effects60. Both pyrolusite and manganite have demonstrated ability to sequester a wide variety of metals and organic ligands. They both were able to sequester Cr(VI) through inner sphere complexation at pH lower than 6 and outer sphere complexes at pH higher than 669. Pyrolusite has also been shown to adsorb organic compounds like humic acid, tannic acid, and other cationic molecules (safranin and ferroin)70. There have also been extensive studies on both minerals for sorption of Zn2+, Pb2+, and Mg2+ along with other metals71-73. The abundance as well as the sorption characteristics of pyrolusite and manganite make them good candidates as substrates to study uranium sorption.

1.4. Adsorption onto Mineral Surfaces

In addition to understanding the nature and structure of each constituent of this study, an understanding of the adsorption process is needed to effectively categorize the reactions taking place. Adsorption reactions depend on a multitude of factors including pH, identity and concentration of sorbate, concentration of substrate, and the presence of competing species74.

Sorption of ions onto mineral surfaces can be categorized into two interactions, electrostatic or chemical. Chemical adsorption (inner sphere sorption) is insensitive to surface charge and ionic strength and involves chemical bonding of sorbate to mineral surface. Electrostatic adsorption

(outer sphere sorption) takes place through coulombic attraction between oppositely charged sites on the surface and the charged sorbate. It is very reliant on pH, which determines the charge of the surface functional groups75. Metal oxide minerals specifically have hydroxyl functional

10 groups that form when water sorbs onto the surface and then subsequently hydrolyzes74. Figure

4 illustrates this formation of reactive functional units bound to the mineral structure.

Figure 4: Cross section of metal-oxide mineral surface. a) unhydrated surface, b) H2O molecules coordinate with surface metal ions, c) Dissociation of protons from adsorbed H2O molecules creates a hydroxylated surface, d) Hydroxylated surface can attract other water molecules or cationic species74.

Due to the functional groups, the solid substrates acquire an electrical charge when they come into contact with an aqueous phase. This electrical charge can attract counterions and could lead to sorption of charged species. For metal oxides, a variable surface charge approach is taken where the charge develops as a result of the dissociation of the functional groups74. The hydroxyl

11 functional groups can become protonated or deprotonated in order to produce positive or negative charges. The reactions to illustrate this are (X=metal)76:

+ 0 + XOH2 = XOH + H

XOH0 = XO- + H+

The control of the surface charge is therefore dependent on the pH of the solution in contact with the mineral. At low pH values, the protonated form of the functional group dominates (positive) and at high pH values, the deprotonated form of the functional group dominates (negative)74. It is important to note that at every pH value, there is a presence of positive, negative, and neutral sites; however, there is a distribution that favors either positive or negative charges. At a specific pH value, the point of zero charge occurs where the surface has a

+ net zero charge as the number of positively and negatively charged sites become equal (XOH2 =

XO-)77. For pyrolusite and manganite, this point of zero charge is pH 6 and 8, respectively78, 79.

This dependence of surface charge on pH directly relates to the observed sorption of cations, as typically cation adsorption increases with increasing pH and anion adsorption decreases with increasing pH80. Other factors that can affect the sorption of metals onto minerals is the complexation by dissolved ligands and the presence of competing sorbate species that can compete for surface sites30.

1.5. Research Objectives

While there has been extensive research on the topic of siderophores and their complexation properties with metals, there are still knowledge gaps within the environmental mobility of these complexes. For uranium specifically, there are only a couple of publications that report the existence of U(VI)-DFOB complex, but no other information is available about the

12 environmental stability of this complex or interaction with natural substrates. For reasons related to natural attenuation and remediation, it is crucial to understand the role that siderophores can play in the mobility of uranium with common minerals like pyrolusite and manganite. In order to fully understand the mobility effects of complexation with DFOB, the following research objectives were set:

1. To determine how uranium adsorbs to the surface of pyrolusite in the absence and

presence of DFOB.

2. To investigate potential mechanisms of DFOB and uranium sorption. This includes

understanding potential degradation and sorption mechanisms of DFOB, a process that is

not currently fully understood.

3. To compare the findings of pyrolusite to that of the more studied manganite. This

includes drawing comparisons about the sorption mechanisms of U(VI) and DFOB.

CHAPTER TWO: METHODOLOGY

2.1. Mineral Synthesis

The pyrolusite, β-MnO2, used in this study was procured through a commercial vendor

(Alfa Aesar, 98% purity) and was in the form of a fine black/gray powder. The manganite used

81 in this study was synthesized according to a published procedure . 11 g of MnSO4 (99%, Acros

Organics) were added to 1750 mL of degassed ultrapure water (resistivity of 18.2 MΩ-cm,

o Barnstead Nanopure). This solution was then heated to 50 C under continuous stirring. Then 6 mL of 30% H2O2 (Alfa Aesar) was added, followed by the slow addition of 200 mL of 0.6235 M

o NH4OH (Fisher). This solution was covered with a watch glass and boiled at 95 C for approximately 6 hours under continuous stirring. After 6 hours, the solution was filtered while still hot with a 0.22 μm nanopore filter (VWR) and was rinsed with hot, deionized water. The solid precipitate was dried at 60oC under approximately 20 mmHg pressure for 72 hours. This dried precipitate was an ashy, brown powder and the structure of manganite was confirmed using infrared spectroscopy (FT/IR-6600, Jasco) and X-Ray diffraction (XRD). The analyses took place at the Materials Characterization Facility (MCF) located at the UCF Main Campus. The

XRD (Empyrean, PANalytical) spectra were collected from 10 to 80° 2θ, with a scan speed of

0.063865 °/s, at a voltage of 45 kV and current 40 mA, using Cu-Kα (λ = 0.15405980 nm). The

IR and XRD in Figure 5 and Figure 6 and is confirmed by published XRD82 and IR83 spectra of manganite. The IR spectrum revealed 3 characteristic OH peaks in the 1050-1150 cm-1 range and a broad OH peak from the 2400-2900 cm-1 range. XRD characteristic peaks of MnOOH were present at 2θ values of 26, 34, 38, and 5583.

Figure 5: IR spectrum of synthesized manganite

Figure 6: XRD of synthesized manganite

2.2. Uranium and Siderophore Solutions Preparation

The DFOB used came in the form of the mesylate salt (C25H48N6O8 • CH4O3S, Acros

Organics, 95%). A stock solution of DFOB (1 mM) was created by adding 0.0657 g of DFOB mesylate salt to 100 mL of ultrapure water at a final pH of 5.5. DFOB was determined experimentally to be stable down to pH 3 and up to pH 1037, which encompasses all pH values that will be used throughout the course of this study. The stock DFOB solution was stored at 4oC when not in use.

A stock solution of U(VI) (1 mM) was prepared by adding 0.0504 g of uranyl nitrate

(UO2(NO3)2 • 6H2O, Fisher Scientific, 98%) to 100 mL of ultrapure water. A U(VI)-DFOB complex was formed by mixing stock aqueous solutions of 1 mM DFOB-mesylate salt

(C25H48N6O8 • CH4O3S, Acros Organics, 95%) and 1 mM of uranium (UO2(NO3)2 • 6H2O,

Fisher Scientific, 98%) at pH 6. This complex is reported to be stable from the range of pH 3-

1137. The existence of U(VI)-DFOB complex was identified by two characteristic peaks of the complex at 490 and 380 nm37 using a double-beam UV-Vis spectrophotometer (Thermo

Scientific, Genesys 50), as seen in Figure 7. The speciation for the U(VI)-DFOB complex was developed with Visual MINTEQ (Figure 8) with the stability constants listed in Table 3.

Figure 7: UV-Vis spectrum for the U(VI)-DFOB complex in the range of 350 – 900 nm

Figure 8: Speciation of U(VI)-DFOB complex in aqueous solution. Modeled using Visual MINTEQ.

Table 3: Logarithm of stability constants for potential aqueous species.

Species Log K

(DFOB)3- + H+ ↔ H(DFOB)2- 10.90a

3- + - a (DFOB) + 2H ↔ H2(DFOB) 20.48

3- + a (DFOB) + 3H ↔ H3(DFOB) 29.48

3- + + a (DFOB) + 4H ↔ H4(DFOB) 37.85

+2 + + b UO2 + H4(DFOB) ↔ UO2DFOBH2 23.78

+ + b UO2OH + H4(DFOB) ↔ UO2DFOBH 18.40

+ - b UO2(OH)2 + H4(DFOB) ↔ UO2OHDFOBH 23.43 a Stability constants determined at background electrolyte of 0.1 M; Hepinstall et. al. (2005)47 b Wolff-Boenisch and Traina (2006)84

2.3. Batch Sorption Experiments

Batch sorption experiments were performed by bringing in contact initially 10 mg of each mineral and 10 mL of ultra-pure water of a specific pH for 48h in order to minimize pH shift during the actual sorption experiment (which will include DFOB and U(VI)-DFOB complex).

The pH values used throughout the course of this study were pH 3.5, 6, and 8. Minor additions of dilute HCl and NaOH were done throughout the 48h to ensure the suspensions remain in the desired pH values. The process was repeated until there was no further observed shift in pH. The reason that this method was employed, as opposed to using buffers to keep the pH constant, is due to the fact that buffers interfere with the analysis of DFOB in the supernatant, as it will be explained further at Section 2.4.2).

After equilibration was achieved U-DFOB was introduced to the vials. Sorption experiments were performed for both pyrolusite and manganite at a constant concentration of 36

µM of the U(VI)-DFOB complex and at pH 3.5, 6, and 8. Controls were also performed in parallel spiked only with U(VI) or DFOB of the same concentration. Ionic strength experiments were performed by bringing in contact 10 mg of each mineral with 10mL ml of U-DFOB 36 µM, pH 6 at three different ionic strengths: 0.001, 0.01, and 1M of NaClO4 (Fisher, 99%). Sodium perchlorate was selected as an inert electrolyte. For isotherms, the equilibrated suspensions of pyrolusite were spiked with U(VI)-DFOB stock solution to create final concentrations of 6, 12,

36, 60, and 78 µM, pH 6. Moreover, isotherm experiments containing only 1 mM UO2(NO3)2

+ and 1 mM H4DFOB were performed in parallel. All samples were allowed to spin in a tube revolver (Fisher) at 15 rpm for 24 hours and then the suspensions were centrifuged for 10 minutes (5,000 rpm, Thermo Sorvall ST 16). Aliquots were isolated from the supernatant for the determination of Mn, U(VI), and DFOB residual concentrations in the supernatant.

Recovery experiments were performed for both minerals by taking samples that had already undergone sorption of the U(VI)-DFOB complex and centrifuging them at 5,000 rpm for

10 minutes, then discarding the supernatant. The solid substrate was dried in a vacuum oven at

60°C for 48 hrs. Then, 10 mL of either ultra-pure water or 0.01 M HCl were added to the dried substrates and the new suspensions were allowed to mix over the course of 1 week and then sampled, as described above.

2.4. Analysis Methods

2.4.1. Determination of Mn(II) and U(VI)

Residual concentration of U(VI) and Mn concentration in the supernatant were determined using Inductively Coupled Plasma – Mass Spectroscopy (ICP-MS - Thermo, iCAP

RQ) The calibration standards for U and Mn were prepared from standard concentrated solutions

(plasma standard solution in 5% HNO3, SpecPure) with a range of concentrations of 0.1-19 ppb and 10-500 ppb, respectively. This method detects total uranium concentration (complexed and un-complexed). Samples were diluted using plasma grade 2% HNO3 (Thermo). Calibration curves for both elements are shown in Figures 9 and 10.

y= 723,590x + 530,068 2 R = 1.0

Figure 9: Calibration Curve for Mn(II). Standards range from 10 ppb to 500 ppb.

y= 900,180x – 110,124 2 R = 1.0

Figure 10: Calibration Curve for U(VI). Standards range from 0.1 ppb to 19 ppb.

2.4.2. U(VI)-DFOB Complex Analysis

The concentration of DFOB present in the supernatant was determined spectrophotometrically by utraviolet-visible (UV-vis) spectroscopy (Thermo Scientific, Genesys

50). This procedure was reported in previous literature and adapted to this experiment85. A solution of 1.8 mM FeCl3 (Alfa Aesar, 98%) in 0.1M HCl was prepared by adding 0.0073 g of

FeCl3 to 25 mL of 0.1 M HCl. In 4.5 mL polystyrene cuvettes, 100 µL of the FeCl3 solution, aliquots of supernatant from sorption experiments, and water was added to reach a final volume of 2.6 mL. The volume of supernatant sampled was determined in order to make dilutions that would place the expected concentration of DFOB remaining in the supernatant into the range of the calibration curve (Figure 11). These samples were capped and allowed to sit for 10 minutes before analysis at 428 nm. This procedure determines [DFOB]Total in the supernatant as the

Fe(III) will complex with any non-complexed DFOB in solution as well as any DFOB complexed with uranium, since Fe(III)-DFOB (log K= 32.02 ± 0.543) is a stronger complex than

U(VI)-DFOB (log K= 17.1 ± 0.237). This was verified experimentally by performing this same analysis on U(VI)-DFOB stock solutions of known concentration, which yielded 100% DFOB recovery. This procedure also requires an absence of buffers since FeCl3 formed a complex in the same wavelength range when added to organic buffers (HEPES and PIPES).

Figure 11: UV-Vis calibration curve for Fe(III)-DFOB at 428 nm

Calibration standards of Fe(III)-DFOB were made from adding stock 1mM DFOB with at least a 1:1 molar equivalent of FeCl3 and diluting to achieve a calibration range of 1.9 µM to

13.5 µM of Fe(III)DFOB. This calibration curve is pictured in Figure 10. The Fe(III)DFOB complex was characterized by the presence of a peak maximum at 428 nm consistent with

86 literature values . In addition, the absorbance spectra of UO2(NO3)2 and were collected to

22 confirm no interferences from that species in the experimentally relevant concentration ranges.

The spectrum for Fe(III)DFOB is shown in Figure 12.

UV-Vis Spectrum for Fe(III)-DFOB - pH 6

0.2

0.1 Absorbance (Au) Absorbance

0.0

400 600 800 1000 Wavelength (nm)

Figure 12: UV-Vis Spectrum of Fe(III)-DFOB at pH 6

2.4.3. Alternative Methods for Quantifying DFOB

While the FeCl3 method was employed in the analysis of DFOB, other techniques were evaluated for their effectiveness for determining DFOB concentration. The first method examined was the direct spectrophotometric determination of DFOB using UV-Vis. This method requires minimal effort since the supernatant can be sampled and scanned directly for analysis at

195 nm. The spectrum for DFOB is shown in Figure 13.

Figure 13: UV-Vis Spectrum of free DFOB at pH 6

This method provided good linearity in the region from 5 µM to 50 µM (Figure 14) but departed from linearity above and below this concentration range. This wavelength region is also susceptible to multiple interferences and needed to be performed in quartz cuvettes. Some interferences are organic buffers (MES, Acetate buffers, HEPES, PIPES, and Citrate buffers), which showed high absorbance values in this region, so if this method is to be used there must be no buffers present in solution.

Figure 14: UV-Vis Calibration curve for free DFOB at 195 nm

The second alternative way to measure DFOB concentration in solution is through the use of Cu-CAS (chrome azurol S) complex. This procedure was published by Shenker et. al

(1995)87, and allows for the determination of free DFOB. The Cu-CAS complex has absorbance at 582 nm and when added to DFOB (or other strong chelating agents), a ligand exchange takes place with Cu exchanging CAS for the more favorable DFOB. The decrease in absorbance at 582 nm can then be measured and used to quantify the concentration of Cu-CAS that has undergone ligand exchange (1:1 ratio with DFOB present in solution). This method was successful in quantifying free DFOB as well as Mn(III)-DFOB stock solutions but was unsuccessful in quantifying U(VI)-DFOB. When U(VI)-DFOB stock solution was added to Cu-CAS solution, a

1:1 ligand exchange was not observed, therefore this method would not be reliable for a system that contains U(VI)-DFOB.

2.4.4. Determination of Acetate and Succinate

Ion chromatography (Dionex Integrion HPIC) was used in order to determine the concentrations of acetate and succinate produced as a result of DFOB sorption on the substrate and subsequent hydrolysis. DFOB is commercially available as a mesylate salt, and the mesylate anion peak interferes with the acetate peak in Ion Chromatography55. To this end, 1mM DFOB

(in the form of mesylate salt) stock solution was run through a Dowex 1 × 2 (chloride form, 200-

400 mesh; Acros Organics) ion exchange resin packed in 0.7 X 4 cm column (Kimble Flex-

Column) using mini peristaltic tubing pump (Thomas Scientific). The complete exchange of mesylate salt with chloride ions and consequently the production of mesylate free solution was verified by the absence of mesylate peak at 3.4 min in IC using 35 mM NaOH eluent. The chromatographic column in Ion Chromatography was a Dionex anion exchange AS20 column and samples were passed through the column at a flow rate of 1 mL/min. Standard solutions of acetate and succinate were created by dissolving the appropriate amount of Sodium acetate anhydrous (Fisher, ≥ 99% purity) and Sodium succinate dibasic (Sigma Aldrich, 98% purity) in ultra-pure water. The concentration range was 0.3 to 5 ppm. The calibration curve for both analytes are shown in Figures 15 and 16.

Figure 15: IC calibration curve for acetate using a 5 mM NaOH eluent, elution time: 6.125 min

Figure 16: Calibration curve for succinate using a 35 mM NaOH eluent, elution time: 4.900 min

The eluent used for both acetate and succinate was ultrapure NaOH (50% wt % solution in water, 49.0 -51.0% purity, Acros Organics), however the concentration differed:5 mM NaOH eluent was used for acetate determination due to the very weak retention of acetate by the column (retention time 6.1 minutes) , whereas for succinate, a 35 mM NaOH eluent was used

(retention time 4.9 minutes). The chromatograms for both acetate and succinate are in Figures 17 and 18.

Figure 17: Chromatogram of succinate calibration standard (5 PPM) with 35 mM NaOH eluent,

elution time: 4.900 min.

Figure 18: Chromatogram of acetate calibration standard (5 PPM) with 5 mM NaOH eluent, elution time: 6.123 min.

CHAPTER THREE: RESULTS

3.1. Uranium Sorption on Pyrolusite in the Presence of DFOB

3.1.1. Effect of pH

To investigate the effect that DFOB has on the ability of pyrolusite to sequester U(VI), sorption experiments needed to be performed at a wide range of pH values. While the most environmentally relevant pH range is 4-1088, doing the experiment at a lower pH value can help draw conclusions about how DFOB interacts with the mineral surface. Adsorption in this study is defined as the amount of the initial U(VI) and DFOB concentration no longer present in the supernatant after equilibration with the mineral. The percent of adsorption is calculated as

퐶 −퐶 푖푛푖푡푖푎푙 푓푖푛푎푙 푥 100 . 퐶푖푛푖푡푖푎푙

DFOB sorption throughout the pH range studied decreases as pH increases both when it is complexed with U(VI) and when it is free in solution (Figure 19). At pH 3.5, roughly all the

DFOB initially introduced into the system is no longer present in the aqueous phase. The interaction of DFOB with the pyrolusite surface leads to the disappearance of intact DFOB molecules in the supernatant and this will be referred to as “unaccounted” DFOB

(DFOBUnaccounted) until further conclusions can be made about the fate of the molecule. This percentage of unaccounted DFOB decreases through pH 6 and pH 8, where approximately 70% of the initial DFOB can still be accounted for in the aqueous phase. There is also no significant difference between the unaccounted DFOB when it is introduced into the system either complexed with U(VI) or by itself. At pH 3.5, the DFOB disappearance percentages when complexed and uncomplexed is 96 ± 3 and 94 ± 1, respectively. At pH 6, the DFOB disappearance percentages are 41 ± 5 when complexed and 50 ± 4 when uncomplexed. For pH 8,

DFOB disappearance percentages when complexed and uncomplexed was 32 ± 2 and 27 ± 5, respectively. This indicates that the U(VI) has no effect on the sorption behavior of pyrolusite

towards DFOB.

% Disappearance %Disappearance

Figure 19: DFOB sequestration a function of pH by pyrolusite at differing pH values. 37 uM initial concentration of U(VI) or U-DFOB complex. Error bars represent relative standard deviation from triplicate samples.

For U(VI), the opposite result was obtained. The residual amount of U(VI) detected in the aqueous phase decreases as the pH is increased, with a noticeable enhancement of the sorption when DFOB is present (Figure 20). The U(VI) no longer present in the aqueous phase is referred to as “adsorbed” since unlike DFOB, there is no possibility for degradation or multiple reaction pathways. At pH 3.5 and 6, there is negligible adsorption of U(VI) when it is free in solution (in the absence of DFOB), whereas at pH 8 sorption of U(VI) was found to be ~25%. One trend that

31 is constant throughout all pH values is the enhancement of U(VI) adsorption when DFOB is present. In the case of pH 3.5 and 6, DFOB is the only source through which U(VI) is sequestered to the minerals surface. At pH 8, while pyrolusite also has the ability to sequester

U(VI) when in the absence of DFOB, there is still an enhancement when DFOB is introduced into the system indicating that multiple mechanisms may be occurring at this pH value.

Figure 20: Uranium sorption vs pH: The percentages of Uranium that is no longer present in the aqueous phase after introduction with pyrolusite at differing pH values. Error bars represent relative standard deviation from triplicate samples.

When comparing the results from both DFOB sorption and U(VI) sorption, it is also noticed that the concentration of each component no longer present in the aqueous phase is not equivalent. For example, at pH 3.5 the concentration of adsorbed U(VI) is 1.3µM while the amount of DFOB unaccounted for in the aqueous phase is 36µM. At pH 6, the adsorbed U(VI) is

8.6µM, while the unaccounted for DFOB is 14µM. This continues at pH 8, where 15µM of

U(VI) is adsorbed while only 12µM of DFOB is unaccounted for. When the U(VI)-DFOB complex forms in solution, it forms in a 1:1 ratio of U(VI) to DFOB39. Because U(VI) and

DFOB do not disappear from the aqueous phase in a 1:1 ratio, the interaction between DFOB and U(VI) responsible for enhanced sorption is not exclusively due to the adsorption of an intact

U(VI)-DFOB complex. For example, if the sole mechanism for sorption for both constituents was through an intact U(VI)-DFOB complex, then the concentrations adsorbed would be a 1:1 ratio. The first conclusion that can be drawn from the pH experiments is that while adsorption as an intact U(VI)-DFOB complex is possible, it cannot be the sole mechanism of adsorption occurring in this system.

The sharp difference in U(VI) sorption that occurs when moving from pH 6 to pH 8 as well as the large change in DFOB disappearance across all pH values demonstrates that pH is an important factor in the sorption ability of pyrolusite. Across these pH ranges, the speciation of both un-complexed DFOB and U-DFOB is unlikely to contribute to the observed difference in sorption. The un-complexed DFOB exists as one species throughout the pH range of this study.

The dissociation constants of the three protonated hydroxamate groups on DFOB are 8.30, 9.00, and 9.36, while the dissociation constant for the terminal amine group is 10.8489. The un- complexed DFOB exists with 3 protonated hydroxamate groups and a protonated terminal amine group with an overall positive charge for all pH values used. Based on the speciation of the

U(VI)-DFOB complex37, we know of the existence of two major U-DFOB species. At pH 3.5 the

+ - predominant species is UO2DFOH2 , at pH 6 and pH 8 the dominant species is UO2OHDFOH .

The dominant species at both pH 6 and 8 is negatively charged; however, the protonated terminal amine group remains as a localized source of positive charge. This protonated amine group could 33 be a contributor in the sorption of DFOB. Adsorption through a protonated amine group was reported for kaolinite, hematite, and montmorillite38, 50, 52.

The speciation of both the un-complexed and complexed DFOB cannot explain the observed trends in the sorption as a result of pH. The point of zero charge for pyrolusite lies between pH 5.9-6.478. The surface charge for the pH 3.5 experiment is mostly positive. The distribution of negative to positive surface sites increases until pH 8 where the surface becomes predominately negative. The un-complexed and U-DFOB complex at low pH values have a net positive charge. If the interaction between the surface of pyrolusite and DFOB were exclusively electrostatic, it would be expected to see very low sorption for low pH and higher sorption as pH is increased for both complexed and un-complexed DFOB. This is opposite of the trend seen in the results of this experiment, because the DFOB disappearance is higher at lower pH values.

Electrostatic interactions should not be neglected since this is still a possibility, specifically at higher pH values; however, there must also be additional mechanisms for sorption that do not rely on pH-dependent characteristics of the substrate. It is reasonable that the uranyl ion in solution in the absences of DFOB would have a higher sorption at pH 8 than the other studied pH values.

The pH experiments allow for important conclusions to be made about the adsorption of both DFOB and U(VI). This first is that the uranyl ion follows characteristic sorption behavior when uncomplexed in solution. The uranyl ion has low adsorption at low pH values (mostly positive surface) and high adsorption at high pH values (mostly negative surface). The second conclusion is that through interaction with DFOB, U(VI) sorption is somehow enhanced. An attempt will be made to elucidate this interaction. The final conclusion is that DFOB must have a

34 mechanism for disappearance that is not purely electrostatic. The disappearance does not follow the trend expected based solely on charge distribution of the mineral surface compared to the charge on the DFOB species for each pH value.

3.1.2. Mechanisms for DFOB and U(VI) Sorption

3.1.2.1. Reported Pathways for DFOB Adsorption

The next goal of this study is to propose the mechanism by which DFOB enhances the adsorption of U(VI). Considering that DFOB is essential for the sorption of U(VI) at most pH values, understanding the sorption mechanism for uncomplexed DFOB is the starting point for understanding how U(VI) sorption occurs. The mechanism of sorption of DFOB has been studied previously on a variety of substrates under different environmental conditions.48, 50-52, 55,

86 Siebner-Freibach et. al. (2006)50 examined the adsorption of free DFOB and Fe(III)-DFOB onto Ca-montmorillonite through thermospectroscopic studies. The free ligand was adsorbed through the use of all active groups including: the hydroxamic groups (both C=O and C-OH),

+ hydrogen bonding using the N-H of the secondary amide groups, and through NH3 (protonated amine group) adsorption directly to the surface or indirectly to the surface through water molecules50. These active groups are highlighted in Figure 21.

Figure 21: Active groups for binding of DFOB

+ When binding occurred through the NH3 or through the secondary amide groups, it was shown through FTIR that the Fe-binding center (hydroxamate groups) was unaffected50. DFOB coordinates with Fe(III) to form an octahedral complex using its three hexadentate hydroxamate

90 2+ 37,91 groups , and it is expected to coordinate to UO2 in a similar manner . If DFOB adsorbs

+ onto the surface of pyrolusite through the NH3 group, the three hydroxamate groups remain

2+ available for complexation and it provides the avenue for UO2 to also become sequestered to the surface while staying complexed to DFOB. In addition to the afore mentioned study, additional studies of the DFOB sorption mechanisms mention the potential for both cation-like sorption from a cationic DFOB species and a site specific sorption incorporating the hydroxamate groups48, 51, 52. When studying the interaction of DFOB with hematite, Eisenlauer and Matijević (1980)53 proposed two configurations of DFOB to explain observed sorption and leaching of Fe(III) into solution. The first configuration suggests that DFOB is fixed flat on the surface with the hydroxamate groups oriented such that they can coordinate to Fe+ sites, which leads to the dissolution of Fe(III)53. They also proposed a second mechanism that occurs with less likelihood with DFOB being electrostatically attracted through the protonated amine group

36 to the FeO- sites53. This mechanism did not result in dissolution and release of Fe(III). Sorption of DFOB has also been attributed to the dissolution and subsequent release of structural Fe(III) and Mn(II)/Mn(III) from Fe(hydr)oxides and Mn oxides respectively, which indicate that DFOB participates in dissolution of minerals through reductive promoted and ligand promoted mechanisms utilizing its hydroxamate groups51, 55-57, 86. These two reported mechanisms are illustrated in Figure 22. By applying sorption models from literature, an effort was made to elucidate the sorption mechanisms in the current study of DFOB and pyrolusite.

Figure 22: Illustration of proposed mechanisms for DFOB adsorption. Proposed mechanism #1 orients the hydroxamate groups towards metal centers to participate in inner-sphere complexation. No sequestration of U(VI) can occur through this mechanism. Proposed mechanism #2 involves orientation of protonated terminal amine groups towards the surface, leaving hydroxamate groups open to complexation with U(VI).

3.1.2.2. Adsorption Using Hydroxamate Groups

The first mechanism that is applied to this system is the coordination of the hydroxamate groups of DFOB to the surface Mn+ sites on pyrolusite. When studied at pH 6, there are two

37 indicators that this mechanism is occurring. The first indicator of an inner sphere binding of

DFOB on pyrolusite, involving its hydroxamate groups, is the presence of oxidation products of

DFOB. The oxidation products of DFOB are mentioned many times in previous studies 86, 92; however, these are hard to determine and have not been successfully identified. To further complicate the identification of the oxidation products, DFOB has been reported to undergo hydrolysis during dissolution experiments with goethite and lepidocrocite93, 94. It is important to note that the oxidation and hydrolysis of DFOB are related processes, although the order of events is unknown. One approach speculates that hydrolysis of DFOB occurs first, and the highly reducing DFOB fragments that remain are the key players in mineral reduction55. Another approach speculates that DFOB first becomes oxidized, and the reduction of mineral is the driving force for hydrolysis57, 86. Whether oxidation or hydrolysis occurs first, the end products are the same and these products are what will be examined for this mechanism. Simanova et.al.

(2010)55 proposes that these hydrolysis products are acetate and a DFOB-hydroxylamine fragment, which upon reduction of Fe(III) produces a nitroso-DFOB fragment55. These proposed products are shown in Figure 23.

Figure 23: Hydrolysis/oxidative products of DFOB as proposed by Simanova et. al. (2010)55.

According to this pathway, acetate, as a hydrolysis product of DFOB should be readily detected in the samples of this study. A degradation pathway proposed by Winkelmann et. al.

(1999)95 though hydrolase reactions with DFOB was also considered in which acetate and succinate are both produced as the DFOB molecule undergoes multi-step fragmentation. The main difference between these two proposed pathways is that the scheme in Figure 23 claims that a hydroxylamine-DFOB fragment remains, whereas the scheme in Figure 24 shows complete degradation of the DFOB molecule. This distinction between the two pathways is important because if the fragmentation is only 1-step (Figure 23), the remaining DFOB fragment would still be capable of complexing with U(VI) or Fe(III). If DFOB undergoes further fragmentation (to produce acetate and succinate), it no longer has the ability to complex with

U(VI) because resulting fragments contain no more than one hydroxamate group. If it no longer has the ability to complex with U(VI) then this mechanism can be eliminated as a potential source of U(VI) adsorption enhancement.

Figure 24: DFOB degradation pathway proposed by Winkelmann (1999)95.

Both acetate and succinate were detected in the supernatant under the conditions studied.

The acetate detected in the pyrolusite samples indicate that hydrolysis and subsequent oxidation of DFOB has occurred. The presence of succinate in the samples indicate that the oxidation/hydrolysis reaction results in a multi-step degradation of DFOB such that the molecule is no longer left intact to complex with metals. The presence of succinate confirms that the

40 mechanism with pyrolusite is not the mechanism from Figure 23 and there is no remaining nitroso-DFOB fragment or hydroxylamine-DFOB fragment either adsorbed onto the surface or in solution. Having succinate in addition to acetate in the samples show that the oxidation/hydrolysis products of DFOB when in contact with pyrolusite are more complex, with the molecule following a multi-step degradation pathway. In our samples, acetate and succinate were detected in equal amounts, indicating that each DFOB molecule that degrades to acetate also degrades to succinate, according to the pathway in Figure 24 (no partial degradation of

DFOB molecules). Therefore, the DFOB undergoing hydrolysis in this case is no longer available for complexation with U(VI) and does not contribute to the observed adsorption enhancement. The amount of acetate and succinate released is shown in Figure 25. In addition, acetate and succinate are not likely to adsorb on pyrolusite surface96, so it can be concluded that the acetate and succinate detected in solution represents all of the hydrolysis products of DFOB and therefore represents every DFOB molecule that participates in this inner-sphere complexation and resulting hydrolysis/reducing reaction. The DFOB that undergoes hydrolysis is not considering to be adsorbed to the surface and will be referred to as DFOBHydrolyzed.

Figure 25: Acetate and succinate released into solution. Error bars represent relative standard deviation from triplicate samples.

The second indicator of an inner-sphere complexation mechanism is the presence of

Mn(II) in the supernatant post-sorption with DFOB acting as the reducing agent of Mn(IV). All

Mn present in the supernatant is assumed to be Mn(II) due to reductive dissolution being the primary pathway for pH ≤ 6.586. The Mn(III)-DFOB complex is also unstable below pH 7.542.

This instability of a Mn(III)-DFOB complex, means that all of the Mn detected in the supernatant must be in the +2 oxidation state. Mn(IV) centers in pyrolusite, upon the sorption of DFOB, will act in a similar manner to the decomposing Mn(III)-DFOB complex, and will undergo an internal electron transfer to give Mn(II) and oxidized DFOB products.

Figure 26 shows this release of Mn(II) into the supernatant. The concentration of DFOB responsible for inner-sphere complexation with the mineral was equated to the concentration of 42 acetate detected in the aqueous phase. When pyrolusite was brought in contact with U(VI)

(absence of DFOB), there was no release of Mn(II) into solution. This corresponds to very low amounts of sorption of U(VI) in the absence of DFOB. The U(VI) is not a reducing agent for

Mn(IV) and also does not promote the dissolution of the mineral. When pyrolusite was brought in contact with un-complexed DFOB, a large release of Mn(II) into solution occurred. The

U(VI)-DFOB complex was then introduced to pyrolusite, which again resulted in a large release of Mn(II). There is a direct correlation between the concentration of DFOB hydrolyzed

([acetate]) and the amount of Mn(II) released, and it is determined that DFOB is the primary source of Mn(II) release into solution. While this linear correlation of inner-spherically complexed DFOB and the subsequent release of Mn(II) is observed, the exact ratio of DFOB to

Mn(II) is unknown. This is because the exact mechanism of the oxidation of DFOB is unknown and oxidized DFOB fragments could contribute to further dissolution of minerals, without an exact understanding of these elementary reactions86. In addition, it is not clear how many hydroxamate groups bind inner-spherically or how many metal centers participate in coordination. In this study, approximately 4 Mn(II) atoms were released into solution per every

DFOB molecule that inner-spherically adsorbed to the surface. These results compare well with previously reported results. With manganite, the Mn(II) released after DFOB promoted dissolution was 3 to 4 times higher than the amount of DFOB participating in inner-sphere complexation86. When DFOB dissolved hematite, ferric ions were released in a 3:1 ratio to

DFOB molecules53. These results confirm that one possible mechanism for DFOB sorption is through inner-sphere complexation and that the mechanism for DFOB is independent of the presence of U(VI).

Figure 26: Concentration of Mn released from the inner-sphere adsorption of DFOB at pH 6 as a function of hydrolyzed DFOB. Hydrolyzed DFOB concentration was set equal to concentration of acetate in samples. Linear fit was applied to obtain the slope. Error bars represent relative standard deviation from triplicate samples.

3.1.2.3. Adsorption Using Terminal Amine Group

After solidifying one potential mechanism for the unaccounted for DFOB, the second proposed adsorption mechanism is examined. The second mechanism is an electrostatic adsorption of DFOB, through its protonated amine group leaving 3 hydroxamate groups open to complex with U(VI). If this holds true, the concentration of U(VI) sequestered during sorption can be set equal to the DFOB that is electrostatically adsorbed to the surface. This will be denoted as DFOBSorbed. Combining the concentration of hydrolyzed DFOB with the concentration of electrostatically adsorbed DFOB should allow for an accounting of DFOB initially introduced into the system. The mass balance will be attempted, and a definition of terms used is:

Table 4: Definition of terms for each portion of DFOB involved in separate proposed mechanisms

Term Definition

DFOBSupernatant DFOB that remained intact in aqueous phase; no interaction with surface

DFOBHydrolyzed DFOB inner-spherically adsorbs and subsequently hydrolyzes; set equal to

[acetate]

DFOBSorbed DFOB left electrostatically adsorbed to surface; set equal to [U(VI)Sorbed]

DFOBInitial DFOB initially introduced into the system

Since we draw the conclusion that there are only 3 possible fates of DFOB, then the

DFOBSupernatant, DFOBHydrolyzed, and DFOBSorbed should equal DFOBInitial. Table 3 shows that the addition of these portions together at pH 6 give comparable values to the known [DFOB]Initial. In this analysis, a 1:1 ratio was assumed for complexation U(VI) to DFOB; however, it is possible that not every DFOB molecule adsorbed electrostatically is also complexed to U(VI). At pH 6, the adsorption of U(VI) to pyrolusite in the absence of DFOB is negligible and was therefore not accounted for in this mass balance.

Table 5: Mass balance of DFOB after sorption onto pyrolusite at pH 6. 1 2 3 [DFOBsupernatant] [DFOBHydrolyzed] [DFOBSorbed] Total of 1, 2, 3 [DFOB]initial (µM) (µM) (µM) (µM) 37 14 24 75 (± 4) 78

26 12 22 60 (± 4) 60

14 11 15 40 (± 1) 37

2.0 8.4 8.6 19 (± 0.5) 12

To further investigate the possibility of two mechanisms for DFOB, a mass balance was attempted at pH 3.5 and pH 8. Both of these experiments were performed with an initial DFOB concentration of 37 µM. For pH 3.5 and pH 6, the amount of U(VI) adsorbed in the absence of

DFOB was negligible (only 3-5% of total U(VI)). However at pH 8, due to the large increase in the sorption of U(VI) when un-complexed (25% of total U(VI)), this value had to be accounted for when determining how much adsorbed U(VI) is solely a result of DFOB in the system. For pH 8, this amount of [U(VI)]Sorbed was then determined by subtracting the U(VI) adsorbed without DFOB present from the total U(VI) adsorbed when DFOB was present in the system.

Using the change in the U(VI) that is attributed to DFOB, a better mass balance is achieved. This indicates that pyrolusite has the ability to sequester U(VI) with and without the assistance of

DFOB, and these two individual mechanisms do not compete with each other.

Table 6: Mass balance of DFOB at pH 3.5 and 8.

1 2 3 pH [DFOBInitial] [DFOBSupernatant] [DFOBHydrolyzed] [DFOBSorbed] Total of (µM) (µM) (µM) 1, 2, 3 (µM) (µM) pH 3.5 37 0 47 1.3 48 (± 7)

pH 8 37 25 7.0 4.8 37 (± 5)

The mass balance for both pH 3.5 and pH 8 can be justified using the two proposed mechanisms of DFOB sorption. For pH 3.5, the largely positive surface would not accommodate electrostatic attraction from either a charged DFOB molecule or U(VI) therefore the inner-sphere mechanism is dominant94. For pH 8, U(VI) can adsorb independently, and electrostatic sorption

46 of DFOB (still complexed to U(VI)) becomes more important. There is still inner-sphere complexation and hydrolysis of DFOB, but the increase in the amount of U(VI) adsorbed as a result of DFOB demonstrates the two mechanisms.

Experiments that altered the ionic strength of the supernatant at pH 6 using NaClO4 were performed in order to investigate if ionic strength affects the ability of U(VI) and U(VI)-DFOB to be sequestered by pyrolusite. For both U(VI) and DFOB, there was no change in the adsorption throughout the range of ionic strength studied (Figure 27). For some metals, increased ionic strength leads to an increase in mobility and less adsorption on to minerals97. This is due to the electric double layer that forms when a charged surface in aqueous environment attracts counterions to balance charge98. Electrostatic adsorption occurs in this double layer and the thickness of the layer is inversely proportional to the ionic strength98. As the ionic strength in this experiment was increased, the double layer thickness decreased, typically making adsorption of metals less likely. This is not observed for U(VI) or DFOB, and these results can be justified based on the proposed mechanisms. For U(VI), the adsorption occurs via DFOB and U(VI) (with the exception of a small amount at pH 8) is not coming into contact with the mineral surface.

Ionic strength would not affect U(VI) alone but would only affect it through the adsorption of

DFOB.

For DFOB, a few explanations exist for the observed behavior. The first is that DFOB is a large organic molecule and because of its size and multiple functional groups, it may not adhere the same behavior observed when small metals are being studied99, 100. The second explanation is the ambiguity for sorption of the protonated amine group. When Siebner-Freibach

(2006)50 studied the infrared spectra of adsorbed DFOB, the presence of characteristic bands

+ + were related to different methods of -NH3 adsorption. The first band showed a fraction of NH3

+ groups adsorbed via a water molecule. The second band showed a fraction of NH3 directly

+ bound to oxygen atoms on the mineral surface. Heating had no effect on the NH3 that was

+ directly bound to oxygen and the NH3 was much more strongly adsorbed when bound directly to

+ the oxygen. The NH3 adsorbing onto pyrolusite could also be interacting through more than one way, with varying strengths of attraction and influence from environmental changes.

Figure 27: DFOB and U(VI) adsorption onto pyrolusite with differing ionic strengths. Ionic strength is plotted as the -log of [NaClO4]. Error bars represent relative standard deviation from triplicate samples.

Recovery experiment were performed at pH 6 and pH 2, by using ultrapure water and

0.01M HCl, respectively. In both cases, there was no DFOB detected in the supernatant after the recovery experiments. For pH 6, this was rather expected since we hypothesized that most DFOB

48 is held on the surface of pyrolusite by inner sphere complexation, and water would not be able to recover DFOB due to its subsequent hydrolysis and fragmentation. This method would only be able to recover the DFOB adsorbed to the surface, which at pH 6 was around half of the unaccounted for DFOB. Furthermore, detection limits of U(VI) is lower than that of DFOB which allows us to detect those smalls amounts that we may not be able to see using DFOB analysis methods. At pH 2, the absence of DFOB cannot be attributed to a specific reason, since due to the high acidity, DFOB may have degraded and hence, the Fe(III) complexation method may not be able to detect it. On the other hand, there was recovery of U(VI) in both cases (Table

7).

Table 7: U(VI) extracted from surface of pyrolusite with deionized water and 0.01 M HCl Solution added to dried Concentration of U(VI) Percentage of sorbed U(VI) substrate recovered (µM) recovered (%) 10 mL, deionized water pH 6 2.7 ± 0.1 11 ± 1

10 mL 0.01 M HCl, pH 2 11 ± 1 51 ± 7

The concentration of U(VI) recovered in the water suspension (pH 6) accounted for 11% of the originally adsorbed U(VI), whereas 51% of the originally adsorbed U(VI) was recovered at pH 2 using HCl. The U(VI)-DFOB complex is strong and stable in pH 6 water. At pH 2, some U(VI) recovered in the aqueous phase could be due to degradation of the complex. Surprisingly, the high concentration of the acid was not enough to provide full recovery of U(VI) back in the aqueous phase, supporting further the argument that U(VI) is sequestered on the surface of pyrolusite by strong binding.

3.1.3 Isotherms

Isotherms for both DFOB, U(VI) and U-DFOB complex were performed at pH 6 at 5 different concentrations. In addition to looking at the complexed vs uncomplexed effect, the order of addition was also studied to see if the enhancement of U(VI) sorption relied on being introduced while complexed to DFOB. Samples spiked first with only un-complexed DFOB were spiked post sorption with U(VI) in a 1:1 ratio to DFOB still present in the supernatant.

Sorption of U(VI) showed no difference whether introduced when complexed or introduced after free DFOB was allowed to interact with the substrate. It was observed that DFOB sorption was unaffected after the addition of U(VI). These results support the hypothesis that U(VI) sorption on pyrolusite is facilitated by DFOB and it is likely due to a ternary surface complex, where

DFOB sorbs to the mineral surface first, then complexes with U(VI). This type of surface ternary complex has been observed before with other ligands101. An example of this ternary complex is the aforementioned electrostatic adsorption of DFOB, which in turn complexes with U(VI). The results also showed a trend of higher percent of both constituents adsorbed as the initial introduced concentration was decreased, with the enhancement of U(VI) consistent at all concentrations. These results are shown in Figure 29 for DFOB and Figure 30 for U(VI).

Figure 28: Sorption of DFOB at pH 6 plotted as % of initial DFOB concentration (µM) adsorbed vs. initial DFOB concentration (µM). Error bars represent relative standard deviation from triplicate samples

Figure 29: Sorption of U(VI) at pH 6 plotted as % of initial concentration (PPM) sorbed vs. initial U(VI) concentration (PPM). Error bars represent relative standard deviation from triplicate samples

A second set of isotherms were developed where the DFOB examined is the DFOB presumed to be remaining adsorbed on the surface of the pyrolusite. This means that the concentration of DFOB hydrolyzed has been subtracted since it no longer remains adsorbed onto the surface of the mineral. After subtracting for DFOB hydrolyzed, the isotherms were plotted

(Uptake as a function of equilibrium concentration) and they are presented in Figures 31 and 32.

The uptake is defined as the milligrams of adsorbed DFOB (or U(VI)) per gram of pyrolusite and

(퐶 −퐶 ) 푥 푉표푙푢푚푒 was calculated as 푄 = 푖푛푖푡푖푎푙 푓푖푛푎푙 . The shapes of each isotherm give information 푚푎푠푠 표푓 푠푢푏푠푡푟푎푡푒 (푔) about the adsorption behavior of both DFOB and U(VI). For DFOB, the trend appears linear and does not reach a plateau. This means that the sorption capacity of pyrolusite for DFOB is large and in the concentration, range studied, there is no saturation of DFOB observed. It exhibits a

“C” type isotherm where the ratio of compound remaining in solution and adsorbed to the surface is constant for all concentration values studied102. For U(VI), a slight departure from linearity is observed which could indicate the formation of a plateau as the equilibrium concentration of DFOB increases. This isotherm exhibits an “L” type or Type I Langmuir isotherm, where the plateau represents a progressive saturation of the substrate102. Since DFOB is the primary sorbate at pH 6 and U(VI) is only adsorbed as a result of DFOB adsorption, it is reasonable that U(VI) could reach saturation before DFOB does. DFOB molecules may adsorb onto pyrolusite and while it has hydroxamate groups open for complexation, this complexation may not be favored due to steric hindrance from other DFOB molecules adsorbed close by.

Figure 30: Isotherm of DFOB sorption excluding hydrolyzed DFOB at pH 6. Q is mg of DFOB adsorbed per gram of pyrolusite. Error bars represent relative standard deviation from triplicate samples.

Figure 31: Isotherm of U(VI) sorption enhanced by DFOB at pH 6. Q is mg of U(VI) adsorbed per gram of pyrolusite. Error bars represent relative standard deviation from triplicate samples.

3.2 Uranium Sorption on Manganite in the Presence of DFOB

3.2.1. Effect of pH

Adsorption of DFOB and U(VI) on manganite as a function of pH was investigated, for both components individually, as well as for the U(VI)-DFOB complex. Adsorption onto manganite revealed very noticeable differences in the behavior of both U(VI) and DFOB compared to pyrolusite. In manganite studies, sorption experiments were limited to pH 6 and 8, since they are more environmentally relevant than pH 3.588. The first noticeable difference is in the adsorption of U(VI) where unlike on pyrolusite, the percentage adsorbed is very high: approximately 84% of U(VI) is adsorbed at both pH values (Figure 33). The second major

54 difference between the two substrates is that there is no significant difference between the adsorption levels for U(VI) in the presence and absence of DFOB. At pH 6, the adsorption percentage was found 84 ± 1% for complexed and 85 ± 2% for uncomplexed U(VI) and at pH 8,

86 ± 3 % U(VI) adsorption and 92 ± 4% U(VI) adsorption were measured for U-DFOB and

U(VI), respectively.. The mobility of U(VI) in the presence of manganite is therefore unaffected by the presence of DFOB.

Figure 32: U(VI) adsorption as a function of pH in the presence and absence of DFOB. Error bars represent relative standard deviation from triplicate samples.

Another notable difference between manganite and pyrolusite is the sorption levels of

DFOB on the minerals. In Figure 34, we make the same distinction as in pyrolusite: we refer to the DFOB as “unaccounted” and not as “adsorbed”. The DFOB disappearance from the supernatant could be attributed to sorption onto the mineral surface or hydrolysis, as it will 55 explained in detail later. The overall disappearance percentages of DFOB observed, are lower in the case of manganite when compared to pyrolusite. At pH 6, DFOB disappearance was found to be 47 ± 2 % when complexed to U(VI), whereas at pH 8, only 22 ± 5% disappearance was measured. Furthermore, for both pH 6 and pH 8, DFOB disappearance is enhanced when U(VI) is present: 82% and 72% respectively. In the case of manganite, it seems that U(VI) is the driver of DFOB adsorption and therefore, the mechanism for DFOB adsorption depends on the presence of U(VI) to an extent.

Figure 33: DFOB sequestration a function of pH by manganite at different pH values. 37 uM initial concentration of U(VI) or U-DFOB complex. Error bars represent relative standard deviation from triplicate samples.

3.2.2. Mechanisms for U(VI) and DFOB Adsorption on Manganite

As previously done with pyrolusite, an attempt was made to elucidate the adsorption mechanisms of both U(VI) and DFOB onto manganite. The differences in the pH trial results from pyrolusite allows for the elimination of some of the mechanisms proposed for pyrolusite.

The first difference is that there are negligible hydrolysis products in the supernatant. As mentioned in the pyrolusite section, acetate is a primary hydrolysis product of DFOB upon inner- sphere complexation with the mineral surface. When the supernatant from the manganite adsorption experiments were examined, there was minimal concentration of acetate present compared to acetate levels detected in the pyrolusite samples. These values are summarized in

Table 8. These low values of acetate allow us to rule out the inner-spherical adsorption of DFOB as a major mechanism for the DFOB disappearance at pH 6 and pH 8.

Table 8: Concentration of acetate detected in manganite suspensions. DFOB initial concentration was 37 µM added to 10 mg of manganite. Experiments were performed in triplicate. pH Acetate (µM)

6 1.1 ± 1

8 0.94 ± 0.5

Moreover, U(VI) adsorption is not affected at both pH values by the presence of DFOB.

Therefore, the possibility of a ternary surface complex in the form of =surface-ligand-metal as seen in pyrolusite can be disregarded. An additional mechanism that can be eliminated is the adsorption of an intact U(VI)-DFOB molecule, based on the following observation: there is not a

1:1 ratio of concentrations of DFOB and U(VI) adsorbed. Since they form a complex in a 1:1 ratio, the disappearance of both entities would be 1:1 if they were adsorbing as an intact species.

The concentration of adsorbed U(VI) and DFOB at pH 6 is 30µM and 16µM, respectively. The difference in adsorbed concentration gets larger at pH 8, with 31µM of U(VI) being adsorbed, while only 7.1µM of DFOB is adsorbed. In addition, the U(VI) adsorption is constant at both pH values, while only the DFOB adsorption is changing with the pH. If the adsorption happened as an intact complex, both entities would be affected as the overall speciation changed as a result of pH.

Since the adsorption of an intact complex does not explain the observed results, and the results cannot be explained by the same surface-ligand-metal “bridging” seen with pyrolusite, new potential mechanisms need to be examined. In literature, there are three frequently mentioned orientations for the adsorption of metal-ligand complexes onto mineral surfaces101, 103-

106, summarized in Figure 34.

Figure 34: Possible adsorption mechanisms of metal-ligand complexes. Type I involves adsorption through the metal, Type II involves adsorption only through ligand active groups, and Type III involves the interaction of both metal and ligand active groups to the surface. “S” represents surface functional sites.

The first orientation is one in which the metal directly adsorbs to the mineral. This can occur if the mineral surface groups can exchange their OH- with the metal-ligand complex or the metal interacts directly with the =SO- groups101. In addition, for this mechanism to occur, the coordination shell of the metal must not be completely filled with donor atoms of the ligand (i.e.

U(VI) would not be complexed with all 3 of DFOB’s hydroxamate groups)101. Adsorption in this manner gives a “metal-like” behavior, with adsorption increasing with an increase in pH103. The second orientation involves a ligand that can be adsorbed directly to the mineral surface and has additional active groups capable of forming a ligand bridge between the surface and the metal cation. Adsorption in this manner is “ligand-like” with increasing adsorption as pH decreases103.

In the final orientation, the metal and ligand can both complex on the mineral surface and those surface sites are in close enough proximity so that the metal can stay complexed to the ligand.

This leads to adsorption that exhibits both “metal-like” and “ligand-like” behavior103.

Since it has already been determined that DFOB does not affect the adsorption of U(VI),

Type II adsorption can be disregarded as a possible mechanism. This leaves Type I and Type II as potential options for the ability of U(VI) to enhance DFOB adsorption. Furthermore, based on the opposing trends between U(VI) adsorption (increasing with increasing pH) and DFOB adsorption (decreasing with increasing pH), Type III seems the more likely possibility as both metal and ligand like adsorption is observed. For Type I to be a valid mechanism, U(VI) would be the driver behind DFOB adsorption, and the trends would be similar between the two. Type

III, on the other hand, would first require that U(VI) directly adsorbs to manganite through negatively charged surface groups and it also interacts with DFOB through only 1 or 2 hydroxamate groups. DFOB is then brought into close proximity to the surface through its complexation with U(VI), but also has additional active groups capable of adsorbing to the 59 surface. As was discussed with pyrolusite, in addition to its hydroxamate groups, DFOB can adsorb through the protonated amine group or through its secondary amide group, without disruption to its metal binding ability50. Either of these active groups could be used to support a

Type III mechanism.

When examining whether a cation-like adsorption through the protonated amine group or hydrogen bonding through the secondary amide group is more likely to be responsible for DFOB adsorption to manganite, the most important factor to consider is the trend of adsorption of

DFOB as the pH increases from 6 to 8. This decrease is consistent with “ligand-like” behavior103.

As seen with pyrolusite, adsorption through the protonated amine group is cation-like and increased with the increasing pH. Since the adsorption is characteristically “ligand-like”, the secondary amide groups could be the source of DFOB adsorption. Secondary amide groups have a demonstrated ability to form hydrogen bonds through their carbonyl oxygen serving as a hydrogen acceptor107, 108.

+1/2 Figure 35: Secondary amine group in DFOB hydrogen bonding with OH2 surface groups on manganite.

The surface charges of manganite can be summarized into one equation:

+1/2 -1/2 =MnOH2 ↔ =MnOH

+1/2 with the proportion of negative sites becoming larger as the pH increases. The MnOH2 group is more favorable to hydrogen bonding because each hydrogen has a smaller share of the electrons than its MnOH-1/2 counterpart104. Therefore, if DFOB adsorbs through hydrogen

+1/2 bonding, adsorption should be higher when the MnOH2 are in greater abundance (i.e. at lower

+1/2 pH values). As MnOH2 sites become sparser, less sites are available for DFOB to hydrogen bond to which could explain why the DFOB adsorption decreases from pH 6 to 8.

The last part of the mechanism that needs to be explored is why the adsorption concentration of DFOB is much lower than that of U(VI). The mechanism can theoretically accommodate 1 DFOB molecule per U(VI) atom adsorbed to the surface, especially since both pH values are below or at the point of zero charge, pH 879, leading to both negative and positive surface sites79. One limitation to this mechanism is that it requires the presence of two adjacent sites that allow for bonding of both the metal and the ligand104. This requirement could eliminate many surface sites if they if they do not have an oppositely charged adjacent site. In addition,

DFOB is a bulky ligand and is prone to steric hindrance84. U(VI) may be able to bring DFOB into close proximity to the manganite surface, however steric hindrance could make it unable to adsorb to the surface in a physically stable manner. This could explain why not every U(VI) atom is complexed to one DFOB molecule.

CHAPTER FOUR: CONCLUSIONS

4.1. Closing Remarks

Due to uranium contamination in the environment and its related negative health effects, there is a need to understand its environmental behavior in order to develop efficient and sustainable remediation and natural attenuation methods. Among these potential methods is adsorption of U(VI) to reactive minerals. It is necessary to understand the interactions between

U(VI), commonly occurring minerals, and naturally occurring organic molecules. This study provided insight into the role that common siderophore DFOB plays in the mobility of U(VI) in and its interaction with manganese oxide minerals.

As part of the first research objective, the adsorption of U(VI) in the presence and absence of DFOB onto pyrolusite was examined. When U(VI) was introduced in the absence of

DFOB at pH 3.5 and pH 6, pyrolusite did not show any retention for U(VI) (only 3-5% adsorption). At pH 8, the adsorption of U(VI) was higher at 25%. This behavior is consistent with expected metal adsorption where adsorption increases with increase in pH due to surface charge distribution becoming more negative. Pyrolusite at higher pH values (above the point of zero charge, 6) has the ability to affect the mobility of U(VI) in aqueous solution but is only capable of sequestering small percentages of the total U(VI). When U(VI) was introduced into the system while complexed to DFOB, there was an enhancement of the amount of U(VI) adsorbed onto pyrolusite at pH 6 and pH 8 leading to a resulting adsorption of 25% and 40% respectively. When DFOB is present in the environment it can facilitate the retention of U(VI) by pyrolusite, contributing this way to enhanced uranium natural attenuation. DFOB itself also undergoes structural changes: at pH 3 and pH 6, DFOB underwent hydrolysis, producing acetate

62 and succinate that were both released into aqueous phase. This degradation of DFOB is important because DFOB is an essential molecule in order to maintain microbial life109.

Pyrolusite can decrease the concentration of DFOB in soil, limiting the potential iron intake by microbes.

The second research objective aimed at proposing mechanisms by which U(VI) and

DFOB adsorb to pyrolusite. These mechanisms were developed by applying DFOB binding modes from literature38, 50. The presence of acetate and succinate is reported to be the result of inner-sphere adsorption between DFOB’s hydroxamate groups with metal centers on the mineral.

This was the dominant mechanism of DFOB adsorption at pH 3.5. This mechanism did not participate in the enhancement of U(VI) adsorption because as the molecule degrades, it loses its ability to complex with U(VI). The second mechanism was attributed to the increased U(VI) and based on increased adsorption with the increasing pH, exhibited cation-like adsorption behavior from the protonated terminal amine group. This electrostatic interaction only involves the terminal amine group and DFOB retains its ability to complex to U(VI). A mass balance was performed by combining these two proposed mechanisms to account for the fate of DFOB and obtained values within 10% of the starting DFOB concentration for both pH 6 and pH 8. The mechanisms for DFOB adsorption to pyrolusite were consistent with literature confirming

DFOB’s ability to adsorb through its hydroxamate groups and terminal amine group50.

The final research objective was to compare the adsorption behavior of U(VI) and DFOB onto pyrolusite with manganite. Because nodules of manganese minerals formed in nature contain many phases, it is useful to examine U(VI) adsorption with other forms of manganese minerals60. In the case of manganite, DFOB had no effect on the sorption of U(VI) at pH 6 and

63 pH 8. If DFOB is present in the soil and in aqueous solution, the mobility of U(VI) is not affected. U(VI), in contrast to pyrolusite, enhances the adsorption of DFOB. The adsorption mechanism for DFOB relies on U(VI) acting as a conduit between the molecule and the mineral surface. In literature, two metal-ligand adsorption mechanisms could be possible103. One in which U(VI) adsorbs to the surface while complexed to DFOB which has no interaction with the surface. This is not likely to be occurring due to the opposite trends in adsorption as a function of pH for each component. The more likely mechanism is one in which through U(VI) adsorption,

DFOB is brought into close contact with manganite, at which point it can use its secondary amide groups to hydrogen bond with positively charged surface sites. This explains the trend of decreasing adsorption as pH increases. Comparing manganite with pyrolusite shows the versatility of DFOB to interact with minerals. It also shows that depending on the substrate, the presence of both DFOB and U(VI) can have effects on the availability of each component. When organic ligands are present, the ability of a sorbate to sequester U(VI) increasing its mobility.

4.2. Future Work

The electrostatic adsorption mechanism that was proposed for DFOB onto pyrolusite can be further investigated by repeating adsorption experiments with desferrioxamine-D1 (DFO-D1), shown in Figure 37. This molecule is a derivative of DFOB where a terminal amine group proton is replaced by an acetyl group. This derivative was used by Kraemer et. al. (1999)110 to investigate if electrostatic adsorption of DFOB onto goethite occurred.

Figure 36: Derivatization of DFOB to DFOD1 by replacement of terminal amine proton with acetyl group110.

By using DFO-D1, the possibility of cation-like adsorption through the terminal amine group is eliminated. If adsorption does not occur (with the exception of inner-sphere adsorption producing hydrolysis products), then the proposed electrostatic adsorption can be further confirmed. If there is still an observed adsorption of DFOB, then the possibility of ligand-like adsorption can be inferred. If ligand-like adsorption did occur with pyrolusite, it would be analogous to the adsorption mechanism proposed for manganite.

LIST OF REFERENCES

1. de Andrade Martins, R., Becquerel and the choice of uranium compounds. Archive for

History of Exact Sciences 1997, 51 (1), 67-81.

2. Monreal, M. J.; Diaconescu, P. L., The riches of uranium. Nature Chemistry 2010, 2 (5),

424-424.

3. Agency, I. E., Key world energy statistics. International Energy Agency Paris: 2007.

4. Association, W. N., Nuclear share figures. 2015.

5. Abdelouas, A., Uranium Mill Tailings: Geochemistry, Mineralogy, and Environmental

Impact. Elements 2006, 2, 335-341.

6. Mitsakou, C.; Eleftheriadis, K.; Housiadas, C.; Lazaridis, M., Modeling of the

Dispersion of depleted uranium aerosol. Health physics 2003, 84, 538-44.

7. Meinrath, A.; Schneider, P.; Meinrath, G., Uranium ores and depleted uranium in the environment, with a reference to uranium in the biosphere from the Erzgebirge/Sachsen,

Germany. Journal of Environmental Radioactivity 2003, 64 (2), 175-193.

8. Krestou, A.; Xenidis, A.; Panias, D., Mechanism of aqueous uranium (VI) uptake by natural zeolitic tuff. Minerals Engineering 2003, 16 (12), 1363-1370.

9. Shuey, C., Uranium Exposure and Public Health in New Mexico and the Navajo Nation:

A Literature Summary. Southwest Research and Information Center, Occupational Exposures and Health …: 2000.

10. Organization, W. H., Guidelines for drinking-water quality. World Health Organization:

1993.

11. WHO, G. J. W. H. O., Guidelines for drinking-water quality. 2011, 216, 303-304.

12. Bhalara, P. D.; Punetha, D.; Balasubramanian, K., A review of potential remediation techniques for uranium(VI) ion retrieval from contaminated aqueous environment. Journal of

Environmental Chemical Engineering 2014, 2 (3), 1621-1634.

13. Choy, C. C.; Korfiatis, G. P.; Meng, X., Removal of depleted uranium from contaminated soils. Journal of Hazardous Materials 2006, 136 (1), 53-60.

14. Anirudhan, T. S.; Bringle, C. D.; Rijith, S., Removal of uranium(VI) from aqueous solutions and nuclear industry effluents using humic acid-immobilized zirconium-pillared clay.

Journal of Environmental Radioactivity 2010, 101 (3), 267-276.

15. Bailey, S. E.; Olin, T. J.; Bricka, R. M.; Adrian, D. D., A review of potentially low-cost sorbents for heavy metals. Water Research 1999, 33 (11), 2469-2479.

16. Hashem, A.; Abdel-Halim, E. S.; El-Tahlawy, K. F.; Hebeish, A., Enhancement of the

Adsorption of Co(II) and Ni(II) Ions onto Peanut Hulls through Esterification Using Citric Acid.

Adsorption Science & Technology 2005, 23 (5), 367-380.

17. Hashem, A.; Akasha, R.; Ghith, A.; Hussein, D. J. E. E. S. T., Adsorbent based on agricultural wastes for heavy metal and dye removal: A review. 2007, 19 (69), e86.

18. Mellah, A.; Chegrouche, S.; Barkat, M., The removal of uranium(VI) from aqueous solutions onto activated carbon: Kinetic and thermodynamic investigations. Journal of Colloid and Interface Science 2006, 296 (2), 434-441.

19. Song, Q.; Ma, L.; Liu, J.; Bai, C.; Geng, J.; Wang, H.; Li, B.; Wang, L.; Li, S.,

Preparation and adsorption performance of 5-azacytosine-functionalized hydrothermal carbon for selective solid-phase extraction of uranium. Journal of Colloid and Interface Science 2012, 386

(1), 291-299.

20. Rao, G. P.; Lu, C.; Su, F., Sorption of divalent metal ions from aqueous solution by carbon nanotubes: A review. Separation and Purification Technology 2007, 58 (1), 224-231.

21. Li, Z.; Chen, F.; Yuan, L.; Liu, Y.; Zhao, Y.; Chai, Z.; Shi, W., Uranium(VI) adsorption on graphene oxide nanosheets from aqueous solutions. Chemical Engineering Journal

2012, 210, 539-546.

22. Hilal, N.; Al-Zoubi, H.; Darwish, N. A.; Mohamma, A. W.; Abu Arabi, M., A comprehensive review of nanofiltration membranes:Treatment, pretreatment, modelling, and atomic force microscopy. Desalination 2004, 170 (3), 281-308.

23. Raff, O.; Wilken, R.-D., Removal of dissolved uranium by nanofiltration. Desalination

1999, 122 (2), 147-150.

24. Farazmand, A.; Ghafourian, H.; Shahamat, M.; Levin, M.; Colwell, R., Uranium accumulation by a bacterium isolated from electroplating effluent. World Journal of

Microbiology and Biotechnology 2002, 18, 295-302.

25. Wu, W.-M.; Carley, J.; Luo, J.; Ginder-Vogel, M. A.; Cardenas, E.; Leigh, M. B.;

Hwang, C.; Kelly, S. D.; Ruan, C.; Wu, L.; Van Nostrand, J.; Gentry, T.; Lowe, K.; Carroll,

S.; Luo, W.; Fields, M. W.; Gu, B.; Watson, D.; Kemner, K. M.; Marsh, T.; Tiedje, J.;

Zhou, J.; Fendorf, S.; Kitanidis, P. K.; Jardine, P. M.; Criddle, C. S., In Situ Bioreduction of

Uranium (VI) to Submicromolar Levels and Reoxidation by Dissolved Oxygen. Environmental

Science & Technology 2007, 41 (16), 5716-5723.

26. Lee, M.; Yang, M., Rhizofiltration using sunflower (Helianthus annuus L.) and bean

(Phaseolus vulgaris L. var. vulgaris) to remediate uranium contaminated groundwater. Journal of

Hazardous Materials 2010, 173 (1), 589-596.

27. Pratas, J.; Rodrigues, N.; Paulo, C., Uranium accumulator plants from the centre of

Portugal - Their potential to phytoremediation. 2006; pp 477-482.

28. Hampel, C. A., Encyclopedia of the chemical elements. 1968.

29. Craig, D. J. W. S. R. C. A., SC, Chemical and radiological toxicity of uranium and its compounds. 2001.

30. Ching-kuo Daniel, H.; Langmuir, D., Adsorption of uranyl onto ferric oxyhydroxides:

Application of the surface complexation site-binding model. Geochimica et Cosmochimica Acta

1985, 49 (9), 1931-1941.

31. Krepelova, A.; Sachs, S.; Bernhard, G., Uranium(VI) sorption onto kaolinite in the presence and absence of humic acid. Radiochimica Acta - RADIOCHIM ACTA 2006, 94, 825-

833.

32. Tinnacher, R. M.; Nico, P. S.; Davis, J. A.; Honeyman, B. D., Effects of Fulvic Acid on

Uranium(VI) Sorption Kinetics. Environmental Science & Technology 2013, 47 (12), 6214-6222.

33. Schmeide, K.; Pompe, S.; Bubner, M.; Heise, K.; Bernhard, G.; Nitsche, H. J. R. A.,

Uranium (VI) sorption onto phyllite and selected minerals in the presence of humic acid. 2000,

88 (9/11), 723-728.

34. Murphy, R. J.; Lenhart, J. J.; Honeyman, B. D., The sorption of thorium (IV) and uranium (VI) to hematite in the presence of natural organic matter. Colloids and Surfaces A:

Physicochemical and Engineering Aspects 1999, 157 (1), 47-62.

35. Johnstone, T.; Nolan, E., Beyond Iron: Non-Classical Functions of Bacterial

Siderophores. Dalton Trans. 2015, 44.

36. Dimkpa, C., Microbial siderophores: Production, detection and application in agriculture and environment. Endocytobiosis and Cell Research 2016, 27, 7-16.

37. Mullen, L.; Gong, C.; Czerwinski, K., Complexation of uranium (VI) with the siderophore desferrioxamine B. Journal of Radioanalytical and Nuclear Chemistry 2007, 273

(3), 683-688.

38. Matijević, E.; Kolak, N., Coagulation of lyophobic colloids by metal chelates. I. Journal of Colloid and Interface Science 1967, 24 (4), 441-450.

39. Mo, K.-F.; Dai, Z.; Wunschel, D. S., Production and Characterization of

Desmalonichrome Relative Binding Affinity for Uranyl Ions in Relation to Other Siderophores.

Journal of Natural Products 2016, 79 (6), 1492-1499.

40. Duckworth, O. W.; Bargar, J. R.; Jarzecki, A. A.; Oyerinde, O.; Spiro, T. G.; Sposito,

G., The exceptionally stable cobalt(III)–desferrioxamine B complex. Marine Chemistry 2009,

113 (1), 114-122.

41. Evers, A.; Hancock, R. D.; Martell, A. E.; Motekaitis, R. J. J. I. C., Metal ion recognition in ligands with negatively charged oxygen donor groups. Complexation of iron (III), gallium (III), indium (III), aluminum (III), and other highly charged metal ions. 1989, 28 (11),

2189-2195.

42. Duckworth, O. W.; Sposito, G., Siderophore-manganese(III) interactions. I. Air-oxidation of manganese(ll) promoted by desferrioxamine B. Environ Sci Technol 2005, 39 (16), 6037-44.

43. Smith, R. M. J. V., NIST Critically Selected Stability Constant of metal complexes database. 1997.

44. Hernlem, B. J.; Vane, L. M.; Sayles, G. D., Stability constants for complexes of the siderophore desferrioxamine B with selected heavy metal cations. Inorganica Chimica Acta

1996, 244 (2), 179-184.

45. Neu, M. P.; Ruggiero, C. E.; Francis, A. J. J. A. i. p. c., Bioinorganic chemistry of plutonium and interactions of plutonium with microorganisms and plants. 2002, 169-211.

46. Neubauer, U.; Furrer, G., The use of voltammetry for sorption studies of heavy metals on mineral surfaces in presence of the siderophore desferrioxamine B. Analytica Chimica Acta

1999, 392 (2), 159-173.

47. Hepinstall, S. E.; Turner, B. F.; Maurice, P. A. J. C.; Minerals, C., Effects of siderophores on Pb and Cd adsorption to kaolinite. 2005, 53 (6), 557-563.

48. Neubauer, U.; Nowack, B.; Furrer, G.; Schulin, R., Heavy Metal Sorption on Clay

Minerals Affected by the Siderophore Desferrioxamine B. Environmental Science & Technology

2000, 34 (13), 2749-2755.

49. Hamidpour, M.; Karamooz, M.; Akhgar, A.; Tajabadipour, A.; Furrer, G., Adsorption of Cadmium and Zinc onto Micaceous Minerals: Effect of Siderophore Desferrioxamine B.

Pedosphere 2019, 29 (5), 590-597.

50. Siebner-Freibach, H.; Hadar, Y.; Yariv, S.; Lapides, I.; Chen, Y., Thermospectroscopic

Study of the Adsorption Mechanism of the Hydroxamic Siderophore Ferrioxamine B by Calcium

Montmorillonite. Journal of Agricultural and Food Chemistry 2006, 54 (4), 1399-1408.

51. Haack, E. A.; Johnston, C. T.; Maurice, P. A., Mechanisms of siderophore sorption to smectite and siderophore-enhanced release of structural Fe3+. Geochimica et Cosmochimica

Acta 2008, 72 (14), 3381-3397.

52. Rosenberg, D. R.; Maurice, P. A., Siderophore adsorption to and dissolution of kaolinite at pH 3 to 7 and 22°C. Geochimica et Cosmochimica Acta 2003, 67 (2), 223-229.

53. Eisenlauer, J.; Matijević, E., Interactions of metal hydrous oxides with chelating agents.

II. α-Fe2O3—low molecular and polymeric hydroxamic acid species. Journal of Colloid and

Interface Science 1980, 75 (1), 199-211.

54. Shirvani, M.; Nourbakhsh, F., Desferrioxamine-B adsorption to and iron dissolution from palygorskite and sepiolite. Applied Clay Science 2010, 48 (3), 393-397.

55. Simanova, A. A.; Persson, P.; Loring, J. S., Evidence for ligand hydrolysis and Fe(III) reduction in the dissolution of goethite by desferrioxamine-B. Geochimica et Cosmochimica

Acta 2010, 74 (23), 6706-6720.

56. Duckworth, O. W.; Sposito, G., Siderophore-promoted dissolution of synthetic and biogenic layer-type Mn oxides. Chemical Geology 2007, 242 (3), 497-508.

57. Peña, J.; Duckworth, O. W.; Bargar, J. R.; Sposito, G., Dissolution of hausmannite

(Mn3O4) in the presence of the trihydroxamate siderophore desferrioxamine B. Geochimica et

Cosmochimica Acta 2007, 71 (23), 5661-5671.

58. Turekian, K. K.; Wedepohl, K. H. J. G. S. o. A. B., Distribution of the Elements in Some

Major Units of the Earth's Crust. 1961, 72, 175.

59. Howe, P. D.; Malcolm, H. M.; Dobson, S.; World Health, O.; International Programme on Chemical, S., Manganese and its compounds : environmental aspects. World Health

Organization: Geneva, 2004.

60. Post, J. E., Manganese oxide minerals: Crystal structures and economic and environmental significance. 1999, 96 (7), 3447-3454.

61. Jenne, E. A., Controls on Mn, Fe, Co, Ni, Cu, and Zn Concentrations in Soils and Water: the Significant Role of Hydrous Mn and Fe Oxides. In Trace Inorganics In Water, AMERICAN

CHEMICAL SOCIETY: 1968; Vol. 73, pp 337-387.

62. Takematsu, N., Sorption of transition metals on manganese and iron oxides, and silicate minerals. Journal of the Oceanographical Society of Japan 1979, 35 (1), 36-42.

63. Grigorova, I., STUDIES AND POSSIBILITIES OF LOW GRADE MANGANESE ORE

BENEFICIATION. 2011.

64. Jabłońska-Czapla, M., Manganese and its speciation in environmental samples using hyphenated techniques: A review. Journal of Elementology 2015, 20.

65. Mason, R., Trace Metals in Aquatic Systems. 2013.

66. Davies, G., Some aspects of the chemistry of manganese(III) in aqueous solution.

Coordination Chemistry Reviews 1969, 4 (2), 199-224.

67. Maphanga, R. R.; Parker, S. C.; Ngoepe, P. E., Atomistic simulation of the surface structure of electrolytic manganese dioxide. Surface Science 2009, 603 (21), 3184-3190.

68. Tompsett, D. A.; Parker, S. C.; Islam, M. S., Rutile (β-)MnO2 Surfaces and Vacancy

Formation for High Electrochemical and Catalytic Performance. Journal of the American

Chemical Society 2014, 136 (4), 1418-1426.

69. Islam, M. A.; Angove, M. J.; Morton, D. W.; Pramanik, B. K.; Awual, M. R., A mechanistic approach of chromium (VI) adsorption onto manganese oxides and boehmite.

Journal of Environmental Chemical Engineering 2020, 8 (2), 103515.

70. Bernard, S.; Chazal, P.; Mazet, M., Removal of organic compounds by adsorption on pyrolusite (β-MnO2). Water Research 1997, 31 (5), 1216-1222.

71. Qin, Y.-w.; Pan, G.; Zhang, M.-m.; Li, X.-l., Adsorption of zinc on manganite (γ-

MnOOH): particle concentration effect and adsorption reversibility. Journal of Environmental

Sciences 2004, 16, 627-630.

72. Bochatay, L.; Persson, P., Metal Ion Coordination at the Water–Manganite (γ-MnOOH)

Interface: II. An EXAFS Study of Zinc(II). Journal of Colloid and Interface Science 2000, 229

(2), 593-599.

73. Ajmal, M.; Rao, R. A. K.; Siddiqui, B. A., Adsorption studies and the removal of dissolved metals using pyrolusite as adsorbent. Environmental Monitoring and Assessment 1995,

38 (1), 25-35.

74. Smith, K. S. J. T. E. G. o. M. D. P. B. C. S.; Topics, R., Metal sorption on mineral surfaces: an overview with examples relating to mineral deposits. 1999, 6, 161-182.

75. Sposito, G. J. P., New York, USA, The Chemistry of Soils Oxford Univ. 1989.

76. Parks, G. J. M.-w. i. g., Surface energy and adsorption at mineral-water interface: An introduction. 1990, 23, 133-175.

77. Lewis-Russ, A., Measurement of surface charge of inorganic geologic materials: techniques and their consequences. In Advances in agronomy, Elsevier: 1991; Vol. 46, pp 199-

243.

78. Cristiano, E.; Hu, Y.-J.; Sigfried, M.; Kaplan, D.; Nitsche, H., A Comparison of Point of Zero Charge Measurement Methodology. Clays and Clay Minerals 2011, 59 (2), 107-115.

79. Ramstedt, M.; Andersson, B. M.; Shchukarev, A.; Sjöberg, S., Surface Properties of

Hydrous Manganite (γ-MnOOH). A Potentiometric, Electroacoustic, and X-ray Photoelectron

Spectroscopy Study. Langmuir 2004, 20 (19), 8224-8229.

80. Davis, J. A.; Hayes, K. F., Geochemical processes at mineral surfaces: an overview. ACS

Publications: 1986.

81. Moore, T. E.; Ellis, M.; Selwood, P. W., Solid Oxides and Hydroxides of Manganese1.

Journal of the American Chemical Society 1950, 72 (2), 856-866.

82. Li, F.; Wu, J.; Qin, Q.; Li, Z.; Huang, X. J. J. o. a.; compounds, Facile synthesis of γ-

MnOOH micro/nanorods and their conversion to β-MnO2, Mn3O4. 2010, 492 (1-2), 339-346.

83. Shen, Y.; Zhang, J.; Sheng, Q.; Zheng, J. J. C. J. o. C., A MnOOH‐Polyaniline

Nanocomposite Modified Gold Electrode for Electrochemical Sensing of Nitrite. 2017, 35 (8),

1317-1321.

84. Wolff-Boenisch, D.; Traina, S. J., A comparative study of the effect of desferrioxamine

B, oxalic acid, and Na-alginate on the desorption of U(VI) from goethite at pH 6 and 25°C.

Geochimica et Cosmochimica Acta 2006, 70 (17), 4356-4366.

85. Frazier, S. W.; Kretzschmar, R.; Kraemer, S. M., Bacterial Siderophores Promote

Dissolution of UO2 under Reducing Conditions. Environmental Science & Technology 2005, 39

(15), 5709-5715.

86. Duckworth, O. W.; Sposito, G., Siderophore−Manganese(III) Interactions II. Manganite

Dissolution Promoted by Desferrioxamine B. Environmental Science & Technology 2005, 39

(16), 6045-6051.

87. Shenker, M.; Chen, Y.; Hadar, Y., Rapid Method for Accurate Determination of

Colorless Siderophores and Synthetic Chelates. 1995, 59 (6), 1612-1618.

88. Schaffer, M.; Licha, T., A guideline for the identification of environmentally relevant, ionizable organic molecule species. Chemosphere 2014, 103, 12-25.

89. Farkas, E.; Enyedy, É. A.; Csóka, H., A comparison between the chelating properties of some dihydroxamic acids, desferrioxamine B and acetohydroxamic acid. Polyhedron 1999, 18

(18), 2391-2398.

90. Dhungana, S.; White, P. S.; Crumbliss, A. L., Crystal structure of ferrioxamine B: a comparative analysis and implications for molecular recognition. JBIC Journal of Biological

Inorganic Chemistry 2001, 6 (8), 810-818.

91. Groenewold, G. S.; Van Stipdonk, M. J.; Gresham, G. L.; Chien, W.; Bulleigh, K.;

Howard, A., Collision-induced dissociation tandem mass spectrometry of desferrioxamine siderophore complexes from electrospray ionization of UO2(2+), Fe3+ and Ca2+ solutions.

Journal of mass spectrometry : JMS 2004, 39 (7), 752-61.

92. Beyer, W. F.; Fridovich, I., Characterization of a superoxide dismutase mimic prepared from desferrioxamine and MnO2. Archives of Biochemistry and Biophysics 1989, 271 (1), 149-

156.

93. Holmén, B. A.; Casey, W. H., Hydroxamate ligands, surface chemistry, and the mechanism of ligand-promoted dissolution of goethite [α-FeOOH(s)]. Geochimica et

Cosmochimica Acta 1996, 60 (22), 4403-4416.

94. Borer, P.; Kraemer, S. M.; Sulzberger, B.; Hug, S. J.; Kretzschmar, R.,

Photodissolution of lepidocrocite (γ-FeOOH) in the presence of desferrioxamine B and aerobactin. Geochimica et Cosmochimica Acta 2009, 73 (16), 4673-4687.

95. Winkelmann, G.; Busch, B.; Hartmann, A.; Kirchhof, G.; Süßmuth, R.; Jung, G.,

Degradation of desferrioxamines by Azospirillum irakense: Assignment of metabolites by

HPLC/electrospray mass spectrometry. Biometals 1999, 12 (3), 255-264.

96. Norén, K.; Persson, P., Adsorption of monocarboxylates at the water/goethite interface:

The importance of hydrogen bonding. Geochimica et Cosmochimica Acta 2007, 71 (23), 5717-

5730.

97. Acosta, J. A.; Jansen, B.; Kalbitz, K.; Faz, A.; Martínez-Martínez, S., Salinity increases mobility of heavy metals in soils. Chemosphere 2011, 85 (8), 1318-1324.

98. Binazadeh, M.; Xu, M.; Zolfaghari, A.; Dehghanpour, H., Effect of Electrostatic

Interactions on Water Uptake of Gas Shales: The Interplay of Solution Ionic Strength and

Electrostatic Double Layer. Energy & Fuels 2016, 30 (2), 992-1001.

99. Guo, Z.-J.; Yu, X.-M.; Guo, F.-H.; Tao, Z.-Y. J. J. o. C.; science, I., Th (IV) adsorption on alumina: Effects of contact time, pH, ionic strength and phosphate. 2005, 288 (1), 14-20.

100. Petruzzelli, G.; Guidi, G.; Lubrano, L. J. C. i. S. S.; Analysis, P., Ionic strength effect on heavy metal adsorption by soil. 1985, 16 (9), 971-986.

101. Nowack, B.; Sigg, L., Adsorption of EDTA and Metal–EDTA Complexes onto Goethite.

Journal of Colloid and Interface Science 1996, 177 (1), 106-121.

102. Limousin, G.; Gaudet, J. P.; Charlet, L.; Szenknect, S.; Barthès, V.; Krimissa, M.,

Sorption isotherms: A review on physical bases, modeling and measurement. Applied

Geochemistry 2007, 22 (2), 249-275.

103. Bowers, A. R.; Huang, C. P., Adsorption characteristics of metal-EDTA complexes onto hydrous oxides. Journal of Colloid and Interface Science 1986, 110 (2), 575-590.

104. Elliott, H. A.; Huang, C. P., The adsorption characteristics of Cu(II) in the presence of chelating agents. Journal of Colloid and Interface Science 1979, 70 (1), 29-45.

105. Benjamin, M. M.; Leckie, J. O., Conceptual model for metal-ligand-surface interactions during adsorption. Environmental Science & Technology 1981, 15 (9), 1050-1057.

106. Davis, J. A.; Leckie, J. O., Effect of adsorbed complexing ligands on trace metal uptake by hydrous oxides. Environmental Science & Technology 1978, 12 (12), 1309-1315.

107. Johansson, A.; Kollman, P.; Rothenberg, S.; McKelvey, J., Hydrogen bonding ability of the amide group. Journal of the American Chemical Society 1974, 96 (12), 3794-3800.

108. Ramiah, K. V.; Puranik, P. G., Hydrogen bonding in alcohols, acids and secondary amides. Proceedings of the Indian Academy of Sciences - Section A 1962, 56 (3), 155-163.

109. Saha, R.; Saha, N.; Donofrio, R. S.; Bestervelt, L. L., Microbial siderophores: a mini review. 2013, 53 (4), 303-317.

110. Kraemer, S. M.; Cheah, S.-F.; Zapf, R.; Xu, J.; Raymond, K. N.; Sposito, G., Effect of hydroxamate siderophores on Fe release and Pb(II) adsorption by goethite. Geochimica et

Cosmochimica Acta 1999, 63 (19), 3003-3008.