Industrial Waste
Management
-------
This Guide provides state-of-the-art tools and
practices to enable you to tailor hands-on
solutions to the industrial waste management
challenges you face.
WHAT'S AVAILABLE
• Quick reference to multimedia methods for handling and disposing of wastes
from all types of industries
• Answers to your technical questions about siting, design, monitoring, operation.
and closure of waste facilities
• Interactive, educational tools, including air and ground water risk assessment
models, fact sheets, and a facility siting tool.
• Best management practices, from risk assessment and public participation to
waste reduction, pollution prevention, and recycling
-------
^DGEMENTS
The fotawng members of the Industrial Waste Focus Group and the Industrial Waste Steering Committe aregrateMy
acknowledged far al of their time ana assistance in the development of this guidance document
'ICUS
.
r-oui own, nie isun viicamuti
Company
Walter Carey. Nestle USA Inc and
New Milford Farms
Rama Chaturvedi Bethlehem Steel
Corporation
H.C. Clark. Rice University
Barbara Dodds. League of Women
Voters
Chuck Feerick. Exxon Mobil
Corporation
Stacey Ford. Exxon Mobil
Corporation
Robert Giraud DuPont Company
John Harney, Citizens Round
Table/PURE
Kyle Isakower. American Petroleum
Institute
Richard Jarman, National Food
Processors Association
James Meiers, Cinergy Power
Generation Services
Scott Murto. General Motors and
American Foundry Society
James Roewer, Edison Electric
Institute
Edward Repa. Environmental
Industry Association
Tim Saybr, International Paper
Amy SchaRer, Weyerhaeuser
Ed Skemote. WMX Technologies. Inc
Michael Wach Western
Environmental Law Center
David Wels, University of South
"*—«• Medical Center
rat fewin. Cherokee Nation of
Oklahoma
rocu?.
wu wom.nu. ^».u uiuu
Brian Forrestal. Laidlaw Waste
Systems
Jonathan Greenberg. Browning-
Ferris Industries
Michael Gregory, Arizona Toxics
Information and Sierra Club
Andrew Mites, The Dexter
Corporation
Gary Robbins, Exxon Company
Kevin Sail. National Paint & Coatings
Association
Bruce Sterer. American Iron & Steel
Lisa Williams. Aluminum Association
arid Territorial Solid Waste" "
Management Officials
Marc Crooks. Washington State
Department of Ecology
Cyndi Darling. Maine Department of
Environmental Protection
Jon Dilliard Montana Department of
Environmental Qualty
Anne Dobbs. Texas Natural
Resources Conservation
Commission
Richard Hammond. New York State
Department of Environmental
Conservation
Elizabeth Haven California State
Waste Resources Control Board
Jim HuD. Missouri Department of
Natural Resources
Jim Knudson. Washington State
Department of Ecology
Chris McGuire. Florida Department
of Environmental Protection
Gene Mitchell Wisconsin
Department of Natural Resources
William Pounds. Pennsylvania
Department of Environmental
Protection
Bjjan Sharafkhani Louisiana
Department of Environmental
Qualty
James Warner, Minnesota Pollution
Control Agency
railKHa l*ujiit, mame LmpwUlRitlt Of
Environmental Protection
NormGumenik Arizona Department
of Environmental Qualty
Steve Jenkins, Alabama Department
of Environmental Management
Jim North Arizona Department of
Environmental Qualty
-------
Industrial waste is generated by the production
of commercial goods, products, or services.
Examples include wastes from the production
of chemicals, iron and steel, and food goods.
-------
EPA 542-R-00-006
June 2000
Abstracts of Remediation
Case Studies
Volume 4
Federal
Remediation
Technologies
Roundtable
Prepared by the
Member Agencies of the
Federal Remediation Technologies Roundtable
-------
Abstracts of Remediation
Case Studies
Volume 4
Prepared by Member Agencies of the
Federal Remediation Technologies Roundtable
Environmental Protection Agency
Department of Defense
U.S. Air Force
U.S. Army
U.S. Navy
Department of Energy
Department of Interior
National Aeronautics and Space Administration
Tennessee Valley Authority
Coast Guard
June 2000
-------
NOTICE
This report and the individual case studies and abstracts were prepared by agencies of the U.S.
Government. Neither the U.S. Government nor any agency thereof, nor any of their employees, makes
any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,
completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents
that its use would not infringe privately-owned rights. Reference herein to any specific commercial
product, process, or service by trade name, trademark, manufacturer, or otherwise does not imply its
endorsement, recommendation, or favoring by the U.S. Government or any agency thereof. The views
and opinions of authors expressed herein do not necessarily state or reflect those of the U.S. Government
or any agency thereof.
Compilation of this material has been funded wholly or in part by the U.S. Environmental Protection
Agency under EPA Contract No. 68-W-99-003.
11
-------
FOREWORD
This report is a collection of abstracts summarizing 78 case studies of site remediation applications
prepared by federal agencies. The case studies, collected under the auspices of the Federal Remediation
Technologies Roundtable, were undertaken to document the results and lessons learned from technology
applications. They will help establish benchmark data on cost and performance which should lead to
greater confidence in the selection and use of cleanup technologies.
The Roundtable was created to exchange information on site remediation technologies, and to consider
cooperative efforts that could lead to a greater application of innovative technologies. Roundtable
member agencies, including the U.S. Environmental Protection Agency, U.S. Department of Defense,
and U.S. Department of Energy, expect to complete many site remediation projects in the near future.
These agencies recognize the importance of documenting the results of these efforts, and the benefits to
be realized from greater coordination.
The case study reports and abstracts are organized by technology in a multi-volume set listed below. The
78 new case studies are available on a CD-ROM, and cover a variety of in situ and ex situ technologies.
Remediation Case Studies, Volumes 1-13, and Abstracts, Volumes 1-3, were published previously, and
contain 140 projects, and are also available on the CD-ROM. Abstracts, Volume 4, covers a wide variety
of technologies, including full-scale remediations and large-scale field demonstrations of soil and
groundwater treatment technologies. In the future, the set will grow as agencies prepare additional case
studies.
2000 Series
Published on CD-ROM, FRTR Cost and Performance Case Studies and Related Information,
EPA-542-C-00-001; June 2000
1998 Series
Volume 7: Ex Situ Soil Treatment Technologies (Bioremediation, Solvent Extraction,
Thermal Desorption), EPA-542-R-98-011; September 1998
Volume 8: In Situ Soil Treatment Technologies (Soil Vapor Extraction, Thermal Processes),
EPA-542-R-98-012; September 1998
Volume 9: Groundwater Pump and Treat (Chlorinated Solvents), EPA-542-R-98-013;
September 1998
Volume 10: Groundwater Pump and Treat (Nonchlorinated Contaminants), EPA-542-R-98-
014; September 1998
Volume 11: Innovative Groundwater Treatment Technologies, EPA-542-R-98-015;
September 1998
Volume 12: On-Site Incineration, EPA-542-R-98-016; September 1998
Volume 13: Debris and Surface Cleaning Technologies, and Other Miscellaneous
Technologies, EPA-542-R-98-017; September 1998
in
-------
1997 Series
Volume 5: Bioremediation and Vitrification, EPA-542-R-97-008; July 1997; PB97-177554
Volume 6: Soil Vapor Extraction and Other In Situ Technologies, EPA-542-R-97-009;
July 1997; PB97-177562
1995 Series
Volume 1: Bioremediation, EPA-542-R-95-002; March 1995; PB95-182911
Volume 2: Groundwater Treatment, EPA-542-R-95-003; March 1995; PB95-182929
Volume 3: Soil Vapor Extraction, EPA-542-R-95-004; March 1995; PB95-182937
Volume 4: Thermal Desorption, Soil Washing, and In Situ Vitrification, EPA-542-R-95-
005; March 1995; PB95-182945
Abstracts
Volume 1: EPA-542-R-95-001; March 1995; PB95-201711
Volume 2: EPA-542-R-97-010; July 1997; PB97-177570
Volume 3: EPA-542-R-98-010; September 1998
Volume 4: EPA-542-R-00-006; June 2000
Accessing Case Studies
The case studies and case study abstracts also are available on the Internet through the Federal
Remediation Technologies Roundtable web site at: http://www.frtr.gov. The Roundtable web site
provides links to individual agency web sites, and includes a search function. The search function allows
users to complete a key word (pick list) search of all the case studies on the web site, and includes pick
lists for media treated, contaminant types, and primary and supplemental technology types. The search
function provides users with basic information about the case studies, and allows them to view or
download abstracts and case studies that meet their requirements.
Users are encouraged to download abstracts and case studies from the Roundtable web site. Some of the
case studies are also available on individual agency web sites, such as for the Department of Energy.
In addition, a limited number of hard copies are available free of charge by mail from NSCEP (allow 4-6
weeks for delivery), at the following address:
U.S. EPA/National Service Center for Environmental Publications (NSCEP)
P.O. Box 42419
Cincinnati, OH 45242
Phone: (513) 489-8190 or
(800)490-9198
Fax: (513)489-8695
IV
-------
TABLE OF CONTENTS
Section Page
INTRODUCTION 1
ABSTRACTS
IN SITU SOIL TREATMENT ABSTRACTS 19
Cometabolic Bioventing at Building 719, Dover Air Force Base, Dover Delaware 20
Bioventing at Multiple Air Force Test Sites 22
In Situ Gaseous Reduction System Demonstrated at White Sands Missile Range, New Mexico . . 24
Electrokinetics at an Active Power Substation (Confidential Location) 26
Electrokinetics at Site 5, Naval Air Weapons Station Point Mugu, California 28
Electrokinetic Extraction at the Unlined Chromic Acid Pit, Sandia National Laboratories,
New Mexico 30
In-Situ Thermal Desorption at the Former Mare Island Naval Shipyard, California 32
Soil Vapor Extraction Enhanced by Six-Phase Soil Heating at Poleline Road Disposal Area,
Fort Richardson, Alaska 34
Phytoremediation at Argonne National Laboratory - West, Waste Area Group 9, Operable
Unit 9-04, Idaho Falls, Idaho 36
Phytoremediation at the Open Burn and Open Detonating Area, Ensign-Bickford Company,
Simsbury, Connecticut 38
Phytoremediation at Twin Cities Army Ammunition Plant, Minneapolis-St. Paul, Minnesota .... 40
EG&G's ™ Aerobic Biofiltration System for the Destruction of Hydrocarbon Vapors from
Fuel-Contaminated Soils 42
Internal Combustion Engines for the Destruction of Hydrocarbon Vapors from Fuel-
Contaminated Soils 44
Purus PADRE® Regenerative Resin System for the Treatment of Hydrocarbon Vapors from Fuel-
Contaminated Soils 46
Barometrically Enhanced Remediation Technology (BERT™) Demonstration at Idaho
National Engineering and Environmental Laboratory, RWMC, Pit 2, Idaho Falls, Idaho 48
INCINERATION ABSTRACTS 51
On-Site Incineration at Weldon Spring Ordnance Works, St. Charles County, Missouri 52
THERMAL DESORPTION ABSTRACTS 55
Thermal Desorption at the Arlington Blending and Packaging Superfund Site
Arlington, Tennessee 56
v
-------
Thermal Desorption at Letterkenny Army Depot Superfund Site, K Areas, OU 1 Chambersburg,
Franklin County, Pennsylvania 58
Low Temperature Thermal Desorption at Longhorn Army Ammunition Plant, Karnack, Texas ... 60
Thermal Desorption at the Rocky Flats Environmental Technology Site, Trenches T-3 and T-4,
Golden, Colorado 62
OTHER EX SITU SOIL TREATMENT ABSTRACTS 65
Slurry Reactor Biotreatment of Explosives-Contaminated Soils at Joliet Army Ammunition
Plant, Joliet, Illinois 66
Joint Small Arms Range Remediation (Physical Separation and Acid Leaching) at Fort Polk
Range 5, Leesville, Louisiana 68
Thermo NUtech's Segmented Gate System at Los Alamos National Laboratory Technical
Area 33, Los Alamos, New Mexico 70
Thermo NUtech's Segmented Gate System at Pantex Plant, Firing Site 5, Amarillo, Texas 72
Thermo NUtech's Segmented Gate System at Sandia National Laboratories, ER Site 16,
Albuquerque, New Mexico 74
Thermo NUtech's Segmented Gate System at Sandia National Laboratories, ER Site 228A,
Albuquerque, New Mexico 76
Thermo NUtech's Segmented Gate System at Tonapah Test Range, Clean Slate 2,
Tonapah, Nevada 78
Chemical Extraction for Uranium Contaminated Soil at the RMI Titanium
Company Extrusion Plant, Ashtabula, Ohio 80
Transportable Vitrification System at Oak Ridge National Laboratory, Oak Ridge, Tennessee ... 82
PUMP AND TREAT ABSTRACTS 85
Groundwater Extraction and Treatment at the Logistics Center Operable Unit, Fort Lewis,
Washington 86
IN SITU GROUNDWATER TREATMENT ABSTRACTS 89
In Situ Bioremediation Using Molasses Injection at an Abandoned Manufacturing Facility,
Emeryville, California 90
In Situ Bioremediation Using Molasses Injection at the Avco Lycoming Superfund Site,
Williamsport, Pennsylvania 92
In Situ Bioremediation Using Bioaugmentation at Area 6 of the Dover Air Force Base,
Dover Delaware 94
Aerobic Degradation at Site 19, Edwards Air Force Base, California 96
In Situ Bioremediation at the Hanford 200 West Area Site, Richland, Washington 98
Aerobic Degradation at Moffett Naval Air Station, Mountain View, California 100
Enhanced In Situ Anaerobic Bioremediation of Fuel-Contaminated Ground Water 102
VI
-------
In Situ Bioremediation (Anaerobic/Aerobic) at Watertown, Massachusetts 104
Methane Enhanced Bioremediation Using Horizontal Wells at the Savannah River Site,
Aiken, South Carolina 106
In Situ Bioremediation at the Texas Gulf Coast Site, Houston, Texas 108
In Situ Redox Manipulation at U.S. DOE Hanford Site, 100-H and 100-D Areas 110
In Situ Chemical Oxidation Using Potassium Permanganate at Portsmouth Gaseous
Diffusion Plant, X-701B Facility 112
Phytoremediation Using Constructed Wetlands at the Milan Army Ammunition Plant,
Milan, Tennessee 114
Multi-Phase Extraction at the 328 Site, Santa Clara, CA 116
Dual Phase Extraction at the Defense Supply Center, Richmond, Virginia 118
Dual Vapor Extraction at Tinkham's Garage Superfund Site, Londonderry, NH 120
Frozen Soil Barrier at Oak Ridge National Laboratory, Oak Ridge, Tennessee 122
Horizontal Wells Demonstrated at U.S. DOE's Savannah River Site and Sandia
National Laboratory 124
In Situ Chemical Oxi-Cleanse Process at the Naval Air Station Pensacola Florida, Operable
Unit 10, Pensacola, Florida 126
In Situ Chemical Oxidation Using Fenton's Reagent at Naval Submarine Base Kings Bay,
Site 11, Camden County, Georgia 128
Six Phase Heating at the Skokie, Illinois Site 130
Hydrous Pyrolysis Oxidation/Dynamic Underground Stripping (HPO/DUS) at Visalia
Superfund Site, CA 132
Intrinsic Remediation at AOCs 43G and 43J, Fort Devens, Massachusetts 134
Monitored Natural Attenuation at Keesler Air Force Base, Mississippi 136
Monitored Natural Attenuation at Kelly Air Force Base , Former Building 2093
Gas Station, Texas 138
In Situ Permeable Reactive Barriers for Contaminated Groundwater at Fry Canyon 140
Permeable Reactive Wall Remediation of Chlorinated Hydrocarbons in Groundwater at
Moffett Field Superfund Site 142
Groundwater Extraction and a Permeable Reactive Treatment Cell at Tacony Warehouse,
Philadelphia, Pennsylvania 144
DEBRIS/SOLID MEDIA TREATMENT ABSTRACTS 147
Direct Chemical Oxidation at Lawrence Livermore National Laboratory Livermore, California . 148
Acid Digestion of Organic Waste at Savannah River Site, Aiken, South Carolina 150
Remotely Operated Scabbling at Argonne National Laboratory-East, Argonne, Illinois 152
Soft Media Blasting at the Fernald Site, Fernald, Ohio 154
Concrete Grinder at the Hanford Site, Richland, Washington 156
vn
-------
Concrete Shaver at the Hanford Site, Richland, Washington 158
Concrete Spaller Demonstration at the Hanford Site, Richland, Washington 160
Stabilization Using Phosphate Bonded Ceramics at Argonne National Laboratory,
Argonne, Illinois 162
Stabilize Ash Using Clemson's Sintering Process at Clemson University, Clemson,
South Carolina 164
Mixed Waste Encapsulation in Polyester Resins at the Hanford Site 166
Innovative Grouting and Retrieval at the Idaho National Engineering and Environmental
Laboratory, Idaho Falls, Idaho 168
Polysiloxane Stabilization at Idaho National Engineering and Environmental Laboratory,
Idaho Falls, Idaho 170
Amalgamation of Mercury-Contaminated Waste using NFS DeHgSM Process, Applied
Technology Laboratories, Erwin, TX 172
Amalgamation of Mercury-Contaminated Waste using ADA Process, Colorado Minerals
Research Institute 174
GTS Duratek (GTSD) Process for Stabilizing Mercury (<260 ppm) Contaminated Mixed
Waste from U.S. DOE's Los Alamos National Laboratory 176
Stabilize High Salt Content Waste Using Sol Gel Process at Pacific Northwest National
Laboratory, Richland, WA 178
ATG Process for Stabilizing Mercury (<260 ppm) Contaminated Mixed Waste from
U.S. DOE's Portsmouth, Ohio Facility 180
Graphite Electrode DC Arc Furnace at the Idaho National Engineering and Environmental
Laboratory, Idaho Falls, Idaho 182
Plasma Hearth Process at the Science and Technology Applications Research (STAR) Center,
Idaho Falls, Idaho 184
Tables
1 Summary of Remediation Case Studies 3
2 Remediation Case Studies: Summary of Cost Data 11
Vlll
-------
INTRODUCTION
Increasing the cost effectiveness of site remediation is a national priority. The selection and use of more
cost-effective remedies requires better access to data on the performance and cost of technologies used in
the field. To make data more widely available, member agencies of the Federal Remediation
Technologies Roundtable (Roundtable) are working jointly to publish case studies of full-scale
remediation and demonstration projects. Previously, the Roundtable published 13 volumes of case study
reports. At this time, the Roundtable is publishing a CD-ROM containing 78 new case study reports,
primarily focused on soil and groundwater cleanup.
The case studies were developed by the U.S. Environmental Protection Agency (EPA), the U.S.
Department of Defense (DoD), and the U.S. Department of Energy (DOE). They were prepared based on
recommended terminology and procedures agreed to by the agencies. These procedures are summarized
in the Guide to Documenting and Managing Cost and Performance Information for Remediation Projects
(EPA 542-B-98-007; October 1998).
The case studies and abstracts present available cost and performance information for full-scale
remediation efforts and several large-scale demonstration projects. They are meant to serve as primary
reference sources, and contain information on site background and setting, contaminants and media
treated, technology, cost and performance, and points of contact for the technology application. The
studies contain varying levels of detail, reflecting the differences in the availability of data and
information. Because full-scale cleanup efforts are not conducted primarily for the purpose of
technology evaluation, data on technology cost and performance may be limited.
The case study abstracts in this volume describe a wide variety of ex situ and in situ soil treatment
technologies for both soil and groundwater. Contaminants treated included chlorinated solvents;
petroleum hydrocarbons and benzene, toluene, ethylbenzene, and xylenes; polycyclic aromatic
hydrocarbons; pesticides and herbicides; explosives/propellants; metals; and radioactivity. Many of the
applications described in the case study reports are ongoing and interim reports are provided
documenting their current status.
Table 1 provides summary information about the technology used, contaminants and media treated, and
project duration for the 78 technology applications in this volume. This table also provides highlights
about each application. Table 2 summarizes cost data, including information on quantity of media
1
-------
treated and quantity of contaminant removed. In addition, Table 2 shows a calculated unit cost for some
projects, and identifies key factors potentially affecting technology cost. (The column showing the
calculated unit costs for treatment provides a dollar value per quantity of media treated and contaminant
removed, as appropriate.) Cost data are shown as reported in the case studies and have not been adjusted
for inflation to a common year basis. The costs should be assumed to be dollars for the time period that
the project was in progress (shown on Table 1 as project duration).
While a summary of project costs is useful, it may be difficult to compare costs for different projects
because of unique site-specific factors. However, by including a recommended reporting format, the
Roundtable is working to standardize the reporting of costs to make data comparable across projects. In
addition, the Roundtable is working to capture information in case study reports that identify and
describe the primary factors that affect cost and performance of a given technology. Factors that may
affect project costs include economies of scale, concentration levels in contaminated media, required
cleanup levels, completion schedules, and matrix characteristics and operating conditions for the
technology.
-------
Table 1. Summary of Remediation Case Studies
Site Name, State (Technology)
Principal
Contaminants*
Chlorinated Solvents
BTEX and/or TPH
Pesticides/Herbicides
Explosives/Propellants
fi
"3
1
Radionuclides
Media
(Quantity Treated* *)
Project
Duration
Highlights
In Situ Soil Treatment
Dover Air Force Base, Building 719, Delaware
(Bioventing)
Multiple Air Force Test Sites, Multiple Locations
(Bioventing)
White Sands Missile Range, SWMU 143, New Mexico
(Chemical Reduction/Oxidation)
Active Power Substation, Confidential Location
(Electrokinetics)
Naval Air Weapons Station Point Mugu, Site 5,
California (Electrokinetics)
Sandia National Laboratories, Unlined Chromic Acid
Pit, New Mexico (Electrokinetics)
Former Mare Island Naval Shipyard, California (In
Situ Thermal Treatment; In Situ Thermal Desorption)
Fort Richardson Poleline Road Disposal Area, OU B,
Alaska (In Situ Thermal Treatment; Six Phase Heating)
Argonne National Laboratory - West, Waste Area
Group 9, OU 9-04, Idaho (Phytoremediation)
Ensign-Bickford Company - OB/OD Area, Connecticut
(Phytoremediation)
Twin Cities Army Ammunition Plant, Minnesota
(Phytoremediation)
•
•
•
•
•
•
•
•
•
Soil (450,000 Ibs)
Soil (200 to 270,000 yd3
per site)
Soil
Soil
Soil
Soil
Soil
Soil (3,910 yd3 or
7, 150 tons)
Soil
Soil
Soil
May 1998 to July 1999
April 1992 to December
1995 (typical test about
1 year)
April 1998 to June 1998
Summer 1998 (6 month
pilot-scale study)
March 1998 to June
1999
May 1996 to November
1996
September 1997 to
December 1997
July 1997 to December
1997 (treatability Study)
May 1998 to October
1998
April 1998 to October
1998
Spring/Summer 1998
Field demonstration of in situ cometabolic
bioventing to treat chlorinated solvents in
soil
Major initiative to demonstrate the
feasibility of bioventing for petroleum-
contaminated soil at 145 AF sites
Demonstrate use of injection of H2S for in
situ reduction of hexavalent chromium
First field demonstration of electrokinetic
remediation in the U. S. for arsenic-
contaminated soil
Field demonstration of electrokinetics for
treatment of metals in a sandy soil
The first field demonstration of
electrokinetics for removal of contaminant
ions from arid soil
Field demonstration of in situ thermal
desorption to treat PCBs in shallow and
deep contaminated soils
Demonstration of SPSH applied to
contamination in saturated soils.
Bench-scale testing of phytoremediation to
treat heavy metals in soil
Phytoremediation of lead in soil using both
phytoextraction and phyto stabilization
Phytoremediation of heavy metals in soil in
a northern climate
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Patrick Air Force Base, Active Base Exchange Service
Station, Florida (Soil Vapor Extraction)
Patrick Air Force Base, Active Base Exchange Service
Station, Florida (Soil Vapor Extraction)
Vandenberg Air Force Base, Base Exchange Service
Station, California (Soil Vapor Extraction)
Idaho National Engineering and Environmental
Laboratory, Pit 2, Idaho (Soil venting, BERT™)
Principal
Contaminants*
Chlorinated Solvents
•
BTEX and/or TPH
•
•
•
Pesticides/Herbicides
Explosives/Propellants
if*
"3
"5
§
Radionuclides
Media
(Quantity Treated* *)
Soil vapors
Soil vapors
Soil vapors
Soil
Project
Duration
January 1 994 to
February 1994
October 1993 to January
1994
February 1994 to June
1994
December 1996 to
January 1999
Highlights
Demonstration of treatment of extracted
vapors from an SVE system using
biofiltration
Demonstration of treatment of extracted
vapors from an SVE system using an
internal combustion engine
Demonstration of treatment of extracted
vapors from an SVE system using resin
adsorption
Demonstrate use of passive soil venting for
remediation of VOC-contamination
Incineration
Former Weldon Springs Ordnance Works, OU 1,
Missouri (Incineration (on-site))
•
Soil (30,000 tons or
18,000yd3)
Wooden pipeline
August 1998 to 1999
Use of on-site incineration for treatment of
nitroaromatic-contaminated materials
Thermal Desorption
Arlington Blending and Packaging Superfund Site,
Tennessee (Thermal Desorption)
Letterkenny Army Depot Superfund Site, K Areas,
OU1, Pennsylvania (Thermal Desorption)
Longhorn Army Ammunition Plant, Burning Ground
No. 3, Texas (Thermal Desorption)
Rocky Flats Environmental Technology Site, Trenches
T-3 and T-4, Colorado (Thermal Desorption)
•
•
•
•
•
•
•
Soil (4 1,431 tons)
Soil (13,986 yd3)
Soil (32,293 yd3 or
5 1,669 tons)
Soil and debris
(3,796 yd3)
January 1996 to June
1996
September 1993 to
October 1994
February 1997 to
December 1997
June 1 996 to August
1996
Application of low temperature thermal
desorption to treat pesticide-contaminated
soil
Thermal desorption to treat VOC-
contaminated soil, including soils with high
oil and grease content
Thermal desorption of soil with high
concentrations of chlorinated solvents
Application of thermal desorption to treat
soils contaminated with VOCs and low
levels of radiation
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Principal
Contaminants*
Chlorinated Solvents
BTEX and/or TPH
Pesticides/Herbicides
Explosives/Propellants
if*
"3
"5
§
Radionuclides
Media
(Quantity Treated* *)
Project
Duration
Highlights
Other Ex Situ Soil Treatment
Joliet Army Ammunition Plant, Illinois
(Bioremediation (ex situ) Slurry Phase)
Fort Polk Range 5, Louisiana (Chemical
Reduction/Oxidation)
Los Alamos National Laboratory, Technical Area 33,
New Mexico (Physical Separation; Segmented Gate
System)
Pantex Plant, Firing Site 5, Texas (Physical Separation;
Segmented Gate System)
Sandia National Laboratories, ER Site 16, New
Mexico (Physical Separation; Segmented Gate System)
Sandia National Laboratories, ER Site 228A, New
Mexico (Physical Separation; Segmented Gate System)
Tonapah Test Range, Clean Slate 2, Nevada (Physical
Separation; Segmented Gate System)
RMI Titanium Company Extrusion Plant, Ohio
(Solvent Extraction)
Oak Ridge National Laboratory, Tennessee
(Vitrification)
•
•
•
•
•
•
•
•
•
•
•
Soil
Soil (1,098 tons)
Soil and debris
(2,526 yd3)
Soil and debris (294 yd3)
Soil (66 1.8 yd3)
Soil (1,352 yd3)
Soil and debris (333 yd3)
Soil (64 ton; 38 batches)
Sludge (1 6,000 Ibs)
July 1994 to August
1995
August 1996 to
December 1996
April 1999 to May 1999
March 1998 to May
1998
February 1998 to March
1998
July 1998 to November
1998
May 1998 to June 1998
January 1997 to
February 1997
October 1997
Use of bioslurry technology for treatment
of explosives wastes
Demonstration of physical separation and
acid leaching to treat metals in soil
Use of a gate system to reduce soil volume
requiring off-site disposal
Use of a gate system to reduce soil volume
requiring off-site disposal
Use of a gate system to reduce soil volume
requiring off-site disposal
Use of a gate system to reduce soil volume
requiring off-site disposal
Use of a gate system to reduce soil volume
requiring off- site disposal
Demonstration of chemical leaching
process for treatment of uranium-
contaminated soil
Demonstration of a transportable
vitrification system to treat low-level mixed
waste sludges
Pump and Treat
Fort Lewis Logistics Center, Washington (Pump and
Treat)
•
Groundwater
(2.147 million gallons)
August 1995 to ongoing
Containment of lateral migration of
contaminants
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Principal
Contaminants*
Chlorinated Solvents
BTEX and/or TPH
Pesticides/Herbicides
Explosives/Propellants
if*
"3
"5
§
Radionuclides
Media
(Quantity Treated* *)
Project
Duration
Highlights
In Situ Groundwater Treatment
Abandoned Manufacturing Facility - Emeryville,
California (Bioremediation (in situ) Groundwater)
Avco Lycoming Superfund Site, Pennsylvania
(Bioremediation (in situ) Groundwater)
Dover Air Force Base, Area 6, Delaware
(Bioremediation (in situ) Groundwater)
Edwards Air Force Base, California (Bioremediation
(in situ) Groundwater)
Hanford 200 West Area, Washington (Bioremediation
(in situ) Groundwater)
Moffett Field Superfund Site, California
(Bioremediation (in situ) Groundwater)
Naval Weapons Station Seal Beach, California
(Bioremediation (in situ) Groundwater)
Watertown Site, Massachusetts (Bioremediation (in
situ) Groundwater)
Savannah River Site, South Carolina (Bioremediation
(in situ) Groundwater)
Texas Gulf Coast Site, Texas (Bioremediation (in situ)
Groundwater)
Hanford Site, 100-H and 100-D Areas, Washington
(Chemical Reduction/Oxidation)
•
•
•
•
•
•
•
•
•
•
•
•
•
Groundwater
Groundwater
Groundwater
Groundwater
(12,132m3 pumped)
Groundwater
Groundwater
Groundwater (in situ),
Soil (in situ), LNAPL
Groundwater
Groundwater and
sediment
Groundwater
Groundwater
Ongoing, data from
April 1997 to October
1998
Ongoing, data through
July 1998
Testing Phase:
September 1996 to June
1999
February 1996 to April
1997
January 1995 to March
1996
September 1986 to
November 1988
September 1997 to
October 1998
Ongoing, data from
November 1996 to
October 1997
February 1992 to April
1993
Ongoing, data from June
1995 to December 1998
September 1995 to
September 1998
Bioremediation of a site contaminated with
both chlorinated solvents and hexavalent
chromium
One of the first applications of molasses
injection technology on a full scale at a
Superfund site
First successful bioaugmentation project
using live bacteria from another site to treat
TCE using reductive dechlorination
Field demonstration using groundwater
recirculation wells to remediate TCE in a
two-aquifer system
In situ bioremediation of chlorinated
solvents and nitrate
One of the earliest field demonstrations of
aerobic in situ bioremediation
Demonstrate anaerobic bioremediation for
treating fuel hydrocarbons
Combined anaerobic/aerobic system for
treatment of chlorinated solvents
Demonstration using horizontal wells and
methane injection
Groundwater recirculation system using
trenches for extraction and injection
Demonstrate in situ redox manipulation for
treatment of hexavalent chromium
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Portsmouth Gaseous Diffusion Plant, X-701B Facility,
Ohio (Chemical Reduction/Oxidation)
Milan Army Ammunition Plant, Tennessee
(Constructed Wetlands)
328 Site, California (Dual-Phase Extraction)
Defense Supply Center, Acid Neutralization Pit,
Virginia (Dual-Phase Extraction)
Tinkham's Garage Superfund Site, New Hampshire
(Dual-Phase Extraction)
Oak Ridge National Laboratory, Tennessee (Frozen
Soil Barrier)
Portsmouth Gaseous Diffusion Plant, X-701B Facility,
Ohio (In Situ Oxidation)
Naval Air Station Pensacola, OU 10, Florida (In Situ
Oxidation; Fenton's Reagent)
Naval Submarine Base Kings Bay, Georgia (In Situ
Oxidation; Fenton's Reagent)
Confidential Manufacturing Facility, Illinois (In Situ
Thermal Treatment; Six Phase Heating)
Visalia Superfund Site, California (In Situ Thermal
Treatment; Dynamic Underground Stripping)
Fort Devens, AOCs 43G and 43 J, Massachusetts
(Monitored Natural Attenuation)
Principal
Contaminants*
Chlorinated Solvents
•
•
•
•
•
•
•
•
BTEX and/or TPH
•
Pesticides/Herbicides
Explosives/Propellants
•
if*
"3
"5
§
Radionuclides
•
Media
(Quantity Treated* *)
Groundwater (in situ)
Ground-water
Soil and Groundwater
Soil, Groundwater
(17 million gallons)
Soil (9,000 yd3)
Groundwater
Soil, Sediment,
Groundwater
Groundwater (in situ)
Groundwater
Groundwater (78,989
gallons)
Soil and groundwater
(34,600 yd3)
Groundwater
Groundwater
Project
Duration
Spring 1997 (operated
for one month)
June 1996 to July 1998
November 1996 to May
1999
July 1997 to July 1998
November 1994 to
September 1995
September 1996 to
September 1998
1988 to 1993
November 1998 to May
1999
November 1998 to
August 1999
June 1998 to April 1999
June 1997 to mid- 1999
March 1997 to June
1999
Highlights
Demonstrate in situ chemical oxidation for
treating chlorinated solvents
Use of constructed wetlands for treatment
of explosives-contaminated groundwater
Use of DPE with pneumatic fracturing for
VOCs in silty clay soils and shallow
groundwater
Use of DPE to treat soil and groundwater
contaminated with chlorinated solvents
Use of DVE to treat soil and groundwater
contaminated with chlorinated solvents
Demonstrate frozen soil barrier for
containment of contaminated surface
impoundment
Demonstrate use of horizontal wells to treat
groundwater at multiple sites and locations
Field demonstration of in situ chemical
oxidation using Fenton's reagent to treat
chlorinated solvents
Use of Fenton's Reagent to remediate
chlorinated solvents in groundwater
Use of SPH to remediate chlorinated
solvents in soil and groundwater
Use of HPO/DUS for treatment of large
quantity of creosote in groundwater
Intrinsic remediation for a site
contaminated with BTEX
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Keesler Air Force Base Service Station, AOC-A
(ST-06), Mississippi (Monitored Natural Attenuation)
Kelly Air Force Base, Former Building 2093 Gas
Station, Texas (Monitored Natural Attenuation)
Fry Canyon, Utah (Permeable Reactive Barrier)
Moffett Field Superfund Site, California (Permeable
Reactive Barrier)
Tacony Warehouse, Pennsylvania (Permeable Reactive
Barrier; Pump and Treat)
Principal
Contaminants*
Chlorinated Solvents
•
•
BTEX and/or TPH
•
•
Pesticides/Herbicides
Explosives/Propellants
if*
"3
"5
§
•
•
Radionuclides
•
Media
(Quantity Treated* *)
Soil, groundwater, and
soil gas
Soil, groundwater, and
soil gas
Groundwater
(33,000 ft3 or 200,000
gallons)
Groundwater
Groundwater (393, 165
gallons during the first
year)
Project
Duration
September 1997 to April
1999
July 1997 to July 1998
Ongoing, data from
September 1997 to
September 1998
April 1996 to December
1997
May 1998 through 2001
(projected)
Highlights
Monitored natural attenuation for a
gasoline contaminated site
Monitored natural attenuation for a
gasoline-contaminated site
Demonstration of three types of PRBs to
treat uranium-contaminated groundwater
Demonstration of PRB to remediate
groundwater contaminated with chlorinated
solvents
Use of an extraction well surrounded by
permeable reactive media at site
contaminated with chlorinated solvents.
Debris/Solid Media Treatment
Lawrence Livermore National Laboratory, California
(Chemical Reduction/Oxidation; Direct Chemical
Oxidation)
Savannah River Site, South Carolina (Chemical
Reduction/Oxidation)
Argonne National Laboratory - East, Illinois (Physical
Separation)
Argonne National Laboratory - East, Illinois (Physical
Separation)
Femald Site, Ohio (Physical Separation)
•
•
•
•
•
•
Waste streams from
LLNL operations
Organic wastes
Debris (concrete)
Debris (concrete floor)
Debris
Not identified
1996 to 1997
August 1997 to
September, 1997
Not identified
August 1996 to
September 1996
Pilot-scale demonstration of the DCO
process to treat a variety of organic
aqueous waste streams
Demonstrate acid digestion of organic
wastes as an alternative to incineration
Demonstration of a remotely-controlled
concrete demolition system to remove
radioactively contaminated concrete
Demonstration of a remotely-operated
scabbier to decontaminate radioactive
concrete flooring
Demonstration of soft blast media to clean
surfaces contaminated with uranium
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Hanford Site, Washington (Physical Separation)
Hanford Site, Washington (Physical Separation)
Hanford Site, Washington (Physical Separation)
Argonne National Laboratory - East, Illinois
(Solidification/Stabilization)
Clemson University, South Carolina
(Solidification/Stabilization)
Hanford Site, Washington (Solidification/Stabilization)
Idaho National Engineering and Environmental
Laboratory, Idaho (Solidification/Stabilization)
Idaho National Engineering and Environmental
Laboratory, Idaho (Solidification/Stabilization)
Idaho National Engineering and Environmental
Laboratory, Idaho (Solidification/Stabilization)
Los Alamos National Laboratory, New Mexico
(Solidification/Stabilization)
Los Alamos National Laboratory, New Mexico
(Solidification/Stabilization)
Principal
Contaminants*
Chlorinated Solvents
•
BTEX and/or TPH
Pesticides/Herbicides
Explosives/Propellants
if*
"3
"5
§
•
•
•
•
•
•
•
Radionuclides
•
•
•
•
•
•
Media
(Quantity Treated* *)
Debris (concrete) (54 ft2)
Debris (concrete)
Debris (contaminated
concrete walls and
floors) (4.6m2)
Salt-containing waste
streams
Incinerator fly ash
Process waste streams
Soil and debris
Process waste streams
Liquid mercury (75 kg)
Liquid mercury (132 kg)
Sludge (1,253 Ibs)
Laboratory Wastes
Project
Duration
November 1997
November 1997
January 1998
Not identified
1995
Not identified
Summer 1994 to
Summer 1996
1997 to 1998
1998
1998
September 1997 to
September 1998
Highlights
Demonstration of a light weight hand-held
grinder to decontaminate radioactive
concrete surfaces
Demonstration of a concrete shaver to
decontaminate radioactive concrete
surfaces
First demonstration of the hand-held
concrete spaller on contaminated surfaces
Demonstration of phosphate-bonded
ceramics to stabilize a variety of high salt-
containing wastes
Treatability study of stabilization of mixed
waste fly ash using a sintering process
Treatability study of various polyester resins
to stabilize high salt-containing mixed waste
Field demonstration of innovative jet
grouting and retrieval techniques that are
applicable to TRU wastes
Demonstration of polysiloxane to
encapsulate high-salt content wastes
Demonstrate amalgamation of elemental
mercury
Demonstrate amalgamation of elemental
mercury
Demonstrate stabilization of low level
mercury in radioactive wastes
-------
Table 1. Summary of Remediation Case Studies (continued)
Site Name, State (Technology)
Pacific Northwest National Laboratory, Washington
(Solidification/Stabilization)
Portsmouth Gaseous Diffusion Plant, Ohio
(Solidification/Stabilization)
Idaho National Engineering and Environmental
Laboratory, Idaho (Vitrification)
STAR Center, Idaho (Vitrification)
Principal
Contaminants*
Chlorinated Solvents
BTEX and/or TPH
1
'3
a
'3
^
Explosives/Propellants
"3
"5
*
*
*
*
Radionuclides
*
*
*
Media
(Quantity Treated* *)
Salt waste surrogates
Ion exchange resin
(160kg)
Wastes - including slag,
plutonium-238 waste,
neutron generators
Fly ash, soil, sludges,
debris
Project
Duration
Not identified
1998
1997 to 1998
1993 to 1997
Highlights
Laboratory testing of the sol gel process to
stabilize high salt content waste
Demonstrate stabilization of low level
mercury in radioactive wastes
Demonstrate DC arc plasma furnace to treat
a variety of wastes from DOE facilities
Demonstration of a plasma hearth furnace
to treat metals and radionuclides in a
variety of waste types
' Principal contaminants are one or more specific constituents within the groups shown that were identified during site investigations.
10
-------
Table 2. Remediation Case Studies: Summary of Cost Data
Site Name, State (Technology)
Technology Cost
(S)1-2
Quantity of Media
Treated
Quantity of
Contaminant
Removed
Calculated Unit Cost
for Treatment 1>2
Key Factors
Potentially Affecting
Technology Costs***
In Situ Soil Treatment
Dover Air Force Base, Building 719, Delaware
(Bioventing)
Multiple Air Force Test Sites, Multiple Locations
(Bioventing)
White Sands Missile Range, SWMU 143, New Mexico
(Chemical Reduction/Oxidation)
Active Power Substation, Confidential Location
(Electrokinetics)
Naval Air Weapons Station Point Mugu, Site 5,
California (Electrokinetics)
Sandia National Laboratories, Unlined Chromic Acid
Pit, New Mexico (Electrokinetics)
Former Mare Island Naval Shipyard, California (In
Situ Thermal Treatment; In Situ Thermal Desorption)
Fort Richardson Poleline Road Disposal Area, OU B,
Alaska (In Situ Thermal Treatment; Six Phase Heating)
Argonne National Laboratory - West, Waste Area
Group 9, OU 9-04, Idaho (Phytoremediation)
Ensign-Bickford Company - OB/OD Area, Connecticut
(Phytoremediation)
Twin Cities Army Ammunition Plant, Minnesota
(Phytoremediation)
Patrick Air Force Base, Active Base Exchange Service
Station, Florida (Soil Vapor Extraction - Biofiltration)
Not provided
P: $92,300
P: $798,163
Not provided
Not provided
Not provided
Not provided
$967,822
P: $2,247,000
Not provided
Not provided
Not provided
450,000 Ibs
200 to 270,000
cubic yards per site
Not provided
Not provided
Not provided
Not provided
Not provided
3,910 cubic yards
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
P: $10 to $60 per cubic
yard
P: $43 to $100 per cubic
yard
Not provided
Not provided
Not provided
$100 to $250 per ton
(vendor estimate)
$189 to $288 per cubic
yard, $726 to $2,552 per
Ib of contaminant
removed
Not provided
Not provided
$30.34 per cubic yard of
soil per year ($153 per
cubic yard over the life of
the project)
$18.66 to $38.06 per kg
(costs estimates were
provided by other
vendors)
Not provided
Volume of soil treated,
with lower costs for
sites with > 10,000 yds3
Size of the waste site
Not provided
Not provided
Not provided
Not provided
Availability and cost for
power
Amount of time needed
to meet goals and size
of area treated
Not provided
Amount of time needed
to meet goals and size
of area treated
Contaminant
concentration and flow
rate
11
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Patrick Air Force Base, Active Base Exchange Service
Station, Florida (Soil Vapor Extraction - Thermal
Destruction)
Vandenberg Air Force Base, Base Exchange Service
Station, California (Soil Vapor Extraction - Resin
Adsorption)
Idaho National Engineering and Environmental
Laboratory, Pit 2, Idaho (Soil venting BERT™)
Technology Cost
(S)1'2
Not provided
DEMO: $36,634
P: $67,860
Quantity of Media
Treated
Not provided
Not provided
Not provided
Quantity of
Contaminant
Removed
Not provided
570 gals of
hydrocarbons
Chlorinated
solvents ranged
from 0.25 to 2.9
gms/day
Calculated Unit Cost
for Treatment ll2
Operating costs of $0.83
to $15. 40 per kg TVH
destroyed, $97 to $550
perkgofBTEX
destroyed
DEMO: $23 per kg of
hydrocarbon removed
P: $100 per cubic yard
Key Factors
Potentially Affecting
Technology Costs***
Contaminant
concentration and
supplemental fuel
requirement
Contaminant
concentration and flow
rate
Size of contaminated
area and length of
operation
Incineration
Former Weldon Springs Ordnance Works, OU 1,
Missouri (Incineration (on-site))
$13,665,997
30,000 tons
(18,000 cubic
yards)
85,230 feet of
pipeline
Not provided
Not provided
Types and properties of
materials treated (such
as moisture content,
BTU value)
Thermal Desorption
Arlington Blending and Packaging Superfund Site,
Tennessee (Thermal Desorption)
Letterkenny Army Depot Superfund Site, K Areas,
OU1, Pennsylvania (Thermal Desorption)
Longhorn Army Ammunition Plant, Burning Ground
No. 3, Texas (Thermal Desorption)
Rocky Flats Environmental Technology Site, Trenches
T-3 and T-4, Colorado (Thermal Desorption)
C: $4,293,893
0: $62,351
$4,647,632
$4,886,978
$1,934,203
41,431 tons
13,986 cubic yards
32,293 cubic yards
(5 1,669 tons)
3,796 cubic yards
Not provided
Not provided
Not provided
Not provided
$105 per ton
$220 per cubic yard
$151 per cubic yard
$350 per cubic yard
Types and properties of
materials treated such
as moisture content and
types of contaminants
(pesticides)
Types and properties of
materials treated such
as moisture content and
types of contaminants
(high oil and grease
content)
Types and properties of
materials treated such
as moisture content and
types of contaminants
(solvents)
Use of radiological
engineering controls
12
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Technology Cost
(S)1'2
Quantity of Media
Treated
Quantity of
Contaminant
Removed
Calculated Unit Cost
for Treatment ll2
Key Factors
Potentially Affecting
Technology Costs***
Other Ex Situ Soil Treatment
Joliet Army Ammunition Plant, Illinois
(Bioremediation (ex situ) Slurry Phase)
Fort Polk Range 5, Louisiana (Physical Separation and
Acid Leaching)
Los Alamos National Laboratory, Technical Area 33,
New Mexico (Physical Separation; Segmented Gate
System)
Pantex Plant, Firing Site 5, Texas (Physical Separation;
Segmented Gate System)
Sandia National Laboratories, ER Site 16, New
Mexico (Physical Separation; Segmented Gate System)
Sandia National Laboratories, ER Site 228A, New
Mexico (Physical Separation; Segmented Gate System)
Tonapah Test Range, Clean Slate 2, Nevada (Physical
Separation; Segmented Gate System)
RMI Titanium Company Extrusion Plant, Ohio
(Solvent Extraction)
Oak Ridge National Laboratory, Tennessee
(Vitrification)
Not provided
DEMO: $1,169,000
P: $1,700,000
$275,745
$203,887
$164,109
$220,040
$138,126
Pilot: $638,670
C: $5,000,000
AO: $10 to $44 per
kg of waste
Not provided
DEMO: 835 tons
PC: 10,000 tons
2,526 cubic yards
294 cubic yards
661.8 cubic yards
1,352 cubic yards
333 cubic yards
64 tons (38 batches)
16,000 Ibs of pond
and neutralization
sludge
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
P: $290 to $350 per
cubic yard
DEMO: $1,400 per ton
P: $170 per ton
$109 per cubic yard
$111 per cubic yard
$236 per cubic yard
$154 per cubic yard
Not provided
P: $250 to $350 per ton
of soil
Not provided
Use of additives and
frequency of
replacement
Volume of waste
treated and level of
treatment required to
regenerate leachate
Quantity of material
processed
Quantity of material
processed
Quantity of material
processed
Quantity of material
processed
Quantity of material
processed
Contaminant
concentrations and
amount of heating
required for solvent
Size of area treated;
energy requirements;
and level of emission
controls required
13
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Technology Cost
(S)1'2
Quantity of Media
Treated
Quantity of
Contaminant
Removed
Calculated Unit Cost
for Treatment ll2
Key Factors
Potentially Affecting
Technology Costs***
Pump and Treat
Fort Lewis Logistics Center, Washington (Pump and
Treat)
$5,208,000
2. 147 million
gallons (through
8/98)
2,772 Ibs of TCE
(through 9/97)
Not provided
Length of system
operation; presence of
DNAPL
In Situ Groundwater Treatment
Abandoned Manufacturing Facility - Emeryville,
California (Bioremediation (in situ) Groundwater)
Avco Lycoming Superfund Site, Pennsylvania
(Bioremediation (in situ) Groundwater)
Dover Air Force Base, Area 6, Delaware
(Bioremediation (in situ) Groundwater)
Edwards Air Force Base, California (Bioremediation
(in situ) Groundwater)
Hanford 200 West Area, Washington (Bioremediation
(in situ) Groundwater)
Moffett Field Superfund Site, California
(Bioremediation (in situ) Groundwater)
Naval Weapons Station Seal Beach, California
(Bioremediation (in situ) Groundwater)
Watertown Site, Massachusetts (Bioremediation (in
situ) Groundwater)
Savannah River Site, South Carolina (Bioremediation
(in situ) Groundwater)
Texas Gulf Coast Site, Texas (Bioremediation (in situ)
Groundwater)
Hanford Site, 100-H and 100-D Areas, Washington
(Chemical Reduction/Oxidation)
$400,000
C: $220,000
AO: $50,000
C: $285,563
O: $522,620 (for 15
months)
C: $323,452
0: $14,354
Not provided
Not provided
DEMO: $875,000
P: $1,085,000
DEMO: $150,000
PC: $452,407
PAO: $236,465
C: $600,000
AO: $100,000
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
12,132 cubic meters
Not provided
Not provided
Not provided
Not provided
17,000 Ibs VOCs
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
P: $5.80 per cubic meter
Not provided
P: $4,340 per gallon of
fuel
Not provided
Not provided
Not provided
Not provided
Size of area treated;
amount and frequency
of molasses injections
required
Size of area treated;
amount and frequency
of molasses injections
required
Size of area treated;
amount and type of
additives
Size of area treated; two
contaminated aquifers
Plume size - cost
effective for small
plumes (100 m
diameter)
Not provided
Size of area treated; for
demo, analytical costs
Not provided
Size of area treated;
DNAPL present
Size of area treated; use
of methanol as additive
Not provided
14
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Portsmouth Gaseous Diffusion Plant, X-701B Facility,
Ohio (Chemical Reduction/Oxidation)
Milan Army Ammunition Plant, Tennessee
(Constructed Wetlands)
328 Site, California (Dual-Phase Extraction)
Defense Supply Center, Acid Neutralization Pit,
Virginia (Dual-Phase Extraction)
Tinkham's Garage Superfund Site, New Hampshire
(Dual-Phase Extraction)
Oak Ridge National Laboratory, Tennessee (Frozen
Soil Barrier)
Savannah River Site, Aiken, South Carolina
(Horizontal Wells)
Naval Air Station Pensacola, OU 10, Florida (In Situ
Oxidation; Fenton's Reagent)
Naval Submarine Base Kings Bay, Georgia (In Situ
Oxidation; Fenton's Reagent)
Confidential Manufacturing Facility, Illinois (In Situ
Thermal Treatment; Six Phase Heating)
Visalia Superfund Site, California (In Situ Thermal
Treatment; Dynamic Underground Stripping)
Fort Devens, AOCs 43G and 43 J, Massachusetts
(Monitored Natural Attenuation)
Technology Cost
(S)1'2
DEMO: $562,000
P: $516,360
P: $3,466,000
C: $300,000
O: $550,000
Treat: $538,490
$1,500,000
DEMO: $1,809,000
Not provided
DEMO
C: $97,018
O: $81,320
Phase 1: $223,000
Not provided
Not provided
$671,642
PAO: $50,000
Quantity of Media
Treated
Not provided
Not provided
Not provided
17 million gallons
ofgroundwater
9,000 cubic yards
Not provided
Not provided
Not provided
Phase 1: 78,989
gallons
Not provided
Not provided
Not provided
Quantity of
Contaminant
Removed
Not provided
Not provided
l,2201bsVOCs
145 Ibs VOCs
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
141,000 gal of
creosote
Not provided
Calculated Unit Cost
for Treatment ll2
P: $64 per cubic yard
P: $1.78 per 1,000
gallons of groundwater
$53 per cubic yard
(based on treatment of
16,000 cubic yards)
Treat: $0.03 per gallon
$170 per cubic yard
Not provided
Not provided
Not provided
Not provided
$32 per cubic yard
P: $39 per cubic yard
Not provided
Key Factors
Potentially Affecting
Technology Costs***
Size of area treated;
DNAPL present
Type of system used
(gravel vs. lagoon-
based), size of area
treated, and climate
Use of pneumatic
fracturing;
contamination in two
aquifer zones
Volume ofgroundwater
treated; contamination
confined to upper
aquifer
Size of area treated;
contamination in two
aquifer zones
Complex hydrogeology
due to presence of
fractured bedrock
Not provided
Volume of reagent
injected and frequency
of injections
Volume of reagent
injected and frequency
of injections
Size of area treated;
power requirements
Groundwater extraction
capacity and plume size
Length of remediation;
monitoring
requirements
15
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Keesler Air Force Base Service Station, AOC-A
(ST-06), Mississippi (Monitored Natural Attenuation)
Kelly Air Force Base, Former Building 2093 Gas
Station, Texas (Monitored Natural Attenuation)
Fry Canyon, Utah (Permeable Reactive Barrier)
Moffett Field Superfund Site, California (Permeable
Reactive Barrier)
Tacony Warehouse, Pennsylvania (Permeable Reactive
Barrier; Pump and Treat)
Debris/Solid Media Treatment
Lawrence Livermore National Laboratory, California
(Chemical Reduction/Oxidation)
Savannah River Site, South Carolina (Chemical
Reduction/Oxidation)
Argonne National Laboratory - East, Argonne, Illinois
(Concrete Scabbling)
Femald Site, Femald, Ohio (Soft Media Blasting)
Hanford Site, Hanford, Washington (Concrete Grinder)
Technology Cost
(S)1'2
PO: $15,000 per
event
Not provided
DEMO C: $674,000
PAO: $55,000 to
$60,000
PC: $4,910,942
PAO: $72,278
$607,336
C: $416,777
AO: $16,880
Other: $132,417
Not provided
P: $2,000,000 to
$8,000,000
C: $165,000
O: $l,995/day
Not provided
C: $854 (purchase);
$75/week (rental)
Quantity of Media
Treated
Not provided
Not provided
33,000 cubic feet
(200,000 gallons
through 9/98)
Not provided
393,165 gallons
during the first year
Quantity of
Contaminant
Removed
Not provided
Not provided
Not provided
Not provided
Not provided
Calculated Unit Cost
for Treatment ll2
Not provided
Not provided
Not provided
Not provided
Not provided
Key Factors
Potentially Affecting
Technology Costs***
Length of remediation;
monitoring
requirements
Not provided
Type of reactive media;
size of PRB
Size of PRB and type of
reactive material;
projected costs assume
PRB constructed in two
sections
Size of PRB and type of
reactive material
Not provided
Not provided
Not provided
Not provided
54 square feet
Not provided
Not provided
Not provided
Not provided
Not provided
P: $9. 88 per kg of carbon
in the waste if oxidant
recycled; $79 per kg of
carbon if not recycled
Not provided
Not provided
DEMO: $4.60 per square
foot
Not provided
Amount of carbon in
waste stream; whether
oxidant is recycled
Physical and chemical
characteristics of waste
stream; volume treated
Area and depth of
concrete surface treated;
extent of particulate
controls used
Grade of media used;
size and depth of
concrete surface treated;
noise protection used
Size and depth of
concrete surface treated
16
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Hanford Site, Washington (Concrete Shaver)
Hanford Site, Washington (Concrete Spaller)
Argonne National Laboratory - East, Illinois
(Phosphate Bonded Ceramic Stabilization)
Clemson University, Clemson, South Carolina
(Stabilization Using Clemson's Sintering Process)
Hanford Site, Hanford, Washington (Polyester Resin
Encapsulation)
Idaho National Engineering and Environmental
Laboratory, Idaho (Innovative Grouting and Retrieval)
Idaho National Engineering and Environmental
Laboratory, Idaho (Polysiloxane Stabilization)
Idaho National Engineering and Environmental
Laboratory, Idaho (Amalgamation of Mercury using
the NFS DeHgSM Process)
Los Alamos National Laboratory, New Mexico
(Amalgamation of Mercury using the ADA Process)
Los Alamos National Laboratory, New Mexico
(Solidification/Stabilization - GTS Duratek Process)
Pacific Northwest National Laboratory, Washington
(Solidification/Stabilization - Sol Gel Process)
Technology Cost
(S)1'2
C: $17,861
Not provided
PC: $2,000,000
PO: $6,5 10 per
cubic meter of waste
Not provided
PC: $2,000,000
PO: $5,940 per
cubic meter of waste
P: $19,000,000
(1-acre);
$64,000,000(4-acre)
Not provided
Not provided
Not provided
Not provided
P: $600,000 to
$1,000,000
Quantity of Media
Treated
Not provided
4.6 square meters
Not provided
Not provided
Not provided
Not provided
Not provided
75 kg of mercury
132kg of mercury
Not provided
Not provided
Quantity of
Contaminant
Removed
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
Not provided
1,253 Ibs of sludge,
3 containers of
laboratory wastes
Not provided
Calculated Unit Cost
for Treatment ll2
$1 .32 per square foot
$128 per square meter
Not provided
Not provided
Not provided
Not provided
$8 per pound ($573 per
cubic foot) of salt waste
P: $300 per kg (based on
treating more than 1,500
kg)
P: $300 per kg (based on
treating more than 1,500
kg)
Not provided
Not provided
Key Factors
Potentially Affecting
Technology Costs***
Size and depth of
concrete surface treated
Size and depth of
concrete surface treated
Salt loading in waste;
types and
concentrations of heavy
metals
Not provided
Salt loading in waste;
types and
concentrations of heavy
metals
Size of area treated;
physical and chemical
characteristics of waste
Salt loading in waste;
types and
concentrations of heavy
metals
Quantity of waste
treated (costs
prohibitive for small
quantities)
Quantity of waste
treated (costs
prohibitive for small
quantities)
Not provided
Salt loading in waste;
types and
concentrations of heavy
metals
17
-------
Table 2. Remediation Case Studies: Summary of Cost Data (continued)
Site Name, State (Technology)
Portsmouth Gaseous Diffusion Plant, Ohio
(Solidification/Stabilization)
Idaho National Engineering and Environmental
Laboratory, Idaho (Graphite Elctrode DC ARC
Furnace)
STAR Center, Idaho (Plasma Hearth Process)
Technology Cost
(S)1'2
PC: 30,000
PO: $95 per hour
PC: $50 to $80
million
PO: $12 to $18
million (startup);
$48 to $62 million
(for 5yrs)
PC: $50 to 86.2
million
PO: $12 to $18
million (startup);
$48 to 62 million
(for 5 yrs)
Quantity of Media
Treated
160 kg of resin
Not provided
Not provided
Quantity of
Contaminant
Removed
Not provided
Not provided
Not provided
Calculated Unit Cost
for Treatment ll2
$1.73 per kg
P: $7,400 to $10,800 per
cubic meter (based on
17,000 cubic meters)
P: $7,400 to $10,800 per
cubic meter
Key Factors
Potentially Affecting
Technology Costs***
Types and
concentrations of heavy
metals
Physical characteristics
of waste (moisture
content); cost of power
Physical characteristics
of waste (moisture
content); cost of power
Actual full-scale costs are reported unless otherwise noted.
Cost abbreviation: AO = annual operation and maintenance (O&M) costs, C = capital costs, D = disposal costs, DEMO = demonstration costs, O = total O&M costs,
P = projected costs, Pilot = pilot-scale costs.
18
-------
IN SITU SOIL TREATMENT ABSTRACTS
19
-------
Cometabolic Bioventing at Building 719,
Dover Air Force Base, Dover Delaware
Site Name:
Dover Air Force Base, Building 719
Location:
Dover, Delaware
Period of Operation:
Propane acclimation period: December 1997 to April 1998
Bioventing operation: May 1998 to July 1999
Cleanup Authority:
CERCLA
Purpose/Significance of Application:
Field demonstration of in situ cometabolic bioventing to treat chlorinated
solvents in soil
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated Solvents
• Maximum concentrations of chlorinated aliphatic hydrocarbons (CAHs) in
soil found during site investigations were TCE (250 mg/kg); TCA (1,000
mg/kg); DCE (20 mg/kg)
• Estimated mass of CAH in test plot - 26 pounds; TCA made up
approximately 70% of the total estimated mass of contaminants
• Soil in the area is sand with varying amounts of clay, silt and gravel. Soil
permeability is 1.9xlO"7 to 7.0xlO"8 cm2.
Waste Source:
Discharges to a drainage ditch and
sanitary sewer; leaks from
underground and above ground tanks
Contacts:
EPA Remedial Project Manager:
Darius Ostrauskas
Remedial Project Manager
U.S. EPA Region 3
1650 Arch Street (3HS50)
Philadelphia, PA 19103
(215)814-3360
ostrauskas.darius@epa.gov
EPA Contact for Demonstration:
Dr. Gregory Sayles
U.S. EPA (mail stop 420)
26 W. Martin Luther King Drive
Cincinnati, OH 45268
(513)569-7607
Fax:(513)569-7105
E-mail: sayles.gregory@epa.gov
Technology:
In Situ Bioremediation; Cometabolic Bioventing
• Test plot - approximately 30 ft long, 20 ft wide, and 10 ft deep with a volume
of 4,500 ft3 of soil
• Three injection wells, screened to a depth of 10 ft bgs
• A blower and a mass flow controller were used to inject a mixture of air and
propane (300 ppm in air) through the three wells at a rate of 1 cfm
• 13 soil gas monitoring points to monitor soil gas conditions throughout the
demonstration. Each soil gas monitoring point was equipped with two gas
probes (one at a depth of 4-5 ft and one at a depth of 8-9 ft bgs); an additional
11 "temporary" soil gas monitoring points were used during initial air
permeability testing, and during system operation, to monitor soil gas
Type/Quantity of Media Treated:
Soil/ 450,000 Ibs, based on an assumed density of 100 lbs/ft3
Regulatory Requirements/Cleanup Goals:
The objectives of the pilot test included evaluating in situ cometabolic bioventing to treat chlorinated solvents in soil and
to collect data for potential full-scale application of the technology at the site
Results:
• After 14 months of operation, concentrations of TCE, TCA, and DCE were reduced in the soil in the test area
• Reductions included TCE from >10 mg/kg to <0.25 mg/kg; TCA from >100mg/kg to <0.5mg/kg; and DCE from
>20mg/kg to <0.25mg/kg
Costs:
Not provided
20
-------
Cometabolic Bioventing at Building 719,
Dover Air Force Base, Dover Delaware
Description:
Dover Air Force Base (AFB), located in Dover, Delaware, is a 4,000 acre military installation that began operating in
1941. Building 719 is a jet engine inspection and maintenance shop where a variety of materials, including solvents and
fuel, were used un base operations. Until the mid-1960s, wastes from the shop were discharged to a drainage ditch and
sanitary sewer. During site investigations, leaking tanks were identified in the area to the northeast of the shop, and soil
and groundwater at the site was found to be contaminated with chlorinated solvents. Dover AFB was listed on the
National Priorities List in March 1989. As part of the interim ROD for the site, a pilot test of in situ cometabolic
bioventing was conducted at Building 719 to evaluate the ability of the technology to remove CAHs. The test plot
selected for the pilot study was an area contaminated with high concentrations of CAHs. Prior to the pilot test, laboratory
tests were performed on soils from the test plot area to evaluate candidate substrates. Propane was selected because of its
ability to stimulate cometabolic activity towards both TCA and TCE.
The bioventing system used for the pilot test included three injection wells, screened to a depth of 10 ft bgs, which was
the lowest expected water table elevation. In addition, soil gas conditions were monitored throughout the demonstration
using soil gas monitoring points. In situ cometabolic bioventing was successful in reducing CAH concentrations in test
plot soils. After 14 months of operation, TCE and DCE were reduced to concentrations of less than 0.25 mg/kg, and TCA
was reduced to concentrations of less than 0.5 mg/kg. According to the researchers for the pilot test, results of laboratory
treatability testing identified propane as a useful cosubstrate for driving the cometabolism of TCE and TCA.
21
-------
Bioventing at Multiple Air Force Test Sites
Site Name:
Multiple Air Force Test Sites (145 total; refer to case study for names and
locations of each test site)
Location:
Multiple locations throughout U.S.
Period of Operation:
• Overall program: April 1992 to December 1995
• Each test: varied by site; typical operation about one year
Cleanup Authority:
Sites are being addressed under
CERCLA, RCRA, and state
underground storage tank programs
Purpose/Significance of Application:
Major initiative to demonstrate the feasibility of bioventing to remediate
petroleum-contaminated soil at 145 Air Force sites
Cleanup Type:
Pilot scale
Contaminants:
Benzene, Toluene, Ethylbenzene, and Xylenes (BTEX) and Total Petroleum
Hydrocarbons (TPH)
• Data provided for average initial concentrations of BTEX and TPH in soil
and soil gas (based on 328 samples from 100 test sites)
• Average BTEX constituent concentrations in soil (soil gas) - benzene -106
mg/kg (88 ppmv); toluene - 250 mg/kg (13 ppmv); ethylbenzene - 276 mg/kg
(64 ppmv); xylenes -1,001 mg/kg (46 ppmv)
• Average TPH concentration in soil - 3,301 mg/kg; Total Volatile
Hydrocarbons (TVH) in soil gas - 22,555 ppmv
Waste Source:
Leaks from underground storage
tanks, including tanks used to store
gasoline, JP-4, diesel fuel, heating
oils, and waste oils
Contacts:
Air Force Contact:
Lt. Col. Ross N. Miller
U.S. Air Force Center for
Environmental Excellence
Brooks, AFB
Texas
Telephone: 210-536-4331
Technology:
In Situ Bioventing
• Specific configuration varied by site for number, depth of vent (air injection)
wells, number of monitoring wells, and blower size and type
• Typical configuration included vent wells (1 to 9 per site; depths -7 to 233
feet below ground surface); vapor monitoring wells (1 to 6 per site); blower
(1 to 5 horsepower; either rotary vane or regenerative)
• Horizontal vent wells used at five sites
Type/Quantity of Media Treated:
Soil
• Quantities treated at each test site ranged from 200 to more than 270,000
cubic yards; based on radius of influence of vent well(s) at each site
• Soil gas permeability - about 20% of the test sites contained greater than 50%
silt and clay fractions; the radius of oxygen influence from a single vent well
was equal to or greater than the contaminated area at about 50% of the test
sites
• Soil pH - pH ranged from 5 to 9 at the majority of sites
• Soil moisture - ranged from 5% to 20% at the majority of sites
• Total Kjeldahl nitrogen - ranged from <50 to 200 mg/kg at the majority of
sites
• Soil temperature - not measured at each site; soil temperatures between 0°C
and 25 °C observed at test sites
Regulatory Requirements/Cleanup Goals:
• The objectives of the bioventing initiative included documenting the ability of bioventing to remediate petroleum-
contaminated soils in a variety of conditions, and obtaining a significant set of bioventing data
• No specific cleanup goals were identified for the test sites
22
-------
Bioventing at Multiple Air Force Test Sites
Results:
Results from data collected after one year of bioventing (328 sampling locations at 100 sites):
• Average reduction in BTEX concentrations of 97% in soil and 85% in soil gas; average TPH concentrations reduced by
24% in soil; average TVH concentrations reduced by 90% in soil gas
• Biodegradation rates measured at the test sites - at start of test ranged from <300 mg/kg/yr to >6000 mg/kg/yr; average
1,200 mg/kg/yr
• Average biodegradation rate decreased to 700 mg/kg/yr, as a result of the decreasing bioavailability of hydrocarbons
over time
• Bioventing was effective in a variety of climate conditions, ranging from 0°C in Alaska to 25 °C in California; higher
biodegradation rates were observed in warmer climates
• A combination of high moisture content and fine-grained soils made bioventing infeasible at only two of the 145 test
sites
Costs:
• The average actual cost for design, installation, and 1-year of operation of pilot-scale bioventing at a single vent well
site was $60,000
• The projected cost of full-scale bioventing generally ranges from $ 10 to $60 per cubic yard of soil treated
• At sites with more than 10,000 cubic yards of contaminated soil, costs are less than $10 per cubic yard; at sites with
less than 500 cubic yards of contaminated soil, costs are greater than $60 per cubic yard
• Projected costs for a typical full-scale bioventing system (defined as an Air Force site with 5,000 cubic yards of soil
contaminated with 3,000 mg/kg of JP-4 fuel; bioventing system consisting of four vent wells at a depth of 15 feet,
operated for two years) - $92,300, including $27,000 for pilot testing and $27,500 for full-scale construction
Description:
In April 1992, the Air Force Center for Environmental Excellence (AFCEE), in cooperation with the Air Force Armstrong
Laboratory and the U.S. Environmental Protection Agency, began an initiative to demonstrate the feasibility of using
bioventing to remediate petroleum contaminated soils in a variety of climatic, soil, and contaminant conditions. Between
April 1992 and December 1995, initial bioventing tests were conducted at 145 Air Force sites throughout the country.
The pilot-scale systems included vent (air injection) wells, monitoring wells, and blowers. The specific configuration
varied by test site, and horizontal vent wells were used at five of the sites. Concentrations of BTEX and TPH were
measured in soil and soil gas from over 300 sampling locations at 100 sites at the start of bioventing operations and after
one year of operation. Results showed that bioventing was effective in reducing concentrations of BTEX and TPH in soil
and soil gas in a variety of site conditions. Soil BTEX and TPH concentrations were reduced by 97% and 24%,
respectively. Soil gas BTEX and TVH concentrations were reduced by 85% and 90%, respectively. According to the
Air Force, the reductions in BTEX are sufficient to meet the most conservative EPA risk-based cleanup criteria for soils,
and regulatory acceptance of this technology was obtained in 38 states and the 10 EPA regions. The pilot-scale systems
have been converted to full-scale systems at about half of the test sites, saving the Air Force an estimated $5 to $ 10
million in design and construction costs.
23
-------
In Situ Gaseous Reduction System Demonstrated at White Sands Missile Range,
New Mexico
Site Name:
White Sands Missile Range, SWMU 143
Location:
NM
Period of Operation:
April - June 1998
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstrate use of injection of H2S for in situ reduction of hexavalent
chromium
Cleanup Type:
Field demonstration
Contaminants:
Heavy metals
• Cr6+
Waste Source:
Spills
Contacts:
Technical Contacts:
Ed Thornton
Pacific Northwest National
Laboratory (PNNL)
(509) 373-0358
DOE Contacts:
Jim Hanson
DOE EM50
(509) 372-4503
James A. Wright
DOE SRS
(803) 725-5608
Technology:
In Situ Gaseous Reduction (ISGR)
• ISGR involves injection of a low concentration H2S gas mixture (100-200
ppmv) into soils, where it reacts with oxidized metals such as Cr6+ and
uranium, followed by extraction of gas containing reduced metals, such as
Cr3+
• System included an injection pump, extraction pump, water knockout tank,
scrubber, one central injection well, and six extraction wells; wells were
completed to approximately 20 ft bgs
• Treatment progress was measured by breakthrough of H2S at the extraction
wells
Type/Quantity of Media Treated:
Soil (in situ)
Regulatory Requirements/Cleanup Goals:
• Objectives of demonstration were to provide technical and cost information about ISGR; obtain operational
information; and determine site air flow characteristics
• No specific cleanup goals were identified
Results:
• After completion of the demonstration, soil samples were collected from nine boreholes; these results showed that
nearly all the Cr6+ in the interval from 4 -10 ft bgs was reduced - this zone contained clean white gypsum sand that
initially contained the highest concentrations of Cr6+
• The mass of Cr6+ did not change appreciably in the 10-16 ft bgs interval, which contained a brownish sand containing
gypsum plus clay
• These results suggested that the effectiveness of ISGR is limited by subsurface heterogeneities, with channeling of the
injected gases in the most permeable white sand
• Comparison of pre- and post-demonstration soil samples showed that >70% of the Cr6+ mass was reduced, and all post-
treatment samples had <30 mg/kg of Cr6+
Costs:
• Projected costs for a full-scale application of ISGR were a total cost of $798,163, or $43/yd3
• Projected unit for ISGR were estimated to range as high as $ 100/yd3, depending on the size of the waste site
24
-------
In Situ Gaseous Reduction System Demonstrated at White Sands Missile Range,
New Mexico
Description:
The White Sands Missile Range lies within the Mexican Highland Section of the Basin and Range Province.
Contamination was discovered at SWMU 143 in January 1990 when greenish-yellow soil was found in a corner of the
equipment yard. A review of facility records indicated that several 5 5-gallon drums of Entec 300 had spilled directly onto
the ground in 1982 or 1983.
In a cooperative effort between DOE and DoD, ISGR was demonstrated at this site in the spring and summer of 1998.
The technology involved injecting 200 ppm H2S mixture into chromate-contaminated soils. Results showed that >70% of
the Cr6+ in the soil was reduced to Cr34" during the demonstration, and that all post-treatment soil samples had <30 mg/kg
of Cr6+. The amount of H2S consumed during the test was greater than the amount predicted in laboratory studies, and is
likely due to interfering reactions in the field or slower reaction kinetics. A life-cycle cost analysis suggested that ISGR
should be a less expensive remedy than excavation, especially for sites where the depth of contamination is more than 15 -
20 ft bgs. During FY 1999-2000, a deployment is planned at the DOE Hanford site to remediate Cr6+-contaminated soils
in the 100 Area.
25
-------
Electrokinetics at an Active Power Substation (Confidential Location)
Site Name:
Active Power Substation (Confidential Location)
Period of Operation:
Summer 1998 (6 month pilot-scale study)
Purpose/Significance of Application:
First field demonstration of electrokinetic remediation in the U. S. by
Electrokinetics, Inc.
Contaminants:
Heavy Metals
• Arsenic concentrations ranged from 1-1,400 mg/kg
Contacts:
Vendor:
Laurie LaChiusa
Vice President
Electrokinetics, Inc.
1 1552 Cedar Part Avenue
Baton Rouge, LA 70809]
Telephone: (225) 753-8004
E-mail: ekinc@pipeline.com
Location:
Southern U.S.
Cleanup Authority:
Not identified
Cleanup Type:
Field demonstration
Waste Source:
Herbicide use
Technology:
Electrokinetics
• Pilot-scale testing was conducted in two adjacent treatment cells - one for
arsenic extraction and one for arsenic stabilization - each measuring 30 ft
long by 20 ft wide by 3 1 ft deep (18,600 ft3)
• Each treatment cell had three anodes spaced 10 ft apart and one cathode
located 30 ft from the middle anode; the cathode was made of carbon steel
and inserted to a depth of 3 1 ft
• In the first cell, a depolarizing agent was pumped in at the cathode to create a
neutral to slightly basic catholyte
• In the second cell, proprietary reactive anodes were used to inject an arsenic-
binding compound into the soil mass
• The first cell (extraction) required 80 kW-hr per yd3; the second cell
(stabilization) 74 kW-hr per yd3; for each cell, the pH was 5 and moisture
content was 25%
• Prior to the pilot-scale tests, bench-scale studies were conducted using soil
samples from several substation sites located in the southeastern U.S.
Type/Quantity of Media Treated:
Soil
• Silty sands without heavy clay
• Soil properties include pH of 5 and hydraulic conductivity of 6 x 10"5 cm/sec
Regulatory Requirements/Cleanup Goals:
• Assess the performance of extraction and stabilization systems, and determine which configuration would yield the best
results for extracting arsenic and preventing off-site migration
Results:
• Bench-scale test results showed that >99% of arsenic was extracted; tests of arsenic -binding compounds showed that
soil passed both the TCLP and SPLP teachability tests
• A final report for the pilot-scale demonstration had not yet been submitted, and performance results are not available
for release to the public
• Results are expected to be available in the first quarter of 2000
Costs:
• Cost data are not yet available for release to the public; these are expected to be
available in the first quarter of 2000
26
-------
Electrokinetics at an Active Power Substation (Confidential Location)
Description:
A large southern U.S. power company performed bench- and pilot-scale studies of electrokinetic extraction and
electrokinetic stabilization for selected arsenic contaminated sites. After extensive analysis of both the results of bench-
scale studies on representative soils and site conditions at several substations, one active power substation site
(confidential location) was selected for pilot-scale electrokinetic treatment. Both electrokinetic extraction and
electrokinetic stabilization configurations were explored at this site.
The pilot-scale demonstration was performed using one treatment cell for arsenic extraction, that used a depolarizing
agent, and one cell for arsenic stabilization, that used proprietary reactive anodes. Results from the bench-scale tests
showed extraction of >99% of the arsenic from the soil, and that soil passed both TCLP and SPLP teachability tests.
Results from the pilot-scale tests are expected to be made available in the first quarter of 2000.
27
-------
Electrokinetics at Site 5, Naval Air Weapons Station Point Mugu, California
Site Name:
Naval Air Weapons Station Point Mugu, Site 5
Location:
Point Mugu, California
Period of Operation:
March 1998 - June 1999
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Field demonstration of electrokinetics for treatment of metals in a sandy soil
Cleanup Type:
Field demonstration
Contaminants:
Heavy metals
• Total concentrations of chromium up to 25,100 mg/kg and cadmium up to
1,810 mg/kg
• TCLP concentrations of chromium were nondetect and cadmium were 10.5
mg/L
Waste Source:
Lagoons used for wastewater
discharges from electroplating and
metal finishing activities
Contacts:
Technology Researcher:
Gene L. Fabian
Mechanical Engineer
US Army Environmental Center
Attn: SFIM-AEC-ETD
5179HoadleyRoad
APG-EA, MD 21010-5401
Telephone: (410) 436-6847
E-mail:
gene.fabian@aec.apgea.army.mil
Vendor:
Lynntech, Inc.
Technology:
Electrokinetics
• Two 1/8-acre test cells; one cell (#1) contained the two former waste lagoons
and the surrounding berms, and was an artificially confined treatment area
• The second cell (#2) was an unconfined treatment area that was open to
groundwater and tidal effects
• Operations within test cell #2 were never initiated due to performance
problems observed in cell #1
• An electrically nonconductive sheet pile barrier wall was installed to a depth
of 20 feet around the perimeter of cell #1
• Three rows of anode wells and two rows of cathode wells were installed to a
depth of 10 ft; initial current density was 0.2 mA/cm2
• By 5/98 (3 months of operation), the size of the test area was reduced (1/16
acre), and the current density was increased from 0.2 mA/cm2 to more than
0.33 mA/cm2
• In 10/98 (22 weeks of operation), the field demonstration was temporarily
suspended
• From January to June 1999, system operation resumed in a further reduced
area (approximately 500 ft2)
Type/Quantity of Media Treated:
Soil
• Soil type was sandy soil and sediment, with 85% sand, 7% gravel, 6% silt,
and 1% clay
• Soil properties included pH of 5.84, total organic carbon of 6,390 mg/kg,
hydraulic conductivity of 0.045 cm/sec, and cation exchange capacity of 3.9
Regulatory Requirements/Cleanup Goals:
• Metals - meet TCLP levels and California state total threshold limit concentration, and soluble threshold limit
concentration levels
Results:
• Analytical results of multiple soil and pore fluid samples were used to track the movement of heavy metals over time
• October 1998 results indicated that chromium was migrating towards the cathode
• June 1999 results indicated that cadmium was moving toward the surface and towards the cathode region, and that
chromium was moving toward the cathode region
• During the demonstration, elevated levels of trihalomethanes and free chlorine were found in the electrolyte solutions
28
-------
Electrokinetics at Site 5, Naval Air Weapons Station Point Mugu, California
Costs:
Not provided
Description:
The U.S. Army Environmental Center and the Engineer Research and Development Center of Waterways Experiment
Station conducted a field demonstration of electrokinetics at a metal-contaminated site at Site 5 of Naval Air Weapons
Station Point Mugu, California. NAWS Point Mugu comprises approximately 4,500 acres, and is located approximately
50 miles northwest of Los Angeles. Site 5, the Old Area 6 Shops, is a large area where electroplating and metal finishing
operations were conducted. The area of study was approximately one-half acre in and around two former waste lagoons
located in the center of Site 5. The lagoons are unlined and were used between 1947 and 1978 to receive wastewater
discharge from electroplating and metal finishing activities. Prior to the field demonstration, extensive laboratory testing
was conducted to assess the potential effectiveness of electrokinetic extraction at NAWS Point Mugu.
Results from laboratory studies showed that electrokinetics could successfully be applied to the demonstration site at
NAWS Point Mugu. During the demonstration, electrokinetics increased the mobility of cadmium and chromium at this
site. Operation of the electrokinetic extraction system at the NAWS Point Mugu site is continuing to identify and further
assess the factors that limit the performance of the technology. According to USAEC, at its current stage of development,
this technology is not considered to be sufficiently developed to be considered as a commercially available technology.
Issues to be resolved prior to full-scale commercialization include formation of trihalomethanes; effects on naturally
occurring ions; a methodology for predicting treatment performance; electrode design and its effects on electric field
shape and intensity; and a methodology for determining the configuration of the electrodes under field conditions.
29
-------
Electrokinetic Extraction at the Unlined Chromic Acid Pit, Sandia National
Laboratories, New Mexico
Site Name:
Sandia National Laboratories (SNL), Unlined Chromic Acid Pit
Period of Operation:
May 15 to November 24, 1996
Purpose/Significance of Application:
The first field demonstration of electrokinetics for removal of contaminant ions
from arid soil
Contaminants:
Heavy metals (chromium)
• Total chromium concentrations were measured in soil as high as 200 mg/kg,
up to 17 ftbgs
• TCLP chromium concentrations were measured in soil as high as 28 mg/L
Contacts:
Technology Researcher:
Dr. Eric R Lindgren
Sandia National Laboratories
P.O. Box 5800
Albuquerque, NM 87185-0719
Telephone: (505) 844-3820
E-mail: erlindg@sandia.gov
Location:
New Mexico
Cleanup Authority:
Not identified
Cleanup Type:
Field demonstration
Waste Source:
Waste pit
Technology:
Electrokinetics
• SNL's patented electrode - constructed of a porous, ceramic outer casing and
an inner, iridium-coated titanium electrode; extracts contaminants by moving
them into water held under tension (a partial vacuum) inside the outer casing
• For demonstration, three rows of electrodes in 144 ft2 area; center row - five
anodes; outer two rows - each five cathodes
• Voltage applied between electrodes - 1,572 kW hrs total; current applied to
each electrode was about 15 amps
• Additional components included a liquid control system, a vacuum control
system, a power application system, and a monitoring system
Type/Quantity of Media Treated:
Soil
• Near surface geology consists of alluvial fan deposits with some eolian
deposits
• Sediments consist of intercalated fine-to-coarse grained, well-sorted to
poorly-sorted sands, gravels, and cobbles
• Water table located 485 ft bgs
• Soil moisture content about 10 weight percent; conductivity is <10 mS/m
Regulatory Requirements/Cleanup Goals:
• Demonstrate extraction of chromate from unsaturated soils without addition of significant amounts of water
Results:
• 13 tests were performed in the demonstration (12 operating conditions; 1 system performance testing)
• A total of approximately 600 grams of hexavalent chromium were removed from the soil after 2700 hours of operation
(0.22 g/hr)
• At the system's preferred operating conditions, approximately 200 grams of hexavalent chromium were removed during
700 hours of operation (0.29 g per hour)
• After treatment, soil samples adjacent to the cathodes had total chromium concentrations of 72 ppm and TCLP
concentrations less than 5 mg/L
• Addition of significant amounts of water was not required
Costs:
Not provided
30
-------
Electrokinetic Extraction at the Unlined Chromic Acid Pit, Sandia National
Laboratories, New Mexico
Description:
Sandia National Laboratories (SNL) is located southeast of Albuquerque, New Mexico, within the boundaries of Kirtland
Air Force Base. The Unlined Chromic Acid Pit is located in the Chemical Waste Landfill at SNL, which is located in
Technical Area III. The chromium disposed of in the Unlined Chromic Acid Pit was in the form of chromic sulfuric
acids. A chromium plume resides in the vadose (unsaturated) zone beneath the pit, with the most contaminated horizon
beneath the pit containing concentrations of chromium higher than 200 mg/kg.
A field demonstration of in situ electrokinetic extraction technology was conducted at the Unlined Chromic Acid Pit to
show that chromate could be extracted from unsaturated soils on a field scale without the addition of significant amounts
of water. The field demonstration targeted the floor of the former pit at a horizon 8 to 14 feet below the surface, with
three rows of electrodes placed in a 12-foot by 12-foot area. Test results met the goal, with the soil samples adjacent to
the cathodes showing total chromium concentrations of 72 ppm and TCLP concentrations less than 5 mg/L. In addition,
the electrokinetic process was found to be stable over long periods of time. While SNL's electrokinetic extraction system
was successful in removing chromium from unsaturated sandy soil, SNL noted that the electrode system was a research
prototype and was not specifically engineered for commercialization. After the 1996 field demonstration, SNL began
developing a passive system, where the system is operated at a lower power, thereby avoiding the expense of actively
cooling the electrokinetic electrode system. The new system uses a solid matrix capture system, eliminating the need for
the liquid control and vacuum systems.
31
-------
In-Situ Thermal Desorption at the Former Mare Island Naval Shipyard, California
Site Name:
Former Mare Island Naval Shipyard
Location:
California
Period of Operation:
September to December 1997
Cleanup Authority:
California EPA
Purpose/Significance of Application:
Field demonstration of in situ thermal desorption to treat PCBs in shallow and
deep contaminated soils
Cleanup Type:
Field demonstration
Contaminants:
PCBs
• PCB levels were measured as high as 2,200 mg/kg, with an average of 220
mg/kg during a RI
Waste Source:
Contaminated wash water discharged
to soil
Contacts:
Vendors:
Mr. Gary R. Brown, P.E.
Project Oversight Manager
RT Environmental Services, Inc.
215 West Church Road
King of Prussia, PA 19406
Telephone: 610-265-1510
Fax 610-265-0587
Email: grbrwn@aol.com
Mr. Vince Fredrick, Project Manager
TerraTherm Environmental Services,
Inc.
19510 Oil Center Blvd.
Houston, TX 77073
Telephone: 281-925-0400
Fax: 281-925-0480
Email: vfredrick@terratherm.com
U.S. Navy Contacts:
Mr. Ken Spielman
EFA West, NAVFAC
Code 182
900 Commodore Drive
San Bruno, CA 94066
Telephone: 650-244-2539
Fax: 650-244-2553
Email:
khspielman@efawest.navfac.navy.mil
Mr. Chris Lonie
EFD Pacific
Env Restoration 258 Makalapa Dr
Pearl Harbor 96860-3134
Telephone: 808-474-5962
Email:
LonieCM@efdpac.navfac.navy.mil
Technology:
In-Situ Thermal Desorption (ISTD)
• Two demonstrations were conducted - a thermal well and a thermal blanket -
using the MU-125 (125 cfm capacity) unit
• 12 thermal/vapor extraction wells, installed to a depth of 14 ft bgs and
screened from 6 inches to 14 ft, used to treat deeper soil
• Two thermal blankets used to treat shallow soils
• Emissions control system included a flameless thermal oxidation unit, a heat
exchanger, and GAC augmented with pelletized calcium hydroxide
• Heating was conducted for a total of 35 days (over a period of 3 months) to
reach the target temperature of 600 °F at four central monitoring locations
• Process flow rates ranged from 38 to 82 scfm
Type/Quantity of Media Treated:
Soil
• Aquifer material - siltstone/fine-grained sandstone
• Groundwater depth - approximately 9 feet to 15 feet bgs
• Moisture content - approximately 20%
• Porosity - approximately 30%
32
-------
In-Situ Thermal Desorption at the Former Mare Island Naval Shipyard, California
Regulatory Requirements/Cleanup Goals:
• The primary performance objective for the demonstration was to treat PCBs in soil to a concentration of less than 2
mg/kg
• Off-gas limits included an HCL emission rate limit of 4.0 Ibs/hr
Results:
• All post-treatment samples had total PCB concentrations below the quantitation limit (10 ug/kg) and met the
performance objective of <2 mg/kg
• On average, the thermal wells reduced total PCBs from 53,540 ug/kg to <10 ug/kg, to a 12 ft depth
• On average, the thermal blankets reduced total PCBs from 20,607 ug/kg to <10 ug/kg, to a 1 ft depth
• The HC1 emission rate limit of 4.0 Ibs/hr was not exceeded during the demonstration
• CO emissions were below 10 ppmV with a mean concentration of approximately 2 ppmV
• Total hydrocarbon emissions ranged from 0 to 8 ppmV with a median discharge rate of less than 0.002 Ib/hr as CH4
• Excess oxygen was > 12%, except during the change over to the thermal blanket
Costs:
• Actual construction and operating costs for this project are not available
• Depending on site-specific factors, the vendor has established an overall cost range of approximately $100 to $250 per
ton
Description:
The Former Mare Island Naval Shipyard includes an electrical workshop, known as Building 866, which was used from
1955 to 1994. From 1955 to 1978, transformers washed in the workshop contained polychlorinated biphenyl (PCB) oils,
which were drained and washed into a 30-gallon sump through floor grates and drains. The liquid waste and sludge that
accumulated in the sump were pumped to a 3,000 gallon grease trap near the western corner of the building. The test site
was located in the area of the former grease trap and adjacent paved areas located at the northwest corner of Building 866.
Levels of PCBs as high as 2,200 mg/kg were identified at the site during the remedial investigation. A demonstration of
In-Situ Thermal Desorption (ISTD) using thermal blankets and thermal wells was conducted in this area by the U.S. Navy
and the Bay Area Defense Conversion Action Team (BADCAT) Environmental Technology Project (ETP).
ISTD is a combination of thermal desorption and vacuum extraction, and is conducted in-situ. Two demonstrations were
conducted (thermal well and thermal blanket) and were found to be effective in treating PCB impacted soils, achieving the
performance objective of 2 mg/kg. The results of the demonstrations suggested minor modifications in well heater
materials, control, and monitoring to aid in more even soil heating and extend heater life and efficiency. The heater
failures experienced on this project were attributable to the use of 316 stainless steel heater strips (rather than 310
stainless steel), and the initially high operating temperature of heaters.
33
-------
Soil Vapor Extraction Enhanced by Six-Phase Soil Heating at Poleline Road
Disposal Area, Fort Richardson, Alaska
Site Name:
Poleline Road Disposal Area (PROA), Operable Unit B
Location:
Fort Richardson, Alaska
Period of Operation:
Treatability Study - July through December 1997
Cleanup Authority:
CERCLA and State
Record of Decision (ROD) date
August 8, 1997
Purpose/Significance of Application:
Treatability study of SVE enhanced with SPSH to treat soil contaminated with
VOCs.
Cleanup Type:
Treatability study
Contaminants:
Organic Compounds
• Volatiles (halogenated)
- 1,1,2,2-Tetrachloroethane (TCA)
- Tetrachloroethene (PCE)
- Trichloroethene (TCE)
- Maximum concentrations: 2,030 mg/kg TCA, 159 mg/kg PCE, 384 mg/kg TCE
Waste Source:
Chlorinated solvents were used
as a carrier for neutralization
chemicals after burning of
materials in disposal trenches
Contacts:
Project Management:
USACE, Alaska District
P.O. Box 898
Anchorage, Alaska 99506-0898
Kevin Gardner
US Army, Dept of Public Works
Fort Richardson, Alaska
(907)384-3175
Vendor:
David Fleming
Current Environmental Services
P.O. Box 50387
Bellevue, Washington 98015
(425) 603-9036
david@cesiweb.com
http://cesiweb.com/index.cfm
Regulatory Contacts:
Lewis Howard
Alaska Department of Environmental
Conservation
555 Cordova
Anchorage, Alaska 99501
(907) 269-7552
Lhoward@envircon.state.ak.us
Matt Wilkening
US EPA Region 10
1200 6th Street
Seattle, Washington 98101
(206) 553-1284
wilkening. matt@epamail. epa.gov
Technology:
Soil Vapor Extraction (SVE) with Six-Phase Soil Heating (SPSH)
• Electrical power was delivered to the soil by steel electrodes inserted
vertically in a circular array. Each electrode served as an SVE vent
• Electric current passed through the soil creating steam and contaminant
vapors
• A blower pulled soil vapors from the SVE vents and through a knockout tank
to a condenser
• The condenser cooled and condensed hot vapors and separated the gas and
liquid phases
• The gas phase passed through a knockout tank and was discharged to the
atmosphere
• The liquid stream was treated by air stripping and was discharged on site
Type/Quantity of Media Treated:
• 3,910 cubic yards or 7,150 tons of soil in situ
• Soil Moisture Content: 7.3 - 13.9%
• Air Permeability (within the soil volume): 1.6 x 10
• Soil Porosity: 21 ^27%
cm
34
-------
Soil Vapor Extraction Enhanced by Six-Phase Soil Heating at Poleline Road
Disposal Area, Fort Richardson, Alaska
Regulatory Requirements/Cleanup Goals:
• System performance was evaluated against three primary criteria:
1. The ability of each of the three six-phase heating arrays to heat soil in-situ
2. Demonstrated removal of contaminants, as measured in the condenser off-gas and condensate
3. Demonstrated reduction of soil contamination, as measured in the pre- and post-treatment soil samples
• The air stripper effluent was compared to the Alaska maximum contaminant levels (MCLs) for drinking water
Results:
• The treatability study met all of the criteria established for system performance
• The air stripper effluent met Alaska MCLs
Costs:
• The total cost for this project was $967,822
• The total cost for treatment ranged from $189 to $288 per CY ($103 to $158 per ton) of soil. The soil treatment costs
ranged from $726 to $2,552 per pound of contaminant removed
• The large power requirement of the treatment equipment was a significant operating cost because the site was in a
remote location and power was provided by diesel generators
Description:
The PRDA was active from approximately 1950 to 1972. Chlorinated solvents were used as a carrier for neutralization
chemicals that were applied after burning of materials in disposal trenches. These materials included chemical warfare
agents, smoke bombs, and Japanese cluster bombs (detonated prior to burial). Four disposal areas have been identified in
an area encompassing approximately 1.5 acres. Two solvents, TCA and TCE, were found in higher concentrations and
over a larger area than any other chemicals detected. PCE was also detected above action levels. A 1996 treatability study
at the PRDA concluded that SVE was capable of removing solvent vapors from the subsurface, but at a rate that would
require more than 10 years of treatment. Based on these results, it was recommended that SVE treatment enhanced with
in-situ soil heating could be used at the site as a means for completing treatment more rapidly.
A treatability study was conducted between July and December 1997 to evaluate SVE enhanced by SPSH. Three arrays
were constructed and operated at PRDA. Two arrays were 27 feet in diameter and one array was 40 feet in diameter.
Each array was operated for six weeks after a shakedown period. The smaller arrays demonstrated over 90% removal of
soil contaminants; the larger array demonstrated over 80% removal of contaminants. These results indicated that there
may be limitations to the size of the array that can effectively treat soil at a particular site. The size of the array is limited
by the resistivity of the soil and power requirements.
35
-------
Phy to remediation at Argonne National Laboratory - West, Waste Area Group 9,
Operable Unit 9-04, Idaho Falls, Idaho
Site Name:
Argonne National Laboratory - West, Waste Area Group 9, Operable Unit 9-04
Location:
Idaho Falls, Idaho
Period of Operation:
May to October 1998
Cleanup Authority:
CERCLA
• ROD dated 9/29/98
Purpose/Significance of Application:
Bench-scale testing of phytoremediation to treat heavy metals in soil
Cleanup Type:
Bench scale
Contaminants:
Heavy metals
• Contaminants of concern included chromium, mercury, selenium, silver, and
zinc.
• Soil concentrations are 44.85 mg/kg Cr, <1.5 mg/kg total Hg, and 56.32
mg/kg total Zn
Waste Source:
Scientific and engineering research
activities
Contacts:
Technology Provider:
Ray Hinchman/M. Cristina Negri
Argonne National Laboratory
9700 S. Cass Avenue
ES-Bldg 362
Argonne, IL 60439
Telephone: (630) 252-3391/9662
E-mail:
hinchman@anl. gov/negri@anl. gov
Site Contact:
Scott D. Lee
Argonne National Laboratory - West
P.O. Box 2058
Idaho Falls, ID 83403-2528
Telephone: (208) 533-7829
Technology:
Phytoremediation
• Greenhouse experiments were performed using contaminated soil and clean
sand
• Three candidate plant species were tested: Prairie Cascade hybrid willow;
canola; and kochia
• For the soil experiment, the soil was spiked with EDTA and citric acid
• For the sand experiment, the soil was spiked with metals (soluble forms of
Cr, Zn, Hg, Ag, and Se)
Type/Quantity of Media Treated:
Soil
• Site is a relatively flat, semi-arid, sagebrush desert
• Climate conditions are a temperature range of 7.9°F - 84.8°F; a growing
season of April to mid-October; and annual average precipitation of 8.7
inches
• Soil texture is loam, with particle size distribution of 47% sand, 34.6% silt,
18.4% clay
• Soil composition is 1.59% organic matter, 5.41% lime, 5,310 mg/kg
extractable Ca, 510 mg/kg extractable Mg, 76 mg/kg extractable Na, 438
mg/kg extractable K, 48 mg/kg extractable P, 71 mg/kg soluble SO4, and 76
mg/kg soluble Na; soil pH is 8.57
Regulatory Requirements/Cleanup Goals:
• Determine uptake rates and metal concentration factors for each plant species
• Determine the most effective, non-hazardous chelating agent to increase the availability of metals from impacted soils
• Evaluate potential maximum uptake of metals by candidate plant species under selected conditions
Results:
• The optimum formulation of chelating agents for treating the metals was determined to be a 0.05 molar solution of 40%
EDTA and 60% citric acid
• In the sand experiment, the best recovery levels for zinc, chromium, mercury, and silver were found in the willow with
96%, 38%, 42%, and 24% recovery, respectively
• Testing using actual soils yielded significantly lower removals than with the sand experiment; the amount of zinc and
chromium removed was 4-5% and 2%, respectively
• The willow roots had better removal of the metals than either kochia or canola
• It was concluded that willows would be used in the field; possible removal rates of up to 14% of Zn and 3 to 4% Cr per
year were predicted, which could result in cleanup times between 6-7 years for Zn and 9 years for Cr
36
-------
Phy to remediation at Argonne National Laboratory - West, Waste Area Group 9,
Operable Unit 9-04, Idaho Falls, Idaho
Costs:
• Use of phytoremediation at full-scale for four sites at ANL-W was projected to cost $2,247,000, including management
- $528,000; documentation - $98,000; construction - $841,000; and O&M - $780,000
• The construction cost consisted of an initial 2-year field test at $300,000 and a contingency of $542,000 for five
additional years of phytoremediation
Description:
The Idaho National Engineering and Environmental Laboratory (INEEL), located in Idaho Falls, Idaho, is a government
facility managed by the U.S. DOE. Various sites at ANL-W are contaminated with wastes generated from the scientific
and engineering research at ANL-W and contain various levels of petroleum products, acids, bases, PCBs, radionuclides,
and heavy metals. The ROD for Waste Area Group 9 identifies seven areas that will undergo remediation and identifies
phytoremediation as the remedy, with a contingent remedy of excavation and disposal. As a pre-condition for
implementing phytoremediation in these areas, bench scale (laboratory and greenhouse) tests were performed to evaluate
the applicability of phytoremediation as well as to determine operating parameters and time frames for full-scale
implementation
The bench-scale tests were conducted in a greenhouse using contaminated soil and sand that was spiked with metals.
Results from these tests showed that use of contaminated soils yielded significantly lower removals than sand, with
removals from soil of chromium - 2% and zinc - 4 to 5%, and that willows were the best species for use at the site. Based
on these results, ANL-W calculated the number of years of phytoremediation that would be required to meet the
remediation goals for several site areas, and these estimates ranged from 6 to 122 years. As a next step, each of five sites
at ANL-W will be treated using phytoremediation during a two-year field test. Each site will be planted with three-foot
tall bare-root willow trees in a grid pattern, and whole tree harvesting (roots and above ground) will occur at the end of
each growing season. Excavated trees will be chipped and transported to an on-site incineration facility for disposal.
37
-------
Phytoremediation at the Open Burn and Open Detonating Area, Ensign-Bickford
Company, Simsbury, Connecticut
Site Name:
Ensign-Bickford Company, Open Burn and Open Detonating Area
Period of Operation:
April - October 1998
Purpose/Significance of Application:
Phytoremediation of lead in soil using both phytoextraction and
phytostabilization
Contaminants:
Lead
• Average concentration of total lead was 635 mg/kg; concentrations were
higher than 1,000 mg/kg in many areas of the site, with some areas exceeding
4,000 mg/kg
• Leachable lead concentrations were higher than 0.015 mg/L using the
Synthetic Precipitation Leaching Procedure (SPLP)
Contacts:
Vendor:
Dr. Michael Blaylock
Edenspace Systems Corp.
11720 Sunrise Valley Drive
Reston, Virginia 20191
Telephone: (703)390-1100
Fax: (703)390-1180
E-mail: SoilRx@aol.com
Location:
Simsbury, Connecticut
Cleanup Authority:
Not identified
Cleanup Type:
Full scale
Waste Source:
Open burn and open detonation
Technology:
Phytoremediation
• Combination of phytoextraction (for treatment of four areas with high lead
concentrations - Areas 1-4) and phytostabilization (for treatment of one area
with low lead concentrations - Area 5) to reduce total soil lead concentrations
and SPLP extractable lead
• Soils were fertilized with nitrogen, phosphorus, and potassium; dolomite lime
was added to adjust soil pH
• Fertilizers and lime were tilled into the soil to a depth of 15 to 20 cm; an
overhead irrigation system was used to provide moisture
• Areas 1-5 were seeded with Indian mustard and sunflower; 3 treatment crops
were planted
• Supplemental foliar fertilizers were added through the irrigation system
• Area 5 also treated with stabilizing amendments
Type/Quantity of Media Treated:
Soil
• Soil type is silly loam with a pH of 6.5 to 7.5
• Water table ranges in depth from 2 to 4 ft below surface soil
• Site drainage is poor; soil remains saturated throughout the growing season
(April to October)
Regulatory Requirements/Cleanup Goals:
• Reduce total lead concentrations; specific cleanup levels not identified
Results:
• Plant growth for each of the treatment crops was generally good
• Some areas within the treatment area remained saturated; these areas exhibited poor plant growth and reduced biomass
yields
• Total lead concentrations in Areas 1-4 decreased from an average of 635 mg/kg (4/98) to 478 mg/kg (10/98); by 10/98,
the highest concentrations in Areas 1-4 had been reduced
• Lead uptake ranged from 342 mg/kg (dry weight) in the Indian mustard in treatment crop 1 to 3252 mg/kg in the Indian
mustard in treatment crop 3
• Average lead uptake measured in the sunflower plant material and Indian mustard were similar, having average lead
concentrations from all crops of approximately 1000 mg/kg (dry weight).
• The average reduction in SPLP lead concentration in Area 5 was 0.95 mg/L.
Costs:
Not provided
38
-------
Phytoremediation at the Open Burn and Open Detonating Area, Ensign-Bickford
Company, Simsbury, Connecticut
Description:
The Ensign-Bickford Company in Simsbury, Connecticut, conducted open burn/open detonation (OB/OD) activities,
resulting in near surface soils in the area becoming contaminated with lead. From 1996 to 1997, Edenspace Systems
Corp. (formerly known as Phytotech, Inc.) conducted phytoremediation treatment of a 1.5 acre area surrounding the
OB/OD area. In 1998, this effort was expanded to include a total of 2.35 acres and to address not only reductions in total
lead concentrations, but also stabilizing leachable lead in the soil.
Phytoremediation was conducted using three treatment crops of Indian mustard and sunflower over a six month period.
Total lead concentrations in a portion of the site decreased from an average of 635 mg/kg (4/98) to 478 mg/kg, with hot
spots also reduced. In the area where phytostabilization also was used, the average reduction in SPLP lead concentration
was 0.95 mg/L. Further treatment is planned during 1999 and 2000.
39
-------
Phytoremediation at Twin Cities Army Ammunition Plant,
Minneapolis-St. Paul, Minnesota
Site Name:
Twin Cities Army Ammunition Plant (Site C and Site 129-3)
Location:
Minneapolis-St. Paul, Minnesota
Period of Operation:
Spring and Summer 1998
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Phytoremediation of heavy metals in soil in a northern climate
Cleanup Type:
Field demonstration
Contaminants:
Heavy metals
• Site C: antimony, arsenic, beryllium, lead, and thallium; average of 2,610
ppm lead in surface soil
• Site 129-3: antimony, barium, chromium, and lead; average of 358 ppm lead
in surface soil
Waste Source:
Burn areas, pits used for wastewater
disposal
Contacts:
Technology Contact:
Ms. Darlene F. Bader
U.S. Army Environmental Center
SFIM-AEC-ETD (Bader
5179 Hoadley Road, Bldg E4430
Aberdeen Proving Ground, MD
21010-5401
Telephone: (410)436-6861
E-mail: t2hotline@aec.apgea.army.mil
Technology:
Phytoremediation
• Demonstration used 0.2-acre plots at Site C and Site 129-3
• Sites were prepared by clearing, fencing, plowing, and installing an irrigation
system
• Two crops were grown on each site; first corn and second white mustard
• Amendments (acetic acid and EDTA) were added to the soil to aid in the
solubilization and uptake of lead
• Each crop was harvested and smelted
Type/Quantity of Media Treated:
Soil
• Climate conditions included an average annual precipitation rate of 28.6
inches and an average annual temperature of 49.6°F; the location also can
have early/late frosts
• Soil type at Site C is peat, underlain by fine sand and sandy clay; at Site 129-
3, fine- to medium-grained sand
• Depth to water table at Site C is 2 to 6 ft bgs; at Site 129-3, 140 to 200 ft bgs
Regulatory Requirements/Cleanup Goals:
• Determine if phytoextraction is a technically and economically feasible means of reducing lead contamination from
near-surface soils; specific cleanup levels not identified
Results:
• Results from the first year's demonstration showed less than anticipated biomass yields and lead uptake in the
harvested plant material
• Corn yielded 2.1 to 3.6 tons per acre, compared to the anticipated yield of 6.0 tons per acre; poor yields were attributed
to agronomically low producing soils at the site and the presence of other soil contaminants
• Lead concentrations in harvested corn averaged 0.65% and 0.13% dry weight for Sites C and 129-3, compared with the
0.85% removal obtained during a prior greenhouse study
• White mustard yielded 1.9 to 2.1 tons per acre; on a per plot basis, the total yields for Site C were half of this value
since the white mustard grew in only about 50% of the plot area
• In the areas where plants grew, the yields were comparable to the expected yield of 2 tons per acre of mustard
• Lead concentrations in harvested white mustard averaged 0.083% and 0.034% dry weight for Sites C and 129-3,
compared with the 1.5% obtained during greenhouse studies
40
-------
Phytoremediation at Twin Cities Army Ammunition Plant,
Minneapolis-St. Paul, Minnesota
Costs:
• USAEC developed a preliminary cost estimate for a typical full-scale phytoextraction project in a northern U.S.
location, with two crops grown per year (one corn and one white mustard), sub-optimal soil conditions for plant
growth, soil lead levels of about 2,500 ppm, and five years of remediation required to meet the regulatory standard
• The projected cost for full-scale phytoextraction was $30.34 per cubic yard of soil per year, or about $153 per cubic
yard of soil over the life of the project
Description:
The Twin Cities Army Ammunition Plant (TCAAP) is a 2,370-acre facility located in Arden Hills, Minnesota,
approximately 10 miles north of Minneapolis-St. Paul, Minnesota. The TCAAP was used for the production and storage
of small arms ammunition, related materials, fuzes, and artillery shell materials. A phytoremediation demonstration was
conducted at areas within Sites C and 129-3 at the TCAAP. Site C was used for burning production materials and
decontamination equipment. Site 129-3 contained pits that were believed to have contained contaminated wastewater
from a lead styphanate production facility. The project is a two-year field demonstration executed under a partnering
agreement among the U.S. Army Environmental Center (USAEC), Tennessee Valley Authority (TVA), TCAAP, and the
U.S. Army's Industrial Operations Command (IOC).
During the first year, phytoremediation was conducted at thee sites using corn and white mustard, and results were less
than anticipated. Changes planned for 1999 to improve performance included use of alternate mustard varieties; use of
higher fertilizer rates to encourage greater biomass; varying the irrigation scheme to encourage rooting and growth;
alternate amendment delivery systems; deep tilling; and alternate EDTA degradation methods.
41
-------
EG&G's ™ Aerobic Biofiltration System for the Destruction of Hydrocarbon
Vapors from Fuel-Contaminated Soils
Site Name:
Patrick Air Force Base, Active Base Exchange (BX) Service Station
Location:
Florida
Period of Operation:
1/15/94 to 2/26/94
Cleanup Authority:
RCRAUST
Purpose/Significance of Application:
Treatment of extracted vapors from an S VE system
Cleanup Type:
Field demonstration
Contaminants:
Volatile hydrocarbons and BTEX
• Initial soil gas contained TVH of 2,400 ppmv in study area
Waste Source:
Leaks from USTs
Contacts:
Vendor:
EG&G
Rotron Division
Saugerties, NY
Air Force Contact:
U.S. Air Force
Center for Environmental Excellence
Technology Transfer Division
Brooks AFB, TX
Technology:
Biocube™ (supplement to SVE)
• Demonstration used an above-ground biofiltration unit, consisting of a
proprietary mixture of inorganic and organic substrate containing active
bacteria
• Unit removed hydrocarbons by adsorption and biodegradation
• Soil vapors from a horizontal vent well at 4 ft bgs passed through knockout
drum and diluted with fresh air to maintain an influent concentration of 1,000
ppm, then passed through a humidifier prior to the Biocube™
• A recirculation loop was installed to allow multiple passes through the
biofilter
• Air emissions from the biofilter were passed through vapor-phase carbon
prior to discharge; three 5 5 -gal drums were used
Type/Quantity of Media Treated:
Soil vapors
• Average depth to water table is 5 ft
Regulatory Requirements/Cleanup Goals:
• Test objectives were to remove BTEX - >90% and TVH - >75%, based on an influent concentration of 1,000 ppmv
Results:
• In first 8 days of operation, at a flow rate of 30 scfm, no measurable differences were detected between Biocube™
influent and effluent, and the system was reconfigured
• Maximum removal efficiencies of 90.8% for BTEX and 29.5% for TVH were achieved at very low loading rates and
flow rate of 3.2 scfm
• At a 49 scfm flow rate, removal efficiencies were BTEX of 40% and TVH of 18%
• Limitations experienced during the demonstration included a relatively slow system acclimation; vacuum leaks and
dilution of process gas; and inaccurate flow measurements
Costs:
• Full-scale cost estimates were not provided based on this demonstration
• Costs provided by other biofiltration vendors showed unit costs of $18.66/kgto $38.06/kg, for TVH influent
concentrations of 1,000 to 2,000 ppmv and flow rates of 20 - 40 scfm
42
-------
EG&G's ™ Aerobic Biofiltration System for the Destruction of Hydrocarbon
Vapors from Fuel-Contaminated Soils
Description:
A field demonstration of the Biocube™ aerobic biofiltration system technology was conducted at Patrick Air Force Base
in Florida to determine the effectiveness of the technology in reducing VOCs in extracted soil vapors prior to release to
the atmosphere. The Biocube™ demonstration was tested on the soil vapors extracted from a single extraction well at the
Base Exchange service station.
The target removal efficiencies could be achieved only when the flow rate and loading were reduced to unpractically low
levels. As such, the Biocube™ could not be used as the primary vapor treatment technology when high BTEX and TVH
removal efficiencies were required.
43
-------
Internal Combustion Engines for the Destruction of Hydrocarbon Vapors from
Fuel-Contaminated Soils
Site Name:
Patrick Air Force Base, Active Base Exchange (BX) Service Station
Location:
Florida
Period of Operation:
10/18/93 to 1/14/94
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Use of an internal combustion engine to treat extracted vapors from an SVE
system
Cleanup Type:
Field demonstration
Contaminants:
Volatile hydrocarbons
• Initial soil gas contained TVH - 26,800 ppmv; benzene - ND; toluene -15
ppmv; ethylbenzene -14 ppmv; xylenes - 200 ppmv; concentrations
decreased after these initial levels were measured
• Low levels of oxygen and BTEX were found in the soil vapors
Waste Source:
Leaks from USTs
Contacts:
Vendor:
Tom Davis
VR Systems
Anaheim, CA
Telephone: (714) 826-0483
Fax: (714) 826-8746
Air Force Contact:
U.S. Air Force
Center for Environmental Excellence
Technology Transfer Division
Brooks AFB, TX
Technology:
Internal combustion engine (ICE, as a supplement to SVE)
• Demonstration used a VR Systems Model V3 Ford Motor Company 460 in3
displacement engine, 55-gallon knockout drum prior to engine, and onboard
computer
• Horizontal vent well installed at 4 ft bgs as part of a bioventing pilot test, and
used in vapor extraction mode for demonstration
• During ICE demonstration, flow rate of 150 scfm and average engine speed
of 1,790 rpm used for first 2 days, followed by a flow rate of 80 scfm
• Propane used as supplemental fuel; 1,925 ft3 used in first 2 days
Type/Quantity of Media Treated:
Soil vapors
• Average depth to water table is 5 ft
Regulatory Requirements/Cleanup Goals:
• The objectives of the demonstration included evaluating the performance and cost of the ICE technology
Results:
• Destruction efficiency was >99% for BTEX and >96% for TVH throughout the test period
• A 4% reduction in TVH destruction efficiency occurred when the engine rings and valves began to wear, allowing a
portion of the propane to pass unburned through the exhaust
Costs:
• Average operating cost was $325/day for first 5 days of operation, including equipment rental, propane, and labor
• Over the course of the test, operating costs ranged from $0.83 to 15.40/kg TVH destroyed, and $97 to 550/kg of BTEX
destroyed
• Costs varied based on soil vapor concentrations and supplemental fuel requirements
44
-------
Internal Combustion Engines for the Destruction of Hydrocarbon Vapors from
Fuel-Contaminated Soils
Description:
A field demonstration of an internal combustion engine (ICE) technology for extracted soil vapors was conducted at
Patrick Air Force Base in Florida. In Florida, soil vapor extraction must include a vapor treatment technology capable of
removing 99% of the VOCs prior to discharge. The ICE demonstration was tested on the soil vapors extracted from a
single extraction well at the Base Exchange service station.
The ICE tested was a Ford Motor Company 460 in3 displacement engine; it was preceded by a 55-gallon knockout drum.
An onboard computer was used to control system operation. For the demonstration, the initial flow rate was 150 scfm
with an average engine speed of 1,790 rpm, followed by a flow rate of 80 scfm for the remainder of the demonstration.
Propane was used as supplemental fuel. The destruction efficiency measured was >99% for BTEX and >96% for TVH
throughout the test period. The researchers found that initial soil gas oxygen levels were low, and they had to adjust flow
rates to maintain an adequate oxygen/fuel ratio. According to the researchers, ICE technology is most effective when
initial soil gas TVH is greater than 40,000 ppm, when the unit can operate without supplemental fuel.
45
-------
Purus PADRE® Regenerative Resin System for the Treatment of Hydrocarbon
Vapors from Fuel-Contaminated Soils
Site Name:
Vandenberg Air Force Base, Base Exchange (BX) Service Station
Location:
Lompoc, CA
Period of Operation:
2/11/94 to 6/1/94
Cleanup Authority:
Santa Barbara County Air Pollution
Control District
California Department of Toxic
Substances
Central Coast Regional Water Quality
Control Board
Purpose/Significance of Application:
Use of resin adsorption to treat extracted vapors from an SVE system
Cleanup Type:
Field demonstration
Contaminants:
Petroleum hydrocarbons and BTEX
• Maximum concentrations in soil: TPH-gasoline - 22,000 mg/kg; benzene -
210 mg/kg; toluene - 2,000 mg/kg; ethylbenzene - 490 mg/kg; xylenes - 2,900
mg/kg
• Maximum concentrations in soil gas: volatile hydrocarbons - 54,000 ppmv;
benzene - 400 ppmv
Waste Source:
Leaks from USTs
Contacts:
Vendor:
Purus Inc.
2713 N. First Street
San Jose, CA 95134-2000
Air Force Contact:
U.S. Air Force
Center for Environmental Excellence
Technology Transfer Division
Brooks AFB, TX
Technology:
Resin Adsorption (as a supplement to SVE)
• Demonstration used a Purus Padre® Model 1.6 vapor treatment system to treat
hydrocarbon vapors removed using soil vapor extraction (5 SVE wells; flow
rates 20 - 49 scfm)
• System used filter beds filled with synthetic polymeric adsorbent (PurSorb -
200); preceded by a water and dirt trap; two beds were used with 180 Ibs
adsorbent/bed; beds were switched between adsorption and desorption cycles
• In the desorption cycle, organic material was volatilized, condensed, and
transferred to a tank; 2-stage condenser operated at 2°C and ^45°C
• Treated soil gas with less than 1,000 ppm total hydrocarbons was returned to
the soil using a perimeter injection trench for in situ biotreatment
Type/Quantity of Media Treated:
Soil vapors
• Contamination found within a permeable silly sand, extending from 3 to 14 ft
bgs
• Depth to groundwater varies from 7 to 9 ft bgs and fluctuates seasonally
• Impermeable clay bed located between 14 and 20 ft bgs
• Soil vapor depleted of oxygen due to fuel biodegradation
Regulatory Requirements/Cleanup Goals:
• Objectives of the demonstration included evaluating the performance and cost of the Purus Padre* technology
• No air emission permit was required; instead operating conditions were established for ambient air, flux emissions, and
site emissions
Results:
• Average soil vapor concentrations reduced by factor of five during first 18 days of treatment and by factor of 20 during
110 days of operation, with increase in oxygen content
• Resin system removal rates averaged > 98% for total hydrocarbons and >99% for benzene
• The system recovered approximately 570 gallons of hydrocarbons (1,600 kgs; 3,520 Ibs) and 70 gallons of water during
the 110 day demonstration
46
-------
Purus PADRE® Regenerative Resin System for the Treatment of Hydrocarbon
Vapors from Fuel-Contaminated Soils
Costs:
• Demonstration costs were $36,634, consisting of setup - $2,500; rental - $25,667; operation labor - $4,500; power -
$1,212; nitrogen - $1,760; and mobilization/demobilization - $1,000
• Total treatment cost corresponded to a unit cost of $23/kg ($ 10.45/lb) of hydrocarbons removed
• A comparison of technologies showed that internal combustion engine technology will be less expensive that Purus
Padre® at fuel spill sites
Description:
In 1985, two 10,000 gallon unleaded gasoline tanks and associated piping were removed from the Vandenburg AFB BX
service station. Two additional gasoline storage tanks and a 250-gallon waste oil tank were removed in 1991.
Hydrocarbon contamination was discovered in soil and groundwater beneath the tanks. A two-phased bioventing pilot
test began on February 11, 1994. During phase one, high levels of hydrocarbon vapors were removed using soil vapor
extraction, treated using a Purus Padre® resin adsorption unit, and returned to the soil using a perimeter injection trench
for in situ biotreatment. When the average soil gas concentrations had been reduced to less than 1,000 ppmv, the Purus
Padre® unit was removed and soil gas returned directly through the trench.
This demonstration used a Purus Padre® Model 1.6 vapor treatment system to treat hydrocarbon vapors removed using
five soil vapor extraction wells, with a total flow rate of 20 - 49 scfm. The system used filter beds filled with PurSorb -
200; 180 Ibs adsorbentfoed were used. System removal rates averaged > 98% for total hydrocarbons and >99% for
benzene, and recovered approximately 570 gallons of hydrocarbons (1,600 kgs). Demonstration costs were $36,634,
corresponding to a unit cost of $23/kg of hydrocarbon removed. This system was found to be an effective method of
controlling vapor emissions.
47
-------
Barometrically Enhanced Remediation Technology (BERT™) Demonstration at
Idaho National Engineering and Environmental Laboratory, RWMC, Pit 2, Idaho
Falls, Idaho
Site Name:
Idaho National Engineering and Environmental Laboratory, Radioactive Waste
Management Complex, Pit 2
Location:
Idaho Falls, Idaho
Period of Operation:
December 1996 to January 1999
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstrated use of passive soil venting for remediation of VOC-
contamination
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated solvents
• Maximum concentration of carbon tetrachloride was 111 ppm
Waste Source:
Waste burial pit
Contacts:
Technical Contacts:
William E. Lowry
Science and Engineering Associates, Inc.
(505) 424-6955
E-mail: blowry@seabase.com
Eric Miller
Lockheed Martin Idaho Technologies
Company
(208) 526-9410
E-mail: ecm@inel.gov
Management Contacts:
William Haslebacher
Federal Energy Technology Center
(304) 285-5435
E-mail: whasle@fetc.doe.gov
Technology:
Barometrically Enhanced Remediation Technology (BERT™)
• BERT™ consists of a large surface area seal, a collection plenum, and a
one-way valve to vent soil gas to the atmosphere at a low rate; the system
operation relies on small changes in atmospheric pressure and wind
effects to displace soil gas
• The system at INEEL used a surface seal 100 ft by 100 ft made of 45-mil
EPDM, a collection plenum filled with YA to l/i inch pea gravel that was
10 ft thick and 30 ft diameter, and a vent pipe 6 ft tall
• In October 1996 (after almost 2 years of operation), the system was
modified to extend the collection plenum to the edges of the surface seal
to expose more soil; this was referred to as the wind-enhanced
configuration
• No boreholes or site electrical power was used in the demonstration
Type/Quantity of Media Treated:
Soil (in situ)
• Surface soils are typically less than 20 ft thick and consist of gravelly
sand and fine-grained eolian deposits; water table is 600 ft bgs
Regulatory Requirements/Cleanup Goals:
• Objectives of demonstration were to obtain technical and cost information about BERT™
• No specific cleanup goals were identified
Results:
• During the initial phase of the demonstration, the average vent flow rate was 9 mVday, with contaminants removed as
follows: TCE - 27.8 ppm and 1.15 g/day, CC14 - 5.2 ppm and 0.25 g/day, and chloroform -19.6 ppm and 0.73 g/day
• During the wind-enhanced phase of the demonstration, the average vent flow rate was 34 nrVday, with contaminants
removed as follows: TCE -18.9 ppm and 2.9 g/day, CC14 - 6.8 ppm and 1.2 g/day, and chloroform - 9.4 ppm and 1.3
g/day
• Results showed that wind speed had a greater effect on vent flow than did changes (drop) in atmospheric pressure
48
-------
Barometrically Enhanced Remediation Technology (BERT™) Demonstration at
Idaho National Engineering and Environmental Laboratory, RWMC, Pit 2, Idaho
Falls, Idaho
Costs:
• Projected costs for a full-scale application of BERT™ were $67,860 total, including materials, labor, and O&M, or
$100/yd3 ($74/ton)
• Unit costs for BERT™ were compared with costs for landfill disposal ($320/yd3, $237/ton), soil vapor extraction
($183/yd3, $136/ton), and thermal desorption ($360/yd3, $267/ton), and found to be lower on both a per cubic yard and
per ton basis
Description:
The Idaho National Engineering and Environmental Laboratory (INEEL) Radioactive Waste Management Complex
(RWMC) contains a Subsurface Disposal Area (SDA). The SDA is a 96 acre fenced disposal area where mixed wastes
containing VOCs and radioactive wastes were buried in shallow waste disposal pits, trenches, and soil vault rows.
Disposal pit 2 at the SDA received barrels of sludge between 1954 and 1965. The primary contaminant in this area was
chlorinated solvents.
The Barometrically Enhanced Remediation Technology (BERT™) was demonstrated at this site. BERT™ induces net
upward displacement of soil gas based on small changes in atmospheric pressure and wind speed. A system was
demonstrated that was 100 ft long by 100 ft wide, and that required no boreholes or site power. During the initial phase of
the demonstration, the average vent flow rate was 9 nf/day, with removals of TCE, CC14, and chloroform ranging from
0.25 to 1.15 g/day. During the wind-enhanced phase of the demonstration, the average vent flow rate was 34 nrVday, with
removals of TCE, CC14, and chloroform ranging from 1.2 to 2.9 g/day. Results showed that wind speed had a greater
effect on vent flow than changes/drops in atmospheric pressure. Unit costs for BERT™ were compared with costs for
landfill disposal, soil vapor extraction, and thermal desorption, and found to be lower on both a per cubic yard and per ton
basis. A BERT™ system is currently under construction at Los Alamos National Laboratory, with operation anticipated
by the end of July 1999.
49
-------
This Page Intentionally Left Blank
50
-------
INCINERATION ABSTRACTS
51
-------
On-Site Incineration at Weldon Spring Ordnance Works, St. Charles County,
Missouri
Site Name:
Former Weldon Spring Ordnance Works (WSOW), Operable Unit 1
Location:
St. Charles County, Missouri
Period of Operation:
• Trial Burn - 8/14/98 to 8/16/98
• Interim Operation - 8/17/98 to 9/18/98
• Full-Scale Operation - 9/19/98 through 1999
Cleanup Authority:
CERCLA and State
Record of Decision (ROD) date - May
1996
Purpose/Significance of Application:
Full-scale, on-site incineration of explosives and propellants
Cleanup Type:
Full scale
Contaminants:
Explosives/Propellants
• 2,4,6-Trinitrotoluene (TNT) and 2,4- and 2,6-Dinitrotoluene (DNT)
• Maximum concentrations:
- 510,632 mg/kg TNT
- 7,100mg/kg2,4-DNT
- 200 mg/kg 2,6-DNT
• Some soil contaminated by lead, asbestos, PCBs, and PAHs
Waste Source:
Discharge and leaks/spills of
contaminated wash water and
wastewater; open burning of
explosives
Contacts:
Project Management:
Mr. Dan Mroz
USACE, Kansas City District
USACE-MD-H
60 IE. 12th Street
Kansas City, KS 64106
(816) 983-3567
Captain Jim Workman
USACE
Big Piney Building 1018
P.O. Box 200
Ft. Leonard Wood, MO 65473
(314)498-5176
Vendor:
Mr. Alan J. Zupko
Roy F. Weston, Inc.
1 Weston Way
West Chester, PA 19380-1499
(610) 701-3623
Regulatory Contacts:
Mr. Tom Lorenz
U.S. EPA Region 7
726 Minnesota Avenue
Kansas City, KS 66101
(913)551-7292
Mr. Ray Strebler
Missouri Department of Natural Resources,
Hazardous Waste Program,
Division of Environmental Quality
P.O. Box 176
Jefferson City, MO 65102-0176
(573)751-7241
Technology:
On-Site Incineration
• Excavated pipeline and combustible debris were shredded
• Soil and shredded materials were fed through a screen to remove large
debris
• The incineration system consisted of a co-current, rotary kiln and a
secondary combustion chamber (SCC)
• The kiln operated at an exit gas temperature above 1626 °F and the SCC
operated above 1823°F
• Hot gases exiting the SCC passed through a two-stage spray tower and
two pulse-jet baghouses in parallel
• Treated soil and fly ash were stockpiled for compliance sampling
• Treated soil and fly ash that met treatment standards were used as fill
material at the site
Type/Quantity of Media Treated:
• An estimated 30,000 tons (18,000 cubic yards) of nitroaromatics-
contaminated soil
• An estimated 85,230 linear feet of nitroaromatics-contaminated wooden
pipeline
• Average Moisture Content: 18%
• BTU Value: 60 Btu/lb
• Pipeline: 1 linear foot weighed approximately 25 pounds; the shredded
density was 0.43 tons/CY
52
-------
On-Site Incineration at Weldon Spring Ordnance Works, St. Charles County,
Missouri
Regulatory Requirements/Cleanup Goals:
• Destruction and Removal Efficiency (ORE) of 99.99% for POHC
• Regulatory limits for treated soil and fly ash after incineration were 57 ppm TNT and 2.5 ppm 2,4- and 2,6-DNT
• Treated soil and fly ash with TCLP values in excess of 5 mg/L lead were stabilized
• Air emission requirements included control of metals, hydrogen chloride, chlorine, 2,3,7,8-tetrachlorinated dibenzo-p-
dioxin toxic equivalents, carbon monoxide, total hydrocarbons, nitrous oxides, paniculate matter and opacity in the
stack gas
Results:
• Sampling of treated soil and fly ash indicated that the soil and pipeline cleanup goals were met
• Emissions data from the trial burn, interim operations and full-scale operations indicated that all emissions standards
were met
Costs:
• The total cost for this project was $13,665,997 including all remedial activities performed at the site, including
incineration
Description:
The former Weldon Spring Ordnance Works included a nitroaromatics manufacturing facility operated by the Army
between 1941 and 1945. Wash water and wastewater generated in the TNT and DNT production plants were discharged to
settling lagoons at the WSOW prior to mid-1942 and to wastewater treatment plants via an underground wooden pipeline
after mid-1942. Leaks and spills occurred at the production buildings and the wastewater pipeline. Open burning was used
to dispose and/or treat off-specification material, surplus product and contaminated soil. Nitroaromatics were detected in
surface soil, shallow subsurface soil, groundwater and springs at the former WSOW. A ROD was signed in September
1996, specifying on-site incineration as the remedial technology for addressing nitroaromatics-contaminated soil and
wooden pipeline at the site. Contaminated soil and pipeline at the former WSOW was identified as Operable Unit (OU) 1.
Site cleanup goals were specified in the ROD.
Site work for construction of the incinerator was commenced in December 1997. Incinerator start up and shake down were
performed in July and August 1997. The trial burn was conducted in August 1998. After receiving approval from EPA
and MDNR of the proposed operating limits, the incinerator was put into full-scale operation in September 1998.
Treatment was completed in April 1999. The incineration system consisted of a co-current, rotary kiln followed by a SCC.
After confirming that treated soil and fly ash met the cleanup criteria, the materials were backfilled at the site.
Demobilization of the incinerator from the site was completed in 1999.
53
-------
This Page Intentionally Left Blank
54
-------
THERMAL DESORPTION ABSTRACTS
55
-------
Thermal Desorption at the Arlington Blending and Packaging Superfund Site
Arlington, Tennessee
Site Name:
Arlington Blending and Packaging Superfund Site
Location:
Arlington, TN
Period of Operation:
January 13 to June 4, 1996
Cleanup Authority:
CERCLA - Remedial Action
• ROD signed June 28, 1991
Purpose/Significance of Application:
Application of low temperature thermal desorption to treat pesticide-
contaminated soil
Cleanup Type:
Full scale
Contaminants:
Pesticides and Metals
• Maximum concentrations during remedial investigation: chlordane (390
mg/kg surface and 120 mg/kg subsurface); endrin (70 mg/kg surface and 20
mg/kg subsurface); pentachlorophenol (130 mg/kg surface and 9.5 mg/kg)
subsurface; arsenic (370 mg/kg surface)
Waste Source:
Leaks and spills of pesticides during
blending and packaging operations;
process wastewater discharged to
drainage ditches at the site
Contacts:
Vendor:
Smith Environmental Technologies
Corporation (formerly Canonie)
EPA Remedial Project Manager:
Derek Matory
U.S. EPA Region 4
345 Courtland Street, NE
Atlanta, GA 30365
Telephone: (404) 562-8800
Fax: (404) 562-8788
E-mail: matory.derek@epa.gov
Additional Contacts:
George Harvell
Memphis Environmental Center
2603 Corporate Avenue, Suite 100
Memphis, TN 38132
Telephone: (901) 345-1788
Fax: (901) 398-4719
Paul Sadler
Senior Project Engineer
Focus Environmental, Inc.
9050 Executive Park Drive
Knoxville, TN 37923
Telephone: (423) 694-7517
E-mail: psadler@focusenv.com
Technology:
Low Temperature Thermal Desorption
• Direct-fired rotating dryer that heated the soil to between 580 and 750°F
using a hot air stream
• Propane gas was used to heat the air stream, and the organic constituents in
the soil were desorbed in the dryer through contact with the heated air
• Off-gas treatment included a cyclone/baghouse system; a low pressure drop
Venturi air scrubber; and vapor-phase carbon adsorption
• A vacuum of 0.10 to 0.18 inches of water was maintained throughout the
process train
Type/Quantity of Media Treated:
Soil-41,431 tons
• Soils primarily silty sands with an average moisture content of 17 wt%
• pH of soil-6.8
Regulatory Requirements/Cleanup Goals:
• Cleanup goals for organics were: chlordane (3.3 mg/kg); heptachlor (0.3 mg/kg); pentachlorophenol (0.635 mg/kg);
endrin (0.608 mg/kg); heptachlor expoxide (0.2 mg/kg)
• Cleanup goal for arsenic initially established at 25 mg/kg in ROD; changed to 100 mg/kg in BSD. All treated soil with a
total arsenic concentration >100 mg/kg was to be disposed of off-site. Any treated soil with total arsenic concentrations
>100 mg/kg and leachable arsenic >5mg/L (determined by the toxicity characteristic leaching procedure) was required
to be identified as hazardous waste and stabilized prior to disposal off-site
• Emission standards for the unit were total hydrocarbons (500 ppmv); particulates (0.08 gr/dscf); and system removal
efficiency (>95%)
56
-------
Thermal Desorption at the Arlington Blending and Packaging Superfund Site
Arlington, Tennessee
Results:
• A total of 84 batches of soil (41,431 tons) were treated
• All but six batches of soil met the cleanup goals for the organics on the first pass through the system
• Three batches exceeded the cleanup levels and were retreated and met the cleanup goals
• An additional three batches were slightly above the cleanup levels for total chlordane; based on the concentrations, EPA
determined that the batches were not required to be retreated
• One batch of treated soil did not meet the 100 mg/kg limit for arsenic and was shipped offsite for disposal in a Subtitle
C landfill; however, because the TCLP level for arsenic was below the 5 mg/L limit, solidification/stabilization prior to
off-site disposal was not required
• Compliance with the emissions standards was verified during the performance test; the unit met all emissions standards
during the three test runs, achieving a system removal efficiency >99.999%
Costs:
• Total project cost was $5,586,376 including $4,356,244 in costs directly associated with the thermal treatment
• Treatment costs included $4,293,893 in capital costs and $62,351 in O&M costs
• The calculated unit cost for this application was $105 per ton, based on 41,431 tons of soil treated
Description:
The Arlington Blending and Packaging Superfund site, located in Arlington, Tennessee, is a 2.3 acre site that was used for
the formulation and packaging of pesticides and herbicides from 1971 to 1978. Chemicals handled at the facility included
the pesticides endrin, aldrin, dieldrin, chlordane, heptachlor, lindane, methyl parathion, and thimet as well as solvents and
emulsifiers used in the formulation operations. Leaks and spills of chemicals occurred during these operations and process
wastewater was discharged to drainage ditches at the site. The site was placed on the National Priorities List (NPL) in July
1987. A remedial investigation (RI), begun in 1988, determined that the main areas of soil contamination at the site were
located around and beneath the process buildings. The ROD, signed in 1991, specified excavation of contaminated soil and
treatment on site using thermal desorption.
Smith's low temperature thermal aeration (LTTA) process was used to treat the contaminated soil at the site. The unit
included a direct-fired rotating dryer that heated the soil using a hot air stream. The heated soil was discharged from the
rotary dryer to an enclosed pugmill where it was quenched with water to cool and rehumidify the soil. The treated soil was
then sampled, and based on the results, backfilled on site or stabilized and shipped off-site for disposal. A total of 41,431
tons of contaminated soil in 84 batches were treated during this application. All but six batches of soil met the cleanup
goals for the organics on the first pass through the system. Three batches exceeded the cleanup levels and were retreated.
Three additional batches slightly exceeded the cleanup goal for total chlordane. EPA determined, based on the
concentrations, that the batches did not have to be retreated. Following confirmation that the cleanup goals had been met,
treated soil was backfilled at the site. Only one batch of treated soil did not meet the total arsenic limit and was shipped
offsite for disposal in a Subtitle C landfill. The original estimate for the soil excavation was 10,000 tons, based on the
results from field-based screening using the Drexil method. Subsequent verification analyses indicated that the results
from this method were not accurate. The site was recharacterized, using immunoassay sampling (results confirmed to be
accurate by an off-site laboratory), and an additional 30,000 tons of soil requiring excavation were identified. The use of
immunoassay sampling saved time by providing real time results (versus 5 to 6 day turnaround time for an off-site
laboratory).
57
-------
Thermal Desorption at Letterkenny Army Depot Superfund Site, K Areas, OU 1
Chambersburg, Franklin County, Pennsylvania
Site Name:
Letterkenny Army Depot Superfund Site K Areas, OU 1
Location:
Chambersburg, Franklin County, Pennsylvania
Period of Operation:
September 1993 to October 1994
Cleanup Authority:
CERCLA
• ROD signed June 28, 1991
• ROD modified August 2, 1991
Purpose/Significance of Application:
Thermal desorption to treat VOC-contaminated soil, including soils with
high oil and grease content
Cleanup Type:
Full scale
Contaminants:
Volatile Organic Compounds and Metals
• Maximum concentrations of TCE of 30,000 mg/kg in soils in K areas
• Maximum concentrations of lead of 10,000 mg/kg in soils in K areas
Waste Source:
Disposal of waste in lagoons; leaks and spills
from waste solvent drum storage area
Contacts:
Vendor:
McLaren/Hart, Inc
300 Stevens Drive
Philadelphia, PA 19113
EPA Contact:
Stacie Driscoll
U.S. EPA Region 3
1650 Arch Street
Philadelphia, PA 19103
Telephone: (215) 814-3368
Facsimile: (215) 814-3001
E-mail: driscoll.stacie@epamail.epa.gov
USACE Contact:
Paul Stone
U.S. Army Corps of Engineers (USACE)
Baltimore District
PO Box 1715
Baltimore, MD 21203-1715
Telephone: (410) 962-4906
Facsimile: (410) 962-6732
E-mail: Paul.R.Stone@nab02.usace.army.mil
Technology:
Low Temperature Thermal Desportion
• Patented I.R.V.-100 LTTD system
• 1.2 million BTU/hr system; six carbon steel treatment chambers (5
cubic yards of soil per chamber capacity)
• Each chamber equipped with 16 propane-fired infrared heaters; soil
temperature of 600° F
• System operated at under a vacuum of 12 to 20 column inches of
water; volumetric air flow of 500 to 1,000 cubic feet per minute per
chamber
• Residence Time- 60 minutes for clay soils and 120 to 150 minutes for
black stained soils
• Off-gas treatment included two cyclones, two air expansion chambers
to cool the temperature of the air from about 120° F to about 90 °F,
and one 4,000 pound activated carbon adsorption unit
Type/Quantity of Media Treated:
Soil -13,986 cubic yards (11,366 cubic yards of clay soil; 2,620 cubic
yards of black stained soil)
Regulatory Requirements/Cleanup Goals:
• ROD specified cleanup goal for TCE in treated soil - 0.05 mg/kg
• RCRA Land Disposal Restriction treatment standards for the following VOCs - acetone, benzene, carbon tetrachloride,
chlorobenzene, o-dichlorobenzene, 1,1,1-trichloroethane, 1,1,2-trichloroethane, tetrachloroethene, ethylbenzene,
toluene, and total xylene) and for metals
• No goals were established for total RCRA metals
• Emissions standards for the unit included an opacity limit of < 10% for 30 minutes, total VOC emissions of < 1
pound/hour, and paniculate matter of < 0.08 grains per dry standard cubic foot
58
-------
Thermal Desorption at Letterkenny Army Depot Superfund Site, K Areas, OU 1
Chambersburg, Franklin County, Pennsylvania
Results:
• A total of 13,986 cubic yards of contaminated soil were treated to the cleanup goals; information on the total number of
batches treated was not provided
• Soil that did not meet the cleanup goals on the first pass were retreated until the goals were met; approximately 10% of
the clay soils and 14% of the black stained soils from the K-l area required retreatment
• Treated soil that exceeded the TCLP limits for metals required to be stabilized and shipped off site for disposal;
treated soils that was excavated from the top 6 feet of the K-l area and the top 3 feet of the K-2 area were stabilized
prior to off-site disposal; a total of about 4,000 cubic yards of treated soils was stabilized prior to disposal off site
• The remaining treated soil was backfilled on-site
Costs:
• Total actual project cost - $5,402,801, including $4,647,632 in actual costs for McLaren/Hart's application of thermal
treatment and other project costs identified by USAGE for design and project remediation ($192,827), design contract
costs ($249,320), and construction contract management ($312,320)
• The unit cost for the application was $220 per cubic yard, based on 13,986 cubic yards of soil treated
• McLaren/Hart's actual costs of $4,647,632 include $2,622,470 for five modifications to the contract
• USAGE subsequently paid McLaren/Hart a total of $3,905,256 for the remediation of the K area soils, as a result of a
settlement agreement regarding costs of the modifications
Description:
The Letterkenny Army Depot is a 19,243-acre U.S. Army facility located in Chambersburg, Franklin County,
Pennsylvania. Since 1942, the Army has used the site to overhaul, rebuild, and test missile systems; store and demilitarize
ammunition; and maintain and refurbish equipment and vehicles. Operations at the facility have included degreasing,
metal plating, painting and paint stripping, steam cleaning, and petroleum storage. Wastes from these operations were
disposed of in landfills, trenches, pits, and surface impoundments at the site. Site investigations identified elevated levels
of volatile organic compounds in soil and groundwater in the site, including three areas of soil contamination, also referred
to as the K areas. K-l was a waste disposal lagoon, K-2 was used as a transfer station, and K-3 was an area used to store
drums of waste solvent. A 1992 remedial investigation identified elevated levels of TCE, polychlorinated biphenyls
(PCBs), metals, and semivolatile organic compounds (SVOCs) in soils in the K areas. A Record of Decision, signed in
June 1991, specified excavation of VOC-contaminated soil and on-site treatment using low temperature thermal
desorption.
A low temperature thermal desorption system (LTTD), model I.R. V.-100 designed by McLaren/Hart, was used to treat the
contaminated soil from the K areas. The 1.2 million BTU/hr system, operated under vacuum, included a total of six
carbon steel treatment chambers used to heat soils to temperatures up to 600°F. The unit operated from September 1993
to October 1994. A total of 13,986 cubic yards of soil were treated during this application, including 2,620 cubic yards of
"black stained" soils that were encountered during the excavation of areas K-l and K-3. The black stained soils contained
heavy oils, greases, and debris and were stockpiled separately from the "clay soils" for treatment. Approximately 10% of
the clay soils and 14% of the black stained soils from the K-l area were retreated. In addition, a total of about 4,000 cubic
yards of treated soils was stabilized prior to disposal. This included treated soil that was above the TCLP metals levels and
from the top 6 feet of K-l area and top 3 feet of K-2 area. The remaining treated soil was backfilled on-site. According to
vendor, the presence of the black stained soils had not been anticipated at the time of the original contract. The adverse
effects of these soils on the operation of the unit, from the heavy hydrocarbons in the soil, were discovered during the first
demonstration test and required modification to the design and operation of the system, including expansion of the
emissions controls. This resulted in increased costs and a delay in the schedule over the original plan.
59
-------
Low Temperature Thermal Desorption at Longhorn Army Ammunition Plant,
Karnack, Texas
Site Name:
Longhorn Army Ammunition Plant, Burning Ground No. 3
Location:
Karnack, Texas
Period of Operation:
• Proof of Performance Test-February 1997
• Full-Scale Operation - February to December 1997
Cleanup Authority:
CERCLA and State
ROD date - May 1995
Purpose/Significance of Application:
Thermal desorption to treat chlorinated solvents in the site soil and source
materials
Cleanup Type:
Full scale
Contaminants:
Organic Compounds - Volatiles (Halogenated)
• Trichlororethylene (TCE) and Methylene Chloride
• Maximum concentrations in mg/kg - TCE (1,000 mg/kg) and Methylene
Chloride (742 mg/kg)
Waste Source:
Open burning, incineration,
evaporation, and burial of pyrotechnic
and combustible solvent wastes
Contacts:
Project Management:
Jonna Polk
USACE, Tulsa District
1645 South 101st Avenue
Tulsa, OK 74128-4629
Oscar Linebaugh
USACE, Ft. Worth District
Eastern Area Office
(318)676-3365x225
David Tolbert
Longhorn/Louisiana Army
Ammunition Plant
Highway 80 East, Gate 4
Doyline, LA71055
(903) 679-2054
Vendor:
Bryan Smith
Radian International LLC
Longhorn Army Ammunition Plant
P.O. Box 107
Karnack, TX 75661
(903) 679-3448
Regulatory Contact:
Chris Villarreal
U.S. EPA Region 6
1445 Ross Avenue, Suite 1200
Dallas, TX 75202-2733
(214) 665-6758
Diane Poteet
TNRCC
Superfund Investigation,
MC-143
12100 Park 35 Circle, Bldg. D
Austin, TX 78753
Technology:
On-Site Low Temperature Thermal Desorption (LTTD)
• Soil was fed through a vibrating screen to remove large debris
• Soil passed counter-current to hot combustion gases in one of two parallel
LTTD units
• Soil was heated between 350 and 650°F
• The gas stream from each LTTD unit passed through a baghouse and then the
two streams were combined
• The combined gas stream was preheated to 680°F prior to entering the
catalytic oxidizer where desorbed VOCs in the gas stream were destroyed
• Hot gases exiting the oxidizer passed through a heat exchanger, multi-stage
quench and packed bed scrubber
• Solids exiting the thermal desorption units and baghouses were stockpiled for
compliance sampling
Type/Quantity of Media Treated:
Soil (ex situ)
• 32,293 cubic yards (51,669 tons) of soil
• Average Clay Content: 31.5%
• Mean Particle Size: 0.032mm
• Average Moisture Content: 17.5 %
• Bulk Soil Density: 1.6 tons per cubic yard
60
-------
Low Temperature Thermal Desorption at Longhorn Army Ammunition Plant,
Karnack, Texas
Regulatory Requirements/Cleanup Goals:
• If TCE or methylene chloride concentrations in the soil were below 40 mg/kg, the treatment objective was to reduce the
concentrations to 2 mg/kg or lower
• If TCE or methylene chloride concentrations in the soil exceeded 40 mg/kg, the treatment objective was to reduce the
concentrations by at least 95%
• Air emission requirements included control of total chemical emissions, paniculate matter and 2,3,7,8-tetrachlorinated
dibenzo-p-dioxin toxic equivalents in the stack gas
Results:
• Sampling of treated soil indicated that all soil cleanup goals were met
• Emissions data from the Proof of Performance test and full-scale operations indicated that all emissions standards were
met
Costs:
• The total cost for this project was $4,886,978
• The total cost for treatment was $ 151 per cubic yard ($95 per ton) of contaminated material
Description:
Burning Ground No. 3 was operational from 1955 to 1997. The site was used for the treatment, storage, and disposal of
pyrotechnic and combustible solvent wastes including open burning, incineration, evaporation and burial. Site
investigations indicated the presence of high concentrations of chlorinated solvents and heavy metals in subsurface soils
and shallow groundwater at the site. In addition, buried sawdust and other solvent-contaminated wastes were encountered.
A ROD was signed in May 1995, specifying LTTD as the remedial technology for addressing soil contamination at the
site. Site soil cleanup goals were specified in the ROD.
Mobilization and set-up of the soil treatment plant (STP) occurred in January 1997. System start-up and shake down and
the Proof of Performance test were conducted in February 1997. After successfully demonstrating that the STP could meet
performance requirements, the STP was put into full production. Soil/source material excavation and full-scale operation
of treatment system was performed between February and December 1997. The STP consisted of a counter-current, LTTD
system followed by a low-temperature, catalytic oxidation system to treat the LTTD off-gas. After confirming that treated
soil met the cleanup criteria, the soil was used as general fill material for landfill caps at two sites at the LHAAP.
Demobilization of the STP from the site was completed in January 1998 and site restoration was completed by June 1998.
61
-------
Thermal Desorption at the Rocky Flats Environmental Technology Site,
Trenches T-3 and T-4, Golden, Colorado
Site Name:
Rocky Flats Environmental Technology Site, Trenches T-3 and T-4
Location:
Golden, Colorado
Period of Operation:
June - August 1996
Cleanup Authority:
CERCLA - Removal
• Action Memorandum
Date - January 18, 1996
Purpose/Significance of Application:
Application of thermal desorption to treat soils contaminated with VOCs and low levels of
radiation
Cleanup Type:
Full scale
Contaminants:
Chlorinated solvents, ketones, and low level radionuclides
• The highest concentrations of VOCs in trench T-3 were TCA at 13,000 mg/kg, acetone at
5,100 mg/kg, methylene chloride at 2,400 mg/kg, and carbon tetrachloride at 700 mg/kg
• The highest concentrations of VOCs in trench T-4 were TCE at 680 mg/kg and acetone at
120 mg/kg
• Subsurface soils contaminated with low levels of radionuclides including uranium,
plutonium, and tritium
Waste Source:
Burial of drums and debris
in trenches on the site
Contacts:
Vendor:
Ronnie Hill
Principal Construction Manager
McLaren-Hart, Inc.
9323 Stockport Place
Charlotte, NC 28273
(704) 587-0003
ronnie_hill@mclaren-hart. com
EPA Contact:
Tim Rehder
Rocky Flats Project Coordinator
U.S. EPA Region 8
999 18th Street, Suite 500
Denver, CO 80202-2466
(303)312-6293
rehder.timothy@epa.gov
State Contact:
Steve Gunderson
CDPHE Rocky Flats Cleanup Agreement Coordinator
4300 Cherry Creek Dr. South
Denver, CO 80246-1530
(303) 692-3367
steve.gunderson@state.co.us
DOE Contact:
Hopi Salomon
Rocky Mountain Remediation Services, LLC
Rocky Flats Environmental Technology Site
P.O. Box 464
Golden, CO 80402-0464
(303) 966-2677
Fax: (303) 966-8244
Technology:
Vacuum-enhanced low temperature thermal desorption
• IRV-100 system manufactured by McLaren-Hart
• 6 treatment chambers (18 feet long, 8 feet wide and 5 feet
high; operating capacity of 5yd3 per chamber)
• Each chamber equipped with 16 propane units
• Energy output of total system (infrared energy) -1.5 MM
Btu/hr
• Vacuum condition in treatment chamber - 500 mm Hg
• Air flow rate -1,000-3,000 cfm
• Residence time - 5.25 hours
• System throughput -1 yd3/hour
• Soil temperature - 250°F
• Emissions controls - two dry paniculate filters (in series), a
condenser, and a granular activated carbon unit
Type/Quantity of Media Treated:
Soil and debris - 3,796 cubic yards
• Soils consist of sandy and clayey gravel
• Moisture content approximately 8%
62
-------
Thermal Desorption at the Rocky Flats Environmental Technology Site,
Trenches T-3 and T-4, Golden, Colorado
Regulatory Requirements/Cleanup Goals:
Cleanup goals specified for 12 VOCs:
• Goal of 6 mg/kg each for TCE, TCA, PCE, DCE, DCA, chloroform, carbon tetrachloride
• Goal of 10 mg/kg each for benzene, ethylbenzene, toluene
• Goal of 160 mg/kg for acetone and 30 mg/kg for methylene chloride
Results:
• A total of 58 batches (3,796 yd3 total) of soil were treated during this application
• Of the 58 batches treated, 52 met the cleanup goals on the first pass, including 20 batches where all 12 VOCs were
below the detection level
• Six batches did not meet the cleanup goals on the first pass, exceeding the level for PCE; these batches were retreated
and met the cleanup goals
• Concentrations of six VOCs (TCA, DCE, DCA, carbon tetrachloride, chloroform, and ethylbenzene) were below the
detection level in all 58 batches
• According to the vendor, there were no exceedances of the applicable air emissions standards
Costs:
• The total cost for this project was $1,934,203, including $1,328,600 in costs directly associated with the thermal
treatment
• The calculated unit cost was $3 50/yd3 based on the treatment of the 3,796 yd3 of contaminated soil and debris
• The original contract cost was $ 1,200,000, based on treating 2,200 yds3 of contaminated soil; two change orders were
issued for the remediation of additional soil volumes, changing the total amount of soil treated from 2,200 yd3 to 3,796
yd3, with a final project cost of $1,934,204
Description:
From 1951 to 1989, the U.S. Department of Energy (DOE) used the Rocky Flats site to process and store plutonium,
manufacture components for nuclear weapons, fabricate, machine, and assemble components from metals, and store
solvents used in the manufacturing processes. Hazardous and radioactive wastes were stored and disposed of at various
locations at the site, including on-site trenches. Waste handling practices at the site also included recycling of hazardous
materials. Trenches T-3 and T-4 were used for the disposal of sanitary sewage sludge contaminated with uranium and
plutonium and miscellaneous debris, primarily flattened drums contaminated with volatile organic compounds (VOCs),
uranium, and plutonium. Subsurface soils in trenches T-3 and T-4 were found to contain elevated levels of VOCs,
semivolatile organic compounds, and metals, along with low-level radiological contamination. A Proposed Action
Memorandum (PAM) was issued in January 1996 calling for thermal treatment of the T3/T4 soils. Prior to treatment, each
load of excavated soil was screened using a Field Instrument for the Detection of Low Energy Radiation (FIDLER) to
identify "potentially radiologically contaminated material". Soil with readings above 5,000 counts per minute (cpm) was
segregated and treated separately from the soil that was not considered to be potentially radioactive. A total of about 380
cubic yards of soil were identified as potentially radioactive.
The thermal desorber used at this site, an infrared radiation-heated unit manufactured by McLaren-Hart (the IRV-100
system), was a modular, batch-operated vacuum system, equipped with six treatment chambers. The system was operated
under a vacuum of approximately 500 mm Hg and soil was heated to temperatures of 250 °F. Thermal treatment
operations were conducted from June to August, 1996. A total of 58 batches (3,796 yd3 total) of soil were treated during
this application. Fifty-two of the batches met the cleanup goals on the first pass. The six batches that did not meet the
cleanup goals were retreated and met the cleanup goals. The total project cost was $1.9 million with the cost for the
thermal treatment application being $1.3 million or $350/yd3 (based on 3,796 yd3 of contaminated soil and debris).
According to vendor, the total project cost would likely be less for a similar application at sites where radiological
engineering controls were not required.
63
-------
This Page Intentionally Left Blank
64
-------
OTHER EX SITU SOIL TREATMENT ABSTRACTS
65
-------
Slurry Reactor Biotreatment of Explosives-Contaminated Soils at Joliet Army
Ammunition Plant, Joliet, Illinois
Site Name:
Joliet Army Ammunition Plant
Location:
Joliet, Illinois
Period of Operation:
July 1994 to August 1995
Cleanup Authority:
CERCLA
• Final ROD scheduled for June 2001
Purpose/Significance of Application:
Use of bioslurry technology for treatment on explosives wastes
Cleanup Type:
Field demonstration
Contaminants:
Explosives
• Excavated soils had concentrations of TNT -1,000 - 6,226 mg/kg; DNT - ND
- 360 mg/kg; TNB - 48 - 360 mg/kg; RDX - ND - 310 mg/kg; and HMX - ND
-215 mg/kg
Waste Source:
Process water from munitions washout
Contacts:
Technical Contacts:
J.F. Manning, Jr., R. Boopathy, and
E.R. Breyfogle
Argonne National Laboratory
Environmental Research Division
Bioremediation Group
9700 South Cass Avenue
Argonne, IL 60439-4843
Mark Hampton
U.S. Army Environmental Center
SFIM-AEC-ETD
Aberdeen Proving Ground, MD
21010-5401
(410) 436-6852
mark.hampton@aec.apgea.army.mil
EPA Remedial Project Manager:
Diana Mally
U.S. EPA Region 5
77 W.Jackson Blvd.
Chicago, IL 60604
(312)886-7275
E-mail: mally.diana@epa.gov
Technology:
In Situ Bioremediation
• Field bioslurry system included a soil screening operation, four 420-gallon
bioslurry reactor tanks (variable speed drive mixer with double impeller); two
slurry dewatering beds; and tanks for water storage
• Bioslurry demonstration was performed in the reactors (350-380 gals/reactor),
with addition of molasses, pH adjustment (to >6), and aerobic-anoxic
operating cycles
• Four reactors were operated: (1) a control with no molasses; (2) a 20% weekly
replacement; (3) a 10% weekly replacement; and (4) a 5% daily replacement
• All reactors were operated with a 10-16% WAV soil slurry in a sequencing
batch mode
• Soil was screened to 40 mesh (0.0165 inch) prior to placement in the reactors
Type/Quantity of Media Treated:
Soil
Regulatory Requirements/Cleanup Goals:
• Determine effectiveness and cost of bioslurry systems for degrading explosives in soil
• Evaluate a field-scale system for mechanical integrity and ability to enrich a microbial consortium, and to analyze
system performance over an extended operating period
• A target goal of 20 mg/kg for TNT was used for the demonstration, since a cleanup goal had not yet been established
66
-------
Slurry Reactor Biotreatment of Explosives-Contaminated Soils at Joliet Army
Ammunition Plant, Joliet, Illinois
Results:
• Removed more than 99% of explosives from the soil
• The 20% weekly replacement reactor (soil retention time of 5 weeks), when operated at >25°C, degraded TNT to <50
mg/kg and DNT to < 100 mg/kg, and RDX and TNB to <10 mg/kg; the report does not indicate if this reactor met the
target goal for TNT
• The 10% weekly replacement reactor (soil retention time of 10 weeks), when operated at >25°C, degraded TNT to <20
mg/kg and DNT to <10 mg/kg, and RDX and TNB to <10 mg/kg
• The 5% daily replacement reactor (soil retention time of 5 weeks) had performance similar to that of the 20% weekly
replacement reactor, and removed TNT to <20 mg/kg
• The control reactor (no molasses addition) showed no explosives removed from the soil
Costs:
• Projected costs for full-scale implementation of the slurry-phase biotreatment system was $290-350/yd3
Description:
Joliet Army Ammunition Plant was constructed in Will County, Illinois, approximately 17 miles south of Joliet, in the early
1940's. JAAP contains two major functional areas - a manufacturing area for production of constituent chemicals and
explosive materials, covering 14 square miles, and a load-assemble-package (LAP) area for munitions filling and assembly
lines, storage magazines, and demilitarization, covering 27 square miles. In April 1989, the LAP area was added to the
NPL. Soil for a field demonstration of bioslurry technology was obtained from Group 61, Site LI of the LAP Area, a
ridge-and-furrow area that received pink water from washout of munitions.
The field demonstration showed that bioslurry technology could reduce concentrations of TNT and other explosives in
soil. The important process parameters are the need for an organic co-substrate (molasses), operation of the reactors in an
aerobic-anoxic sequence, and temperature. In warmer temperatures (>25°C), operation of the system at >20% weekly
replacement will achieve removal of explosives. Colder temperatures did not destroy the microbial activity, but did slow
the metabolic rate. In particular, degradation of TNT continued with the accumulation of DNT. The reactors were
operated successfully at lower replacement rates (< 10% weekly) in colder weather. The treated soil (bioslurry) can be
applied directly to land and will not affect plant growth.
67
-------
Joint Small Arms Range Remediation (Physical Separation and Acid Leaching)
at Fort Polk Range 5, Leesville, Louisiana
Site Name:
Fort Polk Range 5
Location:
Leesville, Louisiana
Period of Operation:
August - December, 1996
Cleanup Authority:
RCRA
Purpose/Significance of Application:
Demonstration of physical separation and acid leaching to treat metals in soil
Cleanup Type:
Field demonstration
Contaminants:
Heavy Metals - Lead
• Stockpiled untreated soil had a lead assay of 0.5%
Waste Source:
Small arms testing
Contacts:
Vendor Contacts:
Acetic Acid Leaching:
Thomas Leggiere
ContracCon Northwest Inc
Hydrochloric Acid Leaching:
Craig Jones
Brice Environmental Corporation
Army Contacts:
Richard O'Donell
Lisa Miller
Army Environmental Center
Technology:
Physical separation and acid leaching
• Demonstration included two vendors - one used physical separation and acetic
acid (weak acid) leaching; the other used physical separation and hydrochloric
acid (strong acid) leaching
• Physical separation for both vendors included screening to remove oversize
debris, including bullets and bullet fragments; hydrodynamic separation;
density separation; froth flotation; and magnetic separation
• Following separation, the soil was mixed with the acid in a tank; for the acetic
acid leaching, three tanks were used in series; for the hydrochloric acid
leaching, one mix tank was used
• The treated soil slurry was separated from the leachate and dewatered (filter
press); leachate was regenerated (preciptiation)
• Average processing rate - 2.8 tons/hr (acetic acid) and 6.3 tons/hr
(hydrochloric acid)
Type/Quantity of Media Treated:
Soil
• Acetic acid leaching process - 263 tons
• Hydrochloric acid leaching process - 835 tons
Regulatory Requirements/Cleanup Goals:
• TCLPforleadof5ug/L
• Total metals concentration for lead, copper, zinc, and antimony -1,000 mg/kg each for acetic acid leaching and 500
mg/kg each for hydrochloric acid leaching
Results:
• Soil from physical separation alone was tested for TCLP lead; did not meet cleanup criteria
• Acetic Acid Leaching:
- Initially, approximately 93% total lead, 93% total copper, 77% total zinc, and 70% total antimony removed
- During the demonstration, both total and leachable lead levels in treated soil rose due to buildup of lead in
regenerated leachate as a result of inadequate precipitation
- Total lead was reduced from 2,828 mg/kg in untreated soil to 122-1,443 mg/kg in processed soil; data on TCLP lead
levels was not provided
• Hydrochloric Acid Leaching:
- Met both total and TCLP lead targets throughout demonstration
- Removed 96% total lead, 97% total copper, 89% total zinc, and 60% total antimony
- Total lead was reduced from 4,117 mg/kg in untreated soil to 165 mg/kg in treated soil
- Average TCLP lead level in treated soil was 2 mg/L
68
-------
Joint Small Arms Range Remediation (Physical Separation and Acid Leaching)
at Fort Polk Range 5, Leesville, Louisiana
Costs:
• Costs from the acid leaching demonstration were not analyzed because of the operational difficulties experienced with
leachate regeneration
• Costs for physical separation and hydrochloric acid leaching demonstration were $ 1,400/ton for the 83 5 tons of soil
processed
• Projected full-scale costs for physical separation and hydrochloric acid leaching are $ 170/ton based on 10,000 tons of
soil
Description:
A demonstration of physical separation and acid leaching of soil from a small arms testing range at Fort Polk was
conducted from August to December, 1996. Two types of acid leaching were demonstrated - one using acetic acid to
demonstrate a weak acid and one using hydrochloric acid to demonstrate a strong acid. Soil containing heavy metals was
excavated from the small arms range and stockpiled for the demonstration. The soil was sent through physical separation
followed by acid leaching. The treated soil was separated from the leachate, and dewatered; the leachate was regenerated
and reused in the process.
Results showed that treating soil using physical separation alone did not meet the cleanup goals. While the acetic acid
leaching initially removed metals, operational problems with the regeneration of the leachate resulted in increasing levels
of lead in the treated soil. The hydrochloric acid leaching process met the cleanup goals for all metals throughout the
demonstration. Projected full-scale costs for physical separation and hydrochloric acid leaching are $ 170/ton based on
10,000 tons of soil treated.
69
-------
Thermo NUtech's Segmented Gate System at Los Alamos National Laboratory
Technical Area 33, Los Alamos, New Mexico
Site Name:
Los Alamos National Laboratory Technical Area 33
(Report also addresses testing from Site TA-15)
Location:
Los Alamos, New Mexico
Period of Operation:
April 28^ May 19, 1999
(soil processing on 15 days)
Cleanup Authority:
Voluntary Corrective Action
Purpose/Significance of Application:
Use of a gate system to reduce volume of radioactive-contaminated soil
requiring off-site disposal
Cleanup Type:
Full scale
Contaminants:
Natural Uranium (NU) and Depleted Uranium (DU)
• Concentrations reported as high as 431.46 pCi/g
Waste Source:
Nuclear weapon production operations
Contacts:
Vendor:
Joe Kimbrell
Thermo NUtech
Albuquerque, NM
(505) 254-0935 ext. 209
Management Support:
Ray Patteson, Sandia National
Laboratories, (505) 884-1904
John McCann, Los Alamos National
Laboratory, (505)665-1091
Technology:
Segmented Gate System (SGS)
• SGS is a combination of conveyor systems, radiation detectors (primarily
gamma radiation), and computer control used to segregate waste by
contamination level
• Detectors monitored radioactivity content of soil traveling on belt and
computer opened specified gates to separate portions of soil based on
radioactivity criteria
• Contaminated soil on conveyor belt was diverted by segmented gates into
stockpiles
• Operating parameters included a belt speed of 30 ft/min, belt length of 16 -18
ft, soil layer thickness of 2 in by width of 30.75 in, and soil density of 1.02
g/cm3
• Total soil processing time was 91.1 hrs; average daily operational time was
6.48 hrs
• Oversize debris and rock pre-screened
Type/Quantity of Media Treated:
Soil and Debris
• 2,526 yds3 of soil were processed
• Soil moisture content estimated as 12-15%
Regulatory Requirements/Cleanup Goals:
• Reduce the volume of contaminated soil by separating soil that was above the specified criteria and that would require
off-site storage and disposal, from soil that was below the criteria
• The sorting criterion was 50 pCi/g
Results:
• Overall volume reduction of contaminated soil was 91.64%; approximately 350 yds3 of above-criteria soil required off-
site disposal
• Average activities for soil from Sites C33-003, C33-010, and C33-007b were: above-criteria 318, 431.46, and 165.89
pCi/g, respectively, and below-criteria soil: 3.2, 44.8, and 9.88 pCi/g
Costs:
• Actual cost for SGS was $275,745, including $6,600 for pre-deployment activities, $46,000 for mobilization, $185,445
for processing, $35,000 for demobilization, and $2,700 for final report
• Additional costs incurred by LANL were $543,400, including for staff, prime contractor, G&A, and soil disposal
• Overall unit cost of SGS was $ 109/yd3 of soil processed
70
-------
Thermo NUtech's Segmented Gate System at Los Alamos National Laboratory
Technical Area 33, Los Alamos, New Mexico
Description:
Los Alamos National Laboratory (LANL) is a 43-square mile multi-disciplinary research facility owned by the U.S. DOE
and located in north-central New Mexico. Technical Area 33 (TA-33), located in the eastern portion of LANL, is an active
testing area. TA-33 was used to test initiators (components of nuclear weapons) from 1947 to the 1950's. This report
focused on remediation of uranium-contaminated soil and debris from Potential Release Sites (PRSs) 33-007(b), 33-
010(c), and C33-003 in TA-33. Historical records indicate that natural uranium (NU) and depleted uranium (DU) are
present at these sites.
A Segmented Gate System (SGS) was used to reduce the volume of radioactive-contaminated soil that required off-site
disposal. SGS is a combination of conveyor systems, radiation detectors, and computer control, where contaminated soil
on a conveyor belt is diverted by segmented gates into stockpiles by contamination level. Detectors monitor the
radioactivity content of the soil traveling on the belt and a computer opens specified gates to separate portions of the soil
based on radioactivity criteria. At this site, the overall volume reduction was measured as 91.64%. The actual cost for the
application was $275,745, including $185,445 for processing. This corresponded to an overall unit cost of $109/yd3 based
on 2,526 yd3. During the demonstration, delays were caused by operational failures from hydraulic systems.
71
-------
Thermo NUtech's Segmented Gate System at Pantex Plant, Firing Site 5, Amarillo,
Texas
Site Name:
Pantex Plant, Firing Site 5
Location:
Amarillo, Texas
Period of Operation:
March 27-May 1, 1998
(soil processing from April 17 -April 19, 1999)
Cleanup Authority:
RCRA Corrective Action
Purpose/Significance of Application:
Use of a gate system to reduce volume of radioactive-contaminated soil
requiring off-site disposal
Cleanup Type:
Full scale
Contaminants:
Depleted Uranium (DU)
- Concentrations reported as high as 567 pCi/g
Waste Source:
Firing range for test shots of depleted
uranium and explosives
Contacts:
Site Contact:
Martin Amos, Battelle Pantex
(806) 477-6458
Vendor:
Scott Rogers, Thermo Nutech
(423) 481-0683
Management Support:
Tom Burford, Sandia National Laboratories,
(505) 845-9893
Technology:
Segmented Gate System (SGS)
• SGS is a combination of conveyor systems, radiation detectors
(primarily gamma radiation), and computer control used to
segregate waste by contamination level
• Detectors monitored radioactivity content of soil traveling on belt
and computer opened specified gates to separate portions of soil
based on radioactivity criteria
• Contaminated soil on conveyor belt was diverted by segmented
gates into stockpiles
• Operating parameters included a belt speed of 30 ft/min, belt length
of 16 -18 ft, soil layer thickness of 2 in by width of 30.75 in, and
soil density of 1.0 g/cm3
• Average daily operational time was 2.67 hrs
• Oversize debris and rock pre-screened
Type/Quantity of Media Treated:
Soil and Debris
• 294 yds3 of soil were processed
• Soil moisture content estimated as 17%
Regulatory Requirements/Cleanup Goals:
• Reduce the volume of contaminated soil by separating soil that was above the specified criteria and that would require
off-site storage and disposal, from soil that was below the criteria
• The sorting criterion was 50 pCi/g
Results:
• Overall volume reduction of contaminated soil was 38.5%; approximately 180.8 yds3 of above-criteria soil required off-
site disposal
• Average activities ranged from 125 - 213 pCi/g for above-criteria soil and 20 - 54 pCi/g for below-criteria soil
Costs:
• Actual cost for SGS was $203,887, including $18,768 for regulatory and compliance issues, $103,015 for mobilization,
$32,594 for soil processing, and $49,510 for demobilization
• Additional costs incurred by LANL were for site preparation, excavation, oversight labor, health physics support,
sample analysis, and waste disposal (specific cost data not provided)
• Unit cost of SGS was $ 11 I/yd3 based on 294 yd3 of soil
72
-------
Thermo NUtech's Segmented Gate System at Pantex Plant, Firing Site 5, Amarillo,
Texas
Description:
Firing Site 5 (FS-5) is within the boundaries of the Pantex Plant, located northeast of Amarillo, Texas. The site was used
to conduct test shots of combined explosives and depleted uranium. The firing site was surrounded on three sides by an
earthen berm 10 ft high and 33 ft thick at the base. Soil at the site was contaminated with depleted uranium (DU).
A Segmented Gate System (SGS) was used to reduce the volume of radioactive-contaminated soil that required off-site
disposal. SGS is a combination of conveyor systems, radiation detectors, and computer control, where contaminated soil
on a conveyor belt is diverted by segmented gates into stockpiles based on contamination level. Detectors monitor the
radioactivity content of the soil traveling on the belt and a computer opens specified gates to separate portions of the soil
based on radioactivity criteria. At this site, the overall volume reduction for the first 294 yds3 of soil treated was measured
as 38.5%, and the results were determined to be unsatisfactory; processing was terminated at this time. The actual cost for
the application was $203,887, including $32,594 for soil processing. This corresponded to a unit cost of $11 I/yd3 for soil
processing. Lessons learned included problems with using a hand survey method for classifying soil, which resulted in
misclassifying soil as above the SGS criterion, and the method used each day to cover and uncover the piles.
73
-------
Thermo NUtech's Segmented Gate System at Sandia National Laboratories, ER Site
16, Albuquerque, New Mexico
Site Name:
Sandia National Laboratories, ER Site 16
Location:
Albuquerque, New Mexico
Period of Operation:
February-March 1998
(soil processing from February 27 -
March 5, 1998)
Cleanup Authority:
RCRA Corrective Action
• Part B permit issued 8/93
Regulatory Authority:
New Mexico Environment Department
Purpose/Significance of Application:
Use of a gate system to reduce volume of radioactive-contaminated soil
requiring off-site disposal
Cleanup Type:
Full scale
Contaminants:
Depleted Uranium (DU)
• Concentrations reported as high as 4,100 pCi/g
Waste Source:
Dump Site
Contacts:
Site Contact:
Tom Burford
Sandia Corporation
DOE/AL
(505) 845-9893
Vendor:
Scott Rogers
Thermo NUtech
A ThermoRetec Company
4501 Indian School Road ME, Suite
G105
Albuquerque, NM 87110
(505) 424-3072
Technical Support:
Sue Collins
Sandia National Laboratories
(505) 284-2546
Technology:
Segmented Gate System (SGS)
• SGS is a combination of conveyor systems, radiation detectors (primarily
gamma radiation), and computer control
• Contaminated soil on conveyor belt was diverted by segmented gates into
stockpiles
• Detectors monitored radioactivity content of soil traveling on belt and
computer opened specified gates to separate portions of soil based on
radioactivity criteria
• Operating parameters included a belt speed of 30 ft/min, belt length of 16 -
18 ft, soil layer thickness of 2 in by width of 30.75 in, and soil density of
1.0 g/cm3
• Average daily processing time was 4.7 hrs, less than the target of 7 hrs
• Oversize debris and rock pre-screened using a field grizzly (vertical bar grate)
and hammermill
Type/Quantity of Media Treated:
Soil
• 661.8 yds3 of soil were processed
• Soil identified as silty sands, containing 35-45% silt and clay; moisture
content estimated as 10%
Regulatory Requirements/Cleanup Goals:
• Reduce the volume of contaminated soil by separating soil that was above the specified criteria and that would require
off-site storage and disposal, from soil that was below the criteria
• The sorting criterion was 54 pCi/g
Results:
• Overall volume reduction of contaminated soil was 99.9%; 358 kg of above-criteria soil required off-site disposal
• After first pass, average activity of above-criteria soil was 406.5 pCi/g and below-criteria soil 4.2 pCi/g
Costs:
• Actual cost was $164,109, including $59,326 for mobilization, $57,770 for operations, and $47,013 for demobilization
• Overall unit cost was $236/yd3 of soil processed, reflecting the relatively small amount of soil processed
• Additional activities included site preparation, operation of crane, excavation, oversight labor, health physics support,
water supply, sample analysis, and waste disposal
74
-------
Thermo NUtech's Segmented Gate System at Sandia National Laboratories, ER Site
16, Albuquerque, New Mexico
Description:
Sandia National Laboratories' Environmental Restoration (ER) Site 16 is located northeast of the Technical Area III/V
complex, within Kirtland Air Force Base. The site covers 25 acres and was an open dumping ground for concrete and
other rubble. The concrete and rubble was presumed to be the source on contamination. Approximately 1/3 acre was
excavated for the project from the side and bottom of an arroyo, after the removal of larger debris.
A Segmented Gate System (SGS) was used to reduce the volume of radioactive-contaminated soil that required off-site
disposal. SGS is a combination of conveyor systems, radiation detectors, and computer control, where contaminated soil
on a conveyor belt is diverted by segmented gates into stockpiles. Detectors monitor the radioactivity content of the soil
traveling on the belt and a computer opens specified gates to separate portions of the soil based on radioactivity criteria.
At this site, the overall volume reduction was measured as 99.9%. The actual cost for the application was $164,109,
including $59,326 for mobilization, $57,700 for operations, and $47,013 for demobilization. This corresponded to an
overall unit cost of $236/yd3. Lessons learned included impacts from startup requirements, jams in the screen/hammermill
caused by larger rocks, and soil buildup in the gas chutes.
75
-------
Thermo NUtech's Segmented Gate System at Sandia National Laboratories, ER Site
228A, Albuquerque, New Mexico
Site Name:
Sandia National Laboratories, ER Site 228A
Location:
Albuquerque, New Mexico
Period of Operation:
July-November 1998
(soil processing from November 6-17, 1998)
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Use of a gate system to reduce volume of radioactive-contaminated soil
requiring off-site disposal
Cleanup Type:
Full scale
Contaminants:
Depleted Uranium (DU)
• Concentration not provided
Waste Source:
Burial pits
Contacts:
Site Contact:
Sandia Corporation
DOE/AL
Vendor:
Thermo NUtech
A ThermoRetec Company
4501 Indian School Road ME, Suite G105
Albuquerque, NM 87110
Technical Support:
Sue Collins
Sandia National Laboratories
(505) 284-2546
Technology:
Segmented Gate System (SGS)
• SGS is a combination of conveyor systems, radiation detectors
(primarily gamma radiation), and computer control
• Contaminated soil on conveyor belt was diverted by segmented gates
into stockpiles
• Detectors monitored radioactivity content of soil traveling on belt and
computer opened specified gates to separate portions of soil based on
radioactivity criteria
• Operating parameters included a belt speed of 30 ft/min, belt length of
16-18 ft, soil layer thickness of 2 in by width of 30.75 in, and soil
density of 1.29 g/cm3
• Average daily processing time was 4.47 hrs, less than the target of
7hrs
• Oversize debris and rock pre-screened using a field grizzly (vertical
bar grate) and hammermill
Type/Quantity of Media Treated:
Soil
• 1,352 yds3 of soil were processed
• Extended 0.4 acres at a depth of 2 ft
• Soil identified as sandy, moisture content estimated as 10%
Regulatory Requirements/Cleanup Goals:
• Reduce the volume of contaminated soil by separating soil that was above the specified criteria and that would require
off-site storage and disposal, from soil that was below the criteria
• The sorting criterion was 27 pCi/g
Results:
• Overall volume reduction was measured as 99.56%; 21 55-gallons drums of above-criteria soil required off-site disposal
• Average activity of above-criteria soil was 223 pCi/g and below-criteria soil 14.77 pCi/g
• 5.2 yds3 of soil from Burn Site showed volume reduction of 99.4%
Costs:
• Actual cost was $220,040, including $29,000 for excavation and pre-screening, $41,300 for mobilization, $117,000 for
operations, and $32,340 for demobilization
• Overall unit cost was $154/yd3, reflecting the relatively small quantity of soil processed
• Project contracted as a lump sum fixed price; did not include excavation, oversight labor, health physics support, water
supply, fuel services, generator support, sample analysis, and waste disposal
76
-------
Thermo NUtech's Segmented Gate System at Sandia National Laboratories, ER Site
228A, Albuquerque, New Mexico
Description:
Environmental Restoration (ER) Site 228A, the Centrifuge Dump Area and Tijeras Arroyo Operative Unit-ADS 1309, is
located 500 ft east of Technical Area II, within Kirtland Air Force Base. In July 1997, heavy rains eroded a portion of a
depleted uranium burial from the Tijeras Arroyo rim. Depleted uranium mixed with soil and debris washed down the
slope.
A Segmented Gate System (SGS) was used to reduce the volume of radioactive-contaminated soil that required off-site
disposal. SGS is a combination of conveyor systems, radiation detectors, and computer control, where contaminated soil
on a conveyor belt is diverted by segmented gates into stockpiles. Detectors monitor the radioactivity content of the soil
traveling on the belt and a computer opens specified gates to separate portions of the soil based on radioactivity criteria.
At this site, the overall volume reduction was measured as 99.56%. The actual cost for the application was $220,040,
including $29,000 for excavation and pre-screening, $41,300 for mobilization, $117,000 for operations, and $32,340 for
demobilization. This corresponded to an overall unit cost of $154/yd3. Lessons learned included impacts from weather
delays and equipment concerns, and difficulties with rocks that were 3 inches in diameter.
77
-------
Thermo NUtech's Segmented Gate System at Tonapah Test Range, Clean Slate 2,
Tonapah, Nevada
Site Name:
Tonapah Test Range, Clean Slate 2
Location:
Tonapah, Nevada
Period of Operation:
May 4^ June 12, 1998
(soil processing from May 18 - June 3, 1999)
Cleanup Authority:
RCRA Corrective Action
Purpose/Significance of Application:
Use of a gate system to reduce volume of radioactive-contaminated soil
requiring off-site disposal
Cleanup Type:
Field demonstration
Contaminants:
Plutonium
- Concentrations reported as high as 1,100 pCi/g
Waste Source:
Weapons test range
Contacts:
Vendor:
Scott Rogers, Thermo Nutech, (423) 481-0683
Management Support:
Tom Burford, Sandia National Laboratories,
(505) 845-9893
Technical Contact:
Mike Hightower, Sandia National Laboratories,
(505) 844-5499
Technology:
Segmented Gate System (SGS)
• SGS is a combination of conveyor systems, radiation detectors
(primarily gamma radiation), and computer control used to segregate
waste by contamination levels
• Detectors monitored radioactivity content of soil traveling on belt
and computer opened specified gates to separate portions of soil
based on radioactivity criteria
• Contaminated soil on conveyor belt was diverted by segmented gates
into stockpiles, based on the criteria
• Operating parameters included a belt speed of 30 ft/min, belt length
of 16 -18 ft, soil layer thickness of 1 - 2 in by width of 30.75 in, and
soil density of 1.0 g/cm3
• Oversize debris and rock were pre-screened
Type/Quantity of Media Treated:
Soil and Debris
• 333 yds3 of soil were processed
• Soil was primarily sand and silt with some gravel and cobbles; soil
type and moisture content optimal for SGS operation
Regulatory Requirements/Cleanup Goals:
• Reduce the volume of contaminated soil by separating soil that was above the specified criteria and that would require
off-site storage and disposal, from soil that was below the criteria
• The sorting criterion was 50 -1,500 pCi/g; demonstration results were to be used to define optimum operating
parameters
Results:
• 79 runs were conducted, each characterized by different soil activity levels, operating parameters, and end points
(sorting criterion)
• Results showed that optimum separation criteria for soils with <400 pCi/g was about 300 pCi/g, resulting in a volume
reduction of 60% and an average clean soil activity of 160 pCi/g
• Soils between 400 - 800 pCi/g did not appear to have an optimum separation criterion, and had a volume reduction of
30 - 40% and an average clean soil activity of 250 pCi/g
• Soils >800 pCi/g did not appear to have an optimum separation criterion, and had a volume reduction of 30% and an
average clean soil activity of 500 pCi/g; this clean soil activity was too high and suggested that processing soil with
>800 pCi/g would probably not be appropriate
Costs:
• Actual cost for SGS was $138,126, including $8,203 for regulatory and compliance issues, $29,614 for mobilization,
$78,545 for physical treatment, and $21,764 for demobilization
78
-------
Thermo NUtech's Segmented Gate System at Tonapah Test Range, Clean Slate 2,
Tonapah, Nevada
Description:
Tonapah Test Range is a DOE and DoD weapons testing range. The Clean Slate-2 soil remediation site of the range is in
the northwest portion of Nellis Air Force Base. In 1963, a series of four nuclear weapons, component, and explosive
vulnerability destruction experiments, known as Operation Roller Coaster, were conducted at the range. These
experiments left varying levels of finely dispersed plutonium at the site. Approximately 32,000 yds3 of soil in Clean Site-2
are contaminated, with the site still being characterized.
A Segmented Gate System (SGS) was used to reduce the volume of radioactive-contaminated soil that required off-site
disposal. SGS is a combination of conveyor systems, radiation detectors, and computer control, where contaminated soil
on a conveyor belt is diverted by segmented gates into stockpiles based on contamination levels. Detectors monitor the
radioactivity content of the soil traveling on the belt and a computer opens specified gates to separate portions of the soil
based on radioactivity criteria. At this site, 79 periods of operation (runs) were conducted, each characterized by different
soil activity levels, operating parameters, and end points (sorting criterion) ranging from 50 to 1,500 pCi/g. Results
showed that optimum separation criteria for soils with <400 pCi/g was about 300 pCi/g, resulting in a volume reduction of
60% and an average clean soil activity of 160 pCi/g. Soils >400 pCi/g did not appear to have an optimum separation
criterion. Results suggested that processing soil with >800 pCi/g would probably not be appropriate for the SGS. Actual
cost for SGS was $138,126, including $78,545 for soil processing. Results from these tests were used to develop potential
treatment scenarios for the SGS at Clean Slate-2. Lessons learned covered topics such as the need for accurate site
characterization data and the benefits of selective excavation of hot spots.
79
-------
Chemical Extraction for Uranium Contaminated Soil at the RMI Titanium
Company Extrusion Plant, Ashtabula, Ohio
Site Name:
RMI Titanium Company Extrusion Plant
Location:
Ashtabula, Ohio
Period of Operation:
January 7, 1997 - February 14, 1997
Cleanup Authority:
NRC
Purpose/Significance of Application:
Demonstration of chemical leaching process for treatment of uranium-
contaminated soil
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides - Uranium
• Most uranium present as U+236
• Uranium levels in feed soil were 74-146 pCi/g
Waste Source:
Particulates from uranium extrusion
operations
Contacts:
DOE Contacts:
Ward Best, DOE Ashtabula Area
Office, (216) 993-1944
Jeff Kulpa, RMI Environmental
Services, (216) 993-2804
Erik Groenendijk, ART pilot project
manager, (813) 264-3529
Mike Hightower
Sandia National Laboratories
Telephone: (505) 844-5499
Fax:(505)844-0116
E-mail: mmhight@sandia.gov
EPA Contact:
Brian Nickel
Ohio EPA
401 East Fifth Street
Dayton, OH 45402-2911
Telephone: (513) 285-6357
Fax:(513)285-6249
Technology:
Chemical Extraction
• Process involves application of heated bicarbonate solution to soil in a rotary
reactor, liquid/soils separation, dewatering, and ion exchange to remove
uranium from liquid
• Solution was 0.2 M NaHCO3 at a 115°F and retention time of 1.5 hrs; reactor
was a 5 yd3 cement mixer
• Processed 1 to 2 tons of soil/batch, using a 30% solids slurry
Type/Quantity of Media Treated:
Soil
• 64 tons (3 8 batches)
• high clay content silt loams and clay loams; low organic material
Regulatory Requirements/Cleanup Goals:
• Evaluate process performance, such as ability to meet a 30 pCi/g free release standard and achieve a significant volume
reduction of the waste
Results:
• Treated soil from two areas of the plant had 12-14 pCi/g of uranium, with removal efficiencies of 87-91%
• Treated soil from another area of the plant had 27-47 pCi/g; the higher concentrations was attributed to high feed
concentrations from a hot spot with 587 pCi/g
• Volume reduction was 95%; less than 5% residual waste required off-site disposal
• Average feed concentration to ion exchange was 16 ppm and output 1.7 ppm, resulting in a 91% removal efficiency
Costs:
• The total cost for the pilot plant was $638,670, including mobilization and preparatory work; monitoring, sampling,
testing, and analysis; chemical treatment; decontamination and decommissioning; disposal commercial; demobilization;
and data compilation and report writing
• The report authors indicate that a linear relationship does not exist between pilot plant and full-scale costs
• Full-scale costs were estimated to range from $250-350 per ton of soil treated
80
-------
Chemical Extraction for Uranium Contaminated Soil at the RMI Titanium
Company Extrusion Plant, Ashtabula, Ohio
Description:
From 1962 to 1988, the RMI Titanium Company (RMI) performed uranium extrusions operations for the U.S. DOE at its
plant in Ashtabula, Ohio. The uranium metal processed at the site included deleted and slightly enriched material that was
used in nuclear and non-nuclear weapons. During the early years of extrusion and machining, paniculate uranium was
generated and discharged from roof vents and stacks and settled on surrounding soils. A test of a carbonate extraction
process was conducted to leach uranium from contaminated soils.
Thirty-eight batches of 1-2 tor^atch were treated in a pilot-scale test of a chemical extraction process, through DOE's
ITRD program. Treated soil had an overall removal efficiency of approximately 82% with a volume reduction of 95%;
less than 5% of residual waste required off-site disposal. Difficulties with meeting the cleanup goal were identified only
when treating soil from a hot spot. The total cost for the pilot plant was $638,670, and full-scale costs were estimated as
$250-3 50/ton.
81
-------
Transportable Vitrification System at Oak Ridge National Laboratory,
Oak Ridge, Tennessee
Site Name:
Oak Ridge National Laboratory (ORNL)
Location:
Oak Ridge, TN
Period of Operation:
October 1997
Cleanup Authority:
RCRA and NRC
Purpose/Significance of Application:
Demonstration of a transportable vitrification system to treat low-level mixed
waste sludges
Cleanup Type:
Field demonstration
Contaminants:
Metals and Radionuclides
Waste Source:
Mixed low-level waste sludges from
DOE operations - included pond
sludge and sludge from a
neutralization facility
Contacts:
Principal Investigator:
Frank Van Ryn
Bechtel Jacobs Company
ORNL
P.O. Box 2003
Oak Ridge, TN 37831
Telephone: 423-574-1907
Fax: 423-574-9786
E-mail: xs2@ornl.gov
DOE Technical Program Manager:
Dave Hutchins
Environmental Technology Group,
EM-93
U.S. DOE, Oak Ridge Operations
Office
P.O. Box 2001
Oak Ridge, TN 37831
Telephone: 423-241-6420
Fax: 423-576-5333
E-mail: hutchinsda@oro.doe.gov
Technology:
Vitrification
Transportable Vitrification System (TVS):
• Waste and Additives and Materials Processing Module - 240-gal melter feed
blend tank equipped with a load cell and agitator, centrifugal pump, feed tank,
melter module, and emission control module
• Melter Module - joule-heated glass melter equipped with molybdenum rod
electrodes and lined with heavy flux contact refractory
• Melter capacity - up to 300 Ib/hr; operating temperature -1,150 to 1,400°C;
heated with a 500,000-BTU/hr propane burner
• Melter equipped with a drain bay chamber to remove waste glass and salt tap
side chamber to remove corrosive salts
• Waste glass poured from drain bay chamber into 8-cubic foot stainless steel
containers
• Emission Control Module included quench tower, packed bed cooler, variable
throat venturi, mist eliminator, reheater, and high-efficiency paniculate air
filters
• Control and Services Module - used to control and monitor equipment
operation
Type/Quantity of Media Treated:
Sludge
• Pond sludge and mixtures of pond and neutralization sludge -16,000 Ibs
Regulatory Requirements/Cleanup Goals:
• RCRA Land Disposal Restriction (LDR) standards and NRC guidelines
• Air emissions limits were specified in a State of Tennessee air permit
82
-------
In situ Bio remediation Using Molasses Injection at an Abandoned Manufacturing
Facility, Emeryville, California
Results:
• The waste form produced by the TVS met the RCRA LDR standards and NRC guidelines, was stable and durable, and
represented a 60% volume reduction of the waste
• The TVS system operated within the required emissions limits
• The melting rate decreased during the demonstration, resulting in lower average throughput rate (450 kg/day versus
expected 900 kg/day); attributed to high iron content of waste which decreased heat transfer characteristics of glass
material
Costs:
• Projected costs for a full-scale system include:
- Capital costs, including all equipment - $5 million
- Operating costs - $10 to $44/kg of waste, assuming analytical expenses similar to those incurred for the
demonstration; assuming less extensive analytical requirements for normal operations, operating costs were estimated
at$7to$17/kgofwaste
Description:
In October 1997, following completion of process development and testing, demonstration of the TVS was conducted at
ORNL, using actual low-level mixed waste containing metals. The waste used for the demonstration was B&C pond
sludge and a mix of B&C pond sludge and sludge from a neutralization facility. The objectives of the demonstration
included meeting the RCRA LDR standards and NRC guidelines for the glass waste form, meeting the air emissions limits
for the operation of the TVS, and collecting operating and performance data for the process for use in scale-up.
The results of the demonstration showed that the TVS was capable of treating low-level mixed waste sludges to the RCRA
LDR levels and the NRC guidelines, and of operating within the required air emission standards. The waste form
produced by the TVS was highly durable with long-term integrity, and significant reductions in waste volumes were
achieved. For different waste compositions from those tested, additional process development would be required to
determine the process controls and scale-up methods needed to achieve optimal glass waste forms, consistent melter
operation, and to avoid adverse melter conditions. Treatability studies are recommended for any waste stream to be treated
using TVS.
83
-------
This Page Intentionally Left Blank
84
-------
PUMP AND TREAT ABSTRACTS
85
-------
Groundwater Extraction and Treatment at the Logistics Center Operable Unit,
Fort Lewis, Washington
Site Name:
Fort Lewis Logistics Center Operable Unit
Location:
Fort Lewis, Washington
Period of Operation:
August 1995 - ongoing
Cleanup Authority:
CERCLA Remedial Action
Record of Decision (ROD) signed on
September 25, 1990
Purpose/Significance of Application:
Use of two groundwater extraction systems to remove VOCs and treat using air
stripping.
Cleanup Type:
Full scale
Contaminants:
Organic Compounds, Halogenated (Chlorinated Solvents) - TCE and DCE
Maximum TCE concentration in groundwater is greater than 100,000 mg/L.
Waste Source:
Disposal of waste solvents in surface
trenches, including disposal of free
liquids and disposal of drums
containing liquids
Contacts:
Project Management:
Project Manager
Mr. Bill Goss
USACE, Seattle District
4735 Marginal Way, South
O&M Contractor:
URS Greiner Woodward-Clyde
2401 Fourth Avenue, Suite 1000
Seattle, Washington 98121
LTM - Kelly Teague
(206)674-7931
O&M-SteffranNeff
(206) 343-7933
Regulatory Contact:
Mr. Bob Kievit
U.S. EPA Region X
Washington Operations Office
300 Desmond Drive, Suite 102
Lacey, Washington 98503
(360) 753-9014
Technology:
• Groundwater is extracted via two well fields located at the suspected main
contaminant source area (The East Gate system), and from a line of wells
located down gradient of the source areas (The 1-5 system)
• Extracted groundwater is treated by air stripping
• Treated groundwater is recharged to the subsurface via wells and infiltration
galleries near each extraction area
Type/Quantity of Media Treated:
• 2.147 Million gallons of water extracted, treated and recharged as of 8/98
• 2772 pounds of TCE removed as of 9/97
Regulatory Requirements/Cleanup Goals:
• Groundwater extracted at the Logistics Center Site is required to be treated to drinking water standards (MCLs) prior to
recharge to the subsurface for the contaminants of concern: TCE - 5 mg/L; DCE - 70 mg/L
• Air emissions from the treatment systems are required to be below 75 pounds per month (1-5) and 325 pounds per
month (East Gate), respectively
Results:
• Effluent sampling at each air stripper indicates that TCE concentrations in the treated groundwater are consistently
below the treatment requirement of 5 mg/L. Several samples collected during the first few months of operation for the
East Gate system contained TCE in concentrations exceeding 5 mg/L, however, operations have since been modified to
improve performance of this system. No results above MCLs have been observed since October 1995. TCE removal
efficiencies for the air strippers have ranged from 96 percent to greater than 99 percent since start up in 1995
• Air emissions have been below allowable limits for both treatment systems since since start up in 1995
86
-------
Groundwater Extraction and Treatment at the Logistics Center Operable Unit,
Fort Lewis, Washington
Costs:
The total cost incurred for design, construction and the first year of O&M for the two extraction and treatment systems
was $5,208,000. The design cost was $1,251,000, and the construction cost was $3,528,000
Description:
The Logistics Center site at Fort Lewis covers 650 acres and is currently an active facility. The site was previously
operated as an ordnance depot from 1942 to 1963 and has been operated as a non-aircraft maintenance facility since 1963.
Groundwater at the Logistics Center has been contaminated with chlorinated organic compounds as the likely result of
past disposal activities that included disposal of waste solvents in trenches excavated at the site. The principle
contaminants of concern at the site are TCE and DCE. In 1990, a ROD was signed for the Logistics Center Operable Unit
specifying that the contaminant plume be monitored and reduced over time, and that migration of groundwater
contamination from the site be minimized.
In response to the ROD, it was determined that two extraction and treatment systems would be installed at the site. One
system (the East Gate system) was designed to reduce the contaminant plume in the source area, and the other system (1-5)
was designed to minimize off-site migration of contaminants. Both systems include treatment of contaminated
groundwater using air stripping, followed by recharge of treated water to the subsurface. Recharge is accomplished using
infiltration galleries located at each site and also by injection wells located at the East Gate site. The treatment systems
have been in operation since 1995, and it is anticipated that treatment will continue for 30 years. The treatment systems
each consistently meet federal and local requirements for treatment of groundwater prior to recharge and for allowable air
emissions.
87
-------
This Page Intentionally Left Blank
88
-------
IN SITU GROUNDWATER TREATMENT ABSTRACTS
89
-------
In Situ Bio remediation Using Molasses Injection at an Abandoned Manufacturing
Facility, Emeryville, California
Site Name:
Abandoned Manufacturing Facility
Period of Operation:
Pilot study - August 1995 to February 1996
Full scale system - ongoing, data available from April 1997 to October 1998
Purpose/Significance of Application:
Bioremediation of a site contaminated with both chlorinated solvents and
hexavalent chromium
Contaminants:
TCE, hexavalent chromium
• Concentrations of TCE reported as high as 12,000 ug/L
Contacts:
Remediation Contractor:
Daniel L. Jacobs
ARCADIS Geraghty & Miller, Inc.
3000 Cabot Boulevard West, Suite
3004
Langhorne, PA 19047
Telephone: (215) 752-6840
Fax: (215) 752-6879
e-mail: Djacobs@gmgw.com
Location:
Emeryville, California
Cleanup Authority:
State voluntary cleanup program
Cleanup Type:
Pilot and Full scale
Waste Source:
Electroplating operations
Technology:
In situ bioremediation
• A pilot study was performed using a mixture of molasses, biologically
inoculated solution (supernatant), and tap water was injected into the
subsurface
• The full-scale system used 91 temporary injection points, installed to 24 ft
bgs with a Geoprobe
• Molasses injection events were performed in April 1997 and February 1998,
which involved a mixture of water, molasses, and a small amount of
supernatant
• During the first injection event, each injection point received 25 gallons of
molasses, 1 gallon of supernatant, and 125 gallons of water
Type/Quantity of Media Treated:
Groundwater
• Site geology consists of interbedded sand and clay units
• Depth to groundwater is approximately 3.5 to 8 ft
• Groundwater velocity is estimated at approximately 60 ft per yr
Regulatory Requirements/Cleanup Goals:
• The pilot study was performed to determine if TCE degradation and metal precipitation could be enhanced by an in situ
reactive zone
• Cleanup goals for the full-scale application were not identified
Results:
• The average TCE concentration in on-site wells has decreased by 99% (3,040 \igfL in April 1995 to 4 ug/L in October
1998) during bioremediation
• The trends for TCE degradation products (cis-l,2-DCE and VC) indicate that TCE has been reductively dechlorinated
to ethene under the engineered anaerobic conditions; initial cis-l,2-DCE and VC concentrations increased following
the first reagent injection, but declined as shown in the October 1998 groundwater monitoring results
• The average concentrations of total chromium and hexavalent chromium in the injection area have been reduced by
approximately 98% and 99%, respectively
Costs:
• The overall project cost was approximately $400,000
90
-------
In Situ Bio remediation Using Molasses Injection at an Abandoned Manufacturing
Facility, Emeryville, California
Description:
Metal plating operations were conducted at a manufacturing facility located in Emeryville, California (actual site name
confidential) from 1952 until 1989. Investigations conducted at the site found groundwater to be contaminated with
chlorinated solvents, primarily TCE, and hexavalent chromium. From August 1995 to February 1996, the site owner
conducted a pilot study of anaerobic reductive dechlorination to evaluate its potential as a groundwater remedy under a
state voluntary cleanup program. Based on the results of the pilot test, a full-scale system was installed and is operating at
the site.
The injection of molasses reagent solution created conditions favorable for the reduction in TCE, DCE, VC, and
chromium concentrations in the subsurface. During an 18-month period of full-scale operation, average concentrations of
TCE were reduced by 99%, from more than 3,000 ug/L to 4 ug/L, and average concentrations of Cr+6 also were reduced
by 99%. The pilot study showed that the rate of reductive dechlorination could be enhanced with the use of an injected
molasses solution.
91
-------
In Situ Bio remediation Using Molasses Injection at the Avco Ly coming
Superfund Site, Williamsport, Pennsylvania
Site Name:
Avco Lycoming Superfund Site
Location:
Williamsport, Pennsylvania
Period of Operation:
Pilot study October 1995 to March 1996;
Full-scale system ongoing, data available through July 1998
Cleanup Authority:
CERCLA
• ROD signed December 1996
Purpose/Significance of Application:
One of the first applications of molasses injection technology on a full scale at a
Superfund site
Cleanup Type:
Pilot and Full scale
Contaminants:
Chlorinated solvents and heavy metals - TCE, DCE, VC, hexavalent chromium,
cadmium
• Maximum concentrations measured in late 1996 were TCE - 700 ug/L,
hexavalent chromium - 3,000 ug/L, and cadmium - 800 ug/L
Waste Source:
Spills and leaks from plating
operations; disposal in lagoons and
wells
Contacts:
Remediation Contractor:
Daniel L. Jacobs
ARCADIS Geraghty & Miller, Inc.
3000 Cabot Boulevard, West, Suite
3004
Langhorne, PA 19047
Telephone: (215) 752-6840
Fax: (215) 752-6879
E-mail: djacobs@gmgw.com
EPA Remedial Project Manager:
Eugene Dennis
U.S. EPA Region 3
1650 Arch Street (3HS21)
Philadelphia, PA 19103-2029
(215)814-3202
E-mail: dennis.eugene@epa.gov
Technology:
In Situ Bioremediation; Anaerobic Reductive Dechlorination
• Pilot studies consisted of molasses injection and air sparging/soil vapor
extraction
• Full scale molasses injection system consists of 20 four-inch diameter
injection wells, ranging in depth from 19 to 30 ft, completed in the
overburden
• Molasses is added two times each day at variable concentrations and rates
• Eight additional wells are used for monitoring system performance
• This is a proprietary technology owned by ARCADIS Geraghty & Miller.
Type/Quantity of Media Treated:
Groundwater
• Site geology consists of a sandy silt overburden overlying a fractured bedrock
and a fractured limestone
• Target area for treatment is the shallow overburden to approximately 25 ft
bgs, covering approximately 2 acres
Regulatory Requirements/Cleanup Goals:
• The 1996 ROD specified the following cleanup goals for groundwater: TCE - 5 ug/L; 1,2-DCE - 70 ug/L; VC - 2 ug/L;
Cd - 3 ug/L; Cr+6 - 32 ug/L; Mn - 50 ug/L
Results:
• The pilot study showed that the technology was able to create strongly reducing conditions
• The baseline sampling event showed that anaerobic, reducing conditions were present only near two of the site
monitoring wells
• Since the injection of reagent, the redox levels have decreased to anaerobic conditions in many of the wells that had
previously indicated an aerobic environment, and cleanup goals have been met in some of the wells
• Analytical results for TCE, DCE, and VC for an area that was converted from aerobic to anaerobic show that TCE was
reduced from 67 to 6.7 ug/L, a 90% reduction. The concentration of DCE initially increased, indicating the successful
dechlorination of TCE, and then decreased to 19 ug/L
• Concentrations of TCE, DCE, and Cr+6 have been reduced to less than their cleanup goals in many of the monitoring
wells at the site
92
-------
In Situ Bio remediation Using Molasses Injection at the Avco Ly coming
Superfund Site, Williamsport, Pennsylvania
Costs:
• ARCADIS Geraghty & Miller reported a total project value of $145,000 for the pilot study application at this site,
including preparation of a work plan
• The costs for the construction of the full scale molasses injection system was approximately $220,000. Operation and
maintenance, including monitoring, is approximately $50,000 per year
Description:
The Avco Lycoming Superfund site (Lycoming) is a 28-acre facility located in Williamsport, Pennsylvania. Since 1929,
various manufacturing companies have operated at the site. Past waste handling practices have contaminated the site,
including disposal of waste in wells and lagoons, and spillage and dumping of wastes from metal plating operations. In
1984, the state identified volatile organic compound (VOC) contamination in the local municipal water authority well
field located 3,000 ft south of the site. A pump and treat system was installed in the mid 1980's. In May 1995, the PRP
proposed the use of in situ bioremediation to replace the pump and treat remedy. Pilot studies of molasses injection and
air sparging/soil vapor extraction (SVE) were conducted from October 1995 to June 1996. A new ROD, issued in
December 1996, replaced the pump and treat remedy with in situ bioremediation, and a full-scale system has been
operating at the site since January 1997. Construction of the air sparging/SVE system was suspended in the Spring of
1998, due to higher than anticipated water levels.
The use of molasses injection was shown to create an anaerobic reactive zone in an 18-month period where concentrations
of TCE, DCE, and hexavalent chromium were reduced. According to the PRP contractor, this technology was shown to
save substantial resources when compared to pump and treat.
93
-------
In Situ Bioremediation Using Bioaugmentation at
Area 6 of the Dover Air Force Base, Dover Delaware
Site Name:
Dover Air Force Base, Area 6
Location:
Dover, Delaware
Period of Operation:
Proof of Technology Test: September 1996 to March 1998
Testing for Technology Scale-up: April 1998 to June 1999 (planned)
Full-scale System: Summer 1999 (planned)
Cleanup Authority:
CERCLA
Purpose/Significance of Application:
The first successful bioaugmentation project using live bacteria from another
site to treat TCE using reductive dechlorination
Cleanup Type:
Field demonstration (pilot proof of
technology test)
Contaminants:
Chlorinated solvents
• Concentrations in the pilot area before the test were PCE - 46 ug/L, TCE -
7,500 ug/L, cis-DCE - 2,000 ug/L, and vinyl chloride - 34 ug/L
Waste Source:
Waste disposal
Contacts:
RTDF Contact:
Dr. David Ellis
DuPont Engineering
Barley Mill Plaza 27-2234
P.O. Box 80027
Wilmington, DE 19880-0027
(302) 892-7445
email: david.e.ellis@usa.dupont.com
ITRC Contact:
Paul Hadley
ITRC In Situ Bioremediation
Technical Task Team Leader
California Environmental Protection
Agency
Department of Toxic Substances
Control
PO Box 806
Sacramento, CA 95814
(916) 324-3823
EPA Remedial Project Manager:
R. Drew Lausch
U.S. EPA Region 3
1650 Arch Street
Philadelphia, PA 191103
(215) 814-3359
email: lausch.robert@epa.gov
Technology:
In Situ Bioremediation
• Groundwater flow and three-dimensional transport models (MODFLOW and
MT3D) were used in designing the pilot system
• The pilot system included three extraction or pumping wells and three
injection wells, each screened to a depth of 38 to 48 ft bgs, and designed to
operate as an isolated or "closed-loop" recirculation cell
• The pumping wells were operated at a combined rate of 3.75 gpm (1.25 gpm
each), providing a residence time of about 60 days for groundwater from the
deep zone of the aquifer
• The extracted groundwater was filtered, and substrate (sodium lactate) and
nutrients (ammonia and phosphate) were injected into the combined
groundwater stream downstream of the filter
• On June 5 and 20, 1997, an aqueous culture (from the DOE's Pinellas site in
Largo, Florida; augmenting solution) was injected into the cell
Type/Quantity of Media Treated:
Groundwater
• The saturated portion of the formation consists of various sands and is about
3 8 feet thick
• The aquifer acts as one unconfined unit that includes three zones
(approximately equal thickness) - an upper zone of fine sand (0 to 12 ft bgs),
an intermediate zone of medium sand (12 to 25 ft bgs), and a deep zone also
of medium sand (25 to 48 ft bgs)
• Groundwater is found in the intermediate and deep zones, starting at 10 to 12
ft bgs.
• Hydraulic conductivity was 60 ft/day and groundwater velocity 140 ft/yr
Regulatory Requirements/Cleanup Goals:
Pilot test goals: 1) demonstrate that TCE and PCE degradation can be stimulated in the deep portion of an aquifer; 2)
confirm that degradation will proceed to nontoxic end products; 3) develop operation and cost data for a full-scale system;
and 4) document the methodology used in the pilot system
94
-------
In Situ Bioremediation Using Bioaugmentation at
Area 6 of the Dover Air Force Base, Dover Delaware
Results:
• During the first five months of operation, the concentration of TCE gradually decreased, cis-DCE showed a slight
increase, and there was no increase for vinyl chloride or ethene, indicating that limited dechlorination was occurring
• For the first 90-days following bioaugmentation, TCE concentrations continued to decrease and DCE concentrations
continued to increase; however, there was no evidence of vinyl chloride or ethene in the groundwater
• By March 1998, all TCE and DCE in the groundwater were converted to ethene and between 75 and 80% of the TCE
and DCE had been recovered as ethene, indicating that the bioaugmentation was successful in destroying TCE by
reductive dechlorination
• From April 1998 through June 1999, the test was focusing on testing of parameters involved with technology scale up
Costs:
• Total capital costs were $285,563
• Total operating costs were $164,962 for the first three months of operation (through November 30, 1996) and $522,620
for the first fifteen months of operation (through November 30, 1997)
• According to the RTDF contact, a typical full-scale bioaugmentation system would cost substantially less than the
system used in the pilot test at Dover
Description:
Dover Air Force Base (AFB), located in Dover, Delaware, is a 4,000 acre military installation that began operating in
1941. An estimated 23,000 cubic feet of waste, including solvents, waste fuels and oils, and a variety of other wastes,
were disposed at the site from 1951 to 1970. Soil and groundwater at the base were found to be contaminated with
volatile organic compounds, including TCE and PCE, and with heavy metals, including arsenic and cadmium. In March
1989, the site was listed on the National Priorities List. During a remedial investigation, "Area 6" was one of the areas at
the base that was determined to have been contaminated with chlorinated solvents; a plume of VOCs was identified in
groundwater in this area. Based on the results of that investigation as well as additional sampling, the area was selected
for pilot testing of a bioaugmentation process. The remediation of Dover AFB is managed by EPA Region 3 and the
Delaware Department of Natural Resources and Environmental Control. Interim RODs were signed in September 1995
that identify the following technologies for remediation at Dover: anaerobic reductive dehalogenation, cometabolic
bioventing, and monitored natural attenuation. The pilot test was performed as part of the Bioremediation Consortium of
the Remediation Technology Development Forum.
Data from the pilot test indicated that an extended period of time was required for the bacteria to exhibit functional
dechlorination. At the start of bioaugmentation, lag periods of about 180 days between bioaugmentation and complete
reduction of TCE and DCE to ethene were observed, including a 90-day lag period before vinyl chloride was first
observed. Injection well plugging was a problem during the pilot test. Several methods were used to keep the wells
unplugged including cleaning the well screens with wire brushes and pumping out residue from the screened interval,
using hydrogen peroxide to clean the wells, and changing substrates from sodium lactate to lactic acid. Hydrogen
peroxide proved the most effective technique for keeping the wells from clogging.
95
-------
Aerobic Degradation at Site 19,
Edwards Air Force Base, California
Site Name:
Edwards Air Force Base
Location:
California
Period of Operation:
February 5, 1996 to April 1, 1997
Cleanup Authority:
CERLCA
Purpose/Significance of Application:
Field demonstration of in situ bioremediation using groundwater recirculation
wells to remediate TCE in a two aquifer system
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated Solvents
• Primary contaminant in groundwater- trichloroethene (TCE)
• Levels as high as 1,150 ug/L found in the groundwater; average TCE
concentration in the upper and lower aquifer of 680 and 750 ug/L, respectively
• No 1,1-DCE found at the site prior to the demonstration
Waste Source:
Equipment cleaning and solvent
degreasing operations
Contacts:
EPA RPM:
Richard Russell
U.S. EPA Region 9
75 Hawthorne Street, SFD-8-1
San Francisco, CA 94105
(415)744-2406
e-mail: russell.richard@epa.gov
Air Force Project Manager:
David Steckel
AFFTC/EMR
5 East Popson Avenue, Building
2650A
Edwards Air Force Base, CA 93524-
1130
(805) 277-1474
fax:(805)277-6145
e-mail: david.steckel@edwards.af.mil
Principal Investigator:
Dr. Perry McCarty
Stanford University
Department of Civil and
Environmental Engineering
Stanford, CA 94305-4020
(650)723-4131
fax: (650) 725-9474
e-mail: mccarty@ce.stanford.edu
Technology:
In Situ Bioremediation; Aerobic Degradation
• Two 8-in diameter, PVC treatment wells installed approximately 24 m deep
and spaced 10m apart; equipped with submersible pumps
• Each treatment well screened in both the upper (15m) and lower aquifers (10
m)
• Groundwater recirculation - one well withdrew water from the upper aquifer
and discharged it into the lower aquifer, while the other well withdrew water
from the lower aquifer and discharged it into the upper aquifer creating a
bioreactive treatment cell
• Initial flow rate - 38 liters per minute (L/min)
• Operation included groundwater pumping, pulsed addition of toluene, and
addition of dissolved oxygen (DO, as gaseous oxygen) and hydrogen
peroxide (H2O2)
• An area of 480 m2 (0.12 acres) was monitored using 20 monitoring wells
• The demonstration included five phases, during which time the operating
parameters were varied: pre-operational studies (days 0 - 33); establishment
of a toluene-degrading consortium (days 34 - 55); pre-steady-state operation
(days 56 -136); steady-state operation (days 145 - 271); and balanced flow
operation (days 317 - 444)
Type/Quantity of Media Treated:
Groundwater
• Volume of water in test area -1,160 m3
• Volume of water pumped -12,132 m3 from upper to lower aquifer; 16,063 m3
from lower to upper aquifer
• Groundwater contaminant plume of approximately 53 acres
• Two relatively homogeneous aquifers - upper, unconfined aquifer is 8 m
thick, and separated by a 2 m aquitard from the lower confined aquifer;
lower, confined aquifer is approximately 5 m thick and lies above weathered
bedrock
Regulatory Requirements/Cleanup Goals:
• The objectives of the field demonstration included evaluate the effectiveness of in situ bioremediation to treat TCE in
groundwater and to collect data for potential full-scale application at the site
• Specific remedial goals were not established for the demonstration
96
-------
Aerobic Degradation at Site 19,
Edwards Air Force Base, California
Results:
• The system was found to be technically feasible for remediation of TCE in a two aquifer system
• TCE concentrations were reduced by 97.7%, from levels of up to 1,150 ug/L to 27 ug/L
• The average reduction of TCE during steady-state operation (days 145 - 271) was 87% in the upper aquifer bioactive
zone and 69% in the lower aquifer adjacent to treatment well Tl discharge screen
• The average reduction of TCE during balanced flow operation (days 365 - 444) was 86% and 83% in the upper and
lower aquifer bioactive zones, respectively
• No information was provided about potential degradation products from this demonstration
Costs:
• The total cost for the demonstration at Edwards AFB was $337,807, including $323,453 in capital costs and $14,354 in
O&M costs
Description:
Edwards Air Force Base covers approximately 301,000 acres, is located on the western portion of the Mojave Desert,
about 60 miles north of Los Angeles, and is used for aircraft research and development. From 1958 through 1967, rocket
engines were maintained in facilities at the site. Spent TCE from maintenance operations was disposed at Site 19, a 53
acre area on the west side of Rogers Dry Lake. The resulting groundwater contaminant plume extends approximately
3,200 ft down-gradient from the source area. The site was added to the National Priorities List in August 1990. A Record
of Decision (ROD) had not been signed for this facility at the time of this report.
Site 19 at Edwards Air Force Base was selected for a field demonstration to evaluate in situ bioremediation for the
treatment of groundwater contaminated with TCE. The system used for the demonstration consisted of two treatment
wells screened in both the upper and lower aquifers. One treatment well was used to withdraw water from the upper
aquifer and discharged it into the lower aquifer, while the other treatment well was used to withdraw water from the lower
aquifer and discharge it into the upper aquifer. This process recirculated the water between the two aquifers creating a
bioreactive treatment cell. Treatment system operation included the pulsed addition of toluene, and the addition of
dissolved oxygen and hydrogen peroxide (H2O2). The demonstration included steady-state and balanced flow operation.
The results of the field demonstration showed that in situ bioremediation using groundwater recirculation was technically
feasible for remediating TCE in a two aquifer system. TCE concentrations were reduced by 97.7%. The average
reduction of TCE during steady-state operation was 69% to 87% in the lower and upper aquifer bioactive zones,
respectively. The average reduction of TCE during balanced flow operation was 83% and 86% in the lower and upper
aquifer bioactive zones, respectively. Prevention of well clogging was found to be an important operational concern for
application of this technology. In this demonstration, site operators used well redevelopment and addition of hydrogen
peroxide to control clogging.
97
-------
In Situ Bioremediation at the Hanford 200 West Area Site, Richland, Washington
Site Name:
Hanford 200 West Area
Location:
Richland, Washington
Period of Operation:
January 1995 to March 1996
Cleanup Authority:
Not identified
Purpose/Significance of Application:
In situ bioremediation of chlorinated solvents and nitrate, including use of a
computer-based tool to aid in system design and operating strategies
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated solvents
• Concentrations in groundwater at the demonstration site were approximately
2 mg/L for carbon tetrachloride (CC14) and about 250 mg/L for nitrate
• Estimated 600,000 kg of CC14 in soil and groundwater at demonstration site
Waste Source:
Chemical processing operations
Contacts:
Technical Contact:
Rod Skeen
Principal Investigator
Pacific Northwest National
Laboratory
(509) 375-2265
Management Contact:
Jim Wright
DOE EM-50
Subsurface Contaminants Focus Area
Manager
(803) 725-5608
Licensing Information:
John Sealock
Technology Transfer
PNNL
(509) 375-3635
Technology:
In Situ Bioremediation
• One injection/extraction well pair (dual multi-screened wells) used to
recirculate groundwater; two monitoring wells located between recirculation
wells; a nutrient injection system; and a groundwater sampling system
• Groundwater was extracted and filtered, nutrients were added, and reinjected
• Nutrients consisted of acetate and nitrate pulses added at 24 hr intervals; the
nitrate pulses were skewed 10 hrs from the acetate pulses
• An Accelerated Bioremediation Design Tool (ABDT) was used to determine
pulse requirements
• Two separate tests were performed - one in the upper aquifer zone and one in
the lower aquifer zone
Type/Quantity of Media Treated:
Groundwater
• The unsaturated zone is 75 m thick and uncontaminated
• Upper aquifer zone occurs at 75 - 78 m bgs; lower aquifer zone occurs at 87 -
92 m bgs; zones separated by low permeability unit and do not interact with
each other significantly
Regulatory Requirements/Cleanup Goals:
Purpose of the demonstration was to evaluate the ability of the technology to degrade chlorinated solvents and to collect
information about the use of ABDT
Results:
• Approximately 2 kg of CC14 were biodegraded during the upper and lower zone tests, with less than 2% conversion to
chloroform
• CC14 biodegradation rate - 0.8 mg/g-biomass/day in upper zone and 0.9 mg/g/day in lower zone
• The concentration of CC14 in the upper zone was reduced from approximately 2.0 to 1.2 mg/L after 100 days
• The upper zone test produced more than 20 kg of bacteria and the lower zone more than 10 kg (dry weight)
• No plugging of the injection well was observed
• The ABDT was used to design and operate an effective in situ bioremediation system for the demonstration
98
-------
In Situ Bioremediation at the Hanford 200 West Area Site, Richland, Washington
Costs:
• An analysis of projected costs showed that the costs for in situ bioremediation were $5.80/m3, compared to $13.30/m3
for the baseline technology of air sparging/GAC; the treatment time was estimated as 1.9 yrs for ISB and 4.5 yrs for
AS/GAC
• In situ bioremediation is cost-effective where plumes or portions of plumes are small enough for volumetric treatment
(100 m diameter range), in aquifers where contaminant plumes exhibit non-equilibrium contaminant partitioning, and in
source area plumes with significant contaminant sorption
Description:
The Hanford Site's mission has been to support national defense efforts through the production of nuclear materials.
From 1944 to 1989, as part of the plutonium recovery processes, a variety of wastes including solvents, metals, and
radionuclides were released to the soil and groundwater. Soil and groundwater at the 200 West Site Area at Hanford,
located approximately 250 ft north of the sanitary tile field and 750 ft west of the 221-T plant, is contaminated with an
estimated 600,000 kg of CC14. The 200 West Site Area was selected for a field-scale demonstration of in situ
bioremediation. The demonstration included two separate tests, which were conducted in distinct, unconnected aquifer
zones at the test site.
A recirculating well in situ bioremediation system was demonstrated at the 200 West Site Area, which showed reductions
in the mass and concentration of CC14 in the two aquifer zones. Lessons learned from the field demonstration included
that effective ISB system design and operational process control requires an ABDT or similar process simulator, and that
use of an ABDT allows quick corrective action (such as changes in the amount/duration of nutrient pulse or the pulse
period) to maintain rapid contaminant destruction during these changes. In addition, ISB was found to yield significant
economic and efficiency gains over conventional baseline technologies for remediation of groundwater contaminated with
VOCs and nitrates, and to be potentially effective for treating plumes caused by dissolution of non-aqueous phase liquids.
99
-------
Aerobic Degradation at Moffett Naval Air Station, Mountain View, California
Site Name:
Moffett Naval Air Station
Location:
Mountain View, California
Period of Operation:
September 1986 to November 1988 (three seasons)
Cleanup Authority:
CERCLA
Purpose/Significance of Application:
One of the earliest field demonstrations of aerobic in situ bioremediation under
varying experimental conditions
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated Solvents
• 1,1,1 -trichloroethane (TCA) and 1,1 -dichloroethane (DCA) found in test
zone; regulatory approval obtained for adding TCE, cis- and trans-DCE, and
VC to the injected groundwater for demonstration
Waste Source:
Leaks and spills from aircraft and
maintenance operations; disposal of
waste in landfills
Contacts:
EPA RPM:
Roberta Blank
U.S. EPA Region 9
75 Hawthorne Street, SFD-8-1
San Francisco, CA 94105
(415)744-2384
e-mail: blank.roberta@epa.gov
Principal Investigator:
Dr. Lewis Semprini
Oregon State University
Department of Civil, Construction,
and Environmental Engineering
202 Apperson Hall
Corvallis, OR 97331-2302
(541) 737-6895
fax: (541) 737-3099
e-mail: Lewis.Semprini@orst.edu
Technology:
In Situ Bioremediation; Aerobic Degradation
• One extraction well and two injection wells used to create groundwater
recirculation treatment cell
• TCE, cis- and trans-DCE, and VC injected into groundwater (regulatory
approval obtained)
• Experiments conducted using native bacteria, methane addition, phenol and
toluene addition, and hyfrogen peroxide addition; bromide tracer tests also
performed
Type/Quantity of Media Treated:
Groundwater
• Test zone located in shallow, confined aquifer -1.5 m thick; approximately 4
to 6 m bgs
• Hydraulic conductivity -0.11 cm/sec; indigenous methanotrophic bacteria
present in aquifer
Regulatory Requirements/Cleanup Goals:
• The objectives of the field demonstration included evaluating the performance of in situ biodegradation of chlorinated
aliphatic hydrocarbons (CAHs) using native bacteria enhanced through addition of methane, toluene, and phenol
• Specific remedial goals were not established for this demonstration
Results:
• Methane addition was required for biodegradation of CAHs
• Removal rates for methane addition - TCE (20 - 30%), cis-DCE (45 - 55%), trans-DCE (80 - 90%), and VC (90- 95%);
rate of TCE reduction remained relatively constant over three seasons of testing
• Use of phenol and toluene achieved higher percent removals of TCE (93 - 94%)
• Presence of 1,1-DCE was toxic to the transforming bacteria
Costs:
Not provided
100
-------
Aerobic Degradation at Moffett Naval Air Station, Mountain View, California
Description:
Moffett Naval Air Station, used for aircraft operations and maintenance, operated from 1933 to 1994, and is located 35
miles south of San Francisco in Santa Clara County. In 1994, the Navy ceased operations and the airfield was transferred
to the National Aeronautics and Space Administration. Soil and groundwater at the site are contaminated with petroleum
products and VOCs, including TCE and PCE. Moffett was selected for a field demonstration of aerobic biodegradation
and a series of experiments were conducted to evaluate the performance of the technology in treating CAHs using native
bacteria enhanced through addition of methane, toluene, and phenol.
Results showed that active use of methane in the treatment zone was required for biodegradation of CAHs, and that
groundwater residence times in the treatment zone of 1-2 days resulted in biodegradation of TCE, DCE, and VC. The use
of phenol and toluene achieved higher percent removals of TCE (93 - 94%) compared with use of methane (19%), and
hydrogen peroxide was found to achieve TCE removals similar to those achieved using oxygen. While 1,1-DCE was
partially transformed in the study with phenol, the transformation products were toxic to the transforming bacteria.
Therefore, the use of this technology when 1,1-DCE is present may not be appropriate. Alternating pulsed addition of
methane and oxygen helped to prevent biofouling in the area near the injection well. According to the researchers, the
relatively low concentration of phosphate in the groundwater did not limit the biodegradation of CAHs at this site; other
phosphate minerals may have dissolved in the groundwater to replenish this mineral as it was being removed by the
bacteria. The results from the field experiments were found to be consistent with the results from batch soil column
laboratory testing using aquifer solids from the test zones.
101
-------
Enhanced In Situ Anaerobic Bio remediation of Fuel-Contaminated Ground Water
Site Name:
Naval Weapons Station (NWS) Seal Beach
Location:
Southern CA
Period of Operation:
9/97 - 10/98
Cleanup Authority:
California Regional Water Quality
Control Board
Regulatory Authority:
Lawrence Vitale
CARWQCB Region 8
2010 Iowa Ave, Suite 100
Riverside, CA 92507-2409
(909) 782-4130
Purpose/Significance of Application:
Demonstrate anaerobic bioremediation for treating fuel hydrocarbons
Cleanup Type:
Field demonstration
Contaminants:
Fuel hydrocarbons and BTEX
• Maximum concentrations in groundwater: benzene - 4,000 ug/L;
ethylbenzene - 250 ug/L; m+p-xylenes - 500 ug/L
Waste Source:
Leaks from USTs
Contacts:
Project Management:
Carmen A. LeBron
Naval Facilities Engineering Service
Center
1100 23rd Ave, ESC 411
Port Hueneme, CA 93043
Telephone: (805) 982-1616
Fax: (805) 982-4304
E-mail: lebronca@nfesc.navy.mil
Principal Investigator:
Martin Reinhard
Dept. of Civil and Environ. Engr.
Stanford University
Stanford, CA 94305 ^4020
Telephone: (650) 723-0308
Fax:(650)725-3162
E-mail: reinhard(@cive.stanford.edu
Technology:
In Situ Bioremediation
• Demonstration used one extraction and three injection wells (three zones of
180 m3 each)
• Extraction rate 4.5 L/min; injection 1.5 L/min/well
• Electron acceptors varied by zone - one zone augmented with sulfate, one
with sulfate and nitrate, one with none; three rounds of augmentations
performed
• Sampling performed with automated system
Type/Quantity of Media Treated:
Groundwater (in situ), Soil (in situ), LNAPL
• Contaminated area 20 acres, demonstration conducted on portion of site
• Groundwater velocity 0.7 cm/sec; transmissivity >2 ftVday; depth to
groundwater low
• Groundwater had been anaerobic for > 10 yrs
Regulatory Requirements/Cleanup Goals:
• Demonstrate the technical viability of the technology to treat petroleum hydrocarbons and to stimulate biodegradation
of BTEX with nitrate and sulfate
• No specific cleanup goals were identified
Results:
• Concentrations of BTEX compounds were reduced, with toluene preferentially degraded
• Ethylbenzene and m+p-xylene degradation stimulated by nitrate, with concentrations reduced from 250 to <10 ug/L for
ethylbenzene and from 500 to <20 ug/L for xylenes
• O-xylene degradation stimulated by sulfate, with concentration reduced from >400 to <10 ug/L
• Benzene removal was mostly due to flushing rather than biodegradation
Costs:
• Demonstration costs were $875,000, including equipment, labor, laboratory supplies, travel, and overhead; >9,000
samples were collected
• Projected present value costs for a full-scale bioremediation application were $ 1,085,000, or $4,340/gallon of fuel
recovered, compared with similar costs for pump and treat of $1,530,000, or $6,120/gallon of fuel recovered
102
-------
Enhanced In Situ Anaerobic Bio remediation of Fuel-Contaminated Ground Water
Description:
In 1984, a fuel leak was discovered at the Naval Weapons Station (NWS) Seal Beach when a steel tank was replaced with
fiberglass tanks. NWS Seal Beach is located in southern California between Long Beach and Huntington Beach. About
5,800 gallons of fuel had leaked and migrated to the groundwater and was a concern for its potential effects on a local
wildlife refuge.
A demonstration of in situ bioremediation was performed in a portion of the contaminated area of this site. The
demonstration evaluated the performance of various concentrations of sulfate and nitrate in three zones between one
extraction well and three injection wells. The results showed that concentrations of BTEX compounds were reduced,
with toluene preferentially degraded. Ethylbenzene and xylenes also were degraded, but benzene was found to be
removed mostly by flushing. Projected full-scale costs for in situ bioremediation were found to be approximately 30%
less than for pump and treat. Lessons learned included the effect of BTEX compounds in a non-aqueous phase, the
demand of non-BTEX fuel hydrocarbons on sulfate and nitrate, and the role of sulfate and nitrate as terminal electron
acceptors.
103
-------
In Situ Bioremediation (Anaerobic/Aerobic) at Watertown, Massachusetts
Site Name:
Not identified
Period of Operation:
Anaerobic: November 1996 to July 1997;
Aerobic: August 1997 to ongoing (data available through October 1997)
Purpose/Significance of Application:
Combined anaerobic/aerobic system for treatment of chlorinated solvents
Contaminants:
Chlorinated Solvents
• TCE, PCE; initial TCE levels were 12 mg/L
Contacts:
Technology Researcher:
Dr. Willard Murray
Harding Lawson Associates
107 AudubonRoad Suite 25
Wakefield,MA01880
(781)245-6606
E-mail: wmurray@harding.com
EPA Contact:
Dr. Ronald Lewis
U.S. Environmental Protection
Agency
26 W. Martin Luther King Dr.
Cincinnati, OH 45268
(573) 569-7856
lewis.ronald@epa.gov
Location:
Watertown, Massachusetts
Cleanup Authority:
Not identified
Cleanup Type:
Field demonstration
Waste Source:
Manufacturing operations
Technology:
In situ bioremediation
• A "two-zone" enhanced bioremediation process that used sequential
anaerobic and aerobic biodegradation processes to degrade PCE and TCE;
anaerobic conditions were used for eight months (through late July 1997),
then changed to aerobic conditions
• The system was a groundwater recirculating cell that consisted of three
injection wells and three extraction wells, and covered a surface area of
approximately 10 ft by 20 ft; with wells screened from 13 to 20 ft bgs
• Nutrients and a carbon source were injected into the groundwater through the
three up-gradient wells and extracted through the three down-gradient wells
• A relatively constant recirculating flow rate of 0.25 gpm was used along with
an amendment injection rate of about four gallons per day (approximately 1%
of the recirculating flow)
• Lactic acid was used in the anaerobic conditions, and ORC socks plus
methane in aerobic conditions
Type/Quantity of Media Treated:
Groundwater
• Soil at the Watertown site consists of about 13 ft of sand and gravel over
approximately 7 ft of silty sand
• Depth to groundwater is approximately 8 ft bgs
Regulatory Requirements/Cleanup Goals:
Purpose of the demonstration was to evaluate the use of a combined anaerobic and aerobic system for treatment of
chlorinated solvent
Results:
• After four to five months of operation of anaerobic operation, significant increases in DCE were observed along with
decreases in TCE concentrations, indicating that reductive dechlorination was occurring; no significant increases in VC
concentrations were observed until July 1997, 8 months after operations began
• By July 1997, TCE concentrations had been reduced from about 12 mg/L at the beginning of the demonstration to less
than 1 mg/L and there was an overall reduction of about 80% in the mass of total VOCs
• During the aerobic phase, levels of DCE and vinyl chloride have started to decrease in the groundwater; in addition,
DCE epoxide, a transient biodegradation product of aerobic degradation of DCE, was detected, indicating that aerobic
VOC-degrading bacteria have been stimulated
Costs:
• The field-scale pilot study has incurred a cost of approximately $150,000 through November 5, 1997
• No estimates were provided about the projected costs for a full-scale system using this technology
104
-------
In Situ Bioremediation (Anaerobic/Aerobic) at Watertown, Massachusetts
Description:
The Watertown site has been used since the late 1800's for a variety of operations, including a coal gas manufacturing
plant, which ceased operations in the 1930's, and a metal plating shop, which ceased operations in 1990. The site is
currently being used as a manufacturing facility for electric switch assembly. Soil and groundwater at the site are
contaminated with chlorinated solvents, including TCE and PCE, from past operations and waste disposal practices. A
field demonstration of the Two-Zone Plume-Interception Treatment Technology, developed by Harding Lawson
Associates (HLA, formerly ABB Environmental Services, Inc.), was conducted at the Watertown site under the Superfund
Innovative Technology Evaluation (SITE) program. The field demonstration is currently ongoing.
Under anaerobic conditions, TCE in groundwater was reduced by reductive dechlorination (from 12 mg/L to less than 1
mg/L) and there was an overall reduction of about 80% of the total VOC mass in one well. Data indicate that
methanogenic conditions were not achieved during the anaerobic phase and most of the reductive dechlorination was
attributed to sulfate-reducing bacteria. A period of about one month was required to establish aerobic conditions after
ORC socks were placed in the wells. This lag time was attributed to the presence of residual carbon that had to be
degraded before aerobic conditions could be established. Initial results indicate that VOC levels, primarily DCE and vinyl
chloride, are decreasing. According to EPA, future applications should consider not starting in the winter, start when the
anaerobic process can go quickly, use a higher level of lactate, and drive the oxidation potential down quickly.
105
-------
Methane Enhanced Bioremediation Using Horizontal Wells
at the Savannah River Site, Aiken, South Carolina
Site Name:
U.S. DOE Savannah River Site
Location:
Aiken, South Carolina
Period of Operation:
February 26, 1992 to April 30, 1993
Cleanup Authority:
CERCLA
Purpose/Significance of Application:
Field demonstration of in situ bioremediation system using horizontal wells and
methane injection
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated Solvents
• TCE and PCE concentrations in groundwater ranged from 10 to 1,031 ug/L
and 3 to 124 ug/1, respectively
• TCE and PCE concentrations in sediment ranged from 0.67 to 6.29 mg/kg
and 0.44 to 1.05 mg/kg, respectively
Waste Source:
Wastewater discharges from
aluminum forming and metal finishing
operations
Contacts:
Principal Investigators:
Dr. Terry C. Hazen
Lawrence Berkeley National
Laboratory
Center for Environmental
Biotechnology
MS70A-3317
One Cyclotron Road
Berkeley, CA
(510)486-6223
fax: (510)486-7152
tchazen@lbl.gov
Brian Looney
Westinghouse Savannah River
Company
PO Box 616
Aiken, SC 29802
(803) 725-64137(803) 725-3692
DOE Integrated Demonstration
Manager:
Kurt Gerdes
U.S. DOE
Office of Environmental Management
Science & Technology Development
Office of Technology Systems
Cloverleaf Room 113 5
Germantown, MD 20874
Telephone: (301) 903-7289
fax: (301)903-7457
Technology:
In Situ Bioremediation
• Methane enhanced bioremediation
• Two horizontal wells used for the demonstration:
- "lower" horizontal injection well - depth of 175 feet (below the water
table); screen length of 310 feet; "upper" horizontal extraction well -
depth of 80 feet (in the vadose zone); screen length of 205 feet
- Air and gas injection rate - 200 scfm; air and contaminant extraction rate -
240 scfm
• Catalytic oxidizer used to treat the extracted vapors
• Demonstration performed in six different operational modes:
- baseline tests of the vapor extraction and injection systems (with and
without air sparging)
- a series of nutrient additions (addition of 1% methane, 4% methane, pulsed
4% methane; and combination of nitrous oxide at 0.007% and triethyl
phosphate at 0.07% in air in combination with pulses of 4% methane)
- a helium tracer test
- an assessment of microbiological assays for monitoring performance
Type/Quantity of Media Treated:
Groundwater and sediment
• VOC plume was estimated to cover about 1200 acres and to be about 150-ft
thick
• Dense nonaqueous phase liquids (DNAPLs) have also been observed
• Depth to groundwater -120 to 135 feet bgs
• Groundwater velocity -15 to 100 feet/year
Regulatory Requirements/Cleanup Goals:
• Cleanup goals for groundwater included TCE (5 ppb) and PCE (5 ppb)
• Information was not provided about cleanup goals for sediment
106
-------
Methane Enhanced Bioremediation Using Horizontal Wells
at the Savannah River Site, Aiken, South Carolina
Results:
• After 384 days of operation, the system removed about 17,000 Ibs of VOCs through a combination of vacuum
extraction and biodegradation - the vacuum component of the system removed 12,096 Ibs of VOCs and the biological
component degraded 4,838 Ibs of VOCs
• After treatment, the total sediment inventory for both TCE and PCE decreased by 24%, with the concentrations of
VOCs in most sediment samples reported to be below the detection limits; concentrations of TCE and PCE in
groundwater were reported to be less than 5 ppb; soil gas concentrations reportedly decreased by more than 99%
• The addition of methane stimulated the growth of methanotrophs - 1% methane addition increased the population of
methanotrophs by several orders of magnitude, to levels close to 100,000 MPN/ml; 4% methane addition initially
increased the population of methanotrophs, which then decreased as a result of nutrient depletion
• The addition of nitrogen and phosphorous nutrients with pulsed methane stimulated microbial activity. This phase was
reported to optimize bioremediation and mineralization of TCE and PCE in groundwater and sediments
• Helium tracer tests indicated that more than 50% of the injected methane was consumed by indigenous microbes before
it reached the extraction well; results were not provided from the microbiological assays
Costs:
• Projected costs for full-scale application at this site were $452,407 for total capital costs (including equipment
amortized over 10 years, well installation, and mobilization) and $236,465 for operation and maintenance (including
monitoring, consumables, and demobilization)
Description:
The U.S. Department of Energy (DOE) Savannah River Site (SRS) is a 300 square mile facility located in Aiken, South
Carolina that has been used for the research and production of nuclear materials. Area M at the facility was used for
aluminum forming and metal finishing operations. Wastewaters from this area containing an estimated 3.5 million pounds
of solvents were discharged to an unlined settling basin, a process sewer line, and a nearby stream from the 1950's to the
1980's. High levels of chlorinated solvents, primarily TCE (up to 1,031 ug/L in groundwater and 6.29 mg/kg im
sediment) and PCE (up to 124 ug/L in groundwater and 1.05 mg/kg in sediment), were found at the site and DNAPLs
were observed. The VOC groundwater plume was estimated to cover about 1200 acres and to be about 150-ft thick.
From February 1992 to April 1993, DOE conducted a field demonstration of in situ methane enhanced bioremediation
using two horizontal wells - one located below the water table and used for injection and one located in the vadose zone
and used for extraction. A catalytic oxidizer was used to treat the extracted vapors. The demonstration was performed in
six different operational modes, varying the type and concentration of nutrients added and the use of pulsing. During the
demonstration, about 17,000 Ibs of VOCs were removed through a combination of vacuum extraction and biodegradation.
The addition of methane stimulated the growth of methanotrophs, with the addition of 1% methane increasing the
population of methanotrophs by several orders of magnitude. Results of a tracer test showed that more than 50% of the
injected methane was consumed by indigenous microbes before it reached the extraction well.
107
-------
In Situ Bioremediation at the Texas Gulf Coast Site, Houston, Texas
Site Name:
Texas Gulf Coast Site (actual site name confidential)
Period of Operation:
Ongoing (data available from June 1995 to December 1998)
Purpose/Significance of Application:
Groundwater recirculation system using trenches for extraction and injection
Contaminants:
TCE, cis-l,2-DCE, VC
• TCE was present at approximately 50 mg/L
Contacts:
Site Contractor:
Susan Tighe Litherland, P. E
David W. Anderson, P.E., P.O.
Roy F. Weston, Inc.
5300 Bee Caves Road, Suite 1-100
Austin, TX 78746
(512)329-8399
fax:(512)329-8348
e-mail: litherls@mail.rfweston.com
e-mail: andersod@mail.rfweston.com
Site Contact:
Not identified
Location:
Houston, Texas
Cleanup Authority:
State of Texas Voluntary Cleanup
Program; administered by TNRCC
Cleanup Type:
Full scale
Waste Source:
Leaks and spills from manufacturing
operations
Technology:
In situ bioremediation
• An extraction-injection recirculation system, completed in May 1995,
consists of an alternating series of four extraction (1,800 linear ft total) and
four injection (1,100 linear ft total) trenches set at a spacing of approximately
100ft
• The extraction trenches were completed to a depth of at least one foot into the
bottom clay layer (20 - 22 ft bgs), and were sloped to a sump
• System operation consists of groundwater circulation and addition of
methanol
• As of January 1999, the recirculation rate averages 6 to 8 gpm, and a total of
12 million gallons have been recirculated through the system (approximately
2.5 pore volumes)
Type/Quantity of Media Treated:
Groundwater
• The area of contaminated groundwater is approximately 600 ft by 700 ft in an
unconsolidated water-bearing zone which occurs at a depth of approximately
12 -20 ft bgs
• Hydraulic conductivity is 1 x 10"4 to 4 x 10"4 cm/sec
• Groundwater velocity is 4 - 18 ft/yr
Regulatory Requirements/Cleanup Goals:
• The primary objectives of the clean up are to actively remediate the contaminated groundwater at this site to a point that
natural attenuation would prevent further migration of the plume, and to discontinue active treatment
• No specific cleanup goals have been identified for groundwater at this site
Results:
• Excluding results from the one potential "source" area, the average decrease in TCE concentrations is approximately
99% (from an average of 12 to 0. 12 mg/L) during a 3 l/i year period
• TCE concentrations in portions of the plume have decreased to below the detection limit (0.005 mg/L).
• Accounting for dilution, the site contractor reported that TCE concentrations were reduced by approximately 2% per
month during a period of nutrient-only addition, and approximately 10% per month during the period of methanol
addition
• The ratio of cis-l,2-DCE to TCE increased from approximately 0.06: 1 to 0.30: 1 after addition of methanol, suggesting
more active dechlorination associated with higher concentrations of substrate.
Costs:
• Capital costs for construction of the extraction/injection trenches and control building were approximately $600,000
• Annual costs for operation, maintenance and monitoring are approximately $100,000
108
-------
In Situ Bioremediation at the Texas Gulf Coast Site, Houston, Texas
Description:
The Texas Gulf Coast site is an abandoned industrial manufacturing facility located near Houston, Texas that operated
between 1952 and 1985. Trichloroethene was used at the facility and was found in the groundwater starting in 1986. In
situ bioremediation is being used to clean up groundwater at the site under the State of Texas Voluntary Cleanup Program.
Methanol addition was found to increase the rate of biodegradation of TCE at this site, based on the reduction of TCE
concentration and increase in the ratio of cis-l,2-DCE to TCE. This site is planning to stop using active bioremediation
after four years of system operation (three years of methanol addition) to allow use of natural attenuation. According to
the site contractor, natural attenuation will be used to prevent future migration of the plume, and to achieve stable or
declining contaminant concentrations. Excessive biomass formation, leading to a reduced flow rate, was found to be a
concern for addition of methanol. Excess biomass was not noted during the period when nutrients alone were added;
however, a significant increase in biomass formation was noted after addition of methanol. To remedy this, the site
contractor modified their methanol addition to a batch system. The site contractor found that it was difficult to balance
the system hydraulics between the extraction and infiltration trenches, and that it required approximately one year of
operating time to achieve a balance. In addition, they found it difficult to interpret the treatment performance data
because of the non-homogeneous nature of the initial groundwater quality, and dilution due to recharge of rainwater and
clean water from beyond the planned treatment area.
109
-------
In Situ Redox Manipulation at U.S. DOE Hanford Site, 100-H and 100-D Areas
Site Name:
U.S. Department of Energy Hanford Site, 100-H and 100-D Areas
Location:
Richland, WA
Period of Operation:
September 1995 to September 1998
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstrate in situ redox manipulation for treatment of hexavalent chromium
Cleanup Type:
Field demonstration
Contaminants:
Chromium
• Initial chromate concentrations 60 ug/L in 100-H area and 910 ug/L in 100-D
area
Waste Source:
Nuclear processing operations
Contacts:
Technical Contacts:
John Fruchter
Pacific Northwest National
Laboratory
(509) 376-3937
Wayne Martin
Pacific Northwest National
Laboratory
(509) 372-4881
Management Contacts:
James A. Wright
DOE SR, Field Manager
(803) 725-5608
Technology:
In Situ Redox Manipulation (ISRM)
• The field demonstration used 20,500 gallons of buffered sodium dithionite
solution (Na2S2O4, also known as hydrosulfite) to react with natural iron in
the subsurface and form reduced iron (Fe2+); the reduced iron reacts with
chromate to form insoluble chromium oxides
• Dithionite solution was injected through one 8-inch diameter
injection/extraction well, allowed to react for 18 hrs, and then withdrawn; this
created a reduced zone 50 ft in diameter
• The withdrawal phase took 83 hrs and 4.8 injection volumes to remove
unreacted reagent, buffer, reaction products, bromide tracer, and mobilized
metals
• 16 two-inch monitoring wells were used to assess physical and chemical
conditions after the test
Type/Quantity of Media Treated:
Groundwater
• Depth to groundwater is 50 ft in 100-H area and 85 ft in 100-D area
• Aquifer is 15-20 ft thick
Regulatory Requirements/Cleanup Goals:
• Evaluate performance of ISRM for treating chromium in groundwater
• No specific cleanup goals were identified
Results:
• Concentrations of chromium in groundwater were reduced to less than 8 ug/L in one month
• 87-90% of the dithionite solution was recovered during the withdrawal phase, along with most of the mobilized metals
(Fe, Mn, Zn)
• Within 25 ft of the injection well, 60-100% of the available iron was reduced; this zone was estimated to have a life of
7-13 yrs
• Two years after treatment was complete, the treatment zone remained anoxic and hexavalent chromium below detection
limits
Costs:
• Projected costs for use of ISRM in a full-scale deployment at this site were identified using two methodologies (one for
a 200 ft barrier and one for a 1,400 ft barrier), both in comparison to projected costs for pump and treat; this analysis
showed cost savings for use of ISRM of $4.6 to 16 million
110
-------
In Situ Redox Manipulation at U.S. DOE Hanford Site, 100-H and 100-D Areas
Description:
The 100 Area of the Hanford site contains nine nuclear reactors, and is located in the north-central portion of the site near
the Columbia River. During reactor operations, chromium was introduced to the soil and groundwater in this area.
A demonstration of in situ redox manipulation (ISRM) was conducted in the 100-H and 100-D areas at Hanford that
consisted of field-scale demonstrations. ISRM is a passive barrier technique that uses injection of buffered sodium
dithionite solution (Na2S2O4) to react with natural iron in the subsurface and form reduced iron (Fe2+); the reduced iron
reacts with chromate to form insoluble chromium oxides. Results from the field demonstration test showed that initial
chromate concentrations of 60 ug/L in the 100-H area and 910 ug/L in the 100-D area were reduced to less than 8 ug/L in
a one month period. In addition, 87-90% of the dithionite solution was recovered during the withdrawal phase, along with
most of the mobilized metals (Fe, Mn, Zn). A full-scale deployment for the Hanford 100-HR-3 operable unit is planned
to begin in late 1999.
111
-------
In Situ Chemical Oxidation Using Potassium Permanganate at Portsmouth Gaseous
Diffusion Plant, X-701B Facility
Site Name:
Portsmouth Gaseous Diffusion Plant, X-701B Facility
Location:
Piketon, OH
Period of Operation:
Spring 1997 (operated for one month)
Cleanup Authority:
RCRA Corrective Action
Purpose/Significance of Application:
Demonstrate in situ chemical oxidation for treating chlorinated solvents
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated solvents
• Initial TCE concentrations in groundwater averaged 176.7 mg/L
Waste Source:
Leaks from USTs
Contacts:
Technical Contacts:
Robert L. Siegrist
Colorado School of Mines and Oak
Ridge Natl. Lab.
(303) 273-3490
Olivia West
Oak Ridge Natl. Lab.
(423) 576-5005
Management Contacts:
Tom Houk
Bechtel Jacobs Company, LLC
(740) 897-6502
James A. Wright
DOE SR, Field Manager
(803) 725-5608
Technology:
In Situ Chemical Oxidation
• Demonstration used a pair of parallel horizontal wells - one to extract
groundwater (6 gpm) and one to reinject after addition of potassium
permanganate (KMnO4)
• Each well had a 200 ft screened section located in a 5 ft thick silty, gravel
aquifer in the center of a plume
• Crystalline KMnO4 was added to the extracted groundwater and reinjected
into the downgradient well 90 ft from the extraction well; a total of 206,000
gals of KMnO4 solution was injected
• Oxidant solution (~2% KMnO4) was recirculated for one month
• Delivery of oxidant solution was not uniform throughout the horizontal well;
a subsequent injection of KMnO4 was made into a nearby vertical well for 8
days to enhance delivery
• System shutdowns were due to heavy rains, well-screen clogging, and repairs
Type/Quantity of Media Treated:
Groundwater (in situ)
• The Gallia sand and gravel unit was the target for the demonstration
• DNAPL compounds (mostly TCE) were located 25-35 ft bgs, 12 ft below top
of water table
• Area of contamination approximately 90 ft by 220 ft by 6 ft (119,000 ft3)
containing 272.7 Ibs of TCE
Regulatory Requirements/Cleanup Goals:
• Evaluate performance of the technology in degrading TCE
• No specific cleanup goals were identified
Results:
• Average concentrations of TCE were 176.7 mg/L before treatment, 110 mg/L at completion of treatment, and 41 mg/L
two weeks after recirculation ended; concentrations increased to 65 mg/L at 8 weeks and 103 mg/L at 12 weeks after
recirculation ended
• Immediately after recirculation ended, concentrations of TCE were low (BDL to low ug/L) in monitoring wells where
KMnO4was also detected
• Residual concentrations of KMnO4were detected at nine monitoring well locations 19 months after the demonstration
ended
112
-------
In Situ Chemical Oxidation Using Potassium Permanganate at Portsmouth Gaseous
Diffusion Plant, X-701B Facility
Costs:
• The estimated cost for the demonstration was $562,000, consisting of project management ($67,440), pre-
demonstration characterization ($162,980), remediation operations/oxidant recirculation ($162,980), resistivity
monitoring ($67,440), and post-demonstration characterization and demobilization ($101,160)
• Projected costs for use of the technology at a full-scale were $516,360, to treat a hot spot area of 22.9 acres in the
central portion of the X-701B plume; this corresponds to $64/yd3
Description:
The Portsmouth Gaseous Diffusion Plant (PORTS), located 80 miles south of Columbus, Ohio, is a 3,714-acre DOE
reservation. It was constructed between 1952 and 1956 and enriches uranium for electrical power generation. The X-
70 IB site, located in the northeastern area of PORTS, contains an unlined 200 ft by 50 ft holding pond. The pond was
used from 1954 to 1988 for neutralization and settling of metal-bearing acidic wastewater and solvent-contaminated
solutions. During a RCRA Facility Investigation, TCE was detected in a groundwater sample at 700 mg/L.
A field demonstration of in situ chemical oxidation was conducted at PORTS using a pair of parallel horizontal wells -
one for extraction and one for reinjection. Crystalline KMnO4 was added to extracted groundwater and reinjected into the
downgradient well 90 ft from the extraction well; a total of 206,000 gals of KMnO4 solution was injected and recirculated
for one month. Results showed that immediately after recirculation ended, concentrations of TCE were low (DDL to low
ug/L) at those locations where KMnO4 was detected in the monitoring well. However, oxidant addition was not uniform
and average concentrations were higher -110 mg/L at completion of treatment, and 41 mg/L two weeks after recirculation
ended. The researchers concluded that the number and pattern of extraction and injections wells must be designed to
ensure maximum coverage of the treatment zone.
113
-------
Phytoremediation Using Constructed Wetlands at the Milan Army Ammunition
Plant, Milan, Tennessee
Site Name:
Milan Army Ammunition Plant
Location:
Milan, Tennessee
Period of Operation:
June 17, 1996 to July 21, 1998
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Use of constructed wetlands for treatment of explosives-contaminated
groundwater
Cleanup Type:
Field demonstration
Contaminants:
Explosives
• Total nitrobody (the sum of the following six explosives: TNT, RDX, HMX,
TNB, 2A-DNT, and 4A-DNT) concentrations in groundwater ranged from
3,250 to 9,200 ppb
• TNT concentrations in groundwater ranged from 1,250 to 4,440 ppb
• RDX concentrations in groundwater ranged from 1,770 to 4,240 ppb
• HMX concentrations in groundwater ranged from 87 to 110 ppb
Waste Source:
Industrial wastewater discharged to
ditches
Contacts:
AEC Project Manager:
Darlene F. Bader
U.S. Army Environmental Center
ATTN: SFIM-AEC-ETD (Bader)
5179HoadleyRoad
APG,MD 21010-5401
(410)436-6861
Fax: (410) 436-6836
E-mail: dfbader@aec.apgea.army.mil
TVA Program Manager:
Richard A. (Rick) Almond
Tennessee Valley Authority
Reservation Road CEB 4C
Muscle Shoals, AL 35661
(256) 386-3030
Fax: (256) 386-3799
E-mail: raalmond@tva.gov
Technology Research Biologist:
Dr. Susan L. Sprecher
Chemical Control Technology
U.S. Army Corps of Engineers
Waterways Experiment Station, ES-p
3 909 Halls Ferry Rd.
Vicksburg, MS 39180-6199
(601) 634-2435
Fax: (601) 634-2617
Technology:
Constructed Wetlands
• Two types of wetlands were demonstrated - a gravel-based system and a
lagoon-based system
• Both systems were designed to retain groundwater for approximately 10 days
at an influent flow rate of 5 gpm per system
• The gravel system consisted of two 4 ft deep gravel-filled beds (cells)
connected in series and planted with emergent plants; the first cell (0.088
acre) was maintained as anaerobic (by carbon addition) and the second cell
(0.030 acre) as aerobic; emergent plants used were canary grass, wool grass,
sweetflag, and parrotfeather
• The lagoon system consisted of two 2 ft deep lagoons (cells) connected in
series and planted with submergent plants
• The demonstration was conducted in three phases - (I) plant screening and
treatability studies; (II) design, construction, and 16 months of monitoring;
and (III) longer-term monitoring and optimization
Type/Quantity of Media Treated:
Groundwater
• Groundwater flow north-northwest
Regulatory Requirements/Cleanup Goals:
• Reduce concentration of TNT to less than 2 ppb, and total nitrobody concentrations (see contaminants) to less than 50
ppb
114
-------
Phytoremediation Using Constructed Wetlands at the Milan Army Ammunition
Plant, Milan, Tennessee
Results:
• The gravel-based system performed better than the lagoon-based system
• The gravel system reduced TNT, RDX, and HMX concentrations to below the cleanup goals during all but the coldest
months; in addition, a sustainable ecosystem was established
• The lagoon system met the cleanup goal for TNT of 2 ppb only during the first 50 days of the demonstration, but did
not remove RDX and HMX or meet the total nitrobody goals; in addition, an adequate plant population was not
maintained within the lagoon system
Costs:
• Projected costs for a 10-acre, full-scale, gravel-based system designed to treat 200 gpm of contaminated groundwater at
Milan AAP were $3,466,000 ($1998).
• Assuming a 95% system availability and 30-yr life, the total cost (capital plus O&M) for use of this system was
estimated as $1.78 per 1,000 gals of groundwater
Description:
The Milan Army Ammunition Plant (MAAP) is a government-owned, contractor-operated military industrial installation
within the U.S. Army Industrial Operations Command. The original facility was constructed during World War II.
MAAP is located on 22,436 acres of land, which include approximately 548 acres for various production lines, 7,930
acres for storage areas, and 1,395 acres for administrative, shop maintenance, housing, recreation, and other functions.
From World War II to 1981, MAAP's production facilities discharged explosives-contaminated wastewater directly into
open ditches that drained from sumps or surface impoundments into local streams. Several of these drainage ditches
became contaminated with explosive residuals which leached into the groundwater. In 1981, the production facility's
wastewaters were redirected to explosives-contaminated wastewater treatment plants.
A wetlands demonstration system was constructed in Area K adjacent to Building K-100. The demonstration consisted of
gravel- and lagoon-based systems, and was conducted over a two-year period. The study found that the gravel-based
system had results better than the lagoon system, and met the goals during all but the coldest months. The lagoon system
did not consistently meet the goals, and had several operational problems, including a severe tadpole infestation and a
hailstorm. The demonstration study authors concluded that a wetland's economic and technical feasibility depends on
site-specific factors such as regional temperature variations, rainfall patterns, groundwater flow characteristics, explosive
type and concentration, the presence of other contaminants, and regulatory requirements. In general, they found that
wetlands perform better in warmer climates with moderate levels of rainfall.
115
-------
Multi-Phase Extraction at the 328 Site, Santa Clara, CA
Site Name:
328 Site
Location:
Santa Clara, CA
Period of Operation:
November 19, 1996 to May 4, 1999
Shutdown period to assess rebound: June 5, 1998 through September 8, 1998
Cleanup Authority:
State of California San Francisco Bay
Regional Water Quality Control Board
Purpose/Significance of Application:
Use of DPE with pneumatic fracturing to remove VOCs from silty clay soils and
shallow groundwater
Cleanup Type:
Full scale
Contaminants:
Chlorinated Solvents
• Trichloroethene (TCE) is the primary contaminant of concern, with the
highest TCE concentration measured in the soil and groundwater during the
remedial investigation at 46 mg/kg and 37,000 ug/L, respectively
Waste Source:
Storage of waste from vehicle
manufacturing operations
Contacts:
Vendor:
Jeffrey C. Bensch, P.E.
HSI GeoTrans
3035 Prospect Park Drive, Suite 40
Rancho Cordova, California 95670,
Tel: 916-853-1800
Fax: 916-853-1860
E-mail: jbensch@hsigeotrans.com
State Contact:
Mr. George Lincoln
State of California
Regional Water Quality Control Board
San Francisco Bay Region
1515 Clay Street, Suite 1400
Oakland, CA 94612
Additional Contacts:
Zahra M. Zahiraleslamzadeh
Environmental Project Manager
FMC Corporation
1125 Coleman Avenue, Gate 1 Annex
P.O. Box 58123
Santa Clara, California 95052
Tel: 408-289-3141
Fax: 408-289-0195
E-mail: zahra_zahir@udlp.com
Technology:
Dual Phase Extraction (DPE) with Pneumatic Fracturing System
• 20 dual phase, single pump extraction wells installed at the source area
• 41 fracture locations (two pneumatic fracture points installed between each
pair of extraction wells)
• Following initial fracturing, a low flow/low pressure compressor provided
continuous air injection into each fracture point
• Groundwater extraction rate - approximately 35 gpm on a continuous basis
• Average vapor flow rate - increased from approximately 39 scfm to over 65
scfm, following pneumatic fracturing
Type/Quantity of Media Treated:
Soil and Groundwater
• Depth to groundwater - 8 ft bgs; the first water-bearing zone (A-level aquifer)
present at 20 to 50 ft bgs; second water-bearing zone (B-level aquifer)
present 50 to 90 ft bgs
• Sediments underlying the site include marine or basinal clays, coarse channel
deposits, and inter-channel silts and clays
Regulatory Requirements/Cleanup Goals:
• Less than 10 mg/L total VOCs in soil.
Results:
• The DPE system removed approximately 1,220 pounds of VOCs from the source area
- VOC mass removed by soil vapor extraction - 782 pounds
• Average source area VOC concentration in groundwater declined from over 12,000 ug/L to less than 800 ug/L
• During first month of operation, about 40% of the mass of VOCs removed was from the vadose zone; by the fifth
month, groundwater extraction was removing more VOC mass than SVE
• DPE system shut down June through August 1998 to assess rebound
- VOC concentrations remained relatively constant during shut down and after restart
• 27 confirmation soil samples averaged 0.93 mg/L total VOCs
116
-------
Multi-Phase Extraction at the 328 Site, Santa Clara, CA
Costs:
• The cost to design and install the DPE system with pneumatic fracturing was approximately $300,000.
• Approximate costs for two years of operation and maintenance services, reporting, and analytical fees were $450,000,
averaging $225,000 per year. Approximately $100,000 was required for the disposal of spent carbon.
• The unit cost for treatment of the 0.5-acre source area from 0 to 20 feet bgs was $53 per cubic yard of soil (for
treatment of 16,000 yd3)
Description:
The 328 Site occupies approximately 27.1 acres in a primarily industrial and commercial area of San Jose and Santa
Clara, California, near the San Jose Airport. The 328 Site was used for manufacturing military tracked vehicles, including
assembly and painting operations, from 1963 through 1998. A former waste storage area was the suspected source of
VOC contamination of soil and groundwater at the site. The cleanup of the 328 Site was performed in anticipation of
future commercial/industrial redevelopment and was conducted by FMC Corporation in accordance with the State of
California San Francisco Bay Regional Water Quality Control Board Final Site Cleanup Requirements Order Number 96-
024.
A DPE system, which included 20 dual phase, single pump extraction wells, was used to remove VOCs from silty clay
soils and shallow groundwater at the site Air flow through the soils was enhanced by pneumatic fracturing (PF) between
DPE extraction wells and by supplying continuous low flow/low pressure air to the fractured soils. Over 40 percent of the
VOC mass removal occurred from the vadose zone during the first month of operation. Groundwater extraction provided
greater mass removal rates than soil vapor extraction by the fifth month of operation. The combination of technologies
has allowed soil vapor extraction to be effective in an area that is not well suited for in-situ remediation.
117
-------
Dual Phase Extraction at the Defense Supply Center, Richmond, Virginia
Site Name:
Defense Supply Center, Acid Neutralization Pit (ANP)
Location:
Richmond, VA
Period of Operation:
July 1997 - July 1998
Cleanup Authority:
CERCLA - Remedial Action
• ROD signed 1992
• BSD signed 1995
Purpose/Significance of Application:
Use of DPE to treat soil and groundwater contaminated with chlorinated
solvents, including PCE and TCE
Cleanup Type:
Treatability study
Contaminants:
Chlorinated Solvents
• The highest concentrations of VOCs detected in the upper aquifer were 3300
micrograms per liter Og/L) for PCE, 890 //g/L for TCE, and 26 //g/L for 1,2-
DCE; VOCs were not detected in the lower aquifer
Waste Source:
Leaks from settling basins that
received wastewater from metal
plating operations
Contacts:
Vendor:
Katy L. Allen, P.E.
Law Engineering and Environmental
Services, Inc.
112 Town Park Drive
Kennesaw, GA30144
Tel: (770)421-3400
Regulatory Contact:
Stephen Mihalko
Remedial Project Manager
Virginia Department of Environmental
Quality
P.O. Box 10009
Richmond, VA 23240
Tel: (804)698-4202
Todd Richardson
U.S. EPA Region 3
1650 Arch Street (MC 3HS50)
Philadelphia, PA 19103-2029
Tel: (215)814-5264
E-mail: richardson.todd@epa.gov
Additional Contacts:
Bill Saddington
DSCR Remedial Project Manager
Defense Supply Center Richmond
8000 Jefferson Davis Highway
Richmond, VA 23297-5000
Tel: (804)279-3781
E-mail: bsaddington@dscr.dla.mil
Technology:
Dual Phase Extraction (DPE)
• 12 DPE wells and six air injection wells arranged in a rectangular grid
• DPE wells installed to depth of 22 to 28 ft bgs (10 ft screen length) and
equipped with an electric, submersible (variable-frequency drive) pump,
• SVE vacuum at blower - 42 in WC; SVE air flow rate - 314 cfm
• Groundwater extraction rate - 37 gpm
• DPE radius of influence - 600 to 800 ft, downgradient
• Air extracted by the SVE blower was vented to the atmosphere. Extracted
groundwater was pumped directly to a low-profile tray type air stripper to
remove VOCs. Air stripper off-gas was released to the atmosphere
• Effluent water was discharged to a storm sewer that flows to a nearby stream.
Type/Quantity of Media Treated:
Soil and Groundwater/17 million gallons of groundwater recovered and treated
• The plume area was estimated to be 16,000 square feet
• Depth to groundwater -10 to 15 ft bgs; hydraulic gradient - 0.001 to 0.002
ft/ft; aquifer transmissivity - 374 to 504 ftVd
118
-------
Dual Phase Extraction at the Defense Supply Center, Richmond, Virginia
Regulatory Requirements/Cleanup Goals:
• Remedial goals for PCE - 5 //g/L and TCE - 5 /wg/L, or attainment of an asymptotic trend in contaminant of concern
concentrations in groundwater (whichever occurs first)
• The purpose of the DPE treatability study was to collect additional operational data to refine system design parameters,
and to evaluate the effectiveness of an air injection system to facilitate air flow through soils exposed by drawdown of
the groundwater surface
Results:
• Total VOC concentrations were reduced by more than 99% in several wells; for example, in two wells located in the
plume center initial concentrations of total VOCs were reduced from 1,980 fj,g/L to 11.9 //g/L and from 1,766 //g/L to
3.5//g/L
• Total mass of VOC removed -145 Ib:
- Groundwater VOC mass removal rate - 28 Ib (0.09 Ib/d) total, including 2 Ib (<0.01 Ib/d) aromatic and 26 Ib (0.08
Ib/d) chlorinated
- Soil VOC mass removal rate -117 Ib (0.37 Ib/d) total, including 70 Ib (0.22 Ib/d) aromatic and 47 Ib (0.15 Ib/d)
chlorinated
• At the completion of the treatability study, PCE and TCE concentrations remained above the remedial goals in several
wells, and increasing VOC concentrations were observed in wells at the outer edge of the radius of influence of the
DPE system
Costs:
• The total cost for the one year treatability study of the DPE system was $538,490, including $134,092 for pre-design
investigations supporting DPE design, $73,198 for engineering design of the DPE system, $205,743 in system
construction costs (equipment only), $24,309 in startup costs, and $101,148 in operation and maintenance, which
included the cost of sample collection and analysis
• The total cost per unit volume of groundwater recovered and treated was $0.03 per gallon (based on 17 million gallons
of groundwater)
Description:
The 640-acre Defense Supply Center Richmond (DSCR) is a military support, service, and storage facility located
approximately 11 miles south of the City of Richmond, VA. Since 1942, DSCR has been furnishing and managing
general military supplies to the Armed Forces and several federal civilian agencies. Historic and current industrial
operations at the DSCR have included repair of equipment, engine rebuilding, and refurbishment of combat helmets and
compressed gas cylinders. The Acid Neutralization Pit (ANP) site, located in the northern section of the DSCR, consists
of two former concrete settling basins that received wastewater from metal cleaning operations conducted at one of the
warehouse buildings. In 1985, when the tanks were closed, they were observed to be cracked and broken. Site
investigations determined that the groundwater was contaminated with chlorinated solvents, primarily tetrachloroethene
(PCE) and trichloroethene (TCE). The site was placed on the National Priorities List. A ROD, signed in 1995,
addressed the contamination at the ANP site, and the results of the Feasibility Study identified DPE as a potentially viable
remediation alternative for the site.
A pilot test of the DPE system, along with aquifer testing, was performed in June 1995 to gather site-specific
hydrogeologic data and data on air extraction rates and SVE mass removal rates. The results of the testing supported the
use of DPE for VOC recovery at DSCR. A full-scale system, consisting of 12 DPE wells and six air injection wells were
installed and a treatability study was conducted for one year to evaluate the effectiveness of the full-scale system,
including collecting operational data to refine system design parameters, and to evaluate the effectiveness of the air
injection system. After one year, the DPE system removed 145 pounds of VOCs, including 117 pounds from the soil
vapor and 28 pounds from the groundwater. Although VOC concentrations were reduced in a number of wells, including
reductions of more than 99% in two wells located withing the plume, concentrations of PCE and TCE remained above the
cleanup goals in several wells. Based on the results of the treatability study, the Army's contractor recommended that the
DPE system continue operation and that additional investigations be done to better define the capture zone of the system.
The unit cost was $0.03 per gallon based on 17 million gallons of groundwater treated during the pilot test.
119
-------
Dual Vapor Extraction at Tinkham's Garage Superfund Site, Londonderry, NH
Site Name:
Tinkham's Garage Superfund Site
Location:
Londonderry, NH
Period of Operation:
November 22, 1994 to September 29, 1995
Cleanup Authority:
CERCLA
• ROD signed 1985
• ROD amended March 1989
Purpose/Significance of Application:
Use of D VE to treat soil and groundwater contaminated with chlorinated
solvents, including PCE and TCE
Cleanup Type:
Full scale
Contaminants:
Chlorinated Solvents
• Tetrachloroethene (PCE) and trichloroethene (TCE)
• Site investigations found total VOCs as high as 652 ppm in soil and 42 ppm
in groundwater
Waste Source:
Discharges of liquids and sludge to
surface soils
Contacts:
Vendor:
Joleen Kealey
Project Manager
Terra Vac, Inc.
213 Rear Broadway
Methuen, MA01844
Tel: (978)688-5280
EPA RPM:
James DiLorenzo
U.S. EPA Region 1 (MC:HBO)
One Congress Street, Suite 1100
Boston, MA 02114-2023
Tel: (617)918-1247
E-mail: dilorenzo.jim@epa.gov
Technology:
Dual Vapor Extraction (DVE)
• 33 DVE wells divided into 25 shallow DVE wells, screened in the
overburden, and 8 deep DVE wells, screened in the upper bedrock and
overburden; five existing pilot test wells were left in place and used for vapor
extraction; the wells were distributed over three manifold lines
• SVE vacuum at blower - 5 in Hg (=68 in WC)
• SVE flow rate - 500 scfm, average
• Vapors treated using activated carbon; recovered groundwater treated using
air stripping to meet the Deny POTW pre-treatment standards; off-gas from
air stripper treated using vapor phase carbon
Type/Quantity of Media Treated:
Soil and Groundwater/9,000 cubic yards of soil treated
• Overburden consisting of inorganic and organic silty clay and sand grading to
fine and medium-grained sand; weathered metamorphic bedrock at
approximately 14feetbgs
• Depth to groundwater - 5 to 6 feet bgs
• Hydraulic conductivity - range 1 ft/d to 10 ft/d
Regulatory Requirements/Cleanup Goals:
• ROD specified 1 ppm total VOCs for soil and 5 ppb each for PCE and TCE in groundwater
Results:
• Soil cleanup goals were achieved within ten months of operation; groundwater cleanup goals were not achieved at the
conclusion of DVE system operation and pump-and-treat has been implemented as the site
• Approximately 53 pounds of VOCs were removed by the DVE system:
- vapor extraction removed approximately 48 pounds; averaging 0.17 pounds per day
- groundwater extraction removed approximately 5 pounds of VOCs (recovered in the aqueous phase); averaging 0.016
pounds per day
• The majority of VOCs recovered were PCE and TCE
• VOCs extracted in the vapor phase were reduced from concentrations as high 16 ppm to below 1 ppm (the soil cleanup
goal)
• Concentrations of VOCs in groundwater in the source area decreased by over 99% in one well and by 64% in a second
well. However, total VOCs concentrations in groundwater remained above the cleanup goals and ranged from 29 to
237 ppb in the source area
120
-------
Dual Vapor Extraction at Tinkham's Garage Superfund Site, Londonderry, NH
Costs:
• The actual cost for the project, not including permitting and oversight, was $1.5 million, or $170/cy (based on 9,000 cy
of soil treated).
• This cost includes an adjustment for inflation
Description:
The Tinkham's Garage Superfund site includes 375 acres of residential and undeveloped land in Londonderry, NH. Site
investigations in 1981 found soil and groundwater contaminated with VOCs, including PCE, and TCE, resulting from
unauthorized surface discharges of liquids and sludge in 1978 and 1979. Several source areas were identified at the site
including areas near a condominium complex and a one acre area located behind Tinkham's Garage ("Garage Area").
The original 1985 ROD for the site specified excavation of contaminated soil with onsite treatment by either thermal
aeration, composting, or soil washing. As a result of the pre-design and pilot studies, the ROD was amended in March
1989 to require the treatment of contaminated soil by DVE. For cost purposes, all VOC impacted soil was consolidated
for treatment. Contaminated soil from the various areas at the site was excavated and hauled to the Garage Area, where it
was and spread and compacted in place.
The DVE system consisted of 33 DVE wells divided into 25 shallow DVE wells, screened in the overburden, and 8 deep
DVE wells, screened in the upper bedrock and overburden. Five existing pilot test wells were left in place and used for
vapor extraction. The wells were distributed over three manifold lines to provide the greatest coverage over the area of
contamination. After 10 months of operation, approximately 53 pounds of VOCs were removed by the DVE system, with
SVE removing about 48 pounds and groundwater extraction removing about 5 pounds. The soil cleanup goals were
achieved. VOCs extracted in the vapor phase were reduced from concentrations as high as 16 ppm to below 1 ppm (the
soil cleanup goal). However, total VOCs concentrations in groundwater remained above the cleanup goals. According to
Terra Vac, DVE was not intended to achieve groundwater remediation goals; rather the extraction and treatment of
groundwater was necessary to target and remediate soil contamination located within the saturated zone. A pump and
treat system is currently operating at the site to provide a long term migration control remedy for groundwater. The actual
project cost was $1.5 million, or $170/cy (based on 9,000 cy treated).
121
-------
Frozen Soil Barrier at Oak Ridge National Laboratory, Oak Ridge, Tennessee
Site Name:
Oak Ridge National Laboratory
Location:
Oak Ridge, Tennessee
Period of Operation:
September 1996 to September 1998
Cleanup Authority:
NRC
Purpose/Significance of Application:
Demonstrate frozen soil barrier for containment of contaminated surface
impoundment
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides
• Initial concentrations in sediment included strontium 90 - 75 Curies (Ci) and
cesium 137 -16 Ci
Waste Source:
Nuclear processing operations
Contacts:
Technology Vendor:
Edward Yarmak
Chief Engineer
Arctic Foundations Inc.
(907) 562-2741
Technical Contacts:
Elizabeth Phillips, Principal
Investigator, DOE Oak Ridge
(423)241-6172
Michael Harper, Co-Principal
Investigator, Bechtel Jacobs Company
LLC
(423) 574-7299
Steven Rock, EPA SITE
(513)569-7149
DOE Contact:
Scott McMullin, DOE Savannah River
(803) 725-5608
Technology:
Frozen Soil Barrier
• The demonstration used an array of 50 sealed thermocouples installed around
the perimeter of the impoundment, on 6 ft centers to a depth of approximately
SOftbgs
• The thermocouples were fabricated from 6 inch schedule 40 steel pipe, and
used carbon dioxide as a working fluid, with an above-ground refrigeration
system, to freeze the soil
• The refrigeration system used R-404A, and thermal expansion valves, to
control the amount of freezing
• The frozen soil barrier was established in 18 wks, had a length of 300 linear
ft, depth of 30 ft, thickness of 12 ft, frozen soil volume of 108,000 ft3, and
contained a volume of 168,750 ft3
• A two-part polyurea coating was spray applied over a non-woven geotextile
fabric to prevent surface water from entering the isolated area
• The system was operated first in a freeze-down phase, where the frozen soil
barrier was created; subsequent operation was in maintenance phase
Type/Quantity of Media Treated:
Soil, Sediment, Groundwater
• Depth to groundwater is 2 to 9 ft bgs
• Groundwater discharges to surface water at several locations around the
impoundment
• Complex hydrology due to presence of fractured bedrock
Regulatory Requirements/Cleanup Goals:
• Evaluate performance of the barrier for isolating and containing contaminants
• No specific cleanup goals were identified
Results:
• Performance was evaluated based on groundwater level monitoring, dye tracer studies, and operation tests
• Groundwater level monitoring and dye tracer studies (eosine and phloxine dies) showed hydraulic isolation of the
impoundment
• A 7-day loss of power test showed that the barrier maintained its integrity during a power outage
122
-------
Frozen Soil Barrier at Oak Ridge National Laboratory, Oak Ridge, Tennessee
Costs:
• The actual cost for the demonstration was $1,809,000, consisting of $43,000 for site infrastructure, surveys, and
maintenance; $1,253,000 for system design, fabrication, procurement, installation, and start-up; $274,000 for ORNL
site support; and $239,000 for barrier verification
• A review of projected costs for frozen soil barriers to grouted barriers showed that frozen soil is less costly for initial
installation and operation, with a break-even point of 8 to 9 yrs
Description:
The demonstration site is a former unlined, earthen impoundment used from 1958 through 1961 for retention/settling of
liquid radioactive wastes generated from operation of a Homogeneous Reactor Experiment (HRE) at DOE's Oak Ridge
facility. The impoundment was 75 ft long by 80 ft wide by 10 ft deep, with a capacity of approximately 310,000 gallons.
In 1970, the impoundment was backfilled with local soils, covered with 8 inches of crushed stone, and capped with
asphalt. A 1986 study found that sediments buried in the impoundment contained strontium 90 and cesium 137, and that
groundwater that moved through this area transported contaminants to surrounding locations, including surface waters.
For the demonstration, a frozen soil barrier was constructed using thermocouple technology. The barrier had a length of
300 linear ft, depth of 30 ft, thickness of 12 ft, frozen soil volume of 108,000 ft3, and contained a volume of 168,750 ft3.
Groundwater level monitoring and dye tracer studies showed that the barrier provided for hydraulic isolation of the
impoundment, and a 7-day loss of power test showed that the barrier maintained its integrity during this time. A cost
analysis comparing projected costs for frozen soil barriers to grouted barriers showed that frozen soil barriers are less
costly for initial installation and operation, with a break-even point of 8 to 9 years.
123
-------
Horizontal Wells Demonstrated at U.S. DOE's Savannah River Site and Sandia
National Laboratory
Site Name:
U.S. DOE's Savannah River Site (SRS) and Sandia National Laboratory; Other
Sites (report focuses on use at SRS)
Location:
Aiken, SC, and Albuquerque, NM
Period of Operation:
1988 - 1993
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstrate use of horizontal wells to treat groundwater at multiple sites and
locations
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated solvents
Waste Source:
Multiple sources, including leaks of
solvents
Contacts:
Technical Contacts:
Dawn Kaback
Colorado Center for Environmental
Management
(303) 297-0180 ext. Ill
E-mail: dsdaback@csn.net
Management Contacts:
Skip Chamberlain
DOE EM50
(301) 903-7248
E-mail:
grover. chamberlain@em. doe .gov
James A. Wright
DOE SRS
(803) 725-5608
E-mail: wrightjamesb@srs.gov
Technology:
Pump and Treat (report focuses on installation of horizontal wells above and
below water table)
• Four different systems were used for directional drilling and horizontal well
installation
• A short radius petroleum industry technology was used to install wells at 65
ft bgs and 150-175 ft bgs; these wells were constructed of steel
• A modified petroleum industry technology was used to install two
comparable wells; these wells were constructed of HOPE
• A mini-rig utility industry/compactional tool drilling technology was used to
install a well at 35-40 ft bgs; this well was constructed of fiberglass
• A mini-rig utility industry technology was used to install two wells at 100 ft
bgs; these wells were constructed of PVC
Type/Quantity of Media Treated:
Groundwater (in situ)
• Geology consists of 200 ft of alternating units of permeable sands with low
fines; water table is 120 ft bgs
Regulatory Requirements/Cleanup Goals:
• Test the feasibility of installing horizontal wells in unconsolidated sediments using directional drilling technology
Results:
• Directional drilling technology was used to install a total of seven wells (steel, stainless steel, PVC, HOPE, and
fiberglass) to depths of 35 -175 ft bgs, with horizontal screen sections ranging from 150 - 400 ft
• The wells were used to demonstrate in situ air stripping, in situ bioremediation, and thermally enhanced soil vapor
extraction; four of the wells were later integrated in a vapor extraction remediation system
Costs:
• Costs for horizontal wells vary widely based on drilling method and size of rig, type of drilling tool, drilling fluid,
guidance system, vertical depth, total well length, site geology, well materials, and number of personnel on site
• Costs for installing a PVC or HOPE well using a small to medium sized utility-type drilling rig are projected as $ 164/m
($50/ft)
• Estimated capital costs for horizontal wells were comparable to the capital cost of five vertical wells; O&M costs for
the one horizontal well were less than one-third of the O&M costs for five vertical wells
124
-------
Horizontal Wells Demonstrated at U.S. DOE's Savannah River Site and Sandia
National Laboratory
Description:
This report describes the installation and use of horizontal wells at several DOE sites, including Savannah River Site
(SRS) and Sandia National Laboratories. At SRS, seven wells were installed at depths of 35 -175 ft bgs, with horizontal
screen sections ranging from 150 - 400 ft, and using the following materials: steel, stainless steel, PVC, HDPE, and
fiberglass. The wells were used to demonstrate in situ air stripping, in situ bioremediation, and thermally enhanced soil
vapor extraction; four of the wells were later integrated in a vapor extraction correction action. The SRS demonstration
identified two important factors for consideration during design of horizontal wells: (1) trips in and out of the well bore
should be minimized; and (2) well materials should be adequately flexible to negotiate curves.
At Sandia, several pieces of commercial machinery were tested and evaluated, including the water-assisted Jet Trac
Boring System, the air-assisted True Trac Boring System, the P-80 rod pusher, and the Pierce Arrow pneumatic hammer
tool. Based on the results from initial testing of these machines, construction was begun on a prototype machine, the X-
810.
125
-------
In Situ Chemical Oxi-Cleanse Process at the Naval Air Station Pensacola Florida,
Operable Unit 10, Pensacola, Florida
Site Name:
Naval Air Station Pensacola Florida, Operable Unit 10
Period of Operation:
November 1998 to May 1999
Purpose/Significance of Application:
Field demonstration of in situ chemical oxidation using Fenton's reagent to treat
chlorinated solvents
Contaminants:
Chlorinated Solvents
• TCE primary target for demonstration
• Maximum concentration of TCE 3,600 ug/L
Contacts:
Vendors:
Mattehew Dingens
Geo-Cleanse International, Inc.
4 Mark Road, Suite D
Kenilworth, NJ 07033
Telephone: (908) 206-1250
E-mail: geocleanse@earthlink.net
Site Contact:
Tom Kelly
Public Works Center
NAS Pensacola
Telephone: (850) 452-8236
U.S. Navy Contacts:
Maxie Keisler/Michael Maughon
Naval Facilities Engineering Command
SOUTHNAVFACENGCOM
P.O. Box 190010
2 155 Eagle Drive
N. Charleston, SC 29419
Telephone: (843) 820-7322/7422
Email:
keislermr@efdsouth.navfac.navy.mil;
maughonmj @ef dsouth. navf ac . navy . mil
Mark Stuckey
Hazardous Waste Regulation Section
2600 Blair Stone Road MS 4560
Tallahassee, FL 32399-2400
(850) 921-9246
E-mail: stuckey_m@dep. state. fl.us
Location:
Pensacola, Florida
Cleanup Authority:
RCRA Corrective Action
Cleanup Type:
Field demonstration
Waste Source:
Unlined sludge drying bed
Technology:
In-Situ Chemical Oxidation using Fenton's Reagent
• Geo-Cleanse's patented process for in situ chemical oxidation conducted
in two phases
• Fenton's reagent - hydrogen peroxide (50%) and an equivalent volume
of ferrous iron catalyst
• Phase I injected 4,089 gallons of hydrogen peroxide and similar
volumes of reagents through 14 injection wells at a depth of 10-40 ft bgs
• Phase 2 injected 6,038 gallons of hydrogen peroxide and similar
volumes of reagent through 10 injection wells (9 old, 1 new), totaling
10,127 gallons; phosphoric acid was added to the reagent mix to
stabilize the hydrogen peroxide
• Operating parameters included injection rate of 0.25 - 3 gpm, injection
pressure of 5 - 1 10 psig, pH <8, and CO2 evolution of 5% - >25%
Type/Quantity of Media Treated:
Groundwater
• 16,500 gallons of groundwater in the source area
• Depth to groundwater 0-4 ft; contaminants detected in groundwater 35-
45 ft bgs
• Soil classified as fine to medium quartz sand
• Properties included porosity >15%; pH 3-6; hydraulic conductivity 2-44
ft/day; dissolved iron >500 mg/L
Regulatory Requirements/Cleanup Goals:
• Evaluate effectiveness of in situ chemical oxidation in treating chlorinated solvents
• No specific cleanup goals were identified
126
-------
In Situ Chemical Oxi-Cleanse Process at the Naval Air Station Pensacola Florida,
Operable Unit 10, Pensacola, Florida
Results:
• Phase I reduced TCE concentrations from as high as 3,600 ug/L to 485 ug/L in source area well
• This was considered insufficient reduction; Phase I performance was attributed to elevated concentrations of ferrous
iron in the treatment area, likely due to a historic spill of sulfuric acid
• Phase II reduced TCE concentrations from 460 ug/L to <5 ug/L in source area well
Costs:
• Actual costs for this demonstration, reported by Geo-Cleanse, were $178,338, consisting of $97,018 for capital and
$81,320 for O&M; these costs do not include electrical power or water supply, which were provided by NAS
Pensacola
Description:
Naval Air Station (NAS) Pensacola is a 5,800-acre naval facility located in the western portion of the Florida panhandle.
Operable Unit (OU) 10, is located on 26 acres of Magazine Point Peninsula in the northeast corner of the NAS, was the
site of the former Industrial Wastewater Treatment Plant (IWWTP). The IWWTP treated wastewater from operations
such as painting and electroplating, as well as organic solvents and acids, and included an unlined sludge drying bed. A
groundwater recovery system had been operated for more than 10 years under a RCRA corrective action program to
control migration of contaminated groundwater. In situ chemical oxidation using the Geo-Cleanse patented process was
evaluated for its ability to reduce concentrations of chlorinated solvents in the source area, such that natural attenuation
would be an effective remedy for down-gradient groundwater.
The Geo-Cleanse process used Fenton's reagent (hydrogen peroxide (50%) and an equivalent volume of ferrous iron
catalyst) and was conducted in two phases at OU-10. A total of 10,127 gallons of hydrogen peroxide and similar volumes
of reagents were injected under pressure through 15 wells at a depth of 10-40 ft bgs. Over the two phases, the
concentration of TCE was reduced from 3,600 ug/L to <5 ug/L, as measured in a source area monitoring well. Elevated
concentrations of ferrous iron in the groundwater, due to a historic sulfuric acid spill, limited the effectiveness of the first
phase of injections. In the second phase, phosphoric acid was added to the reagent mix to help stabilize the hydrogen
peroxide in the presence of elevated ferrous iron concentrations. The actual costs for the demonstration were $178,338,
and additional injections were not planned for this site.
127
-------
In Situ Chemical Oxidation Using Fenton's Reagent
at Naval Submarine Base Kings Bay, Site 11, Camden County, Georgia
Site Name:
Naval Submarine Base Kings Bay, Site 11
Location:
Camden County, GA
Period of Operation:
November 1998 through August 1999 (Phase 1 and 2)
Cleanup Authority:
RCRA Corrective Action
Purpose/Significance of Application:
Use of Fenton's Reagent to remediate chlorinated solvents in groundwater
Cleanup Type:
Full scale
Contaminants:
Chlorinated Solvents
• PCE source was 120 feet long by 40 feet wide; 30 to 40 foot horizon below
ground surface (bgs); PCE concentrations in landfill source area detected as high
as 8,500 ug/L
• TCE, DCE, and VC detected at concentrations of more than 9,000 ug/L in
groundwater within the landfill source area
• Because PCE concentrations were as much as 5 percent (%) of the pure solubility
phase, the presence of dense non-phase aqueous liquids (DNAPL) was inferred
Waste Source:
Leaks from a landfill
Contacts:
Vendor:
Matthew M. Dingens, Vice President Sales
J. Daniel Bryant, Ph.D., Senior Geologist
Geo-Cleanse International, Inc.
4 Mark Road, Suite C
Kenilworth, NJ 07033
Telephone: 908-206-1250
Facsimile: 908-206-1251
E-mail: geocleanse@earthlink.net
State Contact:
Billy Hendricks
Compliance Officer
State of Georgia Environmental Protection
Division
205 Butler Street SE, Suite 1162
Atlanta, GA 30334
Telephone: 404-656-2833
E-mail: billy_hendricks@mail.dnr.state.ga.us
Navy Contact:
Clifton C. Casey, P.E.
Southern Division NAVFAC
Environmental Department (Code 18)
P.O. Box 190010
North Charleston, SC 29419-9010
Telephone: 843-820-5561
E-mail: CaseyCC@EFDSOUTH.NAVFAC.
NAVY.mil
Technology:
In Situ Chemical Oxidation; Fenton's reagent
• Geo-Cleanse's patented process for in situ chemical oxidation using
Fenton's reagent
• Fenton's reagent - hydrogen peroxide (50%) and an equivalent
volume of ferrous iron catalyst were delivered via injection to the
subsurface
• Total of 44 injectors - 23 for Phase 1, including deep (42 ft bgs) and
shallow (32 ft bgs) injectors; 21 injectors added for Phase 2, including
deep (40 ft bgs) and shallow (35 ft bgs) injectors
• Phase 1 - two injections of Fenton's reagent into the subsurface,
totaling 12,045 gallons (8,257 gallons November 2-21, 1998; 3,788
gallons February 8-14, 1999) of solution were injected.
• Phase 2 - two injections of Fenton's reagent into the subsurface,
totaling 11,247 gallons (8,283 gallons June 3-11, 1999; 2,964 gallons
July 12-15, 1999)
Type/Quantity of Media Treated:
Groundwater
• Estimated volume of groundwater treated during the Phase Iwas
78,989 gallons (based on a treatment volume of 1,778 cubic yards
and a porosity of 22%)
• Information on volume of groundwater treated during Phase 2 was not
provided
Regulatory Requirements/Cleanup Goals:
• Cleanup goal for the RCRA corrective action at Site 11 was established by the state at 100 ug/L for total chlorinated
aliphatic compounds (CACs), defined as the sum of PCE, TCE, cis-1,2 DCE, and VC concentrations in groundwater
128
-------
In Situ Chemical Oxidation Using Fenton's Reagent
at Naval Submarine Base Kings Bay, Site 11, Camden County, Georgia
Results:
• Phase 1 - after first injection, total CAC concentrations were reduced to below the cleanup goal in five of the seven
monitoring wells, including one well located within the source area where concentrations had been reduced by >97%.
However, total CAC concentrations remained above the cleanup goal in two downgradient monitoring wells; after
second injection, total CAC concentration remained at or above the cleanup goal in the two downgradient wells and
were found to have increased in other wells. As a result, a second phase of treatment was performed
• Phase 2 - after the first injection, total CAC concentrations were reduced to below the cleanup goal in all but one
downgradient monitoring well; however, concentrations increased above the cleanup goal in two downgradient
injectors. After the second injection, total CAC concentrations were reduced to below the cleanup goal in the
downgradient injectors and remained below the cleanup goal in all wells except for the one downgradient well (total
CAC concentration was primarily DCE)
• Sample results from August 1999 showed elevated concentrations of total CACs in one injector located to the east of
the area of concern. The Navy has determined that there is a previously unknown source of contamination in this area
and is addressing the cleanup of the area separate from the Site 11 area of concern. Data on this cleanup were not
available at the time of this report
Costs:
• Total proposed cost for application of in situ chemical oxidation of Fenton's reagent using the Geo-Cleanse process
was approximately $223,000 for Phase 1, including costs for reagents, mobilization, onsite treatment time, injection
and monitoring equipment, documentation, and injector construction oversight and materials
• No additional cost data were provided
Description:
Naval Submarine Base (NSB) Kings Bay, 16,000 acre facility in Camden County, GA, is the U.S. Atlantic Fleet home
port to the next generation of ballistic submarines, and maintains and operates administration and personnel support
facilities. Site 11 is the location of a former 25-acre landfill at NSB Kings Bay, known as the Old Camden County
landfill, that was operated by the county during the mid-1970s to 1980. A variety of wastes from the local Kings Bay
community and the Navy were disposed of in the landfill, including solvents and municipal waste. Site investigations
found the groundwater in the area to be contaminated with PCE, as well as TCE, DCE, and VC. On March 18, 1994,
NSB Kings Bay entered into a Corrective Action Consent Order with the Georgia Environmental Protection Division to
address prior releases of hazardous constituents from Site 11. The Navy selected in situ chemical oxidation using
Fenton's reagent for this site based on its successful use by the U.S. Department of Energy (DOE) in remediating
chlorinated solvent contaminated groundwater at the Savannah River site. The Navy's approach to the cleanup of Site 11
was to use in situ chemical oxidation to reduce groundwater contaminant concentrations in the source area followed by
natural attenuation to address residual contamination.
For the remeditation of Site 11, the Geo-Cleanse® process, a patented in situ chemical oxidation technology using
Fenton's reagent, was used. The Fenton's reagent consisted of hydrogen peroxide (50%) and an equivalent volume of
ferrous iron catalyst that were injected into the subsurface under pressure. The remediation was performed in two phases.
For Phase 1, 23 injectors were installed in and around the area of concern and there were two injections of Fenton's
reagent into the subsurface, totaling 12,045 gallons. During Phase 2, the system was expanded to add 21 injectors and
there were two injections of Fenton's reagent into the subsurface, totaling 11,247 gallons. After two phases of treatment
using the Geo-Cleanse® process, total CAC concentrations had been reduced to below the cleanup goal of 100 ug/L in all
but one well located downgradient of the area of concern. The total CAC concentrations in this well were primarily DCE.
The first phase of treatment (two injections) reduced total CAC concentrations to below the cleanup goal in five of the
seven monitoring wells, including a reduction of >97% in the well located within the source area. Cost data provided by
Geo-Cleanse indicated that the proposed cost for application of in situ chemical oxidation of Fenton's reagent was
approximately $223,000 for Phase 1. No additional cost data were available.
In August 1999, elevated concentrations of total CACs concentrations were found in an injector located to the east of the
area of concern, indicating the presence of an additional contamination source area in the shallow soil. The soil in this
area has been excavated and the Navy is planning to use chemical oxidation to polish the groundwater in this area.
129
-------
Six Phase Heating at the Skokie, Illinois Site
Site Name:
Confidential Manufacturing Facility
Location:
Skokie, Illinois
Period of Operation:
June 4, 1998 to April 30, 1999
Cleanup Authority:
State Voluntary Cleanup
Purpose/Significance of Application:
Use of SPH to remediate chlorinated solvents in soil and groundwater
Cleanup Type:
Full scale
Contaminants:
Chlorinated Solvents
• Primary contaminants included TCE, TCA, and DCE
• Concentrations in groundwater at start of SPH remediation (June 1998) -
TCE (130 mg/L maximum; 54.4 mg/L average), TCA (150 mg/L maximum;
52.3 mg/L average) and DCE (160 mg/L maximum; 37.6 mg/L average)
• DNAPL present
Waste Source:
Leaks from spill contaminant systems
and underground storage tanks
Contacts:
Vendor:
David Fleming, Corporate
Development Leader
Current Environmental Solutions
P.O. Box 50387
Bellevue, WA 98015
Telephone: 425-603-9036
Fax: 425-643-7590
E-mail: david@cesiweb.com
Greg Beyke, Operations Manager
Current Environmental Solutions
1100 Laurel Crest Way
Marietta, GA 30064
Telephone: 770-794-1168
E-mail: greg@cesiweb.com
EPA Contact:
Stan Komperda
Illinois EPA
Bureau of Land, No. 24
1021 East North Grand Avenue
Springfield, IL 62794-9276
Telephone: 217-782-5504
E-mail: epa4207@epa.state.il.us
PRP Oversight Contractor:
Gregory Smith
ENSR
27755 Diehl Rd.
Warrenville, IL 60555
Telephone: 630-836-1700
Technology:
Six-Phase Heating™ (electrical resistive heating combined with soil vapor
extraction)
• Initial network of 107 electrodes (85 beneath the floor of a warehouse
building) operated from June to November 1998; 78 electrodes added (185
total) and operated from December 1998 to April 1999 to treat additional area
of contamination
• Electrodes designed to be electrically conductive throughout a depth interval
of 11 to 21 feet bgs and to increase the subsurface temperature in the depth
interval of 5 to 24 feet bgs to the boiling point of water
• Electrical power input -13.8 kV local service at 1250 kW; 1,775 megawatt
hours (MW-hrs.) consumed from June 4 to November 20, 1998; information
not provided for Dec. 1998/Jan. 1999 through May 1999
• Temperature -100 °C; operating pressure/vacuum - 7.5 inches of mercury
(Hg)
• Network of 37 soil vapor extraction wells, screened to 5 feet bgs, were used
to capture vapors
• Off-gas was condensed and sent through an air stripper prior to discharge to
the atmosphere
Type/Quantity of Media Treated:
Soil and groundwater
• 23,100 cubic yards treated from June to November 1998
• Additional 11,500 cubic yards treated from December 1998 to April 1999
• Soil at site - heterogeneous silty sands with clay lenses to 18 feet bgs
(hydraulic conductivity -10"4 to 10"5 cm/sec); underlain by dense clay till
aquitard (hydraulic conductivity -10"8 cm/sec)
• Depth to groundwater- 7 feet bgs
130
-------
Six Phase Heating at the Skokie, Illinois Site
Regulatory Requirements/Cleanup Goals:
• Tier III cleanup criteria for groundwater; developed by ENSR and approved by Illinois EPA as the cleanup goals for
the site
• Tier III goals were TCE (17.5 mg/L); TCA (8.85 mg/L); and DCE (35.5 mg/L)
• No criteria established for soil
Results:
Results for the remediation of the initial 23,000 cubic yards of contamination:
• By December 1998 (six months of operation), theTier III cleanup goals were achieved for TCE, TCA, and DCE in all
wells in the initial area of contamination
• During this time, average groundwater concentrations were reduced by more than 99% for TCE (54.4 mg/L to 0.4
mg/L); more than 99% for TCA (52.3 mg/L to 0.2 mg/L), and more than 97% for DCE (37.6 mg/L to 0.8 mg/L)
Results for the remediation of the additional 11,000 cubic yards of contamination:
• By April 1999 (five months of operation), theTier III cleanup goals were achieved for TCE, TCA, and DCE in all wells
in the additional area of contamination
• During this time, average groundwater concentrations were reduced by more than 96% for TCE (4.16 mg/L to 0.15
mg/L); more than 92% for TCA (14 mg/L to 1 mg/L); and more than 90% for DCE (2.39 mg/L to 0.24 mg/L)
Costs:
• Cost data were provided on a unit cost basis; total project cost data were not provided
• The unit cost for this technology of $32 per cubic yard is based on a calculated treatment volume of 23,100 cubic yards,
or a treatment area of 26,000 square ft and a depth of 24 ft bgs
• The unit cost for the treatment from December 1998 through May 1999 also was $32 per cubic yard, based on a
calculated treatment volume of 11,500 cubic yards
Description:
The Skokie site is a former electronics manufacturing facility located in Skokie, Illinois. From 1958 to 1988,
manufacturing operations included machining and electroplating. Soil and groundwater at the site was found to be
contaminated with solvents (TCE and TCA), including large pools of dense nonaqueous phase liquids (DNAPL). The site
is being remediated under Illinois' voluntary Site Remediation Program. From 1991 to 1998, steam injection combined
with groundwater and vapor extraction reduced the area of contamination from about 115,000 square feet to about 23,000
square feet. As of early 1998, the remaining area to be remediated represented four source locations where manmade
subsurface features limited the effectiveness of the previously used steam-based remediation system. To complete the
remediation, the site owner selected Six-Phase Heating™ (SPH).
The SPH process operated at the Skokie site from June 4, 1998 to November 20, 1998 to remediate the initial estimated
23,000 cubic yards of contaminated soil and groundwater. Based on the results of sampling conducted in December 1998
that indicated there was a potential for vinyl chloride to be produced outside the initial treatment area at levels in excess of
the cleanup levels, a decision was made to expand the SPH system to cover an additional 11,500 cubic yard treatment
area. The SPH system restarted in December 1998 and operated until April 30, 1999 when cleanup goals were achieved
in the additional area. The unit cost for this technology was $32 per cubic yard for the initial 23,000 cubic yards of
contaminated soil and groundwater and also for the additional 11,500 cubic yards of contaminated media.
131
-------
Hydrous Pyrolysis Oxidation/Dynamic Underground Stripping (HPO/DUS) at
Visalia Superfund Site, CA
Site Name:
Visalia Superfund Site
(report also includes treatment at DOE Portsmouth Site, Piketon, OH)
Period of Operation:
June 1997 to mid-1999
Purpose/Significance of Application:
Use of HPO/DUS for treatment of large quantity of creosote in groundwater
Contaminants:
Semivolatiles - Halogenated and Nonhalogenated
• Creosote and pentachlorophenol were the primary contaminants
Contacts:
Site Contact:
Southern California Edison
Vendor:
Steam Tech Environmental Services
Technical Contact:
Robin Newmark
Lawrence Livermore National
Laboratory
Telephone: (925) 423-3644
E-mail: newmarkl@llnl.gov
EPA Contact:
Kathi Moore
U.S. EPA Region 9
75 Hawthorne Street
San Francisco, CA 94 105
(415) 744-2221
Other Contacts:
James Wright
DOE SRS
Telephone: (803) 725-5608
E-mail: jamesb.wright@srs.gov
Kathy Kauffman
LLNL
Telephone: (925) 422-2646
Location:
Visalia, CA
Cleanup Authority:
CERCLA
ROD - 6/10/94
Cleanup Type:
Field demonstration
Waste Source:
Wood preservation operations
Technology:
Hydrous Pyrolysis Oxidation/Dynamic Underground Stripping
• DUS involved continuous injection of steam and air into permeable zones
over a 5 month period to create a steam front, which swept contaminants from
the injection wells toward extraction wells; when the steam front collapsed,
groundwater reentered the treatment zone and the steam/vacuum extraction
cycle was repeated in a process called "huff and puff
• System used 1 1 injection and 8 extraction wells; steam and air were injected
to 80 - 100 ft bgs in paired wells; average temperature was 60°C (maximum
140°C), with groundwater extracted at 350 - 400 gpm
• Extracted vapors initially were treated with carbon; however, because of the
expense of the carbon, it was replaced with treatment in steam boilers
• Extracted groundwater was treated with filtration and discharged to a POTW
• HPO occurred after the steam and air injection stopped, when groundwater
returned to the heated zone and mixed with oxygen; contaminants were
rapidly oxidized in this environment
• Underground mapping was performed using 29 electrical resistance
tomography (ERT) wells and thermocouples to track the steam fronts and
heated areas
Type/Quantity of Media Treated:
Groundwater
• Three distinct water-bearing zones are present; shallow aquifers from 35 to 75
ft bgs, an intermediate aquifer from 75 to 105 ft bgs, and a deep aquifer below
120 ft bgs; the HPO/DUS system targeted the intermediate aquifer
Regulatory Requirements/Cleanup Goals:
• Evaluate the performance of DUS/HPO for removing creosote
132
-------
Hydrous Pyrolysis Oxidation/Dynamic Underground Stripping (HPO/DUS) at
Visalia Superfund Site, CA
Results:
• During 25 months of operation, a total of 1,130,000 Ibs (141,000 gals) of creosote were removed or treated (10,400
Ibs/wk)
• Approximately 50% of the contaminants were removed in free phase, 16% as vapors, 16% in an aqueous phase, and
17% destroyed by HPO in situ
• Monitoring the progress of the heating fronts showed that all the aquifer was treated
Costs:
• A comparison of projected costs for use of HPO/DUS and pump and treat at Visalia showed that HPO/DUS would have
larger capital and annual O&M costs, but would be operated for less years, than pump and treat; projected unit costs
were $39/yd3 for HPO/DUS and $110/yd3 for pump and treat
• Key factors affecting the cost analysis include the groundwater extraction capacity and size of plume
Description:
Since the 1920's, the four-acre Visalia Poleyard was the site of a wood preservation treatment plant for power poles. Poles
were dipped into creosote, a pentachlorophenol compound, or both. Soil and groundwater to 100 ft bgs were contaminated
with creosote, pentachlorophenol, and diesel fuel. A pump and treat system was installed in 1975 and several years later a
slurry wall was constructed to contain the plume at its leading edge.
A field demonstration of Hydrous Pyrolysis Oxidation/Dynamic Underground Stripping (HPO/DUS) was conducted at
Visalia over a 25-month period. HPO/DUS is a combination of technologies including steam and air injection, vapor
extraction, pump and treat, and electrical resistance tomography. The system used at Visalia consisted of 11 injection and
8 extraction wells; steam and air were injected to 80 -100 ft bgs in paired wells. Groundwater was extracted at 350-400
gpm. During the 25 months of operation, atotal of 1,130,000 Ibs (141,000 gals) of creosote were removed ortreated
(10,400 Ibs/wk). Approximately 50% of the contaminants were removed in free phase, 16% as vapors, 16% in an aqueous
phase, and 17% treated by HPO in situ.
133
-------
Intrinsic Remediation at AOCs 43G and 43J, Fort Devens, Massachusetts
Site Name:
Areas of Concern (AOCs) 43G and 43 J
Location:
Fort Devens, Massachusetts
Period of Operation:
Intrinsic remediation assessment (IRA) - 3/97 to 6/99
Long-term monitoring -12/99 to 12/11 (AOC 43G) and 12/04 (AOC 43 J).
End dates are estimated.
Cleanup Authority:
CERCLA and State
Record of Decision (ROD) signed on
October 17, 1996
Purpose/Significance of Application:
This project demonstrates that intrinsic remediation is a viable treatment
alternative at sites contaminated with BTEX.
Cleanup Type:
Full scale
Contaminants:
Organic Compounds
• Volatiles (nonhalogenated)
• BTEX (benzene, toluene, ethylbenzene, and xylene)
• Maximum benzene concentrations:
- 2,OOOmg/LatAOC43G
- 300 mg/L at AOC 43J
Waste Source:
Leaks and spills from former gasoline and
waste oil USTs.
Contacts:
Project Management:
Mark Applebee
USAGE, New England Division
696 Virginia Road
Concord, MA 01742-2751
mark.r.applebee@usace.army.mil
Jim Chambers
BRAC Environmental Coordinator
Devens Reserve Forces Training Area
30 Quebec Street
Devens, MA 01432-4429
(978)796-3114
ChambersJ@devens-emhl .army.mil
Vendor:
Gina Nyberg
Stone & Webster Environmental
Technology & Services
245 Summer Street
Boston, MA 02210
(617) 589-2527
gina. nyberg@stoneweb. com
Regulatory Points of Contact:
Jerry Keefe
USEPA, Region 1
1 Congress St., Suite 1100
(Mailcode HBT)
Boston, MA 02114-2023
(617) 918-1393
Keefe.Jerry@epamail.epa.gov
John Regan
MADEP
627 Main Street
Worchester, MA 01605
(978) 792-7653
John.Regan-EQE@state.ma.us
Technology:
Intrinsic Remediation
• Remediation approach requires a demonstration, through intensive site
characterization, that natural biological processes are destroying
contaminants in situ and that the site will reach specified remediation
goals within 30 years
• The demonstration includes:
- Observation of a stable or decreasing contaminant plume over time
- Correlation of contaminant plumes with electron acceptor distribution
- Modeling studies that indicate attenuation due to processes other than
dispersion, volatilization, and sorption
• Eight quarterly sampling rounds were conducted to accumulate the data
necessary for the remediation demonstration
• Annual long-term monitoring is required to confirm that adequate
remediation is occurring
Type/Quantity of Media Treated:
• The contaminant plume at AOC 43G extends 320 feet downgradient
from the source area and is 230 feet wide. The contaminant plume at
AOC 43J extends 250 feet downgradient from the source area and is 190
feet wide. Plume dimensions were calculated based on groundwater
concentrations above the maximum contaminant level (MCL) for
benzene in March 1997
• The aquifer is approximately 5 feet thick at AOC 43G and 10 feet thick
at AOC 43J
• Free product has been detected
• Electron acceptors are present in the groundwater at varying levels
134
-------
Intrinsic Remediation at AOCs 43G and 43J, Fort Devens, Massachusetts
Regulatory Requirements/Cleanup Goals:
• The ROD established the preliminary remediation goals (PRGs) for AOCs 43G and 43 J that must be met within 30
years; most goals were based on MCLs.
• Property boundary performance standards for AOCs 43 G and 43 J were based on the PRGs and the Massachusetts
Contingency Plan (MCP) GW-1 standards for extractable and volatile petroleum hydrocarbons (EPH/VPH)
Results:
• The results of the Mann-Kendall statistical trend analyses on BTEX compounds at both sites indicated that groundwater
concentrations exhibit a statistically significant decreasing trend
• At both sites, there is significant evidence of the utilization of electron acceptors and the appearance of degradation
products, suggesting that contaminants are being biologically degraded and not just physically diluted or dispersed
• Modeling indicates that the contaminant plumes at both sites will be reduced below the applicable MCLs between 8 and
15 years after the ROD was signed
• Fate and transport modeling demonstrated that it was unlikely that the BTEX plumes would move off of Army property
Costs:
• The total cost for the IRA was $671,642
• The anticipated long-term monitoring and reporting costs are $50,000 per year
• The number of wells sampled is a significant cost element because it effects the duration of field sampling events,
analytical expenses, and the effort involved with tracking and assessing data
Description:
AOCs 43G and 43J are two former gasoline stations operated at Fort Devens. These sites were also used for motor pool
operations during World War II. BTEX and TPH contamination in soil and groundwater at these sites is consistent with
the historical use of the areas. The Army determined that intrinsic remediation was the most appropriate remedy for the
contamination at both sites. The remedy consists of intrinsic remediation, IRA data collection and groundwater modeling,
long-term groundwater monitoring and annual reporting, and five-year site reviews
The IRAs for AOCs 43G and 43J demonstrated that intrinsic remediation is working and that the Army will not need to
initiate additional cleanup actions. Specifically, modeling indicates that the concentrations of the contaminants of concern
will be below groundwater cleanup levels in less than 30 years and that they will not migrate off of Army property
135
-------
Monitored Natural Attenuation at Keesler Air Force Base, Mississippi
Site Name:
Keesler Air Force Base (AFB), Base Exchange Service Station, Area of Concern
- A (ST-06)
Location:
Biloxi, Mississippi
Period of Operation:
September 1997 to April 1999
Cleanup Authority:
EPA Region 4 and Mississippi DEQ
Purpose/Significance of Application:
Monitored natural attenuation for a gasoline contaminated site
Cleanup Type:
UST cleanup
Contaminants:
BTEX, Lead
• Soil concentrations measured as high as 166 mg/kg for BTEX and 8.7 mg/kg
for lead
• Groundwater concentrations measured as high as 22,400 ug/L for BTEX and
21ug/Lforlead
Waste Source:
Gasoline USTs and associated piping
Contacts:
Vendor:
John Hicks
Parsons Engineering Science, Inc.
1700 Broadway, Suite 900
Denver, CO 80290
(303)831-8100
john.hicks@parsons.com
Site Contact:
Lisa Noble
81stCES/CEVR
508 L Street
Keesler AFB, MS 39534-2115
(601)377-5803
noblel@ces.kee.aetc.af.mil.
Air Force Contact:
Jim Gonzales
AFCEE/ERT
3207 North Rd., Building 532
Brooks AFB, TX 78235-5363
(210) 536-4324
james.gonzales@hqafcee.brooks.af.mil
EPA Contact:
Robert Pope
USEPA Region 4
61ForsythSt, SW
Atlanta, GA 30303-3104
(404) 562-8506.
State Contact:
Bob Merrill
Mississippi DEQ
P.O. Box 10385
Jackson, MS 39289-0385
(601)961-5171
Technology:
Monitored Natural Attenuation
• Bioventing and density-driven convection in-well aeration were used
previously as source control measures
• Monitoring of 9 groundwater wells planned for five years
• Samples will be analyzed for aromatic volatile organics and geochemical
parameters
Type/Quantity of Media Treated:
Soil, groundwater, and soil gas
• Source area plus dissolved plume covers approximately 4.0 acres
- Fine- to medium-grained sand to 20 ft bgs, underlain by a clay layer of
unknown thickness
- Groundwater present at 5 to 9 ft bgs
- Average hydraulic conductivity of sand zone is 40 ft/day
- Calculated horizontal groundwater flow velocity is 0.8 ft/day
136
-------
Monitored Natural Attenuation at Keesler Air Force Base, Mississippi
Regulatory Requirements/Cleanup Goals:
• Cleanup levels for BTEX was 100 ppm for soil and 18 ppm for groundwater
• Risk-based screening levels for lead was identified as 400 ppm in soil and 15 ug/L in groundwater
• OSHA PELs were used as screening levels for BTEX in soil gas
Results:
• In February 1998, the only contaminant in soil to exceed the cleanup levels was BTEX (1 sample at 166 mg/kg); BTEX
(1 sample at 22.4 mg/L) and lead (3 samples - 21, 21 and 16 ug/L) exceeded the cleanup levels in groundwater. Only
lead in groundwater was identified as a chemical of potential concern for this site
• Data from 1988 to 1998 indicated substantial oscillation in dissolved BTEX concentrations at the plume core since May
1993, but that the total BTEX plume appears to have been relatively stable
Costs:
• The estimated O&M cost for long-term monitoring was identified as $15,000 per event
Description:
In 1987, 10 USTs were removed from the Keesler Air Force Base, in Biloxi, Mississippi. During the removals, there was
evidence that one or more of the tanks had leaked, and site investigations found gasoline components in the soil and
groundwater, including BTEX and lead. A bioventing system was installed in 1993 and operated for three years. A
density-driven convection (DDC) in-well aeration system was installed in 1996 and operated at least through February
1998. Based on a RBCA analysis, the recommended final remedial action was monitored natural attenuation. The
recommendation was based on the finding that the site contamination does not currently (and will not in the future) pose a
significant risk to potential receptors, the dissolved plume is stable and degrading, and institutional controls can be
maintained with a high level of confidence. The RBCA analysis showed that concentrations of target analytes in all
sampled media do not exceed applicable MDEQ RBSLs or OSHA PELs, and that detected concentrations of total lead in
groundwater do not pose a risk to potential receptors.
Geochemical data indicated that biodegradation of fuel hydrocarbons is occurring at the site, primarily via the anaerobic
processes of sulfate reduction, nitrogen fixation, and methanogenesis. Previous and current source removal efforts have
reduced hydrocarbon concentrations in vadose zone and saturated zone soils, and the current system does not have an
adverse effect on the natural attenuation processes at the site. A long-term monitoring plan was negotiated with the MDEQ
and USEPA Region 4 that included monitoring of nine wells for five years. Monitoring will occur quarterly for the first
year and annually for the second through fifth years. The purpose of the monitoring is to verify the effectiveness of
naturally-occurring remediation processes at limiting plume migration and reducing dissolved contaminant concentrations.
137
-------
Monitored Natural Attenuation at Kelly Air Force Base , Former Building 2093
Gas Station, Texas
Site Name:
Kelly Air Force Base (AFB), Former Building 2093 Gas Station
Period of Operation:
July 1997 to July 1998
Purpose/Significance of Application:
Monitored natural attenuation for a gasoline-contaminated site
Contaminants:
Gasoline constituents
• BTEX concentrations in groundwater measured as high as 2,807 ug/L in
November 1997
Contacts:
Vendor:
John Hicks
Parsons Engineering Science, Inc.
1700 Broadway, Suite 900
Denver, CO 80290
(303)831-8100
john.hicks@parsons.com
Site Contact:
Jerry Arriaga
SA-ALC/EMRO
301 Tinker Dr., Suite 2
Bldg. 301
Kelly AFB, TX 78241
(210) 925-1819
garriaga@emgatel.kelly.af.mil.
Air Force Contact:
Jim Gonzales
AFCEE/ERT
3207 North Rd., Building 532
Brooks AFB, TX 78235-5363
(210) 536-4324
james.gonzales@hqafcee.brooks.af.mil.
State Contact:
Antonio Pena
TNRCC
P.O. Box 13087
Austin, TX, 78711-3087
(512)239-2200
APENA@tnrcc.state.tx.us.
Technology:
Location:
Kelly AFB, Texas
Cleanup Authority:
Texas Natural Resource Conservation
Commission
Petroleum Storage Tank Division
Cleanup Type:
UST cleanup
Waste Source:
Leaking gasoline USTs and associated
piping
Monitored Natural Attenuation
• Monitoring network not described
Type/Quantity of Media Treated:
Soil, groundwater, and soil gas
• Source area plus dissolved plume covers 1.5 acres
• The site is underlain by silty clay; with a distinct clay unit from 35
to\ A -A- 1~
40 it DgS
• Groundwater occurs primarily in silt and possibly caliche seams that
produce only small amounts of water; static groundwater levels
range from 5 to 25 feet bgs, depending on location and season
• Hydraulic conductivity of the silty clay unit is 0.2 to 0.5 ft/day based
on slug tests, and the estimated horizontal eroundwater flow
velocity is 3 1 ft/year
Regulatory Requirements/Cleanup Goals:
• TNRCC Plan A target concentrations for Category II aquifers, and TNRCC target concentrations for construction
worker exposure are the cleanup goals for affected groundwater
138
-------
Monitored Natural Attenuation at Kelly Air Force Base , Former Building 2093
Gas Station, Texas
Results:
• Based on a Tier 1 screening, only the Plan A concentration for benzene of 0.0294 mg/L was exceeded, and benzene in
groundwater and soil was identified as a contaminant of potential concern
• Fate and transport modeling using the analytical code BIOSCREEN indicated that the maximum migration distance of
dissolved benzene from the source area will be approximately 300 ft, and that dissolved benzene concentrations will be
below groundwater quality standards within 10 years
• Results of groundwater sampling events indicated that the dissolved contaminant plume is not increasing in areal extent,
and that natural attenuation indicator parameters exhibit trends associated with a plume that is being naturally degraded.
• The site was identified as a candidate for immediate closure according to TNRCC guidance
• The Air Force will restrict use of the shallow groundwater at the site until all dissolved benzene concentrations decrease
below TNRCC Plan A Category II criterion of 0.0294 mg/L
• Maximum-detected concentrations of BTEX in soil gas were compared to the chemical-specific OSHA 8-hour time-
weighted average permissible exposure limits (PELs), and there were no exceedences
Costs:
Not provided
Description:
As a result of UST integrity testing in 1989, the former Building 2093 Gas Station at Kelly Air Force Base, in Texas, was
found to be leaking, and the UST and associated piping were removed in 1991. Site investigations found BTEX
contamination in the groundwater. A 1-year-long bioventing pilot test was concluded in January 1995; the test results
indicated that site soils were not sufficiently permeable to enable use of this in situ source reduction technique. Later in
1995, the dispensing islands and remaining below-grade piping were removed, and 2,750 cubic yards of soil in the area of
the former tank pad and dispensing islands were excavated. Based on a RBCA analysis, the TNRCC issued a no-further-
action memorandum closing the site based on plume stability, the occurrence of natural attenuation of fuel residuals, and
the conclusion that site contamination will not pose a significant risk to potential receptors.
139
-------
In Situ Permeable Reactive Barriers for Contaminated Groundwater
at Fry Canyon
Site Name:
Fry Canyon
Location:
Southeastern Utah
Period of Operation:
September 1997 - ongoing (performance data for first year of demonstration -
September 1997 to September 1998)
Cleanup Authority:
Not applicable
Purpose/Significance of Application:
Field demonstration of three types of PRBs to treat uranium-contaminated
groundwater
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides (uranium) and metals
• Uranium concentrations in groundwater found at levels as high as 16,300 ug/L
• Iron and manganese concentrations found in groundwater at 90 ug/1 and 180
ug/L, respectively
Waste Source:
Subsurface drainage from abandoned
uranium ore mill ponds
Contacts:
EPA Contact:
Ed Feltcorn
U.S. EPA/ORIA
Ariel Rios Building
1200 Pennsylvania Avenue, N.W.
Washington, D.C. 20460
Telephone: 202-564-9422
Fax: 202-565-2037
E-mail: feltcorn.ed@epa.gov
USGS Contact:
David Naftz, Ph.D.
U.S. Geological Survey
2329 West Orton Circle
West Valley City, UT 84119-2047
Telephone: 801-908-5053
Fax: 801-908-5001
E-mail: dlnaftz@usgs.gov
Technology:
Permeable Reactive Barriers (PRBs)
• Three types of PRBs demonstrated - phosphate (PO4), zero valent iron (ZVI),
and amorphous ferric oxyhydroxide (AFO)
• PRBs installed side-by-side and operated concurrently
• Funnel and gate design; each PRB was keyed, along with each of the
impermeable funnels, into the bedrock (Cedar Mesa Sandstone formation)
beneath the colluvial aquifer
• 1.5-foot layer of pea gravel on the upgradient side of the PRBs to facilitate
uniform flow of groundwater into the PRBs
• "As built" volume of reactive material was: PO4 - 67.2 ft3; ZVI - 77.7 ft3, and
AFO - 67.2 ft3
• Each PRB contains a total of 22 monitoring wells, configured in two parallel
"rows" - Row 1 and Row 2
• Estimated range of groundwater velocity through PRBs - 0.2 - 2.5 ft/day
Type/Quantity of Media Treated:
Groundwater - 33,000 cubic feet (about 200,000 gallons)
• Depth to groundwater - 8 feet bgs
• Colluvial aquifer ranges in depth from 2-5 feet
• Groundwater flow rate - 0.2-2.5 ft/day
• Transmissivity -10-200 ft/day
• Hydraulic conductivity - 55-85 ft/day
Regulatory Requirements/Cleanup Goals:
The objective of the demonstration project is to evaluate the use of three types of PRBs in controlling the migration of
uranium and metals in groundwater
Results:
• Performance data were available for the first year (September 1997 to September 1998) of this ongoing demonstration
• The ZVI PRB showed the best removal rate of the three PRBs tested, removing more than 99.9% of the uranium from
the groundwater
• The PO4 PRB initially removed more than 99% of the uranium from the groundwater, with the removal rate decreasing
to 60-70% in January 1998, then increasing to 92% as of September 1998. Available results from tracer tests indicated
that there was no leakage from the ZVI PRB to the PO4 PRB; rather, the increased efficiency in the PO4 PRB is the
result of anoxic conditions caused by the release of PO4
• The AFO PRB had the lowest removal rate, consistently removing less than 90% of the uranium from the groundwater;
with removal rates as low as 37% observed
140
-------
In Situ Permeable Reactive Barriers for Contaminated Groundwater
at Fry Canyon
Costs:
• The cost for the PRB demonstration included $280,000 for site selection, characterization, and PRB material testing;
$148,000 for design of the PRBs; and $246,000 for the installation of the PRBs
• O&M costs were reported as being relatively expensive because of the extensive monitoring performed for the
demonstration compared to full-scale operation. Projected costs for full-scale O&M for a comparable site were
estimated to be $55,000-$60,000 per year
Description:
Fry Canyon, located in southeastern Utah (approximately 60 miles west of Blanding, Utah), is the site of an abandoned
uranium ore milling operation and copper leach operation. From 1957 to 1960, COG Minerals Corporation conducted
uranium upgrading (concentrating) operations at the site, and from 1962 to 1968, the Besinare Company conducted copper
leach operations. Waste from these operation, including tailings, were stored and disposed of at the site. The Utah
Department of Health, Bureaus of Radiation Control and Solid and Hazardous Waste, conducted site visits to Fry Canyon
in 1984 and 1986. Elevated levels of uranium were found in water samples from Fry Creek. The site was selected by the
U.S. Environmental Protection Agency (EPA) in cooperation with the U.S. Geological Survey (USGS), the U.S.
Department of Energy (DOE), BLM, and the Utah Department of Environmental Quality, for a field demonstration of
PRBs to assess their performance in removing uranium from groundwater.
Prior to constructing the PRBs, extensive laboratory investigations were conducted to evaluate the various reactive
materials for each type of PRB and to select the specific reactive materials for the Fry Canyon demonstration. Three types
of PRBs were demonstrated - phosphate (PO4), zero valent iron (ZVI), and amorphous ferric oxyhydroxide (AFO). The
PRBs were constructed side-by-side to allow all three types of materials to be evaluated during the demonstration period.
A funnel and gate design was used and each PRB was keyed into bedrock beneath the colluvial aquifer at the site. After
one year of operation, the ZVI PRB showed the best performance, consistently removing more than 99% of the uranium
from the groundwater. The next best performance was observed for the PO4 PRB. While the removal rate for the PO4
PRB varied throughout the year, decreasing to as low as 62%, as of September 1998, the uranium removal rate for the PO4
PRB at the end of one year of operation was greater than 92%. The AFO PRB initially removed greater than 90% of the
uranium from the groundwater, but dropped to as low as 37% after the first year of operation.
Several problems were encountered during installation of the PRBs. For example, a large bedrock nose was encountered
that caused the PRBs to be rotated such that groundwater entered into the gate structures at an oblique angle rather than
perpendicular, as designed. To prevent this problem for other applications, a more detailed view of the bedrock
topography would be needed during site characterization. Full-scale cost considerations include potential lower costs for
design and operation compared to the demonstration costs, which included three PRBs and a more extensive monitoring
system than would be needed for a non-research application.
141
-------
Permeable Reactive Wall Remediation of Chlorinated Hydrocarbons in
Groundwater at Moffett Field Superfund Site
Site Name:
Naval Air Station, Moffett Field Superfund Site
Location:
Mountain View, CA
Period of Operation:
April 1996 - December 1997
(Monitoring data available June 1996 through December 1997)
Cleanup Authority:
Installation Restoration Program
Purpose/Significance of Application:
Field demonstration of PRB to remediate groundwater contaminated with
chlorinated solvents
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated Solvents
• Groundwater contaminated with chlorinated volatile organic compounds
(CVOCs) including TCE, cis-l,2-DCE, PCE, and 1,1-DCE; TCE is the most
prevalent contaminant at the site
• CVOC plume, located in the near surface A aquifer, is more than 10,0000 feet
long and about 5,000 feet wide
• TCE and PCE concentrations in the A aquifer reported above 20 mg/L and
0.5 mg/L, respectively
Waste Source:
Wastes from operations and waste
management activities, including leaks
from underground storage tanks,
aboveground tanks, and sumps
Contacts:
Navy Contractor:
Arun Gavaskar
Battelle
505 King Avenue
Columbus, OH 43201
614-424-3403
Navy Contacts:
Charles Reeter
Naval Facilities Engineering Service
Center
1100 23rd Avernue
Port Hueneme, CA 93043
805-982-4991
Stephen Chao
U.S. Navy, EFA West
900 Commodore Drive
San Bruno, CA 94066
Technology:
Permeable Reactive Barrier (PRB)
• Funnel-and-gate system; pea gravel added to gate to help distribute
groundwater flow through reactive cell
• PRB is 10 feet long (6 feet of reactive material) by 10 feet wide; installed at
depth from +5 feet bgs to -14 feet bgs; keyed into low-permeability sediments
(sand channel)
• Reactive material - iron (from Peerless Metal Products, Inc.); -8 to +40 mesh
particle size range
• Groundwater monitoring well network includes wells within the PRB as well
as upgradient and downgradient
Type/Quantity of Media Treated:
Groundwater
• The aquifer includes two units - Al which is up to 20 feet thick and is
overlain by a clayey surface layer of varying thickness; and A2 which is up to
20 feet thick and extends to 40 feet below mean sea level
• Aquifer contains multiple channels of sand and gravel; zone is not laterally
homogenous due to the interbraided channel nature of the sediments
• Both units are contaminated; however, the pilot-scale PRB penetrates the Al
unit only
• Al unit - hydraulic gradient ranges from 0.005 to 0.009; hydraulic
conductivity ranges from 0.04 foot/day to 633 feet/day (due to lithographic
variation); groundwater velocity ranges from 0.2 to 5.0 feet/day
Regulatory Requirements/Cleanup Goals:
• Groundwater cleanup goals are the MCLs for PCE (5 mg/L), TCE (5 mg/L), cis-l,2-DCE (70 mg/L), and vinyl chloride
(2 mg/L), as measured in the effluent from the PRB
142
-------
Permeable Reactive Wall Remediation of Chlorinated Hydrocarbons in
Groundwater at Moffett Field Superfund Site
Results:
• The PRB monitored on a quarterly basis from June 1996 to October 1997 (five quarters total)
• By October 1997, TCE, PCE, DCE, and VC were reduced to below the MCLs in the effluent from the PRB
• Data from two wells located within the reactive cell (one upgradient; one downgradient) were used to analyze trends in
TCE and DCE degradation:
- TCE concentrations in both wells remained below the MCL every quarter except for June 1996; possible reasons for
the elevated TCE levels in June 1996 included adsorption-desorption on the iron surfaces and residual contamination
from construction activities attributed to the recent installation of the PRB (April 1996)
- DCE concentrations in both wells remained below the MCL for all five quarters
- Over the five quarters, TCE concentrations were relatively constant in both wells
- There was wider variation in DCE concentrations between the two wells; lower DCE concentrations were observed in
the downgradient well, indicating that DCE degraded more slowly than TCE in the reactive medium.
Costs:
• The total cost associated with the treatment of groundwater during the pilot-scale PRB demonstration was $802,375,
including $652,375 in capital costs and $150,000 in O&M costs
• The projected capital cost for a full-scale PRB at Moffett Field was $4,910,942. O&M costs for a full-scale system
were projected to be $72,278 in annual monitoring costs and $267,538 in barrier maintenance costs, incurred once every
ten years, to replace part of the iron medium
• The projected full-scale costs assumed that the PRB would be constructed in two sections - the first section to capture
and treat the groundwater; the second section, constructed downgradient from the leading edge of the plume to control
further migration of the plume; both sections would extend to the base of the A2 aquifer zone, a depth of about 65 feet
Description:
The Naval Air Station, Moffett Field, located in Mountain View California, was selected by the U.S. Navy as part of the
Installation Restoration Program for a field demonstration of a PRB. Groundwater at Moffett Field is contaminated with
chlorinated solvents, and the site was placed on the National Priorities List in 1987. An area known as the West Side
Plume, a chlorinated solvent plume (primarily TCE) located on the west side of Moffett Field, was used for the
demonstration. Based on the results of laboratory testing, iron from Peerless Metal Powders was selected for the PRB.
The pilot-scale PRB, installed in April 1996, was a funnel-and-gate design, keyed into low-permeability sediments. The
PRB was operated through October 1997, with groundwater monitored quarterly from June 1996 through October 1997
(five quarters total). By October 1997, TCE, PCE, DCE, and VC were reduced to below the cleanup goals in the effluent
from the PRB. Additional data for TCE and DCE collected from wells located within the reactive cell showed that TCE
and DCE concentrations within the PRB were generally below the MCLs, and that DCE degraded more slowly in the
reactive cell than TCE. The projected cost for a full-scale PRB at Moffett Field was $4,910,942 in capital costs and
$72,278 in annual monitoring costs. In addition, the projected O&M costs included $267,538 in barrier maintenance costs
for iron medium replacement, incurred once every ten years.
143
-------
Groundwater Extraction and a Permeable Reactive Treatment Cell at Tacony
Warehouse, Philadelphia, Pennsylvania
Site Name:
Tacony Warehouse (TW)
Location:
Philadelphia, Pennsylvania
Period of Operation:
May 13, 1998 through 2001 (projected)
Cleanup Authority:
CERCLA and State
Record of Decision (ROD) signed on
July 21, 1995
Purpose/Significance of Application:
This project demonstrates that an extraction well that is surrounded by permeable
reactive media (iron filings) is a viable treatment alternative at sites contaminated
with chlorinated solvents.
Cleanup Type:
Full scale
Contaminants:
Organic Compounds
• Volatiles (halogenated)
Maximum concentrations: 4,214 mg/L PCE, 579 mg/L TCE,
2,800 mg/L cis-l,2-DCE, 64.6 mg/L trans-1,2-DCE, 2,000 mg/L vinyl chloride
Waste Source:
The source of chlorinated solvents in
the groundwater is not known.
Contacts:
Project Management:
Russ Marsh
USACE, Baltimore District
10 S. Howard Street
Baltimore, MD 21201
(410) 962-2227
russell.e.marsh@nab02.usace.army.mil
Vendor:
Bob Manzitti
Radian International
7101 Wisconsin Ave., Suite 700
Bethesda, MD 20814
(301) 280-2601
bob_manzitti@radian.com
Regulatory Points of Contact:
Mark Stephens
US EPA, Region 3
1650 Arch Street
Philadelphia, PA 19103-2029
(215) 814-3353
stephens.mark@epamail.epa.gov
Christopher Falker
PADEP
Lee Park Suite 6010
555 North Lane
Conshohocken, PA 19428
(610) 832-5930
Technology:
• Pump and treat using a permeable reactive treatment cell
• Three extraction wells are being used to remove groundwater at the site. The
system extracts an average of 3 gallons per minute
• The Tacony Treatment Cell or TTC is located near the monitoring well with
the highest VOC concentrations (MW-9). The TTC is four feet in diameter
and is filled with 22 tons of zero-valent iron filings around a four-inch
diameter extraction well. The thickness of the iron filings layer was
calculated to provide a 10 hour detention time
• Zero-valent iron reacts with the chlorinated hydrocarbons to form less-
chlorinated and non-chlorinated hydrocarbons
• EW-1 and EW-2 are six-inch extraction we 11s with no reactive media. They
were located to influence the hydraulic capture zone
• Extracted groundwater is discharged to the City of Philadelphia sanitary sewer
system
Type/Quantity of Media Treated:
• Before treatment began, an area in the vicinity of MW-9 was contaminated in
addition to an area approximately 300 feet downgradient of MW-19
• During the first year of operation, approximately 1.8 million gallons were
extracted from the aquifer beneath the site, of which 393,165 gallons were
treated by the TTC
• The contaminated aquifer is between 8 and 35 feet below ground surface
(bgs). The aquifer can be described as heterogeneous and anisotropic, with
hydraulic conductivities ranging from 2.3 to 29.4 gal/day/ft2
Regulatory Requirements/Cleanup Goals:
• PADEP established the groundwater remediation goal of achieving background levels, which are based on the analytical
quantitation limits of EPA SW-846 Test Method 8240. The remediation targets are 5 mg/L for PCE, TCE, and DCE
and 10 mg/L for vinyl chloride
• The City of Philadelphia does not allow water to be discharged to the sewer system at concentrations exceeding 2.13
mg/L of total toxic organics
144
-------
Groundwater Extraction and a Permeable Reactive Treatment Cell at Tacony
Warehouse, Philadelphia, Pennsylvania
Results:
• The TTC is demonstrating conversion of PCE and TCE to less-chlorinated hydrocarbons when compared to untreated
groundwater at MW-9, which is located approximately 15 feet away. PCE and TCE were not detected at the TTC,
however, intermediate reaction products (cis-l,2-DCE and vinyl chloride) were observed
• Three of the six target monitoring wells are meeting the remedial standards and a fourth well met the standards in April
1999 but exceeded these levels in June 1999
• The sewer discharge meets the City of Philadelphia limit on total toxic organics
Costs:
The total project cost was $607,336, which includes the capital costs ($416,777), one year of operation and maintenance
($16,880), and other related costs ($132,417)
Description:
The TW site is located on 14.2 acres of land adjacent to the Delaware River in northeast Philadelphia. The site was
constructed and established as an armor plate assembly facility in 1943. The site was used for warehousing operations
from the 1950s through 1992, when the site was vacated. During this time, there were several periods of inactivity and
numerous changes in accountability for the site.
Site investigations at the TW site indicate that the groundwater in several areas is contaminated with chlorinated solvents
and that soil contamination around MW-9 may be a potential ongoing source of contamination. Use of barriers
constructed from zero-valent iron has been demonstrated to be an effective treatment method at other sites contaminated
with chlorinates solvents. At TW, groundwater in the vicinity of MW-9 is drawn through a bed of iron filings surrounding
an extraction well. As the groundwater passes through the bed, it is treated through reductive dehalogenation reactions.
The treated water is combined with untreated groundwater from two other on-site extraction wells and is discharged to the
city sanitary sewer.
Results from the first year of operation indicate that reductive dehalogenation reactions are occurring, but not to
completion. Permeable reactive extraction wells are applicable for many sites, especially where contamination is migrating
off-site. In these cases, the hydraulic control provided by pumping may be necessary or the installation of an interceptor
wall may not be feasible.
145
-------
This Page Intentionally Left Blank
146
-------
DEBRIS/SOLID MEDIA TREATMENT ABSTRACTS
147
-------
Direct Chemical Oxidation at Lawrence Livermore National Laboratory
Livermore, California
Site Name:
Lawrence Livermore National Laboratory (LLNL)
Location:
Livermore, California
Period of Operation:
Not identified
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Pilot-scale demonstration of the DCO process to treat a variety of organic
aqueous waste streams
Cleanup Type:
Field demonstration
Contaminants:
Chlorinated solvents, PCBs, kerosene, explosives, ion exchange resins
• Solvents - TCE, PCE, methylene chloride, chloroform
• 2,4,6-TNT, triethylamine, ethylene gylcol
Waste Source:
LLNL waste streams or surrogates
containing chlorinated solvents
Contacts:
Product Line Manager:
Vince Maio, Advisory Engineer
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
P.O. Box 1625
Idaho Falls, ID 83415
Telephone: 208-526-3696
Fax: 208-526-1061
E-mail: vmaio@inel.gov
Principal Investigator:
Dr. John Cooper
Chemistry and Materials Science
Directorate, L-352
LLNL
P.O. Box 808
Livermore, CA 94550
Telephone: 925-423-6649
Fax: 925-422-0049
E-mail: cooper3@llnl.gov
Technology:
Direct Chemical Oxidation (DCO)
• Nonthermal, low temperature, ambient pressure, aqueous-based technology
used to oxidize organic compounds in hazardous and mixed waste streams to
carbon dioxide and water
• Oxidizing agent - sodium or ammonium peroxydisulfide
• Five continuously stirred tank reactors (CSTRs) - pretreatment, feed, and
three-stage oxidizer (15L each)
• Hydrolysis used a pretreatment step for highly volatile wastes - for
demonstration, hydrolysis used in tests of PCB waste streams only
• Operating temperature - hydrolysis - < 150 ° C; oxidation - 90 ° C
• Oxidation rate - about 200-kg (as carbon) per cubic meter of reactor per day
• Tests conducted on several types of waste streams including concentrated
waste streams (2,4,6-TNT, kerosene, triethlyamine, Dowex - an ion exchange
resin, ethylene glycol), kerosene(predominately dodecane), chlorinated
solvents (PCE, TCE, methylene chloride, chloroform and a mix of PCE and
chloroform), and low concentrations (45 ppm) of PCBs
• The tests included oxidation and destruction rates for concentrated waste
streams; oxidation time profile for kerosene; oxidation of chlorinated solvents
without hydrolysis pretreatment; and treatment of PCB waste both with and
without hydrolysis pretreatment
Type/Quantity of Media Treated:
Waste streams from LLNL operations
Regulatory Requirements/Cleanup Goals:
• The purpose of the demonstration was to evaluate the DCO process on a variety of organic waste streams, including
concentrated waste streams, under varying conditions
• No specific goals were established for the demonstration
148
-------
Direct Chemical Oxidation at Lawrence Livermore National Laboratory
Livermore, California
Results:
• Concentrated waste streams - the oxidation rate (based on Ka of 0.02-0.04 min' and input concentration of 5N oxidant)
and the destruction rate were calculated for the concentrated waste streams; the oxidation rate was considered to be a
rough estimate for CSTR scaling; oxidation rates ranged from 132 kg/m3/day (TNT and Dowex) to 432 kg/mVday
(ethylene glycol); destruction rates ranged from >98.8 (TNT and triethylamine) to >99.97% (kerosene)
• Kerosene- the oxidation rate profile showed a rapid destruction of kerosene following the addition of the oxidant at
90°C - 99.97% within the first 70 minutes, followed by a slower destruction rate during the reminder of the test, with a
destruction rate of 99.99% after 140 minutes
• Chlorinated solvents - results showed that chlorinated solvents are readily oxidized by the process, without pretreatment.
Data reported on the extent of oxidation after 1 hr ranged from 0.967 to 0.996; however, the pretreatment step avoids
the need to pressurize the oxidation step to avoid entrainment of the volatile solvents in the CO2 offgas
• PCBs - results showed that very dilute solutions of PCBs can be treated to below detection limits by the process, both
with and without pretreatment; little difference was observed with and without pretreatment; pretreatment was
determined not to be necessary since PCBs are not volatile
Costs:
• Projected costs for a full-scale DCO process were calculated for a 50 kg/day plant operating at an 80% capacity factor;
costs were estimated for two scenarios - recycling the expended oxident and not recycling
• If recycled, the projected cost is $9.88/kg of carbon in the waste, including the cost of electrical energy ($2.63), labor
($3), and capital cost ($1.92) plus profit and G&A (30%)
• If not recycled, the projected cost is $79/kg of carbon in the waste based on the equivalent weights of sodium
peroxydisulfate (119 g/equivalent) and carbon (3 g/equivalent), a bulk cost for sodium peroxydisulfate ($0.73/lb), and
an assumed 80% stoichiometric efficiency
Description:
In 1992, researchers at LLNL began developing the DCO process, a nonthermal, low temperature, aqueous based
technology, for use in mixed waste treatment, chemical demilitarization and decontamination, and environmental
remediation. A pilot-scale demonstration of the DCO process was conducted on a number of waste streams including
concentrated wastes such as TNT, kerosene, triethlyamine, ion exchange resins, and ethylene glycol; chlorinated solvents
such as TCE, PCE, methylene chloride, and chloroform; and low concentrations of PCBs in solution. The pilot-scale DCO
process included a pretreatment (hydrolysis) step, used for highly chlorinated volatiles and a three-stage oxidation process
performed in 15L reactors.
The results of the pilot-scale testing showed that the DCO process can treat a variety of organic waste streams. The
destruction rate for the concentrated wastes was >98%, chlorinated solvents were readily oxidized using the three-stage
oxidation only (without hydrolysis), and concentrations of PCBs were reduced to below detection levels both with and
without pretreatment. According to LLNL, further research is not needed before scale-up of the technology, however,
treatability studies are recommended for each candidate waste stream. Considerations in selecting DCO to treat a waste
stream include the matrix and physical properties of the waste, waste composition and characteristics, and the target degree
of oxidation/destruction removal efficiency.
149
-------
Acid Digestion of Organic Waste at Savannah River Site, Aiken, South Carolina
Site Name:
Savannah River Site
Period of Operation:
1996 to 1997
Purpose/Significance of Application:
Demonstrate acid digestion of organic wastes as an alternative to incineration
Contaminants:
Organic wastes and simulated radioactive wastes; no specific contaminants
identified
Contacts:
Principal Investigator:
Robert A. Pierce
Westinghouse Savannah River Co.
P.O. Box 616, Bldg. 773A, Rm. C-137
Aiken, SC 29802
Telephone: (803) 725-3099
E-mail: robert.pierce@srs.gov
DOE Contact:
William Owca
U.S. DOE Idaho Operations Office
850 Energy Drive
Idaho Falls, ID 83401-1563
Telephone: (208) 526-1983
Fax: (208) 526-5964
E-mail: owcawa@id.doe.gov
Location:
Aiken, South Carolina
Cleanup Authority:
Not identified
Cleanup Type:
Bench and pilot scale
Waste Source:
Nuclear processing operations
Technology:
Acid Digestion Process
• Process consists of an oxidation vessel, acid recycle and offgas treatment
system, and acid stabilization and waste immobilization system
• Organic destruction takes place in oxidation vessel; waste is added to a bath
of 14. 8M phosphoric acid containing 0.5 to l.OM nitric acid
• The vessel is heated to 150 to 200°C under pressure of 0 to 20 psig
• Bench-scale tests were conducted in units with 2-5 L capacity and pilot-scale
tests in a 40 L glass reactor
Type/Quantity of Media Treated:
Organic wastes
• Cellulose (240 gms of KimWipes™),
neoprene, polyethylene, and PVC
Regulatory Requirements/Cleanup Goals:
• Determine applicable organic wastes for technology, and related operating conditions
• No specific cleanup goals were identified
Results:
• Tests were conducted on cellulose, neoprene, polyethylene, and PVC
• Tests on cellulose showed that 240 gms of KimWipes™ were oxidized to CO2 and H2O in 70 mL of acid and residual
phosphoric acid was stabilized, providing for a volume reduction of 50 to 100 fold
• Tests showed that dissolution time for organic wastes depends on the type of waste, temperature, pressure, and acid
concentration
• The dissolution rate for mixtures of waste types will be limited by the PVC dissolution rate, even when PVC is present
in small quantities
Costs:
• Projected costs for full-scale acid digestion systems are under preparation, but were estimated to range from $2,000,000
to $8,000,000 for design, construction, and demonstration
150
-------
Acid Digestion of Organic Waste at Savannah River Site, Aiken, South Carolina
Description:
Bench- and pilot-scale tests of an Acid Digestion system were conducted at DOE's Savannah River Site in 1996 and 1997.
This technology was tested using job control wastes - organic waste forms consisting of materials such as cellulose,
neoprene, polyethylene, and PVC. Acid Digestion is one of several Alternative Oxidation Technologies (AOT) under
consideration by SRS for treatment of their plutonium 238 contaminated job control wastes.
Acid Digestion consists of dissolution of organic materials in a solution of nitric acid in phosphoric acid, and is conducted
at operating conditions of 150 to 200°C and 0 to 20 psig. Tests were conducted on cellulose, neoprene, polyethylene, and
PVC, and showed that dissolution time for organic wastes depended on the type of waste, temperature, pressure, and acid
concentration. Further, tests showed that the dissolution rate for mixtures of waste types will be limited by the PVC
dissolution rate, even when PVC is present in small quantities. Because the process involves the use of nitric acid,
controlling the reaction is an important safety consideration. Issues associated with monitoring the oxidation rate and
water content need to be resolved for full-scale deployment of the technology.
151
-------
Remotely Operated Scabbling at Argonne National Laboratory-East
Argonne, Illinois
Site Name:
Argonne National Laboratory-East
Location:
Argonne, Illinois
Period of Operation:
Not identified
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstration of a remotely-operated scabbier to decontaminate radioactive
concrete flooring
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides
• Beta/gamma radiation
Waste Source:
Nuclear processing operations
Contacts:
Technical Contacts:
Linda Lukart-Ewansil
Pentek, Inc
412-262-0725
pentekusa@aol.com
Susan Madaris
Florida International University
305-348-3727
madariss@eng.flu.edu
DOE Contact:
Richard Baker
DOE, Chicago Operations Office
630-252-2647
richard.baker@ch. doe. gov
Technology:
Remotely-Operated Scabbier
• Pentek, Inc. Moose® scabbier
• Consists of three subsystems - scabbling head assembly, on-board, high-
efficiency paniculate (HEPA) vacuum system, and six-wheeler chassis;
remote operation performed using a small control panel attached to the
scabbier by a tether (50-ft used for demonstration)
• Scabbling head - seven 2 1/4-in diameter reciprocating scabbling bits, each
with a 9-point tungsten carbide-tip capable of delivering 1,200 hammer
impacts/min
• HEP A vacuum system - two-stage positive filtration system that deposits
waste into an on-board 23-gal drum
• Chassis - independent skid steering for 360-degree rotation
• During demonstration - average rate of scabbling - 130ft2/hr for a 2-person
crew
Type/Quantity of Media Treated:
Debris (concrete floor)
Regulatory Requirements/Cleanup Goals:
• The objectives of the demonstration were to evaluate the remotely-operated scabbier for concrete flooring contaminated
with beta/gamma radiation
Results:
• During the demonstration, the scabbier removed an average of 1/8-inch concrete from 620ft2 of the concrete floor
• Contamination levels (total beta/gamma radiation) reduced from a maximum of 105,000 dpm/100 cm2 to 3,500 dpm/100
cm2
• Waste generated - 37ft3 mix of powder and small chips of paint and concrete
Costs:
• Costs for the Pentek Moose® - $165,000 equipment cost; $l,995/day labor rate (two trained operators); and $2,400 for
replacement parts
• For the cost analysis, the Pentek Moose® was compared to a baseline technology of manual scabbling, using the
demonstration area (620ft2) and a hypothetical job size of 2,500 ft2 (area requiring one week of effort)
• The Pentek Moose® was more expensive than the baseline technology for the smaller area; but was comparable to the
baseline technology for the larger area
• The report includes a detailed analysis of the effect of labor rates, equipment transportation costs, waste disposal costs,
and other factors on the cost of the technology
152
-------
Remotely Operated Scabbling at Argonne National Laboratory-East
Argonne, Illinois
Description:
The Pentek Moose® is a remotely-operated scabbier used to scarify concrete floors and slabs. A demonstration of the
technology was conducted at the Argonne National Laboratory-East, CP-5 Reactor on a floor area (620ft2) contaminated
with beta/gamma radiation. The Moose® includes a head assembly, on-board, high-efficiency paniculate (HEPA) vacuum
system, and six-wheeler chassis. The scabbier is operated remotely using a small control panel attached to the scabbier by
a tether, 50 to 300 ft in length. A 50-ft tether was used for the demonstration.
A two-person crew, one person to operate the scabbier and one to manage hoses and cords, removed an average of 1/8 in
concrete from an area of 620ft2 or at a rate of 130ft2/hr. Total beta/gamma radiation levels were reduced from a maximum
of 105,000 dpm/100 cm2 to 3,500 dpm/100 cm2 following the demonstration. Approximately 37ft3 of waste was generated
by the scabbling, consisting of a mixture of powder and small pieces of paint chips and concrete. The cost analysis showed
that a number of factors affect the cost of the remotely-operated scabbier compared to the baseline of manual scabbling,
including labor rates, costs to transport equipment, and waste disposal. The system is commercially available; however,
several design improvements were suggested based on the results of the demonstration including eliminating the need for a
second operator, increasing the size of the waste drum from 23-gal to 55-gal, and adding a second vacuum connection to
the rear of the unit to collect small pieces of debris.
153
-------
Soft Media Blasting at the Fernald Site, Fernald, Ohio
Site Name:
Fernald Site
Location:
Fernald, OH
Period of Operation:
August 19 - September 5, 1996
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstration of soft blast media to clean surfaces contaminated with uranium
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides
• Enriched uranium (1.34 wt-%U-235)
• Contaminant levels of 18,000 dpm/100 cm2 measured prior to demonstration
Waste Source:
Residue from enriched uranium
processing operations
Contacts:
Vendor Contact:
Edward Damien
AEA Technologies, Inc.
13245 Reese Blvd, #100
Huntsville, NC 28078
704-875-9573
Technical Contacts:
Larry Stebbins
Fluor Daniel Fernald
513-648-4785
larry.stebbins@fernald.gov
Steve Bossart
Federal Energy Technology Center
304-285-4643
sbossa@fetc.doe.gov
Technology:
Soft Media Blasting
• Compressed air is used to propel soft blast media through a hose onto the
contaminated surface; soft media traps and absorbs contaminants on impact
• Air compressor - minimum requirements (250 ftVmin of air; 120 psi line
pressure at the feed unit); for demonstration- 375 ftVmin, 150 psi
• Feed unit - contains media mixture; connected to a hose (1 1/4-in. diameter;
25-ft long) fitted with a venturi-style tungsten carbide blast nozzle (3/8 in and
1/2 in nozzles tested during demonstration)
• Blast pressure - 45 psi; media flow - 20-25 Ibs
• Six grades of media available (color-coded by grade); two grades of media
were tested - green media containing no abrasive; brown media containing
Starblast® abrasive
• Demonstration involved cleaning a settling tank contaminated with enriched
uranium process residue
Type/Quantity of Media Treated:
Debris (concrete)
Regulatory Requirements/Cleanup Goals:
• Performance objectives included cleaning effectiveness (based on amount of residual radioactivity) and production rate
• Evaluate the technology for use in cleaning radioactive-contaminated surfaces
Results:
• Radiation levels were below the minimum detectable count rate (MDCR) following the demonstration
• Production rate was 92 ftVhr; rate was slower than expected - worker time was limited to 1 hr/day because of the noise
generated by the system (106 to 113 dB)
• Brown media was effective on thick dirt; brown media generated more dust than the green media
Costs:
• Demonstration cost for soft media blasting - $4.60/ft2
• Projected full-scale costs are comparable to baseline technology (high-pressure water washing) for an area of 900ft2 or
larger
154
-------
Soft Media Blasting at the Fernald Site, Fernald, Ohio
Description:
A field demonstration of Soft Media Blasting Technology (SMBT) was performed at the Fernald Site to evaluate the
capability of the technology for cleaning radioactively-contaminated surfaces. SMBT uses compressed air to propel soft
blast media onto the contaminated surface, with the soft media trapping and absorbing contaminants on impact. Six grades
of media are available for the SMBT, manufactured by AEA Technologies, Inc. For the demonstration, two grades were
tested - one containing no abrasive and one containing the Starblast® abrasive. A settling tank contaminated with enriched
uranium process residue was used for the demonstration.
The results of the demonstration showed that the SMBT reduced radiation levels from 18,000 dpm/100 cm2 to MDCR.
The production rate of 92 ft2/hr was slower than the baseline technology of high-pressure washing. Because the system
was noisy, the time an individual could work was limited. The demonstration cost for soft media blasting was $4.60/ft2,
more expensive than the baseline technology. However, the projected full-scale costs for SMBT are comparable to the
baseline technology for an area of 900ft2 or larger. Issues associated with full-scale implementation include the noise level
produced by the system and improving the ergonomic design of the nozzle/hose assembly to make it less awkward to use.
While the media was not recycled during the demonstration, a unit (Classifier Unit) can be added to the system for this
purpose. The decision to not recycle the media during the demonstration was based on a concern that the feed and
classifier units would not be successfully decontaminated following repeated recycling of the contaminated media.
155
-------
Concrete Grinder at the Hanford Site, Richland, Washington
Site Name:
Hanford Site
Location:
Richland, WA
Period of Operation:
November 1997
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstration of a light weight hand-held grinder to decontaminate radioactive
concrete surfaces
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides
• Beta/gamma radiation
Waste Source:
Operation of a nuclear research reactor
Contacts:
Technical Contacts:
Stephen Pulsford, BHI, 509-373-1769
Greg Gervais, USAGE, 206-764-6837
DOE Contacts:
JohnDuda, FETC, 304-285-4217
Jeff Bruggeman, DOE-RL, 509-376-7121
Technology:
Concrete Grinding
• Flex Model LD 1509 FR hand-held concrete grinder (6 Ibs)
• 5-in diamond grinding wheel (10,000 rpm)
• 1.25-in. vacuum port for dust extraction
• Powered by 110 VAC, 11 amps
Type/Quantity of Media Treated:
Debris (concrete) - 54ft2
Regulatory Requirements/Cleanup Goals:
The objectives of the demonstration were to evaluate the capability of a light weight, hand-held grinder in removing
concrete
Results:
• Removed concrete from 54ft2 of walls and floors in the demonstration area to a depth of 1/16 in. at a rate of 48ft2/hr
• Contamination levels following demonstration were below free-release levels
Costs:
• The costs for the Flex LD 1509 FR concrete grinder are - $649 equipment cost plus $205 for a replacement diamond
grinding wheel; grinder can be rented for $25/day or $75/week
• The cost for the hand-held grinder were 40% less than the baseline technologies (scalier and scabbier)
Description:
The Flex concrete grinder is a lightweight, hand-held unit used to remove concrete and coatings from concrete surfaces.
The electric powered grinder is equipped with a diamond grinding wheel and a vacuum port for dust extraction. The
grinder was demonstrated on walls and flooring at the C reactor that were contaminated with beta/gamma radiation.
During the demonstration, the grinder removed concrete to a depth of 1/16 in from a total area of 54ft2. At the end of the
demonstration, radioactivity levels were below free-release levels. The Flex grinder was compared to two baseline
technologies - scabbier and sealer. The Flex grinder was found to be easier to use, more flexible, and more efficient that
the baseline technologies, and overall to cost about 40% less. However, the life of the grinding wheel (manufacturer
recommended change after 500ft2 at a depth of 1/16 in and the cost of a replacement wheel ($205) should be factored into
the decision to use the technology. No specific changes or modifications to the grinder are needed for full-scale
deployment.
156
-------
This Page Intentionally Left Blank
157
-------
Concrete Shaver at the Hanford Site, Richland, Washington
Site Name:
Hanford Site
Location:
Richland, WA
Period of Operation:
November 1997
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstration of a concrete shaver to decontaminate radioactive concrete
surfaces
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides
• Beta/gamma radiation
Waste Source:
Operation of a nuclear research reactor
Contacts:
Vendor Contact:
Ian Bannister
Marcrist Industries Limited
+44 (0) 1302 890888
Technical Contacts:
Stephen Pulsford, BHI, 509-373-1769
Greg Gervais, USAGE, 206-764-6837
DOE Contacts:
JohnDuda, FETC, 304-285-4217
Glenn Richardson, DOE-RL, 509-376-
7121
Technology:
Concrete Shaver
• Marcrist Industries Limited Model DTF25 concrete shaver
• Electric-powered, serf-propelled, walk behind concrete and coating removal
system
• 10-in. wide diamond impregnated shaving drum with 5-in. blades; vacuum
port for dust extraction
• Weighs 330 Ibs; requires 380-480 volt, 3-phase power; minimum 16 amps
• Variable cutting depth up to 0.5 in.; can reach to within 3 in. of wall/floor
interface or obstruction
• Demonstrated on radioactive-contaminated concrete floor
Type/Quantity of Media Treated:
Debris (concrete)
Regulatory Requirements/Cleanup Goals:
The objectives of the demonstration were to evaluate the capability of the shaver in removing contaminated concrete
surfaces
Results
• Removed concrete from 816 ft2 of floor space in the demonstration area to a depth of 1/8 in. at a rate of 128 ftVhr
• Contamination levels following demonstration were below free-release levels:
Costs:
• The costs for the Marcrist Industries Limited Model DTF25 concrete shaver are - $10,700 equipment cost plus $7,161
for a set of replacement blades (100 blades)
• Unit cost of $ 1.32/ft2, assuming a rate of 128 ftVhr
• The cost for the shaver is 50% less than the baseline technology (scabbier)
Description:
The Marcrist Industries Limited Model DTF25 concrete shaver is an electric-powered, serf-propelled, walk behind system
used to remove concrete and coatings from concrete surfaces. The electric powered shaver is equipped with a diamond
impregnated shaving drum and a vacuum port for dust extraction. The shaver was demonstrated on concrete flooring in
two rooms at the C reactor that were contaminated with beta/gamma radiation.
During the demonstration, the shaver removed concrete to a depth of 1/8 in from a total area of 816ft2. At the end of the
demonstration, radioactivity levels were reported to be below free-release levels. The shaver was compared to the baseline
technology - scabbier - and was found to be as much as five times faster, produce less worker fatigue, and save 50%
compared to the baseline technology. The shaver requires the use of a HEP A filtration system and is designed to work on
floors, but not walls. No specific changes or modifications to the shaver are needed for full-scale deployment.
158
-------
This Page Intentionally Left Blank
159
-------
Concrete Spaller Demonstration at the Hanford Site, Richland, Washington
Site Name:
Hanford Site
Location:
Richland, WA
Period of Operation:
January 16 - 27, 1998
Cleanup Authority:
Not identified
Purpose/Significance of Application:
First demonstration of the hand-held concrete spaller on contaminated surfaces
Cleanup Type:
Field demonstration
Contaminants:
Radionuclides
• Beta and gamma radioactivty
Waste Source:
Nuclear processing operations
Contacts:
Technical Contacts:
Stephen Pulsford, BHU, 509-375-
4640
Mark Mitchell, PNNL, 509-372-4069
Gregory Gervais, USAGE, 206-764-
6837
DOE Contacts:
Glenn Richardson, 509-372-9629
Shannon Saget, 509-372-4029
Technology:
Concrete Spaller
• Hand-held unit weighing about 30 Ibs
• Components include spalling bit, removable metal shroud, hydraulic cylinder
rated at 9 tons, and hydraulic pump rated at 10,000 psi
• Pre-drill holes in surface (2.5-cm diameter) in a honeycomb pattern
• Spaller bit inserted into hole, the hydraulic valve opened causing bit to
expand and breaking off a chunk of concrete; concrete chunks were collected
in the metal shroud
• A water spray was used to control dust emissions during the demonstration
Type/Quantity of Media Treated:
• Debris - 4.6m2
• Contaminated concrete walls and floors
Regulatory Requirements/Cleanup Goals:
• The objectives of the demonstration were to evaluate the capabilities and design features of the concrete spaller for
removing contaminated concrete surfaces
• No specific cleanup goals were identified
Results:
• During the demonstration, the concrete spaller removed concrete from an area of 4.6 m2 to a depth of 3 mm to 50 mm;
the removal rate was 1.3m2/hr
• Pre-drilling was relatively slow; however, faster drills are available for this step
• Little dust was generated by the spaller
Costs:
• Operating costs for the demonstration were about 22% higher than the baseline technology (scabbier and sealer) because
of the problems encountered with the drill (slower than expected and inexperienced crew)
• For the cost analysis, operating costs were estimated for an improved concrete spaller technology (adequate drill and
experienced crew) - $128/m2, assuming a depth of 3-mm
• Operating costs for the improved spaller are 15% less the costs for the baseline tools (sealer at $155/nf and scabbier at
$156/m2)
160
-------
Concrete Spaller Demonstration at the Hanford Site, Richland, Washington
Description:
The concrete spaller, developed by the Pacific Northwest National Laboratory, is a hand-held tool used for
decontaminating concrete surfaces. The spaller includes a 9-ton hydraulic cylinder and a patented spalling bit that is run
by a 10,000 psi hydraulic pump. Holes are drilled into the concrete in a honeycomb pattern and the spaller bit inserted into
each hole. The hydraulic valve is opened, expanding the bit, and the concrete is removed in chunks up to 2 inches thick
and collected in a metal shroud attached to the spaller. The unit can be used on flat or slightly curved concrete walls and
floors, and can be equipped with a vacuum filtration unit for paniculate control.
The concrete spaller was demonstrated at DOE's Hanford site in Richland, WA on two wall areas in the fan room of the C
Reactor facility. The walls were contaminated with beta/gamma radioactivity. During the demonstration, the spaller
removed 4.6 m2 of contaminated surface to a depth of 3 mm to 50 mm, which was deeper than the baseline technologies
(sealer and scabbier). The operating cost of the spaller under optimal conditions is $128/m2, which is less than the costs
for the baseline tools (sealer at $155/nf and scabbier at $156/m2). Considerations for future development and use of the
technology include the need for a simplified design or manufacturing technique for the spalling bit (which was found to be
fairly difficult to manufacture), the addition of a water spray nozzle to the drill to eliminate the need for a second worker to
manually apply water during drilling, and the additional of an automatic hydraulic control valve.
161
-------
Stabilization Using Phosphate Bonded Ceramics at Argonne National Laboratory,
Argonne, Illinois
Site Name:
Argonne National Laboratory
Location:
Argonne, IL
Period of Operation:
Not identified
Cleanup Authority:
RCRA and NRC
Purpose/Significance of Application:
Demonstration of phosphate-bonded ceramics to stabilize a variety of high salt-
containing wastes
Cleanup Type:
Development tests
Contaminants:
Metals
• Oxide forms of cadmium, chromium, lead, mercury, and nickel were added to
the waste stream at concentrations of 1,000 mg/kg each
Waste Source:
Surrogate waste streams containing
high levels of nitrate salts and chloride
and sulfates similar to those found at
DOE facilities
Contacts:
Principal Investigator:
Arun S. Wagh, Ph.D.
Argonne National Laboratory
9700 South Case Ave
Argonne, IL 60439
Telephone: 630-252-4295
Fax: 630-252-3604
E-mail: arunwagh@gmgate.anl.gov
MWFA Product Line Manager:
Vince Maio, Advisory Engineer
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
P.O. Box 1625
Idaho Falls, ID 83415
Telephone: 208-526-3696
Fax: 208-526-1061
E-mail: vmaio@inel.gov
Technology:
Stabilization using phosphate bonded ceramics
• 50/50 blend of magnesium oxide and monopotassium phosphate powder
mixed with water, additives, and waste
• Mixed for 20-30 minutes; waste form set for 2 hours, then cured for 14 days
• Initial testing performed to determine effects of different test scenarios on
waste forms conducted on surrogate salt solutions and on surrogate salt waste
streams containing activated carbon and ion exchange resins
- Salt solutions - saturated solutions of NaNO3 (50-wt%) and NaCl (10-
wt%); RCRA metals (Cd, Cr, Pb, and Hg) added at 5,000 mg/kg each;
additives included 50-wt% Class-F fly ash and l-wt% K2S to tie up Hg
- Salt waste streams with activated carbon and ion exchange resins - mix
included nitrate, sulfate, and chloride salts (30%), Na2CO3, and CsCl (to
simulate a radioactive component)
• Based on results, additional tests were performed on two salt surrogates - one
containing a high quantity of nitrate salts (58-wt%); the other high quantities
of chloride and sulfates (70-wt%); RCRA metals (Cd, Cr, Pb, Hg, and Ni)
added at 1,000 mg/kg each;
• Waste forms tested for density, compressive strength, and flammability
(nitrate wastes)
Type/Quantity of Media Treated:
Salt-containing waste streams
Regulatory Requirements/Cleanup Goals:
RCRA Land Disposal Restriction (LDR) standards and NRC guidelines
• Universal Treatment Standards (UTS) for metals
• NRC leach index of 6; compressive strength of 500 psi
162
-------
Stabilization Using Phosphate Bonded Ceramics at Argonne National Laboratory,
Argonne, Illinois
Results:
Waste forms from salt solutions of NaNO3 (50-wt%) and NaCL (10-wt%):
• Densities of 1.8 g/cm3 and 1.72 g/cm3, respectively and compressive strengths of 1,800 psi and 3,500 psi, respectively
• Passed the UTS standards for metals, with the exception of Cd; attributed to the less acidic conditions of the test (pH 4)
that slowed reaction of Cd with the phosphate; Cd was fully stabilized in subsequent tests at lower pH levels
• Marginally passed leach index criteria with leach levels of 6.86 and 6.7, respectively, indicating slow salt leaching;
additional binding or coating techniques may be needed to prevent salt leaching from deteriorating the waste
Waste forms from salt solutions containing activated carbon and ion exchange resins:
• For the 60-wt% and 70-wt% loadings - had densities of 1.24 g/ml and 1.32 g/ml and compressive strengths of 2,224 psi
and 5,809 psi, respectively
• Passed the UTS standards for metals
MWFA salt surrogates:
• Had densities in the range of 1.7-2.0 g/cm3 and compressive strength in the range of 1,400-1,900 psi
• Passed the UTS standards for metals
• Leach index results showed that process was only marginally successful in retaining NO3 and CL anions; modifications
to the basic formulation for the process were made including adding fly ash to the binder and a polymer coating to the
waste form, which increased the leach index to as high as 12.6
Costs:
• Projected cost for full-scale stabilization using phosphate bonded ceramics are capital costs of about $2 million,
including equipment design and development, and operating costs of about $6,510 per cubic meter of waste form,
including labor and materials; disposal costs are estimated to be $2,836 per cubic meter of waste
• Compared to the baseline technology (basic Portland cement), the operating costs are higher ($6,510 versus $4,300 per
cubic meter of waste form), but the disposal costs are lower ($2,836 versus $3,700 per cubic meter of waste)
Description:
A series of development tests were conducted at the Argonne National Laboratory to validate the stabilization of salt-
containing wastes using a patented chemically bonded phosphate ceramics (CBPC) process. The low-temperature process
uses magnesium oxide and monopotassium phosphate to form a low porosity, dense waste form consisting mainly of a
ceramic magnesium potassium phosphate barrier. Various tests were performed using a number of mixed waste surrogates,
including saturated salt solutions, salt surrogate containing activated carbon and ion exchange resin, and two MWFA
recommended dry salt waste surrogates that represented actual wastes found at DOE facilities.
The results of the tests showed that the waste forms produced by the CBPC process met the RCRA UTS standards for
metals and the NRC disposal criteria. Flammability test results showed the waste forms containing oxidizing salts
(nitrates) to be stable and safe. Based on the results of the testing, additional testing of the salt waste form is recommended
before full-scale deployment, such as the effects of salt anion leaching over time. For different waste streams, additional
analytical and development work would be needed to qualify wastes for disposal and to verify the operating parameters for
the specific wastes.
163
-------
Stabilize Ash Using Clemson's Sintering Process at Clemson University,
Clemson, South Carolina
Site Name: Location:
Clemson University Clemson, SC
Period of Operation: Cleanup Authority:
1995 Not identified
Purpose/Significance of Application: Cleanup Type:
Treatability study of stabilization of mixed waste fly ash using a sintering Bench scale
process
Contaminants: Waste Source:
Metals Fly ash from the WERF incinerator at
• Fly ash contained heavy metals -cadmium (5,000 mg/kg), chromium (1,000 INEEL
rag/kg), and lead (35,000 mg/kg)
Contacts:
Principal Investigator:
H. David Leigh, III
Department of Ceramic and Materials
Engineering
Clemson University
P.O. Box 340907
Clemson, SC 29634
Telephone: 864-656-5349
E-mail: david.leigh@eng.clemson.edu
MWFA Product Line Manager:
Vince Maio, Advisory Engineer
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
P.O. Box 1625
Idaho Falls, ID 83415
Telephone: 208-526-3696
Fax: 208-526-1061
E-mail: vmaio@inel.gov
Technology:
Stabilization using Clemson's Sintering Process
• Used a high iron/high potassium alumino silicate clay material - Red Roan
Formation (RRF)
• A preliminary study and three statistically designed experiments performed
evaluate and optimize processing parameters
• Preliminary study - 67 vol% to 50 wt% equivalent fly ash/RRF mixture and
high moisture content (18.1 wt%), pressed at 5,000 psi, then fired at 1,000 °
to produce waste form pellets
• Experiment I - to evaluate the effects of different physical properties on the
to
a
C
waste form included 16 batches to test varying formulations; batch size - 270
grams; material pressed at 1,000 psi then fired between 1,025 and 1,075 °C
• Experiment II - to optimize factors from experiment I included 15 batches
(500 grams each); fired between 1,025 and 1,075°C; TCLP leach testing
performed on waste forms
• Experiment III - to further evaluate effects of four physical properties
(moisture content, waste loading, mixing time, auger speed) involved 27
batches, prepared using varying formulation based on the results of the second
experiment
Type/Quantity of Media Treated:
Incinerator fly ash
Regulatory Requirements/Cleanup Goals:
RCRA Land Disposal Restriction criteria
• TCLP concentrations in mg/L - cadmium (0. 19), chromium (0.86), lead (0.37), and zinc (5.3)
164
-------
Stabilize Ash Using Clemson's Sintering Process at Clemson University,
Clemson, South Carolina
Results:
• Preliminary study - TCLP results were above the limits for cadmium, lead, and zinc
• Experiment I - significant factors affecting the waste form included firing temperature, the RRF particle size
distribution, and waste loading
• Experiment II - TCLP results showed that leach values for metals increased with increased waste loadings and
decreased as the temperature increased; TCLP limits were met when waste loadings were below 20% vol
• Experiment III - TCLP results showed that leach values for metals increased as the waste loading increased, decreased
as moisture content decreased; mixing time and auger speed were not significant factors
Costs:
No methodology has been selected to date to evaluate costs associated with full-scale deployment of the Clemson
stabilization process
Description:
A bench-scale treatability study was conducted at Clemson University in 1995 to determine whether stabilization using a
sintering process could be used to immobilize DOE waste. The study was funded by DOE through a cooperative
agreement with University Programs at the Savannah River Site. The process involves mixing a high iron/high potassium
aluminosilicate clay material with the waste, pressing the material, then firing the material to produce a ceramic waste
form. For this study, Red Roan Formation (RRF) was used as the clay material and fly ash from the WERF incinerator at
INEEL (containing high levels of metals) was used as the waste. A preliminary study and three statistically designed
experiments were performed to evaluate the process and to obtain operating data for use in future pilot-scale testing.
The results of the treatability study showed that the process can produce stable, low porosity waste forms that meet the
RCRA TCLP limits for metals at waste loadings of 20% vol or lower. This waste loading was lower than originally
anticipated. Other significant factors affecting the waste form included firing temperature and the particle size distribution
of the RRF. The process is applicable to most inorganic homogeneous solids and sludges such as ash, soils, and
particulates, but is not well siuted for aqueous and organic liquids or heterogeneous debris. Based on the results of the
treatability study, a pilot-scale demonstration of the process is planned for FY 1999.
165
-------
Mixed Waste Encapsulation in Polyester Resins at the Hanford Site
Site Name:
Hanford Site
Location:
Richland, WA
Period of Operation:
Not identified
Cleanup Authority:
RCRAandNRC
Purpose/Significance of Application:
Treatability study of various polyester resins to stabilize high salt-containing
mixed waste
Cleanup Type:
Treatability study
Contaminants:
Metals and radionuclides
• Spiked metals concentrations in treatability study wastes - arsenic (159.3
mg/kg), barium (154.1 mg/kg), cadmium (119 mg/kg), chromium (151.3
mg/kg), lead (132.7 mg/kg), and selenium (140.9 mg/kg)
• Spiked radionuclide concentrations in treatability study wastes - cesium
(1.2xl05 pCi/L), cobalt (l.lxlO5 pCi/L), strontium (l.lxlO5 pCi/L), and
tecnetium (1.3xl05 pCi/L)
Waste Source:
Salt-containing mixed wastes from
DOE processes and surrogate wastes
Contacts:
Principal Investigator:
Rabindra Biyani
COGEMA Engineering Corporation
P.O. Box 840
Richland, WA 99352
Telephone: 509-376-1004
E-mail: biyani@COGEMA-
Engineering. com
MWFA Product Line Manager:
Vince Maio, Advisory Engineer
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
P.O. Box 1625
Idaho Falls, ID 83415
Telephone: 208-526-3696
Fax: 208-526-1061
E-mail: vmaio@inel.gov
Technology:
Microencapsulation by Polyester Resin
• Four polyester resins tested - polymer (trade name) - orthophthalic (S2293),
isophthalic (Aropol™ 7334), vinyl ester (Hetron® 922-L25), and water
extendable (Aropol™ WEP 662 - proprietary)
• WEP resin was tested on aqueous wastes; other three were tested on dry waste
• Initiator (catalyst) - cobalt naphthenate
• Mixer equipped with a variable speed paddle and sample molds for curing
• Dry waste added as free-flowing powder; aqueous waste was slurried
• Mixing time - 5 to 10 minutes at a low rate to homogenize waste; additional 2
to 5 minutes at a high rate after initiator added (until the temperature rises
indicating the onset of curing)
• Curing molds placed in adiabatic chambers
• Three tests using surrogate wastes; one test using a Hanford waste stream
Type/Quantity of Media Treated:
Process waste streams
Regulatory Requirements/Cleanup Goals:
RCRA Land Disposal Restriction (LDR) and NRC disposal criteria
• Treatability test targeted to TCLP levels for RCRA heavy metals - cadmium (1.0 mg/L), hexavalent chromium (5.0
mg/L), lead (5.0 mg/L) and mercury (0.2 mg/L)
• NRC teachability indices - target of 6 or higher
166
-------
Mixed Waste Encapsulation in Polyester Resins at the Hanford Site
Results:
Orthophthalic, isophthalic, and vinyl ester resins:
• For RCRA metals, TCLP results for resins were below the target levels for all metals except cadmium. Failure was
attributed to the sampling method which required the mold be cut to a smaller size (9mm), possibly destroying the
polyester coating. To counter the effect, fully coated polyester waste form molds of 9mm were specifically prepared for
TCLP testing; this sample passed for all metals including cadmium
• Results were also compared to the UTS criteria - most samples failed for RCRA metals
• Polyester microencapsulation was validated for salt loadings of 30-wt% for all three resins, and for salt loadings of up to
70% for the Orthophthalic resin
WEP resin:
• For RCRA metals, TCLP results were below the targeted levels for all metals
• Results were also compared to the UTS criteria - samples passed for all metals expect for cadmium
• For radionucides, the teachability indices ranged from 10.1 to 10.8
Costs:
• Projected full-scale cost for the polyester resin encapsulation process - capital cost of $2 million including equipment
design and development and operating cost of $5,940/cubic meter of waste form
• Disposal cost of $2,100/cubic meter of waste form
Description:
The Mixed Waste Focus Area, a DOE Environmental Management (EM) -50 program, sponsored the development of five
low-temperature stabilization methods as an alternative to cement grouting to stabilize salt-containing mixed waste. One
of the alternative methods is microencapsulation using polyester resins. COGEMA Engineering Corporation performed a
series of treatabilitiy studies and developmental tests of the technology at the Hanford site. The studies included
encapsulation of salt-containing mixed wastes from the Handford site and with surrogate wastes spiked with contaminants.
Four types of resins were tested: Orthophthalic polyester, isophthalic polyester, and vinyl ester for dry waste, and a water-
extendible polyester resin for aqueous wastes. The cured waste forms were evaluated against the RCRA LDR and NRC
disposal criteria.
The results of the studies showed that the encapsulation of salt-containing mixed waste using polyester resins is applicable
to inorganic, relatively homogeneous low-level mixed wastes containing high levels of salt. Further development is needed
to identify chemical additives to reduce the solubility and toxicity of the RCRA metals. Other factors to be considered in
future development of the process include safety controls to address potential flammable and unstable conditions when
using polyester encapsulation, and additional research into the long-term effectiveness of the technology.
167
-------
Innovative Grouting and Retrieval at the Idaho National Engineering
and Environmental Laboratory, Idaho Falls, Idaho
Site Name:
Idaho National Engineering and Environmental Laboratory (INEEL)
Location:
Idaho Falls, ID
Period of Operation:
Summer of 1994 (innovative grouting and retrieval)
Summer of 1995 (polymer grouting)
Summer of 1996 (variety of grouting materials)
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Field demonstration of innovative jet grouting and retrieval techniques that are
applicable to TRU wastes
Cleanup Type:
Full scale and field demonstrations
Contaminants:
• Radioactive and nonradioactive wastes
• Demonstration used nonradioactive tracer to simulate radioactive materials
Waste Source:
Buried drums and waste from DOE
operations
Contacts:
Technical Contact:
G.G. Loomis
Lockheed Martin Idaho
Technologies Company
INEEL
P.O. Box 1625, MS 3710
Idaho Falls, ID 83415
Telephone: 208-526-9208
E-mail: guy@inel.gov
DOE Contacts:
Skip Chamberlain
Subsurface Contaminants
Focus Area HQ Lead
DOE EM50
Germantown, MD
Telephone: 301-903-7248
James Wright
Subsurface Contaminants
Focus Area Program
Manager
DOE Savannah River
Aiken, SC
Telephone: 803-725-5608
Technology:
Innovative Grouting and Retrieval (IGR)
• Demonstrated on a waste pit (10 ft3), loaded with 55-gal cardboard and steel drums, and
cardboard boxes (4 ft3) filled with waste and rare-earth tracer designed to simulate
transuranic (TRU) pits
• Three phases - jet grouting, application of demolition grout, and retrieval of the waste
• Jet grouting - CASA GRANDE drill system and a high pressure displacement pump used
to inject grout at a nominal 6,000 psi; total of 24 yds3 of Portland cement injected into 36
grout holes, creating a monolith
• Demolition grouting - immediately following jet grouting, thin-walled, spiral-wrapped
tubes were inserted into the holes and allowed to cure, after which the demolition grout
(BRISTAR) was added to the tubes; however, the grout did not expand as planned and
the soil/waste matrix was not fractured
• Retrieval - a backhoe bucket was used to remove the monolith
Wall Stabilization Technique Using Jet Grouting for Hot Spot Removal
• Created a U-shaped wall by jet grouting Portland cement into an existing cold test pit at
INEEL containing drums and boxes
• Jet grouting phase - 52 holes jet grouted to create the wall (30 ft along back and sides of
U extended 8 feet); used jet grouting apparatus at 6,000 psi; total of 24 yds3 of Portland
cement injected
• Stabilization evaluation phase - wall excavated and visually examined; no collapse or
structural damage to wall during excavation and no visible voids; grout mixed with soil
and formed a soilcrete material that filled some voids; neat Portland cement filled other
voids
Jet Grouted Polymer for Waste Stabilization or as an Interim Technique Before Retrieval
• Demonstrated on two waste pits designed to simulate TRU pits containing drums; used
55-gallon drums containing cloth, paper, metal, wood, and sludge; tracer placed in each
drum to simulate plutonium oxide
• Tested two formulations of an acrylic polymer - one to produce a hard, durable material
for long-term encapsulation; one to form a soft material for retrieval
• Hard polymer pit -18 holes jet grouted into 4.5 x 9 x 6 ft pit; after curing, hard polymer
was fractured and removed
• Soft polymer pit -15 holes jet grouted into 4.5 x 9 x 6 ft pit; after curing, removed with a
backhoe
In Situ Stabilization
• Demonstrated variety of grouting materials - TECT grout, WAXFIT, Hermite, water-
based epoxy, and Type H cement; jet grouted to form monoliths of buried waste
168
-------
Innovative Grouting and Retrieval at the Idaho National Engineering
and Environmental Laboratory, Idaho Falls, Idaho
Type/Quantity of Media Treated:
Soil and debris
• Steel drams; cardboard boxes containing metal pipe, wire, and plate steel; paper
Regulatory Requirements/Cleanup Goals:
• Purpose of the demonstrations was to evaluate different jet grouting techniques for use in stabilization and hot-spot
retrieval of waste; nonradioactive wastes used for demonstrations
• No specific cleanup goals were identified
Results:
• IGR - produced stable monolith; monolith was removed in 5 hrs; general soilcrete mix easily removed; grouted waste
that were more difficult to retrieve included grouted boxes containing metal pipe, wire, and plate steel and grouted
computer paper, which disintegrated during removal
• Wall - produced a solid wall with no visible voids; wall was stable and excavated intact
• Soft polymer - soft polymer material was removed easily; however, tracer material was detected at two-orders of
magnitude above background; determined that one of the containers was not penetrated during drilling; but was
punctured during removal releasing tracer
• Hard polymer - produced cured, stabilized monolith with no voids; easily fractured with a backhoe and removed
• Various grout materials - TECT, WAXFIT, and Type H materials are easily jet grouted and produced stable monoliths;
Hermite and water-based epoxy cannot be jet grouted
• In general, grouting techniques did not spread tracer, indicating that release of radioactive particulates would be
minimized during operations
Costs:
• Costs projected for IGR, jet grouting using TECT, and jet grouting using WAXFIT; costs developed for 1-acre; for IGR
costs also developed for 4-acre TRU contaminated site
• IGR - projected cost is $19 million (1-acre) and $64 million (4-acre), including grouting and waste management,
excavation, secondary waste management, and D&D equipment
• TECT - projected cost is $15 million, assuming pit is left in place permanently; includes costs for grouting and waste
management and secondary waste management, but no costs for caps
• WAXFIT - projected costs is $20 million, assuming waste pit is a soft polymer and is retrieved; includes costs for
grouting and waste management and retrieval operations
• Jet grouting technologies were less expensive than the baseline retrieval, packaging, and storage ($200 million for 1-
acre; $305 million for 4-acres)
Description:
Between 1994 and 1996, a number of different innovative jet grouting techniques were demonstrated at INEEL to
determine their potential for use in stabilization and retrieval of buried transuranic (TRU) and other wastes at DOE
facilities. Nonradioactive debris containing a rare-earth tracer were tested on waste pits designed to simulate those found
at TRU sites. Technologies demonstrated included innovative grouting and retrieval, wall stabilization techniques using
jet grouting for hot-spot removal, jet grouted polymer for waste stabilization, and various grouting materials for
stabilization.
The results of the demonstrations indicated that a number of the jet grouting technologies produce stable waste forms that
are generally easy to remove, thus making the technology suitable for stabilization and for hot-spot removal. In addition,
the costs for jet grouting and retrieval is up to 90% less than the costs for the baseline technology of retrieval, packaging,
and storage. Further testing is need of the BRISTAR demolition grout, which did not perform as expected, and long-term
durability studies of the materials are recommended, including development of monitoring systems to ensure complete
encapsulation of the waste.
169
-------
Polysiloxane Stabilization at
Idaho National Engineering and Environmental Laboratory, Idaho Falls, Idaho
Site Name:
Idaho National Engineering and Environmental Laboratory (INEEL)
Location:
Idaho Falls, ID
Period of Operation:
1997 - 1998
Cleanup Authority:
RCRA and NRC
Purpose/Significance of Application:
Demonstration of polysiloxane to encapsulate high-salt content wastes
Cleanup Type:
Field demonstration
Contaminants:
Heavy Metals
• hexavalent chromium -1.045 ppm in one surrogate waste
• oxides of lead, mercury, cadmium, and chromium at 1,000 ppm each in two
surrogate wastes
Waste Source:
Salt-containing wastes designed to
simulate wastes from DOE operations
Contacts:
Vendor Contact:
Dr. Steve Prewett
Orbit Technologies
Palomar Triad One
2011 Palomar Airport Road, Suite 100
Carlsbad, CA
330-794-2122
Principal Investigator:
G.G. Loomis
Lockhead Martin Idaho Technologies
Company
INEEL
P.O. Box 1625 (MS 3710)
Idaho Falls, ID 84315
208-526-9208
guy@inel.gov
MWFA Product Line Manager:
Vince Maio, Advisory Engineer
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
P.O. Box 1625
Idaho Falls, ID 83415
Telephone: 208-526-3696
Fax: 208-526-1061
E-mail: vmaio@inel.gov
Technology:
Stabilization using polysiloxane
• Polysiloxane is a part inorganic part thermosetting polymer; for the
demonstration, Orbit Technology's polysiloxane material was used
• The base chemicals (SiH and SiOH) are mixed with the waste and reacted in
the presence of a platinum catalyst to form the desired thermosetting polymer
and hydrogen gas
• A filler such as quartz can be added to strengthen the waste form
• The resultant vinyl-polydimethyl-siloxane product is gelled, and cured to form
a solid waste form
• For the demonstration, the process was tested on three different salt surrogates
- Pad-A salts from INEEL, one high chloride salt surrogate, and one high
nitrate salt surrogate
Type/Quantity of Media Treated:
Process waste streams
Regulatory Requirements/Cleanup Goals:
RCRA Land Disposal Restriction (LDR) and DOT
• Target TCLP levels for RCRA heavy metals - cadmium (1.0 mg/L), hexavalent chromium (5.0 mg/L), lead (5.0 mg/L)
and mercury (0.2 mg/L); also compared to RCRA universal treatment standards (UTS)
• DOT oxidizer test for nitrate salt wastes
• NRC recommended compressive strength of at least 60 psi
170
-------
Polysiloxane Stabilization at
Idaho National Engineering and Environmental Laboratory, Idaho Falls, Idaho
Results:
• INEEL Pad-A salt surrogate waste form - met the target TCLP levels; but did not meet the UTS standard for chromium;
had a compressive strength of 637 psi
• Chloride salt surrogate waste form - met the target TCLP levels; did not meet the UTS standard for cadmium or
chromium
• Nitrate salt surrogate waste form - met the target TCLP levels; did not meet the UTS for chromium or mercury; passed
the DOT oxidizer test
Costs:
• Cost for full-scale polysiloxzane treatment are about $8/lb or $573 per cubic foot of salt waste
• The cost for polysiloxane encapsulation is competitive with the baseline technology of Portland cement stabilization
Description:
The Mixed Waste Focus Area, a DOE Environmental Management (EM)-50 program, sponsored the development of five
low-temperature stabilization methods as an alternative to cement grouting to stabilize salt-containing mixed waste. One
of the alternative methods is stabilization using polysiloxane. A demonstration of Orbit Technology' s polysiloxane
encapsulation process for high-salt content wastes was performed at INEEL on three salt surrogates, representing wastes
found at DOE facilities.
The results showed that the polysiloxane process produced a durable waste form for all three high-salt content surrogates.
The waste forms met the target TCLP levels for heavy metals, and the more stringent UTS standards for several of the
metals tested. The process is currently limited to nonaqueous solid materials. Treatability testing is recommended for
specific wastes prior to use of this technology. In addition, long-term durability testing of the polysiloxane waste forms is
needed.
171
-------
Amalgamation of Mercury-Contaminated Waste using NFS DeHgSM Process,
Applied Technology Laboratories, Erwin, TX
Site Name:
U.S. DOE INEEL, ETTP, and DSSI Facilities (tests conducted at Applied
Technology Laboratories, Erwin, TN)
Period of Operation:
1998
Purpose/Significance of Application:
Demonstrate amalgamation of elemental mercury
Contaminants:
Heavy metals
• Mercury
Contacts:
Technology Vendor:
Nuclear Fuel Services, Inc.
Erwin, Tennessee
Contacts:
Thomas B. Conley
Oak Ridge National Laboratory
Telephone: (423) 574-6792
Fax: (423) 574-7241
E-mail: tbc@ornl.gov
William Owca
DOE Idaho Operations Office
Telephone: (208) 526-1983
Fax: (208) 526-5964
E-mail: owcawa@inel.gov
Location:
Idaho and Tennessee
Cleanup Authority:
Not identified
Cleanup Type:
Field demonstration
Waste Source:
Nuclear processing operations
Technology:
Amalgamation using the NFS DeHgSM Process
• Prior to amalgamation, waste is sorted, shredded, and slurried to create a
homogeneous mixture
• The first step in the process is to stabilize elemental mercury using one or
more amalgamation agents (agents not specified)
• A possible second step is a chemical stabilization process using a proprietary
reagent to break mercury complexes and allow removal of mercury as a
precipitant; this step is required if the waste fails the cleanup criteria after the
first step
• Treated material is produced as a presscake; filtrate is either recycled to the
reactor or discharged
• Processing was conducted at ambient conditions in a ventilated hood
Type/Quantity of Media Treated:
Liquid mercury
• 5 1 kg from East Tennessee Technology Park, formerly the K-25 Site;
characterized as RCRA Waste Code U151
• 23 kg from INEEL; contained oil at 17% by volume; characterized as RCRA
Waste Code D009
• 1 kg from Diversified Scientific Services (DSSI); this material had been
recovered from a thermal desorption treatability study; also D009
Regulatory Requirements/Cleanup Goals:
• Envirocare of Utah Waste Acceptance Criteria
• For mercury - TCLP leachate concentration of 0.20 mg/L; also considered UTS
of 0.025 mg/L
Results:
• Wastes from INEEL (DSSI wastes were combined with those from INEEL) were treated with two step process; for
mercury - TCLP leachate in presscake from second step averaged 0.05 mg/L (range 0.02 to 0. 12 mg/L); TCLP leachate
in oil phase was 0.03 mg/L; total of 15 amalgams weighed 1 14 kg
• Wastes from ETTP were treated with two step process; for mercury - TCLP leachate in presscake from second step
averaged 0.05 mg/L (range 0.01 to 0.17 mg/L); total of 20 amalgams weighed 238 kg
Costs:
• Projected costs for treating more than 1,500 kg were $300/kg, assuming waste is elemental mercury, and does not
include disposal costs of the treated waste
172
-------
Amalgamation of Mercury-Contaminated Waste using NFS DeHgSM Process,
Applied Technology Laboratories, Erwin, TX
Description:
Nuclear Fuel Services (NFS) conducted a demonstration of an amalgamation technology on wastes containing elemental
mercury. The NFS process consists of a two step process, where mercury is first treated using amalgamation agents and
then with proprietary chemical stabilization agents, and is conducted in a hood at ambient conditions.
Wastes from ETTP, INEEL, and DSSI were tested using this process. Results showed that the process reduced the
concentration of mercury to 0.05 mg/L (on average) for each of 35 batches tested, and that the product met the Envirocare
Waste Acceptance Criteria. Projected costs for use of the technology were $300/kg and costs for treating smaller amounts
of wastes, such as at a specific site, were projected to be prohibitive. The report discusses the possibility of a national
procurement contract to lower the cost of the technology on a unit mass basis.
173
-------
Amalgamation of Mercury-Contaminated Waste using ADA Process,
Colorado Minerals Research Institute
Site Name:
U.S. DOE Los Alamos National Laboratory and Fernald Facilities (tests
conducted at Colorado Minerals Research Institute)
Period of Operation:
1998
Purpose/Significance of Application:
Demonstrate amalgamation of elemental mercury
Contaminants:
Heavy metals
• Mercury
Contacts:
Technology Vendor:
ADA Technologies
Englewood, CO
Contacts:
Thomas B. Conley
Oak Ridge National Laboratory
Telephone: (423) 574-6792
Fax: (423) 574-7241
E-mail: tbc@ornl.gov
William Owca
DOE Idaho Operations Office
Telephone: (208) 526-1983
Fax: (208) 526-5964
E-mail: owcawa@inel.gov
Location:
New Mexico and Ohio
Cleanup Authority:
Not identified
Cleanup Type:
Field demonstration
Waste Source:
Nuclear processing operations at U.S.
DOE Los Alamos National Laboratory
and Fernald Facilities
Technology:
Amalgamation using the ADA Process
• Process consists of combining liquid mercury with a proprietary sulfur
mixture in a pug mill to stabilize the elemental mercury
• The pug mill was a dual shaft mixer 0.9 m long with a 0. 1 nf cross section,
and held 0.06 m3 of material; the mixing blades were 14 cm long and
overlapped; mixing speed was 50 rpm
• Mixing was concluded when the reaction exotherm subsided and free
elemental mercury analysis indicated that more than 99% of the mercury had
reacted
• Air above the pug mill was swept to remove mercury vapors and filtered
through a HEPA filter and a sulfur-impregnated carbon filter to capture
mercury
• Processing was conducted at ambient conditions
Type/Quantity of Media Treated:
Liquid mercury
• 1 12 kg of mercury from LANL and 20 kg from Fernald
• No radioactivity was detected in either waste stream
• The waste from Fernald contained significant amounts of water
Regulatory Requirements/Cleanup Goals:
• Envirocare of Utah Waste Acceptance Criteria
• RCRA TCLP limit for mercury - 0.20 mg/L
Results:
• Wastes were processed in 5 batches (4 from LANL and 1 from Fernald) of 20 to 33 kg/batch
• The amount of free mercury was reduced from 99.87 to 99.98% per batch
• TCLP mercury was less than 0. 1 mg/L in each batch, with a mercury waste loading of 57%
• Product from the amalgamation process was found to meet the Envirocare Waste Acceptance Criteria
• Mercury vapor concentrations above the pug mill were below the TLV of 50 ug/m3
Costs:
• Projected costs for full-scale amalgamation using the ADA Process were $300/kg for more than 1,500 kg, assuming
waste is elemental mercury, and does not include disposal costs of the treated waste
174
-------
Amalgamation of Mercury-Contaminated Waste using ADA Process,
Colorado Minerals Research Institute
Description:
ADA Technologies conducted a demonstration of a proprietary amalgamation technology on wastes containing elemental
mercury from Los Alamos and Fernald. The ADA process consists of combining liquid mercury with a proprietary sulfur
mixture in a pug mill, and is conducted at ambient conditions.
Results showed that the process reduced the free mercury by 99.87 to 99.98%, and that the product met the Envirocare
Waste Acceptance Criteria and passed the RCRA TCLP criteria for mercury. Projected costs for use of the technology
were $300/kg and costs for treating smaller amounts of wastes, such as at a specific site, were projected to be prohibitive.
The report discusses the possibility of a national procurement contract to lower the cost of the technology on a unit mass
basis.
175
-------
GTS Duratek (GTSD) Process for Stabilizing Mercury (<260 ppm) Contaminated
Mixed Waste from U.S. DOE's Los Alamos National Laboratory
Site Name:
U.S. DOE Los Alamos National Laboratory (tests conducted at GTSD Bear
Creek Operations Facility)
Location:
New Mexico
Period of Operation:
September 1997 to September 1998
Cleanup Authority:
Not identified
Purpose/Significance of Application:
Demonstrate stabilization of low level mercury in radioactive wastes
Cleanup Type:
Treatability studies
Contaminants:
Heavy metals, Volatile Organics, and Radionuclides
• Mercury concentration was 230 mg/kg; TCLP 0.0399 to 0.184 mg/L
• DCE concentration was 11,000 mg/kg, vinyl chloride 220 mg/kg, methylene
chloride 12,000 mg/kg
• Radionuclides included plutonium and strontium
Waste Source:
Nuclear processing operations
Contacts:
Technology Vendor:
GTS Duratek
Kingston, Tennessee
Contacts:
Thomas B. Conley
Oak Ridge National Laboratory
Telephone: (423) 241-1839
Fax: (423) 241-2973
E-mail: tbc@ornl.gov
William Owca
DOE Idaho Operations Office
Telephone: (208) 526-1983
Fax: (208) 526-5964
E-mail: owcawa@inel.gov
Technology:
Stabilization
• Stabilization reagents involved addition of water and then cement to form a
grout mixture; the mixture was then blended with sodium metasilicate and
cured for two days
• Bench- and pilot-scale tests were conducted, at high and low waste loadings
• Pilot-scale tests were conducted in drums using a vertical in-drum mixer
Type/Quantity of Media Treated:
Sludge and Laboratory Wastes
• Four 55-gallon drums containing 1,253 Ibs of sludge
• Three containers of lab packs from analysis of the sludge
Regulatory Requirements/Cleanup Goals:
• Land Disposal Restrictions for heavy metals (such as mercury - 0.025 mg/L) and organics
• Envirocare Waste Acceptance Criteria (WAC) for disposal
Results:
• At low waste loadings, mercury concentrations were reduced to values ranging from 0.00127 to 0.0169 mg/L, below the
LDR standard of 0.025 mg/L; at high waste loadings, mercury was reduced to values ranging from 0.0024 to 0.0314
mg/L - one sample contained mercury above the LDR standard
• Several organic compounds and radionuclides were higher than the LDR standards or Envirocare WAC after treatment,
including 1,1,1-trichloroethane, 1,1-dichloroethane, methylene chloride, lindane, DDE, heptachlor epoxide, and
methoxychlor, strontium, and americium
• The vendor indicated that these results re-emphasized the importance of accurate characterization data; the high levels
of organics were not expected based on the original characterization data provided by LANL
• Bench-scale tests showed mercury met LDR level in all 3 low load and 2 of 3 high load samples
Costs:
• Projected costs for a full-scale stabilization system using this technology were not developed
176
-------
GTS Duratek (GTSD) Process for Stabilizing Mercury (<260 ppm) Contaminated
Mixed Waste from U.S. DOE's Los Alamos National Laboratory
Description:
Sludge was generated at the Phase Separation Pits of the TA 35 facility of the Los Alamos National Laboratory (LANL) by
addition of a caustic solution to the condensate and particulates removed from laboratory fume hood exhausts by the phase
separators. The sludge and laboratory wastes from analysis of the sludge, were a mixed waste due to the presence of
radionuclides, heavy metals, and RCRA-listed organic compounds.
Bench- and pilot-scale tests of the GTS Duratek process were conducted to stabilize the contaminants in the sludge and
laboratory wastes. The GTS Duratek process includes addition of water, cement, and sodium metasilicate. The stabilized
product met the LDR standard for mercury in all but one high load test sample. However, several VOCs, pesticides,
herbicides, and radionuclides did not meet the LDR standards or Envirocare WAC after treatment. This result was
attributed to inaccurate characterization data of the waste streams, which did not show the relatively high levels of
organics.
177
-------
Stabilize High Salt Content Waste Using Sol Gel Process at
Pacific Northwest National Laboratory, Richland, WA
Site Name:
Pacific Northwest National Laboratory
Location:
Richland, WA
Period of Operation:
Not identified
Cleanup Authority:
RCRA and NRC
Purpose/Significance of Application:
Laboratory testing of the sol gel process to stabilize high salt content waste
Cleanup Type:
Laboratory scale treatability test
Contaminants:
Metals and salts
• Two salt-containing, nonradioactive surrogates - one with nitrate salts; one
with chloride and sulfate salts
• Both contained 1,000 mg/kg each of lead, chromium, cadmium, and nickel (in
the form of metal oxides)
Waste Source:
Salt waste surrogates that simulated
nonradioactive wastes from DOE
facilities
Contacts:
Technical Contacts:
Dr. Gary L. Smith
Pacific Northwest National Laboratory
P.O. Box 999, MSIN K6-24
Richland, WA 99352
Telephone: 509-372-1957
Fax:509-376-3108
E-mail: Gary.L.Smith@pnl.gov
Dr. Brian Zelinski
Arizona Materials Laboratory
University of Arizona
4715 East Fort Lowell Road
Tuscon, AZ 85712
Telephone: 520-322-2977
Fax: 520-322-2993
E-mail: brianz@engr.arizona,edu
MWFA Product Line Manager:
Vince Maio, Advisory Engineer
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
P.O. Box 1625
Idaho Falls, ID 83415
Telephone: 208-526-3696
Fax: 208-526-1061
E-mail: vmaio@inel.gov
Technology:
Stabilization using the Sol Gel Process
• The Sol Gel processing is a general synthesis technique that uses hydrolysis
and condensation to produce solid matrices from liquids
• Ceramic portion formed after tetraethlyorthosilicate (TEOS) was
prehydrolized with acidified water (0.15M HCL) in tertrahydrofuran (THF)
• The polymer polybutadiene was added and the solution was refluxed for 30
minutes
• Salt waste surrogate was mixed into the solution and stirred until the solution
thickened
• Solution was then transferred to a plastic container, allowed to gel, then
capped (the cap was punctured with small holes to allow gas to escape) and
dried in an oven at 66 ° C for a minimum of 24 hours, then placed in a vacuum
oven at 70 ° C for three hours
• The resulting material was a polyceram waste form
• Process modified after initial test results showed open porosity in sample
waste forms; to minimize open porosity, dried samples were submerged in a
polycream or resin solution and placed under vacuum to allow infiltration,
then dried overnight
Type/Quantity of Media Treated:
Salt waste surrogates - two surrogates tested at waste salt loadings of 50 to 70%
Regulatory Requirements/Cleanup Goals:
• RCRA TCLP criteria for metals
• Leachability index (LI) of at least 6.0 for the salt components
• Compressive strength of salt waste forms of at least 60 psi
• Final waste form must incorporate at least 10-wt% of the salt component
178
-------
Stabilize High Salt Content Waste Using Sol Gel Process at
Pacific Northwest National Laboratory, Richland, WA
Results:
• Initial samples met requirements for compressive strength and LI; however, forms contained open porosity which
exacerbated leaching, resulting in the samples not meeting the TCLP limits for metals
• After process was modified to minimize open porosity, samples were below the TCLP limits for all metals and very near
or below the UTS limits for metals (results for cadmium and chromium were reported slightly above the UTS limits, but
results were below the practical quantification limits of the instrument)
• The second waste form samples contained 50% of the chloride/sulfate salt surrogate; data on compressive strength and
LI were not available; however, report indicated that these samples were expected to be stronger and have a higher LI
than the first samples
Costs:
• To date, no detailed cost analyses have been performed on this process
• The report included an order of magnitude estimate for the Sol-Gel process in the range of $600,000 to $ 1 million for
design, capital equipment, installation, and startup costs, as well as obtaining the required environmental and operating
permits
Description:
At the Pacific Northwest National Laboratory, DOE conducted laboratory scale testing of the Sol Gel process to stabilize
high salt content waste. Two salt-containing, nonradioactive surrogates - one with high levels of nitrate salts and one with
high levels of chloride and sulfate salts - were used for the tests to simulate wastes at DOE facilities. The Sol Gel process
involved combining a polymer (polybutadiene) and an oxide-based ceramic (formed using TEOS, acidified water, and
THF) to produce a solid material referred to as a polyceram. The resulting polyceram waste forms were tested to
determine teachability and compressive strength at salt waste loadings of at least 10-wt%.
While initial samples met the requirements for compressive strength and teachability index, they did not meet the TCLP
criteria because the form contained open porosity. To minimize open porosity, the process was then modified to include
infiltration of dried samples with a resin. Test results for the infiltrated samples were below the TCLP levels and near or
below the UTS levels. While a detailed cost analysis had not been performed on the process, an order of magnitude
estimate indicates that the process would cost in the range of $600,000 to $ 1 million.
179
-------
ATG Process for Stabilizing Mercury (<260 ppm) Contaminated Mixed Waste from
U.S. DOE's Portsmouth, Ohio Facility
Site Name:
U.S. DOE Portsmouth, Ohio (tests conducted at Mountain States Analytical
Laboratory)
Period of Operation:
1998
Purpose/Significance of Application:
Demonstrate stabilization of low level mercury in radioactive wastes
Contaminants:
Heavy metals and Radionuclides
• Mercury concentration was 1.06 mg/mL
• Technetium-99 present at 680 pCi/g
• Other heavy metals included barium, cadmium, and chromium
Contacts:
Technology Vendor:
Allied Technology Group
Fremont, CA
Contacts:
Thomas B. Conley
Oak Ridge National Laboratory
Telephone: (423) 241-1839
Fax: (423) 241-2973
E-mail: tbc@ornl.gov
William Owca
DOE Idaho Operations Office
Telephone: (208) 526-1983
Fax: (208) 526-5964
E-mail: owcawa@inel.gov
Location:
Portsmouth, Ohio
Cleanup Authority:
Not identified
Cleanup Type:
Field demonstration
Waste Source:
Nuclear processing operations
Technology:
Stabilization
• Stabilization reagents included proprietary dithiocarbamate (DTC),
phosphate, polymeric reagents, and generic reagents such as magnesium oxide
and activate carbon
• Bench-scale and field demonstration tests were conducted
• Bench-scale tests showed that DTC without other reagents provided the best
results
• Field demonstration tests were conducted on three-3 3 kg batches of waste,
using a 7 ft3 mortar mixer
Type/Quantity of Media Treated:
Ion exchange resin
• 160 kg of resin (liquid waste) containing <5% solids
Regulatory Requirements/Cleanup Goals:
• Universal treatment standard (UTS) for mercury of 0.025 mg/L
Results:
• Mercury concentrations were reduced on average from 1.06 to 0.0092 mg/L, below the UTS of 0.025 mg/L; 99% of the
mercury was stabilized
• Cadmium, the other heavy metal present at concentrations higher than the UTS, was reduced on average from 0.371 to
0.053 mg/L, below the UTS of 0.11 mg/L; 86% of the cadmium was stabilized
• The average density of the treated waste was 1.17 kg/L, which was a 17% weight increase and a 16% volume increase
from the untreated waste
• No mercury vapors or radioactivity was detected during the demonstration
Costs:
• Projected costs for a 1,200 Ib/hr stabilization system included capital costs of $30,000 and operating costs of $95/hr of
operation
• These correspond to a life cycle cost of $ 1 .73/kg, without decontamination and decommissioning
180
-------
ATG Process for Stabilizing Mercury (<260 ppm) Contaminated Mixed Waste from
U.S. DOE's Portsmouth, Ohio Facility
Description:
Allied Technology Group (ATG) conducted a demonstration of stabilization of mixed wastes containing less than 260 ppm
of mercury. The ATG technology used dithiocarbamate (DTC) to stabilize 160 kg of ion exchange resin containing <5%
solids. The resin was contaminated with heavy metals including mercury and cadmium.
The DTC formulation stabilized mercury and cadmium to concentrations lower than the UTS, with a relatively small
increase in weight and volume. A life cycle cost of $ 1.73/kg of waste was projected for use of this technology at a full
scale.
181
-------
Graphite Electrode DC Arc Furnace at the Idaho National Engineering and
Environmental Laboratory, Idaho Falls, Idaho
Site Name:
Idaho National Engineering and Environmental Laboratory (INEEL)
Period of Operation:
1997-1998
Purpose/Significance of Application:
Determine potential applicability of DC arc plasma furnace to treat a variety of
wastes from DOE facilities
Contaminants:
Metals, Radionulcides
• Plutonium-238 and heavy metals including lead
Contacts:
Principal Investigator:
Ronald Goles
Battelle, Pacific Northwest National
Laboratory
P.O. Box 999, MS K6-24
Richland, WA 99352
Telephone: 509-376-2030
Fax:509-376-3108
E-mail: rwgoles@pnl.gov
MWFA Product Line Manager:
Whitney St. Michael
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
2525 N. Freemont
Idaho Falls, ID 83415
Telephone: 208-526-3206
Fax: 208-526-1061
E-mail: whitney@inel.gov
Technology:
Graphite Electrode DC Arc Furnace
Location:
Idaho Falls, ID
Cleanup Authority:
RCRA and NRC
Cleanup Type:
Bench-scale studies and engineering-
scale furnace (ESF) tests
Waste Source:
Waste streams and surrogates from
various DOE facilities
• ESF system included the furnace, power control systems, feed systems, off-
gas system, and control system
• ESF - 3 .5 ft diameter by 4 ft high stainless steel vessel enclosing the furnace
hearth; graphite crucible was lined with Monofrax K-3 refractory; four
graphite rods threaded into the crucible; layers of porous graphite, firebrick,
and refractory material surround crucible; nitrogen used to prevent oxygen
from attacking the graphite crucible
• ESF included penetrations for glass overflow discharge, furnace offgas, and
pyrometer access; overflow section heated to temperatures as high as 1,500°C
to keep glass molten for pouring
• Outer walls of furnace equipped with air cooling j acket and two cooling coils
- to prevent glass migration throughout refractories and insulation
• Bottom drain - inductively heated/freeze-valve bottom drain for removing
metals and/or slag from the bottom of the furnace
• Bench-scale testing included 43 nonradioactive waste tests and 5 radioactive
waste tests
• Two ESF tests conducted in FY 1997
- one using feed spiked with heavy
metals and with plutonium surrogates; one using nonradioactive debris
• First test feed rate was about 5 kg/hr and about 320 kg of feed material was
processed over an 86-hour period; operational problems caused furnace to be
shut down during second test
• One ESF test conducted in FY 1998 on Pantex neutron generators - process
150 neutron generators over a 21 -hour period at a rate of 27 Ibs/hr
Type/Quantity of Media Treated:
• INEEL wastes including soils, high metal wastes, organics/oils/solvents, and
debris
• Slag from Rocky Flats
• Plutonium-238 waste from SRS
• Neutron generators (tritium and lead)
- about 150
Regulatory Requirements/Cleanup Goals:
RCRA Land Disposal Restriction criteria for metals and NRC disposal criteria
182
-------
Graphite Electrode DC Arc Furnace at the Idaho National Engineering and
Environmental Laboratory, Idaho Falls, Idaho
Results:
• The first 1997 test performed as planned with minor problems such as failure of the overflow heater, which was
corrected; produced a uniform, homogeneous vitrified product with a low leach rate for TCLP metals; the behavior of
Plutonium-238 was identical to that of Plutonium-239, with the majority of the plutonium partitioning in the glass phase
• During the second 1997 test, the furnace failed as a result of current firing through a fracture in the sidewall; the system
was shut down and repaired
• 1998 test results showed that the ESF was capable of processing neutron generators, with the resulting glass form
passing the TCLP test for metals; however, approximately 75% of the available lead partitioned to the off gas system
(attributed to the glass collection problem) and 85% of the available tritium was released through the process stack
• Operational problems with the 1998 test included the inability to operate the bottom drain of the melter and the need to
operate in a continuous overflow mode, causing problems with glass collection
• In general, high water content in sludges (30wt%) increased electrode corrosion, caused problems with feeding via the
solids auger and caused water to collect in the off-gas system
Costs:
• Projected cost for full-scale - $50 to $80 million capital cost; operating costs of $12 to $18 million through the startup
period and $48 to $62 million for a five year operating period
• Projected treatment and disposal costs - $7,400 to $10,800 per cubic meter, based on 17,000 cubic meters of waste
• Total life cycle costs estimated to be $124 to $184 million
Description:
A series of bench-scale tests using radioactive and nonradioactive wastes were conducted at INEEL to determine the
potential for using a DC Arc Furnace for waste treatment. Several types of wastes were tested including Rocky Flats
Pondcrete (slag); INEEL soils, high metals wastes, organics/oils/solvents; and debris; and an SRS 238Pu contaminated
debris waste. A DC Arc ESF system, including the furnace, power control systems, feed systems, off-gas system, and
control system, was used for two sets of tests of radioactive and nonradioactive wastes in 1997, and to test the ability to
process neutron generators in 1998.
The results of the first 1997 test showed that the DC Arc Furnace could produce a solid, homogenous glass form that met
the TCLP criteria for metals. The system was then shutdown during the second test when the furnace failed. Following
repairs, the system was shown to be capable of processing neutron generators, with the glass form meeting the TCLP limits
for metals. However, several operational difficulties led to the partitioning of a majority of the primary contaminants
(tritium and lead) to the off-gas. Since these demonstrations, several design improvements have been made to the
prototype system, including a second generation melter and improvements in the feed system and off-gas treatment
systems.
183
-------
Plasma Hearth Process at the Science and Technology Applications Research
(STAR) Center,Idaho Falls, Idaho
Site Name:
STAR Center
Period of Operation:
1993 through 1997
Purpose/Significance of Application:
Demonstration of a plasma hearth furnace to treat metals and radionuclides in a
variety of waste types
Contaminants:
Metals and radionuclides
• Nonradioactive cerium used in tests to simulate plutonium
• Metals include arsenic, barium, cadmium, chromium, lead, mercury
Contacts:
Principal Investigators:
Ray Geimer
SAIC
545 Shoup Ave.
Idaho Falls, ID 83402
Telephone: 208-528-2144
Fax: 208-528-2194
E-mail: Ray Geimer@cpqm.saic.com
Carla Dwight
Argonne National Laboratory - West
P.O. Box 2528
Idaho Falls, ID
Telephone: 208-533-7651
E-mail: carla.dwight@anl.gov
MWFA Product Line Manager:
Whitney St. Michael
Mixed Waste Focus Area
Lockheed Martin Idaho Technologies
Company
Idaho National Engineering and
Environmental Laboratory
2525 N. Freemont
Idaho Falls, ID 83415
Telephone: 208-526-3206
Fax: 208-526-1061
E-mail: whitney@inel.gov
Location:
Idaho Falls, ID
Cleanup Authority:
RCRA and NRC
Cleanup Type:
Bench scale and pilot scale
Waste Source:
• Wastes from DOE facility
operations and air pollution control
systems
Technology:
Plasma Hearth Process (PHP)
• PHP is a high temperature thermal process that heats waste to a molten form,
which is then cooled into a glass/crystalline waste form; equipped with an air
pollution control system to remove particulates and volatiles in the offgas
• PHP melt temperature - 1 ,650-2,200 ° C;
• Three systems tested - nonradioactive bench-scale system (NBS), radioactive
bench-scale system (RBS), and nonradioactive pilot-scale system (NFS)
• NB S- batch system with a refractory lined fixed hearth vessel equipped with a
150 KW Retech RP75T transferred arc plasma torch; feed rate of 15 Ibs/hr
• RBS - batch system with a plasma chamber equipped with a 150 KW Retech
RP75T transferred arc plasma torch; feed rate of 30 Ibs/hr; holds eight, 1-
gallon waste containers and includes offgas treatment system
• NFS - 6.5 ft by 6.5 ft cylindrical hearth equipped with a 1.2 megawatt Retech
RP600T plasma torch; feed rate of 1,000 - 1,500 Ibs/hr; holds three, 55-gallon
waste drums and includes offgas treatment
Type/Quantity of Media Treated:
• NBS - Fly ash, soil, sludges, debris (concrete, asphalt, sheet rock, steel),
sodium nitrate
• RBS - inorganic and organic sludges, debris (wood, graphite, and fire brick)
• NFS -debris
Regulatory Requirements/Cleanup Goals:
• RCRA Land Disposal Restriction (LDR) standards
• Federal and state air emissions standards
184
-------
Plasma Hearth Process at the Science and Technology Applications Research
(STAR) Center,Idaho Falls, Idaho
Results:
• Slag samples passed the RCRA limits for metals
• Cerium oxide (plutonium oxide surrogate) was found to primarily partition to the vitreous slag; slightly higher retention
rates were noted for sludges as compared to combustible debris
• All high vapor pressure metals (mercury, cadmium, lead), except barium, partitioned to the offgas system, where they
were removed prior to release from the stack
• Stack emissions were generally below the air emission limits, including total particulates and metals, except for mercury
• The process was shown to treat a wide variety of waste types
Costs:
Projected costs for full-scale system include:
• Capitals - $50 to $86.2 million for facility construction and outfitting
• Startup operating cost - $12 to $18 million
• O&M for a 5-yr period - $48 to $62 million
• Assuming 17,000 cubic meters of waste are treated, the projected unit cost for PHP is $7,400 to $10,800 per cubic
meter.
Description:
DOE sponsored a series of bench- and pilot-scale tests of the Plasma Hearth Process (PHP) at the STAR Center in
Idaho Falls, Idaho, conducted between 1993 and 1997. PHP is a high temperature thermal process that heats waste to a
molten form, which is then cooled into a glass/crystalline waste form. Three PHP systems were tested on a wide range of
wastes to evaluate the process for treating different wastes and to determine operating conditions. The three systems were
a nonradioactive bench-scale system (NBS), radioactive bench-scale system (RBS), and nonradioactive pilot-scale system
(NPS). The types of wastes tested included fly ash, organic and inorganic sludges, and a variety of debris; for the RBS
system, nonradioactive cerium was used as a surrogate for plutonium wastes.
The results showed that PHP was capable of treating a wide variety of radioactive and nonradioactive wastes, meeting the
RCRA LDR standards for metals and , with the exception of mercury, operating within the air emission requirements for
the systems. Differences were noted between the behavior of sludges and debris in the system, such as higher retention
rates for cerium oxide for sludges as compared to debris. Additional data are needed to better quantify the treatment of
debris using PHP. Other issues to be considered for full-scale deployment include additional data on the behavior of
radionuclides compared to the cerium surrogate, and a more detailed evaluation of PHP for high organic waste feeds.
185
-------
&EPA
United States
Environmental Protection
Agency
Solid Waste and
Emergency Response
(5102G)
EPA 542-R-99-006
September 1999
clu-in.org
Groundwater Cleanup:
Overview of Operating
Experience at 28 Sites
-------
EPA 542-R-99-006
September 1999
Groundwater Cleanup:
Overview of Operating Experience
at 28 Sites
U.S. Environmental Protection Agency
Office of Solid Waste and Emergency Response
Technology Innovation Office
Washington, DC 20460
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
NOTICE
This document was prepared for the U.S. Environmental Protection Agency (EPA) Technology Innovation Office
(TIO) by Tetra Tech EM Inc. under EPA Contract Number 68-W-99-003. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use. For more information about this project,
please contact: Linda Fiedler, U.S. Environmental Protection Agency, Technology Innovation Office,
401 M Street, S.W. (MC 5102G), Washington, D.C., 20460; telephone: (703) 603-7194; e-mail:
fiedler.linda@epa.gov.
This document may be obtained from EPA's web site at . A limited number of hard
copies of this document are available free of charge by mail from EPA's National Service Center for Environmental
Publications (NSCEP), at or at the following address (please allow four to six weeks for
delivery):
U.S. EPA/National Service Center for Environmental Publications
P.O. Box 42419
Cincinnati, OH 45242
Telephone: (513) 489-8190 or (800) 490-9198
Facsimile: (513) 489-8695
ACKNOWLEDGMENTS
Special acknowledgment is given to the remedial project managers, potentially responsible parties, and vendors involved
at the case study sites for their thoughtful suggestions and support in preparing the individual case studies and in
contributing to this report.
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
EXECUTIVE SUMMARY
This study examined operating experiences at 28 sites across the United States at which
completed or ongoing groundwater cleanup programs are in place. Although not a representative
sample, the sites present a range of the types of cleanups typically performed at sites with
contaminated groundwater. At 21 of the sites, pump-and-treat (P&T) systems were used alone as
the remediation technology; at two of the sites, permeable reactive barriers (PRBs), an in situ
technology, were used alone as the remediation technology. In addition, in situ technologies
were used in conjunction with P&T at five sites, including one site with P&T that was replaced
with a PRB. Individual reports have previously been published for each of the 28 sites by the
Federal Remediation Technology Roundtable and are available at .
Of the 28 case study sites, 24 are Superfund remedial actions, one is a Superfund removal action,
one is a state cleanup, and two are Resource Conservation and Recovery Act (RCRA) corrective
actions. Chlorinated solvents are the type of contaminant most frequently present, found at 21 of
the 28 sites. The sites are located throughout the U.S. and include a range of site types and
hydrogeological conditions. For example, nonaqueous phase liquids (NAPL) were observed or
suspected to be present at 18 of the 28 sites, and hydraulic conductivity varied among the sites by
more than six orders of magnitude.
This report summarizes information about the groundwater remediation systems at the 28 sites,
including: design, operation, and performance of the systems; capital, operating, and unit costs
of the systems; and factors that potentially affect the cost and performance of the systems. Data
from the case studies are compared and contrasted to assist those involved in evaluating and
selecting remedies for groundwater contamination at hazardous waste sites.
Data on performance through late 1997/early 1998 compiled for the report show that total
contaminant removal at the case study sites ranged from seven pounds to more than 510,000
pounds with a median contaminant mass removal of 2,000 pounds. The average annual volume
of groundwater treated ranged from 1.7 million to 550 million gallons (at P&T sites).
Although remediation has been completed at only two of the 28 sites, at the 26 sites with ongoing
remediation, progress has been made toward achieving cleanup goals, including: reducing the
size of a contaminated plume; reducing or eliminating a hot spot within a plume; reducing the
concentrations of contaminants within a plume; removing contaminant mass from a plume; and
achieving containment of a plume.
Capital and operating and maintenance (O&M) costs at the case study sites through late
1997/early 1998 also were compiled for this report. Although P&T and PRB systems may be
designed to accomplish similar remedial goals, the spatial area of groundwater they treat is
generally different; therefore, their costs are presented separately in this report. For the 26 P&T
systems, the approximate median capital cost was $1.9 million and the median average annual
operating cost was $190,000; with median unit costs of $96 of capital cost per average 1,000
gallons of groundwater treated per year and $18 of average annual operating cost per average
1,000 gallons of groundwater treated per year. For the three PRB systems, the approximate
median capital cost was $500,000 and the median average annual operating cost was $85,000;
with median unit costs of $520 of capital cost per average 1,000 gallons of groundwater treated
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
per year and $84 of average annual operating cost per average 1,000 gallons of groundwater
treated per year.
Since the sites summarized in this report were not selected as a representative sample of all
groundwater cleanup sites, the medians, averages, and ranges calculated in this report should not
be used to draw generalizations about cost and performance at other groundwater cleanup sites.
Results of analyses of the case studies showed that the factors affecting cost and performance and
the extent of the effect of those factors varied from site to site. However, based on the
information provided for the 28 case study sites and general observations of groundwater cleanup
as a whole, the following factors have a significant effect on the cost and performance of
groundwater remediation systems.
• Source control factors - Method, timing, and success of source controls to
mitigate contact of NAPLs or other contaminant sources, such as highly
contaminated soil, with groundwater
• Hydrogeologic factors - Aquifer properties that define contaminant transport and
groundwater extraction system design needs, including hydraulic connection of
aquifers that allows for multi-aquifer contamination, aquifer flow parameters,
influences from adjacent surface water bodies on the aquifer system, and
influences of adjacent groundwater production wells on the aquifer system
• Contaminant property factors - Contaminant properties that define the relative
ease that contaminants may be removed from the aquifer, the steps that are
required to treat the extracted groundwater, and the complexity of the contaminant
mixture
• Extent of contamination factors - The magnitude of the contaminated
groundwater plume, including the plume area and depth and the concentrations of
contaminants within the plume
• Remedial goal factors - Regulatory factors that affect the design of a remedial
system and/or the duration that it must be operated, including defining aquifer
restoration or treatment system performance goals and specific system design
requirements, such as disallowing reinjection of treated groundwater or specifying
the treatment technology to be used
• System design and operation factors - The adequacy of a system design to
remediate the site, system downtime, system optimization efforts, the amount and
type of monitoring performed, and the use of in situ technology to replace or
supplement a P&T system
Specific examples of how each of these factors affected the cost and performance of the
groundwater remediation systems at the case study sites are cited within this report.
IV
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
TABLE OF CONTENTS
Section Page
EXECUTIVE SUMMARY iii
LIST OF EXHIBITS vi
ACRONYMS AND ABBREVIATIONS vii
1.0 INTRODUCTION 1-1
2.0 OVERVIEW OF 28 CASE STUDY SITES 2-1
3.0 DESIGN AND OPERATION OF REMEDIAL SYSTEMS
AT 28 CASE STUDY SITES 3-1
3.1 Technology Descriptions 3-1
3.2 Remedial System Designs 3-3
3.3 System Operation 3-6
3.4 System Optimization and Modifications 3-8
4.0 PERFORMANCE OF REMEDIAL SYSTEMS AT 28 CASE STUDY SITES 4-1
4.1 Remedial Goals 4-1
4.2 Progress Toward Goals 4-6
5.0 COST OF REMEDIAL SYSTEMS AT 28 CASE STUDY SITES 5-1
6.0 FACTORS THAT AFFECTED COST AND PERFORMANCE OF REMEDIAL
SYSTEMS AT 28 CASE STUDY SITES 6-1
7.0 REFERENCES 7-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
LIST OF EXHIBITS
Exhibit Title
2-1 Summary of 28 Case Study Sites
2-2 Remediation Systems - Years of Operation
2-3 Site Types and Locations
2-4 Categories of Contaminants Treated at 28 Sites
2-5 Specific Contaminants Treated at 28 Sites
2-6 Initial Volume of Contaminated Groundwater Plumes at 24 Sites
2-7 Presence of NAPLs at 28 Sites
2-8 Pertinent Hydrogeological Data at 28 Sites
3-1 Summary of Technologies Used at 28 Sites
3-2 Remedial Technologies Used at 28 Sites
3-3 Pump-and-Treat System Designs at 26 Sites
3-4 Designs of In Situ Treatment Systems at Seven Sites
3-5 Operation of Remedial Systems at 28 Sites
3-6 Types of Optimization and Modification Efforts at 28 Sites
3-7 System Optimization and Modification Efforts Conducted at 28 Sites
4-1 Summary of System Performance for 28 Sites
4-2 Unit Contaminant Mass Removed at 26 Sites
4-3 Summary of Average Contaminant Concentration Reduction at 17 Sites
4-4 System Performance Summary
5-1 Summary of Cost Data for 28 Sites
5-2 Summary of Remedial Cost and Unit Cost Data for 28 Sites
5-3 Average Operating Cost Per Year at 28 Sites
5-4 Capital Cost Per 1,000 Gallons of Groundwater Treated Per Year
5-5 Average Annual Operating Cost Per 1,000 Gallons of Groundwater
Treated Per Year
6-1 Factors Affecting Cost and Performance of Groundwater Remediation
Systems
VI
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
ACRONYMS AND ABBREVIATIONS
AS Air sparging
ACL Alternate concentration limit
BTEX Benzene, toluene, ethylbenzene, and xylene
DCE Dichloroethene
DNAPL Dense nonaqueous phase liquid
DoD U.S. Department of Defense
DOE U.S. Department of Energy
EPA U.S. Environmental Protection Agency
ISB In situ bioremediation
LNAPL Light nonaqueous phase liquid
MCL Maximum contaminant level
NAPL Nonaqueous phase liquid
NPV Net present value
OSWER Office of Solid Waste and Emergency Response
P&T Pump and treat
PAH Polycyclic aromatic hydrocarbon
PCB Polychlorinated biphenyl
POTW Publicly-owned treatment works
PRB Permeable reactive barrier
RCRA Resource Conservation and Recovery Act
ROD Record of Decision
SVOC Semivolatile organic compound
TCE Trichloroethene
TIO Technology Innovation Office
USCG United States Coast Guard
VCB Vertical containment barrier
VOC Volatile organic compound
VII
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
VIM
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
1.0 INTRODUCTION
Groundwater contamination is present at the majority of Superfund and Resource Conservation
and Recovery Act (RCRA) corrective action sites. Groundwater remediation technologies
currently in use to clean up these sites include pump-and-treat (P&T) systems, and in situ
technologies such as bioremediation, permeable reactive barriers, and air sparging. As part of an
effort by the Federal Remediation Technologies Roundtable,1 the U.S. Environmental Protection
Agency (EPA) has prepared 28 case studies of ongoing and completed groundwater remediation
projects. The Roundtable has published these case studies, along with 112 other case studies
about a wide range of technologies, which are available through the Internet at
, or in hard copy through the EPA National Service Center for
Environmental Publications (NSCEP). Case studies are about 10-20 pages in length and contain
information about site background, extent of contamination, technology design and operation,
performance, cost, observations and lessons learned, and points of contact for further
information.
The objective of this report is to provide a summary of information about the 28 groundwater
remediation case studies, including comparing results among sites, to further assist those
involved in evaluating and selecting remedies for groundwater contamination at hazardous waste
sites. The case studies present a range of the type of cleanups typically performed at
groundwater-contaminated sites, and include 21 sites with P&T systems alone, four sites with
P&T systems supplemented with in situ technologies, one site with P&T that was replaced by an
in situ technology, and two systems with only in situ technologies. The majority of the case
studies are ongoing projects, with remediation completed at two of the sites.
The report presents an overview of each of the case study sites (Section 2); describes the design
and operation of the remediation systems, including efforts to optimize the systems (Section 3);
summarizes the performance of each of the systems at the sites, including final results for
completed remediations and progress towards goals for ongoing projects (Section 4); examines
the costs for these systems, including capital, operating, and unit costs (Section 5); and examines
the factors that potentially affect the cost or performance of the remediation systems (Section 6).
References used to prepare this report are listed in Section 7 and are cited in parentheses.
As described in Section 2 of this report, the case study sites were selected in part on the basis of
availability of information. Therefore, it is important to note that the case studies are not
intended to be a representative sample of groundwater remediation projects; rather, they present a
range of the types of systems that are being used at Superfund and RCRA corrective action sites.
Further, this report is not intended to revise or update EPA policy or guidance on how to clean up
sites with contaminated groundwater.
1 The Federal Remediation Technologies Roundtable consists of senior executives from eight agencies
with an interest in site remediation, including the U.S. Army, the U.S. Navy, the U.S. Air Force, the U.S.
Department of Energy (DOE), and EPA. The Roundtable, which was created to build a more collaborative
atmosphere among federal agencies involved in the remediation of hazardous waste sites, has on ongoing effort to
improve the type and availability of cost and performance information for site remediation technologies. This
information is being provided to assist those involved in evaluating and selecting remedies for hazardous waste
cleanups.
1-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
1-2
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
2.0 OVERVIEW OF 28 CASE STUDY SITES
The 28 groundwater case study sites
included in this report were selected
from a list of candidate sites that was
developed using information from
previous work by EPA and
recommendations by EPA regional staff.
The following criteria were used in
selecting the specific sites:
>• Sites are located throughout the U.S.
and include a range of site types and
hydrogeological conditions
*• At some sites, groundwater
remediation has been ongoing since
the late 1980s
*• Chlorinated solvents are the most
common contaminant
• Sites at which groundwater cleanup systems had been operated for a relatively
long period of time
• Sites for which aquifer cleanup goals (not only containment goals) had been
established
• Sites for which sufficient cost and performance data were available
Exhibit 2-1 summarizes general information about each of the 28 sites, such as duration and
status of remediation, categories of contaminants targeted for treatment, type of cleanup, project
lead, and highlights of the project. Of the 28 sites, 24 are Superfund remedial actions, one is a
Superfund removal action, one is a state cleanup, and two are Resource Conservation and
Recovery Act (RCRA) corrective actions. Of the 25 Superfund sites, four are EPA led, one is
U.S. Navy led, 11 are potentially responsible party (PRP) led, and nine are state led. The sites
have been grouped by the type of contamination that was targeted for cleanup at each (volatile
organic compounds [VOC], VOCs combined with other contaminants, or metals).
Exhibit 2-2 presents the years of operation at each site. Groundwater remediation at most of the
case study sites is ongoing, with systems operating over periods ranging from two years (USCG
Center andMqffetf) to 11 years (DesMoines and Former Intersil). Cleanup has been completed
at two of the sites (Firestone and Gold Coast) and the remediation systems at three other sites
(French Ltd., Sol Lynn, and Sylvester/Gilson Road) have been shut down for various reasons,
although the cleanups at these sites are not considered complete. At one site (Western
Processing), the goal has been changed from restoration to containment. Nine of the 28 systems
have been operating since the late 1980s. For sites at which systems are ongoing, the information
presented in the report is current as of late 1997 or early 1998.
Exhibit 2-3 shows the type of site and the relative location of each site.
2-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-1: Summary of 28 Case Study Sites
Site Name, Location,
CERCLIS ID no.
VOCs CONTAMINATION
City Industries SF Site, FL
(City Industries)
CERCLIS #FLD055945653
Des Moines TCE SF Site, IA
(Des Moines)
CERCLIS #IAD98060687933
Former Firestone Facility SF
Site, CA
(Firestone)
CERCLIS #CAD990793887
Former Intersil Inc., CA
(Intersil)
French, Ltd. SF Site, TX
(French, Ltd.)
CERCLIS #TXD980514814
Gold Coast SF Site, FL
(Gold Coast)
CERCLIS #FLD071 307680
JMT Facility RCRA Site (formerly
Black & Decker), NY
(JMT)
Duration/Years
of System
Operation1
4.0
10.5
7.0
10.5
4.0
3.5
10.0
Remediation
Status
Ongoing
Ongoing
Complete
Ongoing
Monitored
Natural
Attenuation
Complete
Ongoing
Contaminant
Categories Targeted
for Treatment
VOCs
VOCs
VOCs
VOCs
VOCs
VOCs
VOCs
Type of
Cleanup2
SF Remedial
SF Remedial
SF Remedial
State Cleanup
SF Remedial
SF Remedial
RCRACA
Lead(s)
PRP
PRP
PRP
PRP
PRP
EPA
Owner/Operator
Site Highlight(s)
Simple hydrogeology with relatively high
hydraulic conductivity; pumping optimization
modeling used
Approximately 5 billion gallons treated to
date to contain and remediate contaminated
groundwater; dense nonaqueous phase liquid
(DNAPL) suspected
Groundwater cleanup completed in seven
years
Used P&T for eight years; replaced that
technology with permeable reactive barrier to
minimize cost of treatment while increasing
effectiveness of treatment, and to return site to
leasable or sellable conditions
Regulatory requirements set as demonstrating,
through modeling, that cleanup goals would
be met at site boundary via monitored natural
attenuation 10 years after P&T is completed
Air sparging used to remediate recalcitrant
area of contamination at the end of the
cleanup, optimization modeling used
Included use of an artificially produced
fracture zone in the bedrock
Notes:
1 Years of system operation as of end of June 1998
2 SF indicates Superfund site; RCRA CA indicates RCRA corrective action site
2-2
Table Continued.
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-1: Summary of 28 Case Study Sites
Site Name, Location,
CERCLIS ID no.
Keefe Environmental Services SF
Site, NH (Keefe)
CERCLIS #NHD092059112
Moffett Federal Airfield SF Site,
CA (Moffett)
Mystery Bridge at Highway 20 SF
Site, WY (Mystery Bridge)
CERCLIS #WYD98 1546005
Old Mill SF Site, OH
(Old Mill)
CERCLIS #OHD9805 10200
SCRDI Dixiana SF Site, SC
(SCRDI Dixiana)
CERCLIS #SCD98071 1394
Site A (Confidential SF Site), NY
(Site A)
CERCLIS #Confidential
Sol Lynn/Industrial Transformers
SF Site, TX (Sol Lynn)
CERCLIS #1X0980973327
Solid State Circuits SF Site, MO
(Solid State)
CERCLIS #MOD9808854111
U.S. Aviex SF Site, MI
(U.S. Aviex)
CERCLIS #MID980794556
Duration/Years
of System
Operation1
5.5
2.0
4.5
9.0
6.0
3.0
3.0
5.4
5.0
Remediation
Status
Ongoing
Pilot Scale
Ongoing
Ongoing
Ongoing
Ongoing
Ongoing
Shut Down
Pending
Study
Ongoing
Ongoing
Contaminant
Categories Targeted
for Treatment
VOCs
VOCs
VOCs
VOCs
VOCs
VOCs
VOCs
VOCs
VOCs
Type of
Cleanup2
SF Remedial
SF Remedial
SF Remedial
SF Remedial
SF Remedial
SF Remedial
SF Remedial
SF Remedial
SF Remedial
Lead(s)
State
U.S. Navy
EPA
EPA
EPA '92-'94
PRP '95-present
State
State
State
EPA '93-'96
State '96-
present
Site Highlight(s)
Major modifications to system design based
on optimization study
Permeable reactive barrier successful in
reducing trichloroethene (TCE)
concentrations; increased monitoring required
for technology certification and validation
Monitored natural attenuation used for
remedy of the off-site portion of the plume
System of trenches used to extract shallow
groundwater
Complex hydrogeology; major modifications
made in system by PRP
Remedial system included use of P&T
supplemented with air sparging and in situ
bioremediation
Multiaquifer contamination (three aquifers);
additional contamination identified after
remediation began
Complex hydrogeology (leaky artesian system
in a Karst formation)
Performance modeling used for system
optimization
Notes:
1 Years of system operation as of end of June 1998
2 SF indicates Superfund site; RCRA CA indicates RCRA corrective action site
2-3
Table Continued.
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-1: Summary of 28 Case Study Sites
Site Name, Location,
CERCLIS ID no.
Duration/Years
of System
Operation1
Remediation
Status
Contaminant
Categories Targeted
for Treatment
Type of
Cleanup2
Lead(s)
Site Highlight(s)
VOCs COMBINED WITH OTHER CONTAMINANTS
Baird and McGuire SF Site, MA
(Baird and McGuire)
CERCLIS #MADOO 104 1987
King of Prussia Technical
Corporation SF Site, NJ
(King of Prussia)
CERCLIS #NJD980505341
LaSalle Electrical SF Site, IL
(LaSalle)
CERCLIS #SCD980711394
Libby Groundwater SF Site, MT
(Libby)
CERCLIS #MTD980502736
Mid-South Wood Products SF
Site, AR (MSWP)
CERCLIS #ARD092916188
Solvent Recovery Services of New
England, Inc. SF Site, CT
(Solvent Recovery Service)
CERCLIS #CTD009717604
Sylvester/Gilson Road SF Site, NH
(Sylvester/Gilson Road)
CERCLIS #NHD099363541
5.5
3.5
5.5
7.0
9.0
3.0
9.5
Ongoing
Ongoing
Ongoing
Ongoing
Ongoing
Ongoing
Shut Down
Pending
Explanation
of Significant
Difference
(ESD)
VOCs, semivolatile
organic compounds
(SVOCs), pesticides,
metals
VOCs, metals
VOCs,
polychlorinated
biphenyls (PCBs)
VOCs, SVOCs
VOCs, SVOCs
VOCs, metals
VOCs, pesticides,
metals
SF Remedial
SF Remedial
SF Remedial
SF Remedial
SF Remedial
Removal
SF Remedial
EPA
PRP
State
PRP
PRP
PRP
State
Complex mixture of contaminants requiring
extensive treatment train
Complex mixture of contaminants requiring
extensive treatment train
Relatively low groundwater flow; DNAPLs
present
Light nonaqueous phase liquids (LNAPLs)
and DNAPLs perpetuate elevated levels of
contaminants in groundwater
System optimization performed after eight
years of operation; contamination reduced to
one localized area of concern
Complex mixture of contaminants having
various properties led to extensive treatment
train; DNAPLs present
Modifications of the system were costly;
system shut down since 1996, pending an
ESD to raise the alternate concentration limit
(ACL) for 1,1-dichloroethane to greater than
method detection limit
Notes:
1 Years of system operation as of end of June 1998
2 SF indicates Superfund site; RCRA CA indicates RCRA corrective action site
2-4
Table Continued.
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-1: Summary of 28 Case Study Sites
Site Name, Location,
CERCLIS ID no.
USCG Support Center, NC
(USCG Center)
Western Processing SF Site, WA
(Western Processing)
CERCLIS #WAD009487514
Duration/Years
of System
Operation1
2.0
10.0
Remediation
Status
Ongoing
Ongoing
Contaminant
Categories Targeted
for Treatment
VOCs, metals
VOCs, metals
Type of
Cleanup2
RCRACA
SF Remedial
Lead(s)
Owner/Operator
PRP
Site Highlight(s)
Use of PRB to treat groundwater
contaminated with TCE and hexavalent
chromium; extensive sampling conducted to
evaluate
Goals for off-site plume met; on-site system
modified to provide containment of on-site
contamination rather than site restoration;
NAPL observed and suspected in various
areas of the site
METALS CONTAMINATION
Odessa Chromium I SF Site, TX
(Odessa I)
CERCLIS #1X0980867279
Odessa Chromium IIS SF Site, TX
(Odessa IIS)
CERCLIS #1X0980697114
United Chrome SF Site, OR
(United Chrome)
CERCLIS #ORD009043001
4.5
4.5
10.0
Ongoing
Ongoing
Ongoing
metals
metals
metals
SF Remedial
SF Remedial
SF Remedial
State
State
PRP
Low groundwater production; electrochemical
treatment for chromium required by Record of
Decision (ROD)
Relatively low groundwater production;
multiaquifer contamination; electrochemical
treatment for chromium required by ROD
Contaminant concentrations reduced to the
point at which extracted groundwater can be
discharged to the publicly-owned treatment
works (POTW) without on-site treatment;
major modifications made in extraction
system
Notes:
1 Years of system operation as of end of June 1998
2 SF indicates Superfund site; RCRA CA indicates RCRA corrective action site
2-5
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-2: Remediation Systems - Years of Operation
Site Name, Location
Former Firestone Facilty SF Site, CA
Sylvester/Gibon Road SF Site, NH
Former Intersil Inc., CA
Des Moines ICE SF Site, IA
JMT Facility RCRA Site (formerly Black & Decker), NY
United Chrome SF Site, OR
Western Processing SF Site, WA
Old Mill SF Site, OH
Mid-South Wood Products SF Site, AR
Gold Coast SF Site, FL
Libby Groundwater SF Site, MT
French, Ltd. SF Site, TX
SCRDI Dixiana SF Site, SC
LaSalfe Electrical SF Site, IL
Sold State Circuits SF Site, MO
BairdandMcGuireSFSite,MA
Keefe Environmental Services SF Site, NH
U.S.AviexSFSite, Ml
Sol Lynn/Industrial Transformers SF Site, TX
Odessa Chromium 1 SF Site, TX
Odessa Chromium IIS SF Site, TX
Mystery Bridge at Highway 20 SF Site, WY
City Industries SF Site, FL
King of Prussia Technical Corporation SF Site, NJ
1986
Solvent Recovery Services of New England, Inc. SF Site, CT
Site A (Confidential) SF Site, NY
Moffett Federal Airfield SF Site, CA
USCG Support Center, NC
1987
1988
1989
1990
1991
1992
1993
©
1994
^ff^^^^^SS)
1995
1996
1)
(D
^^™^^™^^"™-
^
^*
+*
+»
•^^•^*
Operating
7
9.5
10.5
10.5
10
10
10
9
9
3.5
7
4
6
5.5
5.5
5.5
5.5
5
3
4.5
4.5
4.5
4
3.5
3
3
2
2
Notes:
- Indicates ongoing cleanups.
Groundwater cleanup is complete.
(2) Sylvester/Gilson Road system was shut down pending an Explanation of Significant Difference.
(3) French Limited cleanup continues by natural attenuation since 12/95. Cleanup is not complete.
(4) Sol Lynn system was shut down for maintenance and upgrade in 10/96. Cleanup is not complete.
2-6
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-3: Site Types and Locations
Western Processing SF Site
(waste processing) , -,,
Kent,WA [ ft
Libby Groundwater SF Site
lumber mil/wood preserving)
,MT
Moffett Federal
Airfield SF Site
(service and support
for Navy aircraft)
Mountain View, CA
Former Firestone
Facility SF Site
(manufacturing)
Salnas, CA
Old Mil SF Site
(illegal waste disposal)
Rock Creek, OH
JMT Facilty RCRA Site (formerly
Black & Decker)
(appiance manufacturing)
Brockport, NY
Mystery Bridge at Highway 20 SF
Site
(oil & gas production equipment)
EvansviHe, WY
U.S.AviexSFSite
(automotive fluids
manufacturing)
Miles, Ml
Des Moines TCE SF Site
(metal wheel and brake
manufacturing)
Des Moines, IA
A 1
LaSalje Electrical SF Site
(electrical equipment
manufacturing)
LaSalle.lL
Solid State Circuits SF Site
(circuit board manufacturing)
Republic, MO
Former Intersil Inc.,
(semiconductor manufacturing)
Sunnyvale, CA
A
SCRDI Dixiana SF Site
(waste storage)
Cayce, SC
Sol Lynn/Industrial
Transformers SF Site
(scrap metal and transformer
reclamation)
French, Ltd. SF Site
(industrial waste disposal)
Crosby, TX
Sylvester/Gilson Road SF Site
(ilegal waste disposal)
Nashua, NH i
Keefe Environmental
Services SF Site
(spent solvent
reclamation)
Epping, NH
Baird and McGuire SF Site
(chemical mixing
and batching)
Holbrook, MA
Solvent Recovery Services of
New England, Inc. SF Site
(solvents recovery)
Southington, CT
Site A (Confidential) SF Site
(petroleum bulking, chemical
mixing)
Long Island, NY
King of Prussia Technical
Corporation SF Site
(waste disposal and recycling)
Winslow, NJ
USCG Support Center
(electroplating operations)
Elizabeth City, NC
Odessa Chromium I SF Site
(chrome plating)
Odessa, TX
Odessa Chromium IIS SF Site
(chrome plating)
Odessa, TX
City Industries SF Site
(hazardous waste disposal)
Orlando, FL
.Gold Coast SF Site
(spent oil and solvent reclamation)
Miami, FL
2-7
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-4 summarizes the types of contaminants treated at the 28 sites. The contaminants fall
into the following categories. Multiple contaminant category groups have been targeted for
treatment at some sites.
• Volatile organic compounds (VOCs)
* Chlorinated VOCs
*• Benzene, toluene, ethylbenzene, and xylene (BTEX)
»> Other VOCs
• Semivolatile organic compounds (SVOCs)
>• Pesticides
*• Polycyclic aromatic hydrocarbons (PAHs)
>• Polychlorinated biphenyls (PCBs)
Other SVOCs
• Metals
Chlorinated VOCs were the type of contaminant most frequently present, found at 21 of the 28
sites.
Exhibit 2-5 summarizes information about the specific contaminants addressed at the sites. Only
contaminants that were treated at more than one site are included in this exhibit. Six of the 10
most common contaminants treated were chlorinated VOCs, with trichloroethene (TCE), treated
at 18 sites, the most common. Benzene was the most commonly treated nonchlorinated VOC (at
five sites). Chromium was the most common metal, treated at seven of the sites.
Exhibit 2-6 shows the volume of the contaminated groundwater plume at each site. For most of
the sites, the extent of contamination was quantified by the volume of contaminated groundwater.
Plume volume presented for these sites generally represents one pore volume of the contaminated
plume prior to commencing groundwater cleanup activities at the site. The volume was
calculated by the site contractors or during the preparation of the cost and performance reports by
combining isoconcentration data for groundwater contaminants with reported or typical
hydrogeological data. The volume of contaminated groundwater at the sites ranged from 930,000
gallons (Intersil) to 5.6 billion gallons (Moffetf). The average volume of the contaminated plume
was 440 million gallons, and the median volume was 29 million gallons.
2-8
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-4: Categories of Contaminants Treated at 28 Sites
Chlorinated VOCs
BTEX
Other VOCs
Pesticides
PAHs
PCBs
Other SVOCs
Metals
20
10 15 20
Number of Case Study Sites
25
Exhibit 2-5: Specific Contaminants Treated at 28 Sites
^u
18 -
jjj M 16-
^f 10-
il ?-
-' \-
n
ia
15
9 9 „
—
7 7
n n 5 ^ , ,
II 99999999
HnHnnnnnnnn
/&jy&/3$tf&s/q&
*? ^ ^ ^ *$•
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-6: Initial Volume of Contaminated Groundwater Plumes at 24 Sites1
***. "
«*< :
V*6 •
$< :
*
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-7 summarizes information on the observed and suspected presence of nonaqueous
phase liquid (NAPL) at the sites either as light nonaqueous phase liquid (LNAPL), which
generally floats on the water table, or dense nonaqueous phase liquid (DNAPL), which typically
sinks through permeable media (including saturated materials) to an impermeable barrier. Of the
28 sites: NAPLs were observed or suspected to be present at 18; DNAPL only was observed or
suspected at 12 sites; LNAPL only was observed or suspected at three sites; and both DNAPL
and LNAPL was observed or suspected at three sites. As described in Estimating Potential for
Occurrence of DNAPL at Superfund Sites [6], NAPL can be "suspected" at a site if its
components are present in groundwater at greater than one percent of either their pure-phase
solubility or effective solubility. For the case study sites, NAPL was considered "suspected" if at
lease one component was present at greater than one percent of its pure-phase solubility.
Exhibit 2-7: Presence of NAPLs at 28 Sites
Site Name and Location
Baird and McGuire, MA
City Industries, FL
Des Moines, IA
Firestone, CA
French, Ltd., TX
Gold Coast, FL
Intersil, CA
JMT,NY
Keefe, NH
King of Prussia, NJ
LaSalle, IL
Libby, MT
Moffett, CA
MSWP, AR
Mystery Bridge, WY
Odessa I, TX
Odessa IIS, TX
Old Mill, OH
SCRDI Dixiana, SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent Recovery Service, CT
Sylvester/Gilson Road, NH
U.S. Aviex, MI
United Chrome, OR
USCG Center, NC
Western Processing, WA
Number of Sites
NAPL Observed or Suspected
DNAPL
Observed
•
•
•
•
•
•
•
7
Suspected1
•
•
•
•
•
•
•
•
8
LNAPL
Observed
•
•
•
•
•
•
6
Suspected1
0
Note:
Suspected NAPL was identified in the case study reports when contaminants were present at more
than one percent of their either their pure-phase solubility or effective solubility.
2-11
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 2-8 presents information about the hydrogeologic conditions at the 28 sites. The average
hydraulic conductivity of the contaminated water-bearing layer(s) at the sites varied by more than
six orders of magnitude (0.023 feet per day (ft/day) to 1,200 ft/day). At more than half of the
sites, contamination was present in multiple water-bearing layers or aquifers. Seven of the sites
exhibited vertical groundwater flow between aquifers, 13 sites were influenced by adjacent
bodies of surface water, and eight sites were influenced by the presence of production wells (for
example, municipal). Reported depths to the top of contaminated aquifers ranged from zero (at
ground surface) to 45 ft below ground surface (bgs). Additional detail on the hydrogeologic
conditions, such as aquifer type, lithology, and degree of heterogeneity at the sites can be found
in the case studies.
Exhibit 2-8: Pertinent Hydrogeological Data at 28 Sites
Site Name and Location
Baird and McGuire, MA
City Industries, FL
Des Moines, IA
Firestone, CA
French, Ltd., TX
Gold Coast, FL
Intersil, CA
JMT,NY
Keefe, NH
King Of Prussia, NJ
LaSalle, IL
Libby, MT
Moffett, CA
MSWP, AR
Mystery Bridge, WY
Odessa I, TX
Odessa IIS, TX
Old Mill, OH
SCRDI Dixiana, SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent Recovery
Service, CT
Sylvester/Gilson Road, NH
U.S. Aviex, MI
United Chrome, OR
USCG Center, NC
Western Processing, WA
Hydraulic
Conductivity
Range (ft/day)
3-45
6.39
535
100-1200
0.28-2.8
40
370
0.65-0.93
42.5
Variable
0.22
100-1000
0.3-400
Variable
340
1.6-5.1
1.6-5.1
0.22-1.25
10
53.5
0.14-25.5
0.023-1.62
0.023-300
30-50
9.1-45.4
0.5-60
11.3-25.5
1-100
Multiple
Aquifer
Contamination
Y
Y
N
Y
Y
Y
N
Y
N
N
N
Y
Y
Y
N
Y
Y
N
Y
N
Y
Y
Y
Y
N
Y
N
Y
Vertical
GW
Flow
N
N
N
N
N
N
Y
N
N
N
Y
N
Y
N
N
N
N
N
N
Y
N
Y
N
Y
N
Y
N
N
Surface
Water
Influence
Y
N
Y
N
Y
Y
N
N
N
Y
N
Y
N
N
N
N
N
Y
N
Y
N
Y
Y
Y
N
N
Y
Y
Production
Wells in
Area
N
N
Y
Y
N
Y
N
N
N
N
N
Y
N
N
N
Y
Y
N
N
N
N
Y
N
N
Y
N
N
N
Depth to
Contaminated
Aquifer (ft bgs)
10-15
NR
10-25
NR
10-12
5.0
NR
10.0
NR
15.0
3-5
10-20
5.0
NR
14-42
30-45
30-45
5.0
14.0
15-18
20-25
NR
NR
NR
20.0
0-10
6.0
5-10
Notes:
GW = Groundwater
NR = Not recorded in case studies
2-12
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
3.0 DESIGN AND OPERATION OF REMEDIAL SYSTEMS AT 28 CASE
STUDY SITES
Most of the sites used P&T alone;
five of the sites used P&T in
combination with in situ technologies
Eighteen of the P&T systems at the
sites used air stripping as
aboveground treatment; carbon
adsorption, metal removal, and
biological treatment also were used to
a lesser extent
The volume of groundwater treated
per year of operation at the P&T sites
ranged from 1.7 million to 554
million gallons
Optimization and modification efforts
have been made to some extent at all
of the sites
The design of the groundwater remediation
systems at the 28 sites include: pump-and-treat
(P&T) systems used alone as the remediation
technology at 21 sites; in situ technologies
(permeable reactive barriers [PRB] in these
cases) used alone as the remediation technology
at two sites, and in situ technologies, such as in
situ bioremediation, air sparging, or PRBs, used
in conjunction with or to replace P&T systems at
five sites. Source controls were identified at 23
of the sites. Vertical containment barriers (VCB)
were used at five of the sites to provide hydraulic
control of contaminant plumes.
The technologies used at the 28 case study sites
are described briefly below, followed by a
detailed summary of the remedial system designs implemented at the case study sites.
3.1 Technology Descriptions
Pump-and-Treat
P&T involves extracting contaminated groundwater through recovery wells or trenches and
treating the extracted groundwater by ex situ (aboveground) processes, such as air stripping,
carbon adsorption, biological reactors, or chemical precipitation. Variables in the design of a
typical P&T system include:
• The number and production rate of groundwater extraction points (determined by
such factors as the extent of contamination and the productivity of the
contaminated aquifer)
• The ex situ treatment processes employed (determined by such factors as system
throughput and the contaminants that require remediation)
• The discharge location for treatment plant effluent (determined by such factors as
location of the site and regulatory requirements)
Additional information about the fundamentals of P&T technology can be found in Design
Guidelines for Conventional Pump-and-Treat Systems [1].
3-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Air Sparging
Air sparging (AS) involves injecting a gas (usually air or oxygen) under pressure into the
saturated zone to volatilize contaminants in groundwater. Volatilized vapors migrate into the
vadose zone where they are extracted by vacuum, generally by a soil vapor extraction system.
AS also is used to supplement P&T systems. For example, AS may be added to remediate
specific portions of a contaminated plume that are not treated effectively by P&T alone or to
accelerate cleanups. For the purpose of this report, the use of air to promote biodegradation
(sometimes referred to as "biosparging") in saturated and unsaturated soils by increasing
subsurface concentrations of oxygen is referred to as in situ bioremediation.
Permeable Reactive Barriers
A PRB, or treatment wall, consists of an in-ground trench that is backfilled with a reactive
medium. The selection of the reactive medium is based on the targeted contaminants and the
hydrogeologic setting of the site. Zero-valent iron is the most common medium used in PRBs to
date. Examples of other reactive media include, microorganisms, zeolite, activated carbon, peat,
phosphate, bentonite, limestone, and amorphous ferric oxide. The treatment processes that occur
within the tremch are degradation, sorption, or precipitation of the contaminant. PRB systems
may be configured as "funnel and gate" designs; in such configurations groundwater flow is
routed by two or more impermeable walls through a permeable reactive zone.
PRBs may or may not be similar to P&T systems in both purpose and function. Like a P&T
system, PRBs may be used to treat contaminated groundwater at the boundary of a site, or to
restore the groundwater throughout a site. However, the volume of groundwater treated by a
PRB at a site is typically much lower than it would be for a P&T system at the same site because
PRBs treat only the groundwater that passes through the barrier, while P&T systems actively
extract groundwater from an aquifer, usually at multiple locations throughout the plume.
In Situ Bioremediation
In situ bioremediation (ISB) involves microbial degradation of organic constituents through
aerobic or anaerobic processes. In situ bioremediation includes processes by which nutrients
(such as nitrogen and phosphorus), electron donors (such as methane for aerobic processes or
methanol for anaerobic processes), or electron acceptors (such as oxygen for aerobic processes or
ferric iron for anaerobic processes) are added to the groundwater to enhance the natural
biodegradation processes. The addition of oxygen by biosparging is an example of such a
process.
Source Controls
Source controls include such activities as excavation of soil at hot spots, in situ treatment of soil
(for example, by soil vapor extraction), and installation of VCBs for control of NAPLs. Source
controls are implemented to remove source materials or to isolate them from contact with
groundwater.
3-2
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Vertical Containment Barriers for Hydraulic Control
VCBs, such as slurry or sheet pile walls, also are used to hinder the migration of contaminated
groundwater. VCBs are used in conjunction with groundwater extraction wells in an effort to
gain hydraulic control over a contaminated groundwater plume.
3.2 Remedial System Designs
Exhibit 3-1 lists the remedial technologies used at the 28 subject sites. At most of the sites (26 of
28), P&T systems were used for groundwater remediation. At five of the P&T sites, in situ
technologies were incorporated into the P&T approach. AS was incorporated at two sites, ISB at
three sites, and PRBs at one site. At two sites, PRBs were used alone as the remedial technology.
Source controls were used at most (24) of the sites, and VCBs were used for hydraulic control
five sites.
Exhibit 3-1: Summary of Technologies Used at 28 Sites
Technology
Total P&T Technologies
Total In Situ Technologies
P&T Only
P&T with In Situ Technology or Technologies
Air Sparging
In Situ Bioremediation
Air Sparging and In Situ Bioremediation
Permeable Reactive Barriers (replaced P&T)
In Situ Technology Only
Air Sparging
In Situ Bioremediation
Permeable Reactive Barriers
Vertical Containment Barriers for Hydraulic Control
Source Controls
Number of
Sites
26
7
21
5
1
2
1
1
2
0
0
2
5
24
Exhibit 3-2 identifies the specific remedial technology or technologies used at each of the sites.
Extracted groundwater at the sites was treated using treatment systems varying from an
individual ex situ technology to a complex series of different technologies. The ex situ treatment
technologies included:
Air stripping
Carbon adsorption
Filtration
Electrochemical removal of
metals
Oil/water separation
Chemical or ultraviolet
oxidation
Biological degradation
Neutralization
Equalization
3-3
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Vapor-phase treatment of off-gases from the ex situ technologies was employed at eight of the
sites. At those sites, vapor-phase incineration, carbon adsorption, filtration, or thermal oxidation
were used either individually or in series.
Exhibit 3-2: Remedial Technologies Used at 28 Sites
Site Name and
Location
Baird and McGuire,
MA
City Industries, FL
Des Moines, IA
Firestone, CA
French, Ltd., TX
Gold Coast, FL
Intersil, CA
JMT,NY
Keefe, NH
King of Prussia, NJ
LaSalle, IL
Libby, MT
Moffett, CA
MSWP, AR
Mystery Bridge, WY
Odessa I, TX
Odessa IIS, TX
Old Mill, OH
SCRDI Dixiana, SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent Recovery
Service, CT
Sylvester/Gilson Road,
NH
U.S. Aviex, MI
United Chrome, OR
USCG Center, NC
Western Processing,
WA
Total Sites
Remediation Technology
P&T (with ex situ treatment)
0!
0
•
•
•
•
4
STRIP
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
18
U
3
•
•
•
•
•
•
•
•
•
•
10
METAL
•
•
•
•
•
•
•
•
•
•
•
11
O
5
•
•
•
3
Other
Filtration
Equalization
Filtration
Neutralization;
addition of nutrients
and oxygen
Equalization;
clarification; filtration
Filtration
Filtration
Filtration
Settling; filtration;
addition of nutrients;
pH adjustment of
effluent
pH adjustment
Oxidation; filtration;
pH adjustment
Oxidation; filtration
tt
"f,
•
•
2
CO
to
HH
•
•
•
3
CO
e
•
•
•
3
CO
£
•
•
•
•
•
5
Source Control(s)
Implemented
Concurrent excavation
Prior excavation
Prior excavation
Prior excavation
Prior in situ
bioremediation of soil and
sludges
NA
Prior excavation
Prior excavation
Prior excavation
Prior soil washing
Prior excavation
Prior excavation
NA
Prior excavation
Prior excavation/S VE
NA
Prior excavation
Prior excavation
Prior excavation
Prior excavation
Prior excavation
Prior excavation
Prior excavation
Prior and concurrent
capping/slurry wall/
excavation/S VE
NA
Prior excavation
Prior excavation
NA
Notes:
OWS
STRIP
GAC
METAL =
BIO
Oil/water separation AS
Air stripping ISB
Granular activated carbon adsorption PRB
Physical or chemical removal of metal VCB
Biological treatment NA
Air sparging
In Situ bioremediation
Permeable reactive barrier
Vertical containment barrier
Not Available
3-4
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-3 identifies the extraction system design for each of the P&T sites. Groundwater
extraction designs at the 26 P&T sites varied in magnitude from one production well (JMT) to
several wells combined with trenches (MSWP and Old Mill) and to 210 vacuum wellpoints
(Western Processing). Pumping rates for the P&T systems ranged from 3 gallons per minute
(OldMill) to more than 1,000 gallons per minute (DesMoines).
Exhibit 3-3: Pump-and-Treat System Designs at 26 Sites
Site Name, Location
Baird and McGuire, MA
City Industries, FL
Des Moines, IA
Firestone, CA
French, Ltd., TX
Gold Coast, FL
Intersil, CA
JMT, NY
Keefe, NH
King of Prussia, NJ
LaSalle, IL
Libby, MT
MSWP, AR
Mystery Bridge, WY
Odessa I, TX
Odessa IIS, TX
Old Mill, OH
SCRDI Dixiana, SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent Recovery Service,
CT
Sylvester/Gilson Road, NH
U.S. Aviex, MI
United Chrome, OR
Western Processing, WA
Minimum
Maximum
Average
Median
Number of Wells
Extract
6
13
7
25
109
5
3
1
5
11
0
5
15
3
6
10
3
15
5
12
7
12
14
5
30
210
0
210
21
7
Inject
0
0
0
0
59
3
0
0
0
0
0
11
0
0
6
9
0
0
0
14
0
0
0
0
0
0
0
59
4
0
Number of
Trenches
Extract
0
0
0
0
0
0
0
0
1
0
3
0
8
0
0
0
5
1
0
0
0
0
0
0
0
0
0
8
0.7
0
Inject
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
7
0
0
1
0
7
0.4
0
Pumping
Rate
Average
gpm
60
105
1041
484
189
44
8
11.2
23.4
200
17
6.6
24
103
60
58.5
3.1
40
18
8
34
20
265
220
242
230
3.1
1041
140
51
Number Of
Wells/ Trenches
(by Location)
On-site
6
13
7
15
NR
5
4
1
1
4
3
5
10
3
NR
NR
NR
8
5
NR
4
6
14
1
NR
210
1
210
16
5
Off-
site
0
0
0
10
NR
0
0
0
5
7
0
0
5
0
NR
NR
NR
12
0
NR
3
6
0
4
NR
3
0
12
3
0
Treated
Groundwater
Discharge
E
,1
£.E
•
•
•
•
NR
•
•
•
•
•
•
•
•
•
Surface
Water
•
•
NR
•
•
•
•
•
•
•
•
•
2
•
NR
•
•
Notes:
NR = Not reported
3-5
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-3 shows that treated groundwater was reinjected into the aquifer (11 sites), discharged
to an adjacent surface water by a permitted outfall (10 sites), discharged to a publicly owned
treatment works (POTW) (2 sites), or discharged using a combination of these methods (2 sites).
Exhibit 3-4 describes the remedial systems at the seven case study sites where in situ
technologies were used.
Exhibit 3-4: Designs of In Situ Treatment Systems at Seven Sites
Site Name
and
Location
French,
Ltd., TX
Gold Coast,
FL
Intersil, CA
Libby, MT
Moffett,
CA
Site A, NY
USCG
Center, NC
In Situ
Technology (ies)
Used
ISB
AS
PRB
ISB
PRB
AS and ISB
PRB
Design of In Situ Treatment System
P&T system augmented with ISB. ISB system consisted of the reinjection of treated
groundwater into the contaminated aquifer. The treated groundwater was oxygenated
and amended with nitrogen and phosphorus before reinjection.
AS used only at end of cleanup to mitigate a small area of localized contamination.
Original design was a P&T system, which was turned off in 1995 after PRB was
installed. PRB system consisted of two parallel slurry walls 300 and 235 feet long and
1 3 feet deep used to funnel groundwater through a 40-foot- wide, 4-foot-thick permeable
wall of granular iron.
P&T system complemented with ISB. ISB system consisted of the reinjection of treated
groundwater into the contaminated aquifer. Treated groundwater was aerated and
amended with nitrogen and phosphorus in the treatment plant after removal of NAPL
and before it flowed through a series of fixed-film bioreactors.
PRB consisted of an impermeable "funnel" composed of two 20-foot-long sheet pile
walls. Reactive zone consisted of 6-foot- thick, 10-foot- wide, and 18-foot-high
(beginning 5 feet bgs) zone of granular iron. The reactive zone was located between two
zones of pea gravel, each two feet thick.
P&T system in conjunction with AS and ISB. AS system consisted of air injection
through 44 sparging wells at points approximately 10 feet below the water table, with
vapor collection through 20 soil vapor extraction wells (16 vertical and 4 horizontal).
ISB system consisted of the reinjection of treated groundwater into the contaminated
aquifer. The treated groundwater was amended with nitrogen and phosphorus before it
was discharged to the reinjection trench.
PRB consisted of a 2-foot-thick and 1 52-foot-long zone of approximately 450 tons of
granular zero-valent iron keyed into an underlying low conductivity layer at
approximately 22 feet bgs.
3.3 System Operation
Exhibit 3-5 presents data available on the operation of the remedial systems, including the
volume of groundwater treated and the percent of time the systems were operational. The
volume of groundwater treated per year of operation for the P&T systems ranged from 1.7
million gallons (OldMill) to 554 million gallons (DesMoines). Estimated throughput per year
for the PRB sites ranged from 200,000 gallons (Moffett) to 2.6 million gallons (USCG Center).
3-6
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-5: Operation of Remedial Systems at 28 Sites
Site Name and Location
Baird and McGuire, MA
City Industries, FL
Des Moines, IA
Firestone, CA
French, Ltd., TX
Gold Coast, FL
Intersil (P&T), CA3
Intersil (PRB), CA23
JMT,NY
Keefe, NH
King of Prussia, NJ
LaSalle, IL
Libby, MT
Moffett (PRB), CA2
MSWP, AR
Mystery Bridge, WY
Odessa I, TX
Odessa IIS, TX
Old Mill, OH
SCRDI Dixiana, SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent Recovery Service, CT
Sylvester/Gilson Road, NH
U.S. Aviex,MI
United Chrome, OR
USCG Center (PRB), NC2
Western Processing, WA
Minimum
Maximum
Average
Median
Volume of GW Extracted
(million gallons)1
Total
80
151.7
4900
1800
306
80
36
2
50.1
46
151.5
23
15.1
0.284
100.6
192.8
125
121
13
20.6
8.4
13
257
32.5
1200
329
62
2.6
974
0.284
4900
382.5
80
Per Year
21
50
554
266
76
22
5.0
1.1
5.2
11
57
5.2
2.9
0.2
12
54
30
30
1.7
4.5
6.7
4.3
62
11
126
96
7.2
2.6
119
0.2
554
57
12
Percent of Time
Operational (%)
93
90
95
97
90
95
98
100
90
97
76
75
89
100
NR
100
95
95
99
89
75
69
95
100
88
95
99
100
97
69
100
92
95
Notes:
1 At most of the sites, groundwater cleanups are in progress; therefore the values shown represent a portion
of the total volume treated. Data presented here generally are cumulative as of late 1997 or early 1998.
2 The volume of groundwater for PRB sites is equal to the volume of groundwater treated through the wall
at the site.
3 At the Intersil site, groundwater cleanup began with a P&T system; later a PRB was used. The two
phases were treated above as separate sites.
NR = Not reported
3-7
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
The percent of time that the remedial systems were operational at the sites ranged from 69 to 100
percent. Downtime reportedly was required for routine maintenance (such as changing carbon,
cleaning air stripper media, and backwashing filters) and issues specific to particular sites, including:
• Iron corrosion and clogging of extraction wells (Baird andMcGuire, Des Moines, Odessa I,
Odessa IIS, Mystery Bridge, and Solvent Recovery Services)
• Freezing or fouling of air stripper media (Solid State and City Industries)
• System modifications (Keefe, Solid State, Site A, MSWP, and Sylvester/Gilson Road)
• Equipment failures (Libby, French, Ltd., and King of Prussia)
• Brownouts (Keefe)
3.4 System Optimization and Modifications
Optimization and modification efforts that have been undertaken at the case study sites include
remedy refinement, pre-design modeling and testing, and system modifications. Exhibit 3-6
summarizes the types of optimization and modification efforts reported for the case study sites;
these are generally classified as pre-design and post-design efforts. Pre-design efforts at the case
study sites typically consisted of interim designs or systems (used at 13 sites) and groundwater
modeling (used at 11 sites). Post-design efforts consisted of optimization modeling (used at 13
sites), modifications of the groundwater extraction systems (used at twenty of the sites), and
modifications to the groundwater treatment systems (used at 15 sites). Exhibit 3-7 lists the
specific efforts made at each of the 28 sites.
At the time the case study reports were prepared, some of the sites at which remediation was
ongoing had identified plans for future system modifications. The following examples illustrate
these types of plans:
• At U.S. Aviex, further site characterization is needed and the remediation system may
require expansion.
• At City Industries, concentrations of contaminants in extracted groundwater may be low
enough to allow discharge directly to the POTW without prior treatment.
• At Sol Lynn, the system was shut down when extraction well pipes leaked and fouled, and
the extraction system had lost containment. Currently, the site is being reevaluated to
identify alternative remedial plans to address the issues with the extraction system.
3-8
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-6: Types of Optimization and Modification Efforts at 28 Sites1
Pre-Design
Interim Design or
System
• Pilot-scale system
• Demonstration
system
• Staged approach to
construction
• Treatability testing
• Interim system to
contain plume while
full system is
designed
Pre- or Post-Design
Groundwater/
Optimization Modeling
' MODFLOW
' MT3D
• Quickflow
• Randomwalk
• Biotrans
• (for most sites, the type
of modeling was not
specified)
Post-Design
Optimization/Modifications of Extraction System
Modifications | Purpose/ Objective
• Add extraction points • Increase extraction rate/
contain plume
• Abandon extraction points • Respond to reduction in
extent of contamination
• Resize extraction pumps • Increase efficiency of
system
• Adjust pumping rates • Increase efficiency of
system
• Change type of pump • Reduce shearing or
aeration of extracted
groundwater
• Modify extraction system • Increase efficiency of
design system or respond to
changes in remedial goals
• Use alternate remediation • Allow use of more cost-
method effective method (for
example, AS or natural
attenuation)
• Implement or expand • Respond to new source or
source controls increase efficiency in
treating existing source
areas
• Reduce performance • Reduce O&M costs
monitoring
Optimization/Modifications of Treatment System
Modifications | Purpose/ Objective
• Increase or reduce • Make capacity of treatment
equipment capacity plant match that of
extraction system
• Add chemical • Halt fouling of equipment;
enhancements to system enhance removal of
solids; enhance
biodegradation of
contaminants
•
• Add or replace with in situ • Increase performance or
technologies cost-effectiveness of the
system
• Upgrade process • Increase performance or
equipment cost-effectiveness of the
system
• Add process units to • Address unexpected
treatment train contaminants or increase
performance or efficiency
of system
• Automate treatment • Allow remote monitoring;
system operation reduce O&M cost
• Discontinue treatment • When discharge of
untreated water is
permitted
• Reduce compliance • Reduce O&M costs
monitoring
Notes:
1 Because the focus of the case studies was not on optimization, the types of optimization efforts listed in this table are not necessarily a comprehensive list of optimization efforts performed at all of
the case study sites.
3-9
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-7: System Optimization and Modification Efforts Conducted at 28 Sites1
Site Name
Baird and McGuire
SF Site
City Industries SF
Site
Des Moines TCE
SF Site
Former Firestone
Facility SF Site
Former Intersil, Inc.
Site
Pre-Design
Interim
Design
None identified
None identified
None identified
None identified
One extraction
well w/air
stripper to start
Groundwater
Modeling
System
Yes; provided
no details
None identified
Two-
dimensional
MODFLOW
Yes; provided
no details
Yes; provided
no details
Post-Design
Optimization
Modeling
Yes; provided
no details
Examined
varying
pumping rates
None identified
Yes; provided
no details
Yes; provided
no details
Modifications to Extraction System
Modification
Resized extraction
pumps
Increased pumping
from leading edge of
plume and decreased
pumping from
upgradient wells
None identified
Installed 10 additional
wells off-site/ adjusted
pumping rates/
increased overall
pumping rate for a 2
week period
None identified
Reason
Increase pumping rate to
meet design criteria
Maximize pumping
zones of influence
Not applicable
Prevent migration of
contaminated plume into
intermediate zones
Not applicable
Modifications to Treatment System
Modification
Enlarged sludge
thickener/replaced
bioreactor with air
stripper
None identified
AS media changed
from spherical to
chandelier type/
anti-corrosion and
biofouling agents
added to AS
None identified
P&T system
upgraded/switched
from P&T to PRB
in 1995
Reason
Enable treatment plant
unable to meet design
flowrate/maintain biomass
at design flowrates and
contaminant concentrations
Not applicable
Address iron corrosion and
biofouling of AS media
Not applicable
Reduce treatment costs and
allow for transfer of the
property
Table Continued...
Notes:
1 Because the focus of the case studies was not on optimization, the optimization efforts listed in this table are not necessarily a comprehensive list of optimization efforts performed at the case study
sites.
3-10
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-7: System Optimization and Modification Efforts Conducted at 28 Sites1
Site Name
French, Ltd. SF Site
Gold Coast SF Site
JMT Facility RCRA
Site (formerly Black
& Decker)
Keefe
Environmental
Services SF Site
King of Prussia
Technical
Corporation SF Site
LaSalle Electrical
SF Site
Pre-Design
Interim
Design
Staged
approach
None identified
Pilot test
None identified
None identified
None identified
Groundwater
Modeling
System
None identified
None identified
None identified
Yes; provided
no details
MODFLOW
andMTSD
None identified
Post-Design
Optimization
Modeling
MODFLOW
and Biotrans
None identified
None identified
Yes; provided
no details
Ongoing, using
MODFLOW
andMTSD
None identified
Modifications to Extraction System
Modification
None identified
Enlarged two
extraction wells/shut
down system for four
months/conducted air
sparging in "source"
areas/added peroxide
to wells for a period,
with no effect
Conducted full-scale
rehabilitation of
extraction well/
installed an electrical
and piping box at
extraction well
Constructed two
replacement extraction
wells
None identified
None - original design
considered adequate
Reason
Not applicable
Increase extraction rate/
increase amount of TCE
and PCE desorbing from
soil
Unclog well/minimize
time to perform routine
maintenance checks on
system
Optimize system after
reevaluation because
cleanup goals were not
being met
Not applicable
Not applicable
Modifications to Treatment System
Modification
Added second
sheet-pile wall
around DNAPL/
shut system down
in 12/95
None identified
Constructed
enclosure around
the treatment
system
None identified
None identified
None - original
design considered
adequate
Reason
Address DNAPL detected/
continue remediation of the
site via natural attenuation,
as specified in the site ROD
Not applicable
Consolidate system
operation in one building
Not applicable
Not applicable
Not applicable
Table Continued...
Notes:
1 Because the focus of the case studies was not on optimization, the optimization efforts listed in this table are not necessarily a comprehensive list of optimization efforts performed at the case study
sites.
3-11
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-7: System Optimization and Modification Efforts Conducted at 28 Sites1
Site Name
Libby Groundwater
SF Site
Mid-South Wood
Products SF Site
Moffett Federal
Airfield
Mystery Bridge at
Highway 20 SF
Site, Dow/DSI
Facility
Odessa Chromium I
SF Site
Odessa Chromium
IIS SF Site
Old Mill SF Site
Pre-Design
Interim
Design
Pilot test and
demonstration
for in situ
bioremediation
1 985-89 french
drains
Currently in
pilot-test stage
None identified
30-day pilot
study
30-day pilot
study
None identified
Groundwater
Modeling
System
None identified
None identified
None identified
Quickflow
Randomwalk
and Geoflow
Randomwalk
and Geoflow
None identified
Post-Design
Optimization
Modeling
None identified
None identified
None identified
None identified
Yes; provided
no details
Yes; provided
no details
None identified
Modifications to Extraction System
Modification
Tested and converted
to lower-shear pumps/
four extraction wells
abandoned, and one
new well constructed
Removed five
extraction wells/
continuously adjusted
pumping schedule of
extraction wells
None identified
None identified
Added three injection
wells/converted two
monitoring wells to
recovery wells
Added two injection
wells/installed
recovery well
Added three collection
trenches
Reason
Increase effectiveness of
OWS/address decrease in
areal extent of
contamination
No contaminants
detected in the five wells/
schedule adjusted
according to
concentration results
Not applicable
Not applicable
Achieve higher injection
rate/attempt to fully
capture plume
Achieve higher injection
rate/expedite cleaning of
source area
Address new areas of
contamination
discovered
Modifications to Treatment System
Modification
Peroxide system
for aeration of ISB
source water
replaced with
bubbleless system
Added carbon
treatment system
for one year
None identified
None identified
Added chamber to
reaction tank/
added backwash
unit for filter
Added chamber to
reaction tank/
added a backwash
unit
Replaced two
carbon canisters
with one
Reason
Minimize treatment costs
Allow for use of treated
groundwater in production
facility
Not applicable
Not applicable
Precipitate iron before
stripping and filtering
Precipitate iron before
stripping and filtering
Eliminate over design
Table Continued...
Notes:
1 Because the focus of the case studies was not on optimization, the optimization efforts listed in this table are not necessarily a comprehensive list of optimization efforts performed at the case study
sites.
3-12
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-7: System Optimization and Modification Efforts Conducted at 28 Sites1
Site Name
SCRDI Dixiana SF
Site
Site A (Confidential
SF Site)
Sol Lynn/Industrial
Transformers SF
Site
Solid State Circuits
SF Site
Solvent Recovery
Services of New
England, Inc. SF
Site
Sylvester/Gilson
Road SF Site
Pre-Design
Interim
Design
1992-4 EPA
system/20
wells at 4 gpm
Bioremediation
study
None identified
None identified
None identified
4-well GW
circulation
Groundwater
Modeling
System
None identified
None identified
MODFLOW
None identified
None identified
None identified
Post-Design
Optimization
Modeling
Quickflow
None identified
MODFLOW
None identified
None identified
MODFLOW
Modifications to Extraction System
Modification
Added collection
trench/ reduced
extraction wells by
five (15 remain in
operation)
Expanded system by
adding more sparging
wells
Adjusted pumping
strategy because of
additional
contamination in the
silty aquifer identified
Added three wells off
site
None identified
Added six extraction
wells
Reason
Collect contaminated
groundwater from
shallow zone/achieve
more efficient hydraulic
control
Address additional
contamination
discovered during
demolition activities
Prevent cross-
contamination of zones
and prevent further
migration of
contaminants
Contain plume
Not applicable
Address hot spots
Modifications to Treatment System
Modification
Replaced tower air
stripper with
shallow-tray
stripper
None identified
None identified
Electronically
linked air stripper
blower to transfer
pumps so blower
would shut off
when not pumping
None identified
None identified
Reason
Tower air stripper was
struck by lightning
Not applicable
Not applicable
Prevent freezing problems
with blowers
Not applicable
Not applicable
Table Continued...
Notes:
1 Because the focus of the case studies was not on optimization, the optimization efforts listed in this table are not necessarily a comprehensive list of optimization efforts performed at the case study
sites.
3-13
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 3-7: System Optimization and Modification Efforts Conducted at 28 Sites1
Site Name
U.S. Aviex SF Site
U.S. Coast Guard
Support Center
United Chrome SF
Site
Western Processing
SF Site
Pre-Design
Interim
Design
1983-93
interim
remedial
measure
1994 pilot
study
None identified
None identified
Groundwater
Modeling
System
MODFLOW
and
Randomwalk
None identified
None identified
None identified
Post-Design
Optimization
Modeling
MODFLOW
and
Randomwalk
None identified
None identified
None identified
Modifications to Extraction System
Modification
Adjusted pumping
rates for each well
continuously
None identified
Turned off some
extraction wells/
flushed some areas
Discontinued
operation of 210
shallow well
points/installed deep
wells
Reason
Optimize system on the
basis of concentration
data for each well
Not applicable
Stop treatment in areas
with contaminant
concentrations below
cleanup levels/solubilize
contaminants in areas of
higher contamination
flushed
Address change in
remedial goal from
remediation to
containment
Modifications to Treatment System
Modification
Added pH
adjustment
None identified
Switched to
sending untreated
water to POTW/
injected deep
aquifer water into
upper aquifer
Added metals
precipitation to
treatment system/
replaced carbon
type
Reason
Reduce scaling of
equipment and discharge
piping
Not applicable
Minimize treatment costs/
discontinue rapid
dewatering of upper aquifer
Address severe fouling of
air stripping media/
minimize frequency of
carbon changouts required
Notes:
1 Because the focus of the case studies was not on optimization, the optimization efforts listed in this table are not necessarily a comprehensive list of optimization efforts performed at the case study
sites.
3-14
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
4.0 PERFORMANCE OF REMEDIAL SYSTEMS
AT 28 CASE STUDY SITES
Two of 28 sites have met all aquifer
restoration goals
Most sites have made progress
toward meeting remedial goals,
including reducing or eliminating a
hot spot within a plume, reducing the
mass of contaminants within a plume,
and reducing concentrations of
contaminants within a plume
This section discusses the performance of the
remedial systems used at the 28 sites in terms of
the remedial goals set for the sites and progress
made toward achieving those goals.
4.1 Remedial Goals
Remedial goals for the containment and
mitigation of contamination have been
established at all the case study sites. The
remedial goals for the sites included the restoration of all groundwater beneath the site and any
off-site groundwater that may have been affected by the site, as well as the containment of on-site
contamination, allowing off-site contamination to attenuate naturally. It should be noted that all
of the sites selected for case studies were chosen because they had established aquifer cleanup
goals and not just containment only goals, although goals at one site (Western Processing) have
been changed to containment only since the case studies were prepared. In addition, performance
goals for the treatment systems, such as requirements related to water discharge of water and air
emissions, were established for a number of the sites. Exhibit 4-1 identifies the goals established
for each site and whether the goals have been achieved.
Cleanup goals for the sites were established based on one or more of the following factors:
• Maximum contaminant levels (MCL)
• Primary drinking water standards
• Risk-based cleanup levels
• Approved alternative concentration limits (ACL)
• Optional cleanup levels for a non-time-constrained removal action
• Concentrations of contaminants in adjacent surface waters
For two-thirds of the sites, the aquifer goals established were based on MCLs.
Goals for the containment of contaminated groundwater were established for 25 of the 28 sites.
The manner of containment required at each site varied but typically consisted of containment of
the contaminated groundwater on-site or halting of the continued migration of an existing off-site
plume.
Limits on air emissions were identified for three of the sites (JMT, LaSalle, andLibby). During
preparation of the case study reports, EPA did not focus on whether air emission limits were
established; therefore, it is possible that the reports did not identify limits at some of the other
sites.
4-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 4-1: Summary of System Performance for 28 Sites
Site Name and Location
Baird and McGuire, MA
City Industries, FL
Des Moines, IA
Firestone, CA2
French, Ltd., TX
Gold Coast, FL2
Intersil, CA
JMT,NY
Keefe, NH
King of Prussia, NJ
LaSalle, IL
Libby, MT
Moffett, CA
MSWP, AR
Mystery Bridge, WY
Odessa I, TX
Odessa II, TX
Old Mill, OH
SCRDI Dixiana, SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent Recovery
Service, CT
Sylvester/Gilson Road,
NH
U.S. Aviex,MI
United Chrome, OR
USCG Center, NC
Western Processina. WA
Notes:
o
Q
NR
NE
Remedial
Technology
P&T
P&T
P&T
P&T
P&T, ISB
P&T, AS
P&T, AS,
PRB
P&T
P&T
P&T
P&T
P&T, ISB
PRB
P&T
P&T
P&T
P&T
P&T
P&T
P&T, AS, ISB
P&T
P&T
P&T
P&T
P&T
P&T
PRB
P&T
Remedial Goals
Restore
Aauifer
o
o
o
Q
o
Q
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Contain
Plume
o
Q
Q
Q
Q
Q
Q
Q
Q
Q
NE
NE
Q
Q
NE
Q
Q
Q
Q
Q
o
Q
Q
Q
o
Q
o
Q
CleanuD Level Basis
MCLs, surface water
MCLs
MCLs
MCLs, drinking water
criteria, risk-based
risk-based
MCLs, drinking water
MCLs
MCLs
ACLs
MCLs
drinking water
MCLs, risk-based
drinking water
MCLs, risk-based
MCLs
MCLs
MCLs
risk-based
MCLs
MCLs
MCLs
MCLs
To be set
ACLs
MCLs
MCLs, risk-based
drinking water
MCLs
Goal established but not met
Goal established and met
Goal established, but performance not reported
Goal not established
Performance Goals
Air
Emission
NR
NR
NR
NR
NR
NR
NR
Q
NR
NR
Q
Q
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
Water
Discharge
NR
Q
NR
Q
NR
Q
Q
Q
Q
Q
NR
NR
NE
o
Q
Q
Q
Q
Q
NR
Q
NR
Q
Q
Q
Q
NE
Q
Minimum
Maximum
Average
Median
Contaminant
Mass
Removed1
Total Pounds
2,100
2,700
30,000
500
510,000
2,000
120 (P&T);
15 (PRB)
840
68
5,400
130
37,000
NR
800
21
1,100
130
120
7
5,300
5,000
2,700
4,300
430,000
660
31,000
NR
100.000
7
510,000
43,000
2.000
For sites at which groundwater cleanups are ongoing, the total mass of contaminant removed represents the performance
reported as of late 1997 or early 1998. Contaminant mass removals were calculated based on various types of mass balances
around the site treatment system, not based on groundwater monitoring data. Insufficient data were available to calculate a
removal of contaminant mass by in situ bioremediation; for those sites at which used in situ bioremediation was used, the
contaminant mass removed may be greater than shown here.
Firestone and Gold Coast - remediation has been completed at these two sites.
4-2
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Contaminant Mass Removed
Exhibit 4-1 presents the contaminant mass removed for the case study sites. For 26 of the sites,
the mass of contaminant removed by the remediation systems was reported or could be calculated
from reported data. When concentration and throughput data for the treatment system were
available, these data were used to calculate the mass of the contaminant removed. Contaminant
mass removals calculated based on groundwater monitoring results were not available. Total
mass of contaminant removed ranged from seven pounds (SCRDI Dixiand) to 510,000 pounds
(French, Ltd.}, with an average of approximately 43,000 pounds and a median of approximately
2,000 pounds. For almost one-third of the sites, contaminant mass removed ranged from 1,000
to 10,000 pounds per site.
Because mass removal rates are dependent on many factors, including the extent and
concentration of the contamination, contaminant properties, and the volume of groundwater
treated, they generally are not used to evaluate the achievement of remedial goals. The
variability is demonstrated in Exhibit 4-2 which shows the average mass of contaminant removed
per year and per 1,000 gallons of water treated at each site. Contaminant mass removed per year
for the 26 sites varies from approximately two pounds to more than 100,000 pounds, and from
approximately 0.0001 pounds per 1,000 gallons treated to three pounds per 1,000 gallons treated.
Sites with relatively higher mass removal rates per year do not consistently show relatively
higher mass removal rates per 1,000 gallons treated. This may be due in part to differences in the
concentration of contamination in the extracted groundwater. In addition, while not completed
for this report (due to a lack of available data), a comparison of mass removal rates at a site over
time can generally be useful in evaluating changes in system performance, for example, in
identifying when removal rates are approaching asymptotic values (see the case studies for
Western Processing and Firestone}.
Reduction in Concentrations of Contaminants
Exhibit 4-3 presents the average reductions in concentrations of contaminants at the case study
sites, sorted by the number of years which the remediation system was in operation (for this
report, the number of years of performance data available). Average concentrations of
contaminants could be calculated on the basis of available data for 17 of the 28 case study sites.
For several of the sites, average concentrations of contaminations were reported only for a group
of contaminants, and several others reported average concentrations of contaminants by
individual contaminant. In addition, three of the sites (United Chrome, Odessa II, and French
Ltd.) reported individual average concentrations of contaminants for more than one aquifer.
4-3
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 4-2: Unit Contaminant Mass Removed at 26 Sites
1
SCRDI Dixicna
Mystery Bridge
Intersil
Qd Mill
Keefe
GddCecst
LcSdle
Cdsssall
Des Manes
JMT
MSWP
U.S.Aviex
Cdessal
French, Ltd.
SdidStde
BdrdcndMcGuire
S dvent Reccvery S ervice
Sd Lynn
Kingcf Prussia
Qly Indus fries
United Chrcme
SiteA
Libby
Wes tern P races s i ng
Sylvester/Silscn Rood
Firestcne
0.0001
10
pounds per year
100 1,000 10,000
100,000
1,000,000
0.001
0.01 0.1
pounds per 1,000 gallons
DMass Removed per 1,000 Gallons Treated
I Mass Removed per Year
4-4
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 4-3: Summary of Average Contaminant Concentration Reduction
at 17 Sites
Site Name and Location
Intersil, CA
Des Moines, IA
JMT,NY
United Chrome, OR
MSWP, AR4
Odessa I, TX
Gold Coast, FL
Odessa II, TX4
LaSalle, IL4
French, Ltd., TX
U.S. Aviex, MI
Keefe, NH
LaSalle, IL
City Industries, FL
King of Prussia, NJ
Baird and McGuire, MA
Site A, NY
Contaminant(s)
Basis1
Zone1
VOCs (4 contaminants)
VOCs (3 contaminants)
TCE
VOCs (4 contaminants)
TCE
Chromium
Chromium
Shallow Aquifer
Deep Aquifer
Metals/VOCs (4 contaminants)
As
PCP
Cr
Total PAHs
Chromium
PCE
TCE
Chromium
Chromium
PCBs/ VOCs
1,2-DCA
Vinyl chloride
Benzene
1,1,1-TCA
Perched Aquifer
Trinity Aquifer
Shallow Aquifer
SI Aquifer/TNT Aquifer
SI Aquifer/TNT Aquifer
SI Aquifer/TNT Aquifer
VOCs (10 contaminants)
VOCs (5 contaminants)
PCBs/ VOCs
Deep Aquifer
VOCs and SVOCs (16 contaminants)
Metals (6 contaminants)
VOCs (9 contaminants)
VOCs (specific contaminants not identified)
SVOCs (specific contaminants not identified)
BTEX
Average Contaminant
Concentration (jag/L)2
Start
1,609
87
45
950
450
1,923,000
1,400
140
3
22
30
35
980
176
88
180
400
400
256/917
129/420
516/640
107
158
80
100
3,121
3,500
4,500
500
1,000
160
End
31
10
3
30
7
18,000
110
90
4
11
5
23
540
1
1
190
50
570
0.8/1
1.2/1
0.6/2
40
67
18
6
444
1,500
4,000
420
520
26
Years
of
Data
11.1
9
9
8.6
8.6
8.6
8.6
7.1
7.1
7.1
7.1
7.1
5
4.9
4.9
4.8
4.8
4.2
3.9
3.9
3.9
3.6
3.6
3.5
3.2
3
2.6
2.6
1
1
1
Percent
Reduction3
98
89
93
97
98
99
92
36
-33
50
83
34
45
99
99
-6
88
-43
>99
>99
>99
63
58
78
94
86
57
11
16
48
84
Notes:
Data on average concentrations were reported for 17 of the 28 case study sites; for those sites, data are shown here by contaminant(s) and
zone; zones are noted only for those sites at which concentrations of contaminants were reported for more than one aquifer.
Average concentrations of contaminants are based on a reported geometric mean of all data, as presented in the case study reports; for sites
with ongoing cleanups, average concentrations of contaminants shown at "end" time represent the concentrations reported as of the date
that data were available, typically late 1997 or early 1998.
Percent reduction was calculated as the difference between average concentrations of contaminants at start and end points, divided by the
average concentration at the start.
Negative percent contaminant reductions were measured at three sites. These anomalies are discussed in the case studies for the sites.
4-5
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
4.2 Progress Toward Goals
Exhibit 4-4 lists the number of sites that have met specific remedial and system performance
goals.
Exhibit 4-4: System Performance Summary
Goal
Number of Sites with
Snecified Goals 1
Number of Sites Meeting
Snecified Goals
Remedial Goals
Aquifer Restoration
Containment
28
25
2
22
System Performance Goals
Air emissions 2
Discharge of water
3
19
3
18
Notes:
1 Goals for each site are specified in the case study reports.
2 Air emission goals were specifically identified for only three of the case study sites.
Gold Coast and Firestone are the two case study sites for which all remedial goals have been
met, as described briefly below.
• Gold Coast was a spent oil and solvent recovery facility that operated from 1970
to 1982. In the 1980s, groundwater was determined to be contaminated with
chlorinated and nonchlorinated VOCs at levels as high as 100 milligrams per liter.
A P&T system consisting of five extraction wells (pumping at a total of
approximately 100 gpm) and two air stripping towers was put on line in 1990. By
the end of 1994, concentrations of groundwater contaminants were reduced to
levels lower than cleanup standards, with the exception of one source area. A
limited air sparging effort was able to reduce the contaminant levels in that area to
levels lower than cleanup standards by 1995. The site is located over a porous
limestone aquifer, which facilitated groundwater pumping, and the use of source
controls and in situ technology were identified as key factors in the success of the
cleanup.
• The Firestone facility operated as a tire manufacturing plant from 1963 until
1980. In 1984, a 2.5-mile-long contaminated groundwater plume that contained
chlorinated solvents was identified. The primary target contaminant in the plume
was 1,1-DCE. A P&T system consisting of 35 extraction wells and ex situ air
stripping and carbon adsorption was put on line in 1986. By 1987, the
contaminated plume was contained and by 1992 the concentrations of 1,1-DCE in
the plume had been reduced to levels lower than the cleanup goals and the system
was shut down. During the operation of the groundwater extraction system, the
site operators frequently adjusted it to maintain maximum concentration of
contaminant at the treatment plant influent. That factor was identified as a key
one in the success of the cleanup.
4-6
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
In addition to the two sites listed above at which the specified aquifer cleanup goals have been
met, progress has been made toward meeting the specified remedial goals for most of the sites.
Example successes include:
• Meeting aquifer cleanup goals in one or more zones at the site
AtDesMoines, the cleanup goals for the off-site plume were achieved within two
years of startup of the remediation system. P&T continues to maintain an inward
hydraulic gradient and to remediate on-site groundwater. The aquifer at the site
is a relatively homogeneous formation of sand and gravel that has a relatively
high conductivity.
• Reducing the size of a contaminated plume
At Odessa I, the total plume area was reduced approximately 44 percent in two
years (from 1994 to 1996). On several occasions, the groundwater extraction
system was modified to improve efficiency.
• Reducing the concentrations of contaminants within a plume
At United Chrome, average concentrations of chromium were reduced in the
upper aquifer from more than 1,900 to 18 mg/L over nine years, and in the deep
aquifer from 1.4 to 0.11 mg/L over six years. On several occasions, the
groundwater extraction system was modified to target the more highly
contaminated areas of the plume.
• Removing contaminant mass from a plume
At French, Ltd., the P&T system removed approximately 517,000 pounds of
contaminant (measured as total organic carbon) from January 1992 through
December 1995. The mass was removed through aggressive pumping of
groundwater that contained relatively high concentrations of contaminants
(hundreds of mg/L) from more than 100 recovery wells.
• Achieving containment of a plume
At City Industries, the contaminated groundwater plume has been contained
hydraulically since the P&T system was put on line in 1994.
It is important to note that groundwater cleanup is ongoing at most of the case study sites;
therefore, the system performance presented in this report does not represent the final
performance to be achieved in remediating each of the sites. As discussed earlier, the data
presented in the case studies are generally available through late 1997 or early 1998.
4-7
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
4-8
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
5.0 COST OF REMEDIAL SYSTEMS AT 28 CASE STUDY SITES
This section discusses the costs of the remedial
systems at the 28 case study sites and unit costs
for the groundwater cleanups at these sites.
The costs for the sites typically were reported as
capital costs, operation and maintenance
(operating) costs, remedial design costs, and
other costs. For the purpose of this report,
calculated unit costs are provided as average
annual operating cost, capital cost per 1,000
gallons treated per year, and average annual
operating cost per 1,000 gallons treated per year.
Average annual operating cost as a percentage of
capital cost is also presented. Assumptions used
in reporting cost data are summarized below.
Capital and operating costs were
highly variable from site to site with
key cost drivers, including variable
monitoring requirements, significant
system modifications needed, and
size and complexity of the remedial
systems
The following three types of unit
costs were calculated for each site:
• Average operating cost per year of
operation
• Capital cost per 1,000 gallons
treated per year
• Average annual operating cost per
1,000 gallons treated per year
Cost data presented in the case study reports were based on data provided by EPA
remedial project managers, site owners, or vendors. The costs presented in this
report are based on the cost data in these case study reports. In addition, updated
cost data received in May 1999 for several of the sites (BctirdcmdMcGuire,
Libby, French Ltd., United Chrome, Sylvester/Gilson Road, Western Processing)
was included in this report. When actual cost data were not available, site
contacts provided estimates based on the best data available at the time.
Groundwater cleanup is ongoing at most of the sites; therefore, the operating costs
(and in some cases the capital costs) do not represent the total to be spent to
remediate a site. The data presented here generally are current as of late 1997 or
early 1998, with 1999 data available for the sites identified above.
Because groundwater cleanup is ongoing at most of the sites and the total time
necessary to complete cleanup of a site was not known, a net present value (NPV)
of the remedial costs for the sites was not calculated for this report. The systems
in the 28 case studies had been operating for as few as 2 years and as long as 11
years. While many feasibility studies conducted under Superfund assume a 30-
year duration to estimate the cost of a P&T remedy, the use of this timeframe was
not considered to be applicable for this report because the two completed projects
were completed in 3.5 and 7 years.
5-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Capital and operating costs were extracted from cost data provided in the case
studies based on the Recommended Cost Format in Guide to Documenting and
Managing Cost and Performance Information for Remediation Projects [4].
Capital costs included: technology mobilization, setup, and demobilization;
planning and preparation; site work; equipment and appurtenances; startup and
testing; and other technology capital costs. Operating costs included: labor;
materials; utilities and fuel; equipment ownership, rental, or lease; performance
testing and analysis (although compliance testing was often not separated out);
and other technology operating costs. Source controls, RI/FS, and system design
costs were not included as capital or operating costs. However, VCBs used for
hydraulic control were included as capital costs.
As previously discussed in Section 3.1, PRBs may differ in form and purpose
from P&T and PRB systems, and unit costs for sites at which PRBs were used are
shown separately from costs for sites at which P&T was used. PRBs treat only the
groundwater that passes through the barrier, while P&T actively extracts
groundwater from an aquifer. Therefore, the volume of groundwater treated by a
PRB will be relatively less than by a P&T system for the same size plume.
Cost Data
Exhibit 5-1 presents the cost data for cleanup of contaminated groundwater at each of the case
study sites. The table also identifies the major factors that influenced costs at each of the sites.
Exhibit 5-2 summarizes overall remedial costs and unit costs for P&T and PRB sites,
respectively, including minimum, maximum, average, and median costs for each of the two
groups individually and combined.
Capital costs per P&T site ranged from approximately $250,000 (Gold Coast) to $15 million
(Western Processing and French, Ltd.), and average annual operating costs ranged from
approximately $90,000 (MSWP) to $4.4 million (Western Processing). Average annual operating
costs ranged from 2.9 to 56 percent of the capital costs. The median capital cost was $1.9
million and the median average annual operating cost was $190,000; with median unit costs of
$96 of capital cost per average 1,000 gallons of groundwater treated per year and $ 18 of average
annual operating cost per average 1,000 gallons of groundwater treated per year.
Based on three sites, capital costs per PRB site ranged from approximately $370,000 (Mqffett) to
$600,000 (Intersil [PRB]), and average annual operating costs per PRB site ranged from
approximately $26,000 (Mojfeti) to $95,000 (Intersil [PRB]). Average annual operating costs
ranged from 6.9 to 17 percent of the capital costs. For the PRB systems, the approximate median
capital cost was $500,000 and the median average annual operating cost was $85,000; with
median unit costs of $520 of capital cost per average 1,000 gallons of groundwater treated per
year and $84 of average annual operating cost per average 1,000 gallons of groundwater treated
per year.
The total remedial cost for each site was not projected, since the number of years in which each
system has been operating and the progress of each system toward meeting remedial restoration
goals vary from system to system.
5-2
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-1: Summary of Cost Data for 28 Sites
1,2,3
Site Name and
Location
Years of
Operation
(with data
available)
Average
1,000
Gallons
Treated per
Year
Capital
Cost ($)
Average
Operating
Cost (S)
Per Year
P&T SITES
Baird and
McGuire, MA
City Industries,
FL
Des Moines, IA
Firestone, CA
French, Ltd. ,TX
Gold Coast, FL
Intersil (P&T),
CA
JMT,NY
3.8
3.0
8.8
6.8
4.0
3.7
7.3
9.6
21,000
50,000
554,000
266,000
76,000
22,000
5,000
5,200
11,000,000
1,200,000
1,600,000
4,100,000
15,000,000
250,000
330,000
880,000
2,000,000
170,000
110,000
1,300,000
3,400,000
120,000
140,000
150,000
Average
Operating Cost
as Fraction of
Capital Cost
0.18
0.14
0.07
0.31
0.21
0.49
0.43
0.17
Capital Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
530
23
2.9
15
200
11
65
170
Average Annual
Operating Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
97
3.3
0.21
4.9
43
5.6
28
29
Key Cost Drivers
Operating costs increased due to
the need to monitoring for a
wide range of contaminants and
for several full-time operators to
be onsite
Optimized pump rates;
bio fouling of air stripper
increased system downtime
Unit costs reflect economies of
scale
Frequent modifications to
system were required; cost of
analysis and data management
were high
Large system incorporating P&T
and ISB; oversight costs were
high
Optimized extraction wells; P&T
system required less than four
years to clean up site
Groundwater extraction system
was expanded after three years
of operation, likely increasing
operating costs
Modifications of treatment
system increased capital costs 35
percent; system consisted of one
extraction well
5-3
Table Continued...
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-1: Summary of Cost Data for 28 Sites
1,2,3
Site Name and
Location
Keefe, NH
King of Prussia,
NJ
LaSalle, IL
Libby, MT
MSWP, AR
Mystery Bridge,
WY
Odessa I, TX
Odessa II, TX
Years of
Operation
(with data
available)
4.1
2.7
4.4
5.3
8.3
3.6
4.2
4.1
Average
1,000
Gallons
Treated per
Year
11,000
57,000
5,200
2,900
12,000
54,000
30,000
30,000
Capital
Cost (S)
1,600,000
2,000,000
5,300,000
3,000,000
470,000
310,000
2,000,000
2,000,000
Average
Operating
Cost (S)
Per Year
240,000
390,000
190,000
500,000
91,000
170,000
190,000
140,000
Average
Operating Cost
as Fraction of
Capital Cost
0.15
0.19
0.03
0.17
0.19
0.56
0.10
0.07
Capital Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
140
36
1,000
1,000
38
5.7
65
65
Average Annual
Operating Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
21
6.8
36
170
7.4
3.2
6.3
4.6
Key Cost Drivers
Optimization of the system
pumping rates increased mass
removal efficiency
Electrochemical treatment
increased costs
Complex mixture of
contaminants and DNAPL
contributed to elevated capital
costs
Chemical costs (e.g., hydrogen
peroxide) were high for in situ
bioremediation; monitoring,
sampling, and analysis costs
were high at the beginning of the
project
Use of fabric filters increased
operating life of GAC units
Low concentrations in
groundwater
ROD required that ferrous iron
be produced onsite
electrochemically, limiting
number of appropriate vendors
and increasing capital costs
ROD required that ferrous iron
be produced onsite
electrochemically, limiting
number of appropriate vendors
and increasing capital costs
5-4
Table Continued...
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-1: Summary of Cost Data for 28 Sites
1,2,3
Site Name and
Location
Old Mill, OH
SCRDI Dixiana,
SC
Site A, NY
Sol Lynn, TX
Solid State, MO
Solvent
Recovery
Service, CT
Sylvester/Gilson
Road, NH
U.S. Aviex,MI
Years of
Operation
(with data
available)
7.8
4.6
1.3
3.0
4.2
2.9
9.5
3.4
Average
1,000
Gallons
Treated per
Year
1,700
4,500
6,700
4,300
62,000
11,000
126,000
96,000
Capital
Cost (S)
1,600,000
1,800,000
1,400,000
2,100,000
930,000
4,400,000
7,200,000
1,400,000
Average
Operating
Cost (S)
Per Year
210,000
94,000
290,000
150,000
370,000
400,000
1,900,000
180,000
Average
Operating Cost
as Fraction of
Capital Cost
0.13
0.05
0.20
0.07
0.40
0.09
0.27
0.13
Capital Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
960
410
210
490
15
390
57
15
Average Annual
Operating Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
130
21
43
34
6
36
15
1.9
Key Cost Drivers
Modifications to the system
increased capital costs 22
percent
PRP made major modifications
to the remedial system; relatively
low contaminant concentration
Use of skid-mounted modular
equipment reduced capital costs;
treatment system included air
sparging and in situ
bioremediation
Complex hydrogeology
increased capital costs
Capital costs do not include
costs for installation of four deep
extraction wells installed as part
ofRI/FS
Presence of DNAPL contributed
to elevated capital and operating
costs
Several full-time operators were
on site 24 hours per day, high
costs for fuel oil to operate the
vapor incinerator used for air
emission control
Optimization of interim P&T
system before final remedy
reduced costs
5-5
Table Continued...
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-1: Summary of Cost Data for 28 Sites
1,2,3
Site Name and
Location
United Chrome,
OR
Western
Processing, WA
Years of
Operation
(with data
available)
8.6
8.2
Average
1,000
Gallons
Treated per
Year
7,200
119,000
Capital
Cost ($)
3,300,000
15,000,000
Average
Operating
Cost (S)
Per Year
96,000
4,400,00$
Average
Operating Cost
as Fraction of
Capital Cost
0.03
0.30
Capital Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
460
13?4)
Average Annual
Operating Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
13
3l
Key Cost Drivers
Modular treatment system used
initially, reducing costs
Initially used large complex
system with over 200 vacuum
well points, 24-hour oversight
required; frequent maintenance
to control iron precipitate
buildup
PRB SITES
Intersil (PRB),
CA
Moffett, CA
USCG Center,
NC
1.8
1.2
1.0
1,100
200
2,600
600,000
370,000
500,000
95,000
26,000
85,000
0.16
0.07
0.17
520
1,600
190
83
110
33
P&T was replaced by PRB,
reducing operating cost (see
above)
Demonstration-scale project;
increased performance
monitoring was required for
technology validation
Use of PRB was estimated to
save $4 million over a typical
P&T system
Note:
Groundwater cleanups are ongoing almost sites; data presented here generally are current of late 1997 or early 1998.
Capital and operating costs were extracted from costs provided in the case studies based on the Recommended Cost Format in Guide to
Documenting and Managing Cost and Performance Information for Remediation Projects [4]. Source controls, RI/FS, and system design costs
were not included as capital or operating costs.
Cost data shown in the case study reports were based on data provided by EPA remedial project managers, site owners, or vendors. The costs
presented in this report are based on the total costs available at the time the case study report for the site was prepared and updated cost data
received in May 1999 for several of the sites (Baird andMcGuire, Libby, French Ltd., United Chrome, Sylvester/Gilson Road, Western
Processing). When actual cost data were not available, site contacts provided estimates based on the best data available at the time.
The P&T system at Western processing was changed in response to a change in the remedial goals at the site from aquifer cleanup to
containment. The modified system pumped less than half of the water pumped by the original system. However, for this report, data were not
available to determine the cost implications of the system modification.
Table Continued.
5-6
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-2: Summary of Remedial Cost and Unit Cost Data for 28 Sites
1,2,3
Cost Category
Years of System Operation (with
data available)
Average Volume of Groundwater
Treated Per Year (1,000 Gallons)
Total Capital Cost ($)
Average Operating Cost Per
Year ($)
Average Operating Cost Fraction
of Capital Cost
Capital Cost Per Volume of
Groundwater Treated Per Year
($71,000 Gallons)
Average Annual Operating Cost
Volume of Groundwater Treated
Per Year ($71,000 Gallons)
P&T Sites (26 sites)
Range
1.3 - 9.6
1,700 - 550,000
250,000 - 15,000,000
91,000 - 4,400,000
0.03 - 0.56
2.9 - 1,000
0.21 - 170
Median
Average
4.2
5.3
21,000
63,000
1,900,000
3,500,000
190,000
670,000
0.17
0.20
96
250
18
31
PRB Sites (3 sites)
Range
1.0 - 1.8
230 - 2,600
370,000 - 600,000
26,000 - 95,000
0.07 - 0.17
192 - 1,600
33 - 110
Median
Average
1.2
1.3
1,100
1,300
500,00
490,000
85,000
69,000
0.16
0.13
520
780
84
76
All Sites (28 sites)
Range
1.0 - 9.6
230 - 550,000
250,000 - 15,000,000
26,000 - 4,400,000
0.03 - 0.56
2.9 - 1,600
0.21 - 170
Median
Average
4.1
4.9
12,000
57,000
1,600,000
3,200,000
180,000
610,000
0.17
0.19
140
310
21
36
Notes:
i
Groundwater cleanups are ongoing at most sites; data presented here generally are cumulative as of late 1997 or early 1998.
Capital and operating costs were extracted from costs provided in the case studies based on the Recommended Cost Format in Guide to Documenting and
Managing Cost and Performance Information for Remediation Projects [4]. Source controls, RI/FS, and system design costs were not included as capital
or operating costs.
Cost data shown in the case study reports were based on data provided by EPA remedial project managers, site owners, or vendors. The costs presented
in this report are based on the total costs available at the time the case study report for the site was prepared and updated cost data received in May 1999
for several of the sites (Baird andMcGuire, Libby, French Ltd., United Chrome, Sylvester/Gilson Road, Western Processing). When actual cost data were
not available, site contacts provided estimates based on the best data available at the time.
5-7
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Calculated Unit Costs
Calculated unit costs are used to compare and contrast remediation technologies. Although the
basis and methodology for calculation of unit costs for site cleanups are still a matter of some
debate, some unit costs can be used to compare costs and performance at ongoing and completed
cleanup efforts and in identifying cost-efficient remedial strategies for future cleanups. For this
report, the following three types of unit costs were calculated for each site:
• Average operating cost per year of operation
• Capital cost per 1,000 gallons of groundwater treated per year
• Average annual operating cost per 1,000 gallons of groundwater treated per year
Those unit costs, along with their ranges, averages, and medians are summarized in Exhibits 5-1
and 5-2 and depicted in Exhibits 5-3, 5-4, and 5-5, respectively. The three unit costs summarized
for the case study sites are described briefly below.
Average Operating Cost per Year of Operation
The average operating cost per year is determined by the throughput of the system and the
treatment processes required to treat the extracted groundwater, as well as the operating
efficiency of the system. Since a breakdown of annual operating costs by year was not available
for most of the sites, the change in operating costs over the life of a site's remediation system
could not be evaluated for the purposes of this report. The average annual operating costs were
calculated by dividing the total operating cost to date by the number of years represented by that
cost.
At SCRDI Dixicma, where approximately 40 gpm were pumped from 15 wells through a
relatively simple system and discharged to a POTW, the average annual operating cost
was $94,000 over 4.5 years. At French, Ltd., where approximately 190 gpm were
pumped from more than 100 wells through a more complex treatment system before being
reinjected into the aquifer, the average annual operating cost was more than $3.4 million
per year over 4 years.
Capital Cost per 1,000 Gallons of Groundwater Treated Per Year
The capital cost per 1,000 gallons treated per year represents the relative costs of installing
remedial systems of varying capacity. This unit cost is influenced by factors such as the aquifer
complexity (which influences the size and complexity of the system needed to extract the
contaminated groundwater), the types of contaminants targeted for treatment at the site (which
influences the treatment plant components needed to remove the contaminants), the water and air
discharge limits for the particular site (which is also factor into the treatment plant components
needed), and restoration goals (which reflects the difference between sites where a large volume
of groundwater is treated over a relatively short time frame to clean up an aquifer versus
pumping at a lesser rate to prevent a contaminated plume from migrating from the site).
5-8
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-3: Average Operating Cost Per Year at 28 Sites
Western Processing
French, Ltd.
Baird and McGuire
Sylvester/Gilson Road
Firestone
Libby
Solvent Recovery Service
King of Prussia
Solid State
Site A
Keefe
Old Mill
Odessa 1
LaSalle
U.S. Aviex
Mystery Bridge
City Industries
JMT
Sol Lynn
Intersil
Odessa II
Gold Coast
Des Moines
United Chrome
Intersil (PRB)
SCRDI Dixiana
MSWP
USCG Center (PRB)
Moffett (PRB)
I
I
I
I!
18
I9fi 000
14 400 OP
13 400 000
\9 000 000
1500.00
Unn nnn
I390 000
I370 000
l?9n nnn
i?4n nnn
mn nnn
M9n,nnn
Man, nnn
nan nnn
M?n nnn
M7n,nnn
M fin nnn
M5(i nnn
^•140,000
^•140,000
• 120 000
11110,000
96,000
95,000
94,000
31 ,000
5,000
M 9nn nnn
• 1 300 000
3
0
10,000 100,000 1,000,000 10,000,000
Average Operating Cost ($) Per Year
The following example illustrates the effect of the aquifer complexity and treatment plant
requirements on the capital cost per 1,000 gallons of groundwater treated annually.
At the Gold Coast site, groundwater in a relatively shallow and homogeneous aquifer
contaminated with TCE was extracted and treated by an air stripper alone before it was
discharged to surface water. The capital cost was $11 per 1,000 gallons of water
treated. This compares with a cost of$l, 020 per 1,000 gallons of water treated at the
LaSalle Site, where groundwater was extracted via a horizontal pumping regime and was
treated for a complex range of contaminants by a much more complex system. The
system consisted of two air strippers, both vapor and liquid-phase GAC, oil/water
separation, andpH adjustment.
5-9
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-4: Capital Cost Per 1,000 Gallons of Groundwater Treated
Per Year
Moffett (PRB)
Libby
LaSalle
Old Mill
Baird and McGuire
Intersil (PRB)
Sol Lynn
United Chrome
SCRDI Dixiana
Solvent Recovery Services
Site A
French, Ltd.
USCG Center
JMT
Keefe
Western Processing
Intersil (P&T)
Odessa I
Odessa IIS
Sylvester/Gilson Road
MSWP
King of Prussia
City Industries
Firestone
Solid State
U.S. Aviex
Gold Coast
Mystery Bridge
Des Moines
I 'l.bUU
I b
• 3
i
I b30
I 490
I 460
I 410
1 210
I 2UU
I 190
1 1/U
I 140
1 130
I 65
I 65
I bl
I 38
I 23
I 15
I 15
] 11
1,000
1,000
960
1 10 100 1,000 10,000
Capital Cost Per Volume of Groundwater Treated Per Year ($/1,000 Gallons)
5-10
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Exhibit 5-5: Average Annual Operating Cost per 1,000 Gallons of
Groundwater Treated Per Year
Libby
Old Mill
Moffett (PRB)
City Industries
Intersil (PRB)
French, Ltd.
Site A
Western Processing
LaSalle
Solvent Recovery Services
Sol Lynn
USCG Center
JMT
Intersil (P&T)
Keefe
SCRDI Dixiana
Sylvester/Gilson Road
United Chrome
MSWP
King of Prussia
Odessa 1
Solid State
Gold Coast
Firestone
Odessa IIS
Des Moines
Mystery Bridge
U.S. Aviex
Baird and McGuire
,
I 130
1 110
1
1 43
1 43
1 37
1 36
1 36
1 34
1 33
1 29
1 28
1 21
I ^i
I 0.2
1 15
1 13
1 7
1 7
1 6
1 6
1 6
1 5
1 b
1 3
1 3
1 2
97
83
0.1 1.0 10.0 100.0 1,000.0
Average Annual Operating Cost Per Volume of Groundwater Treated Per Year
($/1,000 Gallons)
5-11
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Average Annual Operating Cost Per 1,000 Gallons of Groundwater Treated
Per Year
The average annual operating cost per 1,000 gallons of groundwater treated per year represents
the relative costs to operate systems of various capacities and complexities. Similar to the capital
cost per 1,000 gallons of groundwater treated per year, this unit cost is highly dependent on site-
specific factors such as the aquifer complexity, the types of contaminants targeted for treatment,
the water and air discharge limits, and the restoration goals.
The following example illustrates the effect of the complexity of a site treatment system on
average annual operating cost per 1,000 gallons of groundwater treated per year.
At Des Moines over 500 million gallons of groundwater were treated per year using a
relatively simple system consisting of an air stripper for an average annual operating
cost of $0.21 per 1,000 gallons of groundwater treated annually. Conversely, at Libby,
2.9 million gallons of groundwater were treated per year using a complex remediation
system consisting of oil/water separation, nutrient addition, and bioreactors for an
average annual operating cost of$l 73 per 1,000 gallons of groundwater treated
annually.
In general, systems that treat a relatively large volume of groundwater per year will cost less in
both capital and annual operating costs per 1,000 gallons of groundwater treated than a similar
system that treats a smaller volume of groundwater per year. While no specific correlation could
be derived based on the available information, the following example shows this trend.
The treatment systems at Des Moines, City Industries, and Mystery Bridge consisted of
P&T using air stripping as aboveground treatment. Des Moines treated a relatively
larger volume of groundwater annually. The following table summarizes some of the
data for these sites.
Site Name
Des Moines, IA
City Industries, FL
Mystery Bridge, WY
Average Volume of
Groundwater
Treated Annually
(1,000 Gallons)
554,000
50,000
54,000
Average Annual
Operating Cost Per
Volume of
Groundwater
Treated Per Year
(S/1,000 Gallons)
0.21
3.3
3.2
The average annual operating cost per volume of groundwater treated annually
exemplifies the above trend.
5-12
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
6.0 FACTORS THAT AFFECTED COST AND PERFORMANCE OF
REMEDIAL SYSTEMS AT 28 CASE STUDY SITES
The factors that affected cost and performance at
the 28 case study sites vary and are specific to
each site. This section discusses the key factors
that affected cost and performance of the
groundwater cleanup at the case study sites
identified from the case studies and industry
knowledge about groundwater remediation. The
factors have been grouped into the categories
summarized in Exhibit 6-1.
The factors that affected cost and
performance at the 28 case study sites
vary and are specific to each site
No single factor was found to be the
most important factor in determining
cost and performance of groundwater
cleanup projects
Exhibit 6-1: Factors Affecting Cost and Performance
of Groundwater Remediation Systems
Category
Source control factors
Hydrogeologic factors
Contaminant property factors
Extent of contamination factors
Remedial goal factors
System design and operation
factors
Factors
Presence of NAPL; application and timing of source controls
Properties of the aquifer; contamination of more than one aquifer;
influence of surface water on aquifer; influence of adjacent groundwater
production wells on aquifer
Treatability of the contaminant; fate and transport properties of the
contaminant
Area and depth of contaminated plume; concentrations of contaminant
within the plume
Restoration of the aquifer rather than plume; MCL rather than less-
stringent cleanup levels; cleanup of the entire aquifer rather than partial
cleanup; time allowed for cleanup
System downtime; system optimization; amount and type of monitoring
performed; use of in situ technology
Each of the categories is discussed below; specific examples of how each factor affected cost or
performance of the groundwater cleanup systems at the case study sites are also presented.
Source Control Factors
Sources of groundwater contamination vary from surface discharges to buried wastes. When
source material comes in contact with groundwater, contaminants begin to dissolve and move
into the groundwater by advection and dispersion mechanisms. In addition, contaminant sources
in the vadose zone may act as continuing sources of groundwater contamination via leaching of
contaminants onto storm water recharge that passes through the contaminated zone.
Biodegradation and volatilization also may contribute to the destruction or dispersion of
contaminants. However, in many cases, the mechanisms may have a negligible impact.
6-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
The solubilities of many common contaminants (such as chlorinated solvents) are relatively low,
and sources of those contaminants may remain in the subsurface for extended periods of time.
EPA has concluded that one of the most effective means of remediating a site at which
contaminated groundwater is present is to remove, or at least isolate, the source material from the
groundwater. The source controls implemented at the case study sites (see Exhibit 3-2) include
such methods as removal of hot spots (soil), soil vapor extraction, capping, and installation of
VCBs.
NAPL has been observed or suspected to be a source of groundwater contamination at a majority
of the case study sites (see Exhibit 2-7). Of the 28 sites, NAPLs were observed or suspected to
be present at 18. At twelve of the sites, only DNAPL was present; at three, only LNAPL was
present; and at another three both DNAPL and LNAPL were present.
At several sites (such as French Ltd., SRS, and Western Processing), efforts were made to
remove or isolate the NAPL from contact with the groundwater. Such efforts often involved
significant capital expenditures.
At Western Processing, both DNAPL and LNAPL were observed in the groundwater. A
slurry wall was constructed around the site to contain the plume and help achieve the
cleanup goals within a limited amount of time. The slurry wall required capital
expenditures of approximately $1.4 million.
If NAPL was not removed or isolated, the groundwater remediation efforts often were hindered.
At Solvent Recovery Service, DNAPL is present in both the overburden and the bedrock
aquifers, and is a source from which a dissolved plume continually forms. Despite three
years of P&T operation, the complex hydrogeology and DNAPL present at this site have
resulted in fluctuating concentrations of total VOCs in the groundwater. Site
representatives indicated that they plan to apply for a technical impracticability (TI)
waiver because of the presence of the persistent source of DNAPL.
Hydrogeologic Factors
Hydrogeologic factors that influence the cost and performance of groundwater remediation
systems include the composition and hydraulic conductivity of a water-bearing layer; the depth to
groundwater; contamination of more than one aquifer; vertical groundwater flow; the influence
of surface water; and the influence of nearby groundwater production wells. These factors can
affect the complexity of the groundwater remediation system as well as the ability of the system
to meet the remedial goals at a site.
This report presents information about the hydrogeologic conditions at the 28 subject sites (see
Exhibit 2-8). The hydraulic conductivity of the contaminated water-bearing layer(s) at the sites
ranged from 0.023 ft/day to 1,200 ft/day, a range of more than six orders of magnitude. The
hydraulic conductivity also often varied within an individual site, such as United Chrome, where
the hydraulic conductivity in the upper aquifer was two orders of magnitude less than in the
lower aquifer. At more than one-half of the sites, contamination was present in more than one
water-bearing layer or aquifer. Seven of the sites exhibited vertical groundwater flow, 13 were
influenced by adjacent bodies of surface water, and production wells (municipal or otherwise)
6-2
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
were located in the vicinity of each of eight of the sites. Reported depths of the water table
ranged from zero (at ground surface) to 45 feet below ground surface.
The following examples illustrate specific cases hydrogeological factors affected the cost or
performance of the groundwater remediation technology implemented at a site.
Hydraulic Conductivity
AtJMT, the hydraulic conductivity in the contaminated bedrock aquifer was relatively
low (0.65 ft/day). To increase hydraulic conductivity, controlled blasting was carried out
to create an artificial fracture zone, which served as an interceptor drain in the bedrock
around the extraction well. While that approach increased the capital cost of the system
(by an undetermined amount), it allowed effective extraction groundwater from the unit
by one well screened in the new fracture zone.
Contamination of More Than One Aquifer
At SCRDI Dixiana, eight distinct soil layers have been identified within the upper 100
feet of soils, including five water-bearing units. Early site characterization work at the
site misidentified the thicknesses and degrees of contamination of several of those units.
Groundwater extraction wells were installed based on the results of that early work. The
wells were screened across two units, thereby presenting a pathway for contaminants to
migrate into a previously uncontaminated aquifer. In addition, the contaminated shallow
sand aquifer at the site was not identified until after the system had been installed,
resulting in the need to modify the remedial system to address multiple contaminated
aquifers.
Vertical Groundwater Flow
At Solid State, the groundwater system is a leaky artesian system in karst formations,
with shallow and deep bedrock zones separated by a semi-confining shale layer.
Groundwater flow at the site is vertical as well as lateral, a condition that has resulted in
contamination of multiple aquifers and the need to extract groundwater at several depths.
Influence of Bodies of Surface Water
At Site A, the groundwater flow is subject to tidal influence in the upper few feet of the
upper-most aquifer. Water levels at the site sometimes have risen, and SVE wells at the
site have been flooded.
Influence of Groundwater Production Wells
AtDesMoines, groundwater flow is to the southeast; however, earlier high-volume
pumping from city wells may have affected the flow direction, facilitating the migration of
the contaminant plume.
6-3
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Contaminant Property Factors
The types and properties of the contaminants being treated at a site, such as whether the
contaminant has a tendency to be removed with extracted groundwater or to stay adsorbed to
subsurface soils, can affect the cost or performance of a remediation system. In addition, the
properties of the contaminants determine what treatment technologies are appropriate and the
complexity of the system required to treat contaminated groundwater ex situ or in situ. Examples
of sites where the contaminants (see Exhibits 2-4 and 2-5) and the contaminant properties
affected the cost or performance of the groundwater remediation are presented below.
Complex Mixture of Contaminants
At Sylvester/Gilson Road, contaminants included chlorinated solvents, such as methylene
chloride; nonchlorinated organics, such as toluene and phenols; and the metal selenium.
The mix of contaminants was treated above ground by a long series of operations,
includingpH adjustment, settling, neutralization, filtration, air stripping with vapor
incineration, and biological treatment.
Single Contaminant That was Relatively Easy to Treat
AtJMT, groundwater was contaminated with chlorinated solvents. The groundwater
treatment system consisted of only an air stripper, which was capable of reducing
contaminant concentrations to a level where the treated groundwater could be
discharged to an adjacent surface water body.
Extent of Contamination Factors
Groundwater contamination concentrated in an isolated areal and vertical extent typically is
easier and cheaper to remediate than the same mass of contaminant when it extends deeper and
spreads out over a larger area. This factor affects the size of the extraction and treatment system
and the system complexity in terms of the quantity of groundwater to be extracted from the
aquifer and treated ex situ. The volumes of contaminant plumes at each of the sites are presented
in Exhibit 2-6. The following examples show the effects of a relatively small and relatively large
extent of contamination at which groundwater remediation has been completed.
At Gold Coast, the initial areal extent of the contaminant plume was estimated to be 0.87
acres, and the initial volume of the plume was estimated to be less than 3 million gallons.
The site was remediated at a cost of less than $700,000.
At the Former Firestone facility, the initial areal extent of plume was estimated to be 100
acres (1,300 feet wide and 3,400 feet long), with an initial volume of as much as 2.9
billion gallons. The cost to remediate this site was nearly $13,000,000.
6-4
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Remedial Goal Factors
Remedial goal factors that may affect the cost and performance of a site cleanup include the
stringency of the cleanup levels, the types of remedial goals, the types of performance
requirements that have been established for the remediation as well as the system complexity
required to meet these goals. The following remedial goal factors can influence the volume or
areal extent of groundwater that must be treated, the type of treatment train that may be used, or
the length of time that a system has to be operated.
• Stringency of the cleanup levels
> maximum contaminant levels
*• approved alternate concentration limits
>• risk factors
*• other criteria
• Types of remedial goals
>• aquifer restoration
*• aquifer restoration and containment
>• other restoration goals
• Types of performance requirements
*• treated wastewater discharge limits
> air emission limits
More stringent cleanup levels can require more complex systems, longer periods of operation,
and larger volumes of groundwater to be treated. The type or stringency of the performance
goals (treatment of extracted groundwater and/or air emissions) affect the manner and the extent
to which extracted groundwater or off-gas from the remediation system must be treated before
discharge. The following examples show the effects of various remedial goals on the cost and
performance of site cleanup.
Types of Remedial Goals
At Western Processing, an aggressive P&T system, consisting of more than 200
groundwater extraction points pumping approximately 265 gpm, was installed to pursue
aquifer restorations goals. After approximately seven years of operation, an ESD was
issued to change the focus of remediation from restoration to containment. As a result of
this change, the system was modified to a system pumping approximately 80 gpm. This
modification significantly reduced the operating cost for the system.
Performance Goals Established
At Solid State, the site engineer identified institutional constraints that restricted the
operator's ability to reinject treated groundwater. Reinjection of groundwater may have
been a more cost efficient method for the disposal of treated groundwater, and could
have increased groundwater flow through the contaminated zone. This restriction is
believed to have increased the time required for site remediation more than any other
single factor.
6-5
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
In addition, as shown at the French, Ltd. site, different remedies specified in a site ROD can
impact cost and performance.
Under the ROD for French Ltd., modeling was used as a basis to select natural
attenuation as a component of the site remedy. Modeling showed that concentrations at
the boundaries of the site would be acceptable after 10 years of natural attenuation, and
that P&T, which was costing more than $3 million annually, could be terminated.
System Design and Operation Factors
In addition to site characteristics and remedial goals, system design and operation can affect cost
and performance during remediation. System operation factors include the amount of time the
system is operational and the adequacy of the system design to handle the nature and extent of
the contaminants. For example, the long percentages of downtime for a system (see Exhibit 3-5)
or problems with system design can increase the cost of a site cleanup. Conversely, various
efforts in system optimization at a site (as detailed in Exhibit 3-7) can reduce the cost of a site
cleanup and/or improve the performance of a system. Described below are examples in which
system operation factors affected the cost or performance at case study sites.
System Downtime
At King of Prussia, the treatment system has been operational approximately 7 6 per cent
of the time. Downtime has been caused by several factors, including the need to shut the
system down for two months to repair a crack in a filter, and has increased operating
costs.
System Optimization and Modification
After two years of operation, site engineers at Keefe performed an optimization study. As
a result, two new wells were installed at locations that would increase groundwater
extraction rates. Also, two existing wells were taken offline. Both extraction rates and
contaminant mass flux to the treatment system increased as a result of the modifications,
leading to more efficient capture of the plume.
When periodic groundwater monitoring results atMSWP indicated that aquifer cleanup
goals were met in five extraction wells, pumping from these wells was stopped and the
pumping rates from the other wells was adjusted to optimize system performance.
6-6
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
Additionally, the use of in situ technologies such as air sparging, ISB, and PRBs (see Exhibits
3-2 and 3-4) can lower the cost and improve performance of a remedial system. Because only
seven of the case study sites used in situ technologies, and similar technologies were used at very
few of these sites, it is not possible to draw significant conclusions about the effect of using in
situ technologies on the cost and performance of groundwater cleanups. However, two specific
examples of the effects of using in situ technologies are described below.
A t Intersil, the site owner replaced a P& T system that had been operating for eight years
with a PRB system. The PRB system continued to remove contaminant mass and reduce
concentrations of the contaminant in the aquifer, while minimizing the cost of treatment
and returning the site to sellable or leasable condition.
At Gold Coast, air sparging was used to mitigate elevated contaminant concentrations
around one well that was in a suspected source area. Once the contaminant levels in this
well were reduced, aquifer cleanup goals were able to be met, and the groundwater
remediation system at the site was able to be shut down.
6-7
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
6-8
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
7.0 REFERENCES
1. Cohen, R.M., J.W. Mercer, and R.M. Greenwald. 1998. EPA Ground Water Issue,
Design Guidelines for Conventional Pump-and-Treat Systems. EPA 540/S-97/504.
. September.
2. Federal Remediation Technologies Roundtable. 1998. Remediation Cost and
Performance Case Studies (28 total), . September.
3. Groundwater Remediation Technologies Analysis Center (GWRTAC). 1998.
Remediation Technologies, . July.
4. EPA. 1998. Guide to Documenting and Managing Cost and Performance Information
for Remediation Projects, Revised Version. EPA 542-B-98-007.
October.
5. EPA, Office of Emergency and Remedial Response (OERR). 1997. Cleaning Up the
Nation's Waste Sites: Markets and Technology Trends. EPA 542-R-96-005.
. April.*
6. U.S. Environmental Protection Agency (EPA), Office of Solid Waste and Emergency
Response (OS WER). 1997. Rules of Thumb for Superfund Remedy Selection. OSWER
Directive 9355.0-69, OSWER 9355.0-69, PB97-963301. EPA 540-R-97-013.
. August.*
7. EPA, OERR. 1996. GroundWater Cleanup at Superfund Sites. Directive 9283.1-11.
EPA 540-K-96/008. .
December.*
8. EPA, OERR. 1996. Presumptive Response Strategy and Ex Situ Treatment Technologies
for Contaminated Ground Water at CERCLA Sites: Final Guidance. Directive 9283.1-12.
EPA 540/R-96/023. .
October.
9. EPA, OERR. 1993. Guidance for Evaluating the Technical Impracticability of
Groundwater Restoration. Directive 9234.2-25.
. September.
10. EPA, R.S. Kerr Environmental Research Laboratory and OSWER. 1992. Estimating
Potential for Occurrence ofDNAPL at Superfund Sites. PB 92-963 338. Publication
9355.4-07FS. .
January. *
Available from the U.S. Department of Commerce National Technical Information
Service, 5285 Port Royal Road, Springfield, Virginia 22151; 1(800)553-6847
7-1
-------
Groundwater Cleanup: Overview of Operating Experience at 28 Sites
7-2
-------
United States
Environmental Protection
Agency
Office of
Research and Development
Washington. DC 20460
EPA625 4-91 025
May 1991
xvEPA Seminar Publication
Design and
Construction of
RCRA/CERCLA
Final Covers
-------
Technology Transfer EPA/625/4-91/025
Seminar Publication
Design and Construction
of RCRA/CERCLA
Final Covers
May 1991
Prepared for:
Center for Environmental Research Information
U.S. Environmental Protection Agency
26 West Martin Luther King Drive
Cincinnati, OH 45268
by;
Eastern Research Group, Inc.
6 Whittemore Street
Arlington, MA 02174
Printed on Recycled Paper
-------
NOTICE
The information in this document has been funded wholly or in part by the United States Environmental
Protection Agency under Contract 68-C8-0011 to Eastern Research Group, Inc. It has been subject to
the Agency's peer and administrative review, and it has been approved for publication as an EPA
document. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
-------
ACKNOWLEDGMENTS
This seminar publication is based wholly on papers presented at the U.S. Environmental Protec-
tion Agency (EPA) Technology Transfer seminars on Design and Construction of Resource Con-
servation and Recovery Act (RCRA) and Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA) Final Covers. These seminars were held in July and
August 1990 in Atlanta, Georgia; Philadelphia, Pennsylvania; Boston, Massachusetts; Dallas,
Texas; Kansas City, Missouri; Denver, Colorado; Newark, New Jersey; Chicago, Illinois; Seattle,
Washington; and Oakland, California.
The authors are:
Robert E, Landreth, U.S. Environmental Protection Agency, Risk Reduction Engineering
Laboratory (RREL), Cincinnati, Ohio (Chapters 1 and 5)
Dr. David E. Daniel, Department of Civil Engineering, University of Texas, Austin, Texas
(Chapters 2 and 6)
Dr. Robert M. Koerner, Drexel University, Geosynthetic Research Institute, Philadelphia,
Pennsylvania (Chapters 3, 4, and 7)
Dr. Paul R. Schroeder, Waterways Experiment Station, U.S. Army/Corps of Engineers,
Vicksburg, Mississippi (Chapters 8, 9, and 10)
Dr. Gregory N. Richardson, GN Richardson & Associates, Raleigh, North Carolina (Chapters
11 and 12}
Daniel J. Murray of EPA's Center for Environmental Research Information directed the project,
providing substantive guidance and review. David A. Carson of EPA's Risk Reduction Engineer-
ing Laboratory (RREL), Cincinnati, Ohio, Edwin F. Earth, Jr., of EPA's Center for Environmental
Research Information, Cincinnati, Ohio, and Kenneth R. Skahn of EPA's Office of Solid Waste
and Emergency Response peer reviewed the document. In addition, Frank Walberg, U.S. Army
Corps of Engineers, served as a special reviewer. Susan Richmond, Linda Saunders, Denise
Short, and Heidi Schultz of Eastern Research Group, Inc., provided editorial and production sup-
port.
-------
PREFACE
Cover systems are an essential part of all land disposal facilities. Covers control moisture infiltration
from the surface into closed facilities and limit the formation of leachate and its migration to ground
water. The Resource Conservation and Recovery Act (RCRA) Subparts G, K, and N form the basic
requirements for cover systems being designed and constructed today. In addition, the Comprehen-
sive Environmental Response, Compensation, and Liability Act (CERCLA) refers to RCRA Subtitle
C regulations, and many states have their own more stringent requirements.
This seminar publication provides regulatory and design personnel with an overview of design, con-
struction, and evaluation requirements for cover systems for RCRA/CERCLA waste management
facilities. It offers practical and detailed information on the design and construction of final covers for
both hazardous and nonhazardous waste landfills that comply with these requirements. As such it
should be valuable both to U.S. Environmental Protection Agency (EPA) regional and state person-
nel involved in evaluating and permitting hazardous waste facility closures and to the environmental
design and construction community.
Chapter One presents an overview of cover systems for waste management facilities, including
recommended designs for RCRA Subtitle C and CERCLA facilities. Chapter Two describes soils
used in typical cover systems and discusses critical parameters for soil liners as well as the effects
of environmental impacts such as frost action and settlement. Chapter Three focuses on geosyn-
thetic design and discusses geonet and geocomposite sheet drains, geopipe and geocomposite
edge drains, geotextile filters, geogrid reinforcement, and methane gas vents. Chapter Four covers
durability and aging of geomembranes, discussing in detail the mechanisms of degradation, as well
as synergistic effects, and accelerated testing methods. Chapter Five presents alternative designs
that meet the intent of regulations while adapting to site-specific concerns. Chapter Six discusses
construction quality assurance for soils, including testing of materials, and construction quality as-
surance during all phases of site preparation and soil placement. Chapter Seven covers construc-
tion quality control for geomembranes from manufacture and shipment to placement of the
geomembrane at the site. This chapter also presents destructive and nondestructive tests for sol-
vent and thermal seams in the field. Chapter Eight discusses evaluation of liquid management sys-
tems for landfills using the Hydrologic Evaluation of Landfill Performance (HELP) model. Chapter
Nine examines design parameter effects on cover performance. In Chapter Ten, gas management
systems are discussed with attention to gas generation, migration, and control strategies. Chapter
Eleven presents case studies of five closures, including RCRA industrial and commercial landfills,
one CERCLA lagoon and one CERCLA landfill, and one municipal solid waste commercial landfill.
The final chapter, Chapter Twelve, discusses postclosure monitoring of ground water, leachate, gas
generation, subsidence, surface erosion, and air quality.
This publication is not a design manual nor does it include all of the latest knowledge concerning
RCRA/CERCLA landfill cover systems; additional sources that provide more detailed information are
available. Some of these sources are referred to in the text of the individual chapters. In addition,
state and local authorities should be consulted for regulations and good management practices ap-
plicable to local areas.
IV
-------
TABLE OF CONTENTS
Page
1. OVERVIEW OF COVER SYSTEMS FOR WASTE MANAGEMENT FACILITIES 1
Introduction 1
Recommended Design for Subtitle C Facilities , 1
Low Hydraulic Conductivity Layer , 2
Compacted Soil Component 2
Geomembrane 2
Drainage Layer 2
Vegetation/Soil Top Layer , 3
Vegetation Layer , 3
Soil Layer 3
Optional Layers 3
Gas Vent Layer 3
Biotic Layer 4
Subtitle D Covers 4
CERCLA Covers 5
Applicability of RCRA Requirements 5
Relevant and Appropriate RCRA Requirements 6
State Equivalency 6
Closure , , 6
Applicability of Closure Requirements , , 6
Relevant and Appropriate Closure Requirements 7
References 7
2. SOILS USED IN COVER SYSTEMS 9
Introduction 9
Typical Cover Systems 9
Flow Rates Through Liners 9
Critical Parameters for Soil Liners 12
Materials :.. 12
Water Content 13
Compactive Energy 14
Size of Clods 16
Bonding of Lifts 18
Effects of Desiccation , , 18
Effects of Frost Action 20
Effects of Settlement 20
Interfacial Shear 23
Drainage Layers , 24
Summary 25
References • 25
3. GEOSYNTHETIC DESIGN FOR LANDFILL COVERS 27
General Comments on Design-by-Function 27
-------
Geomembrane Design Concepts 27
Geomembrane Compatibility 27
Vapor Transmission 27
Biaxial Stresses via Subsidence 28
Planar Stresses via Friction 28
Geonet and Geocomposite Sheet Drain Design Concepts 28
Compatibility 28
Crush Strength 28
Flow Capability 29
Geopipe and Geocomposite Edge Drain Design Concepts 30
Compatibility 30
Crush Strength 30
Flow Rate , 30
Geotextile Filter Design Considerations 30
Compatibility 30
Permeability 30
Geotextile Soil Retention , , 32
Geotextile Clogging Evaluation ., 32
Geogrid, or Geotextile, Cover Soil Reinforcement 32
Geotextile Methane Gas Vent 32
References 33
4. DURABILITY AND AGING OF GEOMEMBRANES 35
Polymers and Foundations 35
Mechanisms of Degradation 35
Ultraviolet Degradation , 35
Radiation Degradation 35
Chemical Degradation.... 35
Swelling Degradation 36
Extraction Degradation 36
Delamination Degradation 36
Oxidation Degradation 36
Biological Degradation 36
Synergistic Effects 37
Elevated Temperature 37
Applied Stresses 37
Long Exposure 37
Accelerated Testing Methods 37
Stress Limit Testing , 37
Rate Process Method for Pipe 37
Rate Process Method for Geomembranes 37
Arrhenius Modeling , 37
Multi-Parameter Prediction 39
Summary and Conclusions ,. ......40
5. ALTERNATIVE COVER DESIGNS 43
Introduction 43
Subtitle C 43
Subtitle D 43
CERCLA 43
Other Cover Designs 44
References 45
6, CONSTRUCTION QUALITY ASSURANCE FOR SOILS , 47
Introduction 47
Materials 47
Atterberg Limits 47
VI
-------
Percentage of Fines 47
Percentage of Gravel 47
Maximum Size of Particles or Clods 48
Requirements for Field Personnel 48
Frequency of Testing 48
Control of Subgrade Preparation 48
Soil Placement , , 48
Soil Compaction , 49
Drainage Layers 49
Barrier Materials 49
Protection of a Completed Lift 55
Sampling Pattern 58
Test Pads,.., 58
Outliers 58
Summary 61
Reference 61
7. CONSTRUCTION QUALITY CONTROL FOR GEOMEMBRANES 65
Preliminary Details 65
Manufacture 65
Fabrication of Panels 65
Storage at Factory 65
Shipment 65
Storage at Site , 65
Subgrade Preparation 65
Deployment of the Geomembrane 66
Geomembrane Field Seams 66
Solvent Seams 66
Thermal Seams 66
Extrusion Seams 68
Destructive Seam Tests 68
Nondestructive Seam Tests 69
Penetrations, Appurtenances, and Miscellaneous Details 70
Reference 71
8. HYDROLOGIC EVALUATION OF LANDFILL PERFORMANCE (HELP)
MODEL FOR DESIGN AND EVALUATION OF LIQUIDS
MANAGEMENT SYSTEMS 73
Introduction 73
Overview 73
Covers 73
Leachate Collection/Liner Systems 74
HELP Model 75
Background , 75
Process Simulation Methods 76
Infiltration 76
Evapotranspiration 77
Subsurface Water Routing 78
Vegetative Growth 79
Accuracy 79
Input Requirements 80
Climatological Data 80
Soil and Design Data 80
Output Description 81
Example Application 81
References 84
vn
-------
9. SENSITIVITY ANALYSIS OF HELP MODEL PARAMETERS 97
Introduction , 97
Comparison of Typical Cover Systems , 97
Design Parameters 97
Results 98
Effects of Vegetation 98
Effects of Topsoil Thickness 99
Effects of Topsoil Type 102
Use of Lateral Drainage Layer 103
Effects of Climate 103
Vegetative Layer Properties
Effects of SCS Runoff Curve Number 103
Effects of Evaporative Depth 104
Effects of Drainable Porosity 104
Effects of Plant Available Water Capacity 105
Liner/Drain Systems 106
Clay Liner/Drain Systems 107
Geomembrane/Drain Systems 109
Double Liner Systems 111
Summary of Sensitivity Analysis 115
References 116
10. GAS MANAGEMENT SYSTEMS 117
Gas Generation ....117
Gas Migration 117
Gas Control Strategies , 118
References 121
11. CASE STUDIES—RCRA/CERCLA CLOSURES 123
Introduction 123
Case 1: RCRA Commercial Landfill 123
Calculation of Localized Subsidence 123
Gas Collection Systems 124
Case 2: RCRA Industrial Landfill. 125
Case 3: CERCLA Lagoon Closure 129
Case 4: CERCLA Landfill Closure 132
Case 5: MSW Commercial Landfill 135
Conclusions 138
References 138
Additional References 140
12. POSTCLOSURE MONITORING... 141
Introduction 141
Ground-Water Monitoring 141
Leachate Monitoring .141
Gas Generation 143
Subsidence Monitoring 144
Surface Erosion 145
Air Quality Monitoring 145
References 145
Appendix A
Stability and Tension Considerations Regarding Cover Soils on Geomembrane-Lined Slopes A-1
Appendix B
Long-term Durability and Aging of Geomembranes B-1
vtn
-------
L/STOFF/G(//?ES
Figure Page
1-1 EPA-reeommended landfill cover design 2
1-2 EPA-recommended landfill cover with options 4
2-1 Soil liner, geomembrane liner, and composite liner 10
2-2 Soil liner and composite liner 12
2-3 Effect of bentonite upon the hydraulic conductivity of a bentonite-amended soil 14
2-4 Hydraulic conductivity and dry unit weight versus molding water content 15
2-5 Highly plastic soil compacted with standard Proctor procedures at a water content of 12% 16
2-6 Highly plastic soil compacted with standard Proctor procedures at a water content of 16% 16
2-7 Highly plastic soil compacted with standard Proctor procedures at a water content of 20% 16
2-8 Rototiller used to mix soil 17
2-9 Blades and teeth on rototiller 17
2-10 influence of compactive effort upon hydraulic conductivity and dry unit weight 19
2-11 Road recycler used to pulverize clods of soil 20
2-12 Passage of road recycler over loose lift of mudstone to reduce size of chunks of mudstone 21
2-13 Effect of imperfect bonding of lifts on hydraulic performance of soil liner 21
2-14 Example of heavy footed roller with long feet 22
2-15 Effect of desiccation upon the hydraulic conductivity of compacted clay 23
2-16 Relationship between distortion and tensile strain 24
2-17 Relationship between shearing characteristics of compacted soils and conditions of compaction 25
3-1 Required strength 28
3-2 Response of common geomembranes to the three-dimensional geomembrane tension test 29
3-3 Required geomembrane tension 29
3-4 Common crush strength behavior for geonets and geocomposites 30
3-5 ASTM D-4716 flow rate test 31
3-6 Crush strength of geopipe and geocomposite edge drain cores 32
3-7 Required strength of geogrid for cover soil reinforcement 33
3-8 Allowable gas flow as adapted from ASTM D-4716 34
4-1 Wavelength spectrum of visible and ultraviolet radiation 36
4-2 Stress limit testing for plastic pipe 38
4-3 Rate process method for testing pipe 38
4-4 Rate process method for testing geomembranes 39
4-5 Testing device for Arrhenius modeling 39
4-6 Reaction rate for impact testing of polyethylene shielding 40
4-7 Experimental and field-measured response curves for multi-parameter lifetime prediction 41
5-1 Resistive layer barrier 44
5-2 Conductive layer barrier 45
5-3 Side view of bioengineered lysimeter. Surface runoff is collected from both engineered surface and
soil surface. Soil moisture content is measured with neutron probe. Water table is measured
in well 45
6-1 Traditional method for specification of acceptable water contents and dry unit weights 50
6-2 Data from Mitchell et al. for silty clay compacted with impact compaction 51
6-3 Compaction data for silty clay (6); solid symbols represent specimens with hydraulic conductivity less
than or equal to 1 x 10~7 cm/s and open symbols represent specimens with hydraulic conductivity
>1 x10"7cm/s , 52
6-4 Contours of constant hydraulic conductivity for silty clay compacted with kneading compaction 52
6-5 Recommended procedure 53
6-6 Use of hydraulic conductivity and shear strength data to define a single, overall acceptable zone 55
6-7 Possible approaches for specifying lower limit of Acceptable Zone: (A) minimum degree of
saturation, S; and (B) line of optimums 56
6-8 Compaction curves for Type A soil from East Borrow area at Oak Ridge Y-12 operations project 57
6-9 Hydraulic conductivity versus molding water content for Type A soil from East Borrow area at Oak
Ridge Y-12 operations project 57
6-10 Acceptable zone for Type A soil from East Borrow area at Oak Ridge Y-12 operations project 58
tx
-------
6-11 Pushing of thin-walled sampling tube with abackhoe 59
6-12 Tilting of sampling tube during push 59
6-13 Placement of hydraulic jack on top of sampling tube 60
6-14 Use of backhoe as a reaction for hydraulic jack , 60
6-15 Checklist of critical variables for CQA of low hydraulic conductivity compacted soil used in a
cover system 62
6-16 Checklist of critical variables for CQA of drainage materials used in a cover system 63
7-1 Shear and peel test for geomembrane seams 68
8-1 Cover and liner edge configuration with example toe drain , 73
8-2 Schematic of a single clay liner system for a landfill 74
8-3 Schematic of a double liner and leak detection system for a landfill 75
8-4 Simulation processes in the HELP model 76
8-5 Typical hazardous waste landfill profile 81
8-6 Completed data form for landfill materials and design 82
8-7 Completed data form for climatologieal data 84
8-8 Example output 85
9-1 Cover designs for sensitivity analysis 99
9-2 Bar graph for three-layer cover design showing effect of surface vegetation, topsoil type,
and location 100
9-3 Bar graph for two-layer cover design showing effect of topsoil depth, surface vegetation,
and location ,, 100
9-4 Effect of saturated hydraulic conductivity on lateral drainage and percolation 109
9-5 Effect of ratio of drainage layer saturated hydraulic conductivity to soil liner saturated hydraulic
conductivity on ratio of lateral drainage to percolation for a steady-state (SS) inflow of
20 cm/yr(8 in./yr) , 110
9-6 Effect of ratio of drainage layer saturated hydraulic conductivity to soil liner saturated hydraulic
conductivity on ratio of lateral drainage to percolation for an unsteady inflow of
127 cm/yr (50 in./yr) 110
9-7 Effect of ratio of drainage length to drainage layer slope on the average saturated depth in
drainage layer (KD=10~2 cm/s) above a soil liner (KP=lO"7cm/s) under a steady-state inflow
rate of 20 cm/yr(8 in./yr) 111
9-8 Synthetic liner leakage fraction as a function of density of holes, size of holes, head on the liner
and saturated hydraulic conductivity of the liner 112
9-9 Effect of leakage fraction on system performance 112
9-10 Liner designs 113
9-11 Percent of inflow to primary leachate collection layer discharging from leakage detection layer and
bottom liner for double liner systems C and E 114
9-12 Percent of inflow to primary leachate collection layer discharging from leakage detection layer and
bottom liner for double liner systems Dand F 114
10-1 Cover with gas vent outlet and vent layer 118
10-2 Gravel vent and gravel-filled trench used to control lateral gas movement in a sanitary landfill 119
10-3 Typical trench barrier system 119
10-4 Gas control barriers , 120
10-5 Gas extraction well for landfill gas control 121
10-6 Gas extraction well design 122
11-1 Case 1 - Cap profile and geometry 124
11-2 Case 1 - General subsidence model 125
11-3 Case 1 - Cumulative subsidence 126
11-4 Case 1 - Geomembrane strains intrench subsidence 126
11-5 Case 1 - Uniaxial and biaxial geomembrane response 127
11-6 Case 1 - Subsidence strain in soil barrier 127
11-7 Case 1 - Ultimate tensile strain in clays 128
11-8 Case 1 - Gas collector system , ,, 129
11-9 Case 2 - Cap profile and geometry 130
11-10 Case 2 - Direct shear data: texture HOPE 130
11-11 Case 2 - Slope factors for soil loss evaluation 131
11-12 Case 2 - Sideslope armoring scheme 131
-------
11-13 Case 3 - Cap profile and geometry 132
11-14a Case3 - Placement of geogrid over geomembrane , 133
11-14b Case 3 - Placement of drainage layer over geogrid 133
11-l5a Case 3 - Outlet detail for sideslope toe surface water drainage layer 134
H-15b Cases- Erosion at drainage layer outlet ...134
11-16 Case 4 - Cap profile and geometry ,...,,135
I1-17a Case 4- Placement of geotextile on asphalt emulsion 136
H-l7b Case 4 - Placement of chip seal on geotextile 136
11-18 Cases - Cap profile and geometry 137
11-19 Case 5 - Profile showing MSW subcells. 138
11-20 CaseS - Gas collector well array 139
11-21 Case 5 - Perimeter gas monitoring well 139
12-1 Monitoring well configuration 142
12-2 Monitoring interbedded aquifer 142
12-3 Impact of biological growth on filters 143
12-4 Gas generation versus time , 143
XI
-------
LIST OF TABLES
Table Page
2-1 Calculated Flow Rates through Soil Liners with 30 cm of Water Ponded on the Liner 10
2-2 Calculated Flow Rates through a Geomembrane with a Head of 30 cm of Water above
the Geomembrane 10
2-3 Calculated Flow Rates for Composite Liners with a Head of Water of 30 cm 11
2-4 Calculated Flow Rates for Soil Liners, Geomembrane Liners, and Composite Liners 13
2-5 Effect of Size of Clods during Processing of Soil upon Hydraulic Conductivity of Soil
after Compaction , 18
3-1 Customary Primary Functions of Geosynthetics Used in Waste Containment Systems 27
4-1 Typical Formulations of Geomembranes 35
6-1 Recommended Materials Tests for Barrier Layers 48
6-2 Recommended Tests and Observations on Subgrade Preparation 49
6-3 Recommended Tests and Observations on Compacted Soil for Barrier Layers 54
7-1 Overview of Geomembrane Field Seams 67
7-2 Overview of Nondestructive Seam Tests 69
9-1 Parameters Selected for Sensitivity Analysis 97
9-2 Climatological Regimes 98
9-3 Effects of Climate and Vegetation 101
9-4 Effects of Climate and Topsoil Thickness 101
9-5 Effects of Climate and Topsoil Types 102
9-6 Effects of Evaporative Depth and Runoff Curve Number 105
11-1 Soil Texture Constant for Soil Loss Evaluation , 131
12-1 Threshold Limits of Air Contamination 144
XII
-------
CHAPTER 1
OVERVIEW OF COVER SYSTEMS FOR WASTE MANAGEMENT FACILITIES
INTRODUCTION
Proper closure is essential to complete a filled hazardous
waste landfill. Research has established minimum re-
quirements needed to meet the stringent, necessary,
closure regulations in the United States. In designing the
landfill cover, the objective is to limit the infiltration of
water to the waste so as to minimize creation of leachate
that could possibly escape to ground-water sources.
Minimizing leachates in a closed waste management unit
requires that liquids be kept out and that the leachate
that does exist be detected, collected, and removed.
Where the waste is above the ground-water zone, a
properly designed and maintained cover can prevent (for
practical purposes) water from entering the landfill and,
thus, minimize the formation of leachate.
The cover system must be devised at the time the site is
selected and the plan and design of the landfill contain-
ment structure is chosen. The location, the availability of
soil with a low permeability or hydraulic conductivity, the
stockpiling of good topsoil, the availability and use of
geosynthetics to improve performance of the cover sys-
tem, the height restrictions to provide stable slopes, and
the use of the site after the postclosure care period are
typical considerations. The goals of the cover system are
to minimize further maintenance and to protect human
health and the environment.
Subparts G, K, and N of the Resource Conservation and
Recovery Act (RCRA) Subtitle C regulations form the
basic requirements for cover systems being designed
and constructed today. Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA)
regulations refer to the RCRA Subtitle C regulations but
other criteria, primarily approved state requirements, also
have to be evaluated for applicability. The proposed
RCRA Subtitle D regulations base cover requirements
primarily on the hydraulic conductivity of the bottom liner.
RECOMMENDED DESIGN FOR SUBTITLE C
FACILITIES
After the hazardous waste management unit is closed,
the U.S. Environmental Protection Agency (EPA) recom-
mends (1) that the final cover (Figure 1-1) consist of,
from bottom to top:
1. A Low Hydraulic Conductivity Geomembrane/Soil
Layer. A 60-cm (24-in.) layer of compacted natural or
amended soil with a hydraulic conductivity of 1 x 10"7
cm/sec in intimate contact with a minimum 0.5-mm
(20-mil) geomembrane liner.
2. A Drainage Layer. A minimum 30-cm (12-in.) soil
layer having a minimum hydraulic conductivity of 1 x
10~2 cm/sec, or a layer of geosynthetic material
having the same characteristics.
3. A Top, Vegetation/Soil Layer. A top layer with
vegetation (or an armored top surface) and a mini-
mum of 60 cm (24 in.) of soil graded at a slope bet-
ween 3 and 5 percent.
Because the design of the final cover must consider the
site, the weather, the character of the waste, and other
site-specific conditions, these minimum recommenda-
tions may be altered providing the alternative design is
equivalent to the EPA-recommended design or will meet
the intent of the regulations. EPA encourages design in-
novation and will accept an alternative design provided
the owner or operator demonstrates the new design's
equivalency. For example, in extremely arid regions, a
gravel top surface might compensate for reduced vegeta-
tion, or the middle drainage layer might be expendable.
Where burrowing animals might damage the
geomembrane/low hydraulic conductivity soil layer, a
biotic barrier layer of large-sized cobbles may be needed
above it. Where the type of waste may create gases, soil
or geosynthetic vent structures would need to be in-
cluded.
Settlement and subsidence should be evaluated for all
covers and accounted for in the final cover plans. The
current operating procedures for RCRA Subtitle C
facilities (e.g., banning of liquids and partially filled drums
of liquids) usually do not present major settlement or sub-
sidence issues. For RCRA Subtitle D facilities, however,
the normal decomposition of the waste will invariably
result in settlement and subsidence. Settlement and sub-
sidence can be significant, and special care may be re-
quired in designing the final cover system. The cover
design process should consider the stability of all the
waste layers and their intermediate soil covers, the soil
and foundation materials beneath the landfill site, all the
-------
vegetation/soil
top layer
drainage layer
low hydraulic conductivity
geomembrane/soil layer
waste
60cm
ipMp4£%:Si%3:J -*- filter layer
30 cm
«*- 0.5-mm (20-mll)
60 cm geomembrane
O
o
0
O
o
o
o
Figure 1-1. EPA-recommended landfill cover design (1).
liner and leachate collection systems, and all the final
cover components. When a significant amount of settle-
ment and subsidence is expected within 2 to 5 years of
closure, an interim cover that protects human health and
the environment might be proposed. Then when settle-
ment/subsidence is essentially complete, the interim
cover could be replaced or incorporated into a final cover.
Low Hydraulic Conductivity Layer
The function of the composite low hydraulic conductivity
layer, composed of soil and a geomembrane, is to
prevent moisture movement downward from the overlying
drainage layer.
Compacted Soil Component
EPA recommends a test pad be constructed before the
low hydraulic conductivity soil layer is put in place to
demonstrate that the compacted soil component can
achieve a maximum hydraulic conductivity of 1 x 1Q~7
cm/sec. To ensure that the design specifications are at-
tainable, a test pad uses the same soil, equipment, and
procedures to be used in constructing the low hydraulic
conductivity layer. For Subtitle D facilities, the test fill
should be constructed on part of the solid waste material
to determine the impact of compacting soil on top of less
resistive municipal solid waste.
The low hydraulic conductivity soil component placed
over the waste should be at least 60-cm (24-in.) deep;
free of detrimental rock, clods, and other soil debris; have
an upper surface with a 3 percent maximum slope; and
be below the maximum frost line. The surface should be
smooth so that no small-scale stress points are created
for the geomembrane.
In designing the low hydraulic conductivity layer, the
causes of failure—subsidence, desiccation cracking, and
freeze/thaw cycling—must be considered. Most of the
settling will have taken place by the time the cover is put
into place, but there is still a potential for further sub-
sidence. Although estimating this potential is difficult, in-
formation about voids and compressible materials in the
underlying waste will aid in calculating subsidence.
A soil with low cracking potential should be selected for
the soil component of the low hydraulic conductivity layer.
The potential for desiccation cracking of compacted clay
depends on the physical properties of the compacted
clay, its moisture content, the local climate, and the mois-
ture content of the underlying waste.
Because freeze/thaw conditions can cause soil cracking,
lessen soil density, and lessen soil strength, this entire
low hydraulic conductivity/geomembrane layer should be
below the depth of the maximum frost penetration. In
northern areas, then, the maximum depth of the top
vegetation/soil layer would be greater than the recom-
mended minimum of 60 cm (24 in.).
Penetrating this low hydraulic conductivity/geomembrane
soil layer with gas vents or drainage pipes should be kept
to a minimum. Where a vent is necessary, there should
be a secure, liquid-tight seal between the vent and the
geomembrane. If settlement or subsidence is a major
concern, this seal must be designed for flexibility to allow
for vertical movement.
Geomembrane
The geomembrane placed on the smooth, even, low
hydraulic conductivity layer should be at least 0.5-mm
(20-mils) thick. The minimum slope surface should be 3
percent after any settlement of the soil layer or sub-base
material. Stress situations such as bridging over sub-
sidence and friction between the geomembrane and
other cover components (i.e., compacted soil, geosyn-
thetic drainage material, etc.), especially on side slopes,
will require special laboratory tests to ensure the design
has incorporated site-specific materials.
Drainage Layer
The drainage layer should be designed to minimize the
time the infiltrated water is in contact with the bottom, low
hydraulic conductivity layer and, hence, to lessen the
-------
potential for the water to reach the waste (see Figure 1 -
1). Water that filters through the top layer is intercepted
and rapidly moved to an exit drain, such as by gravity
flow to a toe drain.
If the granular material in the drainage layer is sand, the
minimum requirements are that it should be at least 30-
cm (12-in.) deep with a hydraulic conductivity of 1 x 10"2
cm/sec or greater. Drainage pipes should not be placed
in any manner that would damage the geomembranes.
If geosynthetic materials are used in the drainage layer,
the same physical and hydraulic requirements should be
met, e.g., equivalency in hydraulic transmissivity, lon-
gevity, compatibility with geomembrane, compressibility,
conformance to surrounding materials, and resistance to
clogging. Geosynthetic materials are gaining increased
use and understanding of their performance. Manufac-
turers are also continuing to improve the basic resin
properties to improve their long-term durability. The net
result is that organizations such as the American Society
of Testing Materials (ASTM) and the Geosynthetic Re-
search Institute (GRI), Drexel University, Philadelphia,
Pennsylvania, are continually developing new evaluation
procedures to better correlate with design and field ex-
periences.
Between the bottom of the top-layer soil and the
drainage-layer sand, a granular or geosynthetic filter
layer should be included to prevent the drainage layer
from clogging by top-layer fines. The criteria established
for the grain size of granular filter sand are designed to
minimize the migration of fines from the overlying top
layer into the drainage layer. (For information on filter
criteria, refer to the EPA Technical Guidance Document
[1].) ASTM test procedures have also been established
to evaluate paniculate clogging potential of geosyn-
thetics.
Vegetation/Soil Top Layer
Vegetation Layer
The upper layer of the two-component top layer (Figure
1-1) should be vegetation (or another surface treatment)
that will allow runoff from major storms while inhibiting
erosion. Vegetation over soil (part of which is topsoil) is
the preferred system, although, in some areas, vegeta-
tion may be unsuitable.
The temperature- and drought-resistant vegetation
should be indigenous; have a root system that does not
extend into the drainage layer; need no maintenance;
survive in low-nutrient soil; and have sufficient density to
control the rate of erosion to the recommended level of
less than 5.5 MT/ha/yr (2 ton/acre/yr).
The surface slope should be the same as that of the un-
derlying soils; at least 3 percent but no greater than 5
percent. To support the vegetation, this top layer should
be at least 60-cm (24-in.) deep and include at least 15-
cm (6-in.) of topsoil. To help the plant roots develop, this
layer should not be compacted. In some northern
climates, this top layer may need to be more than the
minimum 60 cm (24 in.) to ensure that the bottom low
hydraulic conductivity layer remains below the frost zone.
Where vegetation cannot be maintained, particularly in
arid areas, other materials should be selected to prevent
erosion and to allow for surface drainage. Asphalt and
concrete are apt to deteriorate because of thermal-
caused cracking or deform because of subsidence.
Therefore, a surface layer 13 to 25-cm (5 to 10-in.) deep
of 5 to 10-cm (2 to 4-in.) stones or cobbles would be
more effective. Although cobbles are a one-way valve
and allow rain to infiltrate, this phenomenon would be of
less concern in arid areas. In their favor, cobbles resist
wind erosion well.
Soli Layer
The soil in this 60-cm (24-in.) top layer should be capable
of sustaining nonwoody plants, have an adequate water-
holding capacity, and be sufficiently deep to allow for ex-
pected, long-term erosion losses. A medium-textured soil
such as a loam would fit these requirements. If the landfill
site has sufficient topsoil, it should be stockpiled during
excavation for later use.
The final slopes of the cover should be uniform and at
least 3 percent, and should not allow erosion rills and gul-
lies to form. Slopes greater than 5 percent will promote
erosion unless controls are built in to limit erosion to less
than 5.5 MT/ha/yr (2 ton/acre/yr). The U.S. Department of
Agriculture's (USDA's) Universal Soil Loss Equation is
recommended as the tool to evaluate erosion potential.
Optional Layers
Although other layers may be needed on a site-specific
basis, the common optional layers are those for gas
vents and for a biotic barrier layer (Figure 1 -2).
Gas Vent Layer
The gas vent layer should be at least 30-cm (12-in.) thick
and be above the waste and below the low hydraulic con-
ductivity layer. Coarse-grained porous material, similar to
that used in the drainage layer or equivalent-performing
synthetic material, can be used.
The perforated, horizontal venting pipes should channel
gases to a minimum number of vertical risers located at a
high point (in the cross section) to promote gas ventila-
tion. To prevent clogging, a granular or geotextile filter
may be needed between the venting and the low
hydraulic conductivity soil geomembrane layers.
As an alternative, vertical, standpipe gas collectors can
be built up as the landfill is filled with waste. These
standpipes, which may be constructed of concrete, can
be 30 cm (12 in.) or more in diameter and may also be
used to provide access to measure leachate levels in the
landfill.
-------
cobbles/soil
top layer
biotic barrier
(cobbles)
drainage layer
low hydraulic conductivity
geomembrane/soil layer
gas vent layer
waste
0 ^ O 0
o (7
Q 0
0
c>
i 0
0> o r° 0
60cm
30cm
30cm
60cm
30cm
geosynthetic filter
geosynthetic filter
- 0.5-mm {20 mil)
geomembrane
geosynthetic filter
Figure 1-2. EPA-recommended landfill cover with options (1).
Biotic Layer
Plant roots or burrowing animals (collectively called
biointruders) may disrupt the drainage and the low
hydraulic conductivity layers to interfere with the drainage
capability of the layers. A 90-cm (3-ft.) biotic barrier of
cobbles directly beneath the top vegetation layer may
stop the penetration of some deep-rooted plants and the
invasion of burrowing animals. Most research on biotic
barriers has been done in, and is applicable to, arid
areas. Geosynthetic products that incorporate a time-
released herbicide into the matrix or on the surface of the
polymer may also be used to retard plant roots. The lon-
gevity of these products requires evaluation if the cover
system is to serve for longer than 30 to 50 years.
SUBTITLE D COVERS
The cover system in nonhazardous waste landfills (Sub-
title D) will be a function of the bottom liner system and
the liquids management strategy for the specific site. If
the bottom liner system contains a geomembrane, then
the cover system should contain a geomembrane to
prevent the "bathtub" effect. When the bottom liner is less
permeable than the cover system, e.g., geomembrane on
the bottom and natural soil on the top, the facility will "fill
up" with infiltration water (through the cover) unless an
active leachate removal system is in place. Likewise, if
the bottom liner system is a natural soil liner, then the
cover system barrier should be hydraulically equivalent to
or less than the bottom liner system. A geomembrane
used in the cover will prevent the infiltration of moisture to
the waste below and may contribute to the collection of
waste decomposition gases, therefore necessitating a
gas-vent layer.
There are at least two options to consider under a liquids
management strategy, mummification and recirculation.
in the mummification approach the cover system is
designed, constructed, and maintained to prevent mois-
ture infiltration to the waste below. The waste will even-
tually approach and remain in a state of "mummification"
until the cover system is breached and moisture enters
the landfill. A continual maintenance program is neces-
sary to maintain the cover system in a state of good
repair so that the waste does not decompose to generate
ieachate and gas.
The recirculation concept results in the rapid physical,
chemical, and biological stabilization of the waste. To ac-
complish this, a moisture balance is maintained within
the landfill that will accelerate these stabilization proces-
ses. This approach requires geomembranes in both the
bottom and top control systems to prevent leachate from
getting out and excess moisture from getting in. In addi-
tion, the system needs a leachate collection and removal
system on the bottom and a leachate injection system on
the top, maintenance of this system for a number of
years (depending on the size of the facility), and a gas
collection system to remove the waste decomposition
gases. In a modern landfill facility, all of these elements,
except the leachate injection system, would probably be
available. The benefit of this approach is that, after
stabilization, the facility should not require further main-
tenance. A more important advantage is that the decom-
posed and stabilized waste may be removed and used
like compost, the plastics and metals could be recycled,
and the site used again. If properly planned and operated
in this manner, several cells could serve all of a
community's waste management needs.
A natural soil material may be used in a cover system
when the bottom liner system is also natural soils and the
regulatory requirements will permit. A matrix of soil char-
acteristics (using either USDA or USCS) and health, aes-
thetics, and site usage characteristics can be developed
to provide information on which soil or combination of
soils will be the most beneficial.
Health considerations demand the evaluation of each soil
type to minimize vector breeding areas and attractive-
ness to animals. The soil should minimize moisture in-
-------
filtration (best accomplished by fine grain soils) while al-
lowing gas movement (coarse grain soils are best). This
desired combination of seemingly opposite soil properties
suggests a layered system. The soil should also minimize
fire potential.
Aesthetic considerations include minimizing blowing of
paper and other waste, controlling odors, and providing a
sightly appearance. All landfill operators strive to be good
neighbors and these considerations are very important
for community relations.
The landfill site may be used for a variety of activities
after closure. For this reason, cover soils should minimize
settlement and subsidence, maximize compaction, assist
vehicle support and movement, allow for equipment
workability under all weather conditions, and allow heal-
thy vegetation to grow. The future use of the site should
be considered at the initial landfill design stages so that
appropriate end-use design features can be incorporated
into the cover during the active life of the facility.
CERCLA COVERS
The Superfund Amendments and Reauthorization Act of
1986 (SARA) adopts and expands a provision in the
1985 National Contingency Plan (NCR) that remedial ac-
tions must at least attain applicable or relevant and ap-
propriate requirements (ARARs). Section 121 (d) of
CERCLA, as amended by SARA, requires attainment of
federal ARARs and of state ARARs in state environmen-
tal or facility siting laws when the state requirements are
promulgated, more stringent than federal laws, and iden-
tified by the state in a timely manner.
CERCLA facilities require information on whether or not
the site is under the jurisdiction of RCRA regulations. The
cover system design can then be developed based on
appropriate regulations.
RCRA Subtitle C requirements for treatment, storage,
and disposal facilities (TSDFs) will frequently be ARARs
for CERCLA actions, because RCRA regulates the same
or similar wastes as those found at many CERCLA sites,
covers many of the same activities, and addresses
releases and threatened releases similar to those found
at CERCLA sites. When RCRA requirements are ARARs,
only the substantive requirements of RCRA must be met
if a CERCLA action is to be conducted on site. Substan-
tive requirements are those requirements that pertain
directly to actions or conditions in the environment. Ex-
amples include performance standards for incinerators
(40 CFR 264.343), treatment standards for land disposal
of restricted waste (40 CFR 268), and concentration
limits, such as maximum contaminant levels (MCLs). On-
site actions do not require RCRA permits or compliance
with administrative requirements. Administrative require-
ments are those mechanisms that facilitate the im-
plementation of the substantive requirements of a statute
or regulation. Examples include the requirements for
preparing a contingency plan, submitting a petition to
delist a listed hazardous waste, recordkeeping, and con-
sultations. CERCLA actions to be conducted off site must
comply with both substantive and administrative RCRA
requirements.
APPLICABILITY OF RCRA REQUIREMENTS
RCRA Subtitle C requirements for the treatment, storage,
and disposal of hazardous waste are applicable for a Su-
perfund remedial action if the following conditions are
met (2):
1. The waste is a RCRA hazardous waste, and either:
2. The waste was initially treated, stored, or disposed of
after the effective date of the particular RCRA re-
quirement
or
The activity at the CERCLA site constitutes treat-
ment, storage, or disposal, as defined by RCRA.
For RCRA requirements to be applicable, a Superfund
waste must be determined to be a listed or characteristic
hazardous waste under RCRA. A waste that is hazard-
ous because it once exhibited a characteristic (or a
media containing a waste that once exhibited a charac-
teristic) will not be subject to Subtitle C regulation if it no
longer exhibits that characteristic. A listed waste may be
delisted if it can be shown not to be hazardous based on
the standards in 40 CFR 264.22. If such a waste will be
shipped off site, it must be delisted through a rulemaking
process. To delist a RCRA hazardous waste that will
remain on site at a Superfund site, however, only the
substantive requirements for delisting must be met.
Any environmental media (i.e., soil or ground water) con-
taminated with a listed waste is not a hazardous waste,
but must be managed as such until it no longer contains
the listed waste—generally when constituents from the
listed waste are at health-based levels. Delisting is not
required.
To determine whether a waste is a listed waste under
RCRA, it is often necessary to know the source of that
waste. For any Superfund site, if determination cannot be
made that the contamination is from a RCRA hazardous
waste, RCRA requirements will not be applicable. This
determination can be based on testing or on best profes-
sional judgment (based on knowledge of the waste and
its constituents).
A RCRA requirement will be applicable if the hazardous
waste was treated, stored, or disposed of after the effec-
tive date of the particular requirement. The RCRA Sub-
title C regulations that established the hazardous waste
management system first became effective on November
19, 1980. Thus, RCRA regulations will not be applicable
to wastes disposed of before that date, unless the
CERCLA action itself constitutes treatment, storage, or
disposal (see below). Additional standards have been is-
-------
sued since 1980; therefore, applicable requirements may
vary somewhat, depending on the specific date on which
the waste was disposed.
RCRA requirements for hazardous wastes will also be
applicable if the response activity at the Superfund site
constitutes treatment, storage, or disposal, as defined
under RCRA. Because remedial actions frequently in-
volve grading, excavating, dredging, or other measures
that disturb contaminated material, activities at Superfund
sites may constitute disposal, or placement, of hazardous
waste. Disposal of hazardous waste, in particular, trig-
gers a number of significant requirements, including
closure requirements and land disposal restrictions,
which require treatment of wastes prior to land disposal.
(See Guides on Superfund Compliance with Land Dis-
posal Restrictions, OSWER Directives 9347.3-01 FS
through 9237.3-06FS, for a detailed description of these
requirements.)
EPA has determined that disposal occurs when wastes
are placed in a land-based unit. However, movement
within a unit does not constitute disposal or placement,
and at CERCLA sites, an area of contamination (AOC)
can be considered comparable to a unit. Therefore,
movement within an AOC does not constitute placement.
Relevant and Appropriate RCRA Requirements
RCRA requirements that are not applicable may, none-
theless, be relevant and appropriate, based on site-
specific circumstances. For example, if the source or
prior use of a CERCLA waste is not identifiable, but the
waste is similar in composition to a known, listed RCRA
waste, the RCRA requirements may be potentially
relevant and appropriate, depending on other circumstan-
ces at the site. The similarity of the waste at the CERCLA
site to RCRA waste is not the only, nor necessarily the
most important, consideration in the determination. An in-
depth, constituent-by-constituent analysis is generally
neither necessary nor useful, since most RCRA require-
ments are the same for a given activity or unit, regardless
of the specific composition of the hazardous waste.
The determination of relevance and appropriateness of
RCRA requirements is based instead on the circumstan-
ces of the release, including the hazardous properties of
the waste, its composition and matrix, the characteristics
of the site, the nature of the release or threatened
release from the site, and the nature and purpose of the
requirement itself. Some requirements may be relevant
and appropriate for certain areas of the site, but not for
other areas. In addition, some RCRA requirements may
be relevant and appropriate at a site, while others are
not, even for the same waste. For example, at one site
minimum technology requirements may be considered
relevant and appropriate for an area receiving waste be-
cause of the high potential for migration of contaminants
in hazardous levels to ground water, but not for another
area that contains relatively immobile waste. Land dis-
posal restrictions at the same site may not be relevant
and appropriate for either area because the required
treatment technology is not appropriate, given the matrix
of the waste. Only those requirements that are deter-
mined to be both relevant and appropriate must be at-
tained.
State Equivalency
A state may be authorized to administer the RCRA haz-
ardous waste program in lieu of the federal program
provided the state has equivalent authority. Authorization
is granted separately for the basic RCRA Subtitle C
program, which includes permitting and closure of
TSDFs; for regulations promulgated pursuant to the Haz-
ardous and Solid Waste Amendments (HSWA), such as
land disposal restrictions; and for other programs, such
as delisting of hazardous wastes. If a site is located in a
state with an authorized RCRA program, the state's
promulgated RCRA requirements will replace the
equivalent federal requirements as potential ARARs.
An authorized state program may also be more stringent
than the federal program. For example, a state may have
more stringent test methods for characteristic wastes, or
may list more wastes as hazardous than the federal
program does. Therefore, it is important to determine
whether laws in an authorized state go beyond the
federal regulations.
Closure
For each type of unit regulated under RCRA, Subtitle C
regulations contain standards that must be met when a
unit is closed. For treatment and storage units, the
closure standards require that all hazardous waste and
hazardous waste residues be removed. In addition to the
option of closure by removal, called clean closure, units
such as landfills, surface impoundments, and waste piles
may be closed as disposal or landfill units with waste in
place, referred to as landfill closure. Frequently, the
closure requirements for such land-based units will be
either applicable or relevant and appropriate at Super-
fund sites.
Applicability of Closure Requirements
The basic prerequisites for applicability of closure re-
quirements are (1) the waste must be hazardous waste;
and (2) the unit (or AOC) must have received waste after
the RCRA requirements became effective, either be-
cause of the original date of disposal or because the
CERCLA action constitutes disposal. When RCRA
closure requirements are applicable, the regulations
allow only two types of closure:
• Clean Closure. All waste residues and contaminated
containment system components (e.g., liners), con-
taminated subsoils, and structures and equipment con-
taminated with waste leachate must be removed and
managed as hazardous waste or decontaminated
before the site management is completed [see 40 CFR
264.111,264.228(3)].
-------
» Landfill Closure, The unit must be capped with a final
cover designed and constructed to:
• Provide long-term minimization of migration ot li-
quids.
• Function with minimum maintenance,
• Promote drainage and minimize erosion.
• Accommodate settling and subsidence.
• Have a hydraulic conductivity less than or equal to
any bottom liner system or natural subsoils present.
Clean closure standards assume the site will have un-
restricted use and require no maintenance after the
closure has been completed. These standards are often
referred to as the "eatable solid, drinkable leachate"
standards. In contrast, disposal or landfill closure stand-
ards require postclosure care and maintenance of the
unit for at least 30 years after closure. Postclosure care
includes maintenance of the final cover, operation of a
leachate and removal system, and maintenance of a
ground-water monitoring system [see 40 CFR 264.117,
264.228(b)].
EPA has prepared several guidance documents on
closure and final covers (1, 3). These guidance docu-
ments are not ARARs, but are to be considered for
CERCLA actions and may assist in complying with these
regulations. The performance standards in the regulation
may be attained in ways other than those described in
guidance, depending on the specific circumstances of the
site.
Relevant and Appropriate Closure Requirements
If they are not applicable, RCRA closure requirements
may be determined to be relevant and appropriate. There
is more flexibility in designing closure for relevant and ap-
propriate requirements because the Agency has the
flexibility to determine which requirements in the closure
standards are relevant and appropriate. Under this
scenario, a hybrid closure is possible. Depending on the
site circumstances and the remedy selected, clean
closure, landfill closure, or a combination of requirements
from each type of closure may be used.
The proposed revisions to the NCP discuss the concept
of hybrid closure (53 FR 51446). The NCP illustrated the
following possible hybrid closure approaches:
» Hybrid-Clean Closure. Used when leachate will not im-
pact the ground water (even though residual con-
tamination and leachate are above health-based
levels) and contamination does not pose a direct con-
tact threat. With hybrid-clean closure:
• No covers or long-term management are required.
« Fate and transport modeling and model verification
are used to ensure that ground water is usable.
• A property deed notice is used to indicate the
presence of hazardous substances.
• Hybrid-Landfill Closure. Used when residual con-
tamination poses a direct contact threat, but does not
pose a ground-water threat. With hybrid-landfill
closure:
* Covers, which may be permeable, are used to ad-
dress the direct contact threat.
» Limited long-term management includes site and
cover maintenance and minimal ground-water
monitoring.
* Institutional controls (e.g., land-use restrictions or
deed notices) are used as necessary.
The two hybrid closure alternatives are constructs of ap-
plicable laws but are not themselves promulgated at this
time. These alternatives are possible when RCRA re-
quirements are relevant and appropriate, but not when
closure requirements are applicable.
REFERENCES
1. U.S. EPA. 1989. Final covers on hazardous waste
landfills and surface impoundments. Office of Solid
Waste and Emergency Response Technical
Guidance Document EPA 530-SW-89-047, Risk
Reduction Engineering Laboratory, Cincinnati, OH.
2. U.S. EPA. 1989. RCRA ARARs: focus on closure re-
quirements. Office of Solid Waste and Emergency
Response Directive 9234.2-04FS, Office of Solid
Waste and Emergency Response, Washington, DC.
3. U.S. EPA. 1978. Closure and postclosure standards.
Draft RCRA Guidance Manual for Subpart G. EPA
530-SW-78-010. Office of Solid Waste and Emergen-
cy Response, Washington, DC.
-------
CHAPTER 2
SOILS USED IN COVER SYSTEMS
INTRODUCTION
This chapter describes several important aspects of soils
design for cover systems over waste disposal units and
site remediation projects. The chapter focuses on three
critical components of the cover system: composite ac-
tion of soil with a geomembrane liner; design and con-
struction of low hydraulic conductivity layers of
compacted soil; and mechanisms by which low hydraulic
conductivity layers can be damaged. In addition, types of
soils used for liquid drainage or gas collection also will be
discussed.
TYPICAL COVER SYSTEMS
Cover systems perform many functions. One of the prin-
cipal objectives of a cover system is to reduce leaching
of contaminants from buried wastes or contaminated
soils by minimizing water infiltration. Cover systems also
promote good surface drainage and maximize runoff. In
addition, they restrict or control gas migration, or, at
some sites, enhance gas recovery. Finally, cover sys-
tems provide a physical separation between buried
wastes or contaminated materials and animals and plant
roots. When designing a cover system, all of these re-
quirements, plus others, typically must be considered.
As presented and discussed in Chapter 1, Figures 1-1
and 1-2 illustrate two typical cover profiles (see pages 1-
3 and 1-7). Figure 1-1 illustrates the minimum cover
profile recommended by EPA for hazardous waste. Many
of the layers shown in the figure are composed of soils or
have soil components. Each layer has a different pur-
pose and the materials must be selected and the layer
designed to perform the intended function:
• Topsail - The topsoil supports vegetation (which mini-
mizes erosion and maximizes evapotranspiration),
separates the waste from the surface, stores water
that infiltrates the cover system, and protects underly-
ing materials from freezing during winter and from
desiccation during dry periods.
• Filter - The filter separates the underlying drainage
material from the topsoil so that the topsoil will not
plug the drainage material. The filter is often a geotex-
tile, but also can be soil.
• Drainage Layer - The drainage layer (which is not
needed in arid climates) serves to drain away water
that infiltrates the topsoil.
• Geomembrane Liner and Low Hydraulic Conductivity
Soil Layer- The geomembrane and low hydraulic con-
ductivity soil layer form a composite liner that serves
as a hydraulic barrier to impede water infiltration
through the cover system.
Figure 1-2 illustrates an alternative cover profile recom-
mended by EPA for hazardous waste. In Figure 1-2, cob-
bles are placed on the topsoil to provide protection from
erosion. Cobbles, which are normally used only at very
arid sites, allow precipitation to infiltrate underlying
materials, but do not promote evapotranspiration (since
there are no plants present) Figure 1-2 also depicts a
biobarrier between two filters. The biobarrier is usually a
layer of cobbles, approximately 30- to 90-cm (1- to 3-ft)
thick. The biobarrier stops animals from burrowing into
the ground, and, if the cobbles are dry, prevents the
penetration of plant roots. The gas vent layer facilitates
removal of gases that could accumulate in the waste
layer.
The cover profiles shown in Figures 1-1 and 1-2 provide
general guidance only. Depending on the specific cir-
cumstances at a particular site, some of the layers shown
in these figures may not be necessary. For example, at
an extremely arid site, a cover system placed over non-
hazardous, nonputrescible waste may simply consist of a
single layer of topsoil with no drainage layer, no hydraulic
barrier, and no gas vent layer. Conversely, some situa-
tions may require more layers than those shown in these
figures. For example, radioactive waste such as uranium
mill tailings may require a radon-emission-barrier layer. In
addition, the designer may need to include several com-
ponents or layers within the cover system to satisfy multi-
ple objectives. When such objectives lead to conflicting
technical requirements, tradeoffs are frequently neces-
sary.
FLOW RATES THROUGH LINERS
Figure 2-1 illustrates three types of hydraulic barriers
(liners) for cover systems: 1) a low hydraulic conduc-
-------
tivity, compacted soil liner; 2) a geomembrane liner; and
3) a geomembrane/soil composite liner. Flow rates for
each of these types of liners are calculated below for the
purpose of comparing the effectiveness of the barriers.
Flow rates through compacted soil liners are calculated
using Darcy's law, the basic equation used to describe
the flow of fluids through porous materials. Darcy's law
states:
q = ks i A
where q is the flow rate (m3/s); ks represents the
hydraulic conductivity of the soil (m/s); i is the dimension-
less hydraulic gradient; and A is the area (m2) over which
flow occurs. If the soil is saturated and there is no soil
suction, the hydraulic gradient (i) is:
i = (h + D)/D
where the terms are defined in Figure 2-1 (h is the depth
of liquid ponded above a liner with thickness D). For ex-
ample, if 30 cm (1 ft) of water is ponded on a 90-cm (3-ft)
thick liner that has a hydraulic conductivity of 1 x 1CT9 m/s
(1 x 10"7 cm/s), the flow rate is 120 gal (454 L)/acre/day.
If the hydraulic conductivity is increased or decreased,
the flow rate is changed proportionally (Table 2-1).
The second liner depicted in Figure 2-1 is a
geomembrane liner. It is assumed that the geomembrane
has one or more circular holes (defects) in the liner, that
the holes are sufficiently widely spaced that leakage
through each hole occurs independently from the other
holes, that the head of liquid ponded above the liner (h) is
Table 2-1. Calculated Flow Rates through Soil Liners
with 30 cm of Water Ponded on the Liner
Hydraulic Conductivity
(cm/s)
1x10~6
1 x10~7
1 x10'8
1 x 10~9
Rate of Flow
(gal/acre/day)a
1,200
120
12
1
constant, and that the soil that underlies the
geomembrane has a very large hydraulic conductivity
(the subsoil offers no resistance to flow through a hole in
the geomembrane). Giroud and Bonaparte (1) recom-
mend the following equation for estimating flow rates
through holes In geomembranes under these assump-
tions:
q = CB a (2gh)as
where q is the rate of flow (m3/s); CB is a flow coefficient
with a value of approximately 0.6; a is the area (m2) of a
circular hole; g is the acceleration due to gravity (9.81
m/s2); and h is the head (m) above the liner. For ex-
ample, if there is a single hole with an area of 1 cm2
(0.0001 m2) and the head is 30 cm (1 ft) (0.305 m), the
calculated rate of flow is 3,300 gal (12,491 L)/day. If there
is one hole per acre, then the flow rate is 3,300 gal
(12,491 L)/acre/day.
Flow rates for other circumstances are calculated in
Table 2-2. Giroud and Bonaparte report that with good
quality control, one hole per acre is typical (1). With poor
control, 30 holes per acre is typical. They also note that
most defects are small (<0.1 cm2), but that larger holes
are occasionally observed, in calculating the rate of flow
for "No Holes" in Table 2-2, it was assumed that any flux
of liquid was controlled by water vapor transmission; a
Table 2-2, Calculated Flow Rates Through a
Geomembrane with a Head of 30 cm of Water
above the Geomembrane
aL = gal x 3.785
Size of Hole
(cm2)
No holes
0.1
0.1
1
1
10
Number of Holes
Per Acre
-
1
30
1
30
1
Rate of Flow
(gal/acre/day)a
0.01
330
10,000
3,300
1 00,000
33,000
al_ = gal x 3.785
D
Area "a"
Hydraulic Conductivity "k"
o
SOIL LINER GEOMEMBRANE
Figure 2-1. Soil liner, geomembrane liner, and composite liner.
COMPOSITE LINER
10
-------
flux of 0.01 gal/acre/day corresponds to a typical water
vapor transmission rate of geomembrane liner materials.
The third type of liner depicted in Figure 2-1 is a com-
posite liner. Giroud and Bonaparte (2) and Giroud et al.
(3) discuss seepage rates through composite liners. They
recommend the following equation for computing
seepage rates for cases in which the hydraulic seal bet-
ween the geomembrane and soil is poor:
q =1.15h°9a01 ks0.74
where all the parameters and units are as indicated pre-
viously. This equation assumes that the hydraulic
gradient through the soil is 1. If there is a good hydraulic
seal between the geomembrane liner and underlying soil,
the flow rate is approximately one-fifth the value com-
puted from the equation shown above; the constant in the
equation is 0.21 rather than 1.15 for the case of a good
seal. For example, suppose the geomembrane com-
ponent of a composite liner has one hole/acre with an
area of 1 cm2 per hole, the hydraulic conductivity of the
subsoil is 1 x 10"7 cm/s (1 x 10~9 m/s), the head of water
is 30 cm (1 ft) and a poor seal exists between the
geomembrane and soil. The calculated flow rate is 0.8
gal (3 L)/acre/day. Table 2-3 shows other calculated flow
rates for composite liners with a head of water of 30 cm
(1 ft.)
It is useful to compare the three types of liners under a
variety of assumed conditions, as illustrated in Table 2-4.
For discussion purposes, each liner type is classified as
poor, good, or excellent. EPA requires that low per-
meability compacted soil liners used for hazardous
wastes have a hydraulic conductivity no greater than 1 x
10~7 cm/s; therefore, a soil liner with a hydraulic conduc-
tivity of 1 x 10~7 cm/s is described in Table 2-4 as a
"good" liner. A compacted soil liner with a 10-fold higher
hydraulic conductivity is described as a "poor" liner, and
a soil liner with a 10-fold lower hydraulic conductivity is
described as an "excellent" liner.
For geomembrane liners, a liner with a large number of
small holes (30 holes/acre, with each hole having an area
of 0.1 cm2) is described as a "poor" liner because Giroud
and Bonaparte suggest that such a large number of
defects would be expected only with minimal construction
quality control (1). A "good" geomembrane liner was as-
sumed to have been constructed with good quality as-
surance and an "excellent" geomembrane liner was
assumed to have one small hole/acre (1). For all of the
seepage rates computed for composite liners in Table 2-
4, it was assumed that there was poor contact between
the geomembrane and soil.
As Table 2-4 illustrates, a composite liner (even one built
by poor to mediocre standards) significantly outperforms
a soil liner or a geomembrane liner alone. For this
reason, a composite liner is recommended when there is
enough rainfall to warrant a very low-permeability
hydraulic barrier in the cover system.
Table 2-3. Calculated Flow Rates for Composite Liners
with a Head of Water of 30 cm
Hydraulic
Conductivity
of Subsoil
(cm/s)
1 x10'6
1 x10'6
1 xlfJ6
1 xicr6
1 x 1 rj6
1 x 1 0~7
1 x 1 0~7
1 x10"7
1 x 10'7
1 xirj7
1 x 10'8
1 x 1 cr8
1 x 1 cr8
1 x10~8
1 x 1 O'8
1 X 10-9
1 x 1 O'9
1x10'9
1x10'9
1 x10'9
Size of Hole in
Geomembrane
(cm2)
0.1
0.1
1
1
10
0.1
0.1
1
1
10
0.1
0.1
1
1
10
0.1
0.1
1
1
10
Number of
Holes/Acre
1
30
1
30
1
1
30
1
30
1
1
30
1
30
1
1
30
1
30
1
Rate of Flow
(gal/acre/dayf
3
102
4
130
5
0.6
19
0.8
24
1.0
0.1
3
0.1
4
0.2
0.2
0.6
0.03
0.8
0.03
al_ = gal x 3.785
To maximize the effectiveness of a composite liner, the
geomembrane must be placed to achieve a good
hydraulic seal with the underlying layer of low hydraulic
conductivity soil. As shown in Figure 2-2, the composite
liner works by limiting the flow of fluid in the soil to a very
small area. Fluid must not be allowed to spread laterally
along the interface between the geomembrane and soil.
To ensure good hydraulic contact, the soil liner should be
smooth-rolled with a steel-drummed roller before the
geomembrane is placed, and the geomembrane should
have a minimum number of wrinkles when it is finally
covered. In addition, high-permeability material, such as
a sand bedding layer or geotextile, should not be placed
between the geomembrane and low hydraulic conduc-
tivity soil (Figure 2-2) because this will destroy the com-
posite action of the two materials.
If there are concerns that rocks or stones in the soil
material may punch holes in the geomembrane, the
stones should be removed, or a stone-free material with
a low hydraulic conductivity placed on the surface.
Vibratory screens also can be used to sieve stones prior
to placement. Alternatively, mechanical devices that
sieve stones or move them to a row in a loose lift of soil
may be used. A different material, or a differently
11
-------
Clav Liner
Composite Liner
Leachate
Leachate
FML
A = Area of Entire
Liner
Area < Area of Entire
Liner
Leachate
Do.
Leachate
Don't
Figure 2-2. Soil liner and composite liner.
processed material that has fewer and smaller stones,
may be used to construct the uppermost lift of the soil
liner (i.e., the lift that will serve as a foundation for the
geomembrane).
CRITICAL PARAMETERS FOR SOIL LINERS
Materials
The primary requirement for a soil liner material is that it
be capable of being compacted to produce a suitably low
hydraulic conductivity. To meet this requirement, the fol-
lowing conditions should be met:
• Fines - The soil should contain at least 20 percent
fines (fines are defined as the percentage, on a dry-
weight basis, of material passing the No. 200 sieve,
which has openings of 0.075 mm).
Plasticity Index - The soil should have a plasticity
index of at least 10 percent, although some soils with a
slightly lower plasticity index may be suitable. Soils
with plasticity indices less than about 10 percent have
very little clay and usually will not produce the neces-
sary low hydraulic conductivity. Soils with plasticity in-
dices greater than 30 to 40 percent are difficult to work
with, as they form hard chunks when dry and sticky
clods when wet, which make them difficult to work with
in the field. Such soils also tend to have high
shrink/swell potential and may not be suitable for this
12
-------
Table 2-4. Calculated Flow Rates for Soil Liners,
Geomembrane Liners, and Composite Liners
Type of
Liner
Overall
Quality
of Liner
Assumed Rate of
Values of Flow
Key (gal/
Parameters acre/day)8
Compacted
Soil
Geomembrane
Composite
Compacted
Soil
Poor
Poor
Poor
Good
-1-6
ks=1 x 10~°cm/s 1,200
30 holes/acre;
a=0.1 cm2
10,000
ks=1 x10~6em/s 100
30 holes/acre;
a=0.1 cm2
-7.
Geomembrane Good
Composite
Good
ks=1 x 10 cm/s
1 hole/acre;
a=1 cm2
ks=1 x 10 cm/s
1 hole/acre;
a=1 cm2
120
3,300
0.8
Compacted Excellent
Soil
Geomembrane Excellent
k5=1 x10'8cm/s 12
Composite
Excellent
1 hole/acre;
a=0.1 cm2
ks=1 x 10"8cm/s
1 hole/acre;
a=0,1 cm2
330
0.1
aL = gal x 3.785
reason. Soils with plasticity indices between ap-
proximately 10 and 35 percent are generally ideal.
Percentage of Gravel - The percentage of gravel
(defined as material retained on the No. 4 sieve, which
has openings of 4.76 mm) must not be excessive. A
maximum amount of 10 percent gravel is suggested as
a conservative figure. For many soils, however, larger
amounts may not necessarily be deleterious if the
gravel is uniformly distributed in the soil and does not
interfere with compaction by footed rollers. For ex-
ample, Shakoor and Cook found that the hydraulic
conductivity of a compacted, clayey soil was insensi-
tive to the amount of gravel present, as long as the
gravel content did not exceed 50 percent (4). Gravel is
only deleterious if the pores between gravel particles
are not filled with clayey soil and the gravel forms a
continuous pathway through the liner. The key
problem to be avoided is segregation of gravel in pock-
ets that contain little or no fine-grained soil.
• Stones and Rocks - No stones or rocks larger than 2.5
to 5 cm (1 to 2 in.) in diameter should be present in the
iiner material.
If the soil material does not contain enough clay or other
fine-grained minerals to be capable of being compacted
to the desired low hydraulic conductivity, commercially
produced clay minerals, such as sodium bentonite, may
be mixed with the soil. Figure 2-3 shows the relationship
between the percentage of bentonite added to a soil and
the hydraulic conductivity after compaction for a well-
graded, silty soil that was carefully mixed in the
laboratory. The percentage of bentonite is defined as the
dry weight of bentonite divided by the dry weight of soil to
which the bentonite is added (VWWs). For well-graded
soils containing a wide range of grain sizes, adding just a
small amount of bentonite will usually lower the hydraulic
conductivity of the soil to below 1 x 10~7. For poorly
graded soils, e.g., those with a uniform grain size, more
bentonite is often needed.
Bentonite can be added to soil in two ways. One techni-
que is to spread the soil to be amended over an area in a
loose lift approximately 23 to 30 cm (9- to 12-in.) thick.
Bentonite is then applied to the surface at a controlled
rate and mixed into the soil using mechanical mixing
equipment, such as a rototiller or road reclaimer
(recycler). Multiple passes of the mixing equipment are
usually recommended. The second procedure is to mix
the ingredients in a pugmill, which is a large device used
to mix bulk materials such as the ingredients that form
Portland cement concrete. Bulk mixing in a pugmill usual-
ly provides more controlled mixing than combining in-
gredients in place in a loose lift of soil. However, mixing
of bentonite into a loose lift of soil can be adequate if the
mixing is done carefully with multiple passes of mechani-
cal mixers and careful control over rates of application
and depth of mixing. The reason why bulk mixing is
usually recommended is that control over the mixing
process is easier.
Water Content
The water content of the soil at the time it is compacted is
an important variable controlling the engineering proper-
ties of soil liner materials. The lower half of Figure 2-4
shows a soil compaction curve. If soil samples are mixed
at several water contents and then compacted with a
consistent method and energy of compaction, the result
is the relationship between dry unit weight and molding
water content shown in the lower half of Figure 2-4. The
molding water content at which the maximum dry unit
weight is observed is termed the "optimum water content"
and is indicated in Figure 2-4 with a dashed vertical line.
Soils compacted at water contents less than optimum
("dry of optimum") tend to have a relatively high hydraulic
conductivity whereas soils compacted at water contents
greater than optimum ("wet of optimum") tend to have a
low hydraulic conductivity. It is usually preferable to com-
13
-------
C/3
E
o
>»
•*—•
]>
"•+—•
o
"O
c
o
O
o
"5
05
•D
X
Percent Bentonite
Figure 2-3. Effect of benlonite upon the hydraulic conductivity of a bentonits-amended soil.
Wb
(Percent bentonite = TTT- )
ws
pact the soil wet of optimum to achieve minimal hydraulic
conductivity.
Figures 2-5 to 2-7 illustrate for a highly plastic soil why
wet-of-optimum compaction is so effective in achieving
low hydraulic conductivity. These three photographs
show a soil that was compacted with standard Proctor
energy (ASTM D698). The soil had a plasticity index of
41 percent. The optimum water content for this soil and
compaction procedure was 19 percent. The specimen
shown in Figure 2-5 was compacted at a water content of
12 percent (7 percent dry of optimum). This compacted
soil had a very high hydraulic conductivity (1 x 10 crn/s)
because the dry, hard clods of soil were not broken down
and remolded by the energy of compaction. The
specimen shown in Figure 2-6 was compacted at a water
content of 16 percent (3 percent dry of optimum) and had
a hydraulic conductivity of 1 x 10 cm/s; the clods were
still too dry and hard at this water content to permit the
clods to be remolded into a homogeneous mass with low
hydraulic conductivity. The specimen shown in Figure 2-7
was compacted at a water content of 20 percent (1 per-
cent wet of optimum) and had a hydraulic conductivity of
1 x 10"9 cm/s. At this water content, the clods were wet,
soft, and easily remolded into a homogeneous mass that
was free of remnant clods and large inter-clod voids and
pore spaces. The visual differences between specimens
compacted dry versus wet of optimum are usually not as
obvious as they are in Figures 2-5 to 2-7 for soils of lower
plasticity. However, even for low-plasticity clays, ex-
perience has almost always shown that the soil must be
compacted wet of optimum water content to achieve min-
imum hydraulic conductivity.
The water content of the soil must be adjusted to the
proper value prior to compaction and the water should be
uniformly distributed in the soil. If the soil requires addi-
tional water, it can be added with a water truck; care
should be taken to apply the water to the soil in a control-
led, uniform manner, e.g., with a spray bar mounted on
the rear of the trucks. Rototillers (Figure 2-8) are very ef-
fective for mixing wetted soil; these devices distribute the
water uniformly among clods of material. Figure 2-9
depicts the teeth on the blades of a rototiller, which
provide the mixing action. Mechanical mixing to mix
water evenly into the soil is especially important for highly
plastic soils that form large clods of soil.
Compactive Energy
Another important variable controlling the engineering
properties of soil liner materials is the energy of compac-
tion. As shown in Figure 2-10, increasing the energy of
compaction increases the dry unit weight of the soil,
decreases the optimum water content, and reduces
14
-------
Hydraulic
Conductivity
Dry Unit
Weight
Molding Water Content
Figure 2-4. Hydraulic conductivity and dry unit weight versus molding water content.
hydraulic conductivity. The hydraulic conductivity of a soil
that is compacted wet of optimum could be lowered by
one to two orders of magnitude by increasing the energy
of compaction, even though the dry unit weight of the soil
is not increased measurably. More energy of compaction
helps to remold clods of soil, realign soil particles, reduce
the size or degree of connection of the largest pores in
the soil, and lower hydraulic conductivity.
The compactive energy delivered to soil depends on the
weight of the roller, the number of passes of the roller
over a given area, and the thickness of the soil lift being
compacted. Increasing the weight and number of passes,
and decreasing the lift thickness, can increase the com-
pactive effort. The best combination of these factors to
use when compacting low hydraulic conductivity soil
liners depends on the water content of the soil and the
firmness of the subbase.
Heavy rollers cannot be used if the soil is very wet or if
the foundation is weak and compressible (e.g., if
municipal solid waste is located just 30- to 60-cm [1 - to 2-
ft] below the layer to be compacted). Rollers with static
weights of at least 13,608 to 18,144 kg (30,000 to 40,000
pounds) are recommended for compacting low hydraulic
conductivity layers in cover systems. Rollers that weigh
up to 31,752 kg (70,000 pounds) are available and may
be desirable for compacting bottom liners of landfills, but
such rollers are too heavy for many cover systems be-
cause of the presence of compressible waste material a
short distance below the cover.
The roller must make a sufficient number of passes over
a given area to ensure adequate compaction. The mini-
mum number of passes will vary, but at least 5 to 10 pas-
ses are usually required to deliver sufficient compactive
energy and to provide adequate coverage.
15
-------
,J f
r* r
Figure 2-5, Highly plastic soil compacted with standard
Proctor procedures at a water content of 12%.
16 %
STANDARD
PROCTOR
Figure 2-6. Highly plastic soil compacted with standard
Proctor procedures at a water content of 16%.
STANDARD
PROCTOR
Figure 2-7. Highly plastic soil compacted with standard
Proctor procedures at a water content of 20%.
Size of Clods
The clay-rich soils that are usually used to construct soil
liners typically form dry, hard clods of soil or wet, sticky
clods, depending on water content. Highly plastic soils al-
most always form large clods. Soils with low plasticity
(plasticity index less than about 10%) do not form very
large clods. For soils that form clods, the clods must be
remolded into a homogeneous mass that is free of large
inter-clod pores if low hydraulic conductivity is to be
achieved.
Benson and Daniel described the influence of clod size of
a highly plastic soil (plasticity index = 41%) upon
hydraulic conductivity (5). These investigators processed
a clayey soil by breaking clods down to pass either the
No. 4 sieve (4,76 mm or 0.2 in. openings) or the 1.9-cm
(3/4-in.) sieve. The soil was then wetted, allowed to
hydrate at least 24 hours, compacted, and permeated.
Benson and Daniel's (1990) results are summarized in
Table 2-5. The optimum water content was 17 percent for
the clods processed through the sieve with a 0.5-cm (0,2-
in.) opening and 19 percent for the soil processed
through the sieve with a 1.9-cm (3/4-in.) opening. For soil
compacted dry of optimum, the soil with smaller clods
had a hydraulic conductivity that was several orders of
magnitude lower than the soil with larger clods. When the
soils were compacted wet of optimum, the size of clods
had a negligible effect. Size is therefore important for dry,
16
-------
Figure 2-8. Rototiller used to mix soil.
Figure 2-9. Blades and teeth on rototllter.
17
-------
Table 2-5. Effect of Size of Clods during Processing of
Soil upon Hydraulic Conductivity of Soil after
Compaction
Molding Water
Content (%)
12
16
18
20
Hydraulic Conductivity (cm/s)
0,2-in. Clods3 0.75-in. Clods3
2x10"8
2 x 1 O"9
1 x10"8
2xier9
4x 10"4
1 x 1 O"3
8x10"10
7x10-10
acm « in. x 2,540
hard clods (dry of optimum), but not for wet, soft clods
(wet of optimum). When the soil is compacted wet of op-
timum, the clods are sufficiently soft that they are easily
remolded regardless of theiroriginal size.
One way to reduce the size of clods in dry materials is to
use a road reclaimer (also called a road recycler), such
as the one shown in Figure 2-11. This device pulverizes
materials with teeth that rotate on a drum at a high
speed. The device was used with great effectiveness at a
site in Pennsylvania in which a mudslone was used for a
liner material (Figure 2-12). In the figure, the road
reclaimer has made a pass through a loose lift of
material. After just one pass of the road reclaimer, the
size of mudstone clods has been greatly reduced.
Bonding of Lifts
Bonding of lifts is important in achieving a low hydraulic
conductivity in soil liners. The upper half of Figure 2-13 il-
lustrates a cross-section of a soil liner consisting of four
lifts. A borehole has been drilled into the lowest lift, filled
with a dye-stained fluid, left for a period of time, and then
drained. The dye penetrates the soil further along lift in-
terfaces than through the lifts themselves. Due to imper-
fect bonding of lifts, a zone of higher horizontal hydraulic
conductivity exists at lift interfaces in this example.
Lift interfaces have important ramifications with respect to
the overall hydraulic performance of a soil liner. The
lower half of Figure 2-13 depicts a liner consisting of six
lifts. Each lift has a few "hydraulic defects." If the lift in-
terfaces have high hydraulic conductivity, water can flow
downward through the more permeable zones in a lift
and spread laterally along a lift interface until it en-
counters a permeable zone in the underlying lift. This
process repeats for underlying lifts and lift interfaces. In
this way lift interfaces provide hydraulic connection bet-
ween defects in overlying and underlying lifts. Better
overall performance (lower hydraulic conductivity) is
achieved if lifts are bonded together to eliminate high
conductivity at lift interfaces.
To bond lifts together, the surface of the previously
compacted lift should be rough so that the newly
placed lift can effectively blend into the surface. If
necessary, the surface of the previously compacted lift
can be roughened by discing the soil to a depth of ap-
proximately 2.5 cm (1 in). Discing the soil involves
plowing up the soil surface to a shallow depth so that
the surface is rough and so that there will be no abrupt
interface between lifts.
Compactors with long "feet" on the drums are useful in
blending one lift into another. Figure 2-14 shows a
popular heavy compactor (20,000 kg [44,000 pounds])
with feet that are 18 to 23 cm (7 to 9 in.) long. During the
first few passes of the compactor, the feet sink through a
loose lift of soil and compact the newly placed lift into the
surface of the previously compacted lift. Using a roller
with feet that fully penetrate a loose lift of soil is recom-
mended to bond lifts and to minimize high horizontal
hydraulic conductivity at lift interfaces.
If a geomembrane liner will be placed on the compacted
soil liner, the final surface of the soil liner should be com-
pacted with a smooth, steel drum roller to achieve a good
hydraulic seal.
EFFECTS OF DESICCATION
Desiccation of soil liners occurs whenever the soil liner
dries, which can be during or after construction. Desicca-
tion causes soil liner materials to shrink and, potentially,
to crack. Cracking can be disastrous in terms of hydraulic
conductivity because cracked liners are more permeable
than uncracked liners.
Boynton and Daniel desiccated slabs of compacted clay,
trimmed cylindrical test specimens for hydraulic conduc-
tivity testing from the desiccated slabs, and measured the
hydraulic conductivity at different effective confining
stresses (6). In laboratory tests, the confining stress
simulates the weight of overburden soil; the greater the
confining stress, the greater the depth of burial below the
surface that is simulated. Control tests also were per-
formed on soils that had not been desiccated. These
results are summarized in Figure 2-15. At low confining
stress, the desiccated soils were much more permeable
than the control. At high confining stress, however, the
desiccated soils were no more permeable than the con-
trol. It appeared that the application of a large compres-
sive stress (>5 psi, or 35 kPa) closed the desiccation
cracks that had formed and, in combination with hydra-
tion of the soil, essentially fully healed the damage done
by desiccation.
In cover systems, the overburden stress on the liner com-
ponents is controlled by the depth of soil overlying the
liner. Because the thickness of soil overburden above the
liner seldom exceeds a few feet, the overburden stress is
normally low. Soil applies an overburden stress of ap-
proximately 1 psi per foot of depth. Thus, for example, if
18
-------
Q
"D
10
10
-6
O E
-7
cd
g>
"CD
o
Z)
>
Q
10
,o8
116
108
100^
92
Increasing
Compactive
Effort
Optimum
Water
Content
Increasing Compactive
Effort
j i
I I I I
j i
12 14 16 18 20 22 24
Molding Water Content (%)
Figure 2-10, Influence of compactive effort upon hydraulic conductivity and dry unit weight.
60 cm (2 ft) of topsoll overlie a 60-cm (2-ft) thick layer of
compacted clay, the maximum overburden stress at the
bottom of the clay is approximately 4 psi. Based on Boyn-
ton and Daniel's results, if desiccation of the compacted
soil liner occurs in a cover system, even though wetting
of the soil may partly swell the soil and "heal" desiccation
cracks, it is not expected that all the damage done by
desiccation would be self-healing.
Montgomery and Parsons described an example of the
damaging effects of desiccation (7). Test plots were built
at the Omega Hills Landfill near Milwaukee, Wisconsin, in
1985. In both test plots, the cover systems consisted of
122 cm (4 ft) of compacted clay. The clay was overlain by
15 cm (6 in.) of topsoil in one plot and 46 cm (18 in.) of
topsoil in the other. In both test plots, the upper 20 to 25
cm (8 to 10 in.) of compacted clay had weathered and
become blocky after 3 years. Cracks up to 1.3-cm (1/2-
in.) wide extended 89 to 102 cm (35 to 40 in.) into the
compacted clay liner. The 46 cm (18 in.) of topsoil did not
appear to be any more effective than 15 cm (6 in.) in
protecting the underlying clay from desiccation.
The layer of low hydraulic conductivity, compacted soil
placed in a cover system must be protected from the
damaging effects of desiccation both during and after
19
-------
Figure 2-11, Road recycler used to pulverize clods of soil.
construction. During construction, the soil must not be al-
lowed to dry significantly either during or after compac-
tion of each lift, Frequent watering of the soil is usually
the best way to prevent desiccation during construction.
The higher the water content of the soil and the higher
the plasticity of the soil, the greater is the shrinkage
potential from desiccation. There are two ways to provide
the required protection after construction. One way is to
bury the liner beneath an adequate depth of soil overbur-
den; another technique is to place a geomembrane over
the soil. If a geomembrane liner is placed on a soil liner
to form a composite, it is often convenient to overbuild
the soil liner (i.e., make it thicker than necessary) and
then to scrape away a few inches of potentially desic-
cated surficial soil just before the geomembrane is
placed.
EFFECTS OF FROST ACTION
Zimmie and La Plante studied the effects of freezing and
thawing upon the hydraulic conductivity of a compacted
clay by testing soils compacted dry of optimum, at op-
timum, and wet of optimum (8). They found that
freeze/thaw cycles caused an increase in hydraulic con-
ductivity of one to two orders of magnitude in all soils ex-
amined. Most of the damage was done after only one to
two cycles of freezing and thawing. From this and other
work, it is recommended that the low hydraulic conduc-
tivity component of cover systems not be allowed to
freeze. Freezing can be avoided by burying the low
hydraulic conductivity soil layer under an adequately thick
layer of soil.
EFFECTS OF SETTLEMENT
Two types of settlement are of concern with respect to
covers: total settlement and differential settlement. Total
settlement of the surface of a cover is the total downward
movement of a fixed point on the surface. Differential set-
tlement is always measured between two points and is
defined as the difference between the total settlements at
these two points. Distortion is defined as the differential
settlement between two points divided by the distance
along the ground surface between the two points. Exces-
sive differential settlement of underlying waste can
damage a cover system. If differential settlement occurs,
tensile strains develop in cover materials as a result of
bending stresses and/or elongation. Tensile strain is
defined as the amount of stretching of an element divided
by the original length of the element. Anytime the cover
settles differentially, some part of the cover will be sub-
jected to tension and will undergo tensile strain. Tensile
strains are of concern because the larger the stretching
(tensile strain), the greater the possibility that the soil will
crack and that a geomembrane will rupture. Bending
stresses, stresses that occur when an object is bent,
result when covers settle differentially; part of the bent
cover is in tension and part is in compression. Bending
stresses are of concern because the tensile stresses as-
sociated with bending may be large enough to cause the
20
-------
-. .-£•*• *.:«> V'
Figure 2-12. Passage of road recycler over loose lift of mudstone to reduce size of chunks of mudstone.
Borehole
Lift 1
Lift 2
Lift 3
Lift 4
J > C f ^
f' ( t S "\
1 } J ^^.^
A ^ \
r
^ r ^ )
^ c \
Figure 2-13. Effect of Imperfect bonding of lifts on hydraulic performance of soil liner.
21
-------
Figure 2-14. Example of heavy footed roller with long feet.
soil to crack. Geomembranes can generally withstand far
larger tensile strains without failing than soils. The
geomembrane also has the ability to elongate (stretch) a
great deal without rupturing or breaking.
Gilbert and Murphy discuss the prediction and mitigation
of subsidence damage to covers (9). Gilbert and Murphy
developed a relationship between tensile strain in a cover
and distortion, delta/L, where delta is the amount of dif-
ferential settlement that occurs between two points that
are a distance (L) apart. This relationship is shown in
Figure 2-16. As the distortion increases, the tensile strain
in the cover soils increases.
Minor cracking of topsoil or drainage layers as a result of
tensile stresses is of little concern. However, cracking of
a hydraulic barrier, such as a layer of low hydraulic con-
ductivity soil, is of great concern because the hydraulic
integrity of the barrier layer is compromised if it is crack-
ed. The amount of strain that a low hydraulic conductivity,
compacted soil can withstand prior to cracking depends
significantly upon the water content of the soil. As shown
in Figure 2-17, soils compacted wet of optimum are more
ductile than soils compacted dry of optimum. For cover
systems, ductile soils that can withstand significant strain
without cracking are preferred. For this reason, as well as
the hydraulic conductivity considerations discussed ear-
lier, it is preferable to compact low hydraulic conductivity
soil layers wet of optimum. The soil must then be kept
from drying out and cracking, as discussed earlier.
Gilbert and Murphy summarize information concerning
tensile strain at failure for compacted, clayey soils (9).
The available data show that such soils can withstand
maximum tensile strains of 0.1 to 1 percent. If the lower
limit (0.1 percent) is used for design, the maximum allow-
able value of distortion (delta/L) is approximately 0.05
(Figure 2-17).
To put this in perspective, suppose that a circular depres-
sion develops in a cover system. The depression has a
radius of 3 m (10 ft) (diameter=6 m [20 ft]). The maximum
allowable delta/L is 0.05, and L is the radius of the
depression, which is 3 m (10 ft). The maximum allowable
settlement (delta) is 0.05 times 3 m (10 ft), or 15 cm (6
in.). If the settlement at the center of the depression ex-
ceeds 6 in., the clay layer may crack from the tensile
strains caused by the settlement.
Some wastes (such as loose municipal solid waste or un-
consolidated sludge of varying thickness) are so com-
pressible that constructing a cover system above the
waste will almost certainly produce distortions that are far
larger than 0.05. The hydraulic integrity of a low hydraulic
conductivity layer of compacted soil is likely to be
seriously damaged by the distortion caused by large dif-
ferential settlement. If the waste is continuing to settle,
e.g., as a result of decomposition, it may be prudent to
place a temporary cover on the waste and wait for settle-
ment to take place prior to constructing the final cover
system. Alternatives for stabilizing the waste include
22
-------
0
(kPa)
50
100
~ 10
E
o
o
3
•o
c
o
o
o
10
-8
10
-9
I -
So mple Containing
Desiccation
Cracks
"Sample Containing
No Desiccation
Cracks
8
12
16
Ef f ective Confin ing Pressure (psi)
Figure 2-15. Effect of desiccation upon the hydraulic conductivity of compacted clay (6).
deep dynamic compaction, soil preloading, and the use
of wick drains to consolidate sludges. These technologies
for waste stabilization are presently emerging and ap-
propriate descriptions are not available in the literature.
INTERRACIAL SHEAR
The stability of a cover system is controlled by the slope
angle and the friction angles between the various inter-
faces of the cover system components. One potential
problem with covers installed with a sloping surface is the
risk that all or part of the cover system may slide
downhill. The recent failure of a partly completed hazard-
ous waste landfill provides an example of the problem
(10). At this facility, slippage occurred between two com-
ponents of the liner system in the landfill cell. The cell
was filled such that a slope was created on the liner sys-
tem that caused slippage.
The interfacial shearing characteristics of all components
of a cover system, as well as internal shearing
parameters of all soil layers, must be known in order to
evaluate stability. If the soils are fully saturated and
below the free water surface, e.g., during a heavy
rainstorm, the stability is much less than if the soils are
dry. Thus, one must consider both typical and worst-case
23
-------
1.0
0.8
0.6
0.4
0.2
0.0
.1
10
100
Tensile Strain (%)
Figure 2-16. Relationship between distortion and tensile strain (9).
conditions when analyzing the stability of the cover sys-
tem.
Methods of measuring interfacial friction between
geosynthetic/geosynthetic or geosynthetic/soii interfaces
are reviewed in detail by Takasumi et al. (11). No stand-
ard testing method exists, although one is under develop-
ment by ASTM.
Seed and Boulanger (12) measured interfacial friction
angles between a smooth high density polyethylene
(HOPE) geomembrane and a compacted soil-bentonite
mixture that contained 5 percent bentonite by dry weight.
Interfacial friction angles were found to be very sensitive
to compaction water content, dry unit weight, and the
degree of wetting of the soil. For a given dry unit weight,
increasing the molding water content or wetting the com-
pacted soil reduced the interfacial friction angle. Increas-
ing the density typically reduced the interfaciai friction
angle, as well. Unfortunately, the compaction conditions
that would yield minimal hydraulic conductivity (i.e., com-
paction wet of optimum with a high energy of compac-
tion) also yielded the lowest interfacial friction angles.
Seed and Boulanger reported interfacial friction angles
that were typically 5 to 10 degrees for the water
content—unit weight combinations that would typically be
employed to achieve minimal hydraulic conductivity.
The study of interfacial friction problems is an area of ac-
tive research. At the present time, designers are cau-
tioned to give careful consideration to the problem and to
measure friction angles along all potential sliding sur-
faces using the proposed construction materials for test-
ing. If adequate stability is not provided, the designer will
need to consider alternative materials (e.g., rougher
geomembranes with higher interfacial friction angles),
flatter slopes, or reinforcement of the cover, e.g., with
geogrids.
DRAINAGE LAYERS
Drainage layers are high-permeability materials used to
drain fluids (such as infiltrating water) or gas produced
from the waste. A drainage layer installed to drain in-
filtrating water is called a surface water collection and
removal system. The hydraulic conductivity required for
this layer depends upon the rate of infiltration, the slope
of the layer, and the hydraulic conductivity of the underly-
ing barrier layer. However, the efficiency of the drainage
layer improves as the hydraulic conductivity of the
drainage material increases. Thus, high hydraulic con-
ductivity is a requirement for drainage layers.
The single most important factor controlling the hydraulic
conductivity of sands and gravels is the amount of fine-
grained material present. Geotechnical engineers define
fine-grained materials as those materials that will pass
through the openings of a No. 200 sieve (0.075 mm
openings). A relatively small shift in the amount of fines
24
-------
CO
c
0
Q
t
A
C/D
CO
tf)
Water Content
Strain
Figure 2-17. Relationship between shearing characteristics of compacted soils and conditions of compaction.
present in the soil can change the hydraulic conductivity
by several orders of magnitude. The drainage material
should be relatively free of fines if the material is to have
a high hydraulic conductivity.
A minimal amount of compaction of the drainage
materials in a cover is adequate to guard against settle-
ment; excessive compaction is usually not necessary. In
fact, excessive compaction may grind up soil particles,
which would tend to lower the hydraulic conductivity of
the drainage layer. However, sands may bulk if placed in
a damp or wet condition, which can lead to an unaccep-
tably loose material. If significant seismic ground shaking
is possible at a site, compaction of drainage layers may
be needed to minimize the risk of liquefaction-induced
sliding.
Designers often place a highly permeable layer at the
base of a cover system above gas-producing wastes,
such as municipal solid waste. This layer aids in collect-
ing gas and is called a gas collection layer. Adequate fil-
ters above and below the gas collection layer must be
provided so that the collection layer does not become
clogged with fine material. Vent pipes are normally
placed in the gas collection layer at a frequency of ap-
proximately one per acre.
SUMMARY
Soils are used in cover systems to support growth of
vegetation, to separate buried wastes or contaminated
soils from the surface, to minimize the infiltration of water,
and to aid in collecting and removing gases. The most
challenging aspect of utilizing soils in cover systems is
designing, constructing, and maintaining a barrier layer of
low hydraulic conductivity. Soils can be compacted to
achieve a low initial hydraulic conductivity, but the soils
can be damaged by excessive differential settlement,
desiccation, and other environmental stresses. Protecting
a compacted soil liner from damage is therefore the
greatest challenge to the designer.
REFERENCES
1. Giroud, J.P. and R. Bonaparte. 1989a. Leakage
through liners constructed with geomembranes—Part I.
Geomembrane liners. Geotextiles and
Geomembranes. Vol. 8: 27-67.
25
-------
2. Giroud, J.P. and R. Bonaparte. 1989b. Leakage
through liners constructed with geomembranes—Part II.
Composite liners. Geotextiles and Geomembranes.
Vol. 8:71-111.
3. Giroud, J.P., A. Khatami, and K. Badu-Tweneboah.
1989. Evaluation of the rate of leakage through com-
posite liners. Geotextiles and Geomembranes. Vol. 8:
337-340.
4. Shakoor, A. and B.D. Cook. 1990. The effect of stone
content, size, and shape on the engineering proper-
ties of a compacted silty clay. Bulletin of the Associa-
tion of Engineering Geologists. Vol. 27, No. 2:
245-253.
5. Benson, C.H. and D.E. Daniel. 1990. Influence of
clods on hydraulic conductivity of compacted clay.
Journal of Geotechnical Engineering. Vol. 116, No. 8;
1231-1249.
6. Boynton, S.S. and D.E. Daniel. 1985. Hydraulic con-
ductivity tests on compacted clay. Journal of
Geotechnical Engineering. Vol. 111, No. 4: 465-478.
7, Montgomery, R.J. and L.J. Parsons. 1989. The
Omega Hills Final Cover Test Plot Study: Three-
Year Data Summary. Presented at the 1989 Annual
Meeting of the National Solid Waste Management
Association, Washington, DC.
8. Zimmte, T.F. and C. La Plante. 1990. The effect of
freeze-thaw cycles on the permeability of a fine-
grained soil. Proceedings, 22nd Mid-Atlantic
Industrial Waste Conference. Philadelphia, Pennsyl-
vania: Drexel University.
9. Gilbert, P.A. and W.L. Murphy. 1987. Predic-
tion/mitigation of subsidence damage to hazardous
waste landfill covers. EPA/600/2-87/025 (PB87-
175386). Cincinnati, Ohio: U.S. EPA.
10. Seed, R.B., J.K. Mitchell, and H.B. Seed. 1990. Ket-
tlemam Hills waste landfill slope failure. II: Stability
analyses. Journal of Geotechnical Engineering. Vol.
116, No. 4: 669-691.
11. Takasumi, D.L., Green, K.R., and R.D. Holtz. 1991.
Soil-Geosynthetics Interface Strength Charac-
teristics: A Review of State-of-the-Art Testing Proce-
dures. Proceedings, Geosynthetics 91, Vol. 1,
87-100.
12. Seed, R.B., and R.W. Boulanger. 1991. Smooth
HOPE—Clay Interface Shear Strengths: Compaction
Effects. Journal of Geotechnical Engineering. Vol.
117, No. 4, 686-693.
26
-------
CHAPTERS
GEOSYNTHETIC DESIGN FOR LANDFILL COVERS
GENERAL COMMENTS ON
DESIGN-BY-FUNCTION
Geosynthetics (GS), like all engineering materials, are
capable of being evaluated using a design-by-function
approach which includes a traditional factor-of-safety
(FS) concept. The primary function of the geosynthetic
depends upon its location in the facility. The usual re-
quirements for waste containment systems are given in
Table 3-1.
Table 3-1. Customary Primary Functions of Geosyn-
thetics Used in Waste Containment Systems
Primary Function
Type of
Geosynthetic
Separate Reinforce Filter Drain Barrier
Geomembrane
Geotextile
Geonet
(Geo) pipe
Geocomposite
Geogrid
Upon selection of the primary function, the FS should be
calculated in the same manner as for any other engineer-
ing material:
q Allowable (Test) Value
~ Required (Design) Value
(D
In the above equation, the test values usually come from
ASTM Committee D35 on Geosynthetics, or some other
standardization group. The design values come from a
site-specific situation that utilizes relevant aspects of
geotechnical, hydraulic, polymer, or environmental en-
gineering principles or from the appropriate governing
regulations. The actual magnitude of the FS is a reflec-
tion of the degree of certainty of the design as well as the
implications of the system's nonperformance.
GEOMEMBRANE DESIGN CONCEPTS
For the design of a geomembrane in a landfill cover there
are at least four general considerations; liner com-
patibility, vapor transmission (water and methane),
biaxial stresses via subsidence, and planar stresses mo-
bilized by friction. Each will be described separately
below.
Geomembrane Compatibility
Since the liquid interfacing the geomembrane liner is
generally water, there is usually no need for EPA 9090
chemical compatibility testing, except in unusual cir-
cumstances. Some other tests that might be considered,
however, are the following:
• Dimensional stability test via ASTM D-1204
• Resistance to soil burial via ASTM D-3083
• Water extraction test via ASTM D-3083
• Volatile loss test via ASTM D-1203
• Biological resistance test via ASTM G-22
* Fungus resistance test via ASTM G-21
The use of these tests is required on a site-specific and
geomembrane-specific basis.
Vapor Transmission
Testing of water vapor transmission through geomembranes
is performed via ASTM E-96. The EPA technical resources
document of September 1988 (1) gives the following values
for the indicated geosynthetic materials:
PVC (polyvinyl chloride)
- 30 mil -1.9 g/m2-day
CPE (chlorinated polyethylene)
- 40 mil - 0.4 g/m2-day
CSPE (chlorylsulfonated polyethylene)
- 40 mil - 0.4 g/m2-day
HOPE (high density polyethylene)
- 30 mil - 0.02 g/m2-day
HOPE - 98 mil - 0.006 g/m2-day
The conversion from g/m2-day to gal/acre-day is 1 to
1.07.
A related measurement is methane (ChU) gas transmis-
sion through geomembranes. This lignter-than-air gas will
rise up from the waste and interface with the
geomembrane. Methane gas transmission rates for dif-
27
-------
ferent geomembranes have been reported in EPA's tech-
nical resources document as follows:
Yes
HCS
PVC
-10 mil - 4.4 ml/m -day-atm.
- 20 mil - 3.3 ml/m -day-atm.
LLDPE (linear low density polyethylene)
-18 mil - 2.3 ml/m2-day-atm.
CSPE - 32 mil - 0.27 ml/m2-day-atm.
- 34 mil -1.6 ml/m2-day-atm.
HOPE - 24 mil -1.3 ml/m2-day-atm.
- 34 mil -1.4 ml/m2-day-atm.
Biaxial Str0sses via Subsidence
As the waste beneath the closure subsides, differential
settlement is likely to occur. Thus a factor-of-safety for-
mulation of FS = Oaiiow/oreqd is necessary. This situation
has been modeled (see Appendix A, Stability and Ten-
sion Considerations Regarding Cover Soils on
Geomembrane Lined Slopes), giving rise to the following
formula for required strength (o"reqd)):
Greqd •
where ycs
Hcs
2 D L2 ycs Hcs
3 t (D2 + L2)
unit weight of cover soil
= height of cover soil
t = thickness of geomembrane
D, L = see Figure 3-1
The allowable strength aaiiow of the candidate
geomembrane must be evaluated in a closely simulated
test, e.g., GRI's GM-4 entitled "Three Dimensional
Geomembrane Tension Test." Figure 3-2 presents the
response to this test of a number of common
geomembranes used in closure situations.
Planar Stresses via Friction
In addition to the above out-of-plane stresses, the cover
soil over the geomembrane might develop greater fric-
tional stresses than the soil material beneath it. This hap-
pens particularly if a wet-of-optimum clay is placed
beneath. Again a factor-of-safety formulation is formed by
comparing the allowable strength (Taiiow) to the required
strength (Treqd) but now in force units rather than stress
units, e.g., FS = Taiiow/Treqd. The required geomembrane
tension can be obtained by the equation given in Figure
3-3 (see Appendix A for a more detailed discussion).
where cau, Cat
SU.SL
(0
L
W
adhesion of the material upper
and lower of the geomembrane
friction angle of the material
upper and lower of the
geomembrane
slope angle
slope length
unit width of slope
unit weight of cover soil
height of cover soil
When calculated, the value of Treqd in Figure 3-3 is com-
pared to the Taiiow of the candidate geomembrane. This
value is currently taken from ASTM D-4885, the wide-
width tensile test for geomembranes. Note that this value
must be suitably adjusted for creep, long-term degrada-
tion, and any other site-specific situations that are con-
sidered relevant.
GEONET AND GEOCOMPOSITE SHEET DRAIN
DESIGN CONCEPTS
Geonets and/or geocomposite drains are often used as
surface water drains located immediately above the
geomembrane in a landfill closure system. There are
three aspects to the design that require attention:
material compatibility, crush strength, and flow capability.
Compatibility
Since the liquid being conveyed by the geonet or
drainage geocomposite is water, EPA 9090 testing is
usually not warranted. The polymers from which these
products are made are polyethylene (PE), polypropylene
(PP), high-impact styrene (HIS) or other long-chain
molecular structures that have good water resistance and
long-term durability when covered by soil.
Crush Strength
The crush strength of the candidate product must be
evaluated by comparing an allowable strength to a re-
quired stress, i.e., FS = aaiiow/oreqd. The allowable
strength is taken as the rib lay-down for geonets and the
telescoping crush strength for drainage geocomposites.
Figure 3-4 illustrates common behavior for geonets and
geocomposites. The test methods currently recom-
mended are GRI GN-1 for compression behavior of
geonets and GRI GC-4 for drainage geocomposites, i.e.,
for sheet drains.
The required stress is the dead load of the cover soil plus
any live loads that may be imposed, such as construction
and maintenance equipment.
req'd
Figure 3-1. Required strength.
28
-------
5000
20
40 60
STRAIN (%)
80
100
Figure 3-2. Response of common geomembranes to the three-dimensional geomembrane tension test.
req'd
reqd-[VCau \) +
Figure 3-3. Required geomembrane tension.
Flow Capability
The planar flow capability of the geonet or drainage
geocomposite must be assured via a formulation such as
FS = qaiiow/qreqd- The allowable flow rate is obtained
using ASTM D-4716, the transmissivity test. The device
is shown schematically in Figure 3-5, along with the type
of data generated. Note that the value of the flow rate ob-
tained must be modified according to site-specific condi-
tan 5U - tan 5L
L W
tions if these conditions are not properly simulated in the
test.
For the required flow rate, qreqd, site-specific conditions
must also be considered. To obtain this value, a water
balance method that takes into account precipitation,
runoff, evapotranspiration, and infiltration is required. The
HELP model, presented in detail in Chapter 8, is recom-
mended for obtaining this value.
29
-------
Stress
Geocomposite
Strain
Figure 3-4. Common crush strength behavior for geonets and geocomposites.
GEOPIPE AND GEOCOMPOSITE EDGE DRAJN
DESIGN CONCEPTS
AH of the infiltrating surface water that is collected from
the geonet or geocomposite sheet drain Is conveyed to
the perimeter of the closure, where it is collected in a per-
forated pipe or in a geocomposite edge drain. The design
of the geopipe or geocomposite edge drain must con-
sider compatibility, crush strength, and flow rate.
Compatibility
With surface water as the flow medium, EPA 9090
chemical resistance testing is usually not required. Plas-
tic pipe is usually unplasticized PVC or HOPE, and al-
most all edge drains are HOPE. Thus, resistance to water
is very good.
Crush Strength
The formulation for crush strength is straightforward, FS
= aaiiow/o"reqd- The allowable strength for plastic pipe is
ASTM D-2412 and for geocomposite edge drain cores is
GRI's GO4, The response to both types of materials is
very crisp (see Figure 3-6).
The required strength is the dead weight of the cover soil
plus any live loads that might be superimposed onto the
system. Truck traffic around the edge of the closure
should be considered in this regard.
Flow Rate
The required flow rate of the pipe or edge drain is the
cumulative flow coming from the geonet or sheet drain
above the geomembrane. Furthermore, this flow is again
cumulative within the pipe or edge drain and is at its max-
imum at the drainage outlet. Required flow rate will
probably dictate the frequency and location of outlets.
To determine the allowable flow rate of pipes, the Man-
ning formula is usually used with a roughness coefficient
for smooth plastic pipe of 0.015. This is straightforward
hydraulics engineering for pipes flowing partially full. For
geocomposite edge drains, the ASTM D-4716 test
described earlier is recommended,
GEOTEXTILE FILTER DESIGN
CONSIDERATIONS
Geotextiles have the greatest flexibility to serve a number
of different functions. In a landfill closure, one of the most
important is as a filter allowing water to enter a drainage
material composed of stone, geonet, geocomposite, or
perforated pipe. In selecting a geotextile for this purpose,
compatibility, hydraulic conductivity, soil retention, and
long-term clogging all must be addressed.
Compatibility
The vast majority of geogrids are PP or PET. Both of
these polymers are very stable in contact with water and
have demonstrated good durability properties. Generally,
there is no need to perform an EPA 9090 chemical com-
patibility test. The geotextile literature is abundant with
related tests assessing long-term behavior and perfor-
mance (2).
Permeability
A geotextile filter must have sufficient openings to allow
the water to enter the drain without developing excess
pore water pressure in the upstream soil. Such pressures
could mobilize cover soil instability. The flow rate design
is based on permittivity, y = kn/t, where kn = hydraulic
conductivity normal to the fabric, and t = the fabric thick-
ness. As usual, a FS = yaiiow/tyreqc) is formulated. The al-
lowable value is obtained from ASTM D-4491 but the
30
-------
ASTM D-4716 Flow Rate Test
Constant Load
Overflow
Outlet
5000 iQOOO
Normal ftm f* ft1)
tMOO
20000
Figure 3-5. ASTM D-4716 flow rate test.
31
-------
Stress
Strain
Figure 3-6. Crush strength of geopipe and geocomposite edge drain cores.
result must be modified to site-specific conditions using
partial factors of safety. The required value comes from a
water balance method, e.g., the HELP computer code
(see Chapter 8).
Geotextile Soil Retention
The geotextile filter must retain the upstream cover soil,
thereby requiring the voids to be suitably small. Since the
desired retention is the opposite of permittivity, the design
must balance the two conflicting design considerations.
Fortunately, there are about 4,000 commercially available
geotextiles to choose from, thus a design can generally
be satisfied by a number of products. The factor-of-safety
formulation is as follows:
FS= Xd85/095
where X 2 to 4
des = the 95 percent finer soil size
Ogs = the geotextile's 95 percent
opening size
The former value (das) is obtained by dry sieving the
upstream soil, the latter value is obtained by sieving glass
beads through the fabric as per ASTM D-4751.
Geotextile Clogging Evaluation
There are four laboratory tests to evaluate long-term
geotextile clogging by upstream soil particles:
• Gradient ratio test (GR).
• Long-term flow test (LTF).
• Hydraulic conductivity ratio test (HCR).
• Fine fraction filtration test (F3).
The first two tests are established in the literature and are
recommended for geotextiles used in landfill covers. The
latter two tests are experimental and aimed at situations
of relatively severe clogging. In general, use of geotextile
filters in landfill closures as discussed in this chapter is
not a particularly demanding design situation.
GEOGRID, OR GEOTEXTILE, COVER SOIL
REINFORCEMENT
Due to the relatively low interface friction angle of many
geomembranes, there have been some instances of
cover soils sliding off the liner in a gradual (or sometime
abrupt) manner. Many of these situations can be avoided
using geogrid, or geotextile reinforcement. The procedure
is sometimes called "veneer reinforcement." The
design is based on a factor-of-safety formulation of FS=
Taiiow/Treqd. The allowable strength is obtained from a
wide-width tensile strength test such as ASTM D-4595.
This value, however, must be reduced for such site-
specific conditions as installation damage, creep, and
long-term degradation.
Based on an infinite slope analysis, the required strength
of the geogrid (or geotextile) is given by the equation in
Figure 3-7. Note that the analysis includes a seepage
force "S" and an earthquake force "E," Both are, of
course, site specific and, if large, can easily dominate the
design.
GEOTEXTILE METHANE GAS VENT
Beneath the liner system in a landfill closure, gases
lighter than air will accumulate and gradually exert pres-
sure on the underside of the geomembrane. In some
known cases, four feet of cover soil have been cast off of
32
-------
T [Required Geogrid (or GT) Strength]
Tension
Crack
H (Cover Soil Depth
at Unit Weight Y)
Active Wedge
Passive Wedge
L (Slope Length)
-/I
- 1
IH
LH "
2H2
\
r
. 5 )
sin
COS (
where
S = Possible Seepage Force
E = Possible Earthquake Force
Figure 3-7. Required strength of geogrid for cover soil reinforcement.
the geomembrane and a geomembrane "whale" has ap-
peared above the ground surface. To avoid such occur-
rences, these landfill gases (mainly methane) must be
conveyed along the underside of the liner in a uniform up-
ward gradient to a high point where the geomembrane is
penetrated. Here the gases are vented, flared, or cap-
tured for energy use.
While not widely implemented, geotextiles with adequate
planar transmissivity could serve this purpose. The
design-by-function concept is again used to select the
proper material, where FS = qaiiow/qreqd. The required
gas flow rate is very site specific, but is available in the
landfill gas literature. The allowable gas flow rate uses an
adapted form of ASTM D-4716, but with radial rather than
parallel flow (see Figure 3-8).
REFERENCES
1. U.S. EPA 1988. U.S. EPA guide to technical resour-
ces for the design of land disposal facilities. EPA
guidance document: Final covers on hazardous
waste landfills and surface impoundments. EPA/530-
SW-88-047.
2. Koerner, R.M., ed. 1989. Durability and aging of
geosynthetics. Elsevier Applied Science Publishers.
332 pp.
3. Koerner, R.M., J.A. Bove, and J.P. Martin. 1984.
Water and air transmissivity of geomembranes, Vol.
1. No. 1, pp. 57-74.
33
-------
Thick nea
gage*
\
W//M
Fabric -
Load
-*-
Water
head
U, = 3.5 psi
600 1000 1500 2000 2500
Stress (lb./*t?)
Figure 3-8. Allowable gas flow as adapted from ASTM D-4716 (3).
34
-------
CHAPTER 4
DURABILITY AND AGING OF GEOMEMBRANES
POLYMERS AND FOUNDATIONS
There is an almost infinite variety of polymeric
geomembranes or flexible membrane liners—that can be
used in landfill cover situations. The major groups are:
• Thermoset elastomers (which are rarely used due to
seaming difficulties)
• Thermoplastic
• polyvinyl chloride (PVC)
• chlorosulphonated polyethylene (either nonrein-
forced or reinforced)—CSPE or CSPE-R
• ethylene interpolymer alloy (reinforced)—EIA-R
• very iow density polyethylene (VLDPE)
• high density polyethylene (HOPE)
• Bituminous (which are rarely used in the United
States)
Geomembranes in the thermoplastic group are currently
most frequently utilized. These particular polymer resins,
however, are used in a formulation to arrive at the final
compound. Table 4-1 presents the formulations typically
used.
Table 4-1, Typical Formulations of Geomembranes
Geomembrane Resin
Type (%)
Carbon Black
and/or
Plasticizer Filler Additive*
PVC
CSPE
EIA
VLDPE
HOPE
45-50
45-50
70-80
96-98
97-98
35-40
2-5
10-25
0
0
10-15
45-50
5-10
2-3
2-2.5
3-5
2-4
2-5
1-2
<1
"refers to antioxidant, processing aids, and lubricants.
When assessing durability and aging of membranes,
each component of the compound within a particular
geomembrane must be addressed. Obviously, those for-
mulations that contain no plasticizers or fillers have a
less complicated set of mechanisms to consider, e, g. ,
VLDPE or HOPE. There is a wealth of information in the
literature on "resin" behavior, but little on the durability of
specific geomembrane formulations. For this reason, this
section will treat each possible degradation mechanism
individually, before dealing with synergistic effects and
lifetime prediction methods.
MECHANISMS OF DEGRADATION
Eight separate mechanisms of degradation are described
in this section. Some of the major ones, such as
ultraviolet degradation, can be eliminated by soil covering
and others, such as radiation or chemical degradation,
are not very likely due to the geomembrane's position in
the closure system above the waste.
Ultraviolet Degradation
By virtue of its short wavelength components, sunlight
can enter into a polymer system and (with sufficient ener-
gy) cause chain scission and bond breaking. Figure 4-1
shows the wavelength spectrum of visible and ultraviolet
radiation. Superimposed on the figure are the most sensi-
tive wavelengths of some commercially used polymers
for geosynthetic materials:
* PE < 300 nm
• PETS325 nm
• PP < 370 nm
The mechanism of ultraviolet degradation is well under-
stood and two approaches are taken to minimize its ef-
fects. Carbon black is added to the formulation, as a
blocking or screening agent, and chemical stabilizers are
added as scavenging agents. To eliminate degradation of
geomembranes in closure situations, however, the
geomembrane should be placed, seamed, and inspected,
and then covered with soil soon afterward.
Radiation Degradation
Clearly, radiation degradation is only of concern if there is
radioactive waste in the facility. Both y-rays and neutrons
within waste can degrade polymers and cause chain
scission and bond breaking. While there is a real concern
for high-level and transuranic wastes, low-level waste is
substantially less radioactive, and may not be a problem.
Chemical Degradation
Various chemicals can be aggressive to certain types of
geomembranes. For this reason, EPA has developed an
entire test protocol called EPA 9090 testing for assessing
35
-------
270 290 310 330 350 370 390 410 430 4SO 479 490 510 530 5SO 570 590
t4ocrr 1
FEjPfcT jpp Wivttonglh • ntnonwtcn
Figure 4-1. Wavelength spectrum of visibla and ultraviolet radiation.
chemical resistance. While testing is necessary for liners
beneath the waste in contact with leachate, the closure
liner should only interface with water, which comes from
seepage through the cover soil placed above it. Thus,
chemical degradation generally should not be an issue
with the thermoplastic geomembranes used in landfill
closures.
Swelling Degradation
All polymers swell when exposed to liquid, including
water. Generally, HOPE swells the least, PVC swells the
most, and the other geomembranes fall between these
two extremes. The swelling process is largely reversible
and does not necessarily lead to degradation. However,
swelling may cause secondary actions that could lead to
other synergistic effects.
Extraction Degradation
If one, or more, of the components of a geomembrane
formulation are extracted, the remaining material will ob-
viously be compromised. For example, if swelling leads
to bond breaking of the plasticizer within a PVC formula-
tion, the plasticizer could be extracted over time. This
phenomenon would decrease the elongation capability of
the geomembrane with respect to tension, tear, and
puncture modes of failure. The closest tests available to
estimate extraction are the following:
• ASTM D3083 for wafer extraction.
• ASTM D1203 for volatile loss.
Other tests, or test procedures, could be required on the
basis of a site-specific and material-specific basis.
Delamination Degradation
For geomembranes that are fabricated in layers, e. g. ,
scrim reinforced or multi-ply types, there is a possibility
that liquid can enter between the layers causing
delamination and premature failure,
To prevent delamination, the edges of multi-layered
geomembranes should be properly sealed in the factory
and have no scrim exposed. If the sheets are trimmed in
the field, the exposed edges should be "flood coated"
with a heavy bodied solvent. An ASTM test on ply ad-
hesion, ASTM D413, should be performed on the
material.
Oxidation Degradation
Oxidation of polymers caused by the gases or liquids in-
terfacing with the geomembrane is unavoidable. The
oxygen, over time, will enter into the polymer structure
and can react with various components in the particular
formulation. All geomembranes (and all geosynthetics)
are subject to this type of oxidation mechanism. The fol-
lowing equation illustrates this mechanism for
polyethylene degradation:
R ' + O2 -> ROO '
ROO ' + RH -> ROOM + R '
where R • = free radical
ROO' = hydroperoxy free radical
RH = polymer chain
ROOH = oxidized polymer chain
The rate of the reaction is very site and polymer specific
and is usually addressed experimentally. The testing pro-
cedure will be discussed in the section on accelerated
testing methods.
To minimize the oxidation reaction, the polymeric for-
mulation contains various anti-oxidants which scavage
(i.e. , neutralize) the free radicals. The amount of anti-
oxidants that can be added, however, is limited, and
once it is utilized, the oxidation process will proceed
depending on site-specific and geomembrane-specific
conditions.
Biological Degradation
Microorganism degradation of the "resin" portion of the
various geomembrane formulations is probably not a
problem. There is no literature available concerning bac-
36
-------
terial or fungal attack of high molecular weight resins.
Microorganisms may, however, interact with the plas-
ticizers and/or fillers used in certain geomembranes. Two
ASTM tests can be used to detect this type of degrada-
tion: G21 deals with resistance of plastics to fungi and
G22 is the complementary test for bacterial resistance.
Higher forms of biological life, like burrowing animals,
may be a much more serious problem. A muskrat, or
other small mammal, interested in burrowing through a
geomembrane, could easily do so. The hardness of the
geomembrane versus the animal's teeth structure, force,
and hardness need to be considered. If such animals are
in the vicinity of the landfill, one might consider using a
rock "bio-barrier" above the geomembrane as per EPA
guidance.
SYNERGISTIC EFFECTS
The eight degradation mechanisms discussed in the pre-
vious section—ultraviolet, radiation, chemical, swelling,
extraction, delamination, oxidation, and biological—can
also interact to cause numerous complex effects. In addi-
tion, there are three situations that should be addressed
in any discussion on aging of polymeric materials:
elevated temperature, applied stresses, and long ex-
posure.
Elevated Temperature
All of the previously mentioned degradation processes
will be more severe at higher temperatures than at lower
temperatures. Activation energy, as will be discussed in
the section on Arrhenius modeling, is clearly a function of
elevated temperature. Thus, a given formulation of a
specific geomembrane will have a shorter lifetime in the
southern states (all other things than temperature being
equal) than in the northern states. A quantification of this
amount, however, requires experimentation and ap-
propriate modeling of the situation.
Applied Stresses
It seems intuitive that the lifetime of a geomembrane in a
relaxed state would be different than that of the same
geomembrane under stress. Thus, modeling of a given
situation should somehow take stresses into account. At
a minimum, compressive stresses should be assessed;
however, tensile and shear stresses (for liners on side
slopes) and out-of-plane bending stresses (for liners in
closures over subsiding waste) are also likely and should
be considered. Simulating the magnitude and the type of
stresses is very difficult and requires making many as-
sumptions.
Long Exposure
Landfill closures are designed (at the minimum) for 30-
year postclosure care periods. Beyond this time frame,
the facility's status is questionable and inquiries often
arise as to ultimate ownership and responsibility for main-
tenance and repairs. To alleviate some of these con-
cerns, it would be useful to quantify in the planning
stages how long a geomembrane will last before its
properties are degraded beyond serviceability. The
longer the geomembrane's lifetime, the easier it will be to
deal with these issues.
ACCELERATED TESTING METHODS
A number of accelerated testing methods in the open
literature predict the lifetime of polymeric materials. Many
of these sources are available in the plastic pipeline
literature from the Gas Research Institute, the Plastic
Pipe Institute, and related organizations. The reference
list for "Long-Term Durability and Aging of
Geomembranes" (see Appendix B) also contains many
useful sources of information.
Stress Limit Testing
Stress limit testing to obtain plastic pipe "design stress" is
fairly well developed and widely implemented. It requires
simulated environmental testing with sections of pres-
surized capped pipe at constant temperature. Figure 4-2
shows typical results where the service time of the pipe is
selected and the design stress is obtained from the ex-
perimental curve.
Rate Process Method for Pipe
In this experimental method for polyethylene pipe, con-
stant stress tests are conducted at different elevated
temperatures. The design life is selected, intersected with
the site-specific temperature response curve, decreased
by a suitable factor-of-safety, and then extended to ob-
tain the allowable stress. The graph in Figure 4-3 shows
pipe behavior at 80°C, 60°C, and 20°C. The behavior at
20°C must be extrapolated as per the details in the paper
by Koerner, Halse, and Lord in Appendix B. (See Appen-
dix B, "Long-Term Durability and Aging of
Geomembranes.")
Rate Process Method for Geomembranes
The rate process method (RPM) can be used to test
geomembranes if the site-specific conditions are properly
modeled. The curve in Figure 4-4 gives notched constant
load test data in an aggressive incubation liquid. The
method is not developed for utilization at this time.
Arrhenius Modeling
Arrhenius modeling is the method most widely used by
chemists and polymer engineers to predict the lifetime of
polymeric materials. This type of modeling assumes that
elevated temperature can be used to simulate time at a
site-specific (and lower) temperature. This assumption is
sometimes referred to as a temperature-time superposi-
tion concept. Figure 4-5 shows an example of a possible
test device. In Arrhenius modeling, measuring a suitable
reaction rate of the polymer at different temperatures
produces a linear plot on an inverse temperature graph.
37
-------
STRESS LIMIT TESTING
100-3
1CH
0}
Q.
O
o
Design Stress
10
Figure 4-2. Stress limit testing for plastic pipe.
100 1000 10000 100000 1000000
Failure Time (hrs.)
RATE PROCESS METHOD FOR GEOPIPES
log a (Mpa)
Callow = 6.5
20°C
Figure 4-3. Rate process method for testing pipe.
The slope of the curve (if it is linear) can be used to ex-
trapolate to the site-specific temperature. The calcula-
tions then follow along the lines given below.
Information regarding polyethylene shielding of electric
cables is used here to demonstrate this technique. The
reaction rate illustrated in Figure 4-6 is for 50 percent
strength reduction of the original and unaged value using
an impact test. (Any one of a number of tests could have
been used.) The slope of the Arrhenius plot shown is:
Eact
R
In10~5-ln10"2
0.00247-0.00198
= -14,000 CK)
To predict low-temperature behavior, consider the
elevated temperature data point at:
1/T = 0.00213
log time
50 years
T
T
469°K
196°C
and project this reaction rate to the site-specific tempera-
ture of 90°C. Using the Arrhenius equation,
Rjl _ _ Eact f 1 _ 1 1
~ 6
Rr2
where Rn =
Rr2 =
Eact/R =
Ti =
T2 =
R
reaction rate at temperature Ti
reaction rate at temperature Ta
slope of experimental curve
experimental (test) temperature
site-specific (desired) temperature
38
-------
Hydraulic Load Device
wan uauge
Air (7)
Liquid Level
Sight Glass
-f^lPH
' 'Record
S.
^-*
I
r
c
0
o
o
o
o
er o
i —
Leachate '-j
Recirculation «
Pump «
k=
t
1
Liquid
T
1
1
k^^
Sand
5^*:*'3Cx>.^
tSswtftfwv?'*-*-.
0
•« Heal Transfer
* Coils
3 Perforated
j-"" " Plate Press
3
3 .^-- Thermocouples
j^ Test Liner
9
>
»
5""t*3 Drain
Figure 4-4. Rate process method for testing geomembranes.
RATE PROCESS METHODS FOR GMs
100
u)
2
in
~u
0)
>
H
4<
O
U.
0>
CL
10
Experimeniat Data @ 4
Predicted Response @ 25°
—./
Experimental Data @ Sot
*T-7
10
Failure Time (hours)
100
Figure 4-5. Testing device for Arrhenius modeling.
Rf @ 196 _ -14.000
Rr@90
1
469 363
= 6,587
Since the reaction took 1,000 hours to complete at
196°C, the comparable reaction rate at 90°C would be:
Rr@90°C = 6,587(1000)
6,587,000 hrs
752 yrs
Because the temperatures of the experiments used in the
example are quite high and quite limited (i. e. , they are
bunched together), extrapolation down to the site-specific
temperature mentioned may be invalid. One does not
know which, if any, of the geomembrane properties will
be amenable to the Arrhenius approach, but the various
possibilities should be investigated on a project-specific
basis and as a general research area.
Multi-Parameter Prediction
Using a number of experimental and field-measured
response curves, it may be possible to generate a lifetime
prediction method. Required (see the Hoechst reference
in Appendix B) are the constant stress (Figure 4-7a),
stress relaxation (Figure 4-7b), and field-measured strain
39
-------
0.0020 0.0022 0.0024 0.0026
t / TEMPERATURE (1 /•*)
Figure 4-6. Reaction rate for impact testing of polyethylene shielding.
(Figure 4-7c) curves. By superposition of the proper
temperature response curve and the appropriate strain
response curve (laboratory and field), one can possibly
project the lifetime of the considered geomembrane
under three possible assumptions:
• No additional stress relaxation, curve (a)
• Intermediate stress relaxation, curve (b)
• Full stress relaxation, curve (c)
These results are each shown on the graph in Figure 4-
7d. This technique is potentially useful, but requires a
relatively large amount of experimental and field data.
SUMMARY AND CONCLUSIONS
Durability and aging of geomembranes (and all geosyn-
thetics) are important issues, especially when consider-
ing the situation of landfill covers beyond the 30-year
postclosure care period. Fortunately, most degradation
processes are eliminated or greatly reduced by burying
the geomembrane in soil soon after installation. Also, be-
cause the interfacing liquid is water and not leachate, as
with the liner beneath the waste, there is little problem
with chemical degradation. Long-term oxidation,
however, is a degradation mechanism that can only be
retarded (via anti-oxidants), but not eliminated, and, thus,
is a focal parameter for experimental modeling.
Of the predictive models that have been reviewed, the
Arrhenius modeling technique, which is under active in-
vestigation, is in the most widespread use. Equally inter-
esting is the multi-parameter approach, but this method is
much less developed. Whatever techniques are used,
they are only laboratory prediction methods. Field feed-
back is necessary to establish better insight into degrada-
tion and aging issues involving polymeric geomembranes
and other related geosynthetic materials.
40
-------
100
10
10
70
"A
8!
CONSTANT STRISS TESTS
(«.g., NCLT or SCLT)
DTP REGION
IS 1 53 10
TIME (HOURS)
•1 —
... ti
O.I 3
0.7 0
0.
o.s 2
»—
Q
0.5 5
025 |
|
0.2
0.15 .
CONSTANT STRAIN TISTS
i.e., STRESS RELAXATION)
100 idoo i OMO
YEARS
10 I 10" 10'
TiMi IHOURS!
100 1000 10000
YEARS
- SO
to
C 30
t-
(O
a
COUPLE LAB AND FIELD
STRAIN GAUGE DATA
FIELD
PAjT*
1 to 10 10
TIME (HOURS)
10 10 10 10 10
i 10 loo 1000 10000
YEARS
O
UJ
ui
cc
to
LU
90
80
70
60
40
30
20
10
SUPERPOSITION OF SITE SPECIFIC
• CONSTANT STRESS CURVE,
• CONSTANT STRAIN CURVE, and
• FIELD STRAIN CURVE
(a) no add relax.
(b) intermsd, relax.
(c) full relax.
I . L
PROJECTED
LIFETIME
it
10 1 10 10
TIME (HOURS)
10
10" 10 10" 10'
0.9
0.8
0.7
0.6
0.5
*
a.
u.
O
0.3 £.
25
2
0.15 "f
1 0
10
100 1000 10000
YEARS
Figure 4-7. Experimental and field-measured response curves for multi-parameter lifetime prediction.
41
-------
CHAPTER 5
ALTERNATIVE COVER DESIGNS
INTRODUCTION
The hazardous waste landfill cover designs developed
and published by the U.S. Environmental Protection
Agency (EPA) are generic in nature and intended to meet
the regulatory criteria for covers on a national basis. This
section reviews these designs to determine alternatives
that would be acceptable to the regulatory community.
This discussion will also review designs being considered
by other agencies, such as the Nuclear Regulatory Com-
mission (NRC).
SUBTITLE C
The basic acceptable generic design for hazardous waste
landfills incorporates natural soil (clay), geomembrane,
drainage, and vegetation layers. This generic design,
however, does not take into account site-specific con-
cerns, such as siting in arid areas where rainfall is very
low. Under these conditions, a barrier layer composed of
both a natural soil (clay) and a geomembrane layer
probably would not be effective. The natural soil layer is
designed to be placed "wet-of-optimum" to achieve the
minimum hydraulic conductivity. When placed in a rela-
tively dry environment, this layer will dry and crack,
making it less effective. In selected cases, the newer
bentonite blankets may be an acceptable alternative.
From a technical standpoint, the geomembrane may be
the only barrier necessary. Site-specific considerations
such as settlement/subsidence, environmental exposure,
and other physical conditions may influence the thickness
of the geomembrane required.
In selected cases, a vegetation layer alone may be
demonstrated (via the Hydrologic Evaluation of Landfill
Performance (HELP) model—see Chapter 8) to meet the
criteria. In this case, a thicker soil layer may be required
to assist in establishing the natural vegetation and to act
as a storage reservoir for the infrequent but high intensity
rainfall.
In summary, the design criteria were established for a na-
tional generic design. EPA is always interested in review-
ing alternative designs that are innovative and utilize
site-specific information. These alternative designs
should be demonstrated to be equivalent in performance
to the generic design proposed by EPA.
SUBTITLE D
While EPA has proposed some generic design con-
siderations, Subtitle D facility designs will most likely be
approved by individual states. Cover designs should be
incorporated into the overall facility design, taking the
bottom liner and liquids management strategy into ac-
count. Depending on site-specific considerations, designs
based on natural soils as well as designs that resemble
multilayer Subtitle C designs will be developed.
Municipal solid waste landfills usually require a daily
cover of natural soils or other alternative materials. One
possible use for postconsumer paper or unsaleable glass
or glass culls may be for daily cover. Some enterprising
individuals have developed a product made from
shredded paper mixed with other proprietary ingredients
that can be blown onto the surface of the waste to meet
the requirements of daily cover. Foams and other
materials have also been developed and evaluated for
performance. Each of these materials has to be
evaluated economically for site-specific use but may
have advantages technically. For example, if the liquids
management strategy for a landfill includes leachate
recirculation, blown-on materials may allow more
homogeneous distribution of the leachate. Another ad-
vantage is that blown-on materials will lose their barrier
qualities as soon as the next layer is placed in the facility.
Natural soils, on the other hand, do tend to act as bar-
riers, which may cause leachate to seep out the side of
the final cover.
Whatever alternate materials are used, they should be
demonstrated to meet the technical requirements for
daily cover.
CERCLA
CERCLA or Superfund cover designs are more complex
from the standpoint of jurisdiction, where ARARs (dis-
cussed in Chapter 1) play an important part in selecting
the final design. A multilayer cover system may be most
environmentally desirable; however, other site-specific
considerations may allow other types of designs. For ex-
ample, early CERCLA covers have been constructed by
regrading existing cover material, and adding small
amounts of cover soil (usually about 6 inches) and, in
43
-------
some cases, a rock armor. Due to the inequality and with
the ARARs ruling, compliance with a RGRA multilayer
has been more acceptable. Site-specific design changes
have been approved after they were demonstrated to
meet the intent of the regulations.
OTHER COVER DESIGNS
The Department of Energy (DOE) and the NRC are both
considering cover designs for landfills containing low
level radioactive wastes. In general, these designs are
comparable to the EPA's multilayer design, with some
notable exceptions. One of the main criteria differences is
based on the fact that DOE and NRC designs have to
last for thousands of years due to the type of waste they
are covering. The long-term nature of their designs has
minimized the use of geosynthetics, since geosynthetics
are thought to have a finite service life.
DOE will soon publish results of a study designed to
develop an all natural soil cover system with a long ser-
vice life (1). The study considered what type of soil would
best qualify for each design aspect- A matrix was
developed from which completed matrix designs will be
proposed.
NRC also has been reviewing conceptual designs that
use natural soils and have long life. Three cover designs
are currently under investigation: (a) resistive layer bar-
rier, (b) conductive layer barrier, and (c) bioengineering
barrier (2). These designs are being assessed in large
(21 x 14 x 3 m [70 x 45 x 10 in.] each) lysimeters in
Beltsville, Maryland. The resistive layer barrier, shown in
Figure 5-1, consists of compacted natural soils or clay.
The resistive layer depends on the low hydraulic conduc-
tivity of the compacted layer to minimize any potential
moisture interaction with the waste.
The conductive layer barrier, shown in Figure 5-2, makes
use of the capillary barrier phenomena to increase the
moisture content above the interface and to divert water
away from and around the waste (3). The capillary barrier
is established when coarse grain soils are sandwiched
between fine grain sediments. Experiments have shown
that the greater the contrast in the permeability between
the two layers, the more effective the barrier. A second
fine grain soil layer would direct water away from the
gravel layer under saturated conditions.
It should be noted that NRC considers these two concep-
tual designs unacceptable where appreciable subsidence
may take place (2). This failure potential in the above two
designs necessitated the development of an easily
reparable surface barrier to be used until major settle-
ment/subsidence activities had ceased. The surface bar-
rier could be easily repaired during the
settlement/subsidence time period, after which a more
permanent barrier could be installed. The bioengineering
management cover system (Figure 5-3) was the result.
This cover system utilizes a combination of engineered
enhanced runoff and stress vegetation, e.g., Pfitzer
junipers, growing in an overdraft condition to control deep
water percolation through cover systems. Stress vegeta-
tion are grasses, trees, and shrubs that can survive when
under stress, such as lack of water. Early results from the
field indicate this system to be very effective in controlling
liquid movement into or out of the waste management
unit.
Figure 5-1. Resistive layer barrier.
44
-------
Figure 5-2. Conductive layer barrier.
Juniper (Pfitzer)
12 to 14'on Center
V
Access Tube (Neutron Probe)
2* Aluminum-
Gutter
•Gutter
Steel Drums (55 gal.,
1/3 Filled with Gravel) v
4-20 mil. Vinyl Liners
Between 5 Layers of Geotexttte
Geotextite
Figure 5-3. Side view of bioengineered lysimeter. Surface runoff is collected from both engineered surface and soil sur-
face. Soil moisture content is measured with neutron probe. Water table is measured in well.
REFERENCES
1. Identification and ranking of soils for UMTRA and
LLW disposal facility covers. U.S. Army Corps of En-
gineers, Waterways Experiment Station. Un-
published.
2. O'Donnell, E., R.W. Ridky, and R.K Schulz. 1990.
Control of water infiltration into near surface LLW dis-
posal units. Progress report on field experiments at a
humid region site, Beltsville, MD. Waste Manage-
ment '90 Tucson, Arizona, February.
Zunker, J.F. 1930. Das Verhalten des Bodens Zum
Wasser In: E. Blanck, ed. Handbuch Der Bodenlehr.
V. 6. Berlin: Verlag von Julius Springer, pp. 66-220.
45
-------
CHAPTER 6
CONSTRUCTION QUALITY ASSURANCE FOR SOILS
INTRODUCTION
Construction quality assurance (CQA) is critical for
producing engineered cover systems that will perform
satisfactorily. The critical CQA issues for soils used in
cover systems are:
• Control of the soil materials used to build various com-
ponents of the cover system.
• Control of subgrade preparation.
• Control of the placement and compaction of soils.
• Protection of the soil during and after construction.
• Use of test pads in the CQA process.
This chapter examines each of these CQA issues. An
EPA guidance document (1) provides general information
concerning CQA, including responsibilities of the contrac-
tor and inspector. In general, the purpose of CQA is to
provide observations and tests that assist in evaluating
whether the construction has been performed in accord-
ance with specifications. Accordingly, each CQA program
must be tailored to the specific construction specifica-
tions for a given project. The sections that follow discuss
general principles that should be considered when
developing a CQA plan.
MATERIALS
The most important and useful quality control (QC) tests
for soil materials used in cover systems are Atterberg
limits, percentage of fines, percentage of gravel, and the
maximum size of the largest stones or clods of clayey
soil.
Atterberg Limits
The liquid and plastic limits or Atterberg limits of a soil,
which are measured with ASTM Method D4318, can be
useful indicators of the suitability of a soil for a specified
purpose. The plasticity index (PI) of a soil is defined as
the liquid limit minus the plastic limit and is a measure of
the breadth of water content over which the soil behaves
plastically. For hydraulic barrier materials, the soil must
have adequate plasticity; otherwise, the material will be
too deficient in clay to serve as an adequate hydraulic
barrier. For drainage materials, the soil must be free of
clay—drainage materials typically have little or no plas-
ticity. By measuring the plasticity of a soil, Atterberg limits
provide a rapid and convenient means for assessing its
suitability for its intended purpose.
The Atterberg limits of a soil are measured in the
laboratory. Samples for testing can be taken from the
borrow area or from the final construction area. Ex-
perienced field engineers and technicians can often tell
just from examining and handling the soil whether it has
the appropriate Atterberg limits. With questionable soils,
or soils that are variable in the borrow area, it is helpful to
sample the borrow soils on a close grid of test pads and
wait for the test results before proceeding with further soil
processing or placement.
Percentage of Fines
The percentage of fine-grained material in a soil is
defined as the percentage on a dry-weight basis of soil
that will pass through the openings of a No. 200 sieve,
which are 0.075 mm (0.003 in.) wide. Material retained
on the No. 200 sieve is defined as the coarse-grained
fraction and material that will pass through the openings
is the fine-grained fraction. The percentage of fines may
be measured with ASTM Method D422 or D1140, with
sample preparation performed with Method D2217, if
necessary.
As with Atterberg limits, an experienced engineer or tech-
nician can often tell by visual observation whether the soil
has an adequate amount of fines (barrier materials). In
such cases, the QC tests serve primarily to build a formal
record of test results that verifies the observations made
by the field personnel. With sandy drainage materials, it
is usually difficult to determine by visual observation
alone whether the material has an excessive amount of
fines.
Percentage of Gravel
The soil for barrier materials cannot contain an exces-
sively large percentage of gravel. The gravel fraction is
determined by sieving the soil through a No. 4 sieve,
which has 4.76-mm (0.19-in.) wide openings. Sieving of
soil to determine the percentage of gravel is performed
with ASTM Method D422, a method similar to the test for
percentage fines. All material that will not pass through
the openings is defined as gravel, according to the
Unified Soil Classification System (ASTM Standard Prac-
47
-------
tice D2487). (Note: Sieve analysis via ASTM Method
D422 is one of several tests used for soil classification
via the procedure for interpretation of test data given in
Standard Practice D2487. Also, particles larger than 75
mm [3 in.] are cobbles.)
Maximum Size of Panicles or Clods
For barrier materials, the maximum size of stones in the
clayey soil cannot be too large. However, it is impractical
for field personnel to sieve large, representative samples
of soil to determine the largest particle size. In the field,
the problem is probably an occasional oversized stone,
which no formal sampling procedure is likely to detect.
Rather, observations by CQA personnel provide the best
opportunity to detect excessively large stones.
On occasion, the maximum size of soil clods may be
specified in the construction specifications. Again, sieving
wet, clayey soils to determine the clod size distribution is
impractical. Direct measurement of representative clods
by field personnel is probably the simplest and best way
to verify that the clods are not too large.
Requirements for Field Personnel
On large and important projects, where CQA is con-
sidered crucial to the overall success, a full-time inspec-
tion of soil excavation in the borrow area and continuous
classification of excavated soils are recommended. Soils
are variable materials, and the borrow area offers the
best opportunity to detect the presence of unsuitable
materials.
Frequency of Testing
Table 6-1 summarizes the frequency of testing recom-
mended by Daniel (2). These recommendations are
based primarily upon practices reported by Gordon,
Huebner, and Kmet (3) and reiterated by Goldman et al.
(4) for clay liners. Although the recommendations are in-
tended for low hydraulic conductivity liners, they are use-
ful for other soil materials, as well. Experience has shown
that even more frequent testing is helpful during the initial
phases of construction, because this is the period when
problems are most likely to occur.
CONTROL OF SUBGRADE PREPARATION
The subgrade must be properly prepared and compacted
before any component of the cover system that requires
significant compaction can be placed. Typically, the low
hydraulic conductivity soil barrier requires compaction
with heavy equipment, whereas the compaction of other
layers is much less important. Thus, CQA for subgrade
preparation is critical for the low hydraulic conductivity
component of a cover system but may not be necessary
for other components.
Table 6-2 summarizes the recommended tests for sub-
grade preparation. For low hydraulic conductivity soils,
the surface of a previously compacted layer of soil should
be disked ("scarified") prior to placing a new layer of soil.
Table 6-1. Recommended Materials Tests for Barrier
Layers (2)
Parameter Test Method Minimum Testing
Frequency
Percent Fines (1) ASTM D1140 1 per 1,000 yd3 (2)(3)
Percent Gravel (4) ASTM D422 1 per 1,000 yd3 (2)(3)
Liquid & Plastic ASTM D4318 1 per 1,000 yd3 (2)(3)
Limits
Water Content ASTM 04643(5) 1 per 200 yd3 (2)(6)
Water Content (7) ASTMD2216 1 per 1,000 yd3 (7)(3)
Construction
Oversight
Observation
Continuous in borrow
pit on major projects;
continuous in placement
area on smaller projects
Notes:
1. Percent fines is defined as percent passing the No. 200 sieve.
2. In addition, at least one test should be performed each day
that soil is excavated or placed, and additional tests should
be performed on any suspect material observed by QA
personnel.
3. 1,000 yd3 = 836m3.
4. Percent gravel is defined as percent retained on the No. 4
sieve.
5. This is a microwave oven drying method. Other methods
may be used, if more appropriate. Any method used besides
direct drying via ASTM D2216 should be calibrated against
ASTM D2216 for the onsite solids.
6. 200 yd3 = 167m3.
7. Microwave oven drying and other rapid measurement
methods may involve systematic errors. Conventional oven
drying (ASTM D2216) is recommneded on every fifth sample
taken for rapid measurement. The intent is to document any
systematic error in rapid water content measurement.
Requirements for scarification should be set forth in the
construction specification. The scarification should be ob-
served to ensure that the newly placed layer blends in
with the previously compacted layer.
SOIL PLACEMENT
Soils are placed in layers called lifts. The thickness of a
lift is typically specified in the construction documents.
For materials having low hydraulic conductivity, the thick-
ness of a lift should not exceed the specified value.
Otherwise, the lower portion of the lift may be inade-
quately compacted, the bonding of lifts is likely to be
poor, and the hydraulic conductivity could be larger than
desired. Control of lift thickness is critical for low-
hydraulic-conductivity, compacted soil liners. The loose
lift thickness can be checked visually near the edge of a
lift. The exact thickness of a loose lift can be measured
by digging a hole in the soil.
48
-------
Table 6-2. Recommended Tests and Observations on
Subgrade Preparation (2)
Parameter
Test Method
Minimum Testing
Frequency
Percent
Compaction (1)
ASTM D2922 or
ASTMD1556or
ASTM D2937 or
ASTMD2167
1 per acre (2)
Compaction Curve ASTM D698 (3) 1 per 5 acres
Observation
Full coverage
Preparation of
Previously
Compacted Lift
Notes:
1. Percent compaction is defined as the dry density of the
compacted soil divided by the maximum dry density
measured in the laboratory with a specified method of
compaction. The test methods listed are for measurement
of the dry density of the compacted soil.
2. In addition, at least one test should be performed each day
the construction personnel prepare subgrade by compaction.
3. Other laboratory compaction methodologies are often
employed.
4. 1 acre = 0,4 ha,
Fill elevations are usually controlled with grade stakes or
lasers; laser equipment is not currently in widespread
use. If grade stakes are used, care must be taken to
remove them and repair the resulting holes. The CQA in-
spector should make sure that grade stakes are not
buried in the cover system. To accomplish this, an inven-
tory system in which all grade stakes are numbered and
accounted for each day is recommended. One advantage
of ferrous metal grade stakes is that if inadvertently
burled in the cover system, they can be found with a
metal detector. The holes left by grade stakes should be
packed with soil liner material or bentonite tamped into
the hole in layers with a rod.
SOIL COMPACTION
Drainage Layers
Nominal compaction of drainage layers is usually ade-
quate. Rarely is it necessary to control the degree of
compaction of drainage materials. One potential problem
to avoid is "bulking" of wet or damp sands; compaction in
lifts will overcome such problems.
Of greater importance than the degree of compaction is
protecting drainage materials from contamination by
fines. Over-compaction of the drainage materials can
grind up soil and increase the amount of fines. However,
the specifications should not permit use of nondurable
materials that are easily broken down. Field personnel
should observe the amount of fines before and after com-
paction. If there is any question about grinding of the soil
during compaction, the percentage of fines should be
measured after compaction to confirm that the compac-
tion process has not increased the percentage of fines.
Barrier Materials
Quality control of barrier materials usually focuses heavi-
ly on water content and dry unit weight. A typical con-
struction specification might require that the soil be
compacted over a specified range of water content (e.g.,
0 to 4 percent wet of optimum) to a minimum dry unit
weight (e.g., 95 percent of the maximum dry unit weight
from standard Proctor compaction).
The methodology for determining the appropriate com-
paction criteria for CQA has recently been reviewed by
Daniel and Benson (5). Figure 6-1 shows the form of a
typical compaction specification. The "Acceptable Zone"
is based upon a specified range of water content and a
minimum dry unit weight. The zero air voids curve repre-
sents a theoretical upper limit above which points cannot
exist. Figure 6-1 represents the usual format for specify-
ing the compaction requirements for a barrier layer.
However, as indicated by Daniel and Benson, the usual
format represents historical practice for structural fills and
is not necessarily appropriate for low hydraulic conduc-
tivity soil liner or cover systems. The next several graphs
illustrate the problem with the usual form of specification.
Figure 6-2 contains a compaction curve and a plot of
hydraulic conductivity versus molding water content for
three compactive energies. The water content-dry unit
weight points are replotted in Figure 6-3 with solid sym-
bols used for those compacted specimens that had
hydraulic conductivities less than or equal to 1 x 1CT7
cm/s and open symbols for specimens with a hydraulic
conductivity greater than 1 x 1Q~7 cm/s. The Acceptable
Zone, which encompasses the compacted specimens
with low hydraulic conductivity, has a much different
shape from the one shown in Figure 6-1. Figure 6-4
presents contours of values of water content and dry unit
weight that yielded certain hydraulic conductivities for
one particular soil. Also shown in Figure 6-4 is a modified
Proctor compaction curve and a typical specification that
might be written using the procedure suggested in Figure
6-1. In this case, a portion of the typical Acceptable Zone
contains soils with unacceptably large hydraulic conduc-
tivities. Use of the Acceptable Zone in Figure 6-4 based
on typical construction practice, i.e., the practice
sketched in Figure 6-1, does not ensure that the com-
pacted soils have low hydraulic conductivity.
The recommended procedure for defining a suitable
range of water content and dry unit weight is shown in
Figure 6-5. The procedure involves four steps:
1. The soil is compacted with three compactive ener-
gies that span the range of compactive effort ex-
pected in the field. The three energies recommended
by Daniel and Benson (5) are modified Proctor,
standard Proctor, and reduced Proctor. Reduced
49
-------
Zero Air Voids
Acceptable Zone
w
opt
Molding Water Content, w
Figure 8-1. Traditional method for specification of acceptable water contents and dry unit weights (5).
Proctor is the same as standard Proctor except that
only 15 drops of the hammer (rather than the usual
25) are employed per lift. Approximately five to six
samples are compacted with each energy.
2. The specimens are permeated and the hydraulic con-
ductivity of each specimen is determined. Hopefully,
at least some of the test specimens will have
hydraulic conductivities that are less than the design
maximum value. Care should be taken to make sure
that the conditions of permeation appropriately simu-
late field conditions. For cover systems, it is par-
ticularly important that the confining stress used in
the laboratory tests is not significantly larger than the
value expected in the field (which is usually small for
cover systems).
3. The water content-dry unit weight points are
replotted, and an Acceptable Zone is drawn. Some
judgment may be necessary in drawing the Accept-
able Zone.
4. The final step is to modify the Acceptable Zone in
any appropriate manner to take into account other
variables besides hydraulic conductivity, e.g., sus-
ceptibility to desiccation damage, local construction
practices, or shear strength considerations. When the
Acceptable Zone is modified, it is only made smaller,
not larger. Figure 6-6 illustrates how one might com-
bine an Acceptable Zone based on hydraulic conduc-
tivity with one based upon shear strength to develop
a single, overall Acceptable Zone.
The lower limit of the Acceptable Zone will probably be
parallel to a line of constant degree of saturation or to the
line of optimums (Figure 6-7). (The line of optimums is a
curve that connects points of maximum dry unit weight
and optimum water content measured with different ener-
gies of compaction.) It may be possible to use a constant
degree of saturation or a line parallel to the line of op-
timums for the lower limit of the Acceptable Zone.
50
-------
C/)
E
o
>»
."E
•t—•
Q
T3
C
O
O
a
a.
c
ID
10
10
10
10
10
10
-4
-5
-6
-7
-8
-9
H
\ Medium
n
High Effort
10
15
20
25
Molding Water Content (%)
120
110
100
90
(B)
Low Compactive Effort
15
20
25
Molding Water Content (%)
Figure 6-2. Data from Mitchell et al. for silty clay compacted with Impact compaction (6).
51
-------
120
o
Q.
110
CD
100
Q
90
Acceptable Zone .
D
10 15 20 25
Molding Water Content (%)
Figure 6-3. Compaction data for silty clay (6); solid symbols represent specimens with hydraulic conductivity less than or
equal to 1x10~7cm/s and open symbols represent specimens with hydraulic conductivity >1x10"7cm,s.
120
o
Q_
D)
"0
110
C
=3
100
Q
90
* I
Mod.
Proctor
Curve
10-5
Acceptable Zone Based on
Typical Current Practice:
W > 0.9 ^max and
w = 0 - 4% Wet of w0pt
15
20
25
Molding Water Content (%)
Figure 6-4. Contours of constant hydraulic conductivity for silty clay compacted with kneading compaction (6).
52
-------
Molding Water Content
Molding Water Content
'(B)
\
" -a,
x X
* \ l\
o Sisndsiti Pioci
O ReSjced Pmcio
Maximum Allowed
X
V^-Jf^'
e-
Q
Accsplabte
ModWied to Aeeounl
lor Other Factors
Molding Water Content
Molding Water Content
Determine compaction curves with three compactive efforts
Determine hydraulic conductivity of compacted specimens
Replot compaction curves using solid symbols for samples with adequately low hydraulic
conductivity and open symbols for samples with a hydraulic conductivity that is too large-
Modify Acceptable Zone based on other considerations such as shear strength or local
construction practices
Figure 8-5. Recommended procedure.
The water content of the soil at the time the soil is com-
pacted has a significant impact on almost all engineering
properties of the soil. For instance, compaction of the soil
at a low water content leads to a strong, low-compres-
sibility soil that is not as vulnerable to desiccation crack-
ing as wetter soils. However, dry soils are brittle and
crack easily, e.g., if there is differential settlement of the
underlying waste. Soils compacted at high water contents
are softer, more compressible, and more vulnerable to
damage from desiccation. Wet soils, however, are ductile
and can accommodate more differential settlement
without cracking than dry soils can. The water content
range specified for construction should reflect a careful
consideration of these, and possibly other, variables. It is
critical that the CQA program ensure that the soil is com-
pacted to the proper water content.
The procedure summarized in Figure 6-5 was applied to
a cover system constructed at the Oak Ridge Y-12
Operations as described by Daniei and Benson (5). The
compaction curves for one of the two types of soils
(called Type A soils) are shown in Figure 6-8. Hydraulic
conductivities are plotted in Figure 6-9. For this project,
the design called for compaction of the soil to achieve a
hydraulic conductivity of 1 x 10~7 cm/s or less. The Ac-
ceptable Zone, shown in Figure 6-10, shows the range of
water content and dry unit weight that met this require-
ment. In the field, technicians measured the water con-
tent and dry unit weight, checked to see if the point
plotted within the Acceptable Zone, and made a pass-fail
decision based on whether the point was within or out-
side of the Acceptable Zone. If the point was outside the
Acceptable Zone, either the soil was compacted more or
the soil was removed, reprocessed, and recompacted.
For soil bentonite mixes, the following procedure is
recommended:
1. Mix batches of soil at different bentonite contents,
e.g., 0, 2, 4, 6, 8, 10, and 15 percent bentonite (dry
weight basis).
2. Develop standard Proctor compaction curves for
each bentonite content.
3. Compact samples with standard Proctor procedures
at a water content 2 percent wet of optimum for each
bentonite content.
4. Permeate the soils prepared in Step 3 and develop a
plot of hydraulic conductivity versus bentonite con-
tent.
5. Decide how much bentonite to use based on data
from Step 4, taking into account construction
variability. Usually more bentonite is used than Step
4 indicates is necessary, because, in the field, the
bentonite will not always be mixed as uniformly with
the soil as it was in the laboratory.
6. For the average bentonite content expected in the
field, utilize the procedures described earlier and
summarized in Figure 6-5.
The procedures discussed in the preceding pages for
determining an appropriate range of water content and
dry unit weight involve laboratory tests. The compaction
53
-------
Table 6-3. Recommended Tests and Observations on
Compacted Soil for Barrier Layers (2)
Parameter
Test Method
Minimum Testing
Frequency
Water Content (1)
ASTMD3017or
ASTM D4643
Water Content (3) ASTM D2216
5/acre/lift (2)
1/acre/lift (3)
Density (4)
Density (Note 5)
Number of
Passes
Construction
Oversight
Notes:
1. ASTMD3017
ASTM D2922 or
ASTM D2937
D1556
Observation
Observation
is a nuclear method and
5/acre/lift (2)
1 /acre/lift (5)
1 /acre/lift (2)
Continuous
D4643 is microwave
oven drying. Direct water content determination (ASTM
D2216) is the standard against which nuclear, microwave,
or other methods of measurements are calibrated for onsite
soils.
2. In addition, at least one test should be performed each day
soil is compacted, and additional tests should be performed
in areas for which QA personnel have reason to suspect
inadequate compaction.
3. Every fifth sample tested with ASTM D3017 or D4643 also
should be tested by direct oven drying (ASTM D2216) to aid
in identifying any significant, systematic calibration errors
withD3017orD4643.
4. ASTM D2922 is a nuclear method and D2937 is a drive ring
method. These methods, if used, should be calibrated
against the sand cone (ASTM D1556). Alternatively, the
sand cone method can be used directly.
5. Every fifth sample tested with D2922 or D2937 also should
be tested (as close as possible to the same test location)
with the sand cone (ASTM D1556) to aid in identifying any
systematic calibration errors with D2922 or D2937. The
sand cone method may be used in lieu of D2922 and
D2937.
6. 1 acre = 0.4 ha.
procedures used in the laboratory should simulate field
compaction as closely as possible. There is always a
possibility, however, that field construction will produce
macro-scale features (e.g., poor bonding between lifts)
that cannot be duplicated with laboratory compaction.
Large-scale in situ hydraulic conductivity testing of field-
compacted soil is recommended for a test pad; the pur-
pose of such tests is to verify that hydraulic conductivity
objectives can be met on the field scale. Test pads are
discussed further later in this chapter.
Table 6-3 lists the tests and observations recommended
for ensuring that the soil is properly compacted. Gordon
et al. provides the basis for the recommended frequency
of testing (3). Periodic calibration checks are recom-
mended for tests that are performed with microwave
ovens, nuclear devices, drive rings, or other equipment
that may introduce a small, systematic bias in the test
results. Systematic measurement errors, especially for
water content, must be documented.
In Table 6-3, it is recommended that observations of the
number of passes of the compaction equipment be peri-
odically determined. The compactive energy delivered to
low hydraulic conductivity materials has a large influence
on the hydraulic conductivity of the materials after com-
paction. If too little compactive energy is delivered to the
soil, e.g., because too few passes of the compactor are
made over the soil, then the hydraulic conductivity may
not be as low as desired.
Some individuals place great emphasis on hydraulic con-
ductivity tests performed on "undisturbed" samples taken
from a compacted lift of low hydraulic conductivity
material. One sampling procedure is to push a thin-
walled sampling tube (sometimes called a "Shelby tube")
into the soil with a backhoe, as shown in Figure 6-11.
However, with this procedure, the sampling tube usually
rotates during the push (Figure 6-12), which leads to un-
acceptable disturbance of the soil sample. This sampling
procedure is strongly discouraged. A better procedure is
to use a thin-walled sampling tube that is only about 23
cm (9 in.) long. (The tube should never be pushed more
than about 23 cm [9 in.] into the soil because stiff, com-
pacted soils usually plug the sampling tube if a longer
push is used.) As shown in Figure 6-13, a hydraulic jack
is placed on top of a sampling tube. The jack is used to
push the sampling tube straight into the soil. A backhoe
can be used, but only as a reaction for the hydraulic jack
as shown in Figure 6-14. Sampling procedures described
in ASTM Practice D1587 should be followed. Procedures
for preserving and transporting samples should be in ac-
cord with ASTM Practice D4220.
There are several potential problems with laboratory
hydraulic conductivity tests performed on "undisturbed"
samples of the liner material for CQA purposes:
• If there are any rocks or stones in the soil, it may be
virtually impossible to obtain a representative sample
for testing. When stones are present, the sampling
tube drags the stones through part of the soil sample,
which damages the sample. Many samples may have
to be taken and discarded before a sample that does
not contain too many stones is obtained. However, the
value of testing a sample that contains almost no
stones, when most of the soil does contain stones, is
questionable.
• Small samples of soil may not be representative. If
there are cracks, zones of poor compaction, or other
hydraulic defects, the chances that an occasional
small sample will detect those defects are remote. Just
because laboratory hydraulic conductivities are low
54
-------
I
g>
'CD
Overall Acceptable
Zone
Acceptable Zone
Based on Hydraulic
Conductivity
Acceptable Zone
Based on Shear
Strength
Molding Water Content
Figure 6-6. Use of hydraulic conductivity and shear strength data to define a single, overall acceptable zone (5).
does not necessarily mean that the field values are
also low.
• Laboratory hydraulic conductivity tests take from 1 day
to 1 week to complete. The value of this type of test for
CQ purposes is minimized by the long time required to
obtain results. If the completed lift is left exposed while
the CQA team awaits the results of hydraulic conduc-
tivity tests, the whole process may be counterproduc-
tive.
There is no widespread agreement about how CQA offi-
cials should deal with the problems noted above for
laboratory hydraulic conductivity tests. The problems are
noted for the reader's information, and it is hoped that
CQA planners will develop strategies for dealing with
these difficulties.
Finally, any holes made in the soil must be sealed.
Quality assurance personnel should visually inspect the
sealing of some of the holes made for QC testing.
PROTECTION OF A COMPLETED LIFT
Visual observations are recommended to determine if
adequate measures have been taken to protect each lift
of soil from desiccation, freezing, or other damaging for-
ces. Additional tests, e.g., water content tests, should be
required if there is any question that the soil may have
been damaged after compaction.
55
-------
o
Q.
CD
Q
Q.
gj
"(D
Q
140
120
100
80
60
(A)
Zero Air Voids
(S=100%)
S=40%
0
10
20
30
40
Molding Water Content (%)
140
120
100
80
60
(B)
Zero Air Voids
Line of Optimums
0 10 20 30 40
Molding Water Content (%)
Figure 6-7. Possible approaches for specifying lower limit of Acceptable Zone: (A) minimum degree of saturation, S; and
(B) line of optimums (5).
56
-------
g>
'CD
+->
'c
=)
130
120
D Red. Proctor
O Std. Proctor
A Mod. Proctor
Molding Water Content (%)
Figure 6-8. Compaction curves for Type A soil from East Borrow area at Oak Ridge Y-12 operations project (5).
Molding Water Content (%
East Borrow Area
Type A Soil
D Red. Proctor
O Std. Proctor
A Mod. Proctor
Figure 6-9. Hydraulic conductivity versus molding water content for Type A soil from East Borrow area at Oak Ridge Y-12
operations project (5).
57
-------
Q
Q.
Q
c
=)
Acceptable Zone
D Red. Proctor
O Std. Proctor
A Mod. Proctor
Molding Water Content (%)
Figure 6-10. Acceptable zone for Type A soil from East Borrow area at Oak Ridge Y-12 operations project (5).
SAMPLING PATTERN
The CQA plan should detail the procedures for selecting
locations where samples will be taken. A random pattern,
or a sampling pattern utilizing a grid system, is recom-
mended. However, the CQA personnel should have the
authority to request additional tests at any questionable
location.
TEST PADS
Occasionally, test pads are constructed for the purpose
of verifying that materials and methods of compaction will
achieve the desired results, e.g., low in-field hydraulic
conductivity. If a test pad yields adequate results, the
CQA team should utilize the test pad in the CQA
program. One of the purposes of QC testing and QA ob-
servations should be to ensure that the actual cover is
built to standards that equal or exceed those used in
building the test pad.
If a test pad is built, it may be desirable to have a proce-
dural construction specification. The test pad is used to
demonstrate that the construction procedure is ap-
propriate. If the construction procedure is specified, a
critical objective of the CQA program should be to make
an adequate number of observations to verify that the
specified procedure has been followed.
One or more in situ hydraulic conductivity tests are usual-
ly performed on the test pad. Testing procedures are dis-
cussed by Daniel (7) and Sai and Anderson (8). The
most widely used method of measurement is the sealed
double ring infiltrometer (SDRI). Testing procedures for
the SDRI are given in ASTM Method D5093. A bevy of
other tests (water content, dry unit weight, Atterberg
limits, etc.) is usually part of the testing protocol, as well,
to provide full documentation of the test pad.
OUTLIERS
Soils are variable materials. It is inevitable that the soil
materials will fail to meet specifications at some points in
the soil mass. If all QC tests pass, this does not mean
that the soil everywhere meets the project specifications;
it just means that enough tests were not performed to lo-
cate the occasional outlier.
Occasional outliers are not necessarily harmful. For bar-
rier layers, the soils are constructed in multiple lifts in part
as a result of recognition that soils are variable and that
the compaction process is not perfect. For drainage
layers, occasional pockets of low hydraulic conductivity
materials will not harm the performance of reasonably
thick drainage layers—permeating fluids will simply go
around the low hydraulic conductivity material.
58
-------
Figure 6-11. Pushing of thin-walled sampling tube with a backhoe.
Figure 6-12. Tilting of sampling tube during push.
59
-------
Figure 6-13. Placement of hydraulic Jack on top of sampling tube.
Figure 6-14. Use of backhoe as a reaction for hydraulic jack.
60
-------
At least two approaches may be utilized for dealing with
outliers:
* The usual procedure is not to allow any outliers.
However, for the reasons noted above, this approach
is not realistic and frequently causes some manipula-
tion of tests or test results to ensure that unrealistic
specifications are met. In addition, the CQA team at
the end of the project may be left trying to justify the in-
significance of the occasional outlier (after the con-
struction is complete).
• One possible solution is to permit an occasional outlier.
Usually, if a small percentage of tests fail, the effect of
the outliers is nil. Such a specification is more realistic
and tends to discourage manipulation of sampling
locations, tests, or test results to meet unrealistic
specifications.
If a test does fail and it is believed that the failure repre-
sents inadequate materials or inadequate construction
procedures, the extent of the failed area must be defined
and the area must be repaired. The specifications should
prescribe how the area to be repaired will be determined.
SUMMARY
Figure 6-15 is a checklist of critical parameters for low
hydraulic conductivity barrier materials. A checklist for
drainage materials is shown in Figure 6-16. The items
shown on these checklists are intended to ensure that the
materials of construction are adequate and that ap-
propriate methods of construction have been utilized. The
CQA process involves a combination of QC testing and
observation by qualified personnel.
REFERENCES
1. U.S. EPA. 1986. Technical guidance document: con-
struction quality assurance for hazardous waste land
disposal facilities. Office of Solid Waste and Emer-
gency Response, Washington, DC, EPA/530-SW-86-
031.
2. Daniel, D.E. 1990. Summary review of construction
quality control for compacted soil liners. In: R.
Bonaparte, ed. Proceedings, Waste Containment
Systems: Construction, Regulation, and Perfor-
mance. New York: American Society of Civil En-
gineers, pp. 175-189.
3. Gordon, M.E., P.M. Huebner, and P. Kmet. 1984. An
evaluation of the performance of four clay-lined
landfills in Wisconsin. In: Proceedings, Seventh An-
nual Madison Waste Conference on Municipal and In-
dustrial Waste, Madison, Wl. pp. 399-460.
4. U.S. EPA. 1988. Design, construction, and evaluation
of clay liners for waste management facilities.
Washington, DC. EPA/530/SW-86/007F.
5. Daniel, D.E. and C.H. Benson. 1990. Water content-
density criteria for compacted soil liners. Journal of
Geotechnical Engineering, Vol. 116, No. 12, pp.
1811-1830.
6. Mitchell, J.K., D.R. Hooper, and R.G. Campanella.
1965. Permeability of compacted clay. Journal of the
Soil Mechanics and Foundations Division ASCE, Vol.
91, No. 4, pp. 41-65.
7. Daniel, D.E. 1989. In situ hydraulic conductivity tests
for compacted clay. Journal of Geotechnical En-
gineering. Vol. 115, No. 9,1205-1226.
8. Sai, J.O. and D.C. Anderson. 1990. Field hydraulic
conductivity tests for compacted soil liners. Geotech-
nical Testing Journal. Vol. 13, No. 3, pp. 215-225.
61
-------
CRITICAL VARIABLES FOR LOW HYDRAULIC CONDUCTIVITY LAYER
Material:
Minimum Liquid Limit =
Minimum Plasticity Index =
Maximum Particle Size =
Maximum Percentage of Gravel =
Minimum Percentage of Fines =
Water Content/Density Defined (Y/N)
Maximum Clod Size =
Lifts:
Scarify Surface Before Placing (Y/N)
Maximum Loose Lift Thickness =
Maximum Completed Lift Thickness =
Compactor:
Minimum Weight =
Type of Roller Drum
Compaction:
Minimum Number of Passes
Protection:
Protection from Dessication & Freezing (Y/N)
Figure 6-15. Checklist of critical variables for CQA of low hydraulic conductivity compacted soil used in a cover system.
62
-------
CRITICAL VARIABLES FOR DRAINAGE LAYER
Material:
Maximum Percentage of Fines =
Maximum Particle Size =
Lifts:
Maximum Loose Lift Thickness
Compactor:
Maximum Weight =
Type of Roller Drum
Compaction:
Desirable Number of Passes
Grinding of Soil:
Visual inspection? (Y/N)
Protection:
Protection from Contamination by Fines? (Y/N)
Figure 6-16. Checklist of critical variables for CQA of drainage materials used in a cover system.
63
-------
CHAPTER 7
CONSTRUCTION QUALITY CONTROL FOR GEOMEMBRANES
PRELIMINARY DETAILS
A number of preliminary steps must be taken to ensure
that optimal construction quality control (CQC) and con-
struction quality assurance (CQA) can be achieved at the
landfill site. These steps involve manufacture, fabrication,
storage at the factory, shipment, and storage at the site
of the geomembrane. This chapter will focus on
CQC/CQA of geomembranes, but all geosynthetics can
be viewed in a similar manner.
Manufacture
The first consideration in quality control is that the
geomembrane resin, and its entire formulation, must be
appropriate to the site where it will be installed. Ap-
propriateness can be evaluated using EPA 9090 chemi-
cal compatibility testing, or by direct comparison to a
local, state, or federal specification, or to a stand-
ardization group. The material may have to be chemically
"fingerprinted" in some situations to assure that the
delivered geomembrane has identical characteristics to
the approved test samples.
The thickness, width, and length of the geomembrane
also must be verified. This is best done at the
manufacturer's plant, since shipment costs are high and
receiving the wrong geomembrane at the job site can
result in uncomfortable arguments and inconvenient
delays.
Other details such as the diameter and strength of the
windup core, protective covering (if appropriate), and
proper marking and identification need to be assured.
Fabrication of Panels
For certain types of geomembranes, such as polyvinyl
chloride (PVC), chlorylsulfonated polyethylene-reinforced
(CSPE-R), and ethylene interpolymer alloy-reinforced
(EIA-R), the relatively narrow rolls (about 1.8-m [6-ft]
wide) are fabricated into wider panels of about four to six
roll widths. Panel fabrication requires factory seaming,
usually either dielectric or bodied solvent. There should
be a thin sheet of plastic over the seams to prevent one
layer from sticking to the next.
The completed panels are accordion folded in two direc-
tions, wrapped in a heavy cardboard box, and placed on
wooden pallets for shipment. Proper marking and iden-
tification at this stage is necessary to ensure proper
delivery.
Storage at Factory
Geomembranes stored at the manufacturer's or
fabricator's facility should be elevated off the ground.
They also should not be stacked so high as to deform the
lower rolls or layers; warm climates are particularly
troublesome in this regard. An enclosed storage space is
recommended.
Shipment
Rolls or pallets of geomembranes are usually shipped by
truck to sites in the contiguous 48 states. When shipped
in closed trailers, the geomembranes should be loaded
and unloaded by lifting rather than by pushing and pull-
ing. Front-end loaders equipped with long rods (called
"stingers") are used for rolled geomembranes and forklift
loaders are used for palletized geomembranes.
In cases where stacking of the geomembranes might be
of concern, the delivery trailer should be inspected at the
job site for squashed rolls or crushed boxes.
Storage at Site
Unless the geomembrane is used directly as it comes off
the shipping trailer, a safe storage area should be
provided. The rolls of geomembrane should be elevated
off the ground or at least placed on a dry soil area that
does not contain vegetation, stumps, or other sharp ob-
jects. Covering is usually not necessary providing the
geomembranes are installed within a short period of time.
Palletized geomembranes should also be stored on site
on dry, level ground with similar considerations.
When the geomembranes are to be stored on the site for
months or longer, they should be covered and/or have an
enclosure around them for protection.
SUBGRADE PREPARATION
The subgrade must be prepared according to the site-
specific plans and specifications. Thus, line and grade
must have been established and verified before any
geomembrane is brought into the facility and positioned.
There can be no sharp objects of any kind such as grade
stakes, tools, stones, or equipment beneath the
geomembrane.
65
-------
Ruts caused by the compaction equipment or by the
geomembrane placement equipment must be leveled by
hand. Ruts are particularly troublesome if they freeze in
their uneven profile. They must be leveled before the
geomembrane is placed by waiting until the ground thaws
or by breaking the uneven surfaces using pneumatic clay
spades and pavement breakers.
Geomembranes should never be placed in ponded water.
Such a procedure is indicative of a poor sequence of
construction operations. Seaming can never be ac-
complished under such conditions.
DEPLOYMENT OF THE GEOMEMBRANE
The geomembrane should be placed on the entire facility
in accordance with a predetermined roll or panel layout.
Layout is a site-specific consideration, but plans are
generally supplied by the geomembrane manufacturer,
fabricator, or installation firm. Usually the rolls or panels
are ordered in a particular direction.
The construction deployment equipment should be low
ground pressure units in comparison to the subgrade
stability. For landfill covers, this is sometimes difficult to
achieve because the waste beneath the construction
operations can be actively subsiding.
After a roll, or panel, is initially positioned or "spotted," it
usually must be shifted slightly for exact positioning. By
lifting the liner up and allowing air to get beneath some of
it, the liner can sometimes be "floated" into position. If
this is not possible (e.g., with thick geomembrane
sheets), the liner has to be shifted by dragging it along
the subgrade (or on the geosynthetic material beneath it).
The entire roll or panel must then be inspected for
blemishes, scratches, and imperfections. Finally, the roll
or panel is weighted down with sandbags to prevent
movement by wind or any other disturbance.
Complete rolls and panels have been captured by gusts
of wind and unceremoniously dumped in a corner of the
facility. The owner/operator should decide beforehand,
i.e., during preconstruction meetings, whether a liner in
this situation can be used again or not. A number of
damage scenarios should be described to anticipate such
problems, since they are not that uncommon.
GEOMEMBRANE FIELD SEAMS
There are many types of geomembrane seams, most of
which were developed for a particular type of
geomembrane. Table 7-1 shows methods of field seam-
ing currently in use.
Solvent Seams
Solvent seams use a liquid solvent placed by a squeeze
bottle between the two geomembrane sheets to be
joined, followed by pressure to make complete contact.
As with any of the solvent-seaming processes described
in this section, a portion of the two adjacent
geomembranes is actually dissolved, resulting in both li-
quid and gaseous phases. Too much solvent will weaken
the adjoining geomembrane, and too little solvent will
result in a weak seam. Therefore, great care is required
in providing the proper amount of solvent for the par-
ticular type and thickness of geomembrane. Care must
also be exercised in allowing the proper amount of time
to elapse before contacting the two surfaces, and in ap-
plying the proper pressure and duration of rolling. These
seams are used primarily on flexible thermoplastic
materials.
Bodied solvent seams are similar except that 8 to 12 per-
cent of the parent lining material is dissolved in the sol-
vent before the seam is made. The purpose being to
compensate for the lost material while the seam is in a li-
quid state and to create a viscous liquid that can be
brushed on the area to be bonded. Pressure is applied,
and heat guns or radiant heaters are used to aid the
process.
A solvent adhesive uses an adherent left after the solvent
dissipates. The adhesive thus becomes an additional ele-
ment in the system. Sufficient pressure must be used to
affect an adequate seam. Most thermoplastic materials
can be seamed in this manner.
Contact adhesives have a wide applicability to most
geomembrane types. The solution is applied to both
mating surfaces by brush or roller. After reaching the
proper degree of tackiness, the two sheets are placed on
top of one another, and pressure is applied by a roller.
The adhesive forms the bond and is an additional ele-
ment in the system.
Vulcanizing tapes and adhesives are used on very dense
thermoset materials such as butyl and EPDM. In this
process, an uncured tape or adhesive containing the
polymer base of the geomembrane and crosslinking
agents are placed between the two sheets. Upon applica-
tion of heat and pressure, crosslinking occurs, which
gives the necessary bond. Factory seams are made in
large vulcanizing presses or autoclaves, while field
seams require a portable machine to provide the neces-
sary heat and pressure. Since thermoset geomembranes
are rarely used in landfill covers, the technique is not par-
ticularly important for our purposes.
Thermal Seams
There are a number of thermal methods that can be used
on thermoplastic geomembrane materials. In all of them,
the opposing geomembrane surfaces are truly melted
into a liquid state. Temperature, time, and pressure all
play important roles: too much melting weakens the
geomembrane and too little melting results in a weak
seam. The same care as is necessary for solvent seams
must be taken with thermal seams.
Hot air seaming uses a machine consisting of a resis-
tance heater, a blower, and temperature controls to blow
66
-------
Table 7-1. Overview of Geomembrane Field Seams.
Method
Seam
Configuration
Typical Rate* Comments
Solvent
200 ft./hr. Requires tack time
Requires hand rolling
Requires cure time
Bodied
Solvent
150 ft./hr. Requires tack time
Requires hand rolling
Requires cure time
Solvent
Adhesive
150 ft./hr. Requires tack time
Requires hand rolling
Requires cure time
Hot Air
50 ft./hr. Good to tack sheets together
Hand held and automated devices
Air temperature fluctuates greatly
No grinding necessary
Hot Wedge
300 ft./hr. Single and double tracks available
Built in nondestructive test
Cannot be used for close details
Highly automated machine
No grinding necessary
Controlled pressure for squeeze-out
Ultrasonic
300 fL/hr. New technique for geomembranes
Sparse experience in the field
Capable of full automation
No grinding necessary
Fillet Extrusion
100 ft./hr. Upper and lower sheets must be ground
Upper sheet must be beveled
Height and location are hand-controlled
Can be rod or pellet fed
Extrudate must use same polymer
compound
Air heater can preheat sheet
Routinely used for difficult details
Flat Extrusion
50 ft./hr. Highly automated machine
Difficult for side slopes
Cannot be used for close details
Extrudate must use same polymer
compound
Air heater or hot wedge can preheat sheet
*m = ft x .3048
67
-------
air between two sheets to actually melt the opposing sur-
faces. Usually, temperatures greater than 260°C (500°F)
are required. Immediately following the melting of the sur-
faces, pressure is applied by rollers. For some devices,
pressure application is automated by counter-rotating
knurled rollers.
in the hot wedge or hot knife method, an electrically
heated resistance element in the shape of a wedge is
passed between the two sheets to be sealed. As it melts
the opposing surfaces, roller pressure is applied. Most of
these seaming units are automated in terms of tempera-
ture, speed of travel, and amount of pressure applied. An
interesting variation of the technique is the dual-hot-
wedge method, which forms two parallel seams with an
unbonded space between them. This space is sub-
sequently pressurized with air and any lowering of pres-
sure signifies a leak in the seam. Lengths of hundreds of
feet can be field tested in this one step. The hot wedge or
hot knife method will be more fully described in the sec-
tion on nondestructive seam testing.
Dielectric bonding is used only for factory seams of
flexible thermoplastic geomembranes. In this method, an
alternating current, at a frequency of approximately 27
MHz, excites the polymer molecules, creating friction and
thereby generating heat. This heat melts the polymer,
and when followed by pressure, results in a seam. A
variation of this method, called ultrasonic seaming, has
recently been introduced for the manufacture of field
seams on polyethylene liners.
Ultrasonic bonding utilizes a generated wave form of 40
kHz, which produces a mechanical agitation of the op-
posing geomembrane surfaces. Following the melting
process, a set of knurled wheels is used to mix and apply
pressure to the material.
Electric welding is yet another new technique for
polyethylene seaming. In this technique, a stainless steel
wire is placed between overlapping geomembranes and
is energized with approximately 36 volts and 10 to 25
amps current. The hot wire radially melts the entire
region within about 60 seconds, thereby creating a bond.
It is later used as a nondestructive testing method with a
low current and a questioning wire outside of the seamed
region.
Extrusion Seams
Extrusion for fusion) welding is used exclusively on
polyethylene geomembranes. It is directly parallel to
metallurgical welding in that a ribbon of molten polymer is
extruded between or against the two slightly buffed sur-
faces to be joined. The extrudate ribbon causes some of
the sheet material to be liquefied and the entire mass
then fuses together. One patented system has a mixer in
the molten zone that aids in homogenizing the extrudate
and the molten surfaces. The technique is called flat
welding when the extrudate is placed between the two
sheets to be joined and fillet welding when the extrudate
is placed over the leading edge of the seam to be
bonded.
DESTRUCTIVE SEAM TESTS
After a field-seaming crew has seamed a given amount
of material, it is important to evaluate their performance.
One procedure is to cut out a sample, send it to a
laboratory, and test it until failure in either shear or peel
modes (see Figure 7-1). Another option would be to test
it directly at the field site. But considering a
geomembrane sheet layout, where should the seam be
tested and in how many places? Because each seam
sample becomes a hole that must be appropriately
patched and then retested, the number of field-seam
samples is commonly reduced to a bare minimum. Then
only the method of seaming is assessed, not its con-
tinuity. The method includes installation type, tempera-
ture, dwell time (time during which seam is under
pressure), pressure, and other operational details affect-
ing seam quality. Samples will ordinarily be taken at the
start of the seaming operations in the morning and after
the midday break. Thereafter, sampling can be done on a
random or a periodic basis. Haxo (1) recommends a fre-
quency of six samples per km (6/3,300 ft) of seam on a
random basis, or one sample per 150 m (1/500 ft) of
seam on a uniform basis.
There is much current discussion on what constitutes an
acceptable seam. Nearly everyone agrees that the seam
test specimen must not fail within the seamed region it-
self; that is, a failure must be a sheet failure on either
side of the seamed region. This is called a "film-tear
bond" failure. Engineers are not in agreement, however,
as to the magnitude of the force required for failure. For
SHEAR TEST
t
'(
VisSiS&ffcXfiiiifSM
PEEL TEST
Figure 7-1. Shear and peel test for geomembrane seams.
68
-------
seams tested in a shear mode, failure forces of 80 to 100
percent of the unseamed sheet strength are usually
specified. For seams tested in a peel mode, failure forces
of 50 to 80 percent of the unseamed sheet strength are
often specified. These percentages underscore the
severity of peel tests as compared to shear tests. For as-
sessing seam quality, the peel test is preferable.
NONDESTRUCTIVE SEAM TESTS
Although the method of seaming must be properly as-
sessed, the test tells nothing of the continuity and com-
pleteness of the entire seam. It does little good if one
section of a seam has 100 percent of the strength of the
parent material, if the section next to it is missed com-
pletely by the field-seaming crew. Thus, this section dis-
cusses only continuous methods of nondestructive
testing (NOT). In each of these methods, the goal is to
check 100 percent of all seams (see Table 7-2).
The air lance method projects a jet of air at approximately
350 kPa (50 Ib/in.2) pressure through an orifice of 5-mm
(3/16-in.) diameter. The jet is directed beneath the upper
edge of the overlapped seam to detect unbonded areas.
When such an area is located, the air passes through,
causing an inflation and fluttering in the localized area.
This method only works on relatively thin (less than 45
mils [1.1 mm]), flexible geomembranes, and only if the
defect is open at the front edge of the seam, where the
air jet is directed. It is strictly a contractor/installer's tool
to be used in a CQC manner.
In the mechanical point stress or "pick" test, the tester
places a dull tool (such as a blunt screwdriver) under the
top edge of a seam. With care, an individual can detect
an unbonded area, because it is easier to lift than a
properly bonded area. This rapid test depends complete-
ly on the care and sensitivity of the person performing it.
Only relatively thick, stiff, geomembranes are checked by
this method. Detectability is similar to that using the air
lance, but both methods are very operator dependent.
This test also is to be performed only by the installation
contractor and/or geomembrane manufacturer. Design or
inspection engineers should use one or more of the tech-
niques discussed below.
Electric sparking is an old technique used to detect pin-
holes in thermoplastic liners. In this method, a high-vol-
tage (15 to 30 kV) current detects any leakage to ground
(through an unbonded area) by producing sparking. The
method is not very sensitive to overlapped seams of the
type generally used in geomembranes and is used only
rarely for this purpose. Today, the technique has been
revived in a somewhat varied form. In the electric wire
method, a copper or stainless steel wire is placed
between the overlapped geomembrane region and is ac-
tually embedded into the completed seam. After seam-
ing, a charged probe of about 20,000 volts is connected
to one end of the wire and slowly moved over the length
of the seam. A seam defect between the probe and the
embedded wire produces an audible alarm from the unit.
The method is strongly advocated by some installation
Table 7-2. Overview of Nondestructive Seam Tests.
Primary User
General Comments
Nondestructive
Test Method
air lance
pick test
electric wire
dual seam
(positive
pressure)
vacuum chamber
(negative
pressure)
ultrasonic pulse
echo
ultrasonic
impedance
ultrasonic shadow
electric field
acoustic sensing
Contractor
yes
yes
yes
yes
yes
-
yes
yes
Design Engineer
Inspector
yes
yes
yes
yes
yes
yes
yes
yes
Third-Party
Inspector
-
-
yes
yes
yes
yes
yes
Cost of
equipment
$200
nil
$500
$200
$1000
$5000
$7000
$5000
$20,000
$1000
Speed of
tests
fast
fast
fast
fast
slow
moderate
moderate
moderate
slow
fast
Cost of
tests
nil
nil
nil
mod.
very high
high
high
high
high
nil
Type of
Result
yes-no
yes-no
yes-no
yes-no
yes-no
yes-no
qualitative
qualitative
yes-no
automatic
yes-no
Recording
Method
manual
manual
manual
manual
manual
automatic
automatic
automatic
manual and
manual
Operator
Dependency
very high
very high
high
low
high
moderate
unknown
moderate
low
moderate
69
-------
firms, but the literature gives conflicting opinions when
comparing this method to vacuum box testing (discussed
below).
The pressurized dual seam method was mentioned ear-
lier in connection with the double-wedge thermal seaming
method. The air channel that results between the double
seam is inflated using a hypodermic needle and pres-
surized to approximately 200 kPa (30 Ib/in.2) for a length
of 30 to 300 m (100 to 1,000 ft). If no drop on a pressure
gauge occurs, the seam is acceptable; if a drop occurs, a
number of actions can be taken:
• The distance can be systematically halved until the
leak is located.
• The section can be tested by some other leak detec-
tion method.
• A cap strip can be seamed over the entire edge.
The test is an excellent one for long, straight-seam runs.
It is generally performed by the installation contractor, but
often with the designer or inspector observing the proce-
dure and assessing the results.
Vacuum chambers (boxes) are the most common form of
nondestructive test currently used by design engineers
and CQA inspectors. In the vacuum chambers method, a
1-m (3-fl) long box with a transparent top is placed over
the seam and a vacuum of approximately 17 kPa (2.5
Ib/in.2) is applied. When a leak is detected, the soapy
solution originally placed over the seam bubbles, thereby
reducing the vacuum. The vacuum is reduced due to air
entering from beneath the liner and passing through the
unbonded zone. The test is slow to perform and it is often
difficult to make a vacuum-tight joint at the bottom of the
box where the box passes over the seam edges. Due to
upward deformations of the liner into the vacuum box,
only geomembrane thicknesses greater than 30 mils
(0.75 mm) should be tested in this manner. It would be
difficult to test 100 percent of the field seams by this
method, however, because of the large number of field
seams and the amount of time required. The test could
also not inspect around sumps, anchor trenches, and
patches with any degree of assurance. The method is
also essentially impossible to use on side slopes, since
the downward pressure required to make a good seal
cannot be obtained (it is usually done by standing on top
of the box).
A number of ultrasonic methods are available for seam
testing and evaluation. The ultrasonic pulse echo method
is basically a thickness measurement technique and is
only used with semicrystalline geomembranes. In this
method, a high-frequency pulse is sent into the upper
geomembrane and (in the case of a good seam) reflects
off of the bottom of the lower one. If, however, an un-
bonded area is present, the reflection will occur at the un-
bonded interface. The use of two transducers, a pulse
generator, and a CRT monitor are required. The
ultrasonic pulse echo test cannot be used for extrusion
fillet seams, because of their nonuniform thickness.
The ultrasonic impedance plane method works on the
principle of acoustic impedance. A continuous wave of
160 to 185 kHz is sent through the seamed liner, and a
characteristic dot pattern is displayed on a CRT screen.
The dot pattern is calibrated to signify a good seam. The
method has potential for all types of geomembranes but
still needs additional development work.
The ultrasonic shadow method uses two roller
transducers: one sends a signal into the upper
geomembrane and the other receives the signal from the
lower geomembrane on the other side of the seam. The
technique shows an energy transmitted on the display
monitor. HOPE seams with received signals greater than
50 percent full-scale height were all found to be accept-
able by subsequent testing with destructive methods.
Received signals less than 20 percent full-scale height in-
dicated that the seams were not acceptable. The 50 to 20
percent range had mixed results. This technique can be
used for all types of seams, even those in difficult loca-
tions, such as around manholes, sumps, and appur-
tenances. It is best suited to semicrystalline
geomembranes, such as HOPE, and will not work for
scrim-reinforced liners.
The electric field test utilizes a liquid-covered
geomembrane to contain an electric field. For this
method to work, the entire bottom of the lined facility
must be covered with liquid, usually water. The depth can
be nominal, approximately 15 cm (6 in.). Electric field
testing cannot be used where water does not cover the
geomembrane, as on side slopes. This method uses a
current source to inject current across the boundary of
the liner. When a current is applied between the source
and remote current return electrodes, current flows either
around the entire site (if no leak is present) or bypasses
the longer travel path through the leak itself (when one is
present). Potentials measured on the surface are af-
fected by the distributions and can be used to locate the
source of the leak. These potentials are measured by
"walking" a probe in the water. The operator walks on a
predetermined grid layout and marks where anomalies
exist. After the survey is completed, these anomalies can
be rechecked by other methods, such as the vacuum
box. Electric field testing is currently commercially avail-
able.
Acoustic sensing is used in conjunction with the dual
seam air pressure test. It is an effective method to locate
a leak if air pressure is not maintained.
PENETRATIONS, APPURTENANCES, AND
MISCELLANEOUS DETAILS
The various details of a geomembrane landfill closure are
very important in making the entire "system" function as
intended. Clearly, gas venting must be addressed in any
70
-------
closure over biodegrading solid waste landfills. Gas vent- always keep in mind that the waste will subside over time
ing pipes are usually constructed with prefabricated and typically in a very nonuniform and random manner.
"boots" around PVC or HOPE pipes. The geomembrane
is then seamed to the pipe according to the proper techni- REFERENCE
que. If metal pipes are used, a stainless steel clamp -, Haxo HEJ 1986 Qua|jty assurance of
usually makes the connection. geom'embranes used as liners for hazardous waste
Other appurtenances and details might need to be con- containment. Journal of Geotextiles and
sidered and details in the open literature should be Geomembranes. Vol. 3, No. 4, pp. 225-248.
evaluated accordingly. The landfill owner/operator must
71
-------
CHAPTERS
HYDROLOGIC EVALUATION OF LANDFILL PERFORMANCE (HELP) MODEL FOR DESIGN
AND EVALUATION OF LIQUIDS MANAGEMENT SYSTEMS
INTRODUCTION
Liquids management systems are critically important for
limiting leachate generation and migration. Cover sys-
tems control leachate generation by restricting infiltration
of precipitation into the waste layer. Leachate collection
and liner systems restrict migration of leachate from the
waste containment site by limiting leakage through liners
and promoting leachate collection. This chapter looks at
using the Hydrologic Evaluation of Landfill Performance
(HELP) model in the design and evaluation of these sys-
tems.
This chapter presents a brief overview of typical liquids
management systems, a detailed description of the HELP
model, and an example application of the HELP model
simulating a complete landfill system.
OVERVIEW
Landfills typically contain two liquids management sys-
tems. The cover is the principal liquids management sys-
tem for controlling leachate generation. Leachate, as
evaluated by the HELP model, is any rainfall or snowmelt
that combines with liquids in the waste and moves by
gravity forces to the bottom of a landfill. During its migra-
tion through the waste, the liquid takes on pollutants that
are characteristic of the waste. As such, the leachate
quantity and quality is site specific and waste specific.
The HELP model generates estimates of leachate quan-
tity given site-specific descriptions of climate, cover
design, and initial moisture content of the waste and soil
layers. The model does not predict leachate quality or
any contribution to the leachate quantity by subsurface
inflow of ground water. Good landfill design and site
selection would minimize any contribution from ground
water.
Covers
Figures 1-1 and 1-2 (see Chapter 1, pp. 1-3 and 1-7)
show typical cover designs recommended in the U.S.
EPA technical guidance document Final Covers on Haz-
ardous Waste Landfills and Surface Impoundments (1).
The designs are composed of three layers for liquids
management—vegetation/soil or cobbles/soil top layer,
drainage layer, and geomembrane liner/low hydraulic
conductivity soil layer (hydraulic barrier layer). The other
components in the design serve to support or maintain
the functions of these three layers.
The topsoil layer should be designed to promote runoff
from major storms, provide storage for evapotranspira-
tion, and protect the hydraulic barrier layer from frost
drainage layers
geomembrane anchors
, Uepafate anchor trench tor »ach geoiyntrntlc)
low-p«rm«ablHty toll
w»it» ' geomembrane
Figure 8-1, Cover and liner edge configuration with example toe drain.
73
-------
O O'BinagD Matnri.l O -*- D'"1" P10«»
(Not to Sold
Figure 8-2. Schematic of a single clay liner system for a landfill.
penetration and desiccation. Above all other considera-
tions, the topsoil, if vegetated, should have good mois-
ture retention properties, A clayey soil promotes runoff,
slows drainage, and provides storage for evapotranspira-
tion; however, it also can promote desiccation to greater
depths and retard vegetative growth.
Runoff promotes erosion and degradation of the cover
system. Vegetation or cobbles impede erosion and sup-
port the long-term integrity of the cover. The thickness of
the topsoil layer must be sufficient to retain adequate
moisture for maintaining vegetation and to ensure that
frost and desiccation will not penetrate to the hydraulic
barrier layer. Typically, a thickness of 60 to 90 cm (2 to 3
ft) is sufficient, but the actual requirement is site and
design specific.
The drainage layer should be designed to reduce
leakage through the hydraulic barrier layer. This is ac-
complished by draining the zone of saturation above the
barrier to a collection pipe or toe drain, as shown in
Figure 8-1 (2). This design lowers the hydraulic head
driving the leakage and decreases the quantity of water
available to leak through the barrier. All drainage layers
should be designed to have free drainage to a collection
system to maximize the drainage for a given system. Ab-
sence of a toe drain at the base of the side slope of the
cover system also can contribute to slope failures.
The drainage layer also reduces root and animal penetra-
tion of the hydraulic barrier layer and provides additional
depth and a capillary break to lessen desiccation and
frost penetration of the barrier. As shown in Figures 1-1
and 1-2, a filter, either soil or geotextile, must be placed
above the drainage layer to decrease migration of fines
and prevent the fines from clogging the drain. Drainage
layers are not necessary at all sites since some sites may
not have sufficient rainfall and infiltration to produce
standing head on the hydraulic barrier for long durations.
The hydraulic barrier layer or liner should be designed to
minimize the infiltration of water into the waste layer over
the long term. The liner systems shown in Figures 1-1
and 1-2 are both composite systems, but municipal waste
landfills have commonly used only a low hydraulic con-
ductivity soil liner. Use of a geomembrane in conjunction
with low hydraulic conductivity soil greatly improves the
effectiveness of the barrier. In conventional cover
designs, the barrier layer is the most important layer in
controlling leakage through the cover system. All other
layers tend to serve mainly as layers that support and
maintain the barrier. Generally, except in arid or semiarid
areas, the topsoil layer cannot promote sufficient runoff
and evapotranspiration to prevent leakage. Drainage
layers typically cannot drain all of the water passing
through the topsoil layer before it reaches the barrier.
Once water stands on the barrier, leakage occurs at a
rate controlled by the barrier.
Leachate Collection/Liner Systems
The second liquids management system used in a landfill
is the leachate collection/liner system. A typical single
liner system used for leachate collection is shown in
Figure 8-2 (2). It consists of a drainage layer overlain by
a filter, either soil or geosynthetic, and a hydraulic barrier
layer composed of hydraulic conductivity soil. The soil
liner is frequently overlain by a geosynthetic
geomembrane to greatly improve its performance. The
performance of these layers for liquids management is
the same as described above for cover systems except
that the layers are controlling leakage from the landfill in-
stead of infiltration into the waste layer and leachate
generation.
Figure 8-3 is a schematic of a typical double liner system
used both for leachate collection and leakage detection
(2). The top drainage layer and geomembrane is the
primary leachate collection system. The bottom drainage
74
-------
layer and composite liner (low hydraulic conductivity soil
overlain by a geomembrane) is the secondary leachate
collection system and leakage detection system.
HELP MODEL
Background
The HELP model was developed by the U.S. Army En-
gineer Waterways Experiment Station for the U.S. EPA
Office of Solid Waste (OSW) to provide technical support
for the Resource Conservation and Recovery Act
(RCRA) and Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA) programs.
Development of the model began in 1982 and Version 1
was released for public comment in June 1984 (3,4). The
program was a mainframe computer model that ran on
the National Computer Center's IBM system. In 1986, the
program was modified to run on IBM-compatible personal
computers. Additional capabilities and refinements were
included in Version 2 of the model released in 1988 (5,
6). The most current version is Version 2.05, which was
released in July 1989. Version 3 of the model is currently
in preparation for release in 1991.
The HELP model is a quasi-two-dimensional, gradually
varying, deterministic, computer-based water budget
model. It is termed quasi-two-dimensional because it
contains a one-dimensional vertical drainage model and
a one-dimensional lateral drainage model coupled at the
base of lateral drainage layers or the top of liners. The
program computes free vertical drainage down to the top
of a liner, at which point the liner restricts drainage and a
zone of saturation develops. The models for lateral
drainage and leakage or percolation through the liner
then use the height of saturated material above the liner
to compute simultaneously the rates of lateral drainage to
collection systems and vertical leakage through the liner,
respectively. The model is termed gradually varying be-
cause the simulation progresses through time using
analyses that are assumed steady for each time period.
Version 2 of the model uses a time period of 6 hours. The
model is deterministic and numerical as opposed to
stochastic and qualitative. Finally, the HELP model is a
computer-based water budget model; that is, the model
uses a computer to apportion the precipitation and initial
moisture content into estimates of the following water
budget components: surface runoff, evapotranspiration,
changes in snow storage, changes in moisture content,
lateral drainage collected in each drain system, and
leakage or percolation through each liner system. Figure
8-4 shows a schematic of the processes and systems
modeled by the program. Daily, monthly, annual, and
long-term average water budgets can be generated.
The HELP model is a tool developed specifically to aid
permit evaluators and landfill designers in the evaluation
and comparison of alternative landfill designs. The model
was built to evaluate whether alternative designs perform
as well as the minimum technical guidance systems over
a long period of time. Therefore, its primary utility is for
tasks involving comparison of alternative designs and
sensitivity of design parameters. A secondary goal of the
model's development was the accurate prediction of
water budget components. Nevertheless, additional test-
ing, verification studies, and refinements are ongoing to
improve its accuracy. In general, the accuracy and
precision of the model is limited by uncertainty and
variability in the properties of material existing in landfills.
As such, simulation results would be expected to be best
used to rate relative merits of designs rather than to ac-
curately predict the water budget components.
The HELP model is a valuable tool for design and permit
evaluation; however, it requires the user to exercise good
judgment. In particular, the user must have a good under-
standing of landfill design, vegetative systems, and the
model to obtain reliable results and correct conclusions.
The user must ensure the integrity of the design and the
data because the model does not evaluate the data.
Filter Medium
Top Liner Boiiom Composite
(geomembrane) Liner
r.r
R.r
Beir
Low P.rm»b.hly Soil
Secondary Leachate Native Soil Foundation
Collection and
Removal Sysiem
g Proposed as the
v - * i
^r??^?]
Leacnate
Collection
System
Sump
Upper
Component
(geomembrane)
Lower Component
(compacted soil)
Leak Detection System
(Not to Scale)
Figure 8-3. Schematic of a double liner and leak detection system for a landfill.
75
-------
RAINFALL/SNOW
INTERCEPTION .TRANSPIRATION
SNOW
EVAPORATION
SNOW
ACCUMULATION
RUNOFF
SNOW
MELT
VERTICAL PERCOLATION
INTERCEPTION
EVAPORATION
PLANT GROWTH
DEPTH OF
HEAD
BARRIER SOIL
PERCOLATION
Figure 8-4. Simulation processes in the HELP model.
An example of a cover system lacking design and data
integrity is a two-layer system with the following charac-
teristics. The top layer is 5 cm (2 in.) of topsoil vegetated
with a good stand of grass. The lower layer is 5 cm (2 in.)
of compacted clay having a saturated hydraulic conduc-
tivity of 10~8 cm/sec. With this description, the HELP
model would predict a very large quantity of runoff and a
small quantity of evapotranspiration since the topsoil
layer has very little storage capacity. In addition, very lit-
tle leakage through the clay liner would be predicted be-
cause the saturated hydraulic conductivity of the liner is
so low. The actual results would be expected to be quite
different. Construction of 5-cm (2-in.) layers is not prac-
ticable, especially, construction of a thin lift of clay com-
pacted uniformly to achieve an effective hydraulic
conductivity of 1CT8 cm/sec. Besides the difficulties in
construction, the layers would lack integrity to maintain
the described properties. Both layers would quickly form
desiccation cracks, producing much larger hydraulic con-
ductivities. As such, the leakage and evapotranspiration
would be much greater than predicted and the runoff
would be much less.
The primary anticipated use of the HELP model—to per-
form quick comparisons of the long-term performance of
alternative designs, often with very little data—was con-
sidered throughout its development. As such, the follow-
ing approach was taken to select simulation methods and
data entry options. Process simulation methods had to be
well-accepted techniques described in the literature that
are computationally efficient, require minimum data that
are readily available, and account for all major design
and climate conditions. Conservative assumptions are
made when necessary because of uncertainty. The term
"conservative" implies that any resulting error would tend
to result in an overestimation of vertical drainage or
leakage through liners. Options are provided to permit
use of data from a default data base or from user entry.
Guidance and recommendations are given for poorly un-
derstood parameters. In addition, the program is interac-
tive and user friendly and runs on IBM-compatible
personal computers to facilitate widespread use.
Process Simulation Methods
The HELP model was adapted from the Hydrologic
Simulation Model for Estimating Percolation at Solid
Waste Disposal Sites (HSSWDS) of the U.S. EPA (7, 8)
and the Chemical Runoff and Erosion from Agricultural
Management Systems (CREAMS) (9) and Simulator for
Water Resources in Rural Basins (SWRRB) models of
the U.S. Department of Agriculture (USDA) Agricultural
Research Service (ARS) (10). The following sections of
this chapter describe all of the principal hydrologic and
physical processes modeled by the HELP model, includ-
ing a discussion of the assumptions and limitations of the
models of the principal processes. Many of the processes
are shown on a schematic of a closed landfill profile in
Figure 8-4. Understanding of the processes, simulation
methods, and their assumptions and limitations is critical
for the proper application of the HELP model.
Infiltration
Daily infiltration into the landfill is determined indirectly
from a surface water balance. Infiltration equals the sum
of rainfall and snowmelt, minus the sum of runoff and sur-
face evaporation. Runoff and surface evaporation are in
part a function of interception. Precipitation on days
having a mean temperature below 0°C (32°F) is treated
76
-------
as snowfall and is added to the surface snow storage.
Decreases in snow storage occur by snowmelt and sur-
face evaporation.
Daily precipitation is an input parameter. Precipitation
data may be synthetically generated, specified by the
user, or selected from the default data base of historical
rainfall data. The synthetic weather generator will be
described later in this chapter.
Snowmelt is computed using a slightly modified version
of the simple degree-day method with 0°C (32°F) as the
base temperature (11). The in./degree-day snowmelt
constant was increased from 0,06 to 0.10, a value more
typical of open areas. In addition, the modification per-
mits a small quantity of snowmelt to occur at mean daily
temperatures between -5° and 0°C (23° and 32°F) to ac-
count for the variation in temperature during a day and for
the fact that landfills often have higher soil temperatures
because of heat generated from biodegradation. Snow-
melt contributes to runoff, evaporation, and infiltration.
Interception is modeled after the work of Morton (12). In-
terception approaches a maximum value exponentially as
the rainfall increases to about 0,5 cm (0.2 in.). The maxi-
mum interception is a function of the quantity of
aboveground biomass or leaf area index and is limited to
a maximum of 0.13 cm (0.05 in.). The interception
evaporates from the surface and decreases the evapora-
tive demand placed on the plants and soil column.
The HELP model uses the Soil Conservation Service
(SCS) curve number method for estimating surface
runoff, as presented in the Hydrology Section of the Na-
tional Engineering Handbook (11). The SCS curve num-
ber method is an empirical method developed for small
watersheds (about 12.1 to 202.4 hectares [30 to 500
acres]) with mild slopes (about 3 to 7 percent). The
method correlates daily runoff with daily rainfall for water-
sheds with a variety of soils, types of vegetation, land
management practices, and antecedent moisture condi-
tions (levels of prior rainfall). As applied, the technique
accounts for changes in runoff as a function of soil type,
soil moisture, and vegetative conditions. Version 3 of the
model will include a procedure to adjust the curve num-
ber as a function of surface slope since surface slopes
greater than 20 percent can produce significantly greater
runoff.
Many assumptions and limitations exist in the applica-
tion of this method in the HELP model, including the
following:
• The SCS curve number method is applicable for
landfills that are much smaller in area than water-
sheds. Verification studies have shown good agree-
ment between the predicted and observed cumulative
annual volume of runoff.
* Cumulative volume of runoff is independent of rainfall
duration and intensity since over a long simulation
period a variety of precipitation events will occur. The
predicted value represents an average of the
measured runoff for the typical variety of rainfall events
of a given quantity.
• No surface run-on from surrounding areas is permitted
by the model.
» Estimates of runoff greater than predicted by the SCS
curve number method are produced when the surface
soils are saturated or limit infiltration due to very low
hydraulic conductivity.
Evapotranspiration
Evapotranspiration consists of three components:
evaporation of water from the surface, from the soil, and
from the plants. Each component is computed separate-
ly. Evaporation of water from the surface is limited to the
smaller of the potential evapotranspiration and the sum of
the snow storage and interception. The HELP model
uses a modified Penman method to compute potential
evapotranspiration. This method, developed by Ritchie
(13), is also used in the CREAMS program (9). The
potential evapotranspiration is a function of ground cover,
daily temperature, and daily solar radiation. Evaporation
of surface water decreases the evaporative demand
placed on the plants and soil column.
The HELP model uses Ritchie's method of evaporation
from soil (13) as applied in the CREAMS (9) and SWRRB
(10) models. The method uses a two-stage, square root
of time routine. In stage one, the soil evaporation equals
the evaporative demand placed on the soil column.
Demand is based on energy and is equal to the potential
evapotranspiration discounted for surface evaporation
and shading from ground cover. A vegetative growth
model is used to compute the total quantity of vegetation,
both active and dormant, which provides shading. In
stage two, evaporation from the soil column is limited by
low soil moisture and low rates of water vapor transport
to the surface by soil suction. Stage two soil evaporation
is a function of the square root of the length of time that
the soil has been in this dry condition.
The HELP model estimates plant transpiration in the
manner of the CREAMS and SWRRB models (9, 10)
whereby the potential plant transpiration is a linear func-
tion of the potential evapotranspiration and the active leaf
area index. The leaf area index of actively transpiring
plants is computed using a vegetative growth model that
accounts for seasonal variation in active and dormant
aboveground biomass and leaf area index. This model
was extracted from the SWRRB model developed by the
USDA ARS (10). See "Vegetative Growth" later in this
chapter for a complete description of this model.
Many assumptions and limitations exist in the application
of this three-component evapotranspiration method in the
HELP model, including the following:
77
-------
• The potential evapotranspiration is a function solely of
the energy available at the surface and, therefore, is
not a function of energy produced in the landfill, soil
temperature, wind, and humidity. As such, the program
uses a vapor pressure gradient that is a function solely
of mean daily ambient air temperature.
• A constant value is used for the albedo (fraction of in-
cident solar radiation that is reflected). The value is
typical for brown soils and grasses and is modified
only when the surface is covered with snow.
• The program uses a constant evaporative zone depth.
This depth is the maximum depth to which soil suction
can draw water to the surface. The depth is a function
of soil properties, design, vegetation, and climatic con-
ditions.
• Ritchie's two-stage soil evaporation method is ap-
plicable for all materials, not just soils.
• Synthetically generated daily temperature and solar
radiation values are sufficient for estimating potential
evapotranspiration.
• The vegetative growth model produces representative
leaf area indices and biomass estimates that are suffi-
cient to estimate interception, surface shading, and
plant transpiration.
Subsurface Water Routing
Subsurface water routing processes modeled by the
HELP model include vertical unsaturated drainage, per-
colation through saturated soil liners, leakage through
geomembranes, and lateral saturated drainage. In model-
ing these processes, the soil moisture of each layer of
the landfill profile being modeled is computed by sequen-
tial analysis proceeding forward through time. The soil
moisture controls the rate of subsurface water movement
by each of the subsurface processes, but the rates of
water movement by these subsurface processes yield the
resulting soil moisture. Consequently, the soil moisture
and rates of subsurface water routing are computed
simultaneously in the HELP model by an iterative
process after accounting for extractions by soil evapora-
tion and plant transpiration.
The HELP model simulates unsaturated vertical drainage
using a unit hydraulic pressure gradient approach
(saturated Darcy's law) where drainage occurs at a rate
equal to the unsaturated hydraulic conductivity. Under
this approach, vertical water routing is only downward ex-
cept in the evaporative zone where water is removed up-
ward by evapotranspiration. The unsaturated hydraulic
conductivity is computed by the Campbell equation using
Brooks-Corey soil parameters to define the shape of this
power function (14, 15). This approach incorporates the
moisture retention properties (capillarity) of the soil in the
determination. The model considers limited interactions
between layers of materials. As such, the model does not
allow drainage from one layer at a rate greater than the
maximum infiltration rate of the layer below it, allowing
placement of a lower hydraulic conductivity nonliner layer
below a layer of higher hydraulic conductivity.
Future versions may consider soil matrix interactions in
the recommendation of values for the evaporative zone
depth and in the selection of the moisture content where
the drainage from one layer into another will cease.
These additions will better model the physical situations
where fine-grained materials overlie coarse-grained
materials. In these situations, the coarse-grained material
may restrict the depth of evapotranspiration and the fine-
grained material may retain a higher water content before
draining into the coarse-grained material despite very low
water content in the coarse-grained material. Conversely,
coarse-grained material overlying fine-grained material
will restrict the transport of water vapor up from below for
evaporation but will freely drain to very low moisture con-
tents. These phenomena occur because the soil suction
of fine-grained material is much greater than that of
coarse-grained material.
Vertical drainage through saturated soil liners is termed
percolation in the HELP model. The barrier soil liners are
assumed to remain permanently saturated, but percola-
tion occurs only when there is a zone of saturation direct-
ly above the liner. Percolation is computed by Darcy's
law using the saturated hydraulic conductivity of the liner
material. The head loss gradient is equal to the average
head above the base of the liner divided by the thickness
of the liner.
Leakage through geomembranes is modeled as a reduc-
tion of the cross-sectional area of flow through the sub-
soil below the geomembrane. The rate of flow through
the leaking subsoil is computed as the percolation rate
through a saturated barrier soil liner. This method
provides good results for composite liners but is not very
good for just a geomembrane. Therefore, Version 3 will
include an improved leakage model for geomembranes
based on the work of Brown (16) and Giroud et al. (17,
18,19).
The HELP model simulates lateral drainage using a
steady-state analytical approximation of the numerical
solution of the Boussinesq equation (Darcy's law for
saturated lateral flow through unconfined porous media
coupled with the continuity equation). The analytical ap-
proximation was developed by converting the Boussinesq
equation into a nondimensional form and solving it for
two analytical solutions at the extremes in nondimen-
sional average saturated depth. These two solutions
were then fitted with the same value and slope in an ap-
proximation that covers the rest of the range of non-
dimensional depths. The approximation matched the
numerical steady-state solution of the Boussinesq equa-
tion within 1 percent of the predicted drainage rate. The
solution is not linear; therefore, the HELP model uses a
Newton-Raphson method to converge onto the solution
78
-------
of the nonlinear approximation. The model uses the
average depth of saturation in the approximation since
the HELP model is quasi-two-dimensional and, therefore,
cannot determine the saturated depth profile. The lateral
drainage, percolation, and leakage through the
geomembrane are solved simultaneously with the
average depth of saturation using an implicit solution
technique.
Many assumptions and limitations exist in the application
of the subsurface water routing methods, including the
following:
• All flow is considered to follow Darcy's law. As such,
the only driving force for water routing is gravity, and
all movement has a downward component.
• All layers of materials are spatially homogeneous and
uniform. All properties of the layers and materials not
related to soil moisture are assumed to remain con-
stant throughout the simulation.
• No subsurface inflow occurs. The layers being
modeled are above the surrounding water table or cut
off from it.
• Brooks-Corey relationships and the Campbell equation
are applicable for estimating the unsaturated hydraulic
conductivity of all types of materials.
• Percolation and leakage through liner systems occur
only when a zone of saturation (head) lies on the top
surface of the liner. The zone covers the entire area of
the liner and, therefore, percolation and leakage are
spatially uniform.
• The liner and drainage layers cover the entire area of
the landfill since the water routing in the vertical and
lateral directions is performed separately by one-
dimensional models.
• The liner system is permanently saturated and the
pressure head at the base of the liner is zero (liner is
above the water table).
• Synthetic liners (geomembranes) are impermeable ex-
cept through specific failure points (holes, punctures,
cracks, faulty seams, etc.) and function by reducing
the area of flow through the subsoil beneath the liners.
• The rate of leakage through geomembranes is mainly
a function of the number of holes, depth of saturation
above the liner, and the saturated hydraulic conduc-
tivity of the subsoil.
• The saturated depth profile (water table) in the lateral
drainage layer is typical of steady-state drainage and
gradually varies between different steady-state profiles
characteristic for different depths of saturation as the
simulation progresses.
• The lateral drainage rate can be reliably estimated
from the average depth of saturation throughout the
drain layer which is estimated from the average soil
moisture content of the drain layer.
• The depth of saturation at the edge of the drain layer
or at the collector is zero. Therefore, lateral drainage is
not retarded by standing water in the drain trench.
Vegetative Growth
The HELP model accounts for seasonal variation in ac-
tive and dormant aboveground biomass and leaf area
index through a general vegetative growth model. This
model was extracted from the SWRRB model developed
by the USDA ARS (10). The vegetative growth model
computes daily values of biomass and leaf area index
based on a maximum allowable value from input, daily
temperature and solar radiation data, and the beginning
and ending dates of the growing season. The maximum
value of leaf area index depends on type of vegetation,
soil fertility, climate, and management factors. The
program supplies typical values for selected covers;
these range from 0 for bare ground to 5.0 for an excellent
stand of grass. The HELP model maintains a data file
containing mean monthly temperatures and beginning
and ending dates of the growing season for 183 locations
in the United States. Vegetative growth is a linear func-
tion of the available solar radiation during the first 75 per-
cent of the growing season. Growth can be limited by
temperatures below 10°C (50°F) and low soil moisture.
Vegetative decay is modeled as exponential decay and is
also a function of temperature and soil moisture. The
decay process is modeled continuously, during both the
active growing and dormant seasons.
Accuracy
As stated previously, the primary purpose of the model is
to simulate alternatives for comparison, showing relative
value of alternatives and sensitivity of design parameters.
The secondary purpose is to quantify the water budget
components accurately. Generation of accurate predic-
tions requires good understanding of the model,
hydrologic processes, and landfill design and construc-
tion. In addition, accurate data describing the properties
and variability of all materials and the climate are essen-
tial.
Even with the best of data and knowledge of the model
and landfill, significant errors should be expected in the
estimates of the water budget components due to mini-
mum data requirements and limitations in the modeling
techniques. The following error bounds are believed to be
generally achievable when extensive and accurate data
are available to a knowledgeable user (ideal circumstan-
ces). The cumulative annual total for a water budget
component can typically be estimated within the larger of
the following error bounds: 25 percent of the total or 2
percent of the precipitation for the surface runoff com-
ponent, 10 percent of the total or 7 percent of the
precipitation for the evapotranspiration component, and
79
-------
10 percent of the total or 0.1 percent of the precipitation
for the percolation or leakage through liners component.
The error bound for cumulative annual lateral drainage to
collection systems is about 7 percent of the precipitation
and is equal to the sum of the other errors. Its error is de-
pendent on all other errors because those processes
occur first and any excess or shortfall in the extraction by
those processes controls the quantity of water available
for lateral drainage. These error bounds would be several
times larger when simulations are run with poor data and
a poor understanding of the model and landfill design.
Input Requirements
Climatological Data
Required ciimatological data include daily precipitation,
daily mean temperature, daily solar radiation, maximum
leaf area index, growing season, and evaporative zone
depth. Daily precipitation data can be provided by three
options. The user may enter each value into the
precipitation data file. The second option is to select 5
years of daily precipitation data from a default data set
that has data for 102 cities throughout the United States.
The third option is to synthetically generate daily
precipitation data using a synthetic weather generator
available in the program. The program has statistical
coefficients describing the daily precipitation at 139 cities
throughout the United States. Improvement in the data
generation can be obtained by specifying the normal
mean monthly precipitation at the landfill site. The syn-
thetic weather generator was adapted from the WGEN
model developed by the USDA Agricultural Research
Service (20).
Daily mean temperature and daily solar radiation data are
synthetically generated by the program after the user
selects a location with similar weather from a set of cities
having available data. The program contains statistical
coefficients describing temperature and solar radiation
values at 183 cities throughout the United States. The
generation of daily temperature values can be improved
by specifying the normal mean monthly temperatures for
the landfill site. Similarly, the daily solar radiation values
can be improved by specifying the latitude of the landfill
site.
When executing the program, guidance is available from
a permanent data file for the remaining three climate-re-
lated parameters—growing season, maximum leaf area
index, and evaporative zone depth. Typically, the growing
season is that portion of the year when the mean daily
temperature is above about 11°C (53°F). The maximum
leaf area index is dependent on climate, soil fertility,
cover design, and management practices. Thick layers of
fine-grained soils have better fertility and moisture reten-
tion than thin layers and coarse-grained soils and, there-
fore, support better stands of vegetation. The evaporative
zone depth is dependent on climate, vegetation, and soil
properties. Representative evaporative zone depths for
silty, loamy topsoils are given in the program as a func-
tion of location and quantity of vegetation. Typical values
would be greater for thick clayey layers while the values
for sandy layers would be smaller. In addition, thin layers
of materials and the presence of synthetic material near
the surface also may restrict the evaporative zone depth,
Soil and Design Data
The second set of required data consists of a description
of each layer of material and a description of the landfill
design. Material descriptions can be selected from a
default set of material properties or specified individually
by the user. The material properties that must be
specified by the user are porosity, field capacity, wilting
point, and the saturated hydraulic conductivity. Porosity is
defined as the volume of voids in a layer of material (or
volume of water in a saturated layer) divided by the total
volume of the layer. Field capacity is defined as the
volume of water remaining in a layer of material after it
ceases to drain by gravity divided by the total volume of
the layer. It corresponds to the moisture content remain-
ing when the material exerts a soil suction of 1/3 atmos-
pheres. Wilting point is defined as the volume of water
remaining in a layer of material after a plant extracts as
much water as possible and goes into a permanent witt,
divided by the total volume of the layer. It corresponds to
the moisture content remaining when the material ex-
hibits a soil suction of 15 atmospheres. User-specified
descriptions are recommended since material properties
vary greatly within a given soil classification.
The default set of material descriptions contains proper-
ties for 15 soil types ranging from coarse sand to high
plasticity clay. These 15 types also include fine sands,
loams, silts, and low plasticity clays. These soils also
may be specified as compacted, which causes the
porosity, field capacity, and saturated hydraulic conduc-
tivity to be lower. Compaction is recommended only for
the fine-grained materials. In addition to the 15 soils,
there are also two descriptions of very low hydraulic con-
ductivity soils suitable for liners and one description of
municipal waste with daily cover.
When using either option of describing materials, the
user has two options for initializing the moisture content
of the materials. The user may specify an initial moisture
content ranging from porosity to wilting point. This option
is used when the effects of changing moisture storage
are important in the water budget. Under the second op-
tion, the program initializes the moisture content of the
layer to the approximate long-term, steady-state value.
This option is used when changes in water storage are
unimportant or when the long-term, pseudo-steady-state
water budget is desired.
The landfill description consists of the SCS runoff curve
number, surface area, runoff area, and a description of
the layers including the number of layers, their order and
function, and their thickness. Four types of layers, based
80
-------
on how they function, are used by the model—vertical
percolation layers, lateral drainage layers, barrier soil
liners, and geomembranes with barrier soil. Vertical per-
colation layers are layers that serve no purpose other
than water storage. The topsoil layer and waste layers
are typical examples of this type. Lateral drainage layers
are layers designed to promote lateral drainage to a col-
lection system. They typically have very large saturated
hydraulic conductivities and are underlain by a sloped
liner. Barrier soil liners are layers of low hydraulic con-
ductivity porous material designed to restrict vertical per-
colation or leakage. A geomembrane with barrier soil is a
synthetic membrane underlain by subsoil and is used to
restrict vertical percolation and leakage. In addition, the
slope and drain spacing are needed for lateral drainage
layers and the liner leakage fraction is needed for
geomembranes.
Output Description
The output from the HELP model is a listing of the input
data followed by an account of the water budget com-
ponents in a tabular format. Information on precipitation,
surface runoff, evapotranspiration, lateral drainage from
each liner/drain system, percolation or leakage through
each liner and from the bottom of the profile, moisfure
storage, snow accumulation, and depth of saturation on
the surface of liners is reported. A partial listing of the
output for an example application is presented in the next
section of this chapter. Simulation results are available at
several levels of detail. The cumulative quantity of the
water budget components and its variance are tabulated
on an average monthly and annual basis for the entire
simulation period and optionally on a daily, monthly, and
annual basis. In addition, peak daily results during the
entire simulation period and final moisture contents of the
layers are reported.
EXAMPLE APPLICATION
This section presents a simulation of the water balance
for the closed landfill illustrated in Figure 8-5. The landfill
has eight layers—a three-layer cover system, a waste
layer, and a four-layer double liner system. The cover
consists of a topsoil to support a fair stand of grass, a
sand layer to drain excess infiltration, and a clay liner.
The waste layer contains lifts of waste and daily cover.
The double liner system provides leachate collection and
leakage detection. It contains a sand layer for primary
leachate collection and a geomembrane for the primary
liner. A second sand layer serves as the subsoil for the
geomembrane and as the leakage detection or secon-
dary leachate collection layer. The lower liner is a com-
posite liner consisting of both a geomembrane and a clay
liner and is considered to be one layer. Many other
aspects of the design description required by the HELP
model are shown on the schematic.
The example profile can be simulated as having eight or
preferably nine layers. The layers are described in the
®? INFILTRATION
VERTICAL
PERCOLATION LAYER TOPSOIt
@ LATERAL DRAINAGE LAYER ,
. ;
(3) BARRIER SOIL LINER CLAY
FtReOUkTION
i! ®
BARRIER SOIL LINEH
PERCOLATION (L
Figure 8-5. Typical hazardous waste landfill profile.
input from top to bottom; therefore, the first layer is the
topsoil layer and the last layer is the composite liner.
Eight layers are shown on the schematic but nine layers
were used in the simulation. The additional layer comes
by dividing the waste layer into two layers, the top of
which is thick and the lower of which is thin. Dividing thick
layers in this manner minimizes incontinuities in the solu-
tion. Incontinuities result from use of average moisture
contents with greatly different layer thicknesses and
occur when the zone of saturation above a liner extends
from a thin layer into a thick layer. The materials in the
profile were described using the default set of material
properties. A completed data form describing the landfill
materials and design is shown in Figure 8-6. The data
form is available in the user guide (5) and lists the data
requirements in the exact form and order that data entry
is made in the model.
Climatological input was entered using the default set of
rainfall data and the statistical coefficients and default
values for synthetic generation of daily temperature and
solar radiation values for Philadelphia, Pennsylvania.
Values for evaporative zone depth and maximum leaf
81
-------
DEFAULT SOIL AflD DESIGN DATA INPUT
Title:
/Y^/P/f
7 /!&_
Do you want the program to initialize the soil wacer?
Number of layers: j_
Layer daCa:
Layer 1
(a) chickness c2i/~ inches
(b) layer type ,/ (1 or 2)
(c) liner leakage fraction (only for layer cype 4) — (0 Co 1)
(d) soil Cexcure number /£? (1 Co 20)*
(e) compacted? (only for soil textures 1 Co 15) A>Y} (Yes or No)
(f) inicial soil water content (not asked if program is to initialize
Che soil water or if layer type is 3 or 4) /^ oZ/V/* vol/vol
(muse be becween wilting poinc and porosicy)
Laver 2
(a) thickness /J? inches
(b) layer type 2 (1 to 41
(c) liner leakage fraction (only for layer type 4) ~ (0 Cc '. :
(d) soil texture number /_ (1 Co 20)*
(e) compacted? (only for soil textures 1 to 15) /t^> (Yes or No)
(f) initial soil water content (not asked if program is to initialize
Che soil wacer or if layer type is 3 or 4) ^..?^vf^> vol/vol
(must be between wilting point and porosity)
Laver 3
(a) thickness trj< inches
(b) layer type .7
(c) liner leakage fraction (only for layer type 4) • (0 Co 1)
(d) soil Cexcure number //<_ (1 to 20)*
(e) compacted? (only for soil textures 1 to 15) fcr (Yes or No)
(f) initial soil water content (not asked if program is to initialize
Che soil v.itpr or if layer cype Is 3 or 4 ) /?. 'A./Y?/? vol/vol
(must be becween wilting poinc and porosity)
Laver 4 Laver 5 Laver 6
(a) _ .fy.f (a) /J, (a) /^
(b) / (b) / (b) j2_
(c) — (c) — (c) —
(d) /? (d) /'f. (d) /
(e) /_£, («) XY^ . (e) ,/4
Figure 8-6. Completed data form for landfill materials and design.
82
-------
Layer 7
Laver 8
Layer 9
(a)
(c)
(d)
(e)
(f)
(a)
(b)
(c)
(d)
(e)
(f)
If
-------
area index were selected from the recommended values
for a fair stand of grass. A completed data form from the
user guide (5) for climatological data input is shown in
Figure 8-7.
A partial listing of the output giving all available options,
is shown in Figure 8-8. The options include daily, month-
ly, and annual water balances.
REFERENCES
1. U.S. EPA. 1989. Technical guidance document:
Final covers on hazardous waste landfills and sur-
face impoundments. EPA/530-SW-89-047.
2. U.S. EPA. 1988. U.S. EPA guide to technical resour-
ces for the design of land disposal facilities. EPA
Guidance Document: Final Covers on Hazardous
Waste Landfills and Surface Impoundments.
EPA/530-SW-88-047.
3. Schroeder, P.R., J.M. Morgan, T.M. Walski, and A.C.
Gibson. 1984a. Hydrologic Evaluation of Landfill Per-
formance (HELP) Model: Vol. I, User's Guide for
Version 1. EPA/53Q-SW-84-009. U.S. Environmental
Protection Agency, Washington, DC. 120 pp.
4. Schroeder, P.R., A.C. Gibson, and M.D. Smolen.
1984b. Hydrologic Evaluation of Landfill Performance
(HELP) Model: Vol. II. Documentation for Version 1,
EPA/530-SW-84-010. U.S. Environmental Protection
Agency, Washington, DC. 256 pp.
5. Schroeder, P.R., R.L. Peyton, and J. M. Sjostrom.
1988a. Hydrologic Evaluation of Landfill Performance
(HELP) Model: Vol. III. User's Guide for Version 2,
Internal Working Document. USAE Waterways Ex-
periment Station, Vicksburg, MS.
6. Schroeder, P.R., B.M. McEnroe, and R.L. Peyton.
1988b. Hydrologic Evaluation of Landfill Performance
(HELP) Model: Vol. IV. Documentation for Version 2.
Internal Working Document. USAE Waterways Ex-
periment Station, Vicksburg, MS.
7. Perrier, E.R. and A.C. Gibson. 1980. Hydrologic
simulation on solid waste disposal sites. EPA-SW-
868. U.S. Environmental Protection Agency, Cincin-
nati, OH. 111 pp.
8. Schroeder, P.R. and A.C. Gibson. 1982. Supporting
documentation for the Hydrologic Simulation Model
for Estimating Percolation at Solid Waste Disposal
Sites (HSSWDS). Draft Report. U.S. Environmental
Protection Agency, Cincinnati, OH. I53 pp.
9. Knisel, W.G., Editor. 1980. CREAMS, a field-scale
model for chemical runoff and erosion from agricul-
tural management systems. Vols. I, II, and III. USDA-
SEA-AR Conservation Research Report 26. 643 pp.
10. Williams, J.R., A.D. Nicks, and J.G. Arnold. 1985.
SWRRB, a simulator for water resources in rural
basins. Journal of Hydraulic Engineering. ASCE, Vol.
111, No. 6, pp. 970-986.
CLIMATOLOCICAL DATA INPUT
Default Precipitation Option
'^A-As*.
Normal Mean Monthly Temperatures in Degrees Fahrenheit (Option*!)
Jan.
F*b.
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sep.
Oct.
Nov.
Dec.
Are* Index"
Evaporative Zone Depth in Inches:
-------
RCRA COVER SEMINAR
PHILADELPHIA, PENNSYLVANIA
29 MAY 90
FAIR GRASS
LAYER 1
VERTICAL PERCOLATION LAYER
THICKNESS - 24.00 INCHES
POROSITY - 0.3980 VOL/VOL
FIELD CAPACITY - 0.2443 VOL/VOL
WILTING POINT - 0.1361 VOL/VOL
INITIAL SOIL WATER CONTENT - 0.2739 VOL/VOL
SATURATED HYDRAULIC CONDUCTIVITY - 0.000360000005 CM/SEC
LAYER 2
LATERAL DRAINAGE LAYER
THICKNESS - 12.00 INCHES
POROSITY - 0.4170 VOL/VOL
FIELD CAPACITY - 0.0454 VOL/VOL
WILTING POINT - 0.0200 VOL/VOL
INITIAL SOIL WATER CONTENT - 0.3489 VOL/VOL
SATURATED HYDRAULIC CONDUCTIVITY - 0.009999999776 CM/SEC
SLOPE - 3.00 PERCENT
DRAINAGE LENGTH - 500.0 FEET
LAYER 3
BARRIER SOIL LINER
THICKNESS - 36.00 INCHES
POROSITY - 0.4300 VOL/VOL
FIELD CAPACITY - 0.3663 VOL/VOL
WILTING POINT - 0.2802 VOL/VOL
INITIAL SOIL WATER CONTENT - 0.4300 VOL/VOL
SATURATED HYDRAULIC CONDUCTIVITY - 0.000000100000 CM/SEC
Figure 8-8. Example output.
85
-------
LAYER 4
THICKNESS
POROSITY
FIELD CAPACITY
WILTING POINT
INITIAL SOIL WATER CONTENT
SATURATED HYDRAULIC CONDUCTIVITY
VERTICAL PERCOLATION LAYER
588.00 INCHES
0.5200 VOL/VOL
0.2942 VOL/VOL
0.1400 VOL/VOL
0.2840 VOL/VOL
0.000199999995 CM/SEC
LAYER 5
THICKNESS
POROSITY
FIELD CAPACITY
WILTING POINT
INITIAL SOIL WATER CONTENT
SATURATED HYDRAULIC CONDUCTIVITY
VERTICAL PERCOLATION LAYER
12.00 INCHES
0.5200 VOL/VOL
0.2942 VOL/VOL
0.1400 VOL/VOL
0.2852 VOL/VOL
0.000199999995 CM/SEC
LAYER 6
LATERAL DRAINAGE LAYER
THICKNESS
POROSITY
FIELD CAPACITY
WILTING POINT
INITIAL SOIL WATER CONTENT
SATURATED HYDRAULIC CONDUCTIVITY
SLOPE
DRAINAGE LENGTH
12.00 INCHES
0.4170 VOL/VOL
0.0454 VOL/VOL
0.0200 VOL/VOL
0.0454 VOL/VOL
0.009999999776 CM/SEC
5.00 PERCENT
50.0 FEET
LAYER 7
BARRIER SOIL LINER WITH FLEXIBLE MEMBRANE LINER
THICKNESS
POROSITY
FIELD CAPACITY
WILTING POINT
INITIAL SOIL WATER CONTENT
SATURATED HYDRAULIC CONDUCTIVITY
LINER LEAKAGE FRACTION
6.00 INCHES
0.4170 VOL/VOL
0.0454 VOL/VOL
0.0200 VOL/VOL
0.4170 VOL/VOL
0.009999999776 CM/SEC
0.00005000
Figure 8-8. (Continued).
86
-------
LAYER 8
LATERAL DRAINAGE LAYER
THICKNESS
POROSITY
FIELD CAPACITY
WILTING POINT
INITIAL SOIL WATER CONTENT
SATURATED HYDRAULIC CONDUCTIVITY
SLOPE
DRAINAGE LENGTH
12.00 INCHES
0.4170 VOL/VOL
0.0454 VOL/VOL
0.0200 VOL/VOL
0.0478 VOL/VOL
0.009999999776 CM/SEC
5.00 PERCENT
50.0 FEET
LAYER 9
BARRIER SOIL LINER WITH
THICKNESS
POROSITY
FIELD CAPACITY
WILTING POINT
INITIAL SOIL WATER CONTENT
SATURATED HYDRAULIC CONDUCTIVITY
LINER LEAKAGE FRACTION
FLEXIBLE MEMBRANE LINER
36.00 INCHES
0.3777 VOL/VOL
0.2960 VOL/VOL
0.2208 VOL/VOL
0.3777 VOLAOL
0.000001650000 CM/SEC
0.00005000
GENERAL SIMULATION DATA
SCS RUNOFF CURVE NUMBER
TOTAL AREA OF COVER
EVAPORATIVE ZONE DEPTH
UPPER LIMIT VEG. STORAGE
INITIAL VEG. STORAGE
INITIAL SNOW WATER CONTENT
INITIAL TOTAL WATER STORAGE
SOIL AND WASTE LAYERS
85,
- 1000000,
24.
9.
6,
0,
56
SQ FT
00 INCHES
5520 INCHES
INCHES
INCHES
5736
0000
IN
213.8724 INCHES
SOIL WATER CONTENT INITIALIZED BY USER.
CLIMATOLOGICAL DATA
DEFAULT RAINFALL WITH SYNTHETIC DAILY TEMPERATURES AND
SOLAR RADIATION FOR PHILADELPHIA PENNSYLVANIA
MAXIMUM LEAF AREA INDEX - 2.00
START OF GROWING SEASON (JULIAN DATE) - 115
END OF GROWING SEASON (JULIAN DATE) - 296
Figure 8-8. (Continued).
87
-------
NORMAL MEAN MONTHLY TEMPERATURES, DEGREES FAHRENHEIT
JAN/JUL FEB/AUG MAR/SEP APR/OCT MAY/NOV JUN/DEC
31.20
76.50
33.10 41.80 52.90 62.80 71.60
75.30 68.20 56.50 45.80 35.50
VARIABLE 1: HEAD ON TOP OF LAYER 3
VARIABLE 2: PERCOLATION THROUGH LAYER 3
VARIABLE 3: PERCOLATION THROUGH LAYER 7
VARIABLE 4: PERCOLATION THROUGH LAYER 9
VARIABLE 5: LATERAL DRAINAGE FROM LAYER 2
VARIABLE 6: LATERAL DRAINAGE FROM LAYER 8
DAILY OUTPUT FOR YEAR 74
DAY
1
2
3
4*
5
6
7
8
9
10
11
12
13
14*
15*
16*
360
361
362
363
364
365
RAIN
IN.
0.05
0.00
0.50
0.15
0.00
0.00
0.00
0.00
0.62
0.29
0.50
0.00
0.02
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.31
RUNOFF
IN.
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.001
0.000
0.006
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
ET
IN.
0.014
0.043
0.013
0.011
0.040
0.047
0.047
0.043
0.015
0.012
0.014
0.034
0.012
0.000
0.000
0.000
0.084
0.066
0.096
0.090
0.079
0.016
VAR.
1
IN.
9.
9.
9.
9.
9.
9.
10.
10.
10.
10.
10.
11.
13.
15.
16.
18.
4.
4.
4.
4.
4.
5.
8 0
8 0
8 0
8 0
8 0
9 0
0 0
2 0
4 0
5 0
7 0
2 0
2 0
4 0
9 0
1 0
7 0
8 0
8 0
9 0
9 0
0 0
VAR.
2
IN.
.0030
.0042
.0043
.0043
.0043
.0043
.0043
.0044
.0044
.0044
.0044
.0044
.0046
.0047
.0048
.0050
.0038
.0038
.0039
.0039
.0039
.0039
VAR.
3
IN.
0.0037
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0043
0.0042
0.0042
0.0042
0.0042
0.0041
0.0041
0.0041
0.0041
0.0041
0.0041
VAR.
4
IN.
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0 . 0000
0.0000
0 . 0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0 . 0000
0.0000
VAR.
5
IN.
0.012
0.017
0.017
0.017
0.017
0.017
0.017
0.018
0.018
0.018
0.018
0.019
0.020
0.020
0.020
0.020
0.010
0.010
0.010
0.010
0.010
0.010
VAR.
6
IN.
0.003
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
0.004
SOIL
WATER
IN/IN
0.2745
0.2727
0.2867
0.2909
0.2905
0.2894
0.2847
0.2790
0.2934
0.3059
0.3199
0.3133
0.3025
0.3016
0.3005
0.2995
0.2576
0.2530
0.2478
0.2427
0.2380
0.2452
Figure 8-8. (Continued).
88
-------
MONTHLY TOTALS FOR YEAR 74
JAN/JUL FEB/AUG MAR/SEP APR/OCT MAY/NOV JUN/DEC
PRECIPITATION (INCHES)
RUNOFF (INCHES)
EVAPOTRANSPIRATION
(INCHES)
LATERAL DRAINAGE FROM
LAYER 2 (INCHES)
PERCOLATION FROM
LAYER 3 (INCHES)
LATERAL DRAINAGE FROM
LAYER 6 (INCHES)
PERCOLATION FROM
LAYER 7 (INCHES)
LATERAL DRAINAGE FROM
LAYER 8 (INCHES)
PERCOLATION FROM
LAYER 9 (INCHES)
2,95
2.08
0.007
0.002
0.978
2.882
2.14
3.83
0.000
0.231
1.568
2.725
4.91
4.68
0.264
0.085
2.629
4.584
2.77
1.93
0.089
0.121
3.257
1.895
3.21
0.81
0.000
0.000
3.368
1.116
4.43
4.04
0.052
0.179
5.157
1.267
0.5736 0.5438 0.6033 0.5951 0.6026 0.5870
0.5887 0.5094 0.4260 0.3829 0.3204 0.2957
0.1476 0.1456 0.1605 0.1793 0.1683 0.1481
0.1387 0.1329 0.1241 0.1243 0.1169 0.1184
0.0006 0.0005 0.0006 0.0006 0.0006 0.0006
0.0006 0.0006 0.0006 0.0006 0.0006 0.0006
0.1310 0.1169 0.1283 0.1235 0.1275 0.1234
0.1276 0.1277 0.1237 0.1279 0.1237 0.1279
0.1320 0.1172 0.1284 0.1236 0.1274 0.1233
0.1275 0.1276 0.1236 0.1278 0.1237 0.1278
0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
0.0001 0.0001 0.0001 0.0001 0.0001 0.0001
MONTHLY SUMMARIES FOR
AVG. DAILY HEAD ON
LAYER 3 (INCHES)
STD. DEV. OF DAILY HEAD
ON LAYER 3 (INCHES)
AVG. DAILY HEAD ON
LAYER 7 (INCHES)
STD. DEV. OF DAILY HEAD
ON LAYER 7 (INCHES)
15.12
11.24
4.19
0.74
0.00
0.00
0.00
0.00
19.00
9.33
0.20
0.51
0.00
0.00
0.00
0.00
DAILY HEADS
18.83
7.73
2.32
0.42
0.00
0.00
0.00
0.00
27.23
6.40
2.79
0.37
0.00
0.00
0.00
0.00
21.29
5.20
1.60
0.33
0.00
0.00
0.00
0.00
16,06
4.43
1.63
0.28
0.00
0.00
0.00
0.00
Figure 8-8. (Continued).
89
-------
AVG. DAILY HEAD ON 0.08
LAYER 9 (INCHES) 0.07
STD. DEV. OF DAILY HEAD 0.00
ON LAYER 9 (INCHES) 0.00
***********************************
***********************************
ANNUAL TOTALS
PRECIPITATION
RUNOFF
EVAPOTRANSPIRATION
LATERAL DRAINAGE FROM LAYER 2
PERCOLATION FROM LAYER 3
LATERAL DRAINAGE FROM LAYER 6
PERCOLATION FROM LAYER 7
LATERAL DRAINAGE FROM LAYER 8
PERCOLATION FROM LAYER 9
CHANGE IN WATER STORAGE
SOIL WATER AT START OF YEAR
SOIL WATER AT END OF YEAR
SNOW WATER AT START OF YEAR
SNOW WATER AT END OF YEAR
ANNUAL WATER BUDGET BALANCE
0.07 0.07
0.07 0.07
0.00 0.00
0 . 00 0 . 00
************
************
FOR YEAR
(INCHES)
37.78
1.029
31.425
6.0285
1,7047
0.0070
1.5090
1.5097
0.0010
-2.221
213.87
211.65
0.00
0.00
0.00
0.07
0.07
0.00
0.00
***********!
***********•)
74
(CU. FT.)
3148334.
85764.
2618786.
502377.
142062.
586.
125754.
125812.
85.
-185075.
17822700.
17637624.
0.
0.
-1.
0.07 0,07
0.07 0.07
0.00 0.00
0.00 0.00
*************
*************
PERCENT
100.00
2.72
83.18
15.96
4.51
0.02
3.99
4.00
0.00
-5.88
0.00
Figure 8-8. (Continued).
90
-------
AVERAGE MONTHLY VALUES IN INCHES FOR YEARS 74 THROUGH 78
PRECIPITATION
TOTALS
STD. DEVIATIONS
RUNOFF
TOTALS
STD. DEVIATIONS
EVAPOTRANSPIRATION
TOTALS
STD. DEVIATIONS
JAN/JUL
4.59
3.67
2.53
1.95
1.478
0.414
3.008
0.556
0.973
4.199
0.246
1.766
LATERAL DRAINAGE FROM LAYER
FEB/AUG
1.88
4.46
0.66
2.49
0.053
0.212
0.066
0.195
1.485
3.565
0.220
1.593
2
MAR/SEP
4,09
4.17
1.00
2.07
0.290
0.302
0.181
0.596
2.701
2.998
0.089
1.536
APR/OCT
3.03
2.76
1.51
1.21
0.154
0.047
0.216
0.050
3.019
2.027
0.331
0.836
MAY/NOV
3.85
2.68
2.03
2.63
0.156
0.384
0.254
0.706
4.110
1.361
1.020
0.430
JUN/DEC
4.50
3.99
2.17
1.78
0.170
0.339
0.244
0.370
4.833
1.015
1.223
0.195
TOTALS 0.4856 0.4957 0.5418 0.5476 0.5656 0.5414
0.5247 0.4530 0.3789 0.3728 0.3589 0.4268
STD. DEVIATIONS 0.1797 0.1440 0.1524 0.0954 0.0847 0.0987
0.1150 0.1025 0.0900 0.1297 0.1225 0.1686
PERCOLATION FROM LAYER 3
TOTALS 0.1548 0.1502 0.1639 0.1623 0.1567 0.1427
0.1344 0.1291 0.1209 0.1236 0.1204 0.1331
STD. DEVIATIONS 0.0391 0.0353 0.0286 0.0245 0.0182 0.0149
0.0082 0,0069 0.0060 0.0089 0.0101 0.0199
LATERAL DRAINAGE FROM LAYER 6
TOTALS 0.0006 0.0006 0.0006 0.0006 0.0006 0.0006
0.0006 0.0006 0.0006 0.0006 0.0006 0.0006
Figure 8-8. (Continued).
91
-------
STD. DEVIATIONS 0.0000
0.0000
PERCOLATION FROM LAYER 7
TOTALS 0.1301
0.1304
STD. DEVIATIONS 0.0015
0.0022
LATERAL DRAINAGE FROM LAYER
TOTALS
0.0000 0.0000 0.0000 0.0000 0.0000
0.0000 0.0000 0.0000 0.0000 0.0000
0.1181 0.1298 0.1257 0.1300 0.1260
0.1305 0.1264 0.1306 0.1264 0.1306
0.0023 0.0017 0.0018 0.0020 0.0021
0.0022 0.0021 0.0022 0.0021 0.0021
8
0.1302
0.1303
STD. DEVIATIONS 0.0017
0.0022
PERCOLATION FROM LAYER 9
TOTALS 0.0001
0.0001
STD. DEVIATIONS 0.0000
0.0000
0.1181 0.1297 0.1256 0,1299 0.1259
0.1304 0.1263 0,1305 0.1263 0.1305
0.0022 0.0016 0.0018 0.0020 0.0020
0.0022 0.0021 0.0022 0.0021 0.0022
0.0001 0.0001 0.0001 0.0001 0.0001
0.0001 0.0001 0.0001 0.0001 0.0001
0.0000 0.0000 0.0000 0.0000 0.0000
0.0000 0.0000 0.0000 0.0000 0.0000
AVERAGE ANNUAL TOTALS & (STD. DEVIATIONS) FOR YEARS 74 THROUGH 78
(INCHES)
PRECIPITATION
RUNOFF
EVAPOTRANSPIRATION
LATERAL DRAINAGE FROM
LAYER 2
PERCOLATION FROM LAYER 3
LATERAL DRAINAGE FROM
43.67 i
3.998 i
32.287 i
5.6928 1
1.6920 I
0.0073 1
( 7.930)
C 3.685)
C 2.428)
( 1.0786)
: 0.1508)
; 0.0002)
(CU. FT.)
3639167.
333190.
2690580.
474402.
140998.
604.
PERCENT
100.00
9.16
73.93
13.04
3.87
0.02
PERCOLATION FROM LAYER 7
Figure 8-8. (Continued).
1.5347 ( 0.0220)
127890.
3.51
92
-------
LATERAL DRAINAGE FROM
LAYER 8
1.5338 ( 0.0214)
127814.
3.51
PERCOLATION FROM LAYER 9 0.0010 ( 0.0000) 86. 0.00
CHANGE IN WATER STORAGE 0.150 ( 5.089) 12491. 0.34
***********************************************************************
***********************************************************************
PEAK DAILY VALUES FOR YEARS 74 THROUGH 78
PRECIPITATION
RUNOFF
LATERAL DRAINAGE FROM LAYER 2
PERCOLATION FROM LAYER 3
HEAD ON LAYER 3
LATERAL DRAINAGE FROM LAYER 6
PERCOLATION FROM LAYER 7
HEAD ON LAYER 7
LATERAL DRAINAGE FROM LAYER 8
PERCOLATION FROM LAYER 9
HEAD ON LAYER 9
SNOW WATER
(INCHES)
3.99
2.341
0.0209
0.0068
36.1
0.0000
0.0043
0.0
0.0045
0.0000
0.1
4.09
(CU. FT.)
332500.0
195074.9
1744.7
567.3
2.5
359.0
371.3
0.2
340770.0
MAXIMUM VEG. SOIL WATER (VOL/VOL) 0.3980
MINIMUM VEG. SOIL WATER (VOL/VOL) 0.1359
***********************************************************************
Figure 8-8. (Continued).
93
-------
FINAL WATER STORAGE AT END OF YEAR 78
LAYER
1
2
3
4
5
6
7
8
9
SNOW WATER
(INCHES)
6,57
4.19
15.48
167.74
3.42
0.55
2.50
0.57
13.60
0.00
(VOL/VOL)
0.2739
0 . 3488
0.4300
0.2853
0.2853
0.0454
0.4170
0.0478
0.3777
Figure 8-8. (Continued).
94
-------
11. USDA, Soil Conservation Service. 1972. Section 4,
Hydrology. In; National Engineering Handbook, U.S.
Government Printing Office, Washington, DC. 631
pp.
12. Morton, R.E. 1919. Rainfall Interception. Monthly
Weather Review, U.S. Weather Bureau. Vol. 47, No.
9, pp. 603-623.
13. Ritchie, J.T. 1972. A model for predicting evaporation
from a row crop with incomplete cover. Water
Resources Research. Vol. 8, No. 5, pp. 1204-1213.
14. Brooks, R.H. and A.T. Corey. 1964. Hydraulic proper-
ties of porous media. Hydrology Paper No. 3,
Colorado State University. 27 pp.
15. Campbell, G.S. 1974. A simple method for determin-
ing unsaturated hydraulic conductivity from moisture
retention data. Soil Science. Vol. 117, No. 6, pp.
311-314.
16. Brown, K.W., J.C. Thomas, R.L. Lytton, P. Jayawik-
rama, and S.C. Bahrt. 1987. Quantification of Leak
Rates Through Holes in Landfill Liners. EPA/600/S2-
87-062. U.S. Environmental Protection Agency, Cin-
cinnati, OH. 151 pp.
17. Giroud, J.P. and R. Bonaparte. 1989a. Leakage
through liners constructed with geomembranes,
Part I: Geomembrane liners. Geotextiles and
Geomembranes. Vol. 8, No. 1, pp. 27-67.
18- Giroud, J.P. and R. Bonaparte. I989b. Leakage
through liners constructed with geomembranes,
Part II: Composite liners. Geotextiles and
Geomembranes. Vol. 8, No. 2, pp. 71-111.
19. Giroud, J.P., A. Khatami, and K. Badu-Tweneboah.
1989c. Evaluation of the rate of leakage through
composite liners. Geotextiles and Geomembranes.
Vol. 8, No. 4, pp. 337-340.
20. Richardson, C.W. and D.A. Wright. 1984. WGEN: a
model for generating daily weather variables. ARS-8,
Agricultural Research Service, USDA. 83 pp.
95
-------
CHAPTER 9
SENSITIVITY ANAL YSIS OF HELP MODEL PARAMETERS
INTRODUCTION
This chapter examines the sensitivity of landfill water
balance to numerous landfill design variables using the
Hydrologic Evaluation of Landfill Performance (HELP)
model. This information is useful in a variety of ways. It
can aid the design engineer in selecting preliminary
design alternatives for municipal or hazardous waste
landfills. It can serve as a basis for regulatory agencies
to establish and evaluate technical guidelines. It can
also provide additional insight on the importance and in-
teraction of specific design variables on the water
balance. Finally, it can assist in evaluating the suitability
of methodologies used in the computer model. The
analyses include examination of both cover systems and
lateral drainage/liner systems (1). The complete list of
design characteristics examined is given in Table 9-1,
The analysis of landfill cover design is divided into two
parts. First, water balance results are compared for dif-
ferent general design conditions such as climate (loca-
tion), topsoil and vegetative characteristics, and cover
Table 9-1. Parameters Selected for Sensitivity Analysis
Typical Cover Systems
Quantity of vegetation
Cover soil thickness
Topsoil type
Use of lateral drainage layer
Geographical location or climate
Vegetative Layer
SCS runoff curve number
Evaporative depth
Drainable porosity
Plant available water
Municipal vs. hazardous waste cover design
Analysis of Percolation and Drainage Design
Hydraulic conductivity of barrier soil layer
Hydraulic conductivity of lateral drainage layer
Geomembrane leakage factor
Liner type (clay, geomembrane, or composite)
Slope of lateral drainage layer
Drainage length
Double liner system design
design. Then, the effects resulting from changes in
specific characteristics of the vegetative layer, such as
runoff curve number, evaporative depth, and moisture
retention properties, are examined. The water balance
components examined in this chapter are surface runoff,
evapotranspiration, lateral subsurface drainage to collec-
tion systems, and vertical percolation through the soil
liner.
The analysis of liner systems examines the effects of
slope, drain spacing, saturated hydraulic conductivity,
and geomembrane leakage characteristics on leachate
collection and leakage through liners. Two types of verti-
cal inflows to the drain layer are considered. First, an in-
flow rate of 127 cm/yr (50 in./yr) was used to represent
infiltration at an open landfill. This inflow was distributed
in time according to actual rainfall patterns at Shreveport,
Louisiana. Second, an inflow rate of 20 cm/yr (8 in./yr)
uniformly distributed in time was used to represent in-
filtration at a covered landfill.
In the discussion that follows, the effects of the saturated
hydraulic conductivities of the drain layer and liner are
first investigated by holding the slope and drainage
length constant. Then, the slope and drainage length are
examined by holding the hydraulic conductivities con-
stant. In all cases, the thickness of the lateral drainage
layer was greater than the maximum head, and the thick-
ness of the soil liner was 61 cm (24 in.).
COMPARISON OF TYPICAL COVER SYSTEMS
Design Parameters
Three locations were studied to determine the effect of
various climatological regimes on cover performance—
Santa Maria, California; Schenectady, New York; and
Shreveport, Louisiana. These locations represent a wide
range in levels of precipitation, temperature, and solar
radiation as summarized in Table 9-2. Default values for
precipitation, temperature, solar radiation, and leaf area
index are stored in the HELP model for each site and
were used for the sensitivity analysis simulations. The
period of record stored in the HELP model for daily
precipitation is 1974 through 1978.
Two cover designs were examined as shown in Figure 9-1.
One is typical of some newer landfills where 0.61 m (2 ft)
97
-------
Table 9-2. Climatological Regimes
Climatological
Variable
Location
Santa Maria, CA Schenectady, NY
Shreveport, LA
Precipitation1
Mean annual (in.)
Mean winter
(Nov-Apr) (in.)
Mean summer
(May-Oct) (in.)
Temperature
Mean annual (°F)
Mean Jan (°F)
Mean July (°F)
Days with minimum
below 32°F
Solar radiation
Mean daily (langleys)
14
12
2
57
51
62
24
450
48
19
29
49
23
73
129
290
44
22
22
66
47
83
37
410
1These mean values are for the period simulated by the HELP model in this section, 1974-1978.
of topsoil overlies a 0.31-m (1-ft) thick lateral drainage
layer having a saturated hydraulic conductivity of 3 x 10~2
cm/sec, a slope of 0.01 m/m (0.03 ft/ft) and a maximum
drainage length of 61 m (200 ft). The drainage layer is
underlain by a 0,61-m (2-ft) thick soil liner having a
saturated hydraulic conductivity of 1 x 10"7 cm/sec. The
other design is typical of older municipal sanitary landfills
where a topsoil layer overlies a 0.61-m (2-ft) thick soil
liner having a saturated hydraulic conductivity of 1 x 10
cm/sec.
-6
Two types of topsoil were considered In the cover
designs: sandy loam and silty, clayey loam. The sandy
loam characteristics were those of the HELP model
default soil texture 6, which represents Unified Soil Clas-
sification System (USCS) soil class SM and U.S. Depart-
ment of Agriculture (USDA) soil class SL. The silty,
clayey loam characteristics were those of the HELP
model default soil texture 12, which represents USCS
soil class CL and USDA soil class SICL. The topsoil-
type designation was used to select soil porosity, field
capacity, wilting point, and hydraulic conductivity, be-
sides influencing the selection of the runoff curve num-
ber. In addition to two types of topsoil, two thickness of
topsoil were examined—46 cm (18 in.) and 91 cm (36 in.).
The vegetative cover was designated as being either a
good stand of grass or a poor stand of grass. This selec-
tion dictated the values for leaf area index, evaporative
depth, and runoff curve number, and influenced the value
used for the saturated hydraulic conductivity of the top-
soil. For a given vegetative cover and topsoil material,
the runoff curve number was obtained from the HELP
Model User's Guide (2). These numbers were 60 for
good grass on sandy loam; 80 for poor grass on sandy
loam; 81 for good grass on silty, clayey loam; and 92 for
poor grass on silty, clayey loam. These curve numbers
are in agreement with values obtained from Section 4,
Hydrology, National Engineering Handbook (3). The
depth of the evaporative zone was chosen as 18 cm (7
in.) for poor grass and 36 cm (14 in.) for good grass.
Results
Figures 9-2 and 9-3 summarize the results obtained in
the general sensitivity analysis performed on the cover
systems, respectively, with and without lateral drainage.
The height of each bar segment represents the cor-
responding mean annual value of water balance com-
ponent in inches which is given next to each bar
segment. The results provide a comparison of the effects
of varying quantity of vegetation, cover design, topsoil
type, topsoil thickness, and Climatological regime.
Effects of Vegetation
Two levels of vegetation were examined—a poor stand of
grass and a good stand of grass; the latter represents
three times the quantity of vegetation as that of the
former. Table 9-3 presents the water balance results for
both cover systems at all three sites as a function of level
of vegetation. The results are given in units of percent of
the precipitation during the simulation period.
Vegetation reduces surface runoff and increases
evapotranspiration. Evapotranspiration is greater be-
cause the plant demand for moisture and a greater quan-
tity of water is available for evapotranspiration due to
greater infiltration and a greater evaporative zone. Runoff
is less because vegetation increases the minimum in-
filtration rate, drying rate, interception, and surface rough-
ness, which results in a decrease in the runoff curve
98
-------
VEGETATION
LATERAL DRAINAGE
18" or 36'
12"
24"
a) Three Layer Landfill Cover Design
VEGETATION
18" or 36"
24"
b) Two Layer Landfill Cover Design
Figure 9-1. Cover designs for sensitivity analysis.
number. The influence of surface vegetation on the
volume of lateral drainage and percolation or leakage
from the cover is varied. However, the quantity of
vegetation tends to have very little effect on the percola-
tion or leakage through the cover system. For the cover
with lateral drainage, the increase in infiltration with good
grass was greater than the increase in evapotranspira-
tion, resulting in a larger volume of lateral drainage and a
negligible change in percolation. For the cover without
lateral drainage, the increase in infiltration yielded high
heads or depths of saturation above the liner that per-
mitted greater evapotranspiration by maintaining higher
moisture contents in the evaporative zone. Consequent-
ly, the increase in evapotranspiration was greater than
the increase in infiltration. This resulted in a trend toward
a small decrease in percolation for a higher level of
vegetation. The opposite trend may occur for vegetative
layers having lower saturated hydraulic conductivities
and higher plant available water capacities. The results
were similar at all three sites despite quite different
climates. In summary, vegetation decreases runoff and
increases evapotranspiration but tends to have little ef-
fect on the water balance. The magnitude of the effects
is design dependent and to a lesser degree climate de-
pendent. The main function of vegetation is to control
erosion.
Effects of Topsoil Thickness
Two topsoil thicknesses were examined—46 cm (18 in.)
and 91 cm (36 in.). Table 9-4 presents the water balance
results for the two-layer cover system as a function of
topsoil thickness at all three sites. The results are given
in units of percent of the precipitation during the simula-
tion period. The cover system with lateral drainage was
not used in this analysis because lateral drainage would
negate the effects by preventing or minimizing the in-
99
-------
60 J
SS -
« 58 -
en
UJ
o 4
C An
m
D
^ 35 -
^ 30 -
I 25 -
| 20 -
3 IS -
1 "
10 -
5 -
0 -
111
WIT.
n rrl
III
RUNOFF
EVAF
LATE
OTRANSPIRATION
BAL ORAIf
JAQE
PERCOLATION
QQ GOOD GRASS
PQ POOR QRA33
8L 18* OF SANDY LOAM
3ICL 18" OF SILTY
CLAYEY LOAM
H°'°
fi
L
[ 8.2
GG
W.4
7.4
5.,
00 8
|.~I_
PG <
SL
-11.1
10. 4
2. 6
> 0 3
T
3.1
8.8
2.2
Q 3
1G PG
SICL
SANTA MARIA. CA
k
10.1
23.3
t».3
1 A
T-"
2.0
22.
9
18.0
1 4
GG PG
SL
-|3.8
32.8
a, a
1.0
1
».i
28,
5.0
• 0,9
GO PG
SICL
SHREVEPORT, LA
4
i nni-
0.0
24.7
22.1
1.2 (
31.1
24.3
21.
1,2
I
GG PG
SL
4
L
"13.8
31.1
12.3
9.4
21.9
9. a
1.0
I
GG PG
SICL
6CHENECTADY. NY
Figure 9-2. Bar graph for three-layer cover design showing effect of surface vegetation, topsoil type, and location.
60 -
55 -
_ 50 -
1 45 -
o
•z
3 35 -
-i 30 -
D
i 25 -
z 20 -
Ul
2 IS -
10 -
5 -
0 -
P
;
^
|
1.3
'//•>
ty§fc
GQ
pa
is
36
BUNOFF
EVAPOTRANSPIHATION
PERCOLATION
GOOD GRASS
POOR GRASS
18- OF SANDY LOAM
36" OF SANDY LOAM n1Q
a. 3
4.9
GG
p
1.8
7.5
5.3
G
18
G
0.4
7.8
6.1
G P
36
30.8
7.5
8.1
G
SANTA MARIA, CA
G
29.4
13.8
G P
3.3
2S.O
15,8
G
18
G
25.1
*
'
X
18.8 E
1
G P
]2.0
23.3
18.9
G
36
SHREVEPORT, LA
G
" 4.8
28.4
14.0
" e.s
' 25.8
15.8
G PQ
18
it,? r
28.3
19.9
i
" 2.7
, 24.7
2O.8
GG PG
36
SCHENECTAOY, NY
Figure i-3. Bar graph for two-layer cover design showing effect of topsoil depth, surface vegetation, and location.
100
-------
Table 9-3. Effects of Climate and Vegetation
81 cm (36 in.) of Sandy Loam Topsoil
61 cm (24 in.) of 1 x 10"6 cm/sec Clay Liner
Locations
Poor grass
Runoff
Evapotranspiration
Percolation
Good grass
Runoff
Evapotranspiration
Percolation
CA
5.6
51.8
42.6
3.1
55.0
42.9
LA
(Percent of Precipitation)
4.6
53.0
42.4
0.2
57.2
42.6
NY
5.5
52.1
42.4
3.5
55.3
41.2
46 cm (18 in.) of Sandy Loam Topsoil
31 cm (12 in.) of 0.03 cm/sec Sand with 61 m (200 ft) Drain Length at 3% Slope
,-7,
61 cm (24 in.) of 1x10 cm/sec Clay Liner
Locations
Poor grass
Runoff
Evapotranspiration
Lateral drainage
Percolation
Good grass
Runoff
Evapotranspiration
Lateral drainage
Percolation
CA
3.0
51.6
41.2
4.2
0.0
52.6
43.2
4.2
LA
(Percent of Precipitation)
4.4
51.9
40.6
3.1
0.2
53.0
43.7
3.1
NY
2.2
50.3
44.0
2.5
0.0
51.0
45.5
2.5
Table 9-4. Effects of Climate and Topsoil Thickness
Sandy Loam Topsoil with a Poor Stand of Grass
61 cm (24 in.) of 1 x 10"6 cm/sec Clay Liner
46cm (18 in.) of topsoil
Runoff
Evapotranspiration
Percolation
91 cm (36 in.) of topsoil
Runoff
Evapotranspiration
Percolation
CA
11.2
51.9
36.9
5.6
51.8
42.6
Locations
LA
(Percent of Precipitation)
7.5
56.9
35.6
4.6
53.0
42.4
NY
13.4
54.5
32.1
5.5
52.1
42.4
101
-------
trusion of the saturated zone above the liner into the
evaporative zone.
Significant differences existed between the 46- and 91-
cm (18- and 36-in.) topsoil depth simulations in the ab-
sence of lateral drainage. The effects were similar at all
three sites. Runoff and evapotranspiration were greater
for the shallower depth to the liner, indicating that the
head above the barrier soil layer maintained higher mois-
ture contents in the evaporative zone. The percolation
was subsequently less than the cases with greater top-
soil thickness. The 91-cm (36-in.) depth to the liner per-
mits larger heads and longer sustaining heads since a
greater thickness of material below the evaporative zone
is free from abstraction of water by evapotranspiration.
The larger heads provide a greater pressure gradient to
increase the leakage rate through the cover system.
In general, the effects of topsoil thickness vary greatly as
the thickness increases from several inches to several
feet. Throughout the transition, the quantity of runoff
should continue to decrease until the depth to the liner
becomes sufficiently great so as to prevent the zone of
saturation to ever climb into the evaporative zone.
Similarly, the percolation through the liner should con-
tinue to increase until there is no interaction between the
saturation zone and the evaporative zone. The
evapotranspiration is expected to increase initially as the
available storage in the evaporative zone increases, i.e.,
until the depth to the liner equals the maximum depth
that evapotranspiration can reach. At greater depths the
evapotranspiration should continue to decrease until the
depth to the liner is sufficient to prevent any further inter-
actions between the evaporative and saturation zones.
While percolation increases with topsoil thickness given
identical properties for all layers in the cover system,
adequate thickness must be provided in a design to en-
sure the integrity of the cover system. A small topsoil
thickness would not provide adequate water storage to
support vegetation, maintain soil stability, and control
erosion. Similarly, a shallow depth to the liner would
promote desiccation or freezing of the liner, which may
greatly increase its permeability and, therefore, the per-
colation.
Effects of Topsoil Type
Two topsoil types were examined—sandy loam and silty,
clayey loam. Table 9-5 presents the water balance
results for the three-layer cover system as a function of
topsoil type at all three sites. The results are given in
units of percent of the precipitation during the simulation
period. The cover system without lateral drainage was
not used in this analysis because the intrusion of the
saturated zone above the liner into the evaporative zone
would decrease the magnitude of the effects.
The results show that the clayey topsoil significantly in-
creased both runoff and evapotranspiration, which in turn
greatly decreased lateral drainage and percolation. The
results were similar at all three sites. Runoff increased
from about 3 percent to 20 percent of the precipitation,
due primarily to the larger runoff curve number selected
for the clayey soil based on its lower minimum infiltration
rate. Evapotranspiration increased approximately from
51 percent to 61 percent of precipitation, due to the lower
hydraulic conductivity of the clayey soil and, more impor-
tantly, the larger plant available water capacity (field
capacity minus wilting point). The lower hydraulic con-
ductivity of the clayey soil slowed the drainage rate,
maintaining moisture contents above field capacity for
longer periods of time and allowing greater
evapotranspiration. The larger plant available water
capacity of the clayey soil provided a larger moisture
reservoir available for evapotranspiration after gravity
drainage ceased. The lateral drainage was reduced from
about 42 percent to 16 percent of the precipitation and
Table 9-5. Effects of Climate and Topsoil Types
46 cm (18 in.) of Topsoil with a Poor Stand of Grass
31 cm (12 in.) of 0.03 cm/sec Sand with 61 m (200 ft) Drain Length at 3% Slope
,-7.
61 cm (24 in.) of 1 x 10 cm/sec Clay Liner
Sandy loam
Runoff
Evapotranspiration
Lateral drainage
Percolation
Silty, clayey loam
Runoff
Evapotranspiration
Lateral drainage
Percolation
Locations
CA LA NY
(Percent of Precipitation)
3.0 4.4 2.2
51.6 51.9 50.3
41.2 40.6 44.0
4.2 3.1 2.5
21.6 22.3 19.2
61.2 64.4 58.6
15.0 11.3 20.3
2.2 2.0 1.9
102
-------
the percolation was reduced from about 3 percent to 2
percent of precipitation,
Use of Lateral Drainage Layer
Direct comparison of the use of a lateral drainage layer
was not made since different liner systems were used in
the two cover designs. The impact of the use of a lateral
drainage layer was explained briefly above. In general,
the use of a lateral drainage layer would be expected to
decrease the height of the saturation zone above the
liner by draining some of the infiltrated water from the
cover system. As such, percolation through clay liners
would decrease slightly. In addition, runoff and
evapotranspiration also would tend to decrease but the
magnitude of the change would be design dependent.
Topsoil thickness, topsoil type, vegetation, and climate
would have impacts.
Effects of Climate
The effects of climate were examined in each of the pre-
vious sections. As shown in Figures 9-2 and 9-3, climate
affects the absolute magnitude, in inches, of the water
budget components. However, Tables 9-3, 9-4, and 9-5
show that climate has a much smaller effect on the rela-
tive magnitude of the water budget components in terms
of percent of the precipitation. The relative proportions of
the water budget components are primarily design de-
pendent while the magnitudes are strongly dependent on
the magnitude of the precipitation.
The effect of temperature and solar radiation can be
determined by comparing the results for the Louisiana
and New York sites. These two sites have similar annual
rainfall, although the New York site had somewhat higher
annual and summer rainfall. The higher temperature and
solar radiation in Louisiana produced about an inch or
two more evapotranspiration despite the larger quantity
of rainfall in New York. Consequently, the lateral
drainage tended to be slightly less at the Louisiana site.
However, these differences are much smaller than the
differences caused by changes in designs.
Vegetative Layer Properties
The effects of vegetative layer properties on the water
balance of two cover systems are presented below. The
vegetative layer properties examined are runoff curve
number, evaporative depth, drainable porosity, and plant
available water capacity. Vegetated cover designs with
and without lateral drainage were used in the analyses;
the vegetation was assumed to be a fair stand of grass.
The thickness of the vegetative layer was 46 cm (18 in.)
in both designs. The simulations were performed using
climatic data for Santa Maria, California, and Shreveport,
Louisiana, and topsoil properties typical of sandy loam
and silty, clayey loam. Tables 9-6 and 9-7 summarize
the parameter combinations examined under this part of
the sensitivity analysis study and present the results of
the simulations as percentage of precipitation.
Effects of SCS Runoff Curve Number
The SCS runoff curve number was varied from 65 to 90
for the sandy loam and from 75 to 95 for the silty, clayey
loam. The range of curve number was selected to in-
clude values representative of the entire range of pos-
sible slopes and land management practices used at
landfills. The depth of the evaporative zone was 25 cm
(10 in.) in all cases. Simulations for the three-layer cover
design were performed for both soil types, whereas
simulations for the two-layer cover design were per-
formed only for sandy loam. The results are presented in
Table 9-6.
An increase in runoff curve number produced an increase
in runoff and a decrease in evapotranspiration, lateral
drainage, and percolation. The percent increase in runoff
was less for the two-layer cover design than for the three-
layer cover design. This result was due to the higher
average moisture content in the topsoil layer of the two-
layer design caused by the restriction to vertical flow im-
posed by the soil liner in the absence of lateral drainage.
This limited the infiltration capacity of the topsoil, causing
more frequent saturation of the topsoil and, therefore,
more runoff. Thus, runoff volume at low curve numbers
was higher for the two-layer cover compared to the three-
layer cover. This effect was not as great at high curve
numbers because infiltration for both designs was sig-
nificantly reduced by the curve number itself rather than
saturated conditions.
The effects of location or climate on runoff are difficult to
discern from the results; however, results in terms of per-
cent of the precipitation did not differ greatly between the
two sites. For example, in comparing runoff from Santa
Maria and Shreveport, a smaller percentage of precipita-
tion could be expected to drain from the surface as runoff
in Santa Maria due to the higher evaporative demand
combined with lower total precipitation and longer periods
of time between storms. This effect is seen in the data
for the three-layer design, but the difference is not as
large as may have been expected. Only small differen-
ces occur largely because the majority of the rainfall at
Santa Maria occurs during the winter when the evapora-
tive demand is the lowest. In addition, several unusually
large storms occurred at Santa Maria that yielded un-
usually large runoff. However, for simulations of the two-
layer design with low curve numbers, the influence of the
two large storms in Santa Maria caused the runoff per-
centage to exceed that in Shreveport. This would not be
the case if the two storms were excluded.
Summarizing the curve number effects, increasing the
curve number directly causes an increase in runoff and a
decrease in infiltration. The majority of the decrease in
infiltration is reflected as decreases in lateral drainage
and evapotranspiration. The decrease in leakage
through the cover system is generally small. Changes in
slope, vegetation, and land management practices yield
103
-------
only small changes in runoff for soil types and conditions
with curve numbers below 75. The climate, design, and
topsoil characteristics affect the volume of runoff for a
given curve number. The nature of the effects is closely
tied to the potential for evapotranspiration, vertical
drainage from the topsoil, and lateral drainage.
Effects of Evaporative Depth
Evaporative depth as defined by its use in the HELP
model is the thickness of the evapotranspiration zone,
the maximum depth from which water can be extracted to
satisfy evapotranspiration demand. This depth is a func-
tion of soil properties, vegetation, climate, and design.
The evaporative depth was varied from 10 to 46 cm (4 to
18 in.) for both sandy loam and silty, clayey loam. The
runoff curve number was 75 lor the sandy loam and 85
for the silty, clayey loam. Simulations for the three-layer
cover design were performed for both soil types, whereas
simulations for the two-layer cover design were per-
formed only for sandy loam. The results are presented in
Table 9-6.
Evapotranspiration increased with increasing evaporative
depth while lateral drainage and percolation decreased;
the effect on runoff varied. The interrelationship between
these variables is complex and depends on many fac-
tors. The increase in evaporative depth allows
evapotranspiration to deplete soil moisture from greater
depths, generally increasing the total volume of
evapotranspiration. However, since the total
evapotranspiration demand remains constant, a smaller
volume of water depletion occurs per unit depth. Conse-
quently, the average moisture content throughout the
evaporative zone would be higher, resulting in a higher
runoff curve number and, therefore, larger runoff.
However, when the time period between storms is suffi-
ciently long, evapotranspiration demand is able to
deplete soil moisture to equal levels with either small or
large evaporative depths. In this case, runoff volume
could decrease with increasing evaporative depth since
antecedent moisture conditions would remain the same
and the increased storage volume in the deeper evapora-
tive zone would increase the infiltration capacity.
The effect of evaporative depth on the volume of
lateral drainage and percolation is directly related to the
composite effect on evapotranspiration and runoff. In the
examples chosen for Table 9-6, the increase in
evapotranspiration with increased evaporative depth was
greater than any increase in infiltration; therefore, lateral
drainage and percolation always decreased.
An increase in evaporative depth caused an increase in
infiltration for the two-layer cover compared to a slight
decrease for the three-layer cover. This difference re-
lates to the different mechanisms controlling infiltration in
these two cases. For the two-layer cover, the hydraulic
conductivity of the clay liner was much less than the
sandy loam topsoil. Therefore, infiltration tended to
saturate the topsoil layer, and the total volume of infiltra-
tion was dependent primarily on the volume of storage
available in this layer. A larger evaporative depth in-
creased the potential for a larger volume of available
storage and thus for more infiltration. For the three-layer
cover, the lateral drainage layer generally maintained a
free drainage condition at the topsoil/lateral drainage
layer interface. Infiltration was then controlled primarily
by the hydraulic conductivity of the topsoil and the avail-
able storage in the top segment of the subprofile. As ex-
plained above, this condition could result in either an
increase or decrease in infiltration with an increase in
evaporative depth.
Summarizing the effects of evaporative depth, an in-
crease in evaporative depth produces an increase in
evapotranspiration and, therefore, generally a decrease
in lateral drainage and percolation. The effects on runoff
are mixed but typically very small. The size of the chan-
ges are difficult to predict because the effects of evapora-
tive depth changes are indirect. Changing the
evaporative depth changes the potential storage in the
potential storage in the evaporative zone that may not
significantly change the net evapotranspiration. As
evidence of this, the change in evapotranspiration is very
small when the evaporative depth is increased beyond 46
cm (18 in.). In addition, the topsoil characteristics,
climate, and design affect the response to a change in
evaporative depth.
Effects of Dralnable Porosity
Drainable porosity is defined as the difference between
porosity and field capacity; that is, the amount of water
that could be vertically drained from a saturated soil by
gravity forces alone. Values ranged from 0.254 to 0.686
cm/cm (0.100 to 0.270 in./in.) in this study. These values
represent the volume of moisture storage capacity in ex-
cess of field capacity, divided by the bulk volume of soil
including voids. Values for field capacity and wilting point
remained constant at 0.668 and 0.338 cm/cm (0.263 and
0.133 in./in.), respectively. Only sandy loam soil was
considered. The evaporative depth was 25 cm (10 in.),
and the SCS curve number was 75. Both two- and three-
layer cover designs were simulated. The results are
presented in Table 9-7.
An increase in drainable porosity increases the moisture
storage volume above field capacity and decreases un-
saturated hydraulic conductivity for a given moisture con-
tent given a constant saturated hydraulic conductivity.
Therefore, more water can infiltrate and be made avail-
able for evapotranspiration during vertical drainage. This
increases the volume of evapotranspiration and
decreases the volume of lateral drainage and percolation
as shown in Table 9-7. However, the effect of increased
drainable porosity on runoff is varied. For the three-layer
cover, runoff decreased slightly at Santa Maria and in-
creased slightly at Shreveport. For the two-layer cover,
104
-------
Table 9-6. Effects of Evaporative Depth and Runoff Curve Number
Description1
Average Annual Volume (Percent Precipitation)
Three-Layer Cover Design
Two-Layer Cover Design
Site
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
LA
Soil
Type
SL
SL
SL
SICL
SICL
SICL
SL
SL
SL
SICL
SICL
SICL
SL
SL
SL
SICL
SICL
SICL
SL
SL
SL
SICL
SICL
SICL
Evap.
Depth
(in.)
10
10
10
10
10
10
4
10
18
4
10
18
10
10
10
10
10
10
4
10
18
4
10
18
SCS
Curve
Number
65
80
90
75
85
95
75
75
75
85
85
85
65
80
90
75
85
95
75
75
75
85
85
85
Runoff
0.1
2.6
11.3
5.5
12.7
34.4
1.1
1.1
1.3
12.6
12.7
12.0
0.5
4.2
15.3
5.8
13.5
36.5
2.0
2.1
2.3
12.4
13.5
14.3
ET3
52.7
51.9
49.5
70.8
67.6
57.3
41.3
52.4
61.9
53.3
67.6
77.0
52.1
50.9
47.1
71.2
69.6
59.0
38.8
51.6
62.4
55.6
68.1
75.8
Lat.4
Drng.
43.6
41.9
35.9
22.1
18.0
6.4
53.3
42.9
34.1
30.5
18.0
11.2
44.1
41.6
34.5
20.3
14.5
3.0
55.7
43.0
32.0
28.8
14.4
8.1
Liner
Perc.
4.2
4.2
4.1
2.2
2.2
1.6
4.5
4.2
3.9
3.7
2.2
1.2
3.1
3.1
3.0
2.3
2.2
1.4
3.2
3.1
3.0
2.0
2.1
1.2
Runoff
7.1
8.7
14.4
8.9
7.8
6.9
2.0
5.1
15.6
8.2
3.3
3.0
ET3
53.8
53.0
50.4
42.9
53.4
63.8
57.9
55.9
49.1
45.1
57.0
66.5
Liner6
Perc.
39.9
39.1
36.0
48.5
39.6
30.6
39.4
38.3
34.8
45.2
39.0
30.2
1CA = Santa Maria, CA; LA = Shreveport, LA; SL = sandy loam (HELP model default texture 6); SICL = silty, clayey loam (HELP
model default texture 12). Fair grass and 46-cm (18-in.) topsoil layer was used for all cases.
2Change in storage is not included in this table; therefore, the water balance components shown do not always add up to 100.0
percent.
3ET = evapotranspiration.
4Lateral drainage from a 31-cm (12-in.) layer having a slope of 3 percent, a drainage length of 61 m (200 ft), and a hydraulic
conductivity of 3 x 10"2 cm/sec.
5Percolation through 61-cm (24-in.) liner having a hydraulic conductivity of 10~7 cm/sec.
6Percolation through 61-cm (24-in.) liner having a hydraulic conductivity of 10~6 cm/sec.
runoff decreased significantly at both locations since the
relative soil moisture is lower and the available storage is
greater. An increase in drainable porosity reduces the
head or depth of saturation resulting from a fixed quantity
of infiltration. This decreases the lateral drainage while
having only small effects on percolation. The design and
climate affects the magnitudes of the changes in the
water budget components.
Effects of Plant Available Water Capacity
Plant available water capacity is defined as the difference
between field capacity and wilting point, or the amount of
water available for plant uptake after vertical drainage by
gravity has ceased. Values ranged from 0.178 to 0.508
cm/cm (0.070 to 0.200 in./in.) in this analysis. These
values represent the volume of potential moisture storage
between wilting point and field capacity, divided by the
105
-------
Table 9-7. Effects of Drainabie Porosity and Plant Available Water Capacity
Description1
Site
CA
CA
CA
CA
CA
CA
LA
LA
LA
LA
LA
LA
DP
0.18
0.18
0.18
0.10
0.18
0.27
0.18
0.18
0.18
0.10
0.18
0.27
PAWC
0.07
0.13
0.20
0.13
0.13
0.13
0.07
0.13
0.20
0.13
0.13
0.13
Average Annual Volume (Percent Precipitation)2
Three-Layer Cover Design Two-Layer Cover Design
Runoff
1.07
1.14
1.30
1.17
1.14
1.1
2.08
2.15
2.26
2.10
2.15
2.2
ET3
48.51
52.54
56.43
48.87
52.53
55.8
47.38
51.74
55.68
46.93
51.74
55.7
Lat.4
Drng.
46.45
42.83
39.43
47.38
42.81
39.6
47.12
42.86
38.92
47.66
42.86
38.8
Liner5
Perc.
4.31
4.22
4.12
4.33
4.22
4.1
3.12
3.08
3.04
3.12
3.08
3.0
Runoff
8.57
7.87
7.06
10.48
7.87
5.22
4.36
3.45
2.98
6.63
3.45
2.32
ET3
49.78
53.55
57.18
50.40
53.55
57.34
54.57
57.05
59.99
55.24
57.05
59.60
Liner6
Perc.
42.16
39.41
37.02
40.02
39.41
38.20
40.08
38.84
36.69
37.65
38.84
37.49
1CA = Santa Maria, CA; LA = Shreveport, LA; DP = drainable porosity (vol/vol); PAWC = plant available water capacity
(vol/vol). All cases are for 46 cm (18 in.) of sandy loam topsoil (HELP model default texture 6); fair grass; evaporative depth
= 25 crn (10 in,); and curve number = 75.
2Change in storage is not included in this table; therefore, the water balance components shown do not always add up to
100.0 percent.
ET = evapotranspiration.
4Lateral drainage from a 31-cm (12-in.) layer having a slope of 3 percent, a drainage length of 61 m (200 ft), and a hydraulic
conductivity of 3 x 10"2 cm/sec.
5Percolation through 61-cm (24-in.) liner having a hydraulic conductivity of 10~7 em/sec.
6Percolation through 61-cm (24-in.) finer having a hydraulic conductivity of 10" cm/sec.
bulk volume of soil including voids. The values for wilting
point and drainable porosity remained constant at 0.338
and 0.457 cm/cm (0.133 and 0.180 in./in.), respectively.
Only sandy loam soil was considered. The evaporative
depth was 25 cm (10 in.), and the SCS runoff curve num-
ber was 75. Both two- and three-layer cover designs
were simulated. The results are presented in Table 9-7.
Increasing the plant available water capacity provides a
greater volume of water available for evapotranspiration
after vertical drainage has nearly ceased. This results In
larger volumes of evapotranspiration as shown in Table
9-7. Consequently, the lateral drainage and percolation
decreases. The change in the volume of runoff was
design dependent. Since increasing the plant available
water capacity results in an increased moisture content
at field capacity, there is a greater potential for higher an-
tecedent moisture conditions or relative moisture content,
resulting in a higher curve number. As such, the runoff
for the three-layer cover systems increased with increas-
ing plant available water capacity. Runoff decreased for
the two-layer cover systems because infiltration is limited
by the storage volume above the liner. As such, increas-
ing the plant available water capacity increases the
storage volume, reducing the limits on infiltration and the
runoff. As shown in Table 9-7, the runoff from the two-
layer cover approaches the runoff from the three-layer
cover as the storage potential in the two-layer cover
becomes large, that is for large values of drainable
porosity and plant available water capacity. In all
cases the increases in evapotranspiration were
great enough to offset any decrease in runoff; therefore,
leachate drainage and percolation always decreased.
The size of the changes in the water budget components
were dependent on the climate and design. The results
would also be dependent on the type of topsoil.
Liner/Drain Systems
This section examines the effects of liner/drain system
design on the performance of the drain system under
conditions typical of cover systems, and leachate collec-
tion systems in open and closed landfills. Performance
was determined by the apportionment of the drainage
into the drain layer between lateral drainage and percola-
tion through the liner. In addition, the effect of design on
the resulting depth of saturation also was examined. For
106
-------
Table 9-8. Sensitivity of Lateral Drainage and Liner Percolation to Lateral Drainage Slope and Length
Annual1
Infill.
(in.)
50
50
50
50
50
50
50
50
8
8
8
8
8
8
8
8
Avg, Annual Vol.
(% Inflow)
Slope
S
(ft/ft)
0.01
0.01
0.01
0.03
0.03
0.03
0.09
0.09
0.01
0.01
0.01
0.03
0.03
0.03
0.09
0.09
Length
L
(«)
25
75
225
25
75
225
25
75
25
75
225
25
75
225
25
75
S*L
(ft)
0.25
0.75
2.25
0.75
2.25
6.75
2.25
6.75
0.25
0.75
2.25
0.75
2.25
6.75
2.25
6.75
US
(ft)
2,500
7,500
22,500
830
2,500
7,500
280
830
2,500
7,500
22,500
830
2,500
7,500
280
830
Lat.2
Drng,
96.71
95.89
93.43
96.85
96.36
95.10
97.37
96.87
83.73
82.29
78.51
84.16
83.59
82.28
84.35
84.23
Liner3
Perc.
3.29
4.11
6.57
3.15
3.64
4.90
2.63
3.13
16.27
17.71
21.49
15.84
16.41
17.72
15,65
15.77
Max. Head
In Lat.
Drng. Layer
(in.)
13.8
29.7
58.2
12.3
24.8
42,3
8.5
16.2
1.2
3.4
9.4
0.5
1.1
3.5
0.2
0.4
1Value of 50 in./yr represents inflow through an open landfill; the temporal distribution is based on rainfall records for Shreveport,
LA. Value of 8 in./yr represents inflow through landfill cover; the temporal distribution is uniform throughout the year.
2Latera! drainage from a layer having a slope of 3 percent, drainage length of 75 ft, porosity of 0.351 vol/vol, field capacity of
0.174 vol/vol, and a saturated hydraulic conductivity of 10 cm/sec
3Percolation through a 24-in.-thick soil liner having a saturated hydraulic conductivity of 10"7 cm/sec.
the cover system or open landfill the drainage into the
drain layer was 127 cm/yr (50 in./yr), distributed tem-
porally in accordance with the precipitation at
Shreveport. For the closed landfill the drainage into the
drain layer was distributed uniformly through time at a
rate of 20 cm/yr (8 in./yr).
Four types of liner/drain systems are examined in the
various parts of this study to determine their perfor-
mance: a sand drainage layer underlain by a clay liner, a
sand drainage layer underlain by a geomembrane, a
sand drainage layer underlain by a composite liner, and
double liner systems. For the clay liner system this sen-
sitivity analysis determines the effects of the saturated
hydraulic conductivity of the liner and drain layer, slope of
the liner, and drain spacing. For the geomembrane and
composite liner systems, the effects of synthetic liner
leakage fraction and saturated hydraulic conductivity of
the geomembrane's subsoil are examined. The sen-
sitivity of the parameters affecting the synthetic liner
leakage fraction are presented graphically. For the
double liner systems, the effectiveness of several dif-
ferent systems in preventing and detecting leakage from
the primary liner prior to teaking through the secondary
liner was compared. In all systems the thickness of the
drain layer was greater than the peak depth of saturation
in the drain layer, and the thickness of the clay liner or
subsoil below a geomembrane was 61 cm (24 in.),
Clay Liner/Drain Systems
Saturated Hydraulic Conductivities. The liner/drain sys-
tem used in this analysis is shown as Design A in Figure
9-10. The value of KD (the saturated hydraulic conduc-
tivity of the drain layer) ranged from 0.001 to 1 cm/sec
while the value of KP (the saturated hydraulic conduc-
tivity of the clay liner) ranged from 10 to 10~5 cm/sec.
The slope of the liner surface toward the drainage collec-
tor was 3 percent, and the maximum drainage length to
the collector was 23 m (75 ft). The results of the
drainage efficiency determinations for the various com-
binations of KD and KP are shown in Figure 9-4, where
the average annual volumes of lateral drainage and per-
colation expressed as a percentage of annual inflow are
plotted.
For the large unsteady inflows totaling 127 cm/yr (50
in./yr), only designs where the saturated hydraulic con-
ductivity of the liner was equal to or less than 10~7 cm/sec
limited the percolation through the liner to volumes less
than 5 percent of the annual inflow (6.4 cm [2,5 in.]). The
effect of KD on the drainage efficiency for these low per-
meability liners is fairly small. Changing KD from 0.001
cm/sec to 1 cm/sec reduced the percolation from 7 per-
107
-------
cent to 1 percent of the inflow for a KP of 10~7 cm/sec
and from 0.7 percent to 0.1 percent for a KP of 10~8
cm/sec. For a KP value of 10 cm/sec, only a KD value
of 1 cm/sec or greater can reduce the percolation to less
than 10 percent of the annual inflow. Liners having a KP
of 1CT5 cm/sec are largely ineffective no matter how large
the value of KD is,
For smaller steady inflows of 20 cm/yr (8 in./yr) typical
of the infiltration through some cover systems, only
liners having a value of KP equal to or less than 10
cm/sec limited leakage except for designs having a KP
of 10"6 cm/sec and a very large KD value, 1 cm/sec or
greater. As above, the effect of KD on the drainage ef-
ficiency is small. Changing KD from 0.001 cm/sec to 1
cm/sec reduced the percolation from 22 percent to 15
percent of the inflow for a KP of 10~7 cm/sec and from
2.3 percent to 1.5 percent for a KP of 10~8 cm/sec.
Liners having a KP of 10~7 cm/sec leaked between 2.5
and 5.1 cm/yr (1 and 2 in./yr) while liners having a KP
of 10~8 cm/sec leaked between 0.25 and 0.51 cm/yr
(0.1 to 0.2 in./yr).
Summarizing the results shown in Figure 9-4, the
saturated hydraulic conductivity of the liner is the primary
control of leakage through a clay liner. At hydraulic con-
ductivities below about 10"6 cm/sec the leakage is nearly
proportional to the value of KP; that is, an order of mag-
nitude decrease in the value of KP yields nearly an order
of magnitude decrease in percolation. The value of KD
has only a small effect on the leakage through liners
having a KP of 10~7 cm/sec or less. Changing the value
of KD by three orders of magnitude when using these low
permeability liners yields much less than an order of
magnitude change in percolation.
Similar effects are also seen in Figures 9-5 and 9-6
which relate the KD/KP design ratio to the resulting ratio
of lateral drainage to percolation. The curves in Figure 9-
5 are log-least-squares regressions for several ranges of
steady-state heads resulting from a steady-state inflow of
20 cm/yr (8 in./yr). The curves in Figure 9-6 are log-least-
squares regressions for several ranges of peak "y result-
ing from a unsteady inflow of 127 cm/yr (50 in./yr). The
plotted points are QD/QP ratios for the given KD/KP
ratio; their symbols indicate the value of KD used in ob-
taining the result. The actual steady-state "y and peak "y
values were both grouped into four ranges of heads. In
Figure 9-5 steady-state heads ranging from 26 to 30.7
cm (10.2 to 12.1 in.) were grouped together as were
heads ranging from 3.56 to 4.06 cm (1.4 to 1.6 in.),
equaling 0.508 cm (0.2 in.), and less than 0.127 cm (0.05
in.). In Figure 9-6 peak heads ranging from 6.1 to 6.4 cm
(2.4 to 2.5 in.) were grouped together as were heads
ranging from 19.3 to 23.6 cm (7.6 to 9.3 in.), from 41.15
to 69.6 cm (16.2 to 27.4 in.), and from 116.1 to 153.2 cm
(45.7 to 60.3 in.).
Figures 9-5 and 9-6 show that percolation tends to
dominate at ratios of KD/KP below 107. This is par-
ticularly true as the depth of saturation or inflow
decreases. When heads remain constant, the ratio of
lateral drainage to percolation is a linear function of
KD/KP. Using the maximum head allowed by RCRA of
31 cm (12 in.) and the current minimum KD/KP ratio im-
plied by RCRA of 105, a percolation of 2.3 percent of in-
flow results; however, an unusually large steady-state
inflow of 203 cm/yr (80 in./yr) or 0.559 cm/day (0.22
in./day) is required to achieve this condition. When using
the RCRA guidance design, therefore, the peak and
steady-state average heads will be considerably smaller
than 31 cm (12 in.) at virtually all locations.
Slope and Drainage Length. The combinations of slope
and drainage length used in this analysis are listed in
Table 9-8 along with resulting average annual volumes of
lateral drainage and percolation expressed as a percent-
age of annual inflow. The table also contains the result-
ing maximum heads above the soil liner. The slope (S)
ranged from 0.003 to 0.028 cm/cm (0.01 to 0.09 ft/ft) (1 to
9 percent) while the drainage length {[_) ranged from 8 to
69 m (25 to 225 ft). The saturated hydraulic conduc-
tivities of the lateral drainage and soil liners were 10"2
and 10~7 cm/sec, respectively. The product S*L and the
ratio US ranged from 0.76 to 2 m (0.25 to 6 ft) and 85 to
6,858 m (280 to 22,500 ft), respectively. S*L is the head
contributed by the liner at the crest of the drainage layer.
The results indicate that the volumes of lateral drainage
and percolation vary little with changes in slope and
drainage length under both steady and unsteady inflows.
A ninefold increase in slope reduced the percolation by a
maximum of 25 percent for the unsteady inflow and 13
percent for the steady inflow. As the drainage length is
reduced and the slope increased, the lateral drainage
rate increases. As a result, the head decreases and is
maintained at smaller depths for shorter durations. Con-
sequently, the percolation decreases since it is a function
of the head on the liner. A ninefold decrease in drainage
length reduced the percolation by a maximum of 50 per-
cent for the unsteady inflow and 25 percent for the steady
inflow. A ninefold increase in slope and decrease in
length decreased the percolation by about 60 percent for
the unsteady inflow and about 30 percent for the steady
inflow.
The head in the drain layer varies greatly with changes in
slope and drainage length. For a steady inflow the
average head increases linearly with an increase in
drainage length and an increase in the inverse of the
slope, as shown in Figure 9-7. A similar relationship ex-
ists between the peak average head during the simula-
tion and L/S for unsteady inflow. The average head is
slightly influenced by the product of the slope and
drainage length when the head is similar to this product.
108
-------
o
111
CD
or
a:
a
cn
LLJ
f—
or
100
90
80
70
60
50
40
30
10 -
0
V
KQ = I cm/s
KD = 0.1 cm/s V^
KD - 0.0! cm/s \
KD = 0.001 cm/s X
o 50 in./yr Inflow
o 8 In./yr SS Inflow
10"
,-7
10 ' 10
KP Ccm/s)
Figure 9-4. Effect of saturated hydraulic conductivity on lateral drainage and percolation.
Geomembrane/Drain Systems
A single synthetic liner under a drain layer as shown in
Design B in Figure 9-10 is examined in this section. It is
assumed that the synthetic liner was laid directly on a
3-m (10-ft) thick layer of native subsoil. The drainage
layer had a saturated hydraulic conductivity of 1CT2
cm/sec, a slope of 3 percent, and a drainage length of
23 m (75 ft). This case will be used to demonstrate the
influence of the synthetic liner leakage fraction and the
saturated hydraulic conductivity of the native subsoil on
the liner system performance. The properties of the sub-
soil ranged from sand to clay in the analysis.
Liner Leakage Fraction. Brown et al. (4) conducted
laboratory experiments and developed predictive equa-
tions to quantify leakage rates through various size holes
in synthetic liners over soil. They assumed that the
measured leakage rates corresponded to a uniform ver-
tical percolation rate equal to the saturated hydraulic con-
ductivity through a circular cross-sectional area of the soil
liner directly beneath the hole. Using the data relating
leakage and cross-sectional area of flow, Brown et al. (4)
developed predictive equations for the radius or area of
this flow cross section as a function of hole size, depth of
leachate ponding, and saturated hydraulic conductivity of
the soil. Figure 9-8 presents their results. The radius of
saturated flow through the subsoil was significantly
greater than the radius of the hole in the synthetic liner.
In this paper, the cross-sectional area of saturated flow
was multiplied by the number of holes per unit area of
synthetic liner to compute the synthetic liner leakage frac-
tion. Liner leakage fraction is simply defined as the total
horizontal area of saturated flow through the subsoil
beneath all of the liner holes divided by the horizontal
area of the liner.
Liner leakage fraction is a function of many parameters,
some quantitatively defined and others qualitatively
defined. Liner leakage fraction increases linearly with in-
creases in the number of holes of the same size and
shape. Shape also has a strong effect on the leakage;
tears have larger leakage than punctures. Increasing the
size of circular holes yields only a slight increase in the
leakage, while increasing the length of a tear or bad
seam increases the leakage nearly linearly. Leakage
also increases nearly linearly with increases in head or
depth of saturation above the liner. The leakage fraction
also is affected by the gap width between the liner and
the subsoil. Gap width is a measure of the seal between
the liner and the subsoil. The smaller the gap the better
the seal. The sea! is a function of the subsoil, installa-
tion, liner placement, and subsoil preparation. Installa-
tion of the liner on coarse-grained subsoil, clods, debris,
or filter fabric provides a poor seal as will wrinkles in the
liner. Coarse-grained subsoils decrease the leakage
fraction while greatly increasing the leakage. The greater
permeability of coarse materials allows greater flow
through a smaller area of saturated flow, reducing the
109
-------
Q.
GJ
\
D
a
io io io
KD/KP
to
io
Figure 9-5. Effect of ratio of drainage-layer saturated hydraulic conductivity to soil-liner saturated hydraulic conductivity
on ratio of lateral drainage to percolation for steady-state (SS) inflow of 20 cm/yr (8 in./yr).
Q.
a
\
a
a
O KD •• 1 cm/a
a KD - O.1 cm/e
O KD - O.OI cm/a
* KD = O.OOI cm/s
Pack y ss 2.5 In.
Peak y ~ O In.
Peak y = 24 In.
Peak y" = 55 In.
io
KD/KP
io
8
1O
7
Figure 9-6. Effect of ratio of drainage-layer saturated hydraulic conductivity to soil-liner saturated hydraulic conductivity
on ratio of lateral drainage to percolation for unsteady Inflow of 127 cm/yr (50 inJyr).
110
-------
Q
-------
o
cc
CD
Q.
en
O3
c
c
CD
Q.
CD
CD
XI
e
13
10=
10'
10-
10'
10'
10°
Upper bound is for 0.08-cru-d ia. openings
Lower bound is for I.27-cm-dia. openings
10-'
KP = 3.4 x 10
1Q-6
10
i-S
10"
10
-3
10
-2
10"
01
D>
C
C
O
a.
o
c
a>
o
CD
m
O1
c
o
a
a.
ai
43
e
i_
o
- 500 _
10U
Synthetic Liner Leakage Fraction, LF
Figure 9-8. Synthetic liner leakage fraction as a function of density of holes, size of holes, head on the liner, and saturated
hydraulic conductivity of the liner.
100 r— — . =•=• ^j — __i« I 0
2B
O
CL
GJ
to
10-10 10-9 10-8
LF x KP Ccm/s)
Figure 9-9. Effect of leakage fraction on system performance.
112
-------
layer and vertical percolation through each synthetic liner
and each soil liner. These predictions were based on 20
cm/yr (8 in./yr) of infiltration passing through the waste
layer and reaching the primary leachate collection sys-
tem. This inflow was distributed uniformly in time.
Figures 9-11 and 9-12 show the results in terms of lateral
drainage from the secondary drainage layer and vertical
percolation through the bottom soil liner as functions of
synthetic liner leakage fraction of the top membrane.
Design C consists of a primary leachate drainage layer
underlain by a synthetic liner, a secondary drainage
layer, and a soil liner. As shown in Figure 11, this design
is not very effective. Large quantities of leakage oc-
curred at fairly low leakage fractions and no leakage
(lateral drainage) was detected from the secondary
drainage layer until the synthetic liner leakage fraction
exceeded about 10~5. At smaller synthetic liner leakage
fractions, the leachate percolated vertically through the
soil liner as fast as the leakage through the synthetic liner
occurred. The product of the saturated hydraulic conduc-
tivity of the secondary drainage layer times the synthetic
liner leakage fraction must be greater than or ap-
proximately equal to the saturated hydraulic conductivity
of the soil liner before leakage will be detected using this
design. At the time leakage is detected, the vertical per-
colation rate through the soil liner could be about 16 per-
cent of total inflow.
Design D consists of a primary drainage layer underlain
by a synthetic liner, a soil liner, a secondary drainage
layer, and a second soil liner. The soil liner immediately
below the synthetic liner is very effective in minimizing
vertical percolation (leakage through the primary liner);
however, a synthetic liner leakage fraction greater than
10"2 to 10~1 would be required before leachate would be
collected from the secondary drainage layer. Because
the vertical percolation through the first liner is so small,
practically all of the leakage is removed by vertical per-
colation through the bottom soil liner as shown in Figure
9-12. This design is ineffective since the leakage detec-
tion system would not function.
Design E consists of a primary drainage layer underlain
by a synthetic liner, a secondary drainage layer, a
second synthetic liner, and a soil liner. In this case, any
leakage through the upper synthetic liner will readily pass
through the underlying drainage medium to the lower
synthetic liner. Since the lower synthetic liner is under-
lain by a soil liner, most leakage will be collected by
lateral drainage. Figure 9-11 shows that leakage will be
detected far in advance of significant vertical percolation
from the landfill. That is, the leakage fraction of the syn-
thetic liners at which leakage detection will occur is
several orders of magnitude smaller than the leakage
fraction at which significant vertical percolation from the
landfill will occur. The leakage lost by percolation is vir-
DESIGN A
DESIGN B
DESIGN C
',:<= WASTE LAYER'
:• -.^j*-^, -„. -,-0
J_* • f~v ^*-J *'^. .r-> O-
DRAIN LAYER
\
\ \ \ \
SOIL LINER
.DRAIN LAYER •
jr-~SYNTHETIC LINER
\ \ \ '\ \^
\NATIVE SUBSOIL^
\ \ \ \ \
' -.'w A ST E T AY'E R i,-
DRAIN LAYER
; ^^—SYNTHETIC L INER
DRAIN LAYER
\\ \ \ \
\ SOIL LINER \
\ \ \ \ \
DESIGN D
DESIGN E
DESIGN F
^".VA^STE LAYER °
;DRAIN LAYER
^SYNTHETIC L INER
\\ \ \"~\r
v NSOIL LINER \
\ \ \ \ \
DRAIN LAYER
\ \ \ \ \
XSOIL LINER \
\\v\\
.-
1'WASTE LAYER/
••u.'-:vvv--'v°a^
DRAIN LAYER
LAYER;
SYNTHETIC L INER
SOIL LINER \
\ \ \ \
LAYER *
DRAIN LAYER
\\ \\ \"
SOIL LINER \
\ \ \ \ \
VDRAIN LAYERS.
— ':^— SYNTHETIC LINER
\
\ \ \ \
-SOIL LINER \
\ \ \ \
Figure 9-10. Liner designs.
113
-------
100
80
-if eo-
40 -
20-
0
o
**—
c
»*«
«^x
G3
QP
on
uU
.___ Design E
Design C
Design C
Design E
tO"7 ID"8 IO"5 ID"4 ID"3 IO"2 10"' 10°
Synthetic Liner Leakage Fraction. LF
Figure 9-11, Percent of inflow to primary leachate collection layer discharging from leakage detection layer and bottom
liner double-liner systems C and E.
3:
O
(3
O.Oi 0.10 I.00
Synthetic Liner Leakage Fraction, LF
Figure 9-12. Percent of inflow to primary leachate collection layer discharging from leakage detection layer and bottom
liner for double-liner systems D and F.
114
-------
tually the same as for Design D but detection is much
better. This design is effective at minimizing leakage
from the landfill and at detecting leakage through the
primary liner, but significant leakage through the primary
liner may occur at fairly low liner leakage fractions.
Design F consists of a primary drainage layer underlain
by a synthetic liner, a soil liner, a secondary drainage
layer, a second synthetic liner, and a second soil liner.
Figure 9-12 shows that the addition of the lower synthetic
liner improves the system performance in comparison to
the performance of Design D. Leakage is detected
whenever leakage occurs. Even at leakage fractions of
10~3 when only 0.02 percent of the inflow leaks through
the primary liner, half of the leakage is collected in the
secondary drainage layer. The depth of saturation in the
secondary drainage layer is lower than in the primary
layer. This sufficiently reduces the leakage through the
second synthetic liner to permit detection whenever the
primary liner leaks. Design F is a very effective double-
liner design because it minimizes the leakage through
the primary liner and from the landfill and collects
leakage at all leakage fractions.
A comparison of the four designs shows that Design F is
the most effective in detecting the earliest leaks with the
least amount of vertical leakage through the primary liner
and also through the bottom soil liner. Design D yields
the same quantity of leakage through the primary liner;
however, leakage in Design D would probably never be
detected or collected. Therefore, the bottom liner in
Design D is not functional. Designs D and E yield the
same leakage through the bottom liner but Design E
detects leakage through the primary liner at the lowest
leakage fraction. Design C also detects leaks at very
small leakage fractions but allows significant vertical per-
colation through the bottom soil liner before detection.
The leakage through the primary liner in Designs C and
E is large even at low leakage fractions. Therefore, syn-
thetic membranes placed on highly permeable subsoils
are ineffective except for very low inflows and for very
low leakage fractions. Synthetic membranes are best
used in conjunction with a low-permeability soil as a com-
posite liner. Comparison of the results for Designs B and
C demonstrates this point. Both designs are composed
of one synthetic membrane and one soil liner, but the
leakage from the composite liner (Design B) shown in
Figure 9-9 as the curve for 20 cm/yr (8 in./yr) steady in-
flow is much lower than the leakage from the double liner
system (Design C) as shown in Figure 9-11.
It is interesting to compare the single-liner performance
of Design B to the double-liner performance of Design D,
assuming the soil-liner-saturated hydraulic conductivity in
Design B is the same as Design D. The vertical percola-
tion leaving the system in Design B is essentially the
same as that leaving the secondary liner in Design D as
seen by comparing Figure 9-12 to the curve in Figure 9-9
for 20 cm/yr (8 in./yr) steady inflow. The secondary liner
in Design D is nonfunctional since the percolation rate of
the second soil liner is generally equal to or greater than
the leakage rate.
SUMMARY OF SENSITIVITY ANALYSIS
The interrelationship between variables influencing the
hydrologic performance of a landfill cover is complex. It
is difficult to isolate one parameter and exactly predict its
effect on the water balance without first placing restric-
tions (sometimes severe restrictions) on the values of the
remaining parameters. With this qualification in mind, the
following general summary statements are made.
The primary importance of the topsoil depth is to control
the extent or existence of overlap between the evapora-
tive depth and the head in the lateral drainage layer. The
greater this overlap, the greater will be evapotranspira-
tion and runoff. Surface vegetation has a significant ef-
fect on evapotranspiration from soils with long
flow-through travel times and large plant available water
capacities; otherwise, the effect of vegetation on
evapotranspiration is small. The general influence of sur-
face vegetation on lateral drainage and percolation is dif-
ficult to predict outside the context of an individual cover
design. Clay soils increase runoff and evapotranspiration
and decrease lateral drainage and percolation. Simula-
tions of landfills in colder climates and in areas of lower
solar radiation are likely to show less evapotranspiration
and greater lateral drainage and percolation. An in-
crease in the runoff curve number will increase runoff and
decrease evapotranspiration, lateral drainage, and per-
colation. As evaporative depth, drainable porosity, or
plant available water increase, evapotranspiration tends
to increase and lateral drainage and percolation tend to
decrease; the effect on runoff is varied.
The sensitivity analysis shows that the ratio of lateral
drainage to percolation is a positive function of the ratio
of KD/KP and the average head above the liner.
However, the average head is a function of QD/QD and
US. The quantity of lateral drainage, and, therefore, also
the average head, is in turn a function of the infiltration.
Therefore, the ratio of lateral drainage to percolation in-
creases with increases in infiltration and the ratio of
KD/KP for a given drain and liner design. The ration of
lateral drainage to percolation for a given ratio of KD/DP
increases with increases in infiltration and the term S/L.
The percolation and average head above the liner is a
positive function of the term US.
Leakage through geomembrane increases with the num-
ber and size of holes, the depth of water buildup on the
liner, the permeability of the subsoil, and the gap bet-
ween the liner and the subsoil. Geomembranes reduce
leakage through liner systems by reducing the area of
saturated flow through the subsoil. The overall effective-
ness of a geomembrane system is equivalent to a soil
liner having a saturated hydraulic conductivity equal to
the product of the saturated hydraulic conductivity of the
115
-------
subsoil and the ratio of the reduced area of flow through
the subsoil to the area of the liner. Composite liners
provide the best reduction in leakage. Drain systems
that yield tow head buildup on the geomembrane improve
the performance of a geomembrane system.
REFERENCES
1. Schroeder, P. R, and Peyton, R. L 1987. "Verifica-
tion of the Hydrologic Evaluation of Landfill Perfor-
mance (HELP) Model Using Field Data." EPA 600/2-
87-050. EPA Hazardous Waste Engineering Re-
search Laboratory, Cincinnati, OH.
2. Schroeder, P. R., R. L. Peyton, and J. M. Sjostrom.
1988. Hydrologic Evaluation of Landfill Performance
(HELP) Model: Vol. Ill User's Guide for Version 2.
Internal Working Document. USAE Waterways Ex-
periment Station, Vicksburg, MS.
3. U.S. Department of Agriculture, Soil Conservation
Service. 1972. Section 4, Hydrology. In: National
Engineering Handbook, U.S. Government Printing
Office, Washington, DC. 631 pp.
4. Brown, K.W., J.C. Thomas, R.L. Lytton, P. Jayawik-
rama, and S.C. Bahrt. 1987. Quantification of Leak
Rates Through Holes in Landfill Liners. EPA/600/S2-
87-062. EPA Office of Research and Development,
Cincinnati, OH.
5. Peyton, R. L and Schroeder, P. R. 1990. "Evalua-
tion of Landfill-Liner Designs." Vol. 116, No. 3, Jour-
nal of Environmental Engineering Division, American
Society of Civil Engineers.
116
-------
CHAPTER 10
GAS MANAGEMENT SYSTEMS
GAS GENERATION
The information in this chapter applies mainly to Subtitle
D landfills. Hazardous waste landfills (Subtitle C) do not
usually contain significant amounts of organic materials
and, thus, normally have a minimal gas management
system as a component of the final cover.
Gas generation in a landfill system poses several
problems. If allowed to accumulate, gas is an explosion
hazard. It also provides stress to vegetation by lowering
the oxygen content available at the roots, severely affect-
ing the ability of the cover to support vegetation. In the
absence of adequate corridors for the gas to escape, gas
pressures can increase sufficiently to physically disrupt
the cover system as well, generating large cracks and
rupturing the geomembrane. Other problems include
odor, toxic vapors, and uncontrolled gas migration which
can cause deterioration of nearby property values.
Gas generation is a product of anaerobic decomposition
of organic materials placed in the landfill. The decom-
position can be described by the reaction:
wCO2
+ humus (1)
The composition of landfill gas generally is about 50 per-
cent methane, 40 percent carbon dioxide, and 10 percent
other gases including nitrogen products. This particular
mix of gases generally will not occur until after the landfill
becomes anaerobic. During the first year after the
materials are placed in the landfill, the gas is
predominantly carbon dioxide and is unsuitable for
recovery and use. After the methane content rises, the
gas can be mined as a fuel or energy source. However,
the BTU value of landfill gas is about half that of natural
gas and, therefore, is generally too low to substitute
directly for natural gas. Landfill gas requires purification
and is frequently used in conjunction with natural gas.
Waste decomposition rates and hence gas production
rates are moisture dependent. Highest gas production
rates occur at moisture contents ranging from 60 to 80
percent of saturation. In modern landfill design, infiltration
of water into the waste is restricted to a practical mini-
mum; therefore, optimum moisture contents may never
be achieved. Consequently, gas production rates may be
much lower than anticipated, decreasing the attractive-
ness of gas recovery systems. To maximize gas produc-
tion, strategies such as leachate recirculation should be
employed to distribute bacteria, nutrients, and moisture
more uniformly. Typical gas production rates from wet,
anaerobic wastes are about 20 to 50 mL/kg/day. These
high production rates will continue for decades. Produc-
tion at low rates may continue for centuries because of
quantity of material and resistance of some material to
biodegradation.
GAS MIGRATION
Gas migrates from landfills through two mechanisms—
convection and diffusion. Convection is transport induced
by pressure gradients formed by gas production in layers
surrounded by low hydraulic conductivity or saturated
layers. Convection also results from buoyancy forces be-
cause methane is lighter than carbon dioxide and air.
Diffusion is the transport of materials induced by con-
centration gradients. Anaerobic decomposition produces
a gas mixture with concentrations of methane and carbon
dioxide that are much greater than those found in the sur-
rounding air. Therefore, molecules of methane and car-
bon dioxide will diffuse from the landfill gas to the air in
accordance with Pick's law. Diffusion plays a much
smaller role in gas migration than convection.
Many factors affect gas migration. Some of the more im-
portant factors are the landfill design, including refuse cell
construction; final cover design; and incorporation of gas
migration control measures. Low hydraulic conductivity
soil layers and geomembranes are very effective barriers
to gas migration. Sand and gravel layers and void spaces
provide effective corridors for channeling gas migration.
Other channels affecting migration are cracks and fis-
sures between and in lifts of waste or soil due to differen-
tial settlement and subsidence.
Other factors affecting gas migration include the gas
production rate, the presence of natural and artificial con-
duits and barriers adjacent to the landfill, and climatic and
seasonal variations in site conditions. High gas produc-
tion rates increase migration. Corridors at the site ad-
jacent to a landfill such as water conduits, drain culverts,
buried lines, and sand and gravel lenses, promote uncon-
117
-------
trolled migration from the site. Barriers can include clay
deposits; high or perched water tables; roads; and com-
pacted, low hydraulic conductivity soils. Environmental
variations can result from the intermittent occurrence of
saturated or frozen surface soils, which seals the surface
and promotes lateral migration. Barometric pressure
changes also affect the rate of gas release to the surface.
Seasonal changes in moisture content can change the
gas production rate and, therefore, the extent and quan-
tity of migration.
GAS CONTROL STRATEGIES
Two gas control strategies—passive and active—are
available, and may be used at any facility. Passive sys-
tems provide corridors to intercept lateral gas migration
and channel the gas to a collection point or a vent. These
systems use barriers to prevent migration past the inter-
ceptors and the perimeter of the landfill. Active systems
generate a zone of negative pressure to increase the
pressure gradient and, consequently, the flow toward the
zone. Active systems also can be used to create a zone
of high pressure to prevent gas migration toward the
zone.
Typical passive systems are shown in Figures 10-1,10-2,
10-3, and 10-4. Figure 10-1 shows a gas-vent layer used
In conjunction with a composite liner and vent in the
cover system (2). The composite liner prevents uncon-
trolled vertical migration, while the gas-vent layer inter-
cepts all vertical migration and directs it to the vent.
Figure 10-2 shows a gravel vent that runs diagonally
down through the waste material. The gravel vent inter-
cepts both vertical and lateral migration and channels it
to the surface. Figures 10-2,10-3, and 10-4 show gravel-
filled trenches (1, 3, 4). The trenches intercept lateral
migration and direct the gas to the surface where it is
vented or extracted. Gravel-filled trenches on the
perimeter of the landfill are often used with an imperme-
able barrier on the outer side of the trench to prevent
migration from the trench to the surrounding area. These
systems often extend from the surface down to a low
hydraulic conductivity soil layer or other barrier such as
the water table or a geomembrane. The systems may be
as deep as the bottom of the landfill, or even lower if out-
side the landfill. Extreme care should be taken in the
design of all of these vent systems to prevent them from
being a source of infiltration through the cover. Improper
design could allow the vent to intercept surface runoff
and pipe additional infiltration into the leachate collection
system.
Typical active systems are shown in Figures 10-4, 10-5,
and 10-6 (4). All three figures show gas extraction wells
using exhaust blowers. The well is placed in a gravel vent
or gravel-filled trench located in the waste cell (Figures
10-5 and 10-6) or along its perimeter (Figure 10-4). The
gravel vent is sealed to prevent the well from drawing air
from the surface and destroying the suction (zone of
negative pressure) needed to draw gas to the well. The
seal also prevents infiltration of surface water. Imperme-
able barriers in the cover and perimeter walls increase
the efficiency of gas extraction wells since they restrict in-
flow of air that would dissipate the suction. In addition, it
reduces the number of wells needed and increases the
heating value of the gas collected. Typically, gas extrac-
tion wells do not extend to the bottom of the landfill since
the suction is able to draw gas from a sizable zone
beyond the gravel fill.
x—v gas veni
''(£4
drain layer CjSi^?§i
geomembrane — *
nbrane-—* "=-^r^=^^ss^sm^ pooa
ven,,ay.r{|^l|feff™^
perforated pipe ' <3
Figure 10-1. Cover with gas vent outlet and vent layer.
118
-------
-*— SLOPE
FINAL COVER MATERIAL
. I VENTED SAS _.
;,CELL -
Figure 10-2. Gravel vent and gravel-filled trench used to control lateral gas movement in a sanitary landfill (5).
WINTER CLIMATES MAY REQUIRE
COLLECTOR WITH VERTICAL RISERS
AND SURFACE SEAL
GRAVEL BACKFILL
BARRIER
MATERIAL
(IMPERVIOUS
MEMBRANE)
COVER MATERIAL
REFUSE
JL
UNDISTURBED IMPERVIOUS MATERIAL OR WATER TABLE "
Figure 10-3. Typical trench barrier system.
119
-------
"A"
Permeable Trench
Perforated
Pipe
"C"
Pipe Vent
LEGEND
Gas Migration
Refuse
Gravel
Trench Cover
Impermeable Layer
"B"
Impermeable Barrier
Perforated
Pipe
"D"
Induced Exhaust
Gas Control Barriers
Figure 10-4. Gas control barriers (6).
120
-------
GAS FLOW
D
ft GAS FLARE
EXHAUST BLOWER
IMPERVIOUS BACKFILL
^-PERFORATED PIPE
GAS FLOW
PERMEABLE MATERIAL
Figure 10-5. Gas extraction well for landfill gas control.
REFERENCES
1. Lutton, R.J., G.L. Regan, and L.W. Jones. 1979.
Design and construction of covers tor solid waste
landfills. EPA-600/2-79-165 U.S. EPA Municipal En-
vironmental Research Laboratory, Cincinnati, OH.
2. U.S. EPA. 1989. Technical guidance document: final
covers on hazardous waste landfills and surface im-
poundments. EPA/530-SW-89-047.
3. Shafer, R.A., A. Renta-Babb, J. T. Bandy, E. D.
Smith, and P. Malone. 1984. Landfill gas control at
military installations. Technical Report N-173. U.S.
Army Engineer Construction Engineering Research
Laboratory, Champaign, IL.
4. McAneny, C.C., P.G. Tucker, J.M. Morgan, C.R. Lee,
M.F. Kelley, and R.C. Horz. 1985. Covers for uncon-
trolled hazardous waste sites. EPA/540/2-85/002
U.S. EPA Hazardous Waste Engineering Research
Laboratory, Cincinnati, OH.
5. Brunner, D.R. and D.J. Keller. 1971. Sanitary landfill
design and operation. SW-66TS. U.S. Environmental
Protection Agency.
6. Rovers, F.A., J.J. Tremblay, and H. Mooij. 1977. Pro-
cedures for landfill gas monitoring and control. EPS
4-EC-77-4, Waste Management Branch, Environ-
ment Canada, Ottawa.
121
-------
PERFORATED PIPE
REFUSE
FINAL COVER - 2*.6
CLAY PLUS
FINE SANO
COARSE 6RAVE1.
Gas Extraction
Well Design
Figure 10-6. Gas extraction well design (6).
122
-------
CHAPTER 11
CASE STUDIES — RCRA/CERCLA CLOSURES
INTRODUCTION
This chapter presents five waste closure case studies.
Each study examines both design details that confirm the
suitability of the individual cap and those that may
detrimentally affect the long-term service of the facility.
The first four caps have been permitted and placed over
RCRA/CERCLA wastes. The fifth cap has been proposed
for a major municipal solid waste (MSW) landfill and
demonstrates the design problems that may be as-
sociated with such facilities.
Each design example is intended to highlight problems
that may be encountered in satisfying all aspects of
general closure criteria. The criteria that must be satisfied
are:
1. Specific minimum technology guidance (1) (MTG) ap-
plicable or appropriate and relevant to the site-
specific waste. MTG is discussed in greater detail in
Chapter 1.
2. Erosion control to limit the loss of cover soil to less
than 2 ton/acre/year, as discussed in Chapters 1 and
8.
3. Gas control systems to minimize movement of waste-
generated gases off site.
4. Ability for all systems to survive both local and global
subsidence potentials, as discussed in Chapters 1
and 2.
This chapter also raises specific concerns regarding the
use of MTG guidance blindly, without engineering confir-
mation of its suitability.
CASE 1; RCRA COMMERCIAL LANDFILL
The first closure case presents a cap over a commercial
hazardous waste disposal cell that is designed to satisfy
the basic MTG cap profile. Figure 11-1 shows details of
the general cap profile and geometry. Note that the slope
of the cap does exceed the 5 percent maximum con-
tained in the MTG criteria but is significantly flatter than
the caps on the other examples. The use of low slopes on
such facilities recognizes that the solidified waste placed
within them is very stable and will not produce significant
long-term subsidence. Such low slopes cannot be used in
applications where high long-term subsidence is a con-
cern, such as with many CERCLA and MSW closures.
This chapter examines two significant design considera-
tions for such facilities: 1) calculation of localized sub-
sidence and its impact on the cover barrier layer, and 2)
the impact of gas collection systems.
Calculation of Localized Subsidence
During the service life of this facility, it received nearly
10,000 transformers containing PCB oils (TSCA per-
mitted). The regulator expressed concern about the long-
term impact of the loss of oil and eventual collapse of the
transformer cases. Fortunately, available records
provided the location and size of the transformers. The
general subsidence model used to predict the surface
displacement of the cap due to transformer collapse was
adapted from an EPA study by Murphy and Gilbert (2)
(see Figure 11-2). The key assumption in this model is
that the volume of the surface depression is equal to the
volume of the oil leaking from the transformer. This is a
conservative assumption because it neglects the arching
that will occur within both the waste and operational soils
placed around the waste. An additional key assumption
must be made regarding the friction angle of the waste it-
self. For this case, the friction angle was assumed to be
that of the operational soils placed around the waste. For
wastes in general, such values can be measured in ac-
tual field tests (3).
The simple model for subsidence due to a single trans-
former collapse then must be applied to the actual cover
for all 10,000 transformers. The subsidences are ac-
cumulated and plotted as shown on Figure 11-3. By ex-
amining the cap's elevation contour, one can estimate
the maximum long-term relative vertical displacement of
the cap. For this case, the maximum relative displace-
ment is approximately 0.5 m in 6 m (1.8 ft in 20 ft.)
Calculation of the maximum vertical relative displacement
is important only if the designer can estimate the impact
of such displacement on the site-specific cap profile,
MTG barrier systems consist of a geomembrane and a
clay layer, both of which must be separately evaluated for
strain. The strains in the geomembrane can be estimated
using one of two models, depending on the type of an-
ticipated subsidence.
123
-------
VEGETATED TQPSCNL AND ROOT ZONE. 3 FT. MHMUM
SYNTHETIC DRAINAGE MEDIA, OEOTEXTIL6 AND OEONET
SYNTHETIC MEMBRANE (FML! »0 MIL HOPE
COMPACTED SOtt. LINER, 3 FT. MINIMUM
INTERMEDIATE COVER, 6 IN. MINIMUM
WASTE DETAIL OF COVER
SECTION THROUGH LANDFILL
Figure 11-1. Case 1—Cap profile and geometry.
For trench-like subsidence, the strains can be calculated
using the model shown in Figure 11-4. The maximum
strain that the geomembrane can tolerate in such a plane
strain condition is given by the uniaxial test data com-
monly reported by geomembrane manufacturers (see
Figure 11-5).
For spherical-type subsidence, the strains in the
geomembrane can be calculated using the method dis-
cussed in Chapter 3 of this manual. For such an as-
sumed failure mode, the designer must compare the
predicted strain with the ultimate strain limit of the
geomembrane, as obtained from biaxial testing (see
Figure 11-5). Chapter 3 gives additional data on typical
ultimate strains of common geomembranes in biaxial
loading. Most geomembranes can easily tolerate vertical
differential settlement of the cover in excess of 0.9 m in 3
m (3 ft in 10 ft) of run. This results in a factor of safety
based on an ultimate strain of 3.3 for the geomembrane
in Case 1.
The strain in the soil component of the Carrier can be es-
timated using the chart in Figure 11 -6. The specific ul-
timate tensile strain of the onsite soil can be evaluated in
a triaxial Consolidated Isotropic Undrained (CIU) test or
can be estimated from the chart in Figure 11 -7. For this
particular soil barrier, the ultimate relative strain allowable
under this criteria is 0.4 m in 3 m (1.2 ft in 10 ft) of run.
This results in a factor of safety of 1.33 for the clay com-
ponent in Case 1. If the settlements are occurring over an
extended length of time, this low factor of safety may be
acceptable due to the ability of a clay to creep. The creep
deformation of the clay allows long-term strains to
develop in the layer without a comparable increase in
stress. This is commonly referred to as "stress relaxa-
tion."
Gas Collection Systems
Commercial hazardous waste facilities generate minimal
gas due to the solidified nature of the waste. Typically,
gas collection systems for such facilities are simple linear
French drain collectors, as shown in Figure 11 -8.
Extreme caution must be exercised in designing gas
removal systems for wastes that have a long anticipated
lifetime. A gas removal system is a very efficient vehicle
for surface water to gain access to the waste if the vent
pipes become damaged. Thus, if long-term maintenance
of the cap cannot be assured, a gas collection system
may eventually cause failure of the cap to perform its
primary function—preventing surface water from reaching
the waste. Provisions should be made in the permit of
124
-------
HOM2ONTAI DISPLACEMENT
STKAM
(-) I
TENSION
l\
\
-COMPRESSION-
-TENSION-
! I
VERTICAL DISPLACEMENT
(SU8SOENCE CURVE) •
POMT OF MAXMUM
TENSU STIIAM
POMT OF MAXMUM
COMPNES9VE STIIAM
15 Murphy & Gilbert
_L
MINED OPEMNG
Figure 11-2. Case 1—General subsidence model.
such facilities for removing and sealing the gas vents if
postctosure monitoring indicates that no appreciable
quantities of gas are being generated.
As a final comment, the HELP analysis (see Chapters 8
and 9) for such caps must assume an effective leakage
for the geomembrane component of the barrier. This
leakage is commonly calculated by assuming from 9 to
13 penetrations (1 cm) per acre in the geomembrane.
The leakage through such penetrations can then be cal-
culated using the following equation (5):
Q = 3a°75h°-75Kd05
where Q = steady-state leakage rate
(M3/sec)
a = area of hole (M2)
h = head of leachate (M)
k = permeability of underlying soil
(MIS)
A revised version of HELP is being developed that will
accept such penetration data directly.
CASE 2: RCRA INDUSTRIAL LANDFILL
The second case study shows a cap profile that is be-
coming increasingly common in Europe and the United
States due to the high cost per acre of composite lined
landfills. As shown in Figure 11-9, the cap has two sig-
nificant profiles: a steep perimeter that provides for the
volume of the facility and a flatter top that covers the
majority of the waste. Figure 11-9 also shows a detail of
the cap profile which is a typical MTG profile. The key
design problems for this case involve the steep perimeter
of the cap, including both the sliding stability of such
covers and the erosion resistance of their protective sur-
face.
The slope stability of covers, or liner systems, containing
geosynthetic layers is typically of concern if the slope ex-
ceeds 8 degrees. The three horizontal to one vertical
(3H:1V) stopes of the perimeter are 18.4 degrees and,
therefore, of concern. The stability of cover and liner sys-
tems on such slopes is evaluated by performing
laboratory direct shear tests on each suspect interface to
determine the minimum factor of safety against sliding.
125
-------
Figure 11-3. Case 1—Cumulative subsidence.
20
15
10
5
=L
O
1
OfCuUf Tfounh M
-------
4000
JOOO -
S 2000 |f -—-
1000
(KEORNER,RICHARDSON-UNIAXIAl)
(STEFFEN-8IAXIAU
100
200
300
400
SOO
STRAIN. %
Figure 11-5. Case 1—Uniaxlal and biaxial geomembrane response.
1 oo
0.75
0.50
0.25
100.0
Index of maximum settlement A/L vs tensile strain (after Gilbert and Murphy, 1987)
Figure 11-6. Case 1—Subsidence strain in soil barrier.
127
-------
3.5
at
1 2.0.
"TTT 1 !—i—
LEGEND
LEONAHO t BEAM FIXXURE TESTS ! —r—
TSCHE30TARIOFF 1 DIRECT TENSION TESTS I
WES DATA ( DIRECT TENSION TESTS )
FOR THIS STUDY (OU DIRECT TENSION TESTS)
i I i i L
100
1000
PLASTICITY INDEX. %
fensde stran ** ptesuaty maex (att«f GBMrt ana Murony, 1987)
Figure 11-7. Case 1—Ultimate tensile strain In clays.
This testing procedure is described in greater detail in
Chapter 3 (pp. 3-4) and Appendix A. Steep covers requir-
ing a geomembrane commonly use three liner/drainage
layer profiles to provide a stable slope:
1. A textured HOPE or VLDPE geomembrane with
either a sand drainage layer or a drainage layer
formed using a geonet with fitter fabric bonded to
both faces.
2. A geomembrane having nonwoven geotextiles
bonded to both faces and a sand drainage layer.
3. A smooth geomembrane, sand, or geonet drain, with
an added geogrid reinforcement in the cover soil
layer to hold the layer on the slope.
The first two alternatives are examined for this case;
Case 3 discusses the third alternative.
Figure 11-10 shows direct shear data for the first alterna-
tive. Because the nonwoven material used in the bonded
geomembrane will develop the full shear strength of the
adjacent soil, direct shear tests are not commonly per-
formed for this material. Therefore, wrth both the tested
textured and bonded geomembrane, the minimal inter-
face friction angle will be significantly greater than the
18.4 degree sidestope angle. It must be noted that such
interface friction angles must be verified in laboratory
testing; not all textured geomembranes are effective.
The owner/operator must use caution in interpreting
direct shear data from evaluations of interface friction
angles. A recent full-scale field test of various cover
profiles demonstrated that the interface friction angle is
very dependent on the normal force acting on the layers (6).
Thus, direct shear data from tests for cap design using
low normal loads should not be used for designing liner
interfaces where high normal loads are anticipated. This
dependency also makes the use of interface friction
angles obtained from the literature very suspect.
Laboratory direct shear tests should be performed using
the soils, geosynthetics, and normal toads associated
with the site-specific conditions.
The steep perimeter slopes must be verified as satisfying
the MTG criteria of a cover soil loss of less than 2
tons/acre/year. The rate of soil loss is verified using the
Universal Soil Loss Equation (USLE) given by:
A = RKLSCP
where K = soil erosion factor from
Table 11-1
LS = slope constant from
Figure 11-11
R = rainfall and runoff index
C = cover management factor
P = practice factor, 0.3 to 1.0
EPA proposed a procedure to calculate an effective LS
factor for caps having two distinct slopes, as found in this
case (7), This method, however, is not currently recom-
mended because it underestimates true soil toss. For
caps having two very distinct slopes, it is more effective
to evaluate each slope independently and to provide a
runoff collection ditch, e.g., swale, between the slopes to
hydrauiicalty disconnect these features in the field. Thus,
the 3H:1V perimeter slopes of Case 2 should be
evaluated using their maximum slope length and full
128
-------
*^r,,, ^^-'
-------
FINISHED GRADE OF CAP
4" MINIMUM THICK TQPSOIL LAYER
WITH VEGETATIVE COVER.
l'-8" THICK 900T ZONE EMBANKMENT
SYNTHETIC DRAINAGE MEDIA
CAP GEOMEMBRAME
2'-0" MINIMUM THICK COMPACTED
SOIL CAP LINER
6" MINIMUM THICK FINAL INTERMEDIATE
COVER LAYER
LIMITS OF WASTE
TYPICAL CAP SECTION
SCALE: NONE
^ /—-—-—===
'\-^ e ^
''"^-•^""^ N v ^ BOTTOM OF FINAL
^—^ WAS^ ^ INTERMEDIATE COVER
•— — ' ^v
^ SUBGRAOE OF
SOIL LINER
,- TOP OF CAP
/ ELEV. 1220.00
1 SLOPE VARIES
s- TOP OF OPERATIONAL COVER
^/ 2.78%
*~~^— ELEV. 1185.70
SECONDARY
Figure 11-9. Case 2—Cap profile and geometry.
UJ
I
en
Normal Stress
Figure 11-10. Case 2—Direct shear data: texture HOPE.
130
-------
Table 11-1. Soil Texture Constant for Soil Loss
Evaluation
Organic mailer cosiest
Texture class
Sand
Find sand
Very fine sand
Loamy sand
Loamy Tine sand
Loamy very toe sand
Sandy loam
Fine sandy loam
Very fine sandy loam
Loam
Silt loam
Silt
Sandy clay lotm
Clay loam
Siity clay loam
Sandy clay
Silly clay
Clay
< 0.5 percent
K
0.05
0.16
0.42
0.12
0.24
0.44
0.2?
0.35
0.47
0.38
0.48
0.60
0.27
0.28
0.37
0.14
0.25
2 per cent
K
0.03
0.14
0.36
0.10
0.20
0.38
0.24
0.30
0.41
0.34
0.42
0.52
0.25
0.23
0.32
0.13
0.23
0.13-0.29
4 per cent
K
0.02
0.10
0.28
0.08
0,16
0.30
0.19
0.24
0.33
0.29
0.33
0.42
0.21
0.21
0.26
0.12
0.19
* The values shown arc estimated averages of broad ranges of specific-toil values. When a tenure
u near ihe borderline of two texture cium. use the average
-------
Cap Profile
il_4'PtRF P^C
DRAWftGE P^£
LOCATION
185
Figure 11-13. Case 3—Cap profile and geometry.
In addition to having sufficient tensile strength, the
geogrid must be anchored sufficiently to develop this
strength. It cannot be anchored using an anchor trench
without impinging on the waste. The geogrid in Case 3,
therefore, is anchored by running the grid continuously
over the cap and counterbalancing the weights of the
cover soils on opposing faces. While this procedure is
technically simple, it restricts, construction significantly;
the cover soil and drain must be placed in a symmetrical
manner, preferably from the top down, to tension the
geogrid. Figure 11-14a shows the geogrid being placed
over the geomembrane, and Figure 11-14b shows the
drain layer being placed on top of the geogrid.
Water collected in the surface water drainage layer must
be allowed to freely leave that system to avoid building up
head on the liner, and to maintain stability. Figure 11-15a
shows the sideslope toe drainage detail used in Case 3.
From a long-term maintenance standpoint, this drainage
system is very poor. The thin layer of loam topsoil will
readily erode at the surface interface with the geotextile
and trap rock, as shown on Figure 11-15b. Surface water
drainage layers in caps having significant slopes, such as
in Case 3, should outlet using pipe laterals placed at a
minimum of one 40-crn (4-in.) drain pipe every 61 m (200
ft) around the perimeter of the surface drain.
CASE 4: CERCLA LANDFILL CLOSURE
The fourth case study is of a cover placed over an exist-
ing MSW landfill that received 20,000 yd3 (15,292 m3) of
baghouse dust containing cadmium, chromium, and lead.
The baghouse dust was placed on top of the MSW waste
and was, therefore, highly exposed. The landfill itself was
adjacent to a community park and the local youths had
established biking paths over the landfill. Both the state
and the principal responsible party (PRP) wanted to close
the landfill in a manner that prevented surface water from
reaching the dust and discouraged the recreational use
of the cap. For these reasons, they selected a unique
hardened cap profile. Such hardened caps do not
promote recreational use of the cover and, therefore, do
132
-------
Figure 11-14a. Case 3—Placement of geogrid over geomembrane.
Figure 11-14b. Case 3—Placement of drainage layer over geogrid.
133
-------
Surface Water Drainage Layer
,_ PVC Membrane
Zone of High Erosion
12" Trap Rock
Note: cm = in. x 2.540
Figure 11-15a. Case 3—Outlet detail for sideslope toe surface water drainage layer.
sj ;
ft-;vv^ »
Ct**?^- .«Tr : '»*.,,)
Figure 11-15b. Case 3—Erosion at drainage layer outlet.
134
-------
not create an attractive nuisance in terms of maintenance
and security. Figure 11-16 shows the final cap profile and
contours.
The cap profile is significantly different than the MTG cap
in that it uses no drainage or agricultural layers. The as-
phalt and paving fabric form a unique composite barrier
with the compacted clay cap. The chip seal added to the
top of the barrier is provided to protect the asphalt and
paving fabric from ultraviolet (UV) light degradation, not
for erosion control. The "hardened" cap is advantageous
since it is not an attractive nuisance, requires very low
maintenance, and minimizes the problem of volunteer
vegetation.
The geotextile was placed over the asphalt on a surface
of asphalt emulsion (see Figure 11-17a). Rolling the
fabric over the hot emulsion fully impregnated the geotex-
tile so that it acts as a water barrier. The chip seal placed
on top of the geotextile (see Figure 11-17b) is bonded to
the geotextile by the emulsion, in a manner similar to an
industrial roofing system.
While the hardened cap is low maintenance, it does re-
quire an annual inspection and renewal of the chip seal
surface every 5 years. Additionally, the perimeter
drainage must be cleaned regularly to promote surface
water drainage. Allowable differential subsidence criteria
must be established for such caps in the same manner
as described for Case 1.
Similar hardened caps have been used on RCRA
closures in the Southeast. One particular closure at a
Department of Energy facility in Tennessee functions as
a parking lot. This particular cap replaced the agricultural
layer of the MTG profile with an asphalt and subbase
parking surface. While such caps must obviously be in-
spected on a regular basis, they can offer significant
maintenance and land use advantages.
CASE 5: MSW COMMERCIAL LANDFILL
This last example, Case 5, shows how the
basic
RCRA/CERCLA closure profiles are being adapted for
the more common MSW landfills. The cap profile shown
in Figure 11-18 includes a composite barrier layer and a
protective/agricultural soil cover. It does not include a
drainage layer between the barrier layer and the cover
soil. The drainage layer is often omitted in MSW caps. In
particular, states such as New York (8) have chosen not
to require the drainage layer due to concerns regarding
Cap Profile
— 2
20' GATE
/ CULVERT \
Figure 11-16. Case 4—Cap profile and geometry.
135
-------
Figure 11-17a. Case 4—Placement of geotextile on asphalt emulsion.
Figure 11-17b. Case 4—Placement of chip seal on geotextile.
136
-------
the impact of this layer on the agricultural growth placed
over the cap. It should be noted, however, that New York
requires a liner system beneath all new landfills that ac-
tually exceeds RCRA MTG criteria. Thus, the omission of
the drainage layer was not for financial reasons.
Contours for the Case 5 cap are shown in Figure 11-18
and reflect the dual slope profile developed in Case 2.
The general goals for the MSW cap are low maintenance,
minimization of infiltration, and aiding in gas collec-
tion/containment. MSW caps commonly cannot be con-
structed on new landfills in a single stage as can RCRA
and CERCLA caps. The staged construction of a MSW
cap is required because lined MSW landfills typically are
divided into adjacent cells, with each cell built to contain 4
to 6 years of waste. Figure 11-19 shows the profile of the
MSW facility with two cells having a common cover.
Facilities have been permitted with more than 10 such
cells beneath a common cover. Such facilities eliminate
the long-term exposure of the liner system that would
result if a single large cell was constructed, and do not
lose the airspace between the cells that would occur if in-
dividual covers were placed on each cell.
It is necessary to incrementally cap a facility that has
multiple adjacent cells to prevent excessive leachate
generation. Strategies for incremental cap construction
should be reviewed as part of the permitting process.
Such strategies should provide for drainage swales
spaced at intervals of no more than 6.1 m (20 ft) of verti-
cal grade change over the cap to control surface water
runoff.
MSW gas collection systems are commonly either
blanket collectors with passive vents or active systems
Cap Profile
Figure 11-18. Case 5—Cap profile and geometry.
137
-------
using discrete wells. Figure 11-20 shows the proposed
active well array for Case 5. Each well consists of a per-
forated plastic pipe within a gravel screen. The top of the
pipe will penetrate the low hydraulic conductivity barrier,
and must be sealed at the soil barrier using a bentonite
seal and at the geomembrane barrier using a boot/clamp
fixture. As the waste subsides, the gas well pipe will
move upward relative to the cap geomembrane. The
flexible boot between the pipe and the cap geomembrane
must be installed to allow such differential movements.
The boots commonly are improperly installed upside
down, e.g., they allow movement of the pipe downward
relative to the cap geomembrane. This installation,
however, must be avoided to prevent damage to the
geomembrane seal. This seal not only limits surface
water infiltration, but also aids in maintaining the low
vacuum required for active gas removal.
The use of a geomembrane in the liner and the cap will
eliminate the lateral migration of gas if the
geomembranes are intact. Perimeter gas monitoring wells
(see Figure 11-21) provide an indication of the condition
of the liner and the cap. Such wells are installed at 152-
to 305-m (500- to 1,000-ft) spacings around the perimeter
of the landfill. Most states now limit gas concentrations in
such wells to less than 25 percent of the lower explosive
limit of the methane.
CONCLUSIONS
The five case studies presented in this chapter illustrate
the need to closely evaluate the stability of closure sys-
tems related to sliding at the interfaces of the layers
making up the cap, and alternatives for controlling sur-
face erosion. Additionally, these cases highlight the fol-
lowing permit considerations:
1.
2.
The permit should contain requirements for regular
monitoring of cap subsidence, criteria for allowable
differential cap subsidence, and an agreed-upon
method for repair of excessive subsidence.
Poorly maintained gas collection systems can allow
surface water through the cap. Passive vents should
be minimized and protected from damage. Active gas
wells will move upward relative to the cap and may
damage the cap barrier. Such wells should be in-
spected regularly and removed when no longer in
production.
3.
All erosion control systems require maintenance and
regular inspection. The limits of both should be estab-
lished in the permit.
CERCLA caps in particular require careful evaluation to
determine which of the RCRA MTG cover components
are appropriate for the specific site and waste.
REFERENCES
1. U.S. EPA. 1989. Technical guidance document: final
covers on hazardous waste landfills and surface im-
poundments. EPA/530-SW-89-047. July.
2. Murphy, W.L. and P.A. Gilbert. 1987. Prediction of
landfill cover performance. Unpublished study by
COE for EPA-RREL, 1985 through 1987.
Landfill remediation. First
Municipal Solid Waste,
3. Richardson, G.N. 1990.
U.S. Conference on
Washington, DC.
4. U.S. EPA. 1988. Geosynthetic design guidance for
hazardous waste landfill cells and surface impound-
ments. EPA/600/52-87/097.
Figure 11-19. Case 5—Profile showing MSW subcells.
138
-------
Well Collection System
Figure 11-20. Case 5—Gas collector well array.
8'OIA. STEEL PIPE
o
**
PO
O.
UJ
Q
UJ
«/>
Z3
U_
UJ
o;
STEEL PIPE CAP w/
HINGE & LOCK
P.V.C.PIPE CAP
00 NOT CEMENT
CONC. BENTONITE SEAL
I' OIA. SCH. 40 P.V.C. PIPE
W/ 3/16' DIA. (MIN.) SCREEN
HOLES
PEA GRAVEL PACK
P.V.C. END CAP
MIN. BORE OIA.
Figure 11-21. Case 5—Perimeter gas monitoring well.
139
-------
5, Bonaparte, R. et al. 1989. Rates of leakage through
landfill liners. IFAI Geosynthetics '89 Conference,
San Diego.
6. Giroud, J.P. et al. 1990. Stability of cover systems on
geomembrane covers. Proceedings, Fourth Interna-
tional Conference on Geotextiles, The Hague,
Netherlands.
7. U.S. EPA. 1980. Evaluating cover systems for solid
and hazardous waste sites. SW-867.
8. New York - 6 NYCRR Part 360.
ADDITIONAL REFERENCES
U.S. EPA. 1986. Covers for uncontrolled hazardous
waste sites. EPA/540/2-85/002.
U.S. EPA. 1985. Geotextiles for drainage, gas vent-
ing, and erosion control at hazardous waste sites.
EPA/600/2-86/085.
U.S. EPA. 1989. Requirements for hazardous waste
landfill cells and surface impoundments. Seminar
Publication. EPA/600/52-87/097.
140
-------
CHAPTER 12
POSTCLOSURE MONITORING
INTRODUCTION
The owner/operator of a facility must give significant con-
sideration during the closure permit process as to the na-
ture and extent of postciosure monitoring that will be
required. While regulatory postciosure monitoring time
frames range from 30 years for RCRA wastes to 500
years for mixed wastes (10 CFR 61), the actual monitor-
ing period will be influenced by the stability of the waste
and cover system. The permit should establish monitor-
ing procedures, acceptance criteria, and remediation
methods for the following key parameters:
1. Ground-water quality and potentiometric surface
should remain within the limits established in permit-
ting.
2. Leachate quantities and chemical makeup should
remain predictable.
3. Gas release concentrations and general air quality
must remain within guidelines. Such guidelines will
become stricter with time.
4. Differential subsidence of the cover must be limited
and repaired if allowable limits are exceeded.
5. Surface erosion must stay within the 2 ton/year allow-
able and be repaired on an annual basis.
The key elements in the monitoring program that must be
established during permitting are detection methods, al-
lowable limits, and the plan for remediation when limits
are exceeded.
GROUND-WATER MONITORING
Key monitoring variables in a comprehensive ground-
water monitoring program include both changes in the
potentiometric surface that could bring the landfill liner
system in contact with the ground water and the chemical
quality of the ground water that is an indicator of leachate
release. In RCRA facilities, both the potentiometric and
background water quality will be established during per-
mitting of the landfill prior to placement of the waste. For
CERCLA facilities, such information should be estab-
lished during the closure permit process.
A ground-water well network must be established that
both tracks changes in the ground-water potentiometric
surface and detects leakage from the facility. In both
RCRA and CERCLA facilities, the background quality of
the ground water must be documented prior to closure.
Individual monitoring wells must be designed to reflect
both the anticipated contaminant and the site-specific
stratigraphy. Figure 12-1 shows a typical well configura-
tion. The well casing will commonly be PVC for inorganic
contaminants and stainless steel for organic con-
taminants. While monitoring wells have become very
standardized, it is important to specify locking well caps
that prevent tampering of the well, well seals that restrict
surface water flow into the well, and solvent free well pipe
connections that do not contaminate the well.
In CERCLA sites, great care must be taken during the
placement of monitoring wells and during any soil borings
to avoid penetrating an aquiclude (low hydraulic conduc-
tivity soil layer) that may lie underneath contaminated
ground water. Figure 12-2 shows such a potential stratig-
raphy. When placing monitoring wells through an
aquiclude, a casing must first be installed from the
ground to the aquiclude. A grout seal is then established
to hydraulically isolate this casing from the aquiclude,
and the monitoring well is drilled to the lower aquifer
within the casing. In this manner, the contamination from
the upper aquifer will not contaminate the lower aquifer.
Ground water should be sampled at a frequency defined
by the level of anticipated contamination and the site con-
ditions. Generally, it is useful to have monthly ground-
water background data prior to permitting operation or
closure of a facility. Post-operation sampling frequency
then can be decreased to quarterly monitoring, which
should be maintained unless the consistency of measure-
ments and operation justify sampling less frequently,
Postciosure monitoring frequencies commonly range
from quarterly for lined RCRA facilities to annually for
common MSW landfills.
LEACHATE MONITORING
Both the quantity and composition of leachate generated
within a RCRA facility provide significant information on
the performance of the closure system. If the closure sys-
tem is properly designed and installed, the rate of
leachate generation in the primary collector will decrease
with time. If the closure is not complete, then the rate of
141
-------
1 2 3
123 12
Landfill
W.T.
Sand (K = 1 X 1CT2 cm/sec)
Layer 1
Sand (K = 1 x 1(T3 cm/sec)
Layer 3
Figure 12-1. Monitoring well configuration.
8'00 LOCKING PROTECTIVE-
STEEL CAP
SIDE VENTED PVC WELL CASING CAP x
LOCK
7'10 STEEL PROTECTIVE CASING
y/l LONG
4' SCM. 40 PVC RISER PIPE
ISTICK-UPI
ORIGINAL CSOUNO SURFACE
REINFORCED CONCRETE CAP —
UIN. 2' RADIUS '»/ •« REB4R
ON 6-CENTERS
CONCRETE PLUG EXTENDING 36
DOWN BOREHOLE BELOW CAP
4- SCH. 40 PVC FLUSH JOINT CASING
CEMENT/BENTONITE GROUT PLACED
BY SIDE DISCHARGE TREMIE PIPEi
94 LBS.PORTLAND CEMENT
5 LBS. POWDERED 3ENTOMTE
6 GALS. WATER
ILB. CALCIUM CHLORIDE
BENTOMTE PtLLET SEAL
TAUPED AND HYDRA TED
FINE SAND FILTER
SAND PACK, CONSISTING OF
WASHED I GRADED SILICA SANO.
SIZED FOR THE AQUIFER AND
PLACED BY TREMIE PIPE
FACTORY SLOTTED OR CONTNUOUS WIRE •
SCREEN SIZED FOR AQUIFER GRAIN
SIZE DISTRIBUTION
TAILPIECE OR SEDIMENT CUP
CENTRALIZE? (SPACED AS REO'O.) •
PVC END CAP (THREADED!
BOTTOM OF BOREHOLE
-| J' CLEARANCE ~I 1
1 — CONST, ^r oil?
3§a!s
z
0.
O
a
UJ
3
S
UJ
1
• ~5iN72rilAi. '
UJ
:
0
0
I
a.
a
r'-V-'sAXOPACK '
ABOVE SC.RFE" TQP
z
GL
Q
_j
l:
^
O
u-
K
SCREENED INTERVAL 2'-20
NORMALLY V-IO'
1
VARIABLE CUP
J> NORMALLY «••
LENGTH.
Figure 12-2. Monitoring interbedded aquifer.
142
-------
Impact of Biological Growth
CD
<
LLJ
2
cc
LLJ
0.
BIOCIOC BXCXfLOSS
II
BIOCICE BACKrLOSH
12
i 5 T
3 0 T
TIME
Figure 12-3. Impact of biological growth on filters.
100 -i
80-
'«
V)
s
£ *
^ &
(0 x
(0 ~
03
o
20 40 60 80 100 120
5L Rate
5L Acids
Figure 12-4. Gas generation versus time.
leachate generation in the primary collector may reflect
precipitation trends. Therefore, the integrity of the closure
can be verified by evaluating leachate quantity records. A
sudden increase in the quantity of leachate generated will
clearly indicate failure of the closure. Unfortunately, many
CERCLA facilities will lack a liner and primary collector
system.
The concentration of contaminants in a facility's leachate
will increase with time until an equilibrium condition is es-
tablished. A sudden reduction in this level of con-
taminants is a good indication that the cover has been
breached, allowing a slug of surface water to enter the
waste and dilute the leachate. Biological growth,
however, can also have a significant impact on the
monitoring system over the long term. Figure 12-3 shows
the impact of biological growth in municipal solid waste
(MSW) leachate on the permeability of a geotextile filter
commonly used in collector systems (1). The time, T, for
significant reduction in permeability may be as short as 6
weeks. Thus, a long-term decrease in the amount of
leachate generated may indicate biological clogging of
the collector, which may prevent detecting failure of the
closure. Such biological clogging occurred recently in a
MSW landfill in Delaware. A significant head of perched
leachate was discovered within the waste while the quan-
tity of leachate generated was actually decreasing. This
clogging required excavation of the waste and replace-
ment of the primary collector system.
GAS GENERATION
Gas generation within a waste containment system must
be monitored both to ensure that such gas does not
143
-------
Table 12-1. Threshold Limits of Air Contamination
Contaminant
Threshold Limit Values of Selected Air Contaminants"
TLV
Dust
Carbon monoxide
Asbestos
Benzene
Coal dust
Cotton dust
Grain dust
Hydrogen sulfide
Nuisance particulates
Phenol
Vinyl chloride
Wood dust
Hard wood
Soft wood
I mg/m3
50 ppm
0.2 to 2 fibers/cm3 (depending on asbestos type)
10 ppm
2 mg/mj
0.2 mg/m3
4 mg/m3
10 ppm
10 mg/m3
5 ppm
5 ppm
I mg/m3
5 mg/m3
"Values of TLV obtained from the American Conference of Governmental Industrial
Hygienists (I987).
migrate off site and to indicate closure performance. The
rates of gas generation vary from more than 900 liters/kg
waste/year in MSW wastes (2) to insignificant rates in
RCRA commercial landfills. The rate of gas generation in
future MSW landfills is anticipated to decrease as these
landfills are constructed with liners and leachate collec-
tion systems. The addition of a geomembrane in the
cover will significantly decrease the amount of surface
water infiltration and also lead to lower gas generation
rates.
When geomembranes are used in a cover, very little gas
can escape vertically. Therefore, in an unlined facility,
such as a typical CERCLA closure, escaping gas will
move to the perimeter of the cover. Simple gas monitor-
ing wells (described in Chapter 11, Case 5) must be in-
stalled around the perimeter of the cover to detect
laterally moving gas. The level of gas at such wells must
remain below 25 percent of the lower explosive limit
(LEL). The level of gas production can vary significantly
with the weather; therefore, the monitoring frequency
should be increased when the surrounding ground is
saturated or frozen.
Gas odors detected above a closure system that includes
a geomembrane indicate that the geomembrane has a
significant penetration. A regular survey of gas levels on
the surface of the closure is a good method of verifying
the integrity of the cap barrier.
As detailed in Chapter 11, gas removal systems must be
designed with a minimal number of penetrations through
the cover system. Each vent is a potential major leak. For
passive systems, a maximum of one vent per acre should
be included initially. If monitoring of these vents reveals
excessively high gas concentrations, then additional
wells can be installed. In active systems, gas wells must
be removed when they are no longer productive to
prevent damage to the cover.
As with leachate quantities, the rate of gas generation
should also decrease with time if the cover system is
functioning properly such that moisture does not reach
the waste. Figure 12-4 shows the result of laboratory
column gas generation tests (3). In the figure, methane
production rate and the level of carboxylic acids in the
leachate decrease with time. A properly functioning cover
will ensure that the leachate will remain acidic and that
gas production will be low.
SUBSIDENCE MONITORING
Chapter 11 discusses the ability of the cap barrier com-
ponents to tolerate differential settlements due to waste
subsidence. In Case 1, differential settlements as large
as 0.5 m in 6 m (1.8 ft in 20 ft) were tolerated by com-
posite barriers. Thus, the level of differential settlements
of interest during postclosure monitoring can be quite
large. Such levels can commonly be found by walking the
cover after a rain storm and looking for major puddles or
ponding. Subsidence depressions also can be found
through an annual survey of the cover using either con-
ventional or aerial survey methods.
Subsidence depressions must be remediated below the
level of the barrier system to avoid long-term acceleration
of the subsidence due to a "roof ponding" mechanism.
Roof ponding refers to the common failure in flat roof sys-
tems where ponding water causes the roof rafters to
deflect, thus allowing more water to pond, causing more
deflection, and so on. This mechanism continues until the
roof collapses. Remediation requires removing the cover
system in the region of subsidence and backfilling the
depression with lightweight fills. This fill may either be
144
-------
more waste or commercial lightweight aggregates. The
full cover profile must then be rebuilt over the new fill.
SURFACE EROSION
All cover systems will erode and require long-term main-
tenance. Cover systems with moderate slopes and an
agricultural cover will typically require annual main-
tenance of 0.5 percent of their surface area; this percent-
age increases with slope. Thus, all covers that use
agricultural covers require an annual inspection and
repair program. Such repair may include cleaning out sur-
face water swales, replacing cover soil, and reestablish-
ing vegetation. Areas of the cover requiring repeated
repair may benefit from hardening or the use of geosyn-
thetic erosion control blankets. Covers that use hardened
erosion control systems should also be inspected annual-
ly, though annual maintenance should not be required.
The annual inspection should verify that the agricultural
cover is being mowed at least annually to prevent the
growth of deep-rooted volunteer vegetation. In arid
regions of the country or during droughts, full RCRA
covers may not be able to maintain vegetation unless the
plants are very drought resistant. This loss of vegetation
is due to moisture loss in the root zone of the cover soil,
resulting from the underlying drainage system.
AIR QUALITY MONITORING
Air emissions from waste storage facilities will come
under increasing scrutiny in the next decade. Monitoring
techniques will be similar to those used at industrial
facilities and include passive samples obtained using col-
lection media, grab samples obtained in evacuated
sample vessels, and active pump and filter samples. The
most common air contaminants coming from the waste
disposal cell obviously are waste dependent; for MSW
wastes, these are methane, vinyl chloride, and benzene.
Table 12-1 presents typical allowable limits of selected air
contaminants. Such limits are currently undergoing exten-
sive review; significantly lower allowable levels are an-
ticipated for future operations.
The geomembrane component of the MTG cover com-
posite barrier system controls air emissions significantly.
In fact, the presence of emissions indicates that the
geomembrane cover has failed and needs to be repaired
immediately.
REFERENCES
1. Unpublished research at Geosynthetic Research In-
stitute, personal communication with R.M. Koerner.
2. Walsh, J.L. 1988. Handbook on biogas utilization.
Georgia Tech Research. February.
3. Barlaz, M.A. et al. 1989. Bacterial population
development and chemical characteristics of refuse
decomposition in a simulated sanitary landfill. Applied
and Environmental Microbiology. January.
145
-------
APPENDIX A
STABILITY AND TENSION CONSIDERATIONS REGARDING COVER SOILS ON
GEOMEMBRANE-LINED SLOPES
-------
Stability and Tension Considerations Regarding Cover Soils on
Geomembrane Lined Slopes
by
Robert M. Koerner and Bao-Lln Hwu
Geosynthetlc Research Institute
Drexel University
Philadelphia, Pennsylvania' 19104
Abstract
The occurrence of cover soil instability in the form of sliding on geomembranes is far too
frequent. Additionally, there have been cases of wide width tension failures of the underlying
geomembranes when the friction created by the cover soil becomes excessive. While there are
procedures available in the literature regarding rational design of those topics, it is felt that a
unified step-by-step perspective might be worthwhile. It is in this light that this paper is
written. Included are four separate, but closely interrelated, design models. They are the
following;
• cover soil stability on side slopes when placed above a geomembrane,
« cover soil reinforcement provided by either geogrids or geotextlles,
« wide width tension mobilized in the geomembrane caused by the interface friction of
the soils placed above and below the geomembrane, and
» circumferential tension mobilized in the geomembrane by subsidence of the subgrade
material beneath the geomembrane.
Each of these designs are developed in detail and a numeric problem is framed to illustrate the
design procedure. Emphasized throughout the paper is the need for realistic laboratory test
values of Interface friction, in-plane tension and out-of-plane tension of the geomembranes.
By having realistic experimental values of allowable strength they can be compared to the
required, or design, strength for calculation of the resulting factor-of-safety against
instability or failure.
A-l
-------
Stability and Tension Considerations Regarding Cover Soils on
Geomembrane Lined Slopes
Introduction
Geomembrane lined soil slopes are common in many areas of civil engineering
construction but nowhere are they more prevalent than in the environmental related field of
the containment of solid waste. Cover soils on geomembranes placed above the waste as in
landfill caps and closures as well as lined side slopes beneath the waste are commonplace as
the sketches of Figure 1 Indicate. The variations of soil types beneath the geomembrane as well
as above the geomembrane are enormous. They range from moist clays in the form of
composite liners to drainage sands and gravels of very high permeability. The likelihood of
having other geosynthetic materials adjacent to the geomembranes (like geotextiles, geonets
and drainage geocomposites) presents another set of variables to be considered. Lastly, the
existence of many different geomembrane types, having different thicknesses, strengths,
elongations and surface characteristics leads to the necessity of performing a rational design
on such systems. Clearly, the development of design models to evaluate the stability of the
overlying materials as well as the tensile stresses that may be induced in the underlying
geomembranes should always be performed. Fortunately, both the stability of the overlying
soil materials and the reduction of tensile stress in the geomembrane can be accommodated by
reinforcing the cover soil with either geogrids or geotextiles. This is becoming known as
veneer stability reinforcement and is necessitated due to a number of cover soil stability, or
sloughing, failures, some of which are shown In the photographs of Figure 2.
This paper presents several design models and their development Into design equations
for cover soil stability (both without and then with reinforcement) and for the induced tensile
stresses that are mobilized in the underlying geomembrane. The approach taken in this paper
will utilize a single geomembrane, but it should be recognized that double liners are frequently
used beneath solid and liquid waste. Design considerations into the secondary liner, however,
can be handled by reasonable extensions of the material to be presented.
A-2
-------
Geomembrane
Waste
SOIL
SUB-SOIL
COVER S'
SUB-SOIL
^Geotextile
Seonet or Geocomposite
"Geomembrane
Waste'
(a) Landfill Cover with Soil
Above Geomembrane
(b) Landfill Cover with Drainage
Geosynthetic Above Geomembrane
Proposed,
" Waste
Geomembrane
SUB-SOIL
(c) Landfill Liner with Soil
Above Geomembrane
Geotextile'
Geonet or-
Geocomposite
Geomembrane*
(d) Landfill Liner with Drainage
Geosynthetic Above Geomembrane
Figure 1 - Various Solid Waste Geomembrane Covers and Liners Involving
Natural Soils and/or Drainage Geosynthetics
A-3
-------
Figure 2 - Cases of Cover Soil Instability for Case l(a) (upper photo) and for Case He) (lower
photo) as shown In Figure 1
A-4
-------
Interface Friction Considerations
It will be seen that at the heart of the design equations to be developed in this paper are
interface friction values between the geomembrane and the overlying soils or drainage
geosynthetics and also against the underlying soils or drainage geosynthetics. These values
are obtained by direct shear evaluation in simulated laboratory tests. Unfortunately, many
aspects of the direct shear test have not yet been standardized (although ASTM has a Task
Group working on a draft Standard), and many important details must be left to the design
engineer and testing organization. For example, the following items need to be carefully
considered.
minimum or maximum size of shear box
aspect dimensions of the test specimen
type of fixity of the geomembrane to the shear box and to a substrate
moisture conditions during normal stress application
type of liquid to use during sample preparation and testing
method and duration of normal stress application
strain controlled or stress controlled shear application
rate of shear application
moisture conditions and drainage during shear application
duration of test
number of replicate tests at different normal stresses
linearity of resulting failure envelope
Thus the use of reported values in the published literature can only be used with considerable
caution and, at best, for preliminary design.^1"3) For final design and/or permitting, the site
specific conditions and the proposed materials must be used in the tests so as to obtain realistic
values of the shear strength parameters adhesion (ca) and Interface friction (5), Additionally,
tests should also be performed on the soil by Itself so as to obtain a reference value for
comparison to the inclusion of the geomembrane. Calculation of the adhesion efficiency on
soil cohesion and a frictional efficiency to that of the soil by itself are meaningful In assessing
the numeric results of the designs to follow. I.e.
Ec = ca/c(100) (1)
E = tan5/tan(100) (2)
0
where
Ec = efficiency on cohesion
ca = adhesion of soil-to-geomembrane
c = cohesion of soil-to-soil
E. = efficiency on friction
8 = friction angle of soil-to-geomembrane
<)> = friction angle of soil-to-soil
A-5
-------
Past Investigations and Analyses
The Isolation of free body diagrams depleting the site specific situation to be analyzed is
certainly not new. It is a direct extension of geotechnical engineering of soil stability and is
reasonably straightforward since the failure plane against the geornembrane is clearly
defined. Thus a computer search Is generally not necessary to locate the minimum factor-of-
safety stability value. Also it should be recognized that the failure surface Is usually linear,
rather than circular, log spiral, or other complicated geometric shape in that it follows the
surface of the geornembrane itself.
A procedure which nicely accommodates a clearly defined straight line slip surface has
been developed by the U.S. Army Corps of Engineers. W Their wedge analysis procedures form
the essence of the developments to follow. The graphic procedures are outlined in Reference #4
but are developed in this paper into design equations in a more rigorous manner. Also to be
mentioned Is the work of Giroud and Beech® and Giroud, et al. ® in providing excellent
insight into several aspects of the design.
In the first referenced paper, by Giroud and Beech®, a two-part wedge method is utilized
to arrive at a similar equation as ours except without an adhesion term. Also, the treatment at
the top of the slope is slightly different. Their work will be referenced In the second problem of
this set of four examples and a comparison of results will be made. In the second referenced
paper, by Giroud, et al.^, a large overburden stress necessitated the use of arching theory to
recognize that a limiting value will occur when the geornembrane is located beneath deep fills.
This is not the case with the shallow overburden stresses imposed by cover soils placed on
geomembranes that are the focus of this paper. In the fourth example to be presented we will
use the full thickness of the overburden times its unit weight. Additionally, we will not use a
deformation/strain reduction value In the interest of being conservative. The reference cited
by Giroud, et al.^) should be used in this regard.
A-6
-------
Model # 1: Stability of Cover Soil Above a Geomembrane
Consider a cover soil (usually a permeable soil like gravel, sand or silt) placed directly
on a georaembrane at a slope angle of "co". Two discrete zones can be visualized as seen In
Figure 3. Here one sees a small passive wedge resisting a long, thin active wedge extending the
length of the slope. It is assumed that the cover soil is a uniform thickness and constant unit
weight. At the top of the slope, or at an Intermediate berm, a tension crack m the cover soil is
considered to occur thereby breaking communication with additional cover soil at higher
elevations.
Resisting the tendency for the cover soil to slide is the adhesion and/or Interface
friction of the cover soil to the specific type of underlying geomembrane. The values of "ca"
and "5" must be obtained from a simulated laboratory direct shear test as described earlier.
Note that the passive wedge Is assumed to move on the underlying cover soil so that the shear
parameters "c" and "", which come from soll-to-soll friction tests, will also be required.
PASSIVE
WEDGE.
WP
ACTIVE
Geomembrane
Figure 3 - Cross Section of Cover Soil on a Geomembrane Illustrating the Various Forces
Involved on the Active and Passive Wedges
A-7
-------
By taking free bodies of the passive and active wedges with the appropriate forces being
applied, the following formulation for the stability factor-of-safety results, see Equation 3.
Note that the equation is not an explicit solution for the factor-of-safety (FS), and must be
solved indirectly. The complete development of the equation is given In Appendix "A".
(FS)2 [0.5 y LH sin2 (2 co)] - (FS) |y LH cos2 co tan 8 sin (2 co) + c a L cos co sin (2 co)
+Y LH sin2 co tan <)> sin (2 co) + 2 c H cos co +yH 2tan ]
+ [(yLH cos co tan 8 + caL) (tan Q> sin co sin (2 co)] = 0 (3)
Using ax2 + bx + c = 0, where
a = 0.5 Y LH sin22co
b = -[ Y LH cos2co tan 8 sin (2co) + caL cos co sin (2 co)
+ Y LH sin2co tan <(> sin (2co) + 2cH cos co +Y H2 tan $]
c = (Y LH cos co tan 5 + caL) (tan sin co sin (2co))
the resulting factor-of-safety is as follows:
When the calculated factor-of-safety value falls below 1.0. a stability failure of the cover soil
sliding on the geomembrane is to be anticipated. However, it should be recognized that seepage
forces, seismic forces and construction placement forces have not been considered in this
analysis and all of these phenomena tend to lower the factor-of-safety. Thus a value of greater
than 1.0 should be targeted as being the minimum acceptable factor-of-safety. An example
problem illustrating the use of the above equations follows:
Example Problem: Given a soil cover soil slope of co = 18.4° (I.e.. 3 to 1),
L = 300 ft.. H = 3.0 ft. Y = 120 lb/ft3. c = 300 lb/ft2. ca = 0.4 = 32°. 8= 14°,
determine the resulting factor-of-safety
Solution:
a =.0.5 (120) (300) (3) sin2 (36.8°)
= 19.400 lb/ft
b =- [(120) (300) (3) cos2 (18.4°) tan (14°) sin (36.8°)
+ 0 + (120) (300) (3) sin2 (18.4°) tan (32°) sin (36.8°)
+ 2 (300) (3) cos (18.4°) + 120 (9) tan (32°)1
A-8
-------
= - [14523 + 0 + 4028 + 1708 + 6751
= - 20,934 Ib/ft
c = [(120) (300) (3) cos (18.4°) tan (14°) + 0]
[tan (32°) sin (18.4°) sin (36.8°)]
= [25500J[0.118]
= 3019 Ib/ft
_ 20.934 4- y (-20934) - 4(19400) (3019)
= 38,800
FS = 0.91, which signifies that a
stability failure will occur
A-9
-------
Model #2: Reinforcement of Cover Soil on a Geomembrane
Once the cover soil factor-of-safety becomes unacceptably low for the site specific
conditions (as illustrated In the previous problem), a possible solution to the situation is to add
a layer of geogrid or geotextile reinforcement as shown in Figure 4. In the case of landfill
covers, the tensile stresses that are mobilized In the reinforcement are carried over the crown
to (generally) an equal and opposite reaction on the opposing slope. Alternatively, these
stresses can be carried in friction via an anchorage mode of resistance as would occur in an
intermediate berm situation. For a landfill liner, the stresses in the reinforcement are
generally carried to an individual anchor trench extending behind the geomembrane anchor
trench. If the reinforcement is a geogrid it Is placed within the cover soil so that soil can
strike-through the apertures and the maximum amount of anchorage against the transverse
ribs can be mobilized. When using geotextiles, they can be placed directly on the geomembrane.
or embedded within the cover soil so as to mobilize friction in both surfaces.
The tensile stress of the reinforcement layer per unit width is calculated by setting "EA"
equal to "Ep" in Figure 3 and solving for the unbalanced force T* in Figure 4 which Is required
for a factor-of-safety equal to one. This value of T becomes Trec.^ which is given in Equation 5.
The complete development is available in Appendix "B".
yLH sin (o>-6)
cos <{>
cH
. .
+— - - tan <(>
sin co sin 2co
_
cos 8 a cos ( + co)
(5)
This value is now compared to the allowable wide width tensile strength of the particular
geogrid or geotextile under consideration, i.e.,
FS=Tallow/Treqd (6)
Note that the value of "Tallow" must Include such considerations as Installation damage, creep
and long-term degradation from chemical or biological interactions. If the value is obtained
from a test such as ASTM D-4595. the wide width strip tensile test, the use of partial factors-of-
safety is recommended to accommodate the above items. W An example problem using
Equations 5 and 6 follows:
Example Problem: Continue the previous problem of cover soil
instability where a geogrid with allowable wide width tensile strength of
4000 Ib/ft is being considered (I.e., the value Includes the above
mentioned partial factors-of-safety). What Is the resulting overall factor-
of-safety? The parameters are co = 18.4°, L = 300 ft., H = 3.0 ft..
A-10
-------
Reinforcement
•T (Geogrid or
Geotextile)
T (Geogrid or
^Geotextile)
Proposed
Waste
Geomembrane
COVER SOIL
Reinforcement
SUB-SOIL
Figure .4' - Geogrid or Geotextile Reinforcement of a Cover Soil Above
Waste and of a Cover Soil on a Geomembrane Beneath Waste
A-ll
-------
y =120 lb/ft3, c = 300 Ib/ft2, ca = 0, 0 = 32°, 8= 14
Solution:
(120) (300) (3) sin (4.4°)
cos (14°)
(300) (3) (120) (9)
cos
,000
(32
cos (50.4°)
= 8539-0-5292
Treqd =3247 lb/ft
FS =Tallow/Treqd
_ 4000
= 3247
FS = 1.23, which is marginally acceptable and a
stronger reinforcement or a double layer should be
considered,
Note: Using the formulation developed by Glroud and Beech^J with
the soil cohesion equal to zero results in a Treqd = 6890 lb/ft, while
the above formulation adjusted for a zero cohesion results in Treqcj =
7040 lb/ft. Thus the methods appear to be comparable to one
another,
A-12
-------
Model #3: Gcomembrane Tension Stres^ca Due i-~- Unbalanced Friction Forces
The shear stresses from the cover soil above the liner act downward on the underlying
geomembrane and In so doing mobilize upward shear stresses beneath the geomembrane from
the underlying soil. The situation Is shown In the sketch of Figure 5.
T (in Geomembrane)
where
u = Cau+(Wcosc0) tan8u
= caL +(W COSCD ) tan 8L
Figure 5 - Shear and Tensile Stresses Acting on a Covered Geomembrane
Here three different scenarios can be envisioned:
• If tu =t L, the geomembrane goes into a state of pure shear which should not be of great
concern for most types of geomembranes
• If t u TL, the geomembrane goes Into a state of pure shear equal to TL and the balance
of TU- T^must be canted by the geomembrane In tension.
This latter case Is the focus of this part of the design process. The situation generally occurs
when a material with high Interface friction (like sand or gravel) Is placed above the
geomembrane and a material with low Interface friction (like high moisture content clay) Is
placed beneath the geomembrane. The essential equation for the design is as follows where "T"
Is In units offeree per unit width, i,e., T/W. The complete derivation follows In Appendix "C".
T/W = [(caU - CjjJ + jH cos co (tan fy - tan 5j] L
(7)
A-13
-------
The resulting value of force per unit width T/W is then compared to the allowable strength of
the geomembrane which is shown schematically for different geomembranes In Figure 6. The
target values are T^^ for scrim reinforced geomembranes, Ty^y for semi-crystalline
geomembranes and Tauow (at a certain value of strain) for nonrelnforced flexible
geomembranes. Note that these curves should be obtained from a wide width tensile test which
Is currently under development in Committee D-35 on Geosynthetlcs in ASTMJ8'
break
CSPE-R
CPE-R
,EiA-R
FORCE
WIDTH
yield
aflo
E « 20 to 50%
STRAIN
Figure 6 - Tensile Behavior of Various Geomembrane Types
Since there Is generally no reduction for partial factors-of-safety in these values of laboratory
obtained strength, the final factor-of-safety in the design should be quite conservative. An
example problem follows:
Example Rrdblem: Given the same landfill cover as described in the
previous problems with a geomembrane having an allowable strength of
2000 Ib/ft The shear strength parameters of the geomembrane to the
upper soil are cau = 0 Ib/ft2 and Sy = 14° and to the lower soil are %L= 50
Ib/ft2 and SL= 5°. Calculate the tension In the geomembrane and the
A-14
-------
resulting factor-of-safety against geomembrane failure.
Solution:
T/W = [(cau - caj + 7H cos co (tan % - tan 5j] L
= [(0-50) + (120) (3) cos (18.4°) (tan (14°) - tan (5°) 1J300
= [-50 + 55.3] 300
= 15901b/ft
FS = Tallow/Treqd
2000
~ 1590
FS = 1.25, which is barely acceptable.
Note: An alternative design to the above is to bench the cover soil
(thereby decreasing the slope length) or use a liner whose lower surface
has a higher adhesion or a higher friction surface, than the one used in
the example thereby increasing "caL" and/or "5^".
A-15
-------
Model #4: Geomembrane Tension Stresses Due to Subsidence
Whenever subsidence occurs beneath a geomembrane and It Is supporting a cover soil
some induced tensile stresses will occur due to out-of-plane forces from the overburden. Such
subsidence is actually to be expected in closure situations above completed or abandoned
landfills where the underlying waste is generally poorly compacted. The magnitude of the
induced tensile stresses in the geomembrane depends upon the dimensions of the subsidence
zone and on the cover soil properties.
The general scheme is shown In Figure 7 where the critical assumption is the shape of
the deformed geomembrane. In the analysis which is provided In Appendix "D", the deformed
shape is that of a spheroid of gradually decreasing center point along the symmetric axis of the
deformed geomembrane.^ As a worst case assumption, the geomembrane is assumed to be
fixed at the circumference of the subsidence zone. The required tensile force In the
geomembrane can be solved In terms of a force per unit width "Treqcj", or as a stress. l.e. "
The latter will be used In this analysis since it will be compared to a laboratory test method
resulting in the compatible term. The necessary design equation Is as follows where the
specific terms are given In Figure 7.
Jreqd
cs
(8)
Cover Soil
Unit Weight
D
Geomembrane
Figure 7 - Tensile Stresses in a Geomembrane Mobilized by Cover Soil and Caused by
Subsidence
A-16
-------
Upon calculating the value of areq(j for the site specific situation under consideration, it is
compared to an appropriate laboratory simulation test. Recommended at this time is a three-
dimensional, out-of-plane, tension test of the same configuration as Figure 6. It Is available as
GRI Test Method GM-4.^10) Thus the formulation for the final factor-of-safety becomes the
following:
Since the value of Callow13 used directly from the test method without any reduction in the
form of partial factors-of-safety, relatively conservative values should be required. An
example problem follows:
Example Problem: Given the same cover soil situation as in the previous
example, except now a local subsidence occurs which is estimated to be 1.0
ft deep by 3.0 ft radius. The geomembrane is 40 mUs thick has a craiiow°f
1000 lb/in2. Determine the factor-of-safety of the geomembrane against
the mobilized tensile stresses.
Solution:
2DL%SHCS
a,
Jreqd o
3t(D
_ 2 (LO) (3.0)2 (120) (3.0)
3 (0.040/12) [(l.O)2 + (3.0)2]
= 64,800 lb/ft2
<*reqd = 450 lb/in2
FS = CTallow/<%eqd
= 1000/450
= 2.2, which Is acceptable
A-17
-------
Summary and Conclusions
The occurrence of cover soils sliding off geomembrane lined slopes is not an infrequent
incident. While less obvious, but of even greater concern, there are often tensile stresses
imposed on the underlying geomembrane. The occurrence of extensive tensile failures of
geomembranes on side slopes is also known to have occurred.. This paper is focused toward a
series of four design models to be used to analyze various aspects of the situation.
The first model considered the cover soil's stability by itself. The design procedure is
straightforward but it does require a set of carefully generated direct shear tests to realistically
obtain the interface friction parameters.
The growing tendency toward steeper and longer slope angles gives rise to the second
design model which is veneer reinforcement of the cover soil. Geogrids and geotextiles have
shown that they can nicely reinforce the cover soil and the first design example was modified
accordingly. The design leads to the calculation of the required tensile strength of the
reinforcement. This value must then be compared to a laboratory generated wide width tensile
strength of the candidate reinforcement material. It is important in this regard to consider
long term implications which can be addressed by partial factors-of-safety.
Both of the above analyses serve to set up the third design scenario, that being a
calculation procedure for determination of the induced tensile stresses in the underlying
geomembrane brought on by unbalanced friction values. Whenever the frictional
characteristics beneath the liner are low (e.g.. when the liner is placed on a high moisture clay
soil as it Is in a composite liner), this type of analysis should be performed. The tensile stress
in the geomembrane is then compared to the wide width tensile strength of the geomembrane
for its resulting factor-of-safety.
Lastly, a design procedure for calculation of out-of-plane generated tensile stresses in
the geomembrane was developed. This situation could readily arise by subsidence of solid
waste beneath the geomembrane. The resulting tensile stresses in the geomembrane must then
be compared to a properly simulated laboratory test for the factor-of-safety. Such three
dimensional axi-symmetric test procedures are currently available.
Each of the four above described models along with their design/analysis procedures
were illustrated by means of an example problem dealing with a cover soil in a solid waste
closure situation. This type of application is the primary focus of the paper. However, similar
situations can arise elsewhere. For example, the same situation occurs in the case of gravel
covered primary geomembrane liners on the side slopes of unfilled, or partially filled,
landfills. These slopes may have to be exposed to the elements for many years until the waste
A-18
-------
provides sufficient passive resistance and final stability. In the meantime, cover soil
instability will cause sloughing and can expose the geomembrane to ultraviolet light, high
temperatures via direct exposure, and a significantly shortened lifetime.
Hopefully, the use of design models such as presently here (and elsewhere), coupled with
the appropriate test method simulating actual field behavior, will lead to recognition of the
problems encountered and to a widespread rational design of cover soils on geomembrane
lined side slopes.
A-19
-------
References
1, Martin, J, P., Koerner, R, M. and Whitty, J. E., "Experimental Friction Evaluation of
Slippage Between Geomembranes, Geotextiles and Soils," Proc. Intl. Conf, on
Geomembranes, IFAI. Denver, CO, 1984, pp. 191-196.
2. Koerner, R. M., Martin, J. P. and Koemer, G. R., "Shear Strength Parameters Between
Geomembranes and Cohesive Soils," Jour. Geotex. and Geomem., Vol. 4, 1986, pp. 21-30.
3. Mitchell, J. K., Seed, R. B. and Seed, H. B., "Kettleman Hills Waste Landfill Slope Failure. I
- Liner-System Properties," Jour. Geotechnlcal Engineering, Vol. 116, No. 4, April 1990.
pp. 647-660.
4. .Manual EM 1110-1902, U. S. Army Corps of Engineers, Washington. DC, 1960.
5. Glroud, J. P. and Beech, J. F., "Stability of Soil Layers on Geosynthetlc Lining Systems."
Geosynthetics '89 Conference, San Diego, CA. IFAI, 1989, pp. 35-46.
6. Glroud, J. P., Bonaparte, R.. Beech, J. F. and Gross, B. A., "Design of Soil Layer -
Geosynthetic Systems Overlying Voids," Journal of Geotextiles and Geomembranes. Vol. 9.
No. 1. 1990, Elsevler. pp. 11-50.
7. Koerner. R M.. Designing with Geosvnthetlcs. 2nd Edition. Prentice Hall Publ. Co.,
Englewood Cliffs, NJ, 1990, 652 pgs.
8. , Standard Test Method for Determining Performance Strength of Geomembranes by
the Wide Strip Tensile Method." ASTM Draft Designation D35.10.86.02 (In task group
status).
9. Koerner. R M.. Koemer, G. R. and Hwu, B-L. "Three Dimensional, Axl-Symmetrtc
Geomembrane Tension Test," Geosynthetic Testing for Waste Containment Applications,
ASTM STP 1081, Robert M. Koemer. Editor ASTM, Philadelphia. PA. 1990.
10. , GRI Test Method GM-4. Three Dimensional Geomembrane Tension Test,"
Geosynthetic Research Institute, Philadelphia PA. 1989.
A-20
-------
Appendix "A"
Derivation of FS for Cover Soil Stability on a Geomembrane
Active
Wedge
W,
Passive
Wedge
Passive Wedge
y _
*< Si
Y H
sin co cos o> sm 2 to
£~
"5
cos co
C H
FS sin (0
A-21
-------
Passive Wedge
EF = WP
DE EF
WP
sin(90°-<))D)
DE = WP • tan
AD
sin (90° + <|>D) sin(90°-<|)D-i
Ep Cp + Wp • tan D
COS (J>£) COS (<(>JD + 05)
£„ =
cos <)>£)• [Cp 4- WP • tan <)>D]
cos (D + 05)
. T C H Y-
L 05 sin
cos
tan
(cos <|>D cos 05 - sin <{>D sin
(cos 05 - tan <)>D sin
(FS • cos 05 - tan ()> • sin
r C-H
" L FS • sin
y-H
05 2 sin 05 cos 05 FS
2 • sin 05 • cos 05 • FS
2 • sin 05 • cos 05 • FS
A-22
-------
90+
90+co
A-23
-------
EA-EP
T TT , . B x ca' L (2-C-H-cosG5 + y H2- tan)
y • L • H • (sin 05 - cos 03 • tan 5D) =£— = -r=z — r—r-—r—, . ,T
' u FS (FS • cos 05 - tan • sin 03) • (sin 2 03
T TT ( . _ cos 03 • tan S ^ _ Ca- L _ (2 • C • H • cos 03 + y • H2 • tan )
Y lSmG5~ FS J~ FS ~ (FS-cos03-tan<()- sin03) • (sin 20
Y • L • H • (sin 03 • FS - cos 03 • tan 6) - Ca • L _ (2 • C • H • cos 03 + y • H2 • tan <())
FS
(FS • cos 03 • sin 203 - tan <(>• sin 03 • sin 203)
Y • L • H • sin 03 • FS - y • L • H • cos 03 • tan 8 - Ca • L (2 • C • H • cos 03 + y -H2 • tan <]>)
FS
(FS • cos 03 • sin 203 - tan <(> • sin 03 • sin 203)
FS • (y • L • H • sin 03 • cos 03 • sin 203) - FS • (y • L • H • cos 03 • tan 8 • sin 203)
- FS • (Ca • L • cos 03 • sin 203) - FS • (y • L • H • sin203 • tan • sin 203)
+ (Y • L • H • cos 03 • tan 8 + Ca • L) • (tan <|) • sin 03 • sin 203) = FS • (2 • C • H • cos 03 + y • H2 • tan (}))
FS2 • f -r- ' Y' L • H • sin 2 203 J - FS • (y • L • H • cos2G3 • tan 8 • sin 203+ Ca • L • cos 03 • sin 203
+ y • L • H • sin203 • tan (j> • sin 203 + 2 • C • H • cos 03 + y • H2 • tan )
+ (y • L • H • cos 03 • tan 8 + Ca • L) • (tan <)> • sin 03 • sin 203) =0
A-24
-------
Appendix "B"
Derivation of Required Tensile Strength of Geogrid
or Geotextile Reinforcement of Cover Soil on a Geomembrane
Reinforcement
(Geogrid or
Geotextile)
EA =
T = EA =>T = EA-EP
Y • L • H ' sin (G3 - 80)
cos 5
D
Ep =
f_C_ H
"I FS ' sin t
cos (<)>D + CJ)
= 1,5D = 8,
T _
*
7 ' L • H • sin (C5 - 8)
cos 5
C-H y]
sin 05 sin
• tan<{!
cos( + i
A-25
-------
Der'nva«on of
Due tn
"C"
Forces
' ^^* y» v/_ s V . tr
1=0 '
*?" ^ rf-H C3n Of.
^^C.. 4.V.T, U
^
Oiv ""
,«„,„
A-26
-------
Appendix "D"
Derivation of Geomembrane Tensile Stress Due to
Subsidence of Material Beneath the Geomembrane
Cover Soil
Unit Weight-Ycs"
D
Geomembrane
C (circumference) = 2- it • L
t = thickness of the geomembrane
= (R-D)
^R^ = R^-2
R s
D2+L2
2-D
R
R-D
fL
J (2- it • r) • dr • ycs • Hcs • r= 0rcqd- t • C • R
*•%
• JC • L3
- Creqd" t ' (2 ' It ' L) ' R
* Ycs '
f(2-7i-L)-R
' YcS ' HCS
3- t-R
2 • D • L2 • YCS ' H
cs
A-27
-------
APPENDIX B
LONG-TERM DURABILITY AND AGING OF GEOMEMBRANES
-------
GEOMEMBRANES DURABILTTY AND AGING
GRI-96
Long-term Durability and Aging of Geomembranes
Robert M, Koerner1 , Tick H, Halse* and
Arthur E. Lord, Jr.*
Perhaps the most frequently asked question regarding
geomembranes (or any other type of geosynthetic material)
is, "how long will they last"? The answer to this
question is Illusive (in spite of a relatively large data
base on polymer degradation} mainly because of the buried
nature of geomembranes. Soil burial greatly diminishes,
and even eliminates many of the degradation processes and
synergistic effects which have been most widely
investigated by the polymer industry for exposed
plastics. However, different degradation processes coming
from chemical interactions and extremely long time frames
may be Involved via exposure to liquids like leachate for
systems Intended to last for many decades or even
hundreds of years. Thus the lifetimes of buried
geomembranes can be significantly different than exposed
plastics, but a quantitative method to predict "how long"
is still not available.
This paper describes the various degradation
mechanisms of plastics on an individual basis and then
addresses the various synergistic effects which may
accelerate degradation. It will be seen that synergistic
effects greatly complicate the situation. While
accelerated test methods are attractive to assess the
various phenomena, these procedures may significantly
misrepresent the actual long-term performance of
geomembranes. Thus the transfer of information must
proceed with caution.
'Bowman Professor of Civil Engineering and Director,
Geosynthetic Research Institute, Drexel University
Philadelphia, PA, 19104.
'Research Assistant Professor, Geosynthetic Research
Institute, Drexei University, Philadelphia, PA, 19104. ,
'Professor of Physics, Geosynthetic Research Institute,
Drexel University, Philadelphia, PA, 19104.
106
, with observations on field behavior,
nf
,
;„'"., ;r,'..:
project, structure or system
Throughout the 1960's and the 1970' s « tremendous
rise to many v*rla"° and Kay's book (1988) gives
or omembranes,
being
preferred over clay liners.
Regarding the
g*om«mbr«ne types e*isn!J », st protocol, was
reducing the many possible
-------
W
N)
Long-Term Durability and Aging of Geomembranes
and
Robert M. Koerner1 , ylck H. Halse'
Arthur E. Lord, Jr.'
Abstract
Perhaps the most frequently asked question regarding
geomembranes (or any other type of geosynthetic material)
Is, "how long will they last"? The answer to this
question is Illusive (in spite of a relatively large data
base on polymer degradation) mainly because of the buried
nature of geomembranes. Soil burial greatly diminishes,
and even eliminates many of the degradation processes and
synerglstic effects which have been most widely
Investigated by the polymer industry for exposed
plastics. However, different degradation processes coining
from chemical Interactions and extremely long time frames
may be Involved via exposure to liquids like leachate for
systems intended to last for many decades or even
hundreds of years. Thus the lifetimes of burled
geomembranes can be significantly different than exposed
plastics, but a quantitative method to predict "how long"
is still not available.
This paper describes the various degradation
mechanisms of plastics on an individual basis and then
addresses the various synergistic effects which may
accelerate degradation. It will be seen that synerglstic
effects greatly complicate the situation. While
accelerated test methods are attractive to assess the
various phenomena, these procedures may significantly
misrepresent the actual long-term performance of
geomembranea. Thus the transfer of Information must
proceed with caution.
'Bowman Professor of Civil Engineering and Director,
Geosynthetic Research Institute, Drexel University
Philadelphia, PA, 19104.
"Research Assistant Professor, Geosynthetic Research
Institute, Drexel University, Philadelphia, PA, 19104.
"Professor of Physics, Geosynthetic Research Institute,
Drexel University, Philadelphia, PA, 19104.
106
The
GEOMEMBRANES DURABILITY AND AGING
summarizes the different geomembrane
107
cledfwith observations on field behavior.
iff f>f r.eompmbranes
^
subset of the geosynthetics area, we will refer to them
project, structure or system".
Throughout the 1960's
variety .? ---
the 1910's , tre™.dous
rise to many »«latlon'
book
preferred over clay liners.
or
reducing the many possible
-------
108
WASTE CONTAINMENT SYSTEMS
CO
current use, particularly those used for pollution
control. The test method has a parallel document directed
at assessing the test results, which is in the form of an
expert computer system called FLEX (U.S. EPA 1987). In
the next section, characterization of the most widely
used geomembranes is presented.
Types and Properties of Commonly Used Geomembranes
With the initial caution that polymer formulation and
processing is an on-going and constantly changing
technology, we will attempt to categorize the
geomembranes that are in common use. To be noted,
however, is that our perspective is from a geosynthetic
engineering design point of view. A different grouping
would indeed occur from a polymer chemist's or compound
formulator's perspective,
(a) Stiff {gemi-Crystallinei Thermoplastic Geomembranes -
In the stiff, semi-crystalline, thermoplastic category
are those geomembranes with crystallinity near, or above,
50% which results in stiffness values of greater than
1000 g-cm as per the ASTM D-1388 flexural rigidity test.
This is a bending test developed for fabrics, but it can
be readily used to distinguish stiff-from-flexible
geomembranes.
By far the most important geomembrane in this
category is high density polyethylene (HOPE). There are a
number of variations within HOPE, an important one being
the development of textured sheet which results in a high
friction surface. Also coextrusion manufacturing can
result in very Intriguing composite materials with the
basic sheet being HOPE.
As a polymer formulation, HOPS is almost pure
polyethylene resin (about 97%), in the 0.935 to 0.937
g/cc density range. When carbon black is added for
ultraviolet stability, however, the material's overall
density is at, or slightly above, the 0.941 g/cc lower
limit of ASTM definition of high density polyethylene.
Thus it is commonly referred to as HOPE. The balance of
the compound is 2.0% to 2.5% carbon black and the
remaining 1.0% to 0.5% is antioxidant and processing
aide. The processing aides provide viscosity control,
manufacturing lubrication, and prevent adhesion between
various surfaces. They have also been called viscosity
depressants, slip agents and antiblocking agents. They
are generally proprietary as to their exact chemical
compositions. It should also be noted that the
polyethylene resins themselves have certain uniquenesses,
particularly the nature and extent of tie molecule
bonding and branching.
GEOMEMBRANES DURABILITY AND AGING
109
(b) Fiftyihie (Low crysrallinitvJ
tonrnMibranaa-This group of geomembranes ii-
characterized by stiffness values in the ASTM D-1388
flexuMl rigidity test as being significantly lower than
1000 V-crn. The main geomembranes in this categorj,-are
PVC, CPE, CSPE and VLDPE. Some typical values of flexural
rigidity are the following:
• 0.50 mm polyvinyl chloride !PVC) = 4 g-cm
. 0.75 mm chlorinated polyethylene (CPE) =20 g-cm
. 0.75 mm chlorosulfonated polyethylene CSPEI = 25 g-cm
• 0.75 mm very low density polyethylene (VLDPE) - 78 g cm
While all of the above geomembranes have some
crvstallinity, it is very low in comparison to HOPE. Yet,
all are1 indeed thermoplastic polymers and can be seamed
using thermal methods. The formulations of these
materials vary widely. Some typical values follow in
Table 1.
Table 1 - Typical Formulations for Flexible,
Thermoplastic, Geomembranes
Geo-
membrane
PVC
CPE
CSPE
VLDPE**
Resin
45-50
60-75
45-50
96-98
Plasticizer
35-40
10-15
2-5
0
Carbon Black
(, Filler
10-15
20-30
45-50
2-3
Additive*
3-5
3-5
2-4
1-2
•refers to antioxidant, r.~ .,
**note that this formulation is typical of most
polyethylenes, including HOPE
f/.ip»
-------
110
WASTE CONTAINMENT SYSTEMS
OEOMEMBRANES DURABILITY AND AGING
111
03
indicated are via the ASTM D-1388 test ,for flexural
rigidity:
• 0.91 nun chlorinated polyethylene (CPE-R)- 25 g-cm
• 0.91 mm chlorosulfonated polyethylene (CSPE-R)- 30 g-cm
• 0.66 mm ethylene interpolymer alloy (EIA-R)- 60 g-cm
As with the imreinforced materials just mentioned, all
can be seamed using thermal methods and are truly
thermoplastic. The formulations of the polymer component
are the same as given in Table 1,
Regarding the type of fabric used for the scrim
reinforcement of CPE-R and CSPE-R there are many
possibilities. Woven fabrics of high tenacity polyester
or nylon are the most common. They are often in a pattern
of 10 yarns per inch (4 yarns per centimeter) in both
directions, which is called a 10 x 10 scrim. Other
variations are 6x6 and 20 x 20 patterns, as well as
unbalanced variations. For reinforced geomembranes like
EIA-R, a very tightly woven fabric is used. Hot
specifically mentioned are spread coated fabrics where a
variety of nonwoven needlepunched fabrics can be
utilized.
(d) Other Geomembranfta - While the focus in this paper
will be on the geomembranes Just reviewed, there are many
other possibilities. Two complete groups have been
omitted since they are not currently used to any
significant degree in North America. They are the
thermoset elastomers (Matrecon 1988, Kays 1988) and
polymer modified bitumens (Gamski 1984), the latter group
being used quite often in Europe. Other newer materials
which fall within the categories just mentioned are in
the development stage and undoubtedly more will appear in
the future. This review, however, will be sufficient to
build upon in describing the various degradation
processes and synergistic effects which may occur.
Mechanisms of Degradation
This section of the paper describes various polymer
degradation processes which can act within a geomembrane.
Each process and its resulting implication is taken by
itself as though it were acting in isolation. This, of
course, is not field representative, but we feel that it
is necessary to describe the isolated events before
synergistic effects can be considered. Please note at the
outset that our perspective is from a geosynthetic
engineering design point of view. From a chemistry or
polymeric science point of view there are many references
treating the various subjects in a much more rigorous
(and undoubtedly enlightened) manner.
c
* polyethylene = 300 ran
• polyester = 325 nm
* polypropylene = 3TO nm
unacceptable levels.
The above type of degradation is greatly reduced by
°
the gro»th and
Hindered amlne light stabilizers
added to the P°»V"«
radicals liberated
chemicals
Preventing
such additive, are
For ,.o.e.br,ne application,, . i soil back :t -lii o r
other covering «"f »«" "%pV,?d ^o»..brVne. «.
ttSSSf" ""F-%;S"f Eve"? Its'occirrence5
c. of soil cover Is suMici.m to P««"J £ ln tlnely
i: hTe'ved'ifaU except the foUo.in,
1
-------
114
WASTE CONTAINMENT SYSTEMS
polyethylene and polypropylene geomembranes were
unaffected by the radiation. Furthermore, the radiation
did not have a significant effect on other chemical
degradation rates.
With the absence of radioactive materials in the
waste material, however, the subject becomes a moot
point. Fortunately, this is the case for most solid and
liquid waste contained by geomembranes and related
geosynthetic materials.
(c) Chemical Degradation - The reaction of various
geomembranes to chemicals has probably been studied more
than any other liner degradation mechanism. Most of the
work is laboratory oriented via simple immersion tests
but the body of knowledge is so great that a reasonable
confidence level can be associated with manufacturers
listings and recommendations. Complex waste streams like
leachate, however, are usually not addressed and must be
evaluated on a site specific basis. For this reason the
U.S. EPA developed the Method 9090 procedure. Here
samples of the candidate geomembrane are exposed at 23°C
and at 50°C and removed at 30, 60, 90 and 120 days.
Various physical and mechanical tests are performed and
then compared to the unexposed geomembrane. A percent
change in this behavior is calculated. When plotted for
03 the various exposure times, trends can be established and
u» a decision made as to the nature and degree of chemical
resistance.
Depending on the type of leachate vis-a-vis the
polymeric compound from which the geomembrane is made, a
number of reactions may occur.*
•No reaction may occur, which indicates that the
geomembrane is resistant to the leachate; at least for
the time periods and temperatures evaluated.
•Swelling of the geomembrane may occur which In itself
may not be significant. Many polymers can accommodate
liquid in their amorphous regions without a sacrifice of
physical or mechanical properties. Swelling, however, is
often the first stage of subsequent degradation and a
small loss in modulus and strength may occur. The effect
is often reversible when the liquid is removed.
• Change of physical and mechanical properties, of course,
signifies some type of chemical reaction. The variations
are enormous. Quite often the elongation at break in a
tensile test will be the first property to show signs of
change. It will first occur with the 50"C incubation
data, since this can be considered to be an accelerated
test over the 23"*C incubation data.
• A large change of physical and mechanical properties
signifies an unacceptable performance of the
GEOMEMBRANES DURABILITY AND AGING
115
geomembrane. Limits of acceptability are, however, very
subjective. Table 2 gives a number of recommendations as
accumulated by Koerner, 1990. It should be noted that
there is also an expert computer code available to aid
in t^le decision.
Table 2 - Suggested Limits of Different Test Values for
Incubated Geomembranes, (see the Reference by
Koerner, 1990, for other details and complete
references)
(a) For Flexible, Low Crystallinity, Thermoplastic
Polymers (Unreinforced and Reinforced), after Little
Property
Resistant
Not
Resistant
Permeation Rate {water vapor)
Change in weight {%)
Change in volume (%)
Change in tensile strength (%)
Change in elongation at break (%)
Change in 100% or 200% modulus (%) <30
Change in hardness 10 points
<0.9 g/m2/hr >0,9
<20
<30
>20
>30
>30
10 points
(b) For Stiff, Semi-Crystalline, Thermoplastic, Polymers
Property
O'Toole
Little
Koerner
Resis- Not Resis- Not Resis- Not
tant Resis- tant Resis- tant Resis-
tant tant tant
Permeation Rate
(water vapor)
(g/m2-hr)
Change in Weight
(%)
Change in Volume
<0.5
<0.9
<3
2:0.9
23
<0.9
<2
2:0.9
>2
(%) 0.5 <1 21
Change in Yield
strength (%) <10 >20 <20 220
Change in Yield
Elongation {%) - - <20 220
Change in Modulus
Change in Tear
Strength (1) _ _ -
Change in Puncture
Strength (%) _ - - -
<1
<20
<30
<30
<20
<30
21
>20
230
230
220
230
-------
112
WASTE CONTAINMENT SYSTEMS
1290 3i^l3g_350 3I° 39° 4« «0 4S047Q 4M Slo 530 SM 570 590 °
W»vtl«ngth • nanomaMim
Flguri 1 - Th« Wavttangth SfMOrum ol VHK>I« and UV Solar Radiation.
CO
< i.ooo
10.1,000
1,000-10,000 M UM a
C.OOO MUM 3
«tr 10.000
100 1000 10000 100000 1000000
Failure Tinw (hrs.)
- Schematic Plot of Time of Failure versus Pipe Hoop Stress
for Burst Testing of Un-Notched PE Pipe.
GEOMEMBRANES DURABILITY AND AGING
113
situations.
* Surface impoundments above the liquid level and along
their horizontal runout length
*Canai liners above the liquid level and along their
horizontal runout length
•Covers of surface impoundments, i.e., floating covers
•Landfill liners on side slopes which have had their
surfaces exposed by erosion of cover soil and are
inaccessible.
For the above situations of exposed geomembranes some
amount of degradation over time is unavoidable.
(b) Radiation Degradation - There are a number of reviews
on the effects of radiation on polymer properties
(Phillips 1988, Charlesby 1960S . An extremely brief
summary will be given here. The effects of T~rays,
neutrons and p-rays are essentially equivalent when their
different penetrating powers are considered, p-rays
(electrons) penetrate about a millimeter into a polymer,
whereas frays and neutrons penetrate much further, a-rays
(helium nuclei) penetrate only micrometers hence are only
involved with very near surface damage.
The basic mechanical short term properties of a
typical polymer start to change at a total radiation dose
of between 10* and 101 rads (Phillips 1988) . A rad is
equivalent to 100 ergs of absorbed energy per gram of
material. For reference purposes, the lethal dose of
radiation to a human is about 100 to 200 rads. Therefore
it would appear that if a geomembrane is containing low
level nuclear waste of even lower radiation than the
lethal human dose, the time before significant damage
occurs to its short term mechanical properties will be
quite long indeed. Other, more subtle changes may occur.
For example, even very small amounts of local surface
damage in a semi-crystalline geomembrane might cause a
serious reduction in the stress crack resistance of the
material. The effects of radiation on the additives may
be a more severe problem than the effect on the polymer
itself. It Is possible that after a certain irradiation
the material will be more susceptible to other
degradation processes.
While no test protocol exists for evaluation, some
form of incubation test method can be used with suitable
modifications. Whyatt and Fansworth (1990) have evaluated
m number of different geomembranes in simulated short-
term tests in s high pH (^14 weight percent NaOH)
inorganic solution at 90°C and subjected them to
radiation doses up to 39x10* rad. It was found that only
-------
116
WASTE CONTAINMENT SYSTEMS
03
(d) Degradation by Swelling - One indication of a
geomembrane's durability is the amount of swelling that
occurs due to liquid absorption. It should be emphasized
that swelling per se does not necessarily mean chain
scission nor a failed system. It is, however, a bit
disconcerting, and usually results in a change of
physical and mechanical properties, at least on a
temporary basis.
The test for water absorption, which can be modified
for any liquid, is given in ASTM D570. The test is
directed at a quantitative determination of the amount of
water absorbed, but it is also used as a quality control
test on the uniformity of the finished product. The test
procedure cautions that the liquid absorption may be
significantly different through the edge or through the
surface, particularly with laminated products. (This fact
alone suggests that in seaming of laminated geomembranes,
the upper overlap must be protected against moisture
uptake.) Test specimens of 75 by 25 mm are used and
immersed in a number of possible waysi
• For 2 hr., 24 hr., or 2 weeks of constant immersion in
23'C water.
• Under cyclic (repeated) immersion.
•For 0.5 hr. or 2 hr. of constant immersion in 50°C
water.
•For 0.5 hr. or 2 hr. of constant immersion in boiling
water.
The resulting test data are reported as the percentage
increase in weight using deionized and distilled water.
Some typical values for commonly used geomembranes are as
follows (Haxo, Nelson and Miedema 1985);
• PVC - 3 to 4%
• CPE - 1 to 4%
* CSPE - 1 to 5%
• HOPE - negligible
• VLDPE — not evaluated
Swelling due to other liquids is mentioned in the
reference cited.
(e) Dqffrad,atipn by Extraction - Some polymers exhibit
degradation by the long-term extraction of one or more
components of the compound from the polymeric material.
These are usually polymers which have been compounded
with the use of plasticizers and/or fillers. The as-
formulated and compounded mixture of such polymers is
very Intricate and the bonding mechanism is very complex.
When extraction of plasticizers does occur, a sticky
surface on the geomembrane results with the remaining
GEOMEMBRANES DURABILITY AND AGING
117
structure showing signs of increased modulus and
strength, and a lowering of the elongation at failure,
i.e. the material becomes progressively brittle (Doyle
and Baiter 1989) . The long term behavior, however, is
unknown. It is also possible that anti-degradient
components within the polymer may be extracted and leach
out to the surface. This might indicate that the
remaining polymer is somewhat more sensitive to long-term
degradation.
(f) Pogradatign by Delamination - For geomembranes which
are manufactured by calendering or spread coating,
delamination is a possibility. It is usually observed
when liquid enters into the edge of the geomembrane and
is drawn into the interface by capillary tension. This
can occur between geomembrane plies, between reinforcing
scrim and one of the plies, or between the geomembrane
coating of a fabric substrate as in spread coated
geomembranes. When it occurs, the individual components
are separated and composite action is lost. This type of
wicking action has been problematic in the past but
current manufacturing methods and proper CQC/CQA in field
operations have almost eliminated the situation.
(g) Oxidationpegradation - Whenever a free radical is
created, e.g., on a carbon atom in the polyethylene
chain, oxygen can create large scale degradation. The
oxygen combines with the free radical to form a
hydroperony radical, which is passed around within the
molecular structure. It eventually reacts with another
polymer chain creating a new free radical causing chain
scission. The reaction generally accelerates once it is
triggered as shown in the following equations.
R* + O -> ROO"
ROO + RH -» ROOH + R
where
R* - free radical
ROO* - hydroperoxy free radical
RH - polymer chain
ROOH - oxidized polymer chain
Anti-oxidation additives are added to the compound to
scavenge these free radicals in order to halt, or at
least to interfere with, the process. These additives, or
stabilizers, are specific to each type of resin. This
area is very sophisticated and quite advanced with all
resin manufacturers being involved in a meaningful and
positive way. The specific anti-oxidants are usually
proprietary. Removal of oxygen from the geomembrane's
-------
118
WASTE CONTAINMENT SYSTEMS
surface, of course, eliminates the concern. Thus once
placed and covered with waste, or liquid, degradation by
oxidation should be greatly retarded. Conversely, exposed
geomembranes, or those covered by nonsaturated soil, will
be susceptible to the phenomenon.
Biological Degradation - within the various plant
forms of biological life, i.e., bacteria, actinomycetes,
fungi and algae, polymer degradation is essentially
impossible due to the high molecular weight of the common
resins used in geomembranes. In order for such
degradation to occur the chain ends must be accessible
and this is highly unlikely for molecular weights greater
than 1000, let alone 10,000 to 30,000 which is common for
geomembrane resins. Biological degradation might be
possible for plasticizers or additives compounded with
the resin, but information is not authoritative on this
subject.
Within the higher forms of biological life, i.e.,
protozoa, spiders, insects, moles, rats and small
mammals, polymers do not contain food and thus are
unlikely to be consumed. It is possible, however, that an
animal may try to penetrate the geomembrane for access to
to the opposite side. Here hardness of the predator's teeth
oo enamel versus the geomembrane's hardness is the key
comparison. While such events are possible, authoritative
information on geomembranes being penetrated in this
manner is not known to the authors.
Synergiatie Effects
While not degradation mechanisms within, and of,
themselves, there are several phenomena which can readily
work in conjunction with the previously discussed items.
They generally have the effect of accelerating the
specific degradation process and thus are called
"synergistic effects". At the outset it should be noted,
however, that the quantification of these effects is very
complicated and the data base is very weak in this
regard.
(a) Elevated Temperatures - Whenever the temperature at
the surface of, or within, the geomembrane is increased,
the mobility of the polymer (and all of its other
ingredients) is increased and degradation is usually
accelerated. Of the different degradation mechanisms
mentioned earlier, this is the case for all of them with
the possible exception of biological degradation, which
was seen to be of negligible significance. Clearly,
elevated temperatures accelerate ultraviolet degradation
and this phenomenon has perhaps the largest data base.
Thus extreme conservatism is usually taken when testing
GEOMEMBRANES DURABILITY AND AGING
119
for ultraviolet degradation, fts mentioned earlier,
chemical resistance incubation is usually done at an
elevated temperature of 50°C for comparison to the 23°C
"standa-rd" temperature. Invariably, the higher
temperature produces results having greater changes than
the lower temperature. There is an upper limit for such
temperature testing, however, and that value is based
upon polymer modification not representative of realistic
behavior. Its value is undoubtedly resin dependent but
largely unknown.
For geomembranes placed in the field, high
temperatures can generally be avoided by covering them
with soil, liquid or another material. Thus the buried
environment greatly reduces temperatures in most
geomembrane applications. Notable exceptions are surface
impoundment and canal liners (above the liquid surface)
and for floating covers. For all of these cases simulated
testing is absolutely necessary.
(b> Appliedgtressea - Invariably, the testing for
geomembrane degradation is done on unstressed laboratory
samples. Yet, at the very least, the geomembrane will
have compressive stresses imposed, and quite possibly
tensile stresses as well. What these stresses do to the
degradation of the geomembrane as compared to testing in-
isolation is quite unknown. The reason for this lack of
data is obvious. The cost of experimentation at elevated
temperatures is very high in itself and to stress the
geomembrane in some simulated form of biaxial stress
would generally be cost prohibitive. Yet, some
experimentation with stressed geomembranes should be
initiated, at least to note the severity of the effect.
(c) Long Exposure - Long exposure includes a multiplicity
of effects such as ultraviolet, extraction, oxidation,
etc., which can result in synergistic effects beyond each
of the previously discussed phenomenon taken
individually. For materials as inert as polyethylene, for
example, years of exposure at ambient temperature show no
indication of any change of properties. For other
polymers some changes in surface texture or even in
macroscopic properties might occur, but their influence
on the geomembrane's behavior is not clear.
The major authoritative data base on long-term aging
is available from Matrecon, Inc., (1988), but the steady
development of new polymers and compounds makes the
situation elusive to say the least. It should be
mentioned that a number of landfill owners are beginning
to place geomembrane samples (coupons) in retrieveable
locations for annual exhuming and evaluation. Such
studies will eventually be helpful in assessing actual
-------
120
WASTE CONTAINMENT SYSTEMS
degradation and aging although the coupons are rarely, if
ever, in a stressed condition.
Accelerated Testing Methods
Clearly, the long time frames Involved in evaluating
individual degradation mechanisms at field related
temperatures, compounded by synergistlc effects, are not
providing answers regarding geoinembrane behavior fast
enough for the decision making processes of today. Thus
accelerated testing, either by high stress, elevated
temperatures and/or aggressive liquids, is very
compelling. Before reviewing these procedures, however,
it must be clearly recognized that one is assuming that
the high stress, elevated temperature or aggressive
liquids used actually simulates extended lifetimes... an
assumption which is not readily substantiated. Thus it
might £>e that the test procedures to be described here
actually form lower-bound conclusions in predicting
degradation, i.e., the results may be minimum values but
that is not known with any degree of certainty.
(a) Stress Limit Testing - Focusing almost exclusively
on HOPE pipe for natural gas transmission, the Gas
Research Institute, the Plastic Pipe Institute and the
American Gas Association are all very active in various
aspects of plastic pipe research and development. The
three above-mentioned organizations, together with
Battelle Columbus Laboratories sponsor the Plastic Fuel
Gas Symposia which are held on a biennial basis and the
resulting Proceedings contain many interesting papers.
Stress limit testing in the plastic pipe area has
proceeded to a point where there are generally accepted
testing methods and standards. ASTM D1598 describes a
standard experimental procedure and ASTM D2837 gives
guidance on the interpretation of the results of the
D1598 test method.
In ASTM D1598, long pieces of un-notched pipe are
tightly capped and placed in a constant temperature
environment. Room temperature of 23°C is usually used.
The pipes are placed under various internal pressures
which mobilize different values of hoop stress in the
pipe walls, and the pipes are monitored until failure
occurs. This is indicated by a sudden loss of pressure.
Then the values of hoop stress are plotted versus failure
times on a log-log scale, see Figure 2. If the plot is
reasonably linear, a straight line is extrapolated to the
desired, or design, lifetime which is often 105 hours or
11.4 years. The stress at this failure time multiplied by
an appropriate factor is called the hydrostatic design
basis stress. While of interest for pipelines, the stress
state of geomembranes is essentially unknown and is
CEOMEMBRANES DURABILITY AND AGING
12!
extremely difficult to model. Thus the technique is not
of direct value for geomembrane design. It leads,
however, to the next method.
(b) Bate Process Method (.RPM) for Pipes - Research at the
Gas Institute of The Netherlands (Wolters 1987) uses the
method of pipe aging that is most prevalent in Europe. It
is also an Interna'tional Standards Organization (ISO)
tentative standard currently in committee. Note that
other plastic pipe research institutes also are involved
in this type of research. The experiments are again
performed using long pieces of un-notched pipe which are
tightly capped, but now they are placed in various
constant temperature environments. So as to accelerate
the process, elevated temperature baths up to 80°C are
used. Different pressures are put in the pipes at each
temperature so that hoop stress occurs in the pipe walls.
The pipes are monitored until failure occurs, resulting
in sudden loss of pressure. Two distinct types of
failures are found; ductile and brittle. The failure
times corresponding to each applied pressure are
recorded. A response curve is presented by plotting hoop
stress against failure time on a log-log scale.
The rate process method (RPM) is then used to predict
a failure curve at some temperature other than those
tested. I.e., at a lower (field related) temperature than
was evaluated in the high temperature tests. This method
is based on an absolute reaction rate theory as developed
by Tobolsky and Eyring (1943) for viscoelastic
phenomenon. Coleman (1956) has applied it to explain the
failure of polymeric fibers. The relationship between
failure time and stress is expressed in the form as
follows:
where
log t
time to failure
(1)
T - temperature
o • tensile stress on the fiber
A and
constants
Bragaw (1983) has revised the above model on polymeric
fibers and found three additional equations which yield
reasonable correlation to the failure data of HOPE pipe.
These three equations are as follows:
-------
122
WASTE CONTAINMENT SYSTEMS
log tf - A0 +
log tf - AQ +
log tf - AQ +
A2 log P
A,,?'1 log P
(2)
(3)
where
P - internal pipe pressure proportional to the hoop
stress in the pipe
The application of RPM requires a minimum of two
experimental failure curves at different elevated
temperatures generally above 40°C. The equation which
yields the best correlation to these curves Is then used
in the prediction procedure for a response curve at a
field related temperature e.g., 10°C to 25°C. Two
separate extrapolations are required, one for the ductile
response and one for the brittle response. Three
representative points are chosen on the ductile regions
of the two experimental curves. One curve will be
selected for two points, and the other, the remaining
point. This data is substituted into the chosen equation,
i.e., Equation 1, 2 or 3, to obtain the prediction
equation for the ductile response of the curve at the
desired (lower) temperature. The process is now repeated
for the predicted brittle response curve at the same
desired temperature. The intersection of these two lines
defines the transition time.
Figure 3 shows two experimental failure curves which
were conducted at temperatures of 80°C and 60°C along
with the predicted curve at 20°C. The intersection of the
linear portions of the 20°C curve represents the
anticipated time for transition in the HPDE pipe from a
ductile to a brittle behavior of the material. For pipe
design, however, the intersection of the desired service
lifetime, say 50 years, with the brittle curve is the
focal point. A factor of safety is then placed on this
value, e.g., note that it is lowered to 6.5 MPa in Figure
3. This value of stress is used as a limiting value for
internal pressure in the pipe,
CO gate Process Method fRPMl for Geomembranes - A
similar RPM method to that just described for HDPE pipes
can be applied to HDPE geomembranes. The major difference
is the method of stressing the material. The geomembrane
tests are performed using a notched constant load test
(NCLT). In this test, dumbbell shaped specimens are taken
from the geomembrane sheet. A notch is introduced on one
of the surfaces; the notch depth being 20* of the
thickness of the sheet. The full description of the
GEOMEMBRANES DURABILITY AND AGING
123
logo
(Mpa)
allow
= 6.5
log time
50 yrs.
Figure 3 - Burst Tcft Data for MDPE Pipe, The Intersection of the Ductile Portion
of the 20 C Line and 50 Years Has Been Lowered to 6.5 MPa by
Multiplication with the Appropriate Factor of Safety.
(From Ref. 14)
100
a>
i/i
0)
55
-------
124
WASTE CONTAINMENT SYSTEMS
notching process is described by Halse, et al. (1990).
Tensile loads varying from 30% to 70% of the yield stress
of the sheet, are applied to the notched specimens. The
tests are performed in elevated constant temperature
environments (usually from 40°C to 80°) and in a surface
active wetting agent. Often a 10% Igepal and 90% tap
water is used. The data is presented by plotting percent
yield stress against failure time on a log-log scale.
Figure 4 shows typical experimental curves at 50°C and
40°C which are seen to be very similar to the behavior of
HOPE pipe, recall Figure 3. Here distinct ductile and
brittle regions can be seen along with a clearly defined
transition time.
In order to use these elevated temperature curves to
obtain the transition time for a realistic temperature of
a geomembrane beneath solid waste or liquid impoundments,
e.g., at 25°C, only Equations 3 or 4 can be used due to
the data being plotted on a log-log scale. The following
is a step-by-step procedure of how the experimental data
is used in this regard.
Step. 1 Determine the best fit equation among Equations
3 or 4. We will select Equation 3 which has
been found to show the best fit for both the
ductile and brittle data.
Step. 2 Obtain the equation for the ductile region of
the 25°C curve. To predict the ductile region
of the 25°C curve, two data points were chosen
from the ductile region of the 50°C curve and
one from the 40°C curve. The details of their
values are as follows:
Point 1: T! - 323 °K; Pj - 48% = 9.72 MPa;
tx - 1 h
Point 2: T2 - 313 °K; P2 - 52% = 10.53 MPa;
t2 - 5 h
Point 3: T3 - 313 °K; P3 - 56% - 11.34 MPa;
t3 - 1 h
Note that the calculations are in degrees
Kelvin which is degrees Centigrade plus 273
degrees. By substituting the above three seta
of data into Equation (3), and solving the
resulting three simultaneous equations, three
constants are obtained and are listed below:
A - -37.18,
- 18774, and A2 - -21.2
By substituting these constants back into
GEOMEMBRANES DURABILITY AND AGING
125
Equation (3) the following equation results for
predicting the ductile portion of the curve at
any temperature.
log t = -87.18 + 18774/T - 21.2 log P
(5)
Step 3. Obtain the equation for the brittle region of
the 25°C curve. For the brittle region, two
data points were chosen from the brittle region
of the 50°C curve and one point from the 40°C
curve. The details of their values are as
follows:
Point 1:
323 °K;
37% = 7.50 MPa;
t1 = 10 h
Point 2: T2 = 323 °K; P2 = 28% =5.67 MPa;
t2 = 20 h
Point 3: T3 = 313 °K; P3 = 37% =7.50 MPa;
t, - 30 h
The resulting three constants obtained by
simultaneous solution of the resulting three
forms of Equation 3 are listed below:
-11.8,
= 4853, and A2 = -2.5
Therefore, the equation for predicting the
brittle portion of the curve at any temperature
is as follows :
log t •» -11.8 + 4853/T - 2.5 log P
(6)
Step 4. Use Equations (5) and (6) to obtain the
predicted response curves at the desired
temperature. The ductile behavior from Equation
(5) and the brittle behavior from Equation (6)
at the specific temperature of 25°C was used to
plot the desired curves at 25°C in Figure 4.
Step 5. Intersect the two prediction curves for the
ductile-to-brittle transition time. As seen in
Figure 4 this intersection produces a
transition time for the 25°C temperature
response of 60 hours. Note that the ductile-to-
brittle transition time is increased over ten
fold (from 6 to 70 hours) by decreasing 25
degrees in temperature. It should also be noted
that this curve represents the behavior of the
geomembrane in 10* Igepal solution. The
performance of the same geomembrane will
-------
126 WASTE CONTAINMENT SYSTEMS
probably be very different in other
environments, such as in leaehate or water.
Elevated Temperature and Arrhenius Modeling - Using
an experimental chamber as shown in Figure 5, Mitchell
and Spanner (1985) have superimposed compressive stress,
chemical exposure, elevated temperature and long testing
time into one experimental device. For their particular
tests, three duplicate chambers were operated at 18°C,
48°C and 78°C respectively. At the end of the arbitrarily
designated test period (in the example to be described it
was 18 weeks), the geomembrane samples were removed.
Mechanical tests and chemical analyses were performed on
these incubated samples to monitor if any changes in the
various properties of the geomembranes occurred.
The mechanical tests included the following:
• Tensile Strength and Elongation
* Yield Strength and Elongation
• Stress Cracking Behavior
00
fo
The chemical analyses included the following:
• Differential scanning calorimetry (DSC); for
measuring crystallinity and oxidation induction
time (OIT)
* Infra-red spectrometry (IR); for measuring
concentration of carbonyl groups
• Gel permeation chromatography (GPC); for
measuring the molecular weight and molecular
weight distribution
If there were changes in any of the above properties; for
example, in the concentration of the carbonyl group, the
reaction rate (K) was obtained for each experimental test
temperature (T) . These values were now used with the
Arrhenius equation which is as follows (American National
Standard 1986):
K
Ae
-E/RT
(4)
where
K
A
E
T
R
reaction rate for the process considered
a constant for the process considered
reaction activation energy for the process
considered
temperature !°K -°C + 273)
gas constant (8.314 J/mol-°K)
By plotting "In K" against "1/T", a straight line should
be obtained, see Figure 6. The slope of this line is
"-E/R" for the particular property change being
GEOMEMBRANES DURABILITY AND AGING
127
Hydraulic Load Device
Air (7)
Supply
Liquid Level
Sight Glass
n-3C1oH
Record
1 c-v
^-*
L
HI
I
<5
0
c
e
c
c
c
er c
Liquid
_ 1
Sand
__ „,,
3
3
r~
3
1
Leachale
Recirculalion
Pump
-Heat Transfer
Coils
Perforated
"Plate Press
-Thermocouples
-Test Liner
Drain
Figure 5 - Schematic of Accelerated Aging Column
(after Mitchell and Spanner)
_
to
g
o
<9
o>
oc
In A
Governing Equation:
In K = In A - (|)y
i
78°C 48°C 18°C
Inverse Temperature (1/T)
Figure 6 - Graphical Method of Plotting Reaction Rale Values
for Change in a Specific Geomembrane Property.
-------
128
WASTE CONTAINMENT SYSTEMS
monitored. The constant "A" can also be identified but it
drops out of the equation when comparing the responses at
two different temperatures.
For example, Turi (1981) found that the activation
energy (E) of polyethylene is 109,000 J/mol. This is the
energy requirement for an atom to change its position. If
we use this value in Mitchell and Spanner's tests, then
the degradation that occurs between the two tests at
temperatures of 18°C and 78°C is determined as follows:
"18 + 273
- e
(109.000)
8.314
LI L_l
[351 291J
- 2254
Hence, for their particular process occurring in 18 weeks
at 78CC it would take 2264 times 18 weeks, or 784 years,
to occur at 18°C.
(e) Hoechst Multiparameter Approach - The Hoechst
research laboratory in West Germany has been active in
long term testing of HDPE pipe since the 1950's. Recently
they have been applying their expertise and experience to
the long term behavior of HDPE geomembranes (Kork, et al.
1987) . Note, however, that there are major differences
between pipe and sheet. The stress state in pressurized
pipe is well known, whereas the stress state in the
geomembranes in the field is not nearly as well known.
Furthermore, the pipe is under a constant stress state,
whereas stress relaxation can occur in geomembranes. If
these two differences in the state of stress between pipe
and sheet can be resolved, then the tremendous wealth of
data obtained in over 30 years of pipeline testing can be
carried over to the geomembrane area.
The Hoechst group has considered geomembranes for the
case of local subsidence under the liner and the
mobilization of out-of-plane deformation of the sheet
into an multi-axial stress state. Thus their model for
sheet stresses' is biaxial stress. (In the usual
pressurized pipe, the radial stress is twice the value of
the longitudinal stress. Hence the isotropic biaxial
stress state in a normal pipe testing experiment was
achieved by putting the pipes under an additional
longitudinal stress. It was found that the lifetimes in
these tests were the same as in normal pressure burst
testing. Hence if biaxial stress relaxation could be
accounted for, a viable long term testing technique could
very well be developed for sheet from modified pipe
testing. This of course assumes that the different
manufacturing and processing of pipe and sheet produce
GEOMEMBRANES DURABILITY AND AGING
129
essentially the same material properties. (This may not
be completely the case.) The Hoechst long term testing
for geomembrane "sheet" thus consists of the following
procedure:
(1) Modified burst testing of pipe (of the same material
as the sheet), with additional longitudinal stress
to produce an isotropic, biaxial stress state.
Their tests make note that the site-specific liquid
should be used.
(2) Assume a given subsidence strain versus time
profile.
(3) Measurement of stress relaxation curves in sheets
which have been stressed biaxially, at strain values
encountered in field.
(4) Use steps (2) and (3) to predict the stress as a
function of time.
(5) See how these maximum stresses compare with the
stress-lifetime curves determined in the normal
constant stress-lifetime pipe measurements of Step
(1). The constant stress-1ifetime curves are
modified (as in the normal pipe testing) to
accommodate the effects of various chemicals and
even for seams.
(6) If linear accumulation of degradation is assumed,
then the variable stress curves can be used to
predict failure from the constant stress-1ifetime
curves.
The steps (l)-(5) have been performed by Kork, et al.
(1987) . The approach should certainly be considered
seriously. One of the main impediments to its viability
would seem to be that if one produces pipe, the final
product may have different material properties than
sheet. For example, the residual stresses could be quite
different. Other work of a related nature can be found in
Hessell and John (1987) and Gaube, et al. (1976).
Summary and Conclusions
This rather lengthy treatise on durability and aging
has attempted to give insight into long-term performance
of geomembranes by itemizing those mechanisms which can
degrade the polymer resin and/or compound. Table 3 gives
a summary of the individual degradation mechanisms that
were discussed. All of them, taken individually, are
possible to evaluate and/or quantify. In addition, some
suggestions as to preventative measures are offered.
-------
Table 3 - Degradation Phenomena In Ceomembranes (from a Geosynthetlc Engineering Perspective)
Degradation
Classification
Ultraviolet
Radiation
Chemical
Swelling
Extraction
Degradation
Mechanism
• chain
scission
• bond
breaking
• chain
scission
• reaction with
structure
• reaction with
additives
• liquid
absorption
• additive
expulsion
Initial Change
Laboratory"'
• mol. wt
• stress
crack
resist.
• moL wt.
•stress
crack
resist.
•carbonyl
Index
• m
• mol. wt.
•TGA
•TGA
In Material Subsequent Change In Material
Field*2* Laboratory*31 Field**!
•color
•crazing
• color
•crazing
• texture
• color
•crazing
•TGA
• thickness
•color
•texture
• texture
• color
• thickness
• elongation
• modulus
• strength
• elongation
• modulus
• strength
• elongation
• modulus
• strength
• thickness
• modulus
• strength
• elongation
• modulus
• thickness
• color
• cracking
• color
• cracking
• texture
•cracking
• reactions
•thickness
• softness
• texture
• color
Prcventative
Measure
• screening
agent
• antl*
degradlent
• cover
geomembrane
• cover for 3 and
a- rays
• shield for
neutrons
• reduce dosage
for ? rays
• proper resin
•proper
additives
• proper resin
• proper
manufacturing
• proper
compounds
• proper
manufacturing
WASTE CONTAINMENT SYSTEMS
Table 3 - Continued
Degradation
Classification
Delamlna- •
tlon
Oxidation •
Biological •
Notes:
Degradation
Mechanism
adhesion
loss
reaction
with structure
> reactions
with additives
initial Change
Laboratory"1
• thickness
• edge effects
• IS
• carbonyl
index
• orr
• mol. wt.
• carbonyl
index
• m
jp Material Si
Field»
• thickness
• edge effects
• color
• crazing
• texture
• surface
film
ih«.nu*nt ChanceJn Material
Laboratory131
• thickness
• ply
adhesion
• elongation
• modulus
• strength
• elongation
• modulus
• strength
»M-nrin!tno me
Field*4'
• layer
separation
• thickness
• color
• cracking
• texture
• surface
layer
• cracking
Ihods: mol. wt.
Preventatlv*
Measure
• proper
manufacturing
• protect
geomembrane
edges
• anti-oxidant
• cover with
soil
• cover with
liquid
• avoid sensitive
additives
• blocide
= molecular weight. 1R =
OEOMEMBRANES DURABILITY ANt AGINO
|AJ iiuldcM »o*^msi*w*y *^*«*»*g».»» «*« ^ %•,*•*-—'—--/ — * —
infrared spectroscopy, TGA = thermal geometric analysis. O1T . oxidation inducatlon time
(2) Initial Held changes are generally sensed on a qualitative basis.
(3) Subsequent laboratory changes can be sensed by numerous physical and meehanteal tests. Listed to the table are those
considered to be the most sensitive parameters.
(4> Subsequent field changes are a continuation of the initial changes untU physical and mechanical properties being to
visually change.
B-14
-------
132
WASTE CONTAINMENT SYSTEMS
Also described in the paper were various synergistic
effects which greatly complicate the above mechanisms
when acting concurrently. Items such as elevated
temperature, biaxial or triaxial stress and long exposure
time are very difficult to model accurately.
Nevertheless, much laboratory modeling had been done,
although most has been by the plastic pipe industry, fhis
is almost exclusively on polyethylene pipe and the
technology transfer to polyethylene geomembranes is
certainly very valuable. Many of the techniques are being
evaluated for long-term geomembrane performance; for
example, ductile-to-brittle transition time along with
Arrhenius modeling. Other types of geomembranes appear to
be in need of long-term simulation testing as was
discussed in various facets of the paper.
In conclusion, it is felt that case histories (both
positive and negative) give the best insight into the
field behavior of geomembrane lined facilities at this
point in time. Test strips which are exhumed annually and
tested, versus the original material are extremely
valuable. They can be placed along the edge of the
facility or within sump areas for easy removal. Field
failures are also very important to analyze for aiding
and prompting in the modification of existing polymer
formulations and perhaps changing the basic resins
themselves. Lastly, long term laboratory tests under
simulated conditions should be undertaken. Simulated
stress tests under elevated temperature testing and
Arrhenius modeling is very intriguing in this regard as
is some type of modification of the Hoechst procedures.
Answers to the important question of "how long will they
last" may never be known unless the effort begins as soon
as possible,
References
1 u. S. Federal Register, Regulations on Liner
Systems, July 26, 1982.
f U. s. EPA Method 9090, Compatibility fest for
Hastes and Membrane Liners, in Test Methods for
Evaluating Solid Waste Physical/Chemical Methods, SW-
846, 2nd Edition, 1984,
, U.S. EPA Computer Code on Flexible Membrane Liner
Advisory Expert System (FLEX), Cincinnati, Ohio, 1987.
t ^standard for Polymeric Materials - Long Term
Property Evaluations," American National Standard,
ANSI/UL 746B - 1986.
Apse, J. I., "Polyethylene Resins for Geomembrane
Applications, ** purabtlity and Aging nf GeQsynthafc^f;^,. R.
M. Koerner, Ed., Elsevier Appl. Sci. Publ. Ltd., 1989,
pp. 159-176.
GEOMEMBRANES DURABILITY AND AGING
133
Bragaw, G. G., "Service Rating of Polyethylene Piping
Systems by the Rate Process Method," 8th Plastic Fuel
Gas Pipe Symp., Nov. 1983.
Charlesby, A., Atomic Radiation and PQlmiejLa, Oxford
University Press, 1960.
Chuck, R. T., "Largest Butyl Rubber Lined Reservoir,"
Civil Engineering, ASCE, May 1970, pp. 44-47.
Coleman, B. D., "Application of the Theory of Absolute
Reaction Rates to the Creep Failure of Polymeric
Filaments," Jour. Polymer Science, Vol. 20, 1956, pp.
447-455.
Doyle, R. A. and Baker, K. C., "Weathering Tests of
Kormipmhrangg,- purability and Aging of Geoaynthe.ti.cs, R.
M. Koerner, Ed., Elsevier Appl. Sci. Publ. Ltd., 1989,
pp. 152-158.
Gamski, K., ~Geomembranes: Classification, Uses and
Performance," Jour. Geotex. and Geomemb., Elsevier Appl.
Sci. Publ. Ltd., Vol. 1, 1984, pp. 85-117.
Gaube, E., Diedrick, G. and Muller, V., "Pipes of
Thermoplastics; Experience of 20 Years of Pipe testing,
Kunstoffe, Vol. 66, 1976, pp. 2-8.
Geier, F. H. and Morrison, W. R., "Buried Asphalt Membrane
Canal Lining," Research Report No. 12, A Water Resources
Technical Publication, Bureau of Reclamation, Denver,
Colorado, 1968.
Grassie, N. and Scott, G., po.lymftr Degradation and
Stabilization. Cambridge University Press, 1985.
Halse.Y. H., Lord, A. E. Jr. and Koerner, R. M., "Doctile-
to-Brittle Transition Time in Polyethylene Geomembrane
Sheet," Symposium on Geosynthetic Testing for Waste
Containment Applications, ASTM Spec. Tech. Publ., to
appear in 1990. . ,
Haxo, H. E., Nelson, N. A. and Miedema, T. A., "Solubility
Parameters for Predicting Membrane waste Liquid
Compatibility," Proc. EPA Conf. on Hazardous Waste,
Cincinnati, OH, Apr. 1985, pp. 198-215.
Hessel, J. and John, P., "Long Term Strength of Welded
Joints in Polyethylene Sealing Sheets,"
Werkstofftechnik, Vol. 18, 1987, pp. 228-231.
Kays, «. B., Construction of Linings for Reservoirs. Tanks
flnrt Pollution rnncrol Facilities. 2nd Ed., 3. Wiley and
Sons, Inc., NY, 1988. „,,,,,,
Koerner, R. M., Designing with GeosvnthetiC.S. 2n° Ed"
Prentice Hall Publ. Co., Englewood Cliffs, NJ, 1990.
Kork, R., et. al., "Long Term Creep Resistance of Sheets
of Polyethylene Geomembrane, Report TR-88-0054 from
Hoechst A.G., Frankfurt, W. Germany, 1987.
Matrecon, Inc., Lining of Waste Impoundment and Disposal
Facilities, EPA/600/2-8B.052, Sept., 1988.
Mitchell, 0. H and Spanner, G. E., "Field Performance
Assessment of Synthetic Liners for Uranium Tailings
Ponds," Status Report, Battelle PNL, U.S. NRC, NUREG/CR-
4023, PNL-5005, Jan., 1985.
-------
*
c
o
o
134
WASTE CONTAINMENT SYSTEMS
o
o
3
0
Phillips, D, C., "Effects of Radiation on Polymers,"
Materials Science and Technology, Vol. 4, 1988, pp. 85-
91.
Staff, C. E., "The Foundation and Growth of the
Geomerabrane Industry in the United States," Intl. Conf.
Proc. on Geomembranes, Dever, Colorado, IFAI, 1984, pp.
5-8.
Tobolsky, A, and Eyring, H., "Mechanical Properties of
Polymeric Materials," Jour. Ghent, Phys., Vol. 11, 1943,
pp. 125-134.
Turl, A., thermal Characterization of Polymeric Materials.
Academic Press, 1981.
Van Zaten, R. V., Geotextilea and Geornqmbranes in Civil
Engineering. A. A. Balkema Press, Rotterdam and Boston,
1986.
Whystt, G. A. and Farnsworth, R. K., "The Effect of
Radiation on the Properties of HOPE and PP Liners,"
Symposium on Geosynthetic Testing for Waste Containment
Applications, ASTM Spec. Tech. Publ., to appear in 1990.
Wolters, M., "Prediction of Long-Term Strength of Plastic
Pipe," Proc. 10th Plastic Fuel Gas Symp., 1987, New
Orleans, Amer. Gas Assoc. Publ.
AclcQpwl3d.geme.ntJt
The authors would like to express sincere
appreciation to all of the sponsoring member
organizations of the Geosynthetic Research Institute. A
listing of the firms and their contact members is
available from the authors. The Institute is focused on
long term generic research in geosynthetics of the type
described in this paper.
WASTE
CONTAINMENT
SYSTEMS:
Construction, Regulation, and Performance
Proceedings of a Symposium sponsored by the
Committee on Soil Improvement and Beosynthetics
and the Committee on Soil Properties of the
Geotechnical Engineering Division,
American Society of CM! Engineers
in conjunction with the
ASCE National Convention
San Francisco, California
November 6-7,1990
GEQTiCHNICAL SPECIAL PUBLICATION NO. 26
Edited by Rudolph Bonaparte
Published by the
American Society of CMl Engineers
345 East 47th Street
New Vbrk, New York 10017-2398
-------
United States
Environmental' Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268-1072
Official Business
Penalty for Private Use. $300
Please make all necessary changes on the above label,
detach or copy, and return to the address in the upper
left-hand corner
If you do not wish to receive these reports CHECK HERE D,
detach, or copy this cover, and return to the address in the
upper left-hand corner
EPA/625/4-91/025
-------
ASSESSMENT OF
BARRIER CONTAINMENT
TECHNOLOGIES
A Comprehensive Treatment for
Environmental Remediation Applications
edited by
Ralph R. Rumer and James K. Mitchell
prepared under the auspices of
U.S. Department of Energy
U.S. Environmental Protection Agency
DuPont Company
-------
The opinions, findings, and conclusions expressed are those of the author (s) and not
necessarily those of the workshop sponsors. The workshop sponsors assume no
liability with respect to the use of, or for damages resulting from the use of information,
apparatus, method, or process disclosed in this book. The contents of this publication
are not to be used for advertising or promotional purposes. Citation of brand names
does not constitute an official endorsement or approval of the use of such commercial
products or processes.
This publication is a product of the
International Containment Technology Workshop
Baltimore, Maryland, August 29-31,1995
Credits
Artwork: Barbara E. Evans, Art and Photographic Services, SUNYat Buffalo
Index: Kathryn W. Torgeson, Ithaca. NY
Layout Design: Sue A. O'Donnell, Peripheral Vision, Buffalo, NY
Additional copies of this publication can be obtained from:
National Technical Information Service (NTIS)
U.S. Department of Commerce
5288 Port Royal Rd
Springfield, VA 22161
Telephone; (703) 487-4650
[Request publication #PB96-180583]
-------
TABLE OF CONTENTS
6 Caps 119
6,1 Background 119
6.2 State of Practice 122
6.2.1 Introduction 122
6.2.2 Service Life of HOPE Geomembranes 122
6.2.3 Internal Shear Strength of GCLs 123
6.2.4 Properties of Asphalt 124
6,2.5 Properties of Paper Mill Sludge 125
6.2.6 Slope Stability 126
6.2.7 Water Balance Modeling 128
6.3 Field Performance of the Technology 129
6.3.1 Cap Failures 129
6.3.2 Field Study in Hamburg, Germany 130
6.3.3 Field Study in Beltsville, Maryland 131
6.3.4 Field Study in East Wenatchee, Washington 132
6.3.5 Field Studies in Richland, Washington 132
6.3.6 Field Study in Idaho Falls, Idaho 133
6.3.7 Field Studies in Albuquerque, New Mexico 133
6.4 Assessment of the Technology 134
6.5 Needs 136
6.6 Summary/Recommendations 137
6.7 References 138
-------
119
SECTION 6
prepared by
David E. Daniel and Beth A. Gross, University of Texas
with contributions by
Robert C, Bachus, GeoSyntec Consultants
Craig H, Benson, University of Wisconsin
Jack Boschuk, Jr., J&L Testing Co., Inc.
Subijoy Dutta, U.S. Environmental Protection Agency
Lome G. Everett, Geraghty & Miller
Harley Freeman, Pacific Northwest National Laboratory
Glendon W. Gee, Pacific Northwest National Laboratory
Ed. Kavazanjian, Jr., GeoSyntec Consultants
Robert M. Koerner, Drexel University
Robert E. Landreth, U.S. Environmental Protection Agency
W. Eric Limbach, Idaho State University
Stefan Melchior, University of Hamburg, Germany
Robert W. Ridky, University of Maryland
Paul R. Schroeder, U.S. Army Corps of Engineers
Kenneth Skahn, U.S. Environmental Protection Agency
John C. Stormont, Sandia National Laboratories
Andrew Street, RUST Environmental United Kingdom
Francke C. Walberg, U.S. Army Corps of Engineers
Thomas F. Zimmie, Rensselaer Polytechnic Institute
6.1 BACKGROUND
Caps, also called "cover systems" or "surface barriers", are one component
of the engineered systems used to manage wastes at modern waste disposal
units and, not infrequently, at contaminated sites as part of remedial actions.
The purpose of a cap is to perform one or more of the following functions
(Rumer and Ryan, 1995):
• minimize percolation of water into the underlying contaminated
materials;
• raise the ground surface and provide appropriate slopes to promote
surface-water runoff;
* control the release of gas from the contaminated materials; and
* separate the contaminated materials from humans, animals, and plants.
-------
120 CAPS
Caps are composed of six basic components, from top to bottom: surface
layer, protection layer, drainage layer, barrier layer, gas collection layer, and
foundation layer. The layers are illustrated in Figure 6-1, and the primary
functions and potential materials for each layer are listed in Table 6-1. All
layers must have adequate durability so that they function over the design
life of the cap and adequate shear strength so that the cap surface slopes are
stable. Some layers may contain several materials. For example, a hydraulic
barrier layer may consist of a geomembrane upper component and a com-
pacted clay lower component (i.e., a composite). Not all layers are needed
for all sites. For example, a drainage layer may not be needed at an arid site.
All caps, however, require a surface layer.
Figure 6-1 Cap cross section
The current knowledge and practices for cap technology have been re-
cently reviewed and evaluated (Rumer and Ryan, 1995). The review included
an evaluation of the materials that are used to construct caps, a description of
construction quality control procedures used for barrier layers, and an as-
sessment of factors that can impact the overall performance of caps. The
primary factors that influence cap performance were identified as:
• layer types included in the cap, layer materials, and layer thicknesses,
• annual precipitation,
• surface slope angle of the cap,
• compressibility of the underlying waste,
• gas generation, and
• presence of burrowing animals.
-------
6.1 BACKGROUND 121
TABLE 6-1 Cap components
Layer
Surface
Layer
Protection
Layer
Drainage
Layer
Hydraulic
Barrier
Layer
Gas
Collection
Layer
Foundation
Layer
Primary Functions
•Separate underlying layers from the
ground surface
• Resist wind and water erosion
* Reduce temperature and moisture
extremes in underlying layers
•Store water that has infiltrated the
surface layer until the water can be
removed by evapotranspiration
•Separate the waste from humans,
burrowing animals, and plant roots
•Protect underlying layers from wetting-
drying cycles, which may cause cracking
•Protect underlying layers from freezing-
thawing cycles, which may cause cracking
•Reduce the head of water on the barrier
layer, thereby decreasing water
percolation through the cap
•Reduce pore water pressures in the
overlying cap layers, thereby increasing
slope stability
• Reduce the time the overlying layers are
saturated following rainfall events,
thereby decreasing erosion
•Impede percolation of water through the
cap
•Restrict outward movement of gases
from the waste
•Collect and remove gases to reduce the
potential for uncontrolled gas migration
• Collect gas for energy recovery
•Serve as a foundation for the cap,
especially during construction of layers
requiring compaction (e.g., compacted
clay barrier layer)
Potential Materials
Topsoil (vegetated)
Geosynthetic erosion control
layer over topsoil (vegetated)
Cobbles
Paving material
Soil
Cobbles
Recycled or reused waste
(e.g., fly ash, bottom ash,
and paper mill sludge)
Sand or gravel
Geonet or geocomposite
Recycled or reused waste
Compacted clay
Geomembrane
Geosynthetic clay liner
Recycled or reused waste
Asphalt
Sand or gravel capillary barrier
Sand or gravel
Geonet or geocomposite
Geotextile
Recycled or reused waste
Sand or gravel
Soil
Recycled or reused waste
Select waste
In that review, it was noted that little published information was available
concerning the performance of caps. The performance data are limited be-
cause modern multlcomponent caps have only been in existence for about a
decade, giving insufficient time for much data to be gathered. Also, it is diffi-
cult to quantify the percolation of water or the migration of gas through a cap
on the basis of the available data because the data are not normally collected
for the purpose of evaluating the performance of the cap.
-------
122 CAPS
The purpose of this section is to review the current technological status
of caps, including material performance, design methods, Field performance
(i.e., cap failures), and field monitoring. The focus is on field performance
and field monitoring, as this aspect of caps was identified as an area requir-
ing further research in the earlier publication by Rumer and Ryan (1996). To
this end, researchers and practitioners in the U.S. and Europe involved in the
conduct of field evaluations were asked to share both published and unpub-
lished data. These data are summarized in this section.
It should be recognized that much of the current knowledge about caps
has been developed from landfill capping applications rather than from ex-
periences from containment of contaminated lands. It is important, however,
to learn from the landfill capping experience, since this is the principal source
of the best available information.
6.2 STATE OF PRACTICE
6.2.1 Introduction
The design of a cap for a specific application depends on the requirements of
the application. While the cap requirements for hazardous waste landfills
may differ from those for radioactive waste landfills or municipal waste
landfills, the procedures used to design caps for different applications are
generally the same. The current state of practice for the design of "conven-
tional" caps for municipal solid waste (MSW) landfills has been described by
Othman et al. (1995). As outlined in their paper, major design aspects to be
considered relate to: (i) flow of water into and through the cap; (ii) impacts of
waste settlement on the cap performance; (iii) static and dynamic stability of
the cap; and (iv) surfacewater management and erosion control. Other con-
siderations in cap design include: (i) gas collection and removal; (ii) the
need for bio-barriers; and (iii) long-term durability of cap materials.
Several issues related to cap materials and cap design, namely, the serv-
ice life of high density polyethlyene (HOPE) geomembranes, internal shear
strength of geosynthetic clay liners (GCLs), properties of the alternative bar-
rier layer materials asphalt and paper mill sludge, slope stability of caps, and
water balance modeling, are reviewed in this section.
6.2.2 Service Life of HOPE Geomembranes
Polyethlyene was first synthesized in 1933; however, it was not used in liners
until the 1960s. HDPE geomembranes have not been used long enough to
field test their durability. Nevertheless, the service life of HDPE
geomembranes can be estimated using Arrhenius modeling to interpret the
results of laboratory tests conducted at elevated temperatures (see Section 5).
Research on the service life of HDPE geomembrane is currently being con-
-------
6.2 STATE OF PRACTICE 123
ducted by R.M, Koerner and Y. Hsuan, both of the Drexel University
Geosynthetic Research Institute (GRI). The approach is described by Koerner,
et al. (1992). They identified eight factors influencing degradation of HDPE
geomembrane: (i) oxidation; (ii) chemical attack; (iii) hydrolytic effects; (iv)
ultraviolet radiation; (v) nuclear radiation; (vi) biological attack; (vii) stress
effects; and (viii) temperature effects. Of these factors, oxidation, hydrolysis,
stress, and temperature are being considered in their study, Koerner and
Hsuan defined service life as the time for depletion of antioxidants, initiation
of degradation (i.e., induction time), and degradation to half-life of relevant
strength properties. All of these factors may affect the effectiveness of a
geomembrane as a barrier component. Strength properties are used rather
than permeation properties because loss of antioxidant has little effect on
permeation characteristics but can cause embrittlement and major changes in
strength and elongation characteristics. The study is scheduled to be con-
ducted over a 10-year time period. Based on the approximately three years
of data currently available, depletion of antioxidants from HDPE
geomembrane may take from 45 to 115 years and the geomembrane service
life may be about 250 to 900 years, depending on the specific product and in-
place conditions.
6.2.3 Internal Shear Strength of GCLs
As described by Daniel and Koerner (1993), a cap incorporating a GCL bar-
rier layer can outperform a cap with a compacted clay layer. While a clay
layer will frequently crack due to differential settlement, freezing-thawing
cycles, and wetting-drying cycles, a GCL possesses some self-healing capa-
bility when subjected to these stresses. Also, a GCL is easier to install and to
repair if damaged. One of the primary concerns with using GCLs in caps is
the relatively low shear strength at mid-plane, as reported for hydrated speci-
mens of some GCLs when tested in the laboratory (Rumer and Ryan, 1995).
GCLs fabricated with internal reinforcing (e.g., needlepunched GCLs) have a
higher peak strength than unreinforced GCLs, but, have been found to ex-
hibit a significant decrease in strength when subjected in the laboratory to
relatively large shear displacements. A field-scale study is currently underway
to: 1) evaluate the mid-plane shear strength of different types of GCLs, 2)
verify that caps incorporating GCLs will remain stable on 3H:1V slopes, and
3) verify that stable caps have a factor of safety of 1.5 or greater for slope
stability. This study, which is being conducted jointly by the University of
Texas, Drexel University, and GeoSyntec is described below.
Fourteen cap test plots, 20 m (66 ft) long by 3 m (10 ft) wide, were con-
structed with slopes of 3 horizontal: 1 vertical (3H;1V) and 2H:1V. The cap
cross sections consist of, from top to bottom:
• 0.9 m (3 ft) thick cover soil layer, covered with a geosynthetic erosion
mat and grassed;
-------
124 CAPS
• sand (0.3 m, or 1 ft thick) or geocomposite drainage layer;
• textured geomembrane/GCL composite or GCL barrier layer; and
• subsoil.
The GCLs used in the study were geotextile-encased and needlepunched
(i.e., Bentomat, Bentofix I, and Bentofix II), geotextile-encased, adhesive-
bonded, and stitched (i.e., Claymax), and a geomembrane-bentonite composite
(i.e., Gundseal). To try to force shear failure through the GCL, all geosynthetic
layers above the mid-plane of the GCL, including the upper geotextile of
geotextile encased GCLs, were cut at the top of the slope. Also the cover soil
was removed from the toe of the slope to eliminate toe buttressing. The de-
formations and moisture contents of the GCLs are being monitored at se-
lected depths using tell-tales and fiberglass resistance cells, respectively. Two
cap plots on 2H:1V slopes failed as blocks at 21 and 52 days after the cover
soil was placed. In both cases, the GCLs were internally reinforced, with the
failure surface located between the textured geomembrane and the woven
geotextile of the GCL, rather than within the GCL. Based on the test results
to date, the following observations have been made:
• The factor of safety against mid-plane failure of GCLs on 3H: IV slopes
has been at least 1.5, based on the fact that 2H:1V slopes have been
stable. (Theoretically, if a 2H:1V slope has a factor of safety > 1.0,
then the factor of safety for a 3H:1V slope composed of the same
materials is > 1.5.)
• For reinforced GCLs, the mid-plane strength shortly after construc-
tion has been greater than the interface strength between the woven
geotextile of the GCL and the textured geomembrane.
• Bentonite sandwiched between two geomembranes was found be
partially hydrated at one test plot. The mechanism for this wetting is
unclear, though it may be due to water migration along the tell-tale
and fiberglass resistance cell cables. At the two other test plots with
a similar GCL type, the bentonite between the two geomembranes
has remained dry. The GCLs in the remaining 11 test plots became
hydrated within several months after construction as a result of ab-
sorption of moisture from the underlying, wet subsoils.
6.2.4 Properties of Asphalt
A composite asphalt barrier, composed of a 5-mm (200-mil) thick fluid ap-
plied asphalt membrane (FAM) overlying a 0.15-m (0.5-ft) thick hot-mix as-
phalt concrete (HMAC) layer, is being considered for the cap at the Hanford
site as an alternative to traditional barrier layer materials (Freeman et al.,
1994). Asphalt may be preferable to traditional barrier materials, such as
geomembranes and compacted clay layers, since asphalt appears to have a
longer service life than those materials. The expectation for long service life is
-------
6.2 STATE OF PRACTICE 125
based on the fact that asphalt occurs naturally and is known to have existed
naturally in the subsurface for millions of years. Also, asphalt artifacts, up to
several thousand years old, have been found by archaeologists. Asphalt has
also been demonstrated to have a low permeability and be effective at con-
trolling radon emissions. Asphalt barriers have been installed at several sites.
However, in many cases, asphalt may not be selected over other barrier ma-
terials because of its relatively high cost [e.g., $ 96/m2 ($ 80/ydz) for the as-
phalt barrier proposed for the Hanford site compared to less than $ 7/m2 ($
6/yd2) for installation of a 1.5-mm {60-mil) thick HOPE geomembrane]. Two
concerns with asphalt use in a barrier are: (i) the potential for asphalt to creep;
and (ii) asphalt cracking due to age hardening. Asphalt studies are currently
underway at the Pacific Northwest Laboratory (PNL). Two of these recent
asphalt studies are summarized below.
A 0.6 hectare (1.5 acre) prototype cap incorporating the composite as-
phalt barrier and an adjacent 18.6mx8.6m (61 ftx28ft) barrier test pad were
constructed during 1994 at the Hanford site, located near Richland, Washing-
ton. Laboratory tests conducted on five core samples from the HMAC compo-
nent of the barrier gave an average hydraulic conductivity of 1.3 x 10"9 cm/s.
Field falling head tests conducted at five locations and sealed double ring
inflltrometer (SDRI) tests conducted at two locations gave average hydraulic
conductivity values of 3.7 x 10"8cm/s and 1.1 x 10'8cm/s, respectively, for the
HMAC. The hydraulic conductivity of the FAM component of the barrier, as
measured in the laboratory for four field samples, was on the order of 1 x 10'"
cm/s. It should be noted that the potential exists for preferential flow through
the HMAC layer. The highest HMAC hydraulic conductivity of 1.1 x 10'7
cm/s was found for a core sample taken along a vertical asphalt seam. Also,
the SDRI test conducted over a vertical seam gave a somewhat higher hy-
draulic conductivity than the SDRI test performed over unseamed HMAC.
Lateral flow of water within the asphaltic concrete, between horizontal as-
phalt seams, was observed during the SDRI tests.
Accelerated aging tests have been developed at PNL, permitting the
rheological and chemical properties of asphalt to be determined as a function
of age. The procedure is being validated by comparison with ancient asphalt
artifacts, ranging in age from 500 to 4000 years, and asphalt from naturally
occurring seeps. Accelerated aging tests still need to be completed on a number
of asphalts to allow the long-term performance of composite asphalt barriers
to be predicted.
6.2.5 Properties of Paper Mill Sludge
Paper mill sludges have been used since 1995 as the barrier layer for some
caps constructed in Massachusetts and Wisconsin (Moo-Young and Zimrnie,
1995). However, there is little information in the literature on the engineer-
ing properties of this sludge. Moo-Young and Zimmie performed laboratory
tests to determine water content, organic content, specific gravity, permeabil-
-------
126 CAPS
ity, compaction, consolidation, and strength for seven paper mill sludges. They
found that the sludges had a high initial water content ranging from 150 to
270 percent, an initial hydraulic conductivity ranging from about 1 x 10"7 to 5
x 10'6 cm/s, and behaved similarly to a highly organic soil.
Moo-Young and Zimmie (1995) also performed laboratory tests on six
samples of a sludge used as the barrier layer material in a cap. Three samples
were obtained shortly after construction and the other three samples were
taken at nine, 18, and 24 months after construction. The results of the labora-
tory tests on these undisturbed samples indicated that the water content and
hydraulic conductivity of the sludge decreased somewhat over time as the
sludge consolidated and biodegraded (i.e., mineralized to become more like
a soil). The depth of frost penetration in the sludge barrier layer has been
monitored since 1992. To date, the frost layer has not penetrated into the
sludge layer due to the protection provided by the overlying cap layers and
the high water content of the sludge. Based on the results of laboratory tests
over a range of water contents, if a sludge layer is subjected to freezing and
thawing cycles, the hydraulic conductivity of the sludge may increase by one
to two orders of magnitude (Moo-Young and Zimmie, 1995).
6.2.6 Slope Stability
Static and seismic slope stability analyses are typically carried out as part of
cap design. The general procedures for evaluating the static and seismic sta-
bility of landfills caps have been summarized by Othman et al. (1995). One of
the more critical parts of the analyses involves the selection of the appropri-
ate shear strength values to use for the cap materials and material interfaces.
While there are published shear strength values, it is recommended that
project-specific testing be conducted under the expected field conditions (e.g.,
soil moisture content and unit weight, consolidation load and time, interface
wetting conditions, normal stresses, shear direction, shear displacement rate
and magnitude). Other factors which may affect the long-term shear strength
properties of caps, such as freezing-thawing cycles, heating-cooling cycles,
and creep, should also be considered. However, no consistent standard of
practice currently exists for directly addressing the potential effects of these
factors on slope stability. They may be accounted for indirectly through use
of a higher safety factor. Other measures, such as increasing the thickness of
cover soils above the critical layers to provide thermal insulation and isola-
tion from the environment, may be beneficial in some ways but detrimental
in others.
Seismic design of caps has been brought to the attention of the engineer-
ing community with the promulgation of the Resource Conservation and
Recovery Act (RCRA) Subtitle D regulations for municipal solid waste landfills
(40 CFR 258) which became effective in 1993 and with the observed field per-
formance of several landfills during recent earthquakes in California. Under
the Subtitle D regulations, the performance of landfills located in seismic
-------
6.2 STATE OF PRACTICE 127
impact zones must be evaluated, A seismic impact zone is an area with 10
percent or greater probability that the maximum horizontal accelerations in
bedrock will exceed 10% of the earth's gravitational pull within 250 years.
With this definition, seismic impact zones cover most of the western U.S. and
part of the central and eastern U.S. Overall, these zones cover approximately
40 percent of the U.S., including areas where seismic design may not have
been considered in the past. Furthermore, under Subtitle D regulations, a
geomembrane/compacted clay composite barrier layer be incorporated in
caps over certain landfills, e.g., landfills containing composite bottom barri-
ers. Incorporating a geomembrane (as well as other geosynthetics) in the
cap, however, may make the cap more susceptible to instability and deforma-
tions induced by seismic loading due to relatively low interface shear
strengths.
Observations of the seismic performance of numerous (over 30) landfills
in California, based primarily on field inspections after recent earthquakes,
indicate that, in general, the landfills performed well but that some limited
damage did occur to soil covers (Anderson and Kavazanjian, 1995). Typical
impacts of earthquakes on landfills included: cracking of interim cover soils,
some downslope movement of the cover soils, and disruption of gas collec-
tion systems. No observations are available for landfills with geosynthetic
final caps. However, significant displacements could occur along interfaces
between geosynthetics and other geosynthetics or soils.
The state of practice for evaluating the stability and deformation of
landfills under seismic loading has recently been summarized by Anderson
and Kavazanjian (1995). As described in their paper, the seismic design of
caps involves four steps:
• characterization of the ground motions for the design earthquake,
• evaluation of the response of the landfill to the design earthquake,
» calculation of the stability and deformation of the entire waste mass,
and
• evaluation of the ability of the cap to maintain its integrity when sub-
jected to the calculated ground motions.
An important consideration relative to seismic loading is the potential
for the waste mass to amplify free-field ground accelerations. Amplification
is known to occur in soil deposits and has been observed at the OH Landfill
near Los Angeles, the one landfill at which instrument measurements are
available. A threefold amplification of peak ground acceleration from the base
of the landfill to the crest was recorded during the 1992 Landers earthquake
at the landfill. Amplification can be predicted analytically if the design ground
motions and the dynamic properties of the waste are known. However, in
the central and eastern U.S., there is considerable uncertainty about the na-
ture of the design motions. There is also considerable uncertainty about the
dynamic properties of wastes, especially municipal solid waste. Factors in-
-------
128 CAPS
fluencing seismic resistance of the landfill cover are the yield acceleration of
the cover system and the allowable seismic deformations. Yield acceleration
depends on the shear strength of the considered failure mass. For geosynthetic
materials, it is not clear when peak, residual, or deformation-compatible
strengths should be used. Additionally, most interface strength tests involv-
ing geosynthetics are run at displacement rates well below the rates occur-
ring during seismic loading. The interface shear strength may be higher or
lower at the actual displacement rate caused by the earthquake, depending
on the material used and on moisture conditions. It is extremely difficult to
design an unconditionally stable cap using geosynthetic materials in areas of
high seismicity. Even in areas of low to moderate seismicity, unconditional
stability may be difficult to obtain for covers containing geosynthetics that
are steeper than 3H:1V. Allowable seismic deformations of the cap are based
on practical considerations rather than rigorous analysis. For noncritical caps,
seismic deformation of the cover may be handled as a maintenance issue.
6.2.7 Water Balance Modeling
One of the principal functions of a cap is to limit percolation of water into
underlying contaminated materials. The design of the cap to limit percola-
tion of water through it requires a model that can predict water percolation
through caps. Such models are called "water balance models". Because wa-
ter balance models are essential to performance-based design, they received
considerable attention and discussion at the workshop.
Several computer models are available to evaluate the hydraulic perform-
ance of landfill caps. The most widely used model is the USEPA Hydrologic
Evaluation of Landfill Performance (HELP) model (Schroeder et al., 1994a,b).
The HELP model has the advantage over other models in that it contains
default climatic and material properties data, is relatively easy to use, and is
accepted by the regulatory community. The HELP model, however, contains
a number of simplifying assumptions that make the model inappropriate in
certain cases. For instance, the HELP model, which generally assumes a unit
gradient to model vertical drainage, cannot be used to simulate the perform-
ance of partially saturated layers in very dry environments. Another model
that has been used to model caps is UNSAT-H. This model, which was devel-
oped by Pacific Northwest Laboratory by Payer and Jones (1990), does not
contain the default data of the HELP model and requires extensive computer
time to conduct simulations. However, UNSAT-H performs a much more
rigorous analysis of unsaturated flow by solving the relevant partial differen-
tial equation, rather than assuming unit gradient as the HELP model does. A
limited number of field studies and analytical assessments have been per-
formed to evaluate the reliability of models as tools to predict trends and
magnitudes of the different landfill water balance components. However,
the findings from these studies are not in general agreement. For example,
some of the studies found that the models overpredicted percolation in hu-
-------
6.3 FIELD PERFORMANCE OF THE TECHNOLOGY 129
mid climates and underpredicted percolation in arid climates, while other stud-
ies concluded the opposite. In many cases, the models were unable to predict
short-term trends. However, for a number of cases they appeared to give rea-
sonable predictions of cumulative water balances. The developers of the HELP
model report that the approximate annual errors in water balance components
calculated by the HELP model are:
• runoff = ± 25 percent of actual value or ± 2 percent of precipitation,
• evapotranspiration = ± 10 percent of actual value or ± 6 percent of
precipitation,
* percolation = ± 10 percent of actual value or ± 0.3 percent of precipi-
tation, and
• lateral drainage = ± 7 percent of precipitation.
In a study of field-scale caps [30 m x 30 m (100 ft by 100 ft)] in Atlanta,
Georgia and East Wenatchee, Washington, incorporating compacted clay bar-
rier layers, the measured water balances of the caps were compared to the
simulated water balances using the HELP and UNSAT-H models (Khire,
1995). The caps were constructed with two layers: (i) a 60- or 90-cm (2- or 3-
ft) thick compacted clay barrier layer; and (ii) a 15-cm (0.5-ft) thick vegetated
surface layer. Collection of climate, runoff, soil water content, and percola-
tion data for each cap began in 1992 and is ongoing. Runoff and percolation
are collected in tanks and measured, while soil moisture is measured using
time domain reflectrometry. The cumulative percolations measured for the
caps in Georgia and Washington were about 24 cm (9.4 in.) and 3.1 cm (1.2
in.), respectively. Khire found that for the Georgia cap, HELP overpredicted
percolation by about 320 percent and and UNSAT-H underpredicted perco-
lation by about 24 percent. The relatively large error in percolation predicted
by the HELP model was attributed primarily to the underestimation of run-
off. Both models overpredicted percolation by about 43 percent for the cap in
Washington. This deviation was believed to have been caused by preferen-
tial flow through cracks and animal borrows observed in the barrier layer.
Khire also found that while both models captured seasonal variations in run-
off, evapotranspiration, soil water storage, and percolation, UNSAT-H simu-
lated the variations more accurately.
6.3 FIELD PERFORMANCE OF THE TECHNOLOGY
Some cap failures and findings from other field studies being conducted in
the U.S. and Europe are described in this section.
6.3.1 Cap Failures
There have been a number of documented cases of cap failures, with most of
-------
130 CAPS
the failures occurring during or shortly after construction and resulting from
excessive erosion, build up of pore water pressures in the cap layers, lack of a
drainage layer, a drainage layer with insufficient capacity, or incorrect "esti-
mation" of the shear strength between the cap layers (Boschuk, 1991). While
most of these failures did not involve rupture of the barrier layer, they were
costly to repair. In one recent case, severe erosion problems developed be-
cause the cap slopes were relatively long (180 m (600 ft)) and the cap drain-
age layer was designed with an outlet only at the toe. At some locations, the
cap had eroded to the top of the clay barrier layer. The erosion problem was
exacerbated in some cases because the drainage layer outlet at the toe of the
slope had not been constructed. In these cases, the trapped water eventually
caused pore pressures to become excessive, causing sloughing of the overly-
ing soil layers at the toe of the slope. At another landfill, a gabion-lined chan-
nel for surface water slid down the cap slope due to liquifaction of the fine
sand beneath it brought about by high pore pressures. It was noted that most
of the failures occurred in states with relatively restrictive, prescriptive cap
designs rather than more flexible performance objectives, suggesting that, in
these states with prescriptive cap designs, greater attention may have been
given to regulatory compliance than to the design itself.
6.3.2 Field Study in Hamburg, Germany
Six test caps [10 m (33 ft) wide x 50 m (160 ft) long] were constructed in 1987
to evaluate the field performance of different cap configurations (Melchior
and Miehlich, 1989; Melchior et al, 1994). The caps were constructed with a
0.75-m (2.5-ft) thick sandy loam topsoil layer, underlain by a 0.25-m (0.82-ft)
thick fine gravel drainage layer. The drainage layer was underlain by one of
four barrier layer types: 1) a 0.60-m (2.0-ft) thick compacted clay layer, 2) a
HOPE geomembrane/clay composite layer with welded geomembrane pan-
els, 3) a geomembrane/clay composite layer with overlapped geomembrane
panels, and 4) a compacted clay layer overlying a 0.60-m (2.0-ft) thick fine
sand wicking layer and a 0.25-m (0.82-ft) thick coarse sand/fine gravel capil-
lary barrier.
Each of the four cap configurations was constructed on 4 percent or 20
percent slopes, and several configurations were constructed for both slopes.
Climate, lateral drainages from the topsoil and drainage layers, runoff, and
percolation data are being collected. Soil moisture data are also being col-
lected from several test caps using neutron probes and tensiometers. The
preliminary findings of this field study are summarized below.
For the caps with the compacted clay barrier layer, little percolation was
observed for the first 20 months after construction. Beginning in August 1989,
percolation began to increase and show a correlation with precipitation events.
The summer of 1992 was very dry, and tensiometers indicated that the clay
layers had undergone more drying than usual. This drying resulting in an
almost tenfold increase in percolation measured during the fall of 1992 over
-------
6.3 FIELD PERFORMANCE OF THE TECHNOLOGY 131
percolation recorded a year earlier. Flow through the capillary barrier under
one of the compacted clay layers was first observed during this time. When
excavations were made into the caps in 1993, the clay layers were found to
have small fissures and contain plant roots. Since 1993, the network of plant
roots has developed further, contributing to preferential flow paths and des-
iccation cracking. Percolation through the compacted clay layers is still in-
creasing and was about 200 mm (8 in.) in 1994.
The caps constructed with a composite barrier layer performed much bet-
ter, and no percolation has been observed. However, during the summer and
fall months, the matric potential in the clay layers increased and drainage from
the clay layer was recorded. This drainage, which had typically been less than
1 mm/yr (0.04 in./yr), has been attributed to thermal gradients. During the
summer and early fall, the temperature at the top of the clay layer has been
greater than that at the bottom of the clay layer, and water likely flows in liquid
and vapor phase from the hotter to cooler regions resulting in the measured
drainage. The water loss caused by thermal gradients has not caused shrink-
age of the clay, although the potential for future shrinkage exists.
6.3.3 Field Study in Beltsville, Maryland
Six lysimeters [14 m (45 ft) wide x 21 m (70 ft) long x 3.0 m (10 ft) deep] with
20 percent side slopes were constructed between May 1987 and January 1990
to evaluate caps incorporating either a compacted clay barrier layer, a rock
capillary barrier, or bioengineering management, which combines enhanced
runoff and plant transpiration (Schultz et al., 1995). The bioengineering man-
agement option used alternating aluminum and fiberglass panels as the sur-
face layer over about 90 percent of the cap with moisture-stressed vegetation
(i.e., Pfitzer junipers) located along gaps in the panels. This latter option
requires periodic maintenance and is intended to be used when significant
subsidence of the underlying waste is expected. The six lysimeters were con-
structed with the following caps: (1) bioengineering management, with the
initial water level 90 cm (35 in.) above the bottom of the lysimeter; (2) bioen-
gineering management, with the initial water level 190 cm (75 in.) above the
bottom of the lysimeter; (3) reference lysimeter similar to lysimeter 1, except
without the surface panels and vegetated with fescue grass; (4) rip-rap sur-
face layer and gravel drainage layer over a compacted clay layer; (5) veg-
etated soil surface layer, gravel drainage layer, and compacted clay layer over
a gravel capillary barrier; and (6) vegetated soil surface layer and gravel drain-
age layer over a compacted clay layer. All of the caps were constructed over
native soil. Rainfall, runoff, deep drainage, and soil moisture content data
were collected for the lysimeters.
The data collected through 1994 reveal that initially ponded water in
lysimeters 1 and 2 was removed by the plants within two years after con-
struction. The soils in these lysimeters have generally become drier over time.
The initial water level in lysimeter 3 rose until it was near the surface of the
-------
132 CAPS
lysimeter and the water had to be pumped out. Deep drainage has been
measured for this lysimeter every year of the study. Except for the deep drain-
age from lysimeter 5, which occurred during 1994, deep drainage has not
been observed from lysimeters 4 to 6. It has been noted that the moisture
content of the clay layer in lysimeter 4 has been increasing, indicating the
possibility of future seepage through the clay. The moisture contents of the
clay layers in lysimeters 4 and 6 show some seasonal cycling, with the lowest
moisture contents being measured in the summer.
6,3,4 Field Study in East Wenatchee, Washington
Two caps [30 m x 30 m (100 ft x 100 ft)], one with a 60-cm (2-ft) thick compacted
clay barrier layer overlain by a 0.15-m (0.5-ft) thick vegetated soil surface layer
and the other with a 75-cm (2.5-ft) thick sand capillary barrier layer used in lieu
of a clay resistive barrier, were constructed and monitored (Khire, 1995). Cli-
mate, runoff, percolation, and soil moisture data have been continuously col-
lected since November 1992. The collected data show that cumulative
percolations through the resistive and capillary barrier layers have been 3.1
and 0,5 cm (1.2 and 0,2 in.), respectively. Most of the water movement through
the capillary barrier occurred in winter 1993, primarily due to snowmelt from
the relatively high snowfall (169 cm (66,5 in.)) occurring that year.
6.3,5 Field Studies in Richland, Washington
Since 1985, PNL and Westinghouse Hanford Co. have been working to de-
velop a cap design for the Hanford site. Field tests have been conducted for
over the past seven years using lysimeters to evaluate the performance of
different cap materials and configurations (Petersen et al., 1995). Currently,
24 lysimeters are being monitored to assess the effects of varying precipita-
tion, surface soil, and vegetative conditions. No drainage has been measured
from lysimeters with vegetated or nonvegetated silt-loam surfaces under
normal precipitation conditions; however, some drainage has occurred from
lysimeters with nonvegetated silt-loam surfaces under extreme precipitation
conditions (i.e., three times normal). Significant quantities of water have
drained from lysimeters with gravel and sand surface layers. The perform-
ance of one lysimeter with a 1.5-m (5-ft) thick layer of nonvegetated silty
loam was modeled over a six-year period using the HELP and UNSAT-H
models. The HELP model simulation prediction was 1800 percent greater
than the observed drainage, while the UNSAT-H model simulation predic-
tion was at 52 percent of the observed drainage.
A prototype cap [0.6 hectare (1.5 acre)] was also constructed at the site in
1994 using the following layers, from top to bottom (Gee et al., 1994; Wing
and Gee, 1994; Peterson et al., 1995):
* 1.0-m (3-ft) thick silt loam/admix gravel surface layer,
-------
6,3 FIELD PERFORMANCE OF THE TECHNOLOGY 133
• 1.0-m (3-ft) thick silt loam protection layer,
• 0.15-m (0,5-ft) thick sand filter,
* 0.30-m (1-ft) thick gravel filter,
• 1.5-m (5-ft) thick fractured basalt riprap capillary barrier and
biobarrier,
» 0.3-m (1-ft) thick gravel cushion and drainage layer,
• 0.15-m (0.5-ft) thick composite asphalt barrier layer,
* 0.10-m (0.3-ft) thick top course, and
* compacted soil foundation.
The composite asphalt barrier layer was discussed previously. The wa-
ter and wind erosion, biointrusion, revegetation success, and water balance
of the cap continue to be monitored. Water balance components being re-
corded include: precipitation, runoff, snow depth, soil moisture, and percola-
tion. The water balance is being evaluated under normal and stressed (i.e.,
irrigated) conditions.
6.3.6 Field Study in Idaho Falls, Idaho
A replicate field test program is underway at Idaho National Engineering
Laboratories to compare the hydraulic performance of four caps designs: (1)
1-m (3-ft) thick vegetated soil layer over a geomembrane/ compacted clay
composite barrier layer; (2) 2.5-m (8.2-ft) thick vegetated soil layer with a 0.5-
m (1.6-ft) thick biobarrier located within it at 0.5 m (1.6 ft) below the ground
surface; (3) the same design as cap 2, except the biobarrier is located 1 m (3 ft)
below the ground surface; and (4) 2.0-m (6.6-ft) thick vegetated soil layer.
The test plots were constructed in 1993. Two vegetation types have been
used, a native mixed plant community and a monoculture of crested
wheatgrass. Both vegetative covers were considered since planted
monocultures may be reinvaded by native plant species in the future, and a
mixed native plant community may be more resilient to environmental fluc-
tuations. The test plots will, at times, be subjected to burrowing animals,
ants, and high levels of irrigation. Climate, soil moisture, and percolation
data are being collected. Soil moisture is being measured using a neutron
probe and time domain reflectometry. No data are available.
6.3.7 Field Studies in Albuquerque, New Mexico
A large-scale field test program is being conducted at Sandia National Labo-
ratories to compare the performance of three caps [each 13 m (43 ft) wide and
100 m (330 ft) long]; one incorporates a compacted clay, another a
geomembrane/clay composite, and a third has a geomembrane/GCL com-
posite barrier layer. The test caps were constructed and instrumented during
1995 (Dwyer, 1995). The hydrology and the erosion of the caps are being
monitored. Another three caps are scheduled to be constructed in 1996, each
-------
134 CAPS
having capillary barriers with one vegetated to enhance evapotranspiration.
No data on performance had been collected at the time of the workshop. Moni-
toring is expected to continue for the next several years.
In another field study under limited evapotranspiration conditions, the
performances of two gravel capillary barriers were compared (Stormont, 1995).
When a uniformly-graded sand layer was placed between the gravel barrier
and the overlying silty sand, the lateral drainage above the barrier increased
while drainage through the gravel decreased.
6.4 ASSESSMENT OF THE TECHNOLOGY
Several aspects of cap materials and design, namely the service life of HOPE
geomembranes, the internal shear strength of GCLs, the properties of asphalt
and paper mill waste, the slope stability of caps, and water balance modeling,
have been reviewed. The findings from this review are presented below:
• Preliminary results from long-term laboratory testing of HOPE
geomembranes indicate that service life of the geomembranes should be
at least several hundred years, depending on the specific product and
the in-place conditions. If HOPE geomembranes provide this length of
service, they appear to be a good investment relative to other more costly
barrier layer materials such as compacted clay.
• Based on ongoing field-scale tests of prototype caps incorporating GCLs,
the factor of safety against mid-plane failure of GCLs on 3H:1V slopes
has been observed to be 1.5 or greater. These results are encouraging
since a cap with a GCL is a more effective barrier against infiltration than
a cap with a compacted clay layer. If GCLs remain stable on cap slopes,
they appear to be preferable to a compacted clay layer.
• Asphalt and paper mill waste have been used as the barrier layer material
for caps. However, the service life of these materials is uncertain. While
asphalt may have a longer service life than traditional barrier layer
materials, it is significantly more expensive. Also, the significance of flow
along asphalt seams is unknown. Paper mill waste involves reuse of a
waste product; however, it may be affected by some of the same processes
(e.g., freezing/thawing cycles) that adversely impact compacted clays.
• The shear strength properties of cap components should be evaluated by
conducting project-specific tests under the expected field conditions. The
effect of freezing-thawing cycles, heating-cooling cycles, and creep on
the long term shear strength properties should also be considered by using
a higher safety factor or by increasing the thickness of cover soils above
the critical layers to provide thermal insulation and isolation from the
environment.
• Observations of the seismic performance of several landfills have indicated
that, while landfills have performed relatively well, cracking and some
-------
6.3 ASSESSMENT OF THE TECHNOLOGY 135
downslope movement of interim cover soils and disruption of gas collection
systems occurred. The effect of the seismic motions on the integrity of caps
incorporating compacted clay layers or geosynthetics has not been well
documented. However, there is concern that compacted clay layers could
be prone to cracking and that displacements may occur along interfaces
between two geosynthetics and between a geosynthetic material and soil.
• There is considerable uncertainty about the dynamic properties of wastes,
especially unconsolidated municipal solid waste, and about the appropriate
shear strength values to use for geosynthetic materials under dynamic
loading.
* Allowable deformations of caps under seismic conditions are based on
practical considerations rather than rigorous analysis. For non-critical caps,
seismic deformation of caps may be handled as a maintenance issue.
* Water balance models are used to predict the performance of caps in terms
of water percolation through the cap. Available water balance models that
are applied to caps typically contain numerous simplifying assumptions
and have been inadequately verified by field data. Since these models
provide wide-ranging estimates of water infiltration under site-specific
conditions, their main value may be to compare alternative designs utilizing
different cap configurations and materials.
Based on the field monitoring of test caps and cap failures, the following
remarks are made regarding cap performance:
* Several examples were given of inadequately protected compacted clay
barrier layers that degraded after a few years as a result of desiccation, root
penetrations, or both. Although covering a clay layer with a geomembrane
provided greatly improved protection, one case suggested that thermally
induced flow could eventually desiccate even a geomembrane-covered clay
barrier layer.
« There are few data on the performance of caps incorporating geomembranes
and compacted clay layers and even fewer data for caps that include
capillary barriers or employ surface vegetation to enhance
evapotranspiration. Therefore, we are unable to document how well
existing caps are performing.
» The performance life of caps has not been established. Although the service
life of some components of the system can be estimated, the functional life
of the surface layer and barrier layer are not well documented and the long
term performance of constructed cap systems has not been adequately
documented.
• A number of cases of cap failures have been documented; however, most
of these failures could have been prevented through proper design and
construction. It is felt that some failures may have resulted from
preoccupation with regulatory compliance rather than engineering design
considerations.
-------
136 CAPS
The primary factors adversely affecting cap performance for each of the
six basic layer components are summarized in Table 6-2.
TABLE 6-2 Factors Affecting Cap Performance
Layer
Surface Layer
Protection Layer
Drainage Layer
Barrier Layer
Gas Collection Layer
Foundation Layer
Factor
« Erosion
* Evapotranspiration
• Native versus Exotic Vegetation
» Appropriate Armoring for Side Slopes at Arid Sites
• Erosion
• Slope Failure Due to Pore Pressure Buildup
* Animal Burrows
• Clogging
• Insufficient Capacity
* Insufficient Drainage Layer Outlets
« Cracking due to Desiccation, Deformations From
Waste Settlement, or Seismic Motions (Clay, Paper
Mill Sludge)
• Root Penetration
« Resistance to Gas Migration (GCLs)
* Stability
• Creep (asphalt)
• Service Life
« Adequate Cover Over Waste
* Adequate Strength
6.5 NEEDS
The following needs related to cap technology are presented based on the
assessment of caps presented in this section.
• Data are available that demonstrate that the performance of compacted
clay barrier layers in caps will deteriorate over time. Even so, compacted
clay barrier layers are still being used in caps, primarily because they are
specified in regulations. There is a perception that it may be difficult to
obtain regulatory approval to use alternative barrier materials to a clay
layer. This situation could be improved if guidelines were available for
demonstrating the equivalency of performance among the different
options for cap components.
• Few data are available concerning the hydraulic performance of
traditional caps with resistive barriers. There are even fewer performance
-------
6.3 SUMMARY/RECOMMENDATIONS 137
data for caps containing capillary barriers. More data need to be collected
to assess cap performance. Data are especially needed to bring about
regulatory and community acceptance of alternative cap configurations.
While some of these data can be collected from currently instrumented
sites, other additional sites will probably need to be monitored.
The expected performance life of caps is uncertain. Studies are underway
to assess the service life of some individual cap components. However,
the long term service life of these components in a constructed surface
barrier system has not been adequately studied.
There have been a number of documented cases of cap failures; however,
most of these failures could have been avoided through proper design
and construction. There is a need for more guidance on cap design and
for independent peer review of completed designs prior to construction.
Compliance with regulations is not a sufficient check on a completed
design.
More field observations on the effect of seismic motions on the integrity
of caps incorporating compacted clay layers and/or geosynthetics need
to be made.
The shear strength at interfaces between materials is known to be an
important factor affecting the physical stability of caps on slopes. The
shear strength is also affected by freezing-thawing cycles, heating-cooling
cycles, and creep. Standard procedures for evaluating interfacial strength
need to be developed.
More information is needed on the dynamic properties of wastes, and on
the appropriate dynamic shear strength values to be used for geosynthetic
layer materials.
Other alternative barrier layer materials, such as asphalt and paper mill
waste, appear promising for future use. However, more information needs
to be collected on the long-term performance of these materials.
Available computer models for simulating the hydrologic and hydraulic
performance of caps need to be verified by comparison with field data,
and modified as necessary.
6.6 SUMMARY/RECOMMENDATIONS
The purpose of this section was to define the current technological status of
waste containment caps, including: material performance, design methods,
field performance, and field monitoring. With respect to material perform-
ance and design method, the following issues were considered: service life of
HDPE geomembranes; internal shear strength of GCLs; properties of asphalt
and paper mill sludge; slope stability of caps; and water balance modeling.
Cap failure cases and study results from field monitoring at sites in arid and
temperature climates have been reviewed. Based on these considerations and
review, it is concluded that:
-------
138 CAPS
• The service life of caps is uncertain. The performance of a compacted
clay barrier layer in a cap will likely deteriorate over time, while an HOPE
geomembrane barrier layer may perform satisfactorily for several hundred
years. The long-term performance of cap components, and caps as
systems, needs to be studied further.
• The slope stability aspects of caps are complex. The shear strength
properties of cap components and component interfaces may be impacted
by moisture conditions, stress, creep, temperature, and seismic motions.
However, there are no standard procedures for evaluating interfacial
strength or accounting for these effects. In addition, the dynamic
properties of buried wastes are generally not well understood. Under
certain conditions, the waste may amplify seismic-caused ground
movements which can be transmitted to the cap. Standard procedures
need to be developed for evaluating the shear strength of cap materials
and additional research needs to be performed on the seismic properties
of various types of buried wastes.
* There are few published data on the field performance of constructed
cap systems. More data need to be collected. The collected hydraulic
data can also be used to verify currently used models for simulating the
hydraulic performance of caps.
* Over emphasis on regulatory compliance inhibits innovative and creative
cap design, particularly with regard to the selection of alternative materials
for cap components. Greater emphasis needs to placed on how the design
will affect cap performance. Technical guidelines for demonstrating
alternative cap equivalency need to be developed.
» There have been a number of documented cases of cap failures; however,
most of these failures could have been avoided by proper design and
construction. This problem may be minimized by independent peer
review of design and more rigorous QA/QC during construction.
* Nontraditional materials, such as asphalt, paper mill sludge, capillary
drains, capillary barriers, plants with high tranpiration capability, and
GCLs have undergone limited testing in the field but look promising for
future use in caps.
6.7 REFERENCES
Anderson, D.G. and Kavazanjian, Jr., E. (1995). "Performance of Landfills
Under Seismic Loading." Proceedings of the Third International Conference
on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics,
University of Missouri, Rolla, MO.
Boschuk, Jr., J. (1991). "Landfill Covers An Engineering Perspective."
Geotechnical Fabrics Report. 9(4), pp. 23-34.
-------
6.3 REFERENCES 139
Daniel, D.E, and Koerner, R.M. (1993). "Cover Systems." Geotechnical Practice
for Waste Disposal. D.E. Daniel, ed., Chapman & Hall, pp 455-496.
Dwyer, S.F. (1995). "Alternative Landfill Cover Demonstration." Landfill
Closures Environmental Protection and Land Recovery, Geotechnical Special
Publication No. 53, R.J. Dunn and U.P. Singh, Eds., ASCE, pp. 19 - 34.
Payer, M. and Jones. T. (1990). "Unsaturated Soil-Water and Heat Flow Model,
Ver. 2.0." Pacific Northwest Laboratory, Richland, Washington.
Freeman, H.D., Romine, R.A., Zacher, A.H. (1994). "Hanford Permanent
Isolation Barrier Program: Asphalt Technology Data and Status Report -
FY1994." PNL-10194, Pacific Northwest Laboratory, Richland, Washington,
37 p.
Gee G.W., Freeman, H.D., Walters, Jr., W.H., Ligotkr, M.W., Campbell, M.D.,
Ward, A.L., Link, S.O., Smith, S.K., Gilmore, E.G., and Romine, R.A,(1994).
"Hanford Prototype Surface Barrier Status Report: FY 1994." PNL-10275,
Pacific Northwest Laboratory, Richland, Washington.
Khire, M.V. (1995) "Field Hydrology and Water Balance Modeling of Earthen
Final Covers for Waste Containment," Environmental Geotechnics Report
No. 95-5, University of Wisconsin - Madison, 166 p.
Koerner, R. M., Lord, A. E., and Hsuan, Y. H. (1992). "Arrhenius Modeling to
Predict Geosynthetic Degradation." journal of Geotextiles and Geomem-
branes. Vol. 11, pp. 151 -183.
Limbach, WE., Ratzlaff, T.D., Anderson, J.E., Reynolds, T.D., and Laundr, J.W.
(1994) "Design and Implementation of the Protective Cap/Biobarrier
Experiment at the Idaho National Engineering Laboratory," In-Situ
Remediation: Scientific Basis for Current and Future Technologies. G.W.
Gee and N.R. Wing, eds., Battelle Press, pp. 359-377.
Melchior, S. and Miehlich, G, (1989). "Field Studies on the Hydrological
Performance of Multilayered Landfill Caps." Proceedings of the Third
International Conference on New Frontiers for Hazardous Waste Management,
EPA/600/9-89/072, U.S. EPA, pp. 100-107.
Melchior, S., Berger, K., Vielhaber, B., and Miehlich, G. (1994). "Multilayered
Landfill Covers: Field Data on the Water Balance and Liner Performance."
In-Situ Remediation: Scientific Basis for Current and Future Technologies.
G.W. Gee and N.R. Wing, eds., Battelle Press, pp. 411-425.
Moo-Young, H.K. and Zimmie, T.F. (1995). "Design of Landfill Covers Using
-------
140 CAPS
Paper Mill Sludges." Proceedings of Research Transformed into Practice
Implementation of NSF Research, J. Colville and A.M. Amde, eds., ASCE,
NY. pp. 16-28.
Othman, M.A., Bonaparte, R., Gross, B.A., and Schmertmenn, G.R. (1995).
"Design of MSW Landfill Final Cover Systems." Landfill Closures
Environmental Protection and Land Recovery, Geotechnical Special Publication
No. 53, R.J. Dunn and U.P. Singh, Eds., ASCE, NY, pp. 218-257.
Petersen, K.L., Link, S.O., and Gee, G.W., (1995). "Hanford Site Long-Term
Surface Barrier Development Program: Fiscal Year 1994 Highlights." PNL-
10605, Pacific Northwest Laboratory, Richland, Washington.
Rumer, R. R. and Ryan, M. E., eds. (1995). Barrier Containment Technologies
for Environmental Remediation Applications. John Wiley & Sons, Inc.
NY, 170 p.
Schroeder, PR., Lloyd, C.M., and Zappi, P.A. (1994a). "The Hydrologic
Evaluation of Landfill Performance (HELP) Model User's Guide for
Version 3." U.S. Environmental Protection Agency, Office of Research and
Development, Washington, D.C., EPA/600/R-94/168a.
Schroeder. PR., Dozier, T.S., Zappi, P.A., McEnroe, B.M., Sjostrom, J.W., and
Peyton, R.L. (1994b). "The Hydrologic Evaluation of Landfill Performance
(HELP) Model Engineering Documentation for Version 3." U.S.
Environmental Protection Agency Office of Research and Development,
Washington, D.C., EPA/600 R-94/168b, 116 p.
Schultz, R.K., Ridky, R.W. and O'Donnell, E. (1995). "Control of Water
Infiltration into Near Surface LLW Disposal Units Progress Report of Field
Experiments at a Humid Region Site, Beltsville, Maryland." U.S. Nuclear
Regulatory Commission, NUREG/CR4918 Vol. 8, 20 p.
Stormont, J.C. (1995). "The Performance of Two Capillary Barriers During
Constant Infiltration." Landfill Closures Environmental Protection and Land
Recovery, Geotechnical Special Publication No. 53, R.J. Dunn and U.P,
Singh, Eds., ASCE, NY, pp. 77 - 92.
Wing, N.R. and Gee, G.W. (1994). "Quest for the Perfect Cap." Civil
Engineering. ASCE. Oct., pp.38-41.
-------
United States
Environmental Protection
Agency
Office of Solid Waste and
Emergency Response
Washington DC 20460
EPA 530-SW-89-047
July 1989
Technical Guidance
Document:
Final Covers on
Hazardous Waste
Landfills and Surface
Impoundments
REPRODUCED BY
U.S. DEPARTMENT OF COMMERCE
NATIONAL TECHNICAL INFORMATION SERVICE
SPRINGFIELD, VA. 22161
-------
TECHNICAL REPORT DATA
frleesf read Inaruttiorti 'wi lite rertne lit fort camplttfntl
1, REPORT NO,
EPA/530-SW-39-047
4. TITLE AND SUBTITLE
TECHNICAL GUIDANCE DOCUMENT-
3.
FiNfli rnvFfK
WASTE LANDFILLS AND SURFACE IMPOUNDMENTS
ON Hfl7ftRrinil<:;
7. AUTHORIS)
i.S. Environmental Protection Agency
Office of Solid Waste & Risk
Reduction Engineering Lab
9, PERFORMING ORGANIZATION NAME AND ADDRESS
401 M Street, SW
Washington, DC 20460
26 W. Martin Luther King Drive
Cincinnati, OH 45268
12. SPONSORING AGENCY NAME AND ADDRESS
EPA Office of Solid Waste
Washington, DC 20460 and
Risk Reduction Engineering Laboratory
Cincinnati, OH 45268
16. SUPPLEMENTARY NOTES
Project Officer: Robert E. Landreth
l«. ABSTRACT
This document recommends and
describes a
requirements of RCRA regulations. It is
down, of:
0 a top layer of at least 60 cm of soil,
FTS: 684-7836
design for landfi
3, RECIPIENT'S ACCESSION NO,
W8 0 — p S
S, REPORT DATE
July 1989
S 4 S 0 /AS
*, PERFORMING ORGANIZATION CODE
• .PERFORMING ORGANIZATION REPORT
NO.
10. PROGRAM ELEMENT NO.
H. CONTRACT/GRANT NO.
13. TYPE Of REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/14
COMM: 513/569-7836
11 covers that
will meet the
a multi layered system consisting, from the top
either vegetated or armored at
the surface;
0 a granular or geosynthetic drainage layer with a hydraulic transmissivity no less
than 3 x 10"5 cm /sec; and
0 a two-component lo* permeability layer
installed directly on (2)
no greater than 1 x 10~?
Optional layers may be added
on the need.
17.
a compacted
cm/ sec.
comprised of (11
soil
a flexible membrane liner
component with an hydraulic conductivity
, e.g., a biotic barrier layer or a gas vent layer, depending
KtV WORDS AND DOCUMENT ANALYSIS
1. . DESCRIPTORS
It. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
b.lOENTIFIERS/OREN ENDED TERMS
IB. SECURITY CLASS (THil Report)
UNCLASSIFIED
2O. SECURITY CLASS fTIHiptgel
UNCLASSIFIED
c. COSATI Field/Group
•
21. NO. OFAAGES
22. PRICE
EPA F*n* 3720-1 (R.». 4_77)
PREVIOUS COITION IS O*SOLCtC .
-------
EPA/530-SW-89-047
July 1989
TECHNICAL GUIDANCE DOCUMENT
FINAL COVERS ON HAZARDOUS WASTE LANDFILLS
AND SURFACE IMPOUNDMENTS
Office of Solid Waste and Emergency Response
U.S. Environmental Protection Agency
Washington, DC 20460
In cooperation with
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
The preparation of this document has been funded wholly by
the United States Environmental Protection Agency. It has been
subjected to the Agency's peer and admininistrative review, and
it has been approved for publication as an EPA document. Mention
of trade names or commercial products does not constitute
endorsement or recommendation for use.
11
-------
FOREWORD
Today's rapidly developing and changing technologies and
industrial products and practices frequently carry with them the
increased generation of solid and hazardous wastes. These
materials, if improperly dealt with, can threaten both public
health and the environment. Abandoned waste sites and accidental
releases of toxic and hazardous substances to the environment
also have important environmental and public health implications.
The Risk Reduction Engineering Laboratory assists in providing an
authoritative and defensible engineering basis for assessing and
solving these problems. Its products support the policies,
programs, and regulations of the U.S. Environmental Protection
Agency; the permitting and other responsibilities of State and
local governments; and the needs of both large and small
businesses in handling their wastes responsibility and
economically.
This document provides design guidance on final cover
systems for hazardous waste landfills and surface impoundments.
We believe that the final cover, if properly designed and
constructed, can provide long-term protection of the unit from
moisture infiltration due to precipitation. The cover system
presented herein is a multilayer design consisting of a vegetated
top layer, drainage layer, and low-permeability layer. Optional
layers which may be required for site-specific conditions are
also discussed. Rationale is provided for the design parameters
to give designers and permit writers background information and
an understanding of cover systems.
This document is intended for use by organizations involved
in permitting, designing, and constructing hazardous waste land
disposal facilities.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
111
-------
PREFACE
Subtitle C of the Resource Conservation and Recovery Act
(RCRA) requires the U.S. Environmental Protection Agency (EPA) to
establish a Federal hazardous waste management program. This
program must ensure that hazardous wastes are handled safely from
generation until final disposition. EPA issued a series of
hazardous waste regulations under Subtitle C of RCRA that are
published in Title 40 Code of Federal Regulations (40 CFR). The
principal 40 CFR Part 264 and 265 regulations were issued on
July 26, 1982 for treatment, storage, and disposal (TSD)
facilities and establish performance standards for hazardous
waste landfills, surface impoundments, land treatment units, and
waste piles. The regulations have been amended several times
since then.
In support of the regulations, EPA has been developing three
types of documents to assist preparers and reviewers of RCRA
permit applications for hazardous waste TSD facilities. These
include RCRA Technical Guidance Documents, Permit Guidance
Manuals, and Technical Resource Documents (TRDs).
RCRA Technical Guidance Documents, such as this one, present
design and operating parameters or design evaluation techniques
that generally comply with, or demonstrate compliance with, the
Design and Operating Requirements and the Closure and Post-
Closure Requirements of 40 CFR Part 264.
The Permit Guidance Manuals are being developed to describe
the permit application information the Agency seeks, and to
provide guidance to applicants and permit writers in addressing
information requirements. These manuals will include a
discussion of each set of specifications that must be considered
for inclusion in the permit.
The Technical Resource Documents present summaries of state-
of-the-art technologies and evaluation techniques determined by
the Agency to constitute good engineering designs, practices, and
procedures. They support the RCRA Technical Guidance Documents
and Permit Guidance Manuals in certain areas (i.e., liners,
leachate management, final covers, and water balance) by
describing current technologies and methods for designing
hazardous waste facilities, or for evaluating the performance of
a facility design. Although emphasis is given to hazardous waste
facilities, the information presented in these TRDs may be used
iv
-------
for designing and operating nonhazardous waste TSD facilities as
well. Whereas the RCRA Technical Guidance Documents and Permit
Guidance Manuals are directly related to the regulations, the
information in these TRDs covers a broader perspective and should
not be used to interpret the requirements of the regulations.
This document is a Technical Guidance Document prepared by
the Risk Reduction Engineering Laboratory of EPA's Office of
Research and Development in cooperation with the Office of Solid
Waste and Emergency Response. The document has undergone
extensive technical review and has been revised accordingly.
With the issuance of this document, all previous drafts are
obsolete and should be discarded.
Comments are welcome at any time on the accuracy and
usefulness of the information in this document. Comments will be
evaluated, and suggestions will be incorporated, wherever
feasible, before publication of any future revisions. Written
comments should be addressed to EPA RCRA Docket (OS-305), 401 M
Street S.W., Washington, DC 20460. The document for which
comments are being provided should be identified by title and
number.
-------
ABSTRACT
The owner or operator of a landfill, or a surface
impoundment closed as a landfill, must meet the closure
requirements specified under 40 CFR 264.310 (permitted units) or
40 CFR 265.310 (interim status units).
This guidance document addresses landfill covers and
recommends a multilayer final cover design that includes the
following elements, from top to bottom:
o a top layer consisting of two components: (1) a vegetated
or armored surface component, either of which is selected
to minimize erosion and, to the extent possible, promote
drainage off the cover, and (2) a soil component with a
minimum thickness of 60 cm [24 in.] comprised of topsoil
and/or fill soil as appropriate, the surface of which
slopes uniformly at least 3 percent but not more than 5
percent;
o a soil drainage layer with a minimum thickness of 30 cm
(12 in.) and a minimum hydraulic conductivity of 1 x 10"2
cm/sec that will effectively minimize water infiltration
into the low-permeability layer, and a final bottom slope
of at least 3 percent after settlement and subsidence; or
the drainage layer may consist of geosynthetic materials
with equivalent performance characteristics; the drainage
layer also serves as a protective cover for the flexible
membrane liner (FML) component of the underlying low-
permeability layer;
o a two-component low-permeability layer, that limits water
infiltration into the underlying wastes to a rate less
than or equal to the rate of leachate migration out of the
bottom liner system and consists of (1) a 20-mil minimum
thickness [or greater depending on the material and
design] FML component and (2) a 60-cm [24-inch] minimum
thickness compacted soil component with an in-place
saturated hydraulic conductivity no less than 1 x 10"7
cm/sec. (NOTE: The requirement for FMLs in the cover are
for all permitted units and interim status units with an
FML in the bottom. For interim status units with only a
clay bottom liner, an FML may not be required.)
VI
-------
Optional layers that may be used on a site-specific basis
include (1) a gas vent layer to remove gases produced within the
wastes, and/or (2) a biotic barrier layer to protect the cover
from animal or plant intrusion.
The Agency recommends a detailed construction quality
assurance (CQA) program for each layer of the final cover system.
CQA records should document quality and demonstrate compliance
with plans and specifications. The cover design process must
consider many site-specific factors, such as precipitation,
construction materials, freeze-thaw phenomena, waste
characteristics, potential subsidence, and other environmental
factors.
VII
-------
CONTENTS
Disclaimer ii
Foreword iii
Preface iv
Abstract vi
Figures x
Tables xi
Acknowledgment xii
1. Introduction 1
1.1 Purpose 1
1.2 Closure and Post-Closure Regulations 1
1.3 Liquids Management Strategy 4
1.4 General Cover System Recommendations 5
1.4.1 Design Recommendations 5
1.4.2 Construction Quality Assurance .8
1.4.3 Settlement and Subsidence 8
2. Top Layer 11
2.1 Design 11
2.2 Discussion 12
2.2.1 Upper Component of Top Layer 12
2.2.1.1 Vegetation . .- 13
2.2.1.2 Other Erosion-Impeding Materials . 13
2.2.2 Lower Component of Top Layer 14
3. Drainage Layer 16
3.1 Design 16
3.2 Discussion 18
4. Low-Permeability Layer 22
4.1 Design 23
4.2 Discussion 24
4.2.1 FML Component 25
4.2.2 Low-Permeability Compacted Soil Component . 27
Vlll
-------
CONTENTS (continued)
5. Optional Layers 31
5.1 Gas Vent Layer . . 31
5.1.1 Design 31
5.1.2 Discussion 32
5.2 Biotic Barrier Layer 33
5.2.1 Design 33
5.2.2 Discussion 34
References * 36
-------
FIGURES
Number Page
1. EPA-recommended cover design 6
2. EPA-recommended cover design with optional layers ... 7
3a. Cover with sand drainage layer 17
3b. Cover with geogrid drainage layer 17
4. Cover and liner edge configuration with example ....
toe drain 19
5. Detail of FML/soil composite low-permeability layer . . 22
6. Regional depth of frost penetration 30
7. Cover with gas vent and vent layer 31
-------
TABLES
Number Page
1. Closure and Post-Closure Regulatory Requirements .... 3
2. Synopsis of Minimum Technology Guidance for Covers ... 9
XI
-------
ACKNOWLEDGMENT
The EPA Project Manager who directed the draft preparation
of this document was Les Otte with the assistance of Ana Aviles,
both of the Environmental Protection Agency's Office of Solid
Waste, Land Disposal Branch. Early drafts of the document were
prepared by David C. Anderson of K. W. Brown & Associates, Inc.
Later drafts were prepared by Jeffrey D. Magaw, Charles W. Young,
and Clay Spears of Alliance Technologies Corporation. This draft
has been prepared by Robert P. Hartley of EPA's Risk Reduction
Engineering Laboratory, after peer reviews by Dr. Gordon Boutwell
of Soil Testing Engineers, Inc.j Dr. Richard C. Warner,
University of Kentucky? Leo Overman of Colder Associates, Inc.i
Dirk Brunner of E. C. Jordan, Inc.; the Solid and Hazardous Waste
Management Committee of the American Society of Civil Engineers,
Environmental Engineering Division; and by various members of the
EPA's Risk Reduction Engineering Laboratory.
Xll
-------
1. INTRODUCTION
1.1 PURPOSE
This document provides design guidance on final covers for
hazardous waste units. The recommended design satisfies the
requirements of 40 CFR 264 and 265 Subparts G (closure and
post-closure), K (surface impoundments), and N (landfills). The
Environmental Protection Agency (hereafter referred to as "the
Agency") emphasizes that recommendations are guidance only and
not regulations. The Agency acknowledges that other final cover
designs may be acceptable, depending upon site-specific
conditions and upon a determination by the Agency that an
alternative design adequately fulfills the regulatory
requirements. It is, however, the responsibility of the facility
owner or operator to prove that the alternate design will provide
a level of performance that is at least equivalent to that of the
final cover system described in this document.
The Agency's liquids management strategy for landfills, and
the role that final covers serve in that strategy, are outlined
in general terms for background. Regulatory requirements for
landfill and surface impoundment covers are also outlined, as
well as differences in requirements between interim status and
permitted units. The Agency-recommended final cover system
design is presented in detail, as well as considerations for
construction quality assurance. Attention is given to erosion,
settlement, and subsidence, and their potential cover-damaging
effects.
A separate section of this document is devoted to the design
details of each layer of the recommended cover. A discussion of
the rationale for the recommended specifications is included.
1.2 CLOSURE AND POST-CLOSURE REGULATIONS
All of the regulations dealing with hazardous waste landfill
and surface impoundment cover requirements are found in/Title 40,
Parts 264 and 265, of the Code of Federal Regulations-"t40 CFR 264
and 40 CFR 265) . Part 264 deals with permitted fa»cilities and
Part 265 with interim status facilities. Interim status
facilities are, in general, those facilities that were in
existence on November 19, 1980. Three Subparts of each of Parts
264 and 265 deal with general closure requirements: Subpart G -
Closure and Post-Closure; Subpart K - Surface Impoundments; and
-------
Subpart N - Landfills. Each Subpart contains several sections
important to cover planning, design, and construction, as
outlined in Table 1.
There are few difference between permitted and interim
status unit closure and post-closure regulations under Subpart G
of Parts 264 and 265. The major difference is that, for interim
status units, public notice for changes to the approved closure
and post-closure plans is not required. Changes to plans for
permitted units require permit modifications which, in turn,
require public notice and comment.
There are three significant differences between permitted
and interim status unit final cover regulations under Subparts K
and N of Parts 264 and 265. Part 264.303 requires monitoring and
inspection to ensure that synthetic and soil materials used in
the cover are watertight and structurally uniform. Such a
requirement was not included in Part 265 for interim status
units. The Agency recommends that a Construction Quality
Assurance (CQA) program, establishing inspection activities, be
utilized for covers being built at both permitted and interim
status units. The Agency believes that a site-specific CQA
inspection program is necessary to ensure that cover design
specifications are met.
A second difference in requirements is that, while leachate
collection and removal activities are required after closure
under 40 CFR 264.310, for permitted units, they are not required
under Part 265 for interim status units. The absence of a stated
post-closure leachate collection and removal requirement makes
cover performance for interim status units even more important.
It should be noted that, under the broader performance standards
of 40 CFR 265.111, the Agency may still require leachate
collection during post-closure at an interim site.
The third, and perhaps most significant, difference is in
the requirements of 40 CFR 264.310(b)(1)(v) and 40 CFR 265.310
(b)(1)(v). These subsections require that the cover have a
permeability less than or equal to any bottom liner or natural
subsoil present. For interim status units, without an engineered
liner, the cover could presumably be of relatively permeable
materials. But here again, the Agency may impose the standards
of 40 CFR 265.111, and require a more impermeable cover.
For permitted landfills, to meet the requirements of 40 CFR
264.310, the cover must have a permeability no greater than that
of the double liner required under 40 CFR 264.301(c). The Agency
does not consider this to mean that the final cover for a
permitted unit must actually contain a double liner. Rather, the
Agency recommends that the final cover include a layer whose
liquid-rejection performance is equal to or better than the
-------
Table 1. Closure and Post-Closure Regulatory Requirements
Section
Part 264
Part 265
Subpart G - Closure and Post-Closure
111
112
113
115
116
117
118
120
Closure performance
standard.
Closure plan; amendment
of plan.
Time allowed for closure.
Certification of Closure.
Survey plat.
Post-closure care.
Post-closure plan;
amendment of plan.
Certificate of completion
of postclosure care.
Closure performance
standard.
Closure plan; amendment
of plan.
Time allowed for closure.
Certification of closure.
Survey plat.
Post-closure care.
Post-closure plan;
amendment of plan.
Certificate of completion
of post-closure care.
226
228
Subpart K - Surface Impoundments
Monitoring and inspection. Inspections.
Closure and post-closure
care.
Closure and post-
closure care.
301
302
303
310
Subpart N - Landfills
Design and operating
requirements.
N/A
Closure and post-closure
care.
Design requirements
General operating
requirements.
Monitoring and inspection. N/A
Closure and post-
closure care.
-------
bottom composite liner (flexible membrane liner [FML] underlain,
and in full contact with, compacted soil) of the double-liner
system detailed in the "Minimum Technology Guidance on Double
Liner Systems for Landfills and Surface Impoundments - Design,
Construction and Operation" (EPA, 1987i). The Agency-recommended
design for the cover does, in fact, include a composite barrier
layer as outlined in Section 4. In all cases where a FML is used
in the bottom liner, one should also be used in the cover. This
does not mean, however, that the Agency necessarily recommends
the use of exactly the same barrier materials in both the liner
and cover. For example, different FML materials of equivalent
performance may be used, such as high density polyethylene for
the bottom liner and polyvinyl chloride in the cover.
The Agency also recommends using the composite FML/clay
barrier in interim status unit covers. However, for interim
status units, compacted clay with a permeability equal to or less
than 1 x 10'7 cm/sec may be used without a FML if the clay is
less permeable than the landfill bottom liner or natural subsoil
beneath the site. While 40 CFR 265.310(a)(5) might allow a less
effective design, we believe the long-term protection from
infiltration provided by the recommended cover design justifies
its use for all units. With the Agency-recommended composite
design, it is more certain that the cover will be no more
permeable than the bottom of the unit. In addition, the
installation of the composite design on interim status units
takes advantage of the practical opportunity to more effectively
minimize water infiltration, leachate generation, and leachate
migration.
1.3 LIQUIDS MANAGEMENT STRATEGY
The general closure performance standards are specified in
40 CFR 264.111 and 265.111 (Subpart G) for permitted and interim
hazardous waste disposal facilities, respectively. The standards
state that:
"The owner or operator must close the facility in a manner
that: :
a. Minimizes the need for further maintenance; and
b. Controls, minimizes, or eliminates, to the extent
necessary to protect human health and the
environment, post-closure escape of hazardous
waste, hazardous constituents, leachate,
contaminated runoff, or hazardous waste
decomposition products to the ground or surface
waters or to the atmosphere . . "
The requirements apply to hazardous waste landfills and to
hazardous waste surface impoundments closed as landfills.
-------
Landfill closure requirements are based on a two-part
liquids management strategy of (1) minimizing the leachate
generation by keeping liquids out of the unit, and (2) detecting,
collecting, and removing leachate within the unit. Closure
requirements are specified in 40 CFR 264.310 and 40 CFR 265.310
and include a final cover and post-closure care.
The Agency considers keeping water out of the unit to be the
prime element of the strategy. Thus, the Agency believes that a
properly designed and constructed cover becomes, after closure,
the most important feature of the landfill structure. The Agency
requires that the cover be designed and constructed to provide
long-term minimization of the movement of water from the surface
into the closed unit. Where the waste mass lies entirely above
the zone of ground-water saturation, a properly designed and
maintained cover can prevent, for all practical purposes, the
entry of water into the closed unit, and thus minimize the
formation and migration of leachate. In the absence of damage,
the cover design recommended here, including the FML/soil low-
permeability layer, should" restrict infiltration, to the extent
of the design, for the long term.
1.4 GENERAL COVER SYSTEM RECOMMENDATIONS
The cover system should be a major consideration during site
selection, planning, and initial design of the landfill
containment structure. Factors for consideration include
location and availability of low-permeability soil, stockpiling
of topsoil, restricting height to provide stable slopes, and site
use beyond the post-closure care period.
1.4.1 Design Recommendations
The final cover recommended in this guidance document is a
multilayer design (Figure 1) comprised as follows, from top to
bottom:
o a top layer consisting of two components: (1) either
a vegetated or armored surface component, selected to
minimize erosion and, to the extent possible, promote
drainage off the cover, and (2) a soil component with
a minimum thickness of 60 cm [24 in.], comprised of
topsoil and/or fill soil as appropriate, the surface
of which slopes uniformly at least 3 percent but not
more than 5 percent; a soil component of greater
thickness may be required to assure that the
underlying low-permeability layer is below the frost
zone;
o either a soil drainage (and FML-protective bedding)
layer with minimum thickness of 30 cm (122 in.) and a
minimum hydraulic conductivity of 1 x 10' cm/sec that
-------
will effectively minimize water infiltration into the
low-permeability layer, and will have a final slope of
at least 3 percent after settlement and subsidence; or
a drainage layer consisting of geosynthetic materials
with equivalent performance characteristics; and
a two-component low-permeability layer, lying wholly
below the frost zone, that provides long-term
minimization of water infiltration into the underlying
wastes, consisting of (1) a 20-mil [0.5 mm] minimum
thickness flexible membrane liner [FML] component and
(2) a compacted soil component with a minimum
thickness of at least 60 cm [24 in.] and a maximum in-
place saturated hydraulic conductivity of 1 x 10"7
cm/sec.
vegetation/soil
top layer
drainage layer
low-permeability
FML/soil layer
waste
60cm
30cm
_- =r^^r-=-^-t^^-j go cm
filter layer
20-mil FML
O
0
0
O
0 ~ V 0
O Q3
0
Figure l. EPA-recommended cover design.
Optional layers may be used on a site-specific basis.
Figure 2 depicts a cover design that includes optional layers.
Two such layers include (1) a gas vent layer to remove gases that
are produced within the wastes, and/or (2) a biotic barrier layer
to protect the cover from animal or plant intrusion.
Geosynthetic filter materials may also be used to prevent
migration of fine materials from one layer into another or to
prevent clogging of the drainage layer.
The Agency recognizes, for specific cases, that alternative
designs (e.g., fewer layers or optional layers) may be
applicable. For instance, in extremely arid regions, a gravel-
armored top surface component might serve to compensate for a
naturally reduced vegetation coverage and the erosion control
that it provides. Also, in arid regions the drainage layer might
not be required. In areas where burrowing animals may damage the
-------
low-permeability layer, the damage may be prevented by use of an
overlying "biotic barrier" layer of large-size material, such as
cobbles. A gas vent layer between the waste and the low-
permeability layer may be installed, as shown in Figure 2, at
units that produce gases.
Alternative designs must provide long-term performance at
least equivalent to the recommended design outlined in this
guidance. All alternative designs must be approved by the
appropriate Regional Administrator of the Agency.
cobbles/soil -
top layer
biotic barrier
(cobbles)
drainage layer
low-permeability.
FML soil layer
gas vent layer
waste
60cm
geosynthetic filter
geosynthetic filter
20-mil FML
geosynthetic filter
Figure 2. EPA-recommended cover design
with optional layers.
In some cases, where the waste is of such character that
vertical migration of gases is impeded, full-depth vent
structures to the bottom of the waste mass may be needed. These
structures would be designed to prevent the horizontal migration
of gases out of the landfill into the surrounding soil. Active
rather than passive systems may be required in some cases to
adequately remove accumulated gases.
Filter layers are likely to be needed above the drainage
layer and between layers that are comprised of soils of greatly
different particle sizes, to prevent one from migrating into the
other. The filters may 'be constructed of soils of intermediate
grain size, or they may be geosynthetic materials. Three
between-layer locations where geosynthetic filters may be
appropriate are shown in Figure 2.
Table 2 presents a synopsis of the Agency-recommended
components of a landfill and their principal design parameters.
-------
1.4.2 ConstructionQuality Assurance fCQA)
The Agency believes that the landfill owner or operator
should implement a detailed construction quality assurance (CQA)
program for the final cover system based on written plans for
inspecting the quality of construction materials and the
construction practices employed in their placement. The Agency
believes that use of a CQA program is essential for determining,
with a reasonable degree of certainty, whether a completed final
cover system meets or exceeds all design criteria, plans, and
specifications. The Agency has issued technical guidance that
includes final cover CQA (EPA, 1987i).
The Agency has proposed CQA rules for both permitted and
interim status units (EPA, 1987b). These proposed rules would
require a CQA program for installing the following components of
landfills, surface impoundments and waste piles: foundations;
low-permeability soils; FMLs; dikes; leachate detection,
collection, and removal systems; and final covers. The CQA plan
would be site-specific. It should address activities such as
inspecting, monitoring, and sampling of the individual
components. For the cover, the CQA plan should provide assurance
that: l) all layers of the final cover are uniform and damage-
free? 2) the materials for each layer are as specified in the
design specifications; and 3) each layer is constructed as
specified in the design.
1.4.3 Settlement and Subsidence
Settlement within a closed hazardous waste landfill can
disrupt the integrity and function of the final cover system.
Settlement of the waste may be uniformly distributed and may
occur primarily before placement of the final cover. Subsidence,
however, is considered to be an unevenly distributed settlement
(i.e., differential settlement) after closure that can disrupt
the integrity of the final cover by creating depressions and
cracks. In addition, subsidence due to the collapse of drums
(this will occur mainly in older units), the leaching of soluble
waste constituents, or biodegradation of organic matter in the
waste, may not begin until several years after closure or it may
occur gradually over decades.
To reduce the potential for damage from settlement and
subsidence, the final cover should be designed and constructed to
allow for the total estimated settlement. The final grade after
subsidence of the'cover should be at the actual desired design
elevation. The cover design process used to establish the final
grade elevation should include consideration of the following:
o consolidation of all waste layers (the primary
consideration) and daily and intermediate soil covers;
-------
Table 2. Synopsis of Minimum Technology Guidance for Covers
Layer
Thickness
Slope
Requirements
Top Layer
Vegetation —
OR
Surface Armor 5-10 in.
(13-25 cm)
ON
Soil
Drainage Layer
Soil
OR
> 24 .in.
(> 60 cm)
j> 12 in.
(> 30 cm)
Geosynthetic variable
Low-Permeability Layer
FML
ON
> 20 mils
(> 0.5 mm)
Low-Perme-
ability Soil
> 24 in.
(> 60 cm)
3-5%
> 3%
> 3%
> 3%
> 3%
Optional Layers (site-specific design)
Gas Vent Layer > 12 in. > 2%
(> 30 cm)
Biotic Barrier animal or —
root-dependent
Persistent,
drought-resistant,
adapted to local
conditions.
Cobbles, gravel.
Erosion rate
<2 ton/acre/yr
(5.5 MT/ha/yr).
SP (USCS) soil
K > 1 x 10"2 cm/s;
gravel toe drain.
Performance equi-
valent to soil,
hydraulic transmis-
sivity > 3 x 10
m/sec.
-5
In EPA Report No,
EPA 600/2-88-052.
In-place
K < 1 x 10"' cm/s
and test fill.
,-7
Similar to
drainage layer.
Large materials,
e.g., cobbles.
-------
o consolidation of soils and foundation materials
underlying the site;
o consolidation of liner and leachate collection
systems; and
o consolidation of all final cover components.
The Agency has published two technical research reports on
cover settlement and subsidence (EPA, 1985c and 1987d) that
address both the theoretical and practical aspects.
Interim covers have been proposed when a significant amount
of settlement and subsidence is expected in a fairly short time
(say 2-5 years) that could result in the premature failure of a
final cover. An interim cover could be maintained until settling
is judged to be virtually complete. After settlement occurs, the
interim cover could be removed and replaced or overlain by a new
final cover. If components of the interim cover can meet the CQA
requirements for the final cover, the interim cover could be made
an integral part of the final design.
In no case can an interim cover be used that does not
satisfy the performance standards of 40 CFR 264.111 to protect
human health and the environment. Use of an interim cover on a
permitted unit will generally result in a longer closure period
during which the stipulations of 40 CFR 264.113 must be met,
i.e., the applicant must take all necessary steps "to prevent
threats to human health and the environment from the unclosed but
not operating hazardous waste management unit or facility,
including compliance with all applicable permit requirements."
10
-------
2. TOP LAYER
The Agency recommends a two-component top layer for a
landfill cover system (Figure 1). The upper component should be
vegetation or other surface treatment, designed to impede erosion
but allowing surface runoff from major storm events. The Agency
believes that, in most cases, vegetation underlain by soil, at
least part of which is topsoil, will best accomplish these
objectives. However, in some areas the prevailing climate may
inhibit the establishment and maintenance of vegetation, or a
planned alternative use of the site may preclude vegetation. In
those cases, an armored surface without vegetation (Figure 2),
and underlain by fill soil, might be used if it will minimize
erosion and abrasion of the cover and allow, to the maximum
practicable extent, surface drainage off the cover.
2.1 DESIGN
The Agency recommends that the vegetation component of the
top layer meet the following specifications:
o Locally adapted perennial plants.
o Resistant to drought and temperature extremes.
o Roots that will not disrupt the low-permeability
layer.
o Capable of thriving in low-nutrient soil with minimum
nutrient addition.
o Sufficient plant density to minimize cover soil
erosion to no more than 2 tons/acre/year (5.5
MT/ha/yr), calculated using the USDA Universal Soil
Loss Equation.
o Capable of surviving and functioning with little or no
maintenance.
In landfill situations where the environment or other
considerations make it inappropriate for maintaining sufficiently
dense vegetation, armoring material may be substituted as the
upper component of the top layer or in rare cases the whole
layer. It is recommended that the material possess the following
characteristics:
11
-------
o capable of remaining in place and minimizing erosion
of itself and the underlying soil component during
extreme weather events of rainfall and/or wind;
o capable of accommodating settlement of the underlying
material without compromising the purpose of the
component;
o surface slope approximately the same as the underlying
soil (at least 3 percent slope); and
o capable of controlling the rate of soil erosion from
the cover to no more than 2 tons/acre/year (5.5
MT/ha/yr), calculated by using the USDA Universal Soil
Loss Eguation.
Agency-recommended specifications for the lower soil
component of the top layer include the following:
o for vegetation support, a minimum thickness of 60 cm
(24 in.) including at least 15 cm (6 in.) of topsoil
(soil of lower quality may be used beneath an armored
surface); greater total thickness where required,
e.g., where maximum frost penetration exceeds this
depth, or where greater plant-available water storage
is necessary or desirable;
o medium texture to facilitate seed germination and
plant root development;
o final top slope, after allowance for settling and
subsidence, of at least 3 percent, but no greater than
5 percent, to facilitate runoff while minimizing
erosion; and
o minimum compaction to facilitate root development and
sufficient infiltration to maintain growth through
drier periods.
The owner or operator of the landfill should prepare a
separate section specific to monitoring construction of the top
layer to be included in the construction quality assurance (CQA)
plan.
2.2 DISCUSSION
2.2.1 Upper Component of Top Layer
As noted in the design recommendations above, the upper
component of the top layer may be vegetation (Agency-preferred
where possible) or other erosion-impeding materials. These are
discussed separately below.
12
-------
2.2.1.1 Vegetation
Plant species is an important consideration in the
establishment of vegetation when it is selected as the upper
component of the top layer. The use of shrubs and trees is
usually inappropriate because the root systems extend to a depth
that would normally invade the drainage layer or the low-
permeability layer. A large number of suitable plant species
such as grasses and low-growing plants are available for various
climates (EPA, 1983c and 1987c). The timing of seeding is also
very important to successful vegetation establishment.
The Agency advises landfill owners or operators to contact a
consulting agronomist, Cooperative Extension Service agent, or
local university for recommendations of adapted plant varieties
and other guidance on local crop cultivation. Several references
provide lists of available vegetation and discussions on site-
selection criteria (EPA, 1976, 1979, 1983a, 1983c, 1985a, and
1987c; Lee, et al., 1984; Thornburg, 1979; and Wright, 1976).
These references provide essential information about plant
species, seeding rate, time of seeding, and areas of adaptation.
2.2.1.2 Other Erosion-Impeding Materials
In areas where vegetation is inappropriate or difficult to
establish and maintain, other materials may be selected as the
upper component of the top layer. The materials should be
selected to prevent erosion of the cover and yet allow, as much
as practicable, for surface drainage. Several materials have
been suggested for use in lieu of vegetation, including broken
rock or cobbles that may prevent deterioration of the cap due to
wind, heavy rain, or temperature extremes (EPA, 1982b and 1985a;
Nyhan, et al., 1985; Pertusa, 1980). An example of such an upper
component is a layer several (perhaps eight or more) inches thick
comprised of 5- to 10-cm (2- to 4-in.) cobbles of hard durable
rock. The cobbles allow infiltration of rain water but retard
erosion due to water and wind action (see Figure 2). Asphalt or
concrete might be used if promoting runoff is a prime objective,
but they are likely to deteriorate, for example, by cracking due
to thermal effects and subsidence deformation (EPA, 1979 and
1987a) thus causing concern for their long-term performance.
Substantial maintenance could be expected for these materials.
Asphalt can be very permeable unless special attention is given
to eliminating the air voids during mixing and application (Repa,
et al., 1987).
A surface armor component of very coarse materials promotes
infiltration rather than runoff. Thus, it may be more applicable
in arid areas. In those areas, leachate generation due to water
infiltration may not be a major concern, but it can happen during
infrequent short-duration storms of great intensity (EPA, 1987c).
13
-------
2.2.2 Lower Component of Top Layer
When the upper component of the top layer is vegetation, the
EPA recommends that the associated lower component be composed of
at least 60 cm (24 in.) of soil. The soil should be capable of
indefinitely sustaining plant species that will minimize erosion.
The minimum thickness of the soil component is based upon the
Agency's judgment that:
o it accommodates the root systems of most non-woody
plant species (EPA, 1983c);
o for most locales, it provides adequate water-holding
capacity to attenuate rainfall infiltration to the
drainage layer and to sustain vegetation through dry
periods; and
o it provides sufficient soil thickness to allow for
expected long-term erosional losses.
A layer thicker than 60 cm (24 in.) may be required to
prevent freezing and thawing from damaging the low-permeability
layer, or to increase plant-available water storage capacity in
drier climates.
Medium-textured soils such as loam soils, have the best
overall characteristics for seed germination and plant root
system development. Fine-textured soils, such as clays, are
often fertile but may be beset by management problems such as
puddling of water on the surface or difficulty in initial
establishment of plant cover during wet periods. Sandy soils are
often a problem due to low water retention and loss of nutrients
by leaching. It may be cost-effective to stockpile the topsoil
initially removed from a site for later use during cover
construction. Where only a minimum amount of native topsoil can
be saved by stockpiling, the remainder needed to provide at least
the minimum thickness of 60 cm (24 in.) may be made up by
selecting local borrow material having appropriate qualities.
The Agency recommends that the lower component of the top
layer (and thus the entire top layer) be slightly convex, or be
low in height above the surrounding terrain and uniformly sloped.
In non-level terrain, diversion structures should be installed to
prevent the run-on of surface water onto the cover. To prevent
ponding of rainwater due to irregularities of the surface of the
lower component, the final slope should be uniform and at least
3 percent, after allowance for settlement and subsidence (EPA,
1982a, p. 42). Slopes greater than 5 percent, however, are
likely to promote erosion unless controls are included in the
design. The design of surface water controls is well-documented
(EPA, 1979 and 1982b). The Agency believes that slopes greater
than 5 percent will increase erosion, decrease slope stability,
14
-------
and, in general, increase the long-term maintenance of the cover
system. Owners and operators using final slopes based on
site-specific conditions should determine that the slopes will
not result in the formation of erosion rills and gullies and will
limit total erosion to less than 2.0 tons/acre/year (5.5
MT/ha/yr). The U.S. Department of Agriculture's Universal Soil
Loss Equation (USLE) is recommended as the tool for use in
evaluating erosion potential (EPA, 1982a). The Agency believes
that a maximum erosion rate of 2.0 tons/acre/year (5.5 MT/ha/yr)
is realistically achievable for a wide range of soils, climates,
and vegetation. The Agency also believes that reliance on this
criterion will minimize gully development and cover maintenance.
15
-------
3. DRAINAGE LAYER
The recommended final cover design includes a drainage layer
for the removal of water which infiltrates through the top layer
(see Figures 3a and 3b). The drainage layer should be designed
to minimize the amount and residence time of water coming into
contact with the low-permeability layer, thereby decreasing the
potential for leachate generation. In other words, the drainage
layer construction materials and configuration should facilitate
the rapid and efficient removal of water to an exit drain.
The drainage layer should be designed, constructed, and
operated to function without clogging. Physical clogging may be
prevented by incorporating a filter layer of soil or geosynthetic
material between the top layer and the drainage layer. The
prevention of biological clogging may range from limiting
vegetation .to shallow-rooted species to the installation of a
biotic barrier (see Figure 2). Any or all of these features may
be included in a single cover design.
In arid locations, the need for, and design of, a drainage
layer should be based on consideration of precipitation event
frequency and intensity, and sorptive capacity of other soil
layers in the cover system. It may be possible to construct a
top layer that will absorb most, if not all, of the precipitation
that infiltrates into that layer, eliminating the need for a
drainage layer.
3.1 DESIGN
If composed of granular material such as sand (Figure 3a),
the Agency recommends that the cover drainage layer meet the
following specifications:
o Minimum thickness of 30 cm (12 in.) and minimum slope
of 3 percent at the bottom of the layer; greater
thickness and/or slope if necessary to provide
sufficient drainage flow as determined by site-
specific hydrologic (e.g., HELP) modeling.
o Hydraulic conductivity of drainage material should be
no less than 1 x 10"2 cm/sec (hydraulic transmissivity
no less than 3 x 10"5 m2/sec) at the time of
installation.
16
-------
Granular material should be no coarser than 3/8 inch
(0.95 cm), and classified as SP; it should be smooth
and rounded and should contain no debris that could
damage the underlying flexible membrane liner (FML),
nor should it contain fines that might lessen
permeability.
A filter layer (granular or geosynthetic) should be
included between the drainage layer and top layer if
necessary to prevent clogging of the drainage layer by
fine particles.
vegetation/soil
top layer
20-mil FML
>w-perm<
FML/soll layer
wast*
low-permeability -\ t£^El-El-EZ-EZ-E
filter layer
drainage layer
vegetation/soil J
top layer (
000
Q G
waste
filter layer —i
20-mll FML —^- r i=f
low-permeablllty
FML/soll layer
0
0 00
o o
-i geosynthetic
- drainage layer
Figure 3a,
Cover with sand
drainage layer.
Figure 3b.
Cover with geosyn-
thetic drain layer.
If composed of geosynthetic materials (Figure 3b), the
Agency recommends that the drainage layer meet the following
specifications:
Same minimum flow capabilility as a granular drainage
layer in the same situation; hydraulic transmissivity
no less than 3 x
overburden for the design life.
10'5 m2/sec under anticipated
Inclusion of a geosynthetic filter layer above the
drainage material to prevent intrusion and clogging by
the overlying top layer soil material.
Inclusion of geosynthetic bedding beneath the drainage
layer, if necessary, to increase friction and minimize
slippage between the drainage layer and the underlying
FML, and to prevent intrusion, by deformation, of the
FML into the net or grid of the drainage layer.
17
-------
The owner or operator should prepare a written construction
quality assurance (CQA) plan to be used during construction and
installation of the drainage layer (see EPA, 1987b).
3.2 DISCUSSION
The primary functions of the drainage layer are to intercept
water that percolates through the top layer and to transport the
water out of the cover (for example, by gravity flow to an outlet
at the toe of the cover). The Agency believes that the criteria
presented above are the minimums required to provide cover
drainage and FML protection. The criteria for permeability and
FML bedding are equivalent to those cited in "Minimum Technology
Guidance on Double Liner Systems for Landfills and Surface
Impoundments — Design, Construction and Operation" (EPA, I987i)
for the leachate detection, collection, and removal system.
The recommended 30-cm (12-in.) minimum thickness of the
drainage layer allows sufficient cross-sectional area for
transport of drainage in most situations and for protection of
the FML during construction. In some cases, particularly where
unusually long drainage slopes may be part of the design,
drainage layers thicker than 30 cm (12 in.) and/or slopes greater
than 3 percent may be necessary. The minimum value of 1 x 10"2
cm/sec for permeability was chosen because granular materials
widely used as drainage media (i.e., SP soils) can provide this
minimum hydraulic conductivity. In situations where the minimum
criteria are insufficient or questionable, the design should
utilize flow modelling in arriving at the flow-controlling design
parameters. The HELP model (EPA, 1984a) can be of assistance for
this purpose.
Rounded grains with a maximum size of 3/8 inch (0.95 cm)
have been recommended, because they have been shown to be non-
damaging to most FMLs when in direct contact with them (EPA,
1984b). Crushed stone would not normally be appropriate due to
sharpness of the particles.
The drainage layer must slope to an exit drain which allows
percolated water to be efficiently removed. An example of an
exit drain is shown in Figure 4. Further information is provided
in EPA (1985a) and Bureau of Reclamation (1977) publications.
Care should be taken in the design to control the velocity of the
exiting water, within and beyond the exit drains, to prevent soil
loss and destabilization. Large safety factors may be needed to
accommodate unexpected events.
Materials used to construct the drainage layer should be
washed or screened prior to construction to remove fines that may
promote clogging. To further prevent clogging of the drainage
18
-------
drainage layers
FML
FML anchors
(separate anchor trench for each geosynthetlc)
low-permeability soil
waste
FML
Figure 4. Cover and liner edge configuration with example
toe drain»
layer, the Agency recommends that a granular or geosynthetic
filter be placed directly over the drainage layer to minimize the
migration of fines from the overlying topsoil into the drainage
layer. If a graded granular filter is used, care should be taken
to design the relationship of grain sizes according to the
criteria presented below (Cedergren, 1967).
To prevent
piping:
To maintain
permeability:
D,<; Filter
<4-5, and
D85 Top soil layer
D15 Drainage layer
D85 Filter
D15 Filter
D15 Top soil layer
D15 Drainage layer
<4-5
>4-5, and
>4-5
D15 Filter
19
-------
D50 Filter
To achieve uniformity <25, and
of grain size distribution D50 Top soil layer
curves among top soil
layer, filter, and
drainage layer:
D50 Drainage layer
<25
D50 Filter
These criteria are cited by the Array Corps of Engineers for
selection of a filter layer in relation to a soil to prevent the
soil from piping through the filter. D85 refers to the size of
particle in the gradation, below which 85 percent by weight of
the particles have a smaller particle size. D15 and D50 have
similar definitions. The criteria must be satisfied for all
layers or media in the drainage system, including protected soil,
filter media, and drainage media. Criteria for granular and
geotextile filter design are found in numerous references (Horz,
1984; Bureau of Reclamation, 1984 and 1977; EPA, 1987e; and
Koerner, 1989) .
Innovative drainage systems, such as those using
geosynthetic materials (see Figure 3b), may be used if it can be
shown that they are at least equivalent to the recommended
granular system in hydraulic transmissivity, in performance
longevity (transmissivity must be maintained for cover's design
life), and in their ancillary function as FML bedding. Criteria
which should be addressed in determining equivalence of
geosynthetic and soil drainage materials include, but are not
limited to, the following:
o hydraulic transmissivity (the rate at which liquid can
be removed) no less than 3 x 10"5 m2/sec;
o compressibility (the ability to maintain open pore
space and thus transmissivity, under expected
overburden);
o deformation characteristics (the ability to conform to
changes in the shape of the surrounding materials);
o mechanical compatibility with the FML (the tendency
for the drainage material and the FML to deform each
other);
o useful life of the system; and
o ability to resist physical, chemical and biological
clogging.
20
-------
Geosynthetic drainage materials are manufactured in a
variety of configurations, which continue to evolve with
experience in manufacturing and use. "Geonets" and "geogrids"
are drainage components designed for rapid flowthrough. They are
manufactured as single components that usually must be separated
from overlying and underlying soils that could clog them. The
separating materials are also geosynthetics in the form of filter
fabric. The geogrid, and top and bottom filters (which may also
serve as protective bedding and slide-resistant materials), may
all be factory-bonded together in one unit. These bonded-
together materials, one form of "geocomposites," may be applied
in one operation as the entire drainage layer. The various forms
of geocomposites are.well-described by Koerner (1989). In
geosynthetic materials are continually being improved by the
manufacturers for durability and design.
21
-------
4. LOW-PERMEABILITY LAYER
The final cover system is required by 40 CFR 264.228,
264.310, and 265.310 to provide long-term minimization of
migration of liquids through the closed land disposal unit and to
have a permeability less than or equal to the permeability of the
bottom liner system or natural subsoils present. The Agency has
interpreted this to mean that the cover should contain a FML/soil
composite layer (Figure 5) similar in concept (but not
necessarily identical construction materials) to the composite
bottom liner detailed in "Minimum Technology Guidance on Double
Liner Systems for Landfills and Surface Impoundments — Design,
Construction and Operation" (EPA, 1987i) and in proposed
regulations (EPA, 1987a). The two components (FML and soil) of
the low-permeability layer recommended in this document are
considered to function as one system. They should be designed,
constructed, and operated to maximize removal of water by the
' O o o 0°o0 0°c
R_°n° O P O °Q 0C
20-mllFML —*-rf
60-cm soil -j
waste
drainage layer
FML/sol!
low-permeability
layer
Figure 5. Detail of FML/soil composite
, low-permeability layer.
overlying drainage layer and to minimize infiltration of water
into the waste. The low-permeability layer should require little
or no maintenance during and after the post-closure period. The
Agency recommends the same design for both permitted and interim
status units, although it may not be required for some interim
status units.
22
-------
4.1 DESIGN
The Agency recommends that the low-permeability layer be
located below the maximum depth of frost penetration and, at a
minimum, consist of the following two components:
1. An upper FML component with the following
characteristics:
a. The FML should be at least 20 mils (0.5 mm) in
thickness, but some units and/or some FML
materials may require a greater thickness to
prevent failure under potential stress of the
post-closure care period, or during construction.
The Agency recognizes that some types of FMLs must
be thicker to accommodate unique seamability
requirements, or to increase long-term durability
(e.g., increase resistance to puncture).
b. The surface of the FML should have a minimum
3 percent slope after allowance for settlement.
c. There should be no surface unevenness, local
depressions, or small mounding that create
depressions capable of containing or otherwise
impeding the rapid flow and drainage of
infiltrating water.
d. The Agency recommends the use of material and seam
specifications such as those in "Lining of Waste
Containment and other Impoundment Facilities"
(EPA, 1987h).
e. The FML should be protected by an overlying
drainage layer of at least 30 cm (12 in.) of soil
material no coarser than 3/8-in. (0.95-mm)
particle size, Unified Soil Classification System
(USCS) SP sand, free of rock, fractured stone,
debris, cobbles, rubbish, roots, and sudden
changes in grade (slope) that may impair the FML.
The overlying drainage layer should suffice as
bedding in most cases, but care should be taken
that any included drainage pipes are not placed in
a way that will damage the FML.
f. The FML should be in direct contact with the
underlying compacted soil component and should be
installed on a smoothed soil surface.
23
-------
g. The number of penetrations of the FML by designed
structures (e.g., gas vents) should be minimized.
Where penetrations are necessary, the FML should
be sealed securely around the structure.
h. Bridging or similar stressed conditions in the FML
should be avoided by providing slack allowances
for temperature-induced shrinkage of the FML
during installation and during the period prior to
placement of the protective layer or drainage
layer.
i. Slack should not be excessive to the extent that
folds are created that later may crack.
2. A bottom low-permeability soil component with the
following characteristics:
a. The soil should be at least 60 cm (24 in.) of
compacted, low-permeability soil with an in-place
saturated hydraulic conductivity of 1 x 10"7
cm/sec or less.
b. The compacted soil must be free of clods, rock,
.fractured stone, debris, cobbles, rubbish, and
roots, etc., that would increase the hydraulic
conductivity or serve to promote preferential
water flow paths.
c. The upper surface of the compacted soil (which is
in contact with the FML) should have a minimum
slope of 3 percent after allowance for settlement.
d. The soil layer should be constructed so that it
will be entirely below the maximum depth of frost
penetration upon completion of the cover system.
The written CQA plan prepared by the owner/operator should
include a separate section specific to monitoring the
installation of both the FML and compacted soil liner (see EPA,
1987i).
4.2 DISCUSSION
The Agency believes that the recommended two-component low-
permeability layer design (Figure 5) is the best practicable, in
most cases, to minimize infiltration of surface water into the
underlying waste. Both the FML and the compacted soil components
have excellent characteristics to prevent infiltration into
underlying waste over the long term when properly designed,
installed, and operated in accordance with site-specific
conditions. Their characteristics tend to complement each other,
24
-------
so that the long-term effectiveness of the two components
together is greater than each alone. A summary discussion of the
comparative effectiveness of the composite liner in the bottom
liner application appears in a Federal Register notice (EPA,
1987f). A more complete discussion appears in "Background
Document on Bottom Liner Performance in Double-Lined Landfills
and Surface Impoundments" (EPA, 1987g). In short, the FML will
tend to roof over the inconsistencies in the underlying compacted
soil, while the compacted soil will tend to significantly impede
the flow of any leakage through a hole in the overlying FML. In
addition, each component tends to back up the other in the event
of a failure of either.
In the past, due to lack of data on durability, the Agency
has considered the FML to be short-lived compared to compacted
soil. Thus, the Agency has thought of the FML as fulfilling a
function of "short-term prevention" of infiltration, while the
soil provides for "long-term minimization." With increasing
knowledge of FML characteristics and performance, and the
increasing technical ability to custom-tailor FML materials to
the containment need, it is now the consensus that they, too, can
be made to last for very long periods of time (EPA, 1988a). Of
course, this implies that care be taken in the construction, and
later operation of the facility, that all design requirements are
met, that certain waste consolidation conditions are met to
minimize settlement problems, and that physical damage does not
occur. The same implication applies to the soil component even
though the design requirements and potential physical damage are
significantly different.
The following subsections provide more detail on the design
rationale for each of the two components of the low-permeability
layer.
4.2.1 FML Component
The Agency recommends that, in no case, should the thickness
of the FML be less than 20 mils (0.5 mm). The Agency believes
that this is the minimum acceptable thickness to meet cover
objectives and still be sufficiently rugged to withstand expected
stresses during construction and operation. In many, if not
most, cases the thickness should be greater. The adequacy of the
selected thickness should be demonstrated by an evaluation
considering the type, strength, and durability of the proposed
FML material, its seamability, and site-specific factors such as:
steepness of slopes, physical compatibility with the material
used in the underlying and overlying layers, stresses of
installation, expected overburden, climatic conditions,
settlement, and subsidence.
25
-------
FML failure mechanisms are discussed in several reports (EPA
1985a, 1983b, 1987h). Most failures result from inadequacies in
the design and construction processes. It follows then that most
failures can be prevented if a strict quality assurance program
is adhered to during the construction process. The Agency has
placed great emphasis on construction quality assurance,
particularly in the construction of barrier layers, and has
published guidance in that area for landfill waste containment
liners (EPA, 1986).
One of the causes of FML failure in landfill and surface
impoundment lining systems is chemical incompatibility. However,
the FML in a final cover should not come in direct contact with
any wastes and chemical incompatibility should not be of concern.
This makes it possible to accept a wider range of FML materials
in cover systems. It should be remembered here that it was not
the Agency's intent in the regulations that the bottom liner and
cover barrier necessarily be constructed of the same material,
Another of the primary causes of FML failure is damage
during installation or operation. To aid in preventing damage,
such as punctures, rips and tears, at least 30 cm (12 in.) of
bedding material above and below the membrane is recommended.
Since the FML is in direct contact with the low-permeability soil
layer, that layer will serve as the underlying FML bedding. In
most cases, the drainage layer above the membrane will suffice as
the overlying FML bedding. A minimum underlying bedding
thickness of 30 cm (12 in.) is recommended, the same as for the
drainage layer.* The actual bedding thickness should, however, be
based upon consideration of failure mechanisms and construction
methods potentially harmful to the FML (e.g., if construction
equipment or methods are capable of penetrating the 30-cm [12-
in.] drainage layer and tearing, ripping or puncturing the FML,
then the thickness should be increased). If the design
thicknesses for drainage and bedding differ, then the greater
thickness should be used. Geosynthetic drainage materials may
also serve as protective bedding if they can provide equivalent
protection for the design life of the cover system.
Penetration of the FML by gas vents or drainage pipes should
be minimized. Where a vent is necessary, it is essential to
obtain a secure, liquid-tight seal between the structure and the
FML to prevent leakage of water around the vent (see Section 5).
Settlement of the material around the structure may create
destructive stresses in the FML, which should be taken into
account in the design of both the structure and the FML collar.
Differential settlement across the cover may also cause
disruptive stresses that should be accounted for in the FML
design. Care should be taken to make allowance for these and
other stresses. For example, wrinkles and folds might be created
26
-------
intentionally to reduce stress, but they may, in turn, result in
stresses in the folds that can lead to long-term failure of the
FML (EPA, 1988b).
The subgrade for the FML must be carefully prepared and
smoothed so that no small-scale stress points are created due to
protrusions of rocks or other materials. In most cases, this
should cause no difficulty, since the subgrade will be the low-
permeability soil component, comprised of fine material.
Field-seaming of the FML must be done carefully by
technicians qualified and experienced in seaming the particular
FML being installed. Holes can result from discontinuous seams
or those not sufficiently sturdy to withstand unavoidable
stresses. Some FMLs require destructive surface preparation
(e.g., grinding) prior to seaming; all will expand and shrink
with temperature changes. These characteristics may promote
later leakage if not carefully considered in the construction
process. All of the potential failure causes can be minimized or
prevented by using expert installers and adhering to a strict
construction quality assurance program (EPA, 1986) .
4.2.2. Low-Permeability Compacted Soil Component
The Agency believes that a compacted soil component beneath,
and in direct contact with, the FML will:
o minimize, over the long term, liquid migration into
the waste in the event of FML failure or through
imperfections (holes, tears, etc.) inadvertently left
during the construction process;
o provide a firm foundation for the overlying layers of
the cover system;
o serve as bedding material for protection of the
overlying FML; and
o in conjunction with the FML, satisfy the regulatory
requirement for the cover to be no more permeable than
the bottom liner of the facility.
The design of the soil layer will depend on site-specific
factors including the properties and engineering characteristics
of the soil being compacted, the degree of compaction attainable,
the total expected load, and the expected precipitation.
The Agency recommends a minimum thickness of 60 cm (24 in.)
for the low-permeability soil component. The minimum thickness
is based upon constructability considerations and the ability to
provide uniformity in overall permeability. Sixty centimeters
allows for the installation of four lifts (see Figure 5),
27
-------
considered sufficient to overcome any inconsistencies in the
underlying surface. The four lifts also allow the localized
inconsistencies in permeability in one lift to be "sealed" by the
overlying lift.
As in the case of landfill liners, the Agency recommends the
use of a test fill (EPA, 1986, 1987i, and 1988c) prior to actual
construction of the soil component. The purpose is to
demonstrate, where appropriate soil is available, that the
compacted soil component actually can be constructed to an
hydraulic conductivity no greater than 1 x 10"7 cm/sec. (Most of
the ensuing discussion assumes that soil will be available that
can meet the 1 x 10"7 criterion.) The Agency believes that
construction of a test fill utilizing the soil, equipment, and
procedures to be used in construction of the low-permeability
layer will ensure that design specifications are attainable with
the available materials and equipment.
The test fill need not be constructed on the waste. The
Agency believes that, if the waste consolidation or compressive
strength of the waste is insufficient to allow adequate
compaction of the low-permeability soil component, that problem
should be corrected before installation of the compacted soil.
One of the possible solutions is to install an interim cover, as
noted earlier for the mitigation of subsidence. If components of
the interim cover can meet the CQA requirements for the final
cover, the interim cover could be made an integral part of the
final design.
Potential failure mechanisms that must be considered in
evaluating the design of the compacted soil component include
subsidence, dessication cracking, and freeze-thaw cycling.
Subsidence has been discussed in the Introduction. The main
factor of concern in design to counter subsidence is the
consolidation potential remaining at the time of cover
installation. That potential is difficult to estimate, but, in
the estimation, information regarding the presence of voids and
compressible materials in the underlying waste is all-important.
Ordinarily, most of the consolidation that will take place in
hazardous waste landfills has occurred by the time of cessation
of waste placement (EPA, 1985c and 1987d). An important benefit
is the ability of the compacted soil component to deform somewhat
without rupturing, a desirable characteristic related to the
soil's compressive and tensile strengths under expected field
conditions of moisture, density, etc.
The potential for desiccation of the compacted clay
component will depend on the physical properties of the clay,
design moisture content, local climatology, and moisture content
of the underlying waste. The actual clay-size particle content
of the soil, the type of clay, and properties such as liquid
28
-------
limit, plasticity, and shrinkage should be used to select a soil
with low cracking potential or in determining placement
procedures to reduce cracking potential (RTI, 1983; EPA, 1979).
Compaction of the soil component wet of optimum is
recommended by the Agency to assure that the lowest permeability
may be attained with standard Proctor densities (RTI, 1983). To
guard against drying in this case, the applicant may propose
immediate installation of the FML above the soil. If this is
done, it must first be assured that the installation of the soil
is complete, including a smooth surface on which to directly
apply the FML.
Freeze-thaw conditions are an important potential source of
damage to the soil component of the low-permeability layer.
Cycles of freezing and thawing may cause material cracking,
lessening of density, and loss of strength. This is brought
about by volume expansion of liquids in pore spaces during
freezing which, after thawing, increases the accessibility of
liquids to the pore spaces (EPA, 1983b). Cracking may be created
due to the expansion associated with freezing. For these
reasons, the Agency recommends that, upon completion of cover
system construction, the low-permeability layer be entirely below
the maximum depth of frost penetration estimated for the area in
which the facility is constructed. In other words, the top layer
and drainage layer of the cover together should be thicker than
the maximum depth of frost penetration. In northern areas of the
United States this recommendation would necessitate a top layer
thicker than the recommended 60-cm (24-in.) minimum.
Figure 6 is provided to show the variability of mean frost
penetration across the United States (Stewart, et. al., 1975).
The figure is provided only for perspective. It should not be
used to find the maximum depth of frost penetration at any
particular site. In determining the site-specific maximum depth
of frost penetration, advice may be sought from the Soil
Conservation Service, utility companies, construction
contractors, and universities in the area of concern.
Penetration of the low-permeability soil component by gas
vents or drainage pipes should be minimized. Adequate attention
must be given to the design of seals for such penetrations and
possible complications induced by differential settlement of
natural and man-made materials at penetration points. The Agency
has no information,specific to the adequacy of seals in the soil
component of the low-permeability layer.
29
-------
Figure 6. Regional average depth of frost penetration
in inches (Stewart, et al., 1975).
30
-------
5. OPTIONAL LAYERS
The optional layers discussed in this section are the gas
vent and biotic barrier layers. Other layers may be needed on a
site-specific basis. The Agency does not have information on the
performance of these layers in full-scale multilayer cover
systems.
5.1 GAS VENT LAYER ;
The function of a gas vent layer (Figure 7) is to control
combustible or toxic gases released from wastes buried in a
disposal facility. Hazardous waste disposal facilities that are
most likely to require a gas vent layer are co-disposal
facilities that receive organic waste material such as that found
in municipal waste. However, certain chemicals may also emit
gases or vapors in sufficient quantity to require venting.
gas vent
drain layer 1
FML
vent layer {f£l§J|
C?
top layer
low-permeability
FML/soil layer
waste
Figure 7. Cover with gas vent outlet and vent layer.
5.1.1 Design
The Agency offers the following design recommendations,
based upon engineering judgment, for a gas vent layer:
o The layer should be a minimum of 30 cm (12 in.) thick
and should be located between the low-permeability
soil liner and the waste layer (see Figure 2).
31
-------
o Materials used in construction of the gas vent layer
should be coarse-grained, porous materials such as
those used in the drainage layer.
o Geosynthetic materials may be substituted for granular
materials in the vent layer if equivalent performance
can be shown.
o Venting to an exterior collection point for disposal
or treatment should be provided by means such as
horizontal perforated pipes, patterned laterally
throughout the gas vent layer, which channel gases to
vertical risers.
o The number of vertical risers through the cover should
be minimized and located at .high points in the cross-
section, and designed to prevent water infiltration
through and around them.
An alternative design, particularly useful for layered
landfills where vertical migration is impeded, may include
perforated vertical collector pipes penetrating to the bottom of
the landfill. In this case, several cover penetrations may be
required, one for each standpipe. Here again, the pipes should
be securely sealed to the low-permeability layer. The standpipes
may be 30 cm (12 in.) or more in diameter and may be dual
purpose, serving also to provide access for measurement of
leachate levels.
The written CQA plan prepared by the owner/operator should
contain a specific section which covers monitoring the
construction and installation of the gas vent system.
5.1.2 Discussion
Materials used in construction of the gas vent layer should
have specifications similar to the granular material used for the
drainage layer. The materials should, be chosen and placed in a
way that facilitates the emplacement and compaction of the
overlying low-permeability soil component. Once placed, the
granular material should allow free movement of gases to
collection pipes and/or outlet points.
The outlets may consist of pipes or vents allowing the gas
to be collected, vented, or treated. The vent layer and outlet
should be designed to minimize cover penetrations which could
allow possible liquid infiltration through the cover. Outlet
vents should be constructed through the barrier layer at the
highest elevation of the gas vent layer to allow maximum
evacuation of gas (see Figure 7).
32
-------
A minimum thickness of 30 cm (12 in.) is recommended to
assure that a continuous layer of reasonable thickness (to allow
free movement of gases) is provided after placement on a non-
uniform waste surface.
In addition to providing gas removal, the gas vent layer may
provide a protective foundation upon which to construct the
compacted soil liner. The vent layer must be placed over the
waste up to design elevation, allowing for estimated settlement,
prior to placement and compaction of the soil liner. A filter,
either granular or geotextile, may be required between the gas
vent layer and the low-permeability soil to prevent clogging.
Alternative gas layer designs (e.g., using geosynthetic
materials) may be considered if it can be shown that they provide
a level of performance equivalent to a 30-cm (12-in.) granular
layer. Equivalence is based upon the ability of the design to
efficiently remove any gases produced, resist clogging, prevent
infiltration, withstand expected overburden pressures, and
function under the stresses of construction and operation.
Designs for gas vent layers can be found in several EPA
publications (EPA 1979, EPA 1985a).
Alternative, vertical standpipe gas collectors are
constructed of perforated sections, being built up as the unit is
filled with waste. They may be constructed of concrete and
wrapped with geosynthetic filter material to prevent clogging of
the perforations.
5.2 BIOTIC BARRIER LAYER
Plant roots or burrowing animals (collectively described as
biointruders) may disrupt the integrity of the drainage and low-
permeability layers. The drainage layer may be especially
susceptible to the intrusion of plant roots, which could
interfere with the drainage capability of the material. The
danger of FML penetration by plant intrusion has not been proven.
Burrowing animals may be a greater threat to FMLs, if a threat
indeed exists. In the absence of an FML, the low-permeability
soil layer could be exposed to both root and animal penetration.
Physical barriers, such as layers of cobbles or coarse
gravel beneath the top layer, and chemical barriers, have been
proposed to discourage or reduce the threat of biointrusion.
5.2.1 Design
The Agency knows of no full-scale application that would
prove the effectiveness of a biotic barrier in a landfill
situation. Therefore, the design of such a barrier must rely on
the results of small-scale field experiments. Experiments with
33
-------
barrier layers of cobbles have been carried out in arid or semi-
arid situations, with plants unique to such habitats (Cline, 1979
and Cline, et al, 1982). Some research suggests that three feet
(90 cm) of cobbles, or six inches (15 cm) of gravel over 30
inches (75 cm) of cobbles, may be effective in stopping root
penetration of some deep-rooted plants (DePoorter, 1982). It may
also be effective in stopping the invasion of burrowing animals.
The biotic barrier layer would directly underlie the soil
component of the top layer, perhaps separated by a geosynthetic
filter layer.
A polymeric herbicide carrier/delivery (PCD) system, used to
release herbicide, as discussed by Cline, et al. (1981), might be
installed within a cover, also just above the drainage layer to
stop the intrusion of roots below the system. The PCD system
would contain an herbicide designed to be released slowly over
many years. Note here the probable reluctance of the Agency in
approving this alternative, because it may introduce a hazardous
waste to the cover system, and/or it may not last through the
30-year post-closure period.
5.2.2 Discussion
Research by Cline (1979) and Hokanson (1986) found that if
objects, such as cobbles, placed in a burrowing animal's path are
too large or tightly packed, the animal's progress is effectively
stopped. Hokanson also found that large void spaces, which lack
water and nutrients, within the layer of stone, reduced the
intrusion of plant roots. On the other hand, the layer of very
coarse materials, at least in arid areas, may favor the growth of
grasses by impeding the downward percolation of moisture, thus
helping to retain it in the top soil layer.
Cline et al. (1982) also looked at the effectiveness of
several phytotoxins impregnated into polymeric sheets and buried
in soil. Some of them met the goal of being effective in
stopping the downward progress of root growth, with no other
effects. Some of the phytotoxins killed the plants when the
roots encountered the sheet, while others had no effect.
Obviously, a chemical biotic barrier must be chosen carefully, if
at all, to avoid potentially adverse environmental effects.
Most of the research on the effectiveness of biotic barriers
has been done in arid areas. Thus, the results must be used with
caution in areas of greater precipitation. The design and
resulting effectiveness of a biotic barrier are site-specific and
dependent upon the overlying topsoil layer, biotic barrier
material, natural precipitation, and anticipated biointruders.
34
-------
REFERENCES
Bowles, J. E. 1979. Physical and Geotechnical Properties of
Soils. McGraw-Hill, Inc., New York, NY.
Bureau of Reclamation. 1984. Earth Manual. U.S. Department of
the Interior. Government Printing Office, Washington, D.C.
Bureau of Reclamation. 1977. Design of Small Dams, U.S.
Government Printing Office, Washington, D.C.
Cedergren, H. R. 1967. Seepage, Drainage, and Flow Nets. John
Wiley & Sons, Inc., New York, NY.
cline, J. F. 1979. Biobarriers Used in Shallow-Burial Ground
Stabilization. Pacific Northwest Laboratory Report for
U.S. Department of Energy. PNL Report No. 2918.
Cline, J. F., D. A. Cataldo, F. G. Burton, and W. E. Skiens.
1982. Biobarriers Used in Shallow Burial Ground
Stabilization. In Nuclear Technology. Vol. 58, pp. 150-153.
August 1982.
DePoorter, G. L. 1982. Shallow Land Burial Technology
Development - Presentation for the Low-Level Waste
Management Program Review Committee. Los Alamos National
Laboratory, Los Alamos, NM.
Environmental Protection Agency. 1976. Erosion and Sediment
Control; Surface Mining in the Eastern United States,
Planning. Vols. 1 and 2. Report No. EPA-625/3-76-006.
Environmental Protection Agency. 1979. Design and Construction
of Covers for Solid Waste Landfills. Report No.
EPA-600/2-79-165. NTIS PB 80-100381. Municipal
Environmental Research Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1982a. Evaluating Cover
Systems for Solid and Hazardous Waste. Report No. SW-867.
NTIS PB 87-154894. Municipal Environmental Research
Laboratory, Cincinnati, OH.
35
-------
Environmental Protection Agency. 1982b. Remedial Action at
Waste Disposal Sites. Report No. EPA-625/6-82-006. NTIS PB
82-239054. Municipal Environmental Research
Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1983a. Hazardous Waste Land
Treatment. Report No. SW-874. NTIS PB 81-182107.
Municipal Environmental Research Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1983b. Lining of Waste
Impoundment and Disposal Facilities. Report No. SW-870.
NTIS PB 86-192796. Municipal Environmental Research
Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1983c. Standardized Procedures
for Planting Vegetation on Completed Sanitary Landfills.
Report No. EPA 600/2-83-055. NTIS PB 83-241018. Municipal
Environmental Research Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1984a. The Hydrologic
Evaluation of Landfill Performance (HELP) Model. Vol. 1
and 2, User's Guides for Version 1. Report No. EPA/530-SW-
84-009 and EPA/530-SW-84-010. NTIS PB 85-100840 and PB 85-
100832. Municipal Environmental Research Laboratory,
Cincinnati, OH.
Environmental Protection Agency. 1984b. Laboratory Studies of
Soil Bedding Requirements for Flexible Membrane Liners.
Report No. EPA 600/2-84-021. NTIS PB 84-14148. Municipal
Environmental Research Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1985a. Covers for Uncontrolled
Hazardous Waste Sites. Report No. EPA-540/2-85-002. NTIS
PB 87-119483. Hazardous Waste Engineering Research
Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1985b. Remedial Action at
Waste Disposal Sites. Report No. EPA/625/6-85/006. NTIS PB
87-201034. Hazardous Waste Engineering Research Laboratory,
Cincinnati, OH.
Environmental Protection Agency. 1985c. Settlement and Cover
Subsidence of Hazardous Waste Landfills. Report No.
EPA-600/2-85/035. NTIS PB 85-188829. Municipal
Environmental Research Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1986. Technical Guidance
Document: Construction Quality Assurance for Hazardous
Waste Land Disposal Facilities. Report No.
EPA-530/SW-86-031. NTIS PB 87-132825. Office of Solid
Waste and Emergency Reponse, Washington, DC.
36
-------
Environmental Protection Agency. 198?a. Proposed Amendments for
Landfill, Surface Impoundment, and Waste Pile Closures?
proposed amendment to rule. Federal Register, Vol. 52,
No. 53, March 19, 1987, Part ill.
Environmental Protection Agency. I987b. 40 CFR Parts 260, 264,
265, 270, and 271. Liners and Leak Detection for Hazardous
Waste Land Disposal Units? Notice of Proposed Rulemaking.
Federal Register, Vol. 52, No. 103, May 29, 1987, Part 111.
Environmental Protection Agency. I987c. Engineering Guidance
for the Design, Construction and Maintenance of Cover
Systems for Hazardous Waste. Report No. EPA/600/2-87/039.
NTIS PB 87-191656/AS. Hazardous Waste Engineering Research
Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1987d. Prediction/Mitigation
of Subsidence Damage to Hazardous Waste Landfill Covers.
Report No. EPA 600/2-87/025. NTIS PB 87-175378. Hazardous
Waste Waste Engineering Research Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1987e. Geosynthetic Design
Guidance for Hazardous Waste Landfill Cells and Surface
Impoundments. Report No. EPA/600/2-87/097. NTIS PB 88-
131263/AS. Hazardous Waste Engineering Research Laboratory,
Cincinnati, OH.
Environmental Protection Agency. 1987f. Federal Register Notice
re: Bottom Liner Performance in Double-Lined Landfills and
Surface Impoundments. Federal Register, Vol. 52, No. 74,
April 17, 1987. Pages 12566-12575.
Environmental Protection Agency. 1987g. Background Document on
Bottom Liner Performance in Double-Lined Landfills and
Surface Impoundments. Report No. EPA/530-SW-87-013. NTIS
PB-87-182291. Office of Solid Waste and Emergency Response,
Washington, DC.
Environmental Protection Agency. 1987h. Lining of Waste
Containment and Other Impoundment Facilities. Report No.
EPA/600/2-88/052. NTIS PB 89-129670/AS. Risk Reduction
Engineering Laboratory, Cincinnati, OH.
Environmental Protection Agency. 1987i. Minimum Technology
Guidance on Double Liner Systems for Landfills and Surface
Impoundments - Design, Construction and Operation. Report
No. EPA/530-SW-85-014. NTIS PB 87-151072. Office of Solid
Waste and Emergency Response, Washington, DC.
37
-------
Environmental Protection Agency. 1988a. Consensus Report of the
ad hoc Meeting on the Service Life in Landfill Environments
of Flexible Membrane Liners and other Synthetic Polymeric
Materials of Construction. Report No. EPA 600/X-88-252.
Unpublished.
Environmental Protection Agency. 1988b. Use of Construction
Quality Assurance (CQA) Programs. Unpublished memo on
stress cracking. Office of Solid Waste, Washington DC.
Environmental Protection Agency. 1988c. Design, Construction,
and Evaluation of Clay Liners for Waste Management
Facilities. Report No. EPA/530/SW-86/007F. NTIS PB 86-
184496. Office of Solid Waste and Emergency Response,
Washington, DC.
Hokanson, T. E. 1986. Evaluation of Geologic Materials to Limit
Biological Intrusion of Low-Level Radioactive Waste Disposal
Sites. Los Alamos National Laboratory. Report
NO. LA-10286-MS.
Horz, R. C. 1984. Geotextiles for Drainage and Erosion Control
at Hazardous Waste Landfills. Report No. EPA 600/2-86-085.
NTIS PB 87-129557. Hazardous Waste Engineering Research
Laboratory, Cincinnati, OH.
Koerner, R. M. (in press). Designing with Geosynthetics -
Second Edition. Prentice-Hall, Englewood Cliffs, NJ.
Lee, C. R., J. G. Skogerboc, K. Eskew, R. W. Price, N. R. Page,
M. Clar, R, Kort, and H. Hopkins. 1984. Restoration of
Problem Soil Material at Corps of Engineers Construction
Sites. Instruction Report EL-84-1, U.S. Army Engineer
Waterways Experiment Station, Vicksburg, MS.
Nyhan, J. W., W. V. Abeele, B. J. Drennon, W. J. Herrera, E. A.
Lopez, G. J. Langhorst, E. A. Stallings, R. D. Walker, and
J. L. Martinez. 1985. Development of Technology for the
Design of Shallow Land Burial Facilities at Arid Sites.
LA-UR-35-3278. Proceedings of the Seventh Annual
Participants' Information Meeting, DOE, Low Level Waste
Management Program.
Pertusa, M. 1980. Materials to Line or to Cap Disposal Pits for
Low-Level Radioactive Wastes. Geotechnical Engineering
Report GR80-7. University of Texas, Austin, TX.
Repa, E. W., J. G. Herrmann, E. F. Tokarski, and R. R. Eades.
1987. Evaluating Asphalt Cap Effectiveness at a Superfund
Site. Jour. Env. Eng., Vol. 113, No. 3. June 1987. pp.
649-653.
38
-------
Stewart, B. A., et al. 1975. Control of Water Pollution from
Cropland: Volume 1 - A Manual for Guideline Development.
USDA Report No. ARS-H-5-1. U. S. Dept. of Agriculture,
Hyattsville, MD.
Thornburg, A. A. 1979. Plant Materials for Use on Surface Mined
Lands. TP-157 and EPA-6QO/7-79-134. Soil Conservation
Service, U.S. Department of Agriculture, Washington, D.C.
Wright, M. J. (Ed.). 1976. Plant Adaptation to Mineral Stress
in Problem Soils. Cornell University Agricultural
Experiment Station. Ithaca, NY.
ft U-S. GOVERNMENT PRINTING OFFICE: 1989-648-163/00303
39
-------
•O
9""'^ jwr <— %
O H
a u
T3 £j
IH "*
U «'*
U V
u
S a
a
5-° *"
8 a
3 -5
4-» >
u
S 9
=*8
8<-S |
°-o-S.I
r-,
Reproduced by NTIS
National Technical Information Service
US. Department of Commerce
Springfield, VA 22161
This report was printed specifically for your
order from our collection of more than 2 million
technical reports.
For economy and efficiency, NTIS does not maintain stock of its vast
collection of technical reports. Rither, most documents are printed for
each order. Your copy is the best possible reproduction available from
our master archive. If you have any questions concerning this document
or any order you placed with NTIS, please call our Customer Services
Department at (703)487-4660,
Always think of NTIS when you want:
» Access to the technical, scientific, and engineering results generated
by the ongoing multibiliion dollar R&D program of the U.S. Government.
• R&D results from Japan, West Germany, Great Britain, and some 20
other countries, most of it reported in English.
NTIS also operates two centers that can provide you with valuable
information:
• The Federal Computer Products Center - offers software and
datafiles produced by Federal agencies.
• The Center for the Utilization of Federal Technology - gives you
access to the best of Federal technologies and laboratory resources.
For more information about NTIS, send for our FREE NTIS Products
and Services Catalog which describes how you can access this U.S. and
foreign Government technology. Call (703)487-4650 or send this
sheet to NTIS, U.S. Department of Commerce, Springfield, VA 22161.
Ask for catalog, PR-827.
Address,
- Your Source to U.S. and Foreign Government
Research and Technology.
-------
PB91-141846
EPA/600/2-91/002
October 1990
COMPILATION OF INFORMATION ON
ALTERNATIVE BARRIERS FOR LINER AND COVER SYSTEMS
by
David E. Daniel and Paula M. Estornell
University of Texas at Austin
Department of Civil Engineering
Austin, Texas 78712
Cooperative Agreement No. CR-815546-01-0
Project Officer
Walter E. Grube, Jr.
Municipal Solid Waste and Residuals Management Branch
Waste Minimization, Destruction and Disposal Research Division
U.S. Environmental Protection Agency
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
REPRODUCEDBY
U.S. DEPARTMENT OF COMMERCE
NATIONAL TECHNICAL
INFORMATION SERVICE
SPRINGFIELD, VA 22161
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA/600/2-91/002
4. TITLE AND SUBTITLE
COMPILATION OF INFORMATION ON ALTERNATIVE BARRIERS
FOR LINER AND COVER SYSTEMS.
7. AUTHORlSt
D.E. DANIEL AND P.M. ESTORNELL
9. PERFORMING ORGANIZATION NAME AND ADDRESS
DEPARTMENT OF CIVIL ENGINEERING
THE UNIVERSITY OF TEXAS
AUSTIN, TX 78712
12. SPONSORING AGENCY NAME AND ADDRESS
RISK REDUCTION ENGINEERING LABORATORY - L1NL1NNAI1, UH
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY CINTI, OH 45268
3.RE< PB91-1418U6
5. REPORT DATE
October 1990
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
CR-815546
13. TYPE OF REPORT AND PERIOD COVERED
INTERIM
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
PROJECT OFFICER: WALTER E. GRUBE, JR. FTS 684-7798
16. ABSTRACT
Workshop was held in Cincinnati, Ohio to present and discuss alternative barriers for liner and cover
systems. "Alternative" barriers include thin, manufactured, low-permeability materials that are being used and
being proposed for use in liner and cover systems for landfills, waste impoundments, site remediation projects,
secondary containment structures, and other facilities. The materials are being considered as an extra
component of a Uner or cover system, e.g., to back up a flexible membrane lint \FML), and in other cases, as
a substitute for a thicker layer of compacted, low-permeability soil.
This report contains a compilation of available information concerning alternative barrier materials and
summarizes the main points brought out in the woi kshop. There are four main alternative barrier materials
currently being produced. Three of them consist of a thin layer of bentonite sandwiched between two
geotextiles, and the fourth consists of a thin layer of bentonite glued to an FHL. All of the materials appear
to have a very low hydraulic conductivity (between 1 x 10 10 cm/s and 1 x 108 cm/s, depending upon the conditions
of testing). All of the materials are seamed in the field by overlapping sheets of hydrates. Data on the
hydraulic integrity of the seams are much less complete compared to data on the materials themselves. The
expansive nature of bentonite provides the bentonitic blankets with the capability of self-healing small
punctures, cracks, or other defects. The materials have many advantages, including fast installation with
light-weight equipment. The most serious shortcomings are a lack of data, particularly on field performance,
and the low shear strength of bentonite.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
18. DISTRIBUTION STATEMENT
b. IDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (This Report 1
20. SECURITY CLASS (Tliispagel
c. COSATI Field/Group
21 . NO. OF PAGES
93
22 PRICE
EPA Form 2220—1 (Rev. 4—77) PREVIOUS EDITION is OBSOLETE
-------
DISCLAIMER
The information in this document has been funded wholly or in part by the United States
Environmental Protection Agency under assistance agreement number CR-815546-01-0. It
has been subject to the Agency's peer and administrative review and has been approved for
publication as a U. S. EPA document. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial products and
practices frequently carry with them the increased generation of materials that, if improperly
dealt with, can threaten both public health and the environment. The United States
Environmental Protection Agency is charged by Congress with protecting the Nation's land, air,
and water resources. Under a mandate of national environmental laws, the Agency strives to
formulate and implement actions leading to a compatible balance between human activities and
the ability of natural systems to support and nurture life. These laws direct the U.S. EPA to
perform research to define our environmental problems, measure the impacts, and search for
solutions.
The Risk Reduction Engineering Laboratory is responsible for planning, implementing,
and managing research, development, and demonstration programs to provide an authoritative,
defensible engineering basis in support of the policies, programs, and regulations of the U.S.
EPA with respect to drinking water, wastewater, pesticides, toxic substances, solid and
hazardous wastes, and Superfund-reiated activities. This publication is one of the products of
that research and provides a vital communication link between the researcher and the user
community.
This report documents the available information concerning manufactured materials that
might be utilized in liner and cover systems for landfills, impoundments, site remediation
projects, and secondary containment structures. The information compiled in this report was
obtained from literature, from information supplied by manufacturers, and from discussions at
a 2-day workshop held on June 7 and 8 in Cincinnati. This report will be useful to scientists,
engineers, and regulatory staff who are considering use of these types of materials.
E. Timothy Oppelt
Director
Risk Reduction Engineering Laboratory
-------
ABSTRACT
On June 7-8, 1990, a Workshop attended by approximately 75 people was held in
Cincinnati, Ohio, to present and discuss alternative barriers for liner and cover systems.
Alternative barriers include thin, manufactured, low-permeability materials that are being
used and being proposed for use in liner and cover systems for landfills, waste impoundments,
site remediation projects, secondary containment structures, and other facilities. In some
cases, the materials are being considered as an extra component of a liner or cover system, e.g.,
to back up a flexible membrane liner (FML), and in other cases the alternative barriers are
being considered as a substitute for a thicker layer of compacted, low-permeability soil.
This report contains a compilation of information available concerning alternative
barrier materials and summarizes the main points brought out in the workshop. There are four
main alternative barrier materials currently being produced. Three of them consist of a thin
layer of bentonite sandwiched between two geotextiles, and the fourth consists of a thin layer of
bentonite glued to an FML. All of the materials appear to have a very low hydraulic conductivity
to water (between 1 x 10'10 cm/s and 1 x 10~8 cm/s, depending upon the conditions of
testing). All of the materials are seamed in the field by overlapping sheets of the material and
relying upon the bentonite to form its own seal when it hydrates. Data on the hydraulic
integrity of the seams are much less complete compared to data on the materials themselves.
The expansive nature of bentonite provides the bentonitic blankets with the capability of self-
healing small punctures, cracks, or other defects. The materials have many advantages,
including fast installation with light-weight equipment. The most serious shortcomings are a
lack of data, particularly on field performance, and the low shear strength of bentonite.
The advantages of alternative barrier materials are significant, and the materials
warrant further evaluation.
IV
-------
Table of Contents
Foreword iii
Abstract iv
List of Figures vii
List of Tables ix
1. Purpose of Workshop 1
2. Compacted Soil Liners 5
2.1 Materials 5
2.2 Important Variables 6
2.3 Construction of Compacted Soil Liners 1 0
2.3.1 Processing of Soil 1 0
2.3.2 Surface Preparation 1 2
2.3.3 Placement 1 2
2.3.4 Compaction 1 2
2.3.5 Protection 1 3
2.3.6 Quality Control Tests 1 3
2.3.7 Summary 1 3
2.4 Test Pads 13
2.5 Chemical Compatibility 1 4
2.6 Reliability of Soil Liners 14
3. Bentomat® 1 4
3.1 Description 1 4
3.2 Installation 1 5
3.3 Properties 1 6
3.3.1 Shear Strength 1 6
3.3.1.1 Direct Shear Tests 1 7
3.3.1.2 Tilt Table Tests 20
3.3.2 Hydraulic Properties 20
3.3.3 Seams 22
3.3.4 Mechanical Properties 22
3.4 Examples of Use 22
-------
Table of Contents (continued)
4. Claymax® 29
4.1 Description 29
4.2 Installation 31
4.3 Properties 30
4.3.1 Shear Strength 30
4.3.2 Hydraulic Properties 36
4.3.2.1 Tests with Water 3 8
4.3.2.2 Various Liquid and Chemical Leachates 40
4.3.2.3 Effects of Desiccation 4 2
4.3.2.4 Hydraulic Properties of Damaged Claymax 43
4.3.2.5 Composite Action 4 4
4.3.3 Seams 47
4.3.4 Swelling Characteristics 50
4.4 Examples of Use 5 1
5. Gundseal 53
5.1 Description 53
5.2 Installation 54
5.3 Properties 56
5.3.1 Physical Properties 56
5.3.2 Shear Strength 56
5.3.3 Hydraulic Properties 56
5.3.4 Seams 56
5.4 Examples of Use 5 8
6. Bentofix 59
6.1 Description 59
6.2 Installation 59
6.3 Properties 60
6.3.1 Shear Strength 60
6.3.2 Hydraulic Properties 60
6.3.3 Seams 60
6.3.4 Mechanical Properties 6 1
6.4 Examples of Use 61
7. Other Alternative Barrier Materials 63
v i
-------
8. Equivalency 65
9. Concerns 69
10. informational Needs , 71
11. List of References 74
12. Appendix: List of Participants ,. .....78
vn
-------
LIST OF FIGURES
Figure 1.1 Minimum Requirements for Liner Systems for Hazardous 2
Waste Landfills and Surface Impoundments
Figure 1.2 Recommended Design for Cover System for Hazardous Waste 3
Landfills and Surface Impoundments
Figure 2.1 Dry Unit Weight and Hydraulic Conductivity Versus Molding , 7
Water Content for Typical Compacted, Low-permeability Soils
Figure 2.2 Effect of Method of Compaction Upon Hydraulic Conductivity of 8
a Silly Clay Soil
Figure 2.3 Effect of Compactive Energy Upon Hydraulic Conductivity of a 9
Silty Clay Soil
Figure 2.4 Permeabile Inter-Lift Zones Providing Hydraulic Connection 1 1
between High-Hydraulic-Conductivity Zones in Adjacent Lifts
Figure 3.1 Schematic Diagram of Bentomat® 1 5
Figure 3.2 Results of Direct Shear Tests with Interfacial Shear Occurring 1 8
between Dry Bentomat® and Sand
Figure 3.3 Results of Direct Shear Tests with Interfacial Shear Occurring 1 8
between Hydrated Bentomat® and Sand
Figure 3.4 Results of Direct Shear Tests with Interfacial Shear Occurring 1 9
between Dry Bentomat® and Clay
Figure 3.5 Results of Direct Shear Tests with Interfacial Shear Occurring .....1 9
between Hydrated Bentomat® and Clay
Figure 3.6 Schematic Diagram of Tilt Table Tests 2 1
Figure 3.7 Results of Tilt Table Tests on Bentomat® 21
Figure 3.8 Results of Flexible-Wail Hydraulic Conductivity Tests on Bentomat® 23
Figure 4.1 Schematic Diagram of Ciaymax® 29
Figure 4.2 Mohr-Coulomb Failure Envelope for Direct Shear Tests Performed on. 33
Hydrated Bentonite with Shear Plane Passing through the Bentonite
within Ciaymax®
VIII
-------
List of Figures (continued)
Figure 4.3 Mohr-Couiomb Failure Envelope for Direct Shear Tests Performed on 33
Hydrated Bentonite with Shear Plane Passing through the Interface
between Polypropylene Geotextile on Claymax®
Figure 4.4 Mohr-Coulomb Failure Envelope for Direct Shear Tests Performed on 34
Hydrated Bentonite with Shear Plane Passing through the Interface
between Polypropylene Geotextile on Claymax® and a Silty Sand
Figure 4.5 Results of Consolidated-Drained Direct Shear Tests on Claymax® 35
Figure 4.6 Results of Consolidated-Drained Direct Shear Tests on Claymax® ......35
Figure 4.7 Results of Direct Shear Tests on Dry Samples of Claymax® 3 7
Figure 4.8 Results of Direct Shear Tests on Hydrated Samples of Claymax® .....3 7
Figure 4.9 Results of Hydraulic Conductivity Tests on Claymax® Permeated., 38
with Water
Figure 4.10 Schematic Diagram of Test to Evaluate In-Plane Flow... 46
Figure 4.11 Schematic Diagram of Laboratory Test Designed to Evaluate 48
the Hydraulic Conductivity of Overlapped Seam
Figure 4.12 Schematic Diagram of Tanks Being Used to Measure Hydraulic 49
Conductivity of Bentonitic Blankets Containing Overlapped Seams
Figure 4.13 Results of Swelling Tests on Samples of Claymax® 50
Figure 5.1 Schematic Diagram of Paraseal and Gundseal 5 3
Figure 5.2 Overlap of Paraseal 5 5
Figure 5.3 Schematic Diagram of Hydraulic Conductivity Test on Overlapped 57
Seam of Paraseal
Figure 6.1 Schematic Diagram of Bentofix 59
Figure 6.2 Results of Direct Shear Tests on Bentofix ...61
-------
LIST OF TABLES
Table 3.1 Summary of Results of Direct Shear Tests on Bentomat® 2 0
Table 3.2 Summary of Results of Hydraulic Conductivity (K) Tests on Bentomat® 23
Table 3.3 Physical Property Test Results: Dry Bentomat® Containing High... 25
Contaminant-Resistant Bentonite
Table 3.4 Physical Property Test Results: Hydrated Bentomat Containing High 26
Contaminant-Resistant Bentonite
Table 3.5 Physical Property Test Results for NonWoven Geotextile Component 27
of Bentomat®
Table 3.6 Physical Property Test Results for Woven Geotextile Component 28
of Bentomat®
Table 4.1 Material Specifications 30
Table 4.2 Summary of Results of Direct Shear Tests on Ciaymax® 34
Table 4.3 Results of Direct Shear Tests on Ciaymax® 36
Table 4.4 Results of Direct Shear Tests on Ciaymax®..., 36
Table 4.5 Results of Hydraulic Conductivity Tests on Ciaymax® 39
Permeated with Water
Table 4.6 Hydraulic Conductivity of Ciaymax® Permeated with Various 41
Liquids
Table 4.7 Results of Desiccation Studies on Sand Overlying Ciaymax® 42
Table 7.1 Summary of Results of Hydraulic Conductivity Tests on Fibersorb® 64
Table 8.1 Comparison of Differences in Alternative Barrier Materials 66
-------
Section 1
Purpose of Workshop
A 3-ft-(0.9 m) thick layer of low-permeability, compacted soil is a required
component of secondary liners for hazardous waste landfills and surface impoundments
regulated under the Hazardous and Solid Waste Amendments (HSWA) to the Resource
Conservation and Recovery Act (RCRA) [EPA, 1985]. The minimum primary liner for such
facilities must consist of a flexible membrane liner (FML). In addition, a secondary leachate
collection, detection, and recovery system (LCDRS) must be placed between the two liners
beneath hazardous waste landfills and surface impoundments, and, for solid-waste landfills, a
primary leachate collection and removal system (LCRS) must overlie the uppermost liner. The
minimum required components of a RCRA hazardous waste landfill liner system are sketched in
Fig. 1.1.
The recommended designs for cover systems over RCRA hazardous waste landfills and
closed surface impoundments include a 60-cm-thick layer of low-permeability, compacted soil
(EPA, 1989). A typical recommended design profile for a cover system is shown in Fig. 1.2.
Non-hazardous solid wastes are also regulated under RCRA, but requirements have yet to
be published by the EPA. Presently, the states are establishing requirements for liner and
cover systems for non-hazardous waste landfills. Requirements vary, but most minimum
design requirements are similar to the concepts shown in Figs. 1.1 and 1.2.
No minimum design requirements for final covers over Superfund sites have been
established. Typically, however, some type of control of water infiltration is included in the
final cover design. Typically, a layer of low-permeability, compacted soil is part of the cover
design.
Thus, a layer of low-permeability, compacted soil is either a required or recommended
component of most liner systems for hazardous and non-hazardous waste landfills and surface
impoundments, as well as final covers over buried wastes or contaminated soil. Program and
regional officials of the U.S. Environmental Protection Agency (EPA) are currently evaluating
requests to substitute thin, manufactured clay blankets (alternative barriers) for thicker,
low-permeability, compacted soil in liners and covers. Representatives of EPA, as well as state
regulatory personnel and design engineers, need to be aware of the advantages and disadvantages
of the alternative barrier materials and need to have access to the full breadth of available
information about the various materials.
-------
.;, ill .,...
Primary
Liner
. i
S/SS/SS
Secondary
Composite
Liner
Waste
Primary Leachate Collection
and Removal System (LCRS)
Flexible Membrane
Liner (FML)
Secondary LCDRS
(May Use Geosynthetic)
Flexible Membrane
Liner (FML)
Compacted Soil
Permeability < 10-7 cm/s
Thickness > 3 ft
Figure 1.1. Minimum Requirements for Liner Systems for Hazardous Waste Landfills and
Surface Impoundments (from EPA, 1985).
-------
EPA-Recommended Cover Design
for Hazardous Wastes
60cm
30cm
60 cm
KXX>O^*!
*.••••-. I •,••.••.••.•%••.••. ••.•%•*. •*.•'>.•*,••.*%
^ftftftftftWift^iftsaft***^
Top Soil
Filter
Drainage Layer
FML
Low Permeability
Soil Layer
Waste
Figure 1.2. Recommended Design for Cover System for Hazardous Waste Landfills and Surface
Impoundments (from EPA, 1989).
-------
To disseminate information, a workshop was held on June 7 and 8, 1990, at the EPA's
Risk Reduction Engineering Laboratory (RREL) in Cincinnati, Ohio, The purpose of the
workshop was: (1) to present to EPA technical staff and contractors, as well as state regulatory
officials, the latest available information concerning alternative barriers; and (2) to exchange
ideas that might prove useful in making research on alternative barriers consistent with
ongoing, parallel studies and responsive to the needs of permit writers and regulation
developers.
The specific topics discussed at the workshop were as follows:
1. Presentation of background information on conventional, low-permeability,
compacted soil iiners; functions served by compacted soil; performance of compacted
soil; factors to be considered in judging equivalency of other barrier materials
(presented by D. E. Daniel).
2. Description of alternative barriers (presented by Bentomat®, Clayrnax®, and
Gundseal representatives),
3. Discussion of parallel studies; results of various experiments performed on
compacted soil and manufactured barrier materials (presented by D. E. Daniel, P. M.
Estornell, and T, Zimmie).
4, Discussion of technical and regulatory concerns of EPA program offices, regional
offices, and state regulatory agencies about alternative barriers (open discussion).
5. Discussion of the prospects of using alternative barrier materials in liners and
covers for hazardous waste landfills, municipal solid waste landfills, and Superfund
closure sites; discussion of case studies (open discussion).
6. Discussion of research required to address the needs of permit writers and
regulation developers (open discussion).
This report will provide not only a summary of the proceedings of the Alternative
Barriers Workshop but will also document results from experiments recently conducted on
alternative barrier materials.
Background information on compacted soil barriers is provided in Section 2 of this
report. Information about Bentomat®, Claymax®, Gundseai, and Bentofix is presented in
Sections 3 through 6. Other alternative barriers are discussed in Section 7. The equivalency of
the alternative barrier materials is addressed in Section 8. Concerns about the alternative
barriers are summarized in Section 9. Research needs identified during the workshop are listed
in Section 10. A list of attendees is presented in the Appendix.
-------
Section 2
Compacted Soil Liners
Compacted soil liners are constructed primarily from naturally-occurring, low-
permeability soils, although the liner may contain processed materials such bentonite or even
synthetic materials such as polymers. Soil liners usually contain significant quantities of clay
and thus are frequently called "clay liners" even though clay may not be the most abundant
constituent in the liner material. Compacted soil liners are constructed in layers, called
"lifts," that are typically about 9 in. (225 mm) In loose thickness and 6 in. (150 mm) in
compacted thickness. Heavy compactors, or "rollers", are used to compact the soil.
2.1 Materials
The minimum requirements recommended by Daniel (1990) for most low-
permeability, compacted liners constructed from naturally-occurring soils are as follows:
Percentage Fines: 2. 30%
Plasticity Index: 2. 10%
Percentage Gravel: < 10%
Maximum Particle Size: 1 to 2 in. {25 to 50 mm)
Percentage fines is defined as the percent by dry weight passing the No. 200 sieve, which has
openings of 75u.m. Percentage gravel is defined as the percent by dry weight retained on a No. 4
sieve (4.76 mm openings). Local experience may dictate more stringent requirements, and, for
some soils, more restrictive criteria may be appropriate.
If suitable materials are unavailable locally, local soils can be blended with commercial
clays, e.g., bentonite, to achieve a low hydraulic conductivity. However, bentonite can be
attacked by some leachates -- compatibility tests may be required, A relatively small amount
of bentonite can lower hydraulic conductivity by several orders of magnitude (Daniel, 1987).
One should be cautious about using highly plastic soils (soils with plasticity indices >30
to 40%) because these materials form hard clods when the soil is dry and are very sticky when
the soil is wet. Highly plastic soils, for these reasons, are difficult to work with in the field.
However, special techniques, such as addition of lime, can ameliorate some of the problems with
construction utilizing highly plastic soils so that even these soils may be useable.
-------
2.2 Important Variables
Experience has shown that the water content of the soil, method of compaction,
compactive energy, clod size, and degree of bonding between lifts of soil can have a significant
influence on the hydraulic conductivity of compacted soil liners.
The water content of the soil at the time of compaction ("molding water content")
influences the hydraulic conductivity of saturated soil as shown in Fig. 2.1. When soils are
mixed to different water contents and then compacted, the dry unit weight is found to be
maximum at a certain molding water content, which is called the "optimum water content"
(dashed line in Fig. 2.1). Hydraulic conductivity is usually minimum for soils compacted at
molding water contents greater than the optimum. Experience has shown that the primary
causes for differences in hydraulic conductivity are differences in the arrangement of soil
particles (Mitchell, Hooper, and Campanella, 1965) and the fate of clods of clayey soil (Benson
and Daniel, 1990). With dry soils, the clods of soil are hard and difficult to remold. When the
soil is wetted to water contents higher than optimum, the clods are soft and more easily
remolded into a homogeneous mass that is free of large pores between clods. Thus, it is
important that the water content of the liner material be carefully controlled; otherwise,
undesirably large hydraulic conductivity may result, especially if the soil is too dry when it is
placed and compacted.
The method of compaction can influence the hydraulic conductivity of compacted soil.
Laboratory studies have shown that kneading the soil during compaction minimizes the
hydraulic conductivity (Fig. 2.2). Thus, footed rollers are typically utilized to compact soils in
the field; the "feet" from the drum of the roller penetrate into the soil to knead the soil during
compaction.
The energy of compaction is also an important variable. As shown in Fig. 2.3, the larger
the amount of energy delivered to the soil, the lower the hydraulic conductivity. In the field, it
is important to make an adequate number of passes of a heavy roller, and not to use too thick a
lift, to ensure that adequate compactive energy is delivered to the soil. The minimum weight
and number of passes varies with soil and equipment (Daniel, 1987; Herrmann and Elsbury,
1987; and Daniel, 1990).
The size of clods of soil can also influence hydraulic conductivity. Benson and Daniel
(1990) found that pulverization of clods of soil lowered the hydraulic conductivity of one
highly-plastic soil by a factor of 10,000 when the soil was compacted dry of optimum water
content. For wet soil with soft clods, the size of clods had little effect. For dry, hard soils, such
a shales, mudstones, or dry, highly-plastic soils, preprocessing the material with mechanical
pulverization may be required.
-------
Hydraulic
Conductivity
Dry Unit
Weight
Molding Water Content
Figure 2.1. Dry Unit Weight and Hydraulic Conductivity Versus Molding Water Content for
Typical Compacted, Low-permeability Soils.
-------
10
-5
10 6r
E
a
\o7-
10
-8
Optimum w —
Static
Compaction
Kneadinq Compaction
15 19 23 27
Molding w (%)
Figure 2.2. Effect of Method of Compaction Upon Hydraulic Conductivity of a Silly Clay Soil
(from Mitchell, Hooper, and Campanelia, 1965).
-------
i r 1 1 1
Increasing
Compactive
Effort
Optimum
Water
Content
Increasing Compactive
Effort
12 14 16 18 20 22 24
Molding w {%)
Figure 2.3. Effect of Compactive Energy Upon Hydraulic Conductivity of a Silty Clay Soil
(from Mitchell, Hooper, and Campanella, 1965).
-------
Experience has also demonstrated that lifts of soil must be bonded together to minimize
highly permeable zones at lift interfaces. The problem is illustrated in Fig. 2.4 for a liner
composed of 4 to 6 lifts. If each lift contains occasional hydraulic defects, liquid will permeate
primarily through those defects, if there is a highly permeable inter-lift zone, liquid can
spread laterally along the inter-lift zone until a hydraulic defect in the underlying lift is
reached. Thus, permeable inter-lift zones provide hydraulic connection between the more
permeable zones within adjacent lifts. If permeable inter-lift zones are eliminated, hydraulic
connection between "defects" in each lift is destroyed and a lower overall hydraulic conductivity
can be achieved. To maximize bonding between lifts, the surface of a previously-compacted lift
is roughened ("scarified"), and the new lift of soil is compacted with rollers that have feet that
fully penetrate the loose lift (to compact the new lift of soil into the surface of the previous
lift).
2.3 Construction of Compacted Soil Liners
2.3.1 Processing of Soil
Some liner materials need to be processed to break down clods of soil, to sieve out stones
and rocks, to moisten the soil, or to incorporate additives such as bentonite. Clods of soil can be
broken down with mechanical tilling equipment such as a rototiller. Stones can be sieved out of
the soil with large vibratory sieves or mechanized "rock pickers" passed over a loose lift of
soil. Road reclaimers (also called road recyclers) can process soil in a loose lift and crush
stones or large clods.
If the soil must be wetted or dried more than about 2 to 3 percentage points in water
content, the soil should first be spread in a loose lift about 12 in. (300 mm) thick. Water can
be added and mixed into the soil with agricultural tillage equipment or industrial mixers, or the
soil can be disced or tilled to allow it to dry uniformly. It is essential that time be allowed for
the soil to wet or dry uniformly. At least 1-3 days is usually needed for adequate hydration or
dehydration. Frozen soil should never be used to construct a soil liner.
Additives such as bentonite can be introduced in two ways. One technique is to mix soil
and additive in a pugmill. Water can also be added in the pugmill. Alternatively, the soil can be
spread in a loose lift that is 9 to 12 in. (222 - 300 mm) thick, the additive spread over the
surface, and a mechanical tiller or road reclaimer used to mix the materials. Several passes of
the mixer over a given spot may be needed, and the mixer should be operated in at least 2
different directions to minimize the possibility of strips of unmixed material. Water can be
added in the tiller during mixing or later, after mixing is complete. The pugmill is more
10
-------
Borehole
Lift 1
Lift 2
Lift 3
Lift 4
/
1 ^ L r ^
r* C
i ; \
^s X
"^ — ^
"\ r
^ r \ >
^ r (
Figure 2.4. Permeable Inter-Lift Zones Providing Hydraulic Connection between High-
Hydraulic-Conductivity Zones in Adjacent Lifts.
11
-------
reliable in providing thorough mixing, but, done carefully, in-field methods can provide
effective mixing,
2.3.2 Surface Preparation
It is crucial that each lift of a soil liner be effectively bonded to the overlying and
underlying lifts. The surface of a previously-compacted lift must be rough rather than smooth.
If the surface has been smoothed, e.g., with a finish roller at the end of a day's work shift, the
surface should be excavated to a depth of about 1 in. (25 mm) with a disc or other suitable
device before continuing placement of overlying lifts.
2.3.3 Placement
Soil is placed in a loose lift that is no thicker than about 9 in. (225 mm). If grade
stakes are used to gauge thickness, the stakes must be removed and the holes left by the stakes
sealed. Techniques that do not require penetration of the lift, e.g., laser controls, are preferable
to grade stakes. After the soil is placed, a small amount of water may need to be added to offset
evaporative losses, and the soil may be tilled one last time prior to compaction.
2.3.4 Compaction
Heavy, footed compactors with feet that fully penetrate a loose lift of soil are ideal. The
weight of the compactor must be compatible with the soil; relatively dry soils with firm clods
require a very heavy compactor whereas relatively wet soils with soft clods require a roller
that is not so heavy that it becomes bogged down in the soil. Care should be taken to ensure that
an adequate number of passes of the roller are made. Normal compaction specifications
typically require 6 to 8 passes of a roller to achieve the required density. Since the soil liner
is being build as a hydraulic containment structure, it is necessary to apply sufficient number
of passes that every portion of compacted soil receives the compactive energy applied by the feet
on the roller. The footprint area and the number of feet on the roller drum need to be taken into
account to calculate the minimum number of passes required for complete coverage of an area.
Additional passes beyond the theoretical minimum needed for 100 percent coverage should be
provided to account for the footprint overlap likely to occur in field construction. Experience
has shown that as many as 18 to 20 passes are required for some types of footed rollers to
achieve complete coverage. Since a kneading compaction helps to provide minimal hydraulic
conductivity, it is fallacious to use the common "walking out" endpoint to indicate that sufficient
compaction has been achieved. Experience indicates that minimum hydraulic conductivity has
been achieved while there remains some "waving" of the reworked soil ahead of the roller drum.
12
-------
"Walking out" needs to be monitored carefully as it may indicate that the soil is too dry to
achieve hydraulic conductivity objectives.
2.3.5 Protection
After compaction of a lift, the soil must be protected from desiccation and freezing.
Desiccation can be minimized in several ways: the lift can be temporarily covered with a sheet
of plastic, the surface can be smooth-rolled to form a relatively impermeable layer at the
surface, or the soil can be periodically moistened. The compacted lift can be protected from
damage by frost by avoiding construction in freezing weather or by temporarily covering the
lift with an insulating layer of material.
2.3.6 Quality Control Tests
A critical component in construction quality assurance is quality control (QC) testing.
For soil liners, the tests fall into two categories: (1) tests to verify that the materials of
construction are adequate, and (2) tests and observation to verify that the compaction process is
adequate. Great care must be taken to design an adequate program of QC testing and to repair
holes left from destructive QC tests. Details on QC testing are given by EPA (1986), Goldman et
al. (1988), and Daniel (1990).
2.3.7 Summary
Proper construction of soil liners is difficult. Materials must be carefully selected, the
soil may require extensive processing, the moisture content must be in the correct range, the
surface to receive a lift of soil must be prepared properly, the soil must be adequately
compacted, and each compacted lift as well as the entire liner must be protected from damage
caused by desiccation or freezing temperatures. Further information is provided by Daniel
(1987, 1990), Herrmann and Elsbury (1987), and EPA (1986, 1989).
2.4 Test Pads
The construction of a test pad prior to building a full-sized liner has many advantages.
By constructing a test pad, one can experiment with molding water content, construction
equipment, number of passes of the equipment, lift thickness, and other construction variables.
Most importantly, though, one can conduct extensive field-scale destructive testing, including
QC testing and in-situ hydraulic conductivity testing, on the test pad. Test pads are
recommended by the EPA (1985) for confirming that the materials and methods of construction
13
-------
will provide an adequately low hydraulic conductivity for the soil-liner component in' RCRA
hazardous waste landfills and surface impoundments.
The test pad usually has a width of at least 3 construction vehicles (>10 m), and an equal
or greater length. The pad should ideally be the same thickness as the full-sized liner, but the
test pad may be thinner than the full-sized liner. (The full-thickness liner should perform at
least as well as, and probably better than, a thinner test section because defects in any one lift
become less important as the number of lifts increases). The in-situ hydraulic conductivity
may be determined in many ways, the large sealed double-ring infiltrometer is usually the best
large-scale test (Daniel, 1989).
2.5 Chemical Compatibility
The compatibility of low-permeability soil liners with wastes to be retained must be
assured. Daniel (1987) and Goldman et al. (1988) discuss the mechanisms of attack and
summarize available data.
2.6 Reliability of Soil Liners
Examples can be cited of soil liners that had unacceptabiy large hydraulic conductivity
and therefore failed to function effectively as hydraulic barriers (Daniel, 1987; and Goldman et
al., 1988). Inadequate construction or construction quality control have been the main causes
of problems. Good-quality soil liners can be constructed (Gordon et al., 1989) if construction
is carried out very carefully and adequate construction quality is applied.
14
-------
Section 3
Bentomat®
3.1 Description
Bentomat® is manufactured by American Colloid Company, 1500 West Shure Drive,
Arlington Heights, Illinois 60004 (telephone 708-392-4600). The material consists of a
minimum of one pound per square foot (4.9 kg/m2) of dry (maximum 12% moisture),
granular, sodium bentonite sandwiched between two polypropylene geotextiles (Fig. 3.1). The
upper geotextile is woven while the lower geotextile is non-woven. The weights of the
geotextiles can vary but are typically about 3 to 6 oz. per square yard (102 to 204 g/m2).
Fibers from the upper geotextile are needlepunched through the layer of sodium bentonite and
into the lower geotextile (Fig 3.1). Variations of Bentomat® can be custom engineered to meet
site-specific needs. Also, one of four basic types of sodium bentonite may be incorporated into
Bentomat®- Each bentonite grade has different swelling properties and contaminant-resistant
properties. The four bentonites available have the following designations and properties.
"CS"- CS-50 (untreated, granular bentonite)
"SG"- SG-40 (polymer-treated, high-swelling bentonite)
"PL"- PLS-50, (medium-containment-resistant bentonite)
"SS"- SS-100 (high-contaminant-resistant bentonite).
Woven Geotextile
Needle-Punched
Fibers
1
«
I
I
I
%
&*
'*"••
^Sodium Bentonite
i?
f
Non-Woven Geotextile
Figure. 3.1 Schematic Diagram of Bentomat®.
15
-------
The standard roll size of Bentomat® is currently 12 ft (3.6 m) wide and 100 ft (30 m)
long. The thickness of dry Bentomat® is approximately 1/4 in (6 mm).
Bemalux Inc., of Quebec, Canada, originated Bentomat® in 1980. The original
Bentomat® was constructed in the field by laying a sheet of geotextile on a smooth surface,
spreading a layer of sodium bentonite (about 3 Ibs/ft^, or 14.7 kg/m^) over the geotextile, and
covering the bentonite with another geotextile. The prefabricated, needlepunched version of
Bentomat® was introduced in January, 1990, by American Colioid Company, which acquired
the U.S. patent rights for the product from Bemalux, Inc., in 1989.
3.2 Installation
The following discussion summarizes the manufacturer's recommendations for
installation. The subgrade should be compacted such that no rutting is caused by installation
equipment or vehicles. Subgrade or fill material should be free of angular or sharp rocks
larger than 1 inch in diameter. Organics or other deleterious materials should be removed.
Prior to the placement of Bentomat®, the surface should be graded to fill all major voids and
cracks.
Bentomat® is placed, beginning with the side slopes, by anchoring the panels in anchor
trenches and then unrolling the material down the slope. Panels may a%o be puiled up from the
bottom of the slope to the anchor trench. Seams at the base of the slope should be a minimum of
5 ft (1.5 m) away from the toe of the slope. Seams along the side slopes should be
perpendicular to the toe of the slope. Panels on flat surfaces do not require any particular
orientation.
Seams are formed by overlapping one panel on another. Seam overlaps should be a
minimum of 6 in. (150 mm) wide with contacting surfaces that are flat and clear of any large
rocks, dirt, or debris. The panels are printed with 6-in, and 9-in. (150 and 230 mm)
guidelines along both edges to aid in assuring that the minimum overlap width is achieved.
American Colloid Company recommends sprinkling granular bentonite at a rate of
approximately 0.25 pounds per liner foot (35 g/m) over a 3-in. (76 mm) wide swath in the
overlap zone. Fasteners, anchor pins, or adhesives may be used on seams to keep panels in place
during backfilling operations.
For pipe penetrations, a small notch should be cut in the subbase around the
circumference of the pipe. Bentonite should then be packed around the pipe in the area of the
notch to form a thick bentonite seal. The Bentomat® panel should be slit with an "X" in the
center, placed over the penetration, and sealed with bentonite to produce a seal. A second piece
16
-------
of Bentomat® should then be cut and fit around the pipe with bentonite applied between the
overlap and to any gaps that may exist.
A 12-in. (300 mm) thick layer of protective soil should be placed over the Bentomat®
liner, taking care to keep 12 in. (300 mm) of material between the liner and any machinery or
equipment at all times. Sharp turns and quick stops or starts should be avoided to prevent
pinching or moving the liner. When placing riprap on slopes, a layer of heavier geotextile
should be incorporated into the liner for added puncture resistance (American Colloid Company,
1990).
3.3 Properties
3.3.1 Shear Strength
3.3.1.1 Direct Shear Tests
Direct shear tests were performed on soil/Bentomat® interfaces by J&L Testing
Company (1990a). The frictional resistance between Bentomat® and sand and between
Bentomat® and clay was measured in a direct shear device for both dry and hydrated samples.
The tests were apparently designed to cause failure along the sand/Bentomat® or
clay/Bentomat® interface and not to produce failure within the bentonite.
Samples of Bentomat® measuring 100 mm by 100 mm (3.9 in. by 3.9 in.) were placed
against soil in a direct shear box and subjected to a constant rate of displacement of 0.009
in/min (0.24 mm/min). Normal stresses of 150, 300, and 450 psf (7.2, 14.4, and 21.5
kPa) were applied to each of the specimens. No standard method of testing these types of
materials exists; apparatus of the type normally used for soils was apparently utilized. Failure
occurred in most incidences at a horizontal displacement of approximately 0.2 in. (5 mm). The
time to failure is calculated by the authors of this report to be about 20 minutes. It is doubtful
that the rate of shear was slow enough to allow full dissipation of water pressures generated
within hydrated clay or bentonite during shear. For this reason, the test results probably do
not reflect the long-term performance of the materials or interfaces.
Results of direct shear tests are presented in Figs. 3.2 through 3.5 and are summarized
in Table 3.1. The failure envelopes shown in Figs. 3.2 through 3.5 were calculated by linear
regression. The calculated friction angles are between 28° and 41°.
For both the sand/Bentomat® and clay/Bentomat® tests, the friction angles were 7 to
10° higher when the bentonite was hydrated compared to dry bentonite. The authors of this
report would have expected lower friction angles with hydrated bentonite, but the results of the
tests were the opposite of this expectation. No explanation as to the cause for higher friction for
hydrated versus dry bentonite is apparent, except that the tests may more nearly reflect short-
17
-------
CO
co
CD
CO
CO
0)
-C
CO
400
300
200
100
c = 85 psf, Phi = 28 degrees
100 200 300 400
Normal Stress (psf)
500
Figure 3.2. Results of Direct Shear Tests with Interfacial Shear Occurring between Dry
Bentomat® and Sand (J & L Testing Company, 1990a).
CO
CO
CO
CO
CO
Q)
_c
CO
Figure 3.3.
400
300
200
100
c = 10 psf, Phi = 35 degrees
100 200 300 400
Normal Stress (psf)
500
Results of Direct Shear Tests with Interfacial Shear Occurring between Hydrated
Bentomat® and Sand (J & L Testing Company, 1990a).
18
-------
CO
Q.
CO
CO
CD
CO
l_
05
CO
400
300
200
100
1 00
c = 105 psf, Phi = 31 degrees
200
300
400
Normal Stress (psf)
500
Figure 3.4. Results of Direct Shear Tests with Interfacial Shear Occurring between Dry
Bentomat® and Clay (J & L Testing Company, 1990a).
CO
Q.
CO
CO
CO
1_
CO
0)
_c
CO
500
400 -
300 -
200 -
100 -
c = 77 psf, Phi = 41 degrees
100 200 300 400
Normal Stress (psf)
500
Figure 3.5. Results of Direct Shear Tests with Interfacial Shear Occurring between Hydrated
Bentomat® and Clay (J & L Testing Company, 1990a).
19
-------
term, undrained conditions rather than long-term, fully-drained conditions. These data are
specific to the sand and clay soils tested. American Colloid Company recommends against
extrapolating results to other soils; instead, site-specific testing is recommended.
Table 3.1 Summary of Results of Direct Shear Tests on Bentomat®
(J &L Testing Company, 1990a)
Cohesion Friction Angle
Sample (psf) (degrees)
Dry Bentomat® with Sand 85 28
Hydrated Bentomat® with Sand 10 35
Dry Bentomat® with Clay 105 31
Hydrated Bentomat® with Clay 77 41
(Note: 100 psf = 4.8 kPa)
3.3.1.2 Tilt Table Tests
Tilt table tests were performed by J & L Testing Company (1990a). A multi-layered
system composed of sand, high density polyethylene (HOPE) sheet, hydrated Bentomat®' and
geonet was placed on a tilt table (Fig. 3.6) to measure the friction angle along the weakest
interface (between smooth HOPE and hydrated Bentomat®). Normal stresses of 130 and 385
psf (6.2 and 18.4 kPa) were applied to the HDPE/Bentomat® interface, the table was inclined
slowly, and the inclination at which sliding was first observed was recorded. No information on
the time to failure was provided. Results of the tests are presented in Fig. 3.7. The friction
angle between the smooth HOPE sheet and Bentomat® was 13.5°.
3.3.2 Hydraulic Properties
J & L Testing Company (1990b) conducted flexible-wall permeability tests on 6-in.
(150-mm) diameter samples of Bentomat® containing either untreated granular bentonite
20
-------
Weights
Sand
Sliding
Surface
Geonet
HOPE
Hydrated
Bentomat®
Fastener
HOPE
Base
Figure 3.6. Schematic Diagram of Tilt Table Tests (J & L Testing Company, 1990a).
100
co
Q.
CO
CO
CD
CO
0)
.c
co
1 00
200
300
400
Normal Stress (psf)
Figure 3.7. Results of Tilt Table Tests on Bentomat® (J & L Testing Company, 1990a).
21
-------
("CS" grade) and high-contaminant-resistant bentonite ("SS" grade). Test conditions and
results are summarized in Table 3.2, The duration of the tests was not reported. Figure 3.8
presents the relationship between hydraulic conductivity and maximum effective stress,
Hydraulic conductivities ranged from 6 x 10~10 cm/s to 6 x 10~9 cm/s.
3.3.3 Seams
There have been no test results reported on the performance of Bentomat® seams.
Bench-scale hydraulic conductivity tests on seam overlaps are in progress at the University of
Texas at Austin, but no results were available at the time of this writing.
3.3.4 Mechanical Properties
Tests measuring grab strength, elongation, Mullen burst strength, wide width tensile
strength and other mechanical properties of Bentomat® were conducted by J & L Testing
Company, Inc. Tests were conducted according to ASTM standards, where available, on dry and
hydrated Bentomat® as well as on the individual woven and nonwoven geotextile components of
the liner material. Results of the test are summarized in Tables 3.3 through 3.6. Some
slippage of the mat occurred on wide-width tensile testing. Tests will be repeated with modified
grips.
3.4 Examples of Use
Bentomat® has been used in landfills, industrial and decorative lagoons, and as
secondary containment liners in tank farms. However, because of its recent release in January,
1990, the applications of Bentomat® have been limited.
The largest Bentomat® installation to date in the U.S. was a lake liner for a residential
development. A 9-acre (3.6 ha) lake was designed to be built in the midst of homes in the Cove
on Herring Creek development in Delaware. Due to the existence of poor quality native soils,
standing water, steep slopes and rough subgrade, the developer selected Bentomat® to line the
lake. The liner was placed through water and over soft subgrade by placing each panel and then
immediately following with a backhoe to place a foot of protective soil over the installed liner.
The lake was filled in June, 1990.
A contaminant-resistant grade of Bentomat® ("PL" bentonite, which contains a
polymer) was installed as a secondary containment barrier for petroleum tanks at a site in
Oklahoma. Approximately 8400 ft2 (780 m2) of liner material was installed. Lysimeters
were placed prior to all installations and the impoundments were flooded prior to their use to
22
-------
Table 3.2 Summary of Results of Hydraulic Conductivity (K) Tests on Bentomat®
(J&L Testing Company, 1990b)
Grade of Bentonite
High-Contaminant-
Resistant rss")
Untreated Granular
Bentonite ("CS")
E
o
o
3
"O
c
o
O
o
—
10
-8
10
-9
10
-10
Stress
(osn
Maximum
Cell Headwater Tailwater Effective
50 42,2 41.8 8.2
50 44.6
50 47.2
50 42.2
50 44,6
50 47.2
. • i '
• Bentomat CS
• Bentomat SS
D 4
Hydraulic
Conductivity
(crri/s)
2.1 x 10-9
39.4 10.6 7.5 x 10'10
36.8 13.2 5.8 x 10'10
41.8 8.2 5.6 x 10-9
39.4 10.6 1.1 x 10'9
36.8 13.2
w
1,1.
8 12 1
9.8 x 10-10
6
Max. Effective Confining Stress (psi)
Figure 3.8. Results of Flexible-Wall Hydraulic Conductivity Tests on Bentomat® (J&L
Testing Company, 1990b).
23
-------
both activate the bentonite and check for defects. No leaks were apparent through the
Bentomat® during this test.
Bentomat® has been installed in only one municipal landfill to date. A berm that was
built across the center of a large landfill cell was lined with 20,000 ft2 {1900 m2) of
Bentomat®. To insure the liner material would sea! against an HOPE liner, granular bentonite
was used at the HDPE/Bentomat® interface. The "PL" contaminant-resistant grade of
Bentomat® was used on this project.
24
-------
Table 3.3 Physical Property Test Results: Dry Bentomat® Containing High-Contaminant-Resistant Bentonite
(J&L Testing Company, 1990a)
ro
en
TEST
GRAB STRENGTH
MD-lnitial Peak
MD-Secondary Peak
GRAB ELONG.
MD-lnitial Peak
MO-Secondary Peak
MULLEN BURST
TRAPEZOIDAL TEAR
PUNCTURE
WIDE WIDTH
TENSILE
INTERGEOTEXTILE
SHEAR
INTERGEOTEXTILE
PEEL
ASTM
D-4632
D-4632
D-37B6
D-4533
D-4833
D-4595
D-3083(l)
0-413(1)
UNITS
Ibs
%
psi
MD/lbs
CD/lbS
Ibs
MD/lbs
Ibs/in
Ibs/in
REPLICATE NO.
1
96.9
125.1
9.3
102.7
270
64.9
62.4
71.7
316.5
17.2
6.6
Sep.(2)
2
79.4
N/A
12.0
N/A
300
44.3
61.5
90.9
317.3
; /.6
5.7
Sep.(2)
3
90.3
96.7
11.7
125.7
324
51.5
48.0
78.1
339.8
22.1
4.4
Sep.(2)
4
§2.6
134.9
15.0
141.7
333
50.7
75.9
126.4
293.1
26.0
2.4
Sep.(2)
5
99,0
100.9
14,7
110.0
395
63.5
77.9
131.3
345.5
27,9
7,4
Sep.(2)
AVERAGE
91.64
114.40
12.54
120.03
324.4
54.98
65.13
99.68
322.44
22.16
5.30
STD DEV
6.846
16.049
2.108
15.022
41.505
7.943
10.901
24.655
18.739
4.315
1,760
NOTES: (l)lntergootexiile shear and peel performed using 4 Inch wide specimens.
(2)Seam separated completely during test.
-------
Table 3.4 Physical Property Test Results: Hydrated Bentomat® Containing High-Contaminant-Resistant Bentonite
(J&L Testing Company, 1990a)
ro
01
TEST
GRAB STRENGTH
GRAB ELONG.
MULLEN BURST
TRAPEZOIDAL TEAR
PUNCTURE
WIDE WIDTH
TENSILE
INTERGEOTEXTILE
SHEAR
INTERGEOTEXTILE
PEEL
ASTM
0-4632
D-4632
D-3786
D-4533
D-4833
D-4595
D-3083(1)
D-413(1)
UNITS
Ibs
%
psl
MD/lbs
Ibs
MD/lbs
IbsJin
Ibs/in
REPLICATE NO,
1
88.6
26.7
130
44.6
35.9
276.8
42.0
13.4
Sep.(2)
2
82.5
25.0
120
59.6
39.6
238.1
49.1
Sep.(2)
14.0
Sep.(2)
3
84.7
23.3
75
51.5
38.0
288.2
67.0
20.6
Sep.(2)
4
87.1
21.7
135
51.0
39.9
295.7
49.1
Sep(2)
11.9
Sop.(2)
5
103.0
30.0
145
73.5
32.4
264.2
46.0
Sop.(2)
8.9
Sep.(2)
AVERAGE
90.86
25.43
121.0
56.02
37.15
272.58
50.63
13.74
Sep.(2)
STD DEV
7.147
3.200
24.372
9.932
2.785
20.260
8.584
3.835
NOTES: (l)lnlergeolextile shear and peel performed using 4 Inch wide specimens.
(2)Seam separated completely during test.
-------
Table 3.5 Physical Property Test Results for NonWoven Geotextile Component of Bentomat® (J&L Testing Company, 1990a)
ro
TEST
GRAB STRENGTH
GRAB ELONGATION
MULLEN BURST
TRAPEZOIDAL TEAR
PUNCTURE
WIDE WIDTH
TENSILE
ASTM
D-4632
D-4632
D-3786
D-4533
D-4833
D-4595
UNITS
MD/lbs
MO/%
psl
MD/lbs
Ibs
MD/lbs
REPLICATE NO.
1
75.2
188.4
162
34.6
54.0
297.9
2
56.2
120.0
173
45.1
49.2
256.8
3
104.6
126.6
218
39.2
25.0
129.1
4
69.5
116,6
178
31.2
42.4
134,6
5
75.2
175.0
192
13.2
36.9
121.0
AVERAGE
76.14
145.32
184.6
32,66
41.49
187.88
STD DEV
15.833
30.177
19.283
10.790
10.085
74.325
-------
Table 3.6 Physical Property Test Results for Woven Geotexfile Component of Bentomat® (J&L Testing Company, 1990a)
TEST
GRAB STRENGTH
GRAB ELONGATION
MULLEN BURST
TRAPEZOIDAL TEAR
PUNCTURE
WIDE WIDTH
TENSILE
ASTM
D-4632
D-4632
D-3786
D-4533
D-4833
D-459S
UNITS
MD/tbS
MD/%
psi
MD/lbs
Ibs
MD/lbs
REPLICATE NO,
1
151.4
20,0
175
69.6
9.2
524.1
2
152.4
23.3
326
62.2
12.4
576.9
3
145,9
23.8
352
62,4
13,3
555.3
4
147,8
22.5
340
67.7
16.0
588.4
5
166.6
21. 7
343
72.4
11.0
G57.8
AVERAGE
152.82
22.26
307.2
66.86
12.38
580.50
STO DEV
7.282
1,337
66.626
4,013
2.279
44.422
-------
Section 4
CIaymaxc
4.1 Description
Claymax® is manufactured by the James Clem Corporation, 444 North Michigan, Suite
1610, Chicago, Illinois 60611 (telephone 312-321-6255). The material is a flexible mat
consisting of granular sodium bentonite sandwiched between two geotextiles (Fig, 4.1). The
primary backing, or top geotextile, is a slit-film, woven, polypropylene geotextile. The
polypropylene geotextile typically weighs 3 oz. to 6 oz. per square yard (102 to 204 g/m2),
depending on the application, and provides durability and puncture resistance to protect and
support the system during installation. The secondary backing, or bottom layer, is usually a
spun-lace, open-weave polyester that weighs 3/4 oz. per square yard (25 g/m2), although
other materials can be substituted depending on specific requirements. The primary function of
the secondary backing is to hold the bentonite in place during installation. In addition, the open
weave of the backing allows the bentonite to expand when it hydrates and to ooze out between the
openings so that a seal is formed. Information about the geotextiles is given in Table 4.1.
Woven Polypropylene
Geotextile
with an Adhesive
J-r1 _-L_
_!_.
""' ' J 'T
»-S
1 I 1 I I I I 1 I I ! I I I I I 1 1 I I 1 I 1 1 I
Secondary
Backing
Figure 4.1 Schematic Diagram of Claymax13
29
-------
Table 4.1 Material Specifications (Supplied by Manufacturer)
(A) Primary Polypropylene
Substrate
Tensile Strength (ASTM 04632)
Burst Strength (ASTM D3786)
Puncture Strength (ASTM D3787 mod.)
Elongation (ASTM D4632)
Non-Biodegradable, Non-Toxic, Porous, Woven,
Slit-Film, Polypropylene Geotextile
78 Ibs. per inch (1,390 kg/m) minimum
250 psi (1720 kPa)
70 Ibs (32 kg)
15%
(B) Secondary Backing
Description
Highly porous, non-structural, non-woven fabric
that protects and contains the granular bentonite
during installation
(C) Bentonite
Material
Gradation of Bentonite
Amount of Bentonite
Final Moisture Content
Minimum Volumetric Increase
(ASTM E946-83)
Minimum Swell Index
(USP NF XVII, "Bentonite Swelling
Power")
Gradation of Raw Bentonite
Natural Sodium Bentonite Containing a minimum of
90% Montmorilionite
Two gradations: 6 Mesh and 16 Mesh Granules
Minimum of 1 Ib/ft2 (4.9 kg/m2) Measured at
Final Moisture Content
15 to 18% (Typical)
900%
25 ml
30
-------
Sandwiched between the two geotextiles is 1 Ib per square foot (4.9 kg/m2) of sodium
benton'rte adhered to the geotextiles with a water soluble, non-toxic, organic adhesive. The
sodium bentonite consists of a minimum of 90% montmorillonite and is specially graded to have
fine grained and coarse grained granules. Claymax® is manufactured in sheets that measure
approximately 1/4 in. (6 mm) in thickness, 13.5 ft (4.1 m) in width, and 100 ft (30.5 m) in length. The
sheets are placed on rolls, each of which weighs approximately 1400 Ibs (635 kg).
4.2 Installation
The following discussion details the installation procedures recommended by the
manufacturer. Before installation, the surface should be prepared by removing all angular
rocks, roots, grass, vegetation, and foreign materials or protrusions. All cracks and voids
should be filled. The surficial soils should be compacted to at least 90% of modified Proctor
density (ASTM D1557). The prepared surface should be free from loose earth, fully-exposed
rocks larger than 3/4 in. (19 mm) in diameter, rubble, and other foreign matter.
Claymax® is rolled out with the polypropylene side facing upward and with adjoining
rolls overlapping at least 6 in. (150 mm). No soil should be between the rolls in the
overlapped area. In hot, arid conditions, shrinkage may occur soon after placement; to account
for shrinkage, the longitudinal seam overlap should be increased to 9 in. (230 mm) and the
transverse overlaps increased to 4% of the run length plus 6 in. (150 mm). Seams should run
up and down a slope and never horizontally on slopes. Claymax® should not be installed in rain
or standing water; the material must be dry when installed and when covered. The liner should
be installed in a relaxed condition and should be free of tensile stress upon completion of the
installation. The liner may be pulled tight to smooth out creases or irregularities but should
not be stretched to force the liner to fit.
In windy areas, installation should commence at the upwind side of the project area. The
leading edge of the liner should be secured with sandbags or other means to hold the material in
position during installation. Only material that can be anchored and covered in the same day
should be unpackaged and placed in position. A trench should be used at the top of all slopes to
lock the liner in place by placing the end of the roll of Claymax® in the trench and backfilling
it. Irregular shapes or areas to be patched should be covered with sufficient material to provide
a 6-in. (150 mm) overlap in all directions. Patch repairs should not be allowed on slopes
steeper than 10%.
Claymax® must be protected from ultraviolet light and unrestrained hydration by
covering the material with a geomembrane and/or by placing 6 to 12 in. (150 to 300 mm) of
backfill or aggregate on top of Claymax®. If backfill is used, it should be compacted with
31
-------
wheeled rollers. Sheepsfoot compactors should not be used since the feet on the roller might
damage the Claymax®.
4.3 Properties
4.3.1 Shear Strength
Geoservlces Consulting Engineers (1989a) performed three sets of direct shear tests on
selected Claymax® interfaces for the James Clem Corporation. The purpose of the tests was to
evaluate internal shearing and frictional characteristics of fully-hydrated Claymax® placed
against a silty sand and a smooth polyvinyl chloride (PVC) geomembrane. The tests were
performed in 12-in. by 12-in. (300 mm by 300 mm) direct shear boxes which consisted of,
from top to bottom: (1) a layer of silty sand soil, dense sand, or, dense sand and 40 mil (1.0
mm} PVC geomembrane, depending on the specific test being conducted; (2) Claymax® that had
been fully hydrated for 24 hours under 500 psf (24 kPa) normal stress; and (3) a layer of
dense sand. Vertical stresses ranging from 100 to 575 psf (5 to 24 kPa) were used, and
shearing took place at a rate of 0.02 in/min (0.5 mm/min) with the upper half of the shear box
in motion and the lower half fixed. Failure was forced through the bentonite layer for one series
of tests; for the other series, failure was forced through the contact between the Claymax® and
the overlying material (PVC geomembrane or silty sand). It appears that the rate of shearing
may have been too rapid for long-term, fully-drained conditions to have been ensured.
Mohr-Couiomb diagrams are shown in Figs. 4.2, 4.3, and 4.4 for Claymax®,
CIaymax®/PVC, and Claymax®/sand, respectively. Results of the tests are summarized in
Table 4.2. The friction angles were 12° for Claymax® alone, 15° for Claymax®/PVC, and 17°
for CIaymax®/sand.
Chen-Northern (1988) performed direct shear tests on the bentonite layer of samples
of saturated Claymax® for a uranium mill tailings remedial action project (UMTRA) in
Durango, Colorado. Two consolidated-undrained tests and two consolidated-drained tests were
performed by applying strain rates of 0.047 and 0.00013 in/min (1.1 and .003 mm/min),
respectively, under normal stresses of 3, 6, and 12 psi (20.7, 31.3, and 82.7 kPa). The test
specimens were allowed to hydrate for 2 to 3 days prior to shearing. Results are plotted in Figs.
4.5 and 4.6 for undrained and drained tests, respectively. One data point for the consolidated-
undrained tests was left off of Fig. 4.5 because the point was inconsistent with the overall trend
of data (beyond the ordinary limits of variability of test data). The cohesion and friction angles
computed by least-squares regression are summarized in Table 4.3. With undrained conditions,
Claymax® had an average angle of internal friction of 16°. When sheared under drained
conditions, Claymax® samples had an angle of internal friction of 14° and negligible cohesion.
32
-------
en
CL
en
en
2?
CO
CO
03
.c
C/D
0.5
0.4
0.3
0.2
0.1
0.0
c = 0.03 psi, Phi = 12 degrees
0.0 0.2 0.4 0.6' 0.8 1.0 1.2 1.4 1.6 1.8
Normal Stress (psi)
Figure 4.2. Mohr-Coulomb Failure Envelope for Direct Shear Tests Performed on Hydrated
Bentonite with Shear Plane Passing through the Bentonite within Claymax®
(Geoservices, 1989a).
en
a.
en
en
£
,4—I
CD
re
CD
.C
CO
c = 3.4 psi, Phi = 15 degrees
01234567
Normal Stress (psi)
Figure 4.3. Mohr-Couiomb Failure Envelope for Direct Shear Tests Performed on Hydrated
Bentonite with Shear Plane Passing through the Interface between Polypropylene
Geotextile on Claymax® and a 40-mil PVC Geomembrane (Geoservices, 1989a).
33
-------
C/5
CL
CO
CO
CD
CO
1_
CO
CD
x:
CO
c = 3.9 psi, Phi = 17 degrees
2345
Normal Stress (psi)
Figure 4.4. Mohr-Coulomb Failure Envelope for Direct Shear Tests Performed on Hydrated
Bentonite with Shear Plane Passing through the Interface between Polypropylene
Geotextile on Claymax® and a Silty Sand (Geoservices, 1989a).
Table 4.2 Summary of Results of Direct Shear Tests on Claymax® (Geoservices, 1989a)
Sample
Cohesion
(DSf)
Friction Angle
(degrees)
Hydrated Claymax® Alone
1 2
Hydrated Claymax® Against PVC
490
1 5
Hydrated Claymax® Against Silty Sand
560
1 7
34
-------
10
c/>
0
CO
CO
0)
JZ
CO
!c = 1.8 psi, Phi = 16 degrees
8 10 12 14
Normal Stress (psi)
Figure 4.5. Results of Consolidated-Drained Direct Shear Tests on Claymax® (Chen
Northern, 1988).
tn
a.
in
2?
to
h*.
co
o>
-C
CO
c = 0,28 psi, Phi = 14 degrees
0 2 4 6 8 101214
Normal Stress (psi)
Figure 4.6. Results of Consolidated-Drained Direct Shear Tests on Claymax® (Chen
Northern, 1988).
35
-------
Table 4.3 Results of Direct Shear Tests on Claymax® (Chen-Northern, 1988)
Cohesion Friction Angle
Drainage Conditions (psf) (degrees)
Consolidated-Undrained Conditions 260 16
Consolidated-Drained Conditions 40 14
Direct shear tests were also conducted by Shan (1990). Consolidated-drained tests
were conducted on 2.5-in. (64-mm) diameter samples of both dry and fully saturated
Claymax® with constant strain rates of 0.63 and 0.0008 in. per hour (16 and 0.02 mm/hr),
respectively The soaking period was typically 2 to 3 weeks and the time to failure was
approximately 3 to 5 days for the saturated Claymax®. The rate of shearing used by Shan
(1990) appears to have been slow enough to ensure fully-drained failure. Normal stresses
ranged from 575 to 2880 psf (28 and 138 kPa). Results are summarized in Table 4.4. The
Mohr-Coulomb diagrams are shown in Figs. 4.7 and 4.8 for dry and hydrated bentonite,
respectively. The internal angle of friction was found to be 28° for the dry Claymax®, and 9°
for the hydrated Claymax®.
Table 4.4 Results of Direct Shear Tests on Claymax® (Shan, 1990)
Cohesion Friction Angle
Hvdration Condition (psf) (degrees)
Dry Bentonite 550 28
Hydrated Bentonite 90 9
36
-------
20
~ 16
CL
to 12
w
0)
03
0>
x:
03
c = 3.8 psi Phi = 28 degrees
10 15
Normal Stress (psi)
20
25
Figure 4.7. Results of Direct Shear Tests on Dry Samples of Claymax® (Shan, 1990).
en
Ct
V)
&
€>
JC
c = 0.6 ps Phi = 9 degrees
10 20
Normal Stress (psi)
30
Figure 4.8. Results of Direct Shear Tests on Hydrated Samples of Claymax® (Shan, 1990)
37
-------
4.3.2 Hydraulic Properties
4.3.2.1 Tests with Water
Literature published by the James Clem Corporation lists 2 x 10~10 cm/s as the
hydraulic conductivity of Claymax® permeated with deaired water. A summary of published
measurements of the hydraulic conductivity of Claymax® to water is given in Table 4.5.
Results are plotted in Fig. 4.9 in terms of hydraulic conductivity versus effective confining
stress. The results show that the hydraulic conductivity to water varies from just under about
1 x 10~8 cm/s at low effective stress to just above 1 x 1(T10 cm/s at high effective stress.
Estornell (unpublished) permeated an 8 ft by 4 ft (2.4 m by 1.2 m) piece of Claymax®
in a large tank, which is described more fully in Section 4.3.3. This data point is also shown in
Fig. 4.9 and is similar, though slightly larger than, the trend of the other data.
10
-8
o
^
XJ
c
O
O
o
10
-9
10
-1 0
Q Chen-Northern (1988)
• Geoservices (1988a)
• Geoservices (1889d)
x Shan (1990)
<• Shan (Unpub.)
n Estornell (Unpub.)
1 0
1 00
Effective Confining Stress (psi)
Figure 4.9 Results of Hydraulic Conductivity Tests on Claymax® Permeated with Water.
38
-------
Table 4.5 Results of Hydraulic Conductivity Tests on Claymax® Permeated with Water
CO
to
Source of Information
Clem Corp. Literature
Chen-Northern (1988)
Geoservices (1988a)
Geoservices (1989d)
Geoservices (1989d)
Geoservices (1989d)
Geoservices (1989d)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (Unpub.)
Permeameter
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Flex.
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Wall
Backpressure Diameter of
Saturation? Permeant Water Sample (in.)
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
Yes
Deaired Water
- -
Deaired Tap Water
Deaired Tap Water
Deaired Tap Water
Deaired Tap Water
Deaired Tap Water
Distilled Water
Tap Water
Distilled Water
Tap Water
Distilled Water
Distilled Water
Tap Water
2
2
2
2
2
2
4
4
4
4
4
4
1
.5
.8
.8
.8
.8
.8
.0
.0
.0
.0
.0
.0
2
Effective
Stress (psi)
3.5
2
3
3
3
3
2
2
5
5
1
2
2
9
0
0
0
0
0
0
Hydraulic
Conductivity
(cm/s)
2
2
4
8
8
3
7
2
2
1
8
6
3
2
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10-
10
10-
10"
10
-9
10
10
10-10
10-
10-
10
10
10
10
10
-9
-9
-9
10-10
10-
10-
10
10
10-9
-------
4.3.2,2 Various Liquid and Chemical Leachates
The information available concerning hydraulic conductivity of Claymax® permeated
with liquids other than water is summarized in Table 4.6. All of the test specimens that were
hydrated with water and then permeated with chemicals maintained a hydraulic conductivity < 1
x 10"8 cm/s, even for compounds such as diesel fuel and heptane that would normally be very
aggressive to soil liner materials. Brown, Thomas, and Green (1984), for example, found that
the hydraulic conductivity of a compacted, micaceous soil was 1 to 4 orders of magnitude higher
to kerosene, diesel fuel, and gasoline than it was to water. The inconsistency of results reported
in Table 4.6 to the research conducted by Brown and his co-workers may be related to either a
small cumulative pore volumes of flow in the tests on Claymax® or application of a high
compressive stress to the test specimens. The cumulative pore volumes of flow of permeant
liquid was not reported in many of the test referenced in Table 4.6; in many cases, there was
probably an insufficient quantity of flow to determine the full effects of the permeant liquids.
In some tests, a large effective confining stress was used. Broderick and Daniel (1990) found
that one compacted clay was vulnerable to significant alterations in hydraulic conductivity when
compressive stresses were < 5 - 10 psi (34 - 69 kPa) but did not undergo an increase in
hydraulic conductivity when the specimens were permeated with comoressive stresses larger
than 5 to 10 psi (34 to 69 kPa). Brown and his co-workers appliec no compressive stress to
their test specimens.
Tests on specimens of Claymax® that were hydrated with the same liquid as the eventual
permeant liquid (rather than water) showed mixed results. For leachates, a paper pulp sludge,
and simulated seawater, the hydraulic conductivity was found to be < 1 x 10"9 cm/s. However,
the significance of these results is questionable because the duration of the tests was short, the
cumulative pore volumes of flow was not reported, and the applied compressive stress was not
reported. In as-yet unpublished tests by Shan, markedly different results were obtained when
Claymax® was not prehydrated with water. Shan found that when dry Claymax® was permeated
directly with a 50% mixture of water and methanol, with pure methanol, or with heptane, the
bentonite did not hydrate even after several pore volumes of flow, and the hydraulic
conductivity did not drop below 1 x 10"6 cm/s. Shan used a compressive stress of 5 psi (34
kPa). Thus, with concentrated organic liquids, the conditions of hydration appear to play an
important role in determining the ability of the bentonitic blanket to resist the deleterious
action of organic chemicals. The bentonite appears to be more chemically resistant if hydrated
with fresh water before exposure to concentrated organic chemicals.
40
-------
Table 4.6 Hydraulic Conductivity of Claymax® Permeated with Various Liquids
Source of Information
STS Consultants (1988b)
STS Consultants (1988c)
Geoservices (1988b)
STS Consultants (1989a)
STS Consultants (1989b)
Geoservices (1989d)
Geoservices (1989d)
Geoservices (1989d)
Geoservices (1989d)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (1990)
Shan (Unpublished)
Shan (Unpublished)
Shan (Unpublished)
Permeant Liquid
Sewage Leachate
Paper Pulp Sludge
Simulated Seawaler
Landfill Leachate
Ash-Fill Leachate
Diesel Fuel
Jet Fuel
Unleaded Gasoline
Gasahol
50% (Vol) Methanol
Heptane
Sulfiric Acid
0.01 N CaSO4
0.5 N CaCl2
50% (Vol) Methanol
Methanol
Heptane
Hydration Liquid
Sewage Leachate
Paper Pulp Sludge
Simulated Seawater
Landfill Leachate
Ash-Fill Leachate
Water
Water
Water
Water
Water
Water
Water
Water
Water
50% Methanol
Methanol
Heptane
Pore Volumes
of Flow
- -
- -
- -
- -
1.5
2.5
1.6
0.5
2.2
0.2
3.1
2.2
24
4
5.4
4.3
Effective
Confining Stress
(psi)
- -
30
- -
- -
30
30
30
30
5
5
5
5
5
5
5
5
Hydraulic
Conductivity
(cm/sec)
8 x 10'10
2 x 10-10
2 x 10'10
4 x 10'10
1 x 10'10
9 x 10'10
9 x 10-10
3 x ID'10
3 x ID'10
9 x 10'10
1 x 10'10
6 x 10'11
1 x 10'9
8 x 1C-9
5 x 10'6
3 X 10-5
5 x 10-5
-------
4.3.2.3 Effects of Desiccation
The effects of desiccation were investigated by Geoservices (1989e). Three hydrated
samples of Claymax® were placed in a temperature- and humidity-controlled chamber. The
chambers operated on a timed cycle to simulate day and night conditions. The temperature and
humidity during the day cycle were 95°F and 30%, respectively, while the temperature and
humidity during the night cycle were 70°F and 50%, respectively. Samples of Claymax® were
buried below 8 in. (200 mm) and 18 in. (450 mm) of sand, while a third sample was not
buried beneath any sand. Water content samples were obtained from the Claymax® regularly
throughout the 3-month test period.
Results of the Geoservices tests are summarized in Table 4.7. The Claymax® sample left
exposed with no sand overburden underwent severe drying. In comparison, little or no
desiccation appeared to have occurred during the testing period when Claymax® was buried
beneath sand. The sand appeared to provide an adequate buffer to the extremes of temperature
and humidity to protect the Claymax® from desiccation.
Table 4.7 Results of Desiccation Studies on Sand Overlying Claymax® (Geoservices, 1989e)
Depth Below Water Content (%}
Top of Sand (in.) Elapsed Time (Days): 0 4 21 25 47 90
0 1300 690 15 - 6 1
8.5 260 - - - - 280 - - 260
18.5 300 - - - - 265 - - 248
Shan (1990) studied the effects of desiccation on the hydraulic properties of Claymax®
in a different way. Shan measured the hydraulic conductivity of 4-in (100-mm) diameter
samples .of Claymax® that had been subjected to several wet-dry cycles. His experiment
42
-------
involved permeating Claymax® specimens in a flexible-wall permeameter using an effective
stress of 2 psi (14 kPa) and a hydraulic gradient of about 50. The specimens were then
removed from the permeameter and allowed to air dry. At first the specimens were dried on a
laboratory table top with no overburden, but the specimens shrank to much smaller-diameter
circular discs and did not undergo significant cracking. To force desiccation cracks to develop, a
vertical stress of about 0.2 psi (1 kPa) was applied to the specimens while they dried. The
stress was applied by placing a steel cylinder on top of the sample to be desiccated. Numerous
large (2-mm-wide) desiccation cracks were seen in the dried specimens that had the small
overburden stress. The desiccated specimens containing cracks were set up again in a flexible-
wall permeameter and were permeated under the same conditions. After permeation, the
specimens were removed and desiccated/permeated again. It was found that after 3 wet-dry
cycles, the hydraulic conductivity of Claymax® did not change; it remained approximately 2 x
10"9 cm/s. Shan reported that at the beginning of repermeation after drying, the hydraulic
conductivity was on the order of 10"4 cm/s as water flowed through the cracks very easily.
But the cracks closed within a few hours and the flow stopped as bentonite hydrated and took in
water from both influent and effluent ends. It was not until bentonite was fully hydrated that
flow started again.
Chen-Northern (1988) conducted hydraulic conductivity tests in flexible-wall
permeameters on samples of Claymax® that had undergone 0, 3, and 10 cycles of wetting and
drying. The hydraulic conductivity increased approximately 2.6 times after three cycles of
wetting/drying but underwent no further increase with additional wet/dry cycles. Hydraulic
conductivities were 1 x 10~9 cm/s, 2.6 x 1Q~9 cm/s, and 2.3 x 10"9 cm/s, respectively, for
samples subjected to 0, 3, and 10 cycles of wetting and drying.
4.3.2.4 Hydraulic Properties of Damaged Claymax®
Hydraulic conductivity of Claymax® was measured by STS Consultants (1988a) on a
specimen of Claymax® that had been subjected to 15% elongation. The purpose of the
experiment was to study the hydraulic integrity of Claymax® after a specimen had undergone
deformation. The Claymax® specimen (evidently in a dry condition) was first stretched to 15%
elongation. A 2.5-in. (64 mm) diameter piece of the stretched Claymax® was trimmed from
the larger piece that had been stretched and was then placed above approximately 5-1/2 in.
(140 mm) of silica sand in a flexible-wall permeameter. The test specimen was hydrated,
saturated, and permeated with de-aired water. The effective consolidation stress was 0.15
kg/cm2 (125 kPa) and the backpressure was 4.0 kg/cm2 (392 kPa). The test was allowed to
continue until steady state was reached. The hydraulic conductivity of the material that had been
43
-------
subjected to a 15% elongation was determined to be 4 x 1CT10 cm/s. Elongation appeared to
have a negligible effect upon the hydraulic characteristics of the material.
The effects of punctures on the hydraulic conductivity of Claymax® was investigated by
Shan (1990). Punctures were simulated by cutting three holes, each 0.5, 1.0, or 3.0 in. (13,
25, or 75 mm) in diameter, in dry Claymax® specimens. The specimens were permeated with
tap water in a flexible-wall permeameter under an effective stress of 2 psi (14 kPa) and a
hydraulic gradient of about 50. The test specimens that had been punctured with holes 0.5 and
1.0 in. (13 and 25 mm) in diameter had hydraulic conductivities of 3 x 10'9 and 5 x 10"9
cm/s, respectively, which is only slightly larger than the value of 2 x 10~9 cm/s measured on
an intact specimen. With the specimen containing 3 holes each 3-in. (75 mm) in diameter, 2
of the 3 holes did not seal themselves and were left with openings of about 0.5 in. (13 mm) in
diameter. These tests, in conjunction with the tests on desiccated specimens, demonstrate that
the swelling nature of bentonite gives this material the capability of self-healing small defects
or punctures when the material is hydrated with water.
4.3.2.5 Composite Action
Shubert (1987) described various tests on composites of Claymax® placed adjacent to
defective HOPE geomembrane liners. In the first series of tests, Claymax® was placed between
two defective HOPE sheets in a configuration that simulated the usage of the material at a landfill
in the Chicago area. The four separate tests, the upper and lower HOPE sheet were slit over a
length of 1 in. (25 mm) with razor blade or punctured with a large nail, but the Claymax® was
left intact. The composites were tested in a flexible-wall permeameter with a maximum
effective confining stress of 30 psi (207 kPa). Leachate from a hazardous waste landfill was
used as the permeant liquid and was pressurized with 10 psi (69 kPa) to induce permeation
into the Claymax®. No inflow or outflow was recorded after initial pressurization of the system
over the 3-day test duration.
A second series of tests is described by Shubert (1987). Three samples were tested:
Sample 1: Top HOPE: punctured with 0.84-in. (21 mm) diameter hole
Bottom HOPE: punctured with 16-penny nail
Sample 2: Top HOPE: punctured with 0.84-in. (21 mm) diameter hole
Bottom HOPE: Slit with razor blade for 1 in. (25 mm) length
Sample 3: Top HOPE: punctured with 0.84-in. (21 mm) diameter hole
Bottom HOPE: punctured with 0.84-in. (21 mm) diameter hole.
44
-------
Test conditions were the same as in the first series of tests. After 5 days of permeation, the
quantity of inflow was 1.6, 0.8, and 0.8 ml for Samples 1, 2, and 3, respectively. Visual
observations after the tests revealed that the bentonite was significantly wetted near the
proximity of each membrane defect. The wetting of the bentonite, in turn, is reported to have
caused significant swelling of the bentonite, which caused plugging of the defect.
A third series of experiments was conducted on two samples that had holes of 0.84 in.
(21 mm) diameter drilled into the top and bottom HOPE sheets. Hazardous waste landfill
leaehate was introduced under the same testing conditions described earlier. The "apparent"
hydraulic conductivities of the samples were approximately 1 x 10~9 cm/s after more than
100 pore volumes of flow. A control test on Claymax® alone was not performed.
Shan (1990) conducted a test using a flexible:wall permeameter in order to measure
the in-plane hydraulic conductivity of Claymax® in contact with two sheets of high density
polyethylene (HOPE). The test set-up is shown in Fig. 4.10. The effective confining stress was
2 psi and a head of water of 1 ft was applied to one end of Claymax®. No outflow occurred for
about 2 months. When steady flow was finally reached, the computed hydraulic conductivity was
2 x 10"6 cm/s. At least some seal was obtained between the Claymax® and the HOPE because
there was no outflow for two months. However, in view of the high in-plane conductivity, the
seal was evidently imperfect.
Shan (unpublished) permeated 3 samples of 12-in. (300-mm) diameter Claymax® in
flexible-wall permeameters using an effective stress of 2 psi (14 kPa) and backpressure
saturation. Two of the three Claymax® samples were overlain by a sheet of defective HOPE
sheet (the sheet was placed against the polypropylene geotextile, which would normally be the
upper geotextile in the field); the third sample was a control with no HOPE. One of the two
HOPE sheets was punctured with 3 holes, each 1 in. (25 mm) in diameter, and the second was
slit with a 1-mm-wide slit having a length of 6 in. (150 mm). The hydraulic conductivities
were as follows:
Control (No HOPE): Hydraulic Conductivity = 2 x 10'9 cm/s
Composite (3 Holes in HOPE): Hydraulic Conductivity = 4 x 10"9 cm/s
Composite (Slit in HOPE): Hydraulic Conductivity = 4 x 1Q~9 cm/s
It is not known why the hydraulic conductivities of the composites were slightly greater than
those of the control -- the conductivities of the composites should have been less or equal to that
of the control. Nevertheless, the data do not indicate that a particularly good seal developed
between the HOPE and the bentonite. Liquid evidently spread laterally through the geotextile and
45
-------
(A) Control Test for Determining Baseline
Flowrate
HOPE
Claymax
®
(B) Test on Composite Action between
Claymax® and HOPE
Figure 4.10. Schematic Diagram of Test to Evaluate In-Plane Flow (from Shan, 1990).
46
-------
permeated a large percentage of the area of the Claymax®. Better contact might have been
achieved if the other side of the Claymax® (the side with the light-weight, spun-lace
polyester) was placed against the HOPE. This possibility is being evaluated by the authors.
4.3.3 Seams
STS Consultants Ltd. (1984) performed a hydraulic conductivity test on a 2-in. (50
mm) wide overlapped seam of Claymax®. The test arrangement is depicted in Fig. 4.11. The
specimens were placed above a 6-in. (150-mm) thick layer of silica sand in a flexible-wall
permeameter and were backpressure-saturated prior to permeation with deaired water. The
overlapped materials were hydrated and permeated from bottom to top (with flow from the
underlying sand to the Claymax®. No details on the effective compressive stress, hydraulic
gradient, magnitude of backpressure, or duration of test were given. The hydraulic conductivity
of the test materials with the overlapped seam was 7 x 10'10 cm/s using a weighted average
specimen thickness of 0.4 in. (10 mm).
Bench-scale hydraulic conductivity tests on seam overlaps are currently being
conducted at the University of Texas at Austin. The experiments are being performed in three
rectangular steel tanks, shown schematically in Fig. 4.12, that measure 8 ft (2.4 m) in length,
4 ft (1.2 m) in width, and 3 ft (0.9 m) in height. A 1/2-in. (13 mm) diameter drain hole has
been drilled at the center of the base. To conduct a test, a geotextile/geonet/geotextile composite
drainage layer is placed over the bottom of the tank (except that a 3-in. or 75 mm wide gap is
left between the drainage material and edge of the tank to accommodate a bentonite seal that seals
the material being tested to the bottom of the tank). Next, dry bentonite is placed in the 3-in.
(75 mm) wide gap left between the drainage material and the walls of the tank. The bentonitic
blanket being tested is placed over the drainage material and bentonite edge seal, with the edges
of the material going to the edges of the steel tank. Next, a 1-ft (0.3 m) or 2-ft (0.6 m) thick
layer of gravel is placed over the bentonitic blanket. The tank is slowly filled with a depth of
water above the bentonitic blanket of 1 to 2 ft (0.3 to 0.6 m). Effluent water passing through
the drain hole is collected and weighed to determine the flux of water through the material being
tested. The thickness of the material is estimated based on laboratory measurements. Hydraulic
conductivity is calculated from the measured flux and known head and known area and thickness
of the bentonitic blanket.
Tests have recently been completed on three samples of Claymax®. The tests involve:
(1) a 6-in. (150-mm) wide overlap (in accordance with the manufacturer's recommended
minimum overlap width); (2) a 3-in (75-mm) wide overlap (intended to evaluate whether the
recommended 6-in. or 150-mm wide overlap includes a generous factor of safety); and (3) a
47
-------
TOD View:
4 in. (100 mm)
Diameter
Side View:
2 in. (50 mm)
Sand
\
Claymax®
Figure 4.11. Schematic Diagram of Laboratory Test Designed to Evaluate the Hydraulic
Conductivity of Overlapped Seam (from STS Consultants, 1984).
48
-------
Steel Tank
1/2" Diameter Drain
Orthogonal View
Water
Gravel
Bentonitic Blanket
Drainage Hole
IT
Geonet/Geotextile
Drainage Layer
Bentonite Seal
Collection
Container
Cross Sectional View
Figure 4.12. Schematic Diagram of Tanks Being Used to Measure Hydraulic Conductivity of
Bentonitic Blankets Containing Overlapped Seams.
49
-------
control with no overlap. In the first tests, the gravel layer was 1-ft (0.3 m) thick, and the
depth of water above the Claymax® was 2 ft (0.6 m). These conditions were estimated to
produce an effective vertical stress of approximately 75 psf (3.6 kPa) at the top of the
Claymax® and 200 psf (9.6 kPa) at the bottom of the Claymax®. Steady state flow was
achieved after approximately 1.5 months, and the hydraulic conductivities were 9 x 10~9
cm/s, 2 x 10"8 cm/s, and 6 x 10~8 cm/s for the control, 6-in. (150 mm) wide seam, and 3-
in. (75 mm) wide seam, respectively. Hydraulic conductivity was calculated assuming a
thickness of 0.5 in. (13 mm) for the material based on data published by Shan (1990).
Effective overburden stress was later increased to approximately 220 psf (10.5 kPa) at the top
of the Claymax® but steady state conditions have not been reached as of this writing.
4.3.4 Swelling Characteristics
Shan (1990) measured the swelling characteristics of Claymax® as follows. A 2.5-in.
(64-mm) diameter specimen was trimmed, placed in a consolidation ring, compressed with a
controlled vertical stress, and then hydrated with water. The percentage change in height was
monitored until the sample ceased to swell or compress. The test was repeated for several
different levels of stress. Results are plotted in Fig. 4.13. The stress at which no compression
or swelling occurred was found to be approximately 3,000 psf (144 kPa).
D)
0)
IE
15
c
_
O
o
I
0)
05
d
m
-C
O
-1
1 00
1000
10000
100000
Effective Vertical Stress (psf)
Figure 4.13. Results of Swelling Tests on Samples of Claymax® (from Shan, 1990).
so
-------
4.4 Examples of Use
Lutz (1990) describes examples of the use of Claymax® based on information supplied
by the James Clem Corporation. The following was taken from Lutz's discussion.
The Broward County Landfill in Fort Lauderdale, Florida, is a 20 acre (8 ha)
incinerator ash monofill. County officials wanted an alternative to importing clay to construct a
liner for the landfill so they decided to use Claymax®. The components, from bottom to top, are
as follows: 6 in. (150 mm) of bedding sand; a 60 mil (1.5 mm) HOPE liner; 12 in. (300 mm)
of drainage sand; Claymax®; an 80 mil (2 mm) HOPE liner; and 24 in. (600 mm) of drainage
sand.
A 3 acre (1.2 ha) commercial hazardous waste landfill in Calumet City, Illinois,
contains a double composite liner system with a secondary leachate collection and removal
system. The components, from bottom to top, are as follows: 3 ft (0.9 m) of compacted clay; a
60 mil (1.5 mm) HOPE liner; a secondary leachate collection and removal system; Claymax®;
and a 100 mi (2.5 mm) HOPE liner. This landfill was completed in March of 1986, and there
has been no accumulation of leachate in the secondary leachate collection system.
St. Paul Island is in the Bering Sea and acts as a refueling site for fishing vessels. The
fuel storage tank farm needed to be enlarged and relocated. Since the island is a primary
breeding ground for the northern fur seal and has a large seabird population, the environmental
sensitivity of the island was a major concern. The cold, wet, and windy climate of St. Paul
Island makes construction difficult. Claymax® was used because of its ability to form a barrier
to fuel oils as well as its ease of installation. The liner was installed in sections between periods
of inclement weather. Pipe penetrations were sealed by wrapping Claymax® around the pipe at
the penetration. For tank farm applications, the Claymax® must be saturated with water after
it has been covered with the bedding material, otherwise the Claymax® will neither hydrate
properly nor impede the flow of a hydrocarbon spill.
During the open discussion session of the Alternative Barriers Workshop, Steve Walker
of Polyfelt, Inc., and John Boschuk of J & L Testing Company, Inc., described two other cases in
which Claymax® was used as a liquid barrier. A 60-acre (24 ha) ravine in the Hudson Valley
area was proposed as a site to contain PCB's. The existing subbase was a weak material
(standard penetration test N-value of 2) with organic deposits that generated gas. The
requirements for the liner were that it be impervious, collect gas, and act as reinforcement.
Instead of using compacted clay that would have been difficult to impossible to compact on the
existing subbase and would have required 22,000 truck loads of clay, a custom-made Claymax®
product was used which required only 90 tractor trailers at two-thirds the cost. The custom
made Claymax® liner was made using a 10 ounce per square yard (340 g/m2) geotextile cover
51
-------
fabric instead of the typical 1 ounce per square yard (34 g/m2) in order to meet the
requirements of collecting gas and providing reinforcement.
The second case involved a 5 acre (2 ha) pond at a campground. Two feet (600 mm)
below existing mudline was a layer of decomposed rock which was causing water to drain out of
the pond. Campground owners decided to line the pond with Claymax®. An anchor trench was
deemed unnecessary because the maximum side slope at the site was 12:1. A friction anchor was
used instead, consisting of 1 foot (300 mm) of soil placed above the Claymax® liner at the top
of the slope. Some time shortly after construction, a slope failure occurred. On the 16:1 (4°)
slopes, the Claymax® liner slid only a few inches; on the 12:1 (5°) slopes, the liner slid all the
way down the embankment. The sliding surface was between the geotextile and the ground
surface. It is hypothesized that movement of the slightly viscous bentonite in the Claymax®
caused slippage. It is evident from this failure that anchor trenches are important and that
more information is needed concerning the frictional characteristics of Claymax®.
In addition to the examples listed above, Claymax® liners have been used to waterproof
building foundations and to line waste lagoons and irrigation canals. Claymax® has also been
used as an alternative to cutoff walls and slurry walls and has been used as part of the core
material in a dam. This type of information is available in a series of publications supplied by
the manufacturer.
52
-------
Section 5.0
Gundseal
5.1 Description
Gundseal will be manufactured by the team of Gundle Lining Systems, Inc. (18100
Gundle Road, Houston, Texas 77073, telephone 713-443-8564), and Paramount Technical
Products, Inc. {2600 Paramount Drive, Spearfish, South Dakota 57783, telephone 605-642-
4787). The manufacturing facility will be located in Spearfish, South Dakota. Gundseal will be
similar to an existing product, Paraseal, which is manufactured by Paramount Technical
Products, Inc. Paraseal consists of one pound per square foot (4.5 kg/m2) of sodium bentonite
glued to a 20 mil (0.5 mm) HOPE geomembrane (Fig. 5.1), although the liner can be
manufactured with other thicknesses of HOPE. Paraseai can be supplied with or without a
light-weight fabric backing, which helps to prevent spalling of small granules of bentonite.
Paraseal is manufactured in 24-ft (7.3 m) long by 4-ft (1.2 m) wide rolls. Paraseal can be
installed with the HOPE facing upward or downward. The material is available with different
grades of bentonite, depending upon whether the bentonite is to retain fresh water or saline
water. Paraseal is seamed in the field with simple overlaps; although no mechanical seam is
necessary, mechanical seaming of HOPE to HOPE is possible. To date, Paraseal has been used
primarily for waterproofing basement walls, basement slabs, water-retention structures, and
small reservoirs and ponds.
Light-Weight Fabric Backing
fts&b'&fif&f&i-'&ft&ftow
||&?&?^io3um bentonite
Adhesive
High-Density Polyethylene Sheet
Figure 5.1. Schematic Diagram of Paraseal and Gundseal.
53
-------
,.-
Gundseal will be composed of the same materials as Paraseal but will be produced in
rolls that are approximately 17.5-ft {5.3 m) wide and approximately 200-ft (60 m) long.
Each roll will weigh just under 4,000 Ibs (1800 kg) and will have a diameter of
approximately 3 ft (1 m). The thickness of HOPE sheet in Gundseal will initially be 20 mil
(0.5 mm), but greater thicknesses are also expected to be available. Whereas Paraseal is
intended for use primarily in structural waterproofing, Gundseal is designed for applications
involving landfill liners and covers, liquid containment ponds, waste water lagoons, tank farms,
etc. Gundle Lining Systems, Inc., anticipates utilizing Gundseal as a back-up liner for
conventional HOPE geomembrane liners. For such an application, Gundseal would be installed
with the bentonite facing upward, as shown in Fig. 5.1. A conventional geomembrane liner
would then be placed directly on the bentonite. If there are any defects in the geomembrane
liner, such as a pinhole or defective seam, the leakage through the geomembrane would be
minimized by the bentonite layer within the Gundseal.
There are no technical'data currently available on Gundseal because the product has not
yet been produced. The following section presents information about the established Paraseal
liner, which is similar to Gundseal.
5.2 Installation
The following discussion summarizes the manufacturer's fcjommended installation
procedures. The area to be covered by Gundseal or Paraseal must be graded level. All rocks,
sticks, other sharp objects and loose soil should be removed. The product can be installed with
the HOPE side facing either up or down. If used by itself as a composite liner, the HOPE would
normally face up. If the material is used to backup a geomembrane liner that is placed on top of
the Paraseal, the product is installed with the bentonite facing up.
Paraseal is unrolled and placed on the area to be covered, with adjacent rolls overlapping
at least 1.5 to 3 in. (38 to 75 mm). The material is said to be self-seaming; when the
bentonite is hydrated and swells, the bentonite/HDPE contact is hydraulically sealed. Thus, no
mechanical joining of the seams is necessary (Fig. 5.2a), although the overlapped sheets of
HOPE can be mechanically joined with a double-sided tape called Para JT® (Fig. 5.2b). Para
JT® is a proprietary adhesive joint tape compounded "om a family of partially cross-linked
polymeric elastomers. Para JT® is placed on the HOPE along a strip where the bentonite has
been removed from the edge of the roll. This configuration results in a double seal: one seal is
made by the Para JT® between two pieces of HDPE and the second between the bentonite and the
HOPE. Other methods for joining the HDPE sheets, e.g., fillet extrusion welding, could probably
54
-------
Overlap
(A) Overlap of Paraseal
Overlap
ParaJT^
Adhesive Joint Tape
(B) Overlap of Paraseal with Adhesive Joint Tape
Figure 5.2. Overlap of Paraseal.
55
-------
be employed. Paraseal should be anchored in trenches around the perimeter of the site. Upon
completion of the liner, the Paraseal must be covered with soil or other protective material.
5.3 Properties
5.3.1 Physical Properties
The physical properties of Paraseal reported by the manufacturer are summarized as
follows. The HOPE membrane has a tensile strength (ASTM D412) of 4,000 psi (27 MPa).
Elongation at failure (ASTM D638) is reported to be 700%. Puncture resistance (Federal Test
Method Standard No. 101B) is 95 Ibs (43 kg). Permeance is reported to be 2.7 x 10~13
em3/cm2 when the membrane is applied to a porous stone and placed in a permeameter with a
pressure head equivalent to 150 ft (45 m) of water.
5.3.2 Shear Strength
No information is available on the shear strength of Paraseal.
5.3.3 Hydraulic Properties
Pittsburgh Testing Laboratory (1985) conducted a hydraulic conductivity test on a 2,5-
in. (64-mm) diameter sample of Paraseal. A 15-ft (4,6-m) head of water was applied to the
sample, which was soaked for 5 days prior to permeation. A single, falling-head test was
performed, which yielded a hydraulic conductivity reported to be 4 x 10~10 crn/s. Further
details of the test procedures are not available. However, because the direction of flow was
apparently through the HOPE membrane, the test may have provided a measure of sidewall
leakage rather than flow through the material.
5.3.4 Seams
Twin City Testing Corporation (1986) measured the hydraulic conductivity of the
bentonite in overlapped pieces of Paraseai with flow taking place parallel to the HOPE sheets. A
schematic diagram of the test arrangement is shown in Fig. 5.3. Two 1 in. by 4 in. (25 by 100
mm) pieces of Paraseal were placed against one another and clamped between two half-cylinders
of lucite. The assembly was placed in a flexible-wall permeameter. The overlapped pieces of
Paraseal were compressed with a stress of 24 psi (165 kPa), hydrated under a 6-in. (150-
rnm) head of water for 17 days, and permeated with a head of 40 ft (12 m) for 12 days. The
hydraulic conductivity for in-plane flow with this arrangement was 2 x 10~10 cm/s.
Bench-scale hydraulic conductivity tests on seam overlaps are currently being
conducted at The University of Texas at Austin. A description of the apparatus was given in
56
-------
section 4.3.3 and a diagram of the test apparatus is provided in Fig. 4.12. The three tests
currently underway on Paraseal liner are: (1) a 3-in. (75-mm) wide overlap; (2) a 1.5-in.
(38-mm) wide overlap; and (3) a control with no seam overlap. One foot (300 mm) of gravel
was placed over the Paraseal sheets, and 2 ft (0.6 m) of water was ponded on top of the sheets.
There was no outflow from any of the three test specimens over the entire 5-month testing
period.
BENTONITE BACKING SEAL
POROUS STONE
24 PSI EFFECTIVE
CONFINING PRESSURE
MEMBRANE
Figure 5.3. Schematic Diagram of Hydraulic Conductivity Test on Overlapped Seam of
Paraseal (from Twin City Testing Corporation, 1986).
57
-------
5,4 Examples of Use
Paraseal has been used primarily as a waterproofing material for building basements
and, to a lesser extent, to line water-retention ponds.
58
-------
Section 6,0
Bentofix
6.1 Description
Bentofix is manufactured by Naue-Fasertechnik Co, in Lubbecke, Germany. Information
on the material was obtained from Scheu et al. (1990) and from personal communication with
Robert M. Koerner.
Bentofix is a fiber-reinforced, bentonitic mat composed of a neediepunched, nonwoven
geotextile as a cover, bentonite as a sealing element, and a neediepunched, nonwoven geotextile
as a base layer (Fig. 6.1). The bentonite used for the manufacture of Bentofix is an activated
sodium bentonite (a calcium-bentonite modified to a sodium bentonite) containing 70%
montmorillonite. The geotextile layers are neediepunched together through the bentonite layer
with a large amount of single stitches per square inch to form the Bentofix mat (Fig, 6.1).
Granular Bentonite
Neediepunched Fibers
Non-Woven Geotextile
Figure 6.1. Schematic Diagram of Bentofix (from Scheu et al., 1990).
6.2 Installation
Bentofix may be placed on irregular surfaces, like slightly eroded embankments and
channel beds. Larger pot holes must be filled with concrete and exposed buckles must be
removed. Joints are made by overlapping the material. An overlap width of at least 12 in.
(300 mm) is recommended. Wet granular bentonite is placed along the edge of the overlap
sheet using a "U" shaped device that applies bentonite to the overlapped section in order to
59
-------
increase the integrity of the seam. A ballast layer of underwater concrete or crushed stones
must be placed over the Bentofix mats in order to protect the material and keep the mats in
place. The system can be applied throughout the year without any seasonal restrictions.
Underwater installations are possible, as well.
6.3 Properties
6.3.1 Shear Strength
Direct shear tests were conducted by the Franzius Institute for Hydraulic Research and
Coastal Engineering, University of Hannover, to determine the frictional behavior of Bentofix
(Scheu et al., 1990). Bentofix specimens were placed between two sand layers in a direct shear
box. The specimens were then saturated for three days under normal stresses of 50, 100, and
200 kPa and sheared until a total displacement of approximately 30 mm had been achieved. The
time to failure was not reported by Scheu et al. Sliding took place within the bentonite layer,
which caused the needle-punched threads to align themselves according to the direction of
displacement. The Mohr-Coulomb diagram is shown in Figure 6.2. The cohesion was found to be
8 kPa (1.2 psi), and the angle of internal friction was 30°.
6.3.2 Hydraulic Properties
Hydraulic conductivity tests were carried out at the Institute for Foundation
Engineering, Soil and Rock Mechanics of the Technical University, Munich (Scheu et al., 1990).
Bentofix specimens were placed in triaxial cells, back-pressure saturated, and permeated with
de-aired water. The hydraulic conductivity of water through the Bentofix sample was
determined to be 1 x 10"^ cm/s under an unspecified effective confining stress.
6.3.3 Seams
The Franzius Institute (Scheu et al. 1990) conducted hydraulic conductivity tests on
overlapped seams and overlaps with an intermediate bentonite layer. A large box with a drain at
the bottom was used to contain the overlapped Bentofix samples. Water from an upper overflow
reservoir was fed into the box where it then permeated through the overlapped samples and
collected in a measuring glass located beneath the drain. A set of piezometer tubes were used to
measure the change in head through the sample. From these experiments, the hydraulic
conductivity of water through the Bentofix seams was determined to be 1 x 10"^ cm/s for
overlapped sections of Bentofix containing bentonite between the sheets of Bentofix. The
compressive stress applied to the overlapped area was not specified.
60
-------
CO
to
CD
CO
c = 1.2 psi, Phi = 30 degrees
5 10 15
Normal Stress (psi)
Figure 6.2 Results of Direct Shear Tests on Bentofix (from Scheu et a!., 1990)
6.3.4. Mechanical Properties
Data on the mechanical properties of Bentofix could not be located.
6.4. Examples of Use
Bentofix can be used for many lining applications such as water reservoirs, channels,
artificial lakes, dams, and landfills. The installation examples cited by Scheu et al. (1990)
include a dam rehabilitation project and a chemical containment project.
The Lechkanal is a diversion canal built in the 1920's. The canal runs parallel to the
river Lech, in Germany. The weirs and locks integrated in the diversion canal are used to
generate electricity. Some portions of the 70-year-old canal are lined with man-made levees.
Surface erosion and minor piping channels have developed along sections of the levees over a
long period of time. Instability was solved using a double lining system that consisted of 140
mm of asphalt with a filter fabric drainage layer as the primary liner and Bentofix as the
secondary liner.
61
-------
A purification network was recently constructed at the Munich II Airport in Germany in
order to collect and purify the runoff from runways. Deicing of airplanes with a mixture of
glycol and hot water creates a hazardous runoff that has a potential to contaminate the
underlying groundwater. The purification network at the Munich Airport consists of
underground granular filters that support bacteria used to biologically purify the glycol and
water mixture. The sealing element between the purification system and the groundwater is
Bentofix.
62
-------
Section 7.0
Other Alternative Barrier Materials
Some additional barrier materials that were identified during the workshop include:
1. Flyash-bentonite-soil mixtures;
2. Super-absorbant geotextiles (e.g., Fibersorb®);
3. Sprayed-on geomembranes (intended to form a composite with a compacted soil
layer or bentonitic blanket);
4. Custom-made bentonite composites with geomembranes or geotextiles.
Fly-ash was not discussed because it did not appear to fit within the theme of thin, manufactured
materials, which was the main focus of attention at the Workshop. Few details were presented
concerning spayed-on products or custom-made bentonite composites, other than to indicate
that sprayed-on products are promising and that all of the bentonitic blankets can be custom-
designed and fabricated to meet particular needs, e.g., by using a thicker geotextile as a gas
venting medium.
Information concerning Fibersorb® was supplied via a manufacturer's brochure and a
technical report (STS Consultants, 1990). Fibersorb® is a lightweight geotextile containing
thin, nonwoven, superabsorbent fibers and is manufactured by ARCO Chemical Company (3901
West Chester Pike, Newton Square, Pennsylvania 19073, telephone 215-359-5616). When
water contacts the fibers, the resultant swelling fills voids and impedes water flow.
Fibersorb® has been used primarily in protective clothing, in packaging, filters, or humidity
control systems to seal out water or moisture, as industrial wipes, and has even been used as an
emergency heat barrier in the case of a fire due to it's heat absorption capabilities.
STS Consultants (1990) conducted constant-head hydraulic conductivity tests on 4-in.
(100-mm) diameter samples of Fibrosorb®. Four samples were tested:
Sample 1: Fibersorb® alone.
Sample 2: Fibersorb® overlying a 40 mil (1 mm) HOPE membrane that had a 1/8
in. (3 mm) diameter hole punched in it.
63
-------
Sample3: No Fibersorb®; a 40 mil (1 mm) HOPE membrane with a 1/8-in. {3
mm) diameter hole (the membrane placed between two porous discs and
tested in the permeameter).
Sample 4: A 2-in. (50-mm) wide strip of Fibersorb® was placed between two
sheets of 40 mil (1 mm) HOPE sheets to simulate use as a field seaming
material.
The samples were back-pressure saturated and permeated at an effective confining stress of
0.15 kg/cm2 (15 kPa) with a head difference of 50 mm of water. Results are summarized in
Table 7.1.
Table 7.1. Summary of Results of Hydraulic Conductivity Tests on Fibersorb® (from STS
Consultants, 1990).
Sample Test Conditions Hydraulic Conductivity (cm/s)
1 Fibersorb® 3 x 10'8
2 Fibersorb® with Defective HOPE 6 x 1Q'10
3 Defective HOPE 1 x 10'6
4 Fibersorb® between HOPE Overlap 5 x 1Q'9
64
-------
Section 8.0
Equivalency
The main function of low-permeability, compacted soil is either to restrict infiltration
of water into buried waste (in cover systems) or to limit seepage of leachate from the waste (in
liner systems). Other objectives may include enhancement of the efficiency of an overlying
drainage layer, development of composite action with a flexible membrane liner (FML),
adsorption and attenuation of leachate, restriction or gas migration, and others. In the case of a
cover system, compacted soil must also have the ability to withstand subsidence and must be
repairable if damaged by freezing, desiccation, or burrowing animals. For liner systems, the
ability of the liner to withstand chemical degradation from the liquids to be contained. In
addition, low-permeability compacted soil must have adequate shear strength to support itself
on slopes and to support the weight of overlying materials or equipment.
An alternative barrier material, in order to be fully equivalent to a compacted soil
layer, must serve the same functions as compacted soil. Due to inherent differences in the
composition and construction of compacted soil and alternative barriers, the two categories of
materials can never be "equivalent" in all possible respects. For example, compacted soil is
usually from 2 to 5 ft (0.6 to 1.5 m) thick whereas the alternative barriers discussed in this
report are typically no thicker than approximately 1/2 in. (13 mm). Due to differences in
thickness, the alternative barrier material is bound to be more vulnerable to puncture than the
much thicker layer of compacted soil.
Fundamental differences between compacted, low-permeability soil and the alternative
barriers discussed in this report create inevitable differences in hydraulic properties,
attenuation capacity, time of travel of chemicals, strength, desiccation resistance, freeze/thaw
resistance, reaction to settlement, ease of repair, and useful life. Table 8.1 presents a
qualitative list addressing the differences between compacted soil and alternative barrier
materials.
When the potential use of an alternative barrier is evaluated for a particular project,
the critical functions of the barrier should be identified. "Equivalency" should be evaluated on
the basis of the critical parameters and not necessarily upon all potential areas of comparison.
Further, it should be kept in mind that all liner materials have inherent advantages and
disadvantages -- no one type of liner (including low-permeability, compacted soil) is a
panacea. Some of the potential advantages of alternative barriers over low-permeability,
compacted soil are as follows:
65
-------
Table 8.1 Comparison of Differences in Alternative Barrier Materials
Compacted Soil
Thick (2 ft - 5 ft, or 0.6 - 1.5 m)
Field Constructed
Hard to Build Correctly
Impossible to Puncture
Constructed with Heavy Equipment
Often Requires Test Pad at Each Site
Site-Specific Data on Soils Needed
Large Leacnate-Attenuation Capacity
Relatively Long Containment Time
Large Thickness Takes Up Space
Cost Is Highly Variable
Soil Has Low Tensile Strength
Can Desiccate and Crack
Difficult to Repair
Vulnerable to Freeze/Thaw
Damage
Performance Is Highly Dependent
Upon Quality of Construction
Slow Construction
Alternative Barrier Materials
Thin (< 10 mm)
Manufactured
Easy to Build (Unroll & Place)
Possible to Damage and Puncture
Light Construction Equip. Can Be Used
Repeated Field Testing Not Needed
Manufactured Product; Data Available
Small Leachate-Attenuation Capacity
Shorter Containment Time
Little Space Is Taken
More Predictable Cost
Higher Tensile Qtrength
Can't Crack Unin Wetted
(after Construction)
Not Difficult to Repair
Probably Less Vulnerable to
Freeze/Thaw Damage
Hydraulic Properties Are Less
Sensitive to Construction Variabilities
Much Faster Construction
66
-------
The installation of alternative barriers proceeds rapidly and relatively simply
(construction of low-permeability, compacted soil is slower and requires a
much higher level sophistication in construction technique);
Because of the higher level of sophistication required for proper construction
and protection of low-permeability, compacted soil liners, the alternative
barriers may provide a more predictable end-product for situations in which
the quality of construction of a compacted soil liner cannot be assured;
Alternative barriers, which cost approximately $0.50 to $1.00 per square
foot ($5.50 to $11 per square meter) installed, are often less expensive than
compacted soil and can be installed at a more predictable cost than compacted
soil liners;
Alternative barriers occupy much less volume than compacted soil, which has
three ramifications: (1) more space is available in landfills for waste with the
thin, alternative barrier; (2) fewer truckloads of delivered material are
needed for the alternative barrier compared to compacted soil liners, which can
have important implications for transportation impacts when soil must be
obtained from off-site; and (3) because the alternative barrier weighs less
than the thicker compacted soil, less settlement of underlying waste (for cover
applications) would result with alternative barriers;
Alternative barriers can be installed with light-weight equipment, which is
particularly advantageous for placing liners on top of geosynthetic components,
e.g., a primary liner placed on top of a secondary leachate collection and
removal system;
Once an alternative barrier material is thoroughly characterized and field
tested, there should be no need to retest it unless the materials or installation
procedures change;
Some alternative barrier materials possess unique self-healing characteristics
derived from the expansive nature of bentonite.
The alternative barriers are not without caveats. Some of the potential disadvantages of
alternative barriers include the following:
67
-------
ijrf.^... -*.M*rt*WI
There is a general lack of data and independent research on the alternative
barrier materials;
Field experience is very limited for most of the alternative barrier materials,
and field performance data is virtually nonexistent;
Because the alternative barriers are thin, they are vulnerable to damage from
puncture, e.g., from traffic or construction equipment such as bulldozers, over
unprotected or improperly protected sections or during placement of cover
materials;
Sodium bentonite is more vulnerable to adverse chemical reactions from
leachate than the clay minerals found in most compacted soil liners;
The effects of settlement of underlying waste upon the hydraulic integrity of the
materials 'has not been evaluated;
The effects of cyclic wetting and drying of the materials upon bulk shrinkage
has not been adequately investigated;
Characterization of performance of overlapped seams under actual field
conditions is incomplete;
The low shear strength of bentonite raises questions about the stability of
alternative barrier materials containing bentonite when such materials are
placed on slopes.
One of the areas of application that was relatively uncontroversial was the use of
alternative barrier materials as a back-up to a flexible membrane liner in the primarily liner
of a double liner system. The EPA does not require a clay liner in the uppermost liner for
doubly-lined, hazardous waste landfills; an alternative barrier used in this situation involves
placing an extra component beyond the minimum requirements.
68
-------
Section 9.0
Concerns
During a concluding open discussion session of the Workshop, attendees voiced their
concerns regarding the behavior of alternative barriers and discussed informational needs that
would provide a better understanding of the characteristics of these manufactured materials.
The major concerns, expressed in question form, were as follows:
• Concerning stability -- Should alternative barriers be used in landfill caps
having slopes equal to or greater that 10°? And if so, would reinforcing the
cover soil, e.g., with a geogrid, and construction of an anchor eliminate the
instability problem? Do engineers have sufficient experience with and
knowledge of these materials to allow building on slopes with a high level of
confidence?
* Concerning temporary versus permanent use in a cap -- Alternative materials
should be seriously considered as a temporary cap for some RCRA or CERCLA
sites for which settlement that would damage a final cover is anticipated. How
would the alternative barrier material react to significant settlement? Although
the alternative barrier would appear to be easy to repair, are there practical
problems in repairing the materials that have not been anticipated?
* Concerning application in dry climates — Compacted soils have limited self-
healing capability, especially at low stress, and are vulnerable to damage from
desiccation after they are constructed. Alternative barrier materials are less
vulnerable to damage from desiccation after they are installed because they are
installed dry. Should alternative barriers be given stronger consideration for
applications in arid regions? If so, are there other problems with use of
alternative barriers in arid regions, such as bulk shrinkage upon drying, that
might prove to be significant?
» Concerning installation -- What happens when it rains during construction?
What happens to hydraulic conductivity if the material is wetted before
overburden is placed? How much overburden is needed to form an adequate
seam? What if the alternative barrier is placed on a small pebble; will the
bentonite be pushed aside and cause and increase in permeability? A great deal of
69
-------
care would appear to be necessary to install the alternative barrier material
correctly -- is it reasonable to assume that the necessary degree of care will be
exercised in the installation of the alternative barrier material?
Concerning application in the field -- The following concerns were expressed.
There is a strong need to see real performance; laboratory data alone are not
enough. Possibly a slow approach to the use of alternative barrier materials
would be wise. The least controversial applications of alternative barrier
materials appear to be landfill covers on reasonably flat surfaces, primary
lining systems for which there is a conventional FML/compacted clay secondary
liner located beneath the primary liner, and liner or cover systems in arid
regions, in these situations, there is less doubt about performance, less risk
involved, and the performance may be easier to assess.
Concerning field performance -- Routine methods to monitor actual performance
of field installations is badly needed. Installation of large {e.g., 2 m) diameter
collection lysimeters underneath these barrier materials is feasible and is
encouraged to provide a credible base of data on field performance.
70
-------
Section 10.0
Informational Needs
The workshop was concluded by compiling a list of issues for which more information is
needed. The "needs" were established with the goal of generating the data that design engineers,
owner/operators, and regulatory personnel require to have a high level of confidence that
alternative barrier materials will provide the required environmental protection functions in
waste management applications. The following is a condensed version of this list.
1. Shear Strength
a. Interfacial friction with other liner/cover components
b. Long term performance
c. Water diffusion effects; point wetting
d. Standardized testing procedures
e. Laboratory versus field scale
2. Hydraulic Properties
a. Hydraulic conductivity
b. Attenuation capacity
c. Hydration with water versus leachate
d. Composite action; does a composite seal form?
e. Migration of bentonite
f. Laboratory versus field scale
3. Environmental Effects
a. Freeze/thaw resistance
b. Desiccation resistance
c. Effects of settlement
d. Self healing capabilities
e. Effects of rock beneath alternative barrier
f. Laboratory versus field scale
4. Seams
a. Hydraulic properties
b. Strength
c. Effects of settlement
71
-------
mmnHir'Tnitir '••*•'
d. Wrinkled seam (wrinkle of liner above or below and of alternative barrier
itself)
e. Laboratory versus field scale
5. Quality Assurance/Quality Control
a. Manufacture
b. Transportation
c. Installation
6. Applications
a. Caps versus liners
b. Humid versus arid regions
c. Compressible and incompressible waste
7. Thermal Effects
a. Differential expansion of alternative barrier with other liner materials,
especially with HDPE/bentonite composite, that could cause wrinkles or
delamination of materials
b. Shrinkage of materials upon drying, causing a reduction in overlap width
8. Mechanical Properties
9. Comparison with Compacted Soil
1 0. Useful Life; Aging.
The list was prioritized in the following order:
1. Shear strength
2. Hydraulic properties
3. Seams
4. Useful life.
More information about these research needs, plus other issues not listed above, is
expected to become available over the next few months and years. Individuals with information
are encouraged to pass that information along to David E. Daniel, University of Texas,
Department of Civil Engineering, Austin, TX 78712, or to Walter E. Grube, Jr., U. S.
Environmental Protection Agency, Risk Reduction Engineering Laboratory, Cincinnati, OH
45268. Of particular interest are unpublished data, for example, developed for a particular
project such as a DOE cover project, that might otherwise not be widely disseminated.
72
-------
Many individuals attending the Workshop expressed a desire to hold a similar Workshop
in 1 to 2 years to present and to discuss new information. If a significant base of new data is
developed, the new information would likely be the focal point of discussions in the next
Workshop.
73
-------
Section 11.0
List of References
Benson, C. B., and Daniel, D. E. (1990), "Influence of Clods on Hydraulic Conductivity of
Compacted Clay," Journal of Geotechnical Engineering. Vol 116, No. 8, pp. 1231-1248.
Broderick, G. P., and Daniel, D. E. (1990), "Stabilizing Compacted Clay Against Chemical
Attack," Journal of Geotechnical Engineering. Vol. 116, No. 10, pp. 1549-1567.
Brown, K. W.f Thomas, J. C., and Green, J. W. (1984), "Permeability of Compacted Soils to
Solvents Mixtures and Petroleum Products," Proceedings. Tenth Annual Research
Symposium on Land Disposal of Hazardous Waste, U. S. EPA, Cincinnati, Ohio, EPA-600/9-
84-007, pp. 124-137.
Chen-Northern , Inc. (1988), Untitled Letter Report with Results of Laboratory Tests on
Clayrnax®, Denver, Colorado, December 31.
Daniel, D. E. (1987), "Earthen Liners for Land Disposal Facilities," in Geotechnical Practice
for. Waste, Disposal '87. R. D. Woods (ed.), American Society of Civil Engineers, New York,
pp. 21-39.
Daniel, D. E. (1989), "In Situ Hydraulic Conductivity Tests for Compacted Clay," Journal of
Geotechnipal Engineering, Vol. 115, No. 9, pp. 1205-1226.
Daniel, D. E. (1990), "Summary Review of Construction Quality Control for Compacted Soil
Liners," Proceedings. Symposium on Regulation, Performance, Construction, Operation of
Waste Containment Systems, San Francisco, November 6-7, American Society of Civil
Engineers, New York (in press).
Geoservices, Inc. (1988a), "Interim Test Results - Claymax Liner, Freeze-Thaw Hydraulic
Conductivity Tests," Report to James Clem Corporation, November 11, 3 p.
74
-------
Geoservices, Inc. (1988b), "Results of Saline Solution Hydraulic Conductivity Test, Claymax
CR Liner," Report to James Clem Corporation, Norcross Georgia, November 29, 2 p.
Geoservices, Inc. (1989a), "Report on Direct Shear Testing of Selected Claymax CL Interfaces,"
Report to James Clem Corporation, Norcross, Georgia, March 10, 6 p.
Geoservices, Inc. (1989b) "Hydraulic Conductivity of Claymax CR in a Marine Environment,"
Report to James Clem Corporation, Technical Note No. 2, Norcross, Georgia, March, 10 p.
Geoservices, Inc. (1989c), "Freeze-Thaw Effects on Claymax Liner Systems," Report to James
Clem Corporation, Technical Note No. 3, Norcross, Georgia, March, 9 p.
Geoservices, Inc. (1989d), "Final Report, Claymax/Fuel Compatibility Tests," Report to James
Clem Corporation, Norcross, Georgia, July 7, 14 p.
Geoservices, Inc. (1989e), "Report of Moisture Retention Tests, Claymax CR," Report to James
Clem Corp., Norcross, Georgia, July 14, 14 p.
Goldman, L. G., et ai. (1988), "Design, Construction, and Evaluation of Clay Liners for Waste
Management Facilities," U. S. EPA, Washington, DC, EPA/530/SW-86-007F.
Gordon, M. E., Huebner, P. M., and T. J. Miazga (1989), "Hydraulic Conductivity of Three
Landfill Clay Liners," Journal of Geotechnical Engineering. Vol. 115, No. 8, pp. 1148-
1162.
Herrmann, J. G., and Elsbury, B. R. (1987), "Influential Factors in Soil Liner Construction for
Waste Disposal Facilities," in Geotechnical Practice for Waste Disposal '87. R. D. Woods
(ed.), American Society of Civil Engineers, New York, pp. 522-536.
J & L Testing Company, Inc. (1990a), "Physical Properties Test Results, Bentonite NW/SS/W
Composite," Report to American Colloid Company, Canonsburg, Pennsylvania, May 30, 17p.
J & L Testing Company, Inc. (1990b), "Hydraulic Conductivity Tests, Bentomat NW/Cs/W,
Bentomat NW/SS/W," Report to American Colloid Company, Canonsburg, Pennsylania, July
5.
75
-------
Lutz, D.P, (1990), Barrier Equivalency of Liner and Cap Materials, M. S. Thesis, The
University of Texas, Austin, Texas, 90 p.
Mitchell, J, K., Hooper, D. R., and Campanella, R. G. (1965), "Permeability of Compacted
Clay," Journal, of the gpjl Mechanics and FounglatiQQS Division. ASCE, Vol. 91, No. SM4, pp.
41-65.
Pittsburgh Testing Laboratory (1985), "Permeability Test of Paraseal Water Proofing
Membrane," Report to Paramount Technical Products, Inc., Salt Lake City, November 27, 2
P-
STS Consultants Ltd. (1984), "Permeability Testing on Enviromat Material Seam
Applications," Letter Report to Ciem Environmental Corporation, Northbrook, Illinois, July
16, 3p.
STS Consultants Ltd. (1988a), "Hydraulic Conductivity Testing of Clayrnax Liner Material after
Subjecting the Specimen to 15% Elongation," Letter Report to Clem Environmental
Corporation, Northbrook, Illinois, April 25, 4 p.
STS Consultants Ltd. (1988b), "Hydraulic Conductivity Determination of Claymax Utilizing
Sewage Leachate as a Permeant and Hydration Medium," Report to Clem Environmental
Corporation, Northbrook, Illinois, 4 p.
STS Consultants Ltd. (1988c), "Hydraulic Conductivity Determination of Claymax Material
Utilizing Paper Sludge at the Permeant," Report to Clem Environmental Corporation,
Northbrook, Illinois, July 7, 4 p.
STS Consultants Ltd. (1989a), "Hydraulic Conductivity and Compatibility Testing of Claymax,
Baltimore County Landfill Project, Townson, Maryland," Report to Clem Environmental
Corporation, May 11, 6 p.
STS Consultants Ltd. (1989b), "Hydraulic Conductivity Testing, RCRA Sheiton Ash Mono Cell
Landfill," Report to Clem Environmental Corporation, May 11, 6 p.
76
-------
STS Consultant Ltd. (1990), "Hydraulic Conductivity Testing of Arco Fibersorb SA-7000,"
Report to ARCE Chemical Company, Northbrook, Illinois, April 18, 3p.
Scheu, C., Johannben, and Saathoff, F. (1990), "Non-Woven Bentonite Fabrics - A New Fiber
Reinforced Mineral Liner System," Geotextiles. Geomembranes and Related Products. Den
Hoet (ed), Balkema, Rollerdam, pp. 467-472..
Schubert, W.R. (1987), "Bentonite Matting in Composite Lining Systems," in Geotechnical
Practice for Waste Disposal '87. R. D. Woods (ed.), American Society of Civil Engineers
Conference , New York, pp. 784-796..
Shan, H.Y. (1990), "Laboratory Tests on Bentonitic Blanket/ M. S. Thesis, University of
Texas, Austin, Texas, 84 p.
Twin City Testing Corporation (1986), "Laboratory Permeability Testing of Membrane with
Bentonite Backing Seal," Report to Paramount Technical Products, St. Paul, Minnesota, May
30, 6 p.
U. S. Environmental Protection Agency (1985), "Draft Minimum Technology Guidance on
Double Liner Systems for Landfills and Surface Impoundments - Design, Construction, and
Operation," Office of Solid Waste and Emergency Response, Washington, DC, EPA/530-SW-
014, 70 p.
U. S, Environmental Protection Agency (1986), "Technical Guidance Document: Construction
Quality Assurance for Hazardous Waste Land Disposal Facilities," Office of Solid Waste and
Emergency Response, Washington, DC, EPA/53Q-SW-86-031, 88 p.
U, S. Environmental Protection Agency (1989), "Technical Guidance Document: Final Covers on
Hazardous Waste Landfills and Surface Impoundments," Office of Solid Waste and Emergency
Response, Washington, DC, EPA/530-SW-89-047.39 p.
Zimmie, T.F. and Plante, C.L. (1990), "The Effect of Freeze/Thaw Cycles on the Permeability of
a Fine Grained Soil," Proceedings. Mid-Atlantic Industrial Waste Conference, Philadelphia,
July (in press).
77
-------
APPENDIX
LIST OF PARTICIPANTS
78
-------
Registrants — EPA Workshop
June 6-7, 1990 Cincinnati, Ohio
Rodney Aldrich
NY State DEC
50 Wolf Rd.
Albany, NY 12233
Richard Andersen
Texas Water Commission
P.O. Box 13087
Austin, TX 78711
Mary F. Beck
U.S, EPA Region 111
841 Chestnut St.
Pnilsdelphia, PA 19107
Craig H, Benson
Dept. of Civil & Env. Engr.
2214 Engineering Bldg.
University of Wisconsin
Madison, Wl 53706
Jane Bolton
Army Corps of Engineers
CEMRO-ED-GF
215 No. 17th St.
Omaha, NE 68134
John Boschuk, Jr.
J&L Testing Co.
938 S. Central Ave.
Canonsburg, PA 15317
John Bove
Westinghouse Corp.
11785 Highway Dr. #100
Cincinnati, OH 45241
John Bowders
Dept. of Civil Engineering
653D Engr. Sciences Bldg.
West Virginia University
Morgantown, WV 26506-
6101
Karen Brady
Tennessee Valley Authority
NFE-1AF 112-G
Muscle Shoals, AL 35660
Ed Brdicka
Ohio EPA
1800 Watermark Dr.
P.O. Box 1049
Columbus, OH 43266-1049
Mark Cadwallader
Gundle Lining Co.
19103 Gundle Rd.
Houston, TX 77073
David Carson
U.S. EPA
26 West ML King Dr.
Cincinnati, OH 45268
Eric Chiado
Almes & Associates
RD #1, Box 520
Pleasant Valley Rd.
Trafford, PA 15085
Yoon-Jean Choi
U.S. EPA, Region 1
Waste Mgmt. Division
JFK Federal Bldg.
Boston, MA 02203
Glenn N. Coffman
Law Environmental Inc.
112 Town Park Dr.
Kennesaw, GA 30144
Jack Conner
Meredith Brothers
6013 Tulip Hill Rd.
Columbus, OH 43235
Larry C. Cox
Automated Sciences Group
800 Oak Ridge Turnpike
#C102
Oak Ridge, TN 37830
David E. Daniel
Dept. of Civil Engineering
ECJ 9.102E
The University of Texas
Austin, TX 78712
Dennis L. Datin
OK State Dept. of Health
1000 N.E. Tenth St.
Oklahoma City, OK 73152
Annette DeHavilland
OhbEPA
P.O. Box 2198
Columbus, OH 43266-0149
Gary S. Deutshman
OhbEPA
1035 Deulac Grove Dr.
Bowling Green, OH 43402
Michael Dewsbury
ARCO Chemical Co.
3801 West Chester Pike
Newtown Square, PA 19073
Ed Doyle
Waste Mgmt. of N. America
3003 Butterfield Rd.
Oak Brook, IL 60521
R. Jeffrey Dunn
Kleinfelder, Inc.
2121 N. California Blvd.
#570
Walnut Creek, CA 94596
Ron Ebelhar
Westinghouse Environmental
and Geotechnteal Services
11785 Highway Dr. #100
Cincinnati, OH 45241
David Eberly
U.S. EPA-OSW (OS-343)
401 M St., S.W.
Washington, D.C. 20460
Marcia Ellis
NY State DEC
50 Wolf Rd. #230
Albany, NY 12233
79
-------
EPA Registrants - Continued
Page 2
Allan Erickson
CH2M Hill
310 W. Wisconsin Ave. #700
Milwaukee, Wl 53201
Paula Estornell
Dept. of Civil Engineering
ECJ 9.102E
The University of Texas
Austin, TX 78712
Douglas N. Excell
American Colloid
1500 W. Shure Dr.
Arlington Heights, IL 60004
Heidi Facklam
U.S. Army Corps of Engineers
Missouri River Division
P.O. Box103DTS
Omaha, NE 68101-0103
A. A. Freitag
Bechtel Savannah River Inc.
802 E. Martintown Rd.
N. Augusta, SC 29841
John Gardner
Almes & Associates, Inc.
RD #1, Box 520
Pleasant Valley Rd.
Trafford, PA 15085
Sal Gazioglu
Dames & Moore
4141 Blue Lake, Suite 190
Dallas, TX 75244
George F. Gilbert
Division of Waste Management
18 Reilly Rd.
Frankfort, KY 40601
George Gtos
Bechtel Savannah River
Project
802 E. Martintown Rd.
BTC416
N. Augusta, SC 29841
G. Hossein Golshir
Westinghouse Corp.
802 E. Martintown Rd.
North Augusta, SC 29841
Walter E. Grube, Jr.
U.S. EPA Hazardous Waste
Engineering Laboratory
26 West ML King Dr.
Cincinnati, OH 45268
J. Scott Heisey
American Electric Power SC
1 Riverside Plaza
Cincinnati, OH 43215
Ron Hill
U.S. EPA
26 West ML King Dr.
Cincinnati, OH 45268
Michael Kostanian
Bechtel Environmental
50 Beale St.
San Francisco, CA 94119-
3965
Andrew Leung
TAMS Consultants Inc.
300 Broodacres Dr.
Bloomfield, NJ 07003
James D. Liner
Milwhite, Inc.
P.O. Box 15038
Houston, TX 77220-5038
J.T. Massey-Norton
American Electric Power SC
1 Riverside Plaza
Columbus, OH 43215
John D. Holm Tony Maxson
Army Corps of Engineers Chemical Waste Mgmt.
601 E. 12th St. 3001 Butterfield Rd.
Kansas City, MO 64106-2896 Oak Brook, IL 60521
Janet M. Houthoofd
U.S. EPA
26 West ML King Dr.
Cincinnati, OH 45268
Jon Hutchings
Minnesota Pollution Control
520 Lafayette Rd.
St. Paul, MN 55155
Jim Klang
Minnesota Pollution Control
520 Lafayette Rd.
St. Paul, MN 55155
Robert M. Koerner
Drexel University
Geosynthetic Research
Institute
West Wing- Rush Bldg.
Philadelphia, PA 19104
John M. McBee
R.F. Weston, Inc.
5301 Central Ave., N.E. #100
Albuquerque, NM 87108
Patrick McGroarty
Paramount Technical Products
2600 Paramount Dr.
Spearfish, SD 57783
John E. Moylan
Army Corps of Engineers
601 E. 12th St.
Kansas City, MO 64106-2896
Durge S. Nagda
Division of Waste Management
18 Reilly Rd.
Frankfort, KY 40601
80
-------
EPA Registrants - Continued
Page 3
Jeff W. Newell
Charles T. Main, Inc.
4701 Hedgemore Dr.
Charlotte, NC 28209
Gary Oberholtzer
Bechtel Corp.
P.O. Box 3965
San Francisco, CA 94119
Jim Olsta
American Colloid
1500 W. Shure Dr.
Arlington Heights, IL 60004
Mary Osika
Ohio EPA
P.O. Box 1049
1800 Watermark Dr.
Columbus, OH 43266-0149
Donald Perander
ARMCO, Inc.
P.O. Box 600
Middletown, OH 45043
Mark Phifer
Westinghouse Corp.
802 E. Martintown Rd.
North Augusta, SC 29841
Bob Randolph
Soil Stablization Products
P.O. Box 2779
Merced, CA 95344
Gregory Richardson
Westinghouse Corp.
P.O. Box 58069
Raleigh, NC 27658
Charles Rivette
Browning-Ferris Industries
757 N. Eldridge Rd.
Houston, TX 77079
Nancy Roberts
ERCE
P.O. Box 22879
Knoxville, TN 37993
Mary Rogers
Chambers Development Co.
10700 Frandstown Rd.
Pittsburgh, PA 15235
L.H. Schroeder
Bechtel Savannah River Inc.
802 E. Martintown Rd.
N. Augusta, SC 29841
Dana Sheets
American Electric Power
1 Riverside Plaza
Columbus, OH 43216
Martin Simpson
James Clem Corp.
444 N. Michigan, #1610
Chicago, IL 60611
William Simpson
James Clem Corp.
444 N. Michigan, #1610
Chicago, IL 60611
Kenneth Skahn
U.S. EPA (OS-220)
401 M St., S.W.
Washington, D.C. 20460
Thomas Stam
American Colloid
1500 W. Shure Dr.
Arlington Heights, IL 60004
Lindsay Taliaferro III
Ohio EPA
1800 Watermark Dr.
P.O. Box 1049
Columbus, OH 43266-0149
Steve Walker
Polyfelt, Inc.
1000 Abernathy Rd.
Atlanta, GA 30338
Bryan Wilson
American Electric Power SC
1 Riverside Plaza
Cincinnati, OH 43215
Thomas Zimmie
Rensselaer Polytechnic Inst.
Dept. of Civil Engineering
Troy, NY 12180
81
-------
•d
g |
° fc '§ CD
£> CD CD >
^ £ ^ Fi
QJ 2 c/3 U
VH Ci ffD
^ -F-I > CD
e<3 "2 S
M-H CD 51
—« Q,
«5
JH |H
CD
J-H
CD U
00,
o
•d
CD
O O
* *»
*H (H
13 *H
1*** ^^
K^ ^J
**^ «TH
Reproduced by NTIS
National Technical Information Service
U.S. Department of Commerce
Springfield, VA 22161
This report was printed specifically for your
order from our collection of more than 2 million
technical reports.
For economy and efficiency, NTIS does not maintain stock of its vast
collection of technical reports. Rather, most documents are printed for
each order. Your copy is the best possible reproduction available from
our master archive. If you have any questions concerning this document
or any order you placed with NTIS, please call our Customer Services
Department at (703)487-4660.
Always think of NTIS when you want:
• Access to the technical, scientific, and engineering results generated
by the ongoing multibillion dollar R&D program of the U.S. Government.
• R&D results from Japan, West Germany, Great Britain, and some 20
other countries, most of it reported in English.
NTIS also operates two centers that can provide you with valuable
information:
• The Federal Computer Products Center - offers software and
datafiles produced by Federal agencies.
• The Center for the Utilization of Federal Technology - gives you
access to the best of Federal technologies and laboratory resources.
For more information about NTIS, send for our FREE NTIS Products
and Services Catalog which describes how you can access this U.S. and
foreign Government technology. Call (703)487-4650 or send this
sheet to NTIS, U.S. Department of Commerce, Springfield, VA 22161.
Ask for catalog, PR-827.
Address.
- Your Source to U.S. and Foreign Government
Research and Technology.
------- |