Ne-wr E-n.gla.nd. Interstate
Water Pollution Control
Commission
                        255 Ballardvale Street
                        Wilmington
                        Massachusetts
                        O1887
Bulletin 2
November

1997
LUST.
A Report On Federal & State Programs To Control Leaking Underground Storage Tanks
Where Does


The Buck  Stop,


After All?

Indiana Court of Appeals Rules
On the Lingering Issue of Liability

by Mary-Ellen Kendall


  If s an old story that's all too
  familiar to state underground
  storage tank (UST) program
personnel. The story begins dur-
ing the 1950s, '60s, and '70s. A
major oil company owns the USTs
at a corner gas station.
The property on
which the USTs are
located is owned
by a small busi-
nessman, who also
operates the USTs for
the oil company. All
of the product in the
tanks is supplied and owned by the oil com-
pany. The signs at the gas station and  the
employee uniforms are branded with the oil company's
name and logo. The gasoline is delivered to the station in
trucks that display the oil company's name and logo. The oil
company maintains control of both the USTs and the prod-
ucts sold at the gas station.
   But by the 1980s, things would change. The catalyst for
change, in part, was the issue of liability. By the mid-70s, the
oil companies began to take more and more notice of certain
fundamental and troublesome particulars associated with
USTs: The vast majority of the USTs installed prior to 1980
were constructed of steel; the buried bare steel tanks are sub-
ject to corrosion; any corroding tanks eventually leak; the
notion of leak prevention technology was mostly just a

                       • continued on page 2
                                            tank-nically Speakings Piping's Progres
                                             PA Alternative Integrity Assessment
                                            Guidance
                                            Qs and As
                                            Snapshots From The Field
                                            Field Notes: PEI RP100-97
                                            Upward Migration of Vapors
                                            Three-Dimensional Sampling
                                            MTBE Update
                                            HQ Update

-------
LUSTLine Bulletin 27
• Where Does the Buck Stop?
front page 2	
notion;  station employees  were
unable to detect slow leaks from
USTs by sticking the tank and keep-
ing inventory records; the engineer-
ing expertise and financial resources
necessary to solve  the problem of
slow leaks were beyond the capabili-
ties of independent gas station own-
ers;  and leaking gasoline was a
potential liability—it could find its
way into the groundwater and pol-
lute drinking water.
     As these facts became appar-
ent, the oil companies took steps to
address the problem at the stations
that they owned and operated. Cor-
rosion-resistant UST  technology
became more  prevalent. Oil com-
pany personnel were made aware of
leaking UST issues and, in many
cases,  were  trained  to  prevent,
detect, and respond to leaks. This
investment in storage system train-
ing and technology did not, how-
ever,  extend to the  independent
station owners, who continued to
operate their businesses using the
same signs, logos, and products, but
without the  awareness of potential
environmental problems.
           LUSTLine
          KMttof. Eltfn Frw
     In fact, the oil companies began
 to require the low-volume retailers
 to purchase the USTs outright. The
 tanks were often sold to the indepen-
 dent station owners at the bargain
 price of $1 per tank. At the time of
 the  sale,  there was generally no
 assessment of the integrity of the
 storage system or of the presence of
 any contamination. Because these
 retailers did not have the benefit of
 the knowledge about the environ-
 mental consequences  associated
 with bare  steel USTs, they didn't
 realize that the "bargain" they had
 just bought might make them liable
 for tens of thousands of dollars in
 damages and/or cleanup costs.
     The  ownership for many of
 these USTs was  transferred prior
 to the time that federal laws govern-
 ing  the  underground storage of
 petroleum were enacted. The Envi-
 ronmental Protection Agency pro-
 mulgated   regulations  that  set
 minimum standards for new tanks
 and required owners and operators
 of existing tanks to upgrade, replace,
 or close them. In anticipation  of a
 leak occurring, UST owners or oper-
 ators had to be able to demonstrate
 that they had the financial ability to
 pay for the costs associated  with
 cleaning up the release and compen-
 sating third parties. If a leak did
 occur, UST owners and operators
 were  held  responsible  for  the
 cleanup.
     As far as  the oil companies
 were concerned, they  were  not
 responsible for any UST-related lia-
 bility associated with facilities they
 had sold  to independent  station
 owners. It was quite another matter
 for the small retailers who unwit-
 tingly assumed that liability  and
 later found they  had a petroleum
 release on their hands. Many could
 not  afford  to  stay in business.
 Numerous USTs  at the closed gas
 stations  continued  to  leak,  un-
 checked. The oil companies that for-
 merly owned the USTs had not been
 determined to be responsible parties
 for these cleanups... until now.

 Will the Real "Operator"
 Please Stand Up?
 On August 19, 1997, the  Indiana
 Court of Appeals determined that oil
 companies that had supplied gaso-
line to USTs at an  independently
owned gas  station  were liable as
UST operators for groundwater cont-
amination caused by leaks from the
USTs.  (Shell  Oil  Co.  v. Meyer,
No.79A04-9512-CV-470  (Ind.  App.
Aug. 19, 1997).) It marked the first
time that a court had extended oper-
ator liability to gasoline suppliers
that neither owned nor maintained
daily control over the leaking USTs.
     The liability determination was
based on events that began in 1946
and culminated in the discovery of
contaminated drinking water wells in
1989. Fred Smith purchased property
in West Point for use as a gas station.
From 1946 until 1971, Smith leased
the gas station to others who oper-
ated it as a Shell station. The employ-
ees wore  Shell uniforms and sold
only Shell products. When the bulk
plant that supplied the  gasoline
switched from Shell to Union Oil in
1971, Union Oil's signs, uniforms,
and products were used at the sta-
tion. Smith died in 1979. The business
remained in operation as a Union Oil
station until it closed in 1981.
     Kim Meyer lived in West Point.
She noticed a petroleum odor in her
well water in 1989. The local health
department confirmed the presence
of gasoline constituents  in Meyer's
well and  wells  belonging to  five
other homes in the neighborhood.
The Indiana Department of Environ-
mental Management hired a contrac-
tor, who eventually determined that
the source of the contamination was
the   closed  gas station formerly
owned by Fred Smith..The Meyers
and  the other affected landowners
sued Shell Oil Company, Union Oil
Company,  the  current property
owner, and Smith's widow in 1993.
     The lawsuit had six grounds for
liability of the oil companies: negli-
gence, trespass, nuisance, strict lia-
bility for abnormally  dangerous
activities, operator liability, and anti-
dumping  (later  deemed not rele-
vant). The trial judge separated the
case into two parts: common law
grounds and operator  liability. A
jury heard the part of the case that
dealt with liability based on common
law grounds (negligence, trespass,
nuisance,  and  strict liability for
abnormally  dangerous  activities)
and decided that Shell and Union Oil
were not liable for damages.
     The judge then tried  the case
dealing with the issue of the oil com-
panies'  liability  under  Indiana's

-------
                                                                                            LUSTLine Bulletin 27
Underground  Storage  Tank  Act
("USTA").  In 1995,  the  trial  court
found Shell and Union Oil liable under
USTA as UST operators, awarded the
landowners $2,743,660.21 for cleanup
and $1,459,721.25 in attorneys' fees
and costs, and required Shell Oil to
pay 70 percent and Union Oil 30 per-
cent of the award to the landowners.
This August, on appeal, the Indiana
Court of Appeals affirmed the liabil-
ity of the oil companies as operators
oftheUSTs.

Defining the Issue
The court began its  analysis by
reviewing  the  stated purpose  of
USTA, which is to preserve, protect,
and enhance the environment  to
ensure clean air and water. The court
then reviewed the definitions  in
USTA to determine what the legisla-
ture intended. As is the case in many
states, the Indiana USTA adopted the
definition of operator from federal
law and regulations: "An operator is
a person in  control of,  or having
responsibility for, the daily operation
of the UST." In the absence of a statu-
tory definition of the terms "control"
and "responsibility," the court used
the "plain meaning" (i.e., the dictio-
nary definition) of the words.  The
court agreed with the oil companies
that they  did not  exercise  direct
influence (control)  over  the  daily
operation of the USTs.
     The court did  note, however,
that control was only one factor in
determining  liability. Because the
legislature created a two-part test (an
operator could either exercise control
OR have responsibility for the daily
operation of the USTs), the court then
looked  at  "responsibility  for" the
UST, which was defined as "moral,
legal, or mental accountability."
     This decision was the first time
that a court  had determined  what
constitutes operator liability under
state UST  laws, so there  were no
precedents applicable to Shell Oil.
Using the history of the  USTA and
court decisions concerning operator
liability under  both the Resource
Conservation and Recovery Act and
the Comprehensive Environmental
Response,  Compensation, and Lia-
bility Act to support its decision, the
court said that oil companies could
be held  liable as operators if they had
the authority to control a facility,
whether or not they actually exer-
     Using the history of the USTA
  u and court decisions concerning
   operator liability... the court said
  J32""             '     -       '
   that oil companies could he held
   liable as operators if they had the
  rr  authority to control a facility,
 Er  whether or not they actually
   ^   exercised that authority.
cised that authority.
     In applying this rule to the facts
in Shell Oil, the court found that the
oil  companies did have authority
over the USTs, as evidenced by their
ability to force independent stations
to stop selling leaded gasoline. The
court also found that the evidence
showed  that  the  oil   companies
attempted to  avoid liability when
they required the independent sta-
tion owners to purchase USTs in the
1980s. Quoting two prior cases from
the Indiana  Supreme  Court, the
appeals court concluded that:

  A business should bear its own
  costs, burdens, and  expenses of
  operation, and these should lie dis-
  tributed by means of the price of
  the  resulting product  and not
  shifted, particularly,  to  small
  neighboring property owners for
  them to bear alone. We can under-
  stand no sensible or reasonable
  principle of law for shifting such
  expense or loss to persons who are
  not involved in such business ven-
  tures for profit. (Shell Oil at 12.)

     The court found that attorneys'
fees included costs for lawyers, para-
legals, and expenses such as copying,
but that the  amount  approved for
fees by the trial judge was too high.
Although the landowners were enti-
tled to attorneys' fees for the USTA
trial, they were not entitled to attor-
neys' fees for the jury trial, which the
oil  companies won. The case was
remanded back to the trial court for a
redetermination of the issues relating
to the percent allocation  of costs
between the oil companies, the pro-
cedure by which the landowners will
notify the court and the oil compa-
nies each time they withdraw money
from   the  $2.7  million  cleanup
account (set up by the court tor 3]]
the costs incurred, except legal fees),
and the amount  of attorneys' fees
and costs that were incurred for the
USTA trial. It is highly likely that the
final decision on the issues in Shell
Oil will be appealed to the Indiana
Supreme Court.

Time and Many More Court
Cases Will Tell
While this case is significant because
no other court has held product sup-
pliers liable as UST operators under
federal or state UST laws, there are
two Iowa cases in which the major
oil  companies  were also found to
have liability for USTs. (Hagen v. Tex-
aco Refining & Mktg., 526 N.W. 2d 531
(Iowa  1995);  Iowa Comprehensive
Petroleum Underground Storage Tank
Fund Board v. Amoco Oil Co., 883 F.
Supp. 403 (N.D. Iowa 1995).)
     Dean Lerner of the Iowa Attor-
ney General's Office says that, as a
result of these two cases, Iowa was
successful in recovering $33 million
in settlements with 11 major oil com-
panies. Although the issue of opera-
tor liability was not the basis for the
decision in the Iowa cases, Dean says
that Shell Oil and the Iowa cases indi-
cate a willingness on the part of the
courts to find major oil companies
liable for USTs under state UST laws.
     These  three cases  provide a
new ending for  an old story. The sale
of USTs for $1 to unsuspecting inde-
pendent station owners has often
resulted in the transfer of a finan-
cially crippling liability  to either a
small business owner and/or the
taxpayers  (through state  cleanup
funds  or  the federal  LUST  Trust
Fund). If more courts in other states
make similar determinations of lia-
bility for UST releases, oil companies
may find in the future that the buck
stops with them, after all,  •
  Mary-Ellen Kendall, J.D., M.B.A., is
  the Financial Programs Manager for
  the Virginia Department of Environ-
  mental Quality. She is responsible for
  making liability and fund eligibility
    determinations for the Virginia   •
           UST Program.

-------
 LUSTLSae Bulletin 27
                        nically Speaking
                        by Marcel Moreau
                                                                      Marcel Moreau is a nationally
                                                                  ^recognized petroleum storage specialist ,1
                                                                  whose column, Tank-nically Speaking, *
                                                                   is a regular feature ofLUSTLine. As  ]
                                                                    always, we welcome your questions,  ;
                                                                   .! ^opinions, and technical interests.   "^
 PIPING'S  PROGRESS
 While Many Old Problems Have Been Solved, Piping's Not Out of the Hole Yet
A
    decade ago I wrote an article
(LUSTLine #7) entitled "The Weak
Spots in Piping." The article pointed
out that the leaking tank  problem
actually had a lot to do with piping
and that there were five major areas
of concern:
• The use of unions (steel pipe fit-
  tings used to join sections of steel
  pipe). Unions failed frequently
  when subjected to the stresses of
  the underground  environment.
• The use of swing joints (a com-
  bination of steel pipe elbows and
  short lengths of pipe). Swing
  joints were supposed to permit
  flexibility in steel piping, but they
  were often a source of leakage and
  piping failure.
• The use  of fiberglass/metal
  connections. These connections
  evolved during the period when
  there was no integrated system
  available for running piping from
  tank top to  dispenser. Various
  materials needed  to be joined
  together using  less than ideal
  methods.
• Improper installation. Haphaz-
  ard installation practices often led
  to premature, if not immediate,
  piping failure.
• Lack of testing. Testing piping
  prior to placing a system in opera-
  tion was an often overlooked
  installation step that  allowed
  many installation  errors  to go
  unnoticed until much later.

Today, I am  happy to report  that
many  of these problems are on the
way out:
* Because steel piping is rarely seen
  at  new  installations,  unions
  appear only at tank tops  and
  under dispensers, where they are
commonly surrounded by liquid-
tight sumps.
Swing joints have been replaced,
for the most part, by flexible con-
nectors.
Piping systems have evolved from
mongrels—made up of a variety
of materials cobbled together—to
integrated systems, which include
the  piping and all the specially
designed fittings necessary to get
the product from the tank to the
nozzle.
                                                                   4 FIBERGLASS REINFORCED
                                                                         STIC (FRP JPIPING
                                   The  advent of
                                   flexible   piping
                                   systems has sim-
                                   plified installa-
                                   tion practices, by
                                   greatly reducing
                                   the  number of
                                   field-installed Con-
                                   nections.
                                   Testing   piping
                                   prior to  burial
                                   and before plac-
                                   ing the piping in
                                   service is  now
                                   routine.

-------
                                                                                            LUSTLine Bulletin 27
 So are the problems with piping
 solved? While most of the old issues
 have been addressed, I would ven-
 ture to guess that most failures in
 storage systems installed in the last
 decade still originate in the piping.
 While  technological improvements
 have solved many problems, the fact
; remains that knowledgeable, consci-
 entious installers are a prerequisite
 for leak-free underground storage
 systems. This is especially true for
 piping, which is still very dependent
 on proper installation for long-term
 integrity. One of the lessons that a
 great many tank owners have yet to
 learn is that leak-free, quality con-
 struction is  not synonymous  with
 "low bid."
      While the advent of new mate-
 rials and methods in the realm of
 underground piping has addressed
 most of the old problem areas, it has
 also created some entirely new prob-
 lem areas. As often happens  with
 emerging techniques and technolo-
 gies, the solutions to the new prob-
 lems are not quite so clear-cut as the
 solutions to the old problems. Thus,
 we have some new piping questions
 lingering out there in  search  of
 answers. I don't know the answers,
 but there might be some value in
 exploring a few of the  questions.
 Let's try these three:

 •  Where can  low-melting-point
    materials be used?

 •  Is it really necessary to slope
    piping?

 •  How  deep  should piping  be
    buried?

 Where  Can Low-Melting-
 Point Materials Be Used?
 Fires  associated with gasoline are
 fearful events. Fire authorities have
 been concerned with preventing the
 escape of gasoline at  gas stations
 ever since gas stations were invented
 in the early part of this century. The
 problem was fairly benign as long as
 tanks were buried and all pumping
 systems  were suction-based. If a car
 knocked over a suction pump, only a
 little   gasoline  would  be  spilled
 because the pump ceased to function
 when the piping was broken.
      But the introduction of the sub-
 mersible pump  in the  mid-1950s
 upped the ante. With the pump in
 the tank and the piping operating
under pressure, a knocked-over dis-
penser could result in a geyser of
gasoline and, potentially, a serious
hazard. However, a fairly straight-
forward solution to this problem was
developed. Fire codes required the
installation  of  a device variously
called an "impact valve" or "crash
valve" at the base of dispensers.
Crash valves automatically close off
the flow of gasoline at the base of the
dispenser if the dispenser is knocked
about or if a fire develops in the base
of the dispenser. Thus, as long as all
piping was buried, the situation was
again under control.
    If a fire should occur in a sump
     ''*(,!jji£8!as.s.prflerible piping is *
    exposed and if the submersible
    pump continues to operate, the
 ipiping could melt, and the specter of
  ~a geyser of flaming gasoline rears
  r     its ugly head again.
     A new concern has emerged,
however. It is  associated with the
widespread use of secondary con-
tainment and nonmetallic materials
in piping systems. Before the days of
secondary containment, fiberglass
piping usually remained completely
buried in backfill, and there was little
risk that it would be exposed to fire
sufficiently hot to melt the fiberglass
and release more fuel into the fire.
     With the advent of secondary
containment and liquid-tight tank
tops and under dispenser  sumps,
sections  of nonmetallic piping are
now commonly exposed to plain
view (good for leak detection). How-
ever, these piping sections now risk
potential exposure to a flammable
liquid fire in the sump (bad for fire
extinguishing).  If a fire should occur
in a sump where fiberglass or flexi-
ble piping is exposed and if the sub-
mersible pump continues to operate,
the piping could melt,  and  the
specter of a geyser of flaming gaso-
line rears its ugly head again. In this
situation, the crash valve is rendered
useless  because the piping fails
upstream of the crash valve location.
     Underwriter's   Laboratories
(UL) is an organization whose mis-
sion is to evaluate the safety of con-
sumer products. To achieve a UL list-
ing, a product must undergo a battery
of tests designed to establish whether
it will present hazards under condi-
tions similar to those expected to be
encountered when the product is in
service. Although galvanized steel
piping, the traditional storage system
piping material, has never achieved a
UL listing, most piping materials and
piping components in use today do
have a UL listing.
     UL listings come in a variety.of
flavors and colors depending on the
testing that was completed on the
product. In the area of service station
piping, a product is typically listed
for underground, aboveground, or
underground/aboveground use. The
distinction among the listings centers
on whether the product is able to
withstand fire exposure. The UL fire
test consists of exposing the product
to a pool of burning kerosene for 45
minutes. If the product passes the
test, it earns an aboveground/under-
ground listing. If it fails the test, or if
the vendor chooses not to conduct
the test because he or she knows
what will  happen,  the  product
receives an underground use only
listing. None of the nonmetallic pip-
ing (i.e., fiberglass, flexible compos-
ites) materials that I know about has
passed this fire exposure test.
     Flexible connectors, however,
do come in two varieties: one listed
for underground  use only, and the
other listed  for  underground  or
aboveground   use.   The  under-
ground/ aboveground variety of flex
connector is, in general, constructed
entirely of metal; the strictly under-
ground variety uses  nonmetallic
materials to contain the liquid.
     Fire codes initially held to their
traditional restrictions on the use of
low-melting-point materials.  These
codes  stated  that these materials
should be used aboveground only
when 1) suitably protected against
fire exposure, 2)  located such that
leakage resulting from failure would
not unduly expose persons,  build-
ings, or structures, or 3) located
where leakage can readily be con-
trolled  by  operation of accessible,
remotely located valves (National
Fire Protection Association (NFPA)
Code 30, Section 3-3.4, and Uniform
Fire Code, Article 7901.11.1.2).

               • continued on page 6

                               ~5

-------
LUSTUne Bulletin 27
I

^^- — Vent piping
Concrete paving
£ © r 9\
1C 	 /^TT""^ \
L ^ I J I
\^ J


Product
lines?
jf '
	 u — ;-
/ Vapor |
recovery i
piping !
/if ron'H^ :

.....

Rigid Piping Layout

Concrete
paving

l—Sf


	 ^

! [f|




I
* 1

P 1
-•


                                                          Vapor recovery piping
                                                                    (if req'd)
                                                        Flexible Piping Layout
  7//ese diagrams illustrate the evolution of UST piping systems from rigid rectangular layouts with many burled joints to flexible, curved lay-
  outs with minimal joints in the system.
• Piping's Progress from page 5

     Terminating the low-melting-
point material outside the dispenser
sump (i.e., where it is surrounded by
backfill and not exposed to fire)  and
using fire-resistant piping inside the
sump are not what the piping manu-
facturers   intended  when  they
designed their systems. In addition,
this  convoluted  design is rather
inconvenient.
     A possible  solution  to  the
dilemma is to add backfill to the bot-
tom of the sump, thereby burying the
low-melting-point materials. This
approach somewhat defeats the pur-
pose of the sump, however, in that it
no longer allows for direct visual
observation of piping and piping
joints. It also complicates any future
work on the piping because the back-
fill would need to be removed (most
likely with small hand tools) to gain
access to the piping.
     Another solution is to install
miniature fire extinguishers inside
the sumps that automatically  and
immediately extinguish any fire  that
might  occur. One manufacturer of
flexible piping systems now offers
such a device.
     Another concern of fire person-
nel is whether or not a fire would be
sustained in a sump. Theoretically,
the oxygen in the sump would be
consumed in a very short time,  and
the fire would die. So far,  fire
authorities do not appear to be con-
vinced by this  line of reasoning.
     The 1996 edition of NFPA Code
30 now explicitly allows low-melt-
ing-point materials to be  used in
tank-top sumps that  are  usually
remote from dispensers and the gen-
eral public. The  code still prohibits
the use of low-melting-point materi-
als in dispenser sumps that  are in
close proximity  to areas where the
general public is  likely to be present.
The 1997 edition of the Uniform Fire
Code contains no similar provisions.
     Is all this concern justified? How
many fires have  actually burned in
sumps? I don't know. Does anyone have
any stones they'd like to share?

Is It Necessary to Slope
Piping?
Piping slope is another issue that's
steeped in history. When all pump-
ing systems were  suction systems,
piping that carried product had to be
sloped because otherwise it was very
difficult for the pump to remove all
the air in  the piping  and operate
properly. Piping slope became less
critical for pressurized piping, but
high spots could trap air that would
restrict the effective diameter of the
pipe and affect the performance of
the piping system.
     But this concern  arose  in the
days of rigid piping with 90-degree
bends. Recently,  a  representative
from a flexible piping manufacturer
pointed out to me that with the gentle
curves and smaller diameters that are
common in flexible piping systems,
air pockets are unlikely even if the
piping is not sloped uniformly. As a
result, he explained, piping slope is
not nearly as important to the proper
operation of product piping as it used
to be. It made sense to me, but I'd like
to see some  testing to verify  this
point.
    But there are reasons for slop-
ing the piping that go beyond effi-
cient  delivery of the product. For
example, a  uniform  piping slope
facilitates the draining of piping dur-
ing repair work or removal.
    Another rationale for sloping
piping  applies  to double-walled
pipe.  Most secondary-containment
piping systems are designed so that
leaked product flows back to a sump
(low spot) in order for the leak to be
detected. This strategy is especially
important for ducted systems, where
the secondary containment is much
larger in diameter than the primary
pipe,  and the interstitial space has
considerable volume.
     For coaxial secondary contain-
ment  systems, where the secondary
and primary pipe are installed as a
single unit and the interstitial space
is very small, it  is likely that the
small volume of the interstitial space
and the pressure of the product from
the primary pipe would cause the

-------
                                                                                          LUSTLine Bulletin 27
•prodUict to £low quickly to tKe ends
of the pipe, even if it had to flow up a
few small hills.
     The difficulty with sloping pip-
ing uniformly back to the tank top is
that flexible  piping configurations
typically involve a single run of pipe
that goes to  a dispenser and then
continues on to  the next dispenser
and then the next. This piping layout
results in single piping runs that can
be much longer than those of tradi-
tional rigid piping, where dispensers
were fed from branches off of a sin-
gle  main  line.  If the  piping is
required to slope back to the tank top
from the farthest dispenser,  the
depth of dispenser sumps and the
burial depth of  the  tank  could
become excessive.
     For example, piping is tradi-
tionally sloped 1/8 inch per foot of
pipe, or about 1 foot of slope per 100
feet of pipe. A large, multiple-island
facility might entail a 500-foot-long
piping run, for a total slope of about
5 feet. Add to this the minimum bur-
ial depth of the piping (1.5 feet) and
the fact that the piping is likely to be
about a foot above the tank, and you
end up with the top of the tank some
7.5 feet below the surface. An 8-foot-
diameter tank, with 1 foot of bedding
underneath,   would,   therefore,
require a 16.5-foot-deep excavation.
     This situation not only requires
a deeper excavation that is more
likely to encounter groundwater and
need shoring or  some other method
of stabilization,  but also incurs the
expense of excavating and removing
additional native soil as well as
bringing in more proper backfill. The
deep burial depth may also create a
problem with the structural integrity
of the tank; the maximum burial
depth for fiberglass tanks is 7 feet
and the burial depth recommended
for steel tanks is one-half the tank
diameter.  (The  maximum  burial
depth for steel tanks is also a moving
target—see sidebar.) Maintenance on
pumps that  are 6 feet deep is also
very inconvenient and quite possibly
hazardous.
      Industry-recommended prac-
tices such as API 1615 still specify
uniform slope back to the tank. This
requirement  creates conflict between
regulators, who insist that installers
follow the industry practices, and the
installers, who wish to avoid burying
tanks any deeper than necessary.
    Industry-recommended practices
    such as AP11615 still specify
  uniform slope back to the tank. This
  ,  requirement creates conflict
 | between regulators, who insist that
     installers follow the industry
  I practices, and the installers, who
 \  wish to avoid burying tanks any
       deeper than necessary.
     The most recent edition of the
Petroleum Equipment  Institute's
Recommended Practices for Installation
of Underground Liquid Storage Systems
(PEIRP100-97) specifies that product
piping must be sloped, but it does
not require that the piping slope to
the tank. Piping may slope  to dis-
penser sumps or to  intermediate
sumps installed  along the  piping
run. This alternative may result in
the installation of more leak detec-
tion sensors in dispensers and inter-
mediate  sumps,  but it allows for
more traditional burial depths for the
tank—even those with lengthy pip-
ing runs.
How Deep Should Piping
Be Buried?
Based on  industry-recommended
practices, the minimum burial depth
for piping has traditionally been 18
inches from finished grade. Ade-
quate burial depth is important to
protect the piping from traffic load-
ing, frost movement, and accidental
damage during future construction
or remodeling activities. Some pipe
manufacturers' instructions, as well
as the 1996 edition of API 1615, how-
ever, now state that 18 inches is more
than what is required, and they spec-
ify shallower burial depths.
     I believe that this issue relates
to the piping slope issue discussed
earlier in that, if uniform  piping
slope to the tank is required, then a
deeper   minimum burial  depth
requirement for the piping equates
to a deeper burial depth requirement
for the tank. This means deeper tank
excavations and a greater volume of
appropriate tank and piping backfill
materials are required to complete
the installation. Deeper excavations
and more backfill mean  that  the
installation costs more money.
     So how deep  should piping be
buried? I don't know, do you? •
   New Construction Standard for Steel Tanks

   Beware the Maximum  Burial Depth!

   jT\. recent change in UL 58, the construction standard for steel under-
   ground storage tanks, which went into effect on September 30,1997, is
   worth noting. The thickness of steel plate used in building a tank is one
   of the factors that determines the pressure required to collapse the tank
   from external forces. In the old version of UL 58, the thickness of the steel
   plate to be used in constructing different sizes of tanks was specified as
   part of the standard. In the latest edition of UL 58, the thickness of the
   steel plate is left largely to the discretion of the tank manufacturer, with
   the caveat that each tank must bear a label indicating the maximum bur-
   ial depth for that specific tank. The tank must also be able to withstand at
   least a 5-foot burial depth.
       In other words, it is now possible for a tank manufacturer to pro-
   duce an 8,000-gallon tank with a specified maximum burial depth of 5
   feet, and an 8,000-gallon tank with a specified maximum burial depth of
   8 feet. Because of the physics of the situation, the tank with the deeper
   burial depth will need to be constructed of thicker steel. Because steel is
   sold by  the ton, the tank with a maximum burial depth of 8 feet will
   weigh more and cost more than the tank with a maximum burial depth
   of 5 feet. Thus, installers and regulators will need to verify that the
   planned burial depth for a steel tank is less than the maximum permissi-
   ble burial depth indicated on the tank label. •

-------
 LUSTLine Bulletin 27
  Leak Prevention
 EPA  Issues Guidance on Alternative
 Integrity  Assessment Methods  for
 Steel Tanks
   In order to maximize the number
   of USTs that will be upgraded in
   an effective, safe, and affordable
 way, EPA's Office of Underground
 Storage Tanks (OUST) issued a guid-
 ance document on July 25,1997, that
 is meant to strike a balance among
 several competing interests. On the
 one hand, EPA  is recommending
 that implementing agencies adopt a
 consistent policy on this issue; on the
 other hand, each  agency retains the
 flexibility to follow its own policy if
 it chooses.
     The guidance leaves the door
 open for the use of alternatives to
 human entry (i.e., corrosion model-
 ing, video  camera,  remote ultra-
 sound via robot). However, after
 March 1998, the door is open only if
 the alternatives meet certain attain-
 able standards. While this approach
 allows  flexibility on the  part  of
 industry, it discourages fly-by-night
 operators and substandard proce-
 dures. The guidance provides indus-
 try with two options that may be
 used  to  meet minimum recom-
 mended standards and with suffi-
 cient  time  to  accomplish  the
 necessary work. Finally, the guid-
 ance allows methods that are both
 affordable and protective, so  that
 owners can meet the December 1998
 deadline and avoid future leaks.
     In essence, this final guidance
 recommends that agencies continue
 current policies until March 22,1998.
 After that time, integrity assessment
 methods should either (A) meet a
 national code, or (B) be evaluated by
 a third party to meet certain perfor-
 mance criteria. Agencies should not
 require monthly leak detection mon-
 itoring for methods that meet option
 A or B, but they can consider a range
 of other conditions.

 Option A - Standard Codes of
 Practice
 No standard codes currently exist
 (ASTM ES 40 has expired). A draft
replacement  is  currently  in  the
ASTM process under Committee G-
 01. A draft was balloted once and is
 currently being revised. It will prob-
 ably  be reballoted  in  December.
 Thus, it is possible (although uncer-
 tain) that a new ASTM standard will
 be approved by March 1998.

 Option B - Evaluation and
 Certification
 No alternative integrity assessment
 methods have undergone third-party
 evaluation to date.  However, the
 same process as is used for UST leak
 detection systems applies. An EPA
 Quality Assurance Project Plan* that
 was prepared for a research  effort
 provides a viable test protocol. Proto-
 cols that are more specific can be writ-
 ten by interested parties. The process
 will be carried out by the private sec-
 tor. OUST will provide information
 on how evaluators can qualify and
 "short forms"  so  that  evaluation
 results are reported in a concise, con-
 sistent format.  Depending on how
 many vendors choose option B,  states
 and EPA may want to form a  work
 group to review evaluations, similar
 to the National Work Group on Leak
 Detection Evaluations.

 What About Using Leak
 Detection Alone as an
 Integrity Assessment?
 Although OUST  recognizes the
 important role leak detection plays
 in preventing serious releases into
 the environment, it does not recom-
 mend that leak detection alone be
 considered sufficient to assess the
 integrity of USTs  that are 10 years
 old or older.

What Next?
OUST will continue to monitor  all
integrity assessment code develop-
ment and evaluation efforts, and
pass along any new information to
the states. •

*For information about this document,
contact Carolyn Esposito at (732) 906-
6895.
                                    _.  .TV	ǣǥ.'
                                    Tank Bits
 $65 Million Appropriated
 For LUST Trust Fund
 On October 27, President  Clinton
 signed the Environmental Protection
 Agency's fiscal 1998 appropriations
 bill, which included $65,000,000 (to
 remain available until expended) of
 federal LUST Trust funding. The fund
 may be used for necessary expenses
 associated with carrying out leaking
 underground storage tank cleanup
 activities as authorized by the Super-
 fund Amendments and Reauthoriza-
 tion Act of 1986, and for construction,
 alteration, repair, rehabilitation, and
 renovation of  facilities. Congress
 directed that EPA distribute at least
 85 percent of this funding to the states
 and use no more than $7,500,000 for
 administrative expenses. Other por-
 tions of this funding will be used by
 EPA to provide technical assistance to
 states, to assist with the implementa-
 tion of the LUST program on Indian
 lands, and to support EPA Office of
 Research  and Development (ORD)
 research  efforts (primarily MTBE
 research). •

 U.S. Postal Service Pushes
 UST Improvements
 The U.S. Postal Service has taken its
 UST responsibilities very seriously,
 spending more than $200 million since
 1989 on tank replacements, upgrades,
 and closures; its UST population has
 been reduced from 10,000 to around
 3,000 tanks. Postal Service manage-
 ment is pushing hard to make sure
 that all remaining tanks are in compli-
 ance with the 1998 deadline. Manage-
 ment offers its facilities managers both
 "carrots and sticks" to comply: (1)
 headquarters will provide full funding
 for all tank work done before Decem-
 ber 22,1998, but after that facilities will
 have to use money from their own
 budgets; and (2) facility managers can
 be fined (or even fired for gross viola-
 tions) for violating both Postal Service
 and EPA UST regulations.
     The Postal Service continues to
 take a hard look at whether each UST is
 necessary.  A tank—even one installed
 recently and in perfect condition—may
be removed if there is no longer a need
for it, thus avoiding future problems
 and liabilities. The Postal Service is also
converting the majority of its vehicles
to run on alternative fuels to reduce its
need for USTs altogether. •
8

-------
                                                                                            LUSTLine Bulletin 27
       a
                                            Where USTs are concerned, questions do pop up. Our answers are based on a
                                          carefully considered interpretation of the federal rule and, where available, on
                                       EPA guidance. Keep in mind, individual state requirements may differ from federal
                                     requirements. Your questions and comments are welcome.
v_X. Currently, I am using inventory
control with periodic tightness testing to
meet the release detection requirements
of40 CFR §280.41(a). There is one thing
I don't understand. In recording fuel
deliveries for use in inventory reconcilia-
tion, am I supposed to use gross gallons
or net gallons?
       First, a word of explanation
to readers who may not understand
your question. When you receive a
fuel drop, the amount delivered can
be expressed as gross or net gallon-
age. Gross is a purely volumetric
measure — think of gross volume as
how many gallon cans you could fill
from the delivery truck at the time of
delivery. The problem with gasoline,
however,  is that its volume is very
sensitive  to  temperature changes.
Warm gasoline expands; cold gaso-
line contracts. As a result, there is
actually less gasoline by weight in a
gallon of  warm gasoline than in a
cold one — a tanker truck with 10,000
pounds of warm fuel could fill many
more gas  cans than the same truck
filled with 10,000 pounds of cold
fuel. It's the same amount of fuel, but
the dispensed volumes are different.
     Fuel distributors can compen-
sate for this discrepancy by using net
volume. Net represents the volume
of product you would have if the
product were brought to  a tempera-
ture of 60 degrees Fahrenheit. The
difference can be sizable. For exam-
ple, if a tanker delivers 5,000 gross
gallons of gasoline that's at an ambi-
ent  temperature  of  70  degrees
Fahrenheit, its calculated net volume
is only 4,965 gallons.
     EPA  generally  recommends
using gross rather than net measure-
ments. While your distributor may
have the capability of figuring net
gallons, you most likely do not. The
sales you make with your dispensers
and the  tank measurements you
make with your wooden stick are
gross measurements. If  you don't
treat deliveries  as gross measure-
ments, too, then you  will be intro-
ducing errors into your inventory
calculations. In the example above,
for instance, you are already off by
35 gallons on the day of delivery,
and that's only for a 10-degree tem-
perature difference!
     The use of gross gallons is only
a general recommendation. In cold
climates, fuel dispensing as well as
fuel deliveries may be represented in
thermally compensated net gallons,
and the use of net figures may be
more appropriate. Owners and oper-
ators  of petroleum USTs should
check with their local implementing
agency for additional guidance on
the use of  gross  or net measure-
ments.
      'What is the basis for the federal
threshold for suspecting a problem in
inventory control records? I understand
the 1 percent  of throughput part, but
why the 130 gallons?
       The origin of the 1 percent +
130-gallon standard for suspecting a
problem in inventory records lies in
a study funded by EPA in 1988. EPA
commissioned the study in order to
determine the threshold for suspect-
ing a release through inventory con-
trol records.  The  long-standing
American Petroleum Institute (API)
standard of 0.5 percent of sales was
the only standard around at the time,
and EPA  wanted to have it evalu-
ated. In the study, 20,036 individual
computerized  inventory  control
measurements were reviewed, and
some interesting conclusions were
reached.
     One disturbing discovery (from
a regulatory standpoint) was that the
0.5 percent standard put forth by API
resulted in a false-alarm rate of 30
percent. In other words, inventory
records exceeded the 0.5 percent
allowable variance 30 percent of the
time, when no problem existed with
the storage system. In addition to the
"crying wolf" syndrome that this
eventuality might foster in the regu-
lated community, it was simply not
desirable to have 30 percent of the
storage systems using inventory con-
trol reporting releases every month.
     A 1-percent standard would
lower the false-alarm rate, but one
problem remained. Any standard
that is based solely on a fixed per-
centage is going to produce a high
false-alarm rate at small-throughput
facilities. For example, a facility with
a 100,000-gallon per month through-
put is allowed a 1,000-gallon vari-
ance each month, but a facility with a
1,000-gallon throughput  is  only
allowed  a 10-gallon variance. The
whole theory of inventory control is
based on the assumption that inven-
tory errors in nonleaking storage sys-
tems are largely random and will
tend to cancel one another out. Thus,
it should not be too difficult to stay
within the 1,000-gallon limit  for a
facility pumping 100,000 gallons. On
the other hand, keeping inventory
records accurate to within 10 gallons
would be very difficult, because a
single delivery error or measurement
could easily exceed 10 gallons.
     One way to escape the problem
of high false-alarm rates at small
facilities is to add a constant to the
volume percentage that allows the
small facility more wiggle room. The
EPA study concluded that a thresh-
old of 1 percent + 130 gallons would
produce a false-alarm rate of 5 per-
cent  for  inventory records.  This
number  was  consistent with the
false-alarm rate allowed for other
leak-detection methods. The down-
side of this threshold is that the prob-
ability of detection is 95 percent for a
leak of 1.1  gallons per  hour. The
study also noted that these statistics
were valid for retail-type facilities
with storage capacities in the range
of  500  to 50,000  gallons  with
throughputs of 1,000 to 10,000 gal-
lons per month.
               I continued on page 10

-------
LUSTLine Bulletin 27
 • Qs and As from page 9 _

 v_jt.  Owners and operators of USTs
 are required to demonstrate financial
 responsibility  for  taking corrective
 action and for compensating third par-
 ties for bodily injury and property dam-
 age caused by accidental releases. UST
 owners and operators may use one or a
 combination of mechanisms in 40 CFR
 §280, Subpart H, to demonstrate finan-
 cial responsibility. If an UST owner/
 operator decides to  switch from one
financial responsibility mechanism to a
 different one, do any financial responsi-
 bility notification requirements apply?
        There are no specific federal
financial responsibility notification
or reporting requirements associated
with switching from one UST finan-
cial responsibility  mechanism to
another. Owners and operators may
substitute an alternative financial
mechanism at any time, provided
that financial assurance is always
maintained (§280.111(a)).
     UST owners and operators are,
however,  subject  to the  general
reporting  provisions of §280.110.
Under these provisions, owners and
operators must submit evidence of
financial  responsibility  to  their
implementing agency in the event of
a confirmed release or incapacity by
a financial assurance provider.  The
implementing  agency  may   also
require reporting or a demonstration
of compliance at any time.
          SNAPSHOTS FROM TH€FI€LD
Liz Shepherd, an Inspector with the Kentucky UST program, sent us these
snapshots from the field. In September, she attended a corrosion and cathodic
protection testing training program in Bowling Green, Kentucky, for inspec-
tors in EPA Region 4. She writes: "Over a period of three intense days, we
learned basic theories of electricity and applied them in soup cans and then out
in the field at UST facilities. Corrosion protection is  more complicated than
most of us anticipated! It was yet another reminder that it's not just UST facil-
ities that shouldn't wait until '98—the inspectors have some catching up to do
as well!"
                                        Instructor  Marcel Moreau
                                        advises inspectors on how to
                                        interpret the data.
                                                     y Tern  com.
                                                                     Inspectors examine a corrosion
                                                                     protection system test station.
                                      tifyau have any UST/LUST-related snapshots from the field that you
                                      * would like to snare with our readers, please send them to Ellen Frye
10

-------
                                                                                   LUSTLine Bulletin 27
from Robert N. Renkes, Executive Vice President, Petroleum Equipment Institute
PEI's Recommended Practice for UST System Installation
(RP100) Has Been Revised and Is Now Available
      The federal underground storage tank regula-
      tions (40 CFR 280.20) require that all under-
      ground storage tanks and piping subject to
 these rules be properly installed in accordance with
 a code of practice developed by a nationally recog-
 nized association or independent testing laboratory
 and in accordance with the manufacturer's instruc-
 tions.
      Of the three recommended practices men-
 tioned  in 40 CFR, only one, PEI's Recommended
 Practices for Installation of Underground Liquid Stor-
 age Systems (PEIRP100-97), has been updated on a
 regular basis since it was first published. The group
 responsible for writing the recommended practice,
 PEI's Tank Installation Committee, continually
 monitors  changes in equipment and installation
 techniques and revises the  document when circum-
 stances warrant—PEI's RP100 has been revised
 about every three years since it was first issued
 in 1985.
      The recommended practice has been revised
 once again and is now available to individuals and
 firms interested in the latest word on proper tech-
 niques  and methods for installing underground
 storage tank systems.
      Readers of the 1997 document should note
 that the recommended practice has undergone
extensive changes. Three chapters (spill contain-
ment and overfill prevention, secondary contain-
ment,  and leak  detection)  were  completely
rewritten. Significant additions were made to the
flexible piping section of Chapter 9. Twenty figures
were changed in some manner (see diagram), and
extensive editorial revisions were adopted to make
the text more readable and to clarify tihe meaning of
some of the recommended practice's provisions.
    The method PEI uses to  amend the recom-
mended practice is unique among trade associa-
tions in the petroleum marketing industry. While
the Tank Installation Committee consists solely of
PEI members who install underground systems,
everyone involved in the UST community is given
an opportunity to review and submit comments to
revise RP100. For this edition, for example, 50 per-
cent of the comments were received from tank
manufacturers, 27 percent from regulators, 9 per-
cent from tank installers, 7 percent from trade asso-
ciations representing oil marketing groups, and 4
percent from industry consultants.
    RP100-97 supersedes the previous recom-
mended practice of the same name,  published in
1994. Copies are available for $25 (includes postage
and handling) from PEI, P.O. Box 2380, Tulsa,
Oklahoma 74101-2380. Phone: (918) 494-9696. •
This cross section of an UST system illustrates the new graphic detail in RP100-97.


-------
 LUSTUiif Bulletin 27
               Flux =
                         dX
   Flux is the rate of movement of a compound per unit area.
   De is the effective diffusion coefficient in the vadose zone.
   dCjg is the contaminant concentration gradient in the soil vapor.
   dX is the depth interval in the vadose zone.
     Similar to momentum, transfer (e.g., water running
downhill) and heat transfer (movement from hot to cold),
contaminant  transfer by gaseous diffusion moves from
areas of high concentration to areas of low concentration.
The flux will always be down the concentration gradient,
regardless of the orientation of the concentration gradient
with respect to depth below the surface. In the subsurface
environment, diffusional transport occurs in all directions
                                                                         Figure 1
                                                                                      Concentration
  Investigation and Remediation"'
The Upward  Migration of  Vapors

by Blayne Hartman

A Note From The Editor: Recent changes in environmental cleanup ideology and regulations have led to the increased application
of natural attenuation as a remediation strategy and risk-based corrective action (RBCA) as a means for determining the cleanup require-
ments for contaminated sites. As these approaches are implemented, consideration must be given to the fate and transport of contaminant
vapors in the subsurface and the potential risks they pose to human health. In this article, Blayne Hartman examines the processes by which
vapors move through the vadose zone and the potential risk caused by the upward migration of vapors into an overlying building. He con-
cludes by recommending a protocol for determining the upward vapor flux in the field.
     WJiile I recognize that the technical nature of this article—not to mention the preponderance of daunting equations—may scare off
sonic of you (it scared me), I also recognize that this type of timely information will be of value to those of you who are struggling with
remediation issues. For completeness of information, I chose to retain all of the equations submitted with the article. It is easy enough, how-
ever, to skip over the mathematics and still benefit from the discussion. In a future issue of "LUSTLine," Blayne will discuss the potential
risk lo groundwater resulting from the downward migration of vapors.
How Do Contaminants Move in the
Vapor Phase?
A common misconception associated with vapors ema-
nating from a subsurface source of contamination (i.e.,
soil, fractured bedrock, groundwater) is that the vapors
will preferentially rise upward and escape into the atmos-
phere, much like smoke rising from a smokestack. To
understand why this idea is a misconception, you need to
understand how the transfer of contaminants occurs in
the vapor phase.
     There are primarily two types of physical processes
by which contaminants are transported in the vapor
phase: advection and gaseous diffusion. Advection refers
to the bulk movement of the vapor itself (e.g., the move-
ment of vapor by wind). In advective transport, any con-
taminants in the vapor are carried along with the moving
vapor. Advective transport processes can be important in
the movement  of soil vapor through the vadose  zone
(e.g., near the surface due to atmospheric pressure varia-
tions or near buildings that create pressure gradients due
to differential heating).
     Gaseous diffusion refers to the motion of the conta-
minants by molecular processes through a nonmoving
vapor column. It is the primary transport mechanism for
contaminants in the vapor phase through the soil vadose
zone. Contaminant transport by gaseous diffusion is
described by Pick's first law as:
                                                                                             SURFACE
                                                                                              SOURCE
                                                                                     Concentration
                                                                                               DEEP
                                                                                              SOURCE
                                                                                     Concentration
so contaminants
move away from
a  source  in  all
directions, simi-
lar to an expand-
ing balloon. The
key  issues  to
remember are:
•  Contaminant
   transport  by
   gaseous diffu-
   sion does not
   move prefer-
   entially in one
   direction (e.g.,
   up or down)
   but  spreads
   out radially in
   all directions.
•  The direction
   of movement
   is from high
   concentration
   to low con-
   centration re-
   gardless   of
   the  orienta-
   tion with re-
   spect to depth
   in the vadose
   zone. In other
   words,  if high concentrations of a contaminant in the
   vapor phase are midway between the ground surface
   and the groundwater,  the diffusive  flux from the
   source will move both upward and downward from
   the source.

The Upward Migration of Vapors Into
Enclosed Spaces
Benzene is the principal contaminant of concern at most
sites because of its proven carcinogenity and common
occurrence at gasoline-contaminated sites. Other com-
mon compounds of concern at fuel-contaminated  sites
                                                                                             SURFACE
                                                                                              &DEEP
                                                                                             SOURCE
                                                                          Common soil vapor profiles in the vadose
                                                                          zone for different locations of contaminant
                                                                          sources.
12

-------
                                                                                              LUSTLine Bulletin 27
 Figure 2
                                   Room Volume
                                      (V)
 Upward migration of vapors into building.
include chlorinated solvents (e.g., vinyl chloride and
tetrachloroethylene from oil sumps) and methane.
     A simplified environmental fate  and  transport
model for evaluating the inhalation exposure pathway for
a contaminant is summarized in the 1995 ASTM Risk-
Based Corrective Action (RBCA) Standard. The  model
assumes that contaminant vapor transport is by gaseous
diffusion, that the contamination source is constant and
nondiminishing, and that equilibrium conditions exist.
Buildings with basements or subterranean structures,
however, can  create pressure  gradients that initiate
advective transport, which requires different assump-
tions than those presented in the following discussion.
With these limitations in mind, the indoor air concentra-
tion of a contaminant (Cj) is computed as:
                   Slab * Flux * A  Slab * Flux
                      V*E
Height * E
     Cj is the concentration in the room in (jg/m3.
     Slab is the slab attenuation factor (unitless).
     Flux is the contaminant flux into the room (ug/hr-m2).
     A is the room floor area in m2.
     V is the room volume in m3.
     Height is the room height in m.
     E is the indoor air exchange rate (exchanges per hour, 1/hr).
     As you can see, the indoor air concentration is
dependent upon the flux, the height of the room, and the
indoor air exchange rate with outdoor air. For residential
buildings, an indoor air exchange rate of one room vol-
ume every 2 hours (or 12 exchanges per day) is typically
used.  Commercial  buildings  typically  have faster
exchange rates, which are obtainable from the architect or
engineer.
     In addition, the vapor flux is considered to be atten-
uated by the presence of a concrete slab or wall. The net
result of the concrete is to decrease the soil vapor flux. For
new or relatively new concrete slabs, an attenuation fac-
tor of 0.01 is typically used on the  basis that approxi-
mately 1 percent of the slab consists of cracks that offer
unrestricted vapor flow. For older slabs in poor condition,
an attenuation factor of 0.10 gives a more conservative
estimate of the reduction in vapor flux caused by the slab.
     As described previously, the upward contaminant
vapor flux into a building is computed by Pick's first law,
using the gaseous diffusivity, corrected for porosity.
                                           Flux=
De*ACs
   X
                 * 1,000
                                                           Flux is the rate of movement of a compound per unit area
                                                             (ug/hr-m2).
                                                           De is the effective diffusion coefficient in the vadose zone (m2/hr).
                                                           ACsg is the contaminant concentration gradient in the soil vapor
                                                             (MO/L).
                                                           X is the depth below the surface (m).
                           For most cases, the contaminant concentration in
                      the room air is negligible compared with the soil vapor
                      concentration, so the measured soil vapor concentration
                      (Csg) can be plugged directly into the equation.
                           Calculation of the flux requires knowledge of the
                      soil vapor concentration. In the absence of actual soil
                      vapor data, soil vapor concentrations can be calculated
                      from soil and groundwater data, assuming equilibrium
                      conditions, using equations based upon Henry's law con-
                      stants and soil-to-water partitioning constants. These
                      equations (summarized below) can be found in the 1995
                      ASTM RBCA standard.
                      •  Calculating Soil Vapor Concentration from Soil
                         Data
                         The soil vapor concentration (Csg) is computed based
                         upon the equilibrium partitioning between the soil,
                         moisture, and vapor phases as:
                            csq(Mg/L)
                                                                              H*C,
                                                                                   'soil
                                                     BD
                                                                                                   1,000
                           Csoi| is the concentration in the soil for the contaminant of
                           concern (e.g., benzene). If the concentration values for the
                           specific contaminant are not known, then the value may be
                           estimated from its concentration in the fuel product as the
                           mole fraction times the product concentration. For example,
                           the concentration of benzene in soil may be estimated from
                           TPH-gasoline data as the mole fraction of benzene (2.5 per-
                           cent) times the TPH concentration.
                         From Groundwater with Floating Free Product or
                         Soil with Free Product
                         It is assumed that the vapor immediately above the
                         groundwater is in equilibrium with the free product,
                         based upon the contaminant's mole fraction and vapor
                         pressure:
                              Csg(ug/L) =
VP * MW * MF
     RT
        . 1,000,000
                         From Groundwater with Dissolved Contamina-
                         tion (No Free Product)
                         It is assumed that the vapor concentration immedi-
                         ately above the groundwater is in equilibrium with the
                         groundwater, and the concentration is given by the
                         water concentration times the dimensionless Henry's
                         law constant:
                                                       • continued on page 14

-------
Ll/STOw Bulletin 27
  I Upward Migration of Vapors from page 23
                              (Cwater*H)
   For the preceding equations:
   VP is the contaminant vapor pressure in atmospheres.
   MW is the molecular weight of the contaminant in g/mole.
   MF is the mole fraction of the contaminant in the free product.
   RT is the universal gas constant times temperature equivalent to
     22.4 pL/umole at 0 °C & 24 uL/umole at 20 °C.
   H is the dlmensioniess Henry's law constant.
   Cson is the contaminant soil concentration in mg/kg.
   fJwawr is the contaminant water concentration in mg/L.
   BD is the bulk density in g/cm3.
   Pw and Pa are the water porosity and air porosity, respectively
     (unitless).
   Ks is the soil water distribution coefficient in cn^/g.

     Using these equations, it is possible to compute the
maximum soil concentrations, water concentrations, and
soil vapor concentrations versus depth from the surface
that will yield room concentrations that meet acceptable
EPA levels. These values are shown in the following table
for benzene.
Depth bgs Soil Water Soil Vapor
(ft) (ng/kg) (|ig/L) ((ig/L-vapor)
5
10
20
50
100
20
40
80
200
400
120
240
480
1200
2400
25
50
100
250
500
   bgs » below ground surface
   Assumptions used in computing the tabulated values:
   Air porosity:    0.2
   Total porosity:   0.3
   Bulk density:   2.0(g/cm3)
   Slab factor:     0.01
   Exchange rate:   0.5 (1/hr)
   Acceptable room concentration for benzene at the 1 in 1 million
   cancer risk level: 0.24 ug/m3.
     This summary demonstrates that, based upon the
assumptions used in the upward risk calculation, only
modest concentrations in the soil, soil vapor, or water are
required to result in room air concentrations that fail the
acceptable levels.  For some compounds with lower
acceptable room concentrations (e.g., vinyl chloride ~11
pptr)/ the allowable soil and groundwater values can
approach laboratory detection levels.
     You must recognize, however, that the equations
used to calculate the soil vapor concentration from soil
phase data, water phase data, or free product assume
equilibrium partitioning between the phases. Equilibrium
partitioning is obtained only if a system is well mixed.
This condition is very rarely accomplished in the subsur-
face, because there are no blenders or stirrers present to
homogenize the vapor, soil, and groundwater.
     A common analogy used to illustrate this mixing
concept is the preparation of a salad dressing using oil
and vinegar. When the ingredients are initially added to a
container, they fall into separate layers; the container
must be shaken to mix the ingredients. If the container is
not shaken, the oil and vinegar mix very slowly, "equilib-
rium is not reached," and the resulting salad dressing
does not taste very good.
     In addition to the issue of equilibrium partitioning,
the equations do not account for other processes that are
operative in the vadose zone, such as bioattenuation,
advective flow, and soil heterogeneity.
     For these reasons,  calculated soil vapor concentra-
tions generally do not accurately represent actual soil
vapor concentrations, and, in the case of fuels, calculated
values often overestimate actual soil vapor concentra-
tions by 10 or 100 times. The potential error in the calcu-
lated vapor flux introduced by the incorrect vapor
concentration is likely to be greater than errors intro-
duced by other parameters, such as porosity. Thus, in the
event that a site fails the upward risk calculation from
existing soil or water data, direct measurement of actual
soil vapor concentrations near the surface is likely to be
the easiest and fastest way to verify whether concentra-
tions will pass acceptable levels.

Which Soil Vapor Method to Use?
A number of states are currently trying to decide which
soil vapor method is the best one to employ for determin-
ing upward migration risk. Three methods are commonly
employed to measure soil vapor contamination: active,
passive, and surface flux chambers. A full discussion of
the various measurement techniques is beyond the scope
of this article. I will, however, present some summary
thoughts here.
• Active soil vapor methods (withdrawal of the soil
  vapor from the subsurface and subsequent analysis of
  the vapor) give concentration data, which are required
  for calculating the contaminant flux using Pick's first
  law. Further, vertical profiles of the soil vapor concen-
  trations can be obtained to aid  in determining the
  direction and magnitude of the flux. Active soil vapor
  data can  be collected  and measured in real time,
  enabling decisions to be made in the field.
  The problem most often raised with active soil vapor
  data is whether the concentrations measured at any
  given time and day are representative of normal condi-
  tions (i.e., how "stable" are active soil vapor data?).
  Variations  caused by factors such as barometric
  changes or building pressures are known to exist;
  however, they are difficult to quantify. These effects
  are known to lessen with increased depth below the
  surface (or away from the building), and it is generally
  considered that data collected from 3- to 5-foot depths
  are fairly stable.
• Passive soil vapor methods (burial of an adsorbent in
  the ground with subsequent retrieval and measure-
  ment of the  adsorbent) provide a time-integrated mea-
  surement and, therefore,  reduce the  uncertainty
  associated with the  temporal variations described
  above. Passive methods also are generally  easier to
14

-------
                                                                                            LUSTLine Bulletin 27
  implement. However, passive soil vapor methods yield
  soil vapor data only in terms of mass, not concentration.
  Therefore, a "conversion" of the data from mass units
  to concentration units needs to be performed prior to
  determining the health risk. The primary assumption
  required in making the conversion from mass to con-
  centration is the volume of vapor that passed by the
  buried adsorbent during the burial time period. There
  is no easy way to estimate this volume. Thus, the result-
  ing values have a high degree of uncertainty. Further,
  because passive collectors are buried so close to the sur-
  face (generally 2 feet or less), the measured values are
  highly influenced by any near-surface effects.
• Surface flux chambers are enclosures that are placed
  directly on the surface (e.g., ground, floor) for a period
  of time (e.g., generally a few hours to a few days), and
  then the resulting contaminant concentration in the
  enclosure is measured. By dividing the measured con-
  centration by the incubation time, a direct value for the
  flux is determined. This method offers advantages over
  the other two methods because it yields the actual flux
  of the contaminant out of the ground, which eliminates
  some of the assumptions required when calculating the
  flux by Pick's first law (e.g., effective diffusivity, influ-
  ence of a cement slab). This technique, however, is not
  as fast or easy to implement as the other methods, is
  subject to near-surface effects (e.g., the stability of the
  measured fluxes), and provides no clues about what
  may be "hiding" below.

     The bottom line is that each of the soil vapor meth-
ods has advantages and disadvantages for determining
upward vapor risk; however, all are potentially applica-
ble. Which method to use on a given site depends on the
site-specific goals and the logistical limitations. In my
view, the active soil vapor method offers less uncertainty
and more versatility than the other methods in most situ-
ations.

A Protocol for Determining the Upward
Migration  Risk by Soil Vapor Measurement
Based on the discussion above, I recommend the follow-
ing procedure for collecting soil vapor data near the sur-
face for the purpose of determining the upward vapor
flux into a room or enclosure.
1. Collect active soil vapor data at 5 feet below ground
  surface (bgs) at the location of highest contaminant
  concentration. If the location of highest contaminant
  concentration is unknown, collect soil vapor data at 5
  feet bgs spatially,  across the site, to identify the loca-
  tion of highest concentration.
2. Calculate the health risk as outlined above. If the risk
  calculation indicates that upward  vapor poses no
  threat to  human health, then this pathway may be
  eliminated as an exposure route, assuming the plume
  remains stable or diminishes.
3. If the risk calculation indicates that upward vapor
  migration may pose a threat to human health, then col-
  lect soil vapor samples at 5 feet bgs at the corners of the
  building or room to determine the spatial variation of
  the flux across the area of concern.
4. Recalculate the health risk using the average flux from
  all of the soil vapor locations. If the risk calculation
  indicates that upward vapor migration poses no threat
  to human health, then this pathway may be eliminated
  as an exposure route, assuming the plume remains sta-
  ble or diminishes.
5. If the risk calculation indicates that upward vapor
  migration  may pose a threat to human health, then
  repeat steps 1 through 4 at 3 feet bgs.
6. If the risk calculation still indicates that upward vapor
  migration may pose a threat to human health, then the
  soil vapor concentration at a shallower depth (i.e., <3
  feet bgs) needs to be determined. Measured concentra-
  tions this close to the surface can be greatly influenced
  by soil vapor collection technique and atmospheric air
  infiltration caused by barometric pumping. Thus,
  "time-averaged" data may be appropriate to ensure
  that the measured soil vapor values are representative.

     Time-averaged data may be collected using either
active or flux  chamber soil vapor techniques. With active
methods, a sampling tube should be left in the ground
and the soil vapor analyzed multiple times to demonstrate
consistency in concentrations over time. I recommend that
data be collected at 1-foot intervals and from 1 to 3 feet
bgs to ensure that vertical variations are characterized
adequately.  Alternatively,  a flux  chamber  may  be
emplaced to measure the flux directly. The time duration
for the flux chamber should be long enough to enable ade-
quate measurement of the contaminant.
     For subsurface enclosures, such as basements or
utility trenches, the same protocol can be used; however,
soil vapor samples should be collected from 3 to 5 feet
below the floor, rather than below ground surface. It may
also be necessary to consider the potential flux through
the walls as well as through the floor.  In this case, the
total flux into the room would  be  equal to  the flux
through the floor times the combined surface area of the
floor and the  walls. Alternatively, a soil vapor measure-
ment may be made on each side of the wall (3 to 5 feet
away from the wall) so that the flux through the wall can
be computed separately. The total flux into the room
would then be computed by summing the individual
fluxes through the floor and walls.
     Soil vapor data should be collected and analyzed
using protocols that satisfy the local regulatory agency.
Required detection levels are contaminant-specific and
depend on acceptable room air concentrations. For exam-
ple, for benzene, vinyl chloride, and tetrachloroethylene,
detection levels of 0.1 ^g/L-vapor (~30 ppbv), 0.05 ug/L-
vapor (20 ppbv), and 1.0 ^g/L-vapor (144 ppbv), respec-
tively, are required. •
   Blayne Hartman, Ph.D., is Vice President and Technical
 Director of TEG, Inc., in Solana Beach, CA. This article is an
 excerpt from a chapter on soil vapor methods and applications
 written by Dr. Hartman for a book titled "Legal and Techni-
 cal Considerations for Hydrocarbon Contamination," soon to
   be published by Argent Communications Group in Forest
   Hill, CA. For more information, or for a copy of the entire
    chapter,  contact Blayne by e-mail at: bh@tegenv.com.
                                                                                                        -is

-------
LUSTLine Bulletin 27
  Investigation and Remediation
Three-Dimensional  Sampling
A Vertical Perspective on Cleaning Up LUST Sites
by Gary A. Robbins

      Once groundwater contamina-
      tion occurs at an UST site, it
      typically takes a Herculean
effort to bring the site to closure. In
fact> UST sites that have experienced
groundwater  contamination  are
rarely "cleaned up."  A number of
factors contribute to this dismal situ-
ation. Through our research here at
the University of Connecticut, we've
found that an important contributing
factor, if not the most important con-
tributing factor, is the  glaring lack of
a three-dimensional perspective on
contamination.
       If I could give just one piece
of advice to environmental consul-
tants and regulators involved in tack-
ling fuel releases at  UST  sites, it
would be: Conduct multilevel sam-
pling of groundwater  and soil.
Unfortunately, in this respect, com-
mon practice falls short. It is common
practice to use monitoring wells to
assess groundwater contamination
and to locate product. Contamination
varies with depth below the water
table (see page 17), and wells provide
vertically averaged information over
their screened interval. It is also com-
mon practice to evaluate this informa-
tion in a two-dimensional perspective,
known as the "plume map."
       But, such a perspective can
be rather misleading in evaluating
the contamination. Take a look at
any plume map you have in your
office and ask yourself: Is the plume
6 inches thick or 30 feet thick? Is the
plume just below the  water table or
is it very deep? Is the plume plung-
ing  with  depth or rising?  The
answers to these questions  could
well have a significant bearing on
your understanding of the risk posed
by the contamination, the remedia-
tion strategy chosen  and the effec-
tiveness of that strategy. Yet, these
answers cannot be derived from data
that completely ignore the vertical
dimension.
       If you knew that the ground-
water contamination at a site  was
limited vertically to just a foot below

16
the water table, would you attempt
remediation by installing and pump-
ing a near-field well that went 25 feet
below the water table? If you knew
that groundwater at a site was flow-
ing in an upward direction through
the contaminated sediment, would
you put hundreds of thousands of
dollars into drilling monitoring wells
into  underlying bedrock?  If you
knew that product was  trapped
within 2 feet below the water table
and that the water table was only 5
feet down, would you remediate the
site by investing in a series of wells
to inject nutrients to a depth of 20
feet below the water table? I think
not. Yet, as we have observed again
and  again, ill-suited  remediation
schemes for these types of site condi-
tions are all too common. These ill-
suited decisions can, more often than
not, be traced to the lack of a three-
dimensional perspective.

Into the Next Galaxy
As noted above, wells are used to
detect mobile product. Wells cannot,
however, detect immobile product in
the unsaturated zone  or immobile
product trapped below  the water
table. As we have found, however,
immobile product may be the pri-
mary source of groundwater conta-
mination at UST sites.
       Although  multilevel sam-
pling of soil  is common practice,
such sampling is generally restricted
to the unsaturated zone and little
attention is given to sampling the
soil below the water table. It is as
though the water table  represents
some Star Trek-like energy barrier at
the edge of the galaxy that can't be
crossed using soil sampling tools.
     Our research at UST sites indi-
cates that remediation efforts could
be greatly improved  if  multilevel
sampling were conducted  both
above and below the water table.
Since 1989, we have conducted mul-
tilevel groundwater and soil investi-
gations   at sites  where  gasoline
releases have  occurred. In general,
these sites have relatively shallow
depths to groundwater, generally
less than 20 feet. Site geology has
ranged from compacted tills (silt)
with very low permeability to highly
permeable sand and gravel deposits.
    Irrespective of site geology, we
have observed, at site after site, that
product becomes entrapped below
the water table. For that matter, there
have been sites  where most of the
product is entrapped below the water
table. Despite its low density, petro-
leum product may become entrapped
below the water table when the water
table fluctuates. How?
    Because fuels have a very low
solubility in water, they form a con-
cave lens, or meniscus, when in con-
tact with water in soil pore space. On
the water side of the meniscus, the
water adheres to soil grains. For
product to rise to the water table, it
must have enough buoyant force to
overcome the adhesive force holding
the water onto the soil. The strength
of adhesion (or capillarity) is a func-
tion of the  pore size (grain) of the
soil. The water is held more tightly
in fine-grain soils than in large-grain
soils.
    The buoyant force of the prod-
uct is a function of the volume of
product in the pore space. If product
starts out as a mobile layer on the
water table and is smeared into small
globules when the water table falls,
the globules may not have sufficient
buoyancy to float up to the water
table when the water table rises. This
process leads to product entrapment
and immobility below the water
table. Product entrapment can also
occur if the release starts from below
the water table, as is the case with
partially submerged tanks.  In these
cases,  once  entrapped,  product
becomes a  continuous  source of
groundwater contamination, as con-
taminants slowly move from the
product into the groundwater.

Vertically Speaking
To effectively remediate  product
releases at UST sites, one must locate
and delineate entrapped  product
below  the  water table as  well as
evaluate  the  three-dimensional
distribution of the groundwater con-
tamination.  Interestingly, we have
observed that most of the entrapped
product tends to be close to the water
table  within a relatively  narrow

-------
                                                                                             LUSTLine Bulletin 27
   Figure 1
          270—

          269—

       _  268—

       5  267—
       g  265-]

       §  264—

       1  263-

          262—

          261—

          260
                                i    '
              Horizontal Scale:

               Vertical Seals:'
                 Contours In (ppb)
                                               CML-5

                               ----- Where analysis has been extrapolated
   Cross section of benzene soil contamination at an UST site.
                             Horizontal Scale: e
                              Vertical Scaled
                                 Contours In (ppb)
                          CML-S


        	Where analysis has been extrapolated
                   Cross section of benzene contamination of groundwater at an UST site.
vertical zone—on the order of 3 to 6
feet. (See the example in Figure 1.)
The vertical extent  of  entrapped
product appears to correspond to the
zone of  water  table fluctuation or
somewhat less.
     If near-field pumping has dis-
turbed a site, the vertical extent of
contamination  may be greater. For
example, at one site where the water
table was  on the order of 15 feet
below the ground surface, entrapped
gasoline was evident 15 feet below
the water  table.  This deep-seated
contamination  corresponded to the
maximum   depth  of  drawdown
achieved by a  near-field well that
was used years before in an attempt
to recover product. It would appear
that pumping-resulted in product
drainage and entrapment well below
the water table. Again, as this exam-
ple illustrates, if you don't know the
vertical distribution of contamina-
tion, you can do something that can
make matters worse.
     With  respect to the  three-
dimensional distribution of ground-
water  contamination,  we   have
observed that  this contamination
tends also to  be  rather vertically
restricted,  even at significant dis-
tances from the  source. (See the
example in Figure 2—note the verti-
cal exaggeration.) Typically, the con-
tamination exhibits very little vertical
dispersion  and very high vertical
concentration gradients. Near the
source, the vertical extent of ground-
water contamination tends to be sim-
ilar  to  the  product  distribution.
Beyond the source, the contamina-
tion exhibits very little  additional
vertical spreading, on the order of a
few more feet. We have observed that
groundwater contamination typically
plunges with  distance from the
source, owing to the infiltration of
fresh water on top of contaminated
water. The fresh and contaminated
water tend  to exhibit little mixing.

The Technical Wherewithal
In the past, even if one wanted  to
conduct multilevel sampling below
the water table at UST sites, the prac-
tice was inhibited by the costs for
sampling and  laboratory  analysis
and the lack of regulatory  accep-
tance. Today, with the application of
direct-push technologies in conjunc-
tion with field screening, the three-
dimensional characterization of fuel
contamination can be achieved in a
cost-effective manner. A number of
reliable and innovative tools and
methods are available for conducting
multilevel  groundwater sampling
and for soil sampling below the
water table.
     Clearly, there is a growing reg-
ulatory acceptance  of the  three-
dimensional  approach   to  site
assessment. A number of states and
the U.S. Environmental Protection
Agency have published guidance on
applying direct-push technologies
and multilevel sampling and per-
forming expedited site assessments.
We have developed guidelines for
performing multilevel sampling  in
conducting  expedited site  assess-
ments for the Connecticut Depart-
ment of Environmental Protection
(DEP). (See LUSTLine Bulletin #26.)
The guidelines contain the following
information: reviews of methods for
multilevel sampling for groundwater
and soil; practical guidance on the
application of methods; the results of
studies to resolve key issues associ-
ated  with  conducting  multilevel
sampling;  comparisons of results
from different multilevel sampling
methods; assessments of statistical
uncertainties; and  comparisons of
results  obtained by conventional
groundwater monitoring wells and
by multilevel sampling.
     To provide further guidance on
integrating   multilevel  sampling,
field  screening, and three-dimen-
sional data evaluation in the context
of  conducting  an expedited site
assessment, we are developing guid-
ance for the DEP in the form of a CD-
ROM. The  CD, which  should be
available  in September 1998, will
present the guidance material in a
multimedia format that will include
slide presentations, PC video, mod-
ules for performing calculations, and
case  studies.  The CD will  be
designed to serve as a tutorial, a ref-
erence, and a guide for regulators,
their consultants, and the regulated
community  on evaluating  three-
dimensional information  at UST
sites. Such an evaluation should go a
long way in helping to solve prob-
lems  that have  historically defied
resolution. H	

   Gary Robbins is a professor in the
   Department of Geology and Geo-
 physics at the University of Connect^
      cut in Starrs, CT. For more
 information, contact Gary by e-mail at
    Robbins@uconnvm.uconn.edu.
                                                                                                         •V7

-------
LUSTUnc Bulletin 27
 MIKE   Update
New National Science and Technology
Council Report Provides a Comprehensive
Summary of Oxygenated Fuel Issues
This June, the National Science and Technology Council
released its Interagency Assessment of Oxygenated Fuels
report, which presents current understanding of critical
scientific issues related to oxygenates in gasoline. The
assessment was undertaken by the Council in response to
an Environmental Protection Agency request for a com-
prehensive, interagency review of the winter oxygenated
gasoline program for potential health impacts, fuel econ-
omy and performance issues, and benefits. The document
provides an authoritative evaluation of existing informa-
tion and helps to identify areas where the data are too
limited to draw conclusions about the impacts of the oxy-
genated fuels program.
     The report identifies areas where the data are insuf-
ficient to make definitive conclusions about the costs,
benefits, and risks of using oxygenated fuels in place of
conventional gasoline. It also identifies where research is
needed to help reduce uncertainties and allow a more
thorough assessment of human exposure, health risks
and benefits, and environmental effects. To get a handle
on these uncertainties, several federal agencies are devel-
oping plans to expand monitoring and research efforts on
occurrence of oxygenates in drinking water, the extent of
human  exposure to oxygenates, probable effects of
human exposures, site remediation, and impacts on
aquatic life.
     The interagency assessment considers four main
issues associated with the oxygenate program: air quality,
groundwater and drinking water quality, fuel economy
and engine performance, and potential health effects.
Each of these  subjects is addressed in a separate chapter.
Each chapter  underwent extensive external peer review
before it was submitted for review to the National
Research Council of the National Academy of Sciences.
The entire assessment xvas reviewed by the National Sci-
ence and Technology Council.
     To obtain a copy of this report, contact GCRIO User
Services by mail: 2250 Pierce Rd., University Center, MI
48710; by phone: (517) 797-2730; by fax: (517) 797-2622; or
by e-mail: help@gcrio.org. •

OUST to Release Series of MTBE Fact Sheets
OUST is developing a series of  seven fact sheets on
MTBE. Each  of the fact sheets will focus on a specific
topic. Fact sheets 1,2, and 3 are called Overview, Remedia-
tion ofMTBE-Contaminated Soil and Groundwater, and Use
and Distribution of MTBE and Ethanol. The remaining fact
sheets will cover the U.S. EPA health advisory, analytical
methods for fuel oxygenates, impacts of MTBE releases
on state UST  programs, and potential oxygenate substi-
tutes for MTBE. Illustrations and tables will accompany
the text. OUST is planning to distribute hard copies of the
fact sheets to  states (including field offices), regions, and
federal facilities. OUST anticipates  that the first three fact
sheets will be ready for distribution in early 1998. OUST
will also post the fact sheets (in WordPerfect 6.1) on its
web  site  at  (http://www.epa.gov/  OUST/ mtbe/
mtbepubs.htm); you will be able to download them. •

New Testing Requirements Will Help
California Regional Water Quality Control
Board Keep Tabs on Oxygenates
Gordon Lee Boggs, Underground Tank Manager at the
Central Valley Regional Water Quality Control Board in
California, has been learning more about gasoline oxy-
genate compounds than he'd ever hoped to know. He
knew that MTBE and ethanol were principal players in
the gasoline oxygenate arena, but then he found out that
there were other common oxygenates in the arena as
well—specifically, methanol, tertiary butyl alcohol (TEA),
di-isopropyl ether (DIPE), ethyl  tertiary butyl ether
(ETBE), and tertiary amyl methyl ether (TAME). To get a
handle on the situation, the UST program expanded its
testing requirements to include soil and water analyses
for TBA and the ether compounds to determine which
oxygenated compound is present.
     "We learned," says Boggs, "that the oil companies
trade gasoline around the state and around geographic
areas, which means, of course, that these different oxy-
genates are being traded around as well. When a gasoline
release occurs, we won't know what's out there and
where unless we test for it. We're finding out that more
and more things are out there, but we don't know what to
do about them at this time. We decided that by requiring
testing for these constituents, we'll  have the opportunity
to locate problems and compile a database."
     Once the Board got its facts together and established
its testing requirements, Boggs wrote a "To Whom It May
Concern" letter to spell out the facts. His letter, originally
dated 18 July 1997, has undergone a few revisions and has
generated a great deal of interest in the LUST community.
Here is a slightly abbreviated version of the letter:

       Ethanol has been used for several years in Califor-
   nia. MTBE, as you know, has been used as an octane
   enhancer since the late 1970s and is now used at a
   higher percentage as an oxygenate in gasoline. Now
   we have learned that TAME has been added to Califor-
   nia fuels since 1995. DIPE has been used on the East
   Coast, but recently has been found in Southern Califor-
   nia groundwater and San Joaquin County. TBA has
   been found in groundwater at a service station site in
   San Joaquin County. To date, we  have no information
   about the use of ETBE in California.
       The introduction of these additives presents ana-
   lytical problems for laboratories because the multiple
   analytes can co-elute from the column. For instance,
   TAME may co-elute with benzene in the EPA analyti-
   cal method commonly used today.  [See "MTBE—
   Beware the False  Positive"   in LUSTLine #26.]
   Therefore, we believe that Mass Spectrometry (MS) is
   the most definitive procedure for determining oxy-
   genate compounds. MS will likely increase the cost of
   sample analysis; but, until  another  comprehensive
   analytical method is developed  that  can distinguish
   between compounds, we believe that EPA Method
   8260 is the most reliable, readily available procedure
18

-------
                                                                                              LUSTLine Bulletin 27
 for laboratories. Completion of proposed studies on
 analytical procedures for oxygenate compounds by the
 Lawrence Livermore National Laboratory in early 1998
 should provide additional guidance.
     The problem of identifying which oxygenates are
 present is further  complicated because of gasoline
 swapping; the oil refineries ship gasoline around the
 state and then trade gasoline among geographic areas.
 As a result, we don't know what oxygenate compound
 might be in the gasoline at a particular station—and
 what will end up  in the groundwater if there is  a
 release.
     Research recently  completed by DuPont-Dow
 (http://www.dupont.com/products/viton/lkprev.html)
 shows that oxygenates may be incompatible with some
 elastomer seals used on UST piping. One short-term
 test (168 hours) that used several concentrations of
 MTBE showed swelling could occur with some elas-
 tomers at current gasoline mixture levels. Presumably,
 this will be true, to some extent, for all ether oxygenate
 additives and conceivably, over a longer time, cause
 the failure of the seals, thereby  releasing  the  oxy-
 genated gasoline into the environment. We are particu-
 larly concerned that older tank seals or material used
 to upgrade tanks may not be compatible with the oxy-
 genates and may fail due to high concentrations of oxy-
 genates in the alternative fuel sources. [In the  next
 issue of LUSTLine we will have an article on this com-
 patibility issue.]
     On 14 August 1997, a workshop was conducted at
 the Sacramento office of this regional board. It was
 attended by  representatives of regional and state
 boards, local implementing agencies,  analytical labora-
 tories, and the petroleum industry. The objective was
 to provide guidance to the regulated community on
 how and where to analyze for the oxygenated com-
 pounds in gasoline until a definitive protocol can be
 established in several months. The goal was to allow
 closure of underground tank sites with assurances that
 the interim methodology can detect and  quantify oxy-
 genates.
     The workshop  attendees concurred  that the
 methyl and ethyl alcohols can't be detected by  EPA
 Method 8260  with certainty and that detection limits
 for methyl and ethyl alcohols are about two orders of
 magnitude higher than TEA and the ethers. Also,  with
 the exception of one oil company and special, alterna-
 tive fuel vehicles, ethanol and methanol are used infre-
quently in California and can be isolated by station
and the more analytical methods used. Therefore, at
this time, unless ethanol or methanol are specifically
requested, we' are requiring soil and water analysis
only for TEA and the ether compounds by EPA
Method 8260 in order to determine which oxygenated
compound is present.
    Presence or absence of the oxygenate must be
reported whenever gasoline range hydrocarbons are
present. However, because free product or high petro-
leum concentrations raise the detection limits of the
oxygenates, the oxygenates cannot be detected with
certainty. Therefore, at this time, we do not recommend
sampling where product is present on groundwater.
    Quarterly water samples  for oxygenate  com-
pounds are to be taken from all monitoring wells at
sites with ten or fewer wells. At sites with more than
ten monitoring wells, requests to change the sampling
procedure must be approved by the regional board or
local implementing agency. Soil samples should be
analyzed beneath the primary leak source(s) (i.e., tank,
pipeline, dispenser) at regular depth intervals to the
groundwater interface. If only MTBE is found in the
initial sampling/analysis, as confirmed by 8260, the
responsible party may cpntinue analysis for the oxy-
genate by EPA Method 8020. Prior to requesting site
closure, a "confirming round" must be completed for
all oxygenate compounds.
    Laboratories must include all listed oxygenated
standards (TBA and ethers) in their calibration stan-
dards and follow the QA/QC protocol detailed in EPA
Methods 8000 and 8020, or 8240-B or 8260, and the
Code of Federal Regulations (CFR title 40, parts 136.4
and 136.5).
    In addition to the compounds discussed above,
we have found that both tertiary  amyl ethyl ether
(TAEE) and isopropyl alcohol (IPA) may also be added
to gasoline; PRIST® (ethylene glycol monomethyl
ether) is added to aviation fuel and, reportedly, some
diesel fuels to prevent clogging of the fuel lines by
microorganisms. PRIST is registered as a pesticide with
antimicrobial properties: CAS#000109864. These com-
pounds may be added to the analysis list as more infor-
mation becomes available. •

     For more information on this issue, contact
        Gordon Lee Boggs at (916) 255-3139.
       L.U.S.TLINE
   One-year subscription. $30.00.

Q Federal, state, or local government. Exempt from fee. (If you wish to have LUSTLine sent to your home, please submit your request on agency letterhead.)
Q Please take my name off your mailing list.

Q Please send me back issues of LUSTLine. Fill out name and address - no P.O. boxes. Back issues now cost $2.50 per issue or $25.00 for a complete set.
Name	  Company/Agency	
Address	
             Street                                      City/Town                        State        ZIP
Please enclose a check or'money order (drawn on a U.S. bank) and made payable to NEIWPCC.
Send to: New England Interstate Water Pollution Control Commission
       255 Ballardvale Street, 2nd floor • Wilmington, MA 01887-1013
       Phone: 978/658-0500 • Fax: 978/658-5509 • lustline@neiwpcc.org • www.neiwpcc.org
We welcome your comments and suggestions on any of our articles.

-------
  A NOTE ON THE
     STATE FUND
   STATISTICS ON
      PAGE 28 OF
   LUSTLINE #26:
[ Because of the many questions regarding the high value for "average (cleanup) cost per site" in 1997,' Ver-1
| mont Department of Environmental Conservation staff took a closer look at all of the reported values for 1
! individual states. They discovered that, in fact, one state had misinterpreted the question. That state recal- j
 dilated its cost per site and came up with a number that was considerably lower than originally reported. •;
 This new value, in turn, had a significant effect on the national average. Hence, the figure of $107,975 has j
I been reduced to $60,158 as an average cost per site in 1997.                                       *
  Bulk Copies of '98 Deadline
  Materials Available
  With December 1998 only a year
  away, the EPA is sending a letter to
  EPA regional program managers
  and state UST program managers
  to remind them that OUST can pro-
  vide bulk quantities of materials on
  the 1998 deadline.
  New Publications from
  OUST
  • Underground Storage Tanks:
  Requirements   and    Options
  (EPA-510-F-97-005).  This  leaflet
  reminds UST  owners /operators
  who do not sell petroleum products
  but who may fuel vehicles from
  their own USTs (nonmarketers)
  that they need to comply with UST
  requirements. A list of additional
  documents and ordering informa-
  tion are included. All  materials
  listed urge readers to check with
  state and local regulatory authori-
  ties for additional or more stringent
  requirements.
  •  Controlling  UST  Cleanup
  Costs: Fact Sheets (EPA 510-F-97-
                 «EPA HQ UPDATE
                  006). Five fact  sheets originally
                  published in May 1992 are reissued
                  in this new compilation,  which
                  includes an update page noting
                  new materials. The fact sheets have
                  not been altered. They cover: hiring
                  a contractor; negotiating the con-
                  tract; interpreting the bill; manag-
                  ing the process; and understanding
                  contractor code words.

                  • Straight Talk on Tanks: Leak
                  Detection Methods for Petro-
                  leum   Underground   Storage
                  Tanks and Piping (Third Edition)
                  (EPA 510-B-97-007). The Third Edi-
                  tion  of this popular booklet con-
                  tains updated and revised text, a
                  page clarifying the time restrictions
                  applicable to the combination of
                  tank tightness testing with either
                  inventory control or manual tank
                  gauging, more information on find-
                  ing  help  and  free  publications
                  using toll-free numbers or OUST's
                  web site, and notice of the availabil-
                  ity of the publication List of Leak
                  Detection Evaluations for UST Sys-
tems. Please note that the booklet
urges readers to check with state
and local regulatory authorities for
additional   or   more   stringent
requirements.

• Are You Upgrading an Under-
ground Storage  Tank System?
(EPA-510-F-97-009).  This  leaflet,
which is for UST owners and oper-
ators who are facing upgrade deci-
sions, can help them make sound
decisions about  choosing  tank
integrity assessment methods and
upgrading USTs to meet 1998 dead-
line requirements. Please note that
the leaflet urges readers to check
with state  and  local regulatory
authorities for additional or more
stringent requirements.
 All of the materials listed above
   can be downloaded (in Word-
 Perfect 6.1) from OUST's World
     Wide Web home page at
   http://www.epa.gov/ OUST/.
   Printed copies are available
 from NCEPI at (800)490-9198 or
   via EPA's toll-free hotline at
         (800)424-9346.
 LU.ST.UNE
New England Interstate Water
Pollution Control Commission
255 Ballardvale Street
Wilmington, MA 01887
Forwarding and return postage guaranteed.
Address correction requested.
 L.U.S.T. Buster T-Shirts &
       Sweatshirts!

    Tec's M, L, XL, XXL  $9.00pp
    Sweats: M, L, XL, XXL $16.50pp
    Allow 4-6 weeks delivery.
  StnA chKk Of money order Urawn on U.S.
 hmfa only) !
-------