EPA 600/R-08/107 I September 2008 I www.epa.gov/ada
United States
Environmental Protection
Agency
Natural Attenuation of the Lead
Scavengers 1,2-Dibromoethane
(EDB) and 1,2-Dichloroethane
(1,2-DCA) at Motor Fuel Release
Sites and Implications for Risk
Management
B FOR'USE AS A *^m<
MOTOR FUEL ONLY
CONTAINS1^..;;
LEAD
evelopment
National Risk
atory, Ada, Oklahoma 74820
-------
-------
Natural Attenuation of the Lead
Scavengers 1,2-Dibromoethane
(EDB) and 1,2-Dichloroethane
(1,2-DC A) at Motor Fuel Release
Sites and Implications for Risk
Management
John T. Wilson
National Risk Management Research Laboratory
Kenneth Banks
Dynamac Corporation
Robert C. Earle
Shaw Environmental and Infrastructure, Inc.
Yongtian He
Postdoctoral Associate, National Research Council
Tomasz Kuder
University of Oklahoma
Cherri Adair
National Risk Management Research Laboratory
Office of Research and Development
National Risk Management Research Laboratory, Ada, Oklahoma 74820
-------
Notice
The U.S. Environmental Protection Agency through its Office of Research and
Development funded portions of the research described here. Mention of trade
names and commercial products does not constitute endorsement or recommendation
for use. All research projects making conclusions and recommendations based on
environmentally related measurements and funded by the Environmental Protection
Agency are required to participate in the Agency Quality Assurance Program. This
project was conducted under a Quality Assurance Project Plan for Task 5857. Work
performed by U.S. EPA employees or by the U.S. EPA on-site analytical contractor
followed procedures specified in these plans without exception. Information on the
plans and documentation of the quality assurance activities and results are available
from Cherri Adair or John Wilson.
Front Cover photos:
Courtesy of: Ron Falta and John Wilson
#1 Warning on a dispenser for leaded gasoline.
#2 Examining a water sample from a monitoring well to see if the sample has been
collected properly.
#3 Vulnerability of shallow ground water wells to contamination with EDB.
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air,
and water resources. Under a mandate of national environmental laws, the Agency strives to formulate
and implement actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research program is providing data
and technical support for solving environmental problems today and building a science knowledge base
necessary to manage our ecological resources wisely, understand how pollutants affect our health, and
prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for investigation of
technological and management approaches for preventing and reducing risks from pollution that threaten
human health and the environment. The focus of the Laboratory's research program is on methods and
their cost-effectiveness for prevention and control of pollution to air, land, water, and subsurface resources;
protection of water quality in public water systems; remediation of contaminated sites, sediments and ground
water; prevention and control of indoor air pollution; and restoration of ecosystems. NRMRL collaborates
with both public and private sector partners to foster technologies that reduce the cost of compliance and
to anticipate emerging problems. NRMRL's research provides solutions to environmental problems by:
developing and promoting technologies that protect and improve the environment; advancing scientific and
engineering information to support regulatory and policy decisions; and providing the technical support and
information transfer to ensure implementation of environmental regulations and strategies at the national,
state, and community levels.
Tetra-ethyl lead was widely used in leaded automobile gasoline from 1923 until 1987. To prevent lead
deposits from fouling the engine, 1,2-dibromoethane (EDB) and 1,2-dichloroethane (1,2-DCA) were
added to the gasoline to act as lead scavengers. These compounds reacted with lead in the engine to make
volatile compounds that were discharged in the exhaust. If leaded gasoline is spilled to ground water from
a leaking underground storage tank, there is a potential for EDB and 1,2-DCA to partition from the spill and
contaminate ground water. The Maximum Contaminant Levels (MCLs) for EDB and 1,2-DCA are 0.05 and
5.0 (ig/L respectively. The concentrations of EDB and 1,2-DCA that would be expected in ground water in
contact with unweathered leaded automobile gasoline are 1,900 and 3,700 (ig/L respectively.
Lead was effectively banned in gasoline in the USA before the underground storage tank program was fully
implemented. As a result, only a portion of the state agencies that implement the federal UST program
routinely monitor for EDB and 1,2-DCA at gasoline spill sites. In many states, little is known of the risk
from EDB and 1,2-DCA at old leaded gasoline spill sites. Monitored Natural Attenuation (MNA) is widely
used by State Agencies to manage the risk from other fuel components, such as benzene, in ground water.
The appropriate application of MNA requires a solid understanding of the behavior of the contaminants in
ground water.
To provide a technical basis for application of MNA, this report reviews the current knowledge of the
transport and fate of EDB and 1,2-DCA in ground water. This report also provides information on the
distribution of EDB and 1,2-DCA at motor fuel release sites that was collected during a survey of sites
coordinated by the U. S. EPA Office of Underground Storage Tanks and the Association of State and
Territorial Solid Waste Management Officials (ASTSWMO) and evaluates the associated chance of
contaminating ground water.
Srt W. Puls, Acting Director
Ground Water and Ecosystems Restoration Division
National Risk Management Research Laboratory
-------
-------
Notice ii
Foreword iii
Figures vii
Tables ix
Acknowledgments x
Executive Summary' xi
1.0 Introduction 1
1.1 Use of EDB and 1,2-DCA in Leaded Motor Fuel 1
1.2 Regulation of Motor Fuel Storage to Protect Ground Water and Drinking Water 2
1.3 Investigations of EDB and 1,2-DCA at Motor Fuel Release Sites 3
1.3.1 Evaluation of Data from South Carolina Performed at Clemson University 3
1.3.2 EPA/ASTSWMO Lead Scavengers Team ' 3
1.4 Purpose and Scope of This Report 5
2.0 Transport and Fate of EDB and 1,2-DCA at Motor Fuel Release Sites 6
2.1 Conceptual Model of a Motor Fuel Release in the Subsurface 6
2.1.1 Mathematical Description of Rates of Attenuation 7
2.1.2 Relationship between a First Order Rate Constant and a Half Life 7
2.1.3 A Definition of a "'Generally Useful" Rate of Attenuation 8
2.2 Attenuation caused by physical processes 8
2.2.1 Physical Weathering from Fuel Present at Residual Saturation 8
2.2.2 Sorption on Native Organic Matter 10
2.2.3 Rate Constants for Physical Weathering of EDB and 1,2-DCA 11
2.3 Attenuation Caused by Abiotic Transformation or Biodegradation 12
2.3.1 Neutral Hydrolysis of EDB and 1,2-DCA 12
2.3.2 Abiotic Transformation of EDB and 1,2-DCA by Iron(Il) Sulfide 12
2.3.3 Biodegradation of EDB and 1,2-DCA 15
2.3.3.1 Rate Constants for Biodegradation of EDB and 1,2-DCA 16
2.3.3.2 Association of Geochemical Parameters with Removal of EDB and 1,2-DCA .... 17
2.3.3.2.1 Dissolved Oxygen 18
2.3.3.2.2 Nitrate 18
2.3.3.2.3 Sulfate and Sulfide 19
2.3.3.2.4 Methane 20
2.3.3.2.5 pH 21
2.4 Applications of Compound Specific Isotope Analysis (CSIA) to Document Biodegradation
and/or Abiotic Transformation of EDB and 1,2-DCA 22
2.4.1 Theoretical Background for Using CSIA to Estimate the Extent of Biodegradation .... 22
2.4.2 Isotopic Enrichment during Biodegradation of EDB and 1,2-DCA 23
2.4.3 Isotopic Enrichment During Abiotic Transformation of EDB and 1,2-DCA 25
2.4.4 The Initial Value of 813C of EDB or 1,2-DCA Originally Released in Leaded
Motor Fuel 26
2.4.5 Measured Concentrations of EDB and 813C of EDB in Ground Water 26
2.4.6 Predictions of the Extent of Degradation or Transformation 29
3.0 Distribution of EDB and 1,2-DCA at Motor Fuel Release Sites, and the Associated Chance of
Contaminating Ground Water 30
-------
3.1 Existing Distribution of EDB, 1,2-DCA, and Benzene in Ground Water at Selected Motor
Fuel Release Sites 30
3.2 Relative Distribution of EDB, 1,2-DCA, and Benzene at Selected Motor Fuel Release Sites ... 31
3.3 The Chance that Conventional Monitoring Using EPA Method 8260 will fail to detect EDB ... 34
3.4 Co-Distribution of EDB with 1,2-DCA, Benzene, Xylenes, and Ethylbenzene at Selected
Motor Fuel Release Sites 35
3.5 Local Vulnerability of Exposure to EDB based on Past Usage of Leaded Motor Fuel 36
4.0 References 44
Appendix A. Partitioning of EDB and 1,2-DCA Between Gasoline and Water 48
Appendix B. Materials and Methods for Laboratory Studies of Abiotic Degradationof EDB and
1,2-DCA 50
Appendix C. Method for Compound Specific Isotope Analysis to Determine the Ratio of Stable
Carbon Isotopes in EDB and 1,2-DCA 53
Appendix D. Analytical Methods and Quality Assurance 55
Appendix D.I 55
Appendix D.2 55
Appendix D.3 56
Appendix D.4 56
Appendix D.5 56
Appendix D.6 57
Appendix D.7 57
Appendix D.8 58
-------
Figures
Figure 1.1. Consumption of leaded motor fuel, EDB, and 1,2-DCA in the United States from
1949 through 1989
Figure 1.2. Distribution of EDB in ground water from monitoring wells at gasoline release sites
in South Carolina, and in sites in the EPA/ASTSWMO Study 3
Figure 2.1. Relationship between a first order rate constant in units of per year and half-lives in
units of days, weeks, and years 8
Figure 2.2. Distribution of seepage velocity in MTBE plumes in South Carolina (2002) 9
Figure 2.3. Thought experiment on the effect of the rate of ground water flow on the concentration
of EDB and DCA in ground water in contact with residual leaded motor fuel 9
Figure 2.4. Association of concentrations of EDB in the most contaminated wells at each of
ten sites with estimates of the seepage velocity of ground water at the sites 10
Figure 2.5. Rate of decline of concentrations of EDB overtime in 55 monitoring wells in
South Carolina (Bulsara, 2004) 12
Figure 2.6. Effect of concentrations of sulfide at pH 7 on the rate of abiotic transformation of
EDB and DCA 13
Figure 2.7. Removal of EDB or DCA in the presence of biogenic FeS in batch experiments at
pH near 7 14
Figure 2.8. Expected rates of abiotic transformation of EDB or 1,2-DCAby FeS in aquifer
sediment, assuming a water-filled porosity of 25%, and pH near 7 15
Figure 2.9. Co-occurrence of dissolved oxygen and reduced iron in water samples produced from
monitoring wells at gasoline spill sites 18
Figure 2.10. Association of concentrations of nitrate with concentrations of EDB in the most
contaminated monitoring wells at sites 18
Figure 2.11. Association of concentrations of sulfate with concentrations of EDB in the most
contaminated monitoring wells at sites 19
Figure 2.12. Panel A. Distribution of sulfate in the source area of the plume, and in background
ground water, at gasoline spill sites in the Eastern United States. Panel B. Distribution
of sulfide in the source area of the plumes, compared to the concentration of sulfide
expected by the amount of sulfate remove. Data are from the survey of Kolhatkar et al.,
(2000) 20
Figure 2.13. Association of concentrations of methane with concentrations of EDB in the most
contaminated monitoring wells at sites 21
Figure 2.14. Association of pH with concentrations of EDB in the most contaminated monitoring
wells at sites 21
Figure 2.15. Enrichment of the heavy isotope of carbon in EDB during anaerobic biodegradation
of EDB in a microcosm study. (Redrawn from Henderson et al. (2008).) 23
Figure 2.16. Biological or Abiotic Transformations of EDB and 1,2-DCA 25
Figure 2.17. A comparison of the enrichment of the heavy carbon isotope in EDB during anaerobic
biodegradation of EDB in a microcosm study (Henderson et al., 2008) against enrichment
during abiotic transformation of EDB by FeS 25
-------
Figure 2.18. Relationship between the concentrations of EDB in the most contaminated wells at
each of five sites and the 513C of EDB. (Note that the concentration on x-axis decreases
from left to right) 28
Figure 2.19. A conservative estimate of the relationship between the measured value of 513C of
EDB in ground water at a Leaded Motor Fuel Release Site, and the extent of destruction of
EDB by abiotic and biological processes 29
Figure 3.1. Distribution of states participating in the EPA/ASTSWMO Study 30
Figure 3.2. Distribution of the maximum concentrations of EDB, 1,2-DCA, and benzene in ground
water from monitoring wells at motor fuel release sites in the EPA/ASTSWMO study,
compared to the concentrations predicted for ground water in contact with unweathered
leaded gasoline 32
Figure 3.3. Distribution of the relative chance that EDB, 1,2-DCA, and benzene in ground water at
motor fuel release sites in the EPA/ASTSWMO Study will impact a water supply well at
concentrations above the MCL, compared to the relative chance predicted for ground
water in contact with unweathered leaded gasoline 33
Figure 3.4. Comparison of the relative chance that EDB will exceed the MCL in a water supply
well to the relative chance that benzene will exceed the MCL in a water supply well for
ground water from motor fuel release sites in the EPA/ASTSWMO Study 33
Figure 3.5. Distribution of concentrations of EDB at the leaded gasoline spill sites in the
EPA/ASTSWMO Study where the concentrations were above the MCL 34
Figure 3.6. Association of concentrations of EDB with concentrations of ethylbenzene (Panel A)
or total xylenes (Panel B) 35
Figure 3.7. Association of concentrations of EDB with concentrations of 1,2-DCA (Panel A) and
benzene (Panel B) 36
Figure 3.8. Consumption of EDB in leaded motor fuel in five small states in the USA in the
forty-five years before lead was banned in automotive motor fuel 37
Figure 3.9. Consumption of EDB in leaded gasoline in five large states in the USA in the forty-five
years before lead was banned in motor gasoline 37
Figure 3.10. Relationship between heavy use of leaded motor fuel and the use of shallow ground
water for public water supplies in the contiguous USA 40
Figure 3.11. Relationship between heavy use of leaded motor fuel and the use of shallow ground
water for public water supplies in New England, the Mid-Atlantic States, and the Great
Lakes region of the USA 40
Figure 3.12. Relationship between the population of a state in the USA (2000 census) and the sales
of gasoline (1995 data) 43
Figure A.I. Predicted maximum groundwater concentrations of EDB and DCA for a range of
possible concentrations of un-weathered residual gasoline in aquifer sediment 48
Figure A.2. Expected relationship between the fraction of EDB or DCA that is dissolved in ground
water, and can be flushed away by ground water flow, and the concentration of residual
leaded motor fuel 49
Figure B.I. Consumption of sulfate during incubation of microcosms 52
-------
Tables
Table 1.1. Occurrence of EDB in ground water provided by the state agencies that participated in the
EPA/ASTSWMO State Study 4
Table 2.1. Retardation in ground water due to sorption that is expected from the organic matter
content of the aquifer solids 11
Table 2.2. Comparison of the rate of transformation of EDB and DC A by iron(II) sulfide
to the rate or removal of TCE 14
Table 2.3. Comparison of first order rate constants for biodegradation of EDB and 1,2-DCA in
anaerobic aquifer sediment to rate constants for overall removal with ground water flow
in anaerobic aquifers 17
Table 2.4. Comparison of enrichment factors for EDB and DCA to the range of isotopic
enrichment factors (e) for carbon isotopes during reductive dehalogenation of halogenated
organic compounds 24
Table 2.5. Range of 513C in samples of commercial chlorinated solvents 27
Table 2.6. Relationship between the concentrations of EDB in the most contaminated wells at five
motor fuel release sites, enrichment of stable carbon isotopes in EDB, and a conservative
estimate of the fraction of EDB destroyed 28
Table 3.1. The distribution of EDB in the sites included in the EPA/ASTSWMO State Study
compared to distribution in South 31
Table 3.2. Occurrence of EDB, benzene, and DCA in sites sampled during the EPA/ASTSWMO
State Study 31
Table 3.3. Exposure to Drinking Ground Water Contaminated with EDB from Leaded Motor Fuel. ... 38
Table 3.4. Vulnerability to Drinking Ground Water Contaminated with EDB from Leaded
Motor Fuel 41
Table B.I: Distribution of iron, AVS, and CRS along the Column with Mulch and Hematite as
described in Shen and Wilson (2007) 50
Table B.2: Distribution of pore water, AVS, and CRS in microcosms 51
Table C. 1. Reproducibility of 513C values for EDB and 1,2-DCA prepared by a purge and trap
sampler from ground water containing aqueous solutions of EDB and 1,2-DCA and
gasoline 54
-------
Acknowledgment
Peer reviews of this document were provided by Hal White (U.S. EPA Office of Underground Storage
Tanks), Ronald W. Falta (Department of Geological Sciences, Clemson University), Read S. Miner (South
Carolina Dept. of Health & Environmental Control), Harley Hopkins (American Petroleum Institute), George
DeVaull (Shell Global Solutions), Ileana Rhodes (Shell Global Solutions) and an anonymous reviewer.
Significant technical support was provided by Mark Blankenship, Chuck Hoover, John Cox, Tracy Pardue,
Lisa Hudson, Kelly Bates, Vanessa Scroggins, and Lynda Callaway (Shaw Environmental), Daniel Pope
(Dynamac Corporation), and Kevin Smith (Student Contractor). Jim Weaver (U.S. EPA/ORD/NERL)
provided samples of leaded motor fuel for analysis.
-------
Executive Summary
The lead scavengers Ethylene Dibromide (EDB) and 1,2-Dichloroethane (1,2-DCA) were added to leaded
motor gasoline to prevent the buildup of deposits of lead oxide inside internal combustion engines. Recent
studies demonstrate that lead scavengers may persist for long periods of time in certain ground water
environments. Although lead and lead scavengers were phased out in conventional motor gasoline by the
end of the 1980s, the lead scavengers from old releases may continue to contaminate ground water at many
gasoline service station sites. In addition, aviation gasoline (Avgas) contains lead scavengers, and gasoline
containing lead scavengers is still used for certain off-road applications such as automobile racing. There is
a significant possibility that lead scavengers from releases of leaded gasoline pose an ongoing risk to ground
water quality.
Domestic ground water wells and certain small public water supply wells that are in close proximity to sites
where leaded gasoline may have been released should be of particular concern. These wells often produce
ground water from shallow aquifers, which makes them more vulnerable to contamination than larger
municipal water supply wells which usually produce water from deeper aquifers.
EPA has formed a team with the Association of State and Territorial Waste Management Officials to
determine the scope and magnitude of the occurrence of lead scavengers at leaking UST sites. The team
developed a three-phased approach to this problem: (1) developing an understanding of the magnitude
of the potential problem by compiling existing background information, (2) assessing gaps in current
knowledge, based on the findings of Phase 1, and implementing appropriate measures to fill the gaps, and
(3) determining an appropriate response based on evaluation of the results of Phases 1 and 2.
Phase 1 culminated in development of a document entitled Lead Scavengers Compendium: Overview of
Properties, Occurrence, and Remedial Technologies (U.S. EPA, 2006). Phase 2 consisted of collecting
and analyzing ground water samples from 102 old gasoline release sites spread across the 19 states
that chose to participate in the investigation. This report Natural Attenuation of the Lead Scavengers
1,2-Dibromoethane (EDB) and 1,2-Dichloroethane (1,2-DCA) at Motor Fuel Release Sites and Implications
for Risk Management represents the culmination of Phase 2. It fills some of the data gaps on the expected
distribution of lead scavengers at gasoline release sites, it discusses mechanisms for abiotic transformation
and biodegradation of EDB and 1,2-DCA, and it provides new tools to recognize and use natural
transformation and degradation of EDB and 1,2-DCA as part of a risk management strategy.
The survey found that significant concentrations of EDB continue to persist at many old leaded gasoline
spill sites. Both EDB and 1,2-DCA were present at concentrations above their respective Maximum
Concentration Level (MCL) at a significant number of sites; EDB was detected above its MCL of 0.05 ug/L
at 42% of the sites sampled, and 1,2-DCA was detected above its MCL of 5.0 ug/L at 15% of the sites
sampled. Benzene (with an MCL of 5.0 ug/L) was present at 100% of the sites sampled and was the primary
risk driver at 75% of the sites where both benzene and EDB were present in ground water; EDB was the
primary risk driver in the remaining 25% of sites.
The persistence of EDB at UST spill sites is consistent with its expected behavior in ground water. Simple
physical weathering of EDB and 1,2-DCA from residual gasoline is a slow process that may require decades
to centuries to reduce high concentrations of EDB or 1,2-DCA to their MCLs. At some sites, anaerobic
biodegradation can provide substantial reductions in the concentrations of EDB and 1,2-DCA. At some sites,
abiotic degradation caused by reaction with Iron(II) sulfide minerals in aquifer material can also produce
substantial reduction in the concentration of EDB, particularly in ground water at neutral pH.
Although it is theoretically possible that anaerobic biodegradation or abiotic degradation will remove EDB
at a particular site, it is frequently difficult to prove that degradation is occurring based on conventional
monitoring data. Compound Specific Isotope Analysis (CSIA) can be useful to recognize biodegradation and
-------
abiotic transformation of EDB ground water. Degradation is recognized and documented by a change in the
ratio of stable isotopes of carbon in the molecules of EDB that remain in the ground water after degradation.
The change in the ratios can put a conservative boundary on the extent of degradation that has occurred in
the ground water sampled by a particular well. This makes CSIA a useful tool to prove that degradation has
happened at field scale at a particular site.
If the concentrations of EDB and 1,2-DCA in ground water in the source area of plumes do not attenuate,
the hazard associated with these contaminants will persist indefinitely. MNA is most cost effective as
a remedy when the concentrations of contaminants attenuate to their MCLs in a reasonable period of
time. The concentrations of EDB and 1,2-DCA that would be expected in ground water in contact with
unweathered leaded automobile gasoline are 1,900 and 3,700 ug/L respectively. To bring these initial
concentrations to their MCL within 20 years, the first order rate of attenuation in concentration in the
most contaminated well at a site should be 0.5 per year or greater for EDB and 0.33 per year or greater
for 1,2-DCA. At certain sites, and under some circumstances, rates in excess of 0.5 per year for EDB or
0.33 per year for 1,2-DCA can be attained through anaerobic biodegradation or by abiotic reactions. To
apply MNA at a specific site, rate constants for attenuation over time should be extracted from site-specific
data and should be verified and validated by continued long-term monitoring.
Monitoring for concentrations of EDB in ground water can be a major cost of risk management at gasoline
spill sites. The MCL for EDB is one hundred fold lower than the MCLs for Benzene or 1,2-DCA. Because
the MCL for EDB is so low, not all analytical methods can detect EDB when it is present at its MCL. The
EPA Method that is most commonly used to analyze for gasoline constituents in ground water (Method
8260B) has a detection limit for EDB of approximately 3.0 ug/L, which is sixty fold higher than the MCL.
As a result, Method 8260B cannot be used to document that ground water is free of contamination from
EDB. In contrast, EPA Method 8011 has a method detection limit for EDB of approximately 0.01 ug/L,
which is sufficiently sensitive to measure EDB at its MCL.
Method 8260B would have only discovered 40% of the survey sites with concentrations of EDB above its
MCL. At sites where benzene is the primary risk driver, Method 8260B would be appropriate to monitor
the quality of ground water during active remediation. However, to determine if the site has reached the
MCL for EDB, it is necessary to use Method 8011 or its equivalent.
Keywords: EDB, 1,2-dibromoethane, DCA, 1,2-DCA, 1,2-dichloroethane, ground water, UST,
underground storage tank, MNA, Monitored Natural Attenuation
-------
1.0
Introduction
This section reviews the use of the lead scavengers
1,2-dibromoethane (also called ethylene bromide or
EDB) and 1,2-dichloroethane (1,2-DCA) in leaded
motor fuel, and briefly describes the regulatory
framework developed to protect ground water
resources from releases of leaded motor fuel1 stored
in underground storage tanks. Also, this section
describes two investigations of EDB and 1,2-DCA
at motor fuel release sites. Finally, this section
describes the scope and intended purpose of this
report.
1.1 Use of EDB and 1,2-DCA in Leaded
Motor Fuel
Internal combustion engines burn a mixture of
fuel and air to create mechanical energy that turns
a crankshaft. The most common automotive
engine operates on a four-stroke cycle: intake,
compression, combustion, and exhaust. During the
compression cycle, a mixture of air and fuel vapor
is compressed by a piston moving upward in its
cylinder. Ideally, at the height of the compression
cycle, the mixture is ignited by a spark from the
spark plug, thus initiating the "combustion" stroke,
whereby the piston is pushed downward in the
cylinder producing the mechanical energy that
turns a crankshaft. During the next upstroke of the
piston, exhaust gases are expelled from the cylinder.
Sometime during the combustion stroke, pockets
of unburned fuel outside the advancing flame front
within the cylinder are heated and pressurized
leading to sudden ignition ("detonation") resulting
in engine "knock". Engine knock is damaging to
the mechanical parts of the engine and it wastes
fuel.
To reduce the tendency to knock, various additives
have been used to increase the octane of the motor
fuel. These additives have included tetra-ethyl lead
(TEL) since the 1920s, and since the 1960s tetra-
methyl lead (TML), tri-methyl-ethyl lead (TMEL),
di-methyl-di-ethyl lead (DMDEL), and methyl-
tri-ethyl lead (MTEL). The additives to increase
octane also included methyl tertiary-butyl ether
(MTBE) and ethanol.
Tetra-ethyl lead was widely used in motor gasoline
from 1923 to 1987 (Falta, 2004). Lead oxide
deposits produced during the combustion of leaded
motor fuel can accumulate and damage the engine.
To make the lead volatile and thus reduce the
accumulation of lead deposits, the lead scavengers
EDB and 1,2-DCA were added to gasoline along
with the TEL. With these additives, the lead
forms lead dihalides which are volatile and can be
expelled from the engine.
Starting in 1975, automobiles in the U.S. were fitted
with catalytic converters to treat the exhaust gas
and allow the vehicles to meet U.S. EPA standards
for emissions to control air pollution. Because
lead in motor fuel can poison the catalyst and ruin
the catalytic converter, in 1973 EPA (a) required
that one grade of unleaded gasoline be available
to protect catalytic converters that were to appear
on new cars in 1975, and (b) re-proposed annual
reductions in lead content of all other grades of
gasoline to protect public health2. Figure 1.1
presents estimates of gasoline consumption in
the U.S. that were collected and collated by
Falta (2004). It also estimates the consumption
of EDB and 1,2-DCA in gasoline, based on the
estimates of Falta (2004) for lead consumed in
gasoline, and his observation "Since the early
1940s, leaded automotive gasoline has contained
EDB and 1,2-DCA in proportion to the amount
of tetraalkyllead with a molar ratio of Pb:Cl:Br of
1:2:1..."
The proportion of EDB and 1,2-DCA consumed
each year to the total gasoline consumed each
year changed little from 1949 to 1972. The peak
years for use of EDB and 1,2-DCA were 1969
through 1972. After 1972, the total amount of EDB
and 1,2-DCA consumed in automobile gasoline
declined as the content of lead declined in gasoline.
After 1988, much less EDB and 1,2-DCA were
added to conventional automobile gasoline in the
1 "Leaded motor fuel" is a more inclusive term that
includes leaded gasoline for automobiles plus aviation
gasoline, which still contains lead and a lead scavenger
package, and some grades of racing fuel. Where this
report refers more specifically to "gasoline" it is because
the data and information pertain to leaded gasoline for
automotive purposes.
2 Even though leaded gasoline has not been used
for on-road automobiles for nearly two decades, leaded
gasoline (which also contains lead scavengers) is still
in use in aviation gasoline (avgas) and in some off-road
applications such as racing fuel.
-------
United States because leaded gasoline had largely
been phased out.
400
350.
^300.
^250
§200.
1.150.
100.
50.
0
j500,_.
. 450:3"
. 4001
. 3501.
• SOD'S
•250 |
. 200 J
•1501
•1008
CO
•50 O
1940 1950 1960 1970 1980 1990 2000
Date
Figure 1.1. Consumption of leaded motor fuel,
EDB, and 1,2-DCA in the United
States from 1949 through 1989.
Although lead and lead scavengers were phased
out in conventional motor gasoline by the end of
the 1980s, the lead scavengers from old releases
may continue to contaminate ground water at many
gasoline service station sites. In addition, aviation
gasoline (Avgas) contains lead scavengers, and
gasoline containing lead scavengers is still used for
certain off-road applications such as automobile
racing (Aronson and Howard, 2008).
A portion of the EDB produced in the US was used
as pesticide and fumigant (Aronson and Howard,
2008), and ground water contamination continues
in certain agricultural areas from the past use of
EDB as an agricultural chemical. EDB was used
on citrus crops, on vegetables, on grain crops, and
to protect golf courses (U.S. EPA, 2006). In 1977
approximately 136 million kilograms of EDB was
produced in the USA; 127 million kilograms was
used in fuel, approximately 8 million kilograms
was used as a soil fumigant, and approximately
0.9 million kilograms was used to fumigate stored
grain and grain milling machinery, and quarantined
citrus and other tropical fruits (U.S. EPA, 2006).
1.2 Regulation of Motor Fuel Storage to
Protect Ground Water and Drinking
Water
In 1974, Congress passed the Safe Drinking Water
Act, which required U.S. EPA to determine safe
levels of hazardous chemicals in drinking water.
These safe levels are called Maximum Contaminant
Level Goals or MCLGs. Because of the difficulty
in achieving MCLGs, MCLs (Maximum
Contaminant Levels) have been established for
most contaminants; MCLs are a compromise based
on best available treatment technology, limitations
of analytical methods, and cost. In 1989, U.S. EPA
promulgated MCLs for benzene and for 1,2-DCA
of 5 ug/L. In 1992, U.S. EPA promulgated an MCL
of 0.05 ug/L for EDB.
In 1984, Congress added Subtitle I to the Resource
Conservation and Recovery Act (RCRA), which
required U.S. EPA to develop a regulatory program
for underground storage tank systems (USTs)
that contained petroleum or certain hazardous
substances (collectively referred to as "regulated
substances"). The federal underground storage
tank program is administered by the Office of
Underground Storage Tanks (OUST), within the
Office of Solid Waste and Emergency Response
(OSWER). Subtitle I of the Resource Conservation
and Recovery Act (RCRA) allows state UST
programs approved by EPA to operate in lieu of
the federal program. The U.S. EPA has granted
State Program Approval to most of the states and
the others implement their own program under
cooperative agreements with EPA.
Most USTs are used for the storage of motor
fuel (gasoline and diesel fuel) and the regulated
substance that escaped from most leaking
USTs was gasoline. The primary petroleum-
derived contaminants of concern in gasoline are
the aromatic hydrocarbons benzene, toluene,
ethylbenzene, and xylenes (collectively referred
to as "BTEX"). Most state programs treat the
individual MCLs for the BTEX compounds as the
concentration below which the compounds are
not a concern at gasoline release sites. Even today
ground water monitoring at gasoline release sites is
focused on BTEX.
Most state agencies have not routinely monitored
for EDB or 1,2-DCA in ground water. This may
have been due to the fact that lead in gasoline, and
therefore EDB and 1,2-DCA, was being phased
out, or was altogether banned, at the time the state
agencies put their monitoring programs in place.
The South Carolina Department of Health &
Environmental Control (SDHEC) was an exception.
Beginning in 2001, SDHEC started collecting data
on the concentrations of EDB in monitoring wells
at gasoline service stations that were in existence
at a time when leaded gasoline was still available
in the USA. EPA Methods 8260 or 8021, which
are conventionally used for analysis of BTEX
compounds and fuel oxygenates such as MTBE,
do not have adequate sensitivity to determine
concentrations of EDB at its MCL (0.05 ug/L).
-------
The SDHEC required that analyses for EDB be
performed by EPA Method 8011, which has a
method detection limit that is near 0.01 ug/L.
1.3 Investigations of EDB and 1,2-DCA
at Motor Fuel Release Sites
1.3.1 Evaluation of Data from South
Carolina Performed at Clemson
University
Professor R.W. Falta and his students at Clemson
University evaluated the monitoring data on the
distribution of EDB in ground water in South
Carolina, and found that many gasoline release
sites had concentrations of EDB that far exceeded
the MCL (Falta, 2004; Falta et al., 2005). They
analyzed the data available as of December 2004,
and found that 537 underground storage tank sites
had ground water with concentrations in excess of
the MCL (Falta et al., 2005). Figure 2 of Falta et
al., (2005) presented a frequency distribution of the
maximum concentration of EDB in any well at each
individual site. Figure 1.2 plots the data from Falta
et al., (2005). The concentrations of EDB followed
a log-linear distribution with approximately half of
the sites having concentrations of EDB that exceed
the MCL. The median concentration of EDB in
sites where EDB was detected was 4.3 ug/L and
the maximum concentration was 6,550 ug/L.
These results were unexpected and surprised many
ground water scientists and engineers. Staff in
EPA's Region 4 office in Atlanta, GA, brought
Dr. Falta's findings to the attention of the Office
of Underground Storage Tanks (OUST) at EPA's
headquarters in Washington, DC.
100i
90.
80.
70.
60.
50.
40.
30.
20.
10.
0
° South Carolina (Falta Data)
oOUST/ASTSWMO State Survey
0.01 0.1 1 10 100
EDB concentration (u.g/L)
1000 10000
Figure 1.2. Distribution of EDB in ground water
from monitoring wells at gasoline
release sites in South Carolina, and in
sites in the EPA/ASTSWMO Study.
7.3.2 EPA/ASTSWMO Lead Scavengers
Team
The U.S. EPA's OUST and Office of Research
and Development's National Risk Management
Research Laboratory (NRMRL) in association
with the Association of State and Territorial
Solid Waste Management Officials (ATSWMO)
formed a team to determine what problems, if any,
these lead scavengers pose to public health and
the environment. The team's mission statement
outlines an investigation comprised of the
following three phases:
1. Develop an understanding of the potential
problem as it exists today by:
a. Compiling existing background information:
toxicological data; historical usage
information; and occurrence in drinking
water supplies;
b. Evaluating selected state databases and
case files for information on sampling,
monitoring, and remediation at LUST sites;
c. Conducting a study on the effectiveness
and cost of treatment and remediation
technology; and
d. Assess whether or not there are any gaps in
our current knowledge, based on the results
of Phase 1. If so, develop and implement
appropriate measures to fill the gaps.
2. Identify next steps by evaluating the results of
Phases 1 and 2.
Phase 1 culminated in production of a compendium
of information entitled Lead Scavengers
Compendium: Overview of Properties, Occurrence,
and Remedial Technologies (U.S. EPA, 2006). The
compendium represents EPA's state of knowledge
on lead scavengers (through 2005) relating to
historical usage, physical and chemical properties,
analytical methods, environmental fate and
transport, toxicology, occurrence in drinking water
supplies, presence at leaking UST sites, and the
effectiveness and cost of treatment technologies.
In compiling information for the compendium,
some gaps in knowledge were identified, including
the lack of information on the occurrence of
lead scavengers in domestic (private) wells,
the effectiveness of remediation and treatment
technologies, and the magnitude of the occurrence
of lead scavengers in ground water at leaking UST
sites. Filling in this last data gap became the focus
of Phase 2 of the investigation.
To develop information on the distribution of EDB
and 1,2-DCA in ground water at leaking UST
sites in states that did not routinely monitor for
-------
these contaminants, EPA offered to provide free
analysis of samples collected by the states (or their
contractors) from sites that met certain criteria:
• sites that were used for storage and/or
dispensing of leaded gasoline whether or
not they were currently in use (i.e., sites
where USTs were located in 1989 and
earlier) or
• sites where leaded aviation gasoline
(AvGas) or leaded racing fuel was used or
is still being used (i.e., airports, automobile
race tracks) and
• sites that had existing monitoring wells
on-site and were regularly scheduled for
monitoring (this was done to minimize
the burden on states and their contractors;
however, no sites offered as candidates for
sampling were turned down).
Sites meeting the criteria above that were also
within close proximity to a private well or small
community well were of particular interest because
such wells are more vulnerable to contamination
than larger municipal water supply wells.
States that volunteered to participate in the
investigation were provided with a sampling kit for
each candidate site. Typically the kit was shipped
to the contractor that routinely sampled the wells at
the site. Sample vials were filled by the contractor
at the time the wells were sampled for normal
compliance monitoring, and the samples were
returned to NRMRL at the R.S. Kerr Environmental
Research Center in Ada, Oklahoma, for analysis of
EDB by Method 8011, and benzene and 1,2-DCA
by Method 8260. Details of sampling, preservation,
shipment, storage, and analysis are presented in
Appendix A. Samples were provided between
October 2005 and July 2007. A total of 19 states
provided samples from a total of 802 monitoring
wells spread over 102 sites (Table 1.1).
Table 1.1. Occurrence of EDB in ground water provided by the state agencies that participated in the
EPA/ASTSWMO State Study.
State
Arizona
Colorado
Georgia
Maryland
Minnesota
Missouri
Mississippi
Montana
North Carolina
North Dakota
New Hampshire
New Mexico
Oklahoma
South Carolina
South Dakota
Tennessee
Utah
Vermont
Wisconsin
Total
Number Samples
12
107
12
27
29
16
28
31
25
34
63
15
57
50
67
90
37
19
83
802
Number Sites
1
9
2
3
4
8
2
11
6
6
4
1
10
5
5
7
5
3
10
102
Number Sites
EDB detected
1
4
1
2
3
2
2
5
6
2
3
1
6
5
2
3
3
1
2
54
Number Sites
EDB>MCL
1
2
1
2
3
2
2
3
6
0
3
1
5
5
0
1
3
1
2
43
-------
1.4 Purpose and Scope of This Report
This report represents the culmination of Phase 2
of the EPA/ASTSWMO study. It fills some of
the data gaps on the expected distribution of EDB
at gasoline release sites, it identifies a previously
unsuspected role of iron(II) sulfides in the abiotic
transformation of EDB in ground water, and
provides new tools to recognize and use natural
transformation and degradation of EDB and
1,2-DCA as part of a risk management strategy.
-------
2.0
Transport and Fate of EDB and 1,2-DCA at
Motor Fuel Release Sites
This section is intended for technical personnel who
will conduct a risk evaluation for EDB or 1,2-DCA
at specific motor fuel release sites, and for the
technical staff of regulatory agencies that review
the risk evaluations and make decisions concerning
risk management and cleanup of the contamination.
The implementation of Monitored Natural
Attenuation (MNA) as part of a program of risk-
based corrective action for contaminants in ground
water requires a robust understanding of the
exposure to the contaminant, which in turn requires
a robust understanding of the transport and fate
of the contaminant from the point of release to
the receptor. This section presents a conceptual
model for the behavior of a release of EDB and
1,2-DCA in leaded gasoline into the subsurface
environment. It discusses the available information
on the attenuation of EDB and 1,2-DCA caused
by weathering of the original mass of gasoline
released, and the relationship between weathering
and the persistence of EDB and 1,2-DCA in source
areas at motor fuel release sites. This section also
discusses the effect of sorption on the attenuation
of concentrations of EDB and 1,2-DCA along a
flow path in ground water, and it also presents
new information on the rate of abiotic degradation
caused by sulfide in solution in ground water, or
by FeS mineral phases precipitated in the aquifer
matrix as a result of sulfate reducing activity. This
section also discusses prospects for biodegradation
of EDB and 1,2-DCA, and provides data on the rate
of anaerobic biodegradation of EDB and 1,2-DCA
in ground water at gasoline spill sites. It also
describes the application of Compound Specific
Isotope Analysis (CSIA) for documenting the
degradation of EDB and 1,2-DCA.
2.1 Conceptual Model of a Motor Fuel
Release in the Subsurface
Gasoline released from an underground storage
tank seeks the water table. If it is released above
the water table it drains downward by gravity.
If it is released below the water table it rises
by buoyancy. Gasoline can act as a wetting
phase for particles in the unsaturated zone. As a
consequence, it tends to move into the capillary
fringe just above the water table where it is held by
capillary attraction. Overtime, capillary attraction
will re-distribute the gasoline in a roughly circular
or oval shape just above the water table. As the
re-distribution continues, the area contaminated
with liquid phase gasoline gets larger and the
concentration of liquid phase gasoline declines until
the gasoline can no longer maintain a continuous
wetting phase on the aquifer solids. At this point
the gasoline breaks into discrete droplets that are no
longer in contact with each other, the redistribution
of gasoline stops or slows, and the gasoline is said
to be at residual saturation.
There are several ways that soluble contaminants
from gasoline can enter the ground water. Soluble
contaminants can diffuse from the gasoline in
the capillary fringe down into the ground water.
Recharge water draining down through the
unsaturated zone can perfuse the gasoline and carry
soluble contaminants to the ground water. Finally,
variations in the elevation of the water table can
inundate the residual gasoline, allowing soluble
contaminants to partition directly into the ground
water. At most sites, the elevation of the water
table will vary a few inches to a few feet in a year's
time. As the water table moves up and down, the
gasoline in the capillary fringe moves up and down,
creating a "smear zone" that contains gasoline
at residual saturation. Under typical conditions,
the lower portion of the smear zone is below the
average elevation of the water table.
The most contaminated wells at a site are usually
screened in material that has gasoline at residual
saturation. Over time, residual gasoline tends to
accumulate in the geological material with the
finest texture: silt and clays rather than sands or
gravels. This is probably because the material with
-------
finer texture has a higher capillary attraction for the
residual gasoline.
The concentration of soluble contaminants in
ground water produced by the wells is controlled
by equilibrium partitioning of the contaminants
between the ground water and residual gasoline.
As a consequence, the rate of attenuation of
concentrations of EDB overtime in monitoring
wells in the source area of releases is controlled by
the rate at which EDB dissolves from the residual
gasoline into the flowing ground water and is either
flushed away by ground water flow, or destroyed by
biodegradation or abiotic transformation.
The rate of attenuation overtime in sediment
contaminated with residual gasoline determines
how long a release of gasoline can continue to
contaminate ground water. Once a contaminant
enters the flowing ground water and moves away
from the residual fuel in the source area, the
concentration of the contaminate can be attenuated
by processes such as dilution and dispersion,
sorption to native organic matter, biodegradation
by aerobic or anaerobic microorganisms, by neutral
hydrolysis, or by non biological reactions with
either sulfide in solution, or with sulfide minerals.
The rate of removal in the flowing ground water
will determine how far the plume of contamination
can reach.
2.1.1 Mathematical Description of Rates
of Attenuation
The exposure assessment that is conducted as
part the RBCA processes usually incorporates a
mathematical description of the behavior of the
contaminants at a site. This can be as simple as
a few calculations with equations that describe
the transport and fate of the contaminants, or
it can require the calibration of a computer
model to describe the behavior of the plume of
contaminated ground water in time and space. In
any case, the RBCA process requires a quantitative
understanding of the behavior of the contaminant.
The following section reviews the mathematics
typically used to describe the behavior of organic
contaminants in ground water, identifies rates of
removal that are needed for MNA to be a practical
alternative for EDB and 1,2-DCA at many sites,
and reviews the literature on the rates of attenuation
of EDB and 1,2-DCA over time in the source area
of plumes in ground water at gasoline spill sites.
2.1.2 Relationship between a First Order
Rate Constant and a Half Life
Attenuation processes that follow a first order rate
law can be described with either a half life or a
pseudo first order rate constant. Readers that are
familiar with these relationships can skip to the
next section.
When the rate of reaction is proportional to the
concentration of the contaminant, the progress of
the reaction can be described by equation 2.1;
77- = *** (2-1)
^o
where t is the time elapsed during the reaction,
Ct is the concentration after time t, C0 is the
original concentration, and k is the first order rate
constant for the instantaneous rate of change in
concentration over time. When k has a negative
value, concentrations are declining overtime. The
rate constant for the rate of attenuation, the rate of
abiotic transformation, or the rate of biodegradation
is the negative of the rate constant for the rate
of change (k) in Equation 2.1. The solution of
Equation 2.1 for k yields Equation 2.2.
= ln(C,/C0)/,
t
(2.2)
The half life corresponds to the value oft where
one half of the contaminant has been destroyed, as
described in Equation 2.3.
(2.3)
A half life can be converted to a first order rate
constant or vice versa by dividing one parameter
into -0.693 to calculate the other parameter. The
unit for a half life is time (e.g. years), and the unit
for a first order rate constant is reciprocal time (per
year).
Most readers have an intuitive grasp of a half
life, and as a consequence, microbiologists and
other life scientists commonly describe first
order processes with a half life. Engineers tend
to describe first order processes with a first order
rate constant for two important reasons. The
rate of change is directly related to the constant.
In addition, if several first order processes are
operating simultaneously, such as hydrolysis and
biodegradation, the rate constant for the combined
effect is simply the sum of the individual rate
constants. This property is particularly convenient
when calibrating transport and fate models
because several processes can be combined into
one calibration parameter. The remainder of this
report will describe first order processes using first
order rate constants. For readers who are more
comfortable with half lives, and do not have access
to a calculator, Figure 2.1 can be used to convert
first order rate constants in units of per year to half
lives in days, weeks or years.
-------
100,
30
10
3
£ 1.
^ 0.3
£ 0.1 .
0.03
0.01
0.003
0.001
0.0003
0 0001
s YBc""®
* Dai
I'S
•^n
H
0.01 0.03
-•-^
>,
^,
"«»
•*«._,
•^^
0.1 0.3
'•-~^
* L
^*
••*..
n
1 3
-.,.,
10
First Order Rate Constant (per year)
Figure 2.1. Relationship between a first order rate
constant in units of per year and half-
lives in units of days, weeks, and years.
2.1.3 A Definition of a "Generally
Useful" Rate of Attenuation
To put a rate of attenuation into context for natural
attenuation, it is necessary to define a rate that
might be useful for MNA. Any determination
of "useful" is site specific depending on the
hydrological context of a particular release, and the
proximity of receptors. On a site-specific basis,
the useful rate of attenuation is the rate that will
achieve the remedial objectives within a reasonable
time frame (U.S. EPA, 1999). If the degradation
follows first order kinetics, then the first order
rate constant (k) for a "useful" rate of removal is
defined by the relationship in Equation 2.4.
Z«(MCL/Current Concentration)
Time Available
(2.4)
Any release of conventional leaded gasoline is at
least 20 years old. For the purposes of discussion
and comparison, we will arbitrarily define a
"generally useful" rate as the rate that will bring
the concentration of EDB or 1,2-DCA that would
be expected in ground water in contact with
unweathered leaded gasoline to the MCL for EDB
or 1,2-DCA within an additional 20 years.
Falta (2004) used the average composition of
gasoline and partitioning theory to estimate the
concentration of EDB and 1,2-DCA, that would
be expected in ground water in contact with
unweathered leaded gasoline. The expected
concentrations for EDB and 1,2-DCA, were
1,900 ug/L and 3,700 ug/L respectively. The first
order rate of removal that would bring EDB from
the maximum concentration expected (1,900 ug/L)
to the MCL (0.05 ug/L) in 20 years is 0.5 per year,
and the corresponding rate that would bring the
maximum concentration of 1,2-DCA of 3,700 ug/L
to 5 ug/L in 20 years would be 0.33 per year.
Therefore the "generally useful" rate of degradation
of EDB would be 0.5 per year or greater and the
"generally useful" rate of degradation of 1,2-DCA
would be 0.33 per year or greater.
2.2 Attenuation caused by physical
processes
The flow of ground water through the residual
gasoline should weather ("leach") EDB and
1,2-DCA from the residual source material. With
each exchange of the pore water in contact with
residual gasoline, some fraction of the total
amount of EDB and 1,2-DCA would be flushed
away from the source area. Typical values for the
concentration of gasoline at residual saturation
vary between 2,000 and 10,000 mg/kg of Total
Petroleum Hydrocarbon (TPH). Appendix A
provides equations that can be used to predict the
distribution of EDB and 1,2-DCA between residual
gasoline and ground water. At these typical values
for TPH, the fraction of the total amount of EDB
that is dissolved in ground water would vary
between 30% and 7%, and the fraction of 1,2-DCA
in ground water would vary from 40% and 12%.
Because a relatively small proportion of EDB or
1,2-DCA is in the ground water, these contaminants
would be expected to weather slowly from residual
gasoline. This leads to two questions: How fast
is EDB and 1,2-DCA weathered from gasoline?
What are the expected concentrations of EDB and
1,2-DCA in the plume?
2.2.1 Physical Weathering from Fuel
Present at Residual Saturation
The rate of exchange of pore water in contact
with residual gasoline will depend on the seepage
velocity of ground water and on the distance
ground water must move to traverse the region
with residual gasoline. The faster the ground water
moves, the faster the EDB and 1,2-DCA should
be removed, therefore, higher seepage velocities
should be associated with lower concentrations
of EDB. The average seepage velocity of ground
water is usually calculated from an estimate of
hydraulic conductivity that is acquired from a
slug test on a monitoring well. Because most
monitoring wells are screened across materials with
different textures, the average seepage velocity may
underestimate the local seepage velocity through
the material contaminated with residual gasoline.
-------
The South Carolina Dept. of Health &
Environmental Control collected data on the
hydrological characteristics of 150 MTBE plumes
in South Carolina (Personal Communication,
Read Miner, South Carolina Dept. of Health &
Environmental Control). The seepage velocity
at each plume was estimated by multiplying the
hydraulic gradient by the hydraulic conductivity,
then dividing by 0.3 as an estimate of the effective
porosity. Figure 2.2 presents the frequency
distribution of seepage velocity in the plumes. The
median seepage velocity was 4 m/yr, and only
10% of plumes exceeded a velocity of 50 m/yr.
Measurements in the same data set indicated that
the median width of the source area for MTBE
plumes was 100 feet or 30 meters.
1.0,
0.9
0.8
0.7.
0.6.
0.5.
0.4
0.3
0.2.
0.1
0.0
0.1 1 10 100
Seepage Velocity (m/yr)
1000
Figure 2.2. Distribution of seepage velocity in
MTBE plumes in South Carolina
(2002).
The following thought experiment is offered to
put the rate of weathering of EDB and 1,2-DCA in
context for risk management. Assume, following
Falta (2004), that the initial concentrations of EDB
and 1,2-DCA are 1,900 ug/Land 3,700 ug/L.
The distribution of EDB and 1,2-DCA between
water and non-aqueous phase gasoline is controlled
by the partitioning coefficient between water and
gasoline, and by the relative proportions of pore
water and gasoline in the aquifer. Appendix A
derives equations that can be used to predict the
distribution of EDB and 1,2-DCA based on the
parameters. Assume that the concentration of
residual gasoline is near 5,000 mg/kg, and that the
total porosity is 30%. The equations in Appendix
A would predict that 12% of EDB and 20% of
1,2-DCA is removed whenever one pore volume of
ground water in contact with the residual gasoline
is exchanged. The rate of exchange is the length
of the region with residual gasoline divided by the
seepage velocity of ground water.
Assume that the footprint of residual gasoline from
releases of leaded gasoline had the same geometry
as the releases of gasoline with MTBE, and that the
length of the region with residual gasoline in the
direction of ground water flow is equal to the width
perpendicular to the flow. The assumed length is
30 meters.
Figure 2.3 projects the time course of
concentrations of EDB and 1,2-DCA in ground
water in contact with residual gasoline if the
ground water seepage velocity is 4 m/yr (median
velocity in South Carolina) or 50 m/yr (faster than
90% of sites). The year 1987 was taken as the last
year for a plausible release of EDB from leaded
gasoline in an underground storage tank. The
EPA/ASTSWMO study was conducted twenty
years later in 2006 and 2007.
10000
1000
3
-^ 100
I 10
1 1
o
o
0.1
0.01
EDB MCL
1987 1997
2007 2017
Date
2027 2037
Figure 2.3. Thought experiment on the effect of the
rate of ground water flow on the con-
centration of EDB and DCA in ground
water in contact with residual leaded
motor fuel. (Consult the text for as-
sumptions in the comparisons.)
If the seepage velocity of ground water was
50 m/yr, the concentration of EDB in 2007 would
be near 27 ug/L. While this represents a 70-fold
reduction in concentration from the original
concentration of EDB, the concentration is still
540-fold greater than the MCL. The concentration
of EDB would not be expected to reach the MCL
until 2037. If the seepage velocity were 4 m/year,
the concentration of EDB would be reduced by
less than 30% by 2007. It is reasonable to expect
significant concentrations of EDB to remain in
ground water at many releases of leaded gasoline.
In the case of 1,2-DCA, when the seepage velocity
is 50 m/yr, concentrations of 1,2-DCA would be
-------
expected to meet the MCL before samples were
collected for the study in 2006 and 2007. If the
seepage velocity was 4 m/yr, concentrations of
1,2-DCA would only be reduced to 50% of the
original concentration. Because 1,2-DCA partitions
to ground water more readily than EDB, it weathers
more rapidly, but there is little practical difference
in their behavior as it will take decades to centuries
to reach their respective MCLs.
The survey conducted by the EPA/ASTSWMO
Lead Scavengers Team provides a small data set
that can be used to validate the thought experiment.
Case workers in the state agencies were asked
to review files on the sites, and provide either an
estimate of the seepage velocity of ground water
from the file, or data on the hydraulic gradient and
hydraulic conductivity that could be used with a
reasonable estimate of effective porosity to estimate
the seepage velocity of ground water. Data are
available from ten sites.
Figure 2.4 compares the average seepage velocity
of ground water at the sites to the highest
concentrations of EDB at each site in 2006. In
general, the concentrations were lower than would
be expected from the thought experiment. In eight
of the ten sites, the maximum concentration of
EDB was 1 (ig/L or less, regardless of the seepage
velocity of the ground water. This would indicate
that some process other than leaching controlled the
concentrations of EDB at these eight locations. The
sites with the highest concentrations of EDB had
seepage velocities of 11.5 and 71 meters per year.
These velocities are relatively high, exceeding 72%
and 92% of sites in South Carolina respectively
(compare Figure 2.2). Despite these high seepage
velocities, the concentrations of EDB were above
400 (ig/L, consistent with the slow rates of physical
weathering predicted by the thought experiment.
1000
Moo
CO
Q
LU
I
0.1
0.01
10 100
Seepage Velocity (meter per year)
1000
Figure 2.4. Association of concentrations of EDB
in the most contaminated wells at
each of ten sites with estimates of the
seepage velocity of ground water at the
sites.
2.2.2 Sorption on Native Organic Matter
As long as the ground water is in contact with
residual gasoline, the concentration of EDB or
1,2-DCA is controlled by the concentration of EDB
and 1,2-DCA in the gasoline. Once contaminated
ground water moves away from the area with
gasoline at residual saturation, the concentrations
of EDB or 1,2-DCA are controlled by sorption
to solids in the aquifer matrix, by biological
degradation and abiotic transformation, and by
dilution and dispersion in ground water.
Retardation due to sorption is not an important
process contributing to natural attenuation of EDB
or 1,2-DCA in ground water. Table 2.1 compares
the retardation ratio of EDB, 1,2-DCA, MTBE,
benzene, toluene, and xylenes that is expected
from sorption of the contaminant to the native
organic carbon in the aquifer. The retardation ratio
is simply the rate of movement of water divided
by the rate of movement of the contaminants.
As discussed in Wiedemeier et al., (1999), the
estimates of the retardation ratio in Table 2.1 are
based on Equation 2.5;
^oc Joe I™ p.5)
where R is the retardation ratio, pb is the bulk
density, Koc is the partition coefficient of the
organic contaminant between ground water and
native organic matter, foc is the fraction organic
carbon in the aquifer matrix material, and 0 is
the water filled porosity. In Table 2.1, values for
Koc are taken from Wiedemeier et al., (1999),
except for the value of Koc for EDB. Aronson and
Howard (2008) concluded that "Soil-water partition
coefficients (Koc values) for EDB in the solution
-------
Table 2.1. Retardation in ground water due to sorption that is expected from the organic matter content of
the aquifer solids.
Compound
MTBE
Benzene
DCA
EDB
Toluene
Xylenes
Koc (L/kg)
12
38
58
65
135
240
Fraction of Organic Carbon in sediment
0.01%
Low for Aquifers
1.0
1.0
1.0
1.0
1.1
1.1
0. 1%
Median for
Aquifers
1.1
1.2
1.3
1.3
1.7
2.2
1%
High for
Aquifers
1.6
2.9
3.8
4.1
7.6
13
10%
Typical of soils
7
20
29
33
68
120
phase range from 12 to 160 L/kg ... but average
about 50 to 65 L/kg." The calculation in Table 2.1
assumes a value of 65 L/kg for the Koc of EDB.
The extent of sorption of EDB and 1,2-DCA
is intermediate between the extent of sorption
of benzene and toluene. At concentrations of
organic carbon in the aquifer solids that are near
0.01%, there should be little retardation of EDB or
1,2-DCA. At concentrations of organic carbon in
the aquifer solids near 0.1%, retardation of EDB
and 1,2-DCA is of no practical consequence. These
concentrations are typical of aquifers with low or
average concentrations of organic matter in the
aquifer solids. At relatively high concentrations of
organic carbon in the aquifer solids near 1%, the
expected retardation ratio for EDB and 1,2-DCA is
near four to one.
Sorption can reduce concentrations in plumes that
are expanding, but when a plume reaches steady
state, sorption does not influence the concentration
of the contaminant in ground water. As discussed
in Section 1, the use of leaded gasoline for
conventional motor gasoline was minimal after
1988. At the time of this writing, almost every
plume of EDB or 1,2-DCA from a release of
leaded gasoline from an underground storage tank
at a conventional gasoline service station is at
least twenty years old. In this time interval, it is
reasonable to presume that most plumes of EDB or
1,2-DCA have reached (or are approaching) steady
state. Sorption can not be an important mechanism
that will bring high concentrations of EDB or
1,2-DCA to their MCLs.
2.2.3 Rate Constants for Physical
Weathering of EDB and 1,2-DCA
There is very little data in the literature on trends of
EDB concentrations in monitoring wells at gasoline
release sites. Bulsara (2004) fitted first order rate
constants for attenuation of EDB to monitoring
data from 55 wells in South Carolina. The data are
replotted in Figure 2.5. The rates varied widely,
from approximately - 4 per year to + 4 per year. In
32 of the 55 wells, the rate constants were negative,
indicating that the concentrations of EDB increased
over time.
As a comparison, Figure 2.5 plots data for the
attenuation of MTBE in monitoring wells at
gasoline release sites (Wilson et al., 2005). At
6 out of 20 release sites, the rate constants were
negative, indicating that concentrations of MTBE
were increasing. However, the variation in rates of
attenuation was less, from -0.52 to +0.75 per year.
The range in rates of attenuation of MTBE was
much less than the range in rates of attenuation of
EDB. There is no obvious reason why this should
be the case. The compounds have very similar
physical properties.
-------
-5-4-3-2-10 1 2 34 5
First Order Rate Constant for Decline in Concentration (per year)
Figure 2.5. Rate of decline of concentrations of
EDB overtime in 55 monitoring wells
in South Carolina (Bulsara, 2004).
Rate of decline of concentrations of
MTBE at 20 sites in the USA (Wilson
et al, 2005) provided for reference.
Example data provided in Falta (2004) indicated
that the rates for attenuation of EDB were extracted
from three or four measurements extending over a
monitoring period of one or two years. In contrast,
the rates of MTBE attenuation were extracted from
6 to 34 measurements (mean of 14 measurements)
extending over a minimum of two years. The
comparison of the EDB rate constants to the
MTBE rate constants suggests that rate constants
extracted from small data sets collected over short
time periods may have large uncertainty. If a
rate constant extracted from monitoring data at a
particular release site is used to estimate the time
required for concentrations of EDB to decline
below the MCL, the rate should be verified and
validated by continued long term monitoring.
2.3 Attenuation Caused by Abiotic
Transformation or Biodegradation
For many processes that destroy contaminants,
the rate of the reaction is proportional to the
concentration of the contaminant at any moment
in time. This is particularly true for abiotic
reactions such as hydrolysis or abiotic reactions
with minerals. For biodegradation reactions at low
concentrations, the diffusion of the contaminant to
the organism, and association of the contaminant
with the enzymatic machinery of the organism, is
rate limiting. At these lower concentrations, the
rate of reaction is proportional to the concentration
of the contaminant. At higher concentrations, the
processing of the contaminant by the enzymatic
machinery becomes rate limiting, and the rate of
reaction is proportional to the density of cells,
not the concentration of the contaminant. This
transition in the rate limiting step for anaerobic
biodegradation of EDB should be near 1,000 (ig/L.
At typical concentrations at releases of leaded
gasoline, the rate of EDB biodegradation should be
proportional to the concentration of EDB, and the
process can be described with first order kinetics.
2.3.1 Neutral Hydrolysis of EDB and
1,2-DCA
Both EDB and 1,2-DCA are slowly hydrolyzed in
water. Barbash and Reinhard (1989) reported a
half life for neutral hydrolysis of EDB of 22 years
at 15 °C and 4.6 years at 25 °C. These half lives
correspond to first order rate constants of 0.073
per year at 15 °C and 0.15 per year at 25 °C. The
rate of neutral hydrolysis of 1,2-DCA was slower;
the half life was 300 years at 15 °C and 64 years at
25 °C, corresponding to first order rate constants
of 0.0023 per year at 15 °C and 0.0108 per year at
25 °C.
Hydrolysis of EDB can produce either
2-bromoethanol or vinyl bromide (Pignatello and
Cohen, 1990; Aronson and Howard, 2008), with
2-bromoethanol being the dominant product.
Similarly, 1,2-DCA hydrolyzes to 2-chloroethanol
(Jeffers et al., 1989) and vinyl chloride with vinyl
chloride being a minor product (Barbash and
Reinhard, 1989).
Both EDB and 1,2-DCA can undergo base
catalyzed hydrolysis; however, the rate of the base
catalyzed hydrolysis is not important at the pH of
natural ground waters. At pH 9, the base catalyzed
reaction represents only 10% of the total hydrolysis
reaction (Jeffers and Wolfe, 1996).
2.3.2 Abiotic Transformation of EDB and
1,2-DCA by Iron(ll) Sulfide
Both EDB and 1,2-DCA can also react with sulfide
as H2S and HS~ to produce various thiols and
thioethers (Schwarzenbach et al., 1985). Barbash
and Reinhard (1989) compared the expected rates
of hydrolysis and reaction with sulfide at 15 °C and
25 °C. In Figure 2.6, the combined rate constants
for hydrolysis and reaction with sulfide are used
to project the effects of the concentration of total
sulfide at pH 7 on the rate of removal of EDB and
1,2-DCA. In general, the rate of reaction of EDB
is approximately ten times the rate of reaction of
1,2-DCA (compare the scales of the vertical axes
in Figure 2.6). The reaction of both EDB and
1,2-DCA is sensitive to temperature; rates at 25 °C
are approximately five fold faster than at 15 °C.
Reactions with sulfide only become important at
concentrations above 0.2 mg/L for EDB and 1 mg/L
for 1,2-DCA.
-------
10 ,
CO H
t ;
)
3
0 001 •
—
4fl
DC
0.01
W=^
»A
--
0.1
—
• '"^
- -^
^
y '
--
25°
/
T
V
/
''
7^
1 10
Sulfide (mg/L)
/
'
^
100
/
-1
/
'
5(
c
1000
Figure 2.6. Effect of concentrations of sulfide at
pH 7 on the rate of abiotic transforma-
tion of EDB and DCA.
Depending on temperature, the "generally
useful" rate of EDB degradation of 0.5 per year
is attained at concentrations between 2 and 10
mg/L total sulfide at pH 7. The "generally useful"
rate for 1,2-DCA of 0.33 per year is attained at
concentrations between 90 and 400 mg/L of sulfide.
Sulfide produced as the end product of sulfate
reduction can react with iron(III) minerals in the
aquifer matrix to produce various mineral phases of
iron(II) sulfide according to the following reactions
(Shen and Wilson, 2007).
4H2
2Fe+
SO4 2
+ 3S-2
s-2
•2FeS
4H2O
FeS + S° -> FeS2
The overall reaction consumes three moles of
sulfate to produce one mole of FeS and one mole of
FeS2.
12H2 + 3SO4-2 + 2Fe+3 -> FeS + FeS2 + 24H2O
The transformation of chlorinated hydrocarbons
such as TCE and cis-DCE by FeS and FeS2 is well
documented in the literature (Butler and Hayes,
1999; Butler and Hayes, 2001; Shen and Wilson,
2007; Liang et al., 2007). Shen and Wilson (2007)
reported abiotic transformation of TCE in four
laboratory columns constructed with river sand and
shredded plant mulch. Sulfate reduction supported
by the plant mulch produced sulfide, which reacted
with Fe(III) minerals in the river sand to produce
non crystalline or poorly crystalline FeS. Two of
the columns were supplemented with hematite to
encourage precipitation of FeS in the columns. In
these two columns, the first order rate constant for
reaction of TCE with the FeS varied from 0.53 to
2.3 per day per mole of FeS in contact with a liter
of pore water.
The EPA/ASTSWMO Lead Scavengers Team
conducted an experiment to determine whether
EDB or 1,2-DCA could be degraded by FeS. To
evaluate the capacity of biogenic FeS to degrade
EDB and 1,2-DCA, the column described by Shen
and Wilson (2007) as the Column with Mulch and
Hematite was frozen, cut into sections while frozen,
and then the sections were allowed to thaw under
an oxygen free atmosphere in a glove box. To
remove the confounding effects of sorption to the
plant mulch on concentrations of contaminants, the
sections were sieved to remove the plant mulch.
The sediment and pore water were transferred to
20 ml serum vials, dosed with solutions of EDB
or 1,2-DCA, and incubated at room temperature.
Experimental details are provided in Appendix B.
The removal of EDB and 1,2-DCA followed
first order kinetics without a lag, indicating that
removal was an abiotic process that did not
require acclimation of an active biological process
(Figure 2.7). The removal of EDB was more
rapid than the removal of 1,2-DCA. The first
order rate constants for degradation of EDB or
1,2-DCA were extracted from the data as the slope
of a linear regression of the natural logarithm of
concentration on time of incubation. The rate
constants are presented in Table 2.2. At the end of
the incubation, the concentration of Acid Volatile
Sulfide or AVS was measured in each of the vials
(mg/kg dry sediment). The AVS was taken as an
estimate of FeS in the sediment. The water content
of the wet sediment was determined by drying the
sediment. The concentration of AVS in each vial
was expressed as the millimoles AVS exposed to
each liter of pore water. Finally, the rate of removal
was normalized to the millimoles AVS exposed to
the pore water.
-------
100000 n
EDB
100000
5 10 15 20 25 30 35
Time of Incubation (days)
DCA
Concentration (ug
->• o
-* O O
§ § §
100000
10000
1000
100
V "6-17 n§-1§
\ -6-19 '%-]l
X
D 100 200 300
5 10 15 20 25 30 35
Time of Incubation (days)
The normalized rate constants for transformation
of EDB were consistent between the experimental
vials. The average normalized rate for EDB
degradation was 0.285 ± 0.091 per year per
millimole AVS exposed to a liter of water at 95%
confidence. The rate of transformation of EDB
was in the same range as the rates of abiotic
transformation of TCE in the original column.
The rate of 1,2-DCA degradation was an order of
magnitude lower at 0.0263 per year per millimole
AVS exposed to a liter of water.
Analysis of acid volatile sulfide is simple
to perform, and is commercially available.
Figure 2.8 presents the rate of abiotic degradation
of EDB and 1,2-DCA that would be expected
if the kinetics of degradation follow the rates
of degradation presented in Figure 2.7. The
detection limit for the analysis is near 2 to 3 mg/kg.
The rates of abiotic degradation of EDB are
"generally useful" at concentrations of AVS near
10 mg/kg. Concentrations of AVS must approach
100 mg/kg to provide "generally useful" rates of
transformation of 1,2-DCA.
Figure 2.7. Removal of EDB or DCA in the pres-
ence of biogenic FeS in batch experi-
ments at pH near 7.
Table 2.2. Comparison of the rate of transformation of EDB and DCA by iron(II) sulfide to the rate or
removal of TCE.
Compound
EDB
EDB
EDB
EDB
EDB
DCA
DCA
Experimental unit
2-10
2-11
2-12
6-15
6-s4
6-16
6-17
Rate of Removal
yr1
76.4
65.1
64.9
62.6
94.8
10.2
6.7
Concentration FeS
as AVS*
mM**
198
272
278
381
236
293
140
Rate of Removal
yr^mM"1
0.386
0.239
0.233
0.164
0.402
0.0348
0.0479
Data from Table 1 of Shen and Wilson (2007)
TCE
TCE
TCE
Entire Column
Day 383
Entire Column
Day 578
Entire Column
Day 793
80
193
55
138
230
283
0.584
0.840
0.193
*Acid Volatile Sulfide
** millimoles AVS in contact with 1.0 liter pore water.
-------
10
0.1
0.01
— EDB
--DCA
10
Acid Volatile Sulfide (mg/kg)
100
Figure 2.8. Expected rates of abiotic transforma-
tion of EDB or 1,2-DCAby FeS in
aquifer sediment, assuming a water-
filled porosity of 25%, and pH near 7.
The abiotic reaction of EDB or 1,2-DCA occurs on
the surface of the solid iron(II) sulfide. As a result,
the rate of the reaction should be proportional to
the surface area and not to the mass of the iron(II)
sulfide, and the reaction rate will be faster if the
iron(II) sulfide is finely divided. The rate of abiotic
transformation of EDB and 1,2-DCA has only been
determined for this one preparation of biogenic
iron(II) sulfide. It is not known at present if the
properties of biogenic iron(II) sulfide in aquifer
sediments vary from site to site. The projections
in Figure 2.8 are provisional until more data can be
obtained from field sites.
Butler and Hayes (2001) found that the rate of
transformation of TCE on iron(II) sulfide decreased
as the pH decreased. There is no information on
the effect of pH on the rate of transformation of
EDB and 1,2-DCA. However, it is possible that
the rates will be substantially lower at pH less than
6.5. Based on sustained concentrations of sulfide in
ground water and the temperature, reactions with
sulfide can make a major contribution to removal
of EDB and 1,2-DCA in ground water, once the
water moves away from the area of the plume with
residual gasoline.
2.3.3 Biodegradation of EDB and
1,2-DCA
Both EDB and 1,2-DCA can be rapidly degraded if
oxygen is available (Aronson and Howard, 2008).
Oxygen may be available in the far down gradient
portion of the plume where metabolism of the
petroleum hydrocarbons is complete, and diffusion
and dispersion have mixed oxygen from un-
impacted ground water into the plume. However,
the oxygen demand of the petroleum hydrocarbons
make it unlikely that oxygen will be available in
the source area of a plume from a release of leaded
gasoline, or in the mid gradient portions of the
plume.
Methane producing bacteria can co-metabolize
EDB and 1,2-DCA to ethylene when the cells are
grown on molecular hydrogen and carbon dioxide
(Belay and Daniels, 1987). The reaction does
not support growth of cells, and the halogenated
compounds can be harmful to the methane
producing bacteria. Concentrations of EDB
near 1,300 ug/L inhibit growth of the methane
producing bacteria by 90%. In contrast, 1,2-DCA
has less effect; higher concentrations of 1,2-DCA
near 11,000 ug/L are required to inhibit growth.
Bacteria in the Dehalococcoides group can use
EDB or 1,2-DCA as a substrate. These bacteria
grow with molecular hydrogen as a source of
reducing power, and use EDB or 1,2-DCA as
the electron acceptor, basically as something to
breathe. Strain Dehalococcoides BAVI (He et
al, 2003a) can grow while metabolizing EDB to
ethylene, and Dehalococcoides ethanogenes 195
can grow on 1,2-DCA (Maymo-Gatell et al., 1999)
and metabolize either EDB or 1,2-DCA to ethylene
(Maymo-Gatell et al., 1997).
No information is available in the literature on the
effect of concentrations of EDB on the rate of EDB
degradation by Dehalococcoides strains. The effect
is usually described by the half saturation constant,
the concentration at which the rate is one-half of the
maximum possible rate. This is the concentration
where the rate of degradation becomes dependent
on the density of active organisms instead of the
concentration of the substrate. He et al., (2003b)
published half saturation constants for degradation
of vinyl chloride, cis-DCE, and trans-DCE by
the mixed culture from which Dehalococcoides
BAV1 was isolated. The values were 5.8, 8.9, and
8.5 uM for vinyl chloride, cis-DCE, and trans-DCE
respectively. Cupples et al., (2004) determined half
saturation constants for vinyl chloride and cis-DCE
in a mixed culture containing the bacterium VS
strain of Dehalococcoides. The values of the half
saturation constant were 2.6 and 3.3 uM for vinyl
chloride and cis-DCE respectively.
The value of the half saturation constant is
controlled in part by the affinity of the substrate
for microbial tissue, and for the active site of
the enzyme. The affinity is inversely related to
the water solubility of the substrate. The water
solubility of EDB, vinyl chloride, and cis-DCE
are 4,300 mg/L, 2,700 mg/L and 800 mg/L
respectively.
-------
The closest match to EDB is vinyl chloride. If
the half saturation constant of vinyl chloride is
considered the best estimate of the half saturation
constant for EDB, the constant would vary between
5.8 millimolar and 2.6 millimolar, or 1,100 to
490 ng/L.
Considering the effect of EDB on the growth of
methane producing bacteria along with the half
saturation constant for Dehalococcoides strains,
a value of 1,000 ug/L can be considered an upper
boundary where biodegradation of EDB can be
described by a first order rate law.
2.3.3.1 Rate Constants for
Biodegradation of EDB and
1,2-DCA
Table 2.3 compares the first order rate constants for
removal of EDB, 1,2-DCA, and benzene that were
extracted from laboratory studies using anaerobic
aquifer sediment. The table is restricted to data
from sites that were actively methanogenic and
would reflect the geochemical environment in
the area of a release of leaded gasoline that had
residual gasoline, as well as, the area in the plume
that is immediately down gradient of the area
with residual gasoline where BTEX compounds
persisted in the ground water. The rates of
degradation of EDB, 1,2-DCA, and benzene in
the laboratory microcosm studies all had a fairly
narrow range, extending from 17 per year to 0.3 per
year.
In three different studies, the rate of anaerobic
biodegradation of EDB was equivalent to the rate
of biodegradation of benzene. The rate constants
attained in the laboratory studies meet the arbitrary
criteria established earlier for "generally useful"
rate constants for applications to MNA (0.5 per
year for EDB and 0.33 per year for 1,2-DCA).
Anaerobic biodegradation can make an important
contribution to MNA of EDB and 1,2-DCA in
contaminated ground water, particularly under
methane producing conditions.
Table 2.3 also compares rate constants that were
extracted from the distribution of contaminants
along the flow path in the aquifer. In Table 2.3, the
rate constant was calculated from data provided
in Mayer (2006) by multiplying the first order rate
constant for attenuation of concentrations with
distance along the flow path by the upper value of
the range of the ground water seepage velocity at
the site.
Washington and Cameron (2001) warn that field
scale rates of attenuation include the effects of
sorption and dispersion and dilution. Data in
Table 2.3 are from old releases that have reached
their maximum extent and are stable. They have
likely reached sorptive equilibrium. The length of
plumes reported by Henderson et al., (2008) and
Mayer et al., (2006) is relatively short compared
to the likely width of the source area. The rate
constants reported by Ravi et al., (1998) are
corrected for dispersion and dilution. In the data
from Table 2.3, it is most likely that attenuation at
field scale is dominated by biodegradation, which
would explain the reasonable correspondence
between the field scale rates and the laboratory
studies. In general, the rate constants in the field
are smaller than those extracted from the laboratory
studies, corresponding to the lower end of the range
of laboratory rates.
There is one important exception to the reasonable
correspondence in rate constants. Falta (2004)
extracted rates of attenuation of EDB and
benzene from the distribution of their plumes
of contamination that resulted from a release of
aviation gasoline at the Massachusetts Military
Reservation on Cape Cod, Massachusetts. The
rate constant for EDB biodegradation was ten fold
to one hundred fold slower than the rate constants
seen at other sites (Table 2.3). The release was into
an aquifer comprised of poorly sorted fine to coarse
sands. The seepage velocity of ground water was
in the range of 0.3 to 0.6 m per day. There may
have been no opportunity to develop the strongly
reducing conditions that would lead to BTEX
fermentation to produce the H2 that is needed by the
organisms that biodegrade EDB.
The study of Henderson et al., (2008) is careful
and comprehensive and is published in a respected
peer reviewed journal. The study of Mayer
(2006) uses conventional and generally accepted
approaches to analyze data that was collected at
the site for other purposes. These two studies are
the only useful case studies in the literature on
natural biodegradation of EDB and 1,2-DCA at
gasoline release sites. It would be inappropriate
to uncritically extrapolate the behavior of EDB
and 1,2-DCA at these two sites to hundreds of
thousands of sites across the United States. At
present, the general contribution of biodegradation
at sites across the United States is unknown.
The appropriate application of MNA, or risk
management, requires a site-specific knowledge of
the behavior of the contaminants of concern. It is
important to remember that the data in Table 2.3
illustrate what might be possible at other sites.
Data from Table 2.3 should not be used in the place
of site specific data to conduct a risk evaluation at
other gasoline release sites.
-------
Table 2.3. Comparison of first order rate constants for biodegradation of EDB and 1,2-DCA in anaerobic
aquifer sediment to rate constants for overall removal with ground water flow in anaerobic aquifers.
Material
First Order Rate Constant for
Attenuation (per year)
EDB
DCA
Benzene
Reference
Microcosm studies in laboratory, all are conducted with methanogenic material.
sediment from source zone of a spill of
leaded gasoline, South Carolina
sediment from mid gradient zone of a spill
of leaded gasoline, South Carolina
sediment impacted by leachate from a solid
municipal waste landfill, Norman, Oklahoma
sediment impacted by leachate from a solid
municipal waste landfill, Norman, Oklahoma
sediment from manufacturing site
contaminated with DCA in Louisiana
sediment from manufacturing site
contaminated with DCA in Texas
1.5±1.0
5.4±0.3
17
1.3±0.3
0.3±0.1
1.7
4.4
1.2
1.4±0.2
3.5±0.8
2.6
Henderson et al.,
2008, SI
Henderson et al.,
2008, SI
Wilson etal., 1986
Kleckaetal., 1998
Kleckaetal., 1998
Kleckaetal., 1998
Field studies, flow path in aquifer
spill of leaded gasoline, South Carolina
spill of leaded gasoline, North Carolina
(1995 data)
spill of leaded gasoline, North Carolina
(2004 data)
leachate from municipal solid waste landfill,
Michigan
fs-12 spill of aviation gasoline on Cape Cod,
Massachusetts
1.3
0.63
0.22
0.03
0.9
0.71
0.22±0.19
1.0
0.9
0.26
0.42±0.32
0.14
Henderson et al.,
2008, Supporting
Information
Mayer, 2006
Mayer, 2006
Ravi etal., 1998
Falta, 2004
2.3.3.2 Association of Geochemical
Parameters with Removal of
EDB and 1,2-DCA
Both EDB and 1,2-DCA are biodegradable under
aerobic and anaerobic conditions. Based on the
approach taken in the Technical Protocol for
Evaluating Natural Attenuation of Chlorinated
Solvents in Ground Water (Wiedemeier et al.,
1998), it may be possible to use geochemical
parameters to identify environments where
processes that destroy EDB are favored. This sub-
section examines the association of concentrations
of EDB with concentrations of the biogeochemical
parameters nitrate-nitrogen, sulfate and sulfide, and
methane. The association is evaluated to determine
if there is information in these parameters that can
be used to predict concentrations of EDB.
The sub-section is intended for staff of state
regulatory agencies that determine which
parameters are monitored in ground water
contaminated by releases of leaded gasoline. The
cost of sampling ground water and analyzing
the samples for EDB could be reduced if it were
possible to predict beforehand whether high
concentrations of EDB should be expected at a
particular gasoline release site.
To determine whether geochemical or hydrological
parameters can be used to predict concentrations
of EDB, certain sites from the EPA/ASTSWMO
study that were sampled in 2006 were selected to be
resampled in 2007 for concentrations of EDB and
for concentrations of nitrate, sulfate, and methane.
In the data set there was one site each from
Colorado, Missouri, Montana, Tennessee, Utah,
-------
and Vermont. Two sites each were selected from
Oklahoma, New Hampshire, and South Carolina,
and three sites were selected from Minnesota.
2.3.3.2.1 Dissolved Oxygen
Oxygen is particularly difficult to determine in
monitoring wells. Collecting water samples
with a bailer tends to introduce oxygen from the
atmosphere into the sample so that the measured
dissolved oxygen content of the sample is higher
than that of the ambient ground water. This
concern is illustrated in data that were collected
by Kolhatkar et al., (2000) to investigate the role
of various geochemical parameters in the natural
biodegradation of MTBE in ground water. In 1999
and 2000, ground water samples were collected
from wells at 74 sites in Pennsylvania (41 sites),
New Jersey (8), New York (5), Florida (5), Indiana
(7), Maryland (3), Washington DC (2), and Ohio
(3). Analyses of dissolved oxygen, iron(III), and
hydrogen sulfide were performed on site by staff
of the contractors or consultants that normally
sampled the monitoring wells using procedures
for purging the wells that were prescribed by the
appropriate state agencies. Because iron(II) is
rapidly oxidized to iron(III) in the presence of
molecular oxygen, detectable concentrations of
iron(II) and dissolved oxygen should not occur
together in ground water. Yet, in water produced
from the monitoring wells (Figure 2.9), the
sampling technicians reported high concentrations
of oxygen in samples with high concentrations of
iron(II). Because of the difficulty with measuring
dissolved oxygen in water from monitoring wells,
data on dissolved oxygen was not collected during
the survey conducted by the EPA/ASTSWMO
Lead Scavengers Team.
Is.
-|
0)
1
s
in
Q
c
_o
1
0)
c
o
O
16-,
14-
12-
10-
8^
6-
4-
2
n
(
*• » *
* - * 4
: * . * * i
*
** * * * i
L:. . '•• . . :
b « I
•*'..*•. : . . . '
• •', , I :
^ ' '• .' : 4 I
S- !.j: '«': .-t i
!/" .' '*.".' '
:.. ' . " ' •
3 5 10 15 20 25
f^nrmontratinn nf FoMH (mnl\ \
Figure 2.9. Co-occurrence of dissolved oxygen
and reduced iron in water samples
produced from monitoring wells at
gasoline spill sites.
2.3.3.2.2 Nitrate
Concentrations of nitrate were used as a surrogate
for the availability of dissolved oxygen in ground
water. Higher concentrations of nitrate (and
by inference of oxygen) would support aerobic
biodegradation of EDB, and should be correlated
with lower concentrations of EDB.
Figure 2.10 compares concentrations of nitrate to
concentrations of EDB in the most contaminated
well at each release site in the EPA/ASTSWMO
study. Notice that the same data are plotted on an
arithmetic and logarithmic scale for concentrations
of nitrate nitrogen. The sites were selected to
include both sites with high concentrations of EDB
as well as sites with low concentrations of EDB.
At five of the selected sites, the concentration
of EDB was less than the reporting limit of
0.01 ug/L. These sites had low but detectable
concentrations of EDB in 2006.
1000
100
4 10
00
Q
LU
1
0.1
U.U 1
0
1000
100
10
.01 0.1 1
Nitrate Nitrogen (mg/L)
»
10
CD -I
Q '
111
0.1
0.01
234567
Nitrate Nitrogen (mg/L)
Figure 2.10. Association of concentrations of
nitrate with concentrations of EDB
in the most contaminated monitoring
wells at sites.
There was no obvious relationship between
concentrations of EDB and concentrations of
nitrate nitrogen. The concentration of nitrate
nitrogen in three of fifteen sites was less than the
detection limit of 0.01 mg/L, and the concentration
-------
in eleven of fifteen sites was less than 1.0 mg/L.
At these sites, the concentration of EDB was 10
ug/L or less. Contrary to what was expected, in the
two sites with the highest concentration of EDB
(711 and 147 ug/L), the concentration of nitrate
nitrogen was above 1 mg/L. There was no evidence
that high nitrate concentrations (and by inference
high oxygen concentrations) supported aerobic
biodegradation of EDB.
The presence of detectable concentrations of nitrate
nitrogen in the wells with high concentrations
of EDB is problematic. Presumably the high
concentrations of EDB were associated with wells
screened near the source of residual gasoline. It is
possible that the screens of the monitoring wells
intercepted uncontaminated ground water over
some depth interval, and contaminated water over
another depth interval. This resulted in a sample of
mixed water being collected and analyzed. When
waters from different geochemical environments
are mixed in a monitoring well, it will be difficult
or impossible to use geochemical parameters to
interpret both the geochemical nature of the ground
water environment and the contaminant plume.
2.3.3.2.3 Sulfate and Sulfide
The consumption of sulfate during sulfate reduction
can produce reactive HS- that can destroy EDB or
react with iron(III) minerals to produce FeS which
also reacts to degrade EDB. Lower concentrations
of sulfate would indicate more extensive sulfate
reduction and should be correlated with lower
concentrations of EDB. Figure 2.11 compares the
distribution of sulfate and EDB at the sites in the
EPA/ASTSWAMO survey.
In general, the relationship seems to be as expected;
the two highest concentrations of EDB were
associated with concentrations of sulfate below
41 mg/L, and the very high concentrations of
sulfate were associated with EDB concentrations
below 0.1 ug/L. However, for the sites where
the concentrations of both EDB and sulfate were
above the detection limit, the correlation coefficient
between the logarithm of EDB concentration
and the logarithm of sulfate concentration is low
(-0.15) and the sign of the correlation coefficient
is in the wrong direction (concentrations of EDB
go down as concentrations of sulfate go up).
The concentration of sulfate in the wells that are
contaminated with EDB can not be used to predict
concentrations of EDB.
1000
100
a 10
m
P 1
0.1
0.01
0.1
1000J ,
100
1 10 100 1000
Sulfate (mg/L)
1
m
a
HI
10
0.1
o.or
0
100 200 300 400
Sulfate (mg/L)
500 600
Figure 2.11. Association of concentrations of sul-
fate with concentrations of EDB in the
most contaminated monitoring wells at
sites.
In the survey conducted by EPA/ASTSWMO
Lead Scavengers Team, the comparison was made
between the concentrations of EDB and sulfate in
the contaminated ground water. A better estimate
of the extent of sulfate reduction might have been
the reduction in concentration of sulfate between
the contaminated ground water and ambient ground
water up gradient or side gradient of the release
of gasoline. To illustrate the extent of sulfate
reduction that might be expected at gasoline spill
sites, Figure 2.12 compares the distribution of
sulfate and sulfide in the sites sampled by Kolhatkar
et al. (2000). The maximum concentration of
sulfate in any of the wells at a site was considered
the background concentration. The median of
the background concentrations was 107 mg/L,
and 75% of sites had background concentrations
above 57 mg/L. At most of the sites, at least one
contaminated well had concentrations of sulfate
that were below the detection limit of 1 mg/L. In
every site, there was a substantial reduction in the
concentration of sulfate between the contaminated
wells and the background wells.
-------
1.0.
0.9
0.8
w 0.7
| 0.6.
•5 0.5.
I 0-3
£ 0.2.
0.1.
0 0
1.0,
0.9.
0.8.
« 0.7.
i 0.6.
CO
•5 0.5.
§ 0.4.
3 0.3.
£ 0.2.
0.1.
0.0
A „ f-f
/ '
_.«
/
/ • Sulfate Background
j ' = Sulfate in Source
, .' Area
M •"
10 100 1000
B J
n o /
,'
•*
/
•*
°° a I
° 1 Sulfide Measured
! o ,' Sulfide Expected
D ^
n D * **
0.1 1 10 100 1000
Concentration (mg/L)
Figure 2.12. Panel A. Distribution of sulfate in
the source area of the plume, and in
background ground water, at gasoline
spill sites in the Eastern United States.
Panel B. Distribution of sulfide in the
source area of the plumes, compared to
the concentration of sulfide expected
by the amount of sulfate remove. Data
are from the survey of Kolhatkar et al.,
(2000).
The concentration of sulfide expected from sulfate
reduction was calculated by subtracting the
lowest concentration of sulfate in any well from
the greatest concentration of sulfate in any well
at the site, then multiplying by the ratio of the
molecular weight of sulfide and sulfate. Sulfate
was generally depleted in the contaminated
wells in the sites (Figure 2.12, Panel A, and
significant concentrations of sulfide were expected
(Figure 2.12 Panel B).
The concentrations of sulfide actually measured
were much lower than expected (Figure 2.12
Panel B). In half the wells, sulfide was not present
above the detection limit of 0.1 mg/L. Only 10%
of sites had a sulfide concentration above 3 mg/L
and the maximum concentration of sulfide was
10 mg/L. The concentration of iron(II) was high in
ground water at these sites (Figure 2.9). It is likely
that the sulfide precipitated as iron(II) sulfides.
If these minerals persist as Acid Volatile Sulfide
(AVS), they may have a substantial contribution to
natural attenuation of EDB and DCA at gasoline
spill sites.
2.3.3.2.4 Methane
The source area of gasoline release sites usually
contains high concentrations of methane that is
produced by anaerobic metabolism of the BTEX
compounds in gasoline. After the soluble electron
acceptors such as oxygen, nitrate, and sulfate are
depleted, the metabolism of the BTEX compounds
shifts to a fermentation reaction that produces H2
and fatty acids such as acetate. Methane producing
bacteria require H2 or acetate as substrates to
produce methane. The H2 and acetate produced
during the fermentation of the BTEX compounds
create a geochemical environment where EDB or
1,2-DCA can be metabolized by methane producing
bacteria, or by Dehalococcoides strains, or by
both. Higher concentrations of methane should be
associated with lower concentrations of EDB.
Henderson et al. (2008) constructed microcosms
from the source area of a release of leaded gasoline
and from a mid gradient location. They noted
the greatest removal of EDB in microcosms with
the greatest production of methane. Anaerobic
biodegradation of EDB in the microcosms of
Henderson et al. (2008) produced significant
fractions of 2-bromoethanol instead of ethylene.
Bromoethanol was not identified as a product in
the pure culture studies of methanogenic bacteria
conducted by Belay and Daniels (1987), and has
not been reported as a product in mixed culture
studies of Dehalococcoides strains (Tandol et al.,
1994). However, Bouwer and McCarty (1985)
conducted 14C label studies of the fate of EDB in
a mixed methanogenic culture that was provided
with acetate as the primary substrate. A significant
fraction of label (59±6 %) was recovered in
material that had the properties of bromoethanol.
It is possible that the bromoethanol was produced
by methane producing bacteria because certain
aerobic organisms can metabolize EDB and
produce bromoethanol as an intermediate, using
an enzyme that does not require molecular oxygen
(Poelarends etal, 1999; Scholtz etal., 1987).
It is also possible that EDB was metabolized
to bromoethanol by other anaerobic organisms
that are yet to be characterized. In any case, the
activity of methane producing bacteria and strains
of Dehalococcoides can account for the observed
anaerobic biodegradation of EDB and 1,2-DCA.
-------
Figure 2.13 compares the association of methane
and EDB in sites in the EPA/ASTSWMO study.
Methane was present above its detection limit
of 0.001 mg/1 at fourteen of sixteen sites. Five
sites met the criterion of greater than 0.5 mg/L
of methane. In general there were lower
concentrations of methane in the EPA/ASTSWMO
study than was found in the study of Kolhatkar
et al. (2000). This may reflect a depletion of
the supply of BTEX compounds to produce the
H2 and acetate needed as substrates for methane
production.
1000 ,
100 :
g 10 •
m
S 1.
0.1 •
0.01
1000 ,
100;
0.0001 0.001 0.01 0.1
Methane (mg/L)
10
10;
I
m
Q
LU
1,
0.01
0.5
1 1.5
Methane (mg/L)
2.5
Figure 2.13. Association of concentrations of meth-
ane with concentrations of EDB in the
most contaminated monitoring wells at
sites.
On first examination, there appears to be a
relationship between the concentrations of methane
and the concentrations of EDB in the wells in the
EPA/ASTSWMO study. The two sites with the
highest concentrations of EDB (711 and 147 ug/L)
had concentrations of methane less than 0.01 mg/L,
and the five sites with concentrations of methane
greater than 0.5 mg/L had EDB concentrations
of 1 ug/L or less. However, there are only eight
sites where the concentrations of both EDB and
methane were above the detection limit. For these
eight sites, the correlation coefficient between
the logarithm of the EDB concentration and the
logarithm of the methane concentration is -0.13.
The sign of the correlation coefficient is in the
right direction (concentrations of EDB go down
as concentrations of methane go up) but the
association is too weak to have any predictive value
at other sites.
2.3.3.2.5 pH
The pH of ground water has a strong effect on
the two processes that can destroy EDB and
1,2-DCA in ground water. The rate of abiotic
transformation of TCE by iron(II) sulfide increases
as the pH increases (Butler and Hayes, 2001).
Transformation of EDB and 1,2-DCA by iron(II)
sulfide should proceed by the same mechanism and
pH should have the same effect on the rate of their
transformation. The rates at pH of 5 or less should
be one hundred fold lower than rates at pH 7.
Similarly, the primary organism that is responsible
for biological reductive dechlorination of EDB and
1,2-DCA is sensitive to acid conditions (Maymo'-
Gatell et al., 1997). The EPA/ASTSWMO study
compared concentrations of EDB to pH in the most
contaminated well at eleven sites (Figure 2.14).
The concentration of EDB was high at one site that
was strongly acid (pH of 4), and the concentration
of EDB was lower in wells with pH near 7.
However, there were also several sites where the
concentration of EDB was high and the pH was
near 7; indicating that pH by itself cannot be used
to predict the concentration of EDB in old leaded
motor fuel release sites.
1000
i
100
g 10
m
I 1
0.1
0.01
1
4 5
pH
6
8
Figure 2.14. Association of pH with concentra-
tions of EDB in the most contaminated
monitoring wells at sites.
-------
2.4 Applications of Compound Specific
Isotope Analysis (CSIA) to
Document Biodegradation and/or
Abiotic Transformation of EDB and
1,2-DCA
This sub-section reviews the application of
Compound Specific Isotope Analysis (CSIA)
to evaluate the biodegradation and abiotic
transformation of EDB and 1,2-DCA at motor
fuel release sites. This sub-section is intended for
technical staff that will interpret CSIA analyses to
provide an estimate of the extent of biodegradation
and/or abiotic transformation of EDB or 1,2-DCA
in ground water at gasoline release sites.
Frequently, mathematical models of transport and
fate are used to provide an exposure assessment
as part of the risk management process. Because
EDB and 1,2-DCA do not sorb strongly to aquifer
sediments, the transport and fate models are very
sensitive to the rate of biodegradation and the rate
of abiotic transformation of the contaminants. It
is difficult to extract unequivocal estimates of the
rate of biodegradation or abiotic transformation
from conventional field monitoring data, and as a
consequence, there is a large amount of uncertainty
in the calibration of the transport and fate models.
In many circumstances, a determination of
the isotopic ratio of carbon can unequivocally
document the extent of biodegradation or
abiotic transformation of EDB or 1,2-DCA that
has occurred as the contaminant moves along
a flow path in ground water. By providing an
independent estimate of the extent of degradation
or transformation, CSIA analysis can facilitate the
calibration of transport and fate models by reducing
the uncertainty in the estimates of exposure that are
obtained from the models.
2.4.1 Theoretical Background for Using
CSIA to Estimate the Extent of
Biodegradation
There are two naturally occurring stable isotopes
of carbon: (1) Carbon-12, which has a weight of
twelve Daltons, and (2) Carbon-13, which has a
weight of thirteen Daltons. Carbon-12 (12C) is
approximately one hundred times more abundant
than Carbon-13 (13C). During biodegradation or
abiotic transformation, EDB or 1,2-DCA molecules
with two 12C atoms are degraded more rapidly than
molecules with one 13C atom and one 12C atom,
which leads to an increase in the ratio of 13C to 12C
in the residual EDB or 1,2-DCA. The change in the
ratio of 13C to 12C is described as enrichment of the
carbon isotopes during biodegradation or abiotic
transformation. In recent years, enrichment of
stable carbon isotopes has been used to recognize
and describe the natural biodegradation of MTBE
in ground water (Hunkeler et al., 2001; Gray et al.,
2002; Kolhatkar et al., 2002; Kuder et al., 2005a;
Somsamak et al., 2006; Wilson et al., 2005; Zwank
et al., 2005), and benzene in ground water (Mancini
et al., 2003; Fischer et al., 2007).
The ratio of isotopes is determined with an isotope
ratio mass spectrometer. The mass spectrometer
does not measure the ratio of the stable carbon
isotopes to each other. Rather, it measures the
deviation of the ratio in the sample from the ratio
of a standard used to calibrate the instrument.
The conventional notation for the ratio of 13C to
12C in a sample (513C) reports the ratio in terms
of its deviation from the ratio in the standard, as
described in Equation 2.6 (Meckenstock et al.,
2004; Schmidt et al., 2004).
613C =
(13c/12c) -(13c/12c)
V ' /sample \ ' 4
standard
(13c/12c)
V /standard
xlOOO
(2.6)
The units for 513C are parts per thousand, often
represented as %o, or per mil, or per mill. The usual
pronunciation of 513C is "delta thirteen sea." The
substance used as the international standard for
stable carbon isotopes has a ratio of 13C to 12C of
0.0112372. During the course of biodegradation or
abiotic transformation, the compound that is still
remaining will have more of the heavy isotope 13C,
and the value of 513C will become more positive.
The extent of isotopic enrichment is typically
characterized by calculating a linear regression of
the 513C in the compound being degraded versus the
natural logarithm of the fraction of the compound
remaining (C/Co) after biodegradation or
transformation (Meckenstock et al., 2004; Schmidt
et al., 2004). The slope of the regression line is
termed the isotopic enrichment factor (e). When
513C is expressed in units of %o, the unit of 8 is %o.
When an appropriate value of e is available, and
when an estimate of the initial value of 513C in the
compound as originally released in the gasoline is
available, the fraction (F) of material remaining
in ground water after biodegradation or abiotic
transformation can be calculated from the value of
513C determined for the compound in ground water
(Meckenstock et al., 2004; Schmidt et al., 2004;
Wilson et al., 2005).
F = C/Co =
£13,.
-field ~
S13^
(2.7)
-------
The fraction of material degraded is simply one
minus F.
The value of 513C in EDB or 1,2-DCA in ground
water is measured directly in a water sample from
monitoring wells at the release site (513Cfield).
Equation 2.7 also requires the isotopic enrichment
factor (e) and the initial value of 513C (i.e., at
t = 0). The following material will review available
literature on the isotopic enrichment factor for
biodegradation and abiotic transformation of EDB
and 1,2-DCA, followed by a discussion of the
most plausible values for the initial 513C of EDB or
1,2-DCA originally released in leaded gasoline.
2.4.2 Isotopic Enrichment during
Biodegradation of EDB and
1,2-DCA
At gasoline release sites, the oxygen demand
associated with petroleum hydrocarbons
generally consumes all the available dissolved
oxygen in ground water. As a consequence,
biological degradation of EDB or 1,2-DCA in
close proximity to a spill of gasoline must occur
through an anaerobic process. At this writing
(spring 2008), there is only one value of 8 for
anaerobic biodegradation of EDB available in the
literature (Henderson et al., 2008); the primary
data are presented in Figure 2.15. The enrichment
factor was extracted from microcosms that were
constructed with sediment from a UST release site.
The enrichment factor, as determined by the slope
of the regression line, has a value of-5.7 ± 1%0.
-15
-16
-17 i
P -19 i
-20 i
-21 '•
-22
e = -5.7%o
0 -0.2 -0.4 -0.6 -0.8 -1 -1.2
Natural Logarithm of Fraction Remaining [ln(C/Co)]
value of-32.1 ± 1.1%0 and Hirschorn et al. (2007)
reported a value of-25.8 ±3.5 %o. These values
of e for biodegradation of 1,2-DCA are much more
negative (show stronger enrichment) than is the
case for EDB (e = -5.7 ± 1%0).
There are several ways to account for the stronger
enrichment of 1,2-DCA compared to EDB.
Table 2.4 compares isotopic enrichment factors
for biodegradation of a variety of halogenated
compounds. In general, the value of e becomes less
negative as the molecular weight increases. The
value of e for anaerobic biodegradation of EDB is
near the values for biological degradation of PCE
and TCE, compounds that have a molecular weight
similar to EDB.
This relationship between molecular weight and the
enrichment factor has been explained by Nijenhuis
et al. (2005) as the effect of a mass transport
limitation of the compound across the cell wall of
the bacteria that degrade the contaminant. They
showed that the enrichment factor for degradation
of PCE was most negative for the pure PCE
reductive dehalogenase enzyme, was less negative
for cell extracts, and was even less negative for
the living bacteria. The compounds with lower
molecular weight are more soluble in water, and
enter the bacterial cells less readily.
Part of the difference in the value of the
enrichment factor for anaerobic biodegradation
of EDB compared to 1,2-DCA can be attributed
to differences in the mechanisms of degradation.
Figure 2.16 compares three pathways that are
available for anaerobic biodegradation of EDB
or 1,2-DCA. Biodegradation of EDB in the
microcosm study of Henderson et al. (2008)
was primarily through a mechanism of reductive
dehalogenation, also called hydrogenolysis, with
a minor amount of hydrolytic debromination. In
contrast, in the study of Hunkeler et al. (2002)
biodegradation of 1,2-DCA was through a process
of dichloroelimination. Eisner etal. (2005) showed
that a dichloroelimination reaction should produce
more negative values for the enrichment factor than
a hydrogenolysis reaction.
Figure 2.15. Enrichment of the heavy isotope of
carbon in EDB during anaerobic bio-
degradation of EDB in a microcosm
study. (Redrawn from Henderson et al.
(2008).)
There are only two enrichment factors available
in the literature for anaerobic biodegradation
of 1,2-DCA. Hunkeler et al. (2002) reported a
-------
Table 2.4. Comparison of enrichment factors for EDB and DCAto the range of isotopic enrichment factors
(e) for carbon isotopes during reductive dehalogenation of halogenated organic compounds.
Halogenated
hydrocarbon
EDB
EDB
PCE
PCE
TCE
TCE
TCE
1,1,2-TCA
cis-DCE
cis-DCE
cis-DCE
trans-DCE
1,1 -DCE
1,2-DCA
1,2-DCA
VC
vc
VC
Molecular
Weight
(Daltons)
187.9
187.9
165.8
165.8
131.4
131.4
131.4
133.4
96.9
96.9
96.9
96.9
96.9
98.9
98.9
62.5
62.5
62.5
Mechanism
mostly biological
hydrogenolysis, some
hydrolytic debromination
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
biological
dichloroelimination
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
biological
dichloroelimination
biological hydrolytic
dechlorination
biological hydrogenolysis
biological hydrogenolysis
biological hydrogenolysis
13C/12C 8 %o
-5.7
-20.2
-5.2
-2
-6.6
-2.5
-7.1
-13.8
-2.0
-14.1
-16.1
-19.9
-20.4
-30.3
-7.3
-32.1
-25.8
-21.5
-26.6
-22.4
-31.1
Reference
Henderson et al, 2008
This Report
Slater etal., 2001
Hunkeler et al., 1999
Bloom et al., 2000
Sherwood Lollar etal., 1999
Slater etal., 2001
Hunkeler et al., 2002
Bloom et al., 2000
Hunkeler et al., 2002
Slater etal., 2001
Hunkeler et al., 2002
Hunkeler et al., 2002
Hunkeler et al., 2002
Hirschorn et al., 2007
Bloom et al., 2000
Slater etal., 2001
Hunkeler et al., 2002
-------
EDB
Br Br
H-C-C-H
H H
Hydrolytic ' Hvdroqenolvsis Dibromoelimination
Debromination a|SO caned also called
Reductive
Dehalogenation
Dihaloelimination
2-Bromoethanol
Br OH
H-C-C-H
H H
Bromoethane
Br H
H-C-C-H
H H
Ethylene
1,2-Dichloroethane
ci ci
H-C-C-H
H H
Hydrolytic ' Dichlorelimination
Dechlorination Hydrogenolysis also called
also called Dihaloelimination
Reductive
Dechlorination
i
2-Chloroethanol Chloroethane
Ethylene
CI OH
H-C-C-H
H H
CI H
H-C-C-H
H H
Figure 2.16. Biological or Abiotic Transformations
of EDB and 1,2-DCA.
Hirschorn et al. (2007) compared the enrichment
factor for biodegradation under aerobic conditions
to the enrichment factor under nitrate reducing
conditions. The enrichment factor under aerobic
conditions was strongly negative (-25.8 ± 0.4%o).
The values of the enrichment factor under aerobic
and nitrate reducing conditions were essentially
identical. Hirschorn et al. (2007) used the
correspondence in enrichment factors to argue that
anaerobic biodegradation of 1,2-DCA under nitrate
reducing conditions proceeded through the same
hydrolytic dechlorination reaction that was used
under aerobic conditions.
A third possible explanation for the difference
in the enrichment factors during anaerobic
biodegradation of EDB and 1,2-DCA would relate
to the differences between the C-C1 and C-Br
bond. Miller et al. (2001) compared the enrichment
factors produced by aerobic biodegradation of
methyl chloride, methyl bromide, and methyl
iodide. The enrichment factors were essentially the
same; the nature of the halogen in the compound
does not seem to matter.
2.4.3 Isotopic Enrichment During
Abiotic Transformation of EDB and
1,2-DCA
As of this writing (spring 2008) there is no
literature available on the isotopic enrichment of
EDB during abiotic transformation. As described
in Section 2.3.2, a laboratory study was conducted
on the transformation of EDB in the presence of
biogenic FeS. The biogenic FeS was formed in a
laboratory column that simulated a passive reactive
barrier constructed with plant mulch and sand. The
plant mulch supported sulfate reduction to produce
sulfide, and the sulfide reacted with hematite and
other iron minerals in the column to produce FeS
in the column. To determine the kinetics of EDB
transformation by the biogenic FeS and carbon
isotope enrichment during transformation, the
column was sampled and FeS and sand were
separated from the plant mulch under an oxygen
free atmosphere and then incubated with a dose
solution that contained approximately 50 mg/L
EDB. Incubations were conducted in five separate
preparations taken from two locations in the
original column. The 513C of EDB in the porewater
was determined after 1 hour of incubation and after
7, 15, 22, and 29 days of incubation.
The relationship between the extent of removal
of EDB and the 513C of EDB is presented in
Figure 2.17. The isotopic enrichment factor,
as determined by a linear regression of 513C of
EDB on the natural logarithm of the fraction of
EDB remaining (where the data from all five
incubations were pooled) was -20.2 ± 2.23%o at
95% confidence.
o Abiotic transformation
p Biological degradation
0.3 0.1 0.06 0.03
Fraction Remaining (C/Co)
0.01
Figure 2.17. A comparison of the enrichment of the
heavy carbon isotope in EDB during
anaerobic biodegradation of EDB in
a microcosm study (Henderson et al.,
2008) against enrichment during abi-
otic transformation of EDB by FeS.
-------
The enrichment of carbon isotopes during
abiotic degradation of EDB during reaction
with FeS (-20.2 %o) is significantly greater than
the enrichment during biodegradation (-5.7%o).
This relationship has been observed for other
halogenated organic compounds. Liang et al.,
(2007) compared the enrichment of stable carbon
isotopes during biological and nonbiological
degradation of PCE and TCE. Enrichment of PCE
and TCE during transformation by FeS at pH 7
was -30.2 ± 4.3%0 and -33.4 ± 1.5%o respectively.
Enrichment during biodegradation varied with the
culture, ranging from -1.39%o to -7.12%o for PCE
and from -4.07%o to -15.27%o for TCE. In each
case, the enrichment during abiotic transformation
was substantially greater than the enrichment
during biodegradation.
The contrast in behavior of stable carbon isotopes
in EDB during biodegradation and abiotic
transformation parallels the behavior of PCE and
TCE. If the reaction EDB with FeS followed
the dihaloelimination pathway (Figure 2.16)
the enrichment factor should be close to the
enrichment factor observed by Liang et al. (2007)
for reaction of TCE and PCE with FeS. According
to Equation 2.7, as the value 513C of EDB in ground
water becomes less negative, the predicted value
of the fraction remaining (C/Co) is smaller, and the
predicted extent of biodegradation is greater. For
a given change in the value of 513C, the predicted
extent of biodegradation becomes greater as
the value of the enrichment factor (e) becomes
less negative. Liang et al. (2007) warn that the
approach used in Equation 2.7 will over predict the
extent of biodegradation of PCE or TCE if a value
of e that is appropriate for biodegradation is used
to interpret enrichment in ground water samples
where abiotic processes are primarily responsible
for the destruction of PCE or TCE.
The same warning applies to EDB. If a range of
values of e are available for the same process, or
if several processes can operate simultaneously
and they vary widely in the value of 8, a more
conservative estimate of the extent of EDB
destruction is provided by choosing the most
negative value of the isotopic enrichment factor.
2.4.4 The Initial Value of 513C of EDB or
1,2-DCA Originally Released in
Leaded Motor Fuel
To use equation 2.7 to estimate the extent of
degradation of EDB from the 513C of EDB in
ground water, it is necessary to have an initial
value for 513C of EDB in the leaded motor fuel
that was spilled. The analytical techniques for
CSIA were developed in the late 1990s, while
EDB and 1,2-DCA were banned in conventional
motor gasoline after 1980. As a result, there is no
direct information available on the 513C of EDB
or 1,2-DCA originally released in leaded motor
fuel. Five samples of modern leaded motor fuels
were analyzed for 513C of EDB. Four samples
of aviation gasoline (Avgas LL 100 octane)
returned values of-29.6%o, -30.1%o, -30.2%o, and
-30.7%o. One sample of 110 octane automobile
racing fuel returned a value of -30.2%o. In the
absence of additional information, a value of -30%o
is recommended as the best estimate currently
available of the original value of 513C in EDB in
leaded motor fuels.
Similar processes are used to manufacture
both EDB and 1,2-DCA and PCE, TCE, and
1,1,1 -TCA, and they often were produced at the
same manufacturing plants. Table 2.5 summarizes
published values for 513C of PCE, TCE, and
1,1,1 -TCA produced by a number of manufactures
in the past. The most negative value reported was
-37.2%o, the least negative value reported was
-23.2%o.
If the assumed initial value of 513C in EDB is
more negative than the true value, the extent
of degradation predicted by Equation 2.7 will
overestimate the true extent of degradation. To
provide a conservative lower boundary on the
fraction of EDB remaining after degradation, and
a conservative upper boundary on the fraction
degraded, it is most appropriate to use the least
negative value within the plausible range of values
of 513C that would be expected in leaded motor
fuel. In the absence of other direct information,
a value of -23.2%o will be taken as conservative
upper boundary on the value of 513C in EDB or
1,2-DCA in leaded gasoline.
2.4.5 Measured Concentrations of EDB
and 513C of EDB in Ground Water
If measured concentrations of EDB in ground water
are related to the destruction of EDB in ground
water, there should be a relationship between the
concentration of EDB and the 513C of EDB. To
search for a relationship, water samples from
the most contaminated wells at selected sites in
the EPA/ASTSWMO study were analyzed for
concentrations of EDB and for 513C of EDB. The
population of candidate sites was limited by two
factors: (1) access to the sites so that they could be
resampled, and (2) the relatively high concentration
of EDB required for determination of 513C in EDB
in the sample. A concentration of at least 4 ug/L
of EDB is required to attain the quality objective
-------
Table 2.5. Range of 513C in samples of commercial chlorinated solvents.
Compound
PCE
PCE
PCE
PCE
PCE
PCE
PCE
PCE
TCE
TCE
TCE
TCE
TCE
TCE
TCE
TCE
1,1,1-TCA
1,1,1-TCA
1,1,1-TCA
1,1,1-TCA
1,1,1-TCA
Source
Manufacturer A
Manufacturer B
Manufacturer C
Dow
ICI
PPG
Vulcan
Range
Manufacturer A
Manufacturer B
Manufacturer C
Aldrich
Dow
ICI
PPG
Range
Manufacturer A
Dow
ICI
PPG
Vulcan
Range
613C (%o/PDB)
-27.12±0.03
-35.27±0.12
-24.06±0.08
-23.19±0.10
-37.20±0.03
-33.84±0.03
-24.1±0.04
-23. 19 to -37.20
-31.53±0.01
-27.90±0.08
-29.93±0.18
-33.49±0.08
-31.90±0.05
-31.32±0.03
-27.80±0.01
-27.80 to -33.49
-31.64±0.09
-29.42±0.06
-26.64±0.09
-25.80±0.46
-28.42±0.07
-25. 80 to -31.64
Reference
Jendrzejewski et al. (2001)
Jendrzejewski et al. (2001)
Jendrzejewski et al. (2001)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
Jendrzejewski et al. (2001)
Jendrzejewski et al. (2001)
Jendrzejewski et al. (2001)
Jendrzejewski et al. (2001)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
Jendrzejewski et al. (2001)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
van Warmerdam et al. (1995)
Summarized from Wilson (2008).
-------
of a sample standard deviation of 0.5%o or less.
Five sites met both criteria, one each in Montana,
Minnesota, Virginia, and two sites in New
Hampshire. As was expected, the two sites with
lower concentrations of EDB had higher values for
513C of EDB (Figure 2.18 and Table 2.6).
140n
120^
100
^80-
£.60-
20-
0 •
-20-
-40
Expected from Abiotic Transformation
Maximum concentration of
EDB expected, no enrichment
Expected from Biodegradation
10000 1000 100 10
Concentration EDB (ng/L)
Figure 2.18. Relationship between the concentra-
tions of EDB in the most contaminated
wells at each of five sites and the 513C
of EDB. (Note that the concentration
on x-axis decreases from left to right).
Falta (2004) used the expected concentration
of EDB in leaded gasoline, and Raoult's Law
to estimate the concentration of EDB in ground
water that would be in equilibrium with typical
leaded gasoline. The predicted concentration
is 1,900 (ig/L. Equation 2.7 was rearranged to
produce Equation 2.8, and Equation 2.8 was in
turn used to calculate the value of 513C of EDB that
would correspond to various concentrations of EDB
(Table 2.6).
(2.8)
Where 513Cfield is the expected value of 513C of
EDB in ground water in units of %o, Conc.field is
the expected concentration of EDB in ground water
in units of (ig/L, 513Ct=0 is the initial value of 513C
of EDB in the gasoline (assumed to be -30%o)
and e is the appropriate value for the isotopic
enrichment factors in units of %o. Equation 2.8
assumes that destruction of EDB is the only process
that attenuates the concentration of EDB. There
is no allowance for attenuation due to dilution in
the monitoring wells. If there were dilution of
the plume in the monitoring well, the expected
concentration of EDB would shift to a lower value.
In Figure 2.18, a value of -5.7%o was used
for e to predict the relationship dominated by
biodegradation and a value of -20.2%o was used
to predict the relationship dominated by abiotic
transformation. The data best fit the projections
of fractionation that would be expected from
biodegradation of EDB.
In general, the few data in Figure 2.18 are
consistent with the assumption that reductions in
concentration of EDB are caused by processes that
Table 2.6. Relationship between the concentrations of EDB in the most contaminated wells at five motor
fuel release sites, enrichment of stable carbon isotopes in EDB, and a conservative estimate of the fraction
of EDB destroyed.
Location
Station A, New Hampshire
Virginia
Station B, New Hampshire
Montana
Minnesota
Concentration EDB
Mg/L
711
557
147
8.0
6.8
513C
%0
-22.8
-13.9
-18.9
6.1
3.9
Fraction
Remaining*
0.70
0.45
0.58
0.17
0.19
Fraction
Destroyed
0.30
0.55
0.42
0.83
0.81
* Fraction remaining calculated from Equation 2.7 assuming
, = -30.%o and e = -20.2%o
-------
destroy EDB, and can be recognized by analysis
of 513C in EDB. This suggests that CSIA will be a
useful tool to recognize biodegradation or abiotic
transformation of EDB in residual gasoline in the
source area of plumes.
2.4.6 Predictions of the Extent of
Degradation or Transformation
At present there is no straightforward approach
to attribute one portion of EDB destruction to
biodegradation and the remaining portion to abiotic
transformations. As a consequence, the most
conservative approach is to attribute all of the
destruction of EDB to abiotic transformation.
Figure 2.19 presents a prediction of the extent of
destruction of EDB based on various measured
values of 513C in EDB in ground water, and the
assumption that abiotic processes are entirely
responsible for destruction of EDB. Figure
2.9 presents predictions of the fraction of EDB
remaining that correspond to the best estimate of
the initial value of 513C in EDB in the leaded motor
fuel that was originally spilled (-30%o) and to the
most conservative estimate of the initial value
of 513C in EDB in the leaded motor fuel that was
originally spilled (-23.2%o). As an example, at the
best estimate of the initial value of 513C in EDB
in gasoline, a measured value of 513C in EDB in
ground water of +24%o would predict that 7% of
the original quantity of EDB remained. The most
conservative estimate of the initial 513C in EDB
in ground water predicts that 10% of the original
quantity of EDB remains.
Figure 2.19 was used to provide a conservative
boundary on the extent of destruction of EDB at
the five release sites listed in Table 2.6. Even with
a small data set, CSIA could distinguish sites with
minor amounts of EDB destruction from sites
where a major fraction of EDB had been destroyed.
If the data are available, a better approach to
evaluate EDB destruction at a particular site would
use the most negative value of 513C in EDB in any
monitoring well at a site as the local best estimate
of 513C in EDB in gasoline, then use Equation 2.6
to predict the extent of degradation in the other
wells. This approach is particularly recommended
if data are available from wells along a flow path in
the aquifer that are down gradient of the area with
known residual gasoline. Figure 2.19 is offered
to provide a quick and conservative boundary on
the extent of EDB destruction when the available
data are few, or when the hydraulic relationships of
wells are not readily apparent.
0.01
Isotopic Enrichment Factor e = -20.2 %o
O)
c
'c
'ns
0.10
o
13
m
D
m
1.00
-40 -30 -20 -10 0 10 20 30 40 50 60
6"C (%«)
Figure 2.19. A conservative estimate of the rela-
tionship between the measured value
of 513C of EDB in ground water at a
Leaded Motor Fuel Release Site, and
the extent of destruction of EDB by
abiotic and biological processes. The
heavy solid line projects destruction
based on the best available estimate
of the initial value of 513C of EDB in
leaded gasoline. The light solid line
projects destruction based on the most
conservative estimate for the initial
value of 513C of EDB in leaded gaso-
line. The diamonds represent estimates
of the fraction of EDB remaining in
monitoring wells in the spill sites
described in Table 2.6. See text for
further discussion.
-------
3.0
Distribution of EDB and 1,2-DCA at Motor
Fuel Release Sites, and the Associated
Chance of Contaminating Ground Water
This section describes the existing concentrations
of EDB, benzene, and 1,2-DCA in monitoring
wells at gasoline service stations. It is intended
for decision makers and other staff of regulatory
agencies that must apportion resources for
monitoring and risk management of the hazard
associated with EDB, benzene, and 1,2-DCA in
ground water used for drinking water. It is also
useful to case workers in regulatory agencies as a
basis for comparison of the distribution of EDB and
1,2-DCA at their sites to concentrations that were
determined at other sites.
3.1 Existing Distribution of EDB,
1,2-DCA, and Benzene in Ground
Water at Selected Motor Fuel
Release Sites
Figure 3.1 provides a map showing the states
that participated in the study. Participation in the
EPA/ASTSWMO study was entirely voluntary on
the part of the state agencies. The coverage of the
USA was reasonably representative; however, the
states on the Pacific Coast chose not to participate.
The number of stations that were sampled in each
state was variable. Two of the states provided
samples from only one station each, and two of
the states provided samples from ten stations each
(Table 1.1).
Figure 3.1. Distribution of states participating in the EPA/ASTSWMO Study.
-------
Because the sampling locations were not selected
randomly, the samples can not be taken to represent
the population of gasoline service station sites
in the USA. At best, the samples represent a
population of service station sites in the USA where
the relevant case worker in the state agency was
interested in the possibility that EDB was present
in ground water at the site. Nevertheless, there is
a close correspondence between the distribution of
EDB concentrations in South Carolina, and in the
sites that were sampled in the EPA/ASTSWMO
study (Figure 1.2). The total fraction of sites
that exceeded the MCL was very similar, and
the distribution of sites with concentrations
above 20 ug/L was almost identical. Table 3.1
summarizes the data in Figure 1.2.
As discussed earlier, not all the EDB used in the US
was used in motor fuel. Approximately 10% was
used in agriculture. Background wells were not
sampled at all of the sites that participated in the
EPA/ASTSWMO study. As a consequence, it is not
possible to exclude the possibility that a portion of
the EDB detected in wells at the gasoline service
stations resulted from agricultural application.
3.2 Relative Distribution of EDB,
1,2-DCA, and Benzene at Selected
Motor Fuel Release Sites
There were 39 sites in the EPA/ASTSWMO study
with detectable concentrations of both EDB and
benzene. In the 39 sites, there were seven sites
from Colorado, two sites from Maryland, one site
from Montana, three sites from North Carolina,
four sites from North Dakota, two sites from New
Hampshire, two sites from Oklahoma, five sites
from South Carolina, one site from South Dakota,
three sites from Tennessee, two sites from Utah,
and seven sites from Wisconsin.
Data are summarized in Table 3.2. As was
expected, benzene was detected above the MCL
at every site that was sampled. In contrast, EDB
was detected in 53% of sites, but detected above
the MCL at only 42% of sites, and 1,2-DCA was
detected at 23% of sites, but detected above the
MCL at only 15% of sites. Both EDB and 1,2-DCA
were present at concentrations above the MCL at a
significant number of sites.
Table 3.1. The distribution of EDB in the sites included in the EPA/ASTSWMO State Study compared to
distribution in South Carolina.
Maximum Concentration at Site
Above 100 ug/L
Above 50 ug/L
Above 1 ug/L
South Carolina
(Falta data)
Percent of Sites
7%
11%
35%
EPA/ASTSWMO
State Study
6%
10%
25%
Table 3.2. Occurrence of EDB, benzene, and DCA in sites sampled during the EPA/ASTSWMO State Study.
Method 80 11 for EDB
EDB detected
EDB above MCL
Method 8260 for Benzene
Benzene detected
Benzene above MCL
Method 8260 for DCA
DCA detected
DCA above MCL
States
19
19 (100%)
17 (89%)
12
12 (100%)
12 (100%)
12
6 (50%)
5 (42%)
Sites
102
54 (53%)
43 (42%)
39
39 (100%)
39 (100%)
39
10 (26%)
6 (15%)
Samples
802
151 (19%)
103 (13%)
335
118(40%)
105 (36%)
293
18 (6%)
8 (3%)
-------
Figure 3.2 plots the frequency distribution of the
concentrations of benzene, 1,2-DCA, and EDB
independent of each other at a particular site. For
comparison, the figure also plots the concentration
that would be predicted for benzene, 1,2-DCA,
and EDB in ground water that was in equilibrium
with un-weathered leaded gasoline. The predicted
values for concentrations were taken from Table 2
of Falta (2004). He calculated the predicted values
based on the average concentrations of benzene,
1,2-DCA, and EDB in leaded gasoline, and on the
gasoline to water partition coefficient in the case
of EDB and benzene, or based on Raoult's Law for
1,2-DCA.
100000
10000
> 1000
100
10
§ 1
o 0.1
0.01
0.001
°AAAAAAA.
"AAAA
0 20 40 60 80 100
Percent of Sites Exceeding Concentration
Figure 3.2. Distribution of the maximum concen-
trations of EDB, 1,2-DCA, and benzene
in ground water from monitoring wells
at motor fuel release sites in the EPA/
ASTSWMO study, compared to the
concentrations predicted for ground
water in contact with unweathered
leaded gasoline.
The predicted concentration of benzene in ground
water that is equilibrated with leaded gasoline is
ten fold higher than the predicted concentration
of 1,2-DCA, and the predicted concentration of
1,2-DCA is two fold higher than the predicted
concentration of EDB. The actual concentrations
of benzene, 1,2-DCA, and EDB are less than
predicted. There are a number of possible
explanations. Weathering of the residual gasoline
over time may have reduced the concentrations
of benzene, 1,2-DCA, and EDB in the residual
gasoline. Ground water that was exposed to the
residual gasoline in the aquifer and contaminated
with benzene, 1,2-DCA, and EDB may have been
diluted with uncontaminated ground water in
the monitoring well. There may have been mass
transfer limitations on dissolution of benzene,
1,2-DCA, and EDB from the residual gasoline into
flowing ground water. In any case, the ground
water samples that were most contaminated with
EDB or benzene were within a factor of two of the
predicted concentration. The highest measured
concentration of 1,2-DCA was more than 60 fold
lower than the predicted concentration.
With the exception of two lowest concentrations
of benzene in the data set, the concentrations of
benzene, EDB, and 1,2-DCA all follow a log-linear
frequency distribution. There were ten sites that
reported concentrations of 1,2-DCA above the
method detection limit. As a basis of comparison,
the geometric mean concentration was calculated
for the ten sites with the highest concentrations
of benzene, of EDB, and of 1,2-DCA. On
average the concentration of benzene was much
higher than EDB or 1,2-DCA. The geometric
mean concentrations for benzene, EDB, and
1,2-DCA were 7,300 ug/L, 68 ug/L, and 5.0 ug/L
respectively.
The MCL for EDB (0.05 ug/L) is one hundred fold
lower than the MCL for benzene and 1,2-DCA
(5 ug/L). As a result, it can be confusing and
misleading to evaluate the chance for ground water
contamination by comparing the concentrations
in ground water. A better comparison that is
relevant to beneficial use of an aquifer as a ground
water supply is the chance that a source of EDB
or benzene will contaminate a water supply well
at any concentration above the MCL, making
it useless for drinking water unless the water is
treated.
It is possible to estimate a relative chance
that a water supply well will be impacted at
concentrations above the MCL if the following
three assumptions are accepted. First, the chance
that a plume of contamination will impact a water
well is proportional to the surface area of the
plume. Second, the surface area of a plume is
proportional to the square of its length. Third, the
length of the plume is proportional to the logarithm
of the maximum concentration of the contaminant.
Studies of the distribution of contaminants in
ground water indicate that concentrations attenuate
with distance following a pseudo first order rate
law. As a consequence, the length of the plume
is proportional to the logarithm of the ratio of the
maximum concentration to the MCL. If the MCL
is 0.05 ug/L and the maximum concentration is
0.5 ug/L, the length would be some distance X. If
the maximum concentration is 5 ug/L, the length
would be 2X, if the maximum concentration is
50 ug/L, the length would be 3X, if the maximum
-------
concentration is 500 ug/L, the length would be 4X
and so on.
Ignoring the particular hydrogeological context and
cultural context of a release, the relative chance that
a ground water supply well will be contaminated
at concentrations above the MCL is described by
Equation 3.1.
Relative Chance = (Ln[MaximumConc./MCL])2
(3.1)
Figure 3.3 plots the frequency distribution of the
relative chance that EDB, benzene, and 1,2-DCA
from fuel spill sites in the EPA/ASTSWMO Study
will contaminate a water supply well. The data are
ranked independently of each other at a particular
site. For comparison, the figure also plots the
relative chance that would be predicted for EDB,
benzene, and 1,2-DCA in ground water that was
in equilibrium with un-weathered leaded gasoline.
The predicted chance that EDB in ground water
that is equilibrated with leaded gasoline will exceed
the MCL is greater than the predicted chance for
benzene, which in turn is greater than the predicted
chance for 1,2-DCA. The greatest chance to
exceed the MCL for a particular site and particular
contaminant was associated with EDB.
O 120,
|=100J
1180
o 1" 60 J
5 | 40 |
g> - 20 ,
13
I (
« EDB predicted
« EDB measured
A Benzene predicted
A Benzene measured
• DCA predicted
n DCA measured
20 40 60 80 100
Percent of Sites Exceeding Relative Chance
Figure 3.3. Distribution of the relative chance that
EDB, 1,2-DCA, and benzene in ground
water at motor fuel release sites in the
EPA/ASTSWMO Study will impact
a water supply well at concentrations
above the MCL, compared to the rela-
tive chance predicted for ground water
in contact with unweathered leaded
gasoline. The relative chance is propor-
tional to the surface area of the plume
of EDB or benzene, or 1,2-DCA with
concentrations above the MCL. See
text for details.
Notice from Figure 3.3 that for approximately
20% of the sites in the EPA/ASTSWMO study, the
relative chance that EDB would impact a water
supply well was roughly equivalent to the chance
that benzene would impact a water supply well.
However, for the remaining 80% of sites, the chance
of EDB impacting a water supply well was much
less than the chance that benzene would impact a
well. Also notice that in the worst case, the chance
that 1,2-DCA would impact a water supply well was
roughly 10% of the worst case chance for EDB or
benzene, and that the relative chance for 1,2-DCA
fell off rapidly as a percent of the sites in the survey.
Although the relative chance that EDB or benzene
would impact a water supply well is roughly the
same in 20% of the sites in the EPA/ASTSWMO
study, the 20% of sites are not the same sites
for EDB and for benzene. Figure 3.4 compares
the chance associated with the highest reported
concentration of benzene in any well at a particular
site to the chance associated with the highest
reported concentration of EDB in any well at the
same site. At 10 of 39 sites, the chance associated
with EDB was greater, and the chance associated
with benzene was greater at the remaining 29 sites.
EDB was the risk driver for impact to water supply
wells at approximately one fourth of the sites.
120n
o
0 20 40 60 80 100 120
Relative Chance Benzene will Exceed MCL
Figure 3.4. Comparison of the relative chance that
EDB will exceed the MCL in a wa-
ter supply well to the relative chance
that benzene will exceed the MCL in
a water supply well for ground water
from motor fuel release sites in the
EPA/ASTSWMO Study.
The total chance that EDB, 1,2-DCA, or benzene
at the 39 sites in the EPA/ASTSWMO study might
-------
contaminate a water supply well above the MCL
can be estimated by summing the individual
relative chances for each site. The sum over
the 39 sites for EDB is 1100, for benzene is 580
and for 1,2-DCA is 10. In aggregate, in the sites
sampled in the EPA/ASTSWMO study, the chances
of contaminating a water supply well with EDB
or benzene at concentrations above the MCL are
roughly equivalent, and the chance for 1,2-DCA is
orders of magnitude lower. These estimates can be
compared to data released by the U.S. Geological
Survey from their comprehensive survey of the
occurrence of volatile organic compounds in
ground water and drinking-water supply wells in
the USA (Zogorski et al., 2006, see Appendix 7).
They sampled ground water supplying domestic
and public wells; EDB was not detected in
462 samples of ground water supplying public
water supply wells. However, EDB was detected
in 3 of 2,085 samples of ground water supplying
domestic water supply wells. All three detections
were above the MCL for EDB of 0.05 ug/L. In
comparison, benzene was detected in five of
1,095 samples of ground water supplying public
water supply wells, and in five of 2,401 samples
of ground water supplying domestic water supply
wells, but in each case the concentration was less
than the MCL of 5 ug/L. In a sampling effort that
was approximately equivalent, three samples were
contaminated with EDB above the MCL and no
sample was contaminated with benzene above the
MCL.
In summary, there were significant concentrations
of EDB and 1,2-DCA at the sites that were
sampled during the EPA/ASTSWMO study.
As a consequence, EDB was the risk driver at
approximately one quarter of the sites in the study.
3.3 The Chance that Conventional
Monitoring Using EPA Method 8260
will fail to detect EDB
Method 8260B prepares a water sample for
analysis by purge-and-trap, separates the volatile
organic compounds by gas chromatography,
and determines the chemical identity and the
concentration of volatile organic compounds by
mass spectrometry. It is widely applied for analysis
of the petroleum components of gasoline in ground
water. Method 8260B can also determine EDB or
1,2-DCA, however the effective method detection
limit for EDB and 1,2-DCA is near 3 ug/L. This
detection limit is problematic for interpretation of
concentrations of EDB because it is almost one
hundred fold higher than the MCL. The detection
limit is not a problem for determination of
1,2-DCA, which has an MCL of 5 ug/L.
Method 8011 prepares the water sample by micro-
extraction, separates volatile organic compounds
by gas chromatography, and determines the
concentration of the volatile organic compounds by
the response of an electron capture detector. This
detector is exquisitely sensitive to halogenated
compounds, but responds hardly at all to
conventional hydrocarbons. The detection limit for
EDB is 0.01 ug/L, which is a reasonable margin
lower than the MCL of 0.05 ug/L.
Analysis of EDB can be included in the routine
analysis of BTEX compounds by Method 8260B
at minimal extra cost. In contrast, monitoring
for EDB by 8011 and monitoring for the BTEX
compounds by Method 8260B can essentially
double the total cost of analysis. The selection of
one method over the other depends on the goals and
priority in risk management.
Figure 3.5 presents the practical consequence
of the difference in detection limits using
Method 8260B or Method 8011. It compares the
distribution of EDB that was determined in the
EPA/ASTSWMO survey to the distribution that
would have been discovered if the survey had been
conducted using Method 8260B. Method 8260B
would have discovered only 40% of the sites with
concentrations of EDB above the MCL.
o
o 80-
§ 7&
i1 60-
1 so-
g 40-
3 so-
w
H— 20"
"c 10-
aj 0-
\
1
%
^
$
o
4
,0
h — EDB MCL
I
< — Method 8260B detection limit
o Method 8011
• Method 8260B
>»
* * A
*»
>
*»
^ A
°- 0.01 0.1 1 10 100 1000
EDB concentration (|xg/L)
10000
Figure 3.5. Distribution of concentrations of EDB
at the leaded gasoline spill sites in the
EPA/ASTSWMO Study where the con-
centrations were above the MCL. Note
that the concentrations of EDB in only
40% of the sites were above the method
detection limit for EPA Method 8260B.
-------
3.4 Co-Distribution of EDB with
1,2-DCA, Benzene, Xylenes, and
Ethylbenzene at Selected Motor
Fuel Release Sites
At many gasoline release sites no ground water
samples have been analyzed for EDB or 1,2-DCA,
but analyses for BTEX compounds are common.
Is there an association between concentrations
of any of the BTEX compounds and EDB and
1,2-DCA that could be used to predict whether
EDB and 1,2-DCA might be present and at
what concentrations? Using the idea that a co-
occurring compound of gasoline might serve as
a conservative tracer, Wiedemeier et al., (1996)
compared the concentrations of benzene in a
release of JP-4 jet fuel to concentrations of the
xylenes. The xylenes and ethylbenzene are a
major fraction of gasoline, they tend to partition
from gasoline slowly, and they are resistant to
biodegradation under anaerobic conditions. In
a similar manner Falta (2004) used the average
composition of gasoline and partitioning theory
to estimate the concentration of EDB, 1,2-DCA,
benzene, ethylbenzene, and total xylenes that
would be expected in ground water in contact
with un-weathered leaded gasoline. The predicted
concentrations for EDB, 1,2-DCA, benzene,
ethylbenzene, and xylenes were 1,900 mg/L,
3,700 mg/L, 37,100 mg/L, 3,000 mg/L, and
13,100 mg/L respectively.
Figure 3.6 compares the predicted concentrations
in contact with un-weathered gasoline, and the
concentrations that would be predicted from
dilution, to the measured concentrations of EDB,
ethylbenzene, and total xylenes in wells sampled
during the EPA/ASTSWMO study. The method
detection limit for EDB, ethylbenzene, and
individual xylenes was 0.005, 0.05, and 0.15 (ig/L
respectively, and values less than the method
detection limit are plotted at the method detection
limit.
The dotted line in the panels of Figure 3.6 are the
predicted concentrations of EDB and ethylbenzene
(Panel A) or EDB and total xylenes (Panel B)
that would be expected in water based on the
assumptions and calculations of Falta (2004). If
a datum falls above the line, there is more EDB in
the ground water than would be predicted from the
concentration of total xylenes or ethylbenzene.
10000,
1000]
100
10
1
0.1
0.01
« Measurements
- - EDB Expected from
Ethylbenzene
m
o
LU
•5
0
10000
1000
100
10
01
0.1 1 10 100 1000
Concentration of ethylbenzene (ng/L)
• Measurements
— EDB Expected from
Total Xylenes
B ,--'''
x^
• _ .
10000
0.01
0.001
0.1 1 10 100 1000 10000 100000
Concentration of total xylenes (ng/L)
Figure 3.6. Association of concentrations of EDB
with concentrations of ethylbenzene
(Panel A) or total xylenes (Panel B).
The maximum concentrations of ethylbenzene
and total xylenes were close to the concentrations
that should be in contact in leaded gasoline. For
any particular concentration of total xylenes or
ethylbenzene, the highest measured concentrations
of EDB were in good agreement with the
concentrations that would be predicted from the
calculations of Falta (2004). The only exceptions
were for concentrations of ethylbenzene and total
xylenes below 1 (ig/L. The very low concentrations
of ethylbenzene and total xylenes may have been
influenced by biodegradation or by retardation due
to differential sorption of ethylbenzene or total
xylenes compared to EDB.
Most of the measured concentrations of EDB
were many fold lower than the concentrations that
would be predicted from measured concentrations
of total xylenes or ethylbenzene. For a release of
leaded gasoline, at concentrations in the range of
1 (ig/L or higher of ethylbenzene and 0.1 (ig/L of
total xylenes, the dotted lines in Figure 3.6 and the
measured concentration of ethylbenzene or total
xylenes can be used to estimate a conservative
upper boundary on the concentration of EDB.
These estimates could be used to conduct a
-------
preliminary RBCA assessment for EDB if data on
the concentrations of EDB are not available.
Figure 3.7 compares concentrations of EDB to
concentrations of 1,2-DCA and benzene wells
sampled under the EPA/ASTSWMO study. The
method detection limit for EDB, 1,2-DCA,
and benzene was 0.005, 0.13, and 0.07 ug/L
respectively, and values less than the method
detection limit are plotted at the method detection
limit.
10000,
15) 1000
-S
m 100]
Q
•5 10
.1 1]
ition of EDB (iig/L) Cone
0.01
0.001
0
10000
1000
100
10
1
l> .... ~~ -
.1 1 10 100 1000 10000
Concentration of DCA (ug/L)
. Measurements y'
— Expected .j < ,•'
B ,-<^'" '
0.01;
0.001:
«.'
0.01 0.1 1 10 100 1000 10000100000
Concentration of Benzene (|o.g/L)
Figure 3.7. Association of concentrations of EDB
with concentrations of 1,2-DCA (Panel
A) and benzene (Panel B).
When EDB was measured in the ground water,
the concentration of 1,2-DCA was low or below
the detection limit. When 1,2-DCA was measured
in the ground water, the concentration of EDB
was low or below the detection limit. We have
no explanation for the strong negative association
of concentrations of EDB and 1,2-DCA. The
maximum concentration of EDB was near the
maximum that would be predicted for ground water
in contact with residual leaded gasoline (based
on calculations of Falta, 2004). The maximum
concentration of 1,2-DCA was almost two orders
of magnitude lower than would be predicted,
suggesting extensive degradation of 1,2-DCA
compared to degradation of EDB. In any case,
concentrations of 1,2-DCA cannot be used to
predict concentrations of EDB.
The maximum concentrations of benzene were
near the concentrations that would be predicted
for ground water in contact with leaded gasoline.
There were several samples clustered at a benzene
concentration near 400 ug/L where the EDB
concentration was up to an order of magnitude
higher than would be predicted from the
concentration of benzene. This probably reflects
preferential degradation of benzene compared
to EDB. There was a second cluster of samples
at benzene concentrations less than 1 ug/L that
indicated a preferential degradation of benzene
compared to EDB. As a consequence, benzene
concentrations can not be used to calculate
a conservative upper boundary on predicted
concentrations of EDB.
3.5 Local Vulnerability of Exposure
to EDB based on Past Usage of
Leaded Motor Fuel
This section estimates the chance that drinking
water will be contaminated with EDB from
gasoline that was released from a leaking
underground storage tank. In particular, it
estimates on a state wide basis the relative chance
that a resident of one of the forty eight contiguous
states will be affected. This section presents maps
relating the regions with high densities of residents
that drink water from shallow wells to regions with
high historical use of leaded gasoline. This section
is intended for decision makers and other staff of
regulator agencies that must apportion resources
for monitoring and risk management of the hazard
associated with EDB in ground water used for
drinking water.
In 2006, the U.S. Geological Survey released a
comprehensive study of the occurrence of volatile
organic compounds in ground water and drinking-
water supply wells in the USA (Zogorski et al.,
2006). They sampled 462 public water supply
wells for EDB and EDB was not detected in any of
them; however, EDB was detected in three of 2,085
domestic water supply wells that were sampled.
All three detections were above the MCL for EDB
of 0.05 ug/L. In comparison, benzene was detected
nine times in 997 public water supply wells, but in
each case the concentration was less than the MCL
of 5 ug/L. Based on the findings of Zogorski et al.,
(2006), the most likely scenario for contamination
of drinking water by EDB from leaded motor fuel
is contamination of private domestic water wells in
a suburban residential landscape.
-------
The chance that an individual will consume
drinking water that is contaminated with EDB from
a leaking underground storage tank is influenced
by a number of site-specific factors including
the source of drinking water, the proximity of
underground storage tanks, and the amount of
leaded gasoline that was stored in the tanks and
might have leaked to contaminate ground water.
It is impossible to calculate a precise estimate of
the chances for exposure from data that are readily
available. The following approach was taken to
provide a broad estimate of the local vulnerability
of exposure to EDB based on past usage of leaded
gasoline in the state.
The first task was to access data on the content
of lead in leaded gasoline. Beginning in the
1930s, the National Institute for Petroleum and
Energy Research (NIPER) has collected data on
the composition of gasoline. The successor to
NIPER (Northup-Grumman) continues to analyze
approximately 1,000 gasoline samples taken
twice per year. Data are available from 174 areas
around the USA; however, the record is often
incomplete. About 35 locations have been sampled
on a continuous basis. This data base is the most
extensive historical record (that is publically
available) of the composition of conventional
and reformulated gasoline in the USA. The data
base provided concentrations of lead in gasoline
from 1976 through 1995 (approximately 27,800
samples), collected at locations within 101 cities
representing 42 states. The weighted average lead
concentrations (gram per liter) of all octane levels
were calculated for five-year categories as follows:
1976-1980, 1981-1985, 1986-1990, and 1991-
1995. The weighted averages for these categories
are 0.43 g/L, 0.15 g/L, 0.038 g/L, and 0.015 g/L
respectively. These averages are in good agreement
with estimates provided in Figure 2 of Falta et al,
(2005). Estimates provided in Figure 2 of Falta
(2004) were applied as the estimate average lead
concentrations for the 5-year categories 1951-1955,
1956-1960, 1961-1965, 1966-1970, and 1971-1975.
The second task was to estimate the amount
of leaded gasoline sold in the USA. The U.S.
Department of Transportation (2008) has published
data on the volume of gasoline sold in each state in
each year between 1949 and 1995. To estimate the
amount of lead in gasoline sold in each state in each
year category, the volume of gasoline sold in each
state for all the individual years within each year
category was summed, and the resulting sums were
multiplied by the average lead concentration for
each respective year category.
The third task was to convert the amount of lead to
the amount of EDB. In the interval 1951 to 1995,
the molar ratio of lead and EDB in gasoline has
been one to one (Falta, 2004). From stoichiometry,
there was 0.435 gram of EDB for each gram of lead
in leaded motor fuel sold in the USA in the period
of interest. To estimate the amount of EDB in
gasoline sold in each state in each year category, the
amount of lead was multiplied by 0.453. Figure 3.8
compares the total amount of EDB in gasoline sold
in five states with small populations, and Figure 3.9
compares EDB in gasoline sold in states with large
populations. Notice the difference in the scale of
the axis showing EDB sold in gasoline between
the figures. In general, sales of EDB in gasoline
increased more than two fold between 1951-1960
and 1976-1980, and then declined substantially by
1991-1995.
Q-/North Dakota
Del:
Nevada
Vermont
Wyoming
elaware
Figure 3.8. Consumption of EDB in leaded motor
fuel in five small states in the USA in
the forty-five years before lead was
banned in automotive motor fuel.
50,000,001
Date 198° "85
California
Texas
Florida
New York
1991-
1995
Figure 3.9. Consumption of EDB in leaded gaso-
line in five large states in the USA in
the forty-five years before lead was
banned in motor gasoline.
-------
The estimated total amount of EDB in leaded
gasoline sold in each state in the period 1951-
1995 is presented in Table 3.3. Almost all of the
leaded gasoline was stored in underground storage
tanks prior to sale. The amount of EDB in leaded
gasoline sold in a state will be taken as a predictor
of the amount of EDB released from releases of
leaded gasoline from underground storage tanks.
The next task is to estimate the probability of
encountering a release of leaded gasoline in an
aquifer, expressed as the amount of leaded gasoline
released per unit surface area. One obvious
approach would be to normalize the amount
of EDB sold by the surface area of each state.
Table 3.3 lists, by state, the total amount of EDB
sold in gasoline in that state divided by the surface
area of the state. This is intended as an estimate
of the chance that ground water under a suburban
residential landscape has been contaminated by
EDB from a release of leaded gasoline. The
relative exposure of an individual in a particular
state consuming ground water with EDB from
leaded gasoline is estimated by dividing the total
amount of EDB in gasoline sold in the state by the
surface area of the state, and then by multiplying by
the fraction of people in the state that drink ground
water from shallow wells, as tabulated in the
2000 census. The estimate is termed the Relative
Exposure Index.
In Table 3.3, the states are ranked by the value
of the Relative Exposure Index. The older
industrialized states on the Atlantic Coast and
the Great Lakes area rank particularly high. The
states in the mid-continent and in the arid western
USA rank particularly low. Figure 3.10 compares
the amount of EDB sold in gasoline divided by
the surface area of the state to regions where
shallow ground water is used for drinking water.
Figure 3.11 provides a closer view of the New
England and Mid-Atlantic states where the Relative
Exposure Index is particularly high.
Table 3.3. Exposure to Drinking Ground Water Contaminated with EDB from Leaded Motor Fuel.
State
Connecticut
New Jersey
Rhode Island
Delaware
Maryland
Massachusetts
Pennsylvania
Michigan
Ohio
Indiana
North Carolina
New Hampshire
New York
Vermont
Florida
Virginia
Wisconsin
West Virginia
South Carolina
Maine
Total EDB Sold in
Motor Fuel
(kg per State)
27,528,745
66,527,161
8,009,494
5,856,848
37,661,911
50,397,289
106,478,053
93,618,641
107,154,665
57,014,485
59,481,984
7,865,305
125,731,461
5,011,479
125,947,001
52,072,141
45,375,403
18,960,163
31,454,595
11,364,807
EDB Sold in
Motor Fuel
(kg per hectare)
21.35
34.21
29.59
11.00
14.93
9.06
23.80
6.24
10.04
6.05
4.68
3.28
9.99
2.01
8.71
5.05
3.12
3.02
3.93
1.36
Population Drinking
Water from Shallow
Wells
Percent of Total
Population
8.4
3.8
4.2
8.7
6.0
3.2
8.5
11.2
6.7
9.4
11.8
15.9
4.7
21.9
5.0
8.3
12.6
12.0
8.1
21.3
Relative Exposure
Index*
179
130
124
96
90
77
76
70
67
57
55
52
47
44
44
42
39
36
32
29
-------
State
Illinois
Kentucky
Tennessee
Georgia
Missouri
Minnesota
Iowa
Alabama
Louisiana
Arkansas
Oklahoma
Washington
Mississippi
California
Oregon
Texas
Nebraska
Kansas
Idaho
North Dakota
Montana
South Dakota
Colorado
New Mexico
Wyoming
Arizona
Utah
Nevada
Alaska
District of
Columbia
Hawaii
Total EDB Sold in
Motor Fuel
(kg per State)
104,486,891
37,360,830
48,344,985
57,010,862
55,806,135
41,275,045
33,112,275
41,607,861
40,237,026
24,123,265
34,808,511
35,527,773
25,868,121
201,242,695
24,419,926
149,485,144
17,431,282
27,511,908
9,045,453
7,622,837
9,297,566
8,589,734
25,159,205
14,068,219
6,108,986
18,415,405
11,781,586
5,045,588
5,265,059
6,909,170
2,387,037
EDB Sold in
Motor Fuel
(kg per hectare)
7.16
3.58
4.43
3.75
3.08
1.89
2.27
3.11
3.39
1.76
1.92
2.04
2.10
4.92
0.97
2.18
0.87
1.29
0.42
0.42
0.24
0.43
0.93
0.45
0.24
0.63
0.54
0.18
Population Drinking
Water from Shallow
Wells
Percent of Total
Population
3.5
6.9
5.0
5.9
6.3
10.0
7.3
4.9
4.2
6.9
5.3
4.8
4.5
1.5
6.3
2.8
6.4
4.1
9.3
9.0
13.6
7.2
3.1
5.8
9.0
1.7
1.1
1.9
0
Relative Exposure
Index*
25
25
22
22
19
19
17
15
14
12
10
10
9.4
7.4
6.1
6.1
5.6
5.3
3.9
3.7
3.3
3.1
2.9
2.6
2.2
1.1
0.59
0.33
* The Relative Exposure Index is calculated as the product of total EDB sold in gasoline in the state (kg), divided by
the surface area of the state (hectares), and then multiplied by the fraction of the people in the state that drink ground
water from shallow wells.
-------
EDB Density and Shallow Groundwater Drinkers
Based on Estimate of EDB Quantities from 1950 to 1995
250 0 250 500 750 1000
km
All Shallow GW Drinkers
1 Dot =1000
EDB Density (1950-1995)
[kg/ha]
II 4-35
Figure 3.10. Relationship between heavy use of leaded motor fuel and the use of shallow ground water for
public water supplies in the contiguous USA.
EDB Density and Shallow Groundwater Drinkers
Based on Estimate of EDB Quantities from 1950 to 1995
100 0 100 200 300
km
All Shallow GW Drinkers
1 Dot =1000
EDB Density (1950-1 995)
[k9/ha]ao-2
I I 2-4
I I 4-35
Figure 3.11. Relationship between heavy use of leaded motor fuel and the use of shallow ground water for
public water supplies in New England, the Mid-Atlantic States, and the Great Lakes region of
the USA.
-------
The Relative Exposure Index takes no account of
the fact that the states in the USA vary widely in
the use and extent of development of the landscape.
Within a state, some areas are urbanized, some
areas are largely suburban residential land, some
areas are farmland or pasture, and some areas are
military reservations or parkland. As an alternative
to normalizing the total amount of EDB sold
in gasoline by the surface area of the state, the
calculation presented in Table 3.4 normalizes the
total amount sold to the population of the state in
the 2000 census.
The total amount of EDB sold in gasoline in a
state divided by the population of the state in
the 2000 census is intended as an estimate of
the chance that ground water under a suburban
residential landscape has been contaminated by
EDB from a release of leaded gasoline. The
relative vulnerability of an individual in a particular
state consuming ground water with EDB from
leaded gasoline is estimated by dividing the total
amount of EDB in gasoline sold in the state by the
population, and then by multiplying by the fraction
of people in the state that drink ground water from
shallow wells, as tabulated in the 2000 census. The
estimate is termed the Relative Vulnerability Index
in Table 3.4.
There is a reasonably close correlation between the
population of a state and the volume of gasoline
sold. Figure 3.12 compares the liters of gasoline
sold in states (in 1995, the latest data available) to
the population (in 2000, the latest data available).
The ratio varies by less than a factor of two
between the extremes. Gasoline service stations
are clustered near major highways, shopping areas,
and suburban residential areas, which is also where
drinking water sources are located. The current
population of a state should serve as a proxy for the
surface area of the state that contains both gasoline
service stations and suburban residential areas
where the inhabitants might have shallow private
wells.
The Vulnerability Index varies by a factor of 30
across all the states in the USA. It is most sensitive
to the fraction of the population that drinks ground
water from shallow wells. The rankings change for
some of the states between the Relative Exposure
Index and the Relative Vulnerability Index.
However, the general trends are the same. The
older industrialized states on the Atlantic Coast and
the Great Lakes area rank particularly high. The
states in the mid-continent rank lower and the states
in the arid western USA rank particularly low.
Table 3.4. Vulnerability to Drinking Ground Water Contaminated with EDB from Leaded Motor Fuel.
State
Vermont
Maine
New Hampshire
Montana
Wisconsin
West Virginia
North Carolina
Michigan
Minnesota
Indiana
Idaho
North Dakota
Wyoming
Delaware
Pennsylvania
Connecticut
Total EDB Sold in
Motor Fuel
(kg per State)
5,011,479
11,364,807
7,865,305
9,297,566
45,375,403
18,960,163
59,481,984
93,618,641
41,275,045
57,014,485
9,045,453
7,622,837
6,108,986
5,856,848
106,478,053
27,528,745
EDB Sold in
Motor Fuel
(kg per resident)
8.23
8.95
6.34
10.31
8.47
10.48
7.39
9.42
8.39
9.38
7.01
11.87
12.37
7.47
8.66
8.07
Population Drinking Water
from Shallow Wells
Percent of Total
Population
21.9
21.3
15.9
13.6
12.6
12.0
11.8
11.2
10.0
9.4
9.3
9.0
9.0
8.7
8.5
8.4
Relative
Vulnerability
Index*
181
191
101
140
107
126
87
105
84
88
65
107
112
65
73
68
-------
State
Virginia
South Carolina
Iowa
South Dakota
Kentucky
Arkansas
Ohio
Nebraska
Missouri
Oregon
Maryland
Georgia
New Mexico
Oklahoma
Tennessee
Florida
Alabama
Washington
New York
Mississippi
Louisiana
Rhode Island
Kansas
New Jersey
Illinois
Massachusetts
Colorado
Texas
Nevada
Arizona
California
Utah
Alaska
District of
Columbia
Hawaii
Total EDB Sold in
Motor Fuel
(kg per State)
52,072,141
31,454,595
33,112,275
8,589,734
37,360,830
24,123,265
107,154,665
17,431,282
55,806,135
24,419,926
37,661,911
57,010,862
14,068,219
34,808,511
48,344,985
125,947,001
41,607,861
35,527,773
125,731,461
25,868,121
40,237,026
8,009,494
27,511,908
66,527,161
104,486,891
50,397,289
25,159,205
149,485,144
5,045,588
18,415,405
201,242,695
11,781,586
5,265,059
6,909,170
2,387,037
EDB Sold in
Motor Fuel
(kg per resident)
7.35
7.84
11.30
11.38
9.25
9.03
9.40
10.19
9.97
7.14
7.11
6.96
7.73
10.09
8.50
7.87
9.35
6.03
6.62
9.11
9.00
7.63
10.23
7.91
8.43
7.94
5.85
7.15
2.52
3.59
5.94
5.28
8.40
12.08
1.97
Population Drinking Water
from Shallow Wells
Percent of Total
Population
8.3
8.1
7.3
7.2
6.9
6.9
6.7
6.4
6.3
6.3
6.0
5.9
5.8
5.3
5.0
5.0
4.9
4.8
4.7
4.5
4.2
4.2
4.1
3.8
3.5
3.2
3.1
2.8
1.9
1.7
1.5
1.1
0
Relative
Vulnerability
Index*
61
64
82
82
63
62
63
65
63
45
43
41
45
54
43
40
46
29
31
41
38
32
42
30
30
26
18
20
5
6
9
6
*The Relative Vulnerability Index is calculated as the product of total EDB sold in gasoline in the state (kg) in the
period 1951 through 1965, divided by the population of the state (2000 census data), and then multiplied by the
fraction of the people in the state that drink ground water from shallow wells (1990 census data).
-------
100000n
i2
g 10000-1
o
in
CD
™ 1000
c
T3
O
CO
-------
4.0
References
Aronson, D. and P. H. Howard. The Environmental
Behavior of Ethylene Dibromide and
1,2-Dichloroethane in Surface Water, Soil, and
Groundwater. Publication 4774. American
Petroleum Institute, Washington, D.C., 2008.
Barbash, J. E and M. Reinhard. Abiotic
dehalogenation of 1,2-dichloroethane and
1,2-dibromoethane in aqueous solution
containing hydrogen sulfide. Environmental
Science & Technology 23: 1349-1358 (1989).
Belay, N. and L. Daniels. Production of ethane,
ethylene, and acetylene from halogenated
hydrocarbons by methanogenic bacteria.
Applied and Environmental Microbiology 5 3:
1604-1610(1987).
Bloom Y., R. Arvena, D. Hunkeler, E. Edwards
and S. K. Frape. Carbon isotope fractionation
during microbial dechlorination of
trichloroethene, cis-l,2-dichloroethene, and
vinyl chloride: implications for assessment of
natural attenuation. Environmental Science &
Technology 34 (13): 2768-2772 (2000).
Bulsara, N. The Occurrence of the Lead
Scavengers Ethylene Dibromide and
1,2-Dichloroethane in Groundwater from
Leaking Gasoline Underground Storage Tanks
with a Case Study in South Carolina. M.S.
Thesis, Clemson University, 221 pages, 2004.
Bouwer, E. J. and P. L. McCarty. Ethylene
dibromide transformation under methanogenic
conditions. Applied and Environmental
Microbiology 50: 527-528 (1985).
Butler, E. C and K. F. Hayes. Kinetics of
the transformation of trichloroethylene
and tetrachloroethylene by iron sulfide.
Environmental Science & Technology 33 (12):
2021-2027(1999).
Butler, E. C and K. F. Hayes. Factors influencing
rates and products in the transformation of
trichloroethylene by iron sulfide and iron metal.
Environmental Science & Technology 35 (19):
3884-3891 (2001).
Cupples, A. M., A. M. Spormann and P. L.
McCarty. Vinyl chloride and c/'s-dichloroethene
dechlorination kinetics and microorganism
growth under substrate limiting conditions.
Environmental Science & Technology 38 (4):
102-1107(2004).
Eisner, M., L. Zwank, D. Hunkeler, R. P.
Schwarzenbach. A new concept linking
observable stable isotope fractionation to
transformation pathways of organic pollutants.
Environmental Science & Technology. 39 (18):
6896-6916 (2005).
Falta, R. W. The potential for ground water
contamination by the gasoline lead scavengers
ethylene dibromide and 1,2-dichloroethane.
Ground Water Monitoring & Remediation
24: 76-87 (2004).
Falta, R. W, N. Bulsara, J. K. Henderson and R.
A. Mayer. Leaded-gasoline additives still
contaminate groundwater: Ethylene dibromide
and 1,2-dichloroethane persist at high levels
despite a phaseout in the late 1980's, but they
get little attention. Environmental Science &
Technology 39: 379A-384A (2005).
Fischer, A., K. Theuerkorn, N. Stelzer, M. Gehre,
M. Thullner, and H. H. Richnow. Applicability
of Stable Isotope Fractionation Analysis for the
Characterization of Benzene Biodegradation in
a BTEX-contaminated Aquifer. Environmental
Science & Technology 41 (10): 3689-3696
(2007).
Gray, J. R., G. Lacrampe-Couloume, D. Gandhi,
K. M. Scow, R. D. Wilson, D. M. Mackay and
B. S. Lollar. Carbon and hydrogen isotopic
fractionation during biodegradation of methyl
tert-butyl ether. Environmental Science &
Technology 36 (9): 1931-1938 (2002).
He, J., K. M. Ritalahti, K.-L. Yang, S. S.
Koenigsberg and F. E. Loeffler. Detoxification
of vinyl chloride to ethene coupled to growth
of an anaerobic bacterium. Nature 424: 62-65
(2003a).
He, J., K. M. Ritalahti, M. R. Aiello and
F. E. Loffler. Complete Detoxification
of Vinyl Chloride by an Anaerobic
Enrichment Culture and Identification of
the Reductively Dechlorinating Population
as a Dehalococcoides Species. Applied and
Environmental Microbiology 69 (2): 996-1003
(2003b).
-------
He, Y. T., J. T. Wilson and R. T. Wilkin. Iron
mineral transformation in a permeable reactive
biowall: column and field experiments.
Environmental Science & Technology 42 (17):
6690-6696 (2008).
Henderson, J. K., D. L. Freedman, R. W. Falta,
T. Kuder and J. T. Wilson. Anaerobic
Biodegradation of Ethylene Dibromide and
1,2-Dichloroethane in the Presence of Fuel
Hydrocarbons. Environmental Science &
Technology 42 (3): 864-870 (2008).
Hirschorn, S. K., M. J. Dinglasan-Panlilio, E.
A. Edwards, G. Lacrampe-Couloume and B.
Sherwood Lollar. Isotope analysis as a natural
reaction probe to determine mechanisms
of biodegradation of 1,2-dichloroethane.
Environmental Microbiology 9 (7): 1651-7
(2007).
Hunkeler D., R. Aravena and B. J. Butler.
Monitoring microbial dechlorination of
tetrachloroethene (PCE) in ground water
using compound-specific carbon isotope
ratios: microcosms and field experiments.
Environmental Science & Technology 33 (16):
2733-2738 (1999).
Hunkeler, D., B. J. Butler, R. Aravena and J. F.
Barker. Monitoring biodegradation of methyl
tert-butyl ether (MTBE) using compound-
specific carbon isotope analysis. Environmental
Science & Technology 35 (4): 676-681 (2001).
Hunkeler D, R. Aravena and E. Cox. Carbon
isotopes as a tool to evaluate the origin and fate
of vinyl chloride: laboratory experiments and
modeling of isotope evolution. Environmental
Science & Technology 36 (15): 3378-3384
(2002).
Jeffers, P. M. and N. L. Wolfe. Homogenous
hydrolysis rate constants - Part II: Additions,
corrections and halogen effects. Environmental
Toxicology and Chemistry 15: 1066-1070
(1996).
Jeffers, P. M, L. M. Ward, L. M. Woytowitch and
N. L. Wolfe. Homogeneous hydrolysis rate
constants for selected chlorinated methanes,
ethanes, ethenes, and propanes. Environmental
Science & Technology 23: 965-969 (1989).
Jendrzejewski, N., H. G. M. Eggankamp and M.
L. Coleman. Characterization of chlorinated
hydrocarbons from chlorine and carbon
isotopic compositions: scope of application to
environmental problems. Applied Geochemistry
16: 1021-1031(2001).
Kampbell, D. H. and S. A. Vandegrift. Analysis
of Dissolved Methane, Ethane, and Ethylene
in Ground Water by a Standard Gas
Chromatographic Technique. Journal of
Chromatographic Science 36 (5): 253-256
(1998).
Klecka, G. M., C. L. Carpenter and S. J.
Gonsior. Biological transformations of
1,2-dichloroethane in subsurface soils
and groundwater. Journal of Contaminant
Hydrology 34: 139-154 (1998).
Kolhatkar, R., J. T. Wilson and L. E. Dunlap.
"Biodegradation of MTBE at Multiple UST
Sites." In: Petroleum and Organic Chemicals
in Ground Water: Prevention, Detection
and Remediation Conference & Exposition
Anaheim, CA, November 15 17, 2000, 32- 49.
Kolhatkar, R., T. Kuder, P. Philp, J. Allen, and J.
T. Wilson. Use of compound specific stable
carbon isotope analyses to demonstrate
anaerobic biodegradation of MTBE in
groundwater at a gasoline release site.
Environmental Science & Technology 36 (23):
5139-5146(2002).
Kostka, J. E. and G. W. Luther. Partitioning and
speciation of solid phase iron in saltmarsh
sediments. Geochemica et Cosmochemica Acta
58: 1701-1710(1994).
Kuder, T., J. T. Wilson, P. Kaiser, R. Kolhatkar,
P. Philp and J. Allen. Enrichment of stable
carbon and hydrogen isotopes during anaerobic
biodegradation of MTBE: microcosm and field
evidence. Environmental Science & Technology
39(1): 213-220 (2005a).
Kuder, T., J. T. Wilson, P. Kaiser, R. Kolhatkar,
P. Philp and J. Allen. Supporting Information
for Enrichment of stable carbon and hydrogen
isotopes during anaerobic biodegradation
of MTBE: microcosm and field evidence.
Environmental Science & Technology 39
(1): 213-220 (2005b). available on the
Environmental Science & Technology web site.
Liang, X., Y. Dong, T. Kuder, L. R. Krumholz,
R. P. Philp and E. C. Butler. Distinguishing
Abiotic and Biotic Transformation of
Tetrachloroethylene and Trichloroethylene
by Stable Carbon Isotope Fractionation.
Environmental Science & Technology 41 (20):
7094-7100(2007).
Mancini, S. A., A. C. Ulrich, G. Lacrampe-
Couloume, B. E. Sleep, E. A. Edwards and
B. Sherwood Lollar. Carbon and hydrogen
isotopic fractionation during anaerobic
-------
biodegradation of benzene. Applied and
Environmental Microbiology 69: 191-198
(2003).
Mayer R. Analysis of Ground-Water Contamination
by Ethylene Dibromide and 1,2-Dichloroethane
at Leaded Gasoline Release Sites. M.S. Thesis.
Clemson University, 188 pages, 2006.
Maymo'-Gatell, X., Y.-T. Chien, J. M. Gossett
and S. H. Zinder. Isolation of a bacterium that
reductively dechlorinates tetrachloroethene to
ethene. Science 276: 1568-1571 (1997).
Maymo'-Gatell, X., T. Anguish and S. H. Zinder.
Reductive dechlorination of chlorinated ethenes
and 1,2-dichlorothane by "Dehalococcoides
ethenogenes" 195. Applied and Environmental
Microbiology 65 (7): 3108-3113 (1999).
McNab, W. W., D. W. Rice and C. Tuckfield.
Evaluating chlorinated hydrocarbon plume
behavior using historical case population
analyses. Bioremediation Journal 4 (4): 311-
335 (2000).
Meckenstock, R. U., B. Morasch, C. Griebler and
H. H. Richnow. Stable isotope fractionation
analysis as a tool to monitor biodegradation in
contaminated aquifers. Journal of Contaminant
Hydrology. 75: 215-255 (2004).
Miller, L. G., R. M. Kalin, S. E. McCauley, J. T.
G. Hamilton, D. B. Harper, D. B. Millet, R. S.
Oremland and A. H. Goldstein. "Large carbon
isotope fractionation associated with oxidation
of methyl halides by methylotrophic bacteria."
Proceedings of the National Academy of
Sciences of the United States of America 98
(10): 5833-5837 (2001).
Newell, C. J., H. S. Rifai, J. T. Wilson, J.
A. Connor, J. A. Aziz and M. P. Suarez.
Calculation and Use of First-Order Rate
Constants for Monitored Natural Attenuation
Studies. EPA/540/S-02/500 (2002). www.epa.
gov/ahaazvuc/download/issue/540S02500.pdf,
available May 12, 2008.
Nijenhuis, I., J. Andert, K. Beck, M. Kastner, G.
Diekert and H. H. Richnow. Stable isotope
fractionation of tetrachloroethene during
reductive dechlorination by Sulfurospirillum
multivorans and Desulfitobacterium sp
strain PCE-S and abiotic reactions with
cyanocobalamin. Applied and Environmental
Microbiology 71 (7): 3413-3419 (2005).
O'Sullivan, G., G. Boshoff, A. Downey and R. M.
Kalin. "Carbon isotope effect during the abiotic
oxidation of methyl-tert-butyl ether (MTBE)."
In Proceedings of the Seventh International In
Situ and On-Site Bioremediation Symposium,
Orlando, FL, 2003.
Pignatello, J. J. and S. Z. Cohen. Environmental
chemistry of ethylene dibromide in soil and
ground water. Reviews of Environmental
Contamination and Toxicology 112: 2-47
(1990).
Poelarends, G. J., J. E. T. van Hylckama Vlieg, J.
R. Marchesi, L. M. Freitas dos Santos and D.
B. Janssen. Degradation of 1,2-dibromoethane
by Mycobacterium sp. Strain GP1. Journal of
Bacteriology 181: 2050-2058 (1999).
Ravi, V., J.-S. Chen, J. T. Wilson, J. A. Johnson, W.
Gierke and L. Murdie. Evaluation of natural
attenuation of benzene and dichloroethanes
at the KL landfill. Bioremediation Journal
2: 239-258 (1998).
Rixey, W. G. and S. Joshi. Dissolution of MTBE
from a residually trapped gasoline source,
API Soil and Groundwater Research Bulletin
No. 13, American Petroleum Institute, 2000.
Schmidt, T. C., L. Zwank, M. Eisner, M. Berg,
R. U. Meckenstock and S. B. Haderlein.
Compound-specific stable isotope analysis of
organic contaminants in natural environments:
a critical review of the state of the art,
prospects and future challenges. Analytical and
Bioanalytical Chemistry 378: 283-300 (2004).
Scholtz, R., T. Leisinger, F. Suter and A. M.
Cook. Characterization of 1-chlorohexane
halidohydrolase, a dehalogenase of wide
substrate range from an Arthrobacter sp.
Journal of Bacteriology 169: 5016-5021
(1987).
Schwarzenbach, R. P., W. Giger, C. Schaffner,
and O. Wanner. Groundwater contamination
by volatile halogenated alkanes: abiotic
formation of volatile sulfur compounds under
anaerobic conditions. Environmental Science &
Technology 19: 322-327 (1985).
Shen, H. and J. T. Wilson. Trichloroethylene
Removal from Ground Water in Flow-through
Columns Simulating a Permeable Reactive
Barrier Constructed with Plant Mulch.
Environmental Science & Technology 41 (11):
4077-4083 (2007).
Sherwood Lollar, B., G. F. Slater, J. Ahad, B.
E. Sleep, J. Spivack, M. Brennan and P.
MacKenzie. Compound specific carbon
isotope analysis - a new approach to evaluating
intrinsic bioremediation. Organic Geochemisty
30 (8a): 813-820(1999).
-------
Slater, G. F., B. Sherwood Lollar, B. E. Sleep
and E. A. Edwards. Variability of carbon
isotope fractionation during biodegradation
of chlorinated ethenes: implications for
field applications. Environmental Science &
Technology 35 (5): 901-907 (2001).
Smallwood, B. J., R. P. Philp, T. W. Burgoyne
and J. D. Allen. The use of stable isotopes
to differentiate specific source markers for
MTBE. Environmental Forensics 2 (3): 215-
221 (2001).
Somsamak, P., H. H. Richnow and M. H.
Haggblom. Carbon isotope fractionation during
anaerobic degradation of methyl tert-butyl
ether under sulfate-reducing and methanogenic
conditions. Applied and Environmental
Microbiology 72 (2): 1157-1163 (2006).
Tandol, V., T. D. DiStefano, P. A. Bowser, J.
M. Gossett and S. H. Zinder. Reductive
dehalogenation of chlorinated ethenes and
halogenated ethanes by a high-rate anaerobic
enrichment culture. Environmental Science &
Technology 28: 973-979 (1994).
USEPA. Use of Monitored Natural Attenuation
at Superfund, RCRA Corrective Action, and
Underground Storage Tank Sites. Office
of Solid Waste and Emergency Response,
Directive 9200.4-17P, 1999. available athttp://
www.epa.gov/swerustl/directiv/d9200417.htm
USEPA. Lead Scavengers Compendium: Overview
of Properties, Occurrence, and Remedial
Technologies. (2006). available at www.epa.
gov/swerust 1/cat/PBCOMPND .HTM
U.S. Department of Transportation. 2008. Federal
Highway Administration. Office of Highway
Policy Information Management. Highway
Statistics Summary To 1995. Highway use
of gasoline, by State, 1949-1995 (Table MF-
226). available http://www.fhwa.dot.gov/ohim/
summary95/index.html
van Warmerdam, E. M., S. K. Frape, R. Aravena,
R. J. Drimmie, H. Flatt and J. A. Cherry. Stable
chlorine and carbon isotope measurements of
selected chlorinated organic solvents. Applied
Geochemistry 10: 547-552 (1995).
Washington, J. W. and B. A. Cameron. Evaluating
degradation rates of chlorinated organics
in groundwater using analytical models.
Environmental Toxicology and Chemistry 20:
1909-1915 (2001).
Wiedemeier, T. H., J. T. Wilson, D. H. Kampbell,
M. A. Swanson and R. T. Herrington.
Approximation of biodegradation rate constants
for monoaromatic hydrocarbons (BTEX) in
ground water. Ground Water Monitoring &
Remediation 16 (3): 186-194, (Summer, 1996).
Wiedemeier, T. H., M. A. Swanson, D. E. Moutoux,
E. K. Gordon, J. T. Wilson, B. H. Wilson,
D. H. Kampbell, P. E. Haas, R. N. Miller,
J. E. Hansen and F. H. Chapelle. Technical
Protocol for Evaluating Natural Attenuation
of Chlorinated Solvents in Ground Water.
EPA/600/R 98/128 (1998). available at
http://www.epa.gov/ada/reports.html
Wiedemeier, T. H., H. S. Rifai, C. J. Newell
and J. T. Wilson. "Intrinsic Bioremediation
of Petroleum Hydrocarbons." In Natural
Attenuation of Fuels and Chlorinated Solvents
in the Subsurface, New York: John Wiley &
Sons, 1900,189-240.
Wilkin, R.T. and K. J. Bischoff. Coulometric
determination of total sulfur and reduced
inorganic sulfur fractions in environmental
samples. Talanta 70: 766-773 (2006).
Wilson, B. H., G. B. Smith and J. F. Rees.
Biotransformation of selected alkylbenzenes
and halogenated aliphatic hydrocarbons in
methanogenic aquifer material: a microcosm
study. Environmental Science & Technology 20
(10): 997-1002 (1986).
Wilson, J. T. "Bioremediation: Natural
Attenuation." In: SERDP/ESTCP Sponsored
Publication: In Situ Remediation of Dissolved
Chlorinated Solvents in Groundwater. In press,
2008.
Wilson, J. T, P. M. Kaiser, and C. Adair. Monitored
Natural Attenuation of MTBE as a Risk
Management Option at Leaking Underground
Storage Tank Sites, EPA 600/R 04/179.
Cincinnati, OH: U.S. Environmental Protection
Agency, 2005.
Zwank, L., M. Berg, M. Eisner, T. C. Schmidt, R.
P. Schwarzenbach and S. B. Haderlein. New
evaluation scheme for two-dimensional isotope
analysis to decipher biodegradation processes:
application to groundwater contamination by
MTBE. Environmental Science & Technology
39(4): 1018-1029(2005).
Zogorski, J. S., J. M. Carter, Tamara Ivahnenko,
W. W. Lapham, M. J. Moran, B. L. Rowe, P.
J. Squillace and P. L. Toccalino. The quality
of our Nation's waters—Volatile organic
compounds in the Nation's ground water and
drinking-water supply wells. U. S. Geological
Survey Circular 1292, 101 pages, 2006.
-------
Appendix A.
Partitioning of EDB and 1,2-DCA between
Gasoline and Water
When leaded gasoline is spilled into ground water,
the EDB and 1,2-DCA in the residual gasoline
diffuses out of the gasoline into the adjacent
ground water until the concentrations of EDB
and 1,2-DCA come to equilibrium. As EDB and
1,2-DCA are transferred from gasoline to water, the
concentrations remaining in gasoline decline. As a
result, the final concentration of EDB or 1,2-DCA
in ground water depends on the ratio of gasoline
to ground water in the aquifer, as described in
Equation A. 1.
C
water
r
original, gasoline
K +
gasoline,water
e,.
A
gasolin
(A.I)
Where C is the final concentration in ground
water °
water, C , , was the original concentration
' original, gasoline °
in the gasoline that was released, 6 ,. is the
° ' gasoline
porosity filled with residual gasoline (expressed
as the volume of gasoline divided by the total
volume of aquifer), dwater is the water-filled porosity
(expressed as the volume of water divided by the
total volume of aquifer), and K is the
1 7' gasohne,water
distribution coefficient between gasoline and water
(expressed as the concentration in gasoline divided
by the concentration in water).
Equation A. 1 was provided by personal
communication from Dr. William Rixey at the
University of Houston, see also Rixey and Joshi
(2000).
Falta (2004) provided estimates of K for
v / L gasolme,water
EDB and 1,2-DCA of 152 and 84 respectively.
Typical original concentrations of EDB and
1,2-DCA in leaded gasoline were near 290 mg/L
and 310 mg/L (Falta, 2004). At these estimated
values, Equation 2.9 predicts maximum possible
concentrations of EDB and 1,2-DCA of 1,900 ug/L
and 3,700 ug/L respectively.
Figure A. 1 extrapolates the predictions of Equation
A.I for a range of concentrations of un-weathered
residual leaded gasoline. Calculations assume
a total porosity of 0.3. The value of 9 , was
r J gasoline
calculated from the value of Total Petroleum
Hydrocarbons (mg/kg) by multiplying by the
bulk density of aquifer sediment at a porosity of
0.3 (1.855 g/cm3) and dividing by the density of
gasoline (0.78 g/cm3). The value of 9 , was
& v o / water
calculated by subtracting 9 ,. from 0.3.
J ° easolme
3000 -I
2500-
2000-
1500-
— EDB
--DCA
/'
/
f
I
o 1000-
5,000 10,000 15,000
TPH (mg/Kg)
20,000
Figure A. 1. Predicted maximum groundwater
concentrations of EDB and DCA for a
range of possible concentrations of un-
weathered residual gasoline in aquifer
sediment.
Typical values for the concentration of residual
gasoline in unconsolidated sediments vary from
2,000 to 10,000 mg/kg. The corresponding
predicted concentrations of EDB under typical
conditions vary from 1,100 ug/L to 1,400 ug/L and
corresponding concentrations of 1, 2-DCA vary
from 1,700 ug/L to 2,500 ug/L.
As EDB and 1,2-DCA partition from residual
gasoline to ground water, and as the moving ground
water carries EDB and 1,2-DCA away from the
-------
residual gasoline, the concentrations of EDB and
1,2-DCA in ground water will decline overtime.
The rate of this weathering process is controlled
by the distribution of EDB and 1,2-DCA between
residual gasoline and ground water, and by the
seepage velocity of the ground water.
Figure A.2 estimates the fraction of EDB and
1,2-DCA that is dissolved in ground water, and
can be flushed away from the residual gasoline
by the flow of ground water, as a function of the
concentration of residual gasoline. For a given
value of TPH (mg/kg), values of 9 ,. and 9
v •—' CJ/ 7 gasoline water
were calculated as described above, and Cwater was
calculated using Equation A. 1. An estimate of the
concentration of the contaminant in the residual
gasoline (C ,. ) was calculated by multiplying
& v gasoline7 J * J o
C by K , t . Then Equation A.2 was used
water J gasoline, water l
to calculate the fraction of contaminant in ground
water.
Fraction, water =
C *6
water water
__
water gasoline gasoline
(A.2)
At typical values of TPH between 2,000 and 10,000
mg/kg, the fraction of EDB in ground water would
vary between 30% and 7%, and the fraction of
1,2-DCA in ground water would vary from 40%
and 12%. With each exchange of the pore water
in contact with residual gasoline, roughly 12% of
EDB and 20% of 1,2-DCA would be flushed away
from the source area. Because a relatively small
proportion of the EDB or 1,2-DCA is in the ground
water, these contaminants would be expected to
weather slowly from residual gasoline.
5,000 10,000 15,000
TPH (mg/Kg)
20,000
Figure A.2. Expected relationship between the
fraction of EDB or DCA that is dis-
solved in ground water, and can be
flushed away by ground water flow, and
the concentration of residual leaded
motor fuel.
-------
Appendix B
Materials and Methods for Laboratory Studies
of Abiotic Degradation of EDB and 1,2-DCA
Batch experiments were conducted with material
collected from the column described in Shen
and Wilson (2007) as the Column with Mulch
and Hematite (column B3). After 875 days of
operation, Column B3 was put into a freezer for
one week. The column was frozen solid, and the
glass container broke apart. The glass was removed
from the column of frozen pore water, shredded
plant mulch, river sand, and hematite and then
the frozen column was cut into nine sections with
a cross cut saw. The original column was 45 cm
long and 15 cm in diameter, each section was 5 cm
long and 15 cm in diameter. The sections were
numbered 1 through 9, with 1 being adjacent to
the column influent and 9 adjacent to the column
effluent. The sections were returned to the freezer
before they could thaw.
Each section was thawed in an anaerobic glovebox
(N2:H2=92.5%:7.5%), homogenized and separated
into 4 equal subsections, transferred to a plastic
bag with a zip seal, sealed without headspace,
and refrozen until used for chemical analysis or
preparation of batch microcosms.
For analysis of Fe and S partitioning, samples were
thawed in the anaerobic glovebox in sealed plastic
bags without headspace. Wet sediments were
used directly for analysis. Analytical results were
corrected for the water content; the water content
was determined gravimetrically. The analysis of
total iron and HC1 extractable iron was modified
after Kosta and Luther (1994). The analysis of acid
volatile Sulfide (AVS) and Chromium Reducible
Sulfide (CRS) followed procedures used by Wilkin
and Bischoff (2006). Results are presented in
Table B.I. Data are from He et al, (2008).
Acid Volatile Sulfide includes and is expected
to be dominated by FeS, and CRS includes and
is expected to be dominated by FeS2. Batch
microcosms were constructed with column material
from section 2 and section 6. The concentration
of iron associated with AVS and CRS in section 2
is 4706 mg/kg, and the concentration of iron that
was associated with AVS and CRS in section 6
is 3856 mg/kg. This represents 92% and 84% of
the total iron remaining in the columns. Allowing
for uncertainty in the determinations, the iron(III)
Table B.I: Distribution of iron, AVS, and CRS along the Column with Mulch and Hematite
as described in Shen and Wilson (2007).
Section
si
s2
s3
s4
s5
s6
s7
s8
s9
Distance
from inlet
cm
0-5
5-10
10-15
15-20
20-25
25-30
30-35
35-40
40-45
H2O
Volume %
41.36
40.12
30.63
36.40
32.45
33.25
30.79
31.55
33.85
Fe fractionation
HC1*
(mg/kg)
2216±183
4926±291
4303±936
4612±1181
3645±925
2967±123
2897±297
3181±510
1702±125
total Fe**
(mg/kg)
3710±0
5095±1110
5610±268
5110±282
6475±855
4610±311
6755±7
5080±0
5400±0
S fractionation
AVS
(mg/kg)
1020±100
2642±371
2087±101
2158±187
1630±74
1449±153
1295±229
1190±231
577±76
CRS
(mg/kg)
2774±625
104±219
530±254
2120±370
987±275
1517±64
1793±463
955±336
1583±395
Data from He et al., (2008).
*extracted for one hour in 0.5M HC1 at pH <2.0, extracts amorphous Fe oxides, AVS
**extracted for 0.5 hr, microwave digestion in HNO3
-------
minerals originally present in the river sand and
the hematite added to the column were almost
entirely converted to FeS and FeS2.
Material from section 2 and section 6 was selected
to prepare batch microcosms.
To remove the confounding effects of sorption to
the plant mulch on concentrations of contaminants,
the sections were sieved to remove the plant
mulch. The frozen section (contained in a plastic
bag without headspace) was allowed to thaw in a
glovebag filled with oxygen free nitrogen, but no
hydrogen. This was done to minimize enrichment
of anaerobic bacteria in the microcosms. The
atmosphere of the glovebag was exchanged three
times before the bags were opened.
Microcosms were prepared with material from
section 2 and labeled as 2-10, 2-11, and 2-12.
Material from section 6 was used to prepare
microcosms labeled 6-15, 6-16, 6-17, and 6-19.
The microcosms were prepared as follows. To
separate the fine sand and iron sulfide precipitates
from the plant mulch, each of the thawed sections
was sonicated in a Branson 1510 sonicator for two
cycles often minutes each. Then the sediment
and pore water was sieved to remove the shredded
plant mulch. The wet sediment that passed the
sieve was blended thoroughly and then distributed
to 20 ml serum vials. Each vial received 19 ml
of wet sediment, and 1.0 ml of a solution that
contained 14gmper liter of Na2SO4. The serum
vials were then sealed with a Teflon-faced butyl
rubber septum and an aluminum crimp cap.
An additional microcosm was prepared using
material from section 6 and labeled as 6-s4. The
microcosm was prepared as follows. The mulch
from section 6 was placed in boiled reverse
osmosis water and sonicated again for 10 minutes
(500 gm wet mulch, 150 ml water). Then the
mulch was removed from the water and sediment
by sieving. The material that passed the sieve
was allowed to stand to separate the sediment
from the water. The sediment was collected,
while the water was added back to the mulch and
sonicated again for 10 minutes. After the water
and sediment were removed from the mulch be
sieving, the water and sediment were combined
with the previous sediment, and the mixture
was allowed to settle overnight. The water was
decanted, the settled sediment was mixed well, and
then 19 ml of wet sediment was added to a 20 ml
serum vial. The vial received 1.0 ml of a solution
that contained 14 gm per liter of Na2SO4. Then the
serum vial was sealed with a Teflon-faced butyl
rubber septum and an aluminum crimp cap.
The water content of the microcosms and the
concentrations of AVS and CRS are presented in
Table B.2.
Dose solutions of EDB and 1,2-DCA were
prepared as follows. A 165-ml serum bottle was
filled with 3.42 grams of Na2SO4, a magnetic spin
bar, and boiled reverse osmosis water to make a
positive meniscus. Then 53 microliters of EDB
or 92 microliters of 1,2-DCA were added with a
syringe, and the bottle was sealed with a Teflon
faced septum and an aluminum crimp cap. The
dose solution was stirred overnight to dissolve the
EDB or 1,2-DCA.
The dose solution contained Na2SO4 to ensure
that adequate sulfate was available to sustain
sulfate reducing conditions over the course
of the incubations. Figure B.I presents data
for the microcosms dosed with EDB. Sulfate
concentrations above 100 mg/L were sustained on
any date where EDB was detected.
Table B.2: Distribution of pore water, AVS, and CRS in microcosms.
Microcosm
2-10
2-11
2-12
6-15
6-s4
6-16
6-17
Water content
% volume basis
43.5
43.1
43.8
46.0
40.2
48.7
57.4
CRS
mg/kg
2222
2417
3349
6319
3454
3309
2309
AVS
mg/kg
1907
2577
2700
4054
1983
2537
1488
AVS
mM/L*
198
272
278
381
236
279
121
*millimole in contact with a liter of pore water.
-------
The microcosms and the dose solutions were taken
into a glovebag with an atmosphere of oxygen free
nitrogen. The microcosms were opened; 1.0 ml
of standing pore water was removed and replaced
with 1.0 ml of the appropriate dose solution. The
microcosms were sealed with new septa and crimp
caps.
The dosed microcosms were removed from the
glove bag and incubated at room temperature in the
dark on a roller that slowly stirred the contents of
the microcosms. The microcosms were sampled
after one hour and then after one, two, three, and
four weeks of incubation. At each sampling period,
the microcosms were removed from the roller and
allowed to settle. The microcosms were placed
inside the anaerobic glove bag and the septa were
removed. For each microcosm, 0.5 ml of standing
water was collected and transferred to a 1.5 ml
micro centrifuge tube for analysis of sulfate. An
additional 0.5 ml aliquot of standing water was
collected and transferred to a 20 ml serum vial
containing 14.5 ml of reverse osmosis water for
analysis of EDB or 1,2-DCA. Three drops of 1:1
HC1 were added to the 20 ml serum vials to adjust
the pH to less than 2. The vials were then sealed
with a septum and crimp cap and stored at 4 °C
until analyzed.
Concentrations of EDB and 1,2-DCA were
determined as described earlier by head space gas
chromatography/mass spectrometry (GC/MS).
Concentrations of sulfate were determined with a
Waters Quanta 4000 Capillary Ion Analyzer, using
a modification of EPA Method 6500, "Dissolved
Inorganic Anions in Aqueous Matrices by Capillary
Ion Electrophoresis," January 1998. The method
detection limit for sulfate was 0.172 mg/L. The
lowest calibration standard was 1.0 mg/L.
10000,
^>
1000:
W
c
_o
100J
g
o
10
10 15
Time of Incubation (days)
20
25
Figure B.I. Consumption of sulfate during incu-
bation of microcosms.
-------
Appendix C
Method for Compound Specific Isotope
Analysis to Determine the Ratio of Stable
Carbon Isotopes in EDB and 1,2-DCA
This method applies compound specific isotope
ratio mass spectrometry to determine the ratio of
the stable isotopes of carbon in 1,2-dibromethane
(EDB) and in 1,2-dichloroethane (1,2-DCA)
dissolved in ground water that is contaminated with
gasoline.
Because there was no available protocol for
analysis of EDB and 1,2-DCA in ground water,
U.S. EPA contracted with the University of
Oklahoma (Dr. Paul Philp,pphilp@ou.edu) to
develop a protocol. Because the method for
analysis of stable isotopes requires baseline
separation of peaks in the gas chromatogram,
it was necessary to separate EDB from the
other fuel components by two-dimensional gas
chromatography. To our knowledge, at this
writing the University of Oklahoma is the only
commercially available source for analysis of
stable isotopes of carbon in EDB in ground water
contaminated with gasoline. It is possible that other
vendors will choose to provide this service in the
future. U.S. EPA makes no endorsement of the
services provided by the University of Oklahoma.
The analytes were extracted by a purge and trap
(P&T model OI 4660) interfaced to a GC-IRMS
instrument (Finnigan MAT 252 IRMS). Due to
the chromatographic complexity of the samples,
satisfactory resolution of EDB and 1,2-DCA
required a 2-dimensional chromatographic
approach (separation on polar GC phase followed
by separation on non-polar GC phase). The
cryogenic focuser at the P&T-GCIRMS interface
described in Kuder et al., (2005b) was programmed
for collecting 2 minute heart-cuts from the retention
window of EDB or DCA of the sample eluting from
the polar pre-column. The heart-cuts were directed
onto a non-polar phase GC column for final
separation followed by on-line combustion and
analysis of the isotope composition. The analyses
were otherwise performed as described in Kuder et
al., (2005a) and Kuder et al., (2005b).
For determination of the 513C values for EDB
and 1,2-DCA, the analytes were purged from
a 25 ml water sample on to a Vocarb 3000 trap
for 12 minutes. The sample temperature was
50 °C. The sample was desorbed over a 3
minute period. The trap temperature was 25 °C
at purge, 240 °C at desorption, and the trap was
baked for 15 minutes at 260 °C before the next
cycle. The initial GC separation was achieved
on a DB-Wax column, 30 m, 250 urn i.d., film
thickness 0.5 urn with Helium as the carrier gas
at an initial flow of 6 ml/min (constant flow).
The GC temperature program was isothermal at
40 °C during 5 minutes, then 4 °C/min up to the
time of elution of 1,2-DCA or EDB. The second
stage of separation was achieved on a DB-MTBE
column, 60 m, 320 urn i.d, film thickness 1.8 um,
with Helium as the carrier gas at an initial flow of
1.5 ml/min (constant flow). The GC temperature
program was isothermal at 40 °C during 5 minutes,
then 2.5 °C/min up to 120 °C and 25 °C/min up
to 220 °C (hold time 15 min.). The combustion
reactor for combustion of the components to CO2
and water was maintained at 980 °C.
The isotopic composition of the samples was
measured relative to a CO2 standard directly
introduced as a reference into the ion source. A
standard solution of EDB and 1,2-DCA was
run randomly between the tests to check the
reproducibility of the PT-GC-IRMS method.
To eliminate problems with method linearity
(relationship between signal strength and the
obtained 513C), the concentrations of diluted
samples and standards were kept within a narrow
range of concentrations, approximately 12 ug/L of
EDB. For samples that were lower in concentration,
standards were run at a corresponding
concentration. It was observed that a portion of
EDB was degraded on active surfaces of the P&T
and/or the GC-IRMS interface. This resulted in
isotope fractionation, and as a result, the overall
method bias for EDB was larger than the typical
-------
range for volatile organic compounds. The samples
discussed in this report were analyzed in several
batches, with varying degrees of analytical bias
(from +1 to +4 %o). The bias affected the samples
and standards in the same extent. The bias remained
steady over the duration of analytical work (over
several days) and could be corrected based on the
EDB standards.
To evaluate the method, standards of EDB
and 1,2-DCA were purchased, and standard
solutions were prepared in water at concentrations
of 4.2 (ig/L, 8.4 (ig/L, and 12 (ig/L. The
reproducibility was good, the sample standard
deviation of duplicate analyses of 513C in EDB
standards averaged 0.5%o. There were several
outliers with standard deviations of l%o.
To determine whether the presence of gasoline
hydrocarbons would bias the samples, a "standard"
of contaminated ground water was prepared by
dissolving gasoline (4 ul) into methanol (43 ml)
and then by spiking 15 ul of this diluted gasoline
into 25 ml of distilled water. The standards of
EDB and DCA were spiked into the "standard"
contaminated ground water, and measured values
were compared to each other. Results are in
Table C.I. Concentrations of EDB and 1,2-DCA
varied from 3.6 to 48 ug/L. Concentrations at
4.2 ug/L, 8.4 ug/L, and 36 ug/L were determined
in replicate. The values of 513C in EDB in 19
replicate samples varied from -27.0%oto -28.5%o,
and values for 1,2-DCA varied form -27.8%o to
-28.9%o. The sample standard deviation for 513C
in EDB and 1,2-DCA was the highest at the lowest
concentration (4.2 ug/L). A sample standard
deviation of 0.5%o was be taken as the data quality
objective for determination of 513C in EDB and
1,2-DCA, as a consequence, 4 ug/L is the lowest
concentration of EDB and 1,2-DCA that could be
analyzed for 513C with acceptable precision.
Table C. 1. Reproducibility of 513C values for EDB and 1,2-DCA prepared by a purge and trap sampler
from ground water containing aqueous solutions of EDB and 1,2-DCA and gasoline.
Run#
1975
1976
1979
1981
1982
1973
1984
1986
1988
1990
1991
1971
1972
1974
1978
1980
1985
1989
1977
Cone
Mg/L
3.6
4.2
4.2
4.2
4.2
4.8
8.4
8.4
8.4
8.4
8.4
12
21.6
36
36
36
36
36
48
EDB
513C
-27.7
-27.9
-27.5
-27.2
-27.3
-27.8
-27.3
-27.6
-27.3
-27.1
-27
-27.6
-28.2
-28.3
-28.3
-28.5
-28.5
-28.5
-28.0
Mean
-27.475
-27.26
-28.42
Stan dev
n=2
0.5
0.4
0. 1
DCA
513C
-28.4
-28.4
-27.8
-27.9
-28.1
-28.6
-28.2
-28.6
-28.3
-28.8
-28.4
-28.5
-28.4
-28.7
-28.8
-28.7
-28.9
-28.7
-28.5
Mean
-28.05
-28.46
-28.76
Stan dev
n=2
0.4
0.4
0.1
-------
Appendix D.
Analytical Methods and Quality Assurance
Six compounds were considered critical parameters
in the Quality Assurance Project Plan; EDB,
1,2-DCA, benzene, nitrate, sulfate, and methane.
Appendix D.1
1,2-Dibromoethane (EDB) in Water
Samples from Field Sites
When EDB was determined in samples of ground
water from field sites, EDB was determined by
EPA Method 8011. In some analyses, the method
detection limit was 0.010 ug/L, and the lowest
calibration standard was 0.03 ug/L. In other
analyses the method detection limit was 0.005 and
the limit of quantitation was 0.020 ug/L.
The acceptance value for the method blank was a
value less than the method detection limit for the
particular report for that sample set. A total of
507 method blanks were analyzed in sample sets
analyzed on 137 different dates, and all 507 method
blanks were below the method detection limit
reported with the sample set.
A total of 414 continuing calibration checks were
performed for EDB in sample sets analyzed on 130
different dates. The acceptance value was 60% to
140% of the nominal value of the calibration check
standard. The range of reported values was 62.6%
to 255% of the nominal values. Two continuing
calibration checks were out of the acceptable range,
a value of 206% and a value of 255% in samples
reported 05-07-2007.
A total of 112 performance standards (secondary
source standards) were evaluated for EDB in
sample sets analyzed on 84 different dates. The
acceptable range of recoveries was 60% to 140% of
the nominal value of the secondary source standard.
None of the secondary source standards were out of
the acceptable range.
A total of 74 matrix spikes were performed for
EDB on samples analyzed on 58 different dates.
The acceptable range of recoveries was 60% to
140% of the spiked concentration. The actual range
of recoveries was 48% to 309%; five samples were
out of the acceptable range.
A total of 11 laboratory control spikes were
analyzed for EDB on samples analyzed on nine
different dates. The acceptable range of recoveries
was 60% to 140% of the spiked concentration. The
actual range of recoveries was 72.4% to 134%. No
samples were out of the acceptable range.
Appendix D.2
1,2-Dibromoethane (EDB) in Water
Samples from Microcosm Studies
In samples from laboratory microcosms, EDB was
determined by head space gas chromatography/
mass spectrometry (GC/MS) using a modification
of EPA Method 5021 A, "Volatile Organic
Compounds in Various Sample Matrices using
Equilibrium Headspace Analysis," June 2003.
Samples were collected for analysis with an
automated static head space sampler. Analytes
were determined by gas chromatography/mass
spectrometry using an Ion Trap Detector. The
lowest calibration standard was 0.5 ug/L; the
method detection limit was 0.1 ug/L.
The acceptance value for EDB in the method blank
was less than the method detection limit. A total
of 29 method blanks were analyzed in sample sets
analyzed on five different dates, and all 29 method
blanks were below the method detection limit.
A total of 52 continuing calibration checks were
performed for EDB in sample sets analyzed on five
different dates. The acceptance value was 60%
to 140% of the nominal value of the calibration
check standard. The range of reported values
was 88% to 112% of the nominal values. No
continuing calibration check samples were out of
the acceptable range.
A total of 15 matrix spikes were performed for
EDB on samples analyzed on five different dates.
The acceptable range of recoveries was 60% to
140% of the spiked concentration. The actual range
of recoveries was 96% to 107%. No spike samples
were out of the acceptable range. A total of 46
secondary source standards were analyzed for EDB
on samples analyzed on five different dates. The
acceptable range of recoveries was 60% to 140%
of the spiked concentration. The actual range of
recoveries was 90% to 113%. No samples were out
of the acceptable range. A total of seven laboratory
duplicates were analyzed on five different dates.
The acceptable range of agreement was for the
laboratory duplicates to agree with each other
with a relative percent difference of ± 25%. Two
of the seven laboratory duplicates had detectable
-------
concentrations of EDB; the relative percent
difference was 5% and 11%.
Appendix D.3
1,2-Dichloroethane (DCA) in Water
Samples from Field Sites
Concentrations of 1,2-DCA were determined by
head space gas chromatography/mass spectrometry
(GC/MS) using a modification of EPA Method
5021A, "Volatile Organic Compounds in Various
Sample Matrices using Equilibrium Headspace
Analysis," June 2003. Samples were collected
for analysis with an automated static head space
sampler. Analytes were determined by gas
chromatography/mass spectrometry using an Ion
Trap Detector. The lowest calibration standard was
0.5 ug/L; the method detection limit was 0.13 ug/L.
A total of 79 method blanks were analyzed for
1,2-DCA in sample sets analyzed on 44 different
dates. All method blanks were less than the method
detection limit.
A total of 86 continuing calibration checks were
analyzed in sample sets on 50 different dates. The
range of acceptable values was 80% to 120% of the
nominal value of the check samples. The values
of the continuing calibration check samples ranged
from 81% to 120% of the nominal value. All
samples were within the acceptable range.
A total of 39 performance evaluation standards,
or secondary source standards, were analyzed on
29 different dates. The range of acceptable values
was 80% to 120% of the nominal value of the
check samples. The values of the secondary source
standards varied from 90% to 117% of the nominal
value. All samples were within the acceptable
range.
A total of 25 matrix spikes were performed in
water samples analyzed on 19 separate dates. The
acceptable range of recoveries was 70% to 130%
of the spiked concentration. The actual range was
83% to 119% of the spiked concentration. All
samples were within the acceptable range.
Appendix D.4
1,2-Dichloroethane (DCA) in Water
Samples from Microcosm Studies
Samples from microcosms were analyzed using
the same method as described above for samples
from field sites. A total of 29 method blanks were
analyzed for 1,2-DCA in sample sets analyzed on
five different dates. All method blanks were less
than the method detection limit.
A total of 52 continuing calibration checks were
analyzed in sample sets on five different dates. The
range of acceptable values was 80% to 120% of the
nominal value of the check samples. The values
of the continuing calibration check samples ranged
from 91% to 112% of the nominal value. All
samples were within the acceptable range.
A total of 46 performance evaluation standards, or
secondary source standards, were analyzed on five
different dates. The range of acceptable values
was 80% to 120% of the nominal value of the
check samples. The values of the secondary source
standards varied from 91% to 120% of the nominal
value. All samples were within the acceptable
range.
A total of 15 matrix spikes were performed in
water samples analyzed on five separate dates. The
acceptable range of recoveries was 70% to 130%
of the spiked concentration. The actual range was
91% to 115% of the spiked concentration. All
samples were within the acceptable range.
A total of seven laboratory duplicates were
analyzed on five different dates. The acceptable
range of agreement was for the laboratory
duplicates to agree with each other with a relative
percent difference of ± 25%. Five of the seven
laboratory duplicates had detectable concentrations
of 1,2-DCA; the relative percent differences were
6%, 0.8%, 6%, 0.6% and 4.4%.
Appendix D.5
Benzene in Water Samples from
Field Sites
Concentrations of benzene were determined by
headspace gas chromatography/mass spectrometry
(GC/MS) using a modification of EPA Method
5021 A, "Volatile Organic Compounds in Various
Sample Matrices using Equilibrium Headspace
Analysis," June 2003. Samples were collected
for analysis with an automated static head space
sampler. Analytes were determined by gas
chromatography/mass spectrometry using an Ion
Trap Detector. The lowest calibration standard was
0.5 ug/L; the method detection limit was 0.07 ug/L.
A total of 78 method blanks were analyzed for
benzene in sample sets analyzed on 45 different
dates. All method blanks were less than the method
detection limit.
A total of 88 continuing calibration checks were
analyzed in sample sets on 50 different dates. The
range of acceptable values was 80% to 120% of the
nominal value of the check samples. The values
of the continuing calibration check samples ranged
-------
from 89% to 115% of the nominal value. All
samples were within the acceptable range.
A total of 38 performance evaluation standards,
or secondary source standards, were analyzed on
29 different dates. The range of acceptable values
was 80% to 120% of the nominal value of the
check samples. The values of the secondary source
standards varied from 87% to 107% of the nominal
value. All samples were within the acceptable
range.
A total of 25 matrix spikes were performed in
water samples analyzed on 19 separate dates. The
acceptable range of recoveries was 70% to 130%
of the spiked concentration. The actual range was
89% to 109% of the spiked concentration. All
samples were within the acceptable range.
Appendix D.6
Nitrate
Samples were analyzed for nitrate plus nitrite
nitrogen using an automated hydrazine reduction
method based on EPA Method 353.2. The method
detection limit was 0.01 mg/L and the lowest
calibration standard was 1.0 mg/L
A total of 14 method blanks were analyzed for
nitrate plus nitrite nitrogen in sample sets analyzed
on 6 different dates. There were four method
blanks below the method detection limit, and
concentrations in the remainder of the samples were
below 0.047 mg/L. All data were considered of
adequate quality for the purpose of this report.
A total of 19 continuing calibration checks were
analyzed in sample sets on six different dates. The
range of acceptable values was 90% to 110% of the
nominal value of the check samples. The values
of the continuing calibration check samples ranged
from 94% to 105% of the nominal value. All
samples were within the acceptable range.
A total of seven performance evaluation standards,
or secondary source standards, were analyzed on
six different dates. The values of the secondary
source standards varied from 93% to 101% of the
nominal value. No data were flagged as being out
of the acceptable range.
A total of 10 matrix spikes were performed in
water samples analyzed on six separate dates. The
acceptable range of recoveries was 80% to 120%
of the spiked concentration. The actual range was
8 8% to 115% of the spiked concentration. All
matrix spikes were less than the method detection
limit.
A total of 13 laboratory duplicates were performed
on six separate days. The acceptable range was a
percent relative difference of ± 10%. The relative
percent difference ranged from 0.05% to 1.05%.
Appendix D.7
Sulfate
Concentrations of sulfate were determined by two
separate methods. In some samples, concentrations
of sulfate were determined with a Waters Quanta
4000 Capillary Ion Analyzer, using a modification
of EPA Method 6500, "Dissolved Inorganic
Anions in Aqueous Matrices by Capillary Ion
Electrophoresis," January 1998. The method
detection limit for sulfate was 0.172 mg/L. The
lowest calibration standard was 1.0 mg/L.
A total of 14 method blanks were analyzed for
sulfate in sample sets analyzed on seven different
dates. All method blanks were less than the method
detection limit. Two field blanks were analyzed on
one day. Both were less than the method detection
limit.
A total of 14 continuing calibration checks were
analyzed in sample sets on seven different dates.
The range of acceptable values was 90% to 110%
of the nominal value of the check samples. The
values of the continuing calibration check samples
ranged from 94% to 102% of the nominal value.
All samples were within the acceptable range.
A total of seven performance evaluation standards,
or secondary source standards, were analyzed on
six different dates. The values of the secondary
source standards varied from 86% to 97% of the
nominal value. No data were flagged as being out
of the acceptable range.
A total of six matrix spikes were performed in
water samples analyzed on six separate dates. The
acceptable range of recoveries was 80% to 120%
of the spiked concentration. The actual range was
89% to 97% of the spiked concentration. One
matrix spike was out of range by 1%. Despite this
one excursion out of the acceptable range in the
SOP, all data were considered of adequate quality
for the purpose of this report.
A total of six laboratory duplicates were performed
on five separate days. The acceptable range was a
percent relative difference of ± 10%. The relative
percent difference ranged from 0% to 2.74%.
Water samples that had been preserved with HC1
were analyzed by an automated turbidimetric
method based on EPA Method 375.4. The method
detection limit was 1.14 mg/L and the lowest
calibration standard was 2.0 mg/L.
-------
A total of eight method blanks were analyzed for
sulfate in sample sets analyzed on three different
dates. All method blanks were less than the method
detection limit.
A total of 13 continuing calibration checks were
analyzed in sample sets on three different dates.
The range of acceptable values was 90% to 110%
of the nominal value of the check samples. The
values of the continuing calibration check samples
ranged from 93% to 104% of the nominal value.
All samples were within the acceptable range.
A total of four performance evaluation standards, or
secondary source standards, were analyzed on three
different dates. The values of the secondary source
standards varied from 86% to 97% of the nominal
value. No data were flagged as being out of the
acceptable range.
A total of four matrix spikes were performed in
water samples analyzed on two separate dates. The
acceptable range of recoveries was 80% to 120%
of the spiked concentration. The actual range was
85% to 114% of the spiked concentration. All
matrix spikes were less than the method detection
limit.
A total of four laboratory duplicates were
performed on four separate days. The acceptable
range was a percent relative difference of ± 10%.
The relative percent difference ranged from 0.70%
to 7.75%.
Appendix D.8
Methane
Concentrations of methane were analyzed using
a headspace equilibration technique based on
Kampbell and Vandegrift (1998). The method
detection limit was 0.0001 mg/L and the lowest
calibration standard was 0.0010 mg/L.
A total of 18 method blanks were analyzed for
methane in sample sets analyzed on five different
dates. All method blanks were below the method
detection limit.
A total of 27 continuing calibration checks were
analyzed in sample sets on seven different dates.
The range of acceptable values was 85% to 115%
of the nominal value of the check samples. The
values of the continuing calibration check samples
ranged from 93% to 105% of the nominal value.
All samples were within the acceptable range.
A total of six performance evaluation standards,
or secondary source standards, were analyzed on
five different dates. The expected values of the
secondary source standards varied from 85% to
115% of the nominal value. The measured values
of the secondary source standards varied from 99%
to 101% of the nominal value.
A total of eight laboratory duplicates were
performed on six separate days. The acceptable
range was a percent relative difference of ± 20%.
The relative percent difference ranged from 0.95%
to 3.17%. Two field duplicates were analyzed;
however, they both were below the detection limit.
-------
-------
United States
Environmental Protection
Agency
PRESORTED STANDARD
POSTAGES FEES PAID
EPA
PERMIT NO. G-35
Office of Research and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |