«.<*-
&EPA
United States
Environmental Protection
Agency _
Solid Waste and
Emeraency Response
(5305W) '
EPA530-R-97-024
NTIS: PB97-1 76 861
January. 1993
Indexing of Long-term
Effectiveness of Waste
Containment Systems
for a Regulatory Impact
Analysis; Draft
g |. iPA Headquarters Library
Malt code 3201
! NW !
Waihlngton DC 20460
Printed on paper that contains at test 20 percent postconsumer fiber
-------
UNITED STATES - OFFICE OF SOLID WASTE EPA/NUMBER
ENVIRONMENTAL PROTECTION WASHINGTON, DC 20460 - JANUARY, 1993
AGENCY
INDEXING OF LONG-TERM EFFECTIVENESS
OF WASTE CONTAINMENT SYSTEMS FOR A
REGULATORY IMPACT ANALYSIS
DRAFT COPY
A Technical Guidance Document
Office of Solid Waste
U.S. Environmental Protection Agency
Washington, DC 20460
-------
1
I
I
I
I
I
I
I
1
INDEXING OF LONG-TERM EFFECTIVENESS
OF WASTE CONTAINMENT SYSTEMS FOR A
REGULATORY IMPACT ANALYSIS
Authors
Hilary I. Inyang, Ph.D.
Senior Geoenvironmental Engineer
Guy Tomassoni
Engineering Geologist/Hydrogeologist
I
I
Corrective Action Programs Branch
Office of Solid Waste and Emergency Response
U.S. Environmental Protection Agency
Washington, DC 20460
November 1992
I
I
-------
I
I
I
I
DISCLAIMER
i
t
This document is intended to assist U.S. EPA personnel in the assessment of long-term
effectiveness of some containment systems for waste disposal. It is not advocated that the
content of this document be used as the sole basis for decision making. However, this document
sheds some light on the problems associated with the prediction of long-term effectiveness and
presents one approach to relevant assessments.
This guidance is not a regulation (i.e., it does not establish a standard or conduct that can
be enforced by law) and should not be used as such. The contents of this document do not
necessarily reflect the views and policies of the U.S. EPA. Also, mention of trade names,
commercial products, or publications do not constitute an unqualified endorsement of their use.
I
I
-------
I
'
1
1
1
t..
1
.
.
1
1
1
1 '
I
TABLE OF CONTENTS
Section . .
DISCLAIMER. ; ..;
TABLE OF CONTENTS
LIST OF FIGURES AND TABLES
1.0 PURPOSE : '. '.....
2.0 INTRODUCTION '.'.
3.0 APPROACH TO INDEXING..... ;
4.0 INDEXING SYSTEM
4.1 CAPS
4.L1 Clay Caps * ;
4.1.2 Synthetic Caps
4.1.3 Composite Clay and Synthetic Caps.....
4.2 LINERS . \
4.2.1 RCRA Subtitle C Composite Liner Systems
4.2.2 Clay Liners ;....;
4.2.3 Synthetic Liners :..'.
4.3 Si IBSIIRFAfF BARRIERS /
4.3.1 Slurry Walls
4.3.2 HDPE WALLS ;TT; ;
, 5.0 SUMMARY ;...-.
ii
Page
.... ii
iii
. 'l
3
4
7
7
8
....... 10
" ..14
... ' 15
15
22
22
;.... 23
; 23
25
26
27
-------
LIST OF FIGURES
'*'"
Figujg ' , . . - Page
1 An illustration of long range degradation tracks of waste containment systems 6
2 Summary of the one-dimensional freeze/thaw cycle effects on the Niagara
Clay (Ziramie et al. 1992) , r .9
3 The setup of the Electrical Leak Location Method (Laine and Miklas 1989) . 19
4 A comparison of liner thickness equations (U.S. EPA 1984c) . , 20
5 Simulations of cumulative diffusive solute transport through a hypothetical
soil-bentoriite barrier (Molt and Weber, Jr. 1991) 24
.6 Types of locks for HDPE walls (Manassero and Pasqualini 1992) 25
LIST OF TABLES
Table
1
2
4
5
6
Effectiveness of selected waste containment measures
Performance of barrier layer materials (adapted with modifications from
Koemer and Daniel 1992)
Overall benefit of each barrier configuration of cover/liner materials
determined by summing the horizontal rows in Table 2 (adapted with
modifications from Koerner and Daniel 1992)
Case histories of compacted clay liners (Daniel 1987)
Summary of experience with liner performance in the field (Bass et al. 1985)
Leak detection and location survey data for impoundments where the bottom
floor areas were surveyed (Laine and Miklas 1989) ~
Page
11
13
..14
16
17 ,'
21
iii
-------
1.0 PURPOSE .
This document is intended to support a Regulatory Impact Analysis (RIA) required for a
major Corrective Action regulation proposed by the U.S. Environmental Protection Agency.
The objective of the RIA is to evaluate the impact pf the proposed regulations on the
regulated community, and determine whether the regulations will be protective of human
health and the environment The effectiveness indexing scheme described in this paper was
developed to provide input data to a multi-media contaminant fate and transport model used
in the RIA to assess the pollution potential of ground water, surface water and soil. In view
pf the fact that in certain situations, the proposed regulations may allow on-site containment
of hazardous wastes (e.g., capping) as opposed to treatment/destruction (e.g., incineration),
an assessment of the long-term effectiveness of containment systems is necessary. In such
assessments, there exists the difficulty of predicting and verifying long-term effectiveness.
This problem exists because engineered waste containment systems have existed for a
relatively short time in comparison with other engineering structures. Numerical estimates of
design lives and effectiveness of containment systems have been made without the benefit of
field data on the past performance of similar systems over a reasonably long time interval.
Notwithstanding the paucity of long-term performance data on waste containment
systems, numerical assessments are necessary for regulatory purposes. Available predictive"
tools exemplified by models, mostly deal with initial conditions under the tacit assumption
that constructed systems will exist at the same degree of structural integrity over an
extended period of time. Also, such models often treat a few elements of the overall issue of
effectiveness. Effectiveness is herein defined as the ability of the waste containment system ,
-------
to prevent
and direction.
the mobilization and transport of contained waste outside,the system in any phase
A rating (indexing) scheme has been proposed for assessing the long-term
effectiveness of clay caps, synthetic caps, composite clay and synthetic caps, clay liners,
synthetic' liners, the Resource Conservation and Recovery Act (RCRA) Subtitle C liner
systems, and vertical barrier walls. The indexing system is based on a review of literature on
, _ \
initial and potential performance indices of the structures mentioned above.. Due to the
existence of gaps in available information, technical judgment has been introduced into the
selection process of specific rating numbers for long-term effectiveness. One of the most
widely used modelsthe Hydrologic Evaluation of Landfill Performance (HELP) model
allows the input of a "failure rate," which the user specifies on the basis of his/her judgment.
This allowance is an acknowledgment of the uncertainties associated with predicting the field
scale performance of facilities over time frames as long as one hundred years. The rating
scheme proposed is based partly on subjectivity. It should be recognized as a rating scheme
as opposed to a numerical model. Refinements may be necessary on a site-specific basis,
especially, when conditions are such that specific parameters of long-term effectiveness of
the entire containment system can be reasonably analyzed in greater, detail.
-------
1
2.0 INTRODUCTION
It is generally known that the structural integrity of waste containment systems usually
degrades as. time progresses. This situation stems from environmental conditions which
induce stresses on various components of the system. The root causes of these stresses
include thermal processes, biologic processes, geostatic and geodynamic loads, physico-
chemical interactions between waste constituents and containment system materials,
sunlight, and hydraulic processes. The rate at which the effectiveness of a containment
system is affected over a given time period depends on the conservatism of the initial design,
quality control during implementation activities, and the frequency of facility maintenance.
Ideally, it would be desirable to develop a "macro-model" that comprises sub-models,
each of which treats the decay of the effectiveness of specific components of containment
systems with time. Unfortunately, containment systems comprise many components in too
many configurations in numerous hydrological settings. The relatively short experience with
modern waste containment systems is such that there are too many unknowns. Each
component of a containment system is susceptible to a different degree, to the stresses
mentioned above. For example, polymeric materials of a geomembrane are more susceptible
to ultraviolet radiation than clay liner materials. Furthermore, the burrowing activities of
rodents may introduce defects into the system in the long term. The uncertainties associated
with the physical response of the entire containment system to the synergistic stresses from
various physical, chemical and biological processes plague the development of a useful
"macro-model" for precise prediction of the loss, of effectiveness of containment systems with
time. In essence, such models might not attain a level of accuracy that supersedes that of a
numerical indexing (or rating) scheme that is based on a non-quantitative analysis of
-------
/
available prediction schemes, test results and configuration of containment systems. The
latter approach is adopted in this work. Similar approaches .have also been adopted by
f ^
others, e.g., Koerner and DanieP(1992).
. * ' ' « v
3.0 APPROACH TO INDEXING , ' .
It 'is desirable that waste containment systems remain effective for a long time,
reasonably, beyond the closure period. The design life of a facility should be as long as
necessary to effectively restrict hazardous constituents from becoming a threat to human
health and the environment. The durability of various components of containment systems in
the laboratory have been investigated vigorously only, in the past fifteen years. Experimental
/
and modeling approaches to assessing long-term effectiveness have been adopted in such
studies. Although these investigations have shed light on the significant processes and
potential responses of individual components over extended periods, the overall long-term
effectiveness of composite systems still remains a gray area. It may not be feasible to
develop and verify a complete and precise numerical relationship between the effectiveness of
composite systems and time-since-constmction. Considering the diversity of possible
configurations, hydrogeological conditions, and waste types, coupled with .various
maintenance frequencies, such an approach would require an extensive factorial experiment.
\ .'*''
It should also be noted that some of the relevant factors cannot be fully controlled. For
example, the hydrogeology of a site is not normally controllable.
The approach adopted.herein is to use available information on modeling approaches,
laboratory experiments and field studies to develop a reasonable indexing system for
evaluating the long-term effectiveness of waste containment systems for RIA purposes. In
this approach, the entire system is considered rather than a single component However, the
entire system can comprise only one component in some cases. For multi-component
-------
I
I
I
«
systems, the degradation or failure of a single component does not necessarily mean that the
entire system is completely ineffective,
In this approach, the effectiveness of a containment system is assumed to vary with
i
time as illustrated in Figure 1. The figure also illustrates the .potential gain in effectiveness
that occurs with the implementation of an adequate maintenance scheme. A similar concept
has been applied in the development of serviceability index for highway pavement structures,
most of which comprise many components. When built initially, the containment system has
an effectiveness, Eto, which depends on the conservatism of the design and the adequacy of
the construction quality control. Over time, the effectiveness decreases along curve A. For
example, at time t2, the corresponding effectiveness is E(2. However, if maintenance
activities are implemented at time ti, the effectiveness improves to the level Eti and system
,
degradation follows curve B. The time horizon, t
-------
system. Nevertheless, such information has been reviewed and taken into consideration in
the selection of degradation rates proposed herein.
Figure!.
Tlnw Sine* Construction, t (y*ars)
An illustration of long range degradation tracks of
waste containment systems
-------
f«
I
4.0 INDEXING SYSTEM
The proposed effectiveness-time relationships are summarized in Table 1 for various
configurations of waste containment systems. The time periods selected are 0 years, 10
years, 30 years and 100 years. The technical issues discussed above have been considered.
Pertinent technical literature has also been reviewed. Existing models, experimental data
and verbal propositions have been considered: In most cases, judgment is used in the
analysis and adoption of information reviewed by the authors of this proposition. It should be
emphasized that although the degradation-time relationship is influenced by the initial design
of a waste containment system, design approaches are not the direct focus of this discussion:.
For more information on that aspect, the reader is referred to relevant documents such as
Richardson and Koemer (1989), and U.S. EPA (1984a. 1985a, 1989a and 1989b) for covers
and lining systems; Millet and Perez (1981), Barvenik et al (1985), Morgenstern and Amir-
Tahmasseb (1965), Mott and Weber, Jr. (1992), D'Appolonia (1980) and U.S. EPA (1984b)
for slurry walls; and May et al. (1985), Camberfort (1977), Weaver et al. (1992), Ran and
Daemen (1992), USAGE (1973 and 1984), Van Impe (1989), Hausmann (1990) and U.S.
EPA (1985b) for grout curtains.
(' '
4.1 CAPS
i ' "
Low permeability caps are used to cover waste materials to minimize contaminant
migration on the land surface, through the air, and into the ground water. Caps protect ground
i i
water by minimizing the infiltration .of precipitation into the waste which can mobilize
contaminants through leachate generation. There are a variety of cap designs and capping
materials available. Typically, a design will include a single or multiple layers of low-
permeability natural clay or made-made materials (geosynthetic membrane). Generally, all
-------
designs include a high permeability drainage layer above the low-permeability layer(s) to
promote precipitation runoff. All designs also include a vegetative layer or some other
material placed above the drainage layer to minimize erosion. The discussion of effectiveness
presented below focuses on the low permeability layers and refers to the ability of the cap to
restrict infiltration of water into the underlying waste.
4.1.1 Clay Caps
The factors which may reduce the effectiveness of clay caps include subsidence, slope
instabilities, desiccation cracking, burrowing activities of rodents, and vehicle loads. For the
clay cap, effectiveness is herein defined as its ability to prevent the intrusion of moisture into
the contained waste. For a given set of hydrological conditions, the quantity of leachate
generated is directly proportional to the effectiveness .of the clay cap. The latter depends on
both intergranular flow (pore flow) and flow through flaws which may result from any
combinations of the phenomena stated earlier. Due to the uncertainties associated with the
occurrence of degradation events with time, an exact predictive relationship is difficult to
develop and verify. Models exemplified by the Hydrologic Evaluation of Landfill Performance
(HELP) computer model predict pore flow only. In Figure 2, increases in the permeability of a
compacted clay soil layer due to several cycles of freezing and thawing are illustrated.
-------
1
io-5i
I
I
I
I
I
I
1
i
I
i
i
ID4'
10*
0 DftY
0 OPTIMUM
* WTT
0 5 10 15 '26 25 30 35
NUMBER OF FREEZE/THAW CYCLES
Figure 2. Summary of the one-dimensional freeze/thaw cycle effects on
the Niagara Clay (Zimmie et al. 1992)
Such changes iir permeability may be attributable to the generation of fissures, etc.
When macro-flaws develop in a clay cover over an extended time period, Darcy, flow
relationships become inadequate for predicting moisture percolation. Using the approach
* ' - ' '
proposed by Anderson et al. (.1991), the macrofissures increase the hydraulic conductivity of
the soil layer by the amount given in equation (1).
K = R4pg/8uE>2
(1)
K = hydraulic conductivity
R = radius of the channels
g = acceleration due to gravity
u = viscosity of ordinary water
D = spacing of the channels
p = density
I
I
-------
1
In Table 1, a clay, cap that meets minimum standards with respect to hydraulic
conductivity, thickness and construction quality control is assigned an initial effectiveness of
80%. It is assumed that significant flaws exist even in the initial post-construction period of
such cover materials. Consistent with the degradation-time pattern often observed in most
engineering structures, it is assumed that the effectiveness will decrease exponentially with
time. ' ,
'"
. It is estimated that over a service life of ten years immediately following their
implementation, the effectiveness of the clay cap will drop to 75%. The probability of
occurrence of phenomena that cause flaws increases as the duration of service increases. At
30 years, the effectiveness of the clay cap is estimated to be 60%. It is roughly estimated
that after 100 years of service, settlement, rodent activity, etc., would reduce the
effectiveness of single clay covers to 20%.
> i .
Note, however, if an additional.clay cap is added to the system at 100 years, the
effectiveness goes up to 85%. Replacing the cap at such a time when the original cap is no
longer effective is consistent with 'the understanding that engineered structures will not last
in perpetuity. Furthermore, adding a new cap also reflects the premise that, unlike a liner
system, a cap can be easily maintained and, if necessary, replaced, without actually removing
the waste. Adding a new cap also reflects the need to provide long-term reduction of
leachate generation. .
4.1.2 Synthetic Caps
Holes can be introduced into synthetic caps of waste containment systems during
installation. Geomembrane caps may also be degraded by subsidence, ultraviolet radiation,
\ . . ,
thermal stresses and the burrowing activities of animals. Temperature effects on the
10
-------
I
I
I
I
I
I
i
««
B
o
U
s
U-
'S
5
I
J3
53'
1
I
.a
«s
f
en
w S
i-
5
a
11
00
s
s
3
8
in
00
s
S in
^
8
*
'S'S'S'S'S
I1
M
11
-------
durability of polymeric materials-have been described using methods based on the Rate
Process Analysis (Koerner et al., 1990; and Kanninen, 1992). However, the effectiveness of
a synthetic cap material in the field is a function of temperature as well as radiation and other
unpredictable events. Nevertheless, Koerner and Daniel (1992) rate a single geosymhetic
cap above a single clay cap in terms of overall effectiveness. Relevant information is provided
in Tables 2 and 3. Considering the factors rated in Tables 2 and 3 and those discussed above,
a single synthetic cap is assigned an initial effectiveness of 90%. This rating takes into
account the possible existence of imperfect seams through which water can pass.
j.
Unavoidably, the volume of holes increases in synthetic layers over time. Increases in
the hydraulic conductivity of such layers should be expected. The increase in the discharge
(flow rate of such covers due to the presence of large holes) can be estimated using
. Bernoulli's equation as follows. '
Q = Ca(2gz)0-5
(2)
Q = discharge (m3/s)
C = dimensionless coefficient that pertains to the shape of the edges of the opening
(approximately 0.6 for sharp edges)
a = hole surface area (m2) ; .
g = acceleration due to gravity (9.8 m/s2)
z = depth below water level; may be approximated by the design head of infiltrating
water (m).
In the'indexing system proposed in Table 1, the effectiveness of a single synthetic cap is
assumed.to be 15% after 100 years. This assumed effectiveness is greater than the
assumption made" in U.S. EPA (1982) that after 100 years, a synthetic cap will be completely
ineffective. The 15% effectiveness is based on the assumption, that synthetic caps will
12
-------
ft
fa*
ft
^
.
^ ...
^-
ft,
fe_
ft
.
'
ft £
~ . ffv
*+
m^ "?
** 1
-5
m i
-38
!
i -
1 11
« ' H* .
I is
« -3
51
ft- . :s .
- ' -it S
rt El
J §
1|
r
i
. r*
fl
i
I
i
« =
'l
i
S
^
if
1
IB
u
Stllknwnl
v
1
W
U
S
<
|'s.
ji
A
:'e
^5
*5
3
1
i
fr£
> 2
£
'si
i1
J
]
t'
I
1
1
1
1
~ ,*
S
31
5
y
-
rt
e*
-
B
<
ir>
i^
-
m
z
o
-
<*»
J.
in
1
u
rt
»
^r
-
«
^
a
*
^»
*i
»'
!
to
^
m
M
^
_ . r'
i
u.
»
m
*r*
i«
1 OM/OCUOM
o
'
1
U
a
u
|- .
«" ^ *
3 «i v
<. '5 0 S
o a. a
II 'PS
» S § 5
? _ e S
.S g 2 g
1 8.11
S E o o
u 3 « o
.H V i i
1 IM
g II -H »
! !«.
£
g
£
' n
r*i
.» ^
1 -2
it | s
n -2 » " 6
01 -^SB 8
I IHi-
t l§«i
u £ 5 o S.
i S&&I
§ u u w u
8 Y»»j
' . i 55 3555 H
n u H ii
1 giSI
13
-------
Table 3. Overall benefit of each barrier configuration of coverAiner
materials determined by summing the horizontal rows in
Table 2 (adapted with modifications from Koerner and Daniel
1992)
Design
Alternate
Description
Overall
Benefit
Estimated Cost
(dollars/sq. ft.)
Benefit/Cost '
Ratio
Ranking in
Group
One Barrier Layer
A
B
C
CCL
GM
GCL
36
64
46
6.70
0.70'
0.70
51
91
66
3 .
1
'2 , ,
Two Barrier Layers
. D
E
GM7CCL
GM/GCL
58
66
1.40 .
. 1.40
41.
47
2
1
Three Barrier Layers
F
G
GM/CCL/GM
GM/GCL/GM
71
77
2.10
2.10
34
37
2
1
CCL = Single compacted clay liner.
GM = Single geomembrane.
GCL a Single goesynthetic clay liner.
GM/CCL B Two-component composite.
GM/GCL - Two-component composite.
GM/CCL/GM = Three-component composite liner.
GM/GCL/GM = Three-component composite liner.
degrade over time, but due to improvements in polymeric material technology, degradation
will not occur quite as fast as originally predicted by EPA. If a new synthetic cap is installed
at 100 years, the effectiveness of the synthetic cap is assumed to return to the original 90%
estimate. -
,» ' '
4.1.3 Composite Clay and Synthetic Caps ,
N.
The processes and assumptions described above for clay caps and synthetic caps apply
to the composite clay and synthetic cap; however, the composite cap yields a greater
effectiveness than the additive effectiveness of the two individual layers. The clay layer
underlying the geosynthetic cap acts as a low premeable barrier for leakage through holes in
,
I
I
I
I
I
I
I
I
I
I
i
I
I
I
I
I
14
-------
the geosynthetic layer. Furthermore, the effectiveness of the clay-layer improves in this
composite system because the synthetic affords some protection against the processes
(discussed in section 4.1.1) which degrade clay caps. As indicated in Table 1, the initial
effectiveness of the composite cap is assumed to be 95%. At 100 years, the effectiveness of
the composite system is reduced to 35%. At 100 years, it is assumed that an entirely new
composite cap will be installed resulting in an effectiveness of 98%. The increase, from 95% to
98% is assumed to be due to the residual effectiveness of the original clay layer.
. '*" - '
4.2 LINERS
Liners for waste containment systems are generally intended to prevent the migration of
hazardous constituents into the underlying subsurface. Liners are generally constructed of
low premeability soils (clay), geosynthetics, or a composite system consisting ,of both clay
and geosynthetic membranes.
4.2.1 RCRA Subtitle C Composite Liner Systems
The RCRA Subtitle C liner system consists of both clay and polymeric layers. The
minimum requirements for this liner system includes from top to bottom, a leachate collection
system, a primary (upper) geomembrane, a leachate detection system, and a secondary
composite liner consisting of an upper geomembrane directly overlying a clay layer. In
contrast to the situation with caps, leachate compatibility becomes more significant but
settlement (or subsidence) effects become less significant. In developing the effectiveness-
time estimates contained in Table 1, it is assumed that there is no cover over the buried
waste or contaminated materials. Since the RCRA Subtitle C liner is a multi-component
system, the failure of one component does not necessarily imply that the whole system has
become entirely ineffective. Some failures of single and multi-component lining systems have
been reported.in literature. Unfortunately, information is scanty or non-existent on the time
15
-------
Table 4. Case histories of compacted clay liners (Daniel 1987)
Location of
Site
Nature-of Liner
Actual Field k
(cm/sec)
Reference and Comments
Central
Texas
2-acre (0.8 ha) Liner for
Impoundment; l-ft-(30-cm)
Thick Liner Built of Local Clay
Soil
4 x 10'5
(originally)
Sx'10'6
(Reconstructed)
Daniel (1984). Original liner may have desiccated
somewhat. .Reconstructed liner not subjected to
desiccation. Poor CQA. Liner retained fresh
water. .
Northern
Texas
25-acre (10 ha) Liner for
Impoundment; 8-in-(20-cm)
Thick Liner Built from
Sand/Bentonite Mixture
3 x 10'6
Daniel (1984). Virtually no CQA. Liner retained
slightly saline water.
Southern
Texas
1-acre (0.4 ha) Liner for
Impoundment; Liner was 2-ft-
(30-cm) Thick and Built with
Local Clay Soil
1 x 10
r5
Daniel (1984). Liner retained brine solution.
Little CQA.
Northern
Mexico
Test Liner SO x 50 x 0.5 m and
Built with Local Clay Soil
Ix 10-
Auvinet and Espinosa (1981) and Daniel (1984).
Good CQA. Liner tested with fresh water.
Texas A&M
University
Prototype Liners; Each
Prototype Measured 1.5 x 1.5 x
0.15 m; Soils Consisted of
Kaolintte, Mica, and Bentonite
Blended with Sand
1 x :10'6
to
1 x 10:5
Brown. Green, and Thomas (1983). Good CQA.
Soil compacted with hand-operated equipment.
Liquids were xytene and acetone wastes.
University of
, Texas at
Austin
Two Prototype Liners; Each
Liner Measured 20 x 20 x 0.5 ft
(6 x 6 x 0.15 m) and Was.Built
of Local Clay Soil
4 x 10"6
and
9 x 1
-------
I
i
*
**
O)
i
C
i
£
«
1
I*
I!
*.
"if
1*
v
3
i 511-11
'52S55
i 55 1.5 i i i'1-i i i i '55.15'i
.ip^
22222 2555222 |22
,1
13
i e/31
o. 0.5
i i «
<
iS1
u
Q.
Si
CO U.
0,0.
C.OJ
52252.2225-55255-. 552555'2 55555
a u a a
u u a coeaflQ-mu
y"N . <*** ' *S
w y,
HI-
«
ffi f> PH CO <^ <**!
O O
Ri»».
' p
S'.
V3 00 60 00 Q 05
ggga
08Q'8
es
III! »llll Il
> in in
II
co
s
00
"
X
a
8
eo
II
K
§
'II
o
!_
II
o.
I
17
-------
at which the reported failures occurred. Daniel (1987) reported case histories of clay liner
performance in the field. Pertinent information is provided in Table 4. Bass et al. (1985)
provide the data shown in Table 5 on the failures and successes experienced with various
configurations of lining systems. Some of the configurations described in Table 5 are similar
to RCRA Subtitle C liners. Although the data presented in Table 5 have not been
incorporated in any direct numerical manner into the indexing scheme of Table 1, they do
illustrate the fact that some lining systems do fail unpredictably with time. Information
provided by Daniel (1984) indicates that the actual hydraulic conductivities of four clay liners
in Texas are generally ten to a thousand times higher than those measured on^ samples in the
laboratory. Bonaparte and Gross (1990) have presented data on the field performance of
double-liner systems within four to five years of their construction. Some of the measured
leakage rates are attributed to the consolidation of the clay layer component of the lining
systems investigated. For geomembrane layers of liner systems, Giroud and Bonaparte
(1989a and b) suggest the use of a hole frequency of 2 to 5 holes per Hectare and a hole size
of 3 x l(Hm2 to compute flow rates. .
Using the electrical leak detection technique illustrated in Figure 3, Laine and Miklas
(1989) surveyed 61 new or operating waste storage facilities in which geomembrane .liners
are used. An average of 3.2 leaks per 10,000 ft2 within a range of 0.3 to 5.0 leaks per 10,000
ft2 of liner .was detected. Relevant data for bottom floor areas of impoundments surveyed are
presented in Table 6: Of all leaks, 87% occurred through seams while the remainder occurred
i , ' * ,
through internal areas of the geomembranes. Observed flaws ranged from circular holes of up
to 1 inch in diameter to slits from 0.25 to 12 inches in length. Sometimes, flaws as large as 48
square inches Were observed. ,
18
-------
RIMOTI
CURRENT
RETURN
ELECTRODE
CURRENT SOURCE
EUCTRODE
MOVING
MEASUREMENT
ELECTRODES
CURRENT
FLOWUNES
Figure 3. The setup of the Electrical Leak Location Method (Laine and
Miklasl989)
Considering all the factors and data discussed above, the RCRA Subtitle C liner system
(multi-layered) is assigned an initial effectiveness of 98% as indicated in Table 1. In U.S.
EPA (1983), 1% of the area of the synthetic layer component of the lining system is assumed
to have the same vertical permeability as the, underlying soil. This corresponds roughly to
99% effectiveness. However, considering the uncertainties in the distribution of holes at
geomembrane seams, spatial variabilities in the hydraulic conductivity of the clay layer
' ^ \ ,
component of the system, an initial effectiveness of 98% is assigned in Table 1. By the end of
the tenth year of operation, significant leakage would occur through the upper geomembrane
barrier of the lining system. Assuming that some flaws would, have developed in the
secondary geomembrane and the. underlying clay layer in ten years, there would be the
possibility of seepage of fluids through the system. Considering this situation, the
effectiveness of the system after 10 years is rated at 95%. Within the first 30 years, leachate
that passes through degraded membranes and the drainage layer could also travel through
relatively intact portions of the clay layer component of the system by intergranular (pore)
19
-------
flow and/or diffusion. For a clay layer thickness of up to 35 cm (about 13.8 in.), analyses of
fluid transit time predictions by U.S. EPA (1984c) using various percolation equations
indicate that transit times would generally be under one year. These predictions are
/
presented in Figure 4. It should be noted that only the times of first arrival of retained
substances are predicted. Similar analyses by Shackelford (1992) for a 91 cm (35.83 in.)
thick clay layer indicate breakthrough times ranging from 2.6 to 9.8 years for various
«nd
10* £t«r Hilly 1982
- -- traniit ciM elatioM (M* cm)
j .
Q Ma4ifi«d tnatic timt quctioo
O CtMa-ta»t U»U)
10 19 20
LINCM THICKNESS d . «m
Figure 4. A comparison of liner thickness equations (U.S. EPA 1984c)
20
-------
Table 6. Leak detection and location survey data for impoundments where
bottom floor areas were surveyed (Laine and Miklas 1989)
the
Survey No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
, 38
39
40
41
42
43
44
45
46
47
TOTALS
S
Size
Sq. Feet
958
958
958
1. 000
1,798
2,625
3,000
3,000
3,200
4.951
4.951
4.951
5,175
7,007
12.600
18,346
26,016
.26,016
27.297
32,292
43,560
45,345
50,000
50,400
54,500
55.025
58.900
62.500
64.583
65,340
65.369
65.369
65,369
65.500
65.500
74.088
82,500
87,120
87,120
99,050
135.036
150.781
152,460
152,460
157.584
164.085
362,690
2,769,336
Total
Leaks
2
. 3
3
4
0
6
'21
4
0
0
17
2
2 .
4
7
50
7
4
8
25
2
4
6
193
29
12
8
21
29
56
6
. 7
5
7
5
20
18
8
17
18
17
64
' - 2
- 7
12
18
51
811
- Leaks Located In
Bottom Seam Sheet
2
3
3
4
0
6
21
4
, 0
0
17
2
2
4
7
50
7
4
8 '
25
2
4
6
193
29
12
8
21
29
56
6
7
5
7
5
20
18
8
17
18
17
64
2
7
12
18
51
811
2
3
3
3 '
0
6
21
4
0
0
17
' 2
1
4
7 .
35
7
4
6
25
2
4
6
188
18
12
6
19
, 21
55
6
5
3
'. 5
3
19
15
7
17.
14
16
46
2
7
10
16
. 37
709
0
0
0
1 . -
0
0
0
0
0
0
0
0
1
0
0
15
0
0
2
0
0
. 0
0
. 5
11
0
2
2
8
1
0
2
2
2 .
2
1
3
1 '
0
4
1
18
0 .
0
2
. 2
14
102
Leaks Per
10, 000
Sq. Feet
20.9
31.3
31.3
40.0
0.0
22.9
70.0
13.3
0.0
0.0
34.3
4.0
3.9
5.7
5.6
27.3
2.7
1.5 ,
2.9
7.7
0.5
0.9
1.2
38.3
5.3
2.2
1.4
3.4
4.5
8.6
0.9
1.1
0.8
1.1
0.8
2.7
2.2
0.9
2.0
1.8
1.3
4.2
0.1
0.5
0.8
1.1
1.4
2.9
21
-------
assumptions of concentration ratios and relative fluxes of solutes at the bottom end of the
clay liner. The latter analyses also indicate that maximum solute flux through the same clay
liner would be attained in 83 years.
It should be noted that the data and predictions obtained for the cases discussed above
do not necessarily apply to all cases. However, information on the range of breakthrough
times exemplified by the preceding discussions has been considered in developing the
effectiveness index for RCRA Subtitle C liners shown in Table 1: In the indexing scheme
presented in Table 1, the RCRA C liner system is rated at 85%. and 60% effectiveness at 30
and 100 years of operation, respectively.
4.2.2 Clay Liners
j The processes and findings described in section 4.2.1, pertaining to clay liners, result in
' ' ' ,
an assumed initial effectiveness for a single clay liner of 70% (see Table 1). After 100 years,
* the effectiveness of this liner is reduced to 5%.
v " f ,
-* ' '"'.'' ''-'''
-' 4.2.3 Synthetic Liners , . .
,j The processes and findings described in section 4.2.1, pertaining to synthetic liners,
result in an assumed initial effectiveness for a single synthetic liner of 85 percent (see Table
-*'__;'.
-: 1). Contrary to the single clay liner which retains some level of effectiveness at 100 years, it
is assumed that a single .synthetic liner will be completely ineffective (0%) at this time. This
assumption is based on the fact that once the number and size of holes in the synthetic liner
reaches a certain level, all infiltrating liquids will likely move along the liner to low points in
the system. If these low points conincide with holes, then the liner will provide no barrier to
leakage. . ' . -
22
-------
i
1
t
f
I
4.3 SUBSURFACE BARRIERS
The term subsurface barriers refers to a variety of low permeability cut-off walls or
\
diversions installed below ground to contain, capture, or redirect ground- water flow.
Subsurface barriers are commonly used at hazardous waste sites to constrain or restrict the
migration of contaminants from a designated area. To date, the most commonly used
subsurface barriers at hazardous waste sites are slurry walls. Barriers used in the RIA,
however, also include high density polyetheiene (HDPE) interlocking sheets. Although
relatively new in the. market place, these HDPE walls appear to be a viable and effective
alternative to slurry walls for certain applications; therefore, they have been included in the
proposed indexing system.
The effectiveness-time relationships shown in Table 1 for HDPE walls and slurry walls
are proposed assuming no intervening barriers like liners and covers. This implies that these
walls would be the first barrier layers against the migration of contaminants. The
effectiveness of a particular barrier wall depends on site conditions, mix design of wall
materials (for slurry walls), joint integrity (for HDPE walls), wall thickness, durability of wall
materials to chemical attack, and the quality of construction. The effectiveness indices
proposed for barrier walls-in Table 1 are very general. Owing to the novelty of this
containment technology, most of the available data are predictive in nature. Sufficient time ,
has not elapsed for their verification in the field. The examples illustrated herein are also
provided to indicate the numerical regime of the decay of effectiveness with time. Conditions
1 < " '' , " ,
vary widely; hence, the examples provided may not apply to some specific cases.
' ' " ' ' ' c
4.3.1 Slurry Walls ,
In Table 1, the initial effectiveness of slurry walls is rated at 70%. This rating roughly
reflects the difficulties associated with controlling the in-situ permeability of the slurry during
23
-------
'construction. Owing to the segregation of materials, zones of excessive permeability often
develop. Over a period of 10 years, the quantity of materials that diffuse across the barrier
would perhaps be significant. A pollutant detention time of 30 years would be sufficient to
allow the physico-chemical interactions between contaminants and slurry material to cause
increases in wall permeability. At 30 years, the assumed effectiveness of, slurry walls is
reduced to 20%. With the addition of a new wall, however, the effectiveness could be brought
back to the original 70% effectiveness. Figure 5 is an example of the projections of solute
transport across soil-bentonite barriers made by Mott and Weber, Jr. (1991). By the authors'
estimation, slurry walls may not be effective after a hundred years of service, (0% effective).
Pertinent failure mechanisms are discussed by U.S. EPA (1984d).
OJ
0.7
0.6
05-
D,» 0.635 xKT5, £ = 0.38
D,«O.S3SxlO-s. E = 0.«
A 0,« l.07xl
-------
4.3.2 HDPE Walls '
;
The effectiveness of HDPE barriers should depend largely on the tightness of the bonds
between adjacent sheets. Manassero and Pasqualini have furnished illustrations of various
bond configurations as shown in Figure 6. HDPE walls are perhaps the newest vertical
barrier technology. Hence, long-term performance data are not available on them. The
proposed numerical indices which are tabulated in Table 1 are for single vertical sheets
without complementing slurry walls. The initial rating is 65% (lower than that of slurry walls)
but the effectiveness decays at a much slower rate with increase in time. This pattern is
chosen to be consistent with the fact that HDPE walls are not susceptible to most of the
1 ' .
failure mechanisms that plague slurry walls. At 100 years, the effectiveness ,of HDPE walls
is assumed to be reduced to the extent (25%) that the installation of a new wall would likely
be warranted. This installation should increase the effectiveness back to the original 65%
level.
Figure 6. Types of locks for HDPE walls (Manassero and Pasqualini
1992)
25
-------
5.0 SUMMARY .
*° '. ,
In this document, some of the technical issues that pertain to the long-term
effectiveness of waste containment systems have been discussed. A conceptual model of
effectiveness decay with service life has been described and illustrated. Based on that
model, a numerical rating scheme which can be used in a Regulatory Impact Analysis (RIA)
i
as well as other projects has been developed. The reader is reminded that the scheme is a
rating system developed on the basis of a review of available literature and the technical
judgment of the authors. It is general in nature, and a particular facility can exhibit a
degradation pattern that reflects the curve configuration pattern of Figure 1 but defies the
rating scheme. In essence, the proposed scheme will be refined as more field data become.
available for use with models to fill the information gaps that currently exist
26
I
.-,
I
I
I
I
-------
6.0 REFERENCES
1. Anderson, D.C., Lupo, M.J.,.Rehage, J.A.; Sai, J.O., Shiver, R.L., Speake, R.C., Brown,
K.W., and Daniel, D. 1991. Factors controlling minimum soil liner thickness.
Project Summary EPA/600/S2-91/008. Risk Reduction Engineering Laboratory,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
2. Barvenik, M.J., Badge, W.E., and Goldberg, D.T. 1985. Quality Control of hydraulic
conductivity and bentonite content during soil/bentonite cutoff wall construction.
Proc. of the llth Annual Research Symp., Cincinnati, Ohio,'pp. 66-78.
3. Bass, J.M., Lyman, WJ. and Tratnyek, J.P. 1985. Assessment of synthetic membrane
successes and failures at waste storage and disposal sites. Project Summary
EPA/600-S2-85/100. U.S. Environmental Protection Agency, Cincinnati, Ohio.
4. Bonaparte, R. and Gross, B.A. 1990. Field behavior of double liner systems.
Proceed, of ASCE Symp. on Waste Containment Systems, San Francisco,
California, pp. 52-83. ;
5. Camberfort, H. 1977. The principles and.application of grouting. Quarterly Journal of
Engineering Geology, Vol. 10, pp. 57-95.
6. D'Appolonia, DJ. 1980. Soil-bentonite trench cutoffs. Journal of the Geotechnical
Engineering Division, ASCE, Vol. 106, No. GT4, pp. 399-417.
7. Daniel, D.E. 1984. Predicting hydraulic conductivity of clay liners. Journal of the
Geotechnical Engineering Division, ASCE, Vol. 110, No. 2, pp. 285-300.
8. Daniel, D.E. 1987; Earthen liners for land disposal facilities. Proceed, of ASCE Conf.
on Geotechnical Practice for Waste Disposal, Ann Arbor, Michigan, pp. 21-39.
9. Giroud, J.P. and Bonaparte, R., 1989a. Leakage through liners constructed with
geomembranes - Part I. Geomembrane liners. Geotextiles and Geomembranes,
Vol. 8, No. 1, pp; 27-67.
10. Giroud; J.P. and Bonaparte, R, 1989b. Leakage through liners constructed with
geomembranes - Part II. Composite liners. Geotextiles and Geomembranes, Vol.
8, No: 2,0^78-111.
11. HausmaiUV, M.R. * 1990. Engineering principles of ground modification. McGraw-Hill
_, publishing Company, New York. .
12. Kanninenr M.F. 1992. Assuring the durability of HDPE geomembranes. ASTM
Standardization News. pp. 44-49. -
13. Koerner, R.M., Halse, Y.H: and Lord, A.E. 1990. Long-term durability and aging of
geomembranes. Proceed, of ASCE Symp. on Waste Containment Systems, San
Francisco, California, pp. 106-134.
27
-------
14. Koerner, R.M. and Daniel, D.E. 1992. Better cover-ups. Civil Engineering, pp. 55-
'57. .' ' . ..
15, Laine, D.L. and Miklas Jr., M.P. 1989. Detection and location of leaks in
geomembrane liners using an electrical method: case histories. Proceed, of the 10th
National (Superfund) Conference, Washington D.C., pp. 35-40.
16. Manassero, M. and Pasqualini, E. 1992. Ground pollutant containment barriers.
Proceed, of the Mediterranean Conf. on Environmental Geotechnology, Cesme,
Turkey, pp. 195-204.
17. May, J.H., Larson, R.J., Malone, P.O. and Boa, J.A. 1985. Evaluation of chemical
grout injection techniques for hazardous waste containment. Proc. of the llth
Annual Research Symp., Cincinnati, Ohio, pp. 8-18.
18. Millet, R.A. and Perez, J. 1981. Current USA practice: slurry wall specifications.
Journal of Geotechriical Engineering Division, ASCE, Vol. 107, No. GT8, pp. 1041-
1056. ;
19. Morgenstern, N. and Amir-Tahmasseb, I. 1965. The stability of a slurry trench in
cohesibnless soils. Geotechnique, Vol. 15, pp. 387-395.
20. Mott, H.V. and Weber Jr., WJ. 1991. Diffusion of organic contaminants through soil-
bentonite cut-off barriers. Research Journ. WPCF, Vol. 63, No. 2, pp. 166-176.
2.1. . Mott, H.V., and Weber Jr., W.J. 1992., Sorption of low molecular weight organic
contaminants by fly ash: considerations for the enhancement of cutoff barrier
performance. Environmental Science and Technology, Vol. 26, No. 6, pp. 1234-
1242. . - .
22. Ran, C. and Daemen, J.J.K. 1992. Fracture grouting with bentonite slurries. Proc. of
ASCE Conference on Grouting, Soil .Improvement and Geosynthetics, Vol. 1, New
Orleans, Louisiana, pp. 360-371.
23. Richardson, G.N. and Koerner, R.M; 1989. Geosynthetic design guidance for
hazardous waste landfill cells and surface impoundments. Geosynthetic Research
Institute, Drcxel University, Philadelphia, Pennsylvania.
24. Shakelford, C.D. 1992. Performance-based design of soil liners. Proceed, of the
Mediterranean Conf. on Environmental Geotechnology, Cesme, Turkey, pp. 145-
-153;. . - -. : _ ,-'/.- ...,. .. ..
-25. USAGE.' 1973; Chemical grouting. Engineer Manual EM 1110-2-3504. U.S. Army ,
Corps of Engineers, Department of the Army; Washington D.C.
26. . USAGE: 1984; Grouting technology. Engineer Manual EM 1110-2-3506 U.S. Array
Corps of Engineers, Department of the Army, Washington D.C. ;
27. U.S. EPA. 1982. Hydrologic simulation on solid waste disposal sites. SW-868.'
Office of Solid Waste and Emergency Response; U.S.. Environmental Protection
Agency, Washington D.C. .
28. U.S. EPA. 1983. Synthetic cap and liner systems. Draft Technical Report Office of
... Solid Waste, U:S. Environmental Protection Agency, Washington D.C.
-------
1
I
1
1
i
1
I:
1
*
t
I
4
ft
29. U.S. EPA. 1984a. Procedures for modeling flow through clay liners to determine
required liner thickness. Draft Technical Resource Document EP A/530-SW-84-
001. Office of Solid Waste, U.S. Environmental Protection Agency, Washington
D.C
30. U.S. EPA. 1984b. Slurry trench construction for pollution migration control. EPA-
540/2-84-001. Office of Emergency and Remedial Response/ U.S. Environmental
Protection Agency, Washington D.C.
31. U.S. EPA. 1984c. Procedures for modeling flow through clay liners to determine
required liner thickness. Technical Resource Document EPA/5 30-SW-84-011,
Office of Solid Waste, U.S. Environmental Protection Agency, Washington D.C.
32. U.S. EPA. 1984d. Slurry trench construction for pollution migration control. EPA-
540/2-84-001. Office of Emergency, and Remedial Response, U.S. Environmental
Protection Agency, Washington D.C.
33. U.S. EPA. 1985a. Covers for uncontrolled hazardous waste sites. EPA/540/2-
85/002. Hazardous Waste Engineering Research Laboratory, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
34. . U.S. EPA.. 1985b. Leachate plume management EPA/540/2-85/004. Office of Solid
Waste and Emergency Response, U.S. Environmental Protection Agency,
Washington D.C.
35. U.S. EPA. 1989a. Final covers on hazardous waste landfills arid surface
impoundments. Technical Guidance Document EPA/530-SW-89-047. Office of
Solid Waste and Emergency Response, U.S. Environmental Protection Agency,
Washington D.C.
36. U.S. EPA. 1989b. Requirements for hazardous waste landfill design, construction,
and closure. Seminar Publication EPA/625/4-89/022. Center for Environmental
Research Information, U.S.-Environmental Protection Agency, Cincinnati, Ohio,
37.' Van Impe, W.F. 1989; Soil improvement techniques and their evolution.. A.A.
Balkema Publishers, Rotterdam, The Netherlands.
38; Weaver, K.D., Coad, R.M., and Mclntosh, K.R. 1992. Grouting for hazardous waste
site remediation at Necco Park, Niagara Falls; New York. Proc. of ASCE
Conference ow Grouting, Soil Improvement and Geosynthetics, Vol. 2, New
Orleans, Louisiana, pp. 1332-1343.
39. Zimnuef*T.F.; LaPlante, CM: and Branson, D. 199,2. The effects of freezing and
. thawing on the permeability of compacted clay landfill covers and liners. Proceed, of
th&vMediterranean Conf. on Environmental Geotechnology, Cesme, Turkey, pp.
29
------- |