NOTE TO USERS
This reproduction is the best copy available
®
Trang 3THE BINDING AND AVAILABILITY OF 2,4,6-TRINITROTOLUENE (TNT) FROM SAND, CLAY MINERALS, AND SOILS — THE EFFECTS OF AGING
AND THE IMPLICATIONS FOR BIOTREATMENT
A Dissertation
Submitted to the Graduate School of the University of Notre Dame in Partial Fulfillment of the Requirements
for the Degree of
Doctor of Philosophy
by
Kara M Young, B.S.C.E., M.S.C.E
/ 1 WM 7® Director Graduate Program in Civil Engineering and Geological Sciences
Notre Dame, Indiana
Trang 4UMI Number: 3207141
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted Also, if unauthorized copyright material had to be removed, a note will indicate the deletion ® UMI UMI Microform 3207141 Copyright 2006 by ProQuest Information and Learning Company
All rights reserved This microform edition is protected against
unauthorized copying under Title 17, United States Code
ProQuest Information and Learning Company 300 North Zeeb Road
Trang 5THE BINDING AND AVAILABILITY OF 2,4,6-TRINITROTOLUENE (TNT) FROM SAND, CLAY MINERALS, AND SOILS —- THE EFFECTS OF AGING AND THE
IMPLICATIONS FOR BIOTREATMENT
Abstract
by
Kara M Young
For many years, the ultimate fate of the most important explosive, 2,4,6-
trinitrotoluene (TNT), has been of interest to the scientific community Despite its importance, the fate of TNT in the environment remains unknown The objective of this research was to better understand the binding and physical availability of TNT from geosorbents including sand, clay minerals, and soils — the effects of aging and the implications for biotreatability The physical availability of freshly-spiked TNT from geosorbents was investigated using sorption isotherm and thermal programmed
desorption-mass spectrometry (TPD-MS) experiments TPD-MS experiments
Trang 6Kara M Young
indicated a stronger binding affinity and thus a decreased availability with increasing
geosorbent complexity The geosorbents were then aged to assess the effects of aging on
the physical availability of TNT Chemical extractions revealed that a small fraction of
TNT was bound to the soils after only 30-60 days while TNT remained fully extractable
from the clay minerals In contrast, aqueous desorption experiments demonstrated a reduced availability of TNT with only 30 days of aging for all geosorbents, suggesting that sequestration was the dominant mechanism of interaction
The biotreatability of TNT was investigated by composting Anaerobically
digested biosolids were used for the first time as a sole amendment by
blending/composting with the geosorbents TNT was effectively removed in systems
amended with only 2.5% biosolids Microbial growth corresponded to TNT degradation
Trang 8CONTENTS 3700:1212 eee vỉ ¡102.1 ix ACKNOWLEDGMENTTS - - -G c c9 HS KH HH KH nhìn mm v0, xi CHAPTER 1: INTRODUCTION 1
1.1 Problem Definition ee eeeseessesesecsssesessessessessesescnsersssrassssesseesesersesees 1
1.2 Research Objectrve and Hypotheses - - HH HH ưệt 3
CHAPTER 2: LITERATURE REVIEW 6
2.1 Why Are We Concerned with TÌN T”? - «cv HH HH HH HH Hưện 7 2.1.1 Toxicity and Health Effects of TNT - - án nen, 7
2.1.2 A History of TNT Contaminated SŠ1tes - Ăn 8 2.1.3 Environmental Transformations of TÌNT - sen eeg 9
2.1.4 Soil Constituents Involved in TNT Binding - «c2 se 10
Trang 9CHAPTER 3: ASSESSING THE BINDING AND PHYSICAL AVAILABILITY
OE 2,4,6-TRINITROTOLUENE (TNT) EROM GEOSORBENTS 28 Kể án 28 3.2 Experimental SecfIOn ch ng TH ng HH HH He 29 3.2.1 Mat€rlaÌS - - HH TH Họ nh và 29 3.2.2 TPD-MS BackgTrOUnd - - + sgk 32 3.2.3 TPD-MS InstrumenfatiOn - - s Lt.vk ng g ng nrg 32 3.2.4 TPD-MS Spiking ProcedUre Ác ch HH ng ng 34 ° W0 S0, ơn 34 EVI.N ¿PO 09.4 35 3.2.7 TPD-MS Analysis Method - ĩc HH HH re 36 3.2.8 Sorption Isotherm EXxperlmenS - HH HH He 37
3.2.9 TNT Extraction and ÀnaÌyS1S - HH HH ng gkc 39 3.2.10 Brunauer, Emmett, and Teller (BET) Specific Surface Area 40 3.2.11 Atomic Force Microscopy (AFM) ccccsccssccssssssesssessscssseesessseeneesseeees 40 3.2.12 Powder X-Ray Diffraction (XRD) - Án ren 41
3.3 Results and ID1SCUSSIOTI 0 HH HT TH TH TH HH nh 41
3.3.1 TPD-MS Initial Data Processing ác treo 41 3.3.2 TPD-MS Ramp Rate DIÍÍerences ng net 46 3.3.3 TPD-MS Sorbate DIÍferences ĩc ng HH HH ngà, 47 3.3.4 TPD-MS Sorbent DIfferences cà kg 1 0 xen 48 3.3.5 Release Energy from TPD-MS, ác vn HH g0 01 1 ke 53 3.3.6 Sorption IsOtl€rTnS sgk 34
3.4 DISCUSSION cố ố ố e - 57
Trang 10CHAPTER 5: A COMPARISON OF AQUEOUS AND THERMAL
DESORPTION FOR ASSESSING THE PHYSICAL AVAILABITTY OF AGED
2,4,6-TRINITROTOLUENE (TNT) IN GEOSORBENTS 89
bo 89 5.2 Materials and Methods - . ĩc HH HH HT nh 90 5.2.1 Chemical Reagents and GeosorbenIS - - sn Hi, 90
5.2.2 SpIking and Aging of TÌNT - sxnttnHg nHnHHHHHkưy 91
5.2.3 Aqueous Desorption of Aged TNT-Spiked Geosorbens 91 5.2.4 Thermal Desorption of Aged TNT-Spiked Geosorbents 93 5.2.5 TNT Extraction and Àny§1S - co HH cư 94 6c an 94 5.3.1 Effect of Aging Time on Aqueous TT Desorption 94
5.3.2 Effect of Geosorbent Type on Aqueous TNT Desorption 96
5.3.3 Kinetic Modeling Results 2.0.0.0 cccccssecesscesecseeeeeneessecseeeeenetaceneeeeens 102
5.3.4 Effect of Aging Time on TNT Thermal Desorption 109
5.3.5 EXtractions nh e.- - 113
X90 0 on 115
CHAPTER 6: THE EFFECT OF AGING ON THE BIOTREATABILITY OF 2,4,6-TRINITROTOLUENE (TNT) CONTAMINATED GEOSORBENTB 116
6.1 Introduction 116
Trang 11FIGURES
Figure 2.1 — TNT transformation pathways (taken from Spain and Nishino, 2002) 10 Figure 2.2 — Processes Inherent to aging (taken from Reid et al., 2000) 24
Figure 3.1 — Schematic of TPD-MS instrument showing the direct insertion probe and
sample crucible (an exploded VI€W) sgk 33
Figure 3.2 — Raw thermogram for TNT spiked on sand with individual thermogram
for TNT (m/z 210) shown in the inset 0.0.0 eee eseceseenesesseereeeeseerscesessascnsseresesees 42 Figure 3.4 — Comparison of a) the initial TNT thermogram and b) the TNT
thermogram smoothed using the Gaussian smoothing function - 44 Figure 3.5 — Variation in triplicate thermogram runs for TNT spiked on sand
600990) 45 Figure 3.6 - TNT desorption thermograms illustrating the peak changes with
1ncreasing ramp rate for sand (Š500ppim) - «ngàng 46 Figure 3.7 - NAC desorption from glass beads, at 10°/min, illustrating a thermogram
shift to the right with increasing molecular welghi( - sen, 47
Figure 3.8 — Normalized TNT desorption thermograms, at 10°/min, from glass vials, glass beads, sand, kaolinite, and soil-c, illustrating a shift in peak temperature
with Increasing complexIty of geosorbeni( - sếp 48
Figure 3.9 — TNT thermograms for a) glass vials, glass beads, sand, b) kaolinite and
soil-c, illustrating the decreasing intensity of desorption with increasing
complexity of Øg€OSOTeTI( nọ 70g 50
Figure 3.10 — AFM images of a) kaolinite and b) montmorillonite surfaces 52 Figure 3.11 — Yong and Lang plot for glass, sand, kaolinite, and soil-C 33
Figure 3.12 — a) Linear and b) Freundlich adsorption isotherms of TNT on kaolinite,
montmorillonite, and soil-C (I = 10mM KC|Ì) ĩ5 St rseeee 55
Figure 4.1 — Schematic diagram of the experimental design for composting with
it) 66
Figure 4.3 -Experiment #1 - TNT degradation for 0%, 2.5%, 5%, 10%, and 20%
biosolids/sand amended r€aCfOTS - - ng HH HH rệp 73
Figure 4.4 — Experiment #1 - Formation of ADNTs for the 2.5%, 5%, 10%, and 20% biosolids/sand amended r€äCfOTS G9 nH gn ng 74 Figure 4.5 — Experiment #2 - TNT removal for 0%, 0.25%, 0.5%, 1%, and 2.5%
biosolids/sand amended reactors 0.0 cceeccsseecesceseceereesceescessncessecssceseeeseeeasesseveseesaee 75
Figure 4.6 — Experiment #3 - TNT removal for 0%, 2.5%, and 5% biosolids amended
Trang 12Figure 4.7 - TNT degradation pattern in 2.5% amended soil-A in comparison to
controls without amendments and autoclaved-bioso]1ds «-«-ss«cceee 78
Figure 4.8 — TNT degradation pattern in 2.5% amended soil-C in comparison to
controls without amendments and autoclaved-bios1ds - - «sec 78 Figure 4.9 - TNT degradation pattern in 2.5% amended kaolinite in comparison to
controls without amendments and autoclaved-bios1iđs -. -s«ecesee 80 Figure 4.10 — TNT degradation pattern in 2.5% amended montmorillonite in
comparison to controls without amendments and autoclaved-biosolids 80
Figure 4.11 — Total viable microbial population for kaolinite amended with 0%, 2.5%,
and 2.5%A bloSỌ1đS Gọi TH ng 84
Figure 4.12 — Total viable microbial population for montmorillonite amended with
0%, 2.5%, and 2.5%A blosoÌ1ds HH HH TH TH ng HH nọ HH 85 Figure 4.13 — Total viable microbial population for soil-A amended with 0%, 2.5%,
and 2.5%A Đ1OSỌ1ỞS - 0 HH HT TH TH TH TH TH HH 85 Figure 4.14 — Total viable microbial population for soil-C amended with 0%, 2.5%,
aNd 2.5%A DIOSOLAS 0 ố 86
Figure 5.1 — (a) TNT desorption from sand after 0, 30, 60, 90, and 180 days of aging; (b) TNT desorption from kaolinite after 0, 30, 60, 90, and 180 days of aging; (c) TNT desorption from montmorillonite after 0, 30, 60, 90, and 180 days of aging;
(d) TNT desorption from soil-A after 0, 30, 60, 90, and 180 days of aging; and
(e) TNT desorption from soil-C after 0, 30, 60, 90, and 180 days of aging 97 Figure 5.2 — (a) TNT desorption from sand, kaolinite, montmorillonite, soil-A, and
soil-C without aging; (b) TNT desorption from sand, kaolinite, montmorillonite,
soil-A, and soil-C following 30 days of aging; (c) TNT desorption from sand, kaolinite, montmorillonite, soil-A, and soil-C following 60 days of aging; (d)
TNTdesorption from sand, kaolinite, montmorillonite, soil-A, and soil-C following 90 days of aging; and (e) TNT desorption from sand, kaolinite,
montmorillonite, soil-A, and soil-C following 180 days of aging 99 Figure 5.3 — (a) TNT desorption from sand, kaolinite, montmorillonite, soil-A, and
soil-C without aging and normalized to BET SA; (b) TNT desorption from sand, kaolinite, montmorillonite, soil-A, and soil-C following 30 days of aging and normalized to BET SA; (c) TNT desorption from sand, kaolinite,
montmorillonite, soil-A, and soil-C following 60 days of aging and normalized
to BET SA; (d) TNT desorption from sand, kaolinite, montmorillonite, soil-A, and soil-C following 90 days of aging and normalized to BET SA; and (e) TNT desorption from sand, kaolinite, montmorillonite, soil-A, and soil-C following 180 days of aging and normalized to BET ŠA - Ă Gv 101 Figure 5.4 — Fraction of TNT remaining with sand after 0, 30, 60, 90, and 180 days of
aging The model fit is plotted with the experimental data for each aging period .104
Figure 5.5 — Fraction of TNT remaining with kaolinite after 0, 30, 60, 90, and 180
days of aging The model fit is plotted with the experimental data for each aging POTION, 105 Figure 5.6 — Fraction of TNT remaining with montmorillonite after 0, 30, 60, 90, and
Trang 13Figure 5.7 — Fraction of TNT remaining with soil-A after 0, 30, 60, 90, and 180 days
of aging The model fit is plotted with the experimental data for each aging
POTION, 00 107 Figure 5.8 — Fraction of TNT remaining with soil-C after 0, 30, 60, 90, and 180 days
of aging The model fit is plotted with the experimental data for each aging
POTION, 0.0 — 108 Figure 5.9 — (a) Thermograms for desorption of TNT from sand after aging at 30, 60,
90, and 180 days The thermogram without aging is not shown (b) Thermogram
for TNT desorption after 180 days OŸ aØ1ng HH gen 110 Figure 5.10 -Thermograms for desorption of TNT from kaolinite after aging at 30,
60, 90, and 180 days The thermogram without aging is not shown 111 Figure 5.11 -Thermograms for desorption of TNT from soil-A after aging at 30, 60,
90, and 180 days The thermogram without aging 1s not shown - 111 Figure 5.12 -Thermograms for desorption of TNT from soil-C after aging at 30, 60,
90, and 180 days The thermogram without aging 1s not shown -« 112 Figure 5.13 — TNT extraction data for the five geosorbents after aging for 0, 30, 60,
L8 1 114 Figure 6.1 - TNT degradation pattern in 2.5%-amended sand, aged for 30 days, in
comparison to controls without amendments and autoclaved-biosolids 121 Figure 6.2 — TNT degradation pattern in 2.5%-amended kaolinite, aged for 30 days,
in comparison to controls without amendments and autoclaved-biosolids 121
Figure 6.3 — TNT degradation pattern in 2.5%-amended montmorillonite, aged for 30
days, in comparison to controls without amendments and autoclaved-biosolids 122 Figure 6.4 — TNT degradation pattern in 2.5%-amended soil-A, aged for 30 days, in
comparison to controls without amendments and autoclaved-biosolids 124 Figure 6.5 - TNT degradation pattern in 2.5%-amended soil-C, aged for 30 days, in
comparison to controls without amendments and autoclaved-biosolids 124 Figure 6.6 — Total viable microbial population for aged sand amended with 0%, 2.5%,
and 2.5%A DiOSOLS 128 Figure 6.7 — Total viable microbial population for aged kaolinite amended with 0%,
2.5%, and 2.5%%A_ b1OSỌ1ỞS - Q00 1011191119 ng KH v03 E6 129 Figure 6.8 — Total viable microbial population for aged montmorillonite amended
with 0%, 2.5%, and 2.5%%A blosỌ1S Ăn ng ng nh cr 129
Figure 6.9 — Total viable microbial population for aged soil-A amended with 0%,
2.5%, ANd 2.5%A DIOSOLIS .cccessesessccecscssscccvscescecessececscsessceccsceesersecesevsteesevessases 130 Figure 6.10 — Total viable microbial population for aged soil-C amended with 0%,
Trang 14TABLES
TABLE 3.1 - PROPERTIES OF GEOSORBENTS nen 30 TABLE 3.2 - CHEMICAL PROPERTIES OF TNT AND TETRYL - 31 TABLE 3.3 - RELEASE ENERGY VALUES FOR GLASS, SAND, KAOLINITE, AND SOIL-C càng go TH TH TH HT HH HH HA 54 TABLE 3.4 - SORPTION PARAMETERS FIT TO LINEAR AND FREUNDLICH EQUA TIƠNG LH HH HH HH go HH 54 TABLE 4.1 - REACTOR SCHEMES USED FOR THE BIOSOLIDS 9)380/0/2.99/9)6.409 90115 67 TABLE 4.2 - PROPERTIES OF GEOSORBENTS FOR COMPOSTTING 69 TABLE 4.3 - CHARACTERIZATION OF ANAEROBICALLY DIGESTED
BIOSOLIDS nà nàng HH HH Tà tàn TH TH TH Tà Tà TH ni 70 TABLE 4.4 - SEQUENCE RESULTS TAKEN FROM THE DGGE PROFILES OF
WET, DRIED, AND DRIED-AUTOCLAVED BIOSOLIDS .-.- 72 TABLE 4.5 - INITIAL AND FINAL TNT CONCENTRATIONS IN CONTROL,
BIOSOLIDS, AND AUTOCLAVED-BIOSOLIDS AMENDED
GEOSORBENTS - sàn HH Hà HH TH 81
TABLE 4.6 - CORRELATION COEFFICIENTS FOR ZERO-, FIRST-, AND
SECOND-ORDER LINEAR MODELS - Án niên 82 TABLE 4.7 - FIRST-ORDER REACTION RATE CONSTANTS FOR TNT
DEGRADATION IN SOILS AMENDED WITH 0%, 2.5%, AND 2.5%A
BIOSOLIDS - LH HH HH HH 83
TABLE 5.1 - SUMMARY OF THE EXTENT OF DESORPTION AND %
Trang 15TABLE 5.2 - SUMMARY OF KINETIC DESORPTION PARAMETERS FOR
EACH GEOSORBENT AFTER EACH AGING PERIOD 103 TABLE 5.3 - SUMMARY OF THERMAL DESORPTION PARAMETERS FOR
EACH GEOSORBENT AFTER EACH AGING PERIOD 113 TABLE 6.1 - PERCENT RECOVERY OF TNT AFTER 0 AND 30 DAYS OF
, 610 6 120 TABLE 6.2 - INITIAL AND FINAL TNT CONCENTRATIONS IN CONTROL,
BIOSOLIDS, AND AUTOCLAVED-BIOSOLIDS AGED GEOSORBENTS 123 TABLE 6.3 - CORRELATION COEFFICIENTS FOR ZERO-, FIRST-, AND
SECOND-ORDER LINEAR MODELS FOR AGED GEOSORBENTS 126
TABLE 6.4 - FIRST-ORDER REACTION RATE CONSTANTS FOR TNT DEGRADATION IN GEOSORBENTS AGED FOR 0 AND 30 DAYS,
Trang 16ACKNOWLEDGMENTS
I would like to acknowledge the United States Army Corps of Engineers —
Baltimore District and the United States Army Engineer Research and Development
Center — Waterways Experiment Station for their support of this work I would also like to acknowledge funding from the Bayer Predoctoral Research Fellowship Lastly, I
would like to acknowledge the American Association of University Women for awarding
me an Engineering Dissertation Fellowship I am grateful to my advisor, Dr Jeffrey W Talley, and the members of my committee — Dr Patricia Maurice for her laboratory guidance and classroom interactions, Dr Jennifer Woertz for more laboratory guidance and frequent discussions, and Dr Robert Nerenberg for allowing me to present to his research group and engage in thoughtful discussions related to my research I thank both the current and former members of my research group — Sara Nicholl, Erica Pirie,
Guojing Liu, Tim Ruggaber, Caitlyn Shea and Dr Xiangru Zhang I would like to make
Trang 17CHAPTER 1
INTRODUCTION
1.1 Problem Definition
There is a great need to understand soil interactions with nitroaromatic
compounds (NACs) and the effects of these interactions on both availability and
treatability of NACs in the environment This is especially true for the compound 2,4,6- trinitrotoluene (TNT) due to its extensive usage and connection with military bases and
former munitions manufacturing plants (Eriksson et al., 2004) Much of the soil
contaminated with TNT today is a result of the wide scale use of explosives dating back
to World Wars I and II The adherence and transformation of TNT in soils over such an extended period of time creates an obstacle for remediation (Lewis et al., 2004) and
therefore challenges our understanding of associated risks and accepted cleanup standards
(Alexander, 1995) Whether or not these contaminants are truly available for uptake by
plants and organisms and whether they pose a threat to human health and the
Trang 18which the contaminated matrix has been impacted with TNT before any means of
remediation is attempted
In the aging process, compounds can become bound via numerous mechanisms such that their bioavailability is significantly decreased (Kottler and Alexander, 2001) The time-dependent decline in availability may result from compounds becoming bound
by mechanisms that are still not well understood (Chung and Alexander, 2002) Several
mechanisms for aging are described in the literature The mechanisms of greatest
importance for the aging of chemicals in soils and sediments include increasing sorption and partitioning into the soil organic matter For example, aging of herbicide residues
leads to increased sorption and, thus, to decrease in leaching (Walker et al., 2005) The aging process entails prolonged contact with soil organic matter during which time the molecule of interest becomes progressively more tightly bound to the organic matrix and
correspondingly less bioavailable (Barraclough et al., 2005) While the influence of soil
organic matter is strong, it may not be the sole contributor to the process (Chung and
Alexander, 1998, 2002; Bogan and Sullivan, 2003) and the literature seeks to identify
other parameters of interest (i.e., clay minerals, porosity, etc.) Despite work conducted to better understand the importance of aging on the environmental fate of organic
compounds, information about TNT remains limited
Trang 19compounds that are typically only slightly hydrophobic and sparingly water soluble A
comprehensive study that examines the influence of aging of TNT in a fundamental way by isolating the effects of individual soil components (i.e., clay minerals and organic matter) is needed to expand our knowledge of TNT’s binding and availability
With an improved understanding of the binding and availability of TNT, we can better understand the potential effects of contaminant aging on our current risk-
assessment methods Most research on the fate of TNT has been site-specific, focusing
only on soils This research will bring forth new information on the aging of TNT in laboratory-aged sand, clay minerals, and soils It will examine how mostly mineral or organic carbon environments influence TNT binding and availability This information has great implications for the biotreatability of TNT with respect to both natural
attenuation systems and in-situ bioremediation schemes that are typically less energy intensive and less expensive than are more aggressive ex-situ treatment strategies 1.2 Research Objective and Hypotheses
The primary research objective of this dissertation was to better understand the
binding and availability of TNT in contaminated sand, clays, and soils; the effects of aging; and the implications for biotreatment To accomplish this goal, a review of the literature on TNT, availability, biotreatability, and aging was performed (Chapter 2) The
physical availability of TNT was investigated by using model geosorbents in sorption
isotherm experiments and thermal programmed desorption — mass spectrometry (TPD- MS) experiments (Chapter 3) Experiments were conducted that examined the availability
Trang 20systems (Chapter 4) Anaerobically digested biosolids were used as the sole additive, and
for the first time, to investigate the biotreatability of TNT in the presence of different contaminated matrix type (i.e., sand, clay minerals, and soils) Aging experiments were
conducted to assess the effects of aging on the binding and availability of TNT as
determined via chemical extraction, aqueous desorption experiments, and TPD-MS experiments (Chapter 5) The biotreatability of TNT after aging in contaminated sand, clays, and soils was investigated (Chapter 6) Finally, the binding and availability of TNT in sand, clay minerals, and soils was summarized with respect to its environmental fate, the effects of aging, and the implications for biotreatment (Chapter 7)
Specifically, the following hypotheses were investigated for this research: Hypothesis #1 (Chapter 3) — It is hypothesized that the physical availability, as
determined via aqueous sorption and thermal desorption experiments, of TNT
from the studied model geosorbents will be greatest in the order sand > kaolinite >
montmorillonite > soil-C (2.4% organic matter)
Hypothesis #2 (Chapter 4) — It is hypothesized that TNT and its reduced transformation
products can be degraded by the biota present in anaerobically digested biosolids Additionally, the biosolids can reduce the TNT concentrations by incorporating
TNT and its reduced products into the organic matter I hypothesize that the bioaugmented systems, those receiving biosolids, will perform better than control systems with no biosolids and systems receiving autoclaved-biosolids; thus,
Trang 21Hypothesis #3 (Chapter 5) — It is hypothesized that there will be differences in the rate
and extent of sequestration, the rate and extent of desorption, and the release energy of TNT from geosorbents that are characterized as mostly mineral, having low organic carbon, and having high organic carbon as these geosorbents undergo
aging in the lab As the aging period increases and the percent organic carbon
increases, the rate and extent of both sequestration and desorption will decrease while the estimated release energy will increase
Hypothesis #4 (Chapter 6) — It is hypothesized that differences in the concentration of organic carbon (zero, low, high) for aged contaminated geosorbents will affect both the rate and extent of biotreatability and the percent recovery of TNT With
increasing organic carbon, the percent recovery and the rate and extent of
Trang 22CHAPTER 2
LITERATURE REVIEW
The objective of this research was to investigate the binding, availability, and biotreatability of 2,4,6-trinitrotoluene (TNT) in quartz sand, clay minerals (specifically kaolinite and montmorillonite), and soils A major emphasis was placed on how aging affects both the availability and biotreatability as well as how these vary with different components found in TNT-contaminated soils It is important to understand how changes
in the availability of TNT over time could influence its mobility through soils, and its potential to contaminate surface waters or groundwater resources, and the potential for its
biotreatment This literature review provides background information needed to
understand this research, while simultaneously supporting the need for more fundamental research concerning the effects of aging on the binding and availability of TNT in soils To do so, it answers the following four questions:
1) Why are we concerned with TNT?
2) How do we assess the binding and availability of TNT?
3) How amenable is TNT to biotreatment?
Trang 23The answer to the first question discusses TNT in general, providing motivation
for why this research is important In answering the second question, definitions of bound
residues and availability are given and assessment of TNT binding and availability are
discussed The answer to the third question reviews the current status of TNT
bioremediation efforts The answer to the final question reviews aging mechanisms
previously proposed to explain decreased availability of organic contaminants due to aging
2.1 Why Are We Concerned with TNT? 2.1.1 Toxicity and Health Effects of TNT
The toxicity of TNT is well documented (Harvey et al., 1990; Kaplan and Kaplan, 1982; and Won et al., 1976) TNT is known to be toxic towards bacteria, fish, plants, and mammalian cells (Drzyzga et al., 1995; Layton et al., 1987; and Honeycutt, 1996) TNT is also on the list of USEPA Priority Pollutants, is a known mutagen, and can cause pancytopenia as a result of nitrenium ions formed by enzymatic oxidation (Rodgers and
Bunce, 2001) There have been numerous cases of munitions workers developing liver
damage and anemia due to TNT exposure (Sax, 1963; Bridge et al., 1942; Hamilton,
1921; Voegtlin et al., 1920) Soils contaminated with TNT pose a significant risk to
Trang 242.1.2 A History of TNT Contaminated Sites
Soils contaminated with TNT exist at former ordinance production plants as well as existing military ranges throughout the world (Spain and Nishino, 2002; Thiboutot et al., 2002) Much of the contamination at the production sites resulted from wastewaters
discharged into unlined lagoons or directly onto soil (Pennington and Patrick, 1990;
Selim et al., 1995) Although these practices are no longer employed, localized areas of residual contamination still exist (Harvey et al., 1990) TNT is recalcitrant to degradation in aerated and unsaturated soils, and thus contributes to numerous long-term
contamination problems (Fuller et al., 2003) It is well understood that contamination of the environment by explosives like TNT is a major consequence of munitions
development and testing There nevertheless remains an operational need for testing and training with conventional weapons in order to maintain combat readiness in our armed services (Thiboutot et al., 2002) The environmental impact of meeting this need presents
a serious environmental challenge
The estimated cost for the cleanup of Department of Defense (DOD) sites
contaminated with TNT in the U.S alone is $2.66 billion, and may include up to 750,000 cubic yards of soil and 530 billion gallons of groundwater (Spain and Nishino, 2002) Although a number of TNT (see Rodgers and Bunce, 2001 for a review) treatment
technologies have been developed at the bench-scale, these techniques do not result in the
Trang 252.1.3 Environmental Transformations of TNT
TNT can undergo transformation via photochemical and microbiological
processes in soil (Walsh et al., 1995) Environmental transformation of TNT is significant for two reasons First, the transformation products may be as undesirable as TNT because
of their toxicity (Griest et al., 1993; Gibbs et al., 1996) and second, they may have the
potential for immobilization reactions with soil components (Daun et al., 1998; Drzyzga et al., 1998; and Bruns-Nagel et al., 2000)
It is well established in the literature that the three nitro groups of TNT reduce the electron density of the aromatic ring and impede electrophilic attack (Rieger and
Trang 26o0; ⁄ | Nw COOH Hy Hy ON NM, xà NO; ON NO, H H NO, NH, Noy
2-Amino-4,6-dinitro- 4-Amino-2,6-dinitrotoluenc H’-Meisenheimer
benzoic acid Complex
Figure 2.1 — TNT transformation pathways (taken from Spain and Nishino, 2002)
2.1.4 Soil Constituents Involved in TNT Binding
Since the mechanisms controlling TNT availability in soils are likely similar to
the mechanisms fundamentally developed for nonpolar organic compounds, a brief
review of the soil constituents that are important for controlling TNT fate processes (1.e., sorption, desorption, and partitioning) is provided
2.1.4.1 Quartzite Sand and Clay Minerals
In typical soils, minerals make up over 90% of the soil matrix and are composed
of various arrangements of silica, aluminum, oxygen, and iron (Sparks, 1995) The
Trang 27Quartz, the second most abundant mineral in the Earth’s crust, is important to
TNT binding because it dominates in sands, many soils, and some sediments Quartzite
sand is essentially pure S102, but it may contain trace amounts of other elements It consists of a three-dimensional framework of SiO, tetrahedra, with all oxygens shared by two tetrahedra A large number of nitroaromatic compounds (NACs), such as TNT, have minimal interaction with the surface of quartz For example, Haderlein and
Schwarzenbach (1993) showed that NACs were weakly adsorbed to silica The use of quartz sand as a geosorbent in this research provided a good control for demonstrating
minimal TNT binding to more homogeneous surface sites The sorption of TNT to clay
minerals, on the other hand, is of greater significance
Clay minerals are a group of sheet silicates with related atomic structures Most
are hydrated aluminum or magnesium silicates that are products of weathering Clays are usually very fine grained, often less than 1 um in size, which results in the presence of large surface areas available for the exchange of ions and molecules between the solids
and surrounding solutions Two of the most common clay minerals are kaolinite and montmorillonite
Kaolinite is a common secondary mineral, forming after aluminous silicates, and is a component of many soils Kaolinite clays are two-layer structures with a chemical
formula of Al¿(S14O¡o)(OH)s; The atomic structure of kaolinite is based on layers of S104
tetrahedra with OH’ and A!** in between, which leads to two distinctly different potential
Trang 28adsorption of non-polar organic molecules via van der Waals type interactions (Yariv and
Cross, 2002) Given the presence of these relatively hydrophobic surfaces as well as the high surface area of kaolinite, nonpolar organic molecules, such as TNT, are bound more
tightly to kaolinite as compared to quartz sand
Montmorillonite, the other most common clay mineral, dominates modern clay-
rich sediments Montmorillonite is basically a three-layer structure that can take up extra water or other fluids between the layers of its atomic structure In the process it expands;
thus, we sometimes refer to it as an expandable or swelling clay The structure is based on groups of three layers, where single sheets of (Al,Mg)(O,OH)¢ octahedra are
sandwiched between two sheets of SiO, tetrahedra Uzgiris et al (1995) attributed a
recalcitrant fraction of polychlorinated biphenyls (PCBs) in montmorillonite to PCBs trapped in the interstitial layers of the clay
2.1.4.2 Natural Organic Matter
Natural organic matter (NOM) is an assemblage of organic compounds derived
from plants and animals and found in practically every terrestrial environment NOM may include recognizable biopolymers like proteins, lignin, and cellulose, but also a menagerie of macromolecules from the partial degradation and cross-linking of organic residues remaining from organisms or photochemical reactions (Schwarzenbach et al.,
1993) NOM is predominantly made of carbon, but can also consist of almost as many
oxygens as carbon in their structure Naturally, the structure of the material will depend
Trang 29Soil organic matter (SOM) is decomposed plant and microbial material, with the
bulk of SOM consisting of humic substances (HSs) Substances are typically referred to as humic substances (HSs) if they are soluble or extractable in aqueous base, and humin
or kerogen if they are not The humic substances are further subdivided into fulvic acids (FAs) if they are soluble at all pHs and humic acids (HAs) if they are not soluble in acidic
conditions (< pH 2) but are soluble at higher pHs HSs are a refractory mixture or
macromolecules with molecular weights ranging from a few hundred to many tens of
thousands of grams per mole (or daltons) Partitioning into humic matter may be important in the sorption of nonionic organic chemicals, like TNT, and could be a mechanism of its aging in soils and sediments
2.2 How Do We Assess Binding and Availability? 2.2.1 Bound Residues
Bound residues were first defined in 1975 by the American Institute of Biological Sciences — Environmental Task Group as “‘unextractable and chemically unidentifiable.”
Fuhr et al (1998) proposed the following definition for bound residues: “bound residues
represent compounds in soil, plant, or animal which persist in the matrix in the form of the parent substances or its metabolites after extractions in general; the formation of bound residues reduces the accessibility and the bioavailability significantly.” Weller et
al (1998) proposed the following set of definitions:
Trang 30e Soluble covalently bound residues — those that are extracted along with the matrix
by a specific procedure where the covalent bond is preserved (i.e., humic acids extractions);
e Adsorbed residues — parent compounds or main metabolites which are bound to
the matrix by reversible non-covalent interactions; and
e Entrapped residues — parent compounds or main metabolites which are retained within the matrix by steric effects; entrapped residues are essentially bound unless the matrix structure is modified
The term “bound residue” usually corresponds to the sum of covalently bound residues, adsorbed residues, and entrapped residues The binding of a chemical in a fashion that reduces availability may be linked with another constituent or soil property or several acting together, e.g., amount and type of organic matter, amount and type of clay mineral, surface area, nanoporosity, or cation-exchange capacity (CEC) (Alexander and
Alexander, 2000), and the loss of availability is a characteristic of chemicals with many
dissimilar properties (Kottler and Alexander, 2001)
2.2.2 Availability
How we define availability is critical The available portion of a compound in soil is generally described to be the portion of a compound that can be extracted from soil, typically by an organic solvent and without alteration to the chemical structure of the compound as well as the non-extractable portion (Northcott and Jones, 2000) The
bioavailable fraction is further defined as the quantity of a chemical in soil or another
Trang 31Alexander, 2000) Bioavailability, the term as used here and as commonly used among
soil and environmental scientists, does not relate to toxicity (Alexander and Alexander, 2000) Decreasing bioavailability with increasing contaminant-soil contact time has been shown for microorganisms (Kelsey et al., 1997; White et al., 1997), earthworms (Kelsey et al., 1997; White et al., 1997), and other organisms (Leppanen and Kukkonen, 1998) However, in many cases, the fraction of a compound that is not bioavailable does not
inversely relate to the fraction of the compound that is considered as bound residue
(Edwards, 1974) Decreased bioavailability of contaminants in soils may result from chemical oxidation reactions incorporating contaminants into organic matter, slow diffusion into very small pores, and absorption into organic matter — all related to the aging process
It is generally agreed that one necessary aspect to bioavailability is the release of a solid-bound contaminant (Ehlers and Luthy, 2003) Therefore, before a compound can be
bioavailable, it must actually be physically available This release or physical availability
is defined as the capability of a chemical for transport from its bound state on or within a
geosorbent to a new environmental medium Physical availability can be linked to the extent to which a contaminant can partition into the aqueous phase compared to how
much is sequestered (Chung and Alexander, 1998) and more recently, the energy required
Trang 322.2.3 TPD-MS
The theory behind TPD-MS is simple, yet complex The physical availability, as controlled by sorption and diffusion, of a compound can be assessed semi-quantitatively using release energy, with higher energy values indicating a more tightly bound
compound (Talley, 2000; Talley et al., 2004) Release energy is not to be used
synonymously with the term activation energy from the surface science literature because
a release energy actually represents the total energy required to release a compound from its sorbent, including the energy required for desorption from and diffusion through the
sorbent matrix Release energy values are a function of the type of sorbate and sorbent
involved Release energy values can be calculated using both integral and differential approaches These calculations result in values for the pre-exponential factor (v) and
release energy (E) from desorption rate curves that model thermograms produced during
a TPD-MS experiment Nicholl et al (2004) used the analytical solution to the simplified Polanyi-Wigner equation to calculate energy values of polycyclic aromatic hydrocarbons (PAHs) from several geosorbents The results of this method indicate that, as suggested in other reviews of TPD methods (Talley et al., 2004), the ‘coverage dependence’ of both E and v must be addressed in order to determine a release energy value However, this type of modeling may be possible using a differential approach
Trang 33of organic substrates has centered on hydrophobic organic compounds (HOCs) such as
PAHs, PCBs, and dioxins In previous work, it was found that TPD-MS could be used to
assess the availability of PAHs from sediments (Talley et al., 2002, Ghosh et al., 2001) Specifically, it was discovered that a PAH-impacted material had accessible pollutant fractions and a bound residue that was much more biologically unavailable and
immobile The available (treatable) fraction was associated with a low organic carbon environment and lower release energies, while the bound (untreatable) residue was found to be present primarily in a high organic carbon environment and associated with higher
release energies
2.3 How Amenable is TNT to Biotreatment?
Research regarding the biodegradation of TNT dates back to the 1970s Since
TNT was the most widely produced and used military explosive in the world and it is now the primary explosive contaminant of concern to the DOD, tremendous work was and continues to be done in an attempt to achieve mineralization of this compound
However, numerous investigations show that the TNT molecule is very recalcitrant
against microbial mineralization (Gorontzy et al., 1994; Rieger and Knackmuss, 1995; Hughes et al., 1998; Boopathy et al., 1998)
2.3.1 Biodegradation of TNT
Trang 34more nitro groups (Spain, 1995) via nitroso and hydroxylamino intermediates to form
monoamines and diamines However, reduction of TNT in the literature occurs under both anaerobic (Lewis et al., 1996; Funk et al., 1995; Boopathy and Kulpa, 1994; Boopathy et al., 1993; Funk et al., 1993; Preuss et al., 1993) and aerobic conditions (Kalafut et al., 1998; French et al., 1998; Vorbeck et al., 1998; Bruns-Nagel et al., 1996; Bradley et al., 1994; Vorbeck et al., 1994; Boopathy et al., 1994a; Boopathy et al., 1994b; Boopathy et al., 1994c; Duque et al., 1993) via bacteria and fungi Under anaerobic conditions, the reductions occur more rapidly than in the presence of oxygen, and the unstable hydroxylamino intermediates do not accumulate or form linkages In contrast, under aerobic conditions, the production of aminodinitrotoluenes and azoxy compounds
can lead to the formation of azo or azoxy linkages and to dimerization or polymerization
(Carpenter et al., 1978)
In the last decade, aerobic pure cultures that degrade TNT were identified Vorbeck et al (1994) demonstrated that a Mycobacterium sp strain was able to
metabolize TNT under aerobic conditions producing a Meisenheimer complex as its
initial metabolite Montpas et al (1997) isolated a strain of Serratia marcescans from a TNT-contaminated site that could degrade TNT as the sole source of carbon and energy Vorbeck et al (1998) enriched two bacterial strains with TNT as a sole source of nitrogen under aerobic conditions, TN7-8 and TNT-32, which could carry out nitro-group
Trang 35Enterobacter cloacae PB2 was found by French et al (1998) to be capable of slow aerobic growth with TNT as the sole nitrogen source
The anaerobic metabolism of TNT to reduced intermediates was investigated by Funk et al (1993), using starch as a supplemental carbon substrate and incubation under static conditions Clostridium bifermentans was shown to reductively transform TNT under anaerobic conditions (Lewis et al 1996) Anaerobic biotransformation of TNT was carried out by a methanogenic mixed culture and stimulated by the addition of electron donors, such as Hp (Adrian et al 2003) Recently, Fleischmann et al (2004) investigated the degradation of TNT by bovine rumen fluid, a novel source of anaerobic microbes, and found that it reduced TNT via similar pathways already described in the literature;
however, it may have also resulted in metabolites of TNT that were further transformed than previously reported
Not only are bacteria effective in the bioremediation of TNT, so too are fungi
(Barr and Aust, 1994; Bumpus and Tatarko, 1994; Fernando et al., 1990) Several studies using Phanerochaete chrysosporium discuss the mineralization of TNT Fernando et al (1990) investigated the biodegradation of TNT with Phanerochaete chrysosporium and found 18.4% and 19.6% mineralization in water and soil, respectively Hawari et al (1999) investigated the biotransformation of TNT with Phanerochaete chrysosporium, and found an unusually high number of intermediates, but found that mineralization in this case did not exceed 1% Thus, microbial transformation of TNT to carbon dioxide and water has been observed in several studies, but concentrations were generally too low to simulate the actual conditions in polluted sites, where TNT concentrations can reach
Trang 362.3.2 Bioremediation Strategies for TNT
The greatest advantages of bioremediation are that 1) contaminants can be degraded to nontoxic byproducts and some can be mineralized, potentially resulting in the complete removal of the contaminant from the environment; 2) bioremediation uses
natural means to achieve contaminant removal; and 3) bioremediation is perceived as
relatively inexpensive and environmentally friendly (USEPA, 2002) Recommended bioremediation strategies for TNT-contaminated soils include phytoremediation, bioslurry reactors, and composting
2.3.2.1 Phytoremediation
Phytoremediation, the use of green plants to remediate contaminated soil or water (Cole, 1997), is a promising method for the treatment of explosive-contaminated sites
(Hannink et al., 2002) Phytoremediation is considered an inexpensive, low maintenance
biotechnology that can tolerate high concentrations of contaminants better than some
microorganisms (Rodgers and Bunce, 2001) Hughes et al (1997) was the first to confirm the intrinsic ability of the plants Myriophyllum and C roseus to transform TNT
Trang 37to a “green liver” model in which contaminants are transformed and then conjugated into plant molecules
2.3.2.2 Bioslurry Treatment
Bioslurry treatment involves mixing TNT-contaminated soil with water and treating it in a reactor (e.g., tank or lagoon) either anaerobically, aerobically, or anaerobically-aerobically (Rieger and Knackmuss 1995 and Lenke et al 1997) The
advantage of the bioslurry system is the high degree of mixing and effective contact time which leads to often faster and more complete degradation of TNT and formation of
bound residues Bioslurry treatment for TNT is considered to be an effective method of TNT treatment by means of immobilization (e.g., fixation by irreversible binding to the organic soil matrix)
Drzyzga et al (1998) estimated the actual fate of '*C-TNT in a molasses-
supplemented soil bioreactor after a sequential anaerobic-aerobic treatment process that
transformed 97% of the initially added TNT This was the first study to show data that demonstrated the incorporation of nearly 84% of the originally applied radioactivity into the organic soil matrix following the sequential anaerobic-aerobic bioremediation process Based on alkaline hydrolyses of the solvent extracted soil samples taken after nine weeks of treatment, the major part of the radioactivity was found to be strongly bound to the humin fraction At about that same time, Daun et al (1998) investigated the
interaction of TNT and metabolites with soil and concluded that, without the use of 4c
Trang 38researchers conducted laboratory-scale and full-scale experiments to test for biologically
induced immobilization of TNT, but still could not address the relative stability of the covalent binding of reduced TNT metabolites to soil organic matter
2.3.2.3 Composting
Composting, which involves mixing TNT-contaminated soil with bulking agents (e.g., straw, bark, sawdust, and wood chips) and organic amendments (e.g., manure or fruit and vegetable processing waste), is an attractive and economical bioremediation technology The process has been studied extensively (Kaplan and Kaplan 1982, Griest et
al 1991, Williams and Marks 1991) With suitable porosity, temperature, oxygen levels,
pH and moisture content, TNT can be effectively biotransformed and incorporated into
the soil organic matter Toxicity studies have shown that the composting of TNT-
contaminated soils reduced its concentration as well as mutagenicity
Pennington et al (1995) investigated the fate of TNT in a simulated compost system in an attempt to complete an overall mass balance on '*C-TNT At the time, the best method for tracing TNT and its metabolites was the use of a radiolabeled tracer that
could be compartmentalized into the different fractions of soil organic matter This study
indicated that TNT was not mineralized, no volatile organic compounds were produced, and more than half of the radioactivity was recovered in the cellulose plus humin
Trang 39proved a viable technique, offering direct spectroscopic evidence for the formation of
covalent linkages to the organic matter In a two part study, Knicker et al (1999) and Bruns-Nagel et al (2000) applied solid-state 'SN-NMR spectroscopy to examine the nature of nitrogen-containing products that were formed and immobilized during TNT
degradation via composting Thorn and Kennedy (2002) analyzed the reacted products of
TNT and its major reductive degradation products with soil humic acid using °N NMR
and confirmed the ability of the products to form covalent bonds with the soil humic acid
These highly reactive TNT reduction products (amino groups), once bound to the soil organic matter, can significantly influence the availability and risk of TNT residues in
soil
2.4 What Are the Mechanisms of Aging that Influence the Availability and Biotreatability of TNT?
Several mechanisms have been described to explain the aging of chemicals in
soils (Chiou et al., 1983; Steinberg et al., 1987; Brusseau et al., 1991; Brusseau and Rao, 1991) Aging is believed to be strongly affected by sorption, sequestration, partitioning and/or entrapment in organic matter, and covalent bonding, with each process being
dominant and influential at different times during aging The literature presents two opposing opinions on what the term ‘aging’ should include Some authors believe that aging includes the formation of covalent bonds of parent compounds or their degradation
products (Gevao et al., 2000), while others believe aging does not include reactions that alter the structure of the molecule (Hatzinger and Alexander, 1995) The authors do agree
Trang 40small pores, do occur with aging in soil (Alexander, 1995; Pignatello and Xing, 1996;
Hatzinger and Alexander, 1995; Kelsey et al., 1997; Luthy et al., 1997; Mader et al.,
1997; White et al., 1997; Chung and Alexander, 1998; Nam and Alexander, 1998; Piatt