United States
Environmental Protection
Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S2-87/062 Nov. 1987
&EFA Project Summary
Quantification of Leak Rates
Through Holes in Landfill Liners
K. W. Brown, J. C. Thomas, R. L. Lytton, P. Jayawickrama, and S. C. Bahrt
A study was undertaken to evaluate
the rate at which liquids leak through
flaws in the flexible membrane liner
(FML) component of composite FML-
soil liners. The variables studied were:
flaw size and shape, FML type and
thickness, the influence of a geofabric
between the compacted soil and the
FML, the conductivity of the soil sub-
base, the liquid head, and the liquid
characteristics. Testing was done in 60
cm diameter permeameters. Soils were
compacted in the permeameter and
overlain with the FMLs to be tested
with either round holes, slits, or seam
flaws. A 15 cm layer of gravel was
placed over the FML to provide ballast,
and a head chamber was used to apply
as much as 100 cm of head on the
FML. Tests were conducted with a gravel
subbase to determine the influence of
the flaw alone on the flow rate followed
by soil subbases having nominal con-
ductivities of 1 x 10 * cm gr1 and 1 x
10 6 cm s'1. A calculations! procedure
was developed to simulate the flow
rates through the permeameters and
was modified to allow calculation of
leak rates under field conditions.
The flow of liquids through flaws in
FMLs was primarily dependent on the
size and shape of the flaw, the liquid
head, and the hydraulic characteristics
of the subbase. It was nearly indepen-
dent of the liner thickness, liquid pro-
perties, and the presence or absence of
an underlying geotextile.
Variability in flow rates through seam
flaws and slits was much greater than
that through round holes due to the
variable hole sizes that could result if
the seam or one side of a slit was
displaced relative to the other. As a
result, the average leak rates through
slit and seam flaws over a gravel sub-
base increased over twelve fold when
the flaw length was increased by a
factor of 3, from 5 to 15 cm.
For soil subbases, the head loss across
the system may be divided into the
head loss as the liquid enters the hole,
the head loss across the hole in the
FML, the head loss as the liquid flows
laterally between the FML and the soil,
and the head loss through the soil. The
head loss, in the liquid flowing laterally
between the FM L and the soil, depended
on the width of the gap between the
two media. The gap widths for the ia4
and 10 6 soils were estimated from the
permeameter data to be 0.015 and
0.002 cm, respectively. Thus, less
permeable soils containing greater
amounts of clay form a better seal with
the FML and allow less lateral flow of
liquids. Gap widths and resultant flow
rates were also decreased by overburden
pressure from simulated layers of waste
or liquid head.
In the final report evidence is pre-
sented which indicates that erosion of
the subbase can occur just below a flaw
in a FML, particularly when the liquid
head is large, as would occur in a lagoon,
and when the subbase conductivity is
greater than 10"6 cm r1.
Graphs are presented from which
permit writers can estimate the potential
leakage rates from field installations.
Th/» Project Summary was developed
by EPA's Hazardous Waste Engineering
Research Laboratory, Cincinnati, OH, to
announce key findings of the research
protect that Is fully documented In a
separate report of the same title (see
Project Report ordering Information at
back).
-------
Introduction
A variety of flexible membrane liners
(FMLs) have been utilized in the past for
lining landfills and surface impound-
ments. Typically, they range in thickness
from 0.06 to 0.25 cm and are marketed in
various size sheets and rolls. The mate-
rials are installed and seamed in the field
to conform to the shape of the impound-
ment. They are usually installed over a
clean, smooth, compacted soil and often
are covered with a protective soil layer
before waste is placed in the surface
impoundment or landfill. In some installa-
tions, a fabric or geotextile is placed on
top of the subgrade to protect the FML
from being punctured from below.
In the past, despite these precautions,
leaks have been detected in many of the
facilities lined with FMLs. These leaks
may be a result of imperfect seaming,
rips, punctures, tears that occur during or
after installation, or failures that result
from subsidence or shear failure of the
supporting soil after installation. Also,
failure may result from chemical in-
compatability and solvent attack which
may dissolve either the plastic or the
plasticizer. Facilities that have functioned
well for years will FMLs have been known
to fail rapidly when exposed to chemicals
for which they were never intended.
At the present time, there is a dearth of
knowledge on leakage rates through flaws
in FMLs. It is, therefore, important that
these leakage rates be quantified and the
principles governing leakage rates be
understood so that predictions of leakage
rates can be made. The leakage rate may
be affected by the following parameters:
1. The type of FML,
2. The FML thickness,
3. The size and shape of the flaw,
4. The characteristics of the subbase
material,
5. The presence or absense of a geo-
textile between the subbase and
FML,
6. The head of liquid above the flaw,
and
7. The characteristics of the liquid to
be retained.
The thrust of this research was to
evaluate the effect of each of the above
factors on leakage rate through flaws in
FMLs. Physical measurements of flow
rates were made and used to develop a
calculational procedure for predicting
leakage from any given set of input
parameters.
Materials and Methods
Samples of several thicknesses of FML
materials were obtained for testing. The
materials included the following: 0.05
and 0.08 cm (20 and 30 mil) thicknesses
of polyvinylchloride (PVC); 0.08,0.20, and
0.25 cm (30,80 and 100 mil) thicknesses
of high density polyethylene (HOPE); 0.08
cm (30 mil) ethylene propylene rubber
(EPDM); and 0.09 and 0.11 cm (36 and
45 mil) thicknesses of chlorosulfonated
polyethylene (CSPE, trade name Hypalon).
Each was cut into square pieces 66 cm
on a side to fit in specially constructed
round permeameters. Each permeameter
had an inside diameter of 57.2 cm and a
height of 30.5 cm. Each permeameter
was filled with a subbase consisting of
either gravel having a conductivity of 10 1
cm sec'1, a sandy soil having a con-
ductivity of 10'4 cm sec'1, or a clay soil
having a conductivity of 10~6 cm sec"1.
The permeameter with appropriate sub-
base was overlain with a FML section
having the desired flaw and fitted with a
head tank. Fifteen cm of gravel were
placed above the FML to serve as ballast
and as much as 95 cm of the permeating
liquid to be tested was added. Several
permeameters and head tanks were
modified to simulate liquid depths in
excess of 1 m and large overburden pres-
sures. The majority of tests were run
with water as the permeant while a lesser
number employed simulated landfill
leachate and waste xylene as the per-
meant. Each permeameter was equipped
with a water stage recorder used to
measure changes in water level in the
head tank. When the flow rate was very
small, 14 cm diameter stand pipes were
used to increase resolution. Checks of
the conductivity of the subbase soils
without the presence of any FML were
made periodically to document any
intrinsic changes in conductivity.
Because the subbases could not be
compacted to exactly the same conduc-
tivity, a calculational procedure was
developed to smooth the data and adjust
them to selected conductivities for com-
parison. The procedure partitioned the
total head loss into that which occurs as
the liquid enters the hole, the head loss
as the liquid flows laterally between the
FML and the subbase, and the head loss
through the soil. The procedure was
calibrated using the permeameter data
and then modified by removing the
boundary conditions imposed by the
permeameter walls to allow extrapolation
to field conditions.
Results and Discussion
In tests of FMLs. overlying a subbase
having a conductivity of 10~1 cm sec"1,
the size and shape of the flaw were the
primary determining factors in the leakage
rate. Material type had a lesser effect
with the PVC and CSPE materials having
a somewhat slower leakage rate pre-
sumably due to the very flexible nature of
these materials. The thickness of the
FML and the presence or absence of an
underlying geotextile made no difference
in leakage rate. Data on the maximum
anticipated leakage rates from various
size and shape flaws over a very per-
meable gravel subbase are summarized
in Table 1. This information may be used
as a guide when designing the drainage
system for a double lined facility to
estimate the amount of liquid which may
be necessary to remove annually. It can
also be used to estimate the number
and/or size of flaws present in an existing
FML which leaks into a drainage system.
Table 1. Average Leak Rates (M3 YFf1)
From Different Size and Shape
Flaws in 0.08 CM Thick HOPE
Liner Over Gravel at Two Liquid
Heads.
Head (cm)
Hole Size and
Shape
SO 100
m3yr->
0.16 cm diameter 110
0.64 1482
1.27 4257
5 cm slit —
15 cm slit 3866
5 cm seam 404
15 cm seam 4702
145
2208
6780
79
5623
325
7244
In tests employing soil subbases ol
various conductivities, it was found thai
flaw size and subbase conductivity were
the predominant controlling parameters
in determining the leakage rate through
a defective FML. Leakage rates through
slits and seam flaws were much more
variable than those through holes due tc
the possibility of misalignment of the
materials. The flow rates through holes
in FML's overlying soils will be primarily
controlled by the conductivity of the soil
The type of FML material and presence 01
absence of a geotextile had very little
effect on the flow rates. The effects 01
FML thickness and the liquid properties
of the permeant also had little effect or
the flow rates. The calculational proce
dure used to smooth the permeametei
data was modified by removing the con
-------
fining permeameter walls to allow ex-
trapolation to field conditions. Suitable
changes for the liquid properties can be
added to the calculations! procedure if
needed for permeants other than water.
Predicted leakage rates from various sizes
of holes in FMLs overlying soils of dif-
ferent conductivities are summarized in
Table 2. The predicted radius of the wetted
area are also given in Table 3.
Thus, the data presented can be used
to estimate the potential leakage rate
from a damaged FML under a given set of
conditions. The data can also be used to
estimate the number or size of flaws if
the leakage rate is known. For those
persons involved in the design and design
review of new facilities, the calculations!
procedure and program presented in the
final report will be of assistance in
adequately designing the initial system
and in designing an adequate monitoring
system.
Table 2. Calculated Leak Rates (M3 YR'1) for a Range of Hole Sizes in Flexible Membrane
Liners Over Soils of Different Conductivities. The Values are Given for Three Heads
Hole diameter (cm)
Ksajcm/s)
3.40 x JO'4
3.40 x 10-5
3.40x10-*
3.40 x JO-7
3.40 x W4
3.40 x JO-5
3.40 x ro-6
3.40 x IP7
3.40XW4
3.40x1 Or6
3.40 x 1O-6
3.40 x W7
0.08
// =
19.30
4.30
0.54
0.066
H =
42.30
12.80
1.66
0.20
H =
167.0
84.6
14.3
1.8
0.16
0.3 M
31.50
4.88
0.60
0.72
1.0M
87.80
14.80
1.83
0.22
10.0 M
438.0
123.1
15.6
1.9
0.64
43.20
6.28
0.77
0.095
128.00
18.70
2.29
0.28
1.030.00
153.50
18.80
2.30
1.27
50.60
7.30
0.89
0.107
147.00
21.40
2.61
0.32
1.170.00
171.30
21.00
2.60
Table 3.
Calculated Radius of Wetted Area (M) for a Range of Hole Sizes in Flexible
Membrane Liners Over Soils of Different Conductivities. The Values are Given for
Three Heads
Hole diameter (cm)
0.08
0.16
0.64
1.27
H = 0.3M
3.40 x 104
3.40 x JO-5
3.40 x JO-6
3.40 x 10r7
0.24
0.36
0.40
0.44
0.31
0.38
0.42
0.47
0.36
0.43
0.48
0.53
0.39
0.47
0.51
0.57
3.40 x JO-4
3.40 x 10r5
3.40 x JO-6
3.40 x JO'7
H=1.0M
0.36
0.62
0.70
0.78
0.51
0.66
0.74
0.82
0.62
0.75
0.82
0.91
0.66
0.80
0.88
0.97
H=JO.OM
3.40 x 1QT4
3.40 x IP5
3.40 x ia6
3.40 x JO'7
0.70
1.59
2.06
2.29
1.14
1.91
2.15
2.39
1.75
2.14
2.36
2.62
1.86
2.26
2.50
2.77
K. W. Brown, J. C. Thomas, R. L. Lytton, P. Jayawickrama, and S. C. Bahrt
are with Texas A. & M. University. College Station, TX 77843.
Paul R. de Percin is the EPA Project Officer (see below).
The complete report, entitled."Quantification of Leak Rates Through Holes in
Landfill Liners." (Order No. PB 87-227 666/AS; Cost: $18.95, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Officer can be contacted at:
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States Center for Environmental Research
Environmental Protection Information ___—
Agency Cincinnati OH 45268
"~" f^V"S%jr 0£t~t
Official Business '" V r ~ /-* M"AB! --'*
Penalty for Private Use $300 „ V,^_.H'<>' 6.' S• o' ^
EPA/600/S2-87/062
0000129 PS
U S £»*V1R PROTECTION AGENCY
REGION 5 LIBRARY
230 S 06ARBORS STREET
CHICAGO IL 60604
------- |