United States
Environmental Protection
Agency
Municipal Environmental
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-84-058 Apr. 1984
Project Summary
Innovative Concepts for
Detecting and Locating Leaks in
Waste Impoundment Liner
Systems: Acoustic Emission
Monitoring and Time Domain
Reflectometry
J.L. Davis, R. Singh, B.G. Stegman, and M.J. Waller
A project was conducted to develop
reliable, nondestructive methods for
detecting leaks in lined waste impound-
ments before the leachate can seriously
damage the ground water. A second
goal was to find a technique to locate
the precise area where a known leak is
occurring. After a preliminary study in-
volving a literature review, a state-of-
the-art survey, and a ranking matrix ex-
ercise, two techniques were recom-
mended for immediate further study —
acoustic emission monitoring (AEM)
and time domain reflectometry (TOR).
AEM determines the amplitude and fre-
quency responses of sounds emitted
when water flows at different rates
through soils. TDR is a high-frequency
electromagnetic technique that mea-
sures the electrical properties of ma-
terials in and around the conductors of
a transmission line.
Laboratory and field studies were
undertaken to evaluate the potential of
these two techniques. Results indicated
that AEM potentially offers a practical
and inexpensive technique for leak de-
tection and monitoring in both existing
and planned liquid waste impound-
ments.
TDR appears to be a practical method
for detecting and determining the posi-
tion and extent of a leak under a small
waste impoundment liner. Further test-
ing is needed to determine the useful-
ness of both techniques in the field,
however.
This Project Summary was developed
by EPA's Municipal Environmental Re-
search Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
Each year about 35 million metric tons of
hazardous wastes are disposed of in im-
poundments lined with natural or manmade
materials. The liners act as barriers to impede
the flow of the hazardous wastes to the
underlying ground water. But presently, lit-
tle information exists on the effectiveness of
the different liner materials, and monitoring
methods are unsatisfactory. Ground water
is presently monitored through wells placed
hydrologically down stream. Thus wastes
must actually be present in the ground water
before they are detected, and significant
damage may already have occurred. No
methods currently exist that can detect the
source of a leak in a liner before the leachate
reaches the ground water.
The purpose of this project was to develop
reliable, nondestructive methods for detect-
ing significant leaks in lined waste impound-
ments before the leachate can seriously
-------
damage the ground water. A second goal
was to find an economical technique that
can locate the precise area where a known
leak is occurring in the liner.
The first step in the project was to iden-
tify and evaluate a wide range of techniques
for liner leak detection. A preliminary study
was conducted for this purpose (M.J.
Waller, J.L. Davis, and William Kean, 1981.
"Assessment of Innovative Techniques to
Detect Waste Impoundment Liner Failures,"
EPA-600/2-84-041, U.S. Environmental Pro-
tection Agency, Cincinnati, OH). After a
thorough literature review, state-of-the-art
survey, and ranking matrix exercise, this
earlier study recommended two techniques
for immediate further study — AEM and
TDR. AEM determines the amplitude and
frequency responses of sounds emitted
when water flows at different rates through
soils. TDR is a high-frequency electro-
magnetic technique that measures the elec-
trical properties of materials in and around
the conductors of a transmission line.
The present study was undertaken to per-
form laboratory and field studies to evaluate
the potential of these two techniques for
detecting leaks under waste impoundment
liners.
Acoustic Emission Monitoring
AEM techniques have been used for at
least 50 years to evaluate stresses in rocks
and metals, but their application to soil struc-
tures is more recent. Acoustic emission (AE)
techniques have shown potential use in earth
stability monitoring, settlement and deforma-
tion monitoring, and water flow monitoring
through earthen dams. To determine the
potential application of AEM for detecting
leaks from lined waste impoundments, ex-
periments were conducted to measure the
amplitude and frequency responses of
sounds emitted when water flows at dif-
ferent rates through different soils and rips
in a liner material.
Methods and Equipment
Experiments were carried out with three
objectives: (1) to determine whether sounds
are emitted when water flows through soils
and rips in liners, (2) to characterize the
sounds emitted when water flows through
soils and rips in liners, and (3) to determine
the attenuation of these sounds in water.
Laboratory studies were conducted to fulfill
the first two objectives, and field studies
were carried out to achieve the third.
Soil Columns
The laboratory experiments for determin-
ing and characterizing sounds produced
from water flowing through soils and rips in
liners were conducted in soil columns held
by a polyvinylchloride (PVC) duct 4 m long
and 30 cm in diameter (Figure 1). Different
heights of soils could be placed in the col-
umn and water could be passed through the
soils at different known rates. Some soil col-
umn experiments were conducted using
30-mil PVC sheets with either 5- or 15-cm
rips. A microphone was placed in the soil of
the columns and the sounds were amplified,
processed, and displayed. Special care was
taken to reduce unwanted background
noise. A spectrum analyzer capable of cap-
turing transient events was used to process
the data. The amplitude and frequency re-
sponse of the output signal was displayed
on an oscilloscope, recorded on a digital tape
recorder, and plotted on an X-Y plotter.
To Amplifiers,
, _,, Filters, and
*^^~-{'.'?K\. „ . r
' if, Display
30 cm
Diameter
PVC Duct
4m Long
Water Level
Sight Tube
%
_ Acoustic
1£ Insulator
Water
Funnel
To Reservoir
Figure 1. Schematic diagram of the experi-
mental acoustic emission test
arrangement.
Electronic Equipment
The electronic equipment used to detect,
amplify, process, and display the acoustic
data included an acoustic receiver (Figure 2),
a Weston Acoustic Emission Monitor* with
variable gain amplifier of 1 to 5000, a Weston
AEM filter unit modified to cover a frequency
range of 20 Hz to 5 KHz, a Nicolet 446B mini
ubiquitous spectrum analyzer, an analogue
X-Y plotter, a sweep oscillator, and a
loudspeaker.
Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
Measurement Procedure
The method generally used for laboratory
measurements was to place the microphone
in the soil being tested (as shown in Figure
1) and let the water flow through the soil.
The height of the soil in the column varied
from 70 to 150 cm. The microphone was
placed 15 cm above the bottom of the soil
column (i.e., above the 45-mesh screen).
The flow rate and the acoustic spectra were
measured, and the spectral data were plot-
ted on the X-Y recorder. The water level was
recorded on an X-time recorder. After about
every 30- to 40-cm drop in water level, the
flow was stopped and a spectrum of the
background noise spectra was obtained and
subtracted from the response obtained when
the water was flowing through the soil.
Soils Tested
The soils used in these experiments were
20/30 Ottawa sand and a pea gravel, with
an average grain size of 0.7 mm. The grain
size of more than 90 percent of the pea
gravel sample fell between 25 and 5 mm,
with the remaining portion between 5 and
1 mm.
Water Levels
The water level varied from about 3 m
above the soil to just above the soil in each
of the experiments; thus falling water head
levels occurred during the experiments. The
flow rates typically were varied from 0.3 to
1 cm/sec.
Attenuation of Sound in Water
An audio frequency sweep oscillator work-
ing over a frequency range from 20 to 500
Hz was used to drive a loudspeaker specially
designed to operate underwater. The stan-
dard AEM system used throughout these
studies was used as the receiver. The AE
sensor was moved away continuously from
the loudspeaker at a depth of 1 m in a lake
whose depth was 2 meters. The sound am-
plitude over the frequency spectrum was
measured as the distance between the
source and receiver was varied from 0.2 to
5 meters. The attenuation was measured at
10 dB/m over the frequency range of 100 to
500 Hz in the lake. It was observed that the
attenuation of the sounds decreased sig-
nificantly as the frequency was increased
into the kilohertz range.
Results and Conclusions
1. Acoustic signals are emitted when water
flows in a turbulent mode through soils.
Acoustic signals at frequencies up to 500
Hz are emitted when water flows at rates
of 0.3 to 1 cm/sec through a sand and
a pea gravel.
-------
Amplifier
EM
/A
-t-
3andpa
Filter
1
1
J
ss
Nicolet
Spectrum
Analyzer
X-Y
Plotter
Recorder
D
Microphone
(Hydrophone)
Figure 2.
A block diagram of the electronic equipment used to detect acoustic signals emitted
when water flows through soils.
2. The flow must be turbulent for acoustic
signals to be emitted. Turbulent flow oc-
curs at rates down to 0.1 cm/sec (1 gpm
in the 30-cm-diameter test column) with
the sands tested.
3. The amplitude of the sounds increases as
the turbulent flow rate increases. For ex-
ample, at a flow rate of 0.3 cm/sec (4
gpm through the 30-cm-diameter test col-
umn), the amplitude of acoustic signals
through coarse-grained soils is about
equal to the background noise likely to
exist at liquid impoundment sites. But at
an increased flow rate of 1 cm/sec (10
gpm in the 30-cm column), the amplitude
is 100 times greater.
4. At a flow rate of 0.3 cm/sec, the sound
amplitude increases with increasing varia-
tion in soil grain size.
5. At a flow rate of 0.3 cm/sec, the sound
amplitude decreases with increasing soil
density.
6. The sounds emitted when water flows
through soils have significant amplitude
at frequencies up to 500 Hz, and they
have a peak amplitude between 100 and
200 Hz.
7. The attenuation of sounds between 100
and 500 Hz is 10 dB/m near the surface
of a lake.
8. No significant sounds were detected
when water flowed at rates up to 1
cm/sec through either a 5- or 15-cm rip
in a 1-mm-thick PVC liner material.
Recommendations
1. Determine the spectral response for
background sounds emitted at existing
waste impoundment sites at frequencies
up to 500 Hz.
2. Determine the attenuation of sounds at
frequencies from 20 to 500 Hz in a typical
waste impoundment site environment.
3. Determine the advantage of different
arrays of acoustic sensors for detecting
leaks at waste disposal sites.
4. Determine the spectral characteristics for
acoustic signals emitted from leaks with
known flow rates at a field model of a liq-
uid impoundment site.
5. Determine the spectral response of
sounds emitted when plastic liner ma-
terials creep.
6. Determine whether AE occurs when
water flows through a silt or clay soil.
Time Domain Reflectometry
TDR is a wide-frequency bandwidth,
short-pulse-length measurement technique
that is sensitive to the high-frequency elec-
trical properties of the material in and around
conductors of a transmission line. The high-
frequency (106 to 109 Hz) dielectric proper-
ties of water in its liquid state are about 20
times greater than the high-frequency dielec-
tric properties of dry geologic materials: thus
TDR is primarily sensitive to changes of
water content in geologic materials and
relatively insensitive to changes in soil type
and density, temperature, and pore liquid salt
content.* In materials with a relatively
uniform water content, TDR is sensitive to
changes of soil and rock type. This method
is also sensitive to changes in the phase of
the water molecule, and thus frozen or un-
frozen wet soils. In addition, TDR can detect
either water- or air-filled fractures and voids
in the materials along the transmission line.
Leaks from liners of waste impoundments
into the soil below are likely to have different
electrical properties from those of the host
material: thus TDR can measure leachates
in the soil material along a transmission line.
The TDR technique, using transmission lines
placed under a liner offers the significant
potential of being a practical method for
detecting small leaks from waste dump sites.
Topp, G.C., J.L Davis, and A.P. Annan, "Elec-
tromagnetic Determination of Soil Water Content:
Measurements in Coaxial Transmission Lines," Water
Resources Research, Vol. 16, No. 3, June 1990.
The electrical measurements in soils are
carried out by placing a transmission line
consisting of two electrical conductors in the
soils to be tested. Commercial high-fre-
quency transmission line conductors are
usually up to 1 cm apart, but it is intended
to design and use transmission lines whose
conductors are as far apart as possible — on
the order of meters. Before a practical TDR
system can be developed for detecting leaks
under waste impoundment liners, tests are
needed to determine the practical problems
of increasing the spacing between the con-
ductors.
The objective of this study was to deter-
mine the size of an anomaly (simulating a
leachate) that could be detected as the spac-
ing for the transmission line conductor was
varied.
Background
TDR has been applied to measurements
of geologic materials for fewer than 10 years,
and therefore much of the work to date has
been carried out to determine the practical
applications and limitations of the technique.
The main application of TDR for more than
30 years has been for testing transmission
line cables such as those used in radio and
telephone communications and in high-volt-
age transmission lines.
The electrical properties of geologic ma-
terials, especially soils, are very complicated
at first glance. But over the frequency range
we are interested in (10* to 109 Hz), the
dielectric constant is primarily sensitive to
changes in water content and has a relatively
weak sensitivity to changes in soil type and
density, temperature, and soluble salt con-
tent. TDR has shown excellent potential for
carrying out useful electrical property
measurements of geologic materials.
The TDR equipment is coupled to the
geologic medium by a transmission line. A
transverse electromagnetic (TEM) wave pro-
pagates along the transmission line in the
dielectric material between and around the
conductors, which act as guides for the
waves. The propagation velocity and at-
tenuation characteristics of the TDR signal
in the transmission line are dependent on the
electrical properties of the material in the
transmission line.
Methods and Equipment
Experiments were first conducted in the
laboratory to determine how the TDR signal
pulse rise time response varied as the con-
ductor spacing of the transmission line in-
creased. Next, experiments were carried out
to determine the physical size of the anomaly
in the transmission line that could be de-
tected by the TDR system as the conductor
spacing was varied. Finally, tests were con-
-------
ducted to determine the change of TDR
response as the position of the anomaly was
moved in the transmission line.
The equipment consisted of a TDR unit,
an X-Y plotter, a parallel transmission line,
an electrical connector to connect the TDR
unit to the transmission line, and boxes filled
with dry sand to simulate an anomaly. The
TDR instrument used for these experiments
was a Tektronics Model 1502 cable tester,
a portable field unit with pulser, receiver, and
display units in one package. The transmis-
sion line was a parallel wire transmission line.
The two conductors of the line were made
of 6-m-long aluminum tubes with a 5-cm
outside diameter. These tubes were sup-
ported on a wooden frame so that the spac-
ings between the tubes could be varied (10,
20, 40, 60, 80, 100, 120, 150, and 200 cm).
A special electrical connector was designed
and built to connect the TDR unit to the con-
ductors at the different spacings. The con-
nector consisted of a 50- to 200-ohm Balun
transformer and two sheets of metal to con-
nect the transformer to the aluminum tubes.
Figure 3 is a schematic diagram of the TDR
unit, the connector, and the parallel trans-
mission line.
The anomalies that might be found in the
field were simulated in the laboratory by
boxes (15-cm3), filled with dry sand and
placed side by side and on top of each other
when the size of the overall anomaly needed
to be increased. The sand had a dielectric
constant of 3.5. The boxes of sand were the
equivalent of a water-saturated soil in a soil
host material with a volumetric water con-
tent of about 20 percent. The electrical losses
in the air and sand were negligible to the
TDR pulse used. This arrangement provided
a satisfactory initial model of an actual field
setup under a waste impoundment liner.
Results and Conclusions
Findings of the project were as follows:
1. The TDR pulse signal rise time increases
with the spacing between the conductors
of a transmission line. The attenuation to
the higher frequencies in the pulse in-
creases with the frequency, and thus the
TDR system resolution decreases as the
spacing increases. In other words, the
size of the anomaly in the transmission
line must increase with the transmission
line spacing to produce the same TDR-
reflected signal amplitude.
2. The TDR signal amplitude does not vary
when the TDR signal travel time in the
anomaly is less than or equal to the total
travel time necessary for the signal to pro-
pagate through the anomaly. As the elec-
trical length of the anomaly decreases
below the pulse travel time in the anoma-
ly, the signal amplitude decreases.
3. The TDR signal amplitude varies slightly
with the height of the anomaly (i.e., the
dimension whose axis is perpendicular to
the plane of the conductors in the trans-
mission line varies as long as the height
of the anomaly is at least three times
greater than the diameter of the con-
ductors).
4. The amplitude of the TDR signal is
greatest when the anomaly touches both
conductors of the transmission line. The
signal amplitude decreases significantly
as the width of the anomaly decreases
relative to the spacing of the conductors.
The signal amplitude decreases less rap-
idly if the anomaly is in contact with either
of the conductors.
These findings lead to the conclusion that
a cube-shaped anomaly whose sides are
equal to at least half the spacing between
the conductors will have a TDR signal
amplitude of about one-fifth the maximum
signal amplitude that would occur if the
transmission line penetrated an anomaly that
was very large relative to the spacing of the
transmission line conductors. This gener-
alization will be affected by a number of fac-
tors not included in the studies. The electrical
loss properties of the materials likely to be
found in the field will reduce the TDR signal
amplitude. Thus the size of the anomaly will
have to be increased to increase the reflected
TDR signal amplitude. The electrical property
contrast of the leachate-saturated soil rela-
tive to that of the host soil under the waste
impoundment may not be as great as the
Parallel
Transmission
Line
Figure 3. Schematic diagram of the TDR system.
4
electrical property contrast used in the
model. Thus the reflected signal amplitude
may not be as large in the waste site en-
vironment.
The model used in these experiments only
tested the TDR response in transmission
lines whose maximum spacing between the
conductors was 2 m. Larger spacings be-
tween the conductors are practical using the
same TDR unit, but experiments to deter-
mine the maximum spacing of the conduc-
tors still need to be carried out.
The TDR technique appears to be a prac-
tical method for detecting and determining
the position and extent of a leachate under
a small waste impoundment liner. A cube-
shaped leak with dimensions on the order of
one-half the space between the conductors
of the transmission line is about the
minimum practical size that can be detected
using the TDR technique.
Recommendations
The following recommendations should be
followed before a TDR system is placed in
the field under a waste impoundment liner:
1. Carry out experiments using simulated
leaks over transmission lines at different
conductor spacings placed in a sandy soil
about 1 m thick.
2. Based on the results of the above ex-
periments, design a series of transmission
lines with different spacings between the
conductors and placed in a sandy soil
under a lined pond filled with water. Vary
the lengths of the transmission lines up
to about 30 m. Use controlled leaks in the
liner to test the ability of the TDR system
to detect leaks of different sizes and posi-
tions around the transmission lines.
3. Carry out a TDR system field design con-
figuration and cost analysis after the
above tests have been made.
4. Design TDR equipment specifically for
detecting leaks at actual waste impound-
ment sites. Possibly a higher-power pulse
of longer duration would make the TDR
system more economical for detecting
leaks at large waste impoundment sites.
5. Determine the best materials for transmis-
sion line conductors to minimize corro-
sion in a waste dump environment.
6. Determine the high-frequency (1 to 1000
MHz) electrical properties of some leach-
ate-saturated soils and of soils typically
found under waste sites.
The results of the recommended research
will help define the usefulness of the TDR
technique for detecting leaks and for deter-
mining the extent of leaks from a waste im-
poundment liner. Such studies will also aid
in the design of a practical TDR system for
monitoring leaks from waste dump sites.
-------
The full report was submitted in fulfillment
of Contract No. 68-03-3030 by EarthTech
Research Corporation under the sponsorship
of the U.S. Environmental Protection
Agency.
J. L. Davis, R. Singh. B. G. Stegman, andM. J. Waller are with EarthTech Research
Corporation, Baltimore. MD21227.
Carlton Wiles is the EPA Project Officer (see below).
The complete report, entitled "Innovative Concepts for Detecting and Locating
Leaks in Waste Impoundment Liner Systems: Acoustic Emission Monitoring
and Time Domain Reflectometry," (Order No. PB 84-161 819; Cost: $13.00.
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 Project Officer can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
U.S. GOVERNMENT PRINTING OFFICE: 1964-758-102/920
------- |