United States
Environmental
Protection Agency
Office of Solid Waste and
Emergency Response
(5102G)
EPA 542-R-98-005
August 1998
www.clu-in.com
v>EPA
Evaluation of Subsurface
Engineered Barriers
at Waste
:
«-,tv;*'^~M«i»'O' ,""'-' ^i*'-'^*1""'^ ',""*"*'-.'-""*»'' "'' ."i''".*»'-'j>-'^<''. /,'«.>-' A"> '*' ~*f-~'^~*Z!,'Xjs*f,f**js* f,f**t ^j^V^''-'^*1' ;''-<^f '.' ~±r ' i ."^'*.1'«-'^'J'-'/«"*^^
I
s,
_
-------
L-542-R-98-005
July 1998
vwvw.epa.gov
clu-in.coni
EVALUATION OF SUBSURFACE ENGINEERED
BARRIERS AT WASTE SITES
-------
NOTICE
This document was prepared for the U.S. Environmental Protection Agency's (EPA) Office of
Solid Waste and Emergency Response (OSWER) under Contracts No. 68-W5-0055 and No. 68-
W4-0007. The work assignments preparing this document were led by the EPA Office of
Research and Development's (ORD) National Center for Environmental Assessment (NCEA), in
cooperation with the EPA Office of Emergency and Remedial Response (OERR).
Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise does not constitute or imply endorsement,
recommendation, or favoring by EPA.
-------
FOREWORD
Subsurface engineered barriers have been used to isolate hazardous wastes from contact,
precipitation, surface water, and groundwater. The purpose of this report is to provide the U.S.
Environmental Protection Agency's (EPA) waste programs with a national retrospective analysis
of barrier field performance, as well as information that may be useful in developing guidance
on the use and evaluation of barrier systems. The report focuses on vertical barriers; evaluation
of caps was a secondary objective.
The overall approach to the report was to assemble existing performance monitoring results from
a number of sites, and examine those results in light of remedial performance objectives and
factors that may influence performance, specifically, design, construction quality
assurance/construction quality control (CQA/CQC), types of monitoring programs, and
operation and maintenance (O&M) activities.
This performance evaluation report describes the performance of subsurface engineered barriers
at each of 36 sites. The report discusses the performance evaluation process, including the site
identification and selection process, and the availability of information upon which to base
judgment of the performance of existing barriers and presents findings and conclusions,
including observed similarities or trends among sites.
This document is available on the Internet at www.clu-in.com. A limited number or Appendices
(document No. EPA-542-R-98-005a) are available from the EPA National Center for
Publications and Information (NCEPI) at 1-800-490-9198.
-------
CONTENTS
Section Page
List of Abbreviations v
Executive Summary vii
1.0 INTRODUCTION 1
1.1 INTRODUCTION TO ENGINEERED BARRIERS 1
1.2 OBJECTIVE OF THE STUDY 8
1.3 SCOPE OF STUDY 8
1.4 LIMITATION OF THE STUDY 9
2.0 SITE SELECTION 9
2.1 SITE IDENTIFICATION 10
2.2 SITE SELECTION CRITERIA 11
2.3 SELECTION PROCESS 13
2.4 LIMITATIONS OF THE SELECTION PROCESS 14
3.0 DATA COLLECTION AND ANALYSIS - VERTICAL BARRIERS 15
3.1 GENERAL INFORMATION ABOUT THE SITES 15
3.2 DESIGN 18
3.3 CONSTRUCTION QUALITY ASSURANCE AND CONSTRUCTION
QUALITY CONTROL DATA 33
3.4 PERFORMANCE MONITORING 49
3.5 OPERATION AND MAINTENANCE 60
3.6 COST 64
4.0 DATA COLLECTION AND ANALYSIS-CAPS 64
4.1 DESIGN 64
4.2 CONSTRUCTION QUALITY ASSURANCE/CONSTRUCTION QUALITY
CONTROL 71
4.3 PERFORMANCE MONITORING 77
4.4 OPERATION AND MAINTENANCE 79
4.5 COSTS 79
5.0 PERFORMANCE EVALUATION 80
5.1 PERFORMANCE BASIS 80
5.2 PERFORMANCE STAGES 81
5.3 RANGE OF FINDINGS 83
5.4 FACTORS AFFECTING SYSTEM PERFORMANCE 88
6.0 CONCLUSIONS AND RECOMMENDATIONS 89
6.1 CONCLUSIONS 89
6.2 RECOMMENDATIONS 94
REFERENCES 97
GLOSSARY 99
111
-------
CONTENTS (Continued)
TABLES
Table Page
3-1 CONTAINMENT SYSTEMS SUMMARY 16
3-2 MATRIX FOR EVALUATING BARRIER DESIGN AGAINST
ACCEPTABLE INDUSTRY PRACTICES 20
3-3 SUMMARY OF BARRIER DESIGNS 28
3-4 MATRIX FOR EVALUATING BARRIER CQA/CQC AGAINST ACCEPTABLE
INDSUTRY PRACTICES 35
3-5 CONTAINMENT BARRIER CQA/CQC MATRIX 45
3-6 EVALUATION OF CONTAINMENT BARRIER MONITORING CATEGORIES 51
3-7 CONTAINMENT BARRIER MONITORING MATRIX 57
3-8 CONTAINMENT BARRIER AND CAP O&M AND COST MATRIX 61
4-1 CONTAINMENT CAP DESIGN STANDARDS 66
4-2 MATRIX FOR EVALUATING CAP AGAINST ACCEPTABLE INDSUTRY
PRACTICES 72
5-1 CONTAINMENT CATEGORIES AND PERFORMANCE MONITORING 80
5-2 VARIATION IN ACTIVE CONTAINMENT HEAD DIFFERENTIAL 81
5-3 PERFORMANCE EVALUATION MATRIX 84
5-4 PERFORMANCE RATING VERSUS CRITERIA RATING 87
6-1 NUMBER OF SITES BY RATING CRITERIA 90
FIGURES
1-1 SOIL-BENTONITE SLURRY TRENCH COST-SECTION 4
1-2 SHEET PILE BARRIER 7
5-1A TYPICAL ACTIVE CONTAINMENT GROUNDWATER QUALITY AND
HYDRAULIC HEAD RESPONSES 82
5-1B TYPICAL PASSIVE CONTAINMENT GROUNDWATER QUALITY AND
HYDRAULIC HEAD RESPONSES 82
APPENDICES
Appendix
A SUMMARY OF EXISTING SUBSURFACE ENGINEERED BARRIER AND CAP TYPES AND
TYPICAL CONSTRUCTION TECHNIQUES
B SITE SUMMARIES (Volume II)
C FIELD PROTOCOL
IV
-------
LIST OF ABBREVIATIONS
ASTM American Society for Testing and Materials
AWQC Ambient Water Quality Criteria
bgs Below Ground Surface
BOD Biochemical Oxygen Demand
BRA Baseline Risk Assessment
BTEX Benzene. Toluene, Ethylbenzene. and Xylene
CB Cement-Bentonite
CCL Compacted Clay Liner
cm Centimeter
COD Chemical Oxygen Demand
CQA Construction Quality Assurance
CQC Construction Quality Control
DBCP Dibromochloropropane
DCA Dichloroethane
DCE Dichloroethene
DCPD Dicyclopentadiene
DIMP Diisopropylmethyl Phosphonate
DNAPL Dense Nonaqueous-Phase Liquid
EPA U.S. Environmental Protection Agency
FMGP Former Manufactured Gas Plant
FML Flexible Membrane Liner
FS Feasibility Study
ft Foot
GCL Geosynthetic Clay Liner
GM Geomembrane
gpd Gallon Per Day
gpm Gallon Per Minute
HOPE High-Density Polyethylene
hr Hour
IPA Isopropyl Alcohol
ISSM in situ Soil Mixing
L Liter
Ib Pound
LDPE Low Density Polyethylene
LNAPL Light Nonaqueous-Phase Liquid
MCL Maximum Contaminant Level
ing Milligram
msl Mean Sea Level
NAPS Nonaqueous-Phase Substance
NOAA National Oceanic and Atmospheric Administration
NPDES National Pollutant Discharge Elimination System
NPL National Priorities List
NSF National Science Foundation
-------
LIST OF ABBREVIATIONS (Continued)
O&M Operation and Maintenance
OU Operable Unit
PAH Polynuclear Aromatic Hydrocarbon
PC Plastic Concrete
PCA Primary" Containment Area
PCB Polychlorinated Biphenyl
PCE Tetrachloroethene
PCP Pentachlorophenyl
POP Project Operations Plan
ppb Part Per Billion
ppm Part Per Million
PRP Potentially Responsible Party
psi Pound Per Square Inch
PVC Polyvinyl Chloride
QA Quality Assurance
QC Quality Control
RCRA Resource Conservation and Recovery Act
RI Remedial Investigation
ROD Record of Decision
RPM Remedial Project Manager
SB Soil-Bentonite
SCB Soil-Cement-Bentonite
sec Second
SVE Soil Vapor Extraction
SVOC Semivolatile Organic Compound
SWMU Solid Waste Management Unit
TAL Target Analyte List
TCA Trichloroethane
TCE Trichloroethene
TCL Target Compound List
TCLP Toxicity Characteristic Leaching Procedure
THF Tetrahydrofuran
TPHC Total Petroleum Hydrocarbon
TVOC Total Volatile Organic Compound
VLDPE Very Low Density Polyethylene
VOC Volatile Organic Compound
yd Yard
ing Microgram
VI
-------
EXECUTIVE SUMMARY
Subsurface engineered barriers have been used to isolate hazardous wastes from contact,
precipitation, surface water, and groundwater. The objective of this study was to determine the
performance of such barriers installed throughout the United States over the past 20 years to
remediate hazardous waste sites and facilities. The study focused on vertical barriers; evaluation
of caps was a secondary objective. This study provides the U.S. Environmental Protection
Agency's (EPA) waste programs with a national retrospective analysis of barrier field
performance, and information that may be useful in developing guidance on the use and
evaluation of barrier systems.
The overall approach to the study was to assemble existing performance monitoring results from
a number of sites, and examine those results in light of remedial performance objectives and
factors that may influence performance, that is, design, construction quality
assurance/construction quality control (CQA/CQC), types of monitoring programs, and
operation and maintenance (O&M) efforts.
A national search was launched to locate hazardous waste sites (i.e., Superfund sites, Resource
Conservation and Recovery Act [RCRA] facilities, and other hazardous waste management
units) at which vertical barrier walls had been used as the containment method during a remedial
or corrective action. An initial list of 130 sites was developed. A subset of sites was then
selected on the basis of availability of monitoring data to enable a detailed analysis of actual
field performance. Where caps were present at these sites, they were included in the study as
well. Two available nonhazardous waste sites and one cap-only site with extensive data were
also included to further inform the study. A total of 36 sites were analyzed in detail. It should
be noted that because sites were chosen on the basis of sufficient performance-related
information being available to enable detailed analysis, these sites were likely to represent the
better-managed sites nationally and do not necessarily represent all sites.
For the 36 sites selected, data on design, CQA/CQC, monitoring systems, O&M, and
performance results were obtained by contacting regulatory agencies, contractors, and owners of
sites. Cost data were also noted where available. In some cases, owners required anonymity
before releasing data to be used in the study.
Benchmarks for acceptable industry practice were then developed to enable evaluation of design,
CQA/CQC, and monitoring systems. Designation of acceptable industry practices was based on
a literature review, reinforced by discussions with barrier construction contractors, designers of
barriers, university researchers and the best professional judgment of the project team. Each site
was evaluated against acceptable practices for design, CQA/CQC, and monitoring programs.
These factors were then analyzed in light of remedial goals and performance monitoring results
for each site.
Performance objectives varied among the sites, from maintenance of a specific hydraulic head
differential to achievement of a specific groundwater quality standard downgradient. Thus, the
performance of the barriers cannot be compared to an absolute standard. The evidence showed
that of the 36 sites, 8 had met and 17 may have met the performance objectives established by
the owner or regulatory agency for that system. (Of the 17 sites at which performance objectives
may have been met, 4 sites met the remedial objective, but long-term performance data was
unavailable.) Seven may not have met performance objectives, and 6 had insufficient evidence
Vll
-------
to determine if objectives had been met. Of those that had met objectives, acceptable or better
elements of design and CQA/CQC were generally utilized. Of those that had not met objectives,
elements of design, CQA/CQC, and monitoring were less consistently acceptable, with
insufficient monitoring programs being a common problem. Barrier failures were primarily due
to underflow from the key-in horizon, and did not always correlate with insufficient design or
CQA/CQC.
Major differences were found in the monitoring of the containment systems. At some sites, very
little monitoring of groundwater quality and levels was carried out, while at others, monitoring
well networks downgradient of the site were used to measure trends in groundwater quality and
paired piezometers at a given spacing were used within 50 feet of the barrier to monitor
groundwater levels. Essentially no long-term monitoring of physical samples was performed to
examine mechanisms of degradation affecting the barrier. Geophysical surveys along the wall
alignment were used at several sites, but were inconclusive because the available techniques
cannot detect small changes in the permeability of the wall. Stress testing of the wall after
construction was performed infrequently. However, monitoring data allowed the detection of
leaks at four sites, and the leaks were repaired.
Of the 36 sites, 22 had caps in addition to the barrier wall. In many cases, the caps were tied
into the barrier wall. One site had only a cap. Cap design varied little among the sites, and most
sites met the design requirements set forth under RCRA Subtitle C. Monitoring data for caps
generally were not detailed enough to evaluate performance.
Recommendations in this report to improve the performance and evaluation of subsurface
engineered barriers include:
The design of subsurface barriers and caps should be based on more complete
hydrogeological and geotechnical investigations than are usually conducted. In addition,
designs should be more prescriptive (as appropriate) in terms of contaminant diffusion and
compatibility that could affect long-term performance.
The CQA/CQC effort for subsurface barriers requires further development and
standardization, including nondestructive post-construction sampling and testing.
The importance of a systematic monitoring program in evaluating long-term performance of
subsurface barriers cannot be overemphasized.
Measures should be implemented to ensure the integrity of the barrier throughout its life
including comparative data reviews at 5-year intervals. Such reviews should address 1)
hydraulic head data (specifically, the development and maintenance of a gradient inward to
the containment), 2) trends in downgradient groundwater quality, and 3) data from
monitoring points at the key horizon.
A sampling protocol for use in performance evaluation of vertical barriers is provided as an
appendix to this report. The protocol recommends evaluation of the performance of vertical
barriers using proven and innovative monitoring techniques.
Vlll
-------
1.0 INTRODUCTION
Under Contracts No. 68-W5-0055 and 68-W4-0007, support was provided to the U.S.
Environmental Protection Agency's (EPA) Office of Solid Waste and Emergency Response
(OSWER) to conduct nationwide field performance evaluations of existing subsurface engineered
barriers.
The purpose of this report is to provide EPA?s waste programs with a national retrospective
analysis of barrier field performance, and information that may be useful in developing guidance
on the use and evaluation of barrier systems. The overall approach was to compile existing
design, construction quality assurance/construction quality control (CQA/CQC), operation and
maintenance (O&M), and performance monitoring data from a number of sites, and evaluate the
data using benchmarks for acceptable industry practice relating to design, CQA/CQC, and
monitoring systems. Based on this evaluation, the site's performance was determined with
respect to the remedial performance objectives.
This performance evaluation report describes the performance of subsurface engineered barriers
at each of 36 sites. It discusses design, installation, operation and maintenance, and overall
performance. The report describes the performance evaluation process, including the site
identification and selection process, and the availability of information upon which to base
judgment of the performance of existing barriers and presents findings and conclusions,
including observed similarities or trends among sites.
The report contains two volumes. Volume I presents the objective and scope of the evaluation
report; describes the site selection process, collection and analysis of data, and evaluation of
performance of subsurface engineered barriers and caps; and sets forth conclusions and
recommendations. Volume II contains appendices that present a summary of engineered barriers,
cap types, and construction techniques; an individual summary of each of the 36 sites included in
the study; and a field sampling protocol, including sites recommended for future field
investigations.
1.1 INTRODUCTION TO ENGINEERED BARRIERS
Engineered barriers are constructed containment systems that control one of the following:
Horizontal migration of groundwater. Such barriers are referred to herein as vertical
barriers. Vertical barriers typically used to control sources of hazardous waste are
soil-bentonite, soil-cement-bentonite, cement-bentonite, sheet pile (steel or high-
density polyethylene [HDPE]), and clay barriers. Soil-bentonite barriers are the most
widely used in the United States.
Downward migration or seepage of surface runoff and rain. Such barriers are
referred to herein as caps. The caps used satisfy the design requirement set forth
under Subtitle C of the Resource Conservation and Recovery Act (RCRA) and are
built of clay or geosynthetic material.
-------
1.1.1 Historical Development of Engineered Barriers
Historically, vertical barriers have been used on construction projects to prevent inflow of
ground-water into deep excavations, as well as to support excavation. Sheet pile walls (first of
wood and later of steel) have been installed throughout the world for many decades. The 1950s
saw the development of slurry trenching technology, in which bentonite was used to support the
sides of trenches under excavation before they were backfilled. That development took place
independently in Europe and in the United States.
A market existed in Europe for the construction of deep excavations in urban areas adjacent to
existing buildings, even historical structures. That demand created a need to develop
technologies for rigid support systems and for limiting the drawdown of the water table outside
the excavation to minimize subsidence. Secant pile walls first were used after World War II;
later, in the 1950s, concrete slurry wall technology was developed. That development was a
natural evolution of the secant wall technology, with the goal of decreasing the number of joints
between piles, thereby minimizing the risk of blowouts in the mass excavation through faulty
joints. By the end of the 1960s, cement-bentonite cutoff wall technology also had been
developed in Europe to allow deep excavation below the groundwater table for power plants and
locks, or to act as a cutoff through pervious overburden soils on dam projects. In Europe to date,
the use of cement-bentonite (quite often in conjunction with a geomembrane) remains the
preferred technique for seepage control, with applications including hazardous waste sites.
The development of slurry trenching technology in the United States, occurring independently
from its development in Europe, took place in the late 1940s and early 1950s and was based on
the use of the soil-bentonite technique (still unused in Europe). The main goal was to prevent the
flow of water into deep excavations for lock and dam projects, or to minimize seepage beneath
and through dams and dikes. The first industrial application of the soil-bentonite technique took
place in 1950 at the Terminal Island project in California. Slurry trenches then were used
extensively in the 1960s and 1970s for dam projects as permanent cutoff walls and for the
construction of the Tombigbee Waterway.
More recently, by the late 1970s and early 1980s, vertical engineered barriers have been used in
the United States to isolate hazardous wastes from groundwater, as slurry walls, primarily
soil-bentonite cutoffs, began to be used to contain hazardous wastes. Initially, the goal was to
contain contaminated groundwater for a "limited" period of time. A 30-year life span for the
containment was often the objective. By the late 1980s, the concept of establishing a reverse
gradient appeared. In such applications, an extraction or pumping system is installed in the
contaminated zone, in addition to the peripheral cutoff wall. This approach allows maintenance
of an inward flow through the wall at a very low rate. This approach has its advantages, since it
decreases, if not eliminates, the risk arising from deficiency in design or installation or even
localized anomalies in the aquitard layer.
In recent years, new concepts and developments in subsurface engineered systems have been
introduced. Among them are:
The funnel and gate, or permeable reactive wall: A contaminant plume is channeled
between impervious vertical walls, referred to as the funnel, and flows naturally through
a permeable reactive barrier gate, where the pollutants are treated in situ during the flow
process.
-------
The use of slurry trenching technology to install a deep groundwater extraction trench,
instead of an impervious cutoff wall: The slurry used to support the trench is made from
a biodegradable material (instead of bentonite, which would reduce flow to the trench).
After excavation, the trench is backfilled with a pervious material, and the slurry filling
the voids of the pervious material biodegrades. Drains installed by this biopolymer
method typically are from 20 to 50 feet deep, and sometimes deeper.
Quite recently, engineers began to be concerned not only about the hydraulic transport of
contaminants, but also about the diffusion of contaminants through vertical barriers, a chemical
process. This issue is crucial for the long term (usually considered to be well in excess of
30 years), in terms of the integrity of vertical barriers. New technologies are emerging to
increase the sorption capacity of vertical barriers, primarily through the use of additives in the
backfill materials.
In addition, improvements in barrier construction technology allow the installation of vertical
barriers to depths as much as 400 feet, through various soil and rock conditions, and in hostile
environments (such as brackish water and water contaminated with chemicals).
Caps have been used to prevent the downward flow of surface runoff and precipitation inside
contaminated sites. The concept is similar to the use of impervious blankets on the upstream
slope of a dam. At first, caps included clay blankets. The introduction of chemically resistant
geosynthetic materials that have minimal diffusive conductivity has significantly improved the
quality and the ease of installation of caps. Caps have been used at sites as large as 400 acres.
1.1.2 Types of Engineered Barriers
Engineered barriers, as discussed in this report, are vertical barriers and caps. Appendix A
provides details of the design, construction, and construction quality assurance (CQA) and
construction quality control (CQC) for vertical barriers and caps. Significant features of vertical
barriers and caps are discussed below.
Note: This study does not include engineered bottom barriers, a recent development in which an
impervious horizontal stratum is created below a hazardous waste site, when no aquitard exists,
by grouting or other techniques now in the developmental phase.
Vertical Barriers
Vertical barriers control the subsurface flow of water into or out of a hazardous waste site. They
are classified into various categories. The most common ones are briefly discussed below:
Barriers Installed with the Slurry Trenching Technology: Such barriers consist of a vertical
trench excavated along the perimeter of the site, filled with bentonite slurry to support the trench
and subsequently backfilled with a mixture of low-permeability material (1 x 10"6 cm/sec or
lower) (see Figure 1-1). Such walls are keyed into an aquitard, a low-permeability soil or rock
formation, or a few feet below the groundwater elevation when the objective is to contain light
nonaqueous phase liquids (LNAPL). Significant features of a vertical barrier are, at a minimum:
-------
5
03 O
-------
Continuous wall of uniform low permeability
Sufficient thickness to withstand earth stresses and hydraulic gradients and to
provide long-term sorption capacity
Wall backfill compatible with the groundwater quality and chemistry in the vicinity
of the wall
Continuous key. typically of 2 to 5 feet into the low-permeability soil or rock
formation, if feasible (the quality of the key material can be verified continuously
during excavation of the trench)
As stated previously, the most widely used technique for containment is the soil-bentonite slurry
wall. It is typically the most economical, utilizes low-permeability backfill, and usually allows
reuse of all or most of the material excavated during trenching. The low-permeability backfill is
prepared by mixing a bentonite slurry with the excavated native soils. Sometimes, additional
borrow materials or dry bentonite is added to the mixture to meet the design requirements.
Recently, specialty additives have been used to increase the sorption capacity of the backfill.
In the United States, cement has been added to the soil-bentonite backfill on a few projects to
impart strength to the wall, when that attribute was required by the site conditions. Such walls
are called soil-cement-bentonite walls. The technique also has been used on a few conventional
civil engineering projects at which loading conditions require a stronger barrier. Unfortunately,
the addition of the most common types of cement, such as Portland, increases the permeability of
the backfill.
Cement-bentonite and even concrete slurry walls also are used for containment when they are
required by the site conditions. The techniques reduce the length of excavations held open under
slurry at any given time and provide a backfill that exhibits strength. Typical applications would
be trench excavation adjacent to an existing structure or through soft or unstable soil.
Thin Walls: These walls usually are installed by vibrating a beam to the aquitard that injects a
backfill mixture, usually cement grout, when the beam is extracted. The design concept provides
for overlap between beam imprints. Recent technological developments have allowed improved
control of the overlap between joints and increased driving depth and thickness by the use of
jetting. The effective width of such barriers is in the range of 8 to 10 inches.
Deep Soil Mixing: These barriers consist of overlapping columns created by a series of large
diameter counter rotating augers mixing in situ soils with additives, usually bentonite or cement
grout, which is injected through the augers.
Grout Walls: These barriers are installed by grouting or jet-grouting the soils. Grout walls or
"curtains" have been used extensively in the past for civil engineering projects, but less
frequently at hazardous waste sites. They are usually more expensive than other techniques and
the barriers have higher permeability. However, grout walls are capable of extending the key of
other types of subsurface barriers through bedrock.
-------
Sheet Pile Walls: These barriers traditionally have consisted of steel sheeting with some type of
interlock joint. Recently, such sheeting includes an improved interlock design to accommodate
sealing of joints; several innovative techniques have been developed recently to seal and test the
joints between sheet piles. In addition, plastic has been substituted for steel in a number of
applications (see Figure 1-2).
Liners: Liners also have been used as vertical barriers, either alone or in conjunction with slurry-
walls.
A main concern in the application of these technologies is control of the key into the aquitard. It
should be noted that the slurry trench excavation method is the only one that permits visual
inspection of the key material and assurance of the key-in depth during construction. Appendix
A describes construction methods used to install vertical barriers.
The applicability of a particular method at a site depends on the in situ groundwater conditions,
soil and rock conditions, permeability desired, depth of installation, presence of adjacent
structures, and cost considerations. For example, at one site studied, two walls were installed.
The outer wall was constructed by the slurry trench excavation and backfill method as the
primary barrier to migration of contaminated groundwater. An inner wall also was constructed
by the vibrating beam method to limit the area of active (that is, pumping) containment.
Caps
To meet the performance standards set forth under RCRA Subtitle C, the cap design should
include the following layers (EPA 1989):
A base soil layer to support the other layers
A low-permeability layer (1 x 107cm/sec or less)
A drainage layer
A soil cover, including a vegetative layer
A gas collection and venting layer maybe required to control contained gases, such as methane.
The gas collection layer typically is constructed of granular soils and collection and vent pipes.
The low-permeability layer is constructed of clays and geosynthetic material (geosynthetic clay
liners as a substitute for clay and geomembranes). The drainage layer is constructed of granular
soils, or geosynthetic materials, and the soil cover usually is silty soils. The vegetative layer
consists of topsoil that supports vegetation. In industrial areas, a parking lot or other alternative
surface can be constaicted over the soil cover. Appendix A provides details of typical cap
design, construction, CQA/CQC, and cost.
1.1.3 Previous Studies
Previous studies have focused on the performance of caps, especially the stability of caps in cases
in which geosynthetic materials are used (EPA 1993). The performance of different low-
permeability soil mixtures for use in vertical barriers has been studied to a lesser degree, but
some notable works are available (U.S. Army Corps of Engineers [USAGE] 1996; Evans 1994).
-------
DC
LLJ
-------
Tliis study strives to build upon those early works, rather than repeating them, and draws heavily
upon the results of the identification of more than 160 sites and performance of detailed
evaluations of 36 sites.
1.2 OBJECTIVE OF THE STUDY
The objective of this study is to perform a national retrospective analysis of the field performance
and effectiveness of engineered containment barriers, including innovative and emerging
technologies. The emphasis of the study is on the performance of vertical barriers; however, the
performance of caps used in conjunction with vertical barriers also is evaluated. The factors that
affect performance of containment systems, especially design, CQA/CQC, and operation and
maintenance, are evaluated in detail for 36 sites and their performance determined. The study
also analyzes monitoring techniques used to measure performance.
1.3 SCOPE OF STUDY
The scope of this study is to analyze performance of vertical barriers and caps (if present at the
site) at sites for which sufficient data on performance were available. The four primary tasks
completed during the study are described briefly below.
1.3.1 Site Selection
A nationwide search was launched to identify representative contaminated sites at which vertical
barriers or caps had been used as the containment method during the remedial action. To
determine what lessons could be learned about the performance of vertical barriers, a few
nonhazardous waste sites for which large quantities of performance data related to the vertical
barrier were available also were identified. An initial list of 162 sites was developed. On the
basis of availability of performance data and several other criteria, 36 sites were selected from
among them for detailed analysis of actual field performance.
1.3.2 Data Collection
For the 36 sites selected, data on design, CQA/CQC, performance monitoring, operations and
maintenance (O&M), and cost were obtained by contacting regulatory agencies, contractors, and
owners of the sites. Very large quantities of performance data were obtained for certain sites that
had been constructed in the 1980s, while only limited performance data were obtained for other
sites.
1.3.3 Determination of Existing Industry Practices
The project team reviewed available literature, in conjunction with the analysis of data obtained
from the 36 sites. That research was reinforced further by discussions with constructors and
designers of vertical barriers and with university researchers. The team established standard
industry practice through examination of the information collected and the authors' experience
and professional judgment. For the performance evaluation, a matrix of industry practice was
developed and used as a reference in comparing the performance of individual sites.
-------
1.3.4 Performance Evaluation
The performance evaluation included assessment of the performance of barriers at individual
sites in light of regulatory performance standards, as well as comparison of standards and
performance among the several sites. Available historical monitoring data, primarily those on
ground-water quality and hydraulic head, were reviewed to determine the short-term and
long-term environmental improvement provided by the containment and to determine whether
regulatory standards and performance objectives were being met effectively. Performance, as
measured against these standards and objectives, was compared among the sites.
In addition, available data describing criteria for containment design, CQA/CQC, monitoring,
and O&M were evaluated to assess how each criterion affected performance of the containment
barrier. All sites were compared to assess any trends between the site criteria and performance
among the sites studied. The results of the performance evaluation were summarized to identify
common elements among the sites or containment types that contribute to performance as well as
noted pitfalls of performance or contributory criteria.
1.4 LIMITATION OF THE STUDY
This study focused on U.S. sites for which data were readily available. In addition, most barriers
in the study have been in place for fewer than 10 years; therefore, long-tenn performance can
only be extrapolated.
All sites included in the study were existing sites that had vertical barriers and, in many cases,
caps. None of the sites has an engineered bottom barrier. Therefore, the effect of leakage
through aquitards was not evaluated in this study.
Another limitation of the study was that existing available data rather than primary data were
used; therefore, the analysis in this report is based on the assumption mat published or publicly
available data used for this study were collected and reported properly. In addition, surrogate
factors or indicators of performance, such as design, CQA/CQC, and monitoring results, were
used to evaluate sites. However, releases have occurred from sites mat have exemplary design,
CQA/CQC, and monitoring systems. Further, some releases may go undetected.
Some site owners required anonymity before they would release data; therefore, sites evaluated in
this study are not identified by name.
2.0 SITE SELECTION
Selecting sites for the nationwide assessment of field performance of existing subsurface
engineered barriers involved identifying sites at which engineered barriers are in use and
determining whether performance data are available for those sites. After identifying sites that
have engineered barriers, the project team developed criteria for selecting as many as 40 sites for
evaluation of the performance of the barriers. Sites were selected on the basis of the amount and
type of performance data available, as well as pertinent criteria, such as type of barrier
technology, installation techniques, geographic and geologic distribution, age of the barrier
containment svstem, and nature of the contamination at the site.
-------
Tliis section describes the identification and selection process used in choosing the sites at which
the performance of subsurface engineered barriers would be evaluated. This section also
discusses the limitations of the site selection process.
2.1 SITE IDENTIFICATION
To choose appropriate sites that had subsurface engineered barriers, the project team identified
more than 160 sites at which subsurface engineered barriers were known or believed to be in
place. The project team identified the sites through an open international literature search, as
well as through extensive academic, industry, and government contacts and published sources.
2.1.1 Literature Search
An international open literature search was conducted to identify information available about the
use and performance of subsurface engineered barriers. The purpose of the literature search was
to document the status of existing subsurface engineered barriers, as well as to identify available
information on the design, installation, operation and maintenance, and overall performance of
subsurface engineered barriers. Terms commonly associated with subsurface engineered barriers
(such as slurry walls) were used in conducting the literature search.
The DIALOG service of Knight-Ridder Information, Inc. was used to conduct the on-line
literature search. DIALOG provides a one-step search that covers millions of documents drawn
from scientific and technical literature, as well as from trade journals, newspapers, and news
services. According to Knight-Ridder, DIALOG draws from more sources than any other on-line
service.
The group of files selected for the on-line literature search was "all science," consisting of
174 files. Each of the 174 files was searched for the following key words and phrases:
Engineered barriers
Groundwater barriers
Hydraulic barriers
Subsurface barriers
French drains
Slurry wall(s)
Groundwater cutoff wall(s)
Groundwater dewatering
The project team searched for the years 1980 to the present. To identify available international
bibliographic information, the search covered sources in English and in other languages, such as
French, German, and Japanese.
The key words and phrases identified above were found in 81 "all-science" files. Most of those
files showed few matches and were eliminated from further searches. Some files were newsletter
files that, although they identified companies that used subsurface barriers, were not technical in
nature. Those files also were eliminated from further searches. Five files were selected for the
continuation of the on-line search National Technical Information Service (NTIS), Compendex
Plus, Fluidex (Fluid Engineering Abstracts), Energy Science and Technology, and Water
10
-------
Resources Abstracts. In the second step, the matches obtained through the first search were
searched using the following key words:
Performance
Monitoring
Evaluation
Effectiveness
Efficiency
The project team printed and reviewed the bibliographic results of the second search. The team
then requested abstracts of 92 of the documents for further review. In addition to conducting the
on-line literature search, the project team identified a number of applicable reference texts,
guidance documents, and journal articles.
The project team also conducted an on-line search of the EPA Record of Decision System
(RODS) database for the years 1980 to the present, using the same key words that were used for
the DIALOG search. Abstracts obtained from RODS were reviewed to determine the status of
remedy implementation and to identify sites at which engineered barriers had been installed.
2.1.2 Industry Sources
The project team also contacted a number of academic, industry, and government sources to
identify sites at which existing subsurface engineered barriers were in use. Industry sources
included:
Site owners and operators
Barrier design and engineering firms
Barrier installation contractors
Environmental consultants
Those sources, as well as the academic and government sources contacted, not only were able to
identify sites at which subsurface engineered barriers had been installed, but also were able to
identify7 sources of data for the sites and information about the performance of engineered
barriers.
2.2 SITE SELECTION CRITERIA
To select approximately 40 sites for more detailed evaluation, the project team screened the sites
identified to ascertain the extent and type of performance data that were available for each.
Specifically, the criteria that represent factors crucial to the evaluation of the performance of
engineered barriers were applied to screen the sites. The principal criteria represent data on
monitoring, design, and CQA/CQC. Monitoring data provide a measure of the performance of a
barrier at a specific location or point in time. Data on barrier design and CQA/CQC are
extremely important because of the direct relevance that design and CQA/CQC have on
performance of the barrier. Other relevant (but not crucial) criteria include type of barrier;
geologic setting and distribution; whether the barrier is integrated with a cap or active pumping;
and other considerations or features, such as innovative or emerging barrier technologies. The
intent of the selection process was to identify a cross-section of barrier technologies and
locations. The criteria are described below.
11
-------
2.2.1 Availability of Adequate Monitoring, Design and CQA/CQC Data
Accurate and adequate monitoring data are essential to the evaluation of the performance of
subsurface engineered barriers. The extent of monitoring varies from site to site, depending on
the purpose for which the wall was installed. Types of monitoring data collected at sites include:
Hydraulic head, within and outside the wall
Groundwater quality, within and outside the wall
Settlement of the top surface of the wall
Verticality of the wall
At some cutoff walls installed at dam sites, geotechnical instruments, such as inclinometers,
stress cells, electric piezometers, and survey markers, are installed to monitor the long-term
behavior of the wall. However, at most hazardous waste containment sites, only the first two
types of data (that is, hydraulic head and groundwater quality) are collected. The frequency with
which such data are collected depends on how recently the barrier was installed and the stage of
monitoring. Groundwater level and groundwater quality data generally are collected quarterly or
even monthly after installation of the barrier. The frequency of data collection is reduced once a
data trend or new baseline has been established.
Design data also are caicial in evaluating the performance of the barrier. Indeed, barrier design
objectives establish performance standards, as well as performance monitoring approaches. For
most sites, a complete design report is available mat includes the drawings and specifications.
However, the quality of the design report varied from site to site. The report might set forth the
basis of the design, calculations, value engineering, and performance monitoring requirements
after construction.
The performance of a barrier wall is also highly dependent on the CQA/CQC program followed
during installation. For example, if a subsurface barrier is keyed into a substratum incorrectly,
the containment system ultimately could fail, despite adequate design. Therefore, sufficient
CQA data are crucial in assessing the performance of a barrier. The installation contractor's
quality control testing, independent CQA/CQC inspection and testing, and documentation by the
engineer are components of the CQA/CQC program.
2.2.2 Representativeness Regarding Types of Barriers
Several types of barriers were considered for the study, including soil-bentonite walls and
variations of such walls, funnel-and-gate systems, and sheet piling systems. A range of typical
performance characteristics is associated with each type of barrier. Although most of the sites
considered for detailed evaluation had soil-bentonite subsurface barriers, an attempt was made
through application of this criterion to select examples of different barriers.
2.2.3 General Geologic Distribution
Ideally, the performance of subsurface engineered barriers should be evaluated in a variety of
geologic settings. However, a majority of the sites that were identified for this study are located
in the eastern United States, because barrier walls have been used more often for waste
containment at Superfund sites in EPA Regions 1, 2, 3, and 4 than in other areas of the country.
12
-------
Nevertheless, to ensure a wide geographic distribution, some sites in the Midwest. Rocky
Mountain states, California, Washington, and the South were identified and selected. (Because
the geologic setting would be determined during the detailed evaluation phase of the project
geographic distribution was used as a surrogate for geologic setting in screening sites.)
2.2.4 Unique Features
Other features, such as a unique hydrogeologic setting or a relatively new technology
incorporating geomembranes, were considered in selecting the sites. Consideration of those
features was used as a screening criterion to reflect state-of-the-art developments, such as
innovative and emerging technologies. This criterion also includes other factors, such as lessons
learned from the use of a variety of containment systems in specific applications and settings.
In addition, availability of cost information and determination that the engineered barrier was
part of an integrated system (including a cap and active pumping from within the wall) were used
as screening criteria.
2.3 SELECTION PROCESS
The project team was able to identify 162 sites at which engineered barriers were known or
believed to be in place. They included sites listed on the National Priorities List (NPL), as well
as municipal landfills and sites at which corrective actions had taken place under RCRA. Several
sites identified were dam sites or other sites that required cutoff or containment of water, but did
not contain any hazardous constituents or contaminants. The project team assembled a list of
items about which information was to be collected, including the name and location of each site,
the type of barrier present, the type of monitoring conducted, and other pertinent information.
2.3.1 Results of Initial Search
Of the 162 sites identified, 10 had only a cap as an engineered barrier; that is. there was no
vertical barrier at the site. Although one of those sites would be included for detailed evaluation
of performance, it would be evaluated separately from sites having vertical barriers. A number of
other sites were eliminated because of such factors as:
The record of decision (ROD) had been changed to specify a different remedy.
Hydraulic control was achieved through pumping alone (usually because of a change
in remedy); no subsurface engineered barrier existed.
The owner or operator of the site was not obliged to and did not wish to participate
in the study.
In all, 32 sites were eliminated from further consideration. Thus, the number of sites considered
for more detailed evaluation for subsurface engineered barriers became 130.
13
-------
2.3.2 Sites Selected for Detailed Evaluation of Performance
For each criterion discussed above, the site was ranked on a scale from 1. indicating that data
were not available or were not well defined, to 5, indicating that good quality data were
available. For each site, the ratings were assigned by the project team member who was most
familiar with the site. Project team members relied upon professional judgment to determine the
rating under each criterion. As a result, only those sites for which sufficient data were or were
likely to become available for analyzing the performance and effectiveness of the subsurface
engineered barrier were retained for detailed evaluation.
The criteria have different degrees of importance in evaluating the performance of containment
systems. Therefore, a weighting system was used to reflect the relative importance of each
criterion. Availability of monitoring data is crucial to a meaningful evaluation of the
performance of the barrier wall; that criterion therefore was given a weighting of 35 percent. The
weighting of the criteria related to availability of data was established as follows:
Criterion
Monitoring data
CQA/CQC data
Design data
Barrier type
Unique features
Cost data
Geographic distribution
Integrated system
Weighting (percent)
35
15
15
10
10
5
5
5
As a result of the ranking and weighting of criteria, the 130 subsurface engineered barrier sites
were assigned a priority according to their high, medium, or low potential for selection as one of
the 40 sites for which detailed evaluation of performance would be conducted. Assignment of
priority was based on the professional judgment and experience of the project team. In addition,
some sites that had been designated as medium potential were included in the high potential
category (at least through the data collection phase of the project) to increase the geographic
distribution of sites or types of barrier included. Ultimately, 36 sites were selected for detailed
evaluation, including one cap-only site.
2.4 LIMITATIONS OF THE SELECTION PROCESS
The list of 162 sites at which engineered barriers were believed to be in place was not considered
a comprehensive compilation of sites at which subsurface engineered barriers may have been
installed. Rather, the list reflects the consensus of the project team about a representative sample
of enforcement, RCRA, municipal landfill, and dam sites at which subsurface barriers are in use.
Further, because the selection process emphasized the availability of adequate monitoring,
design, and CQA/CQC data (as well as the other criteria), the sites selected for detailed
evaluation generally represent the best monitored and documented sites. Therefore, the sites
selected for detailed evaluation are, by definition, not representative of subsurface engineered
barrier sites. For example, although 10 cap-only sites originally were identified, only 1 cap-only
site was selected for detailed evaluation because little (if any) performance monitoring data were
14
-------
available for the other cap-only sites. In addition, because data collection ceased for a site once it
had been categorized as medium or low priority, it would be difficult to quantify the amount of
data that do not exist for sites not selected for detailed evaluation.
3.0 DATA COLLECTION AND ANALYSIS - VERTICAL BARRIERS
The site selection process described in Section 2 identified 36 sites to be analyzed in detail. Data
collection and analysis were performed for those 36 sites so that site summaries could be
prepared. The summaries, provided in Appendix B, highlight:
Description and history of the site
Geologic setting
Nature and extent of contamination
Containment remedy
Performance evaluation
General site information, as described above, provided the background necessary to evaluate
more specific performance-related criteria. Data were collected that describe four specific
performance-related criteria: design, CQA/CQC, monitoring, and O&M. Those data were
analyzed to determine performance of the containment and compare it with design objectives.
Monitoring data, typically surrogate information about groundwater quality and hydraulic head,
provided the measurements that were analyzed to estimate the containment performance relative
to the established regulatory performance standards. In addition, monitoring was reviewed to
assess the ability of the prescribed monitoring to measure performance adequately. Design,
CQA/CQC, and O&M were analyzed to evaluate their contribution to short- and long-term
performance.
An industry standard baseline was established for design, CQA/CQC, and monitoring to aid in
the comparison and evaluation of the sites. The baseline was determined from experience, data
sources, and published guidelines. The baseline enabled the project team to assign a comparative
rating to specific criteria for an individual site. The evaluation team then could identify unique
aspects of the three criteria that could be viewed readily in the subsequent performance
evaluation. The objective was to recognize the possible contribution of criteria or subcriteria to
containment performance. A secondary objective of the analysis was to better understand the
variability of specific subcriteria among the sites evaluated.
Section 3.1 briefly describes the content of each of the categories listed above. Sections 3.2, 3.3,
3.4, 3.5, and 3.6 describe the design, CQA/CQC, performance monitoring, O&M, and cost of
barriers.
3.1 GENERAL INFORMATION ABOUT THE SITES
Geotechnical and hydrogeological investigation reports, remedial design documents, remedial
action completion reports (including as-built drawings), quarterly monitoring reports, and other
available data were obtained for 36 sites, as described above. The data in those reports were
analyzed to prepare the site summaries, with particular emphasis on performance evaluation
criteria. Table 3-1 provides a summary of the sites by type of barrier, cap, and extraction system.
15
-------
TABLE 3-1
CONTAINMENT SYSTEMS SUMMARY
Site
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Barrier Type
Year
Installed
1989
1981
1987/89
1993
1991
1993
1990
1989/94
1995
1983
1992
1991
1984
1987
1984/89
1981
1991
1994
1982
1996
1994
1991
SB
CB
SCB
Clay
Concrete
Sheet Pile
Existing
Plastic
Concrete
Vibrating
Beam
Cap
RCRA
Under
Construction
Clay
Under
Construc-
tion
Soil
Proposed
Asphalt
Extraction System
Pumping
Wells
Under
Construction
Leachate
Collection
?
Drains
16
-------
TABLE 3-1 (Continued)
CONTAINMENT SYSTEMS SUMMARY
Site
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Barrier Type
Year
Installed
1984
1990
1989
1977
1990
1986
1986
1991
1986
1989
1994
1988
1994
1992
SB
Hanging
wall
Primary
CB
SCB
Treat-
ment
Wall
Clay
Concrete
Sheet Pile
Plastic
Concrete
Vibrating
Beam
Intermediate
Cap
RCRA
Clay
Soil
Asphalt
Extraction System
Pumping
Wells
Intermittent
Leachate
Collection
Drains
Note: SB = Soil bentonite
CB = Cement bentonite
SCB = Soil cement bentonite
RCRA = Resource Conservation and Recovery Act Subtitle C
17
-------
The summaries in Appendix B, Volume II contain the following sections, based on information
available by mid-April 1997.
3.1.1 Site Description and History
This section describes the history of waste disposal activities at the site and the physical setting
of the site. The dates of investigation, remedial design, and remedial action are presented, and
the remedy is described.
3.1.2 Geologic and Hydrogeologic Setting
The geologic setting, site stratigraphy, and hydrogeologic units present at the site are described.
The rate and direction of groundwater movement, the existence of an aquitard formation, and any
surface-water features are indicated in this section.
3.1.3 Nature and Extent of Contamination
The types and concentrations of contaminants found in the soil and groundwater at the site are
described in this section. The amount of contamination found in monitoring wells outside the
containment system also is indicated.
3.1.4 Containment Remedy
Key components of the remedy implemented at the site are identified in this section, with
emphasis on any unique features. The objective of the containment remedy also is stated, so that
the performance could be evaluated on the basis of the remedial objective.
3.1.5 Performance Evaluation
This section contains all the information available about the design, CQA/CQC, monitoring,
O&M, and cost of the remedy, rated against the "established" industry practice for design,
CQA/CQC, and monitoring of vertical barriers and caps (if data on a cap were applicable and
available). The monitoring data were examined, often replotted, and analyzed. Performance was
indicated by monitoring data reviewed and evaluated against the remediation objective, and the
contributions of design, CQA/CQC, monitoring methods, and O&M to performance were
determined.
3.2 DESIGN
The design of an engineered barrier system outlines the functional criteria and objectives of the
system. The design documents provide the constructor of the system with the detailed drawings,
specifications, CQA and CQC requirements during implementation of the design, and O&M
requirements, as well as a monitoring plan and a schedule for the installation. Therefore,
understanding the design and analyzing the design data are important in evaluating the
performance of the barrier.
The importance of proper design to achieving the intended objective and performance of a barrier
wall cannot be overemphasized. USAGE recognized that fact in its guidance document for
barrier walls (1996). This section will define the components of the design and discuss the design
matrix that was used in this study to evaluate the design of the engineered barriers.
18
-------
The primary objective of subsurface barriers is to provide hydraulic isolation of the material
enclosed by the barrier. (As discussed in Section 1.1.1. recent uses of permeable reaction walls
and deep extraction trenches are notable exceptions.) The barrier must limit lateral inflow or
outflow and must neither degrade nor allow diffusion of target contaminants through the barrier
during its design life. The design should consider the following factors, which will be described
in this section.
Hydrogeologic investigation
Determination of feasibility
Geotechnical investigation
Details of the barrier design, such as alignment and depth of key
Development of monitoring program
For this study, those factors and associated subfactors were identified, and the acceptable
industry practice was identified for each factor and subfactor. Acceptable industry practice was
obtained from several sources, including available guidance documents or texts (USAGE 1996;
Evans 1994; D'Appolonia 1980; Barvenik 1992; McCandless and Bodocsi 1987; EPA 1984),
discussions with engineers and contractors, findings of reviews of sites investigated, and the
authors' experience. Table 3-2 presents a matrix summarizing acceptable industry practice. The
matrix was used to evaluate the 36 sites selected for the study to determine whether the design
effort for the site was acceptable, less than acceptable, or better than acceptable, when compared
with industry practices. Subsection 3.2.5 discusses the range of findings about the sites.
The following discussion of barrier design focuses on slurry trench constructed soil-bentonite
barriers because such barriers are most prevalent and represent the majority of the evaluated sites.
However, many of the design subcriteria are equally applicable to other vertical barrier designs.
3.2.1 Hydrogeologic Investigation
The hydrogeologic investigation should define the subsurface stratigraphy and the hydraulic
conductivity of the aquifer and underlying impermeable zones. The hydrogeologic investigation
of a typical site includes, at a minimum:
Soil and rock borings to define stratigraphy, particularly the extent and properties of
an aquitard bottom (that is, a confining unit)
Groundwater sampling from monitoring wells and piezometers to define the water
quality and aquifer heads
Testing of aquifers to define the hydraulic conductivity of the water-bearing zones
and the extent of the contaminant plume
19
-------
TABLE 3-2
MATRIX FOR EVALUATING BARRIER DESIGN AGAINST ACCEPTABLE INDUSTRY PRACTICES
Category
Hydrogeologic Investigation
Feasibility Determination
Geotechnical Design Investigation
Borings along alignment
Geotech. physical testing
Barrier Design
Groundwater modeling
Alignment & key depth+
Wall thickness/hydrofracture
Trench stability & analysis
Backfill permeability
testing/optimization
Trench slurry compatibility
Long term backfill compatibility
Barrier penetration details
Cap/barrier interface
Protection from dessication
Protection from surface loading
Protection from subsurface breach
Sediment & erosion control
Less than Acceptable
None
None
1 boring/>200 ft
None
No Modeling
<2ft
<2ft
None
<3
<3
<3
None
None
<1 ft
None
None
None
Acceptable
Yes*
Yes**
1 boring/1 00-200 ft***
Yes****
Feasibility Modeling
2-4 ft Key
2-4 ft
Analytical
3 Tests
3 Tests
3 Tests
Contractor Designed
Component Overlay
1-2 ft Clay Cap
Spanning Elements
Physical Protection
Contractor Designed
Better than Acceptable
>
>
1 boring/<100ft
>
Design Performance Modeling
>4ft
>4ft
Numerical
>3
>3
>3
Designer Designed
Physical Connection
>2ft
>
>
Designer Designed
Documentation of hydrogeological investigation necessary to establish design parameters was available .
The feasibility determination was based on adequate geological and hydrogeologic site data.
Spacing of borings depends on geologic variability at the site
' Representative gradation, limits, unit weight and key permeability.
Soil key shown. Rock key rated less than acceptable (to fractured bedrock-no grouting), acceptable (0.5-1.0 ft into sound rock)
better than acceptable (more than 1 ft into sound rock).
20
-------
Most remediation projects at hazardous waste sites use a design life of 30 years. The durability
of construction materials used to install barrier walls at contaminated sites still is being evaluated
for that period.
At a minimum, the investigation report should include the direction and rate of groundwater flow
and the extent and properties of the low-permeability zone, as those properties will affect the
key-in of the vertical barrier.
At 23 of the sites, thorough hydrogeological investigations were conducted to identify the aquifer
and aquitards; at 11 of the sites, the hydrogeological investigations was adequate. For 2 of the
sites, the extent of the hydrogeological investigations could not be determined.
3.2.2 Determination of Feasibility
The hydrogeologic investigation provides the data necessary to determine whether a vertical
barrier is technically and economically feasible for the site. Determination of feasibility is
completed before, or sometimes during, the geotechnical investigation.
3.2.3 Geotechnical Investigation
The successful design and construction of a barrier wall requires that geotechnical data be
collected along the alignment of the barrier wall. Typical industry practice is to obtain closely
spaced soil samples from the surface to the bottom of the wall, usually an impermeable layer
necessary to establish a good key-in and prevent underflow. The primary objectives of the
geotechnical investigation are to:
Determine that a continuous aquitard exists and determine its elevation along the
slurry wall alignment
Determine the elevation of the groundwater and the presence of any artesian
conditions
Determine the physical properties of the soils through which the trench will be
excavated
To collect the soil samples, borings usually are drilled at 100- to 200-foot intervals along the
alignment so that the variations in soil horizons can be established. Tests completed on soil
samples generally include those for gradation, Atterberg limits, unit weight and moisture content,
and permeability of the key-in horizon. For sites mat are found to have geologic variability, the
borings are completed at intervals of less than 100 feet, and extensive testing of soil samples is
conducted to establish the subsurface conditions. Similarly, borings may be farther apart if
geologic strata are consistent. For sites having very uniform geology, the spacing between
borings may exceed 200 feet.
The information obtained through the geotechnical investigation is extremely important. It
allows the designer to determine that the use of a vertical barrier is technically and economically
feasible and to select the most appropriate type of barrier. The contractor also uses the
information to select the equipment required for the excavation of the barrier trench, as well as
21
-------
the rate of production, and to estimate whether all or some of the excavated material can be
reused for the impervious backfill.
Of the sites studied. 10 had borings at approximately 100-foot spacing, 4 had borings at 200-foot
spacing, and 1 had borings at 300-foot spacing. For the other sites, the design report did not
specify the spacing of the geotechnical borings; however, for all the sites, the geotechnical
investigations were adequate to thorough. At some sites, a geophysical survey was used to
supplement the geotechnical drilling program.
3.2.4 Detailed Design of the Barrier
The design of the barrier is dependent on the remedial objective for the site. The study identified
several important elements involved in barrier design; those elements are discussed below.
Groundwater Modeling
The extent of groundwater modeling for a site can vary from no modeling to detailed
finite-element modeling to predict the performance of the design. Highly complex hydrogeology
may require extensive modeling, while small sites or simple hydrogeological conditions may
require no modeling. However, acceptable industry practice is to use modeling to establish the
feasibility of constructing the barrier wall. If performance modeling of the design is completed.
and the modeling was further calibrated using postconstruction data, the site was considered
better than acceptable, compared with industry practices.
For 15 of the sites studied, groundwater modeling was completed as part of the design effort.
The extent of the modeling varied from a limited amount to define the flow pathways to a
numerical model to predict the effect of a reactive barrier on groundwater quality downgradient
of the site. The use of modeling and the extent of the modeling effort seems to have been
dependent on site conditions and requirements imposed by the state or federal regulatory
authority.
Alignment of the Wall
The alignment of the vertical barrier should be outside the contaminated zone. Such alignment
is not always possible because of specific constraints, such as presence of adjacent streams or
structures and sharp changes of topographic features. In such cases, the purpose of the
groundwater monitoring system outside the barrier is to verify long-term improvement in
groundwater quality.
Key in theAquitard
An adequate key is crucial to eliminate the risk of leaking of contaminants below the vertical
barriers. Key depth must allow for seating the barrier in competent low-permeability soil or
rock. The key should not provide a preferential pathway for groundwater flow relative to the
remaining barrier or bottom of the site.
The key must be deep enough to accommodate:
22
-------
Localized variations in the elevation and quality of the aquitard (transition from silty
material to clay material, for example)
Variations in the measurement of key depth
Conditions inherent in the type of wall and installation technique (such as thickness
of the wall)
Acceptable depth of the key usually ranges from 2 to 4 feet and depends on site-specific geology
and barrier depth. The design of a slurry wall key into bedrock is a complex issue. The degree
of fissuring and the increased in situ permeability of the upper rock stratum should be assessed.
In addition, the degree of difficulty and cost of excavating into the bedrock should be evaluated.
Keys in the 1- to 3-foot range usually can be achieved relatively economically in most shale and
limestone formations. For more competent rock, no key or only a very small key can be
excavated economically. In some cases it is economical to extend the wall below the key by
grouting.
Most barrier techniques other than slurry walls will have more stringent limitations than slurry
walls on the execution of the key into competent materials or bedrock or at increased depths. For
example, it is difficult to drive a sheet pile or vibrate a beam at depths exceeding 70 to 80 feet.
In addition, barrier techniques other than slurry walls do not provide for continuous visual
inspection of the aquitard formation, as is the case with slurry trenching.
Measures should be specified to ensure that slum- wall keys are cleaned properly before
backfilling. Some designers and contractors increase the key depth to accommodate some
buildup of soil that can settle out of suspension during the construction, but that could not be
removed before backfilling. This practice is not recommended, since such "muck" is pushed
forward during backfilling by the toe of the backfill, similar to a mud wave. Eventually, the
accumulation of soil and sand becomes too great to be displaced by the backfill, leaving higher-
permeability material at the bottom of the trench.
The importance of the key cannot be overemphasized, since most vertical containment barriers
that do not meet the design objective (mat is, that leak) are deficient in either design (usually the
assessment of the quality of the aquitard) or CQA/CQC during excavation and cleaning of the
key.
Of the sites studied, 1 site was not keyed in to an aquitard (hanging wall), 8 sites had 2-foot keys,
14 sites had 3-foot keys, 6 sites had 5-foot keys, and 1 site had an 8-foot key. The industry has
recognized the importance of the key depth. The greatest difficulty in achieving adequate key
depth was encountered at sites at which fractured bedrock occurred at depths of more than 70 feet
below ground surface.
Thickness of the Wall
Under typical conditions the thickness of the slurry wall varies from 2 to 4 feet to provide an
adequate containment barrier. The thickness is determined primarily by head differential across
the barrier and concern for hydrofracture, transport of contaminants, the practical limits of
excavation equipment, and consideration of future settlement. The sorption capacity of the
barrier also should be considered when determining the thickness of the wall. The designer must
23
-------
balance the need for a thick barrier to withstand high hydraulic heads and retard leaking through
advection and diffusion with the need to minimize construction cost. Depending on the site
conditions and construction methods for earthen barriers, some walls can have a thickness of less
than 2 feet and others can have a thickness of more man 4 feet. However, for typical
slurry-based installations, a thickness of less than 2 feet is considered less man acceptable, and a
thickness greater man 4 feet is considered better than acceptable.
The thickness of the wall was in most cases 3 feet and not less than 30 inches at 33 of the 36 sites
evaluated. The exceptions were a sheet piling site, a site at which an emergency action was
undertaken using a 1-foot-wide backhoe bucket, and an interim remedy site at which a vibrating
beam construction method was used, resulting in a 4-inch thick wall. At one site, a 10-foot-thick
clay wall was used.
Analysis of Trench Stability
The stability of a slurry-installed trench is crucial to successful construction of the wall. In most
cases, a bentonite slurry-filled trench will be stable if there is at least 3 to 5 feet of slurry head
above the surrounding groundwater table and artesian conditions are not present. If stable soil or
rock characteristics are encountered, detailed analysis of trench stability may not be required.
Concerns about stability may arise under conditions of soft native soil, high water tables,
openwork gravels, artesian conditions, long open trenches, excessive surcharge (for example,
from adjacent dikes), or construction loads. If any of the above conditions exists and a stability
analysis was not done, the site is considered less than acceptable; if empirical or analytical
techniques were used, the site is considered acceptable; and if numerical techniques were used,
the site is considered better than acceptable.
Alternative barrier types, not based on slurry trench installation, offer inherent advantages in
some cases by eliminating the need for an open trench and the possibility of trench sloughing.
At all the sites studied, trench stability was analyzed; however, at some sites that had steep
slopes or unstable soils, the stability analysis was rigorous and measures were taken to prepare
the site adequately before excavation of the trench.
Compatibility of Trench Slurry
The fresh or new bentonite slurry is prepared by mixing the bentonite with water from an
adequate source. Additives are required in such cases as:
When the water source does not have the required characteristics to make an
adequate bentonite slurry (for example, when the water is too hard)
When chemically active groundwater or contaminants present in the subsurface soils
have the potential to affect the rheological characteristics of the slurry (such as
viscosity, gel strength, and filter loss)
When trenching through contaminated groundwater, which could cause flocculation
of the slurry and instability of the trench
24
-------
In general, three trench slurry compatibility tests should be conducted (unless incompatibility is
known not to exist). Conducting more man three tests was considered better than acceptable, and
fewer than three, less than acceptable.
The compatibility of trench slurry was evaluated at most of the sites studied; the number of tests
varied from 2 to 5.
Testing of Backfill Permeability
The permeability of the backfill used to construct the barrier wall is a key design parameter that
should be tested adequately. For the soil-bentonite technique, the objective is to establish
proportions of on-site or imported materials needed to achieve the target permeability and
physical properties of the barrier backfill. References and sources differed significantly on what
constitutes standard practice. Site conditions, availability of borrow materials, and procedures
for testing permeant compatibility affect the number of tests required. However, the consensus
average was approximately three permeability tests of the backfill (the same or similar batches),
using acceptable laboratory procedures that simulate in situ conditions. Conduct of three tests
was considered acceptable. Conduct of more than three tests was considered better than
acceptable, and of fewer than three, less than acceptable.
The permeability of backfill at the sites studied varied from 1 x 10' to 9 x 10' cm/sec. The
number of tests conducted to verify the permeability varied from 2 to 5.
Long-Term Compatibility of Backfill
Since chemical reaction with contaminants can increase the permeability of the backfill, the long-
term compatibility of backfill with the in situ soils and groundwater should be analyzed.
Typically, several permeability tests of multiple pore volumes are performed to simulate a long-
term condition and identify degradation through changes in permeability with time. Such tests
often are combined with the testing of permeability of the backfill. Conducting three tests was
considered acceptable. Conducting more than three tests was considered better than acceptable,
and fewer than three, less than acceptable.
Compatibility testing was done at all sites at which leachate or contaminants were encountered.
The extent of testing varied from site to site, with rigorous testing done at some sites and very-
limited testing at other sites.
Barrier Penetration
Subsurface utilities present along the barrier wall alignment and located below the water table
must be delineated, rerouted, or protected with watertight connections. If such conditions were
not considered, the site was rated less than acceptable; if the contractor designed solutions during
construction, it was rated acceptable; and if the engineer investigated the problems and designed
solutions during design, it was rated better man acceptable. Barrier penetrations were
encountered at only a few of the sites studied. In all those cases, the barrier penetrations were
investigated and accounted for in the design by the engineer.
25
-------
Interface between Barrier and Surface Cap
The cap and barrier wall form an integrated containment system that minimizes entry of water
into the waste area or its migration out of the area. If no surface cap was provided, the site was
rated less than acceptable; if a cap was provided but the components of the cap simply were laid
over the barrier wall, the site was rated acceptable. If there was a physical connection between
the cap and the barrier wall, the site was rated better than acceptable. The physical connection
usually is a tie-in of the geomembrane and soil components of the cap to the wall or cap.
Of the sites studied, 21 had a cap and barrier interface. At only 2 sites was there no physical
connection of the cap with the barrier.
Protection from Dessication
The surface cap over the barrier wall alignment must protect against erosion, desiccation, and
long-term physical disturbance of the barrier. Since the earthen barrier materials are primarily
clays and bentonite, they are susceptible to desiccation that leads to the development of
macropores and secondary permeability in the upper section of the barrier. If the barrier wall was
protected from desiccation with less than 1 foot of cover soil, the wall was rated less than
acceptable; if the wall was protected with 1 to 2 feet of clay cap, the wall was rated acceptable.
If the wall was protected with more man 2 feet of clay cap placed in a controlled manner, the
wall was rated better than acceptable.
Protection from Surface Loading
Uncemented earthen backfill is generally of low to moderate strength and subject to
consolidation. Protection from static and dynamic loading should be provided through
engineered fill, geosynthetics, structural concrete slabs, or other suitable material. If no
protection from surface loading was designed, the site was considered less than acceptable; if
spanning elements or soil improvement methods were provided, the site was considered
acceptable. If designed structural elements were provided, the site was rated better than
acceptable.
At all the sites studied, a surface cap had been provided over the barrier wall alignment.
Protection from Subsurface Breach
Breaching by subsurface utilities or other structures in the vicinity of the barrier wall can create
permeable zones in the wall and adversely affect performance. If the possibility of such a breach
was not analyzed, the site was rated less than acceptable; if physical protection was designed to
protect the wall from the potential effects of such breaches, the site was rated acceptable. If
redundant systems to prevent such breaches were designed, the site was rated better than
acceptable.
Construction Sediment and Erosion Control
The flow of construction sediment from the vicinity of the trench into the trench can create
permeable windows in the barrier wall. If control of erosion and sediment flow was not
designed, the site was rated less than acceptable; if the contractor designed the control method
during construction, the site was rated acceptable. If the engineer designed the method of flow
26
-------
diversion and sediment and erosion control during design, the site was rated better than
acceptable.
Construction sediment and erosion control had been provided by the contractor at most of the
sites studied.
Weighting
The 18 categories listed in Subsection 3.2.1 were assigned weights, according to the importance
of each to the performance of the barrier wall. As discussed above, the design categories mat are
crucial to performance are the hydrogeology investigation, geotechnical borings along the
alignment, depth of the key, and thickness of the wall. Therefore, each of those categories was
given a weight of 10. Next in importance are the determination of feasibility, geotechnical
physical testing, groundwater modeling, analysis of the trench stability, and long-term
compatibility of the backfill. Each of those categories was given a weight of 5. All other
categories are of approximately equal importance; each was given a weight of 2.
Each site described in Appendix B, Volume II was evaluated in each category as acceptable (2),
less than acceptable (1), or better than acceptable (3). The resulting number (1, 2, or 3) was
multiplied by the weight assigned to that category, and the total for all categories was obtained.
The total was normalized by dividing the total by the total of weights for all categories. A site
that had a normalized total lower than 1.8 was deemed less than acceptable, and any site having a
total higher than 2.2 was deemed better than acceptable. Although this procedure may not reflect
the design weakness at a particular site, it treats all sites alike and represents the weighted
average design ratings for the sites.
3.2.5 Range of Findings
Subsurface barrier design for most sites was either acceptable or better man acceptable, when
evaluated according to the methodology described above. (Table 3-3 summarizes key features
and overall design rating for the sites evaluated.) Only 1 site was rated less than acceptable. At
Site 1, the design was rated less than acceptable because a thorough hydrogeologic investigation
had not been performed, nor had compatibility testing.
For most of the study sites, the geotechnical and hydrogeologic investigation was adequate to
thorough, with spacing of borings varying from 75 to 300 feet. Groundwater modeling had been
performed for sites 7, 10, 18, 19, 20, 21, 22, 26, 28, 29, 30, 31, 32, 33, and 35. Thickness of the
wall varied from 1 foot to 10 feet, with the walls at most sites having a thickness of 3 feet. The
wall at Site 16 had a thickness of 1 foot, and the wall at Site 12 had a thickness of 4 inches
because of equipment constraints. The remedy for Site 16 was an emergency action, and that at
Site 12 was considered an interim remedy that had been constructed by the vibrating beam
method. The wall at Site 5 was a 10-foot-thick shallow barrier wall (10 feet deep), constructed of
clay.
27
-------
Table 3-3
Summary of Barrier Designs
Site
1
2
3
4
5
6
7
8
9
Geotechnical
Investigation
N/A
Borings at 300
ft spacing,
thorough
investigation
Borings at 100
ft centers
Borings at
100ft
Adequate
Thorough
Thorough
Thorough
Adequate
Hydrogeological
Investigation
N/A
Thorough
investigation
Adequate
Thorough
Adequate
Thorough
Thorough
Thorough
Adequate
Groundwater
Modeling
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Detailed
hydrogeologic
models of area
were
completed
Not performed
Not performed
WaU
Thickness
(ft)
3 feet
3 feet
3 feet
40 ft deep,
3 feet
10 feet
2.5 feet
3 feet
Sheet pile
thickness
3 feet
Wall
Depth/Key
15 ft deep, 2
ft kev
Wall is 20
to 30 feet
deep, with 3
foot key
18 ft deep
At least 2
foot key
3 feet
10 ft deep,
2 ft kev
15 to 25 ft
deep,
3 ft key-
Wall is 20
to 70 feet
deep, with a
5 foot key-
Sheet piles
are 65 to 75
feet long,
with 5-foot
key-
Wall is 15
to 45 feet
deep, with
3 -foot key
Permeability
1 x 10'7 cm/sec
1 x 1 0'7 cm/sec
Target: 1 x 10'7
cm/ sec
1 x 10'7 cm/sec
1 x 1 0'7 cm/sec
1 x 1 0'7 cm/sec
1 x 10'8 to
9 x 10'9 cm/sec
Target permeability of
1 x 10'7 cm/sec
1 x 10'7cm sec
Compatibility
Testing
Not performed
N/A
NA
Yes
Not performed
Yes
Leachate-backfill
compatibility
testing
N/A
Backfill-leachate
compatibility-
testing
Cap/Barrier
Interface
Yes
Impermeable fill
cap over cutoff wall
Cap and walls are
not physically-
connected
Yes
Yes
Yes
Yes
N/A
No cap at site
Other
Gas and leachate
collection
Cutoff wall to
dewater landfill site
-
Grout curtain
underneath SB wall
Leachate collection
svstem was
designed to handle
maximum
anticipated flows
from perched GW
table
Waste solidified
within barrier
-
Flood wall and
cutoff wall at site
(both are sheet
piles)
Rating
Less than
Acceptable
Better than
Acceptable
Acceptable
Better than
Acceptable
Acceptable
Better than
Acceptable
Better than
Acceptable
No rating
criteria
Acceptable
28
-------
Table 3-3
Summary of Barrier Designs (Page 2 of 5)
Site
10
11
12
13
14
15
16
17
Geotechnical
Investigation
Thorough
Soil borings at
100 foot
intervals
Soil borings at
100 foot
spacing
< 100 ft spacing
100ft
Adequate
Adequate
Borings at 200-
ft spacing
Hydrogeological
Investigation
Thorough
Thorough
Thorough
Thorough
Thorough
Adequate
Adequate
Thorough
investigation of
aquifers
Groundwater
Modeling
Modeling
supported the
design
Not performed
Not performed
Not performed;
information
available
Yes
Not performed
Not performed
Not performed
WaU
Thickness
(ft)
3 feet
3 feet each
4 inches
3 feet
3 feet
1 foot
3 feet
Wall
Depth/Key
Wall
averages 52
feet in depth
keyed in 5
feet into
bedrock
Walls are 20
feet deep.
with 3 -foot
key
Wall is 19
to 29 feet
deep, with
3 -foot key
2 feet
8 feet in Till
Wall is 20
feet deep,
with 2 fool
kev
Wall is 23
feet deep.
with 2 foot
soil key
Wall is 15
to 33. 5 feet
deep, with
2-foot key
Permeability
Design conductivity
1 x 10"7 cm/sec
Requirement:
1 x 1 0"7 cm/sec
1 x 10"7 cm/sec
1 x 10"' cm/sec
4x 10"7 cm/sec
1 x 10"7 cm/sec
1 xlO'6 cm/sec target
permeability
1 x 10"7 cm/sec
required
Compatibility
Testing
Yes
Thorough testing
Yes
Yes
Yes, with sea
water
NA
Several
hydrogeologic and
feasibility studies
were performed
before
construction
Yes
Cap/Barrier
Interface
No cap at site
Yes
Yes
Yes
No Cap
N/A
N/A
Yes
Other
Site is combination
hydraulic barrier
and cutoff wall
300 feet SCB wall
and 5.240 feet SB
wall
Geophysical
screening survey
was conducted
Seepage cutoff for
deep open pit mine
Inclinometer
installed
Earner was
essentially
contractor-designed
since part of an
emergency action
..
Rating
Acceptable
Acceptable
Acceptable
Acceptable
Better than
Acceptable
Acceptable
Acceptable
Acceptable
29
-------
Table 3-3
Summary of Barrier Designs (Page 3 of 5)
Site
18
19
20
21
22
23
24
Geotechnical
Investigation
Borings at 100
ft spacing
Geophysical
survey along
alignment and
soil borings
Borings spaced
at 90 ft along
barrier
Borings at
100 ft spacing
Borings at 100
to 200 ft
Adequate
Limited amount
of data
collected
Hydrogeological
Investigation
Thorough
Thorough
Thorough
Thorough
Thorough
Adequate
Adequate
Groundwater
Modeling
Modeling to
define flow-
pathways
Feasibility-
study and
design-level
groundwater
modeling
performed
Yes
Yes
Conducted
Not performed
Not performed
WaU
Thickness
(ft)
30 inches
3 feet
3 feet
3 feet
3 feet
3 feet
2.5 feet
Wall
Depth/Key
Wall is 80
to 86 ft
deep, with
0.1 foot key
into bedrock
Averages 50
feet, 0.1 ft
into
weathered
rock
60 to 80
feet, with 3
foot rock
kev
Wall is 40
to 70 feet
deep with 3-
foot kev
Wall is 12
to 19 feet
deep with
3 ft key
Wall ranges
from 10 to
60 feet
deep, with a
3 foot kev
Wall is 35
feet deep,
with 5-foot
key
Permeability
1 x 10'7 cm/sec
1 x 10"' cm/sec
1 x 10'7 cm/ sec
1 x 10'6 cm/sec
Requirement of
1 x 10'7 cm/sec
1 x 10'7 cm/sec
N/A
Compatibility
Testing
Trench slurry-
tested
Significant
compatibility-
testing performed
Rigorous
compatibility-
testing with on-
site GW and
brackish harbor
water
Rigorous
Yes
Yes
No information
found.
Cap/Barrier
Interface
Cap is physically-
connected to barrier
Yes
Yes
Provision for
erosion control
measures
Yes
Cap is yet to be
constructed
Cap covers slurry
wall
Other
Geotechnical
physical testing
conducted; borings
aligned along
barrier at 100ft
intervals
Barrier designed to
reuse excavated
material
Tidal surface water
fluctuation
successfully
managed
-
HDPE membrane
inserted through
center of wall
Piezocone testing
and pumping tests
conducted; wall is
hydraulically
adequate
-
Rating
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Acceptable
Better than
Acceptable
Acceptable
30
-------
Table 3-3
Containment Barrier Design Matrix (Page 4 of 5)
Site
25
26
27
28
29
30
31
32
Geotechnical
Investigation
Thorough
Thorough
Borings at less
than TOO ft
spacing
Thorough
Adequate
Soil borings
ever,' 100 to
200 feet
Soil borings
even' 75 to 200
feet'
Soil boring at
100 foot
spacing
Hydrogeological
Investigation
N/A
Thorough
Thorough
Thorough
Adequate
Thorough
Thorough
Thorough
Groundwater
Modeling
N/A
Yes
Not performed
Yes
Limited
Yes
Regional and
site scale
models were
performed
Hydraulic
modeling
peiformed.
WaU
Thickness
(ft)
30 inches
3 feet
3 feet
3 feet
3 to 11 feet
3 feet
Minimum
thickness 30
inches
3 feet
Wall
Depth/Key
Wall is 40
to 50 feet
deep, not
keyed into
aquitard
Wall ranges
from 20 to
over 45 feet
deep, with
1 -foot key
into bedrock
Wall is 25
feet deep.
with 5-foot
key
3 feet
Wall has 2-
foot kev
Wall is 30
feet deep
with 3 ft-
key
Wall ranges
from 10 to
77 feet
deep, with
2-foot key-
Wall is 50
feet deep,
with 3 -foot
key
Permeability
Target permeability of
1 x 10'7 cm/sec for
cutoff wall
Target permeability of
1 x 10"' cm/sec for
cutoff wall
1 x 10" cm/sec
required
1 x 10'7 cm/sec
1 x 10'7 cm/sec
1 x 10'7 cm/sec
Target:
1 x 10"7 cm/sec
Target: 1 x 10'7
cm/ sec
Compatibility
Testing
More than 3 tests
performed
Not performed
Thorough testing
Yes
NA
Compatibility
study of two SB
slurry wall
backfill mixtures
Conducted
Yes
Cap/Barrier
Interface
N/A
N/A
Yes
Asphalt parking lot
Yes
Yes
N/A
Earth fill cover
overlies the barrier
wall
Other
Hanging wall,
penetrates a silty
clay layer
Most critical
deficiency of wall
design was key-
depth
More than 60 soil
samples were tested
during geotechnical
investigation
Parking Lot
-
Soil vapor
extraction in
progress
Intermittent
pumping
River channel was
relocated 150 feet
west prior to
construction of
barrier wall
"
Rating
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Acceptable
Better than
Acceptable
Acceptable
Acceptable
31
-------
Table 3-3
Summary of Barrier Designs (Page 5 of 5)
Site
33
34
35
36
Geotechnical
Investigation
49 cone
penetrometer
tests were
performed to
develop a
geotechnical
model
Thorough
Thorough
Adequate
Hydrogeological
Investigation
Thorough
Thorough
Thorough
Adequate
Groundwater
Modeling
Numerical GW
flow model
developed
Yes
Modeling
completed
Not performed
WaU
Thickness
(ft)
1.5 ft and 3 ft
3 and 5 feet
32 inches
N/A
Wall
Depth/Key
Wall
excavated to
design
elevation to
correspond
to minimum
2-foot key
76 feet deep
5 fee!
Wall
averages
138 feet in
depth
N/A
Permeability
<5 x lO'6 cm/ sec
1 x 10'6 cm/ sec
Target: 1 x 10'7
cm/sec
N/A
Compatibility
Testing
Long-term tests
performed on
selected CB mixes
Yes
N/A
N/A
Cap/Barrier
Interface
N/A
NA
N/A
N/A
Other
Site contains a
treatment wall
Civil structure
Modeling was
performed to
determine
dimensions of
pumping system
and to predict
deformation of wall
during mass
excavation
Rating
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Notes: Acceptable industry practices in Table 3-2 were used to evaluate site designs. However, only key design parameters are discussed in this table.
N/A Not applicable
NA Not available
32
-------
The key depth at most sites varied from 2 to 3 feet. The only exceptions were sites 18 and 19,
which were deep (50 to 90 feet) and each had a key of 0.1 foot into bedrock, and Site 26, which
had a key of 1 foot into bedrock. For most of the sites, the barrier wall and cap were connected
physically by extending the cap over the top of the barrier wall.
Compatibility of backfill with the contaminated groundwater had been tested at all sites except
sites 1, 5 and 26. The type of compatibility testing to be conducted to ensure long-term
compatibility has not been standardized, and the level of effort varied among the sites.
Site conditions varied among the 36 sites, and barriers were designed to accommodate those
varying conditions. Two of the barriers studied (those at Site 2 and Site 35) were designed to
withstand head differentials greater than 60 feet for several months during dewatering operations.
The barrier at Site 34 also was designed for a high head differential and accommodated
settlement and hydrofracture concerns with a two-stage construction of barriers of different
thickness. At Site 4, because of concern about a permeable bedrock key, the base of the soil-
bentonite barrier was a grout curtain in the native shale bedrock. At Site 26, a pilot barrier was
constructed; later, wing walls to the barrier were designed and constructed to better capture
migrating contaminants. At Site 33, detailed cone penetrometer investigations and groundwater
modeling were used to characterize subsurface conditions, and a test cell was constructed to
prove the reaction wall design. At Site 11, a soil-bentonite and soil-cement-bentonite barrier was
designed to accommodate significant grade changes.
The design for Site 19 was rated above acceptable. Determination of feasibility and design-stage
groundwater modeling had been performed. A geophysical survey had been performed along the
entire barrier alignment and had been supplemented by a thorough geotechnical drilling and
testing program. A significant amount of compatibility testing of slurry and backfill had been
performed. The construction specifications for the barrier wall were based on performance and
design. A bedrock key had been used; however, flow in the bedrock had been underestimated,
and leaking from the key-in horizon had occurred. The leaking subsequently was repaired by
grouting.
The study of designs at 36 sites showed the significant effect of design on field performance.
The key design elements that require the most attention are the investigation of the key horizon,
hydrogeological assessment of groundwater gradients, and compatibility testing of the backfill
with the groundwater at the site.
3.3 CONSTRUCTION QUALITY ASSURANCE AND CONSTRUCTION QUALITY
CONTROL DATA
The CQA/CQC program is important to the successful implementation of the design and to the
performance of the barrier wall. Experience gained over the past 20 years in the installation of
barrier walls and caps at hazardous waste sites has established typical industry practices for
performing CQA/CQC at such sites. This subsection describes the typical industry practices and
the range of findings for the 36 sites analyzed in this study.
CQA refers to quality assurance testing that the designer or independent CQA engineer performs
to confirm that construction complies with the design specifications, while CQC refers to quality
control testing that the constructor performs to verify the constructed product. In the following
evaluation, CQA and CQC have been combined for ease and considered a single criterion.
33
-------
Because postconstruction verification tests for the entire barrier are difficult, if not impossible.
adequate CQA/CQC is a crucial element in achieving the design objective. The typical industry
practice was determined through examination of several sources, including available guidance
documents or texts (USAGE 1996; EPA 1984; EPA 1987a; Xanthakos 1979; American
Petroleum Institute [API] 1980). discussions with engineers and contractors, findings of reviews
of sites investigated, and the authors' experience. Table 3-4 presents a matrix that summarizes
the standard industry practice. The matrix was used to evaluate the 36 sites selected for the study
to determine for each site whether the CQA/CQC effort was acceptable, less than acceptable, or
better than acceptable. The range of findings forthe sites is discussed in Section 3.3.21.
The following discussion of barrier CQA/CQC focuses on slurry trench soil-bentonite barriers
because those barriers are most prevalent and represent the majority of the evaluated sites.
However, several of the subcriteria are equally applicable to other vertical barriers that are
installed using the slurry trenching method.
3.3.1 Experience of the Specialty Contractor
Subsurface barriers require specialized geotechnical construction. The more complex the project,
the more important it is to use a specialty contractor. If the specialty contractor that installed the
wall had completed four to six comparable projects in the recent past, the site was rated
acceptable. If the contractor had completed fewer than four such projects, the site was rated less
than acceptable. If the contractor had completed more than six such projects, the site was rated
better than acceptable. At 75 percent of the sites studied, the barriers were constructed by one of
four major specialty contractors.
3.3.2 Methods of Trench Excavation
The excavation of slurry wall trenches can be accomplished with various equipment, such as a
backhoe, specially-modified clamshell, dragline, chisel, or hydrocutter. Selection of the
trenching equipment depends upon several factors, such as the type, thickness, and depth of the
slurry wall, as well as the nature of the materials to be excavated. Currently, most soil-bentonite
slurry trenches are excavated with backhoes. Some specialty contractors own large backhoes that
can excavate to depths in excess of 80 feet. When the trench is deeper man the reach of the
backhoe, the excavation usually is extended to the final depth by clamshells. CQA/CQC related
to the excavation should address at least the following items:
Excavation Equipment. The equipment selected by the contractor must be able to
excavate the trench according to the design criteria (such as width and depth) and
through the anticipated geological formations. Indeed, to provide for localized
anomalies in the elevation of the aquitard, the excavation equipment should be able
to reach depths deeper than that required by the design. In addition, the digging
power available at the bottom of the trench is important when the key must be
excavated through very dense materials. For example, additional equipment might
be needed to key the wall into bedrock.
At the majority of the sites studied, conventional or extended stick backhoes were used to
excavate barrier trenches.
34
-------
TABLE 3-4
MATRIX FOR EVALUATING BARRIER CQA/CQC AGAINST ACCEPTABLE INDUSTRY PRACTICES
Category
Specialty Contractor Experience
Trench Excavation Methods
Trench Width, Vertically & Continuity *
Trench Sounding (slope & bottom)
Trench Bottom Cleaning
Trench Key Confirmation
Slurry Mixing
Slurry Viscosity Testing
Slurry Viscosity
Slurry Sand Content Tests
Slurry Sand Content
Backfill Slump Testing
Backfill Slump
Backfill Gradation Testing
Backfill Permeability Testing
Backfill Target Permeability
Backfill Mixing/Placement
Capping Confirmation
Barrier Continuity
Post Construe. Barrier Sampling/Testing
As-Built Records
Groundwater Head Monitoring
Final barrier alignment survey
Barrier construction specification
CQA/CQC program & testing spec.
Groundwater Chemistry Monitoring
Less than Acceptable
<4
No Inspection
No Inspection
>20ft
None
No Sampling
<
<2
<40
<2
>15%
<
<3" or >6"
< 1
< 1
>
Loosely Controlled
None
Interrupted
None
None
None
None
None
None
None
Acceptable
4-6 Comparable Projects
Periodic Inspections
Periodic Inspection
per 10-20 ft
Yes*
Sampling every 20ft
Agitation >12 hrs. Hydration
2 per shift
40+seconds (marsh funnel)
2 per shift
<15%
1 per 400-600 cy
Most tests 3"-6"
1 per 400-600 cy
1 per 400-600 cy
5x10-7 - 1x1 0-7 cm/sec
Controlled Mix/Place
Cap confirmed
Continuous
Minimal
Const. Completion Report
Monitored Fluctuation
Surveyed
Barrier
Designer Specified
Minimal
Better than Acceptable
>6
Constant Inspection
Measured
<10ft
>
Sampling < 20 ft
>
>2
40-50 seconds (marsh funnel)
>2
«15%
>
All tests 3"-6"
> 1
> 1
<
Central Mix/Guided Placement
>
Continuous & Confirmed
Regular & Documented
Report, Drawings, Test Results
Periodic & Across Barrier
Surveyed & Monumented
Barrier & CQA Plan
Independent Duplicate QA
Periodic & Across Barrier
* Observation of trench width and equipment vertically.
Note: The categories, slurry sand content and backfill slump, are site-specific, and the numbers given above are typical for soil-bentonite slurry walls.
35
-------
Trench Inspection. The trench should be inspected regularly, to ensure that it is
aligned as specified in the design and to detect any sloughing, since such sloughing
may indicate the need to clean the bottom of the trench or top of the backfill.
Moreover, an inspection will establish whether the trench is continuous through its
full depth.
If the excavation was inspected regularly (for example, daily), the site was rated acceptable. If
no inspection was conducted, the site was rated less than acceptable. If frequent inspection was
provided, the site was rated better than acceptable.
3.3.3 Width and Verticality of the Trench
As a general rule, the trenching tool at a minimum should have the width specified in the design
to ensure that the width of the barrier will conform with the design. Excavation buckets should
be monitored regularly, and such items as teeth and side cutters should be replaced as needed
before they exhibit excessive wear.
Verticality of the trench also must be monitored. Verticality is particularly important when the
design and construction methods involve joints, such as those between slurry wall panels,
stabilized columns, or vibrating beam imprints. For example, monitoring the verticality of the
excavation helps ensure that the minimum design width is achieved at foil depth if adjacent
panels deviate from the vertical in opposite directions. Verticality is less critical for continuous
excavation of the trench if the construction procedure provides for a positive method to control
the continuity of the trench between adjacent excavated sections. If periodic inspections were
conducted to monitor the width and continuity of the trench and verticality of the equipment, the
site was considered acceptable. If no inspections were conducted, the site was rated less than
acceptable. If actual measurements such as physical measurements of width and measurements
of the level of the excavator, were obtained periodically, the site was rated better than acceptable.
Inspection of the width and verticality of the trench was conducted at all the sites studied. The
frequency of inspections varied from one to two times per day, or from 10 to 25 feet of trench
advance. The type of inspection varied from visual to actual measurements of the width of the
trench and verticality of the wall. At one site, a mechanical caliper device was used to measure
width at different depths. Information about the site revealed that the width of the trench
remained relatively constant, except for the upper sections of the excavation and in areas of
sloughing caused by weak soil or the presence of waste.
3.3.4 Confirmation of Key and Aquitard
Confirming the key of the trench into the aquitard is crucial to the successful installation of the
barrier wall and to its subsequent performance. Confirming the key consists of measuring the
depth of the trench (1) when the top of the aquitard is encountered and (2) after completion of the
trench. In addition, samples of the aquitard formation should be taken at regular intervals with
the excavator or some suitable sampling tool. The engineer of record then can use the results of
such sampling to confirm that the key is within the selected formation. If sampling was
performed every 20 feet, the site was rated acceptable. If sampling was not performed and only
sounding was performed, the site was rated less than acceptable. If the sampling was performed
at a frequency of less than 20 feet, the site was rated better than acceptable.
36
-------
At most of the sites studied, confirmation of the trench key was accomplished by visually-
inspecting the trench bottom cuttings. At Site 8, the measured resistance to sheet pile driving
was used to confirm the key. At Site 27, the key was confirmed by inspecting samples of trench
bottom cuttings for every 25 feet of trench advance. Confirmation of the trench key was
dependent on the qualifications of the inspection personnel; at sites at which a distinct aquitard
was not present or weathered bedrock was present, confirmation of the key was difficult. The
importance of a qualified geologist or geotechnical engineer verifying adequate key-in cannot be
overemphasized. Inadequate key-in zones were discovered during postconstruction sampling at
some sites, and appropriate corrective action was taken. At some sites, inadequate key-in was
revealed only when leaking from the bottom occurred.
3.3.5 Sounding and Cleaning of the Bottom of the Trench
During excavation, soil materials become suspended in the slurry. In addition, if no adequate
surface erosion or sediment control barriers are in place, surface sediments can flow into the
slurry-filled trench during storms. These materials can settle from suspension and accumulate at
the bottom of the trench or on the slope of the backfill. Usually, soils and sediments are more
permeable man the backfill and must be removed before backfilling. Therefore, any
accumulation of sediment must be monitored before the trench is backfilled. The depth of the
trench must be measured (sounding) to verify that it is equivalent to the specified key depth. If
any material has accumulated, additional cleaning of the bottom of the trench must be completed
before backfilling. Cleaning the bottom of the trench may be accomplished with excavation
equipment or with air-lift or special pumps. If accumulation of sediment occurred and such
cleaning was performed periodically, the site was rated acceptable. If cleaning was not
performed, the site was rated less than acceptable. If cleaning was performed frequently (once a
day or more), the site was rated better man acceptable.
At many of the sites studied, cleaning of the trench bottom was accomplished with desanding
pumps. When this procedure was not performed regularly, permeable windows were observed
during postconstruction testing. At Site 18, the sand runs into the trench were not detected, and
cleaning of the trench bottom was not performed regularly. The permeable windows in the wall
were repaired by the deep soil mixing method.
3.3.6 Sounding of the Trench and Cleaning of the Backfill Slope
The slope of the backfill in the trench also must be monitored. Sounding of the backfill slope
should be done at a minimum of twice daily, before work in the morning and after work at night,
to detect cave-ins between shifts. Such soundings are relatively imprecise because of the soft
consistency of the soil-bentonite backfill. The periodic measurements allow detection of any
major anomaly in the backfilling process. Some specialty contractors use a special device to
verify mat no sediments have settled on the backfill slope.
Cleaning of the backfill slope is rarely required, if the rheological characteristics of the slurry are
well maintained. Nevertheless, the need for cleaning the backfill slope exists. Since it would be
risky to straddle the open trench with a backhoe, such cleaning often will require the use of a
crane-mounted clamshell or special procedures developed by the specialty contractor.
Note: In light of the above discussion of cleaning the bottom of the trench and the backfill
slope, it is recommended that a crane-mounted clamshell or other approved special
cleaning tool be mobilized or readily available to sites at which there are deep trenches.
37
-------
If the observations and measurements described above were made every 10 to 20 feet along the
barrier wall excavation, the site was rated acceptable. If they were made at intervals of more than
20 feet, the site was rated less man acceptable, and if they were made at intervals of less than 10
feet, the site was rated better than acceptable.
Trench sounding was performed at least daily at all the sites studied. The frequency of trench
sounding varied from 10 to 25 feet of trench advance. At Site 6, the backfill profile was
measured twice daily to verify- that the trench had not sloughed in. At Site 15, the depth of the
trench was determined by measuring the depth of auger in the trench.
3.3.7 Bentonite Slurry
CQA/CQC of the bentonite slurry is important to ensure the constructability, as well as the
performance, of the slurry wall. The slurry plays an important role in determining:
The stability of the trench under excavation
The cleanliness of the trench bottom and backfill slope, as a result of the ability of
the slurry to keep soil material in suspension
The quality of the backfilling operation
Mixing of Fresh Bentonite
The mixing water should be tested to ensure that it is suitable for mixing with the bentonite
material. Typically, tests are performed for pH, hardness, and dissolved solids. Most project
specifications require the use of bentonite materials that meet standards set forth in API 13A and
B. It is good practice to mix the water and bentonite in a high-shear mixer and allow the slurry'
to hydrate fully in storage tanks or ponds for a minimum of 12 to 24 hours. The slurry should be
kept agitated during storage. This procedure will produce a slurry that has the optimum
rheological characteristics (viscosity, gel strength, density, and filter loss). If the agitation of the
slurry was maintained to achieve hydration in more than 12 hours or high-speed shear mixers
were used, the site was rated acceptable. However, if hydration time was significantly less than
12 hours, with very little quality control, the site was rated less than acceptable. If the typical
agitation was such that hydration time was significantly more than 12 hours, the site was rated
better than acceptable.
At all sites for which data were available, slurry was mixed thoroughly in a pond or tank before it
was introduced into the trench. However, for most of the sites studied, rating for this criterion
was not possible because of lack of data.
Ex Situ Testing of Bentonite Slurry
The rheological characteristics of the fresh slurry should be measured before it is introduced into
the trench. On-site testing of the gel strength of the slurry rarely is required because the viscosity
of bentonite slurry is also an indication of its gel strength. The higher the viscosity, the higher
the gel strength.
38
-------
However, gel strength does not always correlate with viscosity for slurry produced with materials
other than bentonite. For example, typical biopolymer slurries used for the installation of
(leachate) collection trenches by the slurry trenching method exhibit viscosity, as a function of
the content of biopolymer material, but have a very low gel strength regardless of the viscosity of
the slurry. Such slurries therefore do not keep solids in suspension.
The Marsh funnel is the standard used to measure the viscosity of slurry; the time required for a
measured amount of slurry to flow through the Marsh funnel is an indication of its viscosity.
Viscosity may be less than 40 seconds if the trench is excavated through stable material. But
typical values specified will be:
Marsh viscosity 3 40 seconds
Density 3 64 lb/ft3
Filter loss # 15 to 25
These rheological characteristics should be tested at least twice per shift before the fresh slurry is
introduced into the trench, and corrective action taken if the results should be lower than the
values set forth above.
If these measures were taken twice per shift, the site was rated acceptable. If they were taken less
often than twice per shift, the site was rated less than acceptable. If they were taken more than
twice per shift, the site was rated better than acceptable.
For the sites studied, the frequency of testing of slurry for viscosity varied from 1 to 4 times per
shift, and conformed to contract specifications for the site. At the sites studied, the Marsh slurry
viscosity generally ranged from 40 to 50 seconds.
Testing of the Bentonite Slurry in the Trench
It is good practice to test the rheological characteristics of the bentonite slurry in the trench at
least twice per shift, to ensure that no degradation of the required characteristics occurred. Such
degradation could be caused by dilution of the slurry with groundwater (that is, from excavation
through material having a high water content, localized artesian conditions, or surface runoff into
the trench), or by chemical reactions. The measurements also are made to ensure that trench
slurry is appropriate for backfilling of the trench. In particular, if the density of the slurry is
similar to that of the backfill, the bentonite slurry may not be displaced completely by the
backfill; pockets of slurry then would remain inside the backfill, creating higher permeability
zones. The slurry therefore should be sampled at different elevations in the trench, including
close to its bottom.
In addition to viscosity, density, and filter loss, the sand content of the slurry also must be
measured, since the action of excavation suspends soil materials in the slurry, thereby increasing
the density of the slurry. As discussed above, increased density of the slurry not only can impair
the backfilling operation, but also can cause sand sedimentation at the bottom of the trench.
Typical maximum sand content as specified varies between 15 and 20 percent. (For concrete
walls, sand content should be less than 5 percent.) There are several ways to solve such
problems. Standard practice is to control the density of the slurry in the trench so that its density
is at least 15 Ib per ft3 less than the density of the backfill. That result is achieved either by
adding fresh slurry in the trench or by withdrawing the slurry at the bottom of the trench and
39
-------
circulating it through a desanding or desilting unit. Cleaning of the trench bottom was discussed
in Subsection 3.3.5.
If the trench slurry was tested twice per shift, the site was rated as acceptable. If fewer than two
tests per shift were performed, the site was rated less than acceptable. If more than two tests per
shift were performed, the site was rated better than acceptable.
At the sites studied, testing of the trench slurry conformed to contract specifications, and the
number of tests varied from 1 to 3 per shift.
For many of the sites studied, no information about the sand content of the slurry was available.
3.3.8 Mixing and Testing of the Backfill
The objective of the CQA/CQC for backfill mixing is to ensure that the mixed backfill meets the
approved design for backfill mix before it is introduced into the trench. Typical specifications
require:
Backfill Mixing Method and Equipment - The method and equipment used should
achieve thorough mixing of the backfill materials into a relatively homogeneous
mass that will meet the gradation and consistency requirements identified for the
selected mix design. Typical methods and equipment for soil-bentonite trenches are
(1) mixing along the trench with the tracks of a bulldozer, (2) remote mixing with a
bulldozer on a pad, or (3) mixing in a plant such as pugmill. Past experience has
shown that well-controlled mixing along the trench, which is the most economical
method for most sites, yields a satisfactory backfill. In general, the use of remote
mixing depends upon considerations related to the particular site, such as lack of
sufficient space adjacent to the trenches.
Gradation Tests - Gradation tests are used to determine that the backfill, in
particular the content of fines, meets the specified gradation requirements. Such tests
should be conducted once for every 400 to 600 cubic yards, or once per shift, if
production is lower.
Dry Bentonite Content - The mix design may specify' the addition of dry bentonite
to the bentonite slurry. If so, the contractor should ensure that the percentage of dry
bentonite required is added to the backfill.
Consistency of the Backfill - Consistency of backfill is verified by slump testing.
If the slump is too low, typically lower than 3 to 4 inches, the backfilling process
may be impaired. That is, the backfill mixture is likely to contain pockets of slurry.
If the slump is too high, higher than 6 to 7 inches, the backfill may not displace the
slurry completely. Moreover, the backfill slope will be excessive (more than 12
horizontal to 1 vertical) and the trenches therefore longer than necessary, increasing
the risk of construction difficulties (such as sloughing and sedimentation). In
addition, high slump increases the potential for excessive consolidation of the
backfill over the long term. Backfill slumps also should be measured once for every
400 to 600 cubic yards, or once per shift, if production is lower.
40
-------
3.3.9 Permeability of the Backfill
The design permeability of a barrier can vary greatly, depending on the type of barrier and the
design objective. Generally, 1 x 10~7 cm/sec ± is an industry-accepted achievable permeability
for soil-bentonite barriers. Permeabilities of 1 x 10~6 cm/sec ± generally are accepted for cement-
bentonite barriers of various types, such as soil-cement-bentonite and cement-bentonite. Grout
barriers have permeabilities of approximately 5 x 10~b cm/sec. Sampling, type of test conducted,
and testing parameters can influence permeability values significantly.
Standard specifications require that an independent approved laboratory perform testing of
backfill permeability. The tests should be run in a flexible-wall permeameter. Typically, the
sample first will be prepared under a consolidation pressure equivalent to half the depth of the
barrier. The frequency of the tests varies according to the project; however, for this analysis, a
test once for every 400 to 600 cubic yards was considered standard.
Note: It takes approximately one week or longer from the time of sampling to obtain the results
of a flexible-wall permeability test. For that reason, a few contractors conduct daily on-
site permeability tests in a fixed-wall permeameter (filter press). That approach was not
used as a rating criterion for this study. However, the project team recommends the
practice, even if such tests are less accurate than laboratory! tests, because its
application allows the detection of deficient backfill within a few hours, rather than a
week.
If all the above tests on the mixed backfill were performed once for every 400 to 600 cubic
yards, the site was rated acceptable. If the tests were performed less frequently, the site was
rated less than acceptable. If the tests were performed more frequently, the site was rated better
than acceptable.
At the sites studied, backfill gradation was tested once for every 400 to 600 cubic yards unless
the backfill borrow material was obtained from a relatively uniform source. In such a case,
testing was less frequent. The backfill slump at most of the sites studied was tested once for
every 400 to 600 cubic yards and varied from 3 to 6 inches. Testing of the backfill permeability
at the sites varied from once ever}' 250 cubic yards to once every 600 cubic yards.
3.3.10 Placement of the Backfill
Control of the placement of the backfill in the trench is an important component of successful
barrier construction.
First, the bottom of the trench should be sounded and approved by the engineer before the
backfill is placed. Once the initial slope of the backfill has been established appropriately, the
mixed backfill should be pushed on top of the backfill previously placed on the top of the trench.
Free-dropping of the backfill through the slurry should not occur. The slope of the backfill
should be measured at least once per shift and, if the backfill operation was stopped for more
than 24 hours, at a minimum, the slope should be sounded prior to backfill placement for
potential sedimentation on its surface.
If the mixing was controlled loosely and the backfill placed in the trench, the site was rated less
than acceptable. If the mixing was controlled at a central location and the backfill placed in a
manner that prevented segregation, the site was rated better than acceptable.
41
-------
Central mixing was used in completing 80 percent of the soil-bentonite barriers.
3.3.11 Confirmation of Capping
The surface of the barrier wall is capped to protect against physical damage, to limit desiccation,
and to accommodate settlement of backfill materials. If the cap was tested to verify placement
and compaction and the design specifications were met, the site was rated acceptable. If the cap
was not placed on the wall surface in a controlled fashion, the site was rated less than acceptable.
If the cap was placed under stringent CQA/CQC requirements, the site was rated better than
acceptable.
For all sites studied, the cap was placed on the surface of the barrier wall in accordance with the
design specification.
3.3.12 Barrier Continuity
The barrier wall should be constructed in a continuous sequence, with joints, seams, or
connections minimized, thereby minimizing possible areas of preferential flow. If the
construction sequence had been interrupted, the site was rated less than acceptable. If the
continuity of the wall had been confirmed through the CQA process, the site was rated better
than acceptable.
At all the sites studied except Site 2, continuity of the barrier was ensured through a continuous
construction sequence. At one site, construction started, stopped, and restarted after a 12- to 18-
month delay, creating potential problems.
3.3.13 Postconstruction Sampling and Testing of the Barrier Wall
The quality of the construction of the barrier wall sometimes is confirmed by sampling the
barrier wall at regular intervals and testing the permeability and physical properties of the
retrieved samples. If a limited amount of postconstruction sampling and testing of the barrier
wall was performed, the site was rated acceptable. If no sampling and testing was performed, the
site was rated less than acceptable. If the sampling and testing was done at approximately 100-
foot intervals along the length and depth of the barrier wall, the site was rated better than
acceptable.
Postconstruction sampling and testing of the barrier varied widely from site to site. At Site 11,
confirmation testing was done every 100 ft for the soil-bentonite wall and every 50 ft for the soil-
cement-bentonite wall, while at Site 1, samples were tested at 500-foot intervals. At Site 23,
17 dissipation tests were performed in addition to the permeability tests on undisturbed samples.
Section 3.4 discusses these differences further.
3.3.14 As-Built Records
If the barrier wall construction was documented in a construction completion report, the site was
rated acceptable. If no construction completion report was prepared, the site was rated less than
acceptable. If the report included drawings and results of analysis of samples and of tests from
both the construction and the postconstruction periods, the site was rated better than acceptable.
For all the sites studied, as-built records documented the construction effort.
42
-------
3.3.15 Monitoring of Groundwater Head
Ground-water heads within and outside the barrier wall should be measured periodically during
and immediately after construction to determine the groundwater response outside the barrier and
groundwater rise within the barrier. Rapid changes in water level can affect the stability and
constructability of the trench. If the heads were monitored to determine the head fluctuation
during and immediately after construction, the site was rated acceptable. If the heads were not
monitored, the site was rated less than acceptable. If the heads were monitored regularly in
paired wells within and outside the barrier and trends analyzed, the site was rated better than
acceptable.
At all the sites studied, the groundwater head was monitored after construction.
3.3.16 Final Survey of Barrier Alignment
The final survey of barrier alignment should be documented because some changes are likely to
occur during construction. If the alignment survey was completed, the site was rated acceptable.
If no survey was completed, the site was rated less than acceptable. If the survey was completed
and monuments were set in place to identify the barrier alignment and facilitate future surveys,
the site was rated better than acceptable.
3.3.17 Construction Specifications for the Barrier
The construction specifications the designer prepares for the barrier can be based on design or
performance, or on a combination of the two. If the specifications were based primarily on
design, the site was rated acceptable. If the specifications were based primarily on performance,
with little specificity, the site was rated less than acceptable. If the specifications were based on
design and detailed, and if they addressed barrier testing and ancillary construction, the site was
rated better than acceptable.
For most of the sites studied, the construction specifications were based on design. Because the
effort at Site 5 was an emergency action, the contractor designed the wall for the site; the
specifications were based on performance.
3.3.18 CQA/CQC Program and Testing Specification
If the contractor maintained a CQC program and an independent consultant provided QA of the
contractor's CQC and performed some additional QA tests, the site was considered acceptable.
If no formal CQA program was used and limited CQC was performed, the site was considered
less than acceptable. The site was considered better than acceptable if the designer specified
CQC and independent CQA and CQC programs were followed strictly, with at least one-third
duplicate CQA testing.
3.3.19 Monitoring of Groundwater Chemistry
The groundwater chemistry downgradient of the site is monitored after construction to determine
whether the wall is preventing the flow of groundwater across the barrier wall. If some
monitoring relative to the constituents of concern was performed downgradient of the wall, the
site was rated acceptable. If such monitoring was not performed, the site was rated less than
43
-------
acceptable. If it was monitored periodically and in paired wells located inside and outside the
barrier wall, the site was rated better than acceptable.
At all the sites studied except the two civil cutoff walls (Sites 34 and 35), the groundwater
chemistry was monitored downgradient of the site. Section 3.4 discusses groundwater
monitoring further.
3.3.20 Weighting of the Elements
The 26 elements described above are of different degrees of importance in the evaluation of
barrier wall CQA/CQC. The elements were weighted to reflect their importance, as determined
by the project team, with trench key confirmation given the greatest weight, 15 points of a total
of 116 points. The next 4 elements, with weights of 10 points each, are use of an experienced
specialty contractor, trench width and verticality, trench sounding, and construction
specifications for the barrier. The next 6 elements, with a weight of 5 points each, are trench
excavation methods, cleaning of the trench bottom, confirmation of capping, postconstruction
sampling and testing of the barrier, as-built records, and CQA/CQC program and testing
specification. The next 3 elements, with a weight of 3 points each, are gradation testing of the
backfill, permeability testing of the backfill, and target permeability of the backfill. The next 10
elements, with a weight of 2 points each, are mixing of the slurry, viscosity testing of the slurry,
viscosity of the slurry, sand content testing of the slurry, sand content of the slurry, slump testing
of the backfill, slump of the backfill, mixing and placement of the backfill, continuity of the
barrier, and head monitoring of the groundwater. The last 2 elements, with a weight of 1 point
each, are final alignment survey of the barrier and monitoring of groundwater chemistry.
For each site, each rating of less man acceptable (1), acceptable (2), or better than acceptable (3)
was multiplied by the weight for the CQA/CQC element and totaled. The ratings then were
normalized by dividing by 116. If the result was lower than 1.8, the site CQA/CQC was rated
less than acceptable; if the result was between 1.8 and 2.2, it was rated acceptable. If the result
was higher than 2.2, it was rated better than acceptable. Table 3-5 shows the ratings for the sites.
3.3.21 Range of Findings
CQA/CQC for the barrier wall differed widely among the sites studied. For 24 sites, CQA/CQC
was rated better than acceptable. For 2 sites, CQA/CQC was rated less than acceptable; for 7
sites, CQA/CQC was rated acceptable.
The CQA/CQC methods used at each site are described in the individual site summaries (see
Appendix B, Volume II). For most sites, a majority of the CQA/CQC elements were applied,
with sites that were rated better than acceptable completing and documenting the
44
-------
Table 3-5
Containment Barrier CQA/CQC Matrix
Site
1
2
3
4
5
6
7
8
9
10
11
Trench
Sounding
Yes
Every 10 feet
Yes
Conducted
every 10 to 20
feet
N/A
No
Daily
N/A
Constant
trench- side
inspections of
trench
Every 10 feet
Daily-
Key
Confirmation
Visually
Visual
Visually
Confirmed every 20
feet
Visually
Visually
Visually
Visual observation
of driving of sheet
piles
Constant trench-side
inspections of key
depth
Sampling every 20
feet
Thorough
Slurry
Testing
Yes
Yes
Yes
USAGE performed
independent tests
on grout and grout
curtain
N/A
Daily-
Yes
N/A
NA
Once to twice per
shift
NA
Backfill
Testing
Yes
Yes
Yes
Yes
Thorough testing of clay-
backfill
Daily-
Yes
N/A
Backfill was mixed in
centrally located, mobile
pubmill to optimize
characteristics ofbackfill
Once per 300 cu yd
Backfill was mixed in
centrally located pugmill;
test of hydraulic conductivity
ofbackfill were conducted
Backfill
Permeability
Met requirement of 1 x 10"
cm/sec
6.3 x 10"8 cm/sec;
target was achieved
1 x 10"7 cm/sec
1 x 10" cm/sec
Wall met permeability
requirement of 1 x 10"7
cm/sec
1 x 10"' cm/sec
1 x 10"8 cm/sec to 9 x 10"'
cm/sec
N/A
NA
1 x 10" cm/sec
Wall met permeability
requirement of 1 x 10"7
cm/sec
Post-Construction
Testing
Samples collected every
500 If
Annual geophysical survey
NA
None
Undisturbed samples every
400 feet
Samples taken to measure
permeability
Post construction borings
Permeability certification
test performed
NA
NA
Confirmation testing
Rating
Less than Acceptable
Acceptable
Acceptable
Better than Acceptable
Better than Acceptable
Better than Acceptable
Better than Acceptable
No rating criteria
Acceptable
Better than Acceptable
Better than Acceptable
45
-------
Table 3-5
Containment Barrier CQA/CQC Matrix (Continued)
Site
12
13
14
15
16
17
18
19
20
Trench
Sounding
Wall continuity
and depth were
demonstrated
by putting a
steel bar
horizontally
through the
entire wall
Thorough
Every 10 feet
Adequate
Adequate
Slurry wall
backfill profile
was measured
twice daily to
verify that the
trench had not
caved in
Every 10 feet
Adequate
Thorough
Key
Confirmation
Thorough
Thorough
Thorough
Adequate
Adequate
Insufficient during
trench excavation
and backfilling,
insitu soil mixing
required to complete
the key in
Thorough
Thorough
Slurry
Testing
Slurry density and
viscosity were
measured every 2
hours during well
construction
Yes
Yes
Yes
Yes
4 tests per day
2 times per shift
Yes
Before and after
trench placement
Backfill
Testing
Thorough
Very elaborate mix design
and QC program
Yes
4 standard-sized bags of
bentonite and 1 1 bags of
cement for each 3 cu. yds. of
backfill
3 tests per day for slump and
density, 3 tests per 400 If for
permeability
28 backfill samples tested
after wall completion
Thorough
Yes - for each batch
Backfill
Permeability
Hydraulic conductivity was
determined eveiy 200 feet
1 x 10"7 cm/sec
4 x 10"7 cm/ sec
3.4 xlO"8 cm/sec
1 x 10'6 cm/sec
All samples had
permeabilities equal to or
less than 1 x 10"7 cm/sec.
1 x 10" cm/sec
requirement met
NA
1 x 10"7 cm/sec
Post-Construction
Testing
None
None
Instrumentation inside and
outside barrier
50 permeability tests
performed on undisturbed
wall samples
None
Samples of the wall taken
for permeability testing
Samples taken every 100
feet along wall.
Samples taken along wall.
Piezocone testing performed
for window detection, pump
testing across barrier, and
geophysical survey along
barrier.
Hydraulic testing
Rating
Better than Acceptable
Better than Acceptable
Belter than Acceptable
Better than Acceptable
Less than Acceptable
Better than Acceptable
Better than Acceptable
Better than Acceptable
Better than Acceptable
46
-------
Table 3-5
Containment Barrier CQA/CQC Matrix (Continued)
Site
21
22
23
24
25
26
27
28
29
30
31
32
33
Trench
Sounding
Auger depths
measured
Daily
Daily
Daily
Adequate
Depth sounding
every 10 feet
Adequate
Yes
Average
Yes
Every 15 to 20
feet
NA
Depth sounding
every 25 feet
Key
Confirmation
Visually
Visually
Visually
Visually
N/A-hanging wall
Yes
Confirmation
sampling every 25
feet
Yes
Yes
Yes
Visual analysis of
trench bottom
samples
NA
Continuous visual
inspection of
cuttings to ensure at
least 2 feet of
penetration into
aquitard
Slurry
Testing
Daily
Yes; 18 hour slum'
mixing period
Yes
Tested for density.
Marsh viscosity,
and filtrate loss
Twice per shift
Yes
Thorough
Yes
Yes
Yes
Twice per shift
NA
Daily testing
Backfill
Testing
Daily
Yes
Yes
Tested for liquid limit,
gradation, and permeability
Slump testing and slope
testing performed on every
250 to 300 cu. yd. of backfill
Yes
Yes
Yes
Slump testing average
Extensive testing
Once eveiy 400 to 600 cy
NA
Daily testing
Backfill
Permeability
1 x 10"6 cm/sec
1 x 10"' cm/sec
1 x 10"7 cm/sec;
requirement met
NA
3 backfill samples tested
using flexible wall
permeameter; all met
1 x 10"' cm/see requirement
1 x 10"7 cm/sec
Avg. permeability = 2 x 10"8
cm/ sec; wall met
permeability- requirement
1 x 10"7 cm/sec
1 x 10"' cm/sec
1 x 10"7 cm/sec
Wall met requirement
NA
Required permeability was
<5 x 10"' cm/sec
3 C-B; 3 S-C-B backfill
tested.
Post-Construction
Testing
Samples taken to verify key
None
17 dissipation tests
performed measure
permeability. Undisturbed
barrier samples collected
every 500 feet
Samples were collected from
completed trench for
permeability testing
N/A
No information available
NA
Stress Test
Stress testing
Sampling and testing after
slurry wall completion
Regular and documented
NA
Unconfined compressive
strength
Rating
Better than Acceptable
Better than Acceptable
Better than Acceptable
Acceptable
Acceptable
Not evaluated; no data
available
Better than Acceptable
Better than Acceptable
Acceptable
Better than Acceptable
Better than Acceptable
Not assigned; no data
available
Better than Acceptable
47
-------
Table 3-5
Containment Barrier CQA/CQC Matrix (Continued)
Site
34
35
36
Trench
Sounding
Every 10 feet
Yes; < 10ft.
N/A
Key
Confirmation
Visually plus top
and bottom
sounding
Visual inspection of
cuttings
N/A
Slurry
Testing
2-3 times per shift
Slurry viscosity
density, sand
content prior to
concreting
N/A
Backfill
Testing
Daily testing
Plastic concrete mix was
designed and tested with
materials and site mixing
plant
N/A
Backfill
Permeability
K <5 x 10"7 cm/sec
Permeability test
N/A
Post-Construction
Testing
Settlement plates
Stress cells
NA
N/A
Rating
Better than Acceptable
Better than Acceptable
Acceptable (Cap only)
Note: Acceptable industry practices in Table 3-4 were used to evaluate site CQA/CQC. However, only CQA/CQC parameters are discussed in this table.
N/A = Not Applicable
NA = Not Available
48
-------
postconstruction sampling and testing of the barrier and ensuring continuity of the barrier. For
sites that were rated less than acceptable, postconstruction sampling and testing of the barrier was
limited, and the barrier wall was interrupted during construction because of obstructions along
the alignment. In general, CQA/CQC problems were encountered in the areas of cleaning and
confirmation of the trench key.
The CQA/CQC matrix presented in Table 3-5 illustrates the differences among the sites studied.
At Site 1, which is rated less than acceptable, confirmatory samples were taken from the barrier
wall at intervals of 500 linear feet (If), while at Site 11, which is rated better than acceptable,
confirmatory samples were taken at intervals of 50 If for the soil-cement-bentonite wall and 100
feet for the soil-bentonite wall. Site 16 is rated less than acceptable because construction was
interrupted when a gas line was encountered during excavation, and because postconstruction
sampling was not performed.
Key confirmation was done visually at most sites. Trench sounding was performed daily at all
sites, with measurements at intervals of 10 to 25 feet. The slurry testing and backfill testing was
adequate at all sites studied, with the number of tests varying according to requirements at the
particular site. Evaluation of the CQA/CQC efforts at the sites studied revealed that the
acceptable industry practice detailed in Table 3-4 was followed at most sites and the differences
are related to the requirements of the Sites. Only at Site 16, where the barrier wall was
constructed for an emergency action, was acceptable industry practice not followed.
3.4 PERFORMANCE MONITORING
Performance monitoring as a performance evaluation criteria is a direct measurement of the
system's ability to meet the containment objective. The preceding sections, 3.2 and 3.3,
discussed containment design and containment CQA/CQC issues that contribute to performance.
Monitoring does not contribute directly to performance, but monitoring approaches may
contribute to the perceived performance of the containment. Therefore, the following discussion
evaluates monitoring methods as measuring mechanisms, both in general and relative to the
studied sites. It describes as well variations in the methods and possible pitfalls in monitoring
containment performance.
Performance monitoring focuses on demonstrating the ability of the containment system to meet
the design objective in both the short term and the long term. The objective of the containment
system monitoring is to measure three factors: cutoff of outflow (contamination); cutoff of
inflow (leachate generation); and maintenance entombment. Therefore, monitoring should
measure flow or flux into and out of the containment system and measure long-term degradation
mechanisms that affect the flow or flux, as well. The following discussion examines the
monitoring methods unique to either barriers or caps, as well as those common to both, for the
investigated sites.
Monitoring to assess performance must be viewed in terms of the established performance
criteria for the site. Performance goals varied widely among the sites studied, even within
similar types of containment categories. These various goals are compared in Section 4.0 and
discussed in the individual site summaries listed in Appendix B, Volume II. Equally important
is the period within which the performance objective is reached. Monitoring must consider time
required to attain performance, with long periods of flow and flux equilibration often needed
before the site meets its performance goal. The reader is referred to the generic model Methods
for Evaluating the Attainment of Cleanup Standards (EPA 1992).
49
-------
To aid in consistently assessing performance monitoring of a site, matrices were developed to
compare and rank the site monitoring (see Section 3.4.1). A qualitative ranking was determined
on the basis of available data, and a reference to an industry baseline standard was established.
The baseline standard was based on personal experience, consultation with industry contacts,
published information, and site review. Discussions of monitoring methods, findings from site
evaluations, and several significant monitored sites follow.
Subsurface barriers traditionally have monitoring elements common with those for caps, but also
require unique monitoring. For subsurface barriers, monitoring immediately after construction is
more important because of the effect barriers have on dynamics of groundwater flow and the
process of flow equilibration. The subsurface nature of barriers makes monitoring more difficult
and performance measurement more ambiguous, compared with monitoring of surface barriers or
caps. Evaluation of monitoring at the selected sites revealed that monitoring methods varied.
Groundwater head and quality were monitored routinely, while monitoring to detect long-term
degradation was less prevalent. The following subsections discuss monitoring methods for
vertical barriers and the range of findings among the studied sites.
3.4.1 Typical Industry Practices
The most successful way to ensure adequate performance in the long term is to conduct thorough
design studies and provide competent field quality control during construction (Evans 1993).
However, monitoring is necessary to confirm and document performance. EPA provides general
guidance for monitoring (EPA 1984) and discusses specific applications of conventional and
some less typical CQA and monitoring techniques (EPA 1987b). In previous industry research
(EPA 1987a), a reasonable performance monitoring program was outlined. It included the
following key features:
The installation of observation wells and periodic performance pumping tests to
measure the average as-built hydraulic conductivity
Periodic comparisons of baseline and postconstruction groundwater quality at key
downgradient locations
Monitoring of deformation when the barrier is subjected to significant loading
Subsequent work additionally advocated the use of noninvasive or nondestructive testing and
predictive mathematical modeling for fate and transport of contaminants across barriers (Rumer
and Ryan 1995). Finally, the range of applicable monitoring methods and a discussion of current
technology are presented below and summarized in proceedings of the first International
Containment Workshop (Rumer and Mitchell 1996).
Table 3-6 describes the monitoring categories for vertical barriers evaluated as part of this study.
The established industry baseline standard also is defined. Individual monitoring categories are
described further below.
50
-------
TABLE 3-6
EVALUATION OF CONTAINMENT BARRIER MONITORING CATEGORIES
Category
Ground-water Quality
Hydraulic Head
Hydraulic Stress Tests
Physical Samples and
Analysis
Surface Water Quality
Monitoring of
Settlement and Earth
Stress
Inclinometer and
Barrier Movement
Nondestructive
Testing (geophysics
and other)
Environmental
Degradation
Below Industry
Baseline Standard
<3 wells
None
None
None
None
None
None
None
None
Industry Baseline
Standard
at least 3 wells
downgradient
Piezometer pairs across
barrier
Postconstruction at
barrier
At select locations
At select locations
Near structures
At problem slopes
None
None
Above Industry
Baseline Standard
>3
Equally spaced pairs
Several periodic tests
At regular intervals
At several locations
Near structures and
regularly
At problem slopes and
regularly
Some
Some
Groundwater Quality
Groundwater quality is the most widely used method of monitoring containment performance.
This surrogate measurement of barrier performance enables us to estimate the flux leaving the
containment. Groundwater quality typically is measured outside the barrier, with more emphasis
in the downgradient direction.
Performance monitoring through evaluation of groundwater quality should consider the number
of locations, well placement, the screen intervals of monitoring wells, frequency of readings, and
laboratory analysis performed. Typically, the sites studied had more than the minimum number
of monitoring wells located downgradient. For the sites studied, an average of one well per 500
If of barrier (ranging from 1 per 50 ft to 1 per 1,440 ft) had been installed. Spatially, an average
of one well per 7 acres (ranging from 1 per 0.3 acre to 1 per 62 acres) had been installed. At a
few sites, strata below the barrier key were monitored. Most wells were sampled and analyzed
quarterly. Of 20 sites, groundwater quality was measured quarterly at 14, semiannually at 3, and
annually at 3. At 1 site, frequency of monitoring decreased with increasing depth below the zone
of interest, and 2 sites showed a decreased frequency when few elevated levels were detected.
The analytes varied with each site and with the contaminants present. For most sites, indicator
compounds had been chosen for monitoring. In addition, at a few of the sites studied, quality of
groundwater was measured inside the containment, and, in most of those cases, the interior
concentrations decreased overtime.
51
-------
Hydraulic Head
Hydraulic head refers to water level or piezometric readings inside or outside the barrier,
preferably in locations paired across the barrier. Hydraulic head is the second most widely used
monitoring technique for measuring the containment performance of a vertical barrier. The
hydraulic head often is used as a direct measurement of advective inflow or outflow by
measuring head differential across the vertical barrier and the consequent gradient.
Performance monitoring with hydraulic head should consider measurement locations (both cross-
barrier and along the barrier), frequency of measurements, precision of measurements, and the
dynamics of groundwater flow. Head should be measured in correlative strata on opposite sides
of the barrier at multiple locations. Locations along the barrier should be more uniform for an
active (intragradient) containment system or more concentrated downgradient for a passive
system. Hydraulic head also should be monitored in the barrier key or sub-key strata, if there is
concern about the uniform low permeability of the keying interval, or floor, of the containment.
The industry baseline standard was established as paired piezometers as described above,
adjusted for active or passive containment.
For 13 of the 36 sites studied, there are active intragradient containment systems. At all but one
of the 13 sites, paired piezometers were used to measure hydraulic head. Most of the paired
piezometers were spaced equally across the barrier and within 25 feet of the barrier. At most of
the 13 sites, pairs were spaced somewhat equally along the barrier, on average every 750 If (160
If minimum to 2,250 If maximum). Only a few of the passive containment sites had paired
piezometers. Hydraulic head was measured monthly at 5 of the 9 sites for which significant
historical head data were available. For 3 of the sites, head was measured and reported quarterly,
while for 1, head was measured hourly and reported quarterly. The measured values were used to
assess whether cross-barrier head differentials specified in the regulation, typically greater than 1
foot, were achieved. At a few sites, control of vertical gradients also was required, and at 1 site,
maintenance of head below the contained waste material was required.
Hydraulic Stress Tests
Pumping tests provide a means to subject a region in the vicinity of pumping to increased stress
conditions, thereby accentuating flow through imperfections in the barrier. Such imperfections
may be zones of high permeability (windows) or inadequate contact between vertical barrier and
key. A stress test should be run adjacent to the barrier to achieve maximum stress, within
allowable limits. Test monitoring should include monitoring of cross-barrier water level near the
pumping center and along the alignment. Tests can be run either inside or outside the
containment. Disposal of water, potential to induce leaks outside the vertical barrier, and
potential for excessive head differential and hydrofracture during testing should be considered.
Hydraulic stress also can be induced through injection of water or natural stress, such as tidal
influence of groundwater outside the vertical barrier. For sites studied, tests have been run
immediately after barrier construction, as part of the 5-year review of the site, or as a diagnostic
test to evaluate suspected leaks identified by another surrogate monitoring technique. The
industry baseline standard for this monitoring method was established as some postconstruction
pumping tests, as Table 3-6 shows.
Evaluation of the selected sites indicated that some form of active stress test was performed after
construction or as part of postconstruction evaluation tests at only 8 of 36 sites. For 4 of those
sites, tests were conducted at multiple locations with monitoring in several strata, including the
52
-------
key stratum. At 3 of the 4 sites at which the more elaborate tests were used, leaks had been
suspected and were confirmed by the testing. Tests were run either inside or outside the
containment. At a few sites, initial site dewatering was used as a stress-testing alternative. At 2
of the non-waste-containment sites, head differential was from 60 to 100 feet and leaks of less
than 100 gpm were measured to confirm the low permeability of the vertical barrier. At 1 site,
pumping and injection tests and flow simulations were used to prove attainment of required
permeability of the vertical barrier. At 1 site, testing was conducted by conventional falling head
permeability testing through the earthen barrier backfill.
Testing varied significantly among the sites. Logistical issues appear to be the deterrent to stress
testing. However, evaluation pumping tests on sites at which leaks were suspected proved
conclusive.
Physical Samples and Analysis
Physical sampling and laboratory testing of vertical barriers, typically earthen barriers, provide a
direct measurement of the constructed permeability. However, such direct measurement is
affected by the representativeness of the sample, differences of scale between laboratory and
field, disturbance of the sample, and concerns about disturbance of the barrier during sampling.
Some additional complications arise from differences in laboratory procedures for permeability
testing. Despite some disadvantages, sampling and testing often are used. Correspondingly, the
industry baseline standard shown in Table 3-6 includes some physical sampling and testing at
select locations.
Physical sampling and testing should include samples collected at different depths in the earthen
barrier and, at a minimum, from zones at which construction difficulties occurred. Several
drilling and sampling methods have been used to retrieve samples with minimal disturbance
(such as Shelby tubes). Testing should include physical inspection, select physical tests
comparable to previous CQA/CQC testing, and permeability testing.
Postconstruction sampling and testing were performed at 9 of the 36 sites studied. Spacing of
borings averaged every 300 If along the barrier alignment, with a maximum of one boring every
100 If and a minimum of one boring every 500 If. At most sites, conventional flexible wall
permeameters were used to test samples for physical parameters and permeability.
Surf ace-Water Quality
Monitoring of adjacent or downgradient surface-water quality monitoring is sometimes used in
addition to or in lieu of groundwater quality monitoring. Dilution factors make this monitoring
method problematic for assessing the direct performance of a vertical barrier. However.
measurements of surface-water quality and comparison of the results with risk-based
surface-water standards can measure the efficacy of the containment in managing risk to
surface-water receptors. Location and analysis of surface-water monitoring vary significantly
according to the features of the site monitored.
The industry baseline standard was established simply as surface-water quality monitoring at
selected locations, if a body of surface water is near the site.
Of the sites studied, approximately one-third were located adjacent to surface-water bodies;
typically, surface water was measured to determine for containment performance. In just a few
53
-------
cases, data on surface-water quality were submitted to regulatory agencies in place of
corresponding groundwater data. In addition, in a few cases, samples of sediment or biota
supplemented surface-water quality data.
Monitoring of Settlement and Earth Stress
Physical distortion of the earthen barriers can be measured by monitoring settlement, either at the
surface or at depth. Earth stress (modified geostatic stress) in an around the barrier also can be
measured. While these are not direct measurements of containment performance, they can
provide data that indicate potential regions of physical distortion or excessively high stress.
Generally, this type of monitoring is more prevalent in cases in which barriers are located near
structures or high loads. Similarly, this monitoring is performed near barriers subjected to high
hydrostatic stresses, such as those associated with dewatering operations or dam cutoffs.
The industry baseline standard for this monitoring was established as performance of some
monitoring of settlement and earth stress when vertical barriers are located adjacent to structures
or in areas of high induced stresses. At 4 of the sites studied, such monitoring was performed.
Of those sites, 3 were dam cutoffs or dewatering operations
Inclinometer and Barrier Movement
Inclinometers or similar devices can be used, either in the barrier or on the adjacent native
ground, to monitor lateral movement. While monitoring of movement is not a direct measure of
performance, it can measure physical distortion that may affect performance. Concerns about
movement may arise because of adjacent steep slopes or excessive head differentials on the
vertical barrier, contributing to slope instability. Conventional containment barrier installations
cause minimal concern for postconstruction instability unless the installation is near unstable
steep slopes. Therefore, the industry baseline standard was established for application at problem
slopes.
The study of 36 selected sites revealed that lateral movement of the barrier was measured at only
5. Two of those sites were dam cutoffs at which movement was monitored because of high
hydrostatic head differentials across the dam core or vertical barrier. At the remaining sites,
movement was measured because of excessive load from adjacent landfilling, concern for
movement of adjacent railroad lines caused by lateral consolidation of the soil-bentonite backfill,
or steepness of the slope downgradient of a containment barrier having high internal heads.
Concerns about lateral movement of barriers is minimal in conventional containment
installations. However, certain conditions warrant such monitoring, which can be accomplished
with several devices, such as inclinometers or survey bench marks.
Nondestructive Tests (Geophysics andPiezocones)
In addition to destructive sampling, nondestructive geophysical techniques can be used to
monitor the subsurface vertical barrier. Such monitoring methods might provide a means to
measure the integrity of the barrier directly, and perhaps continuously, at costs that appear
reasonable, compared with the cost of intrusive sampling. However, such use has been limited.
Geophysical methods of monitoring vertical barriers include electromagnetic, seismic/acoustic,
magnetic, and gravity techniques. There is growing interest in cross-hole seismic imaging,
surface seismic refraction, ground penetrating radar, ultrasonic and electromagnetic/resistivity
54
-------
surveying (Rumer and Mitchell 1996). Of the 36 sites studied, a geophysical method was used
as a monitoring tool at only 3. At 2 of those sites, resistivity surveying was used, and streaming
self potential was used at the third. At 1 of the two sites at which resistivity surveying was
performed, measurements are collected annually to identify changes in resistivity and suspected
leaks. The results of resistivity surveying have been inconclusive. At the site at which streaming
self potential is being performed, this technique is being used a few years after initial
construction to assess leaking. The information is being correlated with information about
groundwater quality and hydraulic head. Use of self potential methods involves measuring
potentials generated by electrochemical reactions between the native materials and groundwater,
with potential differences between areas of contamination and those at which there is no
contamination. No obvious correlative trends were identified through this effort.
While geophysics may provide intriguing monitoring methods, the data revealed only minimal
successes.
Another form of monitoring used is the piezocone and pore pressure dissipation testing. This test
is intrusive but does not necessarily require extraction of a soil or water sample to evaluate the
integrity of the barrier. The system consists of a cone penetrometer fitted to induce and measure
pore pressure dissipation over time. Results offer information about variations in soil or barrier
backfill and estimates of permeability. This monitoring method was used successfully at 2 sites
after initial barrier construction. It proved useful in defining the location of the barrier,
identifying windows, and measuring permeability in situ.
Environmental Degradation of Barrier Construction Materials
Monitoring environmental degradation of barrier construction materials is not practiced widely.
Detection of degradation processes would allow the introduction of corrective measures or
perhaps lead to preventive design modifications. Degradation mechanisms can include chemical
attack (for example, a high concentration of chlorinated solvents), inhibited bentonite hydration
caused by saline or hard water, desiccation of earthen barriers in a cyclic vadose zone, and
corrosion of metal-sheeted structures. The established industry baseline standard for
postconstruction degradation monitoring is that none is performed. Testing for degradation
would involve some form of direct monitoring.
Often during the design phase, chemical compatibility testing is performed if there is concern
about chemical attack on the vertical barrier, especially in the case of earthen barriers. This
laboratory testing typically involves permeating backfill samples with 3 to 5 pore volumes of
contaminated permeant. Such tests may not simulate adequately in situ, long-term conditions at
the barrier. For approximately half the barriers studied, some compatibility test was performed
in the design phase (see Section 3.2). However, postconstruction analysis of chemical
breakthrough and degradation was reported for only 2 of the 36 sites studied. At no site were
periodic long-term degradation monitoring data collected.
Nonearthen barriers, particularly geomembrane and sheeting, are monitored for degradation
differently than are earthen barriers. At 1 site having steel sheeting containment, conventional
ultrasonic testing was performed after years of operation to determine corrosion. No decreased
performance was noted. Vinyl or plastic sheeting offers obvious corrosion advantages over steel.
Geomembrane vertical barriers have benefited from research on landfill liners, including
extensive laboratory simulations and field efforts that involve exhuming installed geosynthetics.
55
-------
Such tests have indicated lives of hundreds of years for geomembranes buried underground and
not subject to degradation caused by ultraviolet (UV) light.
Historical data that define the effects of long-term attack on vertical barriers are necessary to
better understand the true functional life of such barriers.
3.4.2 Range of Findings
Monitoring of the subsurface barrier varied widely among the sites studied, as described in the
previous section. Monitoring efforts at each site were compared in the various monitoring
categories with the industry baseline standard, as described above. A rating was provided by site
for the monitoring effort as a whole. Measured against the established industry baseline
standards, 11 were considered acceptable, 8 sites were considered less than acceptable, and 16
were considered better than acceptable.
The monitoring methods used at each site are described in some detail in the individual site
summaries in Appendix B, Volume II of this report. Table 3-7 presents a summary of the barrier
monitoring performed at the sites studied.
At most sites, hydraulic head and groundwater quality were monitored. Typically, if surface
water was adjacent to the site, it also was monitored. At most of the sites where active
containment was performed, paired piezometers across the barrier were used to measure head
differential. In almost all cases, the pairs were located within 25 feet of the barrier and, in most
cases, were spaced every 200 to 400 feet along the barrier. At approximately one-third of the
sites, postconstruction confirmation borings and analysis had been performed. Several of those
sites were in New Jersey, where closure or permit regulations require confirmation borings and
testing. At only a few of the sites were pumping tests performed to measure the effectiveness of
the barrier during stressed conditions. In only a few cases were geophysical methods used to
estimate the integrity of the barrier. In no case were tracers used to monitor leaks in the barrier.
Among the 36 sites studied, some noteworthy monitoring efforts were revealed. Those efforts
are outlined below and described in more detail in the site summaries in Appendix B. Volume II.
Site 19 was the most extensively monitored site. Significant postconstruction monitoring and
testing, as well as subsequent performance evaluation and testing, had been conducted there.
Notably, monitoring at the site included piezocone testing to detect windows and pumping tests
at three suspect locations, with multilevel cross-barrier monitoring to measure barrier underflow.
Although Site 23 was monitored adequately during construction, postconstruction monitoring
detected problems in groundwater quality. Borings and piezocone dissipation tests were used to
measure the continuity and integrity of the barrier. Subsequent pumping tests showed interior
head response while the outside groundwater was being pumped. A remedy was put in place to
correct the problem.
56
-------
Table 3-7
Containment Barrier Monitoring Matrix
Site
1
2
3
4
5
6
7
8
9
10
11
12
13
Groundwater Quality
3 outside & 1 inside wells
5 outside (1 up and 4 down
gradient)
Semiannual monitoring
8 pairs of nested wells with
screens in shallow and deep
aquifers
Several wells downgradient
Quarterly monitoring in 6
wells
8 wells outside, and
leachate sampled
100 monitoring wells
6 pairs across barrier;
inside leachate collection
system
21 wells both sides of
barrier
Not yet implemented
5 pairs across barrier
49 wells both sides of
barrier
Hydraulic Head
Not conducted
Not measured
N/A
3 inside & 1 outside are monitored
monthly for water level. Hydraulic
gradient could not be determined
because of deficiencies in data
Paired piezometers
N/A
Head outside only, inside head
monitoring pending;
None
6 pairs across barrier
150 wells measuring local head
Not yet implemented
>7 Alluvial monitoring wells
Paired piezometers
Hydraulic Stress Tests
Not conducted
Not conducted
Not conducted
Not conducted
Not conducted
Not conducted
Not conducted
Pump Test to certify permeability
of sheetpile cutoff
Not conducted
Not conducted
Not yet implemented
Not conducted
Not conducted
Physical Samples
Taken during post-
construction verification
None
Post-construction borings
(100 ft centers)
None
Samples obtained at 400 If
spacing
Yes
None
NA
None
Post-construction
permeability tests
CQA undisturbed samples
and testing per 100 If
None
None
Surface
Water
Quality
2 outside
None
None
None
Several
adjacent
points
No
3 locations
Several
locations
None
None
None
None
Several
locations
OTHER
Settlement, movement,
desiccation, stress, NDT-
geophysics, other
Leachate quality
Annual electrical resistivity
survey outside barrier
None
None
None
Waste was solidified
Leachate quality and quantity is
measured
None
None
None
None
None
None
Rating
Acceptable
Better than
Acceptable
Less than
Acceptable
Better than
Acceptable
Better than
Acceptable
Acceptable
Less than
Acceptable
Less than
Acceptable
Less than
Acceptable
Acceptable
Insufficient Data
Better than
Acceptable
Better than
Acceptable
57
-------
Table 3-7
Containment Barrier Monitoring Matrix (Continued)
Site
14
15
16
17
18
19
20
21
22
23
24
Groundwater Quality
N/A
1 1 outside wells
Several wells up and down
gradient; DNAPL &
dissolved phase measured
Quarterly monitoring of
several wells
Quarterly monitoring at 9
wells, 2 inside and 7
outside
38 single level and 33
multi-level wells
Quarterly monitoring of
wells outside barrier
Quarterly monitoring
13 wells, quarterly-
sampling
15 wells outside
3 outside (1 down and 2 up
gradient) in shallow and 4
in deep zones
Hydraulic Head
43 vibrating wire piezometers
1 5 multiple level standpipe
piezometers
Not conducted
4 pairs across cutoff for
measuring gradient
6 pairs across barrier, and 2 wells
downgradient
4 pairs across barrier
Several pairs across barrier to
measure of zero net gradient
differential
16 pairs across barrier measured
hourly, two levels
4 pairs across barrier
Slight inward gradient
21 pairs across barrier
No piezometer pairs; measured
head in 6 pumping wells
Hydraulic Stress Tests
Permanent high gradient across
hairier
Only during ground water
withdrawal
Not conducted
Not conducted
Not conducted
1 CQA post-construction pump
test; 3 evaluation pump tests with
multi-level cross-barrier
monitoring
Pre- and post-barrier pumping
tests with cross barrier monitoring
Not conducted
Not conducted
3 pump test outside the barrier
with multi-level cross barrier
monitoring
Not conducted
Physical Samples
Visual inspection of key
50 taken during post-
construction verification
None
Post-construction backfill
samples taken
28 samples taken after wall
installation
Post-construction
evaluation samples and
breakthrough analysis
None
Samples obtained to verify
key-
None
Samples collected every
500 If
Samples collected for
permeability testing
Surface
Water
Quality
N/A
None
Several down
gradient
None
None
>3 down
gradient
location
Several
locations
outside the
wall
Yes
4 on site
points
None
Several
locations
OTHER
Settlement, movement,
desiccation, stress, NDT-
geophysics, other
12 inclometers/extensometers in
barrier to control deformation.
Monitoring of settlement and
movement on top of wall
Inclinometers
Streaming self-potential
geophysics for detecting leakage,
unsuccessful
None
None
Electrical resistivity, piezocone
sounding for widow detection
Settlement plate monitoring
None
None
Post construction evaluation with
piezocone for continuity and
dissipation tests for permeability
None
Rating
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Acceptable
Better than
Acceptable
Better than
Acceptable
Better than
Acceptable
Less than
Acceptable
Less than
Acceptable
Better than
Acceptable
Less than
Acceptable
58
-------
Table 3-7
Containment Barrier Monitoring Matrix (Continued)
Site
25
26
27
28
29
30
31
32
33
34
35
36
Groundwater Quality
27 wells monitoring
monthly
54 withdrawal wells, 38
recharge wells; quarterly
monitoring
>5 well outside
Yes
6 paired monitoring wells
Monitored
21 wells and piezometers;
quality no measured if head
gradient is met
16 wells inside and outside
1 up gradient, 4 down in
iron filter, 2 downgradient
of filter
N/A
N/A
N/A
Hydraulic Head
24 paired piezometers
10 equally spaced piezometers on
both sides of barrier
8 pairs across barrier measuring
head differential quarterly
Yes
6 pairs of piezometers
Monitored
8 sets of paired wells across the
barrier
4 pairs monitoring cross-barrier;
interior monitoring for >5 ft
differential below waste
Flow through reactive wall
Electric piezometer inside and
outside of wall
Several paired piezometers across
cutoff
N/A
Hydraulic Stress Tests
Not conducted
Not conducted
Groundwater extraction and cross-
barrier monitoring
Yes
Yes
Not conducted
Not conducted
None, except initial sitewide
drawdown.
N/A
Permanent high gradient across
dam
Head differential
across wall is close to 100 feet
N/A
Physical Samples
None
Post-construction backfill
permeabilities
CQA post-construction
sampling and permeability
test per 250 If
Yes
Yes
CQA undisturbed samples
and testing per 400 If at
various depths
None
None
None
None
CQA testing only
N/A
Surface
Water
Quality
6 stations
sampled
monthly
None
None
None
None
Monitored
forNPDES
River
sediments,
fish
None
None
N/A
N/A
N/A
OTHER
Settlement, movement,
desiccation, stress, NDT-
geophysics, other
None
None
Leachate quality and quantity
Pump test for 1 year
None
None
None
Protective cap
Stress cell, settlement plate, and
inclinometer
Dewatering rates, tieback
loading, deformation monitoring
during construction
N/A
Rating
Acceptable
Acceptable
Better than
Acceptable
Better than
Acceptable
Acceptable
Acceptable
Acceptable
Better than
Acceptable
Acceptable
Better than
Acceptable
Better than
Acceptable
Acceptable (Cap
only)
N/A = Not Applicable
NDT = Non-destructive testing
59
-------
Because of concerns about possible outward flow conditions caused by tidal fluctuations in
surface water and groundwater, paired piezometers at Site 20 were monitored hourly to confirm
head differentials. Exterior groundwater and surface-water quality also were measured monthly.
In contrast to the efforts at the sites mentioned above, minimal to no monitoring was performed
at several sites. Although, at several sites, barrier compatibility was investigated during design
efforts, at only 2 of the sites studied was the barrier reevaluated to assess long-term degradation
of the barrier or contaminant breakthrough. The two studies were prompted by the identification
of other problems related to the barrier. Electrical resistivity surveying of the subsurface barrier
is performed at Site 2 annually in an attempt to measure changes over time in the integrity of the
barrier.
3.5 OPERATION AND MAINTENANCE
O&M of subsurface barriers consisted of scheduled preventive maintenance and repair of
components to ensure the continued effectiveness of the remedial system. Most O&M activities
at the sites studied consisted of operating the pumping systems required to maintain the design
gradient across the barrier wall.
O&M activities for subsurface barriers included monthly site inspections, quarterly readings of
water levels in monitoring wells and water quality sampling, and sampling required under the
facility's National Pollutant Discharge Elimination System (NPDES) permit.
3.5.1 Typical Industry Practice
Because O&M activities are site-specific, it is difficult to establish acceptable practice.
However, the elements of O&M that should be conducted for vertical barriers include:
Monitoring the pumping systems within and outside the barrier wall and adjusting
the treatment system for variations in water quality, either monthly or quarterly
Inspecting the surface of the barrier wall for settlement or erosion, either monthly or
quarterly
3.5.2 Range of Findings
For most of the sites studied, O&M activities were related to the pumping and treating of
contaminated water within the barrier wall. Table 3-8 provides a summary of the O&M
information that was available. O&M activities were not compared with acceptable industry
practice; however, sites at which O&M practices were noteworthy are discussed below.
Site 11 was visited 10 years after the installation of the barrier wall and cap. The investigation
involved extensive head monitoring for horizontal and vertical gradients at the barrier, conduct of
pumping tests at the barrier, advancement of 11 borings to obtain data on the integrity of the
barrier, and evaluation of the groundwater flow in the vicinity of the barrier.
As a result of O&M activities at Site 25, it was determined that the concentrations of
contaminants in groundwater within the wall increased with time, and the treatment system
required modification to accommodate the increased concentrations.
60
-------
Table 3-8
Containment Barrier and Cap O&M and Cost Matrix
Site
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Barrier
O&M
Monthly
inspection
NA
NA
Yes
Regular
inspection
Regular
inspection
NA
Cathodic
protection may
be needed
NA
NA
NA
Periodic visual
inspections
NA
NA
None
NA
NA
Cap
O&M
Surface erosion
control
NA
Monthly inspections
Yes
Monthly inspection
Regular inspection
Inspection of cap
N/A
N7A
N7A
NA
Vegetative cover
maintenance
As needed
NA
NA
N7A
Monthly inspection
Pumping
System
O&M
Quarterly effluent
sampling
NA
N/A
On-site treatment
system
Yes
NA
Maintenance of
leachate
pretreatment
facility
NA
N/A
Information not
available
N/A
As needed
NA
Quart erly effluent
sampling
NA
Monthly inspection
Barrier
Capital
Cost
NA
$444,000111 1981
NA
$6.8 million total for grout
curtain, slurry wall, landfill
cap, and treatment plant
NA
NA
$3,120,885=barrier
($7.81/sq. ft)
NA
NA
$242,000 cost estimate for
cutoff wall extension (650
feet)
$8,900.000
$456,000
$4,500,000
NA
NA
$500.000
NA
Cap
Capital
Cost
NA
NA
NA
Included as lump sum
NA
NA
$17,073,616=cap
$4,154,859=gas collection
system
$'l,833,841=leachate
collection system
N/A
N/A
N/A
NA
$432,000
Included in barrier
NA
NA
N/A
NA
Barrier
O&M
Cost
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
$l,200,000/year
$59,000/year total
NA
NA
NA
NA
NA
Cap
O&M
Cost
NA
NA
NA
NA
NA
NA
NA
N/A
N/A
N/A
NA
Information not available
NA
NA
NA
N/A
NA
61
-------
Table 3-8
Containment Barrier and Cap O&M and Cost Matrix (Continued)
Site
18
19
20
21
22
23
24
25
26
27
28
29
Barrier
O&M
Monthly
inspection
Barrier integrity
evaluated after
10 years
Inspection of
barrier surface
Regular
inspection
Quarterly
monitoring for
three years
Regular
inspection
NA
Regular
inspection and
checking water
levels
No information
available
Monthly site
inspections,
quarterly
effluent'
sampling and
well water level
readings, and
annual well
water sampling
NA
Information not
available
Cap
O&M
Monthly inspection
Monthly inspection
NA
Regular inspection
Monthly inspection
Quarterly water level
readings, monthly
site inspections
Quarterly visual
inspections, annual
benchmark surveys,
monthly mowing of
vegetative cover as
needed
N/A
N/A
NA
N/A
Information not
available
Pumping
System
O&M
$200,000/year
Ground water
treatment:
$1,644,000
wells: $50,000
Regular inspection
Monthly
NA
NA
NA
NA
No information
available
NA
NA
Information not
available
Barrier
Capital
Cost
$5.5 million
$1,000,000 -slurrv wall
($4.65 per sq. ft)
$5,000,000
$3,000.000
$2,550,000
NA
$400,000
NA
No information available
$4,500,000 = total
NA
Information not available
Cap
Capital
Cost
$2.7 million
$436,000=cap
$65,000 = gas vents
NA
$1,300,000
NA
NA
NA
N/A
N/A
NA
N/A
Information not available
Barrier
O&M
Cost
Minimal
NA
$400,000/year
NA
NA
NA
NA
NA
No information available
$30,000/year =total
NA
Information not available
Cap
O&M
Cost
$10,000/year
NA
NA
$10,000/year
NA
NA
NA
N/A
N/A
NA
N/A
Information not available
62
-------
Table 3-8
Containment Barrier and Cap O&M and Cost Matrix (Continued)
Site
30
31
32
33
34
35
36
Barrier
O&M
Yes
Weeklv site
inspection;
monthly barrier
inspection
Periodic
pumping to
maintain water
table level
inside
containment
Monthly water
sampling
Continuous
monitoring of
seepage
N/A
Cap
O&M
Yes
N/A
NA
N/A
N/A
N/A
NA
Pumping
System
O&M
Yes
NA
NA
N/A
N/A
N/A
N/A
Barrier
Capital
Cost
$14.200,000
$2.5 million for construction
and CQA/CQC
No cost information
available
NA
$2 million for installation
only ($6/sq ft)
N/A
Cap
Capital
Cost
Included as lump sum
N/A
NA
N/A
N/A
N/A
NA
Barrier
O&M
Cost
Information not available
NA
NA
NA
N/A
N/A
Cap
O&M
Cost
N/A
N/A
N/A
N/A
N/A
NA
Notes: N/A
NA
Not applicable
Not available
63
-------
3.6 COST
The cost information obtained during the study could not be segregated as design and
construction costs. Completion reports for the remedial actions provided lump-sum costs for the
barrier wall, cap, and groundwater treatment system. In some reports, O&M costs were provided
for the barrier wall, cap, and treatment plant. Details are provided in the individual site
summaries in Appendix B, Volume II.
The costs of installing subsurface barriers vary widely, depending on the unique conditions at
each site (see Table 3-8). Costs also vary because of the construction method and the type of
barrier wall. In general, use of the vibrating beam method of barrier wall construction resulted in
the lowest cost, and use of the trench excavation method resulted in the highest cost. However, it
should be emphasized that the vibrating beam method produces a thinner wall than the trench
excavation method. Soil-bentonite walls cost less than cement-bentonite walls. If off-site
borrow clays are used in soil-bentonite walls, the cost is higher than the cost would be if a
bentonite slurry were sufficient.
The capital cost for construction of a barrier wall is expressed in dollars per square foot of
sidewall area. Such costs vary from $5 to $15 per square foot. The O&M cost is usually very
small, unless the design includes a treatment plant. The O&M cost will depend on the
monitoring requirements specified by the state regulatory agency.
4.0 DATA COLLECTION AND ANALYSIS - CAPS
Data collection and analysis for caps was done in conjunction with mat for barriers. The site
summaries in Appendix B provide information for the barrier and cap systems; site 36 was a cap
only site. This section describes the present design, CQA/CQC, performance monitoring, O&M,
and cost of the caps evaluated.
4.1 DESIGN
The design of a cap defines the functional criteria performance objectives of the cap. The design
documents provide the design calculations, detailed drawings and specifications, CQA and CQC
requirements during cap construction, O&M requirements, a performance monitoring plan, as
well as construction costs and schedule. Therefore, understanding the design objective and
analyzing design documents are important for evaluating cap performance.
Guidance has been provided by USEPA (1989, 1993) and USAGE (1993) for the proper design
of caps. This subsection will define the design components of the cap and discuss the matrix that
was used to evaluate the design of caps.
The primary objective of capping is to minimize infiltration into the waste materials and isolate
waste from human contact. At sites for which the cap is a component of the containment system,
the cap typically is interconnected with the vertical barrier wall. The design of the cap should
consider the following factors, which are described in this section.
Stability of the waste
Settlement
Stability of the cap system
64
-------
Drainage
Infiltration
Gas management
The project team identified the factors listed above and subfactors associated with them and
defined the typical industry practice for each factor and subfactor through discussions with
engineers and contractors, the project team's experience, and review of published literature.
Table 4-1 presents a matrix that summarizes standard industry practice. The matrix was used to
evaluate the 36 sites selected for this study to determine whether the design effort was
acceptable, less than acceptable, or better than acceptable. The range of findings from that
evaluation is discussed in Subsection 4.2.5.
4.1.1 Stability of the Waste
The stability of the waste mass must be analyzed to determine whether it will remain stable under
all potential loading conditions. Several analytical and numerical techniques can be used to
analyze global stability of the waste mass. Several loading conditions must be analyzed to
determine the stability of a waste mass. They are:
Cap loads - A cap can vary in thickness from 3 to 5 feet. The effect of that loading
on stability must be analyzed by accepted slope stability methods.
Seismic stability - The National Oceanic and Atmospheric Administration (NOAA)
publishes the earthquake database for all parts of the United States. The seismic
loading recommended by the database should be used to conduct psuedostatic
seismic stability analysis of the waste mass. EPA (1995) should be used for
conducting seismic stability analysis.
Construction loading - During construction of the cap, loads are applied by
stockpiles of soil and construction equipment. The worst-case short-term loading
conditions should be analyzed.
If the loading conditions described above had not been analyzed and the reliability of the data
had not been verified before the stability analysis was performed, the site was rated less than
acceptable. If stability was analyzed and the results incorporated into the design, the site was
rated acceptable. On the other hand, if the reliability of the data was verified during the analysis
and a sensitivity analysis performed, the site was rated better than acceptable.
4.1.2 Settlement Analysis
The foundation and the waste materials may consolidate under the loading conditions described
in Subsection 4.1.1, with settlement of the waste mass differing accordingly. Because such
differential settlement can have adverse effects on the cap, it must be analyzed. The analyses to
be completed are:
65
-------
TABLE 4-1
MATRIX FOR EVALUATING CAP DESIGN AGAINST
ACCEPTABLE INDUSTRY PRACTICES
Design Items
Less Than Acceptable
Acceptable
Better Than
Acceptable
1. Global Waste Stability
General
Under cap loads
Seismic stability
Construction loading
None performed
Not performed
Not performed
Check reliability of input data
Computer modeling done
NOAA data used and pseudostatic
analysis completed.
Consider worst case short term
loading conditions during
construction, apply equipment and
stockpile loads
Sensitivity analysis
on waste properties
Sensitivity analysis
Sensitivity analysis
Sensitivity analysis
2. Settlement Analysis
Refuse materials
Foundation materials
Impact of differential settlement
considered
Not considered
Not considered
Not considered
Representative data used for
analysis/estimate
Traditional soil mechanics analysis
completed
Impact on cover materials
analyzed
Settlement plates
included in
monitoring plan
3. Stability of Cap System
Interface stability analysis
Cover soil tension above geomembrane
lined slope
Stresses within cap components
Impact of differential settlement on
geosynthetics
Stability of slopes steeper than 3:1
No analysis performed or
analysis performed using
assumed data
No analysis performed
No analysis performed
No analysis performed
Not considered
Analysis performed using data
generated from laboratory testing
of site specific geosynthetics and
soils
Analysis performed to determine
need for tension reinforcement
using appropriate laboratory
generated data
Analysis performed using cap
loads and construction loads
Calculate stresses in geosynthetics
due to subsidence
Slope protection designed
Analysis performed
with additional
sensitivity7 checks
performed
66
-------
TABLE 4-1
MATRIX FOR EVALUATING CAP DESIGN AGAINST
ACCEPTABLE INDUSTRY PRACTICES (Continued)
Design Items
Less Than Acceptable
Acceptable
Better Than
Acceptable
4. Drainage Analysis
Drainage capacity of cover system
evaluated (transmissivity)
Geotextile filtration
Runoff control
Erosion potential
No analysis performed
No analysis performed
No analysis performed
No analysis performed
Evaluate drainage requirements of
cap system using models such as
HELP and size drainage medium
accordingly
Evaluate cover soil compatibility
with separation geotextile to
prevent clogging and encourage
filtration
Evaluate piping and discharge
structure requirement for
appropriate peak storm
Universal soil loss equation used
5. Leachate Management
Leachate generation rate
Collection/treatment system
Not considered
Not considered
Analysis completed using site-
specific data
Collection/treatment system
designed and documented
6. Gas Management
Well design/placement
Passive system design
Not considered
Not considered
Pilot data from site used as design
basis
Lateral piping and vertical vents
designed
7. Miscellaneous/Other Items
Frost depth
Puncture vulnerability
Not considered
Not considered
Soil barrier layer placed beneath
maximum frost depth
Potential for waste material and
cover soil to puncture cap system
evaluated
67
-------
Waste materials - The variability of the waste materials must be considered, and
accepted analytical techniques should be used to estimate the long-term settlement of
waste material.
Foundation materials - The properties of the foundation materials should be
determined, and traditional soil mechanics analysis should be completed to estimate
settlement of the foundation.
Effects of differential settlement - The effects of differential settlement on the cap
materials, especially geomembranes, should be analyzed.
If the analyses described above were not performed, the site was considered less than acceptable.
If the analyses were performed, the site was considered acceptable. If the long-term settlement
was analyzed, sensitivity analysis completed, and the results accommodated by the design, the
site was considered better than acceptable.
4.1.3 Stability of the Cap System
The stability of the cap system, which consists of different types of soils and geosynthetic
materials, must be analyzed. Any failure of the components of the cap system can affect
performance adversely. The analyses to be completed to determine the stability of the cap
system are:
Interface stability - The critical interfaces within a cap system are those of the soil
and geosynthetic clay liner (GCL); the GCL and geomembranes; and the
geomembrane, drainage layer, and cover soils. Stability along these interfaces
should be analyzed using soil from the site, and reliable data should be obtained
from independent laboratories that test geosynthetic materials.
Cover soil tension above geosynthetic-clay lined slopes - The cover soil tension
above the geosynthetic-clay lined slope should be analyzed to determine the need for
tension reinforcement. This analysis should use data generated by an independent
laboratory.
Stresses within components of the cap - The stresses within the components of the
cap should be analyzed through the use of cap loads and construction loads to
determine whether the cap system will remain stable.
Effect of differential settlement on geosynthetics - The differential settlement
caused by subsidence of the waste materials can induce stresses in the geosynthetic
materials. The properties of the geosynthetics selected should be appropriate for
those induced stresses.
Stability of slopes - Stability of all slopes should be analyzed; however if slopes
steeper than 3 horizontal to 1 vertical are required, they should be analyzed to
determine whether slope protection measures are needed.
68
-------
If the analyses described above were not performed, the site was considered less than acceptable.
If the analysis were performed, the site was rated acceptable. If the analyses and additional
sensitivity analyses were performed, the site was rated better than acceptable.
4.1.4 Drainage Analysis
Drainage of storm water at the site is important to prevent the buildup of hydraulic head on the
low-permeability impermeable cap. The design of the drainage system should be subjected to the
following analyses:
Drainage capacity analysis - The Hydrologic Evaluation of Landfill Performance
(HELP) model is used widely to determine the drainage capacity of the cap. The
results obtained from the HELP model are used in designing the drainage layer that
overlies the low-permeability cap.
Geotextile filtration - A geotextile is placed over the drainage layer so that the cover
soil does not clog the drainage layer. The compatibility of the cover soil with this
separation geotextile should be analyzed to ensure the maximum efficiency of the
drainage layer.
Runoff control - The 25-year, 24-hour peak storm typically is used to determine the
amount of runoff from the site. Piping and discharge structures should be adequate
to handle the storm.
Erosion control - The universal soil loss equation is used to analyze the amount of
soil erosion at the site. This analysis should be completed to support the design of
erosion control measures.
If the analyses described above were not performed, the site was rated less than acceptable. If the
analyses were performed, the site was rated acceptable. If the analyses were conducted and a
sensitivity analysis of the various input parameters used was performed, the site was rated better
than acceptable.
4.1.5 Infiltration
The prevention of leaching at a hazardous waste site is important to the success of remediation.
The infiltration design of a cap system should be evaluated according to the following
considerations.
The water balance model predicts water percolation through the cap. Several computer models
are available to evaluate the hydraulic performance of the landfill cap. The HELP model
mentioned in Section 4.1.4 is used to estimate the amount of infiltration into the waste. The
effect of compacted clay or the GCL should be analyzed to predict long-term infiltration through
the cap. In addition, any improvement of the cap by use of a geomembrane in addition to the
clay or GCL should be analyzed.
If the HELP analysis did not use the representative values for the different components of the
cap, the site was rated less than acceptable. If the HELP analysis was completed properly, the
69
-------
site was rated acceptable. If the HELP analysis included sensitivity analysis for the various
components of the cap. the site was rated better than acceptable.
4.1.6 Gas Management
Decomposition of waste materials may produce methane and other gases. The gases must be
collected and vented to the atmosphere or treated, depending on the concentration of
contaminants present. The design analyses to be completed are:
Design and placement of wells - A pilot test usually is conducted to determine the
quality of gas from the waste and the rate at which it is generated. Design of the
depth and spacing of the wells and gas treatment are based on the resultant data.
Design of the passive system - If a passive venting system is required, the locations
of lateral piping and vertical vents are designed as the waste characteristics indicate
is necessary.
If generation or management of gas was not considered adequately, the site was rated less than
acceptable. If the analyses used input data that were not site-specific, the design was considered
acceptable. If site-specific input data were used in performing the design analyses described
above, the site was considered better than acceptable.
4.1.7 Other Factors
Other factors important to the successful performance of the cap are frost penetration of the cover
soils and vulnerability of the geosynthetic materials to puncture. The Soil Conservation Service
of the U.S. Department of Agriculture publishes information on the depth of frost in different
parts of the country. If that factor was not accounted for in the design, the site was considered
less than acceptable. Waste material, construction equipment, and cover soils have the potential
to puncture such geosynthetic materials as geotextiles and geomembranes. If these factors were
not considered in the design, the site was rated less than acceptable.
4.1.8 Ran ge of Fin din gs
Of the 36 sites, 22 have caps. Cap design varied little among the sites, and most sites met the
design requirements set forth under RCRA Subtitle C. The major differences among the designs
were in documentation of the settlement analysis, seismic stability analysis, and interface
stability analysis. Infiltration analyses and gas management design was handled as required for
the site.
At most of the sites, geosynthetic materials were used in the cap. Sites 1, 7. 27, and 30 had
compacted clay liners. Sites 28 and 33 were covered with paved parking lots. For most of the
sites, the design was rated acceptable, incorporating the following design elements:
Geotechnical testing of the materials used for the cap
Analysis of global stability of the waste, based on the properties of materials used at
the particular site
Analysis of potential for settlement
Analysis of the stability of the cap system
70
-------
Analysis of drainage and infiltration through application of the HELP model
Adequate design of the gas management system
Protection of the cap against frost penetration
Site 27 was rated less than acceptable because no analyses of seismic stability and interface
stability were performed, and no leachate management system was designed. At Site 18, a test
section was incorporated into the design to determine the long-term behavior of the geosynthetic
clay liner.
Other variations among caps were noted, usually in response to conditions at the particular site.
Site 15 had a soil cover only. At Site 1, the side slopes were covered with clay only and the top
of the landfill area with geomembrane. At Site 36, a capillary break layer was incorporated in
the cap section to control capillary rise. The design for Site 36 was a "consumptive use" cap
consisting of specified soil types and vegetation to promote consumptive use of rainfall and limit
infiltration.
At most sites, the cap design provided an interconnection with the barrier wall to form a
containment system. In addition, erosion control features were designed to protect the cover
soils.
4.2 CONSTRUCTION QUALITY ASSURANCE/CONSTRUCTION QUALITY
CONTROL DATA
On January 29, 1992, EPA issued final regulations for CQA/CQC in 40 Code of Federal
Regulations (CFR) Part 264.19 (57FR3486). Those regulations apply to the design and
construction of surface impounded waste piles and landfills, including the construction of caps.
However, the state of the art of cap installation has changed since 1992, with geosynthetic
materials replacing natural clays and drainage soils. For CQA/CQC, the following factors, which
are described in this section, should be considered:
Borrow soil
Compacted clay liner
Soil drainage layer
Geomembrane barrier
Development of a matrix that describes the typical industry CQA/CQC practices for caps was
based on the experience of team members, review of published literature, and discussions with
academic and industry practitioners; Table 4-2 presents that matrix. The matrix was used to
evaluate the 36 sites selected for the study to determine whether CQA/CQC activities at each site
were less than acceptable, acceptable, or better than acceptable. The range of findings for the
sites is discussed in Subsection 3.3.21.
EPA has funded several research projects and published reports that summarize the state of the
art for installing caps (EPA 1988, 1989, and 1993). The National Sanitation Foundation (NSF)
has published standards for the manufacture of flexible geomembranes, and the American
Society of Testing and Materials (ASTM) has developed several testing methods that are used
widely during the installation of caps. These standards are discussed in the following
subsections.
71
-------
TABLE 4-2
MATRIX FOR EVALUATING CAP CQA/CQC AGAINST ACCEPTABLE INDUSTRY PRACTICES
Category
Less than Acceptable
Acceptable
Better Than Acceptable
BORROW SOIL (Subgrade & Cover Soil)
Soil Prequalification Testing:
Classification ASTM D2487
Compaction Curve ASTM D698/D1557
Soil Construction Testing:
Density Testing/Lift ASTM D2922/1556
Moisture Content Testing/Lift ASTM
D30 17/22 16
Not performed
Not performed
Not performed
Not performed
1 test per 5, 000-6, 580 c.y.
1 test per 5, 000-6, 580 c.y.
5 per acre
5 per acre
More frequent
More frequent
More frequent
More frequent
COMPACTED CLAY LINER
Clay Prequalification Testing:
Classification ASTM D2487
Compaction Curve ASTM D698/D1557
Remolded Permeability Test ASTM
D5084
Clay Test Pad
Construction/Testing/Evaluation
Clay Construction Testing:
Compaction Curve ASTM D698/D1557
Density Testing/Lift ASTM D2922/1556
Moisture Content Testing/Lift ASTM
D30 17/22 16
Undisturbed hydraulic conductivity
ASTM D5084
Atterberg Limits ASTM D4318
Grain Size ASTM D422
Undisturbed Dry Density
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
1 test per 5, 000-6, 580 c.y.
1 test per 5, 000 c.y.
1 test per 10, 000-13, 160 c.y.
Not performed
1 test per 5, 000 c.y.
5 per acre
5 per acre
1 per acre per lift
1 per acre per lift
1 per acre per lift
1 per acre per lift
More frequent
More frequent
More frequent
Performed and well
documented
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
SOIL DRAINAGE LAYER
Prequalification Testing
Grain Size Analysis ASTM D422
Hydraulic Conductivity ASTM D2434
Carbonate Content Testing ASTM
D4373
Construction Testing
Grain Size Analysis ASTM D422
Hydraulic Conductivity ASTM D2434
Carbonate Content Testing ASTM
D4373
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
1 per 1,500-2, 630 c.y.
1 per 2, 630-3, 000 c.y.
1 per 2, 630-3, 000 c.y.
1 per 2.5 acres
1 per 7.5 acres
1 per 2, 630 c.y.
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
72
-------
TABLE 4-2
MATRIX FOR EVALUATING CAP CQA/CQC AGAINST ACCEPTABLE INDUSTRY PRACTICES
Category
Less than Acceptable
Acceptable
Better Than Acceptable
GEOMEMBRANE BARRIER
Manufacturing Quality Control Testing
Resin Testing:
Melt Index ASTMD1 238
Resin Density ASTM D1505
Environmental Stress Crack ASTM
D1 693/5397
Compliance with NSF 54 Standard
Geomembrane Conformance Field
Testing:
Thickness, Tensile, Elongation,
Puncture, Tear
Construction quality Control Inspection:
Material Delivery Inspection
Material Handling and Storage
Inspection
Pre-deployment Panel Layout Diagram
Pre-deployment Written Subgrade
Inspection Certificate
Construction Quality Control Seam
Testing:
Trial Seams Testing
Non-destructive Seam Testing:
Vacuum Box on Extrusion Welded
Seams
Air pressure testing on Fusion Welded
Seams
Destructive Seam Testing:
Peel and Shear Testing ASTM D4437,
D3083, D751
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
Not performed
1/180,000 Ibs
1/180,000 Ibs
1/lot(1, 800,000 Ibs)
Performed
1/1 00,000 s.f.-1/lot
Every roll
Every roll
Every roll
Every roll
a.m. & p.m.
100% of seams
100% of seams
250-750 linear ft. of seam
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
More frequent
73
-------
4.2.1 Borrow Soil
Borrow soil is used for preparing the subgrade and installing the cover soil. Tests are conducted
to prequalify the borrow source, as well as during the placement of the soils. The standard
frequencies for testing the borrow soils are:
Soil classification (prequalification test) - One test per 1.000 to 6,580 cubic yards.
The ASTM D2487 test is conducted, and the frequency is lower (1 per 6,580 cubic
yards) if the borrow soil is relatively uniform.
Compaction test (prequalification test) - One test per 5,000 to 6,580 cubic yards.
The ASTM D698 or D1557 test is conducted, and the frequency is lower (1 per
6,580 cubic yards) if the borrow soil is relatively uniform.
Density test (construction test) - Five per acre per lift placed. The ASTM D2922 or
D1556 test is conducted.
Moisture content test (construction test) - Five per acre per lift. The ASTM
D3017 or D2216 test is conducted.
4.2.2 Compacted Clay Liner
If natural clays are more economical to use than a geosynthetic clay liner, natural clays are used
to construct the low-permeability layer. The source of the clay must be prequalified, and, during
construction, the quality of the layer must be tested to determine that the design permeability is
achieved. A test pad generally is used to determine the moisture requirement and the compactive
effort needed to achieve the design permeability. According to the results from the test pad, the
clay layer is placed, and tests are conducted on each lift placed to confirm the quality of the layer.
The standard frequencies for these tests are set forth below:
Soil classification (prequalification test) - One test per 1,000 to 6,580 cubic yards.
The ASTM D2487 test is conducted on the borrow soil.
Compaction curve (prequalification test) - One test per 5,000 cubic yards, using
the ASTM D696 or D1557 test.
Permeability (prequalification test) - One test for each compaction curve.
Density - Five per acre per lift placed, using the ASTM D2922 or D1556 test.
Moisture content - Five per acre per lift placed, using the ASTM D3017 or D2216
test.
Undisturbed hydraulic conductivity and moisture content - One test per acre per
lift. This test is conducted on a Shelby tube test sample, using the ASTM D5084
test.
Atterberg Limits - One per acre per lift, using the ASTM D4318 test.
74
-------
Grain size test - One per acre per lift, using the ASTM D422 test.
Undisturbed dry density and moisture content - One test per acre per lift. This
test is performed on the sample obtained for the hydraulic conductivity test.
4.2.3 Soil Drainage Layer
The soil drainage layer is used to drain the cover soils and prevent buildup of hydraulic head on
the low-permeability layer. The borrow source for this granular soil must be prequalified, and,
during construction, the quality of this layer must be tested. The standard frequencies of these
tests are set forth below:
Grain size analysis (prequalification test) - One per 1.500 to 2,630 cubic yards.
The ASTM D422 test is performed.
Hydraulic transmissivity (prequalification test) - One per 2.630 to 3,000 cubic
yards. The ASTM D2434 test is performed.
Carbonate content test (prequalification test) - One per 2,630 to 26,300 cubic
yards. The ASTM D4373 test is performed. This test is used for limestone quarry
sources; the frequency depends on the variability of the rock in the quarry.
Grain size analysis - One test per 2.5 acres.
Hydraulic conductivity test - One test per 7.5 acres.
Carbonate content tests - One test per 2,630 cubic yards.
4.2.4 Geomembrane Barrier
The Industrial Fabrics Association and the NSF have developed several standards to ensure
manufacturing quality control for geomembranes. Over the past decade, geomembrane barriers
have been used extensively as either primary or secondary barriers to the infiltration of water into
waste. The EPA has developed CQA/CQC guidelines for the installation of geomembranes
(EPA 1993). The standard frequency of testing for ensuring the quality of the geomembrane
barrier is described below.
Manufacturing Quality Control
The following tests are conducted in the manufacturing plant;
Resin melt index - One per 180,000 pounds, using the ASTM D1238 test.
Resin density - One per 180,000 pounds, using the ASTM D1505 test.
Environmental stress crack - One per lot (1,800,000 pounds), using the ASTM
D1693 or D5397 test.
NSF test - One per lot, using NSF 54 Standard tests.
75
-------
Field Quality Control Inspection and Tests
The field inspection procedures and tests that are conducted to ensure that geomembranes are
installed properly are described below:
Thickness, tensile strength, elongation, puncture and tear resistance - One per
100,000 square feet or 1 per lot. These properties usually are certified by the
manufacturer; however, on large projects, these properties are field tested.
Material delivery inspection - Every roll.
Materials handling and storage inspection - Every roll.
Predeployment panel layout inspection - Every roll. The exact field location of
the roll and its relation to the other rolls is verified, and conformance to the approved
layout diagram is checked.
Predeployment written subgrade inspection certificate - Every roll. This
inspection verifies that there are no defects in subgrade preparation that could
damage the geomembrane. The written inspection certificate is checked before the
membrane is deployed.
Seam testing - The seam is the weakest link in the geomembrane barrier; therefore.
only experienced personnel perform this test. The frequency of such tests and types
of tests conducted are:
Trial seams - Two per day. Trial seams are tested for peel and shear
strength by application of the ASTM D4437, D3083, and D751 tests.
Nondestructive tests - All seams are checked, using the vacuum box
(extrusion welded seam) or air pressure (fusion welded seam) test.
Destructive tests - 1 per 250 to 750 If of seam. Samples of the seam are cut
to run, and peel and shear strength tests are performed, as for trial seams. A
patch of membrane then is fusion-welded over the sampled areas. Such tests
are kept to a minimum.
If testing described in Subsection 4.2.1, Subsection 4.2.2, Subsection 4.2.3, and Subsection 4.2.4
was not performed with the frequency indicated, the site was rated less than acceptable. If the
testing was conducted as described, the site was rated acceptable. If tests were conducted more
frequently than set forth above, the site was rated better than acceptable.
4.2.5 Range of Findings
Of the 36 sites evaluated during the study, 22 had caps in addition to the barrier wall. One site
had only a cap. The CQA/CQC elements (that is, borrow soil, clay liner, drainage layer, and
geomembrane) are crucial to the successful performance of the cap, and most of the sites studied
were rated acceptable. In many cases, the caps were tied in to the barrier wall to form a
containment system.
76
-------
Caps had been installed at 22 sites. For a majority of those, a GCL had been used for the low-
permeability layer, and a geonet for the drainage layer. (At 2 sites, the caps had yet to be
installed.) Four sites had compacted clay liners, and, at each of 2 sites, a parking lot had been
constructed over the site. The CQA/CQC procedures at those sites consisted of the following
elements:
Borrow soil testing to verify that the soil met specifications
Field and laboratory testing to verify that the compactive effort met or exceeded the
design specification and that the design permeability had been achieved
Geomembrane barrier testing that conformed to standard industry practice
For two of the sites, the use of cap test pads had been documented, as well as the use of quality-
control procedures using field permeability tests and undisturbed samples for laboratory
permeability tests. At site 7, the cap CQA consisted of six settlement monitors to evaluate
settlement during and after installation. Compaction and settlement were evaluated to identify
optimum construction practices, such as equipment weight and number of passes.
The soil drainage layer was usually a geonet, even at sites at which the low-permeability layer
was a compacted clay liner. The ease of using geosynthetic materials and improved quality
control at the manufacturing facilities have brought about widespread use of such materials in the
past five years.
The weakest link in the components of a cap is the seam mat connects each roll of geomembrane.
The QC seam testing described in Subsection 4.2.4 is crucial to performance of the cap.
Although both nondestructive and destructive seam-testing techniques are time-consuming, such
tests are necessary to ensure that the cap is performing as designed. At all the sites studied, the
installer had followed standard industry practice, and the regulatory agency had provided
oversight of this key construction activity.
4.3 PERFORMANCE MONITORING
Horizontal surface barriers or caps prevent dermal contact, limit infiltration or vertical inflow,
and contain upward migration (that is, vectors, diffusion, capillary action, or evapotranspiration).
Effective monitoring focuses on those functions. As is the case with vertical barriers, in ensuring
the performance of caps, most emphasis historically has been put on adequate CQA/CQC to
ensure a quality constructed product (EPA 1993). However, postconstruction monitoring is
required by regulations for closure of municipal solid waste and hazardous waste facilities.
Monitoring of groundwater quality, hydraulic head, and surface quality are common elements in
monitoring performance of vertical barriers. Those elements are described in the preceding
subsection. Specific approaches to the monitoring of caps are discussed in subsequent
subsections.
Generally, limited information about monitoring of the cap was available for the sites studied.
Typically, the visible grading and drainage and erosion control features were inspected. Less
typically, settlement and oilier structural features of the cap were monitored. In few cases was
contamination within or above the cap monitored. The following subsection discusses industry
standards and elaborates on the range of cap monitoring efforts identified in this study.
77
-------
4.3.1 Typical Industry Practices
Industry guidance for monitoring caps can be found in several sources, most notably EPA
publications on landfills (EPA 1990 and EPA 1992). The established monitoring categories are
ground-water, leachate generation, gas concentration, subsidence, surface erosion, and air quality.
These and other sources fail to provide specific monitoring requirements because of the site-
specific nature of the monitoring needs. Federal solid waste landfill regulations, 40 CFR
Part 258. describe in some detail requirements for groundwater monitoring (subpart E) and other
closure (subpart F) considerations (40 CFR 264.228, 264.310, 270.2 le). Bagchi (1989) provides
a concise discussion of performance monitoring requirements for landfills.
Monitoring data for caps were not detailed enough for the sites reviewed to support a numerical
ranking similar to that prepared for the vertical barrier monitoring matrix (Table 3-7).
Groundwater quality monitoring data were common to vertical barriers. Other noteworthy
monitoring data for caps are described in the site summaries in Appendix B, Volume II and are
discussed briefly in the following subsection.
4.3.2 Range of Findings
Of the 36 sites having vertical barriers, 22 also had caps in place. In some cases, the uncapped
sites were operating landfills; at others the cap had not yet been constructed. The remainder
simply were not capped. Of the capped sites, periodic visual inspections of the cap surface had
been performed as part of ongoing operation and maintenance procedures, and the usual
problems with erosion, drainage, and vegetation had been documented. Investigation and
compilation of data yielded only limited cap-specific monitoring data, other than those on
groundwater quality.
At Site 36. soil moisture and vegetative growth had been monitored periodically because the
design of the cap was based on consumptive use to limit infiltration into the containment. At
Site 12, water quality in the drainage layer and long-tenn settlement will be monitored regularly
in response to a regulatory requirement.
4.3.3 System Monitoring
In collecting the monitoring data described in previous sections, it became clear that, if data had
been reviewed, they seldom had been evaluated to determine performance of the complete
containment system. Typically, one reporting entity received groundwater quality data, and
another received head and other information specifically related to the barrier.
Generally, at the sites studied, the system as a whole was monitored and data collected that
cannot identify leaks attributable to defects in either cap or barrier. The monitoring categories
reviewed show a wide variety of methods and frequencies, even in light of differences in
applications. The more diagnostic stress tests were found to be run infrequently, typically only
when a problem was detected.
Testing for long-term degradation, physical sampling for diffusion through the barriers, and
testing for the effect of desiccation on permeability effectively were absent.
Design of the containment system is dependent on conditions at the site, particularly the
hydrogeology. Standardized monitoring therefore is difficult. However, measurement of
78
-------
performance could benefit significantly from more consistent monitoring approaches and more
periodic reporting than currently are the norm. Such recommendations are discussed in more
detail in subsequent sections of this document,
4.4 OPERATION AND MAINTENANCE
O&M for caps require correction of surface erosion problems and inspection of the leachate and
gas management systems (EPA 1989). Acceptable industry practice is to follow the O&M plan
that EPA and the pertinent state agency have approved for the site. The O&M plan may include:
Visual inspection of the cap and mowing of the grass during the growing season
Repairs of any erosion or surface slumping
Repairs of any damage to drainage control structures
Quarterly monitoring of leachate levels and monitoring wells
Quarterly or semiannual monitoring of leachate quality
Monthly inspection of the components of the leachate collection system
Cleanout of leachate collection pipes, as required
Monitoring of the gas vents to clear any obstructions
Quarterly or semiannual monitoring of the quality of gas
No unusual O&M practices were identified at the sites studied.
4.5 COSTS
Caps are constructed with compacted clay, geosynthetics, and combinations thereof. The cost of
construction of a cap depends on the local availability of soils for the grading layer and cover
soils. In general, compacted clay liners may cost more than geosynthetic clay liners, especially if
the clay borrow source is a few miles from the site. The cost of a cap also depends on the extent
of leachate and gas management the site requires.
For the sites studied, the capital cost of installation of a cap varied from $200.000 to $400.000
per acre (see Table 3-8). The O&M cost varied from $10,000 to $50,000 per year, excluding
administrative and reporting costs. The O&M cost depended on the amount of groundwater and
surface-water sampling required by the state regulatory agency.
Obtaining cost information on subsurface barrier walls and caps was not the primary objective of
this study. However, when cost information was provided in the site reports, that information is
included in the individual site summary in Appendix B, Volume II. In most cases. O&M costs
could not be obtained. Capital costs mentioned in the reports are engineer's estimates, rather than
actual construction costs. For a few sites, the remedial action report provided the actual
construction costs for the remedy components, such as barrier wall, cap, and wastewater
treatment plant.
79
-------
5.0 PERFORMANCE EVALUATION
A primary objective of the study is to assess performance of containment as a remedy in
controlling contamination in the subsurface. The term "controlling" here refers to holding the
source for subsequent removal, reducing off-site migration, reducing leachate generation, or
simply containing exposure to the source. The general objective of containment is to control
inflow to and outflow from the system and contact with the source to minimize its effect on the
environment for a significant design life. Measurement of a site's specific performance was
based upon data related to site-specific criteria, compared with regular monitoring data, which at
most sites studied consisted of those on hydraulic head and groundwater quality. Because of the
site-specific criteria and types of monitoring data available for the sites studied, a comparative
evaluation is problematic. Nevertheless, the following subsections present evaluation findings by
identifying and describing the basis for measuring performance, presenting a summary and
evaluation based on 4 considerations, and describing factors that contribute to performance.
5.1 PERFORMANCE BASIS
The performance basis varied considerably among the sites evaluated. The performance basis
can be viewed first by containment type and second by specific monitoring standard. Table 5-1
below summarizes the basic containment type and standards for the sites studied.
TABLE 5-1
CONTAINMENT CATEGORIES AND PERFORMANCE MONITORING
Category
Active
Passive
Reactive
Cutoff
Number
of sites
18
12
1
4
Performance Based on
Groundwater Quality
Monitoring
15
12
1
1
Performance Based on
Hydraulic Head
Monitoring
15
2
Performance Based
on Other
Monitoring
3
1
Section 3-4 describes the various monitoring methods used for the sites studied. The monitoring
method used and the performance criteria depend on the category of the site and often on site-
specific features.
Active containment is the most prevalent containment type among the sites studied. Further, site
selection research (see Section 2.0) indicates it is the most prevalent form of containment.
Active containment includes a vertical subsurface barrier, often a surface horizontal barrier (cap),
and some form of groundwater or leachate withdrawal inside the containment to maintain the
groundwater elevation at a level lower inside the barrier than outside. This hydraulic head
differential or gradient criterion varied among the active containment sites, sometimes
influencing performance. Table 5-2 summarizes the head differences among the 18 active sites.
80
-------
TABLE 5-2
VARIATION IN ACTIVE CONTAINMENT HEAD DIFFERENTIAL
Head Differential
Inward, < 1 foot
Inward. 1-10 foot
Inward. >10 foot
Inward, below waste elevation
Inward, unknown difference
Number
of Sites
2
7
5
1
3
Remarks
See Table 3-7 for locations of readings
Hydraulic head monitoring commonly is used to monitor performance at active containment
sites. The necessary groundwater head differential across the vertical barrier is established to
maintain inflow and control flux. Head differential may be sufficient to overcome diffusion
through the vertical barrier or keep contaminants away from the barrier. Vertical gradients may
be induced as well to contain contamination. For example, thin barriers of low permeability may
require increased head differential to counteract diffusion mechanisms (Devlin 1996).
To measure performance, passive and reactive containment types usually rely on monitoring of
groundwater quality, rather than hydraulic head. Often, the requirement for monitoring of
groundwater quality is more rigorous for passive and reactive containment than for active
containment because improvements in groundwater quality are the only measure of performance.
Similarly, monitoring of groundwater quality is important as a measure of whether active
containments achieve and maintain standards for groundwater quality outside the containment.
Section 3-4 presents and Table 3-8 lists other site performance criteria and corresponding
monitoring.
In addition to compliance with performance standards, performance of the containment system
should be measured by its success in meeting the design intent of providing a system that
minimizes long-term O&M costs to the owner. For example, at several active containment sites,
actual groundwater extraction to maintain performance standards was greater than had been
predicted; the system therefore had been more costly to the owner than it had been predicted.
5.2 PERFORMANCE STAGES
Containment performance can be viewed as occurring in stages. The stages have parallels to the
generic model for cleanup of groundwater (EPA 1992). Figures 5-1A and 5-1B present a
graphical display of typical groundwater quality and hydraulic head responses for containment
scenarios. Several stages can be defined: the construction stage, the equilibration stage, the
demonstration stage, the short-term performance stage, and the long-term performance stage.
After the initial construction stage, groundwater inside and outside the containment adjusts to
new flow conditions during the equilibration stage. This stage may be complete within months
to a year after construction, depending on the flow regime. The demonstration period consists of
several quarters of monitoring, during which system operation is fine-tuned and final
equilibration continues. This period generally lasts from 1 to 2 years. The end of the period
triggers operation and performance monitoring, as required by the regulatory agency. The short-
term performance period ends after 5 years of continued operation and generally overlaps
somewhat with the demonstration period. The long-term performance stage includes operations
81
-------
I
I
i
b
T3
CO
CD
Performance Stages
for an Engineered Barrier
Figure 5-1 A
Active Containment
I T
1 2
CD
o
(D
Stage
CD
o
£
£
T3
CO
CD
Figure 5-1 B
Passive Containment
I I
1 2
CD
° 3
- (D
Stage
Head Differential
(+ Outward)
(- Inward)
GW Quality
(+ Contaminant Down Gradient)
Stages: 1 .Construction
2.Equilibration
3.Demonstration
4.Short-Term Performance
S.Long-Term Performance
82
-------
and some monitoring, presumably at a reduced frequency, over a 30-year period. Figure 5-1
illustrates performance response for both active and passive containment. Performance for both
remedies converges toward no or some de minimus leaking during the performance period.
Specific performance results are discussed in the following subsection.
5.3 RANGE OF FINDINGS
Table 5-3 provides a summary of the range of findings. The table lists the remedial objective for
the particular containment system and identifies elements of short- and long-term performance.
The preceding subsection described short- and long-term performance periods. Generally, short-
term performance is covered by monitoring during the first 5 years by sampling and testing after
installation of the containment system. Long-term performance is identified through monitoring
over the design life of the containment system, for example 30 years; possible effects of
degradation of the containment system should be considered. Assessment of long-term
performance may include evaluation of data from hydraulic stress testing, periodic testing of
barrier integrity, and long-term compatibility testing of the barrier materials with the site
contaminants, as well as evaluation of the effectiveness of any redundancy measures installed.
Table 5-4 also presents a rating of the performance of a particular site's containment. The
performance rating was based on available data and considerations related to several containment
systems. Standard 5-year review reports were used, when available. Quarterly and annual
reports of monitoring data also were used. Monitoring data were reviewed to determine whether
regulatory criteria were being met, as required. Hydrogeologic features of the site and the type of
containment were weighed during the evaluation. Site contacts also were queried about the
operational performance of the containment.
The ratings listed in Table 5-4 and in the individual site summaries in Appendix B of Volume II
were based on several considerations:
Did the site meet performance criteria (water quality, head differential, or other)?
Did performance improve continually over time?
Were any problems detected, and, if so, were they remedied readily?
Was the owner satisfied with the performance and protectiveness of the containment?
Using site data and evaluating performance according to the above considerations, the review
team assigned each site one of four ratings, as defined below:
1 = Remedial objective was met.
2 = Evidence suggests that remedial objective may have been met.
3 = Evidence suggests that remedial objective may not have been met.
X = Data are insufficient to determine whether the remedial objective was met.
83
-------
Table 5-3
Performance Evaluation Matrix
Site
1
o
3
4
5
6
7
8
9
10
11
12
13
14
15
Remedial Objective
Surface water and down gradient
water quality to meet MCL; no
head standard
GW cutoff to allow for landfill
excavation
Source control
Prevent offsite migration
Containment of inflow
Intercept landfill seepage; achieve
groundwater gradient and MCL
Minimize offsite migration
No head gradient criteria,
reduce the potential for downward
vertical migration
DNAPL containment
Contain landfill leachate; enable
further landfill expansion
Capture and remove organic
contaminants
Improved surface water and
groundwater quality
Reduce lateral migration of
groundwater contamination from
source trenches
Limit migration of contaminants
Seepage cutoff wall
Maintain inward gradient around
landfill
Containment System
SB hairier, leachate collection,
cap-geomembrane over compacted
clay liner (CCL)
Passive SB hairier
7 SB walls, multimedia cap
Grout curtain, SB barrier; RCRA
cap
Clay barrier, leachate drain, clay
cap, active gas collection
SB barrier, multimedia cap
SB hairier, leachate drain, CCL
cap, active gas collection
Steel pile, GW bioremediation
SB hairier; leachate collection
system
Active system consisting of SB
cutoff wall, withdrawal and
recharge wells
SB & SCB barrier; RCRA type
cap, up gradient groundwater
interceptor; leachate extraction;
gas collection
Passive system consisting of SB
barrier, soil and vegetative cap
SB barrier, cap. pump/treat
leakage collection
Plastic concrete; run-on extraction
wells
SB hairier and leachate collection,
no cap
Short-Term Performance
Groundwater quality
improvement; visible
improvement in adjacent surface
water
Dewatered in 1 year
GW quality outside well has
reached steady state
Pumping rates as designed; down
gradient water quality-
improvement
Shallow groundwater met MCL
standards
Works as designed
Surface and shallow groundwater
improvements
Design criteria met
Undetermined; system not active;
rising leachate levels
Improved groundwater quality, but
groundwater contaminants were
detected moving around both
edges of the linear cutoff wall
Improved surface water
monitoring data unavailable
Inward gradient developed along
the up gradient side; outward on
down gradient side.
GW quality outside improved
Performance as designed for 4
years of mining operation
Groundwater quality improved
Long-Term Performance
No long-term data available
Landfill is still diy after 15 years
GW monitoring program being
modified to assess performance
Insufficient information
Decreasing VOC concentrations
Insufficient information
System not yet able to dewater and
develop inward gradient
Insufficient information
Not applicable
Groundwater quality continues to
improve down gradient of the
cutoff wall
No data
Groundwater quality has showed
little change over time.
GW quality outside improved
No data
Performing as designed since 1984
Remarks
Site has fulfilled the remedial objective
Site is performing as designed
2 hanging walls
No groundwater gradient criteria
Compacted clay hairier
Contained area is solidified
-
DNAPLs monitored in wells
Active landfill
The cutoff" wall was extended in 1990; additional
extraction and recharge wells were added in 1991 to
prevent contaminant movement around the wall
SB/SCB hairier
Vibrating beam thin wall
Extraction (flushing system)
Very well instrumented at site
2 phase barrier installation, 1984 and 1989
84
-------
Table 5-3
Performance Evaluation Matrix (Continued)
Site
16
17
18
19
20
21
22
23
24
25
26
27
28
Remedial Objective
Contain DNAPL; improve surface
water quality; maintain head
gradient across barrier
Maintain inward gradient and
leachate collection
Maintain head differential of 1 ft
Zero net gradient across barrier;
downgradient water quality to meet
MCL"
Inward 0.01 ft head differential
Maintain head differential of 1 ft
Maintain inward gradient
Improve groundwater quality
outside barrier; inward head
gradient differential of 1 ft.
Prevent further degradation of SW
and GW quality-
Inward gradient in shallow
groundwater zone
Intercept, treat, and discharge
contaminated groundwater
Groundwater standard; maintain
regional gradient, source separation
Minimize off-site migration
Containment System
CB barrier; to permit DNAPL
extraction
SB barrier, multimedia cap,
extraction wells, interceptor trench
SB barrier; RCRA cap
SB barrier; RCRA type cap, well
extraction and trench recirculation
SB barrier, multi-media cap,
extraction wells; auto monitor for
head differential
SB barrier, multi-media cap
SB barrier, multimedia cap,
extraction wells
SB barrier, no cap, wells and
drain, groundwater extraction
Active SB barrier, RCRA cap,
extraction wells
Active system consisting of SB
hairier, groundwater infiltration
and extraction system
Active system consisting of SB
cutoff wall, withdrawal wells
SB hairier; CCIVgeomembrane
cap; extraction wells
SB barrier, extraction wells
Short-Term Performance
Immediate surface water
improvement; enabled DNAPL
extraction
Inward gradient established
Inward gradient developed;
pumping rates as designed
Immediate shallow groundwater
improvement
Immediate surface and ground
water improvement; good
hydraulic test response
Inward gradient developed
Inward gradient not developed as
vet
Head differential achieved in most
locations; ground water quality
improvement
Inward gradient maintained
Inward gradient developed
Groundwater contamination
migrated under and through the
wall due to poor key and high
hydraulic gradient
Immediate groundwater
improvement; reduced pumping;
head difference easily met
Performing as designed
Long-Term Performance
Constant head gradient with time;
decrease in down gradient
concentration
Insufficient information
Insufficient information
Possible hairier underflow,
downward gradient to bedrock
aquifer
Well installed in 1996
Well installed in 1994
Remedy is interim measure
Some down gradient
contamination detected; barrier
underflow identified at one
location
Insufficient information
Groundwater quality improved and
inward gradient maintained along
most of the wall
Improved recharge capacity has
decreased the hydraulic gradient to
near zero
Continued monitoring; protective
of human health
Seems to have accomplished
design objective
Remarks
Good example of DNAPL cutoff; emergency
response installation
Site is performing as designed
Post construction testing revealed improper key-in,
which was corrected with in situ soil mixing
Good design and CQA/CQC; extensive 10-year
review
Head differential and surface and groundwater
quality requirements
Fate and transport modeling completed
HOPE membrane placed through center of the wall
2 phase construction; active landfill; containment
leakage easily rectified
Owners seek to change from active to passive
remediation and cap the site
Recent water quality- data is not available but the
removal of the high gradient should curtail or
reduce plume movement beyond the wall
Good CQA/CQC; sustained dewatering of source
Waste inside hairier has been solidified
85
-------
Table 5-3
Performance Evaluation Matrix (Continued)
Site
29
30
31
32
33
34
35
36
Remedial Objective
Maintain inward gradient
Maintain head differential of 1 ft.
Maintain head differential of about
1 ft. to control contaminant source
and groundwater cleanup
Maintain head differential
Channel plume to flow through
reactive wall
Cut off seepage through and below
dam
Dewatering for construction of
power plant
Prevent migration of contaminant
plume
Containment System
Active SB wall, multimedia cap,
extraction wells
SB banier, multi-media cap
Active system consisting of SB
barrier, collection and treatment
system
S-B/RCRA cap with groundwater
extraction
C-B/S-C-B reactive wall
SB barrier
Plastic concrete and dewatering
Multimedia cap, leachate
collection and treatment
Short-Term Performance
Inward gradient established, GW
quality improved
Inward gradient developed on up
gradient side, outward on down
gradient side
Inward gradient developed;
pumping rates as designed
Inward gradient throughout site
Performing as designed
Very effective
Excellent - based on predicted and
actual Dumping rates
Contaminant plume contained
Long-Term Performance
No information
Well installed in 1993 testing
Inward gradient maintained
Inward gradient for 7 years
System operational for only one
year
Maintains large head differential
across cutoff
Well operational for 2 years
Insufficient information
Remarks
-
Outward gradient developed after pumping from
inside wall was stopped in 1995.
System has successfully reduced onsite levels of
contamination and prevented contaminants from
reaching the nearby river.
System performed better than expected by design.
VOC not detectable in downgradient well
Automated electronic monitoring
Plastic concrete wall subject to high gradient
"Cap only" site
Notes:
MCL = Maximum contaminant limit
86
-------
TABLE 5-4
PERFORMANCE RATING VERSUS CRITERIA RATING
Site
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
Design*
3
1
2
1
2
1
1
X
2
2
2
2
2
1
2
2
2
1
1
1
1
2
1
2
1
1
1
1
2
1
2
2
1
1
1
1
CQA/CQC*
3
2
2
1
1
1
1
X
2
1
1
1
1
1
1
3
1
1
1
1
1
1
1
2
2
X
1
1
2
1
1
X
1
1
1
2
Monitoring*
2
1
3
3
1
2
3
3
3
2
X
1
1
1
1
1
2
1
1
1
3
3
1
3
2
2
1
1
2
2
2
1
2
1
1
2
Performances
2
1
X
X
1
1
2
X
X
2
X
2
2
1
1
2
2+
2
3
2
X
3
2+
3
2
2
2+
2+
2
3
2
1
2
1
1
X
Containment
Category
Passive
Passive
Passive
Passive
Passive
Passive
Passive
Passive
Passive
Passive
Passive
Passive
Active
Cutoff
Active
Cutoff
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Active
Reactive
Cutoff
Cutoff
Cap Only
* Criteria Rating: 1= Better than Acceptable; 2= Acceptable; 3= Less than Acceptable;
X = Insufficient Data
# Performance Rating: 1= Remedial objective was met; 2= Evidence suggests objective may have
been met; 3= Evidence suggests objective may not have been met; X= Insufficient data to determine
if remedial objectives have been met.
87
-------
Of the 36 sites studied, 8 are in Category 1, 13 are in Category 2, 4 are in Category 3, and 7 are
in Category X. An additional 4 sites met the remedial objectives. However, because long-term
performance could not be verified, those sites were rated as 2+.
The ratings show that the engineered barriers at 25 of the 36 sites studied generally performed as
designed and significantly improved the quality of groundwater and surface water in the vicinity
of the site. Unfortunately, for many of the sites, data are insufficient to determine long-term
performance.
At 4 of the sites studied, leaks were detected at the key; however, the leaks were repaired with
relative ease. For example, at Site 26, leaks were detected below the key. The problem was
solved by installing a recharge trench. At Site 19, underflow was detected at the bedrock key;
grouting was used to repair the problem. At Site 23, a leak was detected at one place and was
repaired. At Site 27, downgradient leaks were detected; however, the leaks were interrupted and
treated. At Site 30. downgradient leaks were observed when pumping from inside the wall was
stopped. The extent of the leaks and their effect on downgradient water quality are being
evaluated.
At the sites that have caps, the caps were integrated with the barrier walls to minimize the
amount of surface water entering the waste. Installation of a containment system resulted in
immediate reduction of contaminant transport and, in active systems, significantly reduced the
amount of contaminated water pumped to a treatment plant. Leaks detected were located
primarily at the key horizon. Such leaks could be the result of insufficient penetration into the
low-permeability horizon, insufficient quality control, or poor quality of the key material.
5.4 FACTORS AFFECTING SYSTEM PERFORMANCE
Rated containment performance was compared with the evaluation criteria for the sites studied to
determine whether some obvious correlation exists. No definitive correlation could be identified
from the limited data available because both design and CQA/CQC significantly affect
performance. Table 5-4 shows performance ratings for the sites studied and the previously
assigned ratings for design, CQA/CQC, and monitoring criteria. Table 5-4 shows that the
majority of the sites received favorable ratings (1 or 2). The sites rated 2+ appeared to meet the
remedial objective, and data existed to prove the performance of the containment. However, data
were unavailable for several consecutive periods of positive performance. At several,
performance was poor. At several others, indicated by an X, data were insufficient to support
assignment of a performance rating.
Review of the table reveals that most sites having positive ratings, 1 and 2, also were rated
favorably for the other criteria shown. Good efforts at design, CQA/CQC, and monitoring
generally led to positive performance, as often advocated and indicated by the data. However,
less than optimum performance can occur even when significant design, CQA/CQC, and
monitoring efforts are used, as the ratings for sites 19 and 30 indicate. Less than optimum
performance under one subcriterion or criterion also can influence performance significantly. For
example, a problem with keying a vertical barrier, lack of continuity, or low permeability of the
site floor can decrease the performance of the system.
As discussed in Section 3.3 and 3.4, CQA/CQC testing and postconstruction physical testing
showed that the constructed barriers were of consistently low permeability. At several sites,
88
-------
problems with the key of the vertical barrier and the floor, or the continuity of the floor, likely
created leaks and less than optimum performance. For example, at sites 15, 19, and 26, key
problems led to leaking. At Site 22, drawdown could not be achieved to maintain the needed
head differential, a circumstance that most likely occurred because the bottom was more pervious
than predicted.
However, it should be noted that, at most of the sites with known performance problems, leaks
have been or are being rectified to achieve performance objectives. Those efforts are described in
the individual site summaries in Appendix B of Volume II.
The discussion set forth above describes performance and relates performance criteria to
observed performance. However, other criteria or subcriteria can contribute to short- or long-
term performance. Some of the negative contribution mechanisms are diffusion, long-tenn
degradation, abrupt physical failure, altered groundwater flow, and abrupt mechanical failure.
Among positive mechanisms are natural attenuation and adsorptive capacity of the barrier. The
reader is referred to the Rumer and Mitchell (1996) for further explanation of these and other
mechanisms.
6.0 CONCLUSIONS AND RECOMMENDATIONS
This report evaluated engineered barriers used to isolate hazardous sites from their surroundings.
Those barriers included:
Subsurface barriers - that is, vertical cutoffs, that prevent the horizontal migration of
the groundwater across the barriers
Impervious caps that prevent the downward migration of surface runoff
Initially, subsurface engineered barriers were the major component of passive containment
systems to prevent migration of contaminated groundwater from hazardous waste sites. Caps
often were added to completely isolate such sites. In recent years, for the most part, active
containment systems have been installed, where an inward gradient is maintained by extracting
and treating contaminated groundwater from the contained area. Recently, the concept of active
and reactive barriers has been introduced, and more thought has been given to the use of long-
tenn (more than 30 years) passive containment systems mat require much less maintenance than
active systems and that therefore offer lower operating costs.
In this context, it is important to assess the performance of engineered barriers installed since
1980 as short-term and long-term remediation techniques. The conclusions and recommen-
dations of the study are presented below, emphasizing subsurface barriers.
6.1 CONCLUSIONS
Nineteen of the 36 engineered barriers studied for this report met or may have met the intent of
the design and have been effective to date in preventing migration of contaminated groundwater
outside the contained zones or in meeting design objectives. At 4 of the 36 sites studied,
additional corrective measures were needed to meet the design intent. This conclusion is limited
to the scope of this study. In particular:
89
-------
1. Thirty-six sites of the 130 sites identified were analyzed. The other sites were eliminated
from the study because of the lack of availability of design, construction, or monitoring data.
(It is likely that some those sites that were eliminated do not perform as expected.)
2. This study represents a majority of the sites that have been in operation for less than 10
years. It is difficult to extrapolate this performance evaluation for the 30-year design life of
the barrier because of physical and chemical degradation of the backfill resulting from site
contaminants.
Table 6-1 summarizes the ratings assigned to the factors that affect performance for the 36 sites
included in mis study. As the table shows, most sites were rated acceptable or better than
acceptable for design, CQA/CQC, and monitoring. In particular, more sites were rated better
than acceptable for CQA/CQC than for either design or monitoring. (Data were insufficient for
the CQA/CQC factor at the greatest number of sites.) More sites were rated less than acceptable
for the monitoring factor than for either design or CQA/CQC. The distribution of performance
ratings most closely resembled that for the ratings based on monitoring.
Table 6-1
Number of Sites by Rating Criteria
Rating'
1
2+
2
3
X
Design
19
-
15
1
1
CQA/CQC
24
-
7
2
3
Monitoring
16
-
11
8
1
Performance"
8
4
13
4
7
Notes:
Criteria Rating:
1= Better than Acceptable;
3= Less than Acceptable;
2= Acceptable;
X= Insufficient Data
Performance Rating: 1= Remedial objective was met;
2= Evidence suggests objective may have been met;
3= Evidence suggests objective may not have been met;
X= Insufficient data to determine if remedial objectives
have been met.
The performance of a barrier is inferred by monitoring data obtained after installation; however,
the performance depends on the design and CQA/CQC effort during installation. Of the sites
studied, 94 and 86 percent had acceptable or better than acceptable design and CQA/CQC ratings
respectively, but only 75 percent of the sites had a acceptable or better than acceptable
monitoring rating. This finding suggests that monitoring, which is critical for determining the
performance of a barrier, needs to be improved and based on site specific requirements. Sites
that had a below acceptable rating for monitoring might be performing adequately because design
90
-------
and CQA/CQC were acceptable or better than acceptable at those sites; alternatively, some
releases may be undetected
The general major conclusions of this study are:
Based on data from 25 of the 36 sites studied, subsurface engineered barriers are
effective containment systems for the short and middle term, if properly designed
and installed. None of the monitoring data reviewed indicated a decrease in
effectiveness as a function of time.
In the 36 sites studied, monitoring systems for subsurface engineered barriers lack
consistency in terms of scope, design, and implementation. This results in lack of
credible data that can be used to evaluate performance. This study shows the need to
standardize the design and implementation of monitoring systems. This need
becomes even more crucial for engineered barriers used for long-term containment.
The most likely pathway for leaking of continuous subsurface barriers is in the
vicinity of their keys, as a result of defective installation (that is, insufficient
cleaning prior to backfilling, insufficient key as specified by the design or lack of
sounding, or defects in the aquitard layer below the key).
The soil-bentonite slurry wall technology was the most widely used technique for the
sites that were studied. Improvements in barrier technology, such as the
development of active backfill material capable of degrading the site contaminants,
should be encouraged.
The conclusions and the recommendations of the project team are discussed further below.
6.1.1 General Conclusions from Performance Evaluation
Active containment using soil-bentonite slurry walls is the most prevalent type of containment
used at the sites studied. Sites showed positive performance despite variations in methods and
the amounts of interior groundwater withdrawal required. Increased head differential across the
barrier in active containment did not necessarily correlate with improved performance. Passive
containment, without groundwater extraction, also showed positive performance results,
primarily indicated by groundwater quality outside the containment.
Performance is measured by established standards. However, the implemetation of standards
varied dramatically among the sites studied. Design, CQA/CQC, and monitoring can affect
performance directly. Design was performed somewhat consistently among the common types of
containment sites evaluated. CQA/CQC appeared to conform with industry standards, and, with
some exceptions, was evaluated as acceptable or better than acceptable. Types and frequency of
monitoring were less consistent among the sites studied. Since such criteria provide a
measurement of performance, that circumstance implies mat containment performance may be
inconsistent. A more standard approach to monitoring and reporting that allows variations to
accommodate site specific conditions, could improve the measurement of containment
performance and ultimately lead to improvements in containment as a remedy.
91
-------
The following discussions elaborate on the conclusions of the study by describing important
findings related to the various evaluation criteria. Conclusions are justified further in the
pertinent sections of the report. Preliminary recommendations follow the conclusions.
6.1.2 Design of Containment Systems
The design of subsurface barriers and caps is based on specific remedial objectives for the control
of exposure to hazardous waste and prevention of migration of groundwater contamination.
Section 3.2 defines the established standard for barrier design; design at most sites were
determined to be acceptable or better than acceptable. Design standards for vertical barriers
generally followed USAGE guidance, and cap design generally was based on EPA guidance. For
most site designs, hydrogeologic and geotechnical investigations had been undertaken to obtain
data to support the design. In several cases, that effort was insufficient and resulted in poor
containment performance because either the key or site bottom had not been defined adequately.
For many sites, performance modeling was not performed as part of the design. While modeling
results are often debatable, that tool is the best means of assessing effects on groundwater before
actual operation. Compatibility testing of trench slurry and backfill to support the design had
been conducted at most sites. However, the need for vertical barrier integrity and long design
lives suggests that additional study of long-term performance should be performed.
The quality of design indirectly contributes to the performance of the containment system. The
industry has adopted design guidelines. In some cases, the design subcriteria adopted are more
stringent than the published design guidance.
6.1.3 Construction Quality Assurance and Construction Quality Control of Containment
Systems
CQA/CQC contributes to containment performance. Because of the buried barrier elements of
either caps or vertical barriers, postconstruction testing is difficult; further, a quality constructed
product will improve performance. Construction quality assurance plans (CQAP) sometimes are
prepared at the same time as the design documents, and inspection and testing requirements
typically are specified in CQAPs. CQAPs are produced more often for caps than for vertical
barriers. CQA/CQC guidance provided by USAGE (USAGE 1996) and EPA (EPA 1993) for
vertical barriers and caps, respectively, and earlier guidance had been followed at most of the
sites studied. Generally, CQA/CQC procedures for vertical barriers have improved less than
those for caps. In many cases, CQA/CQC adjustments in testing frequency were made to
accommodate site-specific conditions.
Performance at several sites was affected negatively by problems with the vertical barrier that
might have been eliminated if CQA/CQC controls had been more stringent. Two such controls
are:
Trench key confirmation At some sites, sampling and sounding were used to
confirm the key, while, at a majority of the sites, only sounding was used; physical
inspection of the key at regular intervals is important in minimizing poor key
contact, regardless of the type of barrier.
92
-------
Trench bottom cleaning Accumulation of sloughed or settled high-permeability
material may create windows in the vertical barrier; at some sites, cleaning tools or
desanding pumps were used to ensure a clean bottom in the key.
Other findings of the evaluations are:
The most crucial measurable parameter of the vertical barrier, permeability, had been
measured in several ways, producing significant differences
For slurry trench barriers, physical inspection of the key material can be performed,
while key inspection is difficult at most other barriers
Water levels during and immediately after construction were not monitored or
documented well, although such data often can help determine the integrity of the
barrier
Postconstruction testing of the barrier is prevalent and usually involves obtaining
backfill samples at regular intervals. This procedure is performed and sometimes
mandated, despite concern in the industry about the negative effects of such
sampling efforts on the integrity of the barrier. Efforts are being made to develop
nonintrusive post construction monitoring techniques.
Barrier CQA/CQC has not been standardized to a level comparable to mat for caps
6.1.4 Performance Monitoring
Major differences were found among the sites studied in the performance monitoring of the
containment system. At some sites, very little monitoring of groundwater head and quality had
been carried out, while, at others, paired piezometers at close spacing were used within 25 feet of
the barrier to monitor groundwater head, and monitoring well networks downgradient of the sites
were used to measure trends in groundwater quality.
Frequency of monitoring also varied among sites, with monthly monitoring at some sites and
quarterly monitoring at most sites. At some sites, the monitoring data had been analyzed to
determine performance and reports submitted quarterly to the regulatory agency, but at most
sites, the monitoring data had been submitted with little or no analysis of performance.
At only a few sites, key or subkey soils had been monitored to identify any barrier underflow.
Essentially no long-term monitoring of physical samples had been performed to identify
degradation mechanisms. Geophysics had been used at several sites, but the results were not
conclusive. Cone penetrometer and pressure tests had been conducted at two sites and had
yielded useful information. Stress testing and pumping tests after construction had been
performed only infrequently. Stress cells and electronic wire piezometers have been installed in
the barrier on some civil engineering projects. In addition, mere appeared to have been no
consistent method of establishing the groundwater monitoring network to obtain information
about either quality or head.
93
-------
On the basis of the information presented above and discussed in earlier sections of the report,
the following conclusions can be drawn:
Monitoring requirements differ for active, passive, and cutoff containment methods
Rational and consistent monitoring is needed, including standards for well
placement, accuracy of measurement, and frequency of sampling.
Less frequent sampling should be allowed when trends indicated by data are
consistently positive
Long-term performance of containment is not adequately measured
Geophysical methods, while intriguing, have not been demonstrated to be successful,
but should be investigated because of their inherent value in providing spatially
continuous testing
Reporting of monitoring data is inconsistent, and the regulatory community does not
use such data to the fullest extent possible to assess performance
6.1.5 Operation and Maintenance (O&M) at Containment Systems
O&M at containment systems consisted primarily of quarterly inspections of the cap for erosion
and O&M of the treatment plant. The data available did not support measurement of the effect of
O&M practices on performance.
6.2 RECOMMENDATIONS
Evaluation of the 36 sites indicates that containment can be an effective remedy for protection of
human health and the environment. However, the conclusions presented above reveal that
improvements could be made. Recommendations are discussed in general below and discussed
specifically in light of design, CQA/CQC, and monitoring.
Active containment has become more prevalent than passive containment. Active containment
performance standards should be made reasonably consistent. Passive containment should not be
discouraged; it should be evaluated further to understand the efficacy and cost-benefit
relationship, compared with active containment. Passive containment augmented by active
barriers and reactive walls should also be considered. As an alternative, conversion of active
containment systems to passive containment systems after the effectiveness of the system has
been demonstrated could be considered.
Recommendations by containment criteria, focusing on vertical barriers, are presented in the
following sections.
6.2.1 Design
The design of subsurface barriers and caps should be based on more complete hydrogeological
and geotechnical investigations, focusing on depth of key and integrity of the floor. In addition,
94
-------
although cap design has been standardized, a more prescriptive design for a subsurface barrier
should be developed. When appropriate, that design should include:
Design performance groundwater modeling
Geotechnical borings, at a maximum spacing of 200 feet, to define the stratigraphy
and properties of the key-in horizon
Design for the long-term compatibility of the containing barrier with aggressive
contaminants
Design for diffusion and desiccation mechanisms (as appropriate) that could affect
long-term performance
In addition, it is recommended that design using innovative technologies for vertical barriers.
such as trenching technologies, active barriers, and reactive barriers, be scrutinized to ensure that
sound engineering and construction methods are used in their application. In the case of reactive
barriers, groundwater modeling is a crucial element of the design. Monitoring of the
groundwater flow is also required to ensure long-term performance.
6.2.2 CQA/CQC
Standardization of CQA/CQC for caps has been adopted by the industry. The CQA/CQC effort
for subsurface barriers requires further development. Important CQA/CQC elements include:
Trench key confirmation, using samples of the key-in horizon. The trench bottom
and backfill surface should be profiled twice daily by a qualified geologist or
geotechnical engineer.
Consistent cleaning of the trench bottom and backslope to remove sediments from
the slurry trench
Controlled mixing and placing of the backfill to prevent segregation of materials
Prescribed post-construction sampling and testing, preferably before construction
demobilization, of the barrier through an approved method that preserves the
integrity of the barrier or through some proven nondestructive testing methods.
CQA/CQC for vertical barriers should be developed to a level similar to that for caps.
Preparation of construction quality assurance plans and inspection should become commonplace,
as they have been for caps. This recommendation will become increasingly important as more
innovative barrier technologies challenge CQA/CQC conventions.
6.2.3 Monitoring
The importance of a systematic monitoring program in evaluating long-term performance cannot
be overemphasized. The sampling protocol provided in Appendix C details the suggested long-
term monitoring program; monitoring recommendations also are listed below.
95
-------
Groundwater head monitoring in paired piezometers located within and outside the
wall, at a minimum spacing of 400 If along the wall alignment and within 30 feet of
the barrier wall, automatic monthly or quarterly monitoring to assist in early
detection of leaks
Systematic quarterly monitoring of groundwater quality in downgradient wells to
determine the improvement in water quality over time
Hydraulic stress tests of the barrier wall after construction as compared with
intrusive sampling and testing of the barrier to confirm the integrity of the barrier
and identify areas that may require supplemental long-term monitoring
Further development of nondestructive monitoring methods, such as geophysical
surveys or piezocone testing along the barrier wall alignment, to detect permeable
zones in the completed barrier
Collected data must be compiled and used for the purpose intended. A consistent reporting
format should be developed for all regulatory required data. Archiving of data should be done
consistently to allow future access to those data. Periodic reporting should compare required
measurements consistently so meaningful judgment can be made from the data. Data should be
cumulative and demonstrate clearly trends toward improvement or deterioration.
6.2.4 Long-Term Maintenance
Measures should be implemented to ensure the integrity of the barrier throughout its life, such as:
Access should be provided and maintained along the perimeter of the barrier for
periodic inspection
Comparative data reviews should be performed periodically (for example, at 5-year
intervals). Such reviews should address hydraulic head data, trends in groundwater
quality, as well as data from monitoring points at the key horizon. Poor performance
should trigger a pragmatic graduated response (i.e., additional monitoring, non-
destructive testing, destructive sampling and analysis, hydraulic testing, and
replacement).
96
-------
REFERENCES
American Petroleum Institute (API). 1980. Recommended Practice. 8thEd.,13B. Standard Procedure
for Testing Drilling Fluids. April. Page 41.
Bagchi, Amalendu. 1989. Design, Construction, and Monitoring of Sanitary Landfill. John Wiley &
Sons.
Barvenik, M.O.J. 1992. "Design Options Using Vertical Barrier Systems." Presented at the American
Society of Civil Engineers 1992 International Convention & Symposium, Environmental Geotech
Symposium. New York. September 14-17.
Code of Federal Regulations (CFR). 1992. 40CFR Part 264.19. Final Regulations for Construction
Quality Assurance/Construction Quality Control of Landfills. 57FR3486.
CFR. 1992. 40CFR Parts 258, 264.228, 264.310, 270.21e. Regulations for Groundwater Monitoring
and Closure of Landfills.
D'Appolonia, D.J. 1980. "Soil-Bentonite Slurry Trench Cutoffs." Journal of the Geotechnical
Engineering Division. Vol. 106. No. GT4. April.
Delvin, J.F. and B.L. Parker. 1996. Optimum Hydraulic Conductivity to Limit Contaminant Flux
Through Cutoff Walls. Groundwater. Vol. 34. No. 4. July-August.
Evans, J.C. 1994. "Hydraulic Conductivity of Vertical Cutoff Walls." Hydraulic Conductivity and
Waste Contaminant Transport in Soil. ASTM STP 1142. David E. Daniel and Stephen J. Trautwein,
Eds. American Society for Testing and Materials. Philadelphia, PA.
Evans. J.C. 1994. "Vertical Cutoff Walls." American Society of Civil Engineers, New Jersey Section.
June 15.
Evans, Jeff. 1993. Geotechnical Practice for Waste Disposal. Chapman and Hall.
Grube, W.E., Jr. "Slurry Trench Cut-Off Walls for Environmental Pollution Control." ASTM STP 1129.
Millet, R.A, J.Y Perez and R.R Davidson. 1992. "USA Practice Slurry Wall Specifications 10 Years
Later." Slurry Walls: Design Construction and Quality Control. ASTM STP 1129. David B. Paul,
Richard R. Davidson and Nicholas J. Cavalli, Eds. American Society for Testing and Materials.
Philadelphia.
Millet, Richard A., and Jean-Yves Perez. 1980. "Current USA Practice: Slurry Wall Specifications."
Journal of the Geotechnical Engineering Division. American Society of Civil Engineers. Vol. 107.
No. GTS. August.
97
-------
REFERENCES (Continued)
National Sanitation Foundation (NSP). 1991. "Standard 54, Flexible Geomembrane Liners."
Rumer, Ralph R And James K. Mitchell. 1996. "Assessment of Barrier Containment Technologies,
Chapter 12. National Technical Information Service. PB96-180583.
Rumer, Ralph R. And Michael E. Ryan. 1995. Barrier Containment Technologies for Environmental
Remediation Applications. John Wiley & Sons.
U.S. Army Corps of Engineers (USAGE). 1996. "Checklist for Design of Vertical Barrier Walls for
Hazardous Waste Sites." ETL. 1110-1-163. Department of the Army. Washington, DC.
USAGE. 1993. "Checklist for Landfill Cover Design." Department of the Army. Washington, DC.
U.S. Environmental Protection Agency (EPA). 1995. "RCRA Subtitle D (258) Seismic Design Guidance
for Municipal Solid Waste Landfill Facilities. EPA/600/R-95/051.
EPA. 1993. "Technical Guidance Document: Quality Assurance and Quality Control for Waste
Containment Facilities. EPA/600/R-93/182.
EPA. 1992. "Design, Operation, and Closure of Municipal
EPA. 1990. "Design and Construction of RCRA/CERCLA Covers."
EPA. 1989. "Requirement for Hazardous Waste Landfill Design, Construction, and Closure."
EPA/625/4-89/022. August.
EPA. 1988. "Guide to Technical Resources for Design of Land Disposal Facilities." EPA/625/6-88/018.
December
EPA. 1987a. "Investigation of Slurry Cutoff Wall Design and Construction Methods for Containing
Hazardous Wastes." EPA/600/2-87/063. August.
EPA. 1987b. "Construction Quality Control and Post-Construction Performance Verification for the
Gilson Road Hazardous Waste Site Cutoff Wall.:" EPA/600/2-87/065. August.
EPA. 1984. "Slurry Trench Construction for Pollution Migration Control." EPA/540/2-84/001.
February.
Xanthakos, P.P. 1979. Slurry Walls as Structural Systems. Second Edition.
Zappi, Mark E., Richard Shafer, and Donald D. Adrian. 1990. "Compatibility of Ninth Avenue
Superfund Site Groundwater with Two Soil-Bentonite Slurry Wall Backfill Mixtures." U.S. Army
Engineering Waterways Experiment Station. Vicksburg, MS.
98
-------
GLOSSARY
Anchor TrenchThe terminus of most geosyiithetic materials as they exit a waste containment facility,
usually consisting of a small trench in which the geosynthetic material is embedded and backfilled.
Atterberg LimitsLiquid limit and plastic limit of a soil.
Backfill SlumpSettlement of a volume of backfill mix when it is introduced into a measuring device.
Barrier UnderflowGroundwater inflow or outflow under the containment system key.
BentoniteAny commercially processed clay that consists primarily of the mineral group smecite.
CapLandfill cover system, consisting of several layers of various materials, that contains waste and
prevents infiltration of water.
Clamshell ExcavationMethod of excavating a trench that uses a bucket shaped like a clamshell.
Construction Quality Assurance (CQA)Planned system of activities that provide assurance that a
facility was constructed as specified in the design. CQA includes inspections, verifications, audits,
and evaluations of materials and workmanship necessary to determine and document the quality of
the constructed facility. CQA refers to measures taken by the CQA organization to assess whether
the installer or contractor is in compliance with the plans and specifications for a project.
Construction Quality Control (CQC)Planned system of inspections performed to directly monitor
and control the quality of a construction project. CQC is necessary to achieve quality in the
constructed or installed system. CQC refers to measures taken by the installer or contractor to
determine compliance with the requirements for materials and workmanship, as stated in the plans
and specifications for the project.
Contaminant PlumeArea of contaminated groundwater flowing downgradient of the site.
Contaminant TransportMovement of contaminants by groundwater or surface water flow.
Deep Soil MixingConstruction method in which augers are used to mix in place soils with a backfill
shirty.
Drainage LayerPortion of a landfill cap with a permeability of at least 0.01 to 1 centimeters per
second (cm/sec) that promotes the movement of liquids, usually away from the impermeable layer.
Engineered BarrierVertical barrier walls and caps that are constructed to control the inflow of water.
Feasibility DeterminationInvestigation to determine whether construction of a barrier wall is both
technically and economically feasible.
FinesPortion of soil that passes through a No. 200 sieve (openings of 0.075 millimeters).
Foundation MaterialsSoil materials used as a foundation for the layers of the cap.
99
-------
GLOSSARY (Continued)
Funnel and Gate BarrierPermeable reactive barrier that consists of a permeable curtain (gate) that
contains appropriate reactive materials, and a barrier wall (runnel) that directs the groundwater to the
gate.
Gas CollectionSystem to collect landfill gases, typically methane, produced under the cap.
Geosynthetic MaterialsGeneric term for all synthetic materials used in geotechnical engineering
applications.
Geotechnical InvestigationInvestigation of soil mechanics; rock mechanics; and the engineering
aspects of geology, geophysics, hydrology, and related services.
GradationDistribution of physical size in a granular soil.
Groundwater DewateringRemoval of groundwater from within a barrier system; generally, the water
is treated to remove contamination.
Groundwater Cutoff WallAnother term for a vertical subsurface barrier.
GroutingIntroduction of cemenitous materials in porous soil and fractured rock.
Head DifferentialDifference in water elevation within and outside the barrier wall.
Hydraulic ConductivityRate of discharge of water under laminar flow conditions through a unit
cross-sectional area of a porous medium under a unit hydraulic gradient and standard temperature
conditions.
HydrofractureFracture within a vertical barrier wall caused by earth stresses that allows groundwater
flow across the barrier.
Hydrogeologic UnitsWater-bearing geological units.
InclinometersMeasurement device to monitor the movement of soil and rock materials relative to a
fixed point located along an inclined or vertical borehole.
Key-inSection of the vertical barrier where the low-permeability barrier material intersects with in-situ
low-permeability soil or a rock formation to restrict the movement of groundwater, typically at the
greatest depth of the barrier.
Lateral FlowHorizontal movement of groundwater.
Low Permeability LayerPortion of a landfill cover, vertical barrier, or liner that restricts groundwater
flow to less than or equal to 10"7 cm/sec.
MacroporeDiscontinuity in barrier materials that allows groundwater flow.
Marsh FunnelMeasurement device used to determine the viscositv of bentonite slurry.
100
-------
GLOSSARY (Continued)
Monitoring WellGroundwater well used to measure the water level and quality in a water-bearing
horizon.
Operation and MaintenanceScheduled inspections to prevent, repair, and maintain components of a
remedial system to ensure its continued effectiveness.
Performance Monitoring DataData on groundwater head, quality, and other tests used to monitor
performance of the containment system.
PermeabilityCapacity of a material to conduct or transmit fluid.
Permeable WindowPermeable layer or area within an impermeable barrier wall.
PiezoconeType of pentrometer used to measure the field resistance of soil horizons and pure pressure.
PiezometerMonitoring point used to measure static groundwater levels.
Plastic Cement BarrierBarrier system that uses cement and plastic (a material that contains organic
polymeric substances of large molecular weight that is solid in its finished state) to form a flexible
cement barrier.
Pump and Treat SystemGeneric term used to describe the removal of contaminated groundwater and
its subsequent treatment in some type of treatment plant.
Remedial Investigation/Feasibility StudyStages of the remedial process under CERCLA during
which the nature and extent of contamination are determined and remedial action options are
developed and evaluated.
Remedial ActionLast Stage of the CERCLA remedial program, following a remedial design, during
which a permanent remedy is constructed.
Remedial Action Completion ReportsReports that describe how the remedial action was completed,
describing field changes and deviations from remedial design documents; also known as "as-built
records/'
Remedial Design DocumentsPlans that describe how the remedial action will be completed.
Sheet PileSteel or high-density polyethylene geomembrane material used to construct a vertical
subsurface barrier.
Site StratigraphyThe geologic strata or layers present at a site.
SlurrySuspension of bentonite clay and water.
Slurry Trench and BackfillConstruction method in which a backhoe or clamshell bucket is used to
excavate a trench filled with bentonite slurry; subsequently, the trench is filled with a low-
permeability backfill.
101
-------
GLOSSARY (Continued)
Slurry WallVertical subsurface barrier constructed with a bentonite slurry and other low permeability
materials.
Soil HorizonsSoil layers of various compositions.
Soil CoverLayer of landfill cap that supports vegetation.
Source ControlAny of a number of methods that can be used to control the movement of
contaminants.
Standard Industry PracticeDesign, CQA/CQC, and monitoring practices for barrier walls and caps,
as determined by work completed in this study.
Subsurface BarrierAnother term for vertical subsurface barrier.
Venting LayerLayer of a landfill cap that aids the collection and venting of landfill gas.
Vertical Subsurface BarrierEngineered barrier to restrict the horizontal movement of liquids.
Vibrating Beam MethodConstruction method that consists of an I-beam that is vibrated into the
ground and through which bentonite slurry is introduced to form an impermeable barrier wall.
Wall SloughingThe raveling of soil materials from the walls of a trench caused by instability of the
wall.
102
-------
Appendix A
Summary of Existing Subsurface Engineered Barrier
and Cap Types and Typical Construction Techniques
-------
SUMMARY OF EXISTING SUBSURFACE ENGINEERED BARRIER AND CAP TYPES
AND TYPICAL CONSTRUCTION TECHNIQUES
This appendix summarizes existing subsurface engineered barrier and cap types. The summary includes
descriptions of current technologies, applications, design considerations, and construction methods. The
information contained herein is thoroughly documented in current engineering literature.
1.0 SUBSURFACE ENGINEERED BARRIERS
Subsurface engineered barriers can be used (1) as barriers to groundwater flow, (2) to prevent off-site
migration of contaminated groundwater, and (3) to prevent on-site migration of uncontaminated
groundwater. Barriers may be circumferential or open and hanging or keyed. This section describes some
current barrier technologies in terms of particular design, construction, and performance characteristics.
The subsurface engineered barriers (walls) described in this appendix are grouped into five categories:
slurry trench barriers, grouted barriers, deep soil mixed barriers, sheet-pile walls, and treatment walls.
Slurry trench barriers were the most common barrier type identified in this study; therefore, slurry trench
barriers are discussed in greater detail than the other types. In addition, the appendix briefly describes
biopolymer drains which use barrier technology to engineer migration of groundwater.
1.1 SLURRY TRENCH BARRIERS
The most common subsurface barrier is the slurry wall. In general, slurry walls are constructed in a two-
step process. First a trench is excavated, and a slurry is placed in the trench to maintain trench stability.
When the trench is excavated to the designed depth and width, a permanent backfill material is placed in
the trench, displacing the slurry. The permanent backfill forms a hydraulic barrier. A slurry wall can be
constructed as one continuous trench or as a continuous series of panels. A bentonite-water slurry is
commonly used in slurry trenches, although a variety of slurries and backfill materials can be used. Design
considerations common to all slurry walls include the wall depth and key.
Slurry trenches can typically be excavated to depths of 50 to 80 feet using backhoes. Deeper continuous
and panel slurry trenches can be excavated using a crane-mounted drag line or clamshell bucket. Trenches
are usually 2.5 to 3 feet wide (the width of most backhoe buckets) but may be up to 5 feet wide. Unique
site or project considerations, including hydrogeology, chemical compatibility, permeability, and budget,
should be addressed in selecting the type of slurry trench to be used. The following subsections describe
the different types of slurry trench subsurface barriers.
1.1.1 Soil-Bentonite Barriers
Soil-bentonite (SB) barriers are the most common barrier type identified in this study. The backfill used
for SB barriers is 1 to 5 percent bentonite~a montmorillonitic clay that swells when hydrated-blended
with soil fill. SB barriers can reliably achieve permeabilities of 10"7 to 10"8 cm/sec. The trench is
excavated using a backhoe, dragline, or clamshell, depending on depth requirements. Figures A-l and A-2
illustrate a typical slurry wall construction site and a trench cross section, respectively.
Al
-------
Bentonite
Storage
Backfill Mixing
Area
Backhoe
Backfilled
Trench
Backfill
Placement
Area
Area of Active
Excavation
t
Proposed Line
of Excavation
Access
Road
Slurry
Preparation
Equipment
Bentonite
Storage
ooo
Water Tanks
Figure A-1
Typical Slurry Wall
Construction Site
A2
-------
o
i= LU
ffi
c>
LJJ t co
CC Z o
=J P DC
0^0
U- LU ^
m O
A3
-------
During excavation, a bentonite-water slurry typically consisting of 1 to 5 percent bentonite is placed in the
trench to support the sides of the trench. To support the trench, the slurry overcomes the active earth
pressures in the soil adjacent to the trench and forms a filter cake along the trench walls. The bentonite
slurry should be fully hydrated before being placed in the trench, typically for 12 to 24 hours. Therefore,
temporary ponds or tanks are necessary' for hydration and storage of the slurry. The slurry level in the
trench is maintained at or near the top of the trench and above the surrounding groundwater table.
Soil excavated from the trench may be used for trench backfill unless physical or contaminant
characteristics render it unsuitable. Amendments may be added to the soil to improve its physical
characteristics (for example, fines could be added to improve the gradation of a coarse soil). Bentonite-
water slurry is then mixed into the soil to form a mixture of SB backfill with a slump of approximately 4 to
6 inches. The backfill is then placed in the trench in such a way that it flows down a shallow slope (see
Figure A-l). The backfill should not be free-dropped into the trench. Appropriate placement of SB
backfill in the trench is necessary to displace the bentonite-water slurr}' without entrapping lenses of slurry
within the backfill.
1.1.1.1 Design Considerations
This subsection describes some of the criteria that should be considered during the design of an SB barrier.
In general, the design should consider site conditions, barrier requirements to meet design criteria, and
general construction requirements.
Site Conditions. Predesign investigations should include a thorough evaluation of site conditions,
including the (1) site geology and hydrogeology, (2) nature and extent of contamination, and (3)
geotechnical properties of subsurface materials. Soil borings should be drilled along the potential
alignment route or routes, and samples should be collected for geotechnical, physical, and contaminant
analyses. Groundwater modeling may also be necessary.
Site Grade. SB barriers can generally be constructed only where surface grades are less than 1 percent.
Because the slurry will flow, excavating down a slope will result in lower slurry levels within the upslope
portion of the trench, reducing trench stability. Site grades may need adjustment before construction of the
trench begins.
Site Access. The site should have adequate space available to mix, hydrate, and store the slurry, as well as
to mix backfill and place it in the trench.
Trench Stability. Design studies of trench stability should be conducted to ensure proper determination
of slurry parameters and other construction constraints.
Slurry Properties. The design evaluation of slurry mixes should establish weight, viscosity, and filtrate
loss requirements for the slurry. The weight of the slurry should be sufficient to overcome active earth
pressures in order to maintain an open trench, but the slurry must also be light enough to be displaced by
the SB backfill.
Water Quality. The quality of the water that will be used to hydrate the bentonite should be determined.
Compatibility testing of the water and bentonite may be required, although potable water supplies are
usually acceptable.
Backfill Properties. The backfill design should specify criteria for unit weight, slump, gradation, and
A4
-------
permeability. The backfill should consist of a well graded mixture of coarse and fine materials. The
backfill should typically have approximately a 4-to 6-inch slump. Laboratory studies of backfill
permeability should be conducted during the design phase to ensure that performance criteria can be met.
The backfill should be of sufficient strength to prevent hydrofracturing under high stress.
Contaminant Compatibility. Some contaminants, including some solvents and salts, have been shown to
reverse the swelling characteristics of bentonite, which would result in a higher permeability. The
compatibility of contaminated groundwater and SB backfill should be evaluated.
Integration with Other Remedy Components. If the SB barrier is one component of a multifaceted
remedy, then the design should consider the integration of the various units. For example, if a cap is to be
constructed, an appropriate interface between the SB barrier and the cap should be designed. Additionally,
the SB barrier should be able to withstand changes in site conditions caused by other units of the remedy.
such as the increased hydraulic stress resulting from use of a groundwater recovery system.
Barrier Protection. Protective measures should be taken to prevent backfill desiccation, excess surface
loading, and potential subsurface breaches.
1.1.1.2 Construction Quality Assurance and Quality Control Considerations
This subsection describes some of the criteria that should be considered during construction of an SB
barrier.
Testing of Slurry. Before being placed in the trench, the slurry should be sampled and tested for unit
weight, viscosity, and filtrate loss to ensure that these parameters meet the design requirements. The slurry
may also be tested for pH, sand content, and gel strength. Slurry samples should also be collected from the
trench during excavation. Trench slurry samples should be tested for unit weight, viscosity, and filtrate
loss. A mud balance, Marsh funnel, and filter press can be used to make field measurements of unit
weight, viscosity, and filtrate loss, respectively. These tests should be conducted on a frequent and regular
basis as specified in the site construction quality assurance (CQA) plan.
Inspection of the Trench. During construction, the trench should be inspected for width, depth, key-
penetration, verticality, continuity1, stability1, and bottom cleaning. The most critical factor is key
penetration. The excavator operator should inform the inspector when the key stratum is encountered.
Visual inspection of trench spoils can confirm that the key stratum has been encountered if the key stratum
is visibly different from the overlying material. The inspector can measure the depth to the key stratum
when it is encountered and as excavation continues to ensure mat the required penetration is made. The
depth of the trench can be measured with a rigid probe, and the measurements should be made at frequent
intervals along the trench alignment. Repeated measurements made at the same location can also reveal
accumulation of sloughed soil while the trench has remained open. The bottom of the trench should be
cleaned to ensure a tight seal between the key and the backfill. The appearance of tension cracks in the
ground surface parallel to the trench sides would indicate potential failure of the trench sides. Tension
cracks may be avoided by limiting traffic near the trench, increasing the slurry density, or minimizing the
depth to the slurry surface. For circumferential barriers, the end of the wall should be continued a certain
distance into the beginning of the wall to ensure continuity.
Testing of Backfill. Before its placement in the trench, the backfill should be sampled and tested for unit
weight, slump, gradation, and permeability to ensure mat it meets the design requirements. The backfill
unit weight should be at least 15 pounds per cubic foot greater than the slurry's weight to ensure that the
A5
-------
slurry will be displaced during backfill placement. Backfill samples should be collected during backfill
placement on a frequent and regular basis as established in the CQA plan. When the barrier is completed.
backfill samples should be collected at regular intervals and tested for permeability. This test will establish
whether the completed barrier meets the design criteria.
Handling of Contaminated Materials. At sites where contaminated backfill or slurry may be handled.
precautions should be taken to ensure that potential spills or releases are contained and recovered in order
to prevent exposure of site workers or other receptors.
1.1.2 Cement-Bentonite Barriers
Cement-bentonite (CB) slurry trench cutoff walls are excavated using a slurry composed of water, cement,
and bentonite. The bentonite-water slurry is prepared and allowed to fully hydrate before portland cement
is added. Once the cement has been added, the CB slurry is pumped to the trench. The CB slurry is left to
harden in place, forming a hydraulic barrier with a permeability on the order of 10° to 10~6 cm/sec. The
relatively high permeability is the result of the portland cement reducing the swelling properties of the
bentonite. Because of their relatively high permeabilities, CB barriers are typically not used as
contaminant containment applications, which often require permeabilities of less than 10~7 cm/sec.
However. CB barriers are commonly used as cutoff barriers where higher wall strengths are necessary and
low permeability is not required. A CB barrier is a homogenous, isotropic cutoff wall; therefore, the
likelihood of variations being present in the wall is lower than for SB barriers because no separate
backfilling step is necessary.
Alternative cement mixes have been used that display lower permeabilities and improved chemical
compatibility. Ground, granulated blast furnace slag mixed with portland cement at a ratio of 3:1 or 4:1
has displayed permeabilities of 10~7 to 10~8 cm/sec. Bentonite substitutes have also been used. One such
substitute is attapulgite, a clay mineral that is more resistant to chemical degradation man bentonite. The
use of such additives can significantly increase the overall cost of a barrier.
1.1.2.1 Design Considerations
In general, design considerations for CB barriers are similar to those for SB barriers (see Section 1.1.1.1).
Unique aspects of CB barrier design are described below.
Permeability. CB barriers typically exhibit permeability on the order of 10° to 10"° cm/sec. Because of
their relatively high permeabilities, CB barriers are typically not used for contaminant containment
applications.
Wall Strength. CB barriers have higher shear strengths than SB barriers. The hardened trench of a CB
barrier will exhibit the consistency of stiff clay. Therefore, CB barriers can be used where higher strengths
are needed.
Surface Grade. CB barriers can be constructed with steeper surface grades than can SB barriers. Grade
steps can be easily accomplished because the CB slurry hardens daily.
Site Access. Construction of CB barriers does not require as large a working area as construction of SB
barriers because backfill mixing areas are not used.
A6
-------
Construction Method. CB barriers can be constructed continuously or in panels. If a CB barrier is
constructed continuously, retarding agents may have to be added to the CB slurry in order to prevent
premature curing.
Chemical Compatibility. CB barriers typically have a cement to water ratio of 0.15 to 0.25. Therefore,
CB barriers may be susceptible to increases in permeability resulting from the effects of contaminated
fluids. The introduction of certain additives to the slurry, such as siliceous materials, special clays, or
chemical additives, may improve the chemical compatibility of CB barriers.
Handling and Disposal of Excavated Material. Materials excavated from the trench of a CB barrier will
require disposal. If the materials are contaminated, handling and disposal of the materials may represent a
significant cost factor and may require additional design considerations. If applicable, use of these
materials as site fill may be feasible.
1.1.2.2 Construction Quality Assurance Considerations
In general, CQA considerations for CB barriers are similar to those for SB barriers (see Section 1.1.1.2).
Unique aspects of CB barrier construction are described below.
Slurry Testing. The hydrated bentonite-water slurry mixture should be tested for unit weight, viscosity,
and filtrate loss before the addition of cement because the cement will begin to cure and these properties
will change. The slurry should also be tested for the same parameters after the introduction of cement, and
the test results should be compared to the design requirements. CB slurry test cylinders should be prepared
in the field, allowed to cure (typically for 28 days in a 100 percent humidity environment), and tested in
the laboratory for shear strength and permeability.
Trench Continuity. Because the CB slurry hardens in the trench, during each day of construction, a clean
contact should be established with the previous day's hardened slurry. Therefore, at the beginning of each
construction day, the end of the CB trench should be excavated to ensure a clean contact with the new CB
slurry to be added to the trench.
1.1.3 Plastic Concrete Barriers
Plastic concrete (PC) barriers are constructed under a head of bentonite-water slurry and are backfilled
with a lean concrete mix of water, cement, aggregate, and bentonite. PC barriers are usually constructed in
panels (see Figure A-3). The PC backfill is placed by tremie pipe. PC barriers can achieve permeabilities
ranging from 10"° to 10"8 cm/sec.
1.1.3.1 Design Considerations
Design considerations for PC barriers are similar to those for SB barriers (see Section 1.1.1.1) and CB
barriers (see Section 1.1.2.1). Because PC barriers are constructed under a bentonite-water slurry, the
trench and slurry design criteria are similar to those for SB barriers. As with CB barriers, PC barriers (1)
require disposal of excavated material, (2) can be constructed over steeper grades, and (3) have higher
strengths that allow their use in situations requiring structural loading. Unique aspects of PC barrier
design are described below.
A7
-------
Stop Ends
Figure 5(a)
Primary panejs tremie
concreted with stop
ends in place
Figure 5(b)
Stop ends lifted
Slurry
Figure 5(c)
Excavation of midsection
of secondary panel
Figure 5(d)
Excavation of each end
of secondary panel
^
1 Primary*
Panel
_ y- Slurry
K'PrimaryC
Panel
Figure A-3
Panel Construction Method for
a Plastic Concrete Barrier
Figure 5(e)
Tremie concreting
secondary panel
A8
-------
Strength. PC barriers are considerably stronger than SB and CB barriers. Therefore, PC barriers should
be considered for applications where additional strength is desirable.
Compatibility. The limited data available indicates that PC barriers are more resistant to organic
contamination than CB barriers.
Cost. PC barriers are more expensive than SB and CB barriers. The higher costs are the result of
(1) disposal costs associated with excavated materials; (2) panel construction, which is more time-
consuming than continuous trenching; and (3) the added cost of the aggregate used.
1.1.3.2 Construction Quality Assurance Considerations
In general, CQA considerations for PC barriers are similar to those for SB barriers (see Section 1.1.1.2)
and CB barriers (see Section 1.1.2.2). Slurry, trench construction, and backfill parameters should be
maintained in accordance with design requirements.
1.1.4 Geosynthetic Composite Barriers
Geosynthetic composite barriers have been constructed using a combination of geosynthetic materials and
conventional barrier technology. Geosynthetic materials such as high-density polyethylene (HOPE) can
improve the performance of a traditional SB barrier by (1) decreasing the permeability of the barrier by as
much as two orders of magnitude and (2) improving the chemical resistance of the barrier. Interlocking
geomembranes have been developed to improve joint seals (see Figure A-4). Hydrophillic gaskets that
expand when in contact with water can be used in the interlocks.
Current methods for installing geosynthetic materials include the following:
The geosynthetic material (geomembrane liner) is mounted on an installation frame, and
the frame is lowered into the slurry-filled trench (see Figure A-5). Interlocking joints on
either side of the geomembrane are sealed with a hydrophyllic gasket or by the slurry. The
installation frame is men removed.
The bottom of the geomembrane liner is weighted so that the liner sinks into the trench.
Hardened geomembrane panels are driven into the ground using a pile driver
Geosynthetic composite barriers have been used to prevent gas migration from landfills and to protect the
bentonite in SB barriers from moisture changes resulting from fluctuating water tables. A leak detection
system can also be constructed by sandwiching a geonet fabric between two geomembranes.
1.2 GROUTED BARRIERS
Construction of grouted barriers involves injection of a grout into the subsurface. Pressure grouting and jet
grouting are both forms of injection grouting, in which a grout mixture is injected into the pore spaces of
the soil or rock. When a vibrating beam is used, the grout is injected through a special H-pile into the
space created by the driven pile when the pile is removed.
Particulate or chemical grouts may be used for grouted barriers. Particulate grouts include slurries of
bentonite, cement, or both and water. Chemical grouts generally contain a chemical base, a catalyst, and
A9
-------
Figure A-4
Examples of Interlocks for
Goesynthetic Composite Walls
A10
-------
Soil Bentonite
Backfill
Grout At Toe (Optional)
Trench Bottom
Figure A-5
Installation Method for a
Geosynthetic Composite Wall
All
-------
water or another solvent. Common chemical grouts include sodium silicate, acrylate, and urethane.
Particulate grouts have higher viscosities than chemical grouts and are therefore better suited for larger
pore spaces, whereas chemical grouts are better suited for smaller pore spaces. Combinations of
particulate and chemical grouts can also be used.
The following subsections describe the permeation (pressure) grouting, jet grouting, and vibrating beam
technologies.
1.2.1 Permeation Grouting
Permeation grouting involves filling of soil voids with a sealing grout. To achieve low permeability, the
soil voids must be completely filled, and the lateral extent of grout penetration must be controlled. The
design of a permeation grouted barrier must consider soil permeability, grout viscosity, and soil and grout
particle size. In general, soils with permeabilities greater than 10° cm/sec can be permeation grouted with
chemical grouts, while particulate grouts can be used when soil permeabilities are greater than 10"1 cm/sec.
Two permeation grouting methods, point injection and tube-a-manchette, are currently used and are
described below.
Point Injection: The grout casing is advanced to the maximum depth of penetration, and
grout is extruded through the end as the casing is withdrawn. A continuous barrier is
fonned by installing an array of at least two or three rows of overlapping injection holes,
as shown in Figure A-6.
Tube-a-Manchette: A sleeve pipe containing small holes at 1-foot intervals is grouted into
a borehole with a weak grout. The small holes are covered by rubber sleeves (manchettes)
that act as one-way valves, allowing grout to be forced into the formation (Figure A-7). A
double packer is placed in the sleeve pipe in such a way that it straddles the manchettes,
and grout is injected under pressure. If the required containment permeability is not
achieved, the tube-a-manchette method allows for regrouting at the same location. Also,
this method allows different grout types to be used at the same location (for example, a
cement grout to fill larger voids and a chemical grout to fill smaller voids).
The design of a permeation grouted barrier must include a thorough evaluation of the grouting pressure to
be used. Excessive pressure can cause hydrofracturing. If hydrofracturing occurs, the grout will be forced
into the hydrofractures but may not adequately fill the natural soil voids, and therefore the barrier would
not meet the permeability design requirement.
1.2.2 Jet Grouting
Jet grouted walls are constructed by injecting grout at very high pressure (up to 6,000 pounds per square
inch [psi]) into the soil. The high-pressure grout is injected at rates of 800 to 1,000 feet per second, which
cuts and mixes the native soil into a uniform barrier. Typically a portland cement grout is used, although a
variety of grouts can be used. In general, a small-diameter pilot hole is drilled to the total depth of the
barrier. The hole is jet grouted from the bottom to the surface. A column of soil can be created by-
rotating the jet grouting drill rod while it is being removed, and a panel can be created by not
A12
-------
(Numbers indicate order of injection).
Figure A-6
Injection Array and Sequence
for a Permeation Grout Barrier
A13
-------
Weak cement grout in
annular space around
sleeve pipe
Rubber valve
1.5" PVC sleeve pipe
p
*
r
*
f
/
/
X
r
'
$
\
k.
Hi
^^
0
Q.
'd
e
0
Packer
r
(
^i
>
\
\
\
\
\
S
-
Figure A-7
Tube-A-Manchette
Permeation Grout Apparatus
A14
-------
rotating the rod. A horizontally continuous barrier can be created by successive installation of jet grouted
columns or panels.
Jet grouting can be used to stabilize soils ranging from gravels to heavy clays. Jet grouted barriers have
been built to depths of greater than 200 feet, although below about 100 feet the verticality and thus the
continuity of jet grouted barriers are difficult to control or confirm.
Three types of jet grouting, single rod, double rod, and triple rod (all referring to the number of
passageways in the drill rod), have been developed and are described below.
Single Rod: Only grout is pumped through the rod and injected into the soil. After the
soil-grout column is mixed, excess soil and water are displaced to the surface. Single rod
columns can be up to 1.2 meters wide in granular soils and 0.8 meter wide in cohesive
soils.
Double Rod: Air is injected into the soil through the same jet as the grout. The injected
air keeps the jet stream clear by holding back groundwater and the soil cut by the jet and
helps lift the cut soil to the surface. Double rod columns can be up to twice as large as
single rod columns. However, the high air content of a double rod column may result in
higher permeability.
Triple Rod: Air and water are injected through a cutting jet that cuts and lifts the soil,
resulting in removal of most of the native soil. Grout is injected through a second jet,
filling the column as the drill rod is lifted. The resulting column is therefore almost
entirely grout. Triple rod columns can be up to 3 meters wide in granular soils and 1.5
meters wide in cohesive soils.
Like permeation grouting, jet grouting requires that injection pressure and volume be monitored closely. If
spoil materials cannot be expelled to the surface, excess pressure can build and cause hydrofracturing.
Excessive injection volume could indicate grout loss to the formation. Jet grouting can produce large
volumes of spoil materials; if these materials are contaminated, they require special handling and disposal.
1.2.3 Vibrating Beam Technology
A vibrating beam barrier is a grouted barrier that is suitable for shallow soils. A modified H-pile is driven
into the ground with a vibratory pile driver; the H-pile has been modified to inject grout through nozzles at
the bottom end of the pile. During pile driving, a small amount of grout may be injected through the
nozzles to provide lubrication. Grout is injected through the nozzles to fill the void created by withdrawal
of the pile. A continuous barrier is created by overlapping grouted piles.
CB grouts are usually used for vibrating beam barriers, although bituminous (asphalt-based) grouts have
also been used. Vibrating beam walls are only about 2 to 3 inches thick and therefore have a high potential
to hydrofracture. The permeability of a vibrating beam wall depends on the grout used; for example, a
permeability of 10"5 to 10"6 cm/sec may be expected where a CB grout is used.
Vibrating beam barriers have two primary advantages: (1) no handling or disposal of excavated material is
required; and (2) vibrating beam barriers can be installed in relatively tight quarters. The primary
disadvantage of vibrating beam barriers is that the H-piles may deflect from vertical, making the continuity
of the barrier at depth uncertain. The bottom of a vibrating beam barrier cannot be inspected to confirm
verticality or key penetration. Also, pre-auguring may be necessary to ensure ease of penetration into
A15
-------
tighter soils.
1.3 DEEP SOIL MIXED BARRIERS
Deep soil mixing technology was developed in Japan and consists of in situ mixing of soil and a slurry.
The specially designed equipment typically consists of three auger mixing shafts that inject and mix a
water-bentonite or CB slurry into the soil as the augers are advanced, resulting in a column of thoroughly
mixed soil. As shown in Figure A-8, the final mix is usually about 1 percent bentonite. A continuous
barrier is created by overlapping penetrations.
Deep soil mixed barriers can achieve permeabilities of 10"7 cm/sec. As with a vibrating beam barrier, the
bottom of a deep soil mixed barrier cannot be inspected to confirm key penetration. However, deep soil
mixed barriers are considerably wider than vibrating beam barriers and can achieve lower permeabilities.
Because potentially contaminated materials are not excavated, the advantages of using deep soil mixing
technology include reduction of health and safety risks and elimination of costs associated with handling
and disposal of contaminated soils.
1.4 SHEET-PILE WALLS
Sheet-pile walls have long been used for a wide variety of civil engineering applications, but their use in
environmental situations has been limited. Although sheet-pile walls are extremely strong and steel will
not hydrofracture, the interlocking joints present a leakage problem. The ability of steel sheet piling to
meet a typical 10" cm/sec design performance standard depends on the type of material used to seal the
interlocking joints.
1.5 TREATMENT WALLS
The treatment wall is a relatively new technology in which a traditional barrier wall, such as a slurry trench
wall, is used to direct contaminated groundwater through a treatment zone. The groundwater moves
passively through the treatment zone, where the contaminants are degraded, precipitated, or absorbed by
the treatment media. The treatment zone may contain metal-based catalysts for degrading volatile
organics, chelators for immobilizing metals, nutrients and oxygen for microorganisms to enhance
biodegradation, or other agents. Degradation reactions break down the contaminants in the groundwater
into benign by-products. A precipitation wall reacts with the contaminants to form insoluble products that
remain in the wall as the water passes. A sorption wall adsorbs or chelates contaminants to the wall
surface.
To maintain the treatment zone reactions, parameters such as pH, oxidation/reduction potential,
contaminant or nutrient concentrations, and kinetics must be maintained within certain limits. Therefore,
the geologic, hydrogeologic, and contaminant environments must be adequately characterized.
Most treatment walls are designed to operate in situ for years with little or no maintenance. Some
treatment walls are permanent, others are semipermanent, and still others are replaceable. The long-term
stability of these walls has not been determined.
A16
-------
3rd
5th
2m
N
GOcms
1st
2nd
4th
3rd or 5th STROKE
SMW
Augers
Figure A-8
Installation Sequence for a
Deep Soil Mixed Barrier
A17
-------
1.6 BIOPOLYMER DRAINS
Biopolymer drains are briefly discussed here to illustrate the use of conventional slurry trench construction
methods to build permeable drains. Biopolymer drains are constructed using techniques similar to those
used for SB walls, but instead of being barriers of low permeability, biopolymer drains have higher
permeabilities than the surrounding strata.
During construction of the trench for a biopolymer drain, a biopolymer slurry such as a biodegradable
drilling mud is used to maintain trench stability. The trench is backfilled with a coarse, granular material
such as sand or gravel. As the mud biodegrades. the sand- or gravel-filled trench becomes a permeable
drain. Groundwater flowing into the drain can be collected by pumps or gravity-drained to a collection
point.
The advantages of installing biopolymer drains include the following:
Reduced excavation effort because trench-side backslopes and trench shoring are not
necessary
Reduced health and safety risks because workers do not have to enter the trench
Time and cost savings because construction proceeds rapidly
2.0 CAPS
Cap systems are used at landfills and other waste sites to prevent transfer of contaminants to the
atmosphere and to prevent or minimize surface water infiltration. The vegetation and topsoil over a cap
temporarily store the rainwater and ultimately evapotranspirate a large part of it. The remaining moisture
percolates downward toward the waste and must be drained laterally above a liner.
Caps are used by themselves or in conjunction with other waste treatment technologies such as barrier
walls, groundwater pump and treat systems, and in situ treatment. A cap by itself cannot prevent
horizontal flow of groundwater through the waste, only vertical entry of water into the waste. Caps are
most effective where most of the underlying waste is above the water table. Caps serve to isolate untreated
wastes and treated hazardous wastes, prevent generation of contaminated leachate, contain waste while
treatment is being applied, and control gas emissions from underlying waste. Moreover, a cap can be used
to create a land surface capable of supporting vegetation.
Caps may be temporary (interim) or final. An interim cap can be installed before final closure of the site to
minimize generation of leachate until a better remedy is selected. Caps are also used to cover waste masses
too large for treatment, such as tailings piles at mining sites. Caps can be designed to divert water away
from waste areas while minimizing erosion.
Caps have been shown to successfully contain a variety of contaminants, including volatiles, semivolatiles,
metals, radioactive materials, corrosives, oxidizers, and reducers. Storing materials containing these
contaminants in landfills does not reduce their toxicity, mobility, or volume. However, when properly-
designed and maintained, landfills can isolate contaminated wastes from human and environmental
exposure for long periods. Landfill caps and their components are expected to eventually fail, although
their effective lives can be extended by long-term inspection and maintenance. Likely cap upkeep
activities include vegetation control; construction-related repairs; and erosion, settlement, and subsidence
adjustments. Long-term repairs and maintenance can be minimized if a rigorous CQA/CQC program is
A18
-------
followed during cap construction.
The following sections discuss the basic configuration of a cap system, types of caps and materials of
construction, cap design, construction of caps, CQA and CQC for caps, cap performance, and costs of cap
systems.
2.1 BASIC CAP CONFIGURATIONS
Cap configurations range from a one-layer system of vegetated soil to a complex, multilayer system of soils
and geosynthetics. The materials used in construction of caps include low-permeability and high-
permeability soils and geosynthetic products. The low-permeability materials, such as geomembrane or
soil layers, divert water and prevent its passage into the waste. The high-permeability materials, which are
used in the drainage layer, carry away water that percolates into the cap. Other materials are used to
increase cap slope stability.
The basic layout of a multilayer cap includes (from top to bottom) (1) a surface layer, (2) a protection
layer, (3) a drainage layer, (4) a barrier layer, and (5) a gas collection layer. The surface layer usually
promotes vegetative growth and evapotranspiration and consists of topsoil (at a humid site) or cobbles (at
an arid site). Figures A-9 and A-10 show typical landfill configurations in humid and aired climates,
respectively. The protection layer is designed to protect the underlying layers from intrusion, desiccation,
and freeze-thaw damage and is usually made up of mixed soils. The drainage layer drains away infiltrating
water and is made up of sands, gravels, or geotextiles. The barrier layer minimizes infiltration of water
into the underlying waste and may direct gas to an emission control system. Barrier layers usually consist
of compacted clay or geosynthetic liners, geomembranes, or composites. The gas collection layer transmits
gas to collection points for removal or cogeneration and is usually made up of granular materials and
piping.
2.2 CAP TYPES AND MATERIALS
Natural soil drainage materials are used in the construction of the drainage layer and the gas collection
layer as well as a leachate collection layer, a leak detection layer, or drainage trenches as appropriate. Soil
drainage systems are constructed of materials that will maintain high hydraulic conductivity overtime and
resist plugging or clogging. The hydraulic conductivity of drainage materials depends primarily on the
grain size of the finest particles present in the soil. Drainage materials may also be required to serve as
filters to protect a drainage layer from plugging.
Geomembranes used in barrier layers are supplied in large rolls and are available in varying thicknesses
(20 to 140 mils), widths (15 to 100 feet), and lengths (180 to 840 feet). Most geomembranes are either
polyvinyl chloride (PVC) or polystyrenes of various densities. Geomembranes are much less permeable
than clays; geomembrane leakage is generally attributable to improper installation.
Soil barrier layers usually consist of clay that is compacted to a hydraulic conductivity of 1 x 10"6 cm/sec or
less. Compacted soil barriers are generally installed in 6-inch (or smaller) lifts to achieve a thickness of 2
feet or more. Composite barriers use both soil and geomembrane; the geomembrane is essentially
impermeable, but if a leak develops, the soil component prevents significant infiltration. Composite
barriers (liners) have proven to be the most effective in decreasing hydraulic conductivity.
A19
-------
Vegatation/soil _
top layer
/ /
Drainage layer
Low hydraulic
conductivity _
geomembrane/
soil layer
Waste
o
C?
O
60
cm
30
cm
60
cm
Filter layer
0.5-mm
(20-mil)
geomembrane
Figure A-9
Typical Landfill Cover
Design in Humid Regions
A20
-------
Cobbles/soil
top layer
DIOIIC Darner
(cobbles)
Drainage layer
Low hydraulic
i
conductivity.
n^nmQmhran^/
-i i
en avpr
Gas vent layer
Waste
, LTi r^ffi^*"^*!*^*^ -
*
TririrriiiT
00« Ot« 00 1 0-« 0^ o^* 0
>"1 /*V O O*N o.
<7 O Q «s>
06^0?
60
cm
^
-^
30
cm
en
DU
rm
30
cm
Geosynthetic filter
Geosynthetic filter
0.5-mm (20-mil)
geomembrane
Geosynthetic filter
Figure A-10
Typical Landfill Cover
Design in Arid Regions
A21
-------
Geosynthetic clay barriers, which are composed of a thin layer of bentonite between two geosynthetic
materials, can be used in place of both geomembrane and clay components. The bentonite expands to
create a low-permeability, resealable barrier that is self-healing.* The geosynthetic clay barrier material
is supplied in rolls, but unlike geomembranes, it does not require seaming.
Other barrier materials include flyash-bentonite-soil mixtures, super-absorbent geotextiles, sprayed-on
geomembranes and soil-particle binders, and custom-made bentonite composites made with geomembranes
or geotextiles. These materials have various potential advantages, including quick and easy installation,
better quality control, cost savings, reduction in the volume of material used, use of lighter construction
equipment, and some self-healing capabilities.
2.3 CAP DESIGN
Major design factors that influence the effectiveness of a cap include (1) determination of global waste
stability, (2) settlement analysis, (3) analysis of the stability of the cap system, (4) drainage analysis, (5)
leachate management analysis, and (6) gas management analysis. Determination of global waste stability
involves evaluating whether the waste mass will remain stable under all potential loading conditions.
Waste mass stability is analyzed under several different loading conditions, including cap loads, seismic
stresses, and construction loading. Settlement analysis evaluates the potential for the foundation and waste
materials to consolidate under the loading conditions of a cap. Long-term settlement analysis should be
conducted, and a monitoring plan should be developed to measure any settlement.
The stability of the cap system itself should be analyzed to determine its potential for failure. Analyses
should address (1) interface stability, (2) cap soil tension above a geosynthetic-lined slope, (3) various
stresses within cap components, (4) the impact of differential settlement on geosynthetic materials, and (5)
the stability of slopes steeper than 3 horizontal to 1 vertical.
A drainage analysis is necessary to prevent buildup of hydraulic head on the nonpenneable cap. Drainage
analyses should address (1) drainage capacity, (2) geotextile filtration, (3) runoff control, and (4) erosion
control. Leachate management analysis involves evaluation of the leachate generation rate and the
collection and transmission system to prevent buildup of leachate within the waste. Gas management
analysis consists of an evaluation of the proposed gas collection well design and placement and usually
includes a pilot test to determine the gas composition and generation rate. If a passive venting system is
required, the locations of lateral and vertical vents are determined on the basis of waste characteristics (see
Figure A-l 1). Other design considerations that affect the performance of the cap include frost penetration
of cover soils and the puncture vulnerability of geosynthetic materials.
2.4 CAP CONSTRUCTION
Caps are usually constructed in a domed shape to enhance runoff. The base layer, which may be a gas
collection layer, overlies the waste mass. The clay component of the barrier layer is constructed over this
base layer. The clay is spread and compacted in lifts a few inches thick until the desired thickness is
achieved. Each lift is scarified (roughed up) following compaction to remove any trace of a surface
between it and the next higher lift. The top lift is compacted and rolled smooth so mat the geomembrane
can be laid on it with direct and uniform contact. The clay's optimum moisture content must be
maintained during construction to provide the necessary low permeability upon compaction.
A22
-------
CD
>»
CO
±S "O
)^ CD
i- ^_.
92 o
0-3
oo
A23
-------
A geomembrane should be laid without wrinkles or tension. Its seams should be fully and continuously
welded or cemented, and it should be installed before the underlying clay surface can desiccate and crack.
If vent pipes are present, they should be carefully attached to the geomembrane in order to prevent tearing
as a result of subsidence. Punctures and tears should be avoided during the geomembrane handling and
installation procedure. In addition, the effects of air temperature and seasonal variations on the
geomembrane should be taken into account; stiffness and brittleness are associated with low air
temperatures. If installed while the air temperature is high, the geomembrane will expand and can then
shrink to the point where seams rupture.
A geotextile may be laid on the surface of the geomembrane to protect it, particularly if coarse and sharp
granular materials are to be used in the overlying drainage layer. The drainage layer is designed to carry-
away water that percolates down to the barrier layer. The drainage layer may be either a granular soil with
high permeability or a geosynthetic drainage grid or geonet sandwiched between two porous geotextile
layers. Another geotextile may be put on top of the drainage layer to prevent clogging of the drainage layer
by soil from above. Fill soil and topsoil are then applied, and the topsoil is seeded with grass or other
vegetation.
CQA/CQC, including testing, is estimated to increase the cap installation cost and completion time by 10
to 15 percent but is generally acknowledged as improving the performance of the cap. Materials
considered for each of the cap components are tested to ensure their suitability. Both before and during
construction, soils are tested to determine their grain size, Atterberg limits, hydraulic conductivity, and
compaction characteristics. Geomembrane test strip seaming is performed on narrow pieces of excess
geomembrane to ensure high seam quality. The test strips are subjected to strength (shear and peel) testing
to simulate stress from equipment, personnel, or climatic changes. Seams should run up and down slopes
rather than across them to reduce seam stress.
The slope of the cap top should be between 3 and 5 percent. Steeply mounded caps can present difficulties
related to soil compaction, soil erosion, and anchorage of the geomembrane. High air temperatures and dry
conditions during construction may result in loss of moisture from a clay barrier layer, causing desiccation
cracking that can increase hydraulic conductivity. Desiccation cracking can be prevented by adding
moisture to the clay surface and key installing the geomembrane in a composite barrier quickly after
completion of the clay layer.
Construction equipment required during cap installation includes bulldozers, graders, various rollers, and
vibrator}' compactors. Additional equipment is needed for moving, placing, and seaming geosynthetic
materials. Storage areas are needed for the materials used in the cap. If site soils are not adequate for use
in cap construction, other, low-permeability soils have to be trucked in. Water supplies need to be
adequate to ensure that soils used in construction maintain their optimum soil density.
2.5 CAP CONSTRUCTION QUALITY ASSURANCE AND CONSTRUCTION QUALITY
CONTROL
CQA/CQC is widely recognized as being critically important to overall quality management for waste
containment facilities. Preparation of the best designs and adherence to regulatory requirements do not
necessarily result in a superior cap system unless the cap is properly constructed. Additionally, when
geosynthetic materials are to be used, manufacturing quality assurance and manufacturing quality control
of the geosynthetic products are extremely important.
Cover soil and subgrade soil should be classified using American Society for Testing and Materials
(ASTM) Method D-2488 to ensure that the soils have adequate plasticity. Adequately classifying soils is
A24
-------
best accomplished by observing all excavations of borrow soil from the borrow pit. Borrow soil should
also be periodically evaluated using compaction curves as well as testing of soil density and moisture
content.
The soil used to construct the drainage layer should be periodically evaluated to determine its grain size,
hydraulic conductivity, and carbonate content. Care should be taken during placement of the soil drainage
layer to ensure that underlying materials are not punctured, and that fine-grained soil is not accidentally
mixed with drainage material.
CQA/CQC processes for soil liners are intended to ensure that (1) soil liner materials are suitable, (2) soil
liner materials are properly placed and compacted, and (3) the completed liner is properly protected. Clay
prequalification testing is accomplished through periodic soil classification, generation of compaction
curves, and remolded permeability tests. A test pad should be used to demonstrate that the materials and
methods proposed will result in construction of a liner with the required large-scale, in situ hydraulic
conductivity. Construction testing of the soil liner should include generation of compaction curves as well
as testing of density, moisture content, undisturbed hydraulic conductivity, Atterberg limits, grain size,
and undisturbed dry density.
Manufacturing quality control testing should be performed on geomembrane barriers to ensure that the
geomembrane complies with National Science Foundation (NSF) 54 standard and displays the desired melt
index, resin index, and resistance to environmental stress cracks. Field testing should be periodically
conducted to evaluate geomembrane thickness, tensile strength, elongation, and resistance to punctures and
tears. CQA inspections should be performed for every roll of geomembrane at the site. Seam testing
should be performed using nondestructive (vacuum box or air pressure) and destructive (peel and shear
testing) methods.
2.6 CAP PERFORMANCE
Cap systems usually perform well and require minimal maintenance. However, the impact of differential
settlement and the stability of slopes steeper than 3 horizontal to 1 vertical should be monitored quarterly
or semiannually. Additional monitoring of leachate quality, quantity, and leak detection; water infiltration;
gas quality and quantity, and groundwater will ensure that the cap system is performing as intended.
2.7 CAP COSTS
The cost of a cap (0.5 to one acre) can vary from $500,000 for a one-layer system to several million dollars
for a multilayer cap. The cost is highly dependent on the local availability of soils suitable for construction
and the requirements for monitoring, leachate collection, and gas collection.
A25
-------
REFERENCES FOR FIGURES
Figure A-2 U.S. Environmental Protection Agency. Slurry Trench Construction for Pollution
Migration Control. EPA/540/2-84/001. February.
Figure A-5 Ryan, C.R. 1987. Vertical Barriers in Soil for Pollution Containment. Geotechnical
Practice for Waste Disposal '87. Geotechnical Special Publication No. 13. Edited by
Richard D. Woods. ASCE. Pages 182-204.
Figure A-6 Rumer, R.R. and M. E. Ryan. 1995. "'Barrier Containment Technologies for
Environmental Remediation Applications/' John Wiley and Sons. New York.
Figure A-7 Karol, R.H. 1990. "Chemical Grouting." Marcel Dekker. New York.
Figure A-8 Ryan, C.R. 1987. Vertical Barriers in Soil for Pollution Containment. Geotechnical
Practice for Waste Disposal '87. Geotechnical Special Publication No. 13. Edited by
Richard D. Woods. ASCE. Pages 182-204.
A26
-------
Appendix B
Site Summaries
This appendix gives more detailed descriptions of the 36 sites evaulated in the retrospective report.
This is a separate document.
A limited number are available from NCEPI.
Please reference document EPA-542-R-98-005a.
-------
Appendix C
Field Protocol
-------
1.0 INTRODUCTION
This appendix sets forth a proposed approach to the preparation of a detailed protocol for field study of
sites with subsurface engineered barriers. This discussion also summarizes the results of the performance
evaluation report for subsurface barriers, as those results apply to current monitoring of such barriers, and
puts forth a general approach to monitoring existing subsurface barriers for performance as containment
systems.
Results of the evaluation of engineered barriers revealed that these barriers are monitored as containment
systems, not as single entities within a containment system. The performance of a barrier wall therefore
generally is not monitored singly, but the containment system is monitored for overall performance in
meeting the design objective of a remedial system. The evaluation also showed that monitoring plans
vary for passive and active containment systems. These systems are summarized below.
Prevention or cutoff of outflow from the containment system (that is, stop or slow the rate of flow
of the contaminant plume, for example, by installation of a vertical barrier). A containment
system that uses such a design, with no associated groundwater extraction, can be referred to as a
passive containment system.
Maintenance of the entombment of the contaminants (that is, prevent the flow of contaminants
from the system, for example, by the use of active containment measures like maintenance of a
slight hydrologic flow through active groundwater extraction). A containment system that uses
such a design is referred to as an active containment system.
Monitoring systems evaluate the functional performance of the containment systemthat is, whether the
containment system is performing the function that it was designed to perform. The study concluded mat,
at all the sites studied, once the containment barrier had been constructed and had become operational,
performance monitoring consisted of collecting hydraulic head or groundwater quality data or a
combination of the two. Additional testing was not completed unless problems in the performance of a
barrier containment system were indicated. Diagnostic stress tests seldom were conducted, typically
only when a problem affecting the containment system was suspected. Tests for long-term degradation,
physical sampling for diffusion through barriers, and tests of the effect of desiccation on permeability also
were performed very seldom at the sites studied. In addition, the study concluded that monitoring the
performance of an engineered barrier is dependent on the conditions at the site, in particular, on the
hydrogeological characteristics of the site. The site-specific nature of performance monitoring makes the
establishment of a standard monitoring protocol difficult and impractical. However, performance
monitoring can be made more consistent through the establishment of a general approach to monitoring.
The following sections put forth a monitoring or sampling protocol for engineered barrier containment
systems. The protocol is presented in two categories. First, the consistent use and expansion of existing
technology are discussed. Second, a view of more innovative technologies is presented. The application
and field testing of the developed protocol men are reviewed.
Cl
-------
2.0 PROPOSED PERFORMANCE MONITORING PROTOCOL
A review of the 36 sites studied showed significant variation in monitoring methods used to assess the
performance of containment systems. That variation is discussed in greater detail in Section 3.4 of the
report. An even greater inconsistency was identified in the reporting and tracking of performance
monitoring data from the sites studied. The following section presents recommendations for the
collection of performance monitoring data and the systematic review of those data. Figure C-l, Barrier
Wall Containment System Performance Monitoring Decision Tree, presents a step-by-step general
approach to determining the functional performance of a barrier wall containment system. The steps
outlined in the figure are described below. Table C-l presents a summary of studies recommended for
the evaluation of the performance of a containment system. Table C-l depicts the continuation of the
process illustrated in Figure C-l: if the integrity of a barrier wall containment system is in question.
integrity testing of barrier wall components should be completed. Table C-l summarizes the tests and
evaluations that should be completed for assessing the integrity of a barrier wall containment system.
Below is a discussion of the recommended protocol for assessing the performance of existing engineered
barrier containment systems. The discussion below is general in nature and is intended to outline the
methods used to assess performance. As stated earlier, site-specific hydrogeological characteristics
always should be considered before any containment and performance monitoring program is
implemented. The appropriateness of monitoring techniques is site-specific and application-specific.
Therefore, other than the use of hydrological monitoring techniques (monitoring of groundwater quality
and head differential) and stress or pump testing, neither nondestructive nor destructive monitoring
techniques are routinely recommended to assess the performance of barrier containment systems.
2.1 STEP 1: MONITORING OF GROUNDWATER QUALITY AND HEAD
DIFFERENTIAL
Monitoring of groundwater quality and head differential is the recommended method of assessing the
functional performance of barrier containment systems. As seen in the evaluation study, monitoring of
groundwater quality can range in complexity from very little monitoring (frequency of monitoring,
number of monitoring points or wells, and analysis completed) to significant monitoring efforts. The type
of containment system (active or passive) does not have a bearing on the complexity of the monitoring
program. The complexity of the monitoring program is dictated by the hydrogeological characteristics of
the site.
It is recommended that, for passive or partially active containment sites, monitoring of groundwater
quality be used to assess the performance of a barrier wall containment system. For active sites,
groundwater head differentials should be the primary element monitored to assess performance of the
containment system.
Monitoring of Groundwater Quality
The first step in assessing the functional performance of a barrier wall containment system is monitoring
the quality of groundwater outside the containment barrier. The effectiveness of a groundwater
monitoring network in assessing the performance of a containment system depends on hydrogeological
conditions at the site; therefore, no universal approach to the placement of a monitoring network is
recommended. However, it is recommended that the location of monitoring wells for the assessment of
groundwater quality be based upon a probabilistic approach to compliance monitoring. In addition, flow
and transport mechanisms should be evaluated to assist in establishing the minimum necessary number
and locations of monitoring points. Nests of monitoring wells, set at various depths in different strata,
located close to the barrier system also should be used in identifying underflow or downward flow
conditions that may allow the contaminants to migrate from the containment system.
C2
-------
Figure C-l
Barrier Wall Containment System
Monitor for groundwater
quality and hydraulic head.
Continue groundwater
monitoring until
steadv-state is reached.
Is active
containment
(pump and test)
being conducted?
Monitor for groundwater
quality.
Yes
Complete stress test on wall
to confirm wall integrity.
Complete other tests and
repairs, as necessary.
Yes
Is hydraulic \ Yes
head gradient meeting
design objective'.
Is there
evidence of
containment
system breach'-
Yes
Continue groundwater
monitoring.
No
Complete flow and
transport modeling.*
Complete flow and
transport modeling.*
' No
Re-evaluate pump
system design.
Complete groundwater
modeling study.
Yes
*Note:
Flow and transport modeling indicates the natural decay
of contaminants outside the barrier wall.
Has steady-state been
established?
groundwater qualit
decreased
downeradient?
Has
containment beei
achieved as indicated
by flow and transpoi
modeling*'/
Complete stress testing on
wall to confirm wall
integrity.
Continue groundwater
monitoring until steady-
state is reached.
Continue groundwater
quality monitoring.
Re-evaluate containment
system design and complete
necessary testing and
repairs.
C3
-------
Table C-l
Summary of Studies Commonly Used
to Evaluate the Performance of a Containment System
Barrier Walls
I Evaluation of Barrier Wall Integrity
Reevaluate the ground-water modeling completed for design of the containment system.
Complete nondestructive tests that could include one or more of the following: cone
penetration (piezocone soundings), cross-hole seismic surveys, ground-penetrating radar,
seismic surveys, tomography, surface-based seismic refraction survey, or dye tracing.
Conduct destructive sampling of the barrier wall, such as destructive tests through boreholes
or other means, to measure (1) degradation of components of the barrier wall by contaminants
and (2) hydraulic conductivity.
II Evaluation of Containment Leaks
Review CQA/CQC information and groundwater monitoring data.
Conduct stress testing by means of pumping.
Perform modeling of hydraulic efficiency and evaluation of hydraulic heads inside and
outside the barrier wall.
Evaluate groundwater quality through selection of a groundwater monitoring network and
selection of indicator contaminants.
Ill Evaluation of Key-in of the Containment System
Profile the bottom of the barrier wall.
Compare data obtained from evaluation of leaks in the containment system to identify any
leaks through the key-in.
C4
-------
Groundwater quality should be monitored or reviewed regularly, with weekly monitoring recommended,
until the system has reached a steady-state condition. Once a steady-state condition has been reached,
monitoring of groundwater quality should continue less frequently (for example, quarterly or
semiannually). Tolerance of outlying data points and length of time periods for attaining cleanup levels
should be allowed following the attainment of a steady-state condition of the containment system.
Statistical trend analysis may be applicable for assessing the performance of the containment system.
Monitoring of Groundwater Head Differential
For active containment systems, monitoring of groundwater head differentials between piezometers inside
and outside the barrier should be completed to assess functional performance. Monitoring of groundwater
head differentials, along with groundwater quality, is the first step set forth in Figure 1 for the
performance monitoring of active containment systems.
Interior wall or barrier monitoring wells should be used at active containment system sites to better
evaluate the effect of pumping drawdown, if drawdown is occurring. To achieve efficient cross-barrier
monitoring, monitoring well points should be spaced to approximate a line source. Equal spacing of
monitoring wells or piezometers across the barrier has been shown to be the most efficient design for
monitoring of head for active containment systems, but is less crucial for passive containment systems.
However, for passive containment systems, interior water levels must be monitored to track groundwater
flow through or under the vertical barrier. Again, fate and transport modeling should be used to assist in
establishing the minimum necessary number and locations of monitoring points.
Monitoring groundwater head conditions at the bottom of the containment system, confining unit, or rock
stratum, at or immediately underlying the containment system, is recommended. Such monitoring will
help identify the presence of vertical gradients along the alignment of the subsurface barrier.
To better understand the influence of the barrier system on groundwater flow patterns, it is recommended
that gathering of data begin during the design phase for the barrier containment system and continue
through construction. Measurements of groundwater head should be collected frequently during and
immediately after construction of the containment system, with frequency increasing from weekly to
monthly as the containment system reaches a steady-state condition. Head monitoring can be performed
less frequently when data values have become stable or steady. Monitoring of groundwater head through
wells currently is the most widely used method of monitoring head; that approach is recommended.
However, automatic data reporting may prove to be useful and, over the long term of site remediation,
may prove to be more cost-effective.
2.2 STEP 2: HYDRAULIC TESTING
Step 2 in determining the performance of a containment system is to complete a hydraulic stress test. A
stress test is recommended when a steady-state condition has not been reached within an acceptable time
frame, based on design or fate and transport modeling data, or when design flaws or construction
problems are identified or there is other evidence of a breach of the containment system (for example, a
decrease in water quality downgradient of the system or changes in head differentials) that are not
attributed to changes in the active containment system (for example, a decrease in pumping rates).
Deficiencies in a containment system can become amplified when the contained area is stressed during
pumping. When combined with cross-barrier monitoring of groundwater quality and head, stress tests
can identify definitively leaks in the vicinity of the zone of influence. When combined with multilevel
monitoring, stress testing of a vertical barrier can assess underflow effectively within a region of
influence.
In addition to internally performed stress tests, stress tests can be performed by pumping outside the
containment system. The benefits of external stress testing are the reduction of the concentrations of
contaminants in the extracted water and the minimization of disposal requirements. It is recommended
that, for newly constructed sites, stress testing be performed before or as part of the startup of any
pumping system. Such testing will aid in the reporting and future comparison of data to be used in
C5
-------
evaluating the performance of the barrier. In addition, it is recommended that periodic reviews (for
example, a five-year regulatory review of a remedial remedy) include pumping tests or stress tests and
that comparisons be made to evaluate changing hydraulic properties of the barrier (for example,
degradation of components of the barrier system overtime).
2.3 STEP 3: OTHER PERFORMANCE MONITORING METHODS
Once a stress test has been completed, hydraulic efficiency modeling of the head differentials, both inside
and outside the barrier wall, should be completed. If the results of the hydraulic efficiency modeling
indicate that the integrity of the containment system has been breached, destructive and nondestructive
testing of barrier components should be considered. Below is a list of current and evolving monitoring
techniques and technologies that can be used to monitor the performance of a containment system.
It is recommended that destructive and nondestructive testing be completed simultaneously on
containment systems. Physiochemical processes (deterioration of the components of the containment
system) that can lead to failure of the system can be quantified more effectively through destructive
sampling and testing than through nondestructive testing.
Destructive Tests
Destructive sampling of a barrier containment system is recommended to measure degradation of
components of the barrier. Collection of samples of test borings, undisturbed samplers, and other
methods should be implemented. Samples collected should be tested for reconstituted permeability,
gradation, moisture content, and chemical compatibility with components of the system. Determination
of the location and number of samples to be collected should be based on site-specific information. Data
obtained from destructive testing should be evaluated with the results of nondestructive testing.
Nondestructive Tests
Nondestructive tests that could be used to evaluate the integrity of a containment system include those
listed below. (For more detail about the principles of operation of the technologies listed and their
application and limitations, the reader is referred to Rumer and Mitchell (1996).)
Geophysical Testing Systems: Geophysical techniques are based on the response of geomedia
and pore fluids to the electromagnetic spectrum, seismic or acoustic energy, magnetic fields, or
gravitational fields. Geophysical techniques measure the physical and chemical characteristics of
geomedia and pore fluids that may change with chemical contamination or development of
internal voids. Geophysical methods mat can be used to assess the continuity of natural and
emplaced barriers include: (1) cross-hole seismic imaging; (2) surface seismic refraction;
(3) ground penetrating radar; and (4) microwaves, ultrasound, and radio waves. Geophysical
techniques that are useful in tracking the extent of contaminant plumes include:
(1) electromagnetic resistivity and (2) ground penetrating radar.
Electrochemical Systems: Electrochemical sensing systems monitor the changes in the
physiochemical characteristics of the sensor caused by contact with a fluid.
Electrical Systems: These techniques use electric current impulses to monitor subsurface media
or physical interactions between the embedded device and the surrounding media.
C6
-------
Table C-2
General Application of Monitoring Approaches
\ Monitoring
\^ Method
Barrier ^v
Monitoring ^v
Approach \
1 . Barrier
Integrity
2. Barrier
Permeation
Monitoring
3. External
Monitoring
Monitoring Well
Network
N/A
N/A
C
Geophysical
Methods
G
R
G
Electro-
Chemical
Methods
R
G
G
Mechanical and
Electro-
Chemical
Methods
C
R
R
Electrical
Methods
G
C
G
Key: C=conventional, G=growing application. R=rare, N/A=Not applicable
Note: These methods comprise several specific techniques, some of which may not necessarily fit into these three
categories
Many of the technologies described are innovative in nature and require further refinement to achieve
spatial resolutions necessary for containment system monitoring programs. Table C-2 presents a general
application of approaches to performance monitoring.
2.4 REPORTING PROTOCOLS
Consistent and accurate reporting of monitoring data is vital to an accurate assessment of the performance
of a containment system. It is recommended that a reporting format be established early in the
performance monitoring program or, if possible, during design of the monitoring program. Establishment
of a consistent reporting format will allow the comparison of data over time, and the consistency of the
format will facilitate the identification of changes or trends in the barrier wall containment system that
could lead to functional failure.
C7
------- |