New England Interstate
Water Pollution Control
Commission 01852-1124
www.neiwpcc.org/lustline.htm
116 John Street
Lowell, Massachusetts
LUST
51ON06001
Bulletin 52
May
2OO6
A Report On Federal & State Programs To Control Leaking Underground Storage Tanks
So What about
Those HO and K5 Fu
A Pbeusston OM Materials Cowpatibilr
by Edward W. English II
For some, the use of ethanol as a
fuel blend seems like a new and
intriguing concept. For as long as 1
can remember I have been filling my cars with
gasoline refined from crude oil, and not with
the fermentation product of corn. However,
the use of ethanol as a fuel blend is not as
new an idea as some would think. People
have been experimenting successfully with
ethanol fuel since the 18th century. And
although ethanol was the primary fuel source for
Henry Ford and was used to supplement petroleum
feedstocks during World War II, it's only been over the
last 25 to 30 years that research and development of alco-
hol fuels, such as methanol and ethanol, has taken center
stage as a way to reduce vehicle tailpipe emissions and
our reliance on foreign oil (1).
Since the late 1970s, oxygenates such as ethanol
(EtOH) and methyl ferf-butyl ether (MtBE) have been
added to gasoline to improve vehicle emissions. Ethanol
is not only an effective oxygenate, it has the added advan-
tage of improving fuel octane. The use of 10 percent
ethanol (E10) can increase the octane of base gasolines
two to three points (1). This also made ethanol a good
replacement to the lethal additive tetra ethyl lead (TEL)
(2,3). E10 gasoline is currently used in about 30 percent of
the gasoline consumed in the United States. However,
due to the accelerated phaseout of MtBE, ethanol will
soon be used to a much greater extent in gasoline.
For the past several decades, a variety of automotive
manufacturers have been designing and manufacturing
vehicles to run on fuels made of 85 percent ethanol and
15 percent gasoline (E85). Although ethanol fuel is
approximately 25 percent lower in heat energy than con-
• continued on page ',
Inside
is Your US? System Ethanol Compatible?
Ethanol? Good. Yes? Urn...
Here Comes Ultra-Low Sulfur Diesel
Mapping Hydrostratigraphy, Predicting MtBE Plume Diving
OUST Update
Subsurface Vapor Attenuation—Update on Risk Pathway
Operator Training; Boon or Bust?
What a Tank Operator Needs to Know
Update—Cnergy Policy tot
Tanks on Tribal Lands—Nez Perce
FAQs from the NWGLDE
-------
LlTSTLiiie Biillt'fni 52 • Mm/2006
• E10 and E85 from page 1
ventional gasoline, for some the
energy tradeoff at the E85 level is
worth improved vehicle emissions
and reduced consumption of foreign
crude oil (1,4,5).
Since the early 1990s, there has
been an accumulation of information
documenting the compatibility of
alcohols, such as methanol and
ethanol, and ethers, such as MtBE, on
the materials found in automobile
fuel systems and petroleum storage
equipment. However, there is a
resounding lack of clearly defined
industry standards that effectively
and objectively evaluate the effects of
E10 and E85 fuels on the variety of
materials commonly found in the
equipment used to store, distribute,
and dispense ethanol-blended (E-
blend) fuels (6,7,8).
The goal of this article is to shed
some light on the history of ethanol
as a fuel and to provide a glimpse
into the chemical and physical prop-
erties of ethanol that are the basis for
material-compatibility concerns.
L.U.S.T.LJne
_ Hen Fiye, fih'tor
Ridd Pappo, layout
Mweel Moretu, Technical Adviser
Patricia His, fMX Technical Adviser
Ronald Poltalg NlfWBCC Executive Director
tynn Decent, ERA Project Cficer
LUSTlane Is a product of the New England
Intestate Watef PoJiutiop. Coatrol Coimnis-
sit» (NUWPCCI. It is produced through &
cooperative itece«nent (#T-S30380-01)
between MSWCC and the U.S.
EttrifonoKnla Protection Agency.
IttSTUm Is Jsstwd as « cermHttwfcatton
service tot the Subtitle I BCRA
Hazardous & Solid Waste Amendments
rule promulgation process.
UlBTUm is produced to promote
infarwattan exchange on USF/LUST issues.
Die opinions and information stated herein
see tftsee of tj» authors pnd do i»ot neee*-
sanff rtsftsitfcft (^intense! UlttWCC.
be
C.
H1WK3C wis 5esfaHfsted by m Act of
Congress in 1947 and remains the oldest
. a(j«)Mgf ja tfcNeriieast United States
concerned wj™ eoortinatiofl of Ae aiulti-
of the stales! of Consectiteut, Maine,
Massachusetts, Mew Hampshire,
Mew York, Shed* Island, and Vermont
MEIWPCe
116 |ohn Street
Lowell, MA 01852-1124
Telephone: (978) 523-7929
Fax: (97&) 323-7919
lustline@neiwpcc.org
-------
material (8). Therefore, in this type of
solution the activity or chemical
potential of each component will be
equivalent to its concentration or
mole fraction in solution. This is best
illustrated in Figure 1, where we can
observe the activity of MtBE in an
ASTM C test fuel that is composed of
roughly equal proportions of toluene
and isooctane.
MtBE is an ether, and as such,
possesses an oxygen molecule. How-
ever, the surrounding atoms of car-
bon and hydrogen diminish the
electrostatic nature of the oxygen
atom, making it essentially nonpolar.
This allows MtBE to commingle with
the ASTM C fuel without significantly
altering the activity of the solution.
Therefore, individual activities of the
MtBE, toluene, and isooctane are pro-
portional to their respective concen-
trations or mole fractions.
It is the unique physical and
chemical characteristics of gasoline
atom carries a partially negative
charge, while the hydrogen atom car-
ries a partially positive charge. The
electrostatic charges carried by these
two atoms give the hydroxyl group
an overall "polar" characteristic,
analogous to the poles on a magnet.
There are a number of natural and
manmade compounds that possess
this type of functional group—water
can be thought of as a hydrogen atom
attached to a hydroxyl group (H-
OH).
The unique aspect of the
hydroxyl functional group is that it
gives the ethanol molecule the ability
to "hydrogen bond" with other chem-
icals that possess the complementary
functional group—like water and
ethanol. Although weaker than ionic
bonding, hydrogen bonding is still a
much stronger electrostatic force than
the negligible forces that exist
between the nonpolar compounds
found in gasoline. For example,
Mole Fraction MTBE in Fuel C
FIGURE 1. Activity of MtBE, toluene, and isooctane as a function of ether in
ASTM C fuel (8)
hydrogen bonding is what gives
water its surface tension and high
boiling point. It has also been shown
that hydrogen bonding is what allows
alcohols such as neat methanol to self-
associate, forming a tetracyclic and
nonpolar compound (8).
When dilute or low concentra-
tions of polar ethanol molecules are
mixed with nonpolar gasoline, the
ethanol molecules behave much dif-
ferently and influence the activity of
the other constituents, as demon-
strated in Figure 2. Due to the electro-
static nature of the hydroxyl group,
ethanol molecules preferentially
hydrogen bond with each other
rather than uniformly mix with the
gasoline.
Since chemical potential or activ-
ity is directly proportional to temper-
ature, the activity of dilute ethanol
and other fuel constituents is dispro-
portionately greater than their
respective concentrations. The elec-
trostatic interactions between mole-
cules can significantly influence the
chemical potential or activity of a
solution, which plays an important
role in the equilibrium absorption
and permeation of chemicals by a
nonmetallic material (8).
Potential E-blend Fuel Issues
So now that we have a better under-
standing of some of the physical and
chemical differences between gaso-
line and ethanol, how can we apply
what we've learned to better under-
stand potential issues that could arise
from E10 and E85 fuel systems? Gen-
erally, one or more of the following
• continued on page 4
that have permitted petroleum man-
ufacturers to implement design
strategies to make metallic and non-
metallic petroleum equipment com-
patible with gasoline and resistant to
degradation during storage, trans-
portation, and dispensing.
Comparatively, ethanol has a few
characteristics that make it similar to
gasoline. The ethanol molecule is
composed of two carbon atoms
linked together to form a short linear
molecule that possesses six bonding
locations, which are occupied by five
hydrogen atoms and one unique
functional group made up of a single
oxygen and hydrogen atom called a
hydroxyl group (OH). The oxygen
0.8 --
g. 08
O
O
•8
2? O4
I
BO
Toluene
Iswctane
Mole Fraction of EtOH in ASTM Fuel C
FIGURE 2. Activity of MtBE, toluene, and isooctane as a function of ether in ASTM C
fuel (8)
-------
• E10 and £85 from page 3
issues could result when storing, dis-
tributing, or dispensing E10 and E85
fuels—phase separation, solvency,
metal corrosion, and permeation of
nonmetals.
• Phase Separation
Ethanol will mix with gasoline,
but it does so reluctantly. Although
gasoline is nonpolar, it can only hold
up to 0.2 percent dissolved water
before the water "drops" out of solu-
tion to the bottom of the storage ves-
sel as free water. Conversely,
hydrogen bonding allows E10 fuel to
hold much more dissolved water than
gasoline—approximately 0.5 percent.
This is because the energy needed for
ethanol and water to hydrogen bond
is much lower than the higher energy
required to keep ethanol evenly dis-
tributed with gasoline. Because of
this, ethanol and water will continue
to preferentially bond until the
ethanol and water drop out of solu-
tion, a process known as "phase sep-
aration."
When an E10 fuel undergoes
phase separation, a separate layer of
water with a high percentage of
ethanol settles to the bottom of the
fuel system. The remainder of the
fuel in the upper layer is gasoline
containing a small percentage of
ethanol. An E10 fuel system near its
saturation limit may experience
phase separation if there is a sudden
drop in temperature.
Unfortunately, vehicles are not
designed to run on a mixture of
ethanol and water, so if phase-sepa-
rated ethanol and water is pumped
into a vehicle's fuel tank, the vehicle
will become stranded at the pump or
eventually stall after fueling.
Phase separation of E85 fuels
may also occur, but not as often.
Approximately 4 percent water is
needed for phase separation to occur.
This is a significant amount of water
for a fuel system and typically would
not happen. However, more impor-
tant is what a water concentration of
4 percent can do in an E85 system—
namely facilitate galvanic corrosion.
• Solvency
As we know, alcohol has the abil-
ity to dissolve organic material.
Although E10 and E85 fuels have dif-
fering concentrations of ethanol, each
has the ability to dissolve the petro-
leum-based sediment, particulates,
and lacquers found in fuel tanks
tanks previously dispensing gasoline
or diesel fuel. In this case, E85 would
have a greater solvent capability than
to E10. It is very important that used
fuel tanks designated for E10 and E85
conversion are thoroughly cleaned
and inspected by a reputable and
bonded company in order to identify
potential corrosion issues that could
result in tank leaks.
It is also important that appropri-
ate in-line filters are installed. These
filters provide the last line of defense,
protecting automobiles from receiv-
ing contaminated fuel that could seri-
ously affect their drivability.
When selecting an in-line filter,
two very important issues should be
taken into consideration—particle
size and efficiency. Particle size alone
does not guarantee performance.
When selecting a filter, it is just as
important to consider its efficiency
rating. The greater the efficiency rat-
ing, the better the filter will perform
at its stated particle size.
• Metal Corrosion
Metals typically found in UST-
dispensing systems include alu-
minum alloys, brass, copper, steel,
and zinc. As mentioned above, gaso-
line is electrically nonconductive and
generally will not corrode these met-
als. Corrosion mechanisms found in
gasoline fuel systems typically
involve general corrosion and pitting
corrosion, which usually take place at
the bottom of the fuel-storage tank
where the water layer exists.
Conditions that lead to and exac-
erbate corrosion of metals in any
hydrocarbon fuel system include
water, ionic contaminants, pH, and
microbial contamination (8). Water
typically enters the fuel distribution
and storage system either when
newly refined warmer fuel cools in
the distribution pipeline or when
poor water-ballast-stripping tech-
niques are used during offloading
from tankers. Water enters an UST
through a fuel delivery, poor tank
seals, or condensation of water vapor
in the vapor-recovery system (11).
Water entering ethanol fuel sys-
tems becomes a more serious issue.
As discussed above, ethanol and
water have a strong preference to
hydrogen bond, even in the presence
of gasoline. This preferential bonding
is the reason why the ethanol used to
make E10 and E85 fuels is shipped by
railcar or tanker truck, rather than
pipeline, and blended at the terminal
prior to final delivery to the UST
facility.
As water or moisture enters
ethanol fuel, it is drawn up into the
bulk fuel along with ionic contami-
nants, which increases the conductiv-
ity of the solution. As the electrical
conductivity increases, the fuel more
easily facilitates galvanic corrosion
between dissimilar metals—metals
that are widely separated in the gal-
vanic series, such as aluminum and
steel (5,6,7).
Fortunately, in the quarter cen-
tury that E10 fuels have been in the
U.S. market, there have been very
few reported incidents of significant
corrosion of the metals typically
found in fuel-dispensing systems.
Metals recommended for E10 fuels
are provided in Table 1 below.
TABLE 1
Metal Compatibility in E10
(5,14,15)
RECOMMENDED
Aluminum
Black iron
Brass
Bronze
Carbon steel
Stainless steel
NOT RECOMMENDED
Galvanized zinc
On the other hand, E85 fuels are
viewed much differently because
there are other issues to consider. As
discussed above, E85 fuel is capable
of absorbing greater quantities of
water and contaminants. As a result
of the increased water content, E85
fuel becomes much more conductive,
promoting galvanic corrosion more
easily, which in turn has a very dele-
terious effect on active or anodic met-
als in a fuel system.
The effects of galvanic corrosion
on soft or anodic metals were more
readily apparent during the research
and development of M85 fuels,
which contained 85 percent methanol
and 15 percent gasoline. Testing by
the automotive industry in the late
1980s and early 1990s demonstrated
that methanol in M85 fuels was
observed to be very aggressive
toward elastomers, polymers, and
soft metals, extracting salts and com-
-------
pounds from elastomers and poly-
mers and aggressively dissolving soft
metals, such as aluminum nozzles.
The result was an M85 fuel that had a
very high suspended solids content,
creating a gel-like substance made of
aluminum hydroxide, water, and
methanol.
Although the difference between
methanol and ethanol is one carbon,
ethanol is not considered to be as
aggressive as methanol, but many of
the same degradation issues have
been observed to occur over a longer
period of time.
Thus, it is recommended that dis-
similar metals not exist in systems
that store, distribute, or dispense E85
fuels on a continuous basis. The met-
als recommended for E85 systems are
provided in Table 2.
TABLE 2
Metal Compatibility in E85
(5,14,15)
RECOMMENDED
• Bronze
• Black iron
• Mild steel
1 Stainless steel
• Unplated steel
1 Nickel-plated alu-
minum or brass
1 Aluminum alloy
NOT RECOMMENDED
• Brass
• Lead
• Lead solder
• Magnesium
• Lead-tin alloy
(tin-plated steel
• Zinc
• Permeation of Nonmetals
Although the mechanism is not
really understood, the presence of
ethanol in fuel facilitates the perme-
ation of hydrocarbons through cer-
tain elastomers and thermoplastics,
and to a lesser degree, thermoset
products (8,12,13). Permeation refers
to the mass transport of a substance
(or solvent) through a membrane that
is driven by a chemical potential or
activity gradient. Dilute ethanol in an
E10 fuel possesses sufficient chemical
potential or activity to become the
chemical gradient that drives perme-
ation through the material.
Of course, permeation is depen-
dent on several factors, such as
solvent-material interaction, the sur-
face-area-to-thickness ratio of the
nonmetal, and the degree of cross
linking (e.g., elastic contraction) in a
material (8).
For example, we know that
ethanol and hydrocarbons do not
permeate through metals. To a lesser
degree, the same could be said for
thermoset products—fiberglass-rein-
forced polymeric materials (8). How-
ever, the same principles do not
necessarily hold true for elastomers
and thermoplastics—materials that
have very little to no cross-linking.
The lack of cross-linking can affect
the rate of permeation.
The permeation of gasoline or
ethanol through a nonmetal can
result in a change in the physical,
chemical, and mechanical properties
of an elastomer or polymer. For
example, permeation can result in
excessive swelling. This in turn can
cause an elastomer or polymer that
performs an important dynamic
function, such as a pump seal or
pump impeller, to fail. In principle,
the same can be said of elastomers
and polymers that also perform static
functions.
Elastomers and polymers also
face an issue where excessive perme-
ation and swelling can lead to plasti-
cization. In some cases, chemicals
such as antioxidants and heat stabi-
lizers are added to nonmetals to con-
fer certain performance properties.
Since these additives are not chemi-
cally bound, excessive swelling in a
material can eventually lead to the
extraction of these plasticizers as the
solvent passes through the material.
This will lead to a measurable loss in
strength and flexibility of the elas-
tomer or polymer. A sampling of rec-
ommended elastomers and polymers
is provided in Tables 3 to 6.
Okay, What Have
We Learned?
We have learned that there are basi-
cally two types of ethanol fuels. Low-
ethanol fuel blends like E10 that have
been in use since the late 1970s, and
high-ethanol fuel blends like E85 that
have been in use since the 1990s.
There are four major areas of concern
for low- and high-ethanol fuel
blends—phase separation, solvency,
metal corrosion, and permeation of
elastomers and polymers. Low-
ethanol fuel blends have different
material and handling recommenda-
tions compared with high-ethanol
fuel blends.
Low-ethanol fuel blends are sen-
sitive to water and temperature. High
dissolved water content and/or a
TABLES
Elastomer Compatibility Issues
in E10 Fuels (5,14,15)
RECOMMENDED
1 Acrylonitrile (hoses
& gaskets)
1 Fluorocarbons
1 Fluorosilicone
Natural rubber
1 Polychloroprene
(hoses & gaskets)
Polysulfide rubber
NOT RECOMMENDED
• Acrylonitrile (seals)
• Polychloroprene
(seals)
• Urethane rubber
• Zinc
TABLE 4
Elastomer Compatibility Issues
in E85 Fuels (5,14,15)
RECOMMENDED
• Acrylonitrile,
• Nitrile rubbers
• Polychloroprene
• Polytetrafluoroeth-
ylene
• Fluorocarbon
NOT RECOMMENDED
• Natural rubber
• Cork gasket
material
• Leather
TABLE 5
Polymer Compatibility for
E10 Fuels (5,14,15)
RECOMMENDED
Acetyl
1 Polyamides
Polypropylene
Polytetrafluoro-
ethylene
Fiberglass-rein-
forced plastic
NOT RECOMMENDED
Polyurethane
1 Alcohol-based
pipe dope
TABLE 6
Polymer Compatibility
for E85 Fuels (5,14,15)
RECOMMENDED
Polyamide
Polyethylene
Polypropylene
NOT RECOMMENDED
1 Alcohol-based pipe
dope
1 Methyl-methacry-
late
1 Polyamide
1 Polyester bonded
fiberglass lami-
nates
1 Polyurethane
Polyvinyl chloride
sudden drop in temperature can
result in phase separation of ethanol
and water in the fuel tank. Phase-sep-
arated fuels that enter vehicle fuel
systems can result in operational
problems.
• continued on page 6
-------
• E10 and E85 from page 5
Low E-blends can contain up to
10 percent ethanol. If a fuel system
that previously dispensed gasoline or
diesel fuel has not been properly
cleaned prior to conversion, the
ethanol will clean or dissolve a vari-
ety of materials found on the walls of
USTs, piping, dispenser piping,
hoses, and nozzles. The suspended or
dissolved material can enter the bulk-
fuel stream and could create drive-
ability problems for vehicles.
Most of the available literature
indicates that UST systems should
fare well with low E-blend fuels and
not experience metal corrosion. How-
ever, this may not be the case for cer-
tain elastomers and polymers.
Current data suggest that low E-
blend fuels have a higher activity or
gradient than E85 fuels and therefore
a greater potential for permeation,
swelling, and possible performance
degradation of the elastomer or poly-
mer.
E85 presents other operational
and compatibility considerations.
Early work with E85 and M85 test
fuels demonstrated that methanol
was a more aggressive alcohol than
ethanol and the source of many mate-
rial compatibility issues. As a result,
research and development shifted
away from M85 and refocused on
E85. This is not to suggest that E85 is
immune from material compatibility
issues but rather to suggest that cer-
tain material compatibility issues
may take place over a longer period
of time.
Because E85 can absorb a greater
volume of water and ionic contami-
nants, E85 becomes a good electrical
conductor. Metals that are in constant
contact with E85 fuel should be of the
same or similar material and more
stable on the galvanic series. If soft or
anodic metals such as aluminum or
brass must be used, then they should
be nickel plated if they are to be in
continuous contact with E85 fuels. It
is also possible that certain technolo-
gies that rely on the conductive or
capacitance properties of gasoline
may experience degradation in per-
formance accuracy and precision due
to the change in electrical properties
of the E85 fuel.
E-85 fuels have a high solvent
capability. Any retail station owner
considering converting a used UST
over for E85 use must have the UST
and associated piping thoroughly
cleaned, dewatered, and carefully
inspected by a bonded and insured
company. Data and industry experi-
"There are tour major areas of
concern for low- and high-ethanol
fuel blends—phase separation,
solvency, metal corrosion, and
permeation of elastomers and
polymers. Low-ethanol fuel blends
have different material and handling
recommendations compared with
high-ethanol fuel blends."
ence suggest that E85 will dissolve
material from the walls of the tank,
piping, and dispenser.
Because E85 has greater solvency
and increases the conductivity of the
liquid, dissimilar metal piping, syn-
thetic materials, nonmetallic materi-
als, and nozzles made of soft metals
could dissolve, degrade, and corrode,
creating further problems for UST
facility owners and their clients. It is
strongly recommended that the
owner convert equipment over to
"approved E85 equipment."
Degradation of nonmetallic
materials, such as elastomers, poly-
mers, and natural materials can occur
when in constant contact with E85
fuel. Data also indicate the maximum
swell of some polymers occurs at
approximately 15 percent ethanol
and diminishes as the ethanol con-
centration approaches 85 percent.
Permeation of a nonmetal can cause
swelling, plasticization, degradation,
and loss of dynamic and static func-
tions. •
Fdivnrd W En^lisli 11 ;s Vice Pie^iilent
and Technical Director for Fid'/ Qunl-
iti/ Services, Inc. He cnn be reached In/
phone nt (770) 967-9790 or in/ email nt
ecnglish@fqsinc.com.
Ethanol Fuel Resources
Web Links:
• Alternative Fuels Data Center
http Hwww eere energy gov/afdc/altfuel/altfuels.html
• National Ethanol Vehicle Coalition :
http-//ivww.e85fuel.com/mdexphp
• Renewable Fuels Association (RFA)
http Himxrw ethmwlrfa.org/
Suggested Reading:
• New England Interstate Water Pollution Control
Commission, Vol 3, Chapter 5, July 2001, Health,
Environment, and Economic Impacts of Adding Ethanol
to Gasoline in the Northeast States.
• SAE Technical Paper Service, 2005-01-3708, Effects
of Alcohol in Fuel-Line Material ofGasohol Vehicles
• Matthew F Mynth, February 2003, technical white
paper, A Comparison of Fuel and Oil Resistant Proper-
ties of Polymers.
References
1. Fuels and Transportation Division Meridian Cor-
poration (FTDMC), July 1991, Properties of Alcohol
Transportation Fuels, Alcohol Fuels Reference Work
#1
2. Rural Enterprise and Alternative Agricultural
Development Initiative (REAADI) Report #12.,
June 2002, History of Ethanol Production
3. Energy Information Administration (EIA),
Ethanol Energy Timeline, Milestones in Ethanol
Energy History
4. Halvorsen, Ken, C , December 1998, The Necessary
Components of a Dedicated Ethanol Vehicle
5 United States Department of Energy, Guidebook
For Handling, Storage, and Dispensing Fuel Ethanol
6. American Automotive Manufacturers Associa-
tion, August 1995, Fuel Ethanol Compatibility Stan-
dards and Dispensing Equipment List for E85 Fueled
Vehicles
7 American Automotive Manufacturers Associa-
tion, February 1996, Fuel Methanol Compatibility
Standards and Dispensing Equipment List for M85
Fueled Vehicles
8. Westbrook, Paul A , Ph D , January 1999, Compat-
ibility and Permeability of Oxygenated Fuels to Mate-
rials in Underground Storage and Dispensing
Equipment
9. Roggehn, Ernest M , LUSTLme Bulletin #47, June
2004, Pipes and Sumps—As I See Them.
10 Correspondence August 31,2005, California State
Water Resources Control Board, Advisory
Regarding Thermoplastic Flexible Piping
11 Steel Tank Institute, March 2004, Keeping Water
Out Of Your Storage System
12. Westbrook, Paul A., Ph.D., January 1999, Oxy-
genate Compatibility and Permeability Report, UST
Team 1 Report
13 CRC Project No E-65, September 2004, Fuel Per-
meation From Automotive Systems, Final Report
14 Renewable Fuel Association Publication #96051,
December 2003, fuel Ethanol, Industry Guideline,
Specification, and Procedure
15. United States of American Department of
Energy, April 2002, Handbook for Handling Storing
and Dispensing E85
-------
Is Your UST System Ethanol Compatible?
A Regulator's Perspective
byJeffKuhn
Legislation in Montana and
across the United States ban-
ning the oxygenate compound
MtBE has greatly increased the use of
ethanol, another oxygenate slated to
be the frontrunner to take the place of
MtBE. Currently, twenty-six states
and Washoe County, Nevada, have
MtBE bans in place or enacted; six
states considered bans in 2005. Oxy-
genates are a necessary part of gaso-
line formulation and create the
octane ratings for various grades of
fuel used in internal combustion
engines.
Meanwhile, the fuel efficiency
benefits and rising popularity of
hybrid and flexible-fuel vehicles are
also generating a great deal of inter-
est in the use of ethanol, considered a
more environmentally friendly and
nationally acceptable alternative that
may help loosen U.S. reliance on
Middle East oil. According to the
National Ethanol Coalition (http://
www.ethanol.org/produciion.html) there
are currently 94 ethanol production
facilities in the U.S. and 31 more
under construction.
Most folks in Montana seem to
view the move toward ethanol as a
step in the right direction—less
dependence on foreign crude, and a
potential boon to Montana's agricul-
tural community that might provide
the feedstock for future ethanol
plants in the state. And given the
average driving distances here, (it's
not unusual to drive 4 hours in one
direction and still be well within the
state), most Montanans are painfully
aware of rising fuel costs.
However, despite the benefits of
ethanol, there are potential UST-sys-
tem compatibility issues that need to
be considered by owners and opera-
tors. A number of state websites pro-
vide information resources and fact
sheets to assist UST owners and oper-
ators converting to gasoline-ethanol
blends (E-blends), particularly E10.
During my review of information
gleaned from internet sources (e.g.,
industry literature, state websites,
and published research), I noticed a
recurring theme summed up by the
following statement from the Iowa
Department of Natural Resources:
"Without converting to compatible
equipment, your UST system could
degrade, and a product release could
occur. Ultimately, the equipment and
components must be compatible with
the percentage volume of ethanol-
blend you intend to use."
Many of these websites go so far
as to strongly recommend or require
that equipment used to dispense E-
blends be certified by the manufac-
turer or that the owner/operator sign
a "statement of compatibility," verify-
ing that the equipment is compatible
with E-blends. State and federal rules
require that all components and
equipment used for storing and dis-
pensing motor fuels be compatible
with the product stored. Owners and
operators of UST systems in states
using E-blends need to be aware of
potential compatibility problems and
plan to replace equipment reported to
be prone to deterioration from E-
blends.
Susceptible Components
So what materials are potentially at
risk or prone to deterioration from
contact with ethanol? Soft metals,
including brass, aluminum, and zinc,
commonly found in fuel-storage dis-
pensing systems are not compatible
with ethanol, especially at the higher
concentrations found in E-85 motor
fuel (Wisconsin DNR). Some non-
metallic materials, such as natural
rubber, polyurethane, adhesives
(used in older fiberglass piping), cer-
tain elastomers, polymers used in
flex piping, bushings, gaskets,
meters, filters, and materials made of
cork, are prone to deterioration from
ethanol blends over 10 percent by
volume (Office of the Illinois State
Fire Marshal). Copper and plastic in
air-eliminator floats may also not be
compatible with ethanol.
For detailed information regard-
ing specific storage and dispensing
equipment for E-blends, see the New
England Interstate Water Pollution
Control Commission's (NEIWPCC,
2001) Health, Environmental, and Eco-
nomic
Impacts
of Adding
Ethanol
to Gasoline in the Northeast States, July
2001, pp. 70-71 at http://www.neiwpcc.
org/PDF_Docs/ ethvol3.pdf. Also, a list
of E-85 compatible equipment can be
found at http://www.e85fuel.com/infor-
mation/manufacturers.htm. This docu-
ment provides a detailed and
well-researched chapter on the stor-
age and handling of E-blended fuels,
with discussion regarding compati-
bility with specific UST-system com-
ponents.
The California State Water
Resources Control Board (SWRCB)
strongly recommends that UST own-
ers and operators request a written
compatibility statement from respec-
tive equipment manufacturers before
storing E-blends on site. Their web-
site (http://www.swrcb.ca.gov/cwphome/
ust/leakjprevention/ethanol/ethanol.htm}
provides a reference list of equip-
ment manufacturers to contact for
more information. SWRCB also lists
potential compatibility problems
with the following UST-system com-
ponents:
• Single-walled fiberglass tanks
installed prior to 1/1/1984
• Single-walled fiberglass and flex-
ible piping installed prior to
1/1/1984
• Lining material used to line old
single-walled tanks for repairs or
upgrade
• Adhesives, glues, sealants, and
gaskets used around the piping
and other parts of the UST sys-
tem (more of a concern for older
systems, but may be an issue for
new installations if the contractor
failed to use proper material)
• Pump heads and other auxiliary
equipment, including certain
metals (e.g., aluminum, brasses/
bronzes) that come in contact
with the product
• Older models of some leak-detec-
tion equipment that may not
• continued on page 8
-------
• Ethanol Compatible? from page 9
operate properly or with parts
that may wear out with exposure
to E-blend fuels.
The SWRCB advises that if any of
these components are present at a
site, the owner/operator should con-
tact the equipment manufacturer and
installer to determine whether they
are compatible with E-blends.
Affinity for Water
Ethanol, also known as ethyl alcohol,
has a strong affinity for water (David
Korotney, EPA, undated memo). This
propensity to absorb water makes it
even more important that water accu-
mulation is carefully monitored and
that water is routinely removed from
tank bottoms. Absorbed water in fuel
reduces motor fuel BTU and octane
rating and can lead to phase separa-
tion—allowing the alcohol to drop
out of the gasohol and form a layer of
gasoline on top and a layer of ethanol
on the bottom. This phase-separated
alcohol/water bottom encourages
the growth of aerobic bacteria, which
can be detrimental to petroleum fuels
and certain fuel-handling compo-
nents. Phase separation can also be a
problem for vehicle-fuel and ignition-
system components when fuel conta-
minated with water is distributed.
Degradation and
Accelerated Corrosion
Steel UST systems may be adversely
impacted by ethanol due to acceler-
ated corrosion caused by scouring or
loosening of deposits in tanks and
distribution lines. If a corrosion cell
already exists, ethanol can increase
the effect of scoured exposed steel
surfaces and eventually cause a per-
foration of the steel. As mentioned,
ethanol can corrode soft metals such
as zinc, brass, copper, lead, and alu-
minum. These dissolved metals in
the fuel can, in turn, contaminate a
motor vehicle's fuel system.
Conductivity and ATG Probes
Capacitance probes may not work in
E-blend fuels due to the higher
conductivity of ethanol. Owner/
operators should verify that magne-
tostrictive probes are alcohol compat-
ible and that the automatic tank
gauge (ATG) system is properly cali-
brated for E-blends.
8
Tank Linings
Older tank linings seem to pose a
specific concern for the storage of E-
blends because of the incompatibility
of ethanol with epoxy-based tank lin-
ings. "Older epoxy linings used to
line steel USTs, both inside and out,
have been found to soften when
exposed to E-blend (10 percent
ethanol by volume)" (NEIWPCC
2001, page 64. Source: Downstream
Alternatives, Inc., 2000; Archer
Daniels Midland Co., 2000). The
American Petroleum Institute (API,
1985) also found that general purpose
tank linings softened when exposed
to ethanol vapors. Newer formula-
tions of UST lining material may be
compatible with ethanol.
Converting Existing Storage
and Dispensing for E-Blends
The following state websites provide
specific recommendations for owners
and operators who are converting
existing fuel storage systems for E-
blends.
• Iowa
The Iowa Department of Natural
Resources (DNR) has an online
checklist (http://www.iowadnr.com/lan/
ust/technicalresources/ethanol.html) for
upgrading UST systems for compati-
bility with ethanol blends greater
than 10 percent ethanol by volume.
Dispensers must bear the UL mark or
be certified as compatible with the
product stored and dispensed.
Because there are currently no E-
blend-compatible dispensers avail-
able with a UL listing mark, Iowa
allows incompatible dispensers a
two-year phase-in for E-blend use.
However, shear valves or emergency
valves on existing and new UST sys-
tems must be compatible with E-
blend fuel. UST systems installed
after August 1, 2005, must use avail-
able compatible equipment at the dis-
penser if E-85 is stored and
dispensed. The final phase-in
for ethanol-compatible dispensing
equipment in Iowa is July 1, 2007.
Incompatible dispensers may not be
used after that date.
During the phase-in, dispensers
not certified by the manufacturer or
UL marked as compatible for E-
blends must be checked daily for
leaks and equipment failure. Daily
inspections must be completed for
non-compatible dispensers and
visual observations recorded on a
form provided by the DNR. Any
incompatible component that leaks
or does not operate as designed must
be removed and replaced with E-
blend-compatible components. The
DNR must be notified immediately
of any failed component.
• Wisconsin
The Wisconsin Department of
Commerce (WDC) has prepared an
excellent brochure that summarizes
steps that should be taken to prepare
UST systems for conversion to E-
bler\ds:http://commerce.wi.gov/ERpdf/
bst/ProgramLetters_PL/ER-BST-PL-
PreparingForEthanolBrochure.pdf. The
website notes that "the first-time
transition to blends of up to 10 per-
cent ethanol should not be assumed
to be trouble free" and lists a series of
assessments and procedures for own-
ers and operators to follow.
One of the most notable state-
ments on the site is that "no level of
water is acceptable for ethanol-
blended fuel due to the phase-separa-
tion problems." WDC tells owners
and operators to make certain that
"all fittings and connections at the
top of the tank are tight (no vapors
escape and no water enters) and that
all sump and spill-containment cov-
ers prevent water from entering. Any
water intrusion problems must be
corrected."
The brochure cautions tank own-
ers to "clean any tank used to store
ethanol to remove all sludge from the
bottom of the tank. Any sludge or
particulate in the bottom of the tank
will be suspended in the ethanol and
cause problems with filters and fuel
lines."
• Illinois
The Office of the Illinois State
Fire Marshal requires that owners
and operators sign a Statement of
Compatibility form, which can be
found on their website, certifying
that UST systems under their control
that store E-85 are compatible with E-
blends. Owners and operators must
also submit a Notification Form for
Underground Storage Tanks to the
Office of the State Fire Marshal indi-
cating the change of product that will
be stored in the UST. The form must
be signed by the UST-system owner.
(The form may be downloaded from
the OSFM website at wivw.state.
-------
il.us/osfm and then following prompts
to "Division of Petroleum and Chem-
ical Safety" and then to "Download
Applications" and then choose "UST
Notification Form.")
The Future
Most people in the petroleum indus-
try are keenly aware of the ongoing
changes in fuel formulation, due in
part to advances in technology in
petroleum refining, chemical engi-
neering, automobile manufacturing,
and energy conservation. Some of
these changes are also due to recogni-
tion of oxygenate impacts at LUST
sites and the "boutique fuels" with
geographic-specific fuel formulations
to address the needs and require-
ments of different areas of the
country.
As with all advances in technol-
ogy, the benefits of the technology
should outweigh its risks. The
increased use of ethanol means the
decreased use of MtBE and other
similar compounds that, due to their
unique chemistry, have the ability to
travel farther and faster in ground-
water and pose a greater threat to
drinking water supplies.
However, even the best techno-
logical advances have a retooling cost
—both at the refinery level and at
marketing and retailing levels.
Ethanol typically requires bulk trans-
portation and on-site splash blending
at fueling terminals to avoid its
absorption of water in pipelines. Spe-
cial handling practices and precau-
tions must be taken.
Such considerations were a very
significant part of discussion that led
up to the phase-out of MtBE in Cali-
fornia and the phase-in of ethanol in
gasoline. (See Health & Environmental
Assessment of the Use of Ethanol as a
Fuel Oxygenate, California Air
Resources Board, the State Water
Resources Control Board, and the
California Environmental Protection
Agency's Office of Environmental
Health Hazard Assessment, 1999—
http://www-erd.llnl.gov/ethanol/ eto-
hdoc/index.html.)
Another important question is
whether state UST inspectors, during
the transition to E-blends, will even
look for potential compatibility prob-
lems by carefully inspecting specific
components that are prone to degra-
dation from ethanol. Perhaps this
will require additional training or a
specific module in state training pro-
grams (classroom and web-driven)
that focuses on key compatibility
concerns identified by the petroleum-
equipment industry and other infor-
mation sources.
The advent of E-blend fuels in
Montana and other states has been a
longtime coming. The Montana
Department of Environmental Qual-
ity views ethanol as a favorable alter-
native to MtBE and a step in the right
direction toward environmental
stewardship and energy sustainabil-
ity. At this time, a limited number of
Montana distributors provide E-85 or
other E-blends for consumers.
However, as more UST owners
and operators in Montana, and
throughout the nation, consider stor-
ing and distributing E-blends, they
need to plan accordingly and verify
that their fuel storage systems are
compatible with ethanol. Otherwise
we will only succeed in creating
another means for petroleum from
fuel-storage systems to be released
into the environment, thus carrying
on the long legacy of groundwater
contamination that many of us have
spent our careers trying to rectify. •
References
• National Ethanol Coalition,
http //www.ethanol org/productwn.html
• New England Interstate Pollution Control Commis-
sion (NEIWPCC) Health and Environmental Impacts
of Adding Ethanol to Gasoline in the Northeast States,
July 2001, pp 70-71.
http //www.neiwpcc org/PDF_Docs/ethvol3.pdf
• Office of the Illinois Stale Fire Marshal
www state il us/osfhi
• Iowa Department of Natural Resources
http //www iowadnr.com/lan/iist/technicnlresources/eth
anol html
• California EPA SWRCB's advisory to UST Owners
and Operators Regarding EthanoI-BIended Fuel
Compatibility
http //www swrcb ca gov/cwphome/ustfleakjpreven-
twn/ethanol/ethanol.htm
• California Environmental Protection Agency State
Water Resources Control Board
http //www swrcb ca gov/cwphome/ust/leak_preven-
tion/ethanol/ethanol htm
• Wisconsin Department of Commerce "Preparing
For Ethanol Brochure"
http.//cotnmerce.wigov/ERpdf/bst/ProgramLetters_PL/
ER-BST-PL-PrepanngForEthanolBrochurepdf
• Health & Environmental Assessment of the Use of
Ethanol as a Fuel Oxygenate, California Air
Resources Board, the State Water Resources Con-
trol Board (SWRCB), and the California Environ-
mental Protection Agency's Office of
Environmental Health Hazard Assessment
(CalEPA/OEHHA, 1999).
http //wwwerd Hnl gov/ethanol/etohdoc/mdex html
• Water Phase Separation in Oxygenated Gasoline by
David Korotney, EPA, Chemical Engineer, Fuels
Studies and Standards Branch
http //www epa gov/otaq/rcgs/fuels/rfg/waterphs pdf
Recommended Practices and Codes:
• American Petroleum Institute (April 1985) Storing
and Handling Ethanol and Gasolme-Ethanol Blends at
Distribution Terminals and Service. Stations. API Rec-
ommended Practice 1626
• National Fire Protection Association (NFPA) 30,
Flammable and Combustible Liquids Code, 2000
Edition; NFPA 30A Code for Motor fuel Dispensing
facilities and Repair Garages
• Renewable Fuels Association, Fuel Ethanol Indus-
try Guidelines, Specifications and Procedures, RFA
Publication # 960501, Revised December 2003
http. //www ethanolrfa org/final960501 pdf
' US Department of Energy, National Renewable
Energy Laboratory: Handbook for Handling, Stor-
ing, and Dispensing E85, April 2002
http //www eere.energy gov/bwmass/pdfs/30849 pdf
Good-bye Sammy an
Sammy Ng, U.S. EPA Office of Under-
ground Storage Tanks (OUST) Deputy I
Office Director, says good-bye to state
and federal UST/LUST program friends
at the 18th Annual National Tanks
Conference in March in Memphis, |
Tennessee. Sammy is retiring from EPA
and says he plans to look for America...
um...in an old VW van? He joined the |
OUST staff in 1985 as one of the i
original members of the UST/LUST !
program team and served as Acting
OUST Director from 1999 to 2000. i
Have fun, Sammy. We'll miss you.
-------
LLISTLinc Ritlk'tin 52 • Mai/20tlb
Ethanol? Good. Yes? Um...
It started as an ordinary day, but rapidly degraded into an antacid kind of day. Why?
Because ethanol was going to be replacing MtBE as a gasoline oxygenate a lot faster
than any of us expected. Here's the story.
The Valero Energy Corporation
recently purchased the Dela-
ware City Refinery (and that's
fine and dandy). Because the refinery
is within the coastal zone and is sub-
ject to the requirements of the federal
Coastal Zone Act, any new activities
that take place within the facility
must be permitted. So a request was
circulated to various groups within
the Department of Natural Resources
and Environmental Control
(DNREC) to provide comments
about the refinery's request to change
its gasoline oxygenate from MtBE to
ethanol, which requires some piping
changes and modifications to at least
one large aboveground storage tank
(AST), internally lining the AST, seal-
ing roof joints, and installing a jet
mixer.
The ethanol is to be shipped in to
the terminal by barge or tanker,
transferred to storage via an existing
pipeline, and stored in an AST. Prior
to storage, the ethanol will be
blended with 2 percent gasoline as a
denaturant. Additional modifications
will be made to transfer the ethanol
to the truck-loading lanes, where it
will be blended with gasoline before
leaving the terminal.
I was thrilled to hear this
because, after all, Delaware has not
yet managed to ban MtBE. Last year,
our assembly passed a bill to ban
MtBE , which was to become effective
in 2008, but the bill is now "languish-
ing" in a senate committee. The sen-
10
ate passed a bill to form a task force
to study the issue of an MtBE ban. It
was referred to a house committee,
where it now sits. The bills were car-
ried over into the current legislative
session, but they may never get out of
committee. Since the refinery in ques-
tion supplies the majority of gasoline
used in Delaware, we'd at least be
getting rid of a large chunk of poten-
tial MtBE problems, whether or not a
ban bill is passed.
Whoa, Nellie!
As vocal as I've been about the evils
of MtBE over the past years, you'd
think that I'd be thrilled to hear that
our refinery wants to stop adding
MtBE to our gasoline. But the kicker
in the application was the statement:
"As part of its clean fuels strategy for
the Delaware City Refinery, Valero
will no longer use MtBE as a gaso-
line-blending component effective
May I, 2006." Whoa...that's soon!
That's lots sooner than the bill that
proposed banning MtBE in 2008!
Most of the other states that are
using 10 percent ethanol by volume
(E10) in their gasoline seem reason-
ably happy. Either they've been
using it for years because they're
Corn Belt states, or they're using it
because they banned MtBE and have
had a few years to prepare. So far as I
can tell there were no major problems
with switching over. Clean out the
tank ahead of the switch, change fil-
ters more frequently, at least for a
while, and be diligent.
water out of the tank..
Federal regulatil
and operators must
tern that is made of or lineJ
materials that are compatible^
the substance stored in the Uf
tern." Paul Miller of the U.
Office of Underground
Tanks says "to date, EPA has!
on UL testing to help in the compati-
bility determination."
Our regulations state that all
equipment used as part of a tank sys-
tem must be compatible with the sub-
stance stored, as do those of about ten
other states that I checked with,
where I know ethanol is in use. A
quick scan of our database last sum-
mer showed me that we had 121
fiberglass tanks installed before Janu-
ary 1,1984, only one of which is dou-
ble-walled.
California advised their tank
owners that pre-1984 tanks may not
be ethanol compatible. We had 306
fiberglass tanks with gasoline
installed after that date, 133 of which
were single-walled. We haven't run
queries yet on tank piping. What do
we currently have in the ground for
flexible piping and rigid fiberglass
piping, and when were they
installed? Are they certified as being
E10 compatible? Dates for E10 com-
patibility for tanks and lines vary by
years among different manufacturers.
Calls to some of the manufactur-
ers concerning UL approval of some
of the equipment have resulted in
answers such as "Well, it isn't offi-
cially UL approved for use with E10,
but it's probably okay to use it."
"Probably okay" doesn't cut it for
concurrence with the regulations.
-------
How many of our roughly 350 gaso-
line stations can prove that all their
equipment is approved for use with
E10?
Where are the compatibility con-
cerns? Our regulations state that
"The material used in the construc-
tion and/or lining of the UST system
must be compatible with the product
stored." The New England Interstate
Water Pollution Control Commission
published a report in 2001 entitled
Health, Environmental, and Economic
Impacts of Adding Ethanol to Gasoline in
the Northeast States. Volume 3, Chap-
ter 5 deals with ethanol storage and
handling, and provides an excellent
summary of the parts of tank systems
where there may be compatibility
issues. Besides tanks and lines, there
may be compatibility issues with
leak-detection devices, dispensers,
pumps, and almost every part of an
UST system. I'll leave you to read
that excellent summary at http://
www.neiwpcc.org/PDFJDocs/ethvol3.
pdf.
The Heartburn Thickens
So, since a large percentage of the
gasoline sold in Delaware comes from
the Delaware City Refinery, we could
all be driving with ethanol in the very
near future. If a retailer buys gasoline
from that refinery, he'll be getting
E10. If he can't prove that all his
equipment is approved for use with
ethanol, he's in violation of our regu-
lations. He might also be voiding the
warranty on parts of his tank system.
Now supposing there is a release.
Will his insurance company pay for
investigation and remediation if the
equipment warranties have been
voided and he's potentially in viola-
tion of our regulations? I certainly
doubt it! Are other states as con-
cerned about compatibility issues?
New York and Connecticut evidently
recommended that owners clean
their tanks thoroughly before switch-
ing, and check filters frequently. And
what about proving compatibility? I
checked around to find out where
states are with regard to E10 and E85.
Here's what I've learned so far.
• California
In March 2000, California issued
an advisory to tank owners/opera-
tors regarding E-blend fuel compati-
bility. At that time, some parts of the
state were already using E10. The
advisory stated that if the UST sys-
tem was not compatible with this
fuel, there would be a higher risk of
damage to the UST system and the
environment. It urged "that you ver-
ify that your entire system is compat-
ible with the ethanol-blend fuel
before you store it in your UST sys-
tem," and that compatibility informa-
tion may be available from your
equipment manufacturer. As far as I
can tell, this advisory does not
require that you prove or certify that
your entire UST system is compati-
ble. (http://www.swrcb.ca.gov/ust/leak_
prevention/ethanol/ethanol.htm)
• Illinois
Some states seem to be con-
cerned with compatibility issues
when you convert from E10 to E85. A
memo from the Office of the Illinois
State Fire Marshall states that "com-
ponents and equipment used for stor-
ing/dispensing conventional fuels
are time tested for compatibility and
readily available through your petro-
leum supplier. High-percent ethanol,
however, does not have the same
compatibility characteristics of con-
ventional fuels when it comes to stor-
age and dispensing...In order to
store and dispense high-percent
ethanol, fiberglass and steel UST sys-
tems must be listed by Underwriters
Laboratories, Inc. [UL] or certified by
the manufacturer." Illinois says that
if you store E85, you must certify that
you have researched all of the vari-
ous components of your tank system
and certify that they are E85 compati-
ble. (www.state.il.us/osfm/PetroChem-
Saf/Home.htm) I didn't find any
information on their website about
compatibility with E10 blends; this
seems to be a nonissue for Illinois.
• Iowa
Iowa is allowing existing facili-
ties a two-year phase-in period to
upgrade so they are compatible with
E85. This allows retailers to begin
selling E85, but it requires that dis-
pensers and components not certified
as compatible be upgraded within
the two-year period. I guess that this
gets you to replace any parts that
may be starting to get squishy before
they completely fail. (http://www.
iowadnr.com/land/ust/technicalresources/
ethanol.html) Again, I don't know
whether E10 is a non-issue. Both Illi-
nois and Iowa have been using E10
for years.
• New Hampshire
Lynn Woodard, New Hampshire
Department of Environmental Ser-
vices (DBS), realized that ethanol was
coming to the nearest gas station
much sooner than later and spent
considerable time and effort
researching what tank owners
needed to know about an ethanol
changeover. The DES has posted an
ethanol fact sheet for tank owners,
Qs&As, and a Power Point presenta-
tion.on its website at http://
des.nh.govf/mtbetrans.html.
Woodard says tank owner/oper-
ators of facilities changing over to
ethanol need to be aware of the
importance of getting rid of their
existing water-finding paste. They
need to ask their fuel supplier for a
paste designed to detect water in an
ethanol-blend gasoline. Further, they
need to gauge the tank at its lowest
end, allow at least one minute for the
reaction to occur on the paste, and
check the directions to determine
what each resulting color means (i.e.,
is water present or has phase separa-
tion occurred?).
"We also encourage tank owner/
operators to install alcohol-sorbing
filters on the dispensers," says
Woodard. "These water-sensitive fil-
ters absorb water present in the tank.
When the filter's capacity is
exceeded, it slows the flow of gaso-
line to a very low rate or stops it
entirely. The filter then has to be
removed and replaced."
Woodard emphasizes the impor-
tance of making sure the tank is
cleaned of all water and all pathways
for water intrusion are elimineited
prior to receiving the initial drop of
an ethanol-blend gasoline. "Failure to
do so can result in economic disaster
for the tank owner," says Woodard.
If sufficient water remains in the
tank—0.2 to 0.5 percent by volume—
phase separation can occur. Once this
happens, approximately 40 to 60 per-
cent of the ethanol will have
migrated out of the gasoline to the
water. The ethanol-water level will
more than likely exceed the with-
drawal level in the tank, resulting in
disruption in the ability to pump
product or in a worst-case scenario (if
a water-sorbing filter is not in place)
dispensing a water-ethanol solution
to customer vehicles. Repair of the
vehicles can be costly.
• continued on page 12
_
-------
• WanderLUST/rowi page 11
"The economic problems don't
stop there," explains Woodard. "The
octane of the remaining gasoline in
the tank will have diminished sub-
stantially, and the gasoline won't be
suitable for sales. Unless it can be
shipped back to the refinery, it will
have to be disposed of as a hazardous
waste at a rate of two to three dollars
per gallon. This could be the financial
straw that breaks the tank owner's
back."
• Wisconsin
Sheldon Schall of the Wisconsin
Department of Commerce (DOC)
told me that they have stations sell-
ing E10, E20, and E85, and that they
expected more problems than they've
experienced. They did try to mandate
ethanol-compatible dispensers, but
the retrofit is $7,000, and the ethanol
and corn grower's lobby painted
them as "ethanol unfriendly," so they
had to back off. The decision to allow
dispensers that were not UL-listed
for E85 was based on common sense,
experience, and professional judg-
ment, and not the thought of being
painted as "ethanol unfriendly."
Manufacturer Gilbarco told
Schall that they use the same dis-
pensers for ethanol in South America
(Argentina and Brazil)—where cars
may be fueled with anything up to
100 percent ethanol—that we use for
gasoline in the United States. Also,
Minnesota had been using higher-
percentage ethanol for years with no
particular problems.
Wisconsin has had problems
when tank owners didn't clean tanks
before switching to gas with ethanol.
Schall described a fuel-system
episode last spring in tanks that had
not previously held E10. The operator
ran a few tanks that were low on
product, refilling each time with E10,
but still had a problem with the
ethanol loosening buildup on the tank
wall. Within hours DOC had bad gas
calls from owners of stalled cars.
Wisconsin guidance states that
E10 is a normal component of today's
automotive fuel, and it is accurate to
refer to it as "gasoline." Any level
above 10 percent ethanol cannot be
labeled as gasoline. DOC requires
that a form be submitted when there
are plans for a new installation or
plans to convert an existing system
from conventional motor fuels to
12
blends greater than 10 percent
ethanol. The form requires that the
petroleum equipment contractor or a
professional engineer verify that the
system is E85 compatible either by
UL listing or by the component man-
ufacturer. No special certification is
required if the tank system will be
storing gasoline with 10 percent or
less ethanol.
• New York
Certain areas of New York State
have been using E10 since MtBE was
banned on January 1, 2004. Russ
Brauksieck of the New York Depart-
ment of Environmental Conservation
(DEC) says that "generally speaking,
if the owner/operator prepared for
the switch from gasoline with MtBE
to E10 (by cleaning the tank of water
and precipitated materials) there did
not seem to be much of a problem.
"However," cautions Braukseick,
"some operators of nonretail gasoline
USTs (typically municipalities) did
less to prepare for the introduction of
E10 and had some issues with fuel
quality. This has led to maintenance
issues requiring filters to be changed
more frequently. These problems
seem to have been corrected over
time.
"While DEC has not observed
any compatibility issues with tank
systems storing E10 to date," says
Braukseick, "we are still concerned
with long-term exposures of certain
storage tank systems to ethanol. In
particular, it has not been demon-
strated that fiberglass tanks and pip-
ing manufactured in the early 1980s
have long-term compatibility with
ethanol. This has the potential for
weakening the tanks or piping over
time such that a failure is possible."
One New York regulator recently
visited a facility where the owner
didn't clean his tanks when he began
using E10 two years ago. Since then,
he has had to change his filters at
least weekly. He still has sludge in
his tanks and it has cost him as much
in filters as it would have cost to
clean out the tanks. He is planning to
take his tanks out of service to clean
them out in the very near future.
What about Commingling
MtBE and Ethanol?
Delaware gets gasoline from several
refineries in the Philadelphia area,
just across our northern border, and
from refineries in New Jersey, jusl
across the river. Many gasoline sta-
tions receive all their gasoline from
the same refinery, but some buy on
the spot market, from whoever has
the best price. The thing is, gasoline
with MtBE and gasoline with ethano]
should not be mixed. We haven't
banned MtBE in Delaware, so it's
perfectly legal to have it in gasoline
sold in the state. But it shouldn't be
hauled in the same tankers without
cleaning, and it shouldn't be mixed in
the same UST or in your car's gas
tank, because commingling causes a
vapor-pressure increase. How do we
prevent this?
We probably had MtBE in our
gasoline for 20 years before we knew
about it, so we didn't require anyone
to certify that the equipment that
went into the ground was approved
for use with gasoline containing
MtBE. This time, we're going into it
knowing that we're going to have a
change in gasoline composition and
that there may be compatibility
issues. A strict reading of the regula-
tions out there seems to mean that we
shouldn't allow equipment to be
used that isn't officially declared as
compatible.
Okay, Check Your Equipment
and Clean Your Tanks
Since I began working on this article,
we've done an about-face, or a
reevaluation of our position. Even if
we're able to get firm dates for certifi-
cation of E10 compatibility by manu-
facturers, much of the time we don't
know exactly what is in the ground.
For older sites, we don't have instal-
lation plans showing details about
the equipment, and for newer sites,
we may have names of manufactur-
ers and installation dates, but not the
manufactured date. Many gasoline
stations have changed ownership fre-
quently, and I know that the paper-
work and warranties are not always
provided to the new owner. If we
stuck to a strict interpretation of our
regulations, we'd probably have to
shut down large numbers of stations
in the state, and we'd all be walking
or biking to work.
So, we've gone the path of many
of the other states. Letters went out to
the tank owners and operators sug-
gesting that they check that their
equipment is compatible with E10,
and giving suggestions as to how to
-------
prepare for delivery of E10. Press
releases went out with similar infor-
mation, resulting in articles in several
Delaware newspapers.
Some stations will be ready for
ethanol. Many of the majors have
arranged tank cleaning for company-
owned stations; for those with deliv-
ery contracts, filters, water-finding
paste, and information are being pro-
vided. On the other end of the spec-
trum are the suppliers that have
provided no information whatsoever
to their customer stations. At least
one company has notified customers
that unless the operator can docu-
ment that the tanks have been
cleaned, no gasoline will be deliv-
ered. We're expecting at least a few of
our operators to close up shop rather
than bear the expense of the
changeover.
Do We Need a 15th USTCA
Work Group?
Following passage of the Energy Pol-
icy Act of 2005, U.S. EPA formed 14
different workgroups to deal with the
issues related to changes made in the
Underground Storage Tank Compli-
ance Act (USTCA). These changes
were in Title XV,
Part B of the Energy
Bill. Part A of Title
XV (by the way,
Title XV is titled
Ethanol and Motor
Fuels) is the section
that requires an
increase in the vol-
ume of ethanol from
4.0 billion gallons in 2006 to 7.5 billion
gallons in 2012, so it's probably com-
ing soon to a station near you. •
from Robert N. Renkes, Executive Vice President, Petroleum Equipment Institute
And Here Comes Ultra-Low Sulfur Diesel
With all the talk nowadays about getting
underground storage tank systems ready to
store ethanol and biodiesel, the implementa-
tion of new regulations to reduce the sulfur content in
diesel seems to get lost in the shuffle. Yet in June 2006,
the regulation that seemed so far away to many of us
will be front and center. There is a new petroleum
product in town, and its name is ultra-low sulfur
diesel.
The new federal highway diesel-fuel sulfur rule
(finalized in 2001), which requires a 97 percent reduc-
tion in the sulfur content of highway diesel fuel from
its current level of 500 parts per million to 15 parts per
million, provides that over the next four years, two dis-
tinct on-road diesel fuels will be available in the United
States. One will be the traditional low-sulfur diesel
(LSD) that has a maximum sulfur content of 500 parts
per million (ppm). The other will be a new ultra-low
sulfur diesel (ULSD) with a maximum sulfur content
of 15 ppm.
Beginning June 1, 2006, 80 percent of the diesel
produced by refiners must be ULSD, with the remain-
ing 20 percent produced as LSD. The phase-in will con-
tinue until June 2010, when all on-road diesel fuel must
be USLD. Until that time, diesel marketers must make
the choice to sell one of these products, or both.
For owners of diesel tanks, the first thing they
must do is post new labels on their dispensers. The
labeling of dispensers will help state UST regulators
identify what product is being stored by the tank
owner. By June 1,2006, all retailers and wholesale pur-
chasers/consumers of diesel fuel must post labels with
the following language, depending on the product
sold.
• ULTRA-LOW SULFUR HIGHWAY DIESEL FUEL
(15 ppm Sulfur Maximum)
Required for use in all model year 2007 and later
highway diesel vehicles and engines. Recom-
mended for use in all diesel vehicles and engines.
• LOW SULFUR HIGHWAY DIESEL FUEL (500 ppm
Sulfur Maximum)
Federal law prohibits use in model year 2007 and
later highway vehicles and engines. Its use may
damage these vehicles and engines.
From 2006 to 2010, both LSD and ULSD will be avail-
able in the marketplace for on-highway diesel use. Diesel
engines manufactured in 2007 and later will run only on
ULSD, so demand for ULSD will be relatively small in
the early years of the program as the fleets start to turn
over. However, there will be more ULSD-only engines
on the road every year, and retailers will feel more pres-
sure to offer ULSD to this growing market of truck dri-
vers.
What does this mean to underground storage tank
regulators? The answer lies with the choices made by the
diesel marketers. If they choose to sell only LSD with a
maximum sulfur content of 500 ppm, it will be business
as usual, at least until 2010 when they must change to
ULSD. If they decide to sell only USLD with a maximum
sulfur content of 15 ppm, there would be a change in ser-
vice that regulators would want to know about. It will
still be diesel, but with less sulfur.
The most significant change will result if retailers
decide to expand their storage capacity to sell both LSD
and ULSD until 2010. If they choose this option, they will
either have to add tanks and/or convert existing gaso-
line or LSD diesel tanks to handle ULSD.
For most UST regulators, the introduction of ULSD
will be no big whoop. Expect a few new tanks and some
changes in service notifications during the second half of
2006. Anticipate more interest in ULSD later on as the
price stabilizes and the new model diesel engines come
out. And get ready for a lot of questions from owners of
diesel storage tanks as they make the strategic decision
about their diesel business.
For more information on the rule, go to www.epa.gov/
otaq/diesel. •
13
-------
Using Direct-Push Tools to Map
Hydrostratigraphy and Predict
MtBE Plume Diving;
by John T. Wilson, Randall R. Ross, and Steven Acree
MtBE plumes have been documented to dive beneath screened intervals of conventional monitoring-ivell networks at a num-
ber of LUST sites. This behavior makes these plumes difficult both to detect and remediate. Electrical conductivity logging
and pneumatic slug testing performed in temporary push wells are two well-established technologies that are finding new
applicability in characterizing sites with diving plumes. This article describes the use and evaluation of these techniques in finding a
diving plume of MtBE in an aquifer that supplies water to the village of East Alton, Illinois. [Note: This article contains material pre-
viously published in Ground Water Monitoring and Remediation, Vol. 25, issue 3, pages 93-102.]
Now You See It,
Now You Don't
In October 1999, MtBE was detected
in water supply wells 6, 8, and 9
within the East Alton wellfield. (See
Figure 1.) The staff of the Illinois EPA
(ILEPA) identified the plausible
sources of the MtBE contamination as
releases of gasoline from two service
stations located 400 and 500 meters
northeast of the municipal water sup-
ply wells. However, results of the
investigation revealed plume behav-
ior that they could not easily explain.
In the first 100-meter interval along
the groundwater flowpath, the
plume was readily delineated using
shallow wells extending only a few
meters into the water table. In the
second 100-meter interval, however,
MtBE was not detected in shallow
groundwater but was detected at
depths of between 6.5 and 12.5
meters below the water table.
Because there was minimal
opportunity for recharge through the
paving over the first 100-meter inter-
val, burial of the MtBE plume by
recharge could not explain the
absence of MtBE in the shallow wells.
ILEPA recognized that the MtBE
plume may have moved below the
conventional monitoring network. To
find the diving plume, they assumed
it would follow the flow of ground-
water and that groundwater would
flow through the more conductive
material in the aquifer.
They used electrical conductivity
logging, performed with push tech-
nology to characterize the hydro-
stratigraphy of the aquifer. This
technique is affordable and produces
data in real time. Southwest of the
paved area, the water table was con-
tained in a thick layer of silt and clay
extending up to 7 meters into the
14
aquifer. Below
the silt and clay
there was a layer
of sand and
gravel. ILEPA
collected water
samples from the
layer of sand and
gravel and suc-
cessfully located
the diving plume.
Through a
joint agreement
between U.S.
EPA Region 5
and the Office of
Research and
Development to
provide technical
assistance to state
regulatory agen-
cies in the region '
in identifying and predicting diving
plume behavior at groundwater sites
contaminated with MtBE, EPA's
National Risk Management Research
Laboratory (NRMRL, in Ada, OK)
provided a team to conduct a more
detailed characterization of the flow-
path between the possible sources of
MtBE and the impacted water-pro-
duction wells. Our work at the site
was designed to confirm and extend
the site conceptual model developed
by ILEPA and to evaluate their
approach for site characterization to
predict plume diving.
Potential Sources
A major highway (Illinois Route 3)
extends across the catchments of the
East Alton wellfield approximately
500 meters from the impacted wells.
ILEPA identified two former gasoline
service stations located immediately
north and south of Route 3 as poten-
tial sources of the MtBE. (See Figure
Municipal Water
Supply Wells
FIGURE 1. Relationship between possible sources of MtBE and
water supply wells contaminated with MtBE. The dotted line
encloses permanent monitoring wells with detectable concen-
trations of MtBE. The three solid arrows indicate the direction
of ground water flow in separate rounds of sampling in 2001,
2002, and 2003. They extend from the most contaminated per-
manent well at the site.
I.) Monitoring wells at both sites were
sampled in January 2000. The maxi-
mum MtBE concentration in wells to
the north of Route 3 was 156 ug/L
(although it increased to 1,800 ug/L
in a subsequent round of sampling);
in wells to the south, the maximum
concentration was 2,200 ug/L.
The maximum concentrations of
MtBE in East Alton water supply
wells 6, 8, and 9 were 32, 61, and 560
ug/L, respectively. The concentra-
tion in water supply well 9 peaked in
summer 2000 (Figure 2) and a year
later the concentration was an order
of magnitude lower. As part of
cleanup activities at these two sites,
nearly 30,000 tons of contaminated
backfill and dirt were removed from
the two sites and the holes refilled
with clean material.
Hydrogeologic
Characteristics of the Site
Data were available from a short-
-------
600
500 -
400 -
^300 -
£200 -
'100 -
Jun-9
Jun-00
Jun-01 Jun-02
Date of Sampling
Jun-Q3
Jun-04
FIGURE 2. Concentration ofMtBEin the most contaminated
water supply well.
term pumping test (300 minutes at
6240 L/min) conducted on well 1,
one of the pumping wells in the well-
field. Transmissivity was estimated
based on analysis of drawdown
curves from two observation wells
using the Theis-type curve for a con-
fined aquifer for the pumping data
and the Theis-recovery method for
the recovery data. These estimates of
transmissivity varied from 2,320 to
2,590 m2/d. The transmissivity esti-
mated from the specific capacity of
the pumped well was 1,510 m2/d.
The best estimate of the saturated
thickness of the aquifer was 19.5
meters. The corresponding estimate
of the hydraulic conductivity ranged
from 77 to 134 meters per day.
In 1999, ILEPA simulated the
municipal wellfield using MOD-
FLOW. Their model assumed an
average rate of recharge of 20
cm/year, an average porosity of 0.30,
and an average hydraulic conductiv-
ity of 66 meters per day. The
expected travel time of water from a
point midway between the two pos-
sible source areas to water supply
well 9 was five years. The concentra-
tion of MtBE reached a maximum
concentration in the water supply
wells in 2000. If the release of MtBE to
the aquifer occurred in 1994, there is
a good correspondence between the
arrival time of MtBE in the monitor-
ing record and the predictions of the
model.
Ground water Flow
The general direction of groundwater
flow at the site is toward the south-
west from the potential sources, then
under a shopping
mall, and then
under a grassy
undeveloped area
adjacent to the
municipal water
supply wellfield.
The area to the
north and east of
the wellfield con-
sists of single-
family residential
housing. To esti-
mate the direction
of groundwater
flow from the
possible source
areas and the
magnitude of the
hydraulic gradi-
ent, we used regression analysis to fit
a plane to the elevation of the water
table in permanent monitoring wells
in the source areas. We fit the regres-
sion using the Optimal Well Locator
(OWL) software (Srinivasan et al.,
2004). Results are presented in Table 1.
Over the three rounds of sam-
pling, the direction of groundwater
flow varied by 23 degrees and the
hydraulic gradient varied by 45 per-
cent. The fit to the regression was
particularly good for data collected
on August 9,2001, and September 19,
2002, although the fit was not as good
for data collected on January 8,2003.
In September 2002, the depth to
water was determined in most of the
permanent wells at the site. Figure 3
compares the projected groundwater
contours from the regression to the
actual water elevations on September
19, 2002. All the wells at the site fit the
assumption that the water table was a
plane. Presumably, the MtBE that first
reached the water supply wells would
have moved through the most con-
ductive material in the aquifer.
As will be discussed later, the
highest hydraulic conductivity deter-
mined at any depth interval in the
aquifer was 51 meters per day. The
three solid arrows in Figure 1 repre-
FIGURE 3. Correspondence between
measured elevation (meters) of the
water table in permanent wells on
September 19,2002, and the contours
produced by regression to fit a plane.
All the elevations fell between 123.0
and 123.7m.
sent the direction of groundwater
flow as estimated from the water ele-
vation data collected in 2001, 2002,
and 2003. The length of the arrows is
the distance the groundwater would
have moved in one year under the
prevailing hydraulic gradient if the
effective porosity of the aquifer was
0.30 and the hydraulic conductivity
was 51 meters per day.
If these assumptions of porosity
and hydraulic conductivity are valid,
the MtBE plume near the possible
source areas is moving toward the
municipal wellfield at a velocity on
the order of 86 to 133 meters per year.
Thus, a plume from either of the two
potential sources could reach the
wellfield in three to six years.
Illinois EPA's Evidence for
Plume Diving
ILEPA mapped the distribution of
MtBE in the aquifer by sampling per-
manent wells installed northeast of
the shopping center and by using
• continued on page 16
TABLE 1 Hydraulic Gradient and Direction of Ground Water
Flow Near the Source Areas of the MtBE Plume
Date of
Measurement
August 9, 2001
September 19, 2002
January 8, 2003
Number of Wells
Measured
13
19
7
Hydraulic
Gradient (m/m)
0.00138
000214
000159
Direction of Flow
(degrees clockwise
from north)
240.5
2332
217.9
Correlation
Coefficient r2
0.97
0.98
077
15
-------
• MtBE Plume Diving
from page 15
push technology to sample ground-
water southwest of the shopping cen-
ter. (See Figure 4.) The permanent
wells were installed through asphalt
paving. They are screened to a depth
ranging from 9.1 to 10.7 meters below
land surface. The depth to water is
about 8.5 meters below land surface.
Water samples from temporary push
wells were acquired in the paved
area southwest of the shopping mall.
They were acquired at depths of
about 9, 15, and 21 meters below
ground surface or 0.5, 6.5, and 12.5
meters below the water table.
The permanent wells are coded
in Figure 4 according to the maxi-
mum concentration of MtBE in any of
four rounds of sampling extending
from January 2000 to September
2002. The plume of MtBE was easily
detected by the shallow monitoring
wells in the parking lot northeast of
MtBE greater than 1,000 ug/L. In
contrast to the behavior of the plume
northeast of the shopping center,
MtBE was not detected in shallow
groundwater southwest of the shop-
ping center. (See Figure 4.)
Although MtBE was absent in
the shallow groundwater, it was
detected at two push-well locations
at depths of 6.5 and 12.5 meters
below the water table. The area above
the plume of MtBE was paved or
roofed. There is little possibility that
burial of the plume by recharge could
explain the diving plume.
To better understand the influ-
ence of hydrostratigraphy on the
behavior of the plume of MtBE,
ILEPA surveyed the aquifer using an
electrical conductivity probe mounted
on direct-push tools. (The electrical
conductivity of sandy material that
readily transmits groundwater is
lower than the conductivity of silts
and clays.) The electrical conductivity
survey revealed that the water table in
the aquifer.
Building on the work of the
ILEPA, our team from NRMRL con-
ducted a more detailed characteriza-
tion of the flowpath between the
possible sources of MtBE and the
impacted water production wells. We
performed the characterization at
locations A, B, C, D, and E in Figure
5. Our work at the site was designed
to confirm and extend the site con-
ceptual model developed by ILEPA
and to evaluate their approach for
site characterization to predict plume
diving.
Evaluating the
Hydrostratigraphy
Plume diving is controlled either by
recharge to the aquifer, which
"buries" the plume, or by the hydros-
tratigraphy of the aquifer. The
approach to site characterization
used by ILEPA is appropriate for
sites where plume diving is con-
Permanent Wells: MTBE • >1000 fig/L Q >10 /ig/L
Deep Push Wells: MTBEH >500 jig/L B >10
Shallow Push Wells A MTBE <10
O <10
Q
FIGURE 4. Distribution of MtBE in shallow ground water near
the possible sources and in deeper ground water farther down-
gradient of the sources.
FIGURE 5. Locations used for detailed characterizations of
hydrostratigraphy using electrical conductivity of the vertical
extent of MtBE contamination and hydraulic conductivity.
the shopping mall. Although the
screened intervals of the wells were
very shallow, there is no indication of
plume diving. The plume extended
in the direction of groundwater flow
that was predicted from water table
elevations in the monitoring wells.
The centerline of the plume con-
tained wells with concentrations of
16
the area south-
west of the shop-
ping center was
contained within
an interval domi-
nated by silts and
clays. The silt and
clay interval ex-
tended about 18.3
meters below land surface and 10.7
meters below the water table.
Apparently, groundwater moved
beneath the layer of silt and clay as it
flowed toward the municipal water
supply wells, thus the MtBE plume
"dived" below the elevation of the
shallow groundwater samples as it
followed the flow of groundwater in
trolled by hydrostratigraphy.
To better define the hydros-
tratigraphy of the aquifer, we
repeated the electrical conductivity
logging and extended it as far into
the aquifer as the tools would allow.
We defined the vertical distribution
of hydraulic conductivity in the
aquifer with a downhole-flowmeter
survey in a well that was screened
across most of the aquifer. We also
evaluated the correspondence
between hydrostratigraphy and
hydraulic conductivity by compar-
ing the electrical conductivity log to
the vertical distribution of hydraulic
conductivity determined with the
flowmeter test.
-------
I 2
I
¥
127 •
122 •
117 •
112 •
0 100 200 0 100 200 0 100 200 0 100 200 0 100 200
Electrical Conductivity (mS/m)
FIGURE 6. Distribution of electrical conductivity along the flowpath from the possible
source areas of the MtBE plume to the water production wells.
In recent years, pneumatic-slug
testing performed in temporary push
wells has emerged as an affordable
alternative to slug testing of perma-
nent wells. To evaluate this approach,
we obtained estimates of hydraulic
conductivity from pneumatic-slug
tests and compared them to the down-
hole-flowmeter test. The vertical dis-
tribution of the MtBE plume was
defined at high resolution by sam-
pling water every 3.3 meters, extend-
ing from the water table to the point of
refusal of the push sampling tool.
We evaluated the inferences
made about the texture of aquifer
material from electrical conductivity
logs by acquiring core samples from
a depth interval that had high con-
centrations of MtBE, high hydraulic
conductivity, and low electrical con-
ductivity; and a second interval of 3.3
meters above the first that had low
concentrations of MtBE, low
hydraulic conductivity, and high
electrical conductivity.
Finally, we characterized the
geochemistry of the MtBE plume to
evaluate the prospects for natural
biodegradation of MtBE along the
flowpath. If conditions for natural
biological degradation of MtBE in the
aquifer were unfavorable, MtBE con-
tamination would persist along the
flowpath from the potential source
areas to the municipal wells. (For
more details on the materials and
methods used in this evaluation see
http://wivw.epa.gov/OUST/mtbe/Direct
push tools to predict MTBE plume div-
ing.pdf)
Changes in Hydrostrati-
graphy along the Flowpath
The hydrostratigraphy was charac-
terized using electrical conductivity
logging along an inferred flowpath,
extending from the potential sources
of MtBE contamination to the conta-
minated water supply wells. The
sampling locations are depicted in
Figure 5. Data from the electrical con-
ductivity log are presented in Figure
6. In the experience of Butler et al.
(1999) and Christy et al. (1994), an
electrical conductivity of less than 20
millisiemens per meter (mS/m) is
indicative of sand and gravel, while
an electrical conductivity greater
than 100 mS/m is indicative of clay
and silt.
At locations A, B, and C, the
MtBE plume was readily detected by
conventional monitoring wells with
shallow-well screens. As inferred
from the low value for electrical con-
ductivity, the water table at locations
A, B, and C is contained within sandy
material. In contrast, the MtBE plume
was not detected in the shallow
groundwater at locations D or E. At
location D, the electrical conductivity
log indicates a layer of clay and silt
extending from the water table an
additional 9 meters into the aquifer.
Similarly, at location E, the electrical
conductivity log indicates a clay layer
extending from the water table an
additional 5.2 meters into the aquifer.
Notice that at locations B, C, and D
there is a two- to threefold increase in
the electrical conductivity at the
water table. The marked increase in
electrical conductivity at location D
at an elevation of 122 meters proba-
bly reflects the influence of the water
table and does not indicate a change
in the texture of the sediment.
To confirm the inferences made
from the electrical conductivity logs,
core samples were acquired from
location D at elevations extending
from 112.2 to 113.1 meters (in a zone
of low electrical conductivity) and
from 113.7 to 114.6 meters (in a zone
of high electrical conductivity). The
intervals are identified with arrows
in panel D of Figure 6. The sediment
recovered from the higher elevation
with high electrical conductivity was
plastic clay. (See Figure 7.) The sedi-
ment recovered from the lower eleva-
tion was coarse sand in a matrix of
medium to fine sand.
Location D was the first location
along the flowpath between the
potential sources and the impacted
water supply wells where the water
table occurred in fine-textured mater-
ial. The panel on the left side of
Figure 8 compares the vertical distri-
bution of electrical conductivity and
hydraulic conductivity at Location D.
There was an inverse correspondence
• continued on page 18
113.7 meters amsl, location D
112.8 amsl, location D
FIGURE 7. Texture of materials recovered in two core samples from location D.
Compare Figure 6 for the electrical conductivity of the depth interval sampled. Note
that the sample from an absolute elevation of 113.7m retained the imprint of the
teeth of the core retainer.
17
-------
• MtBE Plume Diving
from page 17
between electrical and hydraulic con-
ductivity.
As determined from a pneu-
matic-slug test, the average hydraulic
conductivity in the well used for the
flowmeter test was 14.3 meters per
day. In the first 9.1 meters of the
aquifer, the hydraulic conductivity as
revealed by the downhole-flowmeter
test was low—0.4 meters per day or
less. When the lithology transitioned
from silt and clay to sand and gravel
at an elevation of about 112.8 meters
above mean sea level (amsl), the
hydraulic conductivity increased dra-
matically. The hydraulic conductivity
at an elevation of 111.3 meters was
51.2 meters per day, compared with
0.27 meters per day at an elevation of
114.3 meters. As revealed by the
downhole-flowmeter test, 99 percent
of the transmissivity in the interval
between elevations of 120.4 and 106.7
meters amsl was associated with the
sandy material that extended
between 114.0 and 106.7 meters amsl.
The pneumatic-slug tests con-
ducted in the temporary push wells
also revealed the low hydraulic con-
ductivity of the shallow silts and clays
and the sharp increase in hydraulic
conductivity in the interval between
114.3 and 111.3 meters amsl (panel on
the right side of Figure 8). The
hydraulic conductivity estimated
from the pneumatic-slug test at an
elevation of 111.3 meters was 12.5
meters per day, compared to 0.33
meters per day at an elevation of
114.3 meters.
The hydraulic conductivity as
estimated by downhole-flowmeter
testing was approximately two to
three times higher than the hydraulic
conductivity estimated by pneu-
matic-slug testing in temporary
wells. (See Figure 8.) Water was not
added to the rods of the temporary
push-technology wells to equalize
the pressure across the screens before
the sheath was pulled away from the
screen. As discussed by Butler et al.
2002, this can result in partial plug-
ging of the screens. No attempt was
made to develop the temporary push
wells. Both of these practices may
have contributed to the lower
hydraulic conductivity determined in
the temporary wells. In any case, the
agreement between the estimate of
18
Hydraulic Conductivity from Flow Meter
Test (meters per day)
Elevation
(meters) 114
Hydraulic Conductivity from Flow Meter Test (meters
per day)
40
0 50 100 150
Eiectncal Conductivity(mS/m)
113
110
108
Hydraulic Conductivity from Pneumatic Slug Test
(meters per day)
FIGURE 8. Inverse correspondence between electrical conductivity and hydraulic
conductivity. The hydraulic conductivity distribution, measured with a downhole-
flowmeter test, is depicted by the solid shape in both panels. The left panel compares
the electrical hydraulic conductivity measured by a downhole flowmeter at location D
in Figure 5. The right panel compares the hydraulic conductivity measured by a
flowmeter test to the distribution of hydraulic conductivity measured by a pneumatic-
slug test. The vertical extent of the dark lines in the right panel represents the
screened interval of the temporary push well subjected to a pneumatic-slug test.
Concentration MTBE (ng/l)
0 200 400 600 800
Concentrator MTBE (ng/l)
0 100 200 300
Concentration MTBE (njfl)
0 200 400 600
0 50 100 150 200
Electrical Conductivity (rnSAn)
0 50 100 150 200
Electrical Conductivity (mS'm)
0 50 100 150 200
Electrical Conductivity (mS/m)
FIGURE 9. Association of higher concentrations of MtBE with sandy aquifer material
(low electrical conductivity) along an inferred ground water flowpath from the
potential sources of MtBE (location B), to a location where the plume dived below
the water table (location D), to the vicinity of a municipal well (location E). Electrical
conductivity logs are the solid lines; concentrations of MtBE are the dashed lines.
hydraulic conductivity from the slug
tests and the flowmeter tests were
acceptable.
MtBE Plume Diving Predicted
from Hydrostratigraphy
Along the inferred flowpath between
the potential sources and the
impacted water supply wells, the
highest concentrations of MtBE were
associated with sandy material with
low electrical conductivity. (See Fig-
ure 9.) Near the potential sources, the
highest concentrations of MtBE were
found near the water table (location
-------
B). In the location where the MtBE
plume first dived below the water
table, the highest concentrations of
MtBE were found at the contact
between the shallow clay unit and
the underlying sand unit (location
D). Further along the flowpath (loca-
tion E), the highest concentration of
MtBE was found more to the center
of the sandy unit. Apparently, the
plume of MtBE followed the hydro-
stratigraphic units with highest
hydraulic conductivity.
Absence of MtBE
Biodegradation
There is a general perception that
MtBE does not biologically degrade
in groundwater, even though MtBE
has been shown to biologically
degrade under aerobic and denitrify-
ing conditions, iron-reducing condi-
tions, sulfate-reducing conditions,
and methanogenic conditions (Wil-
son 2003).
The rates of degradation under
aerobic and denitrifying conditions
are fast (Borden et al. 1997; Salanitro
et al. 2000; Bradley et al. 2001a). The
rates of degradation under iron-
reducing conditions may be fast
when readily available iron is sup-
plied to iron-reducing bacteria
(Finneran and Lovley 2001), but the
rate under iron-reducing conditions
in an aquifer appears to be slow
(Landmeyer et al. 1998). Laboratory
studies show that MtBE may degrade
under sulfate-reducing conditions
(Somsamak et al. 2001; Bradley et al.
2001b), but the field studies available
to date indicate that MtBE degrades
slowly in aquifers under sulfate-
reducing conditions (Wilson 2003).
At two field sites, the rate of MtBE
degradation to tertiary-butyl ether
(TEA) was rapid under methanogenic
conditions (Wilson et al. 2000; Kol-
hatkar et al. 2002). Many MtBE
plumes are much longer than the
associated plumes of BTEX, probably
because the BTEX compounds were
biologically degraded but MtBE failed
to degrade or degraded very slowly
(Amerson and Johnson 2002; Land-
meyer et al. 1998). This pattern is most
likely when the MtBE is contained in
groundwater devoid of oxygen and
nitrate, in water with low concentra-
tions of methane (<0.5 mg/L), and in
water with low concentrations of TEA
compared with MtBE.
TABLE 2 Concentration of Contaminants and Geochemical Parameters
at the Most Contaminated Depth Interval at Three Locations
along the Flow/path. Samples Collected in August 2001 and November 2001
Parameter
MTBE (ug/L)
TBA (uq/L)
Benzene (ug/L)
BTEX (ug/L)
Methane (mg/L)
Sulfate (mg/L)
Iron II (mg/L)
Nitrate-N (mg/L)
Oxygen
B
(near potential source)
695
<10
<0.5
22
006
232
15
<0.1
0.25
D
(downgradient)
197
<10
<0.5
0.75
0.17
462
10
<0.1
015
E
(farther downgradient)
553
<10
<0.5
<0.5
031
46.5
NA
NA
NA
Table 2 compares the concentra-
tions of MtBE, TBA, and BTEX com-
pounds near the potential source of
contamination, at the first location
where plume diving was noticed,
and further along the flowpath near
the water supply wells. Although
MtBE persists along the flowpath, the
BTEX compounds are depleted.
The groundwater was essentially
devoid of oxygen and nitrate, con-
tained little methane, and the concen-
tration of TBA was below the
detection limit, which was much
lower than the concentrations of
MtBE. Based on the geochemical
environment, MtBE should degrade
slowly or not at all in the groundwa-
ter at East Alton. Because it persists,
there is an opportunity for the MtBE
to move with the flow of groundwa-
ter to the water supply wellfield.
An Overall Thumbs Up
So, to sum it up, the MtBE plume
stayed near the water table for the
first 100 meters from the potential
sources and then dived below con-
ventional monitoring over the next
100 meters. At the location where the
plume dived, the depth to water was
9.1 meters below land surface. The
first 10 meters of material below the
water table had an electrical conduc-
tivity near 100 mS/m, indicating silts
and clays. An electrical conductivity
near 25 mS/m, indicating sands or
gravels, was encountered at a depth
of 10.6 m below the water table, and
the sands and gravel extended to a
depth of at least 15.2 m below the
water table.
Pneumatic-slug tests measured
low hydraulic conductivity in the
interval of silt and clay (0.34 and
0.012 m/d) and higher hydraulic con-
ductivity in the interval with sands
and gravels (12.5, 11.6, and 11.3
m/d). Groundwater with the highest
concentration of MtBE was produced
just below the contact between the
silt and clay and the sands and
gravel.
Two properties of the aquifer at
East Alton are responsible for move-
ment of the MtBE plume to below the
water table, which gave the appear-
ance that the plume dived into the
aquifer. First, the geochemistry of the
groundwater prevented rapid
biodegradation of the MtBE and
unacceptable concentrations of MtBE
persisted along the flowpath. As a
rule of thumb, long-term persistence
of MtBE is possible in groundwater
depleted of oxygen and nitrate but
not accumulating significant concen-
trations of methane. Groundwater
that meets this geochemical profile is
vulnerable to plume diving caused
by hydrostratigraphic influences.
Characterization of the hydrostratig-
raphy and vertical distribution of
hydraulic conductivity may be neces-
sary to manage risk from MtBE cont-
amination in these aquifers.
Second, the water table at the
UST sites is in sandy material, but
downgradient of the spill, the water
table is in silts and clays. The natural
flow of groundwater in the aquifer
found its way into a deep layer of
sand and gravels lying below the
layer of silts and clays at the water
table. The hydrostratigraphy oi the
aquifer controlled the vertical distrib-
ution of MtBE contamination. The
plume of MtBE simply followed the
natural flow of groundwater. This
• continued on page 20
19
-------
/ USTline Bulletin 52 • Mm/ 2(K)b
• MtBE Plume Diving
from page 19
study validated the conceptual
model of the diving MtBE plume that
was developed by ILEPA.
In this case, electrical conductiv-
ity logs proved to be an effective tool
for recognizing the vertical distribu-
tion of hydrostratigraphic features
that control the movement of water
in the aquifer. However, we have
conducted electrical conductivity
logs at other sites where the logs
failed to recognize the controlling
hydrostratigraphic features. At these
other sites, cone penetration testing
revealed the hydrostratigraphic fea-
tures that controlled the flow of
groundwater. In any case, it is worth-
while to recover and evaluate core
samples to calibrate the log response
at each site.
The downhole-flowmeter test
was conducted as a research activity
to provide a benchmark for the elec-
trical conductivity log and the pneu-
matic-slug tests. Strictly speaking,
flowmeter tests require a fully
screened well across the aquifer of
interest. However, such wells are
expensive and are rarely available at
UST-release sites.
The pneumatic-slug test offers a
realistic alternative for mapping the
vertical distribution of hydraulic con-
ductivity, and identifying the opti-
mum depth intervals for taking push
samples or locating screens for per-
manent monitoring wells. Schulmeis-
ter et al. 2003, Butler 2002, and Butler
et al. 2002 present a detailed descrip-
tion and evaluation of pneumatic-
slug testing at sites in Kansas that is
very similar to the aquifer at East
Alton. Pneumatic-slug testing is well
developed and can be considered a
routine tool for site characterization.
These site-characterization tools
are cost effective. During our investi-
gation, a two-person crew set up the
equipment for electrical conductivity
logging, logged 80 feet of subsurface
material, and then recovered and
cleaned the tools in an average time
period of two hours. A two-person
crew installed and recovered the
push tools for the pneumatic-slug
tests, while a third person conducted
the tests. One set of push tools was
tested, while a second set was recov-
ered, cleaned, and reinstalled at the
next location.
20
On average, the three-person
crew conducted a pneumatic-slug
test every two hours. If time had been
taken to develop the temporary push
wells before they were slug tested,
the three-person crew would have
conducted a pneumatic-slug test
every three to four hours. However,
at this site, it was not necessary to
develop the temporary push wells to
discern the sharp contrast in
hydraulic conductivity between the
silts and clays at the water table and
the sands and gravels that carried the
plume of MtBE. •
John T. Wilson, I'li.D , is n research
nncrobioJo^ist until the U.S. EPA
Office of Research and Development.
He is assigned to the Subsurface Reme-
diation Brandt in the Ground Water
and Ecosystems Restoration Division of
the National j<;s/<- Management 1 abora-
ton/. Currently, Dr. Wilson /s leading
an evaluation of the natural biological
processes that degrade MtBE and TRA
in groundwater. He can be reached at
wilson .johnt@epa.gov.
Randall R. Ross, Ph D., is a In/drologist
with the U.S. EPA Office of Research
and Development. He is assigned to the
Applied Research and Technical Sup-
port Branch in the Ground Water and
Ecosystems Restoration Division of the
National Risk Management Labora-
tory. He can be reached at ross.ran-
dall@epa.gov.
Steven D. Acrcc is a hydrologist with
the U.S. EPA Office of Research and
Development. He is assigned to the
Applied Research and Technical Sup-
port Branch in the Ground Water and
Ecosi/stems Restoration Division of
the National Risk Management
Laboratory He can be reached at
acree.steven@epa.gov.
Acknowledgments
This work was supported through a
Regional Applied Research Effort
Project sponsored by U.S. EPA
Region 5. Gilberto Alvarez of Region
5 provided support and guidance.
Gina Search, a geologist with the Illi-
nois EPA, developed the original
conceptual model of the hydrology of
the site. She provided advice and
guidance during the field sampling
for this research effort. Karl Kaiser
with the LUST Section of the Illinois
EPA is the case manager for the site.
He shared the case files and helped
coordinate the effort. James Butler,
with the University of Kansas and the
Kansas Geological Survey, provided
a helpful and detailed review.
Note: U.S. EPA, through its Office oj
Research and Development, partially
funded and collaborated in the research
described here under an in-house project.
It has not been subjected to agency review
and therefore does not necessarily reflect
the views of the agency, and no official
endorsement should be inferred.
References
Amerson, I., and R.L. Johnson. 2002. A natural gradi-
ent tracer test to evaluate natural attenuation of
MTBE under anaerobic conditions. Ground Water
Monitoring and Remediation 23, no 1. 54-61
Borden, R C, R.A. Daniel, L.E. LeBrun IV, and C.W
Davis. 1997. Intrinsic biodegradation of MTBE and
BTEX in a gasoline-contaminated aquifer Water
Resources Research 33, no. 5.1105-1115.
Bradley, P M., F H. Chapelle, and J.E. Landmeyer.
2001a. Methyl t-Butyl Ether mineralization in sur-
face-water sediment microcosms under denitrify-
ing conditions. Applied and Environmental
Microbiology 67, no. 4:1975-1978.
Bradley, P.M , F.H Chapelle, and J E Landmeyer.
2001b. Effect of redox conditions on MTBE
biodegradation in surface water sediments. Envi-
ronmental Science and Technology 35, no 23
4643-4647.
Butler, JJ 2002. A simple correction for slug tests in
small-diameter wells. Ground Water 40, no. 3:
303-307.
Butler, J.J. 1998. The Design, Performance, and Analy-
sis of Slug Tests. Boca Raton, Florida: Lewis Pub-
lishers.
Butler, J.J., and E.J. Gamett. 2000. Simple procedures
for analysis of slug tests in formations of high
hydraulic conductivity using spreadsheet and sci-
entific graphics software. Kansas Geological Survey
Open-File Report 2000-40. Lawrence, Kansas
Kansas Geological Survey. Available at
http-//urww.kgsku.e
-------
Salamtro, J P, P C Johnson, G E. Spinnler, P M
Maner, H.L. Wisniewski, and C Bruce 2000 Field-
scale demonstration of enhanced MTBE bioremedi-
ation through aquifer bioaugmentation and
oxygenation. Environmental Science and Technology
34, no 19 4152-4162
Schulmeister, M K., J J. Butler, J.M. Healey, L. Zheng,
D A Wysocki, and G.W McCall. 2003 Direct-push
electrical conductivity logging for high-resolution
hydrostratigraphic characterization. Ground Water
Monitoring and Remediation 23, no. 3 52-62
Somsamak, P., R M Cowan, and M.M Haggblom.
2001 Anaerobic biotransformahon of fuel oxy-
genates under sulfate- reducing conditions. FEMS
Microbiology Ecology 37, no. 3 259-264.
Srinivasan, P , D F Pope, and E Stnz. 2004 Optimal
Well Location (OWL) A screening tool for evaluat-
ing locations of monitoring wells. EPA 600/C-
04/017. Available under Software at
http //www. epa gov/aa/ (accesssed June 25,2005)
Weaver, J.W, and J T Wilson 2000. Diving plumes
and vertical migration at petroleum hydrocarbon
release sites.
LUSTLme, a report on federal & state programs to
control leaking underground storage tanks. Bul-
letin 36. Lowell, Massachusetts New England
Interstate Water Pollution Control Commission
Wilson, JT 2003 Fate and transport of MTBE and
other gasoline oxygenates In Handbook for Man-
aging Releases of Gasoline Containing MTBE, ed.
E.MoyerandP Kostecki, 19-61 Amherst, Massa-
chusetts Amherst Scientific Publishers.
Wilson, J.T., J A. Vardy, J.S Cho, and B.H. Wilson.
2000. Natural attenuation of MTBE in the subsur-
face under methanogemc conditions. EPA/600/R-
00/006. Available at http //www epa gov/ada/
pubs/reports html (accessed June 25,2004).
Young, S.C., H E Julian, H S Pearson, F J Molz, and
G.K Boman. 1998. Application of the electromag-
netic borehole flowmeter EPA/600/R-98/058
Washington, DC. U S EPA Office of Research and
Development. Available at http'//
wimv epa gov/acta/pubs/reports html
A Short Course on Plume Diving
To control the costs of monitoring, most LUST sites are characterized with
conventional monitoring wells that are screened a few feet above and below
the water table or by temporary push samples that are taken near the water
table. Occasionally contamination is missed because of the plume movement
beneath the screens of the monitoring wells, a situation termed "plume div-
ing." Weaver ind Wilson (2000) discuss several environmental processes
that may cause plume diving, illustrate the behavior with a case study on
Long Island, New York, and offer recommendations to recognize and react to
plume diving. The most widely recognized process that can produce plume
diving is the burial of a plume by the recharge of clean water above it. U.S.
EPA provides a calculator on its website that can be used to estimate plume
diving caused by recharge (EPA On-line Tools for Site Assessment Calcula-
tion: Plume Diving;
www,epa.gov/athens/fearn2model/part-two/onsite/dMng),
Revised Operating and
Maintaining UST Systems
Manual Now Available
U.S. EPA's revised Operating And
Maintaining UST Systems: Practical
Help And Checklists (EPA 510-B-05-
002) is now printed and in stock at
NSCEP in Cincinnati. Users can
order printed copies by calling
NSCEP toll free at (800) 490-9198.
The publication is also available
for downloading from the OUST
website at http://www.epa.gov/
oust/pubs/index.htm.
UST Program Meets All FY
2005 GPRA Goals
The LUST program had four
national GPRA goals for FY 2005:
(1) complete 14,500 cleanups, (2)
complete 30 cleanups in Indian
Country, (3) increase the signifi-
cant operational compliance rate
of UST facilities to 65 percent, and
(4) decrease newly reported con-
firmed releases to fewer than
10,000.
In a memo dated December
15, OUST Director Cliff Rothen-
stein reported that the UST pro-
gram met all of these goals. EPA
and state tank programs: (1) com-
pleted 14,583 cleanups in states
and territories, (2) completed 53
cleanups in Indian Country, (3)
achieved operational compliance
at 66 percent of all UST facilities,
and (4) reported 7,421 confirmed
releases.
Since the beginning of the
UST program, almost 74 percent
of all reported releases (332,799)
have been cleaned up, and the
national UST cleanup backlog has
been reduced to 119,240, which is
an 8 percent drop from last year.
The full end-of-year report is
posted on OUST's website at
www.epa.gov/oust/.
Updated UST Program
Directory Now Available:
The Underground Storage Tank Pro-
gram Directory, which contains
EPA and state UST program con-
tact information, has been
updated and is now available on
the OUST website at http://www.
epa.gov/oust/pubs/reglist.htm.
Congress Appropriates $8
Million for UST Sites
Impacted by Hurricanes
Katrina and Rita
Congress has appropriated $8 mil-
lion in supplemental funding to
address UST sites impacted by
hurricanes Katrina and Rita. This
funding will be used to conduct
site assessments and, if necessary,
clean up releases. As many as 800
UST facilities may have had hurri-
cane-related damage that resulted
in releases to the environment.
This supplemental funding was
part of a larger Katrina supple-
mental appropriations bill that
was attached to the defense
appropriations bill recently signed
by President Bush.
21
-------
Vapor Attenuation in the Subsurface from
Petroleum Hydrocarbon Sources
An Update and Discussion on the Ramifications
of the Vapor-Intrusion Risk Pathway
by Robin Davis
Petroleum-contaminated soil and groundw ater caused by leaking underground storage tanks (LUSTs) can result in subsurface
sources of vapor contamination. These vapors move slowly by diffusion through the subsurface and, if not attenuated, can
intrude into overlying buildings, resulting in a complete vapor-intrusion pathway. Thankfully, there are very few reported
cases of vapor intrusion, despite the thousands of LUST sites in this country. Even so, LUST project managers like me are saddled
with the task of determining when and if the vapor-intrusion pathway is complete and justifying cleanup expenses since funds are
limited. The purpose of this article is to show that natural biodegradation processes, accompanied by certain predictable characteris-
tics, attenuate contaminant vapors and that project managers can determine when the pathway may be complete, based on those char-
acteristics and contaminant concentrations.
1 serve as a member of a U.S. EPA work group formed in early 2004 to study the behavior of subsurface petroleum hydrocarbons
and the vapor-intrusion-to-indoor-air-exposure pathway. The decision to form the workgroup was sparked by U.S. EPA.'s draft
guidance, Evaluating the Vapor Intrusion to Indoor Air Pathway from Groundwater and Soils (U.S. EPA, 2002).
This guide uses the Johnson-Ettinger model (J&E) (Johnson and Ettinger, 1991) to calculate groundwater-screening levels for the
vapor-intrusion pathway that are expected to be protective of overlying receptors. The guide also presents conservative shallow and
deep soil-gas concentrations that are expected to be equally protective (Table 2c, U.S. EPA, 2002). U.S. EPA wisely recognized that
the J&E model often predicts that the vapor-intrusion pathway is complete, when it may not be, because the model does not account
for biodegradation.
To explore this issue further, U.S. EPA's workgroup generated a repository of published literature on subsurface petroleum
hydrocarbon vapor attenuation. As a member of the workgroup, I reviewed and compiled data from those publications, and published
my initial findings in LUSTLine Bulletin 49, March 2005. Those findings were based on a relatively small set of vapor data but
showed significant attenuation in a majority of the sample events.
That data set indicated significant attenuation when accompanied by certain predictable signature characteristics of vapor
biodegradation, specifically the presence of clean overlying soil and at least 5 percent oxygen in the vadose-zone soil gas. This paper
presents the findings of a much larger data set, which continues to show that petroleum hydrocarbon vapors attenuate if there is clean,
oxygenated soil overlying the contaminant source.
More Data Compilation
Since my last discourse on the subject
of subsurface petroleum hydrocar-
bon vapor attenuation, I have
amassed a larger data set, which cur-
rently includes 161 benzene events
and 136 TPH events in which vapor
samples were collected at multiple
depths (Figure 1), including 28 events
beneath buildings ("sub-slab"). Data
for 59 benzene vapor events and 65
TPH vapor events were not useful for
evaluating attenuation because they
either took place in uncontaminated
areas or they took place at depths or
intervals of five or more feet and not
within the zone of biodegradation.
The data set for evaluating vapor
attenuation was subsequently re-
duced to 102 benzene events (Figure
2) and 71 events for total petroleum
hydrocarbons (TPH).
Subsurface attenuation factors
(not the same as attenuation factors
into buildings) were calculated by
dividing the vapor concentration at
22
FIGURE 1. Locations of events where samples were collected at multiple depths.
Number of Sampling Events Evaluated
TOTAL: Benzene 161,TPH 136
-------
the shallow depth by that at the
deeper depth. The data set was cate-
gorized and evaluated according to
the following site characteristics:
• Multi-depth and sub-slab vapor-
phase benzene, TPH, oxygen,
and carbon dioxide concentra-
tions
• Depth to groundwater
• Depth to source*
• Source strength (NAPL, adsorbed
and dissolved-phase benzene, TPH)*
• Presence of free product on
groundwater
• Type of ground surface cover
(paved, unpaved, buildings)
• Soil type
• Presence or absence of complete
vapor attenuation*
• Sampling dates
• All media sampled correlated to
date of collection
* Data categories added since Davis, 2005.
Attenuation factors at or below
0.1 are considered significant attenu-
ation and those greater than 0.1 rep-
resent insignificant attenuation.
Attenuation Happenings
The findings of this study show sig-
nificant attenuation of vapors in 98
percent of the benzene sample events
and in 92 percent of the TPH sample
events (Figure 3).
Figure 4 shows a representative
example of significant attenuation
from a multi-depth vapor sample
point (Ririe, et al., 2002), where five
feet of clean coarse-grained soil over-
lie a contaminant source and about 3
percent oxygen is present.
Biodegradation of the vapors is
evidenced by oxygen depletion at
depth accompanied by carbon diox-
ide enrichment, and a rebound of
oxygen to near-atmospheric condi-
tions above the contaminant source
with concomitant reduced carbon
dioxide production. The data indicate
that only two feet of clean fine-
grained overlying soil is necessary to
attenuate petroleum vapors.
Soil types were divided into two
groups according to permeability:
coarse-grained (>10_9 cm2) and fine-
grained (<10~9 cm2) (Freeze and
Cherry, 1979). All of the benzene and
TPH events in fine-grained soil
exhibited significant attenuation. The
22 S Benzene Vapor Sample Event!
II MuftMtepth Benzene Vapor Sample Point
I RF <0.1 lolid line
I Af>a.iaail>edlne
;' [ Coarse-grained
>:-;- Fine-grained
•Building Conduction
sag - stab on grade
bsnrrt° basement
FIGURE 2. Benzene multi-depth vapor sample events (102)
120 -
100-
80 -
n 60 -
40 -
20 -
0
98%
Significant AF<0.1
Benzene
TPH
Insignificant AF>0.1
SIGNATURE CHARACTERISTICS OF BIO-ATTENUATION
- 5 feet clean coarse-grained or 2 feet of fine-grained soil overlie contaminant source
- Vapor concentrations decrease significantly vertically away from source
- O2 depleted and CO2 enriched near the source, O2 enriched and CO2
depleted with increasing distance from the source
- O2 minimum range 3% to 5%
FIGURE 3. Study data set sample events and % attenuation of benzene and TPH
study data set events (102 benzene, 71 TPH; n=# vapor sample events).
few events that exhibited little or no
attenuation took place in coarse-
grained soil.
Benzene events exhibiting signif-
icant attenuation occurred beneath
buildings, pavement, and bare soil.
Of the two events that did not exhibit
attenuation, one was beneath a build-
ing basement (Roggemans, et al.,
2001) and one beneath pavement
with no clean soil overlying a shallow
contaminant source (Roggemans, et
al., 2001).
The event exhibiting insignificant
attenuation may not be representa-
tive of attenuation beneath buildings
because the depth at which vapor
samples were collected is in dispute
(Kremesec, 2004). All of the TPH
events beneath a bare soil ground
cover exhibited significant attenua-
tion. TPH events that did not exhibit
attenuation were overlain by pave-
ment (Roggemans et al., 2001) and
one building basement (Laubacher, et
• continued on page 24
23
-------
CtiHchella, California, COA-2 (Rirfe, ct »!., ZO»Z>
Lnavcd. Free product an GW
DTWMS, Benwn* .M-aO.OOOl
COJ uit O2
5 10
Free Product
\ tt, \
O2, % bj volume BcKMIK V«(»r CafuottraMlHU. ae'm.l
•COX % by wl
FIGURE 4. Fyp/ca/ signature of significant attenuation due to ~5 ft clean overlying
coarse soil (depleting benzene, strong 02 supply and C02 production).
0.01 0.001 0.0001 0.00001
Attenuation Factors
95 percentite <0.01
FIGURE 5. Magnitude of significant attenuation for benzene and TPH
(n = # vapor sample events).
m Vapor Attenuation from page 23
al., 1997). According to the reports for
which sub-slab data were available,
none of the buildings evaluated had
vapor intrusion from the underlying
petroleum source.
The above-referenced building
events where significant attenuation
was not observed were the only
24
events in this data set that exhibited
strong depletion of oxygen, or an
anoxic zone beneath the basement
slabs of the buildings. All other
events beneath buildings, where sub-
slab soil gas oxygen was analyzed,
and depth to contaminant sources
ranged from about five to twenty feet
below grade, exhibited near-atmos-
pheric sub-slab oxygen concentra-
tions (Environ, 2005; Fischer, et al.,
1996; Hers, 2000; Sanders and Hers,
2006; Secor, 2004). These findings,
therefore, indicate that the area
beneath most buildings is sufficiently
oxygenated to biodegrade petroleum
hydrocarbon vapors if there are at
least five feet of overlying clean
coarse-grained soil or two feet of fine-
grained soil.
The magnitude of petroleum
vapor attenuation is shown in Figure
5. The greatest number of events
where benzene and TPH vapors
attenuate have attenuation factors
ranging from 0.01 to 0.00001. At the
95 percentile, the benzene attenua-
tion factor is less than 0.01.
Concentration versus
Distance
The larger data set includes vapor-
sample events over a diverse range of
site conditions and -beneath build-
ings. The data indicate significant
attenuation of benzene vapors (98
percent of the events) and TPH
vapors (92 percent of the events). The
data presented in this study
strengthen the case that the vapor
intrusion to the indoor air pathway is
not likely complete for petroleum
vapors if there are at least five feet of
clean coarse-grained soil or two feet
of fine-grained soil overlying the con-
taminant source.
Vapor samples should therefore
be collected at depths and intervals
less than five feet in order to deter-
mine if vapors are being degraded
and to verify attenuation. If the path-
way needs further evaluation, it is
reasonable to permit vapor sampling
within 10 feet of a building footprint
to avoid intrusive activities like sub-
slab sampling. This approach has
also been suggested by Hartman
(2005,2006) and NJDEP (2005).
Finally, data presented in this
paper show that a conservative sub-
surface bio-attenuation factor of 0.01
can be applied to the dissolved- and
vapor-phase concentrations for
petroleum hydrocarbons. Hence, the
allowable concentrations shown in
Table 2c of the EPA 2002 guidance
can be increased by a factor of 100
(Table 1). If vapor and groundwater
concentrations exceed those levels,
the vapor-intrusion pathway may
require further evaluation.
-------
LE 1 J&E-Predicted Residential Generic Screening Levels for Soil Gas
and Groundwater (EPA, 2002, Table 2c Question 4) and Proposed Screening
Level Concentrations Based on Observed Field Attenuation Factor of 100
Indoor
An
Concentra
tion,10-
06Rrsk
Level and
ID
Chemical Hazard
Index
ueta3
Benzene
Total
Petroleum
Hydrocarbons
(1,3,5-
Tnmethyfljenz
ene surrogate)
6DDE-00
Depth
Below
Grade or
Building
Slab
feet
i
10
5
ID
J&E Predicted Concentrations to Meet
Target Indoor Air Concentrations
Coarse-
Gramed
Fine-
Grained
Soil Gas
SIDE-
DO
310E+
Dl
6DOE+
01
6DOE+
D2
30DE+0
2
5DDE-K1
2
5DOE+D
3
9 DOE-HI
3
Coarse-
Grained
Fine-
Gnuned
Gzoiuidwater
Concentration 1
UE/L
231E-DD
2SDE-00
2 50E-H11
9 87E-H11
312E+
01
323E+
01
102E+
03
106E+
03
Proposed Screening Level Concentrations
Attenuation Factor = IDO(DDl)
Coarse-
Grained
Fine-
G rained
Soil Gas
Ughn3
31DE+
02
310E+
03
6DDE+
03
600E+
04
3DOE+04
5DOE-H14
5DDE+05
9DDE+05
Coarse-
Gnuned
Fme-
Gnuned
Gronndsrater
Concentration
231E+
02
28DE+
02
25DE+
03
9S1E+
03
3 12E-W3
3 23E+03
102E-HJ5
1D6E+05
The conclusions from my data set
are consistent with recent conclusions
by Abreu and Johnson (2005), using 3-
dimensional modeling. They have
shown that a 10-fold reduction in the
source strength from 200 mg/L-vapor
to 20 mg/L-vapor leads to a 6 order of
magnitude reduction in the attenua-
tion factor. They also show that for
the 20 mg/L-vapor scenario, vapors
are reduced by a factor of 1,000 within
a few feet of the source. Since the typi-
cal soil-gas concentrations in my data
set are less than 2 mg/L-vapor for
benzene and 20 mg/L-vapor for TPH,
it is reasonable to expect that vapor
intrusion from typical hydrocarbon
sources is not likely to occur unless
the source is within a couple of feet of
the receptor.
Some states have accounted for
bioattenuation by reducing EPA's
proposed horizontal distance criteria
of 100 feet to a building to a lesser
distance (30 feet has been proposed
by NJDEP, 2005; NHDES, 2005;
UDEQ, 1997). In my opinion, these
criteria are still too conservative and
setting fixed distances may not be the
best approach for determining when
the vapor-intrusion pathway is com-
plete and needs to be assessed.
My data set shows that vapors
are biodegraded and completely
attenuated beneath most of the build-
ings, even those that overlie very
strong sources, when there is suffi-
cient clean soil overlying sources.
Because distance is critical in the
attenuation of petroleum sources, a
better method of determining appro-
priate distances at which certain
source concentrations may or may
not constitute a vapor intrusion risk
is to plot source concentration versus
distance from the receptor. I am cur-
rently compiling my data set to gen-
erate such a plot, and I will report on
it in a future issue of LUSTLine, so
stay tuned.il
Robin Davis is a project manager until
the Utah Department of Environmental
Quality, Leaking Underground Storage
Tank program and member of EPA's
petroleum hydrocarbon vapor intrusion
workgroup. She specializes in fate and
transport of petroleum hydrocarbons,
and data acquisition, reduction and
analysis, most iccentli/ for the vapor
intrusion exposure pathway. Robin can
be reached at (801) 536-4177,
rvdavis@utah.gov
Disclaimer
Any opinion expressed herein is that
of the author and does not represent
opinions of the State of Utah, U.S.
EPA, or authors cited.
Acknowledgements
Dr. Blayne Hartman of H&P Mobile
Geochemistry provided extensive
review and contributions to this arti-
cle. I am also grateful for the support
and technical discussions provided
by Joe Vescio and Henry Schuver,
U.S. Environmental Protection
Agency, John Menatti, Utah Depart-
ment of Environmental Quality, Dr.
Ian Hers, Tom Higgins, Minnesota
Pollution Control, Eric Nichols, Dr.
Todd Ririe, and Dr. Robert Sweeney.
References
Abreu, L. and Johnson, P., 2005, Modeling the effect
of aerobic biodegradation on vapor intrusion. Pre-
sented at the National Ground Water Association
and American Petroleum Institute Petroleum
Vapor Intrusion Workshop, Costa Mesa, California,
August 17, 2005
Davis, R., 2005, Making sense of subsurface vapor
attenuation in petroleum hydrocarbon sources.
LUSTLme Bulletin 49, March 2005
Environ, 2005, Health Risk Assessment-On-Terminal
Areas, Mission Valley Terminal, 9950 and 9966 San
Diego Mission Road, San Diego, California. Sep-
tember 8,2005
Fischer, M.L, Bentley, A.L , Dunkm, K A., A.T , Hodg-
son, Nazaroff, W.W, Sextro, R.G., and Daisey, J.M.,
1996, Factors Affecting Indoor Air Concentrations
of Volatile Organic Compounds at a Site of Subsur-
face Gasoline Contamination. Environmental Science
and Technology 30.10. 2448-2957
Freeze, R.A and Cherry, J A., 1979, Groundwater,
Prentice-Hall, Inc publishers
Hartman, B., 2005, 2006, personal communication
Hers, I, Atwater, J , Li, L and Zapf-Gilje, R 2000.
Evaluation of vadose zone biodegradation of BTX
vapours, Journal of Contaminant Hydrology, 46 (2000)
233-264.
Johnson, PC and Ettmger, R A., 1991, Heuristic
model for predicting the intrusion rate of contami-
nant vapors into buildings Environmental Science
Technology 25-1445-1452.
Kremesec, V., 2004, personal communication
Laubacher, R. C , Bartholomae, P., Velasco, P. and
Reismger, H. J., 1997, An evaluation of the vapor
profile in the vadose zone above a gasoline plume,
Proceedings of the Petroleum Hydrocarbons and
Organic Chemicals in Ground Water, November
New Jersey Department of Environmental Protection
(NJDEP), 2005, Vapor Intrusion Guidance, October
2005
New Hampshire Department of Environmental Ser-
vices (NHDES), 2000, Draft Residential Indoor Air
Assessment Guidance Document March 2000 Revi-
sions.
Ririe, G T., Sweeney, R E , and Daugherty, S J ,
2002, A comparison of hydrocarbon vapor attenua-
tion in the field with predictions from vapor diffu-
sion models Soil and Sediment Contamination, AEHS
publishers, No 11(4) 529-554.
Roggemans, S , Bruce, C.L, Johnson, P C and John-
son, R.L., 2001, Vadose zone natural attenuation of
hydrocarbon vapors: An empirical assessment of
soil gas vertical profile data. American Petroleum
Institute, December 2001, No 15
Sanders, P and Hers, I 2006. Vapor intrusion in
homes over gasoline-contaminated groundwater.
Ground Water Monitoring & Remediation 26, no. I/
Winter 2006/pages 63-72.
SECOR International, 2004, Soil vapor sampling
report for Hal's Chevron, 138 W. Main St, Green
River, Utah, March 5, 2004, prepared for the Utah
Department of Environmental Quality, Leaking
Underground Storage Tank Section.
U S. Environmental Protection Agency (EPA), 2002,
Draft Guidance for Evaluating the Vapor Intrusion to
Indoor Air Pathway from Groundwater and Soils (Sub-
surface Vapor Intrusion Guidance, November).
Utah Department of Environmental Quality (UDEQ),
1997, Guidelines for Utah's Tier 1 Risk-Based Corrective
Action Utah's Guide for Screening Petroleum-Contami-
nated Sites
~25
-------
nically Speaking
by Marcel Moreau
Marcel Moreau is a nationally
recognized petroleum storage specialist
whose column, Tank-nically Speaking,
is a regular feature o/LUSTLine. As
always, we welcome your comments and
questions. If there are technical issues
that you would like to have Marcel
discuss, let him know at
marcel.moreau@juno.com
Operator Training: Boon or Bust?
Among other things, the Energy Act of 2005 requires what
many regulators have long considered a pipe dream - manda-
tory UST operator training. But sometimes a dream come true
can turn into a living nightmare. Whether the operator training pro-
visions of the Energy Act ultimately prove to be a boon or a bust will
be determined largely by how these programs are designed. Without
proper attention to the implementation aspects of training hundreds
of thousands of people each year in the fairly esoteric world of UST
operation, I fear many owners, operators, and regulators will come to
rue the day the Energy Act became law.
Tuning the Tank Rules
The Energy Act of 2005 was signed
some 21 years after President Reagan
enacted Subtitle I of the Resource
Conservation and Recovery Act that
created the federal (and consequently
most state) environmentally based
underground tank regulatory pro-
grams. Since then, Congress has paid
scant attention to underground tank
regulations.
One could surmise that the
return of this congressional focus on
underground tanks portends some
significant changes in the program.
Interestingly enough, except for the
secondary containment/insurance
provisions for storage systems
installed in proximity to water sup-
plies, the new law does little to
change the fundamental structure of
the 1984 Subtitle I or the 1988 EPA
tank regulations. The focus appears
to be better implementation of the
1988 rules:
The periodic inspection require-
ment is aimed at leveling the
playing field by being sure
everyone is in compliance.
The delivery prohibition provi-
sions are an attempt to facilitate
enforcement of the existing 1998
upgrade requirements.
The owner/operator training
provisions make a requirement
out of what the initial rules had
(mistakenly) taken for granted:
that owner/operators would
know how to operate their
storage systems.
So just what are the benefits of
training operators and what are the
challenges to accomplishing this
goal? Let me see...
The Dream
Well-trained UST operators could, in
theory, provide substantial benefits
to UST programs. Knowledgeable
operators would understand how
their UST systems worked, be able to
identify problems, and know what to
do when their systems weren't work-
ing. They would be able to conduct
routine inspections for leaks (e.g.,
inside dispensers) and detect condi-
tions that need attention (e.g., debris-
filled spill buckets, leaking hoses and
nozzles). They would know the regu-
latory requirements and have the
required paperwork readily available
and neatly organized for inspection.
By responding appropriately and in a
timely manner to alarm conditions,
they would be able to detect releases
sooner rather than later. The UST
inspectors' job would be a cakewalk.
The Numerical Challenge
Achieving the benefits of a well-
trained UST operator population will
not be easy. First off, there is the sheer
volume of people who would need to
receive some form of training.
Assuming roughly 250,000 facilities
and two to three people per facility (a
conservative estimate), we're talking
about 500,000 to 750,000 people. And
with convenience-store employee
turnover at slightly over 100 percent
per year, we're talking about provid-
ing roughly this same amount of
training each year. This is not a trivial
task. There are many questions to be
answered:
• Who will provide this training?
Regulators? Private-sector in-
structors? Employers?
• How will this training be delivered?
Classrooms? Self-study work-
books? On the job? On the Web?
On the fly?
• How will this training be docu-
mented (surely the requirement will
be meaningless if there is no way to
distinguish the trained from the
untrained)? Certificates? ID
cards? Employment records?
Electronic databases?
26
-------
All of these options are possible;
indeed all of these options will likely
be needed to meet the challenge of
training this widely dispersed popu-
lation with training needs that are
variable, depending on the idiosyn-
cratic characteristics of the storage
systems they operate.
The Great "Who Is the
Operator?" Challenge
The Energy Act specifies that there
should be three levels of UST-opera-
tor competency, presumably based
on the level of responsibility of differ-
ent types of tank operator. There will
be a strong temptation to make every
UST operator scenario fit this statu-
tory "mold." But there are so many
different possible distributions of
responsibility that it would be a mis-
take to create a caste system of tank
operators and try to force every situa-
tion to fit it.
There will be mom-and-pop
retail operations or commercial-fleet
fueling facilities (e.g., construction
companies, car-rental agencies)
where a single person is responsible
for all facets of operating a specific
storage system. In this situation, all
categories of UST operator will be
collapsed into a single person.
There will be other operations
where the fuel ordering is handled
remotely by a third party, the leak
detection is handled by a remote
monitoring service, and equipment
maintenance is handled by yet
another third-party provider, so that
the on-site personnel are left primar-
ily with the responsibility of knowing
how to identify different levels of
emergencies and whom to call to
respond to them. In this situation,
there may be many more than three
categories of tank operator.
There are likely many thousands
of different variations on the theme
of "tank operator," so how do you
decide who the operator is and what
he or she is responsible for?
The "What Should the
Operator Know?" Challenge
I expect a roomful of regulators or
tank owners could argue at great
length about this topic, and I expect
such arguments will occur with
increasing frequency in the next few
years. I submit that only information
relevant to the facility for which the
operator is responsible should be
taught.
There is no credible reason to try
to teach an operator the ins and outs
of groundwater monitoring when he
is in charge of a facility with no such
system. There is no reason anyone
should know about leak detection for
a suction-pumping system if she is
responsible for a facility with pres-
surized piping (granted, it's impor-
tant know at a basic level how to tell
which type of pumping system is
present at a given facility). This
approach, however, would require
what amounts to individualized
training for each UST operator, a
daunting proposition when the num-
ber of people to be trained is consid-
ered.
"Without proper attention to the
implementation aspects of training
hundreds of thousands of people
each year In the fairly esoteric world
of UST operation, I fear many
owners, operators, and regulators
will come to rue the day the Energy
Act Became law."
The alternative—teaching large
classrooms of operators all aspects of
the rule and all variations of the mul-
titude of UST technologies in use
today—will only produce bored and
confused operators rather than
knowledgeable ones. I would also
submit that operators need to know
much more about storage systems
than what is contained in 40CFR280
or state-level regulations. The Energy
Act presents a great opportunity for
regulators and the regulated commu-
nity to jointly define what an operator
needs to know to ensure that facilities
are both meeting regulatory require-
ments and being operated safely.
The Recordkeeping
Challenge
If not carefully implemented, the
operator-training requirement may
simply become another record-keep-
ing nightmare for tank owners and
an enforcement quagmire for regula-
tors. Tank owners will have to retain
some type of documentation show-
ing that each employee who falls in
some category of "tank operator" has
received the appropriate training.
Regulators will need to add to their
inspection checklists verification that
all current tank operators have been
properly trained. And what reason is
there to believe that compliance with
the training provisions of the Energy
Act will be any better than compli-
ance with the leak detection or
record-keeping provisions of the
existing tank rules?
Knowing versus Caring
The theory behind the training provi-
sions of the Energy Act appears to be
that if operator ignorance is the prob-
lem, then operator training is the
answer. I would suggest, however,
that making tank operators knowl-
edgeable (which I acknowledge is a
noble goal) is very different than
making them better tank operators.
I happen to be pretty ignorant
about most matters relating to profes-
sional sports. You could possibly
remedy my ignorance about sports
with some special training, but that is
not going to make me care any more
about who plays in next year's Super
Bowl. In other words, it is not going
to change my behavior.
Like the ol' horse and the water
trough, you can lead me there, but
you can't make me into a football fan
any more than you can make a con-
venience-store manager into a tank
operator by simply plopping the
right information in front of us.
If training is going to have any
effect on UST system operation
besides multiplying several-fold the
amount of paper records that are
supposed to be maintained, tank
operators will need to be held
responsible for what they have been
trained to do. But somehow I don't
think bringing enforcement actions
against individual UST operators is
going to be a very popular activity
among regulators.
Consistency versus Chaos
By delegating the implementation of
UST-operator training to states, the
Energy Act has created what could
become a major headache for the tank
owner who owns tanks in multiple
states. If each state creates state-spe-
cific operator-training requirements
(e.g., different training content, differ-
ent record-keeping requirements, dif-
• continued on page 28
27
-------
• Tank-nically Speaking
from page 27
ferent amount of time before a new
hire must be trained), it will create a
patchwork of regulatory perplexity
that will make compliance, even by
willing tank owners, a Herculean
task. Consistency of requirements
among states would simplify the tank
owner's life and ultimately that of the
tank regulator who must enforce the
requirements.
What's My Solution?
I believe the path to increasing UST-
operator knowledge and motivation
lies in creating a class of professional
UST operators. This professional UST
operator becomes the person with
primary responsibility for the USTs
in his or her care. This person also
could assume responsibility for des-
ignating who among the on-site per-
sonnel should shoulder what
responsibilities as well as provide the
site-specific training required to carry
out these responsibilities.
This approach is much like Cali-
fornia's recently implemented desig-
nated UST-operator program. For a
more complete description of this
type of approach, see LUSTline #40,
Of Square Pegs and Round
Tanks...What If Tank Operators Really
Knew How to Operate Tanks?
What's Your Solution?
Whether you are an UST regulator,
UST owner, or UST operator, write
and let me know at marcel.moreau®
juno.comM
Marcel's Short List of
What a Tank Operator Needs to Know
How to respond to small spill incidents
How to respond to substantial spill incidents
How to respond to major spill incidents
How to respond to accidents involving fueling equipment (e.g., drive-offs, colli-
sions with dispensers)
How to shut down or disable faulty equipment, ranging from individual nozzles,
individual dispensers, individual products, or the entire fueling facility
How to respond to customer reports of equipment malfunctions (e.g., nozzle not
shutting off)
How to respond to customer reports of slow product flow
How to identify and respond to inappropriate customer fueling behavior (e.g.,
filling glass containers with gasoline, or filling plastic gas cans in the back of a
pickup truck)
How to respond to various fuel-related ATG alarms (e.g., overfill, fuel alarm, high
product, low product)
How to inspect hanging hardware for proper condition/operation
How to look inside a dispenser for problems
How to inspect spill-containment manholes
How to keep inventory-control records
How to tell how much fuel an underground tank can safely hold
Type(s) of leak detection present (e.g., automatic tank gauge, secondary contain-
ment, line leak detector, etc.) at a facility and some basic information on how it
functions and how a leak is indicated (e.g., alarm, slow flow)
How to perform leak detection if it is not automatic (e.g., schedule a tightness test,
sample groundwater wells)
What leak detection records to keep, how to obtain/create them, where to store
them, and how long to keep them
How to calculate how much fuel to order
Some basic information about how the storage system works (tank, piping,
dispenser, pump)
Missouri Launches First-ln-Nation Effort to Educate Fuel Tank Operators
Missouri has taken a bold step to help fuel storage system owners and operators understand and operate their equip-
ment. Missouri's Petroleum Storage Tank Insurance Fund (PSTIF) insures 85 percent of the UST facilities in the state. "The
key to preventing leaks or detecting them early is knowing how your equipment works. But it's very difficult for any company
or local government that owns tank sites to train its employees in all of the many areas they are supposed to know," says
Carol Eighmey, Executive Director of PSTIF.
"We are a state trust fund," says Eighmey, "so we have an obligation to the citizens of Missouri to do what we can to
help prevent pollution. That means we need to keep fuel in the tanks and out of the ground as much as possible. It's a win-
win-win situation for the tank owners, the trust fund, and the environment"
PSTIF set out to solve the problem of educating the people responsible for fuel-storage systems about how their sys-
tems work, how their leak-detection equipment functions, and how to respond when an alarm sounds. To get the maximum
number of people involved, the PSTIF Board decided to purchase the Petroleum Equipment Institute's (PEI) on-line Learning
Center courses in bulk and offer them for free to PSTIF participants.
"If we succeed in preventing only one gasoline leak, we will more than make up the cost of providing the courses," says
Eighmey. Missouri is the first state to tackle the problem by offering free courses to people who work at facilities insured by
the state's tank fund.
Anyone responsible for fuel tanks insured by PSTIF can take courses for free by logging on to www.pstif.org. To find out
more about the PEI Learning Center go to www.pei.org/learn. •
28
-------
Update—ENERGY POLICY ACT
by Ellen Frye
Title XV, Subtitle B of the
Energy Policy Act (EPAct) of
2005 (entitled the Under-
ground Storage Tank Compliance
Act of 2005) contains amendments to
Subtitle I of the Resource Conserva-
tion and Recovery Act—the original
legislation that created the UST pro-
gram. This new law significantly
affects federal and state underground
storage tank programs and requires
major changes to them. Gas station
owners and operators, as well as
other non-marketers who own
underground storage tanks, will be
impacted by the changes U.S. EPA
and states make in their tank pro-
grams as a result of the law.
Some of the provisions require
implementation by August 2006; oth-
ers will require implementation in
subsequent years. To implement the
new law, the U.S. EPA Office of
Underground Storage Tanks (OUST)
and states are working closely with
tribes, other federal agencies, tank
owners and operators, and other
stakeholders to bring about the man-
dated changes affecting under-
ground storage tank facilities. At this
point, however, virtually everything
associated with implementing and
funding the provisions of the EPAct
is still being finalized. So, until we
can report hard-and-fast decisions on
the EPAct, we will simply offer an
ever-so-brief update.
Work Groups Addressing
Grant Guidelines
Since Fall 2005, OUST has been work-
ing with the states through the
Association of State and Territorial
Solid Waste Management Officials
(ASTSWMO) to implement the EPAct.
Fourteen work groups, composed of
representatives from state and EPA
regional and headquarters UST/
LUST programs, as well as tribes for
the tribal strategy, were organized to
develop grant guidelines that will
apply to any state receiving federal
UST money from EPA.
The work groups cover the fol-
lowing specific issues associated with
the EPAct: Cost Recovery, Delivery
Prohibition, Financial Responsibility
and Installer Certification, Inspec-
tions, LUST Allocations, LUST Guid-
ance, Operator Training, Public
Records, Secondary Containment,
State Fund Diversion, State Fund
Soundness, Tribal Strategy and
Report to Congress, Federal UST
Compliance Report, State UST Com-
pliance Report.
As draft guidelines are com-
pleted they will be posted on OUST's
website for a 30-day public comment
period. The first drafts will likely
appear in May. Keep tabs on
progress by visiting http://www.epa.
gov/oust/fedlaws/nrg05_01.htm.
A New Twist on Funding
Among other things, the EPAct
expanded eligible uses of the Leaking
Underground Storage Tank (LUST)
Trust Fund, including meeting the
requirements for preventing as well
as cleaning up leaks. But now it looks
as though FY 2007 LUST Trust Funds
can only be used to perform tradi-
tional Subtitle I tank cleanup tasks,
not to meet any of the new leak-pre-
vention provisions in the EPAct. The
reason for this restriction is a clause
inserted into a Transportation law
enacted soon after the Energy Policy
Act. This provision prevents further
funding of the LUST Trust Fund if
any of this money is spent on any
uses not authorized in the original
legislation (1986).
With the congressional FY 2007
appropriations cycle under way, U.S.
EPA is asking for a little less than last
year's $73 million in LUST funds in
its FY 2007 budget; this year's total
request was for $72,759 million. The
agency is also seeking an additional
$26 million with the baseline $11 mil-
lion normally found in the UST
STAG grants to cover non-cleanup
UST functions—primarily those of
enforcement and compliance.
EPA's FY 2007 budget request is
looking primarily at the catch-up
inspections required in the first two
years after enactment of the EPAct.
This funding is also to be used to
support work in initiating other new
tasks mandated by the EPAct for
operator training, delivery prohibi-
tion, and secondary containment or
expanded financial assurance. The
scope and universe of some of these
new activities have not been estab-
lished yet (e.g., the number of catch-
up inspections that must be
completed before August 2007).
OUST has other issues related to
the new UST funding that will need
to be sorted out over the next few
months, including how EPA should
distribute funds—by the same
amount to all states, by a formula
based on work to be accomplished,
on the basis of need (e.g., number of
catch-up inspections required), or
some combination (e.g., expanded
equal core amount plus some work-
load and current need formula).
All or Nothing?...
Not Necessarily
EPA originally thought the language
in the EPAct meant that any state
receiving funding under the new
amendments to Subtitle I would be
required to accomplish all of the new
and existing mandates, and that a rec-
ognized failure to accomplish any of
the mandates could result in the loss
of all federal funding provided under
the authority of Subtitle I (the "all-or-
nothing" scenario). EPA has now con-
cluded that a failure to meet a
requirement in the law does not nec-
essarily mean the agency must either
terminate a funding arrangement or
automatically withhold future fund-
ing to a state. Rather, if a state materi-
ally fails to meet a statutory
obligation, EPA may use the discre-
tion afforded it under the enforce-
ment provisions of the existing grant
regulations (40 CFR Section 31.43).
These enforcement provisions
provide the agency with considerable
flexibility. It's important to note that
states receiving funding under Subti-
tle I are still required to meet the new
mandates, but the ramification of
missing a deadline is not necessarily
the loss of all Subtitle I funding. EPA
will be able to consider state-specific
circumstances, including a state's
progress and plan for meeting the
requirements.
To see the full text of the new leg-
islation, go to: hitp://frwebgate.access.
gpo.gov/cgibin/getdoc.cgi?dbname=109_
cong_public_laws&docid=f:publ058.109.
pdf (scroll to Title XV - Ethanol and
Motor Fuels, Subtitle B - Under-
ground Storage Tank Compliance,
pages 500-513 of the pdf file) for the
UST/LUST provisions. •
29
-------
by Kevin Bmckney
The Nez Perce Tribe has a Direct
Implementation Tribal Cooper-
ative Agreement with EPA
Region 10 to conduct an UST compli-
ance assistance program on the reser-
vation. The 750,000-acre north Idaho
reservation has 20 active UST facili-
ties, the second highest number of
any tribe in the Region.
The Nez Perce Tribe has two
half-time UST inspectors who have
completed or are currently enrolled
in both a Region 10 and an Intertribal
Council of Arizona UST inspector
certification-training program. In
addition to the UST program, the
inspectors share duties with the
CERCLA 128(a) State and Tribal
Response Program to address
RCRA/CERCLA issues other than
USTs (e.g., aboveground storage
tanks). The tribe's Water Resources
Division hosts both of these inspec-
tion and capacity development pro-
grams.
The tribal inspectors are cur-
rently providing compliance assis-
tance to UST owner-operators, while
EPA is conducting federal enforce-
ment activities. In addition, the tribe
has developed a management strat-
egy of having at least two people
working to maintain program conti-
nuity in case something happens to
one employee—after all, none of us is
here permanently!
The tribe has successfully negoti-
ated the cleanup of three abandoned
and one inability-to-pay UST facili-
ties on the reservation, utilizing the
services of the Alaska-based, tribally
owned national UST contractor, Bris-
30
tol Environmental & Engineering
Services Corporation. A total of 13
tanks will be decommissioned during
summer 2006. Any associated soil
cleanup and remediation will utilize
landfarming to biodegrade the conta-
minants. Groundwater contamina-
tion does not at present appear to be
a serious issue at these facilities.
The Nez Perce Tribe agreed to
pursue the compliance assistance
program primarily to protect the
water resources of the reservation.
The Nez Perce are historically a fish-
ing tribe, utilizing anadromous
salmon and steelhead trout that
annually returned from the Pacific
Ocean to spawn in their natal streams
in northern Idaho. The tribe is now
playing a pivotal role in the restora-
tion of these threatened and endan-
gered fish runs, which have been
decimated by downstream dams,
spawning habitat degradation, over-
fishing, and changing environmental
conditions in the ocean.
Much of the reservation overlies
a federally designated sole-source
aquifer in the Columbia River Basalt
Group. Petroleum contamination of
these bedrock aquifers would be dif-
ficult to remediate, and replacement
of water wells would be expensive.
The groundwater is hydrologically
interconnected with surface water
and readily moves back and forth
based on local hydrogeologic condi-
tions.
The Nez Perce are developing
environmental codes to regulate
USTs and other potential contami-
nant sources affecting soil, surface
and groundwater, and air quality on
the reservation. The tribe is propos-
ing a nondegradation policy for
aquatic resources. This has implica-
tions for the UST Compliance Act 01
2005, which requires additional mea-
sures to protect groundwater wher
USTs are located within 1,000 feet ol
a water supply.
However, the Nez Perce believe
that all potable water resources
should be protected—not jus)
resources that are adjacent to a watei
well—using the nondegradation pol-
icy. Potable water is an increasingly
valuable commodity, and many
water-rich areas of the country are
experiencing water shortages. One
need only compare the price of bot-
tled water to that of gasoline to know
that potable water is a precious
resource that should not be squan-
dered.
The Nez Perce Tribe subscribes
to the philosophy of responsibility
for its actions to the seventh genera-
tion of children yet unborn. This is
not meant to imply that the reserva-
tion should be wilderness; indeed,
the Nez Perce rely on agriculture,
timber, and gaming in addition to
fishing, hunting, and gathering roots
and berries to support the Tribe and
its members. The tribe is located in a
rural area and depends on petroleum
to fuel its equipment and cars.
The tribe thanks EPA Region 10
for supporting its capacity develop-
ment program to safely manage USTs
on the reservation and for allowing
members to play a direct role in
protecting our shared natural
resources. •
-------
•
AftifHifl
a
Ji ,
in complltnei, en thtlr fill '~
TanWetpef ts tht ijfjriliaiet •
and fie private §«ofV fl» was f pint
from U,S. iPA,
by tfte DIQ UST Unit
roattonttehnoIpB fflviiten, ai«J
tive,,LLC, a wfwlly ownei subsidiary of ifwiifuniw
provider MIC IGOV), Twkftrtper it
In aft* months wt wJH haw the opportunity pro-
gram,* says Bill ftite of tht: atQ UST (ML
somi reviews from kfiwlfdpabji replatofs to iiip its
improve thi Don't to in f hi uit or
work with any lieilty," The Mforination is pullii 'mA ttt
datotosi Iirtormttion fe secure, Yoy can also
Department of Transportotton site with '
brule@mt.gov
U. S. Settles UST System Lawsuit
Against New York City
The United States Attorney for the Southern District of
New York and U. S. EPA Region 2 recently settled a civil
lawsuit against the City of New York involving UST viola-
tions in connection with the City's UST systems. The set-
tlement, filed in Manhattan federal court, requires the City
to pay $1.3 million in civil penalties and to bring substan-
dard tank systems into compliance with federal law
(RCRA Subtitle I). The Consent Decree also requires the
City to undertake an additional environmental project to
improve the City's ability to identify releases from its
USTs.
The United States charged in the lawsuit that, from at
least 1997, the City has been violating federal UST
requirements in connection with its UST systems. As
alleged In the complaint, New York City owns at least
1,600 USTs in at least 400 locations throughout the met-
ropolitan area, including all five boroughs.
The lawsuit charged that New York City has for many
years committed numerous violations of the federal UST
regulations issued, including failing to upgrade or close
noncompliant UST systems; provide proper methods to
detect releases of hazardous substances; and report,
investigate, and confirm suspected releases of regulated
substances.
As part of the settlement, the City is required to undertake
a multi-year project to monitor releases and suspected
releases from a central location for USTs owned and
operated by the Police Department, Fire Department, and
Department of Transportation. In addition, the Consent
Decree requires the City to upgrade or close noncompli-
ant USTs. •
LAJ.S.T.LINE Subscription Form
Name
Company/Agency _
Mailing Address _
E-mail Address
Q One-year subscription. $18.00.
Q Federal, state, or local government. Exempt from fee (For home dehvery.mclude request on agency letterhead.)
Please enclose a check or money order (drawn on a U.S. bank) made payable to NEIWPCC.
Send to: New England Interstate Water Pollution Control Commission
116 John Street, Lowell, MA 01852-1124
Phone. (978) 323-7929 • Fax: (978) 323-7919 • lustline@neiwpcc.org • www.neiwpcc.org
31
-------
FAQs from the NWGLDE... All you ever wanted to know about fyak detection, but were afraid to ask.
Automatic Mechanical Line Leak Detectors (MLLD) - Part I
In this installment of FAQs from the National Work Group on Leak
Detection Evaluations (NWGLDE), we discuss the operation of auto-
matic mechanical line leak detectors (MLLDs) with regard to product
type. The next part in this series will address the effects of piping type
and installation location on MLLD operations. This article does NOT
apply to electronic line leak detectors (ELLD). (Please note: The views
expressed in this column represent those of the work group and not
necessarily those of any implementing agency.)
Q. Why do some MLLD manufacturers offer MLLDs that are
certified for multiproduct (e.g., gasoline, diesel), and oth-
ers for diesel only?
A. When MLLDs were first introduced (circa 1960), they were
not tested to evaluate their performance with different prod-
ucts. At the time, there was only one type of piping in wide-
spread use: galvanized steel. Since 1991, MLLDs have been
required to meet the performance standard specified in the
U.S. EPA rule—to detect a leak of 3 gph at 10 psi with a
minimum 95 percent probability of detection and a maxir
mum 5 percent probability of false alarm. Because of this
requirement, some manufacturers introduced MLLDs
designed for use with diesel only to more reliably meet this
performance criterion.
Q. Can MLLDs be interchanged (e.g., diesel to gasoline or vice
versa)?
A. First, let's focus on the function of a MLLD. During the
pumping phase, a MLLD detects line leaks by metering the
flow into the line through a precisely sized orifice at a flow
rate slightly less than the U.S. EPA standard of 3 gph. If this
metered flow through the MLLD is greater than any down-
stream line leak present, the pressure in the line increases
and the MLLD opens to full flow. If this metered flow is less
than any downstream line leak present, the pressure in the
line does not increase, and the MLLD restricts the flow to 3
gpm when a dispenser nozzle is opened. When the flow is
restricted, the MLLD is said to have "tripped."
As long as the fuel being metered by the MLLD has the
same physical properties as the fuel the MLLD was
designed for, the MLLD will be able to properly detect leaks
in accordance with the EPA standard. For example, MLLD
units designed for use with diesel should be able to detect
leaks in tanks holding fuels with similar characteristics,
such as kerosene or jet fuels, because they have similar vis-
cosity signatures. However, if a MLLD unit designed for use
with diesel fuel were placed in a gasoline product pipeline,
the lower viscosity of the gasoline would cause the MLLD to
meter the gasoline into the pipeline at a flow rate above the
EPA standard, and a 3 gph leak would not be able to be
detected.
MLLDs designed for use with gasoline will detect leaks
within EPA standards for fuels with thicker viscosity signa-
tures, such as diesel or kerosene, because the fuel will
meter through the MLLDs at a flow rate belowthe 3 gph
EPA standard. This means that MLLDs designed for use
with gasoline and operating in diesel fuel, will be more sen-
sitive to leaks.
For example, a MLLD designed for diesel and placed in a
gasoline system might only be able to detect a 4 or 5 gph
leak at 10 psf and would therefore not meet the regulatory
requirements for flow rate. Conversely, a MLLD designed
for gasoline and placed in a diesel system might be able to
detect a teak of 1 or 2 gph at 10 psi, which does meet the
regulatory requirements for flow rate, and, in theory at least,
be able to detect even smaller leaks in diesel fuel than is
required by the EPA standard. Since all MLLDs have the
same diameter base and use the same thread, they can be
installed for use with inappropriate fuel types.
Therefore, it is important to review the NWLDE listings to
determine what products a MLLD can handle to ensure that
it is able to detect a leak within the EPA standard of 3 gph at
10 psi with a minimum 95 percent probability of detection
and a maximum 5 percent probability of false alarm. As an
inspector, the violation to watch out for would be a diesel
MLLD installed on a gasoline storage system.
(To be continued)
About NWGLDE
NWGLDE is an independent work group comprising 10 mem-
bers, including eight state and two U.S. EPA members. This col-
umn provides answers to frequently asked questions (FAQs)
NWGLDE receives from regulators and people in the industry on
leak detection. If you have questions for the group, please con-
tact them at questions@nwglde.org.
LUSJ.UNE
New England Interstate Water
Pollution Control Commission
116 John Street
Lowell, MA 01852-1124
Non-ProfitOrg.
U.S. Postage
PAID
Wilmington, MA
Permit No.
200
------- |