vvEPA
United States
Environmental Protection
Agency
Horizontal Configuration of the
Lasagna™ Treatment
Technology
User Guide
-------
-------
EPA/600/R-02/033
June 2002
Horizontal Configuration of the
Lasagna™ Treatment Technology
User Guide
by
Mike H. Roulier and Mark C. Kemper
Land Remediation and Pollution Control Division
National Risk Management Research Laboratory
Cincinnati, Ohio 45268
and
Phillip R. Cluxton
Cluxton Instruments, Inc.
Martinsville, Ohio 45146
NATIONAL RISK MANAGEMENT RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free.
-------
Notice
This document has been reviewed in accordance with U.S. Environmental Protection Agency policy and
approved for publication as an EPA document. Mention of trade names or commercial products does
not constitute an endorsement or recommendation for use.
-------
Foreword
The U.S. 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. To meet this mandate, EPA's research program
is providing data and technical support for solving environmental problems today and building a
science knowledge base necessary to manage our ecological resources wisely, understand how
pollutants affect our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for investigation of
technological and management approaches for preventing and reducing risks from pollution that
threatens human health and the environment. The focus of the Laboratory's research program is on
methods and their cost-effectiveness for prevention and control of pollution to air, land, water, and
subsurface resources; protection of water quality in public water systems; remediation of
contaminated sites, sediments and ground water; prevention and control of indoor air pollution; and
restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's
research provides solutions to environmental problems by: developing and promoting technologies
that protect and improve the environment; advancing scientific and engineering information to support
regulatory and policy decisions; and providing the technical support and information transfer to ensure
implementation of environmental regulations and strategies at the national, state, and community
levels.
This publication has been produced as part of the Laboratory's strategic long-term research plan. It
is published and made available by EPA's Office of Research and Development to assist the user
community and to link researchers with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
in
-------
Abstract
This report is a user's guide that discusses the technology and operations unique to the installation
and operation of the horizontal configuration of the Lasagna™ integrated soil remediation technology.
This technology, called Lasagna™ because of the layers of electrodes and treatment zones, has been
developed to combine electrokinetics with treatment zones for use in low-permeability soils where
rates of hydraulic and electrokinetic transport are too low to be useful for remediation of contaminants.
The technology was developed by two groups, one involving industrial partners and the U.S.
Department of Energy and another involving U.S. Environmental Protection Agency and the University
of Cincinnati, who each pursued different electrode geometries. This report deals with the horizontal
configuration where electrodes and treatment zones are installed by hydraulic fracturing in soil. This
report covers a period from October 1993 to August 2001 and work was completed September 29,
2001.
IV
-------
Contents
Section
Foreword iii
Abstract iv
Figures vi
Acronyms and Abbreviations ... — vii
1.0 Introduction 1
1.1 Technology Description 1
1.2 Technology status 2
1.3 Scope of this report 4
2.0 Installing Electrodes and Treatment Zones 5
- 2.1 Feasibility testing 5
2.2 Pre-installation soil sampling 5
2.3 Access wells for electrodes 5
2.4 Electrode material 7
2.5 Electrode spacing 8
2.6 Treatment zones 10
3. Fluid Management in Electrodes 12
4. Electrical Power Management 14
4.1 Power supply 14
4.2 Temperature protection 15
4.3 Connections 15
5. Process Monitoring , 18
6. Technology Niche 21
7. References .23
-------
Figures
Figure Paae
1-1 Lasagna™ vertical configuration • 2
1-2 Horizontal configuration Lasagna™ cell 3
2-1 Plan view of hydraulic fracture propagating from injection point (I.P.)
toward casing; shadow (unfractured area) develops behind casing 7
2-2 Electrode well dimensions 9
2-3 Connection to graphite electrode fracture. Connection and PVC casing 9
drawn to scale
3-1 Simultaneous increases in liquid depth and temperature in a cathode well 13
5-1 Linear voltage profile in soil • 19
5-2 Voltage profile in soil 19
5-3 Voltage probe 19
VI
-------
Acronyms and Abbreviations
Adc
AFB
AISI
ANGB
bpf
C
Center Hill
cm/sec
cm
cm2
CPVC
CRADA
DC
DOE
EK
EPA
IP.
kPa
kW.
m
NPT
POTW
PVC
PVDF
RCRA
ROD
S/m
SCR
SIMCO
TCE
TM
UC
UIC
V/m
Vdc
Amperes, direct current
Air Force Base
American Iron and Steel Institute
Air National Guard Base
Blows per foot
Centigrade
U.S. EPA Center Hill Facility
Centimeters per second
Centimeter
Square centimeter
Chlorinated polyvinyl chloride
Cooperative research and development agreement
Direct current
U.S. Department of Energy
Electrokinetics
U.S. Environmental Protection Agency
Injection point
Kilopascal
Kilowatt
Meter
National pipe thread
Publicly owned treatment works
Polyvinyl chloride
Polyvinylenefloride
Resource Conservation and Recovery Act
Record of decision
Siemen per meter
Silicon-controlled rectifier
Southern Iowa Manufacturing Company
Trichloroethylene
Trademark
University of Cincinnati
Underground Injection Control
Volts per meter
Volts, direct current
VII
-------
-------
1.0 Introduction
1.1 Technology Description
This report discusses the hardware and operations that are unique to the horizontal configuration of the Lasagna™
technology for in-place treatment of contaminated soil. The general topic of electrokinetic remediation is discussed in other
publications. The U.S. Environmental Protection Agency's National Risk Management Research Laboratory (referred to
here as EPA) entered into a Cooperative Research and Development Agreement (CRADA) with a Consortium of DuPont,
General Electric, and Monsanto companies for this project. The goal was to develop and field test an innovative
combination of technologies proposed by the Consortium for in-situ clean up of soils contaminated by hazardous waste
(1). Investigators at the University of Cincinnati (UC) collaborated on the effort through the sponsorship of U.S. EPA. The
U.S. Air Force is also collaborating with EPA on a large scale demonstration of the horizontal configuration of the
technology.
Called Lasagna™ because of its layered configuration, the technology compensates for the low rate of movement
of fluids and contaminants in response to an electric field by inserting closely-spaced treatment zones between electrodes.
This shortens treatment time because liquids and contaminants have to move only a short distance before reaching a
treatment zone. The treatment zones may use a variety of processes for trapping or treating contaminants in place and
the electrodes may be emplaced either vertically or horizontally by several different geotechnical processes. Monsanto
patented (2,3) the concept of placing treatment zones between sheet electrodes in subsurface soil and then using
electrokinetics (EK) to move water (by electroosmosis) and ions (by electromigration) into the treatment zones. When a
direct current electrical field is applied to soil it causes all water (and contained contaminants) and ions to move, thus
allowing mobilization from fine-grained soil deposits where hydraulic flow alone would not be effective. Placement of
treatment zones between sheet electrodes is the critical element in the Lasagna™ process.
Lasagna™ may be implemented in either a vertical or a horizontal configuration. The Consortium members
investigated the costs and basic factors of the integrated technology (4,5) and demonstrated it at full scale (6,7) in a
vertical configuration using a mandrel/tremie tube system to install electrodes and treatment zones downward from the
soil surface (Figure 1-1) for inducing horizontal electrokinetic flow.
EPA agreed to develop and test the horizontal configuration because it offered the advantages of application at
greater depths and the ability to use hydraulic flow in addition to electrokinetic flow. EPA also had experience with
hydraulic fracturing, a process that can be used in soil to install the horizontal layers of materials needed for the electrodes
and treatment zones. Hydraulic fracturing creates layers about 1 cm thick, 6 to 9 meters in diameter, and parallel to the
soil surface (Figure 1-2). The hydraulic fracturing process had been adapted (from oil field practices) at the EPA Center
Hill Facility for installation of horizontal layers of granular materials in certain types of soils (8-13).
A direct current power supply is connected to the electrode layers, the upper electrode being positive (anode) and
the lower electrode negative (cathode). When an electrical field is applied to soil, the water, ions, and other contaminants
in soil within the field move uniformly by electroosmosis or electromigration. This allows extraction of contaminants from
areas of soil that would be bypassed by hydraulic flow. Liquids are pumped from the cathode to maintain downward flow
and enhance EK transport of contaminants and water toward the cathode. Contaminants are degraded as they are
brought into contact with the treatment layers by electromigration (ion movement relative to bulk water) or electroosmosis
(advective movement of water and contaminants):
One of the likely effects of EK is to move contaminants out of low-permeability zones that would be bypassed by
hydraulic flow. See, for example, the laboratory studies with isolated lumps of clay in columns of sand (4). Once in higher
permeability zones, contaminants can be transported by the more rapid hydraulic flow. Volumetric fluxes from EK were
-------
To Rectifier
150Vd.c.
1.9cm Steel Rod
as Primary
Conductor
(6 at each
electrode)
::::::::!:!:;::;:: : ^*~ . ..
•'•'•t«!'!«!'!'!'!llt!!i i ^ I
ij;j!j!;jijj:!:;i Water Recycled by Gravity
iiiiijilijiiiiiji Thru PVC Pipe to Sump
:cc Iron Filings
Jlron Filings:^
+ Coke |g
n
Hi Water and TCE Move in Soil
: i!:!:;:!:; •:-•,'•. :
ij 3 Treatment Zones
"
Figure 1-1. Lasagna™ vertical configuration.
on the order of 10'8 crn/sec (4) in clay soils. The horizontal configuration can take advantage of both hydraulic and EK
transport processes because water is injected at the upper electrode (anode) and pumped out of the lower electrode
(cathode). In the vertical configuration, EK is the dominant transport process because both electrodes are at the same
elevation.
1.2 Technology status
The Consortium conducted small- and large-scale tests of the Lasagna™ vertical configuration at the DOE
Gaseous Diffusion Plant in Paducah, KY. Both tests removed substantial amounts of trichloroethylene (TCE) and
demonstrated that the technology was successful and economically competitive with other soil treatment technologies
(6,7). After intensive scrutiny by both DOE and the EPA Region IV in several technical meetings during and after the tests
at Paducah, EPA Region IV signed a Record of Decision (ROD) in August 1998, specifying the vertical configuration of
Lasagna™ as the backup technology for the remaining TCE-contaminated soil.
The parties involved in the work at the Paducah plant believe that the vertical configuration of Lasagna™ offers
significant cost advantages for cleanup of low permeability (clayey and silty) soils contaminated with weakly bound
(relatively soluble) compounds such as TCE (7). The technology has been proven at several scales in field tests and is
available for application from a commercial source, the Terran Corporation in Dayton, OH. It has support from a number
of groups who were involved in its development and has the potential to improve the speed and cost-effectiveness of future
cleanup activity.
-------
Liquids Pumped Out
Water Added
Electrode
f
Treatment
Zones
Electroosmptic
and Gravitational
Liquid Flow
Electrode
Figure 1-2. Horizontal configuration Lasagna™ cell.
The Lasagna™ horizontal configuration has been tested at small scale and a large-scale test is currently
underway. EPA scientists conducted laboratory and field studies and funded assistance agreements with the University
of Cincinnati to work on the emplacement and operation of horizontal electrodes and treatment zones using
hydraulicfra'cturing. The Ohio Air National Guard, the U.S. Air Force, and Chemical and Environmental Control Systems
(CECOS) International provided materials, infrastructure, and access to sites for.work on the horizontal configuration.
Under assistance agreements with EPA, the University of Cincinnati conducted cooperative laboratory work and
field studies in clean soils (14,15,16,17) at the EPA Center Hill Facility in Cincinnati, OH and a site near Williamsburg, OH
to develop and test the technology and procedures for using horizontal hydraulic fractures as electrodes. Horizontal
Lasagna™ cells with zero-valent iron treatment zones and biological treatment zones were also installed and operated,
first by UC and later by EPA, in soils contaminated with trichloroethylene (TCE) near Columbus, OH (18,19). This work
has developed and tested the use of hydraulic fracturing for installing the layers, making electrical and hydraulic contact
with the electrode layers, and methods for operating the Lasagna™ cells. The U.S. Air Force has provided a site and
facility support.for a large-scale test of the horizontal Lasagna™ process for treating TCE-contaminated soil at Offutt Air
Force Base (AFB) near Omaha, NE.. The feasibility tests were conducted at this site by UC; the cells were installed by
EPA and are currently in operation. The development history, technology and operational experience, and results from
pilot studies in contaminated soil were summarized and reviewed; these indicated that it was feasible to implement the
horizontal configuration in a field installation (20).
-------
The vertical configuration of Lasagna™ has been demonstrated on a larger scale than the horizontal configuration
and the package of skills needed for the vertical configuration are available from a single source (Terran Corp., Dayton,
OH). The feasibility of installing and operating a horizontal configuration has been well demonstrated (20); the cost-
effectiveness and expected contaminant removal efficiencies are not as well established. The unit operations for
implementing horizontal Lasagna™ are available from several different commercial sources; they would have to be
gathered into a team to operate an installation.
1.3 Scope of this report
This report discusses the technology and operations that are unique to the horizontal configuration of Lasagna™.
The general topics of electrokinetic remediation and hydraulic fracturing are covered elsewhere and are not discussed in
this report. See references 21 through 29 for information on electrokinetic remediation and references 8 through 13 for
information on hydraulic fracturing. Hydraulic fracturing has been used to create horizontal layers with a variety of
materials, and no major problems have been encountered installing horizontal electrodes or treatment zones. The major
problems are maintaining flow of liquids and electrical current in the electrode layers.
The remainder of the report covers five significant aspects of use of the horizontal configuration. Section 2
discusses the design, materials, and testing needed for installation of electrodes and treatment zones. The handling of
fluids that are added to or removed from electrodes is covered in Section 3. The next major topic, in Section 4, is the
conditioning, control, and supply of electrical power to the electrodes. Section 5 discusses monitoring of the process to
guide adjustments and maintenance. Section 6 provides some basis for selecting either the vertical or horizontal
configuration of Lasagna™ at a particular site.
Very little of the information in this report has been published elsewhere. It is a summary of the authors'
experiences with the design, installation, and operation of nine horizontal configuration cells at four different locations over
a period of seven years. These have ranged in size from small (10 ft. diameter) research installations to the full-scale (30
ft. diameter) cells that were tested at Offutt AFB, NE. A few data from these cells are presented in this 'report to
demonstrate how such data may be used to identify problems and to assess the progress of treatment. The data are
intentionally limited in scope and are not presented to test the hypothesis that the horizontal configuration can effectively
treat contaminated soil. That objective will be addressed in a later paper.
-------
2.0 Installing Electrodes and Treatment Zones
2.1 Feasibility testing
It is necessary to conduct a hydraulic fracturing feasibility test at a site being considered for a horizontal Lasagna™
installation because hydraulic fracturing will be used to install the electrode and treatment layers. The ideal approach is
to install sand-filled fractures at a location that is either near the area proposed for the horizontal Lasagna™ installation
or at least in soils that have the same physical characteristics such as state of stress, particle size distribution, and bearing
strength as measured by blow counts. Feasibility testing is most convenient in uncontaminated soil because soil samples
must be collected and examined. The information to be gathered during feasibility testing is:
depths at which horizontal fractures can be created
fracture inclination
fracture thickness
fracture shape (in plan view)
It must be possible to create horizontal hydraulic fractures at the depths of contamination if a horizontal
configuration is to be installed, at the site. If test fractures rise to the surface near the injection point (vent) they will be too
small and steeply dipping to be useful as electrodes or treatment zones. Fracture inclination (dip) is characteristic of a site
and is influenced by soil properties in ways that are not well understood. Fracture inclination is determined by measuring
the depth to the fracture at several distances from the injection point. The inclination of fractures is important because it
affects the ability to control the spacing between electrode fractures. See Section 2.5 below. The thickness of a fracture
is not usually an issue for graphite-filled electrode fractures. Relatively thin (0.5 to 1.0 cm) fractures seem capable of
satisfactorily transmitting liquids and electrical power. Fracture thickness may be a critical issue for treatment zones if
these need to contain some minimum mass of material. See the Section 2.6 (below) on treatment zones.
Hydraulic fractures are rarely symmetrical. The ratio of the longest dimension to the shortest may range from 1.1
to 1.8, with 1.2 being the most common value. If test fractures are highly elongated, the shape will be a consideration in
determining the number of cells that will provide coverage of the contaminated area. Fracture shape is inferred from
measurements of the changes in ground surface elevation after a fracture is installed. In most soils the surface uplift will
approximate the thickness of the,subsurface fractures with the highest point some distance from the injection point.
2.2 Pre-installation soil sampling
Care must be taken during any pre-installation soil sampling to avoid creating places where materials can vent/flow
to the surface during hydraulic fracturing. Thus, any pre-installation soil sampling will have to balance the competing needs
to gather information about contaminant distribution in soil and the requirement to prevent venting during fracturing.
Fracturing can be conducted successfully in the vicinity of bore holes that have been completely plugged. In the horizontal
configuration cells at Offutt AFB there were several locations that had been sampled and backfilled with bentonite pellets
a year earlier; these did not vent to the surface (leak) during installation of the electrode and treatment zone layers. We
do not know any way of predicting when bore holes have been completely plugged; therefore it is desirable to avoid pre-
installation soil sampling in the area where a horizontal cell will be installed. Site records should also be used to identify
sampling locations from previous investigations. If other factors are equal, small borings by machines such as a Geoprobe
may be less of a problem than the larger hollowstem auger borings.
2.3 Access wells for electrodes
The casing for electrode access wells must be made of a non-conductive material because the casing will be in
a strong electrical field and could short circuit between the electrode layers or conduct electricity to the surface where it
-------
would be a hazard to personnel. The most likely material for such casing will be polyvinyl chloride (PVC) or chlorinated
polyvinyl chloride (CPVC). CPVC retains strength at higher temperatures. We have not yet encountered any temperature-
related problems with PVC electrode well casing and it has been used successfully in a number of installations (14,15,
16,17,18,19,20).
The casing should be at least 3 inches nominal diameter for the cathode well to allow space for pumps, sensors,
and electrical connections to fit inside it; the anode casing may be smaller, depending on the type of electrical connection
to the granular graphite and the sensors employed. The casing should also have flush (smooth) threaded joints and be
either schedule 80 or 120 because it will be driven some distance into the soil. It is advisable to select the size of the
casing after all the other down-hole components have been identified and their sizes are known. If there is very little extra
space it would be useful to add devices to hold components to one side to allow space for those installed later.
The upper electrode (anode) should be installed first, then the treatment zones, and finally, the lower electrode
(cathode). This recommendation is not unique to Lasagna™; it applies to any installation where horizontal hydraulic
fractures are placed one above the other (8). When a hydraulic fracture is propagating outward from the injection point
and encounters an existing boring, even one that is thought to be well sealed, there is a risk that the fracture will vent to
the surface. When a casing has been driven directly into soil or into an undersized hole, a good seal is created and there
is only a slight risk of another hydraulic fracture venting upward along the side of the casing. This risk can be avoided
entirely by always creating a new fracture below the depth of any existing fracture. The other potential problem with
fracturing at the depth of an existing casing is the shadow effect (see Figure 2-1). When a growing fracture encounters
a vertical casing or other large object it propagates around it but often does not close upon itself on the other side of the
object. This is a minor problem for an electrode fracture because a sheet electrode will create a satisfactory electrical field
even if the electrode has some holes in it. (Chen, Jiann-Long, Personal Communication) It could be a more significant
problem for a treatment zone since water and contaminants could pass through the shadow zone or hole without being
treated.
Most soils in which horizontal hydraulic fractures can be created are relatively stiff, on the order of 15 blows per
foot (bpf) with a 140 pound hammer dropping 30 inches onto a 2-inch split spoon sampler. In such soils, PVC casing of
sufficient size for electrode wells cannot be driven without some preparation of the hole. We have successfully installed
PVC casing by beveling the leading edge and pushing the casing into a hole approximately the same size as the outside
diameter of the casing. This hole is terminated about 4 feet above the depth at which the fracture will be created. The
casing is sealed into the soil and driven the remaining distance by boring inside the casing either by hand or with a drill
rig. A hole is bored about a foot ahead of the casing, the casing pushed or hammered to refusal, and then another short
hole is bored ahead of the casing. This is repeated until the casing has reached the depth at which the fracture will be
created. Boring inside the casing and driving it into an undersized hole insures a seal between the casing and soil and
prevents leakage up the outside of the casing during fracturing. Although schedule 80 or 120 PVC casing is both strong
and rigid, it has enough flexibility to conform to slightly irregular holes and provide a good pressure seal. When the casing
has been driven to the desired depth, an open (un-cased) hole should be bored about 4 inches deeper to allow space for
notching the soil and creating the hydraulic fracture.
The soils at Offutt AFB were extremely soft, 1 to 2 bpf. In these soils, a hollow stem auger was advanced about
10 feet from the surface in order to bypass and seal off the sandy materials near the surface. The casing was driven inside
the auger the remaining distance using a hydraulic hammer. The end of the casing was sealed with a pointed plastic plug
pushed about 4 inches deeper with a drill rod once the casing had been driven to the desired depth.
These methods for installing access wells require some skill and experience. An obvious and apparently simpler
approach would be to drill a hole larger .than the casing and then grout the annulus between the casing and the soil to
prevent venting to the surface during hydraulic fracturing. Even when a seal could be maintained by using a non-shrinking
grout, the installation resulted in problems with notching the soil at the bottom of the hole or with the fracturing process
itself. Eventually we abandoned this approach and used some variation on a driven casing to install electrode wells.
6 .
-------
o
Casing
shadow
Figure 2-1. Plan view of hydraulic fracture propagating from injection point (I.P.) toward casing; shadow (unfractured area)
develops behind casing
2.4 Electrode material
In the large-scale demonstration of the vertical configuration (7), the electrodes were 1.5 inch (3.8 cm) sheets of
50/50 by volume iron filings and Loresco coke with 1.9 cm steel rods driven into the electrode material at six locations to
make electrical connection. The electrical conductivity of the electrodes was 4.5 S/m, about 200 times greater than the
electrical conductivity of the surrounding soil.
This kind of material could not be used for electrodes in the horizontal configuration because of problems related
to the thickness of the electrode. Hydraulic fractures may be several cm thick near the injection point (electrode access
well) but usually decrease in thickness to 0.5 to 1.0 cm at several meters from the well. The material in a fracture of this
thickness must conduct both water and electricity for the horizontal configuration Lasagna™ cell to function properly.
Several different sizes of carbon materials were tested as potential electrode materials (20). Graphite, 20 x 35 mesh size,
was used most frequently. Its unloaded hydraulic conductivity was 1.2x10"3 cm/sec (14). This has been sufficient for
adding or removing liquids from an electrode and for minimizing blockage of hydraulic flow by gases generated in
electrolysis. The wet electrical conductivity ranged from 55 to 630 S/m at confining stresses from zero to 30 kPa. (20),
This electrical conductivity was great enough to provide a sharp contrast between the electrical conductivity of the
electrode fracture and surrounding soil and to allow electrical connection between the power supply and the electrode at
only one point, in the electrode well.
Graphite is the most satisfactory material we have tested. Its electrical conductivity is high enough to create a
strong contrast between the electrical conductivity of the electrode fracture and the surrounding soil, so that most of the
electrical power will be conducted through the electrode fracture and only one contact point between the electrode fracture
and the power supply is needed. Larger particle size (e.g. 10 x 20 mesh) might provide better water flow and release of
electrolysis gases. Particle size is not an issue for hydraulic fracturing. We have created fractures using material that
passed a 1/4" (0.64 cm) screen and recovered the larger particles in soil samples 12 feet (7.4 m) from the injection point.
As the particle size is increased, the grain-to-grain contact may decrease enough to reduce the bulk electrical conductivity.
..7
-------
If the use of a larger particle size is being considered, tests of wet electrical conductivity are recommended and some of
these tests should be under confining stresses similar to the (soil) overburden stresses at the depths where the electrode
fractures will be installed.
2.5 Electrode spacing
The distance between the electrodes affects the intensity of the EK effect and the speed of the treatment process.
There are four factors to consider in selecting the spacing between electrodes for the horizontal configuration:
Lack of control of fracture slope
Relationship between electric field strength (V/m), and processing .time
* Heating limits on power supplied to an electrode
• Treatment zone installation
The majority of hydraulic fractures exhibit some slope upward as they propagate away from the injection point.
This slope may range from 0.8 to 25 degrees. Unpublished data from the University of Cincinnati suggest that at the 95%
confidence level the slope of any fracture at a particular site will be within 10 degrees of the average slope for all fractures
at the site. The slope appears to depend strongly on conditions at the site; which site conditions affect the slope are not
known. Soils with sharp contrasts in particle size (texture) are the exception. In these soils, fractures will often propagate
along the contact between two layers of soil when there is a difference in particle size. If the soil layers are horizontal, the
fractures will follow the layer and remain horizontal. In any case, the slope of the fracture cannot be controlled; soil
conditions control the slope. If soil conditions change within relatively short vertical distances, fractures at the same
location may have different slopes at different depths. Thus the distance between two electrode fractures may increase,
remain the same, or decrease as distance from the injection point increases.
The rate of movement of liquids and ions is directly proportional to the electric field strength, the applied voltage
divided by the distance between the electrodes. Treatment time is affected by the rate of movement. There is a direct
relation between electric field strength and treatment time in the vertical configuration where EK is the primary transport
mechanism. The relation is less clear in the horizontal configuration because both hydraulic flow and EK are present as
transport mechanisms. An electric field strength of 50 V/m is often recommended as a starting point for design but lower
field strengths may be sufficient in the horizontal configuration because EK mobilizes contaminants from low-permeability
zones and may need to move them only short distances before they encounter gravitationally induced hydraulic flow. See
Alshawabkeh et al. (24) for a discussion of electrode design and treatment time and a graph of treatment time versus
electric field strength.
Heating in the well for the electrode may limit the power that can be supplied to a horizontal configuration
Lasagna™ cell and, hence, the maximum electrode spacing. See Section 4.2 (Temperature Protection) below. In the
vertical configuration (Figure 1-1) the top of the electrode is accessible and a sufficient number of connection points can
be inserted to supply and distribute the power without problems. In the horizontal configuration (Figures 2-2 and 2-3) the
space in the electrode well is limited and power is supplied to the graphite fracture at a single point. Even under optimum
conditions this is a small contact point and will have a finite capacity to transmit power without overheating.
In the horizontal cells at Offutt AFB we were using a power supply that fixed the applied voltage and allowed the
current to float i.e., vary in response to conditions in the cell (see Section 4.1 below). Voltage was adjusted by changing
the connections on a multi-tap isolation transformer. We initially set the potential at 68 volts which corresponded to an
electric field strength of 44 V/m and the cell drew 72 amperes. After several days the cathode connection began to
overheat so the potential was reduced to 49 volts which corresponded to 32 V/m and the cell operated satisfactorily for
a long period, drawing 50 amperes. Selection of electrode spacing must take into consideration the amount of power that
can be supplied to the cell.
-------
A = 40.7 ciri2
(73.9 crm)
A = 90.5 crm.
(121.9 cm*)
Graphite Fracture
Figure 2-2. Electrode Well Dimensions
Electrode Well
— 7.2 cm—^
(9.7 cm)
Dimensions for 3-inch well;
dimension for 4-inch well
in parentheses
1 cm
Connection to
Power Supply
Thermocouple
- Stainless Steel Pipe
—PVC Casing
Figure 2-3. Connection to Graphite Electrode Fracture. Connection and PVC Casing Drawn to Scale
-------
The final consideration in selecting the distance between electrodes is the space needed for installation of the
treatment zones As noted in Section 2.6 (below), the minimum vertical spacing between hydraulic fractures is between
9 and 12 inches (23 to 30 cm). The amount of material in a treatment zone (fracture) will be limited by the thickness of
the fracture and the thickness can be modified only slightly changes in the hydraulic fracturing process. If the treatment
scheme requires a certain mass of material (e.g. zero-valent iron), there must be enough space between the electrode
fractures to install the number of treatment fractures that will contain the needed total amount of treatment material.
2.6 Treatment zones
Treatment zones may be installed with or without permanent access wells depending on how the zone will function.
The biological treatment zone at Rickenbacker Air National Guard Base (ANGB) (19) was connected to the surface by
several wells for injecting gases in the center of the zone and venting around the edges. This treatment zone required
periodic additions of methane to stimulate the methanotrophic microorganisms that had been inoculated into the zone to
degrade trichloroethylene. For a treatment zone that requires a permanent connection to the surface, it is best to leave
the injection well in place and use it for access to the zone. There is often a problem in installing an access well after the
fracture has been created. Smearing of soil at the junction between the well and the fracture can be difficult to remove
when the fracture is at a substantial depth and smearing will block the transfer of liquids and gases between the access
well and the fracture.
The zero-valent iron treatment zones at Rickenbacker ANGB (18) and at Offutt AFB (20) did not have permanent
access wells because the treatment process did not require any. For treatment zones such as these, a well may be
installed for creating the treatment fracture and then removed. Another successful approach uses a direct push soil
sampler as the access for creating the fracture. We used an Earthprobe 2000 on a SIMCO drill rig but other samplers such
as a Geoprobe may be used in a similar fashion. The Earthprobe has an outer casing about 1.25 inches in diameter. The
outer casing and an inner rod with a point to seal the bottom end of the casing were driven together to the depth for
fracturing The inner rod was pushed a few inches deeper to create a cavity below the casing for notching the soil and
injecting the treatment zone material. The top end of the casing was threaded for 1.25 inch male pipe thread so the
fracturing well head could be attached. After one treatment zone was installed, all materials were removed from the outer
casing by flushing with water, the inner rod was replaced and the casing and rod driven to the depth for the next treatment
zone When all treatment zones had been installed, the Earthprobe casing was removed and the hole backfilled with
bentonite pellets. Other methods for pumping fluid grouts are available and would likely work as well as the bentonite
pellets.
The total mass of treatment zone material installed by hydraulic fracturing is limited by the installation process
more than is the case for treatment zones in the vertical configuration. The thickness of an individual fracture can be
increased slightly by adjusting the viscosity of the gel and increasing the solids content of the mix but most fractures will
be about 1 cm thick. The vertical spacing is also limited. In many soils, fractures can be installed with 9 inches (23 cm)
vertical spacing but it is safer to install them at intervals of 12 inches (30 cm) or more to avoid unplanned intersection of
the fractures Given these physical limitations of the process, it may not be possible to place sufficient mass of treatment
zone material in soil to achieve the desired residence (contact) time in the treatment zone for water and contained
contaminants before they reach the cathode and are removed from the treatment system. If this is a problem the system
can be shut down (both the power and pumping from the cathode) periodically to increase residence time in the treatment
zone or the fluids can be passed through a filter/treatment system after they are pumped from the cathode.
Whether the hydraulic fracturing machine operates in a batch or continuous-mix mode, there will be a system for
feeding solids into the mix. When working with iron powder or filings for a treatment zone, care must be taken to clean
the solids feed system thoroughly after fracturing. Some granular iron will remain in the system and will gradually corrode
though contact with moisture in the air and fuse into a solid mass. A year .after we installed the Lasagna cells at Offutt
AFB we were installing some sand-filled fractures and encountered a major problem with the machinery jamming and
stalling the pump engines. When the feed system was disassembled we found that solidified masses of iron powder had
10
-------
reduced the feed path to about 5% of its normal size. The iron was removed by breaking it up with a crow bar It came
out easily in chunks from 1 -inch thick to the size of a hand. It would have been much easier and more convenient to have
cleaned the feed system before the iron fused into a solid mass.
11
-------
3. Fluid Management in Electrodes
The horizontal configuration (see Figure 1 -2) is most efficient if the anode is the upper (positive) electrode and the
lowe^negative) electrode. With this arrangement of electrodes the EK and gravitational fields move water
and have an additive effect. For this process to function, water must be supplied to the anode and removed
fromhe cathode, both of which are porous. If water is not removed from the cathode the f^'"^^^
accumulate at the cathode and gradually build a hydraulic head that opposes movement by EK and stops the process (14,
15,16). Water movement must be maintained because it transports contaminants to treatment zones.
When the anode is installed below a water table, water will be supplied naturally and no additional water is needed
In our installations, the water table was near the anode level but provision was made for water supply because of
anticipated seasonal fluctuations. A pressure transducer was set in the welf and connected to a system or mamta.nm
the water level at a point above the transducer. The set point was adjustable to allow tests of the effect of water level or,
temperature in the well.
Adding excess water to the anode increases the risk that contaminants will be transported laterally out of the cell
as well as vertically downward through the treatment zones. The depth of water maintained in the anode well should be
selected based on the potential for such lateral transport. Several piezometers in the cell just above the anode fracture
and another piezometer outside the cell at the same depth will provide the information needed to calcu ate the laterd
hydraulic head gradient (driving force). Multiplying the lateral hydraulic head gradient by the lateral hydraulic conduclMty
Ks the lateral flow rate. An acceptable lateral flow rate is specific to each site, depending on considerations such a,
distance to adjacent uncontaminated soil, monitoring wells, and the need to track the mass balance of contaminants. Once
an acceptable lateral flow rate has been estimated, it is possible to back-calculate to determine an acceptable waterMeve
for piezometers in the Lasagna cell. Note that piezometers must be installed in soil outside the electrode because.water
levels measured in electrode wells are not a reliable measure of water levels in soils at the same level. Gases arc
generated in both electrodes by electrolysis of water. This gas generation interferes with water movement in the cathode
to a gXte extent than in the anode because the gas production is greater (20). The set point for the water level in the
anode well must be adjusted until the water level in the piezometer(s) in the cell is achieved. A pressure transduce ni a
pTezornetermightbeused to control water addition to the anode well but some kind of control system would <*Uberequn*d
in the anode well to prevent over-filling which would present an electrical hazard or running dry which would interfere with
the electroosmotic process and cause overheating of the anode.
In the cathode well another pressure transducer measured the water level and a pneumatic bladder pump (Cluxton .
Instruments Inc) was used to lower the liquid level and induce flow from the cathode into the well. There are many other
pump^hat serSe Equally well. See, for instance, the pump on page 60 of reference (28). The only requirements are that
52K^Srial8qbe Distant to high temperatures and PH that may be as great as^l 0^ See the l.st of temperate and
PH-resistant materials in Section 4.2 (Temperature Protection). This pumping of cathode fluids insures that EK flow will
not be blocked and contaminants will continue to move through the treatment zones.
Water pumped in from soil surrounding the cathode also counteracts the resistance heating atthe contact between
thecablefrom the power supply and the granular graphite in the electrode fracture. Because finegraphite particles caused
problems in the pump check valves and in the volume measuring system above ground, a screen was wrapped arou d
fee pump intake To prevent clogging of this pump screen, water was injected periodically ms.de the screen tc.back-flu ,h
it We noted temperature increases in the cathode well that coincided with the pump screen flushing cycle (Figure 3-1).
We speculate that the temporary increase in fluid level in the well blocked inflow from the graphite electrode and resu led
in the observed temperature spikes. The interval between pump screen flushes was gradually increased from 30 minutes
to 6 hours; this kept the pump operating and greatly decreased heating associated with the pump cycle.
Gases are generated in the electrode fractures due to electrolysis of water. Oxygen is generated at the anode
and hydrogen at the cathode with the volume of hydrogen being twice that of the oxygen. These gases block pore space
12
-------
West Cathode 10/17/00
2.5
2
1.5
Time (24 hre)
West Cathode 10/17/00
Time (24 hrs)
Figure 3-1. Simultaneous Increases in Fluid Depth and Temperature in a Cathode Well
in the electrodes and reduce their fluid transmissivity. The installation at Offutt AFB is at twice the depth of previous
horizontal configuration cells (20). We speculate that the increased depth is the reason why restriction of fluid flow by
gases is less of a problem than at previous installations but we do not yet have data to test this hypothesis.
In the vertical configuration demonstration (Figure 1 -1), fluids were recycled from the cathode back to the anode.
This was accomplished with a gravity drain located slightly below the original ground surface. Since the fluids were never
brought above the ground surface, the treatment was completely in situ, Thus the operation was not subject to the
provisions of the Resource Conservation and Recovery Act (RCRA) or the Underground Injection Control (UIC) program
which would have required removal/destruction of contaminants before fluids were re-injected into the anode..
In the horizontal configuration demonstration at Offutt AFB we are not recycling fluids from the cathode to the
anode because of concern about plugging the graphite pack in the well and the anode fracture. The space in an electrode
well is quite limited (Figures 2-2 and 2-3) and is easily plugged if the fluid pumped from the cathode contains particulates,
precipitates, floes etc. We have observed these in previous small-scale studies (14-19) and have experienced problems
with recharge of fluids into the anode due to plugging.
Fluids pumped from the cathode are considered groundwater. Groundwater itself is neither a solid waste nor a
hazardous waste. When the groundwater contains hazardous constituents above health-based levels, it is subject to all
applicable RCRA requirements. If groundwater is passed through a filtering system (e.g. activated carbon) and all
constituents are reduced below health-based levels, the groundwater no longer contains hazardous wastes and does not
have to be managed as hazardous under RCRA. The recharge of treated cathode fluids to the anode would still be
regulated by the UIC program and the state or municipality will have their own requirements that have to be met for
discharge to a publicly owned treatment works (POTW) or storm sewer. We have found the requirements for discharge
to a POTW (40CFR304.1 - 304.6) to be less restrictive than those for the UIC program (40CFR144-148).
13
-------
4. Electrical Power Management
A direct current (dc) power supply is connected to the Lasagna™ cell (see Figure 1-2) to induce EK transport,
electroosmosis for water and electromigration for ions. When an electrical field is applied to soil, the water, ions, and other
contaminants in soil within the field move uniformly. This allows extraction of contaminants from areas of soil that would
be bypassed by hydraulic flow, the majority of which occur in large pores and channels between soil structural units. (21,
22,23,24).
4.1 Power supply
A series of power supplies were built for the different sites, increasing in size from 9 kW in the beginning tests at
Center Hill (16), to 40 kW at the Offutt AFB site. These power supplies share four common elements: (1) a fused safety
disconnect switch connected to the utility line power, (2) some type of controller, (3) an isolation transformer and (4), a
rectifier to convert to direct current. These power supplies were custom-built to keep costs down, save time, and because
no suitable units were available commercially. The sizes of the power supplies vary because of the increasing scale of
the field tests.
The controller in the 9 kW power supply was an electromagnetic coil design, using a saturable reactor coil, that
could be adjusted to set the current (Adc) delivered to the Lasagna™ cell. The potential (Vdc) was allowed to change in
response to the effective resistance of the cell. Operation in this constant-current mode was selected because much of
the bench scale work done by others used constant-current mode (21,22, 23,24) where it was unclear how much current
would be drawn from the power supply. This experiment had not been conducted previously in an open system field test
without boundaries. This type of magnetic coil design is very robust, and could operate without damage to the power
supply even with a direct short in the cell. This first power supply used single-phase 208 Vac line power. This design was
chosen to minimize the cost of the expensive magnetic elements. A three-phase design would have provided smoother
direct current output but would have cost about three times as much because three controllers (saturable reactor coils)
would have been required.
When the size of the power supply was scaled up, the saturable reactor became less useful as a controller. The
redesign substituted an electronic silicon controlled rectifier-type (SCR) power controller for the electromagnetic coil. This
change was made because of the prohibitive cost of a larger coil, because the SCR controller was expected to provide
more precise control, and because it had the added feature of operating in either constant-current or constant-potential
mode. This controller includes a current limit circuit to prevent damage to the unit by a short circuit at the cell. The design
was modified to use 208, 230, or 240 Vac, depending on what was available at a particular site. Operation in constant-
potential mode, rather than constant-current mode, became the preferred method, since the constant voltage gradient
appeared to stabilize the system. The maximum current supplied by the power supply has ranged from 40 to 200 Adc
depending on the size and number of Lasagna™ cells that were being operated.
For the RickenbackerANGB site at Columbus, OH (18,19) the design included a thermocouple at the connection
to the graphite fracture in the electrode well. This was connected to a control circuit in the power supply that shut down
power when the temperature in the electrode well reached 80°C and restored power when temperature decreased to 75°C.
This prevented boiling in the electrode wells and fluidization of the granular graphite with subsequent loss of electrical
contact. The power supply for Offutt AFB includes a similar temperature controller, and a computer data acquisition and
control system, with remote operation.
The third element, the multi-tap isolation transformer, is usually not provided in an off-the-shelf rectifier power
supply. It is essential in this application for several reasons.
It is required to separate the system from the utility line power, so that the potentials at the cell could
"float" as needed with respect to far-field ground
14
-------
It reduces the available short-circuit current from the utility line to a safer level and greatly reduces the
hazard of electric shock to workers at the site
The multi-tap feature also allows adjustment for maximum efficiency, which is important when using an
SCR-type controller '
The fourth element, the full wave bridge rectifier, was designed with over-size diodes, and over-size cooling fins,
and fans, to make it robust enough for the changing conditions at field sites. Computer-grade capacitor banks with about
0.4 Farad capacity are used to smooth the output wave to direct current.
4.2 Temperature protection
Electrical resistance is inherent in the system. The major resistance will be where electricity is applied to the
graphite in the electrode well and at the grain-to-grain contact between graphite particles in the electrode fracture. Heat
will be generated at these points and if left unchecked will result in boiling of fluids in the electrode well. This boiling
disrupts the grain-to-grain contact and the electrical contact between the graphite and the power supply. Once contact
is lost it may be quite difficult to reestablish.
Over-heating is avoided by monitoring temperature and shutting off or reducing power when temperature
approaches some critical value. Our experience suggests that shutting off the power between 75°C and 80°C will be
satisfactory. The temperature sensor should be placed as close as possible to the contact between the power conduit and
the graphite pack. When we used stainless steel pipe for the power conduit, the thermocouples were placed inside the
pipe as near to the bottom as possible (Figure 2-3). When the thermocouples were placed too high in the pipe the
temperature at the critical point reached boiling before power was reduced and the electrical contact was lost.
The temperature and pressure monitors, pumps, and any other monitoring or control equipment to be placed in
the electrode wells, must be fabricated using materials that are resistant to high temperatures and to the extremes in pH
(2 to 10). We have used type T copper and constantan thermocouples to measure temperature in all our installations.
This type is most accurate in the range 0 - 100°C (30). Other materials that have been used in horizontal configuration
installations include:
Teflon for the thermocouple insulation
Chlorinated Poly Vinyl Chloride (CPVC) for the pump body
Poly Vinyldenefloride (PVDF) check valve body
VITON® seals
Polypropylene tubing and screen
Nylon cable ties .
Neoprene hose
Polyolefin heat shrink tubing
Poly Vinyl Chloride (PVC) insulated 105°C rated electrical cable
All these materials are heat resistant to 100°C, and are very resistant to extremes of pH. Materials such as duct tape and
standard PVC electrical tape should not be used in electrode wells. • • •
4.3 Connections
As noted above, the point at which the cable from the power supply connects to the graphite fracture (electrode)
is subject to contact resistance, local heating, corrosion, and loss of electrical energy. After working with a variety of
materials and designs we have settled on stainless steel pipe as the connection that is easiest to fabricate, install,.and
maintain. American Iron and Steel Institute (AISI) type 304 stainless steel is most readily available but AISI type 316 is
more resistant to the corrosive environment, particularly in the anode. We used nominal 3/4 inch schedule 120 pipe in type
15
-------
316 stainless steel. There was enough space inside the pipe to allow insertion of a thermocouple completely to the bottom
to monitor temperature at the contact between the stainless steel and the graphite pack. The wall thickness was sufficient
to allow the cutting of alternating male and female inch National Pipe Threads (NPT) so sections of the pipe could be
screwed together with flush joints. The flush joints gave slightly more space and made it easier to place pumps, pressure
transducers, and water lines in the well after the stainless steel pipe connection had been driven into the graphite pack.
See Figure 2-3 for a scaled diagram of the stainless pipe inserted into the graphite pack in an electrode; this diagram
illustrates the limited space in an electrode well.
The threaded joints in the stainless steel pipe also made it easier to remove and replace pumps for periodic
servicing. Our installations had a fixed enclosure over the electrode wells with the inside clearance limited to about 5.5
feet. This made threaded connections necessary in order to add and remove the pipe in sections of 5 feet or less. It there
had not been a fixed enclosure over the electrode wells, the pipe connection could be in longer pieces or even one piece
if the depth to the graphite was less than 20 feet.
In reference (20) it was concluded that stainless steel connections for the cathode are relatively trouble-free while
stainless steel connections for the anode would require periodic maintenance. Subsequent experience supports this
conclusion. Cathode connections at Offutt AFB operated for long periods with no major increases in temperature. When
they were removed for inspection they showed no corrosion and only a small amount of mineral and graphite deposition
that was easily removed with emery paper or a steel brush. The performance of anode connections was more erratic.
In one cell the AISI type 304 schedule 40 stainless steel connection operated for over a year during which time the current
decreased from 54 to 42 amps at 48 volts DC and the temperature increased by only about 20C. In the other cell, which
drew only about 25 amps, two consecutive AISI type 316 schedule 120 stainless steel connections corroded completely
in less than two months each. We speculate that industrial chemicals present in the soil around the anode were involved
in the increased rate of corrosion. Increases in temperature and decreases in current gave clear advance warning that
the anode connections were about to fail.
Another type of electrode connection was used successfully on a number of installations. This was a solid graphite
rod (tip) that was machined to a point on one end and drilled and tapped on the other. See Figure 8 in reference (20) for
a model of this type of connection. The cable from the power supply was soldered to a short copper rod which screwed
into the base of this graphite point. The cable and rod were inside a CPVC pipe to insulate the connection. The CPVC
pipe screwed into the graphite tip around the copper rod and cable to seal out electrode fluids from the connection. See
also Figure 3 in reference (17). These graphite connections were durable, but they had to be pounded, rotated, or
withdrawn at intervals ranging from a week to two months to reestablish contact and reduce resistance. These graphite
connections cannot be driven with as much force as the stainless steel pipe. When inserted after fracturing, the graphite
into which the connections were driven was the softer (less compact) material in the upper part of the access well. We
also experimented with suspending the contact point in the access well and injecting the graphite around it during creation
of the graphite fracture electrode. Both types of connection worked well but still required periodic maintenance.
The electrical contact will usually be driven or pushed into the graphite pack to make connection. The upper
portion of the graphite pack is usually softer and less compact than the pack near the point where the graphite fracture
intersects the well. This softer material will provide a less satisfactory contact point and exhibit greater electrical resistance
and heating than the deeper material. A convenient way to determine when the contact is satisfactory is to connect the
power supply and observe the increase in current as the contact is driven deeper into the graphite. When there is little
change with increasing depth of insertion, the contact is satisfactory. At that point the temperature in the connection should
be observed for several hours. A sharp and continuing increase in temperature (ca. 20C per hour) indicates contact
resistance and an unsatisfactory contact.
When fracturing has been completed some graphite may have to be removed from the well but this should be done
carefully and after careful consideration of the effects. First, allow the graphite to settle and consolidate in the access well
for at least 24 hours. Then measure the distance to the surface of the graphite and note whether the graphite surface is
below the static water level. If this point is more than 6 feet (1.8 m) from the fracture level, we recommend removing some
graphite to minimize resistance to fluid flow. It is recommended that about 4 feet (1.2 m) of graphite pack be left in the
16
-------
electrode well. In the cathode well, the level of the graphite must be below the static water level because liquids are
pumped from the cathode to maintain downward flow in the cell.
For reasons we do not understand, the electrical connection in some electrodes will gradually degrade with the
electrical current decreasing and temperature in the electrode increasing. When this happens, first try pounding the
connection slightly deeper into the graphite pack and rotating it slightly. In several cases this has re-established electrical
contact for a few months. If this is not successful then re-fracturing is necessary. Withdraw any pumps, sensors, and
electrical contacts from the electrode well, remove as much of the old graphite as possible, and pump in a small amount
(about 100 Ibs) of new graphite. In every case where we have done this it re-establishes electrical contact for an extended
period.
Other applications of electroosmosis for remediation of contaminated soil have used electrode systems that were
buffered against pH changes. These systems appear to avoid the problems with loss of electrical contact and corrosion
at the anode that we encountered. See references (28, 29) for examples of these connections.
17
-------
5. Process Monitoring
Conditions in the horizontal Lasagna™ cell must be checked periodically to determine whether the process is
proceeding satisfactorily and to provide a basis for adjustment and maintenance. The critical parameters are potential
(voltage), current, temperature and fluid levels in the electrode wells.
The discussion of monitoring of electrical potential and current (Vdc, Adc) in a horizontal Lasagna™ installation
assumes that the power supply controls potential and allows current to float in response to the resistance in various parts
of the cell. A fixed current/floating potential power supply was used in some of the early work but we have the most
experience with using the fixed potential design. The fixed potential design was used in most of the work because the fixed
potential gradient in soil between the electrodes appeared to stabilize conditions in the cell.
Potential was adjusted by changing the connections on the multi-tap isolation transformer, selecting a value that
would provide minimum heating problems at the electrode connections and maximum efficiency, which is important when
using an SCR-type controller. Potential was measured regularly on the output side to assure that the power supply was
functioning as intended. No problems with control of potential have been noted. The computer monitoring system at Offutt
AFB recorded potential, current, temperature etc. once per minute. These files were downloaded daily to the computer
in Cincinnati. The apparent potential varied less than 1.0 volt. Some of this variation is noise inherent in the system, as
discussed below. When potential was measured at the power supply with a multimeter the variation was less than 0.1 volt.
Potential is also measured in the soil between the electrode fractures in the cell. If the soil electrical properties
are uniform between the electrodes and if there is good (low resistance) contact between the cable from the power supply
and the graphite in the electrode well, a plot of electrical potential versus depth will be roughly linear. See Figure 5-1 for
an example of a satisfactory (linear) voltage profile in soil. This is the desired condition. A uniform potential gradient
between the electrodes insures that water and contaminants are being affected as much as possible in all parts of the soil.
A uniform potential gradient also indicates good contact at the electrodes and a low risk of overheating and disruption of
grain-to-grain contact in the graphite.
Figure 5-2 illustrates another type of information that may be obtained from voltage measurements in soil. The
voltage probe from which these data were obtained was located about 12 feet from the injection point at the center of the
cell. The electrode fractures were installed at approximately 18 and 24 feet. The sharp increase in voltage above 16 feel:
indicates that the.upper electrode (anode) fracture has risen (dipped upward) as it propagated away from the injection
point. This phenomenon (steep up-dip) is observed in many hydraulic fractures.
Potential in soil is measured by driving a plastic casing, with its lower end sealed, below the depth of the lower
electrode (cathode). This casing has stainless steel screws through the wall at 6-inch (15 cm) intervals up the casing over
the interval corresponding to distance between the electrode fractures. The outer end of the screw contacts the soil; a
slider is inserted inside the casing to contact the inner end of the screw. A multimeter is connected to the slider and
measures the potential difference between each screw and the cathode of the nearest cell (Figure. 5-3). A plot of these
measurements versus depth shows the average slope (potential gradient) and any deviations from linearity (see Figures
5-1 and 5-2).
When a fixed-potential power supply design is used, then current (Adc) varies inversely with resistance in the cell.
The majority of changes in resistance occur at the contact between the cable from the power supply and the graphite
fracture. Monitoring current thus becomes one of the primary means for assessing the state of this connection. Decreases
In current to a cell give advance warning that the connection is deteriorating and such decreases are invariably followed
by increases in temperature. Detecting this behavior before the temperature approaches the boiling point provides an
opportunity to clean and reestablish the connection before it deteriorates enough to do irreversible damage to the packing
and grain-to-grain contact of graphite in the electrode well and the electrode fracture in the vicinity of the well.
18
-------
16
Jj 18
§• 20
Q
22
24,
; ;. ' |.
• • • ).
,*'""
..-•-*""'
". C """"
i . i . 'i . i
10 20 30 40
Volts
16 -
JA 20 -
1
O 22
24 -
261 1 1 1 1 1 1 '. I . I . I . I , .
14 16 18 20 22 24 26 28 3
Volts
Figure 5-1. Linear Voltage Profile in Soil
Figure 5-2. Voltage Profile in Soil
To Cathode (-)
in Lasagna Cell
PVC
Casing
Ground Surface
Wire to
Multimeter
Stainless
Steel
Screws
PVC Pipe
Metal Tip
Figure 5-3. Voltage Probe
19
-------
Monitoring temperature in the electrode wells is also a way to assess the condition of the connection between the
power conduit and the graphite electrode fracture. When temperature rises sharply, the connection is deteriorating and
resistance is increasing. However, the primary importance of temperature is its effect on the granular graphite in the
electrode well and electrode fracture. As indicated above, boiling disrupts the grain-to-grain contact and makes it difficult
to reestablish a satisfactory (low resistance) connection. Temperature is monitored primarily so this problem can be
avoided. As also noted above, temperature should be measured as closely as possible to the contact between the power
conduit and the granular graphite in the electrode well.
Fluid levels are monitored in all electrode wells to determine the need for adding or removing fluids and to check
the operation of systems for accomplishing this. The cathode (lower electrode) wells are pumped to remove fluids
accumulating due to EK transport of water through soil. If this is not done, a hydraulic head would gradually develop that
opposes EK transport and either halts it or greatly reduces efficiency. Water is added to anode (upper electrode) wells
to replace that which is moved downward by EK. Water levels in the anode well should not be increased so far above
static (far-field) levels that a groundwater mound develops and transports contaminants laterally out of the Lasagna cell.
We used down-well pressure transducers for monitoring fluid levels; other systems could be used to accomplish the same
thing.
While all this process monitoring could be done manually by personnel at the site, it is highly advantageous to be
able to accomplish this automatically and have access to the information from a remote location. Automatic monitoring
makes it possible to develop on-site control systems that will, for instance, turn off power if temperature is too high, add
water to maintain desired levels in anode wells, or transmit alert messages if fluid levels in the cathode wells rise too high.
When process monitoring is accomplished automatically, the signals from current, temperature, and voltage
sensors may be noisy (vary erratically) for several reasons. Very long cable lengths, the need to use isolation interface
modules (to protect the computer or data acquisition system from accidental direct connection to high voltages at the field
site), and an open framework of the data acquisition rack and power supply, may all contribute to noise. When a multimeter
is used to monitor current, it is connected to the current shunt of a power supply and noise reduction techniques are
incorporated in the multimeter. This arrangement provides a signal that is much more stable than a signal from a
multimeter connected through an isolation interface module and data acquisition rack on a computer or other data
acquisition system. For indirect, noisy connections it is advisable to use software to calculate a running average over a
5- to 7-second period so the true value of temperature, current, or potential is not obscured by variation that is an artifact
of the system design. See the temperature data in Figure 3-1 for an example of a noisy signal.
20
-------
6. Technology Niche
The applicability, in general, of electrokinetic extraction is discussed in reference (24). An electric field is much
more effective than a hydraulic gradient for moving contaminants in fine-grained soils of low hydraulic conductivity. In
coarse-grained soils of high hydraulic conductivity, the hydraulic gradient is a more effective driving force. In
heterogeneous soils, an electric field in conjunction with a hydraulic gradient can achieve nearly complete removal of
contaminants where the hydraulic gradient alone could not. See, for example, the laboratory studies of electrokinetics in
sand columns containing isolated lumps of contaminated clay (4). Although contaminant transport rates and efficiencies
are strongly affected by soil type, electrokinetic extraction can be used, with some modification, in most soils. If the
contaminant exists as a strongly sorbed phase on the soil particle surface or as precipitates in the soil pore, this may be
a significant limitation.
The following are some conditions which affect the usefulness of the Lasagna™ horizontal configuration relative
to the vertical configuration. They provide some basis for selecting the configuration that will be most effective in a
particular setting.
In normally consolidated soils with no significant stratification (layering), the vertical configuration is the obvious .
choice. Hydraulic fracturing is unlikely to be successful and the horizontal configuration cannot be installed. When there
is doubt about soil properties, install some test fractures. The installation at Offutt AFB is a case where the soil was
normally consolidated and very soft, but fracturing was successful in a part of the soil profile.
Where soil is contaminated completely to the surface, and is relatively shallow, either configuration can be made
to work. The horizontal configuration would use a metal mesh on the surface as the anode and a graphite fracture
electrode as the cathode. This was used successfully at the Columbus, OH site (18,19). The shallower the contaminated
soil the easier it will be to insert the electrodes and treatment zones for the vertical configuration. Where contamination
is deep and is overlain by uncontaminated soil, the horizontal configuration has an advantage because the electrodes and
treatment zones can be placed just where they are needed.
When the contamination extends from the surface to great depth, either configuration could be workable. If the
horizontal configuration is used, it could be necessary to install a number of pairs of electrodes to treat the entire interval,
operating them sequentially with the cathode (lower electrode) being used as the anode for the lower cell when treatment
is completed on the upper cell. This sequential arrangement is necessary because the spacing of the electrodes must
be such that the applied voltage results in a gradient of 0.3 to 0.6 V/cm. The design of the power supply and heating at
the electrodes may limit the voltage that can be applied.
The vertical configuration can use only the electric field to transport water and contaminants; the horizontal
configuration uses both the electric field and gravity. Water is added to the upper electrode (anode) and liquids are
pumped from the lower electrode (cathode). In soils with a wide distribution in pore size or some remaining free product
(liquid phase contaminant), the horizontal configuration has an advantage because it can use the pumping from the
cathode to remove free product before turning on the power to use EK. As indicated above, one of the likely effects of EK
is to move contaminants out of low-permeability zones that would be bypassed by hydraulic flow. Once in higher
permeability zones, contaminants can be transported by the relatively more-rapid hydraulic flow. Only the horizontal
configuration can take advantage.of hydraulic flow as a transport mechanism.
An additional consideration in choosing between the vertical and horizontal configurations is the need to operate
completely in-situ. A vertical configuration can be installed so that liquids can be pumped from the cathode and reinjected
into the anode without ever being brought above the land surface. This meets various regulatory requirements pertaining
to disposition of waters containing hazardous constituents. The top of a vertical configuration anode is accessible for
dealing with any plugging problems resulting from reinjection of cathode fluids. In the horizontal configuration most of the
anode is below ground and inaccessible. Cathode fluids must be treated to remove particulates and floes that would plug
the interface between the bottom of the anode well and the anode. It is difficult to accomplish this entirely below land
21
-------
surface. If fluids are brought above land surface they must be treated to remove contaminants in order to comply with
regulatory requirements and permits obtained for reinjection of the fluids.
Finally, where surface disturbance, needs to be avoided, the horizontal configuration has a definite advantage.
The well heads' and associated equipment are a relatively small installation; most of the control apparatus can be placed
at some distance from the well heads. All of the electrodes and treatment zones are below the ground surface and are
left in place after completion of treatment,
22
-------
7. References
1. U. S. Environmental Protection Agency. 1996. Lasagna™ Public-Private Partnership. (EPA542-F-97-012A,
September 1996), Office of Research and Development and Office of Solid Waste and Emergency Response,
Washington, D.C. 20460, 4p.
2. Brodsky, P.M., and S. V. Ho. 1995. U.S. Patent 5,398,756.
3. Ho, S.V., and P. H. Brodsky. 1995. U.S. Patent 5,476,992.
4. Ho, S. V., P. W. Sheridan, C. J. Athmer, M. A. Heittkamp, J. M. Brackin, D. Weber, and P. H. Brodsky. 1995.
Integrated In Situ Soil Remediation Technology - The Lasagna Process, Environ. Sci. Technol.29(10):2528-2534.
5. Ho, S. V., C. J. Athmer, P. W. Sheridan, and A. P. Shapiro. 1997. Scale-up Aspects of the Lasagna ™ Process
for In-Situ Soil Decontamination. J. Haz. Mat. 55(1-3):39-60.
6. Ho, Sa V., C. Athmer, P. W. Sheridan, B. M. Hughes, R. Orth, D. McKenzie, P. H. Brodsky, A. Shapiro, R.
Thornton, J. Salvo, D. Schultz, R. Landis, R. Griffith, and S.Shoemaker. 1999. The Lasagna Technology for In
Situ Soil Remediation: 1, Small Field Test, Environ. Sci. Technol., 33(7):1086-1091.
7. Ho, Sa V., C. Athmer, P. W. Sheridan, B. M. Hughes, R. Orth, D. McKenzie, P. H. Brodsky, A. Shapiro, R.
Thornton, J. Salvo, D. Schultz, R. Landis, R. Griffith, and S.Shoemaker. 1999. The Lasagna Technology for In
Situ Soil Remediation: 2. Large Field Test, Environ. Sci. Technol., 33(7):1092-1099.
8. U.S: Environmental Protection Agency. 1993. Hydraulic Fracturing Technology: Application Analysis and
Technology Evaluation Report (EPA/540/R-93/505 September 1993) Risk Reduction Engineering Laboratory,
Cincinnati, OH 45268, 129p.
9. U.S. Environmental Protection Agency. 1995. In Situ Remediation Technology Status Report: Hydraulic and
Pneumatic Fracturing. (EPA542-K-94-005, April 1995), Office of Solid Waste and Emergency Response,
Technology Innovation Office, Washington, D.C. 20460; 19p.
10. Murdoch, L C. 1993. Hydraulic Fracturing of Soil During Laboratory Experiments: Part I: Methods and
Observations, Geotechnique, 43(2):255-265.
11. Murdoch, L. C. 1993. Hydraulic Fracturing of Soil During Laboratory Experiments: Part II: Propagation,
Geotechnique, 43(2):267-276.
12. Murdoch, L. C. 1993. Hydraulic Fracturing of Soil During Laboratory Experiments: Part III: Theoretical Analysis,
Geotechnique, 43(2) :277-287.
13. Murdoch, L. C. 1995. Forms of Hydraulic Fractures Created During a Field Test in Overconsolidated Glacial Drift.
Quarterly J. Eng. Geol., Vol. 28, pp.23-35.
14. Chen, J. L. 1997. Characterization of In Situ Electroosmosis Using Conductive Fractures. Ph.D. dissertation,
University of Cincinnati, OH. Bell & Howell Information and Learning (#9818595), Ann Arbor, Ml 48106
15. Chen, J. L., and L C. Murdoch.. 1997. Field Demonstration of In Situ Electroosmosis Between Horizontal
Electrodes. In: Proceedings of the Conference - In situ Remediation of the Geoenvironment, Geotechnical Special
Publication No. 71, American Society of Civil Engineers, Reston, VA. pp 545-559.
23
-------
16. Murdoch, L C., and J. L Chen. 1997. Effects of Conductive Fractures During In-Situ Electroosmosis. J. Haz.
Mat. 55(1-3):239-262.
17. Chen, J. L, and L C. Murdoch. 1999. Effects of Electroosmosis on Natural Soil: Field Test. J. of Geotech. and
Geoenviron. Engrng., ASCE, 125(12):109Q-1098.
18. Kemper, M. C., S. R. AI-Abed, M..H. Roulier, and J-L Chen. 1999. In Situ Degradation of Trichloroethylene
Utilizing Zero Valent Iron-Filled Fractures in a Vertical Electrokinetic Field. Presented at the 1999 Spring Meeting
of the American Geophysical Union, June 1-4,1999, Boston, MA. .
19. Davis-Hoover, W. J., L T. Bryndzia, M. H. Roulier, L. C. Murdoch, M. Kemper, P. Cluxton, S. AI-Abed., W. W.
Slack, and S. J. Vesper. 1999. In Situ Bioremediation utilizing Horizontal LASAGNA™ . In Leeson, A. and
Alleman, B. C. (Ed.) Engineered Approaches for In situ Bioremediation of Chlorinated Solvent Contamination -
5(2), Battelle Press, Columbus, OH. pp 263-267.
20. Roulier, M., M. Kemper, S. AI-Abed, L. Murdoch, P. Cluxton, J-L. Chen, and W. Davis-Hoover. 2000. Feasibility
of electrokinetic soil remediation in horizontal Lasagna™ cells. J. Haz. Mat. B77 (2000) 161-176.
21. Acar, Y. B.; Alshawabkeh, A.N.; 1993. Principles of Electrokinetic Remediation," Environ. Sci. Technol.,
27(13):2638-2647.
22 Probstein, R.F.; Hicks, R.E.; 1993. Removal of Contaminants frorrj Soils by Electric Fields, Science, 260:498-503.
23. . Schultz, D. S. 1997. Electroosmosis technology for soil remediation: laboratory results, field trial, and economic
modeling. J. Haz. Mat. 55(1-3):81-91.
24. Alshawabkeh, A.N.; Yeung A. T.; Bricka M. R.; 1999. Practical aspects of in-situ electrokinetic extraction," J. Env.
Engrng., 125:27-35.
25. Vane, L. N. and G. M. Zang; 1997. Effect of aqueous phase properties on clay particle zeta potential and electro-
osmotic permeability: Implications for electro-kinetic soil remediation processes, J. Haz. Mat., 55:1-22.
26. U. S. Environmental Protection Agency, Office of Solid Waste and Emergency Response. 1995. In Situ
Remediation Technology: Electrokinetics, EPA 542-K-94-007, April 1995. Washington, DC 20460.
27. U. S. Environmental Protection Agency, Office of Radiation and Indoor Air. 1997. Electrokinetic Laboratory and
Field Processes Applicable to Radioactive and Hazardous Mixed Waste in Soil and Groundwater, EPA402-R-97-
006, July 1997. Washington, D.C. 20460
28. Lindgren, E. R., M. G. Hankins, E. D. Mattson, and P. M. Duda. 1998. Electrokinetic Demonstration at the Unlined
Chromic Acid Pit (SAND97-2592), Sandia National Laboratories, Albuquerque, NM 87185.
29. Weitzman, L., I. A. Jefcoat, and B. R. Kim. 1997. In Situ Electrochemically Induced Processes, (Volume 2,
Chapter 2), Innovative Site Remediation Technology, Chemical Treatment, (EPA 542-B-97-005) USEPA Office
of Solid Waste and Emergency Response, Washington, D.C. 20460
30. Natl. Inst. Stand. Technol. 1993. Temperature-Electromotive Force Reference Functions and Tables for the
Letter-Designated Thermocouple Types Based on the ITS-90,. Monograph 175; 630 p.
24
-------
-------
United States
Environmental Protection
Agency
National Risk Management
Research Laboratory
Cincinnati, OH 45268
Official Business
Penalty for Private Use
$300
Please make all necessary changes on the below 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 HEREdJ;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
EPA/600/R-02/033
June 2002
------- |