STATE OF TECHNOLOGY REVIEW
SOIL VAPOR EXTRACTION SYSTEMS
by
Neil J. Hutzler
Blane E. Murphy
John S. Gierke
Department of Civil Engineering
Michigan Technological University
Houghton, MI 49931
Cooperative Agreement No. CR-814319-01-1
Project Officer
Paul R. de Percin
Superfund Technology Demonstration Division
Risk Reduction Engineering Laboratory
Cincinnati, OH 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45268
-------
DISCLAIMER
The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency under assistance agreement CR-
814319-01-1 to Michigan Technological University. It has been subject to the
Agency's peer and administrative review and has been approved for publication
as an EPA document. Mention of trade names or commercial product 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
materials that, if improperly dealt with, can threaten both public health and
the environment. The U.S. Environmental Protection Agency is charged by
Congress with protecting the Nation's land, air, and water systems. 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 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,
implementation, and management of research, development, and demonstration
programs to provide an authoritative, defensible engineering basis in support
of the policies, programs, and regulations of the EPA with respect to drinking
water, wastewater, pesticides, toxic substances, solid and hazardous wastes,
and Superfund-related activities. This publication is one of the products of
that research and provides a vital communication link between the researcher
and the user community.
s
Based on the current state of the technology of soil vapor extraction
system, a number of conclusions can be made. Soil vapor extraction can be
effectively used for removing a wide range of volatile chemicals over a wide
range of conditions. The design and operation of these systems is flexible
enough to allow for rapid change in operation thus, optimizing the removal of
chemicals. While a number of variables intuitively effect the rates of
chemical extraction, no extensive study to correlate variables to extraction
rates has been identified.
E. Timothy Oppelt/ Director
Risk Reduction Engineering Laboratory
-------
ABSTRACT
Soil vapor extraction is a cost-effective technique for the removal of
volatile organic chemicals (VOCs) from contaminated soils. Among the
advantages of the soil air extraction processes are that they creates a
minimal disturbance of the contaminated soil, they can be constructed from
standard equipment, there is demonstrated experience with soil vapor
extraction at pilot- and field-scale, they can be used to treat larger volumes
of soil than can be practically excavated, and there is a potential for
product1recovery.
Unfortunately, there are few guidelines for the optimal design,
installation, and operation of soil vapor extraction systems. A large number
of pilot- and full-scale soil vapor extraction systems have been constructed
and studied under a wide range of conditions. The major objectives of this
report are to critically review available documents that describe current
practices and to summarize this information as concisely as possible. A brief
description of a typical vapor extraction system is presented. The experience
with existing extraction systems has been reviewed, and information about each
system is briefly summarized.
A soil vapor extraction system involves extraction of air containing
volatile chemicals from unsaturated soil. Fresh air is injected or flows into
the subsurface at locations around a spill site, and the vapor-laden air is
withdrawn under vacuum from recovery or extraction wells. A typical soil
vapor extraction system consists of: (1) one or more extraction wells, (2) one
or more air inlet or injection wells (optional), (3) piping or air headers,
(4) vacuum pumps or air blowers, (5) flow meters and controllers, (6) vacuum
gauges, (7) sampling ports, (8) air/water separator (optional), (9) vapor
treatment (optional), and (10) a cap (optional).
Based on the current state of the technology of soil vapor extraction
systems, a number of conclusions can be made. Soil vapor extraction can be
effectively used for removing a wide range of volatile chemicals over a wide
range of conditions. The design and operation of these systems is flexible
enough to allow for rapid changes in operation, thus, optimizing the removal
of chemicals. Intermittent blower operation is probably more efficient in
terms of removing the most chemical with the least energy, especially in
systems where chemical transport is limited by diffusion through air or water*.
Air injection and capping a site have the advantage of controlling air
movement, but injection systems need to be carefully designed. Incremental
installation of wells, while probably more expensive, allows for a greater
degree of freedom in design. While a number of variables intuitively affect
the rate of chemical extraction, no extensive study to correlate variables to
extraction rates has been identified.
iv
-------
CONTENTS
Disclaimer ii
Foreword iii
Abstract iv
Figures vi
Tables vii
1. Introduction 1
2. Process Description 3
System Components 4
System Operation 5
System Variables 5
3. State of the Technology Review 8
Soil Vapor Extraction System Design 8
Well Design and Placement 10
Piping and Blower Systems 14
Miscellaneous Components 16
Site Conditions 19
Soil and Geological Conditions 19
Types and Magnitude of Contamination 21
Extraction System Operation 21
4. Conclusions 25
5. References 27
Appendix -- Soil Vapor Extraction Systems
Site Data Summaries 32
-------
•IGURC:
1. Soil Vapor Extraction System 3
2. Typical Extraction/Air Inlet Well Construction ...... 12
3. Air Flow Patterns in Vicinity of a Single
Extraction Well -- No Cap 13
4. Air Flow from Injection Wells 14
5. Piping Structures 16
6. Air Flow Patterns With Impermeable Cap in Place 18
VI
-------
1. Soil Vapor Extraction System Variables 6
2. List of Typical Pilot and Field
Soil Vapor Extraction Systems 9
3. Site Data Sheet Format 10
4. Pilot and Field Soil Vapor Extraction Systems
-- Well Design and Placement 11
5. Pilot and Field Soil Vapor Extraction Systems
-- Piping and Blower Systems 15
6. Pilot and Field Soil Vapor Extraction Systems
-- Miscellaneous Components 17
7. Pilot and Field Soil Vapor Extraction Systems
-- Soil and Geological Conditions 20
8. Pilot and Field Soil Vapor Extraction Systems
-- Types and Magnitude of Contamination 22
9. Dimensionless Henry's Constants for Typical
Organic Compounds 23
VII
-------
SECTION 1
INTRODUCTION
Soil may become contaminated with volatile organic chemicals such as
industrial solvents and gasoline components in a number of ways. The sources
of contamination at or near the earth's surface include intentional disposal,
leaking underground storage tanks, and accidental spills. Contamination of
groundwater from these sources can continue even after discharge has stopped
because the unsaturated zone above a groundwater aquifer can retain a portion
or all of the contaminant discharge. As rain infiltrates, chemicals elute
from the contaminated soil and migrate towards groundwater.
Alternatives for decontaminating unsaturated soil include excavation with
on-site or off-site treatment or disposal, biological degradation, and soil
washing. Soil vapor extraction is also an accepted, cost-effective technique
for the removal of volatile organic chemicals (VOCs) from contaminated soils
(Bennedsen, 1987; Malot and Wood, 1985; Payne et a/., 1986). Among the
advantages of the soil air extraction process are that it creates a minimal
disturbance of the contaminated soil, it can be constructed from standard
equipment, there is demonstrated experience with the process at pilot- and
field-scale, it can be used to treat larger volumes of soil than are practical
for excavation, and there is a potential for product recovery. With vapor
extraction, it is possible to clean up spills before the chemicals reach the
groundwater table. Soil vapor extraction technology is often used in
conjunction with other clean up technologies to provide complete restoration
of contaminated sites (Malot and Wood, 1985; Oster and Wenck, 1988; CH2M-Hill,
1987).
Unfortunately, there are few guidelines for the optimal design,
installation, and operation of soil vapor extraction systems (Bennedsen,
1987). Theoretically-based design equations which define the limits of this
technology are especially lacking. Because of this, the design of these
systems is mostly empirical. Alternative designs can only be compared by the
actual construction, operation, and monitoring of each design.
A large number of pilot- and full-scale soil vapor extraction systems
have been constructed and studied under a wide range.of conditions. The
information gathered from this experience can be used to deduce the
effectiveness of this technology. One of the major objectives of this report
is to critically review available documents that describe current practices
and to summarize this information as concisely as possible. A brief
description of a typical vapor extraction system is presented. The experience
with existing extraction systems has been reviewed, and information about each
system is briefly summarized in a standard form. The information is further
-------
summarized in several tables, which form the basis for a discussion of the
design, installation, and operation of these systems. Because soil vapor
extraction is a relatively new soil remediation technology, this document will
evolve as more information becomes available.
-------
SECTION 2
PROCESS DESCRIPTION
A soil vapor extraction, forced air venting, or in-situ air stripping
system, such as the one shown conceptually in Figure 1, revolves around the
extraction of air containing volatile chemicals from unsaturated soil. Fresh
air is injected or flows into the subsurface at locations around a spill site,
and the vapor-laden air is withdrawn under vacuum from recovery or extraction
wel1s.
Inlet
Well
Vapor
Treatment
Extraction
Well
Air/Water
Separator
Contaminated
Soil
Groundwater
Table
Figure 1. Soil Vapor Extraction System
3
-------
SYSTEM COMPONENTS
A typical soil vapor extraction system such as the one shown in Figure 1
consists of- (1) one or more extraction wells. (2) one or more air inlet or
injection «dl;> (optional), (3) piping or air headers, (4) vacuum pumps ur air
blowers, (5) flow meters and controllers, (6) vacuum gauges, (7) sampling
ports, (8) air/water separator (optional), (9) vapor treatment (optional), and
(10) a cap (optional). Extraction wells are typically designed to fully
penetrate the unsaturated zone to the capillary fringe. If the groundwater is
at a shallow depth or if the contamination is confined to near-surface soils,
then the extraction wells may be placed horizontally. Extraction wells
usually consist of slotted, plastic pipe placed in permeable packing. The
surface of the augured column for vertical wells or the trench for horizontal
wells is usually grouted to prevent the direct inflow of air from the surface
along the well casing or through the trench.
It may be desirable to also install air inlet or injection wells to
control air flow through zones of maximum contamination. They are constructed
similarly to the extraction wells. Inlet wells or vents are passive and allow
air to be drawn into the ground at specific locations. Injection wells force
air into the ground and can be used in closed-loop systems (Payne et a?.,
1986). The function of inlet and injection wells is to enhance air movement
in strategic locations and promote horizontal air flow to the extraction
wells.
Piping material connecting the wells to headers is usually plastic. The
headers are connected to the blowers or pumps and may be plastic or steel.
Pipes and headers may be buried or wrapped with heat tape and insulated in
northern climates to prevent freezing of condensate.
The pumps or blowers reduce gas pressure in the extraction wells and
induce air flow to the wells. The pressure from the outlet side of the pumps
or blowers can be used to push the exit gas through a treatment system and
back into the ground if injection wells are used.
Gas flow meters are installed to measure the volume of extracted air.
Ball or butterfly valves are used to adjust flow from or into individual
wells. Pressure losses in the overall system are measured with vacuum gauges.
Sampling ports may be installed in the system at each well head, at the
blower, and after vapor treatment. In addition, vapor and pressure monitoring
probes may be placed to measure soil vapor concentrations and the radius of
influence of the vacuum in the extraction wells.
To protect the blowers or pumps and to increase the efficiency of vapor
treatment systems, an air/water separator may need to'be installed. The
condensate may then have to be treated as a hazardous waste depending on the
types and concentrations of contaminants. The need for a separator may be
eliminated by covering the treatment area with an impermeable cap or by
designing the extraction wells to separate water from air within the well
packing. An impermeable cap serves to cover the treatment site to minimize
infiltration and controls the horizontal movement of inlet air.
-------
Vapor treatment may not be required if the emission rates of chemicals
are low or if they are easily degraded in the atmosphere. Typical treatment
systems include liquid/vapor condensation, incineration, catalytic conversion,
or granular activated carbon adsorption.
SYSTEM OPERATION
During operation, the blower is turned on, and the air flows come to
equilibrium. The flows that are finally established are a function of the
equipment, the flow control devices, the geometry of well layout, the site
characteristics, and the air-permeability of the soil. Exhaust air is sampled
on a routine basis and used along with flow measurements to determine the rate
of VOC extraction. Typically, the rate of chemical extraction is high at
first and subsequently decreases because the rate becomes limited by diffusion
of VOC out of immobile air and water zones. In those cases where extraction
is diffusion-limited, the blower can be turned on and off to conserve energy.
At the end of operation, the final distribution of VOCs in the soil can be
measured to ensure decontamination of the site.
SYSTEM VARIABLES
A number of variables characterize the successful design and operation of
a vapor extraction system. They may be classified as site conditions, soil
properties, chemical characteristics, control variables, and response
variables (Anastos et a/., 1985; Enviresponse, 1987). Table 1 lists specific
variables that belong to these groups.
Most site conditions can not be changed. The extent to which VOCs are
dispersed in the soil, vertically and horizontally, is an important
consideration in deciding if vapor extraction is preferable to other methods.
Soil excavation and treatment is probably more cost effective when only a few
hundred cubic yards of near-surface soils are contaminated (Bennedsen, 1987).
If the spill has penetrated more than 20 or 30 feet or has spread through an
area over several hundred square feet at a particular depth or if the spill
volume is in excess of 500 cubic yards, then excavation costs begin to exceed
those associated with a vapor extraction system (CHoM-Hill, 1985; Payne et
a/., 1986). The depth to groundwater is also important. Where groundwater is
at depths of more than 40 feet and the contamination extends to the
groundwater, use of soil vapor extraction systems may be one of the few ways
to remove VOCs from the soil (Malot and Wood, 1985). Groundwater depth in
some cases may be lowered to increase the volume of the unsaturated zone. The
water infiltration rate can be controlled by placing an impermeable cap over
the site. Heterogeneities influence air movement as well as the location of
chemical, and the presence of heterogeneities make it more difficult to
position extraction and inlet wells. There generally-will be significant
differences in the air conductivity of the various strata of a stratified
soil. A horizontally- stratified soil may be favorable for vapor extraction
because the relatively impervious strata will limit the rate of vertical
inflow from the ground surface and will tend to extend the influence of the
applied vacuum horizontally from the point of extraction. The specific
location of the contaminant on a property and the type and extent of
development in the vicinity of the contamination, may favor the installation
-------
TABLE 1. SOIL VAPOR EXTRACTION SYSTEM VARIABLES
Sice Conditions
Distribution of VOCs
Depth to groundwater
Infiltration rate
Location of Heterogeneities
Temperature
Atmospheric pressure
Soil Properties
Permeability (air and water)
Porosity
Organic carbon content
Soil structure
Soil moisture characteristics
Particle size distribution
Chemical Properties
Henry's constant
Solubility
Adsorption equilibrium
Diffusivity (air and water)
Density
Viscosity
Control Variables
Air withdrawal rate
Well configuration
Extraction well spacing
Vent well spacing
Ground surface covering
Pumping duration
Inlet air VOC concentration
and moisture content
Response Variables
Pressure gradients
Final distribution of VOCs
Final moisture content
Extracted air concentration
Extracted air moisture
Extracted air temperature
Power usage
of a soil vapor extraction system. For example, if the contamination extends
across property lines, beneath a building or beneath an extensive utility
trench network, vapor extraction should be considered. Temperature affects
the performance of soil vapor extraction system primarily because of its
influence on chemical properties such as Henry's constant, solubility, and
sorption capacity. In most cases, extraction systems are operated at ambient
temperatures. Atmospheric pressure fluctuations can affect air movement and
depth of the groundwater table (Weeks, 1979).
The soil characteristics at a particular site will have a significant
effect on the applicability of vapor extraction systems." Air conductivity
controls the rate at which air can be drawn from soil by the applied vacuum.
Grain size, moisture content, soil aggregation, and stratification probably
are the most important properties (Bennedsen et a/., 1985; Hutzler et a/.,
1988). The soil moisture content or degree of saturation is also important in
that it is easier to draw air through drier soils. As the size of a soil
aggregate increases, the time required for diffusion of the chemical out of
the immobile regions also increases. However, even clayey or silty soils may
be effectively ventilated by the usual levels of vacuum developed in a soil
vapor extraction system (Camp, Dresser, and McKee, 1987; Terra Vac, 1986b).
The success of the soil vapor extraction in these soils may depend on the
-------
presence of more conductive strata, as would be expected in alluvial settings,
or on relatively low moisture contents in the finer-grained soils.
In conjunction with site conditions and soil properties, chemical
properties will[dictate whether d :>oil vapor extraction vystem is feasible. A
vapor-phase vacuum extraction system is most effective at removing compounds
that exhibit significant volatility at the ambient temperatures in soil. Low
molecular weight, volatile compounds are favored, and vapor extraction is
likely to be most effective at new sites where the more volatile compounds are
still present. It has been suggested that compounds exhibiting vapor
pressures over 0.5 mm of mercury can most likely be extracted with soil air
(Bennedsen et a/., 1985). When expressed in terms of the air-water
partitioning coefficient, compounds which have values of dimensionless Henry's
Law constants greater than 0.01 are more likely to be removed in vapor
extraction systems. Examples of compounds which have been effectively removed
by vapor extraction include trichloroethene, trichloroethane,
tetrachloroethene, and most gasoline constituents. Compounds which are less
applicable to removal include trichlorobenzene, acetone, and heavier petroleum
fuels (Payne et a/., 1986; Bennedsen et a7., 1985; Texas Research Institute,
1980). Soluble compounds tend to travel farther in soils where the
infiltration rate is high. The movement of chemicals with affinity for soil
organic material or mineral adsorption sites will be retarded. In drier
soils, chemical density and viscosity have the greatest impact on organic
liquid movement, however, in most current systems, the contamination is old
enough that no further movement of free product occurs.
Soil vapor extraction processes are flexible in that several variables
can be adjusted during design or operation. These variables include the air
withdrawal rate, the well spacing and configuration, the control of water
infiltration by capping, and the pumping duration. Higher air flow rates tend
to increase vapor removal because the zone of influence is increased and air
is forced through more of the air-filled pores. More wells will allow better
control of air flow but will also increase construction and operation costs.
Intermittent operation of the blowers will allow time for chemicals to diffuse
from immobile water and air and permit removal at higher concentrations.
Parameters responding to soil vapor extraction system performance
include: air pressure gradients, VOC concentrations, moisture content, and
power usage. The rate of vapor removal is expected to be primarily affected
by the chemical's volatility, its sorptive capacity onto soil, the air flow
rate, the distribution of air flow, the initial distribution of chemical, soil
stratification or aggregation, and the soil moisture content.
-------
SECTION 3
STATE OF THE TECHNOLOGY REVIEW
As a part of this investigation, information on 7 pilot-scale, and 10
field (full-scale) studies have been reviewed with respect to the design and
operational variables listed in Table 1. These sites along with their
location, the study type, the duration of study or date the study began, and
the project status are listed in Table 2. This list is by no means complete.
A number of other full-scale soil vapor extraction systems have since been
identified. Full reports were either not available or the available reports
had not been reviewed at the time of this report.
While this technology has been referred to by several names, including
"subsurface venting", "vacuum extraction", "in situ soil air stripping", and
"soil venting", the term "soil vapor extraction" seems to be most descriptive
and is used in this report. Soil vapor extraction technology seems rather
simple in concept, but its application appears to be relatively recent as
indicated by the dates of the available reports. There is a wide variety of
system designs and operating conditions.
From Table 2, it can be seen that soil vapor extraction systems have been
installed at locations across the United States and have been observed over
periods ranging from several weeks to several years. Projects ranging in
status from being complete to being in the preliminary design stage have been
identified. Some of the studies were too short to fully assess the
effectiveness of this technology. Brief descriptions of each study were
compiled in a standard format document of two to six pages (Table 3) to
systematically catalog the information contained in the various reports.
The information on the site data sheets is further summarized in a
number of tables to make it easier to compare specific design and
operational variables. This leads to a more detailed discussion of the
design, installation, and operation of soil vapor extraction systems.
SOIL VAPOR EXTRACTION SYSTEM DESIGN
Tables 4, 5, and 6 summarize the design and operation of the major
components of the pilot- and field-scale systems reviewed for this report.
These include extraction well design and placement, piping and blower
systems and the miscellaneous components discussed previously.
8
-------
SITE
FUEL MARKETING
TERMINAL
VALLEY
MANUFACTURING
INDUSTRIAL
TANK FARM
TIME OIL
COMPANY
SOLVENTS
STORAGE TANK
TCAAP
PILOT 1
TCAAP
PILOT 2
TCAAP
SITE 0
TCAAP
SITE G
GAS
STATION
UNION 76
GAS STATION
SOUTH PACIFIC
RAILROAD
CUSTOM
PRODUCTS
ELECTRONIC
MANUFACTURING
PAINT
STORAGE
THOMAS SOLVENT
COMPANY
HILL AFB
VERTICAL VENTS
HILL AFB
LATERAL SYSTEM
HILL AFB
SOIL PILE
LOCATION
Granger
Indiana
Groveland
Massachusetts
San Juan
Puerto Rico
Tacoma
Washington
Cupertino
California
New Brighton
Minnesota
•i
•I
H
unknown
Bellview
Florida
Benson
Arizona
Stevensville
Michigan
Santa Clara
Valley, CA
Dayton
Ohio
Battle Creek
Michigan
Hill AFB
(3 parallel)
(extraction)
(systems)
STUDY
TYPE
pilot
pilot
pilot/
field
pi lot/
field
pi lot/
field
pilot
pilot
field
field
field
field
field
field
field
- field
field
field
H
•i
DATE OR
DURATION
12 days
10 days
15 days
Jan-Apr 88
30 months
11 days
(Aug 1985)
several
months
67 days
78 days
Feb 1986
Feb 1986
?
7 months
7 months
Dec 1988
>280 days
3 yrs
since
July 1987
since
Jan 1988
Fall 1988
M
H
STATUS
completed
data being
compiled
completed?
pilot
completed
completed?
completed
completed
ongoing
ongoing
completed?
ongoing
completed
completed?
completed?
ongoing
ongoing
one- we 1 1
vent test
completed
REFERENCES
Crow et a I.. 1987
Amer Petr Inst, 1985
Envi response. 1987
Malot & Wood. 1985
Malot, 1985
Woodward-Clyde. 1985
Bennedsen, 1987
Anastos et al.. 1985
Anas t os et a I., 1985
Wenck, 1985
Oster & Wenck. 1988
Wenck, 1985
Oster & Wenck, 1988
Malot & Wood. 1985
Camp, Dresser, &
McKee, 1987, 1988
Johnson, 1988
Johnson & Sterrett, 1988
Payne et al., 1986
Payne & Ltsiecki, 1988
Bennedsen, 1985
Payne ft Lisiecki, 1988
ClUM-Hill. 1987
£
1
Oak Ridge National
Lab, 1988
Radian. Corp., 1987
NAME USED FOR
SYSTEM
Subsurface
Venting
Vacuum
Extraction
Vacuum
Extraction
Soil Gas Vapor
Extraction
Soil Gas Vapor
Extraction
In-situ
Venting
H
«
M
Vacuum
Extraction
Vacuum
Extraction
In-situ Soil
Air Stripping
Forced Air
Circulation
Vapor
Extraction
Enhanced
Volatilization
Vacuum
Extraction
Soil Venting
H
M
-------
TABLE 3. SITE DATA SHEET FORMAT
ASSESSMENT OF PILOT- OR FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Project Name: Each site or project is identified by the name of the
particular site or study.
Principal Investigator(s): In this section, full names of individuals
performing the studies, along with their business addresses are given.
Investigative Report(s): Each report or paper pertaining to the particular
site or project is listed.
Location of Work/Study: The general location of each site or project is
given.
Nature of Contamination: Information on the source of contamination, the
volume of the spill or contaminated soil, the type of contaminants, the levels
of contamination, and the location is summarized as well as possible.
Soil/Site Conditions: Details on the type of soil strata at the site, the
depth of the groundwater table, the porosity and permeability, the moisture
content, and any obstructions present at the site are presented.
Experimental/System Design: The basic components of the particular soil vapor
extraction system are listed. A detailed discussion of the specific system
design follows, including system drawings, if available.
Status of Experimental/Site Clean-up: This section details the final or
current clean-up levels at the site.
Well Design and Placement
Table 4 summarizes information on the design and placement of
extraction and injection wells at the sites listed in Table 2.
Extraction Wells --
Typically, extraction wells are designed to fully penetrate the
unsaturated soil zone or the geologic stratum to be cleaned. An extraction
well usually is constructed of slotted plastic pipe. The well screen is
placed in a permeable packing as shown in Figure 2. Wells may be aligned
vertically or horizontally. Vertical alignment is typical for deeper
contamination zones and results in radial flow patterns. If the depth of the
contaminated soil or the depth to the groundwater table is less than 10 to 15
feet, it may be more practical to dig a trench across the area of
contamination and install perforated piping in the trench bottom versus
10
-------
TABLE 4. PILOT AND FIELD SOIL VAPOR EXTRACTION SYSTEMS -- WELL DESIGN AND PLACEMENT
SITE
FUEL MARKETING
TERMINAL
VALLEY
MANUFACTURING
INDUSTRIAL
TANK FARM
TIME OIL
COMPANY
SOLVENTS
STORAGE TANK
TCAAP
PILOT 1
TCAAP
PILOT 2
TCAAP
SITE 0
TCAAP
SITE G
CAS
STATION
UNION 76
GAS STATION
SOUTH PACIFIC
RAILROAD
CUSTOM
PRODUCTS
ELECTRONIC
MANUFACTURING
PAINT
STORAGE
THOMAS SOLVENT
COMPANY
HILL AFB
VERTICAL VENTS
HILL AFB
LATERAL SYSTEM
HILL AFB
SOIL PILE
NUMBER AND
TYPE
2 wells
8 wells
4 sh. 4 deep
3 wells
7 wells
1 well
9 well
grid
9 well
grid
39 wells
89 wells
vertical &
horizontal
6 wells
3 sh, 3 deep
79 wells
1 well
1 to 2
wells
over
20 wells
14 wells
15 vertical
vent wells
6 laterals
8 laterals
AIW -- air inlet well
BLS -- below land surface
EXTRACTION
WELL
MATERIAL
2" PVC
4" PVC
7
2" PVC
?
3" PVC
3" PVC
3" PVC
3" PVC
?
4" PVC
2" PVC
2" galv.
steel
2" diam.
. galv.
steel
4" PVC
4" PVC
4" poly-
ethylene
4" poly-
ethylene
WELLS
WELL
CONSTRUCTION
screened
14 to 20 ft BLS
up to
30 ft deep
25 to 75 ft BLS
ft at 300 ft BLS
screened
6 to 25 ft BLS
7
grav. pack, slotted
5 to 20 ft BLS
grav. pack, slotted
5 to 20 ft BLS
grav. pack, slotted
5 to 25 - 35 ft BLS
grav. pack, slotted
5 - 25 to 35 ft BLS
?
slotted
10 to 15 ft
15 to 25 ft deep
gravel pack
8 to 25 ft BLS
?
?
7
screened
10 to 30 ft BLS
20 ft BLS
5 ft above
pile bottom
GUT -- ground water table
na -- not applicable
WELL NUMBER AND
SPACING TYPE
20, 40, 4 air inlet
& 100 ft wells
20 ft surface
? surface
40-90 ft surface
na 1 air inlet
well
20 ft 4 wells
50 ft 4 wells
25 ft surface
or air inlet
25 ft surface
or air inlet
? surface
14-50 ft surface
variable surface
& injection
50-70 ft 6 air
inj. wells
? 1 to 2
air inlets
? large no.
of wells
? surface
20 and surface
40 ft or air inlet
15 ft surface
or air inlet
18 ft surface
AIR INPUT
WELL
MATERIAL
2" PVC
na
na
na
?
3" PVC
3" PVC
vents can be
air inlets
vents can be
air inlets
na
na
21 vents were
used as AIW
1.25"
PVC
2" diam.
•
poly-
ethylene
na
vents can be
air inlets
laterals can
be inlets
na
WELL
CONSTRUCTION
screened
14 - 20 ft BLS
na
na
na
?
slotted
15 - 20 ft BLS
slotted
15 - 20 ft BLS
same as
extraction
same as
extraction
na
na
same as
extraction
gravel pack
15 to 25 ft BLS
7
7
na
same as
extraction
same as
extraction
na
? -- no information
sh — shallow
11
-------
installing vertical extraction wells (Oak Ridge National Lab, 1988; Connor,
1988). Usually several wells are installed at a site, especially if soil
strata are highly variable in terms of permeability. In stratified systems,
more than one well may be installed in the same location, each venting a given
strata (Camp, Dresser, and McK.ee, 1987, 1980;. Extraction we!:s can be
installed incrementally starting with installation in the area of highest
contamination (Payne and Lisiecki, 1988; Johnson and Sterrett, 1988). This
allows the system to be brought on-line as soon as possible.
Well spacing is usually based on an estimate of the radius of influence
of an individual extraction well (Malot and Wood, 1985; Wenck, 1985; Oak Ridge
National Lab, 1988). In the studies reviewed, well spacing has ranged from 15
to 100 feet. Johnson and Sterrett (1988) suggest that well spacing should be
decreased as soil bulk density increases or the porosity of the soil
decreases.
Riser
2" to 4" PVC Casing
Concrete
Ground
Surface
Cement-Bentonite Grout
— Bentonite Pellets
Packing Material
Centralizer (optional)
10" Auger Hole
PVC Cap
Figure 2. Typical Extraction/Air Inlet Well Construction.
12
-------
One of the major differences noted between systems was the soil boring
diameter. Larger borings are preferred to provide air/water separation in the
packing.
Air Input -•
In the simplest soil vapor extraction systems, air flows to an extraction
well from the ground surface as depicted in Figure 3. To enhance air flow
through zones of maximum contamination, it may be desirable to include air
inlet wells in the installation. Injection wells or air vents may be located
at numerous places around the site. The function of inlet wells and caps is
to control the flow of air into a contaminated zone. Air vents are passive;
whereas, injection wells force air into the ground at the edge of a site, as
depicted in Figure 4, so as not to force contamination away from the
extraction wells. In addition, injection wells are often installed between
adjacent extraction wells to ensure pressure gradients in the direction of the
extraction wells (Payne et a/., 1986). Typically, injection wells and air
vents are similar in construction to extraction wells. In some installations,
extraction wells have been designed so they can be also be used as air inlets
(Wenck, 1985; Oak Ridge National Lab, 1988).
Figure 3. Air Flow Patterns in Vicinity of a Single Extraction Well -- No
Cap.
Usually, only a fraction of extracted air comes from air inlets (American
Petroleum Institute, 1985; Crow et a7., 1987; Ellgas and Marachi, 1988). This
indicates that air drawn from the surface is the predominant source of clean
air.
Thortan et a7. (1984) investigated the effects of air flow rate, and the
configuration of the inlet and extraction wells on gasoline recovery from an
artificial aquifer. They determined that screening geometry only had an
effect at the low air flow rates. At low flow rates, higher recovery rates
resulted when the screen was placed near the water table versus being screened
13
-------
Figure 4. Air Flow from Injection Wells,
the full depth of the aquifer. A similar assessment was made by Woodward-
Clyde Consultants (1985) at the Time Oil Company site. Woodward-Clyde
engineers suggested that the wells should be constructed with approximately 20
feet of solid pipe between the top of the screen and the soil surface to
prevent the short circuiting of air and to aid in the extraction of deep
contamination.
Piping and Blower Systems
Table 5 summarizes information on the design of piping systems and the
selection of blowers for vapor extraction systems.
Piping --
Piping materials connecting the wells to headers as well as the headers
themselves are usually plastic or steel. Wenck (1985) suggests that headers
be constructed of steel for durability, especially in colder climates.
Headers may be configured as manifold or in a grid as shown in Figure 5,
although, manifold construction appears to be the most common. Pipes and
headers are usually buried or wrapped with heat tape and insulated in northern
climates to prevent freezing of condensate (Wenck, 1985).
Valving --
A control/shut-off valve is usually installed at each wellhead and at
other critical locations, such as lateral/header connections, to provide
operational flexibility and optimize extraction rates. Typically, ball or
butterfly valves are used because they provide better-flow control.
Vacuum Source --
The vacuum for extracting soil air is developed by an ordinary positive
displacement industrial blower, a rotary blower, vacuum or aspirator pump, or
a turbine. There are a large number of commercially available blower models.
In the studies reported herein, the blowers have had ratings ranging from 100
to 6,000 cubic feet per minute at vacuums up to about 30 inches Hg gauge as
14
-------
TABLE 5. PILOT AND FIELD SOIL VAPOR EXTRACTION STSTEHS -- PIPING AND BLOWER SYSTEMS
SITE
FUEL MARKETING
TERMINAL
VALLEY
MANUFACTURING
INDUSTRIAL
TANK FARM
TIME OIL
COMPANY
SOLVENTS
STORAGE TANK
TCAAP
PILOT 1
TCAAP
PILOT 2
TCAAP
SITE D
TCAAP
SITE G
GAS
STATION
UNION 76
GAS STATION
SOUTH PACIFIC
RAILROAD
CUSTOM
PRODUCTS
ELECTRONIC
MANUFACTURING
PAINT
STORAGE
THOMAS SOLVENT
COMPANY
HILL AFB
VERTICAL VENTS
HILL AFB
LATERAL SYSTEM
HILL AFB
SOIL PILE
PIPING
1 £ 2" PVC
PVC manifold
heated
7
2" PVC
manifold
?
3" PVC grid
insulated
3" PVC grid
insulated
8 to 18" steel
insul. manifold
heated
12 to 24" steel
insul. manifold
heated
7
manifold
4" PVC
manifold
2" galv.
steel
duct
galv. St.. heat
manifolds
7
10-16"
metal
manifold
same
same
VACUUM
SOURCE
2 liquid ring
vacuum pumps
blower
vacuum
pump
blower
blowers
2 blowers
1 extr., 1 inj.
2 blowers
1 extr., 1 inj.
up to 4 blowers
variable speed
up to 4 blowers
variable speed
vacuum
pump
vacuum
pomp
3 blowers
separate systems
rotary vane
vac. pump
2 blowers
8 blowers
blower
common source
3 rotary lobe
blowers
1000 cfm
each
250 cfm
aux. blower
AIR
FLOW VACUUM
23 cfm 0.4" Hg
18 cfra 0.3" Hg
40 cfm 0.9" Hg
3 to 800 cfm? 0-29" Hg
18 cfra 25-30" Hg
150 cfm
210 cfm ?
30 cf in/well
10 cfm 0.24" Hg
100 cfm 6" Hg
40 - 55 cfm ?
200-220 to ?
100 to 50 cfm
2200 cfm 1.8" Hg
per blower
5700 cfm 1.8" Hg
per blower
? ?
? ?
86 - 250 0.7-0.6" Hg
cfm
10.2 cfm 4.5" Hg
10 cfm to 0.2 to
100 cfm 3" Hg
? ?
? ?
•up to 9" Hg i
3000 cfm
GAS FLOW
METER
pi tot tube w/
diff. press, roeas
X
?
pi tot tube w/
diff. press, rneas
?
X
X
totalizing
flow meter
totalizing
flow meter
?
?
none
X
7
?
7
orifices
with
Magnehelic
differential
pressure gauges
or U-tube
manometers
X -- listed component present, no detailed information
? -- no information
15
-------
a) Manifold
b) Grid
Figure 5. Piping Structures
shown in Table 5. Ratings of the electric drive motors are usually 10
horsepower or less. The pressure from the outlet side of the pumps or blowers
is usually used to push the exit gas through a treatment system and can be
used to force air back into the ground if injection wells are used (Payne et
a/., 1986), although, it is more common to use a separate blower for injection
(Anastos et a/., 1985). Vapor treatment efficiency can be improved by
installing the blower between the moisture separator and the vapor treatment
system to take advantage of the heat generated by the blower. The blower or
blowers are usually housed in a temporary building on-site.
Gas Flow Meter --
A flow meter should be installed to monitor the volume of extracted air.
This measurement is used in conjunction with gas analysis to determine the
total mass of vapor extracted from the soil. Flow measurements from
individual wells are useful for optimizing extraction system operation. A
flowmeter consisting of an orifice plate and manometer, together with the
appropriate rating curve, will yield the system discharge air flow rate.
Miscellaneous Components
In addition to the basic well, piping, and blower components, a soil
vapor extraction system may require a cover, air/water separator, and vapor
treatment. Table 6 summarizes the range of design of miscellaneous components
at the various pilot- and field systems.
Impermeable Cap --
The surface of the entire site may be sealed with plastic sheeting, clay,
concrete, or asphalt as indicated in Table 6. If movement of the air toward
the extraction well is desired to be more radial than vertical, then an
impermeable cap should be added. The cap controls the air flow pathway so
that make-up or clean air is more likely to come from air vents or injection
16
-------
TABLE o. PILOT AND FIELD SOIL VAPOR EXTRACTION SYSTEMS -- MISCELLANEOUS COMPONENTS
SITE
FUEL MARKETING
TERMINAL
VALLEY
MANUFACTURING
INDUSTRIAL
TANK FARM
TIME OIL
COMPANY
SOLVENTS
STORAGE TANK
TCAAP
PILOT 1
TCAAP
PILOT 2
TCAAP
SITE D
TCAAP
SITE G
GAS
STATION
UNION 76
GAS STATION
SOUTH PACIFIC
RAILROAD
CUSTOM
PRODUCTS
ELECTRONIC
MANUFACTURING
PAINT
STORAGE
THOMAS SOLVENT
COMPANY
HILL AFB
VERTICAL VENTS
HILL AFB
LATERAL SYSTEM
HILL AFB
SOIL PILE
IMPERMEABLE
CAP
plastic
membrane
none
none
none
none
7
7
18" clay
18" clay
concrete
pavement
existing
pavement
none
6 mil-poly-
ethylene
none
clay cover
& concrete
none
.80'x HO'
plastic
concrete
tank pad
none
AIR/WATER
SEPARATOR
none
500 gallon
condenser
55 gallon
tank
55 gallon
tank
none
none
none
none
condenser?
gas/water
separator
none
liquid
trap
none
trap w/pump
. to tank
none
50 gallon
knock-out
drum
M
II
VAPOR
TREATMENT
none
GAC
recovery
tank
none
none
GAC
GAC
none
none/GAC
none
none
none
GAC
none
combustion
GAC
catalytic
incinerator
GAC?
M
GAUGES
vacuum
temperature
vacuum
?
vacuum
temperature
?
vacuum
temperature
vacuum
temperature
vacuum
vacuum
?
7
?
vacuum
vacuum
vacuum
temperature
vacuum
temperature
vacuum
temperature
humidity
•i
M
SAMPLING
PORTS
well heads
exhaust port
well head
system lines
?
well heads
exhaust port
?
inlet ports
exhaust port
inlet ports
exhaust port
well heads
central header
well heads
central header
?
?
7
before and
after GAC
exhaust
we 1 1 heads
vapor, water
exhaust
GAC outlets
well heads
exhaust
H
M
TYPES OF
MONITORING
monitoring well
vapor probes
exhaust gas
monitoring wells
exhaust gas
monitoring wells
soil borings
exhaust gas
exhaust gas
soil borings
air monitoring
soil borings
air monitoring
soil vapor
air monitoring
exhaust gas
soil vapor
air monitoring
exhaust gas
monitoring wells
monitoring wells
soil borings
vapor probes
monitoring wells
soil borings
exhaust gas
soil samples
exhaust gas
monitoring wells
soil borings
monitoring wells
soil borings
air monitoring
pressure
monitoring
wells
soil
borings
GAC -- granular activated carbon
? -- no information
17
-------
wells. This is depicted in Figures 3 and 6. Without the cap (Figure 3), a
more vertical movement of air from the soil surface takes place. But when an
impermeable cap is in place, the radius of influence around the extraction
well is extended (Figure 6). Thus, more of the contaminated soil may be
cleansed by the air flow. If direct flow of air from the ground surface to
the extraction well limits the effectiveness of the extraction system, it may
be necessary to cap or cover the surface. The use of a polyethylene cover
will also prevent or minimize infiltration, which, in turn, reduces the
moisture content and further chemical migration. With little or no
infiltration, water is less likely to be extracted from the system, thus
reducing the need for an air/water separator. In very dry climates, a
reduction of moisture content below which partial drying of the soil occurs,
extraction system efficiency may be reduced due to increased adsorption
capacity of the dry soil (Johnson and Sterrett, 1988).
Air/Water Separator --
If water is pulled from the extraction wells, an air/water separator is
required to protect the blowers or pumps and to increase the efficiency of
vapor treatment systems. The condensate may then have to be treated as a
hazardous waste depending on the types and concentrations of contaminants.
The need for a separator may be eliminated by covering the treatment area with
an impermeable cap. In some cases, a gasoline/water separator may be used in
conjunction with a combination vapor extraction/pumping system for gasoline
product recovery (Malot and Wood, 1985; Thornton et aJ., 1984).
Figure 6. Air Flow Patterns With Impermeable Cap in Place.
Vapor Treatment --
Air emission problems should not be created solving a soil contamination
problem. Vapor treatment may not be required for systems that produce a very
low emission rate of easily degradable chemicals. The decision to treat vapor
must be made in conjunction with air quality regulators. There are several
18
-------
treatment systems available that limit or control air emissions. These
include liquid/vapor condensers, incinerators, catalytic converters, and gas-
phase granular activated carbon (GAC). If air emissions control or vapor
treatment is required for an installation, a vapor phase activated carbon
adsorber system probably will be "the most-practical system' depending on
chemical emission rates and VOC levels, although catalytic oxidation units
have produced favorable results (Bennedsen, 1985). Gas-phase GAC may require
heating of the extracted air to control the relative humidity in order to
optimize the carbon usage rate. As the fraction of water increases, the
capacity for the target chemical decreases and the carbon replacement rate
increases. The spent carbon may be considered as a hazardous waste depending
on the contaminants (Enviresponse, 1987). On one project, where the initial
extraction rate of volatiles was over 200 pounds per day, the extracted gas
was able to be piped to the combustion'air intake zone of a nearby industrial
boiler that was in continuous operation (Bennedsen, 1985). Laboratory
analyses did not detect unwanted volatiles in the boiler emissions.
Incineration can be self-sustained combustion if the vapor contains high
concentrations of hydrocarbons or combustible volatile chemicals. Usually
there is a lag time to achieve a high concentration of combustibles.
Concentration of volatiles in the air stream might be increased by
intermittent blower operation or by intermittently operating different
extraction wells. Some systems have auxiliary fuels to maintain a desired
exhaust temperature.
Pi tot Tubes and Pressure Gauges --
Various monitoring devices such as sampling ports, vacuum gauges, and
pi tot tubes for estimating vapor discharges are required. Pressure gauges are
required to monitor the pressure losses in the overall system to optimize air
f1ows.
Sampling Ports --
Sampling ports are usually installed at each well head, at the blower,
and after gas treatment. The basic measurements required to assess soil vapor
extraction system performance are the system air flow rate and the
concentration of volatile organic chemicals in the extracted flow. A gas
chromatograph equipped with an appropriate detector for the compounds expected
to be present in the exhaust gas is typically used to provide VOC
concentration data.
Monitoring Systems --
Vapor and pressure monitoring probes may be placed in the soil
surrounding the extraction system to measure vapor concentrations and the
radius of influence of the extraction wells. The monitoring wells are usually
required to assess the final clean-up of a particular site.
SITE CONDITIONS
Soil and Geological Conditions
Table 7 briefly summarizes the geologic conditions at the various pilot
and field sites. Although, it has been suggested that soil vapor extraction
systems should be used primarily in highly permeable soils, they have been
19
-------
TABLE 7. PILOT AND FIELD SOIL VAPOR EXTRACTION SYSTEMS -- SOIL AND GEOLOGICAL CONDITIONS
======== - =-=====-
SITE
FUEL MARKETING
TERMINAL
VALLEY
MANUFACTURING
INDUSTRIAL
TANK FARM
TIME OIL
COMPANY
SOLVENTS
STORAGE TANK
TCAAP
PILOT 1
TCAAP
PILOT 2
TCAAP
SITE D
TCAAP
SITE G
GAS
STATION
UNION 76
GAS STATION
SOUTH PACIFIC
RAILROAD
CUSTOM
PRODUCTS
ELECTRONIC
MANUFACTURING
PAINT
STORAGE
THOMAS SOLVENT
COMPANY
HILL AFB .
VERTICAL VEMTS
HILL AFB
LATERAL SYSTEM
HILL AFB
SOIL PILE
SOIL/ GUT
GEOLOGY DEPTH
sand and fine sand layers 25 ft
w/ traces of clay and silt
5 - 12 ft of sand over 27-52 ft
5 - 10 ft of clay
over glacial till
40 - 210 ft clayey silts 300 ft
900 ft limestone
sand and gravel >30 ft
with some silt
unknown 85 ft
4 - 6 ft sand and loamy sand 170 ft
fill over stained low
permeability sediments over sand
same as Pilot 1 "
same as Pilot 1 "
up to 135 ft sand over 130 ft
glacial till and sand
6 - 12 ft clayey soil 8-10 ft
grading to silt & sand
18 - 21 ft clayey sand over 48-53 ft
5-13 ft gumbo clay over 28-42 ft
silty sand over limestone
20 - 25 ft silt & sand, gravel 240 ft
layers, 50 ft silty clay a 40 ft
30 ft of fine sand 30 ft
alluvial clayey 90 ft
silts and sands
sandy soil with clay strata 40-50 ft
sands and gravels
a sand and gravel alluvial 22 ft
deposit over sandstone
4 ft silty sand 600 ft
underlain by 16 - 31 ft
of sand underlain by
discontinuous "
sand and clay layers
mixture of sand "
and silty sand
SOIL HYDRAULIC MOISTURE
POROSITY CONDUCTIVITY CONTENT
0.38 10~3 Ctn/s ?
? permeable to perched
impermeable water
? very ?
permeable
? 3x1 O"3 cm/s ?
? ? ?
? very ?
permeable
? " ?
? " ?
? very ?
permeable
? impermeable? ?
? ? ?
0.1 - 0.3 10~4 cm/s 2 - 5X
777
? relatively ?
impervious
? ? ?
? sand - 0.1 cm/s ?
bedrock-
0.06 cm/s
? permeable to perched
impermeable water
7 M If
? permeable ?
AREA
AFFECTED
two 60 ft2
areas
?
4,400,000
cu yds
30,000 sq
unknown
3800 to
33000
cu yds
it
ii
7
7
unknown
60 x 70 ft
50 acres
7
?
?
90 x 14 ft
II
?
? -- no information
GUT -- groundwater table
20
-------
installed In soils with a wide range of permeabilities. The range of areas
and volumes of soil vented by vapor extraction systems is large. Soil vapor
extraction systems have been used in shallow as well as deep unsaturated
zones. Much of the information needed to fully assess the effects of soil
properties (moisture content, organic carbon content, and porosity) on vapor
extraction is not available.
As the permeability of the soil decreases, more time is required for
extraction and decontamination. In addition to permeability, the presence of
heterogeneities make it more difficult to position inlet and extraction wells.
The effect of clay lens at the Groveland site resulted in perched water table.
During high rainfall periods, the contaminant seeped over the lip of this clay
lens and spread further. Extraction wells had to be installed below this clay
lens to assure an effective extraction operation. Varying strata was also a
concern at the gas station site in Florida (Camp, Dresser, and McKee, 1987).
Some layering of soil can make it easier to extract VOCs from soils where air
channeling occurs through sand layers with subsequent VOC diffusion from less
permeable layers.
The soil moisture content or degree of saturation is also important in
that it is easier to draw air through drier soils. A case in point is that of
the South Pacific Transportation site in Arizona where the soil was relatively
dry (Johnson, 1988; Johnson and Sterrett, 1988). The moisture content was
only 2 to 5 percent. After seven months, 6500 kg of dichloropropene had been
extracted using a moderate air flow rate of 85 to 250 cfm. Higher air flow
rates tend to increase vapor removal because the radius of influence increases
and more air is forced through the air filled pores. In addition, more air is
pulled through the soil in a shorter time period.
Types and Magnitude of Contamination
The types and magnitude of chemical contamination encountered at the
various sites are summarized in Table 8. The common chemical contaminants
extracted were trichloroethylene, 1,1,1-trichloroethane, methylene chloride,
carbon tetrachloride, tetrachloroethylene, dichloroethylene, toluene, 1,3-
dichloropropene, and gasoline along with its constituents (benzene, toluene,
ethylbenzene, and xylene). Most chemicals that have been successfully
extracted have a low molecular weight and high volatility. Another common
screening tool is the air-water partitioning coefficient, expressed in
dimensionless terms as Henry's Law constant (See Table 9). Most of the
compounds have values of Henry's Law constants greater than 0.01. Vapor
extraction can be used to remove large quantities of volatile chemicals as
demonstrated at several sites.
EXTRACTION SYSTEM OPERATION
At most sites, the initial VOC recovery rates were relatively high and
then decreased asymptotically to zero with time (Oster and Wenck, 1988; Payne
et a/., 1985; Payne and Lisiecki, 1988; Terra Vac, 1987b). Vapor extraction
is more effective at those sites where the more volatile chemicals are still
present than when the spill is relatively recent. Several studies have
21
-------
TABLE 8. PILOT AND FIELD SOIL VAPOR EXTRACTION SYSTEMS -- TYPES AND MAGNITUDE OF CONTAMINATION
SITE
FUEL MARKETING
TERMINAL
VALLEY
MANUFACTURING
INDUSTRIAL
TANK FARM
TIME OIL
COMPANY
SOLVENTS
STORAGE TANK
TCAAP
PILOT 1
TCAAP
PILOT 2
TCAAP
SITE 0
TCAA-*
SITE 1
CAS
STATICH
UNION 76
GAS STATION
SOUTH PACIFIC
RAILROAD
CUSTOM
PRODUCTS
ELECTRONIC
MANUFACTURING
PAINT
STORAGE
THOMAS SOLVENT
COMPANY
HILL AFB
VERTICAL VENTS
HILL AFB
'LATERAL SYSTEM
HILL AFB
SOIL PILE
CHEMICALS
IDENTIFIED
gasoline
hydrocarbons
TCE. PCE. MC
OCE. TCA
carb.-n
tetrachloricte
TCE, PCE, TTCJ>
MC, TCA, OCE
TCA, TCE
OCA, OCE
TCE. TCA
OCE, toluene
» others
w
M
TCE. TCA
OCE. tolu*ne
* others
gasoline
benzene, toluene
xylene, KCs
dichloropropenz
PCE
TCA
chl. solvents
acetone, ketones
toluene,, xylenes
PCE. TCE. TCA
jet fuel
CJP4)
M
M
SPILL INITIAL CONTAMINATION
VOLUME LEVELS
>100000 gal 1.6 ft product on GUT
60-110 ppnv a 16 ft, 3500-28000
3 20 ft, 11000-51000 a 21 ft
unknown nax cones: 2500*qTCE/kg
40 ngPCE/kg. 12 njOCE/kg
200, COO 7GX of carbon tet
Ibs contained in insat. zone
•rknown fro* 5 pen at 30 ft
to over. 1000 ppn at 6 in
unk.ot« >10 irgTCA/n3. 1 agTCE/*3 i
unknown 5 - SO ngVOC/kg
itainvd sediments 4-40 ft 3LS
TCE up to 8000 og/ks
• M
U «
unlrno-n >1000 BgVOC/kg
u-Jcw*; up to 10 in of gasoline
Of. OVT. no MC 3 If. ft 815
unknown 0.2 to 12.4 ngBTEX/kg
highest cone. *t IS ft
'50,000 ;bs 30 to 60S of initial spil!
remaining in soil
<50GO 3 to 5600 anPCE/kg soil
cu yd soil 920CJ mg/m in exhaust
unknown 2000 ppnv organics
in initial extracted gas
over 400,000 Total VOC in CU froa
cu yds soil 1 to 62C.OOO ug/L
? 1700 Ibs VOC in 1984
> 25, 000 gal up to 6200 ng/kg fuel
total in upper 5 ft of soil
200 - 900 mg/kg
" between 5 - 10 ft deep
below detection
below SO ft deep
* soil vapor cone, up to
80000 ppb in top 10 ft
FINAL CONTAMINATION
LEVELS
7
being measured
7
initial rate * 2SO Ib/day
current status
unknown
unknown, extract iun rate
decreased with time
not determined
H
"
not determined
no free product
93X reduction of MC in C.
less than 10 ppx in 40
soit sanples
17 ugPCE/kg soi I
after 200 d
50 ppnv in exhaust
(target is 20 pprav)
Total VOC in GU from
net detected to 10 ug/L
not determined
system not yet
operational
AMOUNT
EXTRACTED
190 gallons
being
measured
>70X of spill
volume
240 Ibs
Illb/day
-1000 Ibs
7
>84,000 Ibs
VOCs
> 85. 000 Ibs
VOCs
1200 Ibs of
I gasoline
22,000 Ibs
:n 123 days
90,000 tbs
62 - 76 kg
in 35 days
> 12000 Ibs
VOCs
>7800 Ibs
after 16S days
'
1600 Ibs
in one-
well vent
test
BTEX -- benzene, toluene, ethylbenzene. and xylene
OCA -- dichloroethane
OCE -- dichloroethene
HC -- hydrocarbon
MC -- methylene chloride
PCE -- tetrachlorethene (perchloroethylene)
TCA -- trichloroethane
TCE -- trichloroethene
TTCA -- tetrachloroethane
VOC -- volatile organic chemical
na -- not applicable
ppnv -- parts per million by volume
22
-------
TABLE 9. OIMENSIONLESS HENRY'S LAW CONSTANTS FOR TYPICAL ORGANIC COMPOUNDS.
Component
nonane
n-hexane
2-methylpentane
cyclohexane
chlorobenzene
1 , 2- di chlorobenzene
1 ,3-dichlorobenzene
1 ,4-dichlorobenzene
o-xylene
p-xylene
m-xylene
propy I benzene
ethylbenzene
toluene
benzene
methyl ethylbenzene
1 , 1 -dichloroethane
1,2-dichloroethane
1,1, 1 -trichloroethane
1,1,2- trichloroethane
cis-1,2-dichloroethylene
trans- 1,2-dichloroethyl ene
tet rach loroethy I ene
trichloroethylene
tetralin
decal in
vinyl chloride
chloroethane
hexach I oroethane
carbon tetrachloride
1,3,5-trimethylbenzene
ethyl ene di bromide
1 , 1 -dichloroethylene
methylene chloride
chloroform
1,1,2,2-tetrachloroethane
1 , 2 - d i ch I oropr opane
di brotnoch 1 oromethane
1 ,2,4- tri chlorobenzene
2,4-dimethylphenol
1,1, 2- tr ichlorotr i f luoroethane
methyl ethyl ketone
methyl isobutyl ketone
methyl cellosolve
tri ch lorof luoromethane
10°C
17.21519
10.24304
29.99747
4.43291
0.10501
0.07015
0.09511
0.09124
0.12266
0.18076
0.17689
0.24446
0.14030
0.16397
0.14203
0.15106
0.15838
0.05035
0.41532
0.01678
0.11620
0.25390
0.36410
0.23154
0.03228
3.01266
0.64557
0.32666
0.25522
0.63696
0.17344
0.01291
0.66278
0.06025
0.07403
0.01420
0.05251
0.01635
0.05552
0.35678
6.62785
0.01205
0.02841
1.89798
2.30684
15°C
20.97643
17.46626
29.35008
5.32869
0.11884
0.06048
0.09769
0.09177
0.15267
0.20427
0.20976
0.30915
0.19073
0.20807
0.16409
0.17762
0.19200
0.05498
0.48635
0.02664
0.13787
0.29815
0.46943
0.28208
0.04441
3.53977
0.71049
0:40515
0.23641
0.80776
0.19454
0.02030
0.85851
0.07147
0.09854
0.00846
0.05329
0.01903
0.04441
0.28504
9.09260
0.01649
0.01565
1.53517
2.87580
20°C
13.80119
36.70619
26.31372
5.81978
0,14175
0.06984
0.12222
0.10767
0.19704
0.26813
0.24859
0.36623
0.24983
0.23071
0.18790
0.20910
0.23404
0.06111
0.60692
0.03076
0.14965
0.35625
0.58614
0.35002
0.05654
4.40641
0.90207
0.45727
0.24568
0.96442
0.23736
0.02536
0.90622
0.10143
0.13801
0.03035
0.07898
0.04282
0.07607
0.41986
10.18462
0.00790
0.01206
4.82210
3.34222
25°C
16.92131
31.39026
33.72000
7.23447
0.14714
0.06417
0.11649
0.12957
0.19905
0.30409
0.30409
0.44143
0.32208
0.26240
0.21581
0.22807
0.25545
0.05763
0.71119
0.03719
0.18556
0.38625
0.69892
0.41690
0.07643
4.78211
1.08313
0.49456
0.34129
1.20575
0.27507
0.02657
1.05860
0.12098
0.17207
0.01022
0.14592
0.04823
0.07848
0.20150
13.03840
0.00531
0.01594
1.26297
4.12815
30°C
18.69235
62.70981
34.08841
8.96429
0.19014
0.09527
0.16964
0.15637
0.25164
0.37988
0.35656
0.55072
0.42209
0.32480
0.28943
0.30953
0.31194
0.06995
0.84819
0.05346
0.23114
0.48640
0.98487
0.51454
0.10773
7.99952.
1.12556
0.57484
0.41405
1.51951
0.38711
0.03216
1.27832
0.14512
0.22270
0.02814
0.11497
0.06110
0.11939
0.15074
12.90375
0.00442
0.02734
1.53277
4.90423
Adapted from Howe et al. (1986)
23
-------
indicated that intermittent venting from individual wells is probably more
efficient in terms of mass of VOC extracted per unit of energy expended (Crow
et a/., 1987; Oster and Wenck, 1988; Payne and Lisiecki, 1988). This is
especially true when extracting from soils where mass transfer is limited by
diffusion out of immobile air and water. Optimal operation of a soil vapor
extraction system may involve taking individual wells in and out of service to
allow time for liquid and gas diffusion and to change air flow patterns in the
region being vented. Little work has been done to study this.
One of the major problems in the operation of a soil vapor extraction
system is determining when the site is sufficiently clean to cease operation.
Mass balances using initial and final soil borings have not been particularly
successful in predicting the amount of.chemical actually removed in a system
(Anastos et a/., 1985; Camp, Dresser, and McKee, 1988). Soil vapor
measurements in conjunction with soil boring and groundwater monitoring may be
useful in determining the amount of chemical remaining in the soil. Risk
analysis has been used to evaluate final clean up in at least one system
(Ellgas and Marachi, 1988). Payne and Lisiecki (1988) suggest intermittent
operation near the end of clean up. If there ceases to be a significant
increase in vapor concentration upon restart, one can assume the site has been
decontaminated.
Malot and Wood (1985) discuss use of in-situ soil air extraction in
conjunction with groundwater pumping and treatment as a low-cost alternative
for the clean up of petroleum and solvent spills. Large quantities of organic
chemicals can be retained in the vadose zone by capillary forces, dissolution
in soil water, volatilization, and sorption. If this product can be removed
before it reaches the groundwater then the problem is mitigated. Since vapor
transport is diffusion-controlled in the absence of air extraction, the vapor
spreads horizontally, and a concentration gradient is established in the
vertical direction as vapor diffuses back to the surface. Malot and Wood
(1985) indicate that vapor extraction is effective in removing organic
chemical vapor, sorbed chemical, and free product at the water table. This
suggests that the soil should be decontaminated by vapor extraction before
groundwater clean up can be completed. Vapor extraction becomes more cost-
effective as the depth to groundwater increases, primarily because the cost of
excavation becomes prohibitive.
The design and operation of soil vapor extraction systems can be quite
flexible, allowing for changes to be made during the course of operation, with
regard to well placement or blower size, and air flows from individual wells.
If the system is not operating effectively, changes in the well placement or
the capping the surface may improve it. At one site, the blowers were housed
in modules with quick disconnect attachments. This allowed for portability,
thus improving the removal efficiency by allowing for1 the blowers to be moved
about the site to particular locations where extraction was required the most.
24
-------
SECTION 4
CONCLUSIONS
Based on the current state of the technology of soil vapor extraction
systems, the following conclusions can be made.
1. Soil vapor extraction can be effectively used for removing a wide range of
volatile chemicals over a wide range of conditions.
2. The design and operation of these systems is flexible enough to allow for
rapid changes in operation, thus, optimizing the removal of chemicals.
3. Intermittent blower operation is probably more efficient in terms of
removing the most chemical with the least energy, especially in systems where
chemical transport is limited by diffusion through air or water.
4. Volatile chemicals can be extracted from clays and silts but at a slower
rate. Intermittent operation is certainly more efficient under these
conditions.
5. Air injection and capping a site have the advantage of controlling air
movement, but injection systems need to be carefully designed.
6. Extraction wells are usually screened from a depth of from 5 to 10 below
the surface to the groundwater table. For thick zones of unsaturation,
maximum screen lengths of 20 to 30 feet are specified.
7. Air/water separators are simple to construct and should probably be
installed in every system.
8. Installation of a cap over the area to be vented reduces the chance of
extracting water and extends the path that air follows from the ground sur-
face, thereby increasing the volume of soil treated.
9. Incremental installation of wells, while probably more expensive, allows
for a greater degree of freedom in design. Modular construction, where the
most contaminated zones are vented first, is preferable.
10. Use of soil vapor probes in conjunction with soil borings to assess final
clean up is less expensive than use of soil borings alone. It is usually
impossible to do a complete materials balance on a given site because most
sites have an unknown amount of VOC on the soil and in the groundwater.
25
-------
11. Soil vapor extraction systems are usually only part of a site remediation
system.
12. While a number of variables intuitively affect the rate of chemical
extraction., no extensive study to correlate variables to extraction rates has
been identified.
26
-------
SECTION 5
REFERENCES
Alliance Technologies Corp. 1987. Quality Assurance Project Plan, Terra Vac
Inc., In-Situ Vacuum Extraction Technology, SITE Demonstration Project, Valley
Manufactured Products Site, Grovel and,'MA. Contract No. 68-03-3255. Bedford,
MA. September 1, 1987.
American Petroleum Institute. 1984. Forced Venting To Remove Gasoline Vapor
From a Large-Scale Model Aquifer. American Petroleum Institute. Environmental
Affairs Department. Washington, D.C. June, 1984.
American Petroleum Institute. 1985. Subsurface Venting Of Hydrocarbon Vapors
From an Underground Aquifer. API Publication No. 4410. Health and
Environmental Sciences Department. Washington, D.C. September, 1985.
Anastos, G.J., P.O. Marks, M.H. Corbin, and M.F. Coia. 1985. Task 11. In
Situ Air Stripping of Soils, Pilot Study, Final Report. Report No. AMXTH-TE-
TR-85026. U.S. Army Toxic & Hazardous Material Agency. Aberdeen Proving
Grounds. Edgewood, MD. 88 pp. October 1985.
AWARE, Inc. 1987. Phase I - Zone 1 Soil Decontamination Through In-ihtu
Vapor Stripping Processes, Final Report. Contract Number 68-02-4446. U.S.
Environmental Protection Agency, Small Business Innovative Research Program,
Washington, D.C. April 1987.
Bennedsen, M.B. 1987. Vacuum VOC's from Soil. Pollution Engineering,
19(2):66-68.
Bennedsen, M.B., J.P. Scott, and J.D. Hartley. 1985. Use of Vapor Extraction
Systems for In Situ Removal of Volatile Organic Compounds from Soil.
Proceedings of National Conference on Hazardous Wastes and Hazardous
Materials, HMCRI, pp. 92-95.
Camp, Dresser, and McKee, Inc. 1987. Interim Report For Field Evaluation of
Terra Vac Corrective Action Technology at a Florida Lust Site. Contract No.
68-03-3409. U.S. Environmental Protection Agency. Edison, NJ. December 21,
1987.
CHpM-Hill, Inc. Remedial Planning/Field Investigation Team. 1985. Verona
Well Field - Thomas Solvent Company, Battle Creek, Michigan, Operable Unit
Feasibility Study. Contract No. 68-01-6692. U.S. Environmental Protection
Agency. Chicago, IL. June 17, 1985.
27
-------
CH2M-Hill, Inc. 1987a. Operable Unit Remedial Action, Soil Vapor Extraction
at Thomas Solvents Raymond Road Facility, Battle Creek, MI, Quality Assurance
Project Plan. U.S. Environmental Protection Agency. Chicago, IL. August
1987.
CHoM-Hill, Inc. 1987b. Appendix B - Sampling Plan, Operable Unit Remedial
Action; Creek, MI. U.S. Environmental Protection Agency. Chicago, IL.
October, 1987.
Connor, R. 1988. Case Study of Soil Venting. Pollution Engineering.
20(7):74-78.
Crow, W.L., E.P. Anderson, and E.M. Minugh. 1987. Subsurface Venting of
Vapors Emanating from Hydrocarbon Product on Ground Water. Ground Water
Monitoring Review. 7(l):51-57.
Dynamac Corporation. 1986. Literature Review of Forced Air Venting to Remove
Subsurface Organic Vapors from Aquifers and Soil. Subtask Statement No. 3.
U.S. Air Force Engineering and Services Center, Tyndall AFB, FL, 30 pp., July
28, 1986.
Ellgas, R.A., and N.D. Marachi. 1988. Vacuum Extraction of Trichloroethylene
and Fate Assessment in Soils and Groundwater: Case Study in California.
Proceedings of the 1988 Joint CSCE-ASCE National Conference on Environmental
Engineering. Vancouver, B.C., Canada, pp. 794-801. July 13-15, 1988.
Enviresponse, Inc. 1987. Demonstration Test Plan, In-Situ Vacuum Extraction
Technology, Terra Vac Inc., SITE Program, Groveland Wells Superfund Site,
Grovel and, MA. EERU Contract No. 68-03-3255, Work Assignment 1-R18,
Enviresponse No. 3-70-06340098. Edison, NJ. November 20, 1987.
Howe, G.B., M.E. Mull ins, and T.N. Rogers. 1986. Evaluation and Prediction
of Henry's Law Constants and Aqueous Solubilities for Solvents and Hydrocarbon
Fuel Components Volume I: Technical Discussion. USAFESC Report No. ESL-86-66.
U.S. Air Force Engineering and Services Center, Tyndall AFB, FL. 86 pp.
Hutzler, N.J., J.S. Gierke, and L.C. Krause. 1988. Movement of Volatile
Organic Chemicals in Soils, in Reactions and Movement of Organic Chemicals in
Soils, B.L. Sawhney, ed., Soil Science Society of America, Madison, WI, in
press.
Johnson, J.J. 1988. In-Situ Soil Air Stripping: Analysis of Data From A
Project Near Benson, Arizona. M.S. Thesis, Colorado School of Mines, Golden,
CO. March 16, 1988.
Johnson, J.J., and R.J. Sterrett. 1988. Analysis of In Situ Soil Air
Stripping Data. Proceedings of the 5th National Conference on Hazardous
Wastes and Hazardous Materials. HMCRI. Las Vegas, NV. April 19-21, 1988.
28
-------
Malot, J.J., and P.R. Wood. 1985. Low Cost, Site Specific, Total Approach to
Decontamination. Conference on Environmental and Public Health Effects of
Soils Contaminated with Petroleum Products, University of Massachusetts,
Amherst, MA. October 30-31, 1985.
Malot, J.J., J.C. Agrelot, and M.J. Visser. 1985. Vacuum: Defense System for
Ground.Water VOC Contamination. Fifth National Symposium on Aquifer
Restoration and Ground Water Monitoring, Columbus, Ohio. May 21-24, 1985.
Malot, J.J. 1985. Unsaturated Zone Monitoring and Recovery of Underground
Contamination. Fifth National Symposium on Aquifer Restoration and Ground
Water Monitoring, Columbus, Ohio. May 21-24, 1985.
Markley, D.E. 1988. Cost Effective Investigation and Remediation of
Volatile-Organic Contaminated Sites. Proceedings of the 5th National
Conference on Hazardous Wastes and Hazardous Materials, HMCRI, Las Vegas, NV.
April 19-21, 1988.
Oak Ridge National Laboratory. 1987. Draft: Preliminary Test Plan, In-Situ
Soil Venting Demonstration, Hill AFB, Utah. U.S. Air Force Engineering and
Services Center, Tyndall AFB, FL.
Oster, C.C., and N.C. Wenck. 1988. Vacuum Extraction of Volatile Organics
from Soils. Proceedings of the 1988 Joint CSCE-ASCE National Conference on
Environmental Engineering, Vancouver, B.C., Canada, pp. 809-817. July 13-
15, 1988.
Payne, F.C., C.P. Cubbage, G.L. Kilmer, and L.H. Fish. 1986. In Situ Removal
of Purgeable Organic Compounds from Vadose Zone Soils. Purdue Industrial
Waste Conference. May 14, 1986.
Payne, F.C., and J.B. Lisiecki. 1988. Enhanced Volatilization for Removal of
Hazardous Waste form Soil. Proceedings of the 5th National Conference on
Hazardous Wastes and Hazardous Materials, HMCRI, Las Vegas, NV. April 19-21,
1988.
Radian Corp. 1987. Installation Restoration Program Phase II Draft Report.
U.S. Air Force, Hill AFB, UT. July 1987.
Rollins, Brown, and Gunnell, Inc. 1985. Subsurface Investigation and
Remedial Action, Hill AFB JP-4 Fuel Spill, Provo, Utah. U.S. Air Force, Hill
AFB, UT. December 1985.
Terra Vac, Inc. 1987. Demonstration Test Plan In-Situ Vacuum Extraction
Technology. ' Enviresponse No. 3-70-06340098. Terra Vac, Inc., SITE Program,
Groveland Wells Superfund Site, Groveland, MA. November, 1987.
Terra Vac, Inc. 1987. Union 76 Gas Station Clean-up, Bellview, Florida.
Florida Department of Environmental Regulation, Tallahassee, FL.
29
-------
Texas Research Institute. 1986. Examination of Venting For Removal of
Gasoline Vapors From Contaminated Soil. American Petroleum Institute,
March,1980, (Reprinted in 1986).
Thornton, S.J., and W.L. Wootan. 1982. Venting for the Removal o*
Hydrocarbon Vapors from Gasoline Contaminated Soil. J. Environmental Science
and Health, A17(l):31-44.
Thornton, S.J., R.E. Montgomery, T. Voynick, and W.L. Wootan. 1984. Removal
of Gasoline Vapor from Aquifers by Forced Venting. Proceedings of the 1984
Hazardous Materials Spills Conference, Nashville, TN. pp. 279-286. April
1984.
Towers, D.S., M.J. Dent, and D.G. Van Arnam. 1988. Evaluation of In Situ
Technologies for VHOs Contaminated Soil. Proceedings of the 5th National
Conference on Hazardous Wastes and Hazardous Materials, HMCRI, Las Vegas, NV.
April 19-21, 1988.
Treweek, G.P., and J. Wogec. 1988. Soil Remediation by Air/Steam Stripping.
Proceedings of the 5th National Conference on Hazardous Wastes and Hazardous
Materials, HMCRI, Las Vegas, NV. April 19-21, 1988.
U.S. Army. 1986a. "Twin Cities Army Ammunition Plant In-Situ Volatilization
System, Site G, First Week Operations Report", Twin Cities Army Ammunition
Plant, New Brighton, MN, March 1986.
U.S. Army. 1986b. "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site D, Operations Report," Twin Cities Army Ammunition Plant, New
Brighton, MN, September 8, 1986.
U.S. Army. 1987a. "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site D Operations Report," Twin Cities Army Ammunition Plant, New
Brighton, MN, September 1, 1987.
U.S. Army. 1987b. "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Brighton, MN, October 2, 1987.
U.S. Army. 1987c. "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site G, Emissions Control System Operations Report," Twin Cities Army
Ammunition Plant, New, Brighton, MN, September 1, 1987.
U.S. Army. 1987d. "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site G, Emissions Control System Operations Report," Twin Cities Army
Ammunition Plant, New Brighton, MN, October 2, 1987.
Weeks, E.P. 1979. Barometric Fluctuations in Wells Tapping Deep Unconfined
Aquifers. Water Resources Research. 15(5):1167-1176.
Wenck Associates, Inc. 1985. "Project Documentation: Work Plan,
Volatilization, Sites D and G, Twin Cities Army Ammunition Plant,"
Cartridge Corporation, New Brighton, MN. September 1985.
ISV/In-Situ
Federal
30
-------
Weston, Roy F., Inc. 1985. Appendices -- Task 11, In-Situ Solvent Stripping
From Soils Pilot Study. Installation Restoration General Environmental
Technology Development Contract DAAK11-82-C-0017. U.S. Army Toxic and
Hazardous Materials Agency, Aberdeen Proving Grounds, MD. May 1985.
Woodward-Clyde Consultants. 1985. Performance Evaluation Pilot Scale
Installation and Operation Soil Gas Vapor Extraction System Time Oil Company
Site Tacoma, Washington, South Tacoma Channel, Well 12A Project. Work
Assignment No. 74-ON14.1, Walnut Creek, CA. December 1985.
Wootan, W.L., and T. Voynick. 1984. Forced Venting to Remove Gasoline Vapor
from a Large-Scale Model Aquifer. Texas Research Institute, Inc., Final
Report to American Petroleum Institute.
31
-------
APPENDIX
FILE NAME
API.SHT
AWARE.SHT
BELLVIEW.SHT
BENSON.SHT
CLARA.SHT
CUPER.SHT
DAYTON.SHT
GASSTA.SHT
GROVE.SHT
HILL.SHT
SANJUAN.SHT
STEVEN.SHT
TACOMA.SHT
TCAAP.SHT
TRI.SHT
TYSON.SHT
VERONA.SHT
SOIL VAPOR EXTRACTION SYSTEMS
SITE DATA SUMMARIES
SITE
Petroleum Fuels Marketing Terminal, Granger, IN
AWARE Laboratory Study
Union 76 Station, Bell view, FL
Southern Pacific Chemical Spill, Benson, AZ
Electronics Firm, Santa Clara Valley, CA
Leaking Spent Solvents Storage Tank, Cupertino, CA
Paint Storage Warehouse, Dayton, OH
Gasoline Station, Unknown Location
Valley Manufactured Products, Grovel and, MA
Hill Air Force Base, UT
Industrial Tank Farm, San Juan, PR
Custom Products, Stevensville, MI
Time Oil Company, Tacoma, WA
Twin Cities Army Ammunition Plant, New Brighton, MN
Texas Research Institute Laboratory Study
Tysons Lagoon, PA
Verona Well Field, Battle Creek, MI
-------
API.SHT
ASSESSMENT OF PILOT-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.iect Name: Petroleum Fuels Marketing Terminal, Granger, IN
Principal Investigators: Walter L. Crow
Radian Corporation
P.O. Box 9948
Austin, TX 78766
Edward M. Minugh
Riedel Environmental Services, Inc.
P.O. Box 9948
Portland, OR 97208
Investigative Reports:
American Petroleum Institute, "Subsurface Venting Of Hydrocarbon
Vapors From An Underground Aquifer," Health and Environmental Sciences
Department, Washington, D.C., API Publication No. 4410, September,
1985.
Crow, W.L., E.P. Anderson, and E.M. Minugh, "Subsurface Venting of
Vapors Emanating from Hydrocarbon Product on Ground Water," Ground
Water Monitoring Review. Vol. 7(1), pp 51-57, 1987.
Dynamac Corporation, "Literature Review of Forced Air Venting to Remove
Subsurface Organic Vapors from Aquifers and Soil," Subtask Statement
No. 3, prepared for U.S. Air Force Engineering and Services Center,
Tyndall AFB, FL, 30 pp., July 28, 1986.
Location of Work/Study: A Petroleum Fuels Marketing Terminal located in
Granger, Indiana.
Nature of Contamination: In 1980, a valve rupture caused the loss of
approximately 100,000 gallons of gasoline. Much of the lost gasoline was
recovered using conventional liquid recovery techniques. The remaining
gasoline had migrated to the shallow ground water surface at a depth of
approximately 25 feet. This floating hydrocarbon layer had a thickness of
approximately 1.6 feet in the center of the test cell and extended out
beyond the venting cells. Baseline hydrocarbon vapor concentrations
measured at a depth of 16 feet, varied between 58 and 113 parts per million
by volume (ppmv). At depths between 20 and 20.5 feet, the hydrocarbon vapor
concentrations varied between 3450 and 28,000 ppmv. The soil hydrocarbon
vapors were mcst concentrated at a dspth of 21 feet (1 foot above the
capillary zone). Between depths of 21 to 21.5 feet, the hydrocarbon vapor
concentrations varied from 11,000 to 51,500 ppmv.
-------
API.SHT
Soil/Site Conditions: The soil was quite porous, moderately permeable,
and conducive for evaluating the effectiveness of subsurface venting for
eliminating hydrocarbon vapors in soil. Average soil porosity was 38% and
the average soil permeability was 9.8 x 10~4cm/sec. The soil consisted of
sand and fine sand fractions with some traces of silt and clay interspersed
with traces of coarse gravel.
Experimental/System Design: The soil vapor extraction system consisted of
air inlet and extraction wells; a plastic membrane cap; 2 vacuum pumps;
various plumbing, valves, gauges, sampling ports, and probes; and monitoring
wells. Two parallel test cells with identical dimensions were located
approximately 100 yards southwest of the spill site (See Figure 1). The
venting system was installed in a rectangular area, 110' x 60'. Each test
cell contained one vapor recovery well (VRW) positioned between 2 air inlet
wells (AIW) in a straight line (Figures 2 and 3) with the distance between
first AIW and VRW equal to 20 feet and distance between the second AIW and
VRW equal to 40 feet. Each cell contained 16 vapor sampling probes which
were installed at depths that varied from 16 to 21.5 feet. The vapor probes
were used to obtain discrete samples in the vadose zone at heights ranging
from 0.5 to 5 feet above the capillary fringe, which was at a depth 20 feet.
A groundwater observation well was drilled to a depth of 35 feet and located
midway between the 2 Test Cells. This observation well was used to measure
the depth to the hydrocarbon product and the GWT, which was at 24.6 feet.
Thermocouples attached to the well casing were used to measure soil
temperatures at depths of 16, 20, and 21 feet. Three additional vapor
monitoring probes were installed midway between the 2 cells at 13, 20, and
21 feet to monitor the "patenttal tnflueTYce of each of the "cells on the
other. The vapor recovery and air inlet wells were constructed of 2-inch
PVC casing from the surface to 14 feet below the land surface. A screened
interval from 14 to 20 feet in each AIW and VRW consisted of 0.010 inch
slotted PVC well casing. The bottom of each well screen was fitted with a
solid cap.
The the vapor recovery and air inlet wells were designed to promote
lateral flow of air and vapors through the unsaturated zone. Thus, depths
typical of basements, substructures, and utility vaults could be simulated,
and the control of vapors studied. In order to minimize rainfall
infiltration, simulate a surface structure, and insure that the air inlet
wells were the primary sources of atmospheric air, the two test cells were
covered with a plastic membrane.
Two liquid ring pumps were used as the vacuum source during venting
(Figure 4). Both pumps required circulating water to provide a liquid seal.
This reduced the possibility of vapor ignition due to sparks (because there
were no metal-metal contacts). During operation, the suction side of the
pump was connected to the vapor recovery well by a 2-inch PVC pipe. A
standard pi tot tube was installed between the wellhead and a vacuum control
valve to measure the air flow rate. A septum port located on the vapor
recovery well exhaust line was used to collect vapor samples from each
recovery welK Vapor-treatment was not required because of the relatively
small quantities of hydrocarbons being released to the atmosphere.
-------
API.SHT
Site Monitoring: The following parameters were monitored on a periodic
basis:
- Hydrocarbon concentration and composition from the vapor
recovery well exhaust
- Soil vapor concentration and composition at discrete depths
and distances from vapor recovery well and air inlet wells
- Volumetric flow rates at each vapor recovery and air inlet
well
- Static vacuum measurements at each vapor sampling
probe and vapor recovery well
- Physical characterization of soil core samples
a) porosity
b) soil moisture
c) permeability
d) PH
e) conductivity
f) particle size distribution
- Soil temperatures at three soil depths
- Environmental parameters (rain, temperature, barometric
pressure)
- Liquid hydrocarbon layer thickness and depth to GWT in monitoring
well
Status of Experiment/Site Clean-up: Three experiments were conducted at
different flow rates. During each experiment, only one test cell was
vented. For 12 days during the first experiment, cell A was vented
continuously with an average flow rate of 22.7 ± 2.0 standard cubic feet per
minute (scfm). During the second experiment, a flow rate of 18.5 ± 2.3 scfm
was used to vent cell B for ten days. During the third experiment, a higher
flow rate of 39.8 ±1.0 scfm was used to vent cell A again for 15 days. In
addition, the air inlet wells to cell A were capped after 10 days to
determine if the use of air vents would decrease the system's effectiveness.
The rate of reduction in hydrocarbon vapor concentration was initially
high during the first one or two days of venting. The rate then reduced
dramatically and became relatively constant. The most dramatic reductions
occurred along the lines between the extraction and air inlet wells. The
time that elasped between the end of the first experiment (cell A vented at
the low flow rate) and the beginning of the third experiment (cell A vented
at the high flow rate) was 14 days. At the end of this 14 day rest period,
the omeentratioa of hydrocarbons in the .soil^vapor. AWS measured 4o
determine how much hydrocarbon vapor diffused back to the unsaturated zone.
The average concentration was 62% of the original. The time required for
the vapor concentrations to return to equilibrium was longer than that
required for the reduction of hydrocarbon vapors in each cell. This
-------
suggests that intermittent blower operation is more economical than yet as
effective as continuous venting. The estimated combined volume of recovered
hydrocarbon product during the three experiments was 186 gallons. The
cumulative product recovery rates at a deptr. of 40 feet ranged from 31.2
gallons in cell B (18.5 scfm) to 87.9 gallons in cell A (39.8 scfm). At the
20 to 20.5 feet depth (1 foot above the capillary zone), the soil
hydrocarbon vapor reduction was 99.2%.
The radial influence of the applied vacuum was approximately 30 to 40
feet, and it increased with the increase in flow rate. Even though the flow
rate increased, however, the product recovery rate did not. Only 9 to 11
percent of the air to the extraction wells came from the air inlet wells.
Capping the air inlet wells had no significant effect on the product recovery
rate. It might be better to provide more air inlet wells and to use a
pulsed venting procedure to improve recovery rates. Overall, this pilot
scale system was effective in controlling and recovering hydrocarbon vapors
in sand or gravel formations of high porosity and moderate permeability.
Figure 1. Granger Petroleum Fuels Marketing Terminal Site Plan (Source:
American Petroleum Institute, 1985).
-------
CENTERLINE A
CENTERLINE B
AA1
AB1
QAIW
AA2
A A3
AA6AA4
A AS
AA7
I
o
CM
A AS
• A9(VRW)-
AA10
AA11
A14A AA12 AA15 ~
AA13
AA16
OAIW —
A M1
A M2
A M3
Product
Monitoring
Wells
OAIW
AB2
83 A
LB6 84A AB7
BSA
AB8
• 89 (VRW)
A 810
B11 A
B14A B12A A315
B13A
A Bir.
AA17
I
100 It.
A817
I
A Vapor Sampling Probes
• VRW = Vapor Recovery Well
O AIW = Air Inlet Well
& Product Monitoring Well
A Vapor Monitoring Probe
Probe Depth
13 (eel BLS
16 leet BLS
U-20feetBLS
20 leet BLS
21 leet BLS
Probe Number
Ml
3. 11.
9
1. 2. 4. 6. 7. 8. 10. 12. 14. 15. 16. 17. M2
5. 13. M3
Note: M probes were located midway between centerllnes A and B.
Figure 2. Venting Configuration for Test Cells A and B, Granger, IN
{Source: American Petroleum Institute, 19f35).
-------
A VAPOH SAMPI ING PHODE TO 16 F 1 HLS
8. VAPOH SAMPLING PHOBE TO 20 FT 8LS
C VAI'OH SAMPl ING PHOBE TO 21 FT BLS
SANO/SOIL COVER
FOR MEMUfUNt
CAP
MEMBRANE CAP
ON TEST CELL
FREE PRODUCT
LENS AND
CAPIUAHY ZONE
WATER SATURATED SOIL
BELOW WATER TABLE
Figure 3. Cross-section of Subsurface Venting System, Granger, IN (Source:
American Petroleum Institute, 1985).
r
vftOCJIf
/" ro*t
/
y
~ «T«MOM*
Figure 4. Soil Vapor Extraction System for Granger, IN, Site (Source:
American Petroleum Institute, 1985).
-------
AWARE.SHT
ASSESSMENT OF LAB-SCALE VAPOR EXTRACTION SYSTEMS
-SUMMARY SH4--T
Site/Pro.iect Name: AWARE Lab Study
Principal Investigator: AWARE Incorporated
621 Mainstream Drive
Suite 200, Metro Center
Nashville, TN
Investigative Report:
AWARE, Inc., "Phase I - Zone I Soil Decontamination Through In-Situ
Vapor Stripping Processes," (Contract Number 68-02-4446), Final Report,
Prepared for: U.S. Environmental Protection Agency, Small Business
Innovative Research Program , Washington, D.C. Prepared by: AWARE,
Inc., 621 Mainstream Drive, Suite 200, Metro Center, Nashville, TN
37228, April 1987.
Location of Work/Study: AWARE's Nashville Laboratory
Nature of Contamination: Three chemicals were studied
A) Trichloroethene (TCE)
B) Acetone
C) Chlorobenzene
These chemicals were studied as pure chemical and as chemical saturated
with water.
Soil/Site Conditions: Each chemical was studied on 2 soils:
A) Western Tennessee Loess
B) New Jersey Cohansey Sand
Two laboratory columns were used in parallel for this study with each
packed with one of the soils. Figure 1 shows the column setup.
Experimental/System Design: This VES lab study set-up consisted of two
glass soil columns (with intake and exit ports); a dual vacuum pump; two
flow meters; a humidifying flask; and various valves, sensors, sampling
ports and plumbing. The vapor stripping equipment was designed to allow for
continuous parallel operation of both soil columns. A Cole-Parmer aspirator
pump with dual vacuum sources was used to induce a steady air flow through
each column. Flowrates ranged from 4 to 6 mL/min. A water flask was
attached to the inlet gas end of the column to provide saturated air.
System plumbing allowed for flow measurement and air sampling of the column
exit gases. The air sampling ports were designed to-allow insertion of
Supelco volatile organics purge traps so that direct thermal desorption and
gas chromatography analysis of the stripped volatile organic contaminants
could be used.
-------
AWARE.SHT
All organic concentrations were measured using either methanol
extraction of the soil or desorption of adsorbed vapor from Supelco
multiphase traps. Analyses were performed on a Tracer Model 560 gas
chromatograph using a Hall detector with an overall system detection limit
of 5 nanograms of chemical.
Status of Experiment/Site Clean-up: A total of five experiments were
performed. Each experiment contained three major procedures. The first
procedure consisted of spiking the soil and then performing an analysis of
the companion columns (Note: A companion column is a simple glass column
with a screen covered glass support for the soil and bottom drain tube. The
height-to-diameter ratio was the same as used in the parallel primary
stripping columns. There was one companion column for each primary
stripping column. These companion columns were spiked and then drained in
the same way as the parallel primary stripping columns. Immediately after
stripping was initiated, these companion columns were sacrificed and the
soil was analyzed to determine the initial concentrations in the primary
stripping columns. Therefore, the primary columns remained undisturbed
until the stripping test period was completed. During the stripping period,
the second major procedure of air sampling was performed on the parallel
primary columns. At the end of the test period, the last major procedure
consisted of performing the final soil analysis of each primary column run
was completed.
EXPERIMENT 1: Cohansey Sands and TCE
The duration of the experiment was four weeks. Column 1 was spiked
with pure TCE and Column 2 was spiked with TCE saturated with water. The
initial conditions were as follows:
Column 1 Column 2
% Moisture 4.1 17.4
% Volatiles 0.3 0.3
TCE Cone, (ug/gm) 8,850 15
The residual chemicals (those chemicals that remain in the soil after
spiking/draining and later after the experiment is completed) from the pure
TCE spike resulted in higher soil concentrations and higher air
concentrations (often outside the quantifiable range). Based on residual
soil concentrations (the amount of chemical still remaining after the
experiment was completed and the soil analyzed), TCE removal from the soil
columns was excellent both in the absence and presence of water, 99% and
94%, respectively, over the 4 week period.
The volatiles purge trap apparently developed defects upon excessive
handling so difficulty existed in obtaining reproducible results.
Compaction of the adsorbent and gas-flow short circuiting also lead to
sampling error. In addition, early in the run the exit air concentrations
were frequently-high-enough to exceed-±he-ajuantifiable range of the GC or
the trap. Thus, more credibility should be afforded to the initial and
final bulk soil residue analysis. It is important to point out that Aware
then reduced the sampling periods to the shortest duration which allowed for
reproducible procedures or that two different sampling times (15 and 60
-------
AWARE.SHT
seconds) were used for each sample in order to bracket a usable result.
EXPERIMENT 2r - Western"-*eimess«e -tees-s and TCE
The initial conditions were as follows:
Column 1 Column 2
% Moisture 10.2 24.4
% Volatiles 0.28 0.24
TCE Cone, (ug/gm) 4,010 4.2
As in the experiment 1, a significantly higher TCE residue remained in
the column spiked with pure TCE. However, the removal rate was still ten
times higher than the water/TCE system at the experiment's termination (a
similar situation existed in the first experiment).
Ave. Removal Rate Specific Removal Rate
Column 1 2,349 ug TCE/day 0.9 ug TCE/gm Air-day
Experiment 1
Column 2 31 ug TCE/day 0.012 ug TCE/gm Air-day
Column 1 22,478 ug TCE/day 9.0 ug TCE/gm Air-day
Experiment 2
Column 2 2,349 ug TCE/day 0.9 ug TCE/gm Air-day
This experiment was terminated after 1 month. The column spiked with
pure TCE again exhibited a higher air concentration. Therefore, the average
removal rate was higher. Excellent removal was evidenced in the column
which was spiked with pure TCE. The data from the water saturated system
requires further interpretation, since the data from column 2 are suspect.
EXPERIMENT 3: New Jersey Cohansey Sands and 50/50% V/V Acetone/Water
This experiment was run in duplicate to assess reproducibility. The
initial conditions were as follows:
Column 1 Column 2
% Moisture 14.6 12.6
% Volatiles 0.3 0.28
Acetone Cone, (ug/gm) 43,350 46,050
For column 1, the final soil analysis revealed 97% removal. However,
the data from Column 2 data are suspect.
-------
AWARE.SHT
EXPERIMENT 4: Western Tennessee Loess and 50/50% V/V Acetone/Water
The initial-comiitions^were as-follows:
Column 1 Column 2
% Moisture 7.73 6.30
% Volatiles 1.05 0.97
Acetone Cone, (ug/gm) 87,450 87,300
Experimental difficulty with the traps for both columns as the
experiment progressed, has essentially eliminated data utilization with the
exception of the broad statement that Acetone does strip from the soil based
on 86 to 99% removal in only 12 days.
EXPERIMENT 5: Pure Chlorobenzene On Both Test Soils
Both soils were run simultaneously for 8 days with the following
initial conditions:
Column 1 Column 2
(Tennessee Loess) (Cohansey Sands)
% Moisture 4.4 10.2
% Volatiles 0.38 1.12
Average Chlorobenzene
Concentration In Soil 13,000 mg/kg 33,000 mg/kg
The air flow rate ranged from 3.1 mL/min to 5.6 mL/min with an average
flow rate of 4.9 mL/min. The residual (weighted bulk average) in loess was
271 mg/kg for a 98% reduction while the sand had a residual of 18,170 mg/kg
after eight days for a 45% reduction.
Note: The 60 second sampling period with traps versus the 15 second
period provided the more consistent data. The Chlorobenzene did not exceed
the quantifiable range even at the 60 second sampling interval. The removal
data obtained from the traps were essentially the same indicating
Chlorobenzene removal at the same rate.
Removal rates ranged from 45 to 99% after eight days based on initial
bulk and final bulk residual values.
-------
BtLLVltW.bHI
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SJTE DATA -SHE€T
Site/Pro.iect Name: Union 76 Gasoline Station, Bellview, FL
Principal Investigators: Camp, Dresser, and McKee, Inc.
One Center Plaza
Boston, MA 02108
Terra Vac Corporation
4923 W. Waters Avenue
Tampa, FL 33614
Investigative Reports:
Camp, Dresser, and McKee, Inc., "Interim Report For Field Evaluation of
Terra Vac Corrective Action Technology at a Florida Lust Site,"
Prepared for: U.S. Environmental Protection Agency, Edison, NJ,
Contract No. 68-03-3409, Prepared by: Camp, Dresser, and McKee, Inc.,
One Center Plaza, Boston, MA 02108, December 21, 1987.
Terra Vac Corporation, "Union 76 Gas Station Clean-Up, Bellview,
Florida," Tampa, FL, 1987.
Location of Work/Study: Bellview, Florida
Nature of Contamination: Previous investigations directed by the Florida
Department of Environmental Regulation (FDER) indicated that an underground
storage tank at the Union 76 gas station in Bell view, Florida, was a source
of subsurface hydrocarbons. In December 1986, four 6-1/4" soil borings were
made at the site, and soil samples were obtained every 2-1/2 feet. The
maximum depths sampled were from 52 to 60 feet. In addition, soil samples
were obtained with a hand auger at three locations in the storage tank area.
All samples were analyzed on-site for gasoline components. Benzene,
toluene, ethylbenzene, and xylenes (BTEX) and total volatile hydrocarbons
were quantified by gas chromatography. Initial soil concentrations of BTEX
were as high as 97 mg/kg, and total hydrocarbons were as high as 230 mg/kg.
The highest concentrations of benzene, toluene, and xylene were observed at
approximate depths of 10 to 20 feet, above a clay layer and perched water
table.
Soil/Site Conditions: The soil borings revealed four distinct stratigraphic
zones. Clayey sands were observed from the surface to depths of 18 to 21
feet, where Gumbo (a plastic clay) was encountered-. The thickness of this
clay layer varied from 5 to 13 feet. Beneath this clay layer, silty sand
was observed to depths of 28 to 42 feet below the surface. The silty sand
layer was underlain by a weathered limestone which consisted of sand,
shells, and cavities in the upper portion. Geologists noted that this
-------
BELLVIEW.SHT
limestone layer is probably part of the Upper Eocene Ocala formation
Groundwater was encountered at depths between 48 and 53 feet below the
surface, while perched groundwater was--observed above the clay lav^r in
wells VE-1, ME-1 and ME-2 (Figure 1).
Experimental/System Design: The soil vapor extraction system consisted of
six extraction wells; a vacuum pump; a gas flow meter; various plumbing, valves
gauges, and sampling ports; a gasoline/water separator; and monitoring '
wells (Figure 2). Pavement, which was already in place, was used as a cap A vacuum
extraction/monitoring well was installed in each borehole (see Figure 2 for
locations). The wells, VE-1 and VE-2 were used primarily for subsurface
hydrocarbon vacuum extraction. Multi-level, dual purpose wells, which could
monitor the subsurface vacuum as well as extract hydrocarbons from two to
three hydrogeologic zones, were installed at the other two boreholes (ME-1
and ME-2). Well ME-1, which consisted of three monitoring wells, was
capable of monitoring the subsoil at depths of 13, 35 and 50 feet. Well ME-
2 consisted of two monitoring wells at depths of 16 and 58 feet. Each of
the wells was connected to the vacuum extraction unit by way of a manifold
system. So as not to interfere with the continuous operation of the service
station, well heads were installed in underground valveboxes, and the vacuum
extraction manifold was covered by concrete. The system was modified to
include an in-line air/water separator to separate small quantities of
gasoline product and water that were being extracted from the subsoils along
with the hydrocarbon vapor.
Status of Experiment/Site Clean-uo: A pilot test of the vacuum extraction
operation started on January 29, 1987. By the end of February 1987, the
pilot test results indicated that it was necessary to operate the system
continuously in order to estimate the time required for clean-up. Due to
power outages and numerous administrative problems (for example, approval of
permit to discharge extracted water), the system experienced limited
operation during the months of March, April, and May. During June and July,
the system was operating on a nearly continuous basis. Initial extraction
rates for gasoline hydrocarbons ranged from 295 pounds (39 gallons) per day
in VE-1 to 1950 pounds (260 gallons) per day in ME-1-50. During this pilot
test, wellhead concentrations decreased with time which would indicate the
subsoils were being cleaned up. A total of 22,027 pounds (2937 gallons) of
gasoline hydrocarbons had been extracted from the site as of August 1987.
An independent evaluation of this system was conducted in September through
October 1987 for EPA by Camp, Dresser, and McKee, Inc. Additional soil
borings, soil vapor samples, and groundwater samples were collected. During
this 25 day evaluation period, 22 additional pounds of BTEX were removed.
Taking into account other volatile gasoline components removed, an
additional 200 pounds of hydrocarbons were removed. No significant changes
in the soil, soil vapor, or groundwater samples were noted.
-------
VE-2
M£-1
13'SO'
VE-1
UE-2
16' sa'
VE-7 VE-S
IO
2O
40
SO
6O
Clayey Sand
W«othar«d Limestone
10
20
k.
U*
b.
3O
40
30
-16O
Figure 1. Cross-section of Soil Vapor Extraction System at Bellview, FL
(Source: Terra Vac, 1987).
n.rwssufC Mtuion
WI-VtMTUni LCTCa At CMWI
vHi-vf NIU« tAitnriiru wttts
Figure 2. Soil Vapor Extraction System at Bellview, FL (Source: Camp
Dresser, and McKee, 1988).
-------
UNION 7«
SERVICE STATION
M£-f
Underground
Pip« Una
S E Abshier Avanue
Figure 3. Extraction Well Locations at Bellview, FL (Source: Terra Vac,
1987).
-------
BENSON.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHtET
Slte/Pro.lect Name; Southern Pacific Transportation Company spill
site near Benson, AZ
Principal Investigators: International Technology Corporation
Jeffrey 0. Johnson and R. J. Sterrett
Department of Geological Engineering
Colorado School of Mines
Investigative Reports:
Johnson, J.J., "In-Situ Soil Air Stripping: Analysis of Data From A
Project Near Benson, Arizona," M.S. Thesis, Colorado School of Mines,
March 16, 1988.
Johnson, J.J., and R.J. Sterrett, "Analysis of In Situ Soil Air
Stripping Data," Proceedings of the 5th National Conference on
Hazardous Wastes and Hazardous Materials, HMCRI, Las Vegas, NV, April
19-21, 1988.
"Subsurface Investigation and Assessment of Remedial Action
Alternatives, Benson Train Derailment", Project No. 846292,
International Technology Corporation, San Francisco, CA, 1984.
Location of Work/Study: Three miles east of Benson in Southern Arizona.
Nature of Contamination: On April 8, 1984, a Southern Pacific
Transportation Company train derailed three miles east of Benson, AZ. One
of the derailed cars contained 1,3-dichloropropene (DCP), a VOC used as a
pesticide. Approximately 150,000 pounds of the pesticide leaked into the
soil. Approximately 600 cubic yards of the contaminated soil was excavated
and piled south of the tracks.
Soil borings revealed that the highest DCP concentration was 54,500 ppm
at a depth of 5 feet (approximately 20 feet northwest of the spill). The
area of contamination was 60 feet by 70 feet and centered under the railroad
tracks. At a depth of 20 feet, the area of contamination decreased to 20
feet by 25 feet. At depths between 20 and 25 feet, the DCP concentrations
in the soil were below 10 ppm (which was the EPA established clean-up
level). See Figures 1 and 2 for location of soil borings and site cross-
section with contamination levels.
It was estimated that approximately 45,000 to 90,000 pounds of the
original 150,000 pounds of DCP that had spilled, remained in the soil and
the excavated soil pile. Approximately 60,000 to 105,000 gallons of the
-------
BENSON.SHT
DCP had volatilized into the atmosphere.
Soil/Site Conditions: Heavily stratified deposits, that consisted of silty,
well graded, medium dense to dense dry sands exist from the ground surface
to a depth of about 20 to 25 feet. The overall moisture content ranged from
2 to 5 percent. Occasional gravel layers up to six inches thick were also
present. A 50 foot thick silty clay layer was forty feet below the ground
surface. This layer was typically dry and hard. The groundwater table was
encountered at 240 feet below the ground surface. The soils in the
unsaturated zone were nonhomogeneous based on sieve analysis. The mass of
particles less than 0.002 mm diameter present ranged from 7 to 29 percent by
weight. The average porosity was 30 percent. The average hydraulic
conductivity was 10"4 cm/sec.
Experimental/System Design: The soil vapor extraction system consisted of
79 extraction wells installed incrementally; 3 separate blower systems; various
plumbing, valves, pitot tubes, and monitoring wells. Figures 3 and 4 show
the system design and well layout. The extraction wells varied in depth
from 15 to 25 feet and most of the wells were installed at angles under the
tracks. The wells were made of 2" PVC casing connected to a 2" screen with
0.128 inch slots. The screens extended from a depth of 5 feet to the bottom
of the well. Exhaust gas was vented to the atmosphere. The design of the
blower systems allowed for alterations or the addition of additional wells.
Each extraction well was connected to a 4" PVC manifold header by a 2" PVC
pipe. This header was connected to the intake side of the blower. The 3
blowers were operated independently and were connected to different numbers
of extraction wells (21 connected to the "north" system, 23 to the "south",
and 35 to the "west"). The north blower system was converted to air
injection wells toward the end of the project.
The radius of influence extended beyond the contamination area. The
lowest static pressure was less than 0.04 inches of mercury, in observation
well number 8. On the average, the static pressure from all wells ranged
from 0.04 to 0.48 inches of mercury. Air flow rates varied daily from 86 to
250 cfm.
Status of Experiment/Site Clean-up: This system operated continuously for
seven months. At the end of February, 1985, 44 soil samples were collected
and analyzed. Only four of the samples exceeded 10 ppm DCP, which was the
EPA's clean-up criteria. The maximum concentration recorded was 89 ppm.
Approximately 6500 kilograms of DCP were extracted from the spill site.
Overall, the capital costs for this project were approximately $25,000 and
$500,000 was spent for operational costs.
-------
CLARA.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.iect Name: Electronics Manufacturing Facility in Santa
Clara Valley, CA
Principal Investigators: Woodward-Clyde Consultants
One Walnut Creek Center
100 Pringle Avenue
Walnut Creek, CA
Investigative Report:
Bennedsen, M.B., J.P. Scott, and J.D. Hartley, "Use of Vapor Extraction
Systems for In Situ Removal of Volatile Organic Compounds from Soil,"
Proceedings of National Conference on Hazardous Wastes and Hazardous
Materials, HMCRI, pp. 92-95, 1985.
Location of Work/Study; Santa Clara Valley, California
Nature of Contamination: Soil was contaminated by a spent solvents storage
tank that had been leaking. Several chlorinated solvents were stored in the
tank, but an estimated 80% of the total mass of chemical lost was 1,1,1-
trichloroethane (TCA). An initial boring beneath the tank revealed a high
concentration of solvents at a depth of about 40 feet below the ground
surface. Analysis of the gas that was initially extracted showed that it
contained over 2000 ppmv of organics. The total mass of solvents lost to
the soil was unknown.
Soil/Site Conditions: The soils in the area are predominantly alluvial
clayey silts and sands. Overall, the subsurface soil system was judged to
be relatively impervious. The GWT was at a depth of about 90 feet.
Experimental/System Design: The soil vapor extraction system consisted of
two air inlet and two extraction wells; two blowers; a gas flow meter; and
various plumbing, valves, gauges, and sampling ports. Two borings were
drilled on opposite sides of the tank location. One of the borings was
completed for operation as a vacuum extraction well and the other boring as
an air inlet well. Both well casings were 2 inches in diameter. A nearby
building was operated under a continuous low vacuum to prevent accumulation
of solvent vapors in the work area. A duct to the building ventilation
blower was connected to the extraction well. The resulting vacuum created
at the well head was about 0.2 inches of mercury, and the gas flow rate from
the well was about 10 cubic feet per minute (cfm). A vacuum gauge was
connected to the air inlet well and gave a reading of about 0.012 inches
of mercury when the well was capped to prevent inflow. These data indicate
that the applied vacuum of 0.2 inches of mercury was effective in inducing a
flow of air through the soil over a distance of about 30 feet from the
-------
CLARA.SHT
extraction well. A dedicated blower assembly was added to the system and
the extraction rate was increased to about 100 cfm. An additional
extraction well was installed nearer the-Jeak source and was connected to
the blower. With both wells in service, a vacuum of about 3 inches of
mercury was sufficient to induce a flow rate of 100 cfm through the soil.
Status of Experiment/Site Clean-Up; Approximately 12,000 Ib of VOC's were
extracted over a period of three years of operation. During that time, the
concentration of TCA in the extracted gas decreased from over 2000 to about
50 ppmv. It was expected that this system would remain in operation until
the concentration of VOC's in the extracted gas was less than 20 ppmv.
Since the system was installed prior to 1985, it is presumed that this site
has been cleaned up.
-------
CUPER.SHT
ASSESSMENT OF PILOT-SCALE VAPOR EXTRACTION SYSTEMS
S!TE DATA-SHftT
Site/Project Name: Spent solvents storage tank in Cupertino, CA
Principal Investigator; M.B. Bennedsen
Woodward-Clyde Consultants
One Walnut Creek Center
100 Pr ingle Avenue
Walnut Creek, CA
Investigative Report:
Bennedsen, M.B., "Vacuum VOC's from Soil," Pollution Engineering.
11(2), pp 66-68, Feb. 1987.
Location of Work/Study; A major electronics firm in Cupertino, CA.
Nature of Contamination: In 1983, it was discovered that a storage tank had
leaked. Woodward -Clyde Consultants were hired to perform the site
investigation and determine the appropriate remedial measures. During the
investigation a vacuum extraction system was installed to test its removal
efficiency and to determine the extent of the contamination. A series of
laboratory analyzes of the extracted soil gas, performed over the first
several months of operation indicated that it typically contained over
10,000 ug/cubic meter of 1,1,1-trichloroethane plus a total of about 1000
ug/cubic meter of other VOCs (trichloroethylene, 1,1-dichloroethylene, and
1,2-dichloroethane.
Soil/Site Conditions: The groundwater table (GWT) at this site is
approximately 85 feet below the surface of the ground. No information on
the soil or geological conditions was provided
Experimental /System Design: The soil vapor extraction system consisted of
air inlet and extraction wells; a blower; a gas flow meter; various
plumbing, valves, gauges, and sampling ports; and an air/water separator.
One of the required soil borings was completed with a section of perforated
well casing above the groundwater table, and a vacuum was applied. On the
opposite side of the former tank location, another boring was constructed in
a similar fashion but to act as an air inlet well to the contamination zone
when the vacuum was applied to the extraction well. Initially, 3-in.
diameter, valved PVC pipe was installed connecting the wellhead of the
extraction "well to a 24-in. diameter duct delivering air to the- system's
vacuum source. The test site's vacuum source was a building ventilation
blower, mounted on the roof of an adjacent building. It operated
continuously for several months at an intake vacuum of approximately 0.24
inches of mercury. It pulled approximately 10 cfm of gas from the
-------
CUPER.SHT
surrounding soil upon connection to the 3 in. diameter well casing. The
system extracted VOCs at a rate of approximately 5 kg/day.
The system's capacity was increased in November 1985 by the addition of
a positive displacement blower designed to operate at a vacuum of 6 inches
of mercury and produce an extraction rate of approximately 100 cfm. The
blower speed and capacity rating could be varied by changing pulley sizes on
the belt drive. As installed, the blower had a capacity of about 100 cfm,
but the 5 hp electric drive motor selected for the unit was capable of
driving the blower at its maximum rated capacity of about 300 cfm. At the
same time, the system was connected to additional wells and extended to
connect to other wells located in an area where the presence of
contamination of the soil had been detected. By January 1986, a stable new
operating configuration had been developed. Analysis of the soil gas that
was extracted revealed that during this test a progressive decrease in the
concentration of the extracted soil gas's VOCs occurred.
The blower assembly included a silencer and a sound absorbing cover
because the unit was located in an area where noise levels were of concern.
Also, a 55 gallon air-water separator tank was installed ahead of the blower
to trap water extracted with the soil gas. Bennedsen (1987) noted that air
in soil will normally be nearly saturated with water vapor and that, when
the air expands due to the application of a vacuum, the temperature
decreases enough to cause condensation. Therefore, the air-water separator
is required to protect the blower from the water in the extracted air.
In order to measure the quantity of soil gas extracted, an orifice
plate was connected to a U-tube manometer and installed on the blower
discharge pipe. The concentration of volatiles in the extracted soil gas
were determined by laboratory analysis of samples drawn from a sampling port
on the blower discharge line.
Status of Experiment/Site Clean-up: No information on the current status of
this site clean-up was available.
-------
UAYIUN.bHI
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA
Site/Project Name: Paint Storage Warehouse
Principal Investigators: Dr. Frederick Payne
Midwest Water Resources, Inc.
Charlotte, MI
Investigative Reports:
Payne, F.C., and J.B. Lisiecki, "Enhanced Volatilization for Removal of
Hazardous Waste form Soil," Proceedings of the 5th National Conference
on Hazardous Wastes and Hazardous Materials, HMCRI, Las Vegas, NV,
April 19-21, 1988.
Site visit made by Dr. Neil J. Hutzler, MTU and Mr. Paul de Percin, EPA
Project Officer on December 30, 1987.
Location of Work/Study: Dayton, Ohio
Nature of Contamination:- faint solvents, primarily toluene
Site/Soil Conditions: Sandy soil with clay strata
Experimental/System Design: The soil vapor extraction system consisted of
injection and extraction wells; heated headers from extraction wells; 8
blowers; various plumbing, valves, gauges, and sampling ports; an air/water
separator; and monitoring wells. Concrete, which was already in place, and
a clay cover were used to provide an impermeable cap. Initially, vapor
treatment at this site consisted of burning the extracted vapor. Injection
and extraction wells were installed in two or three separate geologic
strata. The system was installed as four identical modules with 2 blowers
each -- one for extraction and one for injection. The installation of
modules proceeded from the most contaminated area to.those areas with less
contamination. The most contaminated area was capped with a clay cover.
The remaining area is beneath a concrete floor. Vapor treatment was
discontinued when the emission rates decreased to State of Ohio approved
levels. Even though most of the contaminated soil is below the concrete
slab, the concrete did not act as an impermeable barrier. An air/water
separator had to be installed when excessive water was pulled from the soil.
Status of Experiment/Site Clean-up: The site is currently in active
treatment stage and should be nearing completion sometime in 1988.
-------
DAYTON.SHT
Chronological Sequence of Events:
May 1987 - Paint warehouse fire
June 1987 - Design proposed
July 1987 - First unit installed
December 30, 1987 - Site visit made by Dr. N.J. Hutzler of Michigan Tech
and Mr. Paul de Percin of the EPA.
-------
GASSTA.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
S-iiTE OA>A ^HnET
Slte/Pro.iect Name: Gasoline Station
Principal Investigators: James J. Malot
Terra Vac Corporation
356 Fortaleza St.
Box 1591
San Juan, Puerto Rico 00903
Paul R. Wood
Applied Technologies Group
2200 N.E. 124th. Street
North Miami, FL 33181
Investigative Reports:
Malot, J.J., and P.R. Wood, "Low Cost, Site Specific, Total Approach
to Decontamination," presented at the Conference on Environmental and
Public Health Effects of Soils Contaminated with Petroleum Products,
University of Massachusetts, Amherst, MA, October 30-31, 1985.
Visser, M.L., and J.J. Malot, "Removal of Volatile Contaminants from
the Vadose Zone of Contaminated Ground," U.S. Patent No. 4,593,760,
June 10, 1986.
Location of Work/Study: Unknown
Nature of Contamination: An electrical utility company installed a new
manhole and conduit line adjacent to a gasoline station. Company personnel
discovered that explosive vapors had penetrated the new conduit system prior
to making electrical, connections. Testing of the gasoline tanks about 50
feet away confirmed that the source of the subsurface hydrocarbons was a
small leak in one of the storage tanks. A site investigation was initiated
with five test borings around the leaking tank. On-site analysis of soil
samples revealed the highest concentrations of hydrocarbons were located at
or just above the GWT, which was between 8 and 11 feet below the ground
surface. Hydrocarbons were generally nondetectable at depths of about 10 to •
15 feet below the GWT. Based on five test borings around the leaky storage
tank, the lateral extent of contamination in the unsaturated zone was
estimated to be limited to a radius of less than 80 feet from the tank.
Soil/Site Conditions: Concrete pavement covered the site. Beneath that,
clayey soils extended to.a depth of 6 to 15.feet where subsoils graded to
silt and gradually into a fine sandy saprolite below. Wells were installed
in the test borings and were designed so they could be used for both vacuum
extraction and the monitoring of water quality. One well near the tanks had
about 10 inches of gasoline floating on the groundwater.
-------
GASSTA.SHT
Experimental/System Design: The soil vapor extraction system consisted of
an extraction well; a vacuum pump; various plumbing and valves; an air/water
separator; and monitoring wells. The objective of the subsurface clean-up
operation was rapid removal of free gasoline and residual hydrocarbons that
were the source of subsurface vapors and groundwater contamination. After
careful consideration of the entire subsurface problem, a vacuum recovery
system was designed and mobilized to the site. The process proved effective
for all four phases of subsurface hydrocarbon contamination problem by
removing the free product, the residual hydrocarbons above the water table,
subsurface vapors, and contaminated groundwater.
Status of Experiment/Site Clean-Up: The day after the vacuum system was
installed, 22 inches of rain fell during the next three days causing the GWT
to rise to within 1 foot of the surface. As a result of all the rainfall,
the unsaturated zone was reduced to a depth of 1 foot and thus was located
above the intake points of the extraction wells. The vacuum system was used
to recover hydrocarbons from the utility conduit until the water table was
lowered. Careful pumping from one extraction well, without lowering the GWT
below the pre-storm depth of 8 to 11 feet below the ground surface, yielded
about 4600 gallons of water. This dewatered the soil enough for the
vacuum system to commence operation.
When operation started, there was 6 inches of gasoline in one well. At
first the system was used for a three-phase recovery method. The
simultaneous recovery of free gasoline, hydrocarbon vapor, and contaminated
groundwater eliminated the free gasoline within 2 weeks. Hydrocarbon vapor
concentrations dropped 80% as the vacuum operation continued.
The endpoint criteria for unsaturated clean-up dictated by the utility
company was that hydrocarbons be nondetectable or less than 5% of the lower
explosive limit in the utility conduit. A horizontal extraction system was
installed near the utility line and connected to the vacuum extraction
system in order to speed up the clean-up of the unsaturated zone.
After eight weeks of operation, the utility company tested the manhole
and conduit line for hydrocarbon vapors. Even though the utility company's
safety personnel could not detect hydrocarbon vapors in the manhole and
conduit, there was concern that the vapors would return once the vacuum
system was shut off. To ensure the complete removal of hydrocarbons,
operation of the soil vapor extraction system was terminated for one week.
Hydrocarbon vapors were still not detected in the manhole and conduit.
Overall, the vacuum operation removed 1600 pounds of hydrocarbons from
the soils and the groundwater at the site. During this time, all traces of
free gasoline were removed from the shallow GWT. The vapor problem was
eliminated because the residual hydrocarbons and free gasoline had been
effectively removed. In addition, hydrocarbon concentrations in the
groundwater were-reduced over -
-------
GROVE.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
•Si TE DA FA SHEET
Site/Pro.iect Name:
Grovel and Wells
EPA Superfund Site
Principal Investigators:
Terra Vac Corporation
356 Fortaleza St.
Box 1591
San Juan, Puerto Rico 00903
Peter Michaels, Technology Evaluation Manager
Enviresponse Inc.
Building 209, Bay F
GSA Raritan Depot
Woodbridge Avenue
Edison, New Jersey 08837
Investigative Reports:
Enviresponse, Inc., "Demonstration Test Plan, In-Situ Vacuum Extraction
Technology, Terra Vac Inc., SITE Program, Groveland Wells Superfund
Site, Groveland, MA," EERU Contract No. 68-03-3255, Work Assignment 1-
R18, Enviresponse No.'3-70-06340090, "EdTson, NJ, November 20, 1987
Alliance Technologies Corp., "Quality Assurance Project Plan, Terra Vac
Inc., In-Situ Vacuum Extraction Technology, SITE Demonstration Project,
Valley Manufactured Products Site, Groveland, MA," Contract No. 68-03-
3255, Bedford, MA, September 1, 1987
Location Of Work/Study: Valley Manufactured Products Co., Inc., Groveland,
MA, which is located on the north side of Washington Street and is
approximately 400 feet west of Mill Pond. The building is 285' long by 105'
wide.
Nature of Contamination; Valley Manufactured Products Co., Inc. has been in
operation since 1964 and has used different types of cutting oils and
degreasing solvents including trichloroethylene (TCE), tetrachloroethylene
(PCE), trans-l,2-dichloroethylene (DCE), methylene chloride, and 1,1,1-
trichloroethane. The sources of contamination were a leaking underground
storage tank and the previous improper storage and handling practices of
these solvents and oils. The highest concentration is located beneath the
building. The highest concentrations at the location of the oil storage
area were 2500 mg/kg of TCE, 40 mg/kg of PCE and 12 mg/kg of 1,2-DCE at 4 to
12 feet deep. -Theses-depths--genaipally -lie above a-clay- lens. Total VOC
contamination levels ranged from nondetectable to 9 mg/kg to 20.4 mg/kg
within the clay lens. A dense sand layer located below this clay lens and
above the groundwater table had VOC contamination levels up to 20.4 ug/g.
The contamination plume is moving in a north-easterly direction towards Mill
-------
GROVE.SHT
Pond. Two of Groveland's municipal wells have already been contaminated.
Soil/Site Conditions: The ground surface slopes downward northeasterly
towards the Mill Pond, with Johnson Creek and Mill Pond acting as discharge
zones for groundwater flow from the Valley site. The clay lens is
approximately 5 to 12 feet below the surface and has an average thickness of
5 to 10 feet. It extends under the area where the highest levels of VOC
contamination were detected. During periods of increased rainfall, the
contamination under the building is transported vertically through the clay
lens and percolates downward into subsequent subsoil strata due to rising
and falling groundwater tables. The GWT varies from 27' to 52' below the
land surface. Once this contamination reaches the groundwater table, it
moves along with the average groundwater flow field from the Valley site
northeasterly toward Mill Pond.
Experimental/System Design: The demonstration soil vapor extraction system
consisted of extraction wells; an air/water separator; a vacuum pump;
various plumbing, valves, gauges, and sampling ports; and monitoring wells.
Vapor treatment at this site consisted of GAC adsorption. Figure 1 shows
the overall system layout along with the building location and approximate
location of the VOC plume. A conventional industrial blower was used to
create the vacuum applied to the extraction wells located in the
contamination zone. The perforated extraction wells were 24 feet in depth
and were connected by piping to an air/water separator. The exhaust gases
were treated by an activated carbon adsorption system. This absorption
system, located upstream of the blower Intake, consists of manifolded
activated carbon canisters. Backup canisters were provided as insurance
against possible contaminant breakthrough into the exhaust. Subsurface
vacuum and vapor concentrations were monitored by strategically located
monitoring wells which also served to provide supply air to the
contamination zone. These monitoring/injection wells were located around
the extraction wells as shown in Figure 1.
Status of Experiment/Site Clean-up: The demonstration system was in an
active treatment stage with part of the contamination being cleaned up under
the EPA Superfund SITE Program. The demonstration project was temporarily
suspended due to operational problems. The system was restarted, and the
SITE demonstration continued into April 1988. The major operational
problems were that too much water was being extracted from soil and that
carbon usage was excessive. The carbon canisters should have been placed on
exhaust side of blower to heat air to reduce relative humidity and increase
carbon capacity.
A site visit was made by Dr. N.J Hutzler of Michigan Tech on 1/15/88.
-------
. BACKUP CAMOM
MIHAM CAMON
(IIIAUSI
HW-I - GAS MONITORING
WELLS/BORINGS
• El.EZ - ENGINEERING
PARAMETER BORINGS
- EXTRACTION WELLS
x - SURFACE GAS MONITORING
LOCATIONS
Figure 1. Site Plan and Soil Vapor Extraction System Design for Groveland,
MA (Source: Enviresponse, 1987).
-------
HILL.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.iect Name: Hill Air Force Base, Refueling Area, Bldg. 914
Principal Investigators:
David DePaoli and Steve Herbes
Oak Ridge National Laboratory
Martin Marietta Energy Systems, Inc.
Oak Ridge, TN 37831
Lt. Michael Elliot/Capt. Ed Heyse
RDVW
U.S. Air Force Engineering and Services Center
Tyndall AFB, FL 32403
Rollins, Brown and Gunnel1. Inc.
1435 West 820 North
P.O. Box 711
Provo, Utah
Investigative Reports:
Oak Ridge National Laboratory, "Draft: Preliminary Test Plan, In-Situ
Soil Venting Demonstration, Hill AFB, Utah," conducted by Oak Ridge
National Laboratory, for the Air Force Engineering and Services Center,
Tyndall AFB, FL, 1987.
Radian Corp., "Installation Restoration Program Phase II Draft Report,"
prepared for Hill AFB, UT, July 1987.
Rollins, Brown, and Gunnell, Inc., "Subsurface Investigation and
Remedial Action, Hill AFB JP-4 Fuel Spill," Provo, Utah, December
1985.
Location of Work/Study: The contamination area is about 100 yards west of
the north end of Building 914 on Hill Air Force Base, Utah. Building 914 is
about one mile north and 3/4 of a mile west of the South Gate, which is the
Visitor's Entrance. Hill AFB is about 20 miles north of Salt Lake City,
Utah, and about 8 miles south of Ogden, Utah, just a few miles east of 1-15.
Nature of Contamination: The area is contaminated with over 25,000 gallons
of JP-4 (jet fuel). The spill occurred on January 9, 1985. Most of the
fuel infiltrated below the overflowing fuel storage tanks, and the remainder
flowed to the west across an area of approximately 90 feet by 140 feet.
Soil/Site Conditions: The contaminated area is relatively flat and isolated
-------
HILL.SHT
from traffic, residential buildings, and surface water. The area directly
below the tanks is now covered by a concrete pad placed at an elevation of
approximately 15 feet below grade. The soil from the surface to a depth of
approximately 50 feet is a fine sand mixed with gravel. It is relatively
dry. Compact clay layers lies below the sand and can be found to depths of
approximately 600 feet. A groundwater aquifer is below the clay.
Experimental/System Design: This project is still in the preliminary
stages. The soil vapor extraction system is being installed. The site,
when completed, will consist of three extraction configurations operating
in parallel as depicted in Figures 1 and 2. Each extraction system will
consist of extraction wells; a gas flow meter; various plumbing, valves,
gauges, and sampling ports; along with monitoring wells and sensors. A
common blower and catalytic incinerators for vapor treatment will serve all
three systems. The soil excavated from below the tanks was piled on-site on
a plastic membrane liner. The pile is approximately 10 feet high, 52 feet
wide, and 100 feet long (52,000 cubic yards). Eight pipes are buried
horizontally, equally spaced and perpendicular to the long axis of the pile.
The pile is to be covered to prevent blowing dirt.
Prior to the installation of the concrete pad on the area below the
tanks, soil samples were taken to a depth of 25 feet below the excavation.
Vapor and pressure sensors were installed in the augered holes. In
addition, 6 trenches were dug to 5 feet below the excavation, and vent pipes
were installed.
The area west of the tank area, where the fuel seeped into the soil, is
to be vented with a grid work of vertical extraction wells. This system was
designed after a preliminary one-well venting test was performed. The one-
well test equipment consisted of a single extraction well surrounded by
vapor and pressure sensors at various distances from the extraction well and
at various depths and was designed to determine the radius of influence of
the extraction well and the air permeability of the soil. Fourteen
additional extraction wells and 25 additional pressure monitoring points are
being installed. The extraction wells are designed such that they can be
operated as passive air inlets.
The blower/emission control system is common to the three venting
subsystems, will provide vacuum for inducing air flow, and will treat
emissions as necessary to meet regulatory requirements. The catalytic
incinerator unit can be operated at air flows up to 1000 cfm until the
extraction gas reaches a hydrocarbon concentration of 0.0002 g/L, at which
time the gas can be discharged directly. This unit will destroy over 99% of
the hydrocarbons in the extracted gas and will be capable of handling fumes
ranging from 0.25 g/L to 0.0002 g/L. Granular activated carbon may be used
for vapor treatment at lower hydrocarbon concentrations.
Status of Experiment/Site Clean-up: As of summer 1988, this site is in the
pretreatment stage—Horizontal,wells in_the.pile and .under- the -tank pad -are
in place but not operating. Vapor and pressure sensors are in place under
the pad. The pad is in place, and the tanks have been replaced. The pile
has been exposed to the atmosphere for several months. One vertical
extraction well is in place for the radius of influence and hydrocarbon
-------
concentration study. Several vapor and pressure sensors are in place around
the extraction well in a semi-circle at different distances and depths. No
free product has been found and is thought to be immobilized in the soil.
ANALYTICAL
TRAILER
EMISSION
CONTROL
EXCAVATED
SOIL
VERTICAL WELL
ARRAY
LATERAL
PIPE ARRAY
Figure 1. Conceptual Diagram of Soil Vapor Extraction Demonstration Project
at Hill Air Force Base (Source: Oak Ridge National Lab, 1988).
Blo»er/Emssion
Control Systen
Excavated So< We Subsysten
HH-H
o za 40
Verticol Vent Subsysten
100
Figure 2. Soil Vapor Extraction System Layout at Hill Air Force Base
(Source: Oak Ridge National Laboratory, 1988)
-------
HILL.SHT
Chronological Sequence of Events:
01/09/85 -- Spillage of -26,000 gallons of JP-4, 1000 gallons recovered.
12/85 -- Remedial Investigation by Rollins, Brown and Gunnel!. Concluded no
imminent danger.
07/87 -- Radian Corporation investigated site further. Recommended
additional monitoring.
07/22/87 -- D. DePaoli and S. Herbes of ORNL and Capt. E. Heyse of AFESC,
Tyndall AFB visited and selected Hill AFB as study site.
08/18/87 -- OePaoli proposed design for horizontal extraction wells in pile
and under tank pad for installation during tank excavation and pad
construction.
08/31/87 -- DePaoli and Herbes propose soil sampling scheme.
10/87 -- Tanks were excavated, horizontal pipes installed, and soil borings
performed.
01/10-22/88 -- D. DePaoli, H. Jennings, J. Wilson, D. Gillespie, and Capt.
E. Heyse conducted one-well vent test.
01/14-17/88 -- J. Gierke (MTU) visited site and aided in one-well test
preparation.
04/21/88 -- Test plan developed by ORNL reviewed by N.J. Hutzler and
others and tentatively approved by U.S.A.F. Tentative starting date is late
summer 1988.
06/17/88 -- Specifications of vent wells, pressure monitoring wells, and
plastic cover prepared by ORNL.
-------
SANJAUN.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.iect Name: Industrial Tank Farm in San Juan, Puerto Rico
Principal Investigators: James J. Malot
Terra Vac Corporation
356 Fortaleza St.
Box 1591
San Juan, Puerto Rico 00903
Paul R. Wood
Applied Technologies Group
2200 N.E. 124th. Street
North Miami, FL 33181
Investigative Reports:
Malot, J.J., "Unsaturated Zone Monitoring and Recovery of Underground
Contamination," Terra Vac, Inc., P.O. Box 550, Dorado, Puerto Rico
00646, Presented at the Fifth National Symposium on Aquifer Restoration
and Ground Water Monitoring, May 21-24, 1985.
Malot, J.J., Jose C. Agrelot, and Melvin J. Visser, "Vacuum: Defense
System For Ground Water VOC Contamination," Presented at the Fifth
National Symposium on Aquifer Restoration and Ground Water Monitoring,
Columbus, Ohio, May 21-24, 1985.
Malot, J.J., and P.R. Wood, "Low Cost, Site Specific, Total Approach
to Decontamination," presented at the Conference on Environmental and
Public Health Effects of Soils Contaminated with Petroleum Products,
University of Massachusetts, Amherst, MA, October 30-31, 1985.
Location of Work/Study: Approximately 60 km. west of San Juan, Puerto
Rico, on the north side of the island
Nature of Contamination: In mid August 1982, approximately 15,000 gallons
(200,000 pounds) of carbon tetrachloride leaked from-an underground storage
tank. Subsurface investigation revealed extensive contamination in the
unsaturated zone and widespread contamination of the aquifer. Roughly
4,400,000 cubic yards of soil and bedrock were estimated to be contaminated
within the unsaturated zone. Groundwater VOC concentrations reached 3000
ppbv in a water supply well 3000 feet away.
Soil/Site Conditions: Between 40 and 2iO feet of layers of clayey silts and
silty clays with fine grained sands interspersed are present beneath the
tank. About 900 feet of the Aymamon and Aquada Limestone Formations
underlie the fine grained soils. These limestone formations are 90 to 98%
calcium carbonate, very permeable, and riddled with solution channels. They
-------
SANJAUN.SHT
contain bedding planes that dip in the direction of groundwater flow (north-
northeast) at a 4-degree angle. Below these limestone layers are the Ciabo
and Lares Formations which contain an artesian aquifer. The top of the
unconfined aquifer is about 300 feet below the ground surface.
Experimental/System Design: Two systems (Vacuum System I and Vacuum System
II) were installed. Each system consisted of one extraction well; a vacuum
pump; various plumbing and valves; a condenser; and monitoring wells.
While groundwater recovery operations were beginning, a pilot vacuum
extraction system was designed and installed. A vacuum pump, cold water
condenser, and recovery tank made up the above-ground vacuum system. The
extraction well's intake was located and sealed in the unsaturated zone
between depths of 25 and 75 feet. A concrete cover was constructed over the
tank farm to reduce the possibility of contaminant migration into the
aquifer due to infiltration from rainfall.
Status of Experiment/Site-Cleanup: Vacuum System I was installed in
contaminated clayey soils to depths ranging from 75 to 180 feet. The
initial vacuum reading in the clayey soil was 29.9 inches of mercury.
Approximately 3 weeks after this vacuum was applied to the wells, the radius
of influence had moved out three feet and the vacuum was now 26 inches of
mercury at the well head. As a result of subsurface vacuum and vapor flow
rate monitoring of the extraction well, it was determined that the
development of an effective radius of influence of more than a few feet took
several weeks. This subsurface vacuum stabilized about 90 days later with
an influence radius greater than 10 feet. The flow rate of recovered
contaminants increased as the subterranean pressure gradient continued to
propagate. After the first 3 weeks, carbon tetrachloride was being
extracted at a rate of 250 pounds per day.
A similar system, Vacuum II, was developed for vacuum extraction from
the fractured bedrock in the unsaturated zone up to 300 feet deep. Vacuum
System II used air dispersion exhaust stacks to treat the recovered
contaminants. The operation of the two vacuum systems together proved to be
the most cost effective source control alternative at the site.
During the development of the vacuum extraction system, contaminated
groundwater was recovered from the deep aquifer. Initially a downgradient
water supply well was sacrificed to control the migration of the contaminate
in the aquifer. Later a second well was installed to cleanup the existing
contamination in the aquifer.
-------
STEVEN.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.lect Name: Custom Products, Inc.
Stevensville, Michigan
Principal Investigator: Dr. Frederick C. Payne
Midwest Water Resources, Inc.
Charlotte, Michigan
Investigative Reports:
Payne, F.C., C.P. Cubbage, G.L. Kilmer, and L.H. Fish, "In Situ Removal
of Purgeable Organic Compounds from Vadose Zone Soils," presented at
the Purdue Industrial Waste Conference, May 14, 1986.
Payne, F.C., and J.B. Lisiecki, "Enhanced Volatilization for Removal of
Hazardous Waste form Soil," Proceedings of the 5th National Conference
on Hazardous Wastes and Hazardous Materials, HMCRI, Las Vegas, NV,
April 19-21, 1988.
Location of Work/Study: Custom Products, Inc., Stevensville, Michigan.
Nature of Contamination: During mid-1984, a VOC plume was discovered in a
useable aquifer. Perchloroethylene (PCE) was found to be the principal
contaminant reaching levels in excess of 100 ug/L in domestic water wells
and levels of 800 ug/L in nearby industrial production wells. PCE levels on
the unsaturated soil ranged from 8.3 to 5,600 mg/kg dry weight. A PCE tank
sludge discharge location outside an exterior wall had a soil surface
contaminant loading of 110 mg/kg. The estimated volume of contaminated soil
ranged from 1000 to 2000 cubic yards.
Soil/Site Conditions: The unsaturated zone extended to 30-feet below the
surface and consisted of fine sand throughout.
Experimental/System Design: The soil vapor extraction system consisted of
six injection wells and one extraction well; a 6 mil-polyethylene
impermeable cap; a rotary vane vacuum pump; a gas flow meter; various
plumbing, valves, gauges and sampling ports; and GAC vapor treatment.
The system forced clean air into the unsaturated zone at a radius outside
the contaminated region and pulled the clean air toward the center of the
contaminated area where it was withdrawn under reduced pressure. See Figure
1. The removal system's central vacuum extraction well was installed in a
5-inch borehole, 25 feet deep, at the location where the PCE sludge had been
applied to the soil surface. It was gravel-packed from the 17 to 25-foot
depths, and a 2-inch galvanized casing was installed from one foot above
grade to the 17-foot depth. The casing was then gravel-packed from the 8 to
-------
STEVEN.SHT
17-foot depth and cemented from the ground surface to the 8-foot depth.
Five air injection wells were constructed at a 50-foot radius from the
central extraction well on the contaminated side of the building. A sixth
injection well was located at the opposite side of the building at a
distance of 70 feet. Each injection well consisted of a 5-inch diameter,
25-foot deep borehole which was gravel-packed from 19 to 25 feet. One and
one quarter-inch PVC casing was installed from 1 foot above grade to the 19-
foot depth and was then gravel-packed to the 15-foot depth. The injection
wells were cemented from the ground surface down to the 15-foot depth to
allow clean air to be blown into the deeper strata.
The surface of the entire site was sealed with 6-mil polyethylene
sheeting. The cap was covered with sand to secure it to the ground surface
and protect it against puncture. The polyethylene cover provided better
control over the pathway that the air took, resulting in radial movement of
air toward the extraction well rather than vertically from the soil surface.
The blower/activated carbon system was constructed inside a building
on-site. Air flowed from the extraction well through a 2-inch galvanized
pipe to a filtration bed filled with 1200 pounds of granular BPL activated
carbon. The extracted air entered the bottom of the filter bed through a
liquid trap and exited through the tank top. Exhaust air from the filter
bed moved through a 2-inch galvanized pipe to a 1-horsepower oilless rotary
vane vacuum pump. The discharge air from the vacuum pump was distributed to
the injection wells through a manifold with individual valves for each line
that allowed balancing of injection pressures. Overall system cost was
about $60,000.
Status of Experiment/Site Clean-Up; Operation of this system began on
December 11, 1985. The vacuum at the central extraction well was 4.5 inches
of mercury, and the air flow rate was 10.4 cfm. During the first 48 hours
of pumping, PCE recovery was extremely high. Approximately 1.7 pounds of
PCE was removed in liquid form. By 48 hours, levels of the recovered
gaseous PCE reached 92,000 mg/nr and then declined to 6,000 mg/nr by 72
hours. During these three days of operation of this system, a peak PCE
concentration of 180,000 mg/nr air was reached and then declined to 5,000
mg/nr air. Through day 12, gaseous PCE levels remained near 5,000 mg/nr in
the central extraction well. On the 19th day. the collected PCE sample was
1,000 mg/m and by day 35 declined to 10 mg/nr. Figure 2 shows the
accumulated PCE levels recovered at the end of the 35 days. At the end of
45 days PCE levels declined to 10 mg/nr air and the contaminant level in
soil was less than 1 mg/kg soil. According to the authors, the operational
cost of this system was less than 20 percent of the projected excavation
cost.
-------
SrtVEN.SHl
Atr Flow Path
Plaslic Cover
Gr;)*pi Pack '
Soil Cover
Ccmcnl Grout
Air
Inieclion
Well
Withdrawal
Well
m
Air
Injection
Well
Figure 1. Closed Loop Soil Vapor Extraction System at Stevensville, MI
(Source: Payne et al., 1986)
UJ
cr
u
QC
UJ
Q.
60
400
800
PUMPING DURATION Ihrl
Figure 2. Perchloroethylene Removal at Stevensville Site (Source: Payne et
al., 1986).
-------
TACOMA.SHT
ASSESSMENT OF PILOT-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Project Name: South Tacoma Channel Well 12A
Time Oil Company
Principal Investigators: Woodward-Clyde Consultants
One Walnut Creek Center
100 Pringle Avenue
Walnut Creek, CA
Investigative Report:
Woodward-Clyde Consultants, "Performance Evaluation Pilot Scale
Installation and Operation Soil Gas Vapor Extraction System Time Oil
Company Site Tacoma, Washington, South Tacoma Channel, Well 12A
Project," Work Assignment No. 74-ON14.1, Walnut Creek, California,
December 1985.
Location of Work/Study: Time Oil Company
Tacoma, Washington
Nature of Contamination: A remedial investigation by the consulting firm of
CH2M-Hill determined that soils in the vicinity of Time Oil Company property
were contaminated with a large number of VOCs, primarily straight-chain,
polychlorinated, volatile organic solvents such as tetrachloroethane and
trichloroethylene, as listed in Table 1. Contamination levels in the soil
ranged from concentrations up to 5000 ppb at depths to 30 feet, which was
the approximate surface of the GWT at the time of sampling, to values
greater than one million ppb at 6 inches below the ground surface. The
contaminant appeared to cover an approximate area of 30,000 square feet,
including the area covered by a foundation slab that served to support
several storage tanks. The groundwater beneath the site contained
significant levels of VOCs, most likely due to leaching by infiltrating
precipitation percolating through the contaminated soil. It was expected
that, without correction, VOCs in the soil would continue to leach into the
groundwater.
Soil/Site Conditions: The soil at this site is described as sandy and
gravely with some silt and is generally relatively permeable with a
hydraulic conductivity of approximately 3.0 x 10"3 centimeters/second.
Experimeotal/System Design: The system as installed consisted-of.seven
extraction wells; a vacuum pump; a gas flow meter; and various plumbing,
valves, gauges, and sampling ports. The system was designed to allow all of
the wells to operate as either extraction or air-inlet wells. Figure 1
shows the overall well layout, and Figures 2 and 3 diagram the system
-------
TACOMA.SHT
design. The soil-gas extraction wells were constructed of 2 inch PVC casing
and screened from approximately 5 feet above the GWT to 6 feet below the
ground surface. Due to the high permeability of the soils at this site,
short-circuiting of air through the upper six feet appears to have been a
significant problem. The installation of the VES was completed on 7 August
1985. A temporary discharge waiver was granted for operation while data on
the actual discharges from this site were being collected during the month
of August. On 8 August 1985, the system was operated about eight hours per
day for three days to check flows and pressures within the system.
Continuous operation began on 13 August 1985. Data collection began on 14
August 1985 and continued for 10 days. During this period, the system was
operated under balanced flow conditions, with approximately 30 cfm drawn
from each well. Eighty-four vapor samples were collected during this 10 day
period. Samples were analyzed by gas chromatography and mass spectrometry
(GC/MS) to permit detection of the widest possible range of chemicals. At
the on-site mobile Laboratory, a Varian Model 330 Gas Chromatograph was
fitted with an electron capture detector (GC/ECD) for optimum quantification
of the major known soil contaminants: 1,1,2,2-tetrachloroethane, 1,1,2-
trichloroethane, trichloroethylene, and 1,2-dichloroethylene.
An average total volatiles extraction rate of approximately 22 pounds
per day and a peak extraction rate of 25 pounds per day were achieved at
this site. The pilot-scale system was designed to utilize seven soil
borings which had been located at the site for the purpose of determining
soil and contamination characteristics. Since the soil-gas extraction wells
could not be located in areas of highest contamination until the initial
site characterization data were available and the project was limited by the
time available, the wells for the pilot-scale VES project were simply
installed in the existing soil bore holes as an expedient and cost-saving
measure. To optimize this system's operation, the extraction wells should
be located in those areas within the site that have the highest
contamination while, the air-inlet wells should be located in the adjacent
areas of lowest concentration to maximize the air-sweeping effect of the
system.
To enhance system performance, Woodward-Clyde suggests that the wells
should have been constructed with approximately 20 feet of casing between
the top of the screen and the soil surface or that the soil surface should
be capped with plastic or asphalt. In addition, the wells should be
screened to the water table surface to assist in the extraction of deep
contamination.
Status of Fxperiment/Site Clean-up: It was determined that extraction rates
in excess of 25 pounds per day of VOCs could be achieved with the existing
pilot system. Based on these results, Woodward-Clyde engineers recommended
that the existing system be expanded and modified and that it be operated
continuously until the concentration of volatiles in the extracted soil-gas
decreased to levels acceptable with local health officials. The suggested
modifications included the addition of more extraction wells, air inlets,
and-vapor treatment, _S_ix new-exjtr action well s wUh- perforations- fr-om 20
feet below the ground surface to the water table at its lowest seasonal
elevation - approximately 35 feet below the ground surface - were
recommended. Four new air inlet wells with perforations extending from
approximately 10 feet to 35 feet below the ground surface were proposed for
-------
TACOMA.SHT
the system. It was recommended that most of the existing wells should be
operated as air inlet wells, since they were perforated to approximately 5
feet below the ground- surface. A vapor ^yhase carbon-bed-adsorption system
should be added to control emissions to the air." The existing blower
assembly was considered satisfactory for continued operation of the system.
It was further recommended that the existing system be restored to operation
as soon as possible in order to minimize the potential for additional VOCs
being transported to the groundwater. The current status of the project is
unknown.
Tin* O» Bultthg
Concrete T«nk F«rm Pad
WC1B-4
Figure 1. Time Oil Company Site Plan with Well and Piping Locations
(Source: Woodward-Clyde, 1985).
-------
Q.I Inltl fiam Wtlh WCSB-1 TMough WCSB-7
• IH holt B>«>< ••! V.lv.
I hch PVC B«l V.V.
2-I/] hcli PVC Pb»-
14 f««l long
V
B«l V*N* >v
VE8 Mirtlokl-^
r1
66 O.loo
W«l«r 8«p««lor
Figure 2.
Time Oil Company Vapor Extraction System Design (Source:
Woodward-Clyde, 1985).
Figure 3. Time Oil Company Well Vault Installation (Source: Woodward-Clyde,
1985).
-------
TACOMA.SHT
TABLE 1
CONTAMINANTS IDENTIFIED IN SOILS FROM THE TIME OIL COMPANY PROPERTY
MAXIMUM MEASURED CONCENTRATIONS
micrograms/kilogram (ppb)
COMPOUND
Volatiles
1,1,2,2-tetrachloroethane
(1,1,2,2-tetrachloroethane -«-
tetrachloroethylene)
trichloroethylene
trans-1,1-dichloroethylene
methylene chloride
vinyl chloride
toluene
acetone
1,1,1-tri chloroethane
chlorobenzene
2-butanone
1,1-dichloroethylene
(1,1,2-trichloroethane +
cis-l,3-dichloropropene •»•
chl orodibromomethane)
(1,1,1-trichloroethane +
2-chloroethylvinyl ether)
carbon tetrachloride
carbon disulfide
methylbronide
Base Neutrals and Acids
(1,2-dichlorobenzene +
1,4-di chlorobenzene)
2-methyl naphthalene
naphthalene
210,000
1,030,000
106,000
3,920
26,100
28
720
900
3,080
1,430
780
3
3,210
108
5
27
2,200
12,000
7,200
-------
TCAAP.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA- SWEET
Site/Pro.iect Name: Twin Cities Army Ammunition Plant (TCAAP)
Principal Investigators: G.J.Anastos
Roy F. Westdn, Inc.
West Chester, PA
Federal Cartridge
U.S. Army
Investigative Reports:
Anastos, G.J., P.J. Marks, M.H. Corbin, and M.F. Coia, Task 11. In Situ
Air Stripping of Soils, Pilot Study. Final Report, Report No. AMXTH-TE-
TR-85026, U.S. Army Toxic & Hazardous Material Agency, Aberdeen Proving
Ground, Edgewood, MD, 88 pp., October 1985.
Oster, C.C., and N.C. Wenck, "Vacuum Extraction of Volatile Organics
from Soils," Proceedings of the 1988 Joint CSCE-ASCE National
Conference on Environmental Engineering, Vancouver, B.C., Canada, pp.
809-817, July 13-15, 1988.
U.S. Army, "Twin Cities Army Ammunition Plant In-Situ Volatilization
System, Site G, First Week Operations Report", Twin Cities Army
Ammunition Plant, New Brighton, MN, March 1986.
U.S. Army, "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site D, Operations Report," Twin Cities Army Ammunition Plant,
New Brighton, MN, September 8, 1986.
U.S. Army, "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site D Operations Report," Twin Cities Army Ammunition Plant,
New Brighton, MN, September 1, 1987.
U.S. Army, "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site D, Operations Report," Twin Cities Army Ammunition Plant,
New Brighton, MN, October 2, 1987.
U.S. Army, "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site G, Emissions Control System Operations Report," Twin Cities
Army Ammunition Plant, New Brighton, MN, September 1, 1987.
U.S. Army, "Twin Cities Army Ammunition Plant In-Situ Volatilization
System Site G, Emissions Control System Operations Report," Twin
Cities Army Ammunition Plant, New Brighton, MN, October 2, 1987.
-------
TCAAP.SHT
Wenck Associates, Inc., "Project Documentation: Work Plan, ISV/In-Situ
Volatilization, Sites 0 and G, Twin Cities Army Ammunition Plant," for
Federal Cartridge Corporation, New Brighton, MM, prepared by: Wenck
Associates, Inc., (WAI), Twelve Oaks Center, 15500 Wayzata Blvd.,
Wayzata, MN, September 1985.
Weston, Roy F., Inc., Installation Restoration General Environmental
Technology Development Contract DAAK11-82-C-0017, Appendices -- "Task
11, In-Situ Solvent Stripping From Soils Pilot Study," prepared for
U.S. Army Toxic and Hazardous Materials Agency, Aberdeen Proving
Ground, MO, prepared by Roy F. Weston, Inc., West Chester, PA, May
1985.
Location of Work/Study: Twin Cities Army Ammunition Plant
New Brighton, MN
Nature of Contamination: The two prominent sites of interest are sites D
and G. Site D is located on the Arsenal Sand kame deposit and is believed
to have been used for open burning prior to 1970. The extent of the stained
sediments (gray-black), presumed to indicate burning or disposal, was mapped
by Weston (4 June 1984 and May 1985). Soil samples indicate volatile
organics, with TCE the most prominent, at concentrations up to 8,000 ppm.
This was mainly in the upper 10 feet of the soil. However, soil borings as
deep as 120 and 140 feet revealed TCE levels at 1,000 and 400 ppm,
respectively. In addition, excessive levels of barium; chromium, lead,
phenolics, and PCB's have also been found in the soils at site D.
Site G is located on the boundary between the Arsenal Sand kame deposit
and the Twin Cities Formation. From the 1940's into the 1970's, site G
was used as an open dump. Magnetic abnormalities in two regions of the
site indicate buried drums or other metallic waste. Volatile organic
concentrations up to 1,000 ppm were observed in soil samples taken along the
eastern and southeastern boundary of the fill area in the upper 20 feet of
the site. The predominant compounds were TCE and its degradation
product, 1, 2-dichloroethylene. Excessive levels of cadmium, chromium,
lead, and phenolics were discovered in site G soil samples.
Soil/Site Conditions: Both sites have homogeneous sandy soils, Arsenal
Sand, as deep as 120 feet below the ground surface, well above the
groundwater table. The groundwater table is located at a depth of 165 feet.
This Arsenal Sand is in a kame deposit (a poorly-sorted, glaciofluvial, sand
and gravel formation) that consists of brown-gray, fine to coarse sand and
gravel. The Arsenal Sand has a 45 foot thick layer of Hillside Sand
underneath it.
Experimental/System Design:-Pilot- acd f-ull-scale studies-have been
conducted at TCAAP. The pilot study was conducted at Site D. The soil
vapor extraction system for the pilot study consisted of injection and
extraction wells; 2 blowers; a gas flow meter; various plumbing; insulation
for the plumbing; various valves, gauges, and sampling ports; GAC vapor
-------
TCAAP.SHT
treatment; and monitoring wells along with probes (see Figure 1). Two
separate pilot systems were installed at Site D. The first pilot system
covered an area-of-2,500- square feet and was~designed to- evaluate the
effectiveness of vapor extraction for TCE removal from soil that had
contamination levels of less than 5 mg/kg. The second system was installed
over a 10,000 square feet area that had TCE concentrations greater than 100
mg/kg. Besides the differences in size and initial TCE contamination
levels, pilot system no. 1 had an extraction rate of between 40 and 55 cfm.
The vent pipe spacing for System 1 was 20 feet, while System 2 had a vent
pipe spacing of 50 feet and an extraction rate of between 200 and 220 cfm.
Monitoring was continuous for both systems, with air flow rate,
moisture content, temperature, pressure, and hourly TCE concentrations being
the primary pieces of data collected. Two full-scale field systems were
installed at Sites D and G. The designs were basically the same except that
there are no injection wells and that there was no vapor treatment at Site
D.
Status of Experiment/Site Clean-up: The following is a summary of the pilot
study and the status of the field system.
1. System Number 1 was in operation for 67 days. During that time,
1,874 grams (1.9 Ib) of TCE was extracted from the 8,000 cubic feet of
contaminated soil. The total daily TCE extraction rate was originally
greater than 70 mg/day, but decreased to less than 10 mg/day during the last
week of the pilot study.
2. System Number 2 was in operation for 78 days. During that time, 730
kilograms (1,609 Ibs.) of TCE were extracted from the 50,000 cubic feet
of contaminated soil. On the average, 11 kg/day was removed by this
pilot system.
3. Termination of air injection did not effect the extraction rate or the
extracted TCE concentrations. Reduction of the extraction flow rates,
affected system number 2 by a reduction in the extracted TCE
concentrations.
4. Less than 1% of the initial soil moisture was removed during venting
for both pilot systems. The temperature range of the extacted air was from
40° to 485.
5. Vapor extraction costs were estimated to be approximately $15 to $20
per cubic yard, including the costs for soil vapor extraction system
hardware, extraction air carbon adsorption system, and soil sampling.
6. Higher recovery rates were observed when the vents were more closely
spaced.
The full-scale operation started during the winter of 1986 and is
nearing final clean-up.
-------
TCAAP.SHT
1 Electric Air Flow dealer
2 Forced Drall Injection Fan
3 Injection Air Oypnss Valve
4 Injection Air Sampling Port
5 Injection Air Flow Meter
6 Extraction Manifold
7 Injection Manifold
8 Slotted Vertical Extraction Vent Pipe (lyp)
9 Slotted Vet lical Injection Venl Pipe (lyp)
10 Extraction Air Sampling Port
11 Extraction Air Plow Muter
12 Extraction Air Bypass VjslvB
13 Induced Uiall Extraction I nn
14 Vapor Carbon Package treatment Unit
Figure 1. Soil Vapor Extraction System Design at TCAAP (Source: Anastos et
al., 1985).
-------
TRI.SHT
ASSESSMENT OF LAB-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.iect Name: Texas Research Institute Lab Study
Principal Investigator: W. L. Wootan, Jr.
Texas Research Institute (TRI), Inc.
Austin, TX
Investigative Reports:
Thornton, S.J., R.E. Montgomery, T. Voynick, and W.L. Wootan, "Removal
of Gasoline Vapor from Aquifers by Forced Venting," in Proceedings of
the 1984 Hazardous Materials Spills Conference, Nashville, TN, pp 279-
286, April 1984.
Wootan, W.L., and T. Voynick, "Forced Venting to Remove Gasoline Vapor
from a Large-Scale Model Aquifer," Texas Research Institute, Inc.,
Final Report to American Petroleum Institute, 1984.
Location of Work/Study: Texas Research Institute, Inc.
9063 Bee Caves Road
Austin, TX 78746
Nature of Contamination: The feasibility of vapor extraction systems was
studied in model aquifers designed and constructed by the Texas Research
Institute. A total of 80 liters of unleaded gasoline was intentionally
spilled in two separate tanks at four points (see Figure 1). Gasoline was
added to the sand just above the capillary zone and was allowed to spread
through the model aquifer before vapor extraction was started.
Site/Soil Conditions: Each model aquifer consisted of a 3 meters by 3
meters by 1.2 meters deep concrete tank packed with washed river sand.
Each tank was insulated, and the interior temperature was maintained at
13°C. Each of the tanks had flowing water with a water table set at 8 to 30
cm deep as shown in Figure 2.
Experimental/System Design: The lab-scale soil vapor extraction system
consisted of air inlet vents and a central extraction well; a concrete cap;
a blower; flow meters; various plumbing, valves, and sampling ports; a
gasoline/water separator; and various thermocouples, vapor, and observation
wells. A steel partition separated the-two tanks. The partition was sealed-
with silicone caulk, and the seams and joints were covered with fiberglass
and polyester resin. Water entered the sand pack near the steel partition
through a perforated PVC pipe, 2.5 meters long. This PVC pipe was suspended
20 cm from the bottom of the tank. A 15 cm high standpipe/extraction well
-------
TRI.SHT
at the center of the opposite wall allowed water and gasoline to flow out of
the tank and into the separator. In the separator, the raw gasoline was
collected for measurement. The effluent water was pumped to a holding tank
from which it was periodically transported to a sewage treatment facility.
The eight observation wells were made from 3.2 cm PVC pipe slotted from
the floor of the tank to 60 cm. Four of these observation wells were also
used as gasoline spill sites as shown on Figure 1. The vapor wells were
placed at depths of 20 and 65 cm. A Sutton, Model P6-55A blower was used to
create a pressure drop of about 0.44 inches of mercury at each extraction
pipe.
Four experiments were run. The first two experiments (A 4 B) compared
the air flow paths of two test cells. Cell A had a flow path from the inlet
well 56 cm tall, extending from the top of the capillary zone to just below
the top of the tank. The flow path in cell B was modified by screening
only 15 cm of the air inlet well just above the capillary zone. See Figure
3 for the air flow patterns for experiments A & B, Phase I. Air flow rates
of 4 and 16 liters per minute were used for both experiments (4 L/min for
one week for both test cells and then 16 L/min for one week, again for both
test cells). The third and fourth experiments used thicker flow paths (75
and 58 cm., respectively, for Experiments C and D) to vent sand packs of two
different permeabilities (medium-grained Finish sand with a hydraulic
conductivity of 4.2 x 10"z cm/sec, and fine-grained mortar sand with a
hydraulic conductivity of 3.1 x 10"' cm/sec) at air flow rates of 0.4, 1.0,
and 4.0 L/min. Figures 4, 5, and 6 show the tank set-up and airflow
patterns for Phase II of the experiments. Each tank had instruments that
could monitor temperature, water levels, water flow rates, air flow rates,
and gasoline vapor concentration in the sand. Three types of gasoline
removal were studied: 1. Raw gasoline flowing out of the tank on top of
the effluent water, 2. Dissolved gasoline flowing out in the effluent
water, and 3. Gasoline vapor swept out of the sand pack with the effluent
air. Residual gasoline levels still in the sand, were estimated by analysis
of core samples of the sand pack.
Status of Experiment/Site Clean-up: All venting geometries and flow rates
were effective in removing gasoline from the sand packs. Both geometries
and flow rates were also effective in reducing the gasoline vapor
concentrations in the upper half of the tank to less than 1000 ppmv
hydrocarbons. This represents a total gasoline vapor recovery of 21.6
liters (27% of the original gasoline present) over the 22 days that the
experiment was performed. Screening geometry only had an effect at low air
flow rates where it was shown that screen placement near the water table
resulted in higher recovery rates than when the well was screened the full
depth of the unsaturated zone. Coarse-grained materials were more amenable
to vapor extraction because the VOCs were able to diffuse out of the
immobile zones at a faster rate. The rate of removal was higher for the
higher air flows.
-------
IKl.bH!
Air Intake Pipe
with Flow Meter
(See Text)
O Thermocouple
Pipe
eter
(Two/Site)
por Well
xt)
Well
Well and
111 Site
e
Water /
Input
Steel /
Divider
0 A 0
'
o o
A
A" 0' A
A* A ;
A
A A'
0 ;
0
"A" 0* A /O A 0* A (5)'
^^
.^(Vx^Alr Exhauat Pipe
/^^> ''
To Blower !'
0'
o o
0 O
"'•••• •"••;•;. .•: /• :'. •.,•_-_•_ :::: —:•„:::... ::•'.„. :. '..•./.-.. .•...:••::..•.-: •:•:-;:•:•>- :^-:. ;:•/ --..•:-:>x-:v.v.".^-:-.-.v"'^i • •< :"•:•.. .;J
Standplpe and
Water Output
XUrethane Fou
InaulatIon
'Coticrata Tank
Figure 1. Plan View of Experimental Tank Set-up for Phase I Study (Source:
Wootan and Voynick, 1984).
EXPERIMENT B
EXPERIMENT A
Flou
Meter
x ^^
M
Watec To Blower
Input Typical
//\ Obaervatton Concrete Cap
Typical Set of f I Uell /
Vapor Wei la ! / /
X. U i-i .Theraocouplea / • .Thermocouple
\ MM iLi / ri I
Steel Divider.
Sealed with
Flberglaaa
Cooling Tube*
on Slda of Tank
Typical Air Intaka/ \Typlcal Atr Intak
(Experiment B) (Experl-ent A)
ni Spout for
ICaaollne
Collection
Exhauat Pipe, Experiment A.
(Pipe for Experiment B la
alotted Ilka air lntak« pipe.)
Water
Punplug Station,
Then to Holding
Tank
Figure 2. Cross-section of Experimental Tank for Phase I Study (Source:
Wootan and Voynick, 1984).
-------
TRI.SHT
1^
1
fe
t
fe
fc.
w
J
£
LJ
Water Table
I
-
t
I
Experiment A
1
t
1
* ! ! *
,-f H- I-
Experiment B
Figure 3. Air Flow Patterns for Phase I Study (Source: Wootan and Voynick,
1984).
A Vapor Well
O Obaei
o The r«x>coupl *
Input
Air Input
time LCn-
n and final
ting)
11
Ion tUll
upl.
Input
Coollne
Input
Steel
Divider
-Air Eshautt Pipe
1 '> To Blower (g)
A o 0
A
O
A
A
0 * 0
0 O
o
A A
0
A A
8>Ai o A o AQ
r and C«>olln>
Recovery
X. Ur«th«n« Tomm
Inatilaclon
T*nk
Figure 4. Plan View of Experimental Tank Set-up for Phase II Study (Source:
Wootan and Voynick, 1984).
-------
TRI.SHT
EXPERIMENT D
O2 Dare/ S«nd)
Typical Sic of Vopoi Uclla
for Experiment D. Nuebera
•re Depths In Centl«elera
below Concrete Cap.
EXPERIKEHT C
(U D.rcy 8>nd)
Typical Set of Vapor Well*
for Experiment C. Hunbera
are dcptha In Cintlmetera
below Concrete Cap.
To Blower Wnter
A A /Input
Typical
Obaervat ion
10
/
-IHHH
m\
2
4li
56
/
11
=
I
0
P
f
—
-
I
$'
1 C«aolln
' .S Input
r^r^
;«; [
HHU:'1. ,:i>-:,Xv:,^::
I
II" 28
50
72
K!.>S:-N-x.:.>Xv>»MWl«y;s>::
/
-^
Typical /I
Thermocouple / /
\4 fc-
j-m^,-^. .-.;-<-•
1 -
/
1^
Coollnf Tub«a
on Side of Tank.
HI-IT™ J
V::»y>:::;x:x^-M;.::i,^^:^
Concrete Cap
/ Flow
/Meter
~IxAlr Intake
• Urethaoa Fo«»
Air Exhauat for Eip«rU«nc C
(rlou Path 75 c. IVaap)
Air Eihauat for Experiment D
(Plow Path 58 c« Deep)
Water Drain to
Pumping Station,
Than to Bo Id In«
Tank
Figure 5. Cross-section of Experimental Tank for Phase II Study (Source:
Wootan and Voynick, 1984).
I
i I
Water Table
I
1
Water Table
Proposed for Phase II Proposed for Phase II
"(Arrows Represent Air Flow Patterns)
Figure 6. Air Flow Patterns for Phase II Study (Source: Wootan and Voynick,
1984).
-------
TYSON.SHT
ASSESSMENT OF LAB-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Site/Pro.lect Name: Tysons Lagoon Superfund Site
Principal Investigator: AWARE Incorporated
621 Mainstream Drive
Suite 200, Metro Center
Nashville, TN
Site Investigation Report:
AWARE, Inc., "Phase I - Zone I Soil Decontamination Through In-Situ
Vapor Stripping Processes," (Contract Number 68-02-4446), Final Report,
Prepared for: U.S. Environmental Protection Agency, Small Business
Innovative Research Program , Washington, D.C. Prepared by: AWARE,
Inc., 621 Mainstream Drive, Suite 200, Metro Center, Nashville, TN
37228, April 1987.
Location of Work/Study: CIBA-GEIGY Corporation spill site at Tysons Lagoon,
Pennsylvania
Nature of Contamination: Soil Column 1 -- Significant quantities of TCE,
trichloropropane (TCP), toluene, ethyl benzene, and xylene plus another
compound at extremely high concentrations tentatively identified as
1,1,1,2-tetrachloroethane. Soil Column 2 -- Soil concentrations
significantly lower than those from Column 1 with toluene, ethyl benzene and
xylene being the major components stripped.
Soil/Site Conditions: Soil type = sandy
Moisture content = 16%
Organic content = 4.4%
% Porosity = 54%
The soil samples tended to clump together and were difficult to pack
even though sandy and at moderately low moisture content.
Experimental/System Design: Same setup as the AWARE study (see AWARE.SHT)
with the exceptions that the soil in column 1 was 15 mm deep and the soil in
column 2 was 300 mm deep. The initial air flow rate was 2 mL/min. Two air
samples (20 mL and 5 mL) were collected from each column immediately after
stripping began. Initial volatile organic emissions were extremely high and
resulted in a multitudE of_peaks.on.the_gas-,chnDmatograph.which could not be
easily identified or quantified. After 24 hours of stripping, a multitude
of peaks again occurred on the gas chromatograph, so the air flow rates were
increased to 3.5 mL/min to speed up the soil stripping process. The
stripping process was continued at this rate for the experiment's duration
-------
TYSON.SHT
of 11 days.
Status of Experiment/Site Clean-up: Significant quantities of TCE,
trichloropropane (TCP), toluene, ethyl benzene, and xylene were stripped
from Column 1 during the 11 days of this study as summarized in Table 1.
The gas concentrations extracted from Column 2 were significantly lower than
those from column 1. This was expected, since the initial concentrations in
column 2 were considerably lower at the beginning of this study. Toluene,
ethyl benzene, and xylene were the major compounds stripped from column 2.
Table 1 also records the percent removals for each of the columns.
A full-scale soil vapor extraction system has reportedly been installed
at the Tyson site (Peter Michaels, personal communication, June 1988).
Table 1: Methanol Extraction of Soil Samples
Soil Analysis For Column 1
Compound Initial Concentration Final Concentration Percent
(ug/g) (ug/g) Removals
Toluene 600 10 98
Ethyl Benzene 1,100 60 95
p, m-Xylene 11,700 334 97
o-xylene 3,700 192 95
1,2,3-TCP 2,600 10 99
Soil Analysis For Column 2
Compound Initial Concentration Final Concentration Percent
(ug/g) (ug/g) Removals
Toluene 37 10 73
Ethyl Benzene 74 41 45
p, m-Xylene 970 255 74
o-Xylene 280 100 64
1,2,3-TCP 100 10 90
(Source: AWARE, 1987)
-------
VERONA.SHT
ASSESSMENT OF FULL-SCALE VAPOR EXTRACTION SYSTEMS
SITE DATA SHEET
Slte/Pro.lect Name: Verona Well Field/Thomas Solvents Raymond Road
Facility -- Superfund Site
Principal Investigators: John Tanaka, Project Manager
USEPA, Region V
230 S. Dearborn St.
Chicago, IL 60604
CHoM-Hill
Reston, VA
Investigative Reports:
CHoM-Hill, Inc., Remedial Planning/Field Investigation Team, "Verona
Well Field - Thomas Solvent Company, Battle Creek, Michigan, Operable
Unit Feasibility Study," Contract No. 68-01-6692, June 17, 1985.
CH2M-Hill, Inc., "Operable Unit Remedial Action, Soil Vapor Extraction
at Thomas Solvents Raymond Road Facility, Battle Creek, MI, Quality
Assurance Project Plan," October, 1987a.
CHoM-Hill, Inc., "Appendix B - Sampling Plan, Operable Unit Remedial
Action; Soil Vapor Extraction At Thomas Solvents Raymond Road
Facility, Battle Creek, MI," October, 1987b.
Location of Work/Study: Thomas Solvents Company Raymond Road Facility,
Battle Creek, MI. See Figure 1 for location of site.
Nature of Contamination: The major public potable water source in Battle
Creek, Michigan, is the Verona Well Field, which serves 35,000 residents and
a number of commercial and industrial establishments. In August 1981, the
Calhoun County Health Department discovered and later verified that nearly
one-half of Battle Creek's 30 potable wells were contaminated with VOCs.
By January 1984, all but six of the city's wells were affected by the
groundwater contamination plume.
Two of the major contaminant sources were the Thomas Solvents sites
(Raymond Road and the Emmett Street Annex). Thomas Solvents Company
stored, transferred, and packaged chlorinated and nonchlorinated solvents.
In addition, Thomas Solvents Company handled liquid industrial wastes. In
all, there were 21 underground storage tanks on the site. The contamination
resuUed from tank leakage and from spillage from-above-ground-transfer of
chemicals. Another source of contamination was the Grand Trunk Western
Railroad's marshaling yard where DOWCLENE (a commercial solvent formulated
from PCE and 1,1,1-TCA) was disposed of during rail-car operations.
-------
VERONA.SHT
The major constituents of both plumes were PCE (tetrachloroethylene or
perch!oroethylene) and 1,1,1-Trichloroethane (1,1,1-TCA). In addition, the
combined Thomas Solvents' plume contains-TCE. Table 1 lists all chemical
contaminants encountered at these sites. It is possible that some of the
chemicals were the the result of the degradation of PCE, TCE and 1,1,1-TCA.
Sixty-eight percent (3,900 pounds of VOCs at levels exceeding 100,000 ppb)
of the total contaminant mass in the southern plume is located beneath the
Raymond Road Facility. The total VOC mass in the unsaturated zone on the
Raymond Road Facility is estimated to be approximately 1,700 pounds. It is
estimated that there are 440 pounds of VOCs in the saturated zone directly
underneath the property
Soil/Site Conditions: In general, the aquifer consists of two units: a sand
and gravel alluvial deposit overlying the sandstone of the Marshall
Formation (bedrock). The municipal well field is within this bedrock layer.
The sand and gravel layer varies in thickness from 13.5 feet to a maximum of
45 feet and consists of fine to medium sand with 3 to 6% silt and clay size
grains. The average hydraulic conductivity of the sand is approximately 0.1
cm/sec (274 feet/day). However, a conservative estimate of 0.04 cm/sec (120
feet/day) was used for determining groundwater clean-up rates. The current
groundwater flow through the sand and gravel layer is to the northwest
across the property and is estimated to be 1 to 2 feet/day. The Marshall
Formation (bedrock) layer consists of fine to medium grained, well-cemented
sandstone with the upper 5 to 20 feet weathered. The hydraulic conductivity
of the upper bedrock is approximately 0.06 cm/sec (170 feet/day). The
aquifer's vertical hydraulic conductivity is lower than the horizontal
hydraulic conductivity. Based on modeling by CHoM-Hill, the resistance to
vertical flow was thought of as a low conductivity layer between the sand
and gravel layer; and the sandstone bedrock layer with a hydraulic
resistance of 0.01/day. There is a shale formation at about a depth of 140
feet to provide the only natural means of confinement.
Experimental/System Design: The soil vapor extraction system consists of
extraction wells; a vacuum pump; a gas flow meter; various plumbing, valves,
gauges, and sampling ports; GAC vapor treatment; and monitoring wells.
The vapor extraction system is being used to remove VOCs from the
unsaturated zone at Thomas Solvents Raymond Road site, while groundwater
pumping with subsequent GAC water treatment is being utilized to treat the
contaminated groundwater. The system consists of fourteen 4-inch PVC wells with perfo
well screens extending into the groundwater, packed with gravel, and sealed
at the top with bentonite to prevent .short circuiting of air flows. The
extraction wells are connected to the suction side of a 25-hp, 960 cfm
vacuum extraction unit via a surface collection manifold. The operation
of the vapor extraction system induces clean air flow from the atmosphere
into the subsoils. This system is designed to decrease the pressure in
the soil voids, resulting in additional VOCs being released. The
extracted soil gas flows through the manifold system to vapor phase
activated carbon adsorption canisters. The extracted gas is then
discharged through a 30-foot stack into the atmosphere. A backup carbon
.system is on-ljne continuously to prevent carbon br.eajkjthr
-------
VERONA.SHT
Status of Experiment/Site Clean-up; This site is currently undergoing
active treatment utilizing Terra Vac system.
Table 1. Chemicals Present in Verona Well Field Site Soils (Source: CHoM-
Hill, 1985)
Chlorinated Hydrocarbons
methylene chloride
chloroform
carbon tetrachloride
1,2-dichloroethane (DCA)
1,1,1-trichloroethane (TCA)
vinyl chloride
1,1-dichloroethylene (DCE)
trans-l,2-dichloroethylene
trichloroethylene (TCE)
tetrachloroethylene (PCE)
Aromatic Hydrocarbons
benzene
toluene
xylene
ethyl benzene
naphthalene
Ketones
acetone
methyl ethyl ketone
methyl isobutyl ketone
------- |