EPA/540/R-93/505
September 1993
HYDRAULIC FRACTURING TECHNOLOGY
TECHNOLOGY EVALUATION REPORT
UNIVERSITY OF CINCINNATI/RISK REDUCTION ENGINEERING LABORATORY
CINCINNATI, OHIO
U.S. Environmental Protection Agency
Risk Reduction Engineering Laboratory
Office of Research and Development
Cincinnati, OH 45268
{£& Printed on Recycled Paper
-------
NOTICE
The information in this document has been prepared for the U.S. Environmental Protection Agency's
(EPA) Superfund Innovative Technology Evaluation (SITE) program under Contract No. 68-CO-0047.
This document has been subjected to EPA's peer and administrative reviews and has been approved
for publication as an EPA document. Mention of trade names or commercial products does not
constitute an endorsement or recommendation for use.
-------
FOREWORD
The SITE program was authorized by the Superfund Amendments and Reauthorization Act (SARA) of
1986. The program is administered by the EPA Office of Research and Development (ORD). The
purpose of the SITE program is to accelerate the development and use of innovative cleanup
technologies applicable to Superfund and other hazardous waste sites. This purpose is accomplished
through technology demonstrations designed to provide performance and cost data on selected
technologies.
This project consisted of two pilot-scale demonstrations conducted under the SITE program to
evaluate the hydraulic fracturing technology developed by the University of Cincinnati (UC) and
EPA's Risk Reduction Engineering Laboratory (RREL) A full-scale demonstration using an EPA
approved Quality Assurance Project Plan (QAPP) has not been conducted for this technology. The
technology demonstrations were conducted at a Xerox Corporation (Xerox) vapor extraction site in
Oak Brook, Illinois (Xerox Oak Brook site); and at a bioremediation site near Dayton, Ohio (the
Dayton site). The demonstrations provided information on the performance and cost of the
technology. Tests to determine the performance of hydraulic fractures over a 1-year period were
conducted at an uncontaminated site at the UC Center Hill Solid and Hazardous Waste Research
(Center Hill) Facility in Cincinnati, Ohio. This technology evaluation report (TER) describes the
development, demonstration, and evaluation of the hydraulic fracturing technology.
Copies of the TER can be purchased from the National Technical Information Service (NTIS),
Ravenswortb Building, Springfield, Virginia 22161, 703/487-4600. Reference copies will be
available at EPA libraries in the Hazardous Waste Collection.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
-------
ABSTRACT
This report evaluates the effectiveness of the hydraulic fracturing technology developed jointly by the
UC and EPA's RREL in enhancing the permeability of contaminated silty clays and presents technical
data from tests conducted at the UC Center Hill Facility and from two pilot-scale SITE
demonstrations,
The hydraulic fracturing technology creates sand-filled fractures up to 1 inch thick and 30 feet (ft) in
radius. These fractures are placed at multiple depths ranging from 5 to 40 ft below ground surface
(bgs) to enhance the efficiency of treatment technologies such as soil vapor extraction (SVE), in situ
bioremediation, and pump-and-treat systems.
Tests were conducted at the Center Hill Facility by UC to determine the factors affecting soil vapor
flow through sand-filled hydraulic fractures. A significant finding from these tests is that rainfall
decreases vapor yield and increases the suction head of fractured wells. The zone of pneumatic
control of a fractured well was 15 to 30 times farther from the well than that of an unfractured well at
the Center Hill Facility, and vapor yield was about an order of magnitude higher than from an
unfractured well.
The hydraulic fracturing technology was demonstrated in 1991 and 1992 at the Xerox Oak Brook site,
where SVE was being conducted. On-site soil contamination included ethylbenzene; 1,1-
dichloroethane (DCA); trichloroethene (TCE); tetrachloroethene (PCE); 1,1, 1-trichloroethane (TCA);
toluene; and xylene. The vapor flow rates, soil vacuums, and contaminant yields of two hydraulically
fractured and two unfractured wells were compared. The fractured wells were fractured at 6, 10, and
15 ft bgs. The vapor yield from fractured wells was one order of magnitude greater than from
unfractured wells. This higher yield was obtained in an area 30 times greater than the area affected
by the unfractured well. The contaminant mass recovery from fractured wells was 7 to 14 times
greater than that from the unfractured well.
Another pilot-scale demonstration was conducted in 1991 and 1992 at the Dayton site, where
bioremediation was being conducted. Site contamination included benzene, toluene, ethylbenzene,
and xylene (BTEX); and total petroleum hydrocarbons (TPH). Fractures were created at 7, 8, 10,
in
-------
and 12 ft bgs at one of two on-site wells. Water containing hydrogen peroxide and nutrients was
pumped into the hydraulically fractured well and into one unfractured well 50 ft from the fractured
well. The injection rates, soil moisture contents, microbial metabolic activity, numbers of colony
forming units (CPU), and rates of bioremediation at the fractured and unfractured wells were
compared. In the fractured well, the injection rate was 25 to 40 times greater, and moisture content
increased 2 to 4 times near the fracture. Comparison of microbial metabolic activity, CPU, and rates
of bioremediation were inconclusive.
Economic data indicate that the capital cost for hydraulic fracturing equipment is $92,900 and the cost
of renting the equipment is $1,000 per day. Rental, operating, and monitoring costs for creating a
fracture range from $950 to $1,425, depending on site-specific conditions. Typically, two to three
fractures are created per well, and four to six fractures can be created in 1 day. The cost of creating
a fracture is not materially affected by the depth of fracture for depths ranging from 5 to 40 ft bgs.
The cost is also unaffected by the type of soil encountered.
IV
-------
TABLE OF CONTENTS
Section Page
NOTICE i
FOREWORD 11
ABSTRACT in
LIST OF ACRONYMS AND ABBREVIATIONS ix
ACKNOWLEDGEMENTS xi
EXECUTIVE SUMMARY ES-1
1.0 INTRODUCTION 1
1.1 SITE PROGRAM 1
1.2 DEMONSTRATION PROGRAM OBJECTIVES 2
1.3 PURPOSE OF THE TER 3
1.4 REPORT ORGANIZATION 3
2.0 DESCRIPTION OF TREATMENT TECHNOLOGY 4
2.1 HYDRAULIC FRACTURING IN SHALLOW DEPOSITS 4
2.2 HYDRAULIC FRACTURING METHOD 6
3.0 CENTERHILL FACILITY TESTS 13
3.1 HYDRAULIC FRACTURING ACTIVITIES 13
3.2 DATA ACQUISITION 15
3.3 FRACTURING RESULTS 18
3.4 CONCLUSIONS 24
4.0 XEROX OAK BROOK SITE TESTS 26
4.1 HYDRAULIC FRACTURING ACTIVITIES 26
4.2 DATA ACQUISITION 31
4.3 FRACTURING RESULTS 32
4.3.1 Vapor Discharge 32
4.3.2 Contaminant Recovery 33
4.3.3 Suction Head 37
4.4 CONCLUSIONS 39
5.0 DAYTON SITE TESTS 40
5.1 FRACTURING ACTIVITIES 40
5.2 DATA ACQUISITION 44
5.3 FRACTURING RESULTS 46
5.4 CONCLUSIONS 6
-------
TABLE OF CONTENTS (Continued)
Section Page
6.0 QA/QC ANALYSES 48
6.1 CENTER HILL FACILITY 49
6.1.1 Sampling and Analytical Procedures . . . . , 49
6.1.2 Measurement of Injection Pressure and Ground Surface Uplift 51
6.1.3 Calibration Frequency 52
6.1.4 Data Validation 52
6.2 XEROX OAK BROOK SITE 52
6.3 DAYTON SITE 54
7.0 COST AND BENEFIT COMPARISON 55
i
8.0 CONCLUSIONS 58
8.1 CENTER fflLL FACILITY TESTS 58
8.2 XEROX OAK BROOK SITE TESTS 58
8.3 DAYTON SITE TESTS 58
9.0 REFERENCES 59
LIST OF TABLES
Table Page
3-1 FRACTURE CHARACTERISTICS AT THE CENTER HILL FACILITY 15
3-2 PERFORMANCE OF WELLS AT THE CENTER HILL FACILITY 24
4-1 FRACTURE CHARACTERISTICS AT THE XEROX OAK BROOK SITE 30
4-2 WELL DISCHARGE READINGS AT THE XEROX OAK BROOK SITE 33
5-1 FRACTURE CHARACTERISTICS AT THE DAYTON SITE 41
5-2 ANALYTICAL RESULTS OF DAYTON SITE SAMPLES 43
5-3 CONTAMINANTS REMOVED AT THE DAYTON SITE 45
6-1 QA OBJECTIVES FOR CRITICAL MEASUREMENTS OF FRACTURING FLUID ... 50
7-1 ESTIMATED COSTS ASSOCIATED WITH HYDRAULIC FRACTURING 56
vi
-------
TABLE OF CONTENTS (Continued)
LIST OF FIGURES
Figure Page
2-1 INJECTION PRESSURE VERSUS TIME 5
2-2 FRACTURING LANCE USED TO PREPARE BOREHOLES FOR HYDRAULIC
FRACTURING 7
2-3 SLURRY MIXING EQUIPMENT MOUNTED ON TRAILERS 8
2-4 SEQUENCE OF OPERATIONS FOR CREATING HYDRAULIC FRACTURES 9
2-5 LAYOUT OF GEMS 11
2-6 PRESSURE VERSUS TIME DURING THE CREATION OF A HYDRAULIC FRACTURE 12
3-1 WELL LOCATIONS AT THE CENTER HILL FACILITY 14
3-2 WELL NO. CHF1 UPLIFT CONTOURS AND
PNEUMATIC PIEZOMETER LOCATIONS 16
3-3 WELLS NO. CHF2, CHF3, CHC1, AND CHC2 UPLIFT CONTOURS AND
PNEUMATIC PIEZOMETER LOCATIONS 17
3-4 WELL YIELD AS A FUNCTION OF TIME FOR WELLS NO. CHF1 AND CHC1-
JANUARY 20 TO MARCH 2, 1992 19
3-5 WELL YIELD AS A FUNCTION OF TIME AND RAINFALL FOR WELLS
NO. CHF2, CHF3, AND CHC2 - JUNE 8 TO JULY 7,1992 20
3-6 PRESSURE DISTRIBUTION NEAR WELL NO. CHF1 21
3-7 PRESSURE DISTRIBUTION VERSUS TIME AND RAINFALL NEAR WELL
NO. CHF2 - JUNE 8 TO JUNE 25, 1992 22
3-8 PRESSURE DISTRIBUTION VERSUS DEPTH NEAR WELL NO. 23
4-1 XEROX OAK BROOK SITE PIPING PLAN FOR VAPOR EXTRACTION SYSTEM . . 27
4-2 SCHEMATIC DIAGRAM OF XEROX OAK BROOK SITE VAPOR EXTRACTION
SYSTEM 28
4-3 EXTRACTION WELL AND PIEZOMETER LOCATIONS 29
vn
-------
TABLE OF CONTENTS (CONTINUED)
Figure Page
TABLE OF CONTENTS (Continued)
Figure Page
4-4 VAPOR FLOW RATE VERSUS TIME IN WELLS NO. RW2, RW3, AND RW4
- JUNE 23 TO NOVEMBER 30, 1992 34
4-5 WATER RECOVERY RATE VERSUS TIME
- JUNE 23 TO NOVEMBER 30, 1992 35
4-6 RAINFALL VERSUS TIME AT THE XEROX OAK BROOK SITE
- JUNE 23 TO NOVEMBER 30, 1992 36
4-1 CONTAMINANTS REMOVED FROM WELLS NO. RW2, RW3, AND RW4
- JUNE 23 TO NOVEMBER 30, 1992 38
5-1 FRACTURED AND INJECTION WELL LOCATIONS AT THE DAYTON SITE 42
5-2 FLOW VOLUMES OF WATER INJECTED IN WELLS NO. SAD2 AND SAD4 .... 47
Vlll
-------
LIST OF ACRONYMS AND ABBREVIATIONS
ASTM
bgs
BTEX
Center Hill Facility
Cfm
CPU
cm/sec
DCA
EPA
ft
ft3
GC
GEMS
gpm
mm
NA
ND
NI
ORD
OSWER
PCE
ppb
psi
PVC
QAPP
QA/QC
RREL
SARA
SITE
American Society for Testing and Materials
Below ground surface
Benzene, toluene, ethylbenzene, and xylene
Center Hill Solid and Hazardous Waste Research Facility
Cubic foot per minute
Colony forming unit
Centimeter per second
1,1-Dichloroethane
U.S. Environmental Protection Agency
foot
Cubic foot
Gas chromatography
Ground Elevation Measurement System
Gallon per day
Gallon per minute
Millimeter
Not applicable
Not detected
No impact
Office of Research and Development
Office of Solid Waste and Emergency Response
Tetrachloroethene
Part per billion
Pound per square inch
Polyvinyl chloride
Quality Assurance Project Plan
Quality assurance and quality control
Risk Reduction Engineering Laboratory
Superfund Amendments and Reauthorization Act of 1986
Super-fund Innovative Technology Evaluation
-------
SVE Soil vapor extraction
TCA 1,1,1-Trichloroethane
TCE Trichloroethene
TER Technology evaluation report
TPH Total petroleum hydrocarbon
UC University of Cincinnati
fig/kg Microgram per kilogram
UST Underground storage tank
V O C Volatile organic compound
Xerox Xerox Corporation
-------
ACKNOWLEDGEMENTS
This report was prepared under the direction of Naomi Barkley, the EPA SITE project manager at
EPA's RREL in Cincinnati, Ohio. Contributors and reviewers for this report were Messrs. Allen
Wolf and Mark Kemper, Dr. Lawrence Murdoch, and Dr. Steven Vesper of UC in Cincinnati, Ohio;
and Mr. Elliott Duffney of Xerox in Rochester, New York. Peer reviewers were Dr. Ronald Lewis
and Robert Stenburg of RREL.
This report was prepared by Drs. V. Rajaram and Pinaki Banerjee of PRC Environmental
Management, Inc. (PRC). The report was typed by Ms. Cheryl Vaccarello, edited by Ms. Shelley
Fu, and reviewed by Dr. Kenneth Partymiller and Mr. Stanley Labunski, all of PRC.
XI
-------
EXECUTIVE SUMMARY
The cleanup of volatile organic chemical (VOC) and petroleum hydrocarbon spills in low permeability
soils is a significant problem in many parts of the United States. In situ techniques for cleaning up
these sites are preferable because they minimize the risk of spreading the contamination and are cost-
effective. However, these techniques are hampered by the low permeability of the soils. A method
of enhancing the permeability of the contaminated soil will enhance in situ remediation.
The University of Cincinnati (UC) and U.S. Environmental Protection Agency (EPA) Risk Reduction
Engineering Laboratory (RREL) developed the hydraulic fracturing technology to enhance the
permeability of silty clays. This technology was evaluated under the EPA Superfund Innovative
Technology Evaluation (SITE) program. The hydraulic fracturing technology was developed at the
Center Hill Solid and Hazardous Waste Research (Center Hill) Facility. Tests were conducted at the
Center Hill Facility to determine factors affecting the performance of the hydraulic fractures.
Demonstrations were conducted during 1991 and 1992 at two sites, a Xerox Corporation (Xerox) site
in Oak Brook, Illinois (the Xerox Oak Brook site), where soil vapor extraction (SVE) was being
conducted, and at a site in Dayton, Ohio (the Dayton site), where bioremediation was being
conducted.
The technology demonstrations had the following objectives:
To establish the viability of creating sand-filled hydraulic fractures in low permeability
silty clays
To study the factors that affect the performance of the fractures over time
To compare the vapor flow rates in wells in fractured and unfracturecl soils
To compare the water flow rates and moisture contents in fractured and unfractured
soils
To develop information required to estimate the operating costs of the technology
The results obtained at the Center Hill Facility and the two SITE demonstrations are discussed below.
ES-1
-------
Center Hill Facility Tests
Researchers from UC conducted field testing during the winter and summer of 1992 to determine the
effect of sand-filled hydraulic fractures on S VE performance. The tests were conducted next to the
Center Hill Facility. The soil at the testing location is underlain by silty clay with minor amounts of
sand and gravel. Five wells were used during the tests. Wells No. CHF1, CHF2, and CHF3
intersected hydraulic fractures. Well No. CHF1 was screened with a 2-inch-diameter screen centered
on the fracture in ground fractured at 5 and 10 feet (ft) below ground surface (bgs); Well No. CHF2
was screened in ground fractured at 5 ft bgs and the fracture reached the ground surface (vented); and
Well No. CHF3 was screened in ground fractured at 5 ft bgs. Two conventional wells, Wells No.
CHC1 and CHC2, were screened in unfractured ground. Well No. CHC1 was screened from 51 to
69 and 111 to 129 inches bgs, and Well No. CHC2 was screened from 47 to 60 inches bgs.
Comparison of the performance of fractured wells individually and with the performance of
conventional wells over two distinct periods (January 20 to March 2, 1992; and June 8 to
July 7, 1992) yielded the following results:
The air flow from a fractured well was about an order of magnitude higher than from
an unfractured well.
The zone of pneumatic control extended more than 10 times farther from the fractured
well than from the unfractured well.
Rainfall decreased the air flow and increased the suction head of fractured wells, and
unfractured wells were not affected by rainfall.
The effect of a vented fractured well was not significantly different than that of an
unvented fractured well.
Xerox Oak Brook Site Demonstration
At the Xerox Oak Brook site, contaminants consisting of trichloroethene (TCE);
1,1, 1-trichloroethane (TCA); 1,1-dichloroethane (DCA), tetrachloroethene (PCE) and other solvents
are present in silty clay till to depths of 20 ft bgs. Xerox installed a two-phase vapor extraction
system to clean up the site. In an effort to enhance SVE, Xerox requested UC to create hydraulic
ES-2
-------
fractures. Fractures were created at depths of 6, 10, and 15 ft bgs at two locations; these fractures
measured about 20 ft in length. Multilevel recovery wells, Wells No. RW3 and RW4, were installed
to access each fracture individually. The vapor recovery rates from these wells were compared to
rates from unfractured Wells No. RW1 and RW2. A multilevel monitoring system consisting of as
many as six pneumatic piezometers per borehole was installed at radial distances of 5, 10, 15, and 20
ft from each recovery well.
The vapor flow rates and contaminant concentrations were measured on a continuous basis using
vortex shedding flow meters and on-line gas chromatography, respectively. Other parameters
measured included water quality in monitoring wells around the site, water discharge from the vapor
extraction system, soil moisture content, and soil vacuum at the recovery wells and monitoring holes.
Results obtained from December 1991 through December 1992 led to the following conclusions:
Fractured wells yielded vapor flow rates 15 to 30 times greater than unfractured
wells.
Vapor flow rates were adversely affected by precipitation.
Contaminant yields from the fractured wells were one order of magnitude greater than
from comparable zones in the unfractured wells.
The zone of pneumatic control extended more than 10 times farther from the fractured
well than from the unfractured well.
Dayton Site Demonstration
At the Dayton site, six underground storage tanks (UST) were removed in December 1989.
Laboratory analysis of the soil samples collected from the UST excavations indicated that benzene,
toluene, ethylbenzene, xylene (BTEX); and total petroleum hydrocarbons (TPH) are present. The site
consists of stiff sandy to silty clay with traces of gravel. A remedial action contractor initiated
bioremediation activities at the site in 1991. UC created hydraulic fractures at the site in August 1991
at depths of 7, 8, 10, and 12 ft bgs. Water containing hydrogen peroxide and nutrients was
introduced into a fractured well, Well No. SAD-2, and an unfractured well, Well No. SAD-4, from
December 1991 to August 1992.
ES-3
-------
A set of samples was collected in September, 1991 to establish initial contaminant concentrations. Soil
samples taken at 5, 10, and 15 ft north of the wells were analyzed for moisture, BTEX, and TPHs.
In February 1992, after about 30 days of bioremediation, soil samples were obtained at the same
locations and analyzed for moisture content, BTEX, TPHs, number of hydrocarbon degraders (colony
forming units [CPU]), and microbial metabolic activity. A similar round of sampling and analysis
was conducted in July 1992.
Comparison of the data obtained from the fractured and unfractured well yielded the following
findings:
Moisture content increased in the vicinity of the fractured well, especially in the
fractured zones. No change in moisture content was detected in the unfractured well.
The flow of water was about 25 to 40 times greater in the fractured well than in the
unfractured well.
Conclusions
Pilot-scale demonstrations at the Xerox Oak Brook two-phase SVE site and at the Dayton
bioremediation site indicate the significant benefits of hydraulic fracturing in remediating
contaminated sites. The cost and time needed to create the sand-filled hydraulic fractures is small;
therefore, the benefit-to-cost ratio is high.
The technology of creating and monitoring the location of sand-filled hydraulic fractures was
established over the past 2 years. Future work in improving this technology should focus on creating
vertical fractures to connect the in situ horizontal fractures and further increase the overall
permeability of the soil mass.
ES-4
-------
1.0 INTRODUCTION
PRC Environmental Management, Inc. (PRC), was awarded a work assignment under the Superfund
Innovative Technology Evaluation (SITE) program to monitor and prepare a technology evaluation
report (TER) on hydraulic fracturing. This technology has the potential to significantly improve the
performance of in situ vapor extraction systems and bioremediation processes in low-permeability silty
clays. The technology was included in the U.S. Environmental Protection Agency's (EPA) SITE
program in 1991. The technology was j ointly developed by the Department of Civil and
Environmental Engineering at the University of Cincinnati (UC) and EPA's Risk Reduction
Engineering Laboratory (RREL).
Field studies at the Center Hill Solid and Hazardous Waste Research (Center Hill) Facility were
conducted to determine the performance of hydraulic fracturing in silty clays. UC also created sand-
filled hydraulic fractures at a contaminated site owned by the Xerox Corporation (Xerox) in Oak
Brook, Illinois (the Xerox Oak Brook site). The performance of these fractures in remediating the
site was studied in 1991 and 1992. Another contaminated site in Dayton, Ohio (the Dayton site), was
studied from September 1991 to July 1992 to determine the effectiveness of this technology in
enhancing bioremediation. The SITE program, demonstration program objectives, purpose of the
TER, and TER report organization are discussed below.
1.1 SITE PROGRAM
In response to the Superfund Amendments and Reauthorization Act of 1986 (SARA), EPA's Office of
Research and Development (ORD) and Office of Solid Waste and Emergency Response (OSWER)
established the SITE program to (1) accelerate the development, demonstration, and use of new or
innovative technologies to clean up Superfund sites; (2) foster further investigation and development
of treatment technologies that are still at the laboratory scale; and (3) demonstrate and evaluate new
or innovative measurement and monitoring technologies
The primary purpose of the SITE program is to enhance the development and demonstration, and
thereby promote the commercial availability, of innovative technologies applicable to Superfund sites
Major goals of the SITE program are the following:
-------
Identify and remove impediments to the development and commercial use of
alternative technologies
Demonstrate more promising innovative technologies in order to establish reliable
performance and cost information for site cleanup decision making
Develop procedures and policies that encourage selection of available alternative
treatment remedies at Superfund sites
Structure a development program that nurtures emerging technologies
EPA recognizes that a number of forces inhibit the expanded use of alternative technologies at
Superfund sites. One of the objectives of the program is to identify these impediments and remove
them or develop methods to promote the expanded use of alternative technologies.
Another objective of the SITE program is to demonstrate and evaluate selected technologies. This
objective is a significant ongoing effort involving ORD, OSWER, EPA Regions, and the private
sector. The demonstration program tests field-ready technologies and provides Superfund decision
makers with information necessary to evaluate the use of these technologies for future cleanup
actions.
Other aspects of the SITE program include developing procedures and policies that match available
technologies with wastes, media, and sites requiring actual remediation; and providing assistance in
nurturing the development of emerging innovative technologies from the laboratory- or bench-scale to
the full-scale stage.
Technologies chosen for a SITE demonstration must be innovative, pilot- or full-scale applications,
and offer some advantage over existing technologies. Mobile technologies are of particular interest.
1.2 DEMONSTRATION PROGRAM OBJECTIVES
The objectives of the SITE Demonstration Program for the hydraulic fracturing technology are as
follows:
-------
To establish the viability of creating sand-filled hydraulic fractures in low permeability
silty clays
To study the factors that affect these fractures over time
To compare vapor flow rates in fractured and unfractured wells
To compar tbs water flow rates and moisture content in fractured and unfractured
wells
To develop information required to estimate operating costs for hydraulic fracturing
1.3 PURPOSE OF THE TER
Thi sTE R provides a comprehensive description of the demonstrations at the Xerox Oak Brook and
Dayton sites and their results. This report is intended for individuals performing a detailed evaluation
of the hydraulic fracturing technology for a specific site and waste situation The purpose of these.
technical evaluations is to obtain a detailed understanding of the performance of the technology during
the demonstrations, and to ascertain the advantages, risks, and costs of the technology for the given
applications. This information is used to produce conceptual designs of sufficient detail to enable the
preparation of preliminary cost estimates for the demonstrated technology.
1.4 REPORT ORGANIZATION
This report provides an independent assessment of the technology and data from the Center Hill
Facility and the Xerox Oak Brook and Dayton sites Section 2.0 describes the hydraulic fracturing
technology. Sections 3.0, 4.0, and 5.0 provide details of the studies conducted at the Center Hill
Facility, the Xerox Oak Brook site, and the Dayton site Section 6.0 describes the quality assurance
and quality control (QA/QC) analyses conducted on this project. Section 7.0 discusses the costs of
hydraulic fracturing and details the benefits of the technology. Conclusions are provided in Section
8.0. References are provided in Section 9.0.
-------
To establish the viability of creating sand-filled hydraulic fractures in low permeability
silty clays
To study the factors that affect these fractures over time
To compare vapor flow rates in fractured and unfractured wells
To compare the water flow rates and moisture content in fractured and unfractured
wells
To develop information required to estimate operating costs for hydraulic fracturing
1.3 PURPOSE OF THE TER
This TER provides a comprehensive description of the demonstrations at the Xerox Oak Brook and
Dayton sites and their results. This report is intended for individuals performing a detailed evaluation
of the hydraulic fracturing technology for a specific site and waste situation, The purpose of these
technical evaluations is to obtain a detailed understanding of the performance of the technology during
the demonstrations, and to ascertain the advantages, risks, and costs of the technology for the given
applications. This information is used to produce conceptual designs of sufficient detail to enable the
preparation of preliminary cost estimates for the demonstrated technology.
1.4 REPORT ORGANIZATION
This report provides an independent assessment of the technology and data from the Center Hill
Facility and the Xerox Oak Brook and Dayton sites. Section 2.0 describes the hydraulic fracturing
technology. Sections 3.0, 4.0, and 5.0 provide details of the studies conducted at the Center Hill
Facility, the Xerox Oak Brook site, and the Dayton site* Section 6.0 describes the quality assurance
and quality control (QA/QC) analyses conducted on this project. Section 7.0 discusses the costs of
hydraulic fracturing and details the benefits of the technology. Conclusions are provided in Section
8.0. References are provided in Section 9.0.
-------
2.0 DESCRIPTION OF TREATMENT TECHNOLOGY
The removal of contaminants in shallow soil through vapor extraction or bioremediation is hampered
by the low permeability of silty clays. In certain low permeability clays, in situ cleanup is impossible
without enhancing soil permeability. Pneumatic fracturing has been attempted to increase the
permeability of clay layers. The disadvantage of pneumatic fracturing is that pneumatic fractures may
close as time passes. UC has been developing the hydraulic fracturing technique at the Center Hill
Facility since 1990 with funding from EPA's RREL. Pilot-scale demonstrations to determine the
technology's effectiveness in enhancing vapor extraction and bioremediation have been conducted in
1991 and 1992.
The hydraulic fracturing technique consists of creating fractures in low permeability clays by pumping
a gel containing coarse sand into the zone to be fractured. Sand is deposited into the fractures and
enhances the permeability of the contaminated soil. Hydraulic fracturing in shallow deposits and the
hydraulic fracturing method are described in this section.
2.1 HYDRAULIC FRACTURING IN SHALLOW DEPOSITS
Hydraulic fracturing has been successfully used in the oil industry to enhance oil recovery from deep,
tight (low permeability) oil reserves. The mechanics of hydraulic fracturing in rock formations is
well understood (Hubbert and Willis, 1957). Application of hydraulic fracturing in cohesive soil
formations has been attempted to increase soil permeability (Murdoch, 1990). When filled with
permeable sand, the horizontal fractures created in an impermeable material improve the rate of flow
through the material.
Laboratory experiments have been conducted at the Center Hill Facility to determine the effect of slot
length (also known as "initial fracture") and soil moisture content on the propagation of fractures in
silty clay (Murdoch, 1993). A typical record of injection pressure versus time obtained during
laboratory tests is presented in Figure 2-1. This record indicates that fully developed (stable
propagation) fracturing occurs at the break in slope, denoted by fa. The injection pressure at this
point is the critical pressure required to initiate fracturing. During the period of stable fracture
propagation, the injection pressure increases. After reaching a peak pressure, the pressure decreases,
-------
UJ
a:
CO
u
oc:
a.
o
F
Pa:
00 D.
o
55 o
LJ
J- Q-
00 O
zo:
:D a.
df
TIME
Source: Modified from Murdoch, 1993
Figure 2-I. Injection Pressure Versus Time.
-------
and this period indicates unstable fracture propagation, when the fracture continues to propagate at
decreasing pressure. The pressure at the onset of propagation depends on the length of the initial
slot, the water content of the soil, and other factors. Injection pressure diminishes markedly as initial
slot length and water content increase. A theoretical analysis of laboratory observations indicate that
pore fluid infiltrating into the tip of a propagating fracture can control many details of hydraulic
fracture development in soil (Murdoch, 1993).
2.2 HYDRAULIC FRACTURING METHOD
The equipment and materials used for creating hydraulic fractures include the following:
A piston pump or a progressive cavity pump to inject slurry, which consists of up to
one part solid to two parts liquid
A continuous mixer for creating the slurry
A fracturing lance composed of an outer casing and an inner rod, both of which are
tipped with hardened cutting surfaces that form a conical point to prepare boreholes
used for hydraulic fracturing (see Figure 2-2)
Steel tubing with a narrow orifice at one end
Guar gum gel with a borate cross-linker and an enzyme breaker to carry the coarse
sand proppant
A rubber-tired trailer on which the slurry mixing equipment is mounted (see Figure 2-3)
The sequence of operations for creating hydraulic fractures follows Steps 1 through 5 in Figure 2-4.
First, the lance is driven to the desired depth. Individual segments of the rod and casing are 5 feet
(ft) long and are threaded together as required by borehole depth. Next, the lance is removed,
leaving soil exposed at the bottom of the casing. Steel tubing with a narrow orifice at one end is then
inserted in the casing.
Next, water is pumped through the steel tubing and into the narrow orifice, forming a jet that cuts
laterally into the soil. The jetting device is rotated, producing a disc-shaped notch extending 4 to 6
inches away from the borehole. A simple measuring apparatus, built from a steel tape extending the
length of the tube and making a right angle bend at the end of the tube, is inserted into the casing to
measure the radius of the slot.
-------
CASING
"*>,
- - DRIVE HEAD
- CUTTING SLEEVE
Source; Modified from UC, 1991g
Figure 2 2. Fracturing Lance Used to Prepare Boreholes for Hydraulic Fracturing.
7
-------
Figure 2-3. Surry Mixing Equipment Mounted on Trailers.
-------
CO
-'' .i''*''->l'-% *,' '-'"- '/'-.
CM
§
O
in
O
CL
O
O
o
c
o
f
10
3 CS
e «
8 3
8 8.
w C
-------
Hydraulic fractures are created by injecting the cross-linked guar gum gel and sand slurry into the
casing. Injection rates of 16 to 24 gallons per minute (gpm) are used. Lateral pressure of the soil on
the outer wall of the casing effectively seals the casing and prevents leakage of the slurry. The
fractures nucleate at the notch and grow away from the borehole. The direction and distance of
propagation of the fracture from the wall of the borehole is measured by monitoring the uplift of the
ground surface. A leveling telescope is used to measure ground elevation at an array of points before
and after each fracture is created to determine the location and net uplift resulting from the fracture.
A laser system called the Ground Elevation Measurement System (GEMS) was developed by UC to
measure uplift in real time during hydraulic fracturing. The system uses a laser and an array of
sensors to track the displacement of each point in the array with time (see Figure 2-5).
A typical pressure versus time plot during hydraulic fracturing is presented in Figure 2-6. The peak
pressure indicates the onset of fracturing, and the subsequent reduction of pressure with time denotes
the period of fracture propagation.
10
-------
Figure 2-5. Layout of GEMS.
11
-------
to
OT
Q.
cr
w
a.
T3
o
Q.
UJ
cc
00
CO
LU
CC
a.
60-
50H
40H
3CH
20H
10H
0-
12
14
16
18
Source: Modified from Wolf and Murdoch. 1992
TIME (minutes)
Figure 26. Pressure Versus Time During the Creation of a Hydraulic Fracture.
-------
3.0 CENTER HILL FACILITY TESTS
UC conducted field testing during the winter and summer of 1992 to determine the effect of sand-
filled hydraulic fractures on SVE performance. The testing location is next to the Center Hill Facility
in Cincinnati, Ohio. The testing location is underlain by glacial drift that is predominantly composed
of silty clay with lesser amounts of sand and gravel. Five wells were used during the tests. Three
wells, Wells No. CHF1, CHF2, and CHF3, intersected hydraulic fractures. These wells will be
referred to in this report as "fractured wells." Two conventional wells, Wells No. CHC1 and CHC2,
were screened in unfractured ground. These two wells will be referred to as "unfractured wells."
The locations of the wells are shown in Figure 3-1.
Well No. CHF1 intersects hydraulic fractures at 5 and 10 ft below ground surface (bgs). Wells No.
CHF2 and CHF3 both intersect fractures at 5 ft bgs. The principle difference between the two wells
is that the fracture at Well No. CHF2 reached the ground surface 23 ft from the well, whereas the
fracture at Well No. CHF3 remained in the subsurface. All five fractured and conventional wells
were monitored to accomplish the following objectives:
Compare the performance of fractured and unfractured wells
Determine the effect of venting on fracture performance
Assess the difference in performance of fractured wells with one or two fractures per
well
Hydraulic fracturing activities, data acquisition, fracturing results, and conclusions from Center Hill
Facility tests are discussed below.
3.1 HYDRAULIC FRACTURING ACTIVITIES
Hydraulic fractures were created at the Center Hill Facility using the technique described in Section
2.0. The fractures are shallow dipping layers of sand several tenths of an inch thick that extend from
the borehole for 10 to 30 ft. The maximum pressure, maximum uplift, size of the fractured zone,
and volume of sand pumped into the fracture are detailed in Table 3-1.
13
-------
GRAVEL
CHF1 ^CHC2 GRASS
CHF2
CHF3
TOPSOIL
LEGEND
© WELL LOCATION
0
UPLIFT CONTOUR
VENTED FRACTURE
NOTE: WELL NO. CHF2 IS VENTED TO THE SURFACE.
NOT TO SCALE
Source: Modified from Wolf ond Murdoch, 1992
figure 3-I. Well Locations at the Center Hill Facility.
14
-------
Table 3-1. Fracture Characteristics at the Center Hill Facility
Fracture
Designation
No.
CHF1 - 5
CHF1 - 10
CHF2-5
CHF3 - 5
Maximum
Pressure
(psi)
50
Not available
59
68
Maximum
Uplift
(inches)
1.02
0.87
0.71
0.63
Size
(ft)
22 by 27
20 by 27
22 by 33
Not available
Volume of Sand
(cubic feet, ft3)
7
11
9
5
The maximum uplift was measured at stations along four radial lines. Measurement of uplift of the
ground surface can be correlated to thickness of sand in the fracture. The thickness of the sand was
also verified by collecting split-spoon samples. The GEMS was used to obtain real-time uplift data
during hydraulic fracturing. The data obtained from this system correlated well with measurements
made with a leveling telescope.
3.2 DATA ACQUISITION
The wells were designed primarily for vapor extraction, but are also capable of removing liquid.
These two-phase extraction wells are similar to the wells installed by Xerox at the Xerox Oak Brook
site. The wells consist of a 2-inchdiameter polyvinyl chloride (PVC) casing and screen into which a
0.5-inch-diameter tube is placed to the bottom of the well. Vacuum is applied to the 0.5-inch tube.
Water at the bottom of the well is removed along with the vapor when a valve connected to the
annulus between the 0.5-inch-diameter tube and the casing is opened. Pneumatic piezometers were
used to determine the distribution of pressure as a function of radial distance from the well. The
locations of pneumatic piezometers and uplift contours in the vicinity of Well No. CHF1 are shown in
Figure 3-2, and the locations of piezometers and uplift contours in the vicinity of Wells No. CHF2,
CHF3, CHC1, and CHC2 are shown in Figure 3-3.
A blower capable of generating 120 inches of water suction head was used to apply vacuum pressure
to the wells. A 20-gallon vapor-liquid separator was used to remove the contained water. Variable
area flow meters were used to measure air flow upstream of the vapor-liquid separator.
15
-------
19
20
t 15
18
21
LEGEND
PNEUMATIC PIEZOMETER LOCATION AND NUMBER
© WELL LOCATION AND NUMBER
- UPLIFT CONTOUR
NOTES: UPLIFT CONTOURS ARE SHOWN IN
MILLIMETERS (mm).
PIEZOMETERS WITH TWO NUMBERS ARE
SCREENED AT TWO DEPTHS.
Source: Modified from Wolf ond Murdoch. 1992
NOT TO SCALE
Figure 3-2. Well No. CHF1 Uplift Contours and Pneumatic Piezometer Locations.
16
-------
1234
».
I
'17
1
LEGEND
PNEUMATIC PIEZOMETER LOCATION AND NUMBER
© WELL LOCATION AND NUMBER
^ UPLIFT CONTOUR
" lililJI VENTED FRACTURE
NOTE: UPLIFT CONTOURS ARE SHOWN IN mm,
Source: Modified from Wolf ond Murdoch,1992
Figure 3-3. We||s No.CHF2. CHF3, CHC1, and CHC2 Uplift Contours and
Pneumatic Piezometer Locations.
17
NOT TO SCALE
-------
3.3 FRACTURING RESULTS
Two distinct periods of testing occurred at the Center Hill Facility, one during the winter from
January 20 to March 2, 1992, and the other during the summer, from June 8 to July 7, 1992. Air
yield and pressure distribution were measured for the five wells. These results are presented below.
Well yield as a function of time and rainfall for Wells No. CHF1 and CHC1 is presented in Figure 3-
4, and well yield as a function of time for Wells No. CHF2, CHF3, and CHC2 is presented in Figure
3-5. Figure 3-4 demonstrates that yields from Well No. CHC1 are about an order of magnitude less
than the yield from fractured Well No. CHF1. Sharp increases in yield follow the removal of water
from Well No. CHF1. The unfractured well was unaffected by rainfall and did not produce water.
Figure 3-5 demonstrates that the fractured wells yielded air flows about an order of magnitude higher
than unfractured Well No. CHC2 during the summer testing period. The air yields from fractured
wells decreased after rainfall, and the vented fracture was more affected by rainfall than the unvented
fracture. The vented fracture is connected to the ground surface and therefore produces higher yields
of both air and water than the unvented fracture.
The suction head (soil vacuum) measured by each piezometer varied throughout the tests and typically
increased after rainfall. Suction head near fractured wells was several times to roughly an order of
magnitude greater than at similar locations around the unfractured wells. Figure 3-6 presents the
pressure distribution near Well No. CHF1 and near conventional wells. The pressure drops off
rapidly near the conventional wells and is about 1 inch of water within 3 ft of the wells. The
pressure near the fractured well drops gradually and extends to a distance of 25 ft from the well.
Figure 3-7 depicts pressure distribution as a function of time and rainfall measured by a pneumatic
piezometer 10 ft from Well No. CHF2. Apparently, suction head decreases over time as soil dries,
and increases significantly after heavy precipitation. As shown in Figure 3-8, pressure also increases
with depth from the ground surface to the fracture and then decreases rapidly below the fracture.
The average yield, maximum yield, and the average radial distance of influence of each well is
summarized in Table 3-2.
18
-------
0)
*-r
3
c
0)
0.
0)
«*_
o
jQ
U
N_X
Q
U
I-
10.22
Rainfall (in.)
0.25
C3G3DDB OBOCl-
-Q-
B
BQBBSQ
0
T1IT
S~~iniiiiiiiiiiiiiniiiiiiniiiiininir~r
10 15 20 25 30 35 40 45
LEGEND
TIME (days)
WELL N O . CHF1
WELL NO. CHC1
Source: Modified from Wolf ond Murdoch. 1992
figure 3-4. Well Yield as a Function of Time and Rolnfall for Wells No. CHF1
and CHC1 January 20, to March 2. 1992,
-------
to
o
o
v^/
O
8'
7
6
5
4'
3
2
1
0.23
Rainfall (in.)
0.53 o.67
0.24
10 15 20
TIME (days)
25
30
35
LEGEND
CD WELL NO. CHF2 (VENTED. FRACTURED)
n WELL NO. CHF3 (FRACTURED)
A WELL NO. CHC2 (UNFRACTURED)
Source: Modified from Wolf and Murdoch. 1992
FIgur« 35. Well Yield as a Function of Time and Rainfall for Wells No.
CHF2. CHF3. and CHC2June 8 to July 7. 1992.
-------
k,
a>
| 10
"o
C/$
CD -«
u ^*
,E
a:
w 0.1
00
ijj
oC
n n 1
. -T, __T__T_..-r__^__1 -T ., r -p r T -T T- ~| -T T- r r j- ,__,_,,__,
] Approximate
" ^» limit of fracture
" \ V U"^rav~- --n
v ^Z.
cA
\
0
5
10
15
20
RADIUS WELL (ft)
w
' H»
r ;t)
O
V
D
LEGEND
CONVENTIONAL WELL PIEZOMETER READING
WELL NO, CHF1 PIEZOMETER READING
ABOVE FRACTURE
WELL NO. CHF1 PIEZOMETER READING
IN
Source: Modified from Wolf ond Murdoch, 1992
Figure 3-8, Pressure Distribution Near Well No. CHF'L
21
-------
NJ
0)
4-»
O
w
0>
O
c
LJ
o:
D
C/)
00
LU
o:
Q.
2
O
p
O
0.23~inches rainfall
Source: Modlflbd from Wolf ond Murdoch, 1992
TIME (days)
Figure 3-7. Pressure Distribution Versus Time and Rainfall Near
Well No. CHFP--June to 25. 1992.
-------
0)
--
o
5
0)
o
c
UJ
or
co
CO
UJ
n:
a.
o
p
o
CO
0
DEPTH (ft bgs)
LEGEND
A PIEZOMETER READING DURING DRY CONDITIONS
BB PIEZOMETER READING AFTER 1.3 INCHES OF RAINFALL
Source: Modified from Wolf and Murdoch. 1992
Figure 3-8. Pressure Distribution Versus Depth Near Well No. CHF2.
23
-------
Table 3-2. Performance of Wells at the Center Hill Facility
Well
No.
CHF1
CHF2
CHF3
CHC1
CHC2
Average Yield
(cfm)
3.7
6.7
3.4
0.33
0.59
Maximum Yield
(cfm)
6.1
7.2
4.05
0.38
1.25
Average Zone of Pneumatic
Control*
(ft)
25 to 30
20 to 2s
Is to 20
0.5 to 1
Less than 1
*Zone in which the pressure distribution can be controlled by varying the applied suction
head.
3.4 CONCLUSIONS
The air yield and zone of pneumatic control of extraction wells increase significantly after the creation
of sand-filled hydraulic fractures. The air yield at the Center Hill Facility tests increased by one
order of magnitude, and the fracture remained effective for 1 year at Well No. CHF1. The zone of
pneumatic control of the fractured wells was more than 10 times greater than that of the unfractured
wells
Rainfall affected the performance of vapor extraction wells by decreasing the yield and increasing the
suction head. The fractured wells yielded a larger amount of water than the unfractured wells, which
never produced water. This difference indicates that continuous recovery of both liquid and vapor
phases is essential to maximize yield from fractured vapor extraction wells. Also, infiltration of
water into the area where vapor extraction is conducted should be minimized to increase the efficiency
of vapor extraction.
Fractures that vent to the surface have gradients that drive flow through the soil toward the fracture.
However, such vented fractures increase infiltration, which reduces air yield from the well. Because
of site-specific conditions that increased water content in unvented fractured Well No. CHF3, vented
fractured Well No. CHF2 appeared to yield significantly higher flows than the unvented fractured
24
-------
well. Therefore, the effect of a vented fracture is not significantly different from that of an unvented
fracture.
25
-------
4.0 XEROX OAK BROOK SITE TESTS
At the Xerox Oak Brook site, contaminants consisting of trichloroethene (TCE); 1, 1,1-trichloroethane
(TCA); 1,1-dichloroethane (DCA), tetrachloroethene (PCE), and other solvents are present in silty
clay till to depths of 20 ft bgs. Xerox investigated the site in 1987. In 1991, a two-phase soil vapor
extraction (SVE) system was installed. The layout of the two-phase vapor extraction system is shown
in Figures 4-1 and 4-2. Vacuum is applied to the wells by a pump, and the water and vapor in the
soils around the vapor extraction wells is withdrawn. An inlet separator removes the water and a
discharge separator removes the water vapor. The remaining vapor is treated in a vapor-phase carbon
adsorption unit and clean air is vented out of the treatment building. The water is passed through a
liquid-phase carbon adsorption unit and discharged to the sewer system. The discharge water is
sampled to ensure that it meets sanitary sewer permit requirements.
The hydraulic conductivity at the site varies from 10"7 to 10"8 centimeters per second (cm/sec). This
low conductivity hampered the rate of vapor extraction. In an effort to enhance vapor extraction,
fractures were created at the site during the week of July 15, 1991. A work plan prepared by UC's
Center Hill Facility describes the pilot-scale study (UC, 199la). The pilot-scale demonstration
consisted of creating six hydraulic fractures at two locations. Figure 4-3 presents extraction well and
piezometer locations. Wells No. RW1 and RW2 are recovery wells in unfractured ground, and Wells
No. RW3 and RW4 are recovery wells in fractured ground. Before fracturing, soil samples were
obtained in the vicinity of the four wells to a depth of 15 ft bgs. Soil moisture content was measured
every foot bgs, and two samples from each borehole were analyzed for volatile organic compounds
(VOC). This work was performed in accordance with the Quality Assurance Plan prepared by
Xerox's subcontractor, Woodward-Clyde Consultants (Woodward-Clyde Consultants, 1991).
Hydraulic fracturing activities, data acquisition, fracturing results, and conclusions for the Xerox Oak
Brook site tests are discussed below.
4.1 HYDRAULIC FRACTURING ACTIVITIES
Hydraulic fractures were created at Wells No. RW3 and RW4 at depths of 6, 10, and 15 ft bgs.
Most of the fractures were gently dipping and 10 to 15-ft in radius, except the fracture at Well No.
RW4 at 6 ft bgs vented to the surface. Ground surface uplift measurements of up to 1.04 inches were
measured at radii of 11.5 and 16.4 ft from the fracturing hole.
26
-------
| VCLL NO,
J
XEROX FACILITY
TO VAPOR
EXTRACTION
SYSTEM
WFI I NO
J
- 3-INCH-DIAMETCR PIPING
2-INCH-DIAMETER PIPING
V.LL NO.
' , \WELL NO,
% RW4
LOW SHRUBBERY TO
OBSCURE PIPING
WINDSOR DRIVE
NOT TO SCALE
Source: Modified from Xerox, 1992
Figure 41, Xerox Oak Brook Site Piping Plan for Vapor Extraction System,
2?
-------
VAPOR
VAPOR
TREATMENT
VAPOR
VACUUM
PUMP
VAPOR
AND
LIQUID
VAPOR
AND
LIQUID
DISCHARGE
SEPARATOR
LIQUID
LIQUID
TREATMENT
LIQUID
WELL
SYSTEM
Source: Modified from Xerox. 1992
Figure 4-2. Schemotlc Diagram of Xerox Oak Brook Site Vapor Extraction System.
28
-------
TRUCK
DOCK
FACILITY
RW3
WD
* » H RW2
n AND
Ssyrss; edified from Xerox, 1092
NOT TO
29
Hgur« 43. Extraction
-------
A week after the fractures were created, recovery wells and monitoring boreholes were drilled.
Multilevel recovery wells consisting of separate screens and risers for each fracture were installed to
allow individual access to each fracture. Multilevel monitoring boreholes containing as many as six
pneumatic piezometers were installed at radial distances of 5, 10, 15, and 20 ft from each recovery
well (see Figure 4-3).
The six fractures were created on the same day, and each fracture required 1.5 to 2 hours to
complete. Essential details of the fractures are summarized in the table below. The details include
the depth bgs at the point where the fracture was created, the bulk volume of sand pumped into the
fracture, the volume of gel in the fracture, the maximum pressure at the point of injection, the
pressure at the end of pumping, the maximum uplift (typically not at the point of injection), and the
approximate radius of the uplifted area over the fracture. The radius of each fracture depends on the
amount of slurry pumped into the fracture.
Table 4-1. Fracture Characteristics at the Xerox Oak Brook Site
Fracture
Designation
No.
OXP1F1
OXP1F2
OXP1F3
OXP2F1
OXP2F2
OXP2F3
Depth
(ft bgs)
6.0
10.0
15.0
6.0
10.0
15.0
Sand
(ft3)
NA*
12
13
6
12
14
Gel
(gallons)
20
130
150
100
140
150
Maximum
Pressure
(psi)
22
38
55
25
45
72
End
Pressure
(psi)
20
8
34
8
10
35
Maximum
Uplift
(inches)
0.12
0.8
0.96
1.04
0.75
1.2
Radius
(ft)
NA*
13.1
16.4
11.5
13.1
14.8
Comment
Vented to surface
Recovery Well
No. RW4
Recovery
Well No. RW4
Recovery Well
No. RW3
Recovery
Well No. RW3
Recovery Well
*Not Applicable
30
-------
Xerox monitored the following parameters for the two fractured recovery wells (Wells No. RW3 and
RW4) and the two unfractured recovery wells (Wells No. RW 1 and RW2) from December 199 1 to
November 1992:
Water discharge from the system
Soil moisture content at depths of 4, 8, and 12 ft bgs, and at lateral distances of 5,
10, 15, and 20 ft north of the wells
Soil vacuum (suction head) at recovery wells and monitoring points
Vapor flow rates from recovery wells
On-line gas chromatography (GC) analyses of DCA, TCA, TCE, toluene, PCE,
ethylbenzene, and xylenes
The UC Center Hill Facility coordinated data acquisition with Woodward-Clyde Consultants for vapor
discharge and suction head from June to November 1992.
4.2 DATA ACQUISITION
Xerox's data acquisition system records the actual vapor flow rates from individual recovery wells.
Valving arrangements were available to measure flows from individual fractures in Wells No. RW3
and RW4. The wells were screened, as follows.
Well No. RW1 screened from 5 to 15 ft bgs
Well No. RW2 screened at three 1-foot intervals at 6, 10, and 15 ft bgs
Well No. RW3 screened at three 1-foot intervals at 6, 10, and 15 ft bgs
Well No. RW4 screened at three 1-foot intervals at 6, 10, and 15 ft bgs
UC Center Hill Facility researchers coordinated data acquisition with Woodward-Clyde Consultants
for vapor discharge from and suction head in Wells No. RW2, RW3, and RW4. The pneumatic
piezometers shown in Figure 4-3 were used to measure the suction head (soil vacuum), and variable
area flow meters were used to measure vapor discharge. Pressure readings in the piezometers were
obtained with a hand-held digital manometer with an accuracy of ±_ 0.2 inches of water. The
variable area flow meters measured vapor discharge from 6, 10, and 15 feet bgs in all four wells.
31
-------
Data were not obtained from Well No. RW1 since a leak existed in the annulus between the riser and
the borehole wall, allowing air from the surface to flow into the well.
Vapor discharge data from variable area flow meters was generally higher than data from vortex
shedding flow meters. After consultation with the meter manufacturer, it was concluded that vortex
shedding flow meters cannot accurately measure two-phase flow. Installation of liquid-vapor
separators upstream of the flowmeter to remove the liquid phase improved the performance of these
meters. However, because an inadequate number of vortex shedding flow meters were available to
automatically record the flow from each zone in each well, only variable area flow meter data were
used in data analysis. In order to improve the accuracy of variable area flow meter readings, a
demister pot was used to remove liquid from the airstream before it entered the meter.
4.3 FRACTURING RESULTS
The vapor discharge, contaminant recovery, and suction head results presented below were collected
from June to November 1992 and analyzed by UC Center Hill Facility researchers. The contaminant
concentrations obtained by Xerox were reviewed by the UC Center Hill Facility, and are summarized
below.
4.3.1 Vapor Discharge
The vapor discharge data for Wells No. RW2, RW3, and RW4 are presented in Table 4-2. The data
indicate that discharge from fractured Wells No. RW3 and RW4 is 15 to 20 times greater than from
unfractured Well No. RW2. The discharge rate versus time is plotted in Figure 4-4. The discharge
from fractured wells tends to fluctuate, and the discharge from unfractured Well No. RW2 is more
consistent. These fluctuations may be due to changes in the subsurface caused by precipitation
events.
The relationship between vapor discharge and precipitation was studied by plotting the water recovery
rate for the system over the same period (see Figure 4-5). The water recovery rate was obtained by
dividing the total water discharge from the system during a specific period by the number of days in
that period (typically 3 to 4 days). The water recovery rate fluctuated widely, ranging from 20 to as
many as 500 gallons per day (gpd). Higher water recovery rates generally produced low vapor
discharge rates. The inverse relationship between water recovery rate and vapor discharge rate
32
-------
Table 4-2. Well Discharge Readings at the Xerox Oak Brook Site
Well No.
RW2
RW3
RW4*
RW4**
Discharge
Rate
(aver age cfm)
0.1 to 4.6
2.2 to 22.0
27.9 to 42.7
17.1 to 29.7
Discharge
Rate
(average cfm)
1.1
14.3
34.2
22.6
Discharge
Percentage
(at 6 ft bgs)
46.3
61.2
36.0
Not
Applicable
Discharge
Percentage
(at 10 ft bgs)
27.3
8.4
41.0
Not Available
Discharge
Percentage
(at 15 ft bgs)
23.2
30.4
23.0
Not
Available
The 6-foot-deep fracture at Well No. RW4 vented to the surface. The data forthis
well includes discharge when suction is applied to all three of the fractures.
Well discharge average when suction was applied to the 10- and 15-foot-deep
fractures only; hence, well discharge smaller than when suction applied to all three of
the fractures
is demonstrated on days 116, 120, and 136 in Figure 4-5. The increased water recovery rate, in
general, related to significant rainfall events (see Figure 4-6). Therefore, Xerox decided to cover the
site with an impermeable membrane to minimize infiltration of water into the subsurface.
4.3.2 Contaminant Recovery
The mass recovery rate of a particular contaminant was determined as follows:
Mass Recovery Rate = Concentration x Flow x Molecular Weight of Contaminant x (1.53 x 1010)
where
Mass Recovery Rate is in pounds per hour
Concentration is in parts per billion (ppb) measured by the GC
Flow is in cfm
Molecular weight of contaminant is in grams per mole
1.53xlO'°is a constant
(4-1)
33
-------
UJ
<
o
cr
o
CL
<
50
40
30
20 4
10
0
0
50
100
150
TIME
Source; Modifi«
-------
UJ
500 t.
400 f-
300
CL
cn
QC
UJ
f~
0
20 40
80
80 100 120 140
Sourcts:
from UC. 1S93
Figure
Water Recovery Rat» Versus Tim*Jun* 23 to November 30, 1992,
-------
i .
U)
en
1.0 h
i
C
-s- 0.8 H
s i=
1 0,6 £
%_^ i.
3 0,4 t-
£ 1-
1 0,2 h
t
n n t,, ,..
f ., ,
i
1
i
1 i
u.
f
1,
«J=W
I , 1
;
ulJj
*
1
1
I ,1
1
J
s
i
i
i i
1 1?
JH
i
«JL_i_JJ
1
i
|
i
|
1
*JM
'
|
i
j
j_»L_»_i_i
1
i
I
\
1
I
_j
-j
"J
j
j
J
J
H
i i
I -j
-*_iliLi__(L3
0
20 40 60 80 100 120 140
Seure»: Modified from UC, 1893
TiME
4-6, at the Ook 23
to 30,
-------
The mass recovery rates for the seven target compounds were then added to give the total mass
recovery rate. Total rates were added for each fracture to give the total mass recovery rate for each
well. The total mass recovery rate was multiplied by the number of hours the well was operated per
day to obtain the cumulative contaminant mass recovered per well. The cumulative mass of
contaminants removed from Wells No. RW2, RW3, and RW4 versus time is presented in Figure 4-7.
The mass recovery rates from hydraulically fractured Wells No. RW3 and RW4 are approximately
one order of magnitude greater than that from unfractured Well No. RW2. The mass recovery rate
from all wells decreased with time. The mass recovery rates from the two fractured wells suggest a
difference between fractures that remain in the subsurface and those that vent to the surface.
Recovery rates from Well No. RW3, where all three fractures remained in the subsurface, was
approximately twice the recovery rates from Well No. RW4, where the 6-foot-deep fracture vented to
the ground surface. The high volumetric recovery rate in Well No. RW4 is apparently from flow
through the 6-foot-deep fracture which is not in contact with contaminated ground. Nevertheless, the
recovery from Well No. RW4 is markedly greater than from Well No. RW2.
Xerox monitored the recovery of contaminants from the site from December 1991 through November
1992. The recovery rate of contaminants decreased exponentially throughout the study period in a
manner consistent with SVE results at other sites. Maximum recovery occurred before June 1992,
and cumulative mass recovery from December 1991 through November 1992 was an order of
magnitude higher than the recovery rates presented in Figure 4-7.
4.3.3 Suction Head
Suction at the well heads and bgs near unfractured Wells No. RW1 and RW2 varied little over the
6-month period from June to November 1992. Suction head decreased abruptly with distance from
the well, from 245 to 285 inches of water to a few tenths of an inch of water at piezometers 5 feet
from Wells No. RW1 and RW2.
In fractured Well No. RW3, however, the suction head decreased gradually from 16 to 13 inches
between radial distances of 5 and 10 ft from the well at about 5 ft bgs, and decreased from 13 to 3
inches between radial distances of 10 to 15 ft from the well. The suction head was 1.2 inches, 25 ft
from the well. Therefore, creation of a hydraulic fracture apparently increased the distance where
suction head is affected by a well from a few feet to about 25 ft at the Xerox site.
37
-------
50
00
CO
a.
g
UJ
cr
20
10
0
iJ
0 20 40
80 100 120 140 160
riyn
WilkL
H WELL NO, RW2
A NO, RW3
WELL NO,
Sourcs: Modified from UC, 1993
TIME
Figure 4-7, Contaminants No, and
June 23 to 30,
-------
Suction head in the subsurface changed during the study period near fractured Wells No. RW3 and,
RW4. UC Center Hill Facility researchers concluded that the changes could be related to infiltration
of rainwater. Similar distributions of suction head were observed around fractured wells in silty clays
during the Center Hill Facility tests, and these distributions are consistent with theoretical analysis of
air flow near sand-filled hydraulic fractures (UC, 1993).
4.4 CONCLUSIONS
The measurement of vapor discharge, contaminant recovery, and suction head at the Xerox Oak
Brook site led to the following conclusions:
1. Vapor discharge from unfractured Well No. RW2 averaged 1.1 cfm, whereas it
averaged 14.3 cfm from fractured Well No. RW3 and 34.2 cfm from fractured Well
No. RW4. The difference in vapor discharge in Wells No. RW3 and RW4 appears to
result from air drawn from the ground surface through the 6-ft-deep fracture in Well
No. RW4, which vented to the surface. Fractured wells increased vapor discharge by
15 to 30 times higher than from unfractured wells.
2. The contaminant mass recovery rate from fractured wells was 7 to 14 times greater
than from the unfractured well.
3. Suction head was essentially nonexistent within a few feet of the unfractured well but
was detected at 25 ft from fractured Well No. RW3, demonstrating that the zone of
remediation may extend for distances of up to 25 ft from a fractured well; therefore,
fewer wells will be required to remediate a site.
4. Vapor discharge is inversely related to the amount of water recovered from the
subsurface.
39
-------
5.0 DAYTON SITE TESTS
At the Dayton site, six underground storage tanks (UST) were removed in December 1989. Three
tanks contained gasoline, one tank contained No. 2 fuel oil, and two tanks contained kerosene.
Laboratory analyses of soil samples collected from the UST excavations indicate that benzene
concentrations ranged from not detected (ND) to 622 microgram per kilogram (/tg/kg). Ethylbenzene
concentrations ranged from ND to 3,800 jig/kg; toluene concentrations from ND to 10,400 /ig/kg;
and xylene concentrations from ND to 41,900 /ig/kg. Total petroleum hydrocarbon (TPH)
compounds ranged in concentration from 32 to 8,550 /ig/kg; and total lead concentrations from 21 to
150jig/kg.
A remedial action contractor investigated the extent of contamination at the site in 1990. The
investigation revealed the following site characteristics:
The site is underlain by stiff, sandy to silty clay with traces of gravel.
The bedrock is shallow, at depths ranging from 15.5 to 17.0 ft bgs, and consists of
claystone and limestone.
The horizontal extent of hydrocarbons is limited to the tank excavation area and the
area east of the former tanks.
The vertical extent of hydrocarbons appears limited to the upper 6.5 to 16.0 ft bgs in
soils.
Fracturing activities, data acquisition, fracturing results, and conclusions for the Dayton site
demonstration are discussed below.
5.1 FRACTURING ACTIVITIES
The remedial action contractor initiated bioremediation activities at the site in 199 1. In July 199 1,
UC's Center Hill Facility proposed an investigation to determine the extent to which creating sand-
filled hydraulic fractures would enhance bioremediation of the site. A Quality Assurance Project Plan
(QAPP) was prepared by UC (UC, 1991b). The delivery of water containing hydrogen peroxide and
nutrients to sustain microorganisms through fractured wells was compared to the delivery of similar
water through conventional unfractured wells.
40
-------
Field tests were conducted from August 16 through 21, 1991. The tests consisted of a feasibility
study, where two hydraulic fractures were created at a location in uncontaminated ground near Well
No. SAD1 to verify that fractures could be successfully created in the contaminated areas. The
feasibility study was followed by the creation of seven fractures at two locations in contaminated
ground near Wells No. SAD2 and SAD3. After fracturing was completed, it was discovered that no
contamination existed near Well No. SAD-3. Figure 5-1 shows a site layout and these fracturing
locations. Essential details of these fractures are summarized below in Table 5-1 and include the
depth in ft bgs at the point where the fracture was created, the bulk volume of sand pumped into the
fracture, the volume of gel in the fracture, the maximum pressure at the point of injection, the
pressure after pumping, the maximum uplift (typically not at the point of injection), and the
approximate radius of the uplifted area over the fracture.
Table 5-1. Fracture Characteristics at the Dayton Site
Fracture
Designation
No.
SAD1-6
SAD1-12
SAD2-7
SAD2-8
SAD2-10
SAD2-12
SAD3-5
SAD3-7
SAD3-9
Depth
(ft bgs)
6
12
7
8
10
12
5
7
9
Sand
Volume
(ft3)
5
10
6
6
9
9
5
8
9
Gel
Volume
(gallons)
90
125
110
100
110
125
85
100
115
Maximum
Pressure
(psi)
27
60
42
17
37
42
19
43
39
End
Pressure
(psi)
7 to 12
15 to 22
7 to 11
7 to 15
10 to 20
18 to 26
3 to 7
7 to 10
12 to 17
Maximum
Uplift
(inches)
0.92
0.56
0.88
0.8
0.68
0.48
0.72
0.68
0.52
Radius
(ft)
14.8
21.0
15.1
14.8
16.4
23.0
14.8
15.4
23.0
Comment
Test fracture in
uncontaminatedl
soil
Contaminated
soil fractures
No contamination
detected
No contamination
detected
No contamination
detected
During the first week of September 1991, injection Wells No. SAD2 and SAD3 were installed. Soil
samples were obtained using a 2-inch-diameter split-spoon sampler at depths of 6 and 8 ft bgs and were
analyzed for moisture content, BTEX, and TPH. Results of samples collected during the first round of
sampling and also during second and third rounds of sampling are presented in Table 5-2.
41
-------
i
N
A
- FENCE
U. 0SA01 EXCAVATION
INTERCEPTION
THENCH
- RETURN
PUMP
o
LEGEND
INJECTION WEIL LOCATION AND
FRACTURED INJECTION WELL LOCATION
AND NUMBER
FLEXIBLE PVC PIPE
FRACTURED WELL LOCATION AND NUMBER
IN ^CONTAMINATED GROUND
NOT TO SCALE
Source: Modified from DC. 19916
figure 5-1, Fractured and Injection Well Locations at the Dayton Site,
42
-------
Samples were collected 5, 10, and 15 ft north of fractured Well No. SAD2 and unfractured Well No.
SAD4, and 10 ft south of Well No. SAD2.
Water containing hydrogen peroxide and nutrients was introduced into Well Nos. SAD2 and SAD4 in
December 1991. The unfractured well, Well No. SAD4, was filled with sand, and the water was gravity
fed by a 1-inch-diameter pipe grouted into place for delivery at 5 ft bgs. The water was gravity fed into
the fractured well which was screened from 6 to 12 ft and accessed fractures at depths of 7, 8, 10, and 12
ft bgs. The same head was applied during injection into fractured and unfractured wells. A system of
interception trenches and a return pump were installed to recover water injected into the wells (see
Figure 5-1).
5.2 DATA ACQUISITION
In February 1992, a second round of soil core samples were collected with a 2-inch-diameter by 2-ft-long
split-spoon sampler near the fractured and unfractured wells. The bottom 0.6 inch of soil from each core
was placed in a jar containing 0.08 gallon of methanol and was later analyzed for BTEX using the
methods described in EPA Test Methods for Evaluating Solid Waste, SW-846 (EPA, 1986). The
remainder of the core was wrapped tightly in sterile plastic and returned to a laboratory for analysis
within 72 hours. In the laboratory, the cores were aseptically cut into 1-inch-long sections. Alternating
sections were analyzed to quantify the number of microbes that have the capacity to degrade hydrocarbons
(expressed in CPUs), moisture content, and microbial activity. Moisture content was determined in
accordance with American Society for Testing and Materials (ASTM) Standard D2216 (ASTM, 1991).
A third round of sampling and analysis was conducted in July 1992 to measure the same parameters
measured in the second round and to determine the progress of bioremediation. Water flow rates into
Wells No. SAD2 and SAD4 were also measured throughout the demonstration.
The contaminant removal percentages for the fractured and unfractured wells are presented in Table 5-3.
Measurements of BTEX, TPH, CPU, and microbial activity produced erratic results for the second and
third rounds of sampling because water was not fed continuously during the period of the demonstration
(December 1991 through July 1992). The remediation contractor encountered mechanical problems
during the demonstration, and UC was not provided data on when the water was shut off and when it was
44
-------
Table 5-2. Analytical Results of Dayton Site Samples
Well No.
Fractured Well
No. SAD2
Unfractured
Well No. SAD4
Fractured Well
No. SAD2
Unfractured
Well No. SAD4
Fractured Well
No. SAD2
Unfractured
Well No. SAD4
Fractured Well
No. SAD2
Sampling
Location
5 ft north
of well
5 ft north
of well
10 ft north
of well
10 ft north
of well
15 ft north
of well
15 ft north
of well
10 ft north
of well
Sampling
Round
First
Second
Third
First
Second
Third
First
Second
Third
First
Second
Third
First
Second
Third
First
Second
Third
First
Second
Third
Average Concentration (mg/Kg)
Benzene
4.0
4.4
0.8
3.7
4.9
5.0
6.0
3.2
5.3
0.8
0.7
0.6
9.8
3.5
6.1
0.3
0.3
0.7
1.0
1.0
0.9
Ethyl-
benzene
15.0
0.4
6.0
8.9
8.1
6.6
20.0
4.3
20.0
2.1
0.6
0.2
26.0
7.1
11.4
0.2
TL
0.7
0.9
TL
2.3
Toluene
0.2
TLb
1.0
0.8
1.6
0.5
0.5
TL
2.8
0.1
TL
0.2
1.2
0.7
2.7
TL
TL
TL
TL
TL
1.1
TPH
490
112
143
230
235
104
235
98
108
75
55
25
385
188
123
8
11
6
290
131
57
Notes:
mg/Kg = milligrams per Kilogram
TL = Too low to measure
43
-------
Samples were collected 5, 10, and 15 ft north of fractured Well No. SAD2 and unfractured Well No.
SAD4, and 10 ft south of Well No. SAD2.
Water containing hydrogen peroxide and nutrients was introduced into Well Nos. SAD2 and SAD4 in
December 1991. The unfractured well, Well No. SAD4, was filled with sand, and the water was gravity
fed by a 1-inchdiameter pipe grouted into place for delivery at 5 ft bgs. The water was gravity fed into
the fractured well which was screened from 6 to 12 ft and accessed fractures at depths of 7, 8, 10, and 12
ft bgs. The same head was applied during injection into fractured and unfractured wells. A system of
interception trenches and a return pump were installed to recover water injected into the wells (see
Figure 5-1).
5.2 DATA ACQUISITION
In February 1992, a second round of soil core samples were collected with a 2-inch-diameter by 2-ft-long
split-spoon sampler near the fractured and unfractured wells. The bottom 0.6 inch of soil from each core
was placed in a jar containing 0.08 gallon of methanol and was later analyzed for BTEX using the
methods described in EPA Test Methods for Evaluating Solid Waste, SW-846 (EPA, 1986). The
remainder of the core was wrapped tightly in sterile plastic and returned to a laboratory for analysis
within 72 hours. In the laboratory, the cores were aseptically cut into 1-inch-long sections. Alternating
sections were analyzed to quantify the number of microbes that have the capacity to degrade hydrocarbons
(expressed in CPUs), moisture content, and microbial activity. Moisture content was determined in
accordance with American Society for Testing and Materials (ASTM) Standard D2216 (ASTM, 1991).
A third round of sampling and analysis was conducted in July 1992 to measure the same parameters
measured in the second round and to determine the progress of bioremediation. Water flow rates into
Wells No. SAD2 and SAD4 were also measured throughout the demonstration.
The contaminant removal percentages for the fractured and unfractured wells are presented in Table 5-3.
Measurements of BTEX, TPH, CPU, and microbial activity produced erratic results for the second and
third rounds of sampling because water was not fed continuously during the period of the demonstration
(December 1991 through July 1992). The remediation contractor encountered mechanical problems
during the demonstration, and UC was not provided data on when the water was shut off and when it was
44
-------
Table 5-3. Contaminants Removed at the Dayton Site
Treatment
Fractured Well
No. SAD2
Unfractured
Well No. SAD4
Fractured Well
No. SAD2
Unfractured
Well No. SAD4
Fractured Well
No. SAD2
Unfractured
Well No. SAD4
Fractured Well
No. SAD2
Location
From Well
Sftnorth
5 ft north
10 ft north
10 ft north
15 ft north
15 ft north
10 ft south
Sampling
Round
Second
Third
Second
Third
Second
Third
Second
Third
Second
Third
Second
Third
Second
Third
Benzene, Ethylbenzene, and
Toluene Removal
(Percent compared to first round)
Benzene
NI*
80
NI
NI
46.7
11.7
NI
NI
64.3
37.8
NI
NI
NI
NI
Ethyl-
benzene
97
60
7.9
37.0
78.5
NI
71.4
90.5
72.7
56.2
NI
NI
NI
NI
Toluene
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
TPH
Removed
(Percent
compared to
first round)
77%
71%
0%
55%
58%
54%
27%
67%
51%
68%
0%
25%
55%
80%
* No impact
45
-------
restarted. In addition, UC was unable to sample the water recovered from the trench. Hence, it is not
possible to determine the relative role of soil flushing and bioremediation in the removal of contaminants
at the site.
5.3 FRACTURING RESULTS
Flow rates in the unfractured and fractured wells are presented in Figure 5-2. Water flow was about 25
to 40 times greater in the fractured well. This increased flow resulted in higher moisture content near the
fractured well.
Results from the sampling location 5 ft north of the wells indicate that fractured Well No. SAD2 had
moisture contents of 1.4 to 4 times greater than the unfractured well. Moisture contents were generally
higher near the fracture, with the highest increase near the top fracture. This trend of increasing moisture
contents was also present at sampling locations 10 and 15 ft north of fractured Well No. SAD2.
The contaminant removal percentages near the fractured and unfractured wells shown in Table 5-3
indicate that benzene, ethylbenzene, and TPHs were significantly remediated near the fractured well. The
variability in removal percentages observed in the second and third rounds of sampling resulted from the
system not being optimized.
CONCLUSIONS
Fluid flow rates in the fractured well was 25 to 40 times higher than in the unfractured well. On certain
days, water flow rate near the unfractured well was minimal, but significant flow passed through the soil
around the fractured well. Fluid flow increased moisture content around the fractured well, with a
fourfold increase near the fractures. Hydrocarbon concentrations decreased in the soils around the
hydraulically fractured well, whereas they remained nearly the same near the unfractured well.
The bioremediation activities were conducted by an independent contractor, and UC had no control over
the operating parameters. Hence, reliable results on contaminant removal could not be obtained from the
pilot-scale demonstration. However, increased permeability near the fractured well was demonstrated by
higher flow rates and increased moisture contents.
46
-------
- v,^y~
o
r^ 0 1 s _,
u- V« 1 V
ce
UJ
i
^ o 1 ft. -
-^ v. 1 V
0
L£G£ND
+ WELL NO,
I NO,
Source; Modified from Vsspar, 1992
10
20 30
Time
47
40
50
60
Figure 5-2. Row of In No, and SAD4.
-------
6.0 QA/QC ANALYSIS
The objective of this study was to evaluate the effectiveness of sand-filled hydraulic fractures in enhancing
the remediation of contaminated clay soils. The study was conducted by UC Center Hill Facility
researchers in Cincinnati, at the Xerox Oak Brook site, and at the Dayton bioremediation site. The UC
Center Hill Facility prepared a QAPP (Project Category IV) that was approved by RREL for use at the
Center Hill Facility and the Dayton site (UC, 1990 and 1991b). At the Xerox Oak Brook site, \
Woodward-Clyde Consultants was responsible for data collection and prepared a quality assurance and
quality control (QA/QC) plan to be used in conjunction with the UC Center Hill Facility work plan for
the site (UC, 1991a and Woodward-Clyde Consultants, 1991).
Sampling and analysis at the Center Hill Facility was conducted solely by UC without independent
verification. Sampling and analysis at the Xerox Oak Brook site was conducted by Woodward Clyde
Consultants. The data obtained from variable area flow meters and pressure gauges at the Xerox Oak
Brook site by Woodward-Clyde Consultants were independently verified by UC researchers. These data
were the only critical parameters at the Xerox Oak Brook site. All data obtained at the Dayton site were
collected by UC without independent verification.
Hydraulic fracturing is a technology that enhances the effectiveness of other remediation technologies in
low permeability soils. The evaluation of this technology required measurement of vapor and water flow
rates, soil moisture contents, and soil vacuum pressures. Chemical analysis of soil and vapor samples
were primarily conducted to measure the progress of remediation. Hence, the QA/QC analyses were not
as rigorous as for technology evaluations that require extensive chemical analyses.
The QAPP developed by UC and the QA/QC plan prepared by Woodward-Clyde Consultants did not
specify the number of QA/QC samples to be collected for critical parameters during the pilot-scale
demonstrations. Also, the number of QA/QC measurements to be conducted for flow rate and suction
head were not provided for the Center Hill Facility tests and the two pilot-scale demonstrations. QA/QC
sample analyses for the Center Hill Facility and the Xerox Oak Brook and Dayton sites are discussed
below.
48
-------
6.1 CENTER HILL FACILITY
The work at the Center Hill Facility included the design, fabrication, and demonstration of hydraulic
fracturing field equipment. This work was conducted using a Project Category IV QAPP prepared by the
UC Center Hill Facility (UC, 1990). The objectives of the QA/QC analysis were to ensure that the
fracturing fluid was appropriate for use in fracturing and that the injection pressure and ground surface
uplift were accurately measured. QA objectives for critical measurements of the fracturing fluid are
provided in Table 6-1. The suction head was measured with a Meri Cal* Model DP 2001 gauge having
an accuracy of i. 0.1 percent. Air flow through the wells was measured with variable area flow meters
having an accuracy of ±_ 2 percent.
Sampling and analytical procedures, measurement of injection pressure and ground surface uplift,
calibration frequency, and data validation used for Center Hill Facility results are discussed below.
6.1.1 Sampling and Analytical Procedures
No laboratory measurements are associated with the Center Hill Facility tests. All measurements are field
measurements and are performed as the fractures are being created (except the suction head and air flow).
Measurements to determine the optimum fracturing fluid were made using nonstandard methods since no
standard methods exist for these measurements.
Grab samples of fracturing fluid were obtained at various points in the mixing procedures. The samples
were obtained by filling a bucket with fluid from the desired locations. Turbulence and in-line mixers
were used to ensure the fluid additives were well mixed so that representative samples could be collected.
Duplicate samples were collected at critical points in the fracturing procedure. If both samples were not
within the range specified or within specified precision objectives, a second duplicate was tested. If
results were still unacceptable, adjustments to the mix were made and additional duplicate samples were
tested until the mix met the stated specifications.
Samples were tested immediately after collection and results were recorded on a field data sheet. The
significant measurement in the mixing operation was the guar gum gel concentration (see Table 6-1). A
standard Marsh funnel was used to determine the time required for a 0.3-gallon sample to pass through
49
-------
Table 6-1. QA Objectives for Critical Measurements of Fracturing Fluid
Measurement
Degree of Mixing
Hydration Time
Guar Gum Gel Concentration
Guar Gum Post-Cross-link
Viscosity
Sand Concentration
Method
Qualitative Visual
Inspection
Minimum Hydration
Time: 25 minutes
Marsh Funnel Test:
44 to 46 seconds
Nationwise Funnel Test:
37 to 68 seconds
Weight of 0.053 gallon
sample: 0.573 to 0.705
pounds
Precision
Range"
NAk
NA
2 seconds
31 seconds
0.132 pounds
Accuracy
Percentage
NA
NA
90
90
90
Notes:
a Maximum absolute range, duplicates must both be within designated range
b NA = Not applicable
Source: Modified from UC, 1990
50
-------
the funnel. A Marsh funnel time of 44 to 46 seconds indicated optimum guar gum concentration. Other
parameters monitored during the preparation of the fracturing fluid include hydration time, guar gum gel
viscosity after the addition of cross-linker (an aqueous solution of borax), and sand concentration to
produce the desired guar gum gel-and-sand slurry.
The most important measurable property of the gel (guar gum) is its viscosity . The gel must be viscous
enough to keep the proppant (sand) in suspension, but not so viscous that it cannot be pumped effectively.
A field test was devised using a modified Marsh funnel (called a Nationwise funnel) to determine the
viscosity of the cross-linked gel. The Nationwide funnel is a long, steep-sided funnel having an orifice
with an inside diameter of 0.8 7 inch. Three-tenths gallon of the gel sample was poured into the funnel,
and the time taken for 0.24 gallon of the gel to flow through the funnel was measured. An acceptable
range of 37 to 68 seconds was established from these tests (see Table 6-1).
The final step in the production of the fracturing fluid is the addition of the sand proppant to produce the
gel-and-sand slurry. A high quality, well rounded, well sorted, 12/20 silica fracturing sand was used. A
sand concentration of 8 to 12 pounds per gallon provides a pumpable slurry with enough sand to prevent
fracture closure. Quality control of the slurry involves visual inspection of the sand as well as checks to
ensure that the sand is in the acceptable concentration range. Crosslinked gel and sand have specific
gravities of 0.98 and 2.65, respectively. The range of acceptable sand concentration is a weight of 0.573
to 0.705 pounds for a 0.053 gallon sample. Field measurements of the specified volume of fracturing
fluid ensured that the acceptable sand concentration was achieved.
6.1. 2 Measurement of Injection Pressure and Ground Surface Uplift
Injection pressure and ground surface uplift are noncritical parameters measured during hydraulic
fracturing . Injection pressures were measured at the ground surface using Druck* transducers interfaced
with a data acquisition system and laptop computer. The transducer calibration charts were obtained from
the manufacturer. A manual mechanical pressure gauge was available as a backup for the transducers.
The surface uplift was measured using standard surveying techniques. Elevations of points on a square
grid around the fracturing hole were measured using a leveling rod and a Deitzen* dumpy level . Uplift
51
-------
measurements were also made using a borehole extensiometer and GEMS. The readings obtained by
these systems were comparable.
6.1.3 Calibration Frequency
The flowmeter calibration was done prior to the field tests, and was checked whenever the gels and
slurries were found to be outside the acceptable range. The laboratory scale used for weighing guar gum
samples and slurries is calibrated and cleaned by a professional calibration service on a yearly basis, and
routine calibration checks with standard weights are made throughout the year. Manufacturer
recommended calibration procedures are used during the calibration.
6.1.4 Data Validation
Data quality was assessed continuously during the field fracturing process to ensure reliability of the data
collected. Fracturing fluid test data was monitored continuously to ensure that proper concentrations and
viscosities were obtained. Injection pressure plots were monitored in the field to determine anomalous
conditions. Surface elevation data were verified by using base station reference points. Three base
station points were situated as far as possible from the fracturing borehole to ensure that their ground
elevations were not affected by ground tilt resulting from the hydraulic fracturing operation. All data was
reviewed by the project principal investigator before acceptance.
After the initial runs, all fracturing fluid samples met the QA objectives. The injection pressures and
surface elevation data provided details on the orientation, thickness, and length of the fractures. There
were no deviations from the QAPP.
6.2 XEROX OAK BROOK SITE
The primary objective of Woodward-Clyde Consultants' field QA/QC program was to generate
scientifically representative, legally defensible data (Woodward-Clyde Consultants, 1991). Before the
startup of the demonstration, split-spoon soil samples and groundwater samples were collected from
fractured and unfractured wells. During well installation, continuous 2-ft-long split-spoon samples were
collected by Woodward-Clyde Consultants. These samples were divided into two equal portions. The
52
-------
first sample was prepared for VOC analysis by EPA Methods 8010 and 8020, and the second sample was
prepared for headspace screening and visual classification.
During the pilot-scale demonstration, continuous vapor samples were collected for VOC analysis by GC.
Tedlar bag vapor samples were also obtained for VOC analysis. Flow rates were measured using vortex
shedding flow meters and variable area flow meters. The soil vacuum pressure was measured using
pneumatic piezometers. Sampling and analytical procedures used by Woodward-Clyde Consultants are
described below. Calibration frequency and data validation procedures used by UC during hydraulic
fracturing are described in Sections 6.1.3 and 6.1.4.
Split-spoon soil samples were collected using a drill rig. Samples collected in jars were immediately
covered with aluminum foil, dull side down, and the tops of the jars were screwed on. After the samples
reached room temperature, the jar was unscrewed, and an analytical probe was punched through the
aluminum foil for headspace reading. Samples sent to the laboratory were tagged and labeled and
analyzed for VOCs.
Soil samples established the background levels of contamination at the site. The important parameters
measured to evaluate the effectiveness of hydraulic fracturing were flow rates from the fractured and
unfractured wells, and soil vacuum near the wells. Significant difficulty arose in measuring two-phase
flow from the wells using vortex shedding flow meters. As discussed in Section 4.2, a liquid separator
placed in line with these flow meters improved their performance but did not yield reliable, consistent
data. Hence, variable area flow meters with an accuracy of i 2 percent were used to measure vapor
flow from screened intervals in the wells. A demister pot used in conjunction with variable area flow
meters improved flow measurement accuracy. The soil vacuum readings were obtained with a pressure
gauge having .+.0.1 percent accuracy.
Vapor samples collected continuously for GC analysis and with Tedlar bags for analysis by EPA Methods
8010 and 8020 established the concentration of contaminants recovered from the site. However, the
concentration of contaminants recovered also depended on factors unrelated to the hydraulic fracturing
technology.
There were no deviations from the Quality Assurance Plan developed by Woodward-Clyde Consultants
53
-------
6.3 DAYTON SITE
Soil samples were collected by UC at 5, 10, and 15 ft north of fractured Well No. SAD2 and unfractured
Well No. SAD4. UC also collected soil samples 10 ft south of Well No. SAD2. These samples were
obtained in September 1991 (before the demonstration) and February and July 1992. Samples were
analyzed for moisture content, CPUs, microbial metabolic activity, and chemical composition. The
chemical composition analyses included benzene, toluene, ethylbenzene, and TPHs.
The hydraulic fracturing technology was evaluated by measuring the flow of water containing hydrogen
peroxide and nutrients through fractured and unfractured wells. The only critical parameter was the flow
rate. Parameters related to bioremediation activity were measured, and chemical analyses were conducted
to determine the extent of bioremediation near the fractured and unfractured wells. The parameters
related to bioremediation and chemical analyses did not yield statistically significant data because UC did
not have control on the introduction of water into the subsurface, and the general heterogeneity of the
soil.
The flow rate in the fractured and unfractured wells was measured with a flowmeter having an accuracy
of +. 2 percent. The flowmeter was calibrated prior to the pilot-scale demonstration, and at least once a
week during the duration of the demonstration. The QA/QC plan was followed for all non-critical
parameters. There were no deviations from the QA/QC plan.
54
-------
7.0 COST AND BENEFIT COMPARISON
The application of hydraulic fracturing to enhance vapor extraction at the Xerox Oak Brook site and
bioremediation at the Dayton site yielded the following benefits:
Vapor Extraction
Increased air flow from a radius of up to 25 ft from recovery wells, which is 15 to 30
times greater than flow observed in conventional wells
Increased mass removal rates of contaminants in vapor extraction system recovery wells
by about 7 to 14 times higher compared to conventional unfractured wells
Bioremediation
Increased flow in fractured wells of 25 to 40 times higher compared to conventional
unfractured wells; flow is directed mainly into the fractures, which extend up to 25 feet
from the inj ection well
Moisture content increased by 1.4 to 4 times near the fractures compared to no impact
near the conventional well
These improvements in remedial activities have been accomplished at a low cost increase associated with
the creation of sand-filled hydraulic fractures. The costs associated with the hydraulic fracturing
operation are detailed in Table 7-1. These costs are based on a SVE site because cost information is
available from the pilot-scale demonstration conducted at the Xerox Oak Brook site. The cost of drilling
the conventional well is not included. The number of wells required to remediate a site can be
significantly reduced by using hydraulic fracturing.
The site preparation cost includes the mobilization of equipment and rental of a bobcat to move material.
Permitting and regulatory costs are based on costs incurred at the Xerox Oak Brook site. There are no
startup costs because the trailer-mounted hydraulic fracturing equipment can be brought to a site and
immediately begin operation. There are no utility costs since hydraulic fracturing uses diesel- or
gasoline-powered pumps and the cost of fuel is included in the supply and consumable costs, The
technology does not treat wastes; therefore, there are no costs associated with effluent treatment and
55
-------
Table 7-1. Estimated Costs Associated With Hydraulic Fracturing
Estimated Daily Cost
Cost Category Q993 Dollars^)
1. Site Preparation $ 1,000
2. Permitting and Regulatory* 5,000
3. Capital Equipment Rental1" 1,000
4. startup 0
5. Labor 2,000
6. Supply and Consumables 1,000
7. Utilities 0
8. Effluent Treatment and Disposal 0
9. Residual and Waste Shipping and Handling 0
10. Analytical and Monitoring 700
11. Maintenance and Modifications 0
12. Demobilization* 400
Total One-Time Costs $ 5,400
Total Daily Costs $ 5,700
Estimated Cost per Fracture" $950 to $1,425
Notes:
One time costs
Capital equipment includes the following:
Equipment trailer Notching pump and accessories
Slurry mixer and pump Pressure transducer and display
Mixing pumps, tanks, and hose Uplift survey equipment
Fracturing lance and wellhead Scale
assembly Miscellaneous tools and hardware
Rental cost is based on 30 rentals per year and a depreciation of the $92,900
capital cost over 3 years.
Total daily costs (excluding one-time costs) divided by 4 or 6 fractures per
day
56
-------
disposal and residual and waste shipping and handling. Labor costs include the cost of four to five
persons and their per diem expenses. Supplies and consumables include sand proppant, guar gum
gel, enzyme, and diesel or gasoline (for running the pumps). Analytical and monitoring costs include
the cost of pneumatic piezometer installation near the fractured wells. Equipment maintenance and
modification costs would be incurred by the technology vendor and would be included in the rental
fee. The demobilization costs are estimated to be about $400 to move the equipment from Chicago,
Illinois, to Cincinnati, Ohio.
The cost per fracture is estimated to be $950 to $1,425, based on creating 4 to 6 fractures per day.
This cost is small compared to the benefits of enhanced remediation and the reduced number of wells
needed to complete the remediation.
57
-------
8.0 CONCLUSIONS
Conclusions from the tests conducted at the Center Hill Facility and the pilot-scale demonstrations
completed at the Xerox Oak Brook and the Dayton sites are discussed below.
8.1 CENTER HILL FACILITY TESTS
The vapor yield from a fractured well was about an order of magnitude higher than
from an unfractured well.
The zone of pneumatic control extended more than 10 times farther from the fractured
well than from the unfractured well.
Rainfall decreased vapor yield and increased suction head of fractured wells.
Unfractured wells were not affected by rainfall.
The effect of a vented fractured well was not significantly different from that of an
unvented fractured well.
8.2 XEROX OAK BROOK SITE TESTS
Fractured wells yielded vapor flow rates 15 to 30 times greater than unfractured
wells.
The vapor flow rate from fractured wells was adversely affected by precipitation.
The contaminant yields from the fractured well zones were 7 to 14 times greater than
from comparable zones in the unfractured wells.
8.3 DAYTON SITE TESTS
The flow of water was about 25 to 40 times greater in the fractured well than in the
unfractured well.
Moisture content increased in the vicinity of the fractured well, especially in the
fractured zones. Only a minor change in moisture content was detected in the
unfractured well.
58
-------
9.0 REFERENCES
American Society for Testing and Materials (ASTM), 1991, Annual Book of ASTM Standards.
Volume 4.08, Soil and Rock; Dimension Stone; Geosynthetics. ASTM, Philadelphia, PA
19103-1187, Standard D2216, Pages 278 to 281.
Hubbert, M.K. andD.G. Willis, 1957, Mechanics of Hydraulic Fracturing. Petroleum Transactions
AIME. Volume 210, Pages 153 through 168.
Murdoch, L.C., 1990, A Field Test of Hydraulic Fracturing in Glacial Till. Proc. 15th Annual
Research Symposium, Ohio. U.S. Environmental Protection Agency (EPA) 600/9-90-006.
Murdoch, L.C., 1993, Hydraulic Fracturing of Soil During Laboratory Experiments, Part I: Methods
and Observations; Part II: Propagation; Part III: Theoretical Analysis, Geotechnique, Volume
43, No. 2. Institution of Civil Engineers, London, Pages 255 to 287
University of Cincinnati (UC), 1990, Quality Assurance Project Plan (QAPP), Hydraulic Fracturing
Field Demonstration. Prepared December 27.
UC, 199la, Work Plan for Hydraulic Fracturing at the Xerox Oak Brook Site in Oak Brook, Illinois.
May 27.
UC, 1991b, QAPP, In Situ Biodegradation and Delivery of Oxygen and Nutrients to Subsurface
Microorganisms. Prepared August 15.
UC, 1993, Superfund Innovative Technology Evaluation (SITE) Field Demonstrations of Hydraulic
Fracturing. UC Center Hill Solid and Hazardous Waste Research (Center Hill) Facility
Unpublished Report.
U.S. Environmental Protection Agency (EPA), 1986, Test Methods for Evaluating Solid Waste,
Volumes IA-IC; Laboratory Manual, Physical/Chemical Methods; and Volume II Field
Manual, Physical/Chemical Methods. SW-846, Third Edition, Offtce of Solid Waste and
Emergency Response. Washington, D.C.
Vesper, S., 1992, Report on Bioremediation Studies at the Dayton Site. UC Center Hill Facility
Unpublished Report.
Woodward-Clyde Consultants, 1991, Hydraulic Fracturing Pilot Study, Quality Assurance and Quality
Control (QA/QC) Plan. Xerox Facility in Oak Brook, Illinois. June 1991.
Wolf, A. and L.C. Murdoch, 1992, The Effect of Sand-Filled Hydraulic Fractures on Subsurface Air
Flow: Summary of SVE Field Tests Conducted at the Center Hill Research Facility. UC
Center Hill Facility Unpublished Report.
Xerox Corporation (Xerox), 1992, Graphs Provided by Mr. Elliott Duffney, Environmental
Engineering Department. Webster, New York.
59
-------
HYDRAULIC FRACTURING TECHNOLOGY
APPLICATIONS ANALYSIS REPORT
UNIVERSITY OF CINCINNATI/RISK REDUCTION ENGINEERING LABORATORY
CINCINNATI, OHIO
U.S. Environmental Protection Agency
Risk Reduction Engineering Laboratory
Office of Research and Development
Cincinnati, OH 45268
-------
NOTICE
The information in this document has been prepared for the U.S. Environmental Protection Agency's
(EPA) Superfund Innovative Technology Evaluate (SITE) program under Contract No. 68-CO-0047.
This document has been subjected to the Agency's peer and administrative reviews and it has been
approved for publication as an EPA document. Mention of trade names or commercial products does
not constitute an endorsement or recommendation for use.
-------
FOREWORD
The SITE program was authorized by the Superfund Amendments and Reauthorization Act (SARA) of
1986. The program is administered by the EPA Office of Research and Development (ORD). The
purpose of the SITE program is to accelerate the development and use of innovative cleanup
technologies applicable to Superfund and other hazardous waste sites. This purpose is accomplished
through technology demonstrations designed to provide performance and cost data on selected
technologies.
This project consists of two pilot-scale demonstrations conducted under the SITE program to evaluate
the hydraulic fracturing technology developed by the University of Cincinnati (UC) and EPA's Risk
Reduction Engineering Laboratory (RREL). A full-scale demonstration using an EPA approved
Quality Assurance Project Plan (QAPP) has not been conducted for this technology. The technology
demonstrations were conducted at a Xerox Corporation vapor extraction site in Oak Brook, Illinois
(Xerox Oak Brook site); and at a bioremediation site near Dayton, Ohio (the Dayton site). The
demonstrations provided information on the performance and cost of the hydraulic fracturing
technology. Tests to determine the performance of fractures over a 1-year period were conducted at
an uncontaminated site at the Center Hill Research Facility (Center Hill) Cincinnati, Ohio. This
Applications Analysis Report provides an interpretation of the data and discusses the potential
applicability of the technology.
Copies of this report can be purchased from the National Technical Information Service (NTIS),
Ravensworth Building, Springfield, Virginia 22161, 703/487-4600. Requests should include the EPA
document number found on the report's cover. Reference copies of this report will be available at
EPA libraries as part of the Hazardous Waste Collection.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
in
-------
ABSTRACT
This report evaluates the effectiveness of the hydraulic fracturing technology, developed by the UC
and EPA's RREL, in enhancing the permeability of contaminated silty clays and presents economic
data from two pilot-scale SITE demonstrations.
The hydraulic fracturing technology creates sand-filled fractures up to 1 inch thick and 20 feet (ft) in
radius. These fractures are placed at multiple depths ranging from 5 to 40 ft below ground surface
(bgs) to enhance the efficiency of treatment technologies such as soil vapor extraction (SVE), in situ
bioremediation, and pump-and-treat systems.
The hydraulic fracturing technology was demonstrated in 1991 and 1992 at the Xerox Oak Brook site,
where SVE was in progress. On-site soil contamination included ethylbenzene; 1, Idichloroethane
(DCA); trichloroethene(TCE); perchloroethane (PCA); 1,1,1-trichloroethane (TCA); toluene; and
xylene. The vapor flow rates, soil vacuums, and contaminant yields of two hydraulically fractured
and two unfractured wells were compared. The fractured wells were fractured at 6, 10, and 15 ft
bgs. The vapor'yield from fractured wells was one order of magnitude greater than from unfractured
wells. This yield was obtained from an area 30 times greater than the area affected by the
unfractured well.
Another pilot-scale demonstration was conducted in 1991 and 1992 at the Dayton site where
bioremediation was being conducted. Site contamination included benzene, toluene, ethylbenzene,
and xylene (BTEX), and petroleum hydrocarbons. Fractures were created at 7, 8, 10, and 12 ft bgs
at one of two on-site wells. Water containing hydrogen peroxide and nutrients was pumped into the
hydraulically fractured well and into one unfractured well 50 ft from the fractured well. The injection
flow rates, soil moisture contents, microbial metabolic activity, numbers of colony forming units
(CPU), and rates of bioremediation at the fractured and unfractured wells were compared. In the
fractured well, the injection flow rate was 25 to 40 times greater and the rate of bioremediation was
higher for benzene, ethylbenzene, and petroleum hydrocarbons.
Possible sites where this technology is applicable include Superfund and other hazardous waste sites
that have soil and ground water contaminated with organic compounds. The technology is to be used
IV
-------
in conjunction with remediation techniques such as SVE, in situ bioremediation, and pump-and-treat
systems to enhance their effectiveness. Economic data indicate that the capital cost for hydraulic
fracturing equipment is $92,900 and the cost of renting the equipment is $1,000 per day. Rental,
operating, and monitoring costs for creating a fracture range from $950 to $1,425, depending on site-
specific conditions. Typically, two to three fractures are created per well, and four to six fractures
can be created in 1 day. The cost of creating a fracture is not materially affected by the depth of
fracture for depths ranging from 5 to 40 ft bgs. The cost is also unaffected by the type of soil
encountered,
-------
TABLE OF CONTENTS
Section Page
FOREWORD iii
ABSTRACT iv
LIST OF ACRONYMS AND ABBREVIATIONS ix
ACKNOWLEDGEMENTS x
EXECUTIVE SUMMARY ES-1
1.0 INTRODUCTION -\.-\
1.1 PURPOSE, HISTORY, AND GOALS OF THE SITE PROGRAM 14
1.2 DOCUMENTATION OF THE SITE DEMONSTRATION RESULTS' i_2
1.3 PURPOSE OF THE AAR 1-3
1.4 TECHNOLOGY DESCRIPTION 1-3
1.5 FRACTURING PROCEDURE 1-4
1.6 KEY CONTACTS 1-8
2.0 TECHNOLOGY APPLICATIONS ANALYSIS 2-1
2.1 SITE DEMONSTRATION OBJECTIVES AND CONCLUSIONS 2-1
2.1.1 Center Hill Tests 2-1
2.1.2 Xerox Oak Brook Site Tests 2-2
2.1.3 Dayton Bioremediation Site Tests 2-3
2.2 OTHER DEMONSTRATION RESULTS 2-3
2.3 EFFECTIVENESS OF THE HYDRAULIC FRACTURING TECHNOLOGY . 2-4
2.4 FACTORS INFLUENCING EFFECTIVENESS 2-5
2.4.1 Site Characteristics 2-5
2.4.2 Rainfall Infiltrating Into the Site 2-6
2.4.3 Operating Parameters 2-6
2.5 PERSONNEL REQUIREMENTS 2-7
2.6 POTENTIAL REGULATORY REQUIREMENTS 2-7
2.7 APPROPRIATE WASTE AND SITE CONDITIONS' 2-8
3.0 ECONOMIC ANALYSIS 3-1
3.1 SITE-SPECIFIC FACTORS AFFECTING COST 3.3
3.2 BASIS OF ECONOMIC ANALYSIS 3.3
3.3 COST CATEGORIES 3.4
vi
-------
TABLE OF CONTENTS
Section Page
3.3.1 Site Preparation Costs 3-4
3.3.2 Permitting and Regulatory Costs 3-4
3.3.3 Capital Equipment Costs 3-5
3.3.4 Labor Costs 3-5
3.3.5 Supply and Consumables Costs 3-6
3.3.6 Analytical and Monitoring Costs 3-6
3.3.7 Demobilization and Decontamination Costs 3-6
3.4 COSTS PER FRACTURE 3-6
4.0 REFERENCES 4-1
Appendix
A VENDOR'S CLAIM REGARDING HYDRAULIC FRACTURING
B SITE DEMONSTRATION RESULTS
C DESCRIPTION OF FRACTURING TESTS
LIST OF TABLES
Table Page
3-1 ESTIMATED COSTS ASSOCIATED WITH HYDRAULIC FRACTURES. 3-2
B-l FRACTURE CHARACTERISTICS AT XEROX OAKBROOK SITE B-6
B-2 SUMMARY OF WELL DISCHARGE READINGS B-7
B-3 FRACTURE CHARACTERISTICS AT THE DAYTON SITE B-12
B-4 CONTAMINANTS REMOVED AT THE DAYTON SITE B-16
vn
-------
LIST OF FIGURES
Figure Page
1-1 SEQUENCE OF OPERATIONS FOR CREATING HYDRAULIC FRACTURES 1-5
1-2 FRACTURING LANCE USED TO PREPARE BOREHOLES FOR HYDRAULIC
FRACTURES 1-6
1-3 SLURRY MIXING AND PUMPING EQUIPMENT MOUNTED ON TRAILERS 1-7
1-4 PRESSURE VERSUS TIME DURING THE CREATION OF
HYDRAULICFRACTURES 1-9
B-l XEROX OAK BROOK SITE PIPING SYSTEM PLAN FOR VAPOR EXTRACTION . . B-2
B-2 SCHEMATIC DIAGRAM OF XEROX OAK BROOK SITE VAPOR EXTRACTION
SYSTEM B-3
B-3 EXTRACTION WELL AND PIEZOMETER LOCATIONS B-4
B-4 VAPOR FLOW RATES IN WELLS NO. RW2, RW3, AND RW4 B-9
B-S CONTAMINANTS REMOVED FROM WELLS NO. RW2, RW3, AND RW4 B-10
B-6 DAYTON SITE FRACTURING AND INJECTION WELL LOCATIONS B-14
B-7 FLOW VOLUMES OF INJECTED WATER IN WELLS NO. SAD2 AND SAD4 ... B-l5
C-l HYDRAULIC FRACTURE AND TEST WELL LOCATIONS, ADDISON, ILLINOIS . C-2
C-2 LOCATION OF PROPOSED WELLS AND MONITORING PROBES,
GRAND LEDGE, MICHIGAN , C-5
C-3 PRESSURE VERSUS TIME DURING THE CREATION OF A HYDRAULIC
FRACTURE, GRAND LEDGE, MICHIGAN , C-6
₯111
-------
LIST OF ACRONYMS AND ABBREVIATIONS
AAR
ARAR
bgs
BTEX
Center Hill
Cfm
CPU
cm/sec
DCA
EPA
FDA
ft
GEMS
gpm
min
NA
ND
NI
OD
ORE)
OSHA
OSWER
PCA
PPb
psi
RCRA
RREL
SARA
SITE
SVE
Applications Analysis Report
Applicable or relevant and appropriate requirements
Below ground surface
Benzene, toluene, ethylbenzene, and xylene
Center Hill Research Facility
Cubic feet per minute
Colony forming units
Centimeter per second
1,1-Dichloroethane
U.S. Environmental Protection Agency
Fluorescein Diacetate Analysis
feet
Ground Elevation Measurement Systems
gallons per minute
Minute
Not applicable
Not detected
No impact
outside diameter
Office of Research and Development
Occupational Safety and Health Administration
Office of Solid Waste and Emergency Response
Perchloroethane
Parts per billion
Pounds per square inch
Resource Conservation and Recovery Act
Risk Reduction Engineering Laboratory
Superfund Amendments and Reauthorization Act
Superfund Innovative Technology Evaluation
Soil vapor extraction
IX
-------
TCA 1,1, 1-Trichloroethane
TCE Trichloroethene
TER Technology Evaluation Report
TPH Total petroleum hydrocarbon
UC University of Cincinnati
jig/kg Microgram per kilogram
UST underground storage tank
V O C Volatile organic compounds
Xerox Xerox Corporation
-------
ACKNOWLEDGEMENTS
This report was prepared under the direction of Naomi Barkley, the EPA SITE project manager at
RREL in Cincinnati, Ohio. Contributors and reviewers for this report were Messrs. Allen Wolf and
Mark Kemper, Dr. Lawrence Murdoch, and Dr. Steven Vesper of the UC in Cincinnati, Ohio; and
Mr. Elliott Duffney of the Xerox Corporation in Rochester, New York. Peer reviewers were Dr.
Ronald Lewis and Robert Stenburg of RREL.
This report was prepared by Drs. V. Rajaram and Pinaki Banerjee, of PRC Environmental
Management, Inc. (PRC). The report was typed by Ms. Cheryl Vaccarello, edited by Ms. Shelley
Fu, and reviewed by Dr. Kenneth Partymiller and Mr. Stanley Labunski, all of PRC.
XI
-------
EXECUTIVE SUMMARY
Hydraulic fracturing has been used by the petroleum industry to create fractures in deep rock
formations to enhance the flow of oil and gas to recovery wells. The University of Cincinnati (UC)
and the U.S. Environmental Protection Agency (EPA) Risk Reduction Engineering Laboratory
(RREL) developed the hydraulic fracturing technology in 1990 for creating fractures in low
permeability soils to enhance the efficiency of treatment technologies such as soil vapor extraction
(SVE), bioremediation, and pump-and-treat systems. The hydraulic fracturing technology creates
sand-filled fractures up to 1-inch thick and 20 feet (ft) in radius at multiple depths ranging from 5 to
40 ft below ground surface (bgs).
The hydraulic fracturing technology was evaluated under EPA Superfund Innovative Technology
Evaluation (SITE) program. Pilot-scale demonstrations were conducted in 1991 and 1992, at a Xerox
Corporation (Xerox) site in Oak Brook, Illinois (the Xerox Oak Brook site), where vapor extraction
was being conducted, and at a site near Dayton, Ohio (the Dayton site), where bioremediation was
used. Testing was also conducted by UC researchers at the Center Hill Research facility in 1992 to
study the factors affecting the performance of hydraulic fractures in enhancing air flow through silty
clays.
Hydraulic fracturing involves mixing a granular solid (termed a proppant), which is usually sand,
with a viscous fluid (guar gum and water mixture), and pumping the mixture into a fracture as it
grows away from the well. An enzyme added to the viscous fluid breaks down the gel, leaving the
sand to hold open the fracture. Hydraulic fracturing equipment can be mounted on a trailer and
includes a mixer, a slurry pump, and storage tanks. The fracturing process begins by the use of a
high-pressure water jet to cut a disk-shaped notch extending 0.5 ft from the borehole wall of a well at
which fractures are to be made. The slurry pump injects a mixture of sand and guar gum, water, and
enzyme into the notch at a rate of 10 to 25 gallons per minute (gpm) until a critical pressure is
reached and a fracture is propagated. The process is repeated at greater depths to produce a stack of
multiple sand-filled hydraulic fractures.
ES-1
-------
The purpose of this Applications Analysis Report (AAR) is to present information from the two SITE
pilot-scale demonstrations that can be used to implement the hydraulic fracturing technology at
Superfund and Resource Conservation and Recovery Act (RCRA) hazardous waste sites.
Section 1.0, the introduction, discusses the purpose, history, and goals of the SITE program;
discusses documentation of SITE demonstration results; presents the purpose of this AAR; provides a
technology description; describes the fracturing procedure; and presents a list of key contacts.
Section 2.0 presents a technology applications analysis that discusses other demonstration results,
factors influencing the technology's effectiveness, SITE demonstration objectives and conclusions,
personnel requirements, potential regulatory requirements, and appropriate waste and site conditions.
Section 3.0, the economic analysis, summarizes site-specific factors affecting costs, the basis of the
economic analysis, the cost categories used to determine hydraulic fracturing costs, and costs per
fracture. References are presented in Section 4.0. Appendix A provides the vendor's claims
regarding the hydraulic fracturing technology, Appendix B summarizes the SITE demonstration
results, and Appendix C describes fracturing tests conducted at two sites.
The rest of this Executive Summary presents an overview of the SITE demonstrations, results from
the demonstrations, waste applicability, an economic analysis, and conclusions for the hydraulic
fracturing technology.
OVERVIEW OF THE SITE DEMONSTRATIONS
The SITE demonstrations at the Xerox Oak Brook site and the Dayton site, and the tests conducted at
Center Hill had the following objectives:
To assess the technology's ability to create sand-filled hydraulic fractures in silty clays
and study the factors that affect these fractures over a period of 1 year
To evaluate the technology's ability to significantly enhance SVE and contaminant
removal at the Xerox Oak Brook site
To determine the efficiency of hydraulic fracturing in delivering water containing
hydrogen peroxide and nutrients to the Dayton site, which is contaminated with
petroleum products
ES-2
-------
To develop information required to estimate the costs for the technology
The Center Hill tests were conducted for over 1 year in uncontaminated silty clays to determine the
effects of single and multiple depth hydraulic fractures on the enhancement of air flow through the
soil. In addition, the effect of rainfall on fracture performance was studied.
The Xerox Oak Brook site contained silty clays contaminated with ethylbenzene, 1,1-dichloroethane
(DCA), trichloroethene (TCE), perchloroethane (PCA), 1, 1, 1-trichloroethane (TCA), toluene, and
xylene. Two out of four wells being used for two-phase soil vapor extraction (S VE) were fractured at
depths of 6, 10, and 15 ft bgs. Over a period of one year, the soil vapor flow rates, suction head,
and contaminant removal rates were measured and compared for the fractured and unfractured wells.
The Dayton site contamination included benzene, toluene, ethylbenzene, and xylene (BTEX), and
petroleum hydrocarbons. One out of two wells was fractured at depths of 7, 8, 10, and 12 ft bgs.
Water containing hydrogen peroxide and nutrients was gravity fed into these wells intermittently for
about 6 months. The site operator was responsible for this activity, and UC Center Hill was
responsible for monitoring the progress of bioremediation in the vicinity of the fractured and
unfractured wells. Two rounds of sampling were conducted at locations 5, 10, and 15 ft north of the
fractured and unfractured wells after bioremediation was in progress for 1 and 6 months. Soil
samples only were obtained and analyzed for moisture content, microbial metabolic activity, number
of colony forming units (CPU), BTEX, and petroleum hydrocarbons.
RESULTS FROM THE SITE DEMONSTRATIONS
The Center Hill tests show that the pneumatic control zone, which is the zone in which the pressure
distribution can be controlled by varying the applied suction head, extended more than 10 times
farther from the fractured well than from the unfractured well. The air yield from the fractured well
was one order of magnitude higher than that from the unfractured well. Rainfall affected the
performance of vapor extraction wells by decreasing the air yield and increasing the suction head.
The vapor extraction demonstration at the Xerox Oak Brook site involved two-phase vapor extraction
that separated water from vapor. The vapor yield from hydraulically fractured wells was
ES-3
-------
approximately one order of magnitude greater than from unfractured wells. The hydraulically
fractured wells enhanced remediation over an area more than 10 times than that remediated by
unfractured wells. The contaminant yields from the fractured well zones were approximately an order
of magnitude greater than from comparable zones in the unfractured wells. Results from the Xerox
Oak Brook site agreed with findings from Center Hill tests on the adverse impacts of rainfall.
The Dayton site results show that the water flow rate into the fractured well is 25 to 40 times greater
than into the unfractured well. Because UC Center Hill did not control the bioremediation activities
at the site, the amount of water fed into the fractured and unfractured wells was erratic during the 6-
month testing period, resulting in anomalous findings from the second and third rounds of sampling.
However, in the fractured well, the rate of bioremediation was higher for benzene, ethylbenzene, and
petroleum hydrocarbons.
WASTE APPLICABILITY
The hydraulic fracturing technology can be applied to low permeability (less than 10"7 cm/s) silty
clays or rock, and used to improve remedial methods that target organic compounds. The technology
is effective up to depths of 40 ft bgs, and minimizes the number of wells needed for in situ
remediation of the site. Potential sites for applying this technology to contaminated soils include
Superfund and RCRA corrective action sites where solvents and/or petroleum hydrocarbons have
spilled. Horizontal compressive stress that is greater than vertical stress in overconsolidated clays
favors the propagation of horizontal fractures. Horizontal fractures are effective in increasing the
permeability of the soil over larger radial distances than steeply dipping fractures; hence,
overconsolidated clays are preferred sites for application of the hydraulic fracturing technology.
ECONOMIC ANALYSIS
The economic analysis was performed to determine the costs of creating a fracture using the pilot-
scale, trailer-mounted equipment assembled by UC. The cost of creating a fracture varies from $950
to $1,425, depending on site-specific conditions. Four to six fractures can be created per day at one
location, and typically; two to three fractures are created in one well.
ES-4
-------
CONCLUSIONS
Creating sand-filled hydraulic fractures in the vicinity of a vapor extraction well significantly affected
both the vapor yield and the area influenced by the well. The vapor yield increased one order of
magnitude for a fractured well, and the distance influenced by the well was more than 10 times
greater than for a well without hydraulic fractures. Rainfall adversely affected the performance of
vapor extraction wells by decreasing yield and increasing the suction head.
The amount of water introduced in the vicinity of a fractured well was 25 to 40 times greater than
that in the vicinity of an unfractured well, significantly enhancing in situ bioremediation of soils in the
vicinity of the fractured well.
ES-5
-------
1.0 INTRODUCTION
This section provides the purpose, history, and goals of the SITE program; documentation of the
SITE demonstration results; the purpose of this AAR; the hydraulic fracturing technology; a
description of the fracturing procedures; and a list of contacts.
1.1 PURPOSE, HISTORY, AND GOALS OF THE SITE PROGRAM
In response to the Superfund Amendments and Reauthorization Act of 1986 (SARA), EPA's Office of
Research and Development (ORD) and Office of Solid Waste and Emergency Response (OSWER)
established the SITE program to (1) accelerate the development, demonstration, and use of new or
innovative technologies to clean up Superfund sites; (2) foster further investigation and development
of treatment technologies that are still at the laboratory scale; and (3) demonstrate and evaluate new
or innovative measurement and monitoring technologies.
The primary purpose of the SITE program is to enhance the development and demonstration of
innovative technologies applicable to Super-fond sites to promote their commercial availability. Major
goals of the SITE program are as follows:
Identify and remove impediments to the development and commercial use of
alternative technologies
Demonstrate the more promising innovative technologies to establish reliable
performance and cost information for site cleanup decision making
Develop procedures and policies that encourage selection of available alternative
treatment remedies at Superfund sites
Structure a development program that nurtures emerging technologies
EPA recognizes that a number of forces inhibit the expanded use of alternative technologies at
Superfund sites. One of the objectives of the program is to identify these impediments and remove
them or to develop methods to promote the expanded use of alternative technologies.
1-1
-------
Another objective of the SITE program is to demonstrate and evaluate selected technologies. This
significant ongoing effort involves ORD, OSWER, EPA regions, and the private sector. The
demonstration program tests field-ready technologies and provides Superfund decision makers with the
information necessary to evaluate the use of these technologies for future cleanup actions.
Other aspects of the SITE program include developing procedures and policies that match available
technologies with wastes, media, and sites for actual remediation, and assisting in the development of
emerging innovative technologies from the laboratory- or bench-scale to the full-scale stage.
Technologies chosen for a SITE demonstration must be innovative, pilot- or full-scale applications,
and offer some advantage over existing technologies. Mobile technologies are of particular interest.
1.2 DOCUMENTATION OF THE SITE DEMONSTRATION RESULTS
The results of each SITE demonstration are incorporated in two documents: the technology
evaluation report (TER) and the AAR. The TER provides a comprehensive description of the
demonstration and its results. A likely audience for the TER are engineers responsible for performing
an in-depth evaluation of the technology for a specific site and waste situation. These technical
evaluators seek to understand the performance of the technology in detail during the demonstration
and the advantages, risks, and costs of the technology for the given application. This information is
used to produce conceptual designs of sufficient detail for evaluators to estimate preliminary costs for
the demonstrated technology.
The AAR is intended for technical decision makers responsible for screening available remedial
alternatives. The AAR discusses factors such as site and waste characteristics that have a major
impact on cost and performance. If the candidate technology appears to meet the needs of the site
engineers, a more thorough analysis will be conducted based on the TER, AAR, and information
from remedial investigations for the specific site.
1-2
-------
1.3 PURPOSE OF THE AAR
To encourage the general use of demonstrated technologies, EPA provides information in the AAR on
each technology's applicability to certain sites and wastes and presents the costs of these applications
based on data from pilot- and full-scale demonstrations. These AARs synthesize available information
on the technology and draw reasonable conclusions about the technology's broad range of
applicability. The AAR is useful to those considering the technology for Superfund and other
hazardous waste site cleanups and represents a critical step in the development and commercialization
of the treatment technology.
Each SITE demonstration evaluates a technology's performance in remediating a site contaminated
with a particular waste Thus, the successful demonstration of a technology at one site does not
ensure that it will work equally well at other sites. Data obtained from the demonstration should be
used along with other information and case studies to estimate the total operating range over which
the technology performs satisfactorily.
1.4 TECHNOLOGY DESCRIPTION
Hydraulic fracturing has been successfully used by the oil industry to enhance oil recovery from deep,
low permeability rock formations. The mechanics of hydraulic fracturing in rock formations is well
understoo d(Hubbert and Willis, 1957). In 1990, a team led by Dr. Murdoch of UC completed
theoretical and laboratory investigations of hydraulic fracturing in low permeability soils. With
funding from EPA 's RREL, the team from UC conducted field experiments, and in 1991, submitted a
proposal for demonstrating this technology under the SITE program. EPA selected the hydraulic
fracturing technology for demonstration under the SITE program in July 1991.
The hydraulic fracturing technology is designed to create sand-filled fractures up to 1-inch thick and
30 ft in radius. These fractures, when created at several depths from 5 to 40 ft bgs, increase the in
situ permeability of contaminated soil. This increased permeability promotes the flow of vapors and
liquids through the soil and enhances the effectiveness of SVE, bioremediation, and pump-and-treat
remediation techniques.
1-3
-------
The hydraulic fracturing equipment consists of a continuous slurry-mixer and positive displacement
pump mounted on a trailer. A typical sequence of operations for creating hydraulic fractures is
shown in Figure 1-1. Equipment and material required is as follows:
A piston pump or a progressive cavity pump to inject slurry
A continuous mixer for creating the slurry, which consists of up to one part of
granular solid and two parts of viscous fluid
A fracturing lance composed of an outer casing and an inner rod, both of which are
tipped with hardened cutting surfaces that form a conical point, to prepare boreholes
used for hydraulic fracturing (see Figure 1-2)
Steel tubing with a narrow orifice at one end
Granular solid, termed proppant, which is usually a coarse sand
A viscous fluid to carry the proppant into the fracture. This fluid is a mixture of guar
gum gel, water, and an enzyme that breaks down the gel after the proppant has been
deposited into the fracture.
A trailer on which the slurry mixer and pump are mounted (see Figure 1-3)
1.5 FRACTURING PROCEDURE
A borehole is drilled using 6 or 8-inch outside diameter (OD) hollow-stem augers. Individual
segments of the rod and casing are 5 ft long and are threaded together as required by fracture depth.
The tip of the fracturing lance is driven to a depth where a fracture is to be created. The lance is
removed, leaving soil exposed at the bottom of the casing (see Figure 1-1). Steel tubing with a
narrow orifice at one end is inserted into the casing, and water is pumped through the tubing to create
a high-pressure water jet. The water jet, which has a pressure of about 3,500 pounds per square inch
(psi), is rotated within the borehole and produces a disc-shaped notch extending 4 to 6 inches from
the borehole (see Figure 1-1). A simple measuring apparatus comprised of a steel tape extending the
length of the tube and making a right angle bend at the end of the tube can be inserted into the casing
to measure the radius of the notch.
1-4
-------
1
f
I- a
j:
Rod
tip
!
2 "I^
of
4
I
o
Cutting jet
Source: Mod!fi«) from University of rir,--irmnti
1991
11, of for
rrn
O
from so''
1-5
-------
Source: Modified from University of Cincinnati, 1991
Figure 1-2. Fracturing Lance Used to Prepare Boreholes for Hydraulic Fracturing
I-6
-------
Figure 1-3. Slurry Mixing and Pumping Equipment Mounted on Trailers
I-7
-------
Sand slurry is produced by mixing one part of granular solid with two parts of viscous fluid in the
continuous mixer. A hydraulic fracture is created by pumping a predetermined volume of slurry at
rates of 10 to 25 gallons per minute (gpm). Lateral pressures from the soil on the outer wall of the
casing effectively seals the casing and prevents leakage of the slurry. The fracture nucleates at the
notch and grows up to 30 ft from the borehole wall.
The direction and distance of propagation of the fracture is measured by monitoring the uplift of the
ground surface. Several stakes are placed along different radial directions around the borehole prior
to fracturing. After fracturing, a leveling telescope can be used to measure the change in elevation of
preexisting marks on the stakes to determine the location and net uplift of the ground surface resulting
from the fracture. A laser system called the Ground Elevation Measurement System (GEMS) was
developed by UC to measure uplift in real time during hydraulic fracturing. The system uses a laser
and an array of sensors to track the displacement of each point in the array with time.
A typical pressure versus time plot obtained during hydraulic fracturing is presented in Figure 1-4.
The maximum pressure indicates the onset of fracturing, and the subsequent reduction of pressure
with time denotes the period of fracture propagation. The rapid pressure oscillations shown in Figure
1-4 result from the cycling of the piston pump, and are absent when a progressive cavity pump is
used to inject the slurry.
1.6 KEY CONTACTS
Additional information on the hydraulic fracturing technology and the SITE program can be
obtained from the following sources:
Hydraulic Fracturing Technology
Dr. Lawrence C. Murdoch
Director of Research
Department of Civil and Environmental Engineering
University of Cincinnati
5995 Center Hill Road
Cincinnati, OH 45224
Telephone No. (513) 569-7897
-------
VO
CO
Q.
UJ
tr
rs
CO
a:
a.
60T
H
j
50
301
20-j
_j
1
o-
10
- j,!1,'-i..-'! l>-»""':- -'s.i' -,i'
: if; ',..'; ./..»' , .
1 ''I'llJUiflll
14
16
from ond
14, the of a
-------
The SITE Program
Ms. Naomi Barkley
of Research and Development
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
26 West Martin Luther King Drive
Cincinnati, OH 45268
Telephone No. (513) 569-7854
1-10
-------
2.0 TECHNOLOGY APPLICATIONS ANALYSIS
This section describes SITE demonstration objectives and conclusions including UC Center Hill tests,
other demonstration results, factors influencing the effectiveness of the hydraulic fracturing
technology, personnel requirements, potential regulatory requirements, and appropriate waste and site
conditions. The vendor's claims regarding the applicability and performance of the hydraulic
fracturing technology are included in Appendix A. The technology's applicability is based on the
results of two pilot-scale demonstrations conducted under the SITE program. The SITE
demonstration results are presented in Appendix B and other applications of the technology are
presented in Appendix C.
2.1 SITE DEMONSTRATION OBJECTIVES AND CONCLUSIONS
The SITE demonstrations were conducted at the Xerox Oak Brook site where SVE was being used,
and the Dayton site where bioremediation was in progress. The technology developer also conducted
tests at an uncontaminated site at Center Hill to determine factors affecting air flow through hydraulic
fractures. The objectives of the two demonstrations and the Center Hill tests were as follows:
To assess the technology's ability to create sand-filled hydraulic fractures in silty clays
and study the factors that affect these fractures over a period of 1 year
To evaluate the technology's ability to significantly enhance SVE and contaminant
removal at the Xerox Oak Brook site
To determine the efficiency of hydraulic fracturing in delivering water containing
hydrogen peroxide and nutrients to the Dayton site, which is contaminated with
petroleum products
To develop information required to estimate the costs for the technology
2.1.1 Center Hill Tests
Fractures were created in three wells at the Center Hill facility site, and air flow through the three
fractured wells over a 1-year period was compared to the air flow through two unfractured
conventional wells. In one well, a fracture was created at 5 ft bgs. In the second well, fractures
2-1
-------
were created at 5 and 10 ft bgs, and in the third well, a fracture was created at 5 ft bgs and vented to
the surface.
A suction head of 120 inches of water was applied to the five wells, and the suction head was
measured at several dozen pneumatic piezometers near each of the wells. The air yield from each of
the wells was also measured. The impacts of rainfall on the suction head and air yield from the five
wells was monitored during the winter and summer of 1992.
Conclusions from the Center Hill tests are as follows:
The vapor yield from a fractured well was about an order of magnitude higher than
from an unfractured well.
The zone of pneumatic control of the fractured well was more than 10 times greater
than that of the unfractured well.
Rainfall decreased vapor yield and increased suction head of fractured wells.
Unfractured wells were not affected by rainfall.
The effect of a vented fractured well was not significantly different from that of an
unvented fractured well.
2.1.2 Xerox Oak Brook Site Tests
The Xerox Oak Brook site contains silty clays contaminated with ethylbenzene, DCA, TCE, PCA,
TCA, toluene, and xylene. Two out of four wells used for two-phase SVE were fractured at depths
of 6, 10, and 15 ft bgs. A suction head was applied to the four wells, an in-line separator removed
the water before the vapor flow rate was measured, and gas chromatograph analysis of the vapor was
conducted. Over a period of 1 year, the vapor flow rates, suction head, and contaminant removal
rates were measured for the fractured and unfractured wells.
Conclusions from the Xerox Oak Brook site tests are as follows:
Fractured wells yielded vapor flow rates 15 to 30 times greater than unfractured
wells.
2-2
-------
The vapor flow rate from fractured wells was adversely affected by precipitation.
The contaminant yields from the fractured well zones were 7 to 14 times greater than
from comparable zones in the unfractured wells.
2.1.3 Dayton Bioremediation Site Tests
Dayton site contamination included BTEX and petroleum hydrocarbons. One of two wells was
fractured at depths of 6, 8, 10, and 12 ft bgs. Water containing hydrogen peroxide and nutrients was
gravity fed into these wells intermittently for 6 months. Two rounds of soil sampling were conducted
at locations 5, 10, and 15 ft north of the fractured and unfractured well s after bioremediation was in
progress for 1 and 6 months.
Conclusions from the Dayton site tests are as follows:
Moisture content increased in the vicinity of the fractured well, especially in the
fractured zones. Only a minor change in moisture content was detected in the
unfractured well.
The flow of water was about 25 to 40 times greater in the fractured well than in the
unfractured well.
Benzene, ethylbenzene, and petroleum hydrocarbon removal was higher in the
fractured well than in the unfractured well.
2.2 OTHER DEMONSTRATIO N RESULTS
The research team from UC used hydraulic fracturing to enhance remediation activities at two other
sites. Results from the two sites are summarized in Appendix C A brief summary of the
effectiveness of the hydraulic fracturing technology at these two sites is presented below.
The first site is an inactive gasoline retail facility located in Addison, Illinois. The site is
contaminated with petroleum hydrocarbons to a depth of 10 ft bgs. The site is underlain by silty clay
till. Groundwater is present at depths ranging from 5 to 8 ft bgs.
2-3
-------
Three sand-filled hydraulic fractures were created at depths of 6.5, 9, and 11.75 ft bgs at two
locations, and SVE wells were installed with screens intersecting these fractured depths. The
performance of these wells was compared to the performance of two vapor extraction wells installed
in unfractured ground. A suction head of up to 10 inches of mercury was applied to the wells.
The soil around the fractured well had greater permeability throughout the formation than the soil
around the unfractured well; however, saturated soils prevented the flow of vapor in both fractured
and unfractured wells. Hence, hydraulic fracturing does not enhance vapor extraction in saturated
soils
The second site, which is located in Grand Ledge, Michigan, is contaminated with petroleum
hydrocarbons resulting from the leakage of gasoline and waste oil from underground storage tanks
(UST). The site is underlain by silty clay with occasional sand and silt seams. The boring logs
indicate that the soils are underconsolidated and softer than at sites where hydraulic fracturing has
been successfully completed.
Fracturing was attempted at uncontaminated areas of the site at depths varying from 18 to 30 ft bgs.
At one location, the fractures vented to the surface. At another location, the stress gradients in the
soil resulted in discontinuous, steeply dipping fractures. Hence, the fractures created in
underconsolidated clays where vertical stress is more than horizontal stress, are steeply dipping and
may be discontinuous. These fractures do not enhance SVE because the permeability increase from
such fractures is significantly less than from gently dipping or horizontal fractures.
2.3 EFFECTIVENESS OF THE HYDRAULIC FRACTURING TECHNOLOGY
Hydraulic fracturing is an innovative technology that increases the permeability of silty clay and rock
formations. The technology creates sand-filled fractures in the formation that are up to 1 inch thick
and 30 ft long. Thus, the technology permanently enhances the flow of vapor or liquid through the
formation.
Analytical results indicate that the hydraulic fracturing technology increases vapor flow by one order
of magnitude, from distances of up to 30 ft from the recovery well. The water flow rate in a
2-4
-------
fractured well was 25 to 40 times that in a conventional unfractured well. This increased flow rate
enhanced bioremediation of contaminated soil
2.4 FACTORS INFLUENCING EFFECTIVENESS
Several factors influence the effectiveness of the hydraulic fracturing technology. These factors
include (1) site characteristics, (2) rainfall infiltrating into the site, and (3) operating parameters,
Each of these factors is discussed below.
2.4.1 Site Characteristics
Hydraulic fractures can be created in both rock and relatively uniform silty clays that are
overconsolidated and have low permeability (less than 10"7 cm/sec). Saturated sandy lenses in a clay
layer may increase the water content of the soil and inhibit the flow of vapor during SVE. Care
should be taken when creating fractures in the vicinity of sensitive structures such as precision
manufacturing plants that may be damaged by deformations of the ground surface. Because the
ground uplift is generally less than 1.5 inches, fractures can be created in the vicinity of roads, most
buildings, and USTs.
Hydraulic fracturing is a permeability enhancement technique used in conjunction with other soil
remediation methods. Sandy soils are permeable to liquid and vapor flow. Therefore, silty clays that
have low in situ permeabilities are best suited for the use of hydraulic fracturing. The horizontal
stress should be greater than the vertical stress at areas where hydraulic fracturing is to be
implemented because this stress condition permits fractures to propagate in a horizontal orientation.
Fractures that remain horizontal can grow to significant lengths, thereby enhancing flow in the
subsurface.
Hydraulic fracturing is ineffective in normally consolidated clays. Demonstrations of hydraulic
fracturing in such clays created fractures that were steeply dipping and vented to the surface. The
presence of water decreases the efficiency of SVE; hence, the use of hydraulic fracturing to enhance
SVE should be limited to unsaturated clays with moisture contents ranging from 20 to 30 percent.
2-5
-------
2.4.2 Rainfall Infiltrating Into the Site
The amount of rainfall infiltrating into the site has a direct bearing on the effectiveness of SVE
systems. The permeability enhancement produced by hydraulic fracturing improves liquid withdrawal
from the subsurface, and until most of the pore water is recovered, relative vapor permeability will be
negligible. Hence, any additional water introduced into the subsurface by rainfall will adversely
affect vapor extraction from the site. Tests conducted for over 1 year at Center Hill and the Xerox
Oak Brook site demonstrate the inverse relationship between rainfall and the SVE rates.
The use of a membrane that will prevent infiltration of rainfall into the subsurface, but that will allow
inflow of air is recommended. SVE tests conducted at a site in Addison, Illinois, demonstrate that
saturated soils must be dewatered before significant vapor flow rates can be achieved. Preventing
rainfall from infiltrating into the site minimizes dewatering efforts and maximizes vapor recovery.
2.4.3 Operating Parameters
Several operating parameters affect hydraulic fracturing. The important parameters that are controlled
during hydraulic fracturing are injection rate and gel-to-sand ratio. A brief summary of the manner in
which these parameters affect hydraulic fracturing is presented below.
The injection rate affects the maximum and final pump pressures (see Figure 1-4). The maximum
pressure depends on initial slot length, in situ stress at the fracture location, and the water content of
the soil. For a fracture created at a depth of 15 ft bgs, the maximum pressure can vary from 55 to
70 psi, and the final pressure is about 35 psi.
The gel-to-sand ratio in the slurry is adjusted to propagate the fracture and to move the sand into the
fracture. The amount of the gel is reduced when a possibility exists of the fracture venting to the
surface. In cases where the fracture propagates horizontally, the sand content is increased during
pumping to increase the thickness and length of the fracture. The gel-to-sand ratio in the slurry is
adjusted from fracture to fracture, depending on depth and site-specific soil conditions. For a fracture
created at 15 ft bgs, about 150 gallons of gel and 14 cubic ft of sand are used.
2-6
-------
2.5 PERSONNEL REQUIREMENTS
Equipment requirements for hydraulic fracturing include a slurry mixing tank, a slurry pump, a high-
pressure water pump for creating the notch, a fracturing lance and well-head assembly, a pressure
transducer and display terminal, and miscellaneous tools. A surveyor's level or the GEMS equipment
is necessary for monitoring ground uplift. The major pieces of equipment, including the slurry
mixing tank and pumps, are usually mounted on a trailer for ease of transport.
Assuming that a borehole has been drilled and is available for fracturing, only a qualified technician
and two assistants are required to complete a set of hydraulic fractures from the borehole. The
technician should be able to (1) keep the pumps and other equipment operational, (2) monitor and
interpret the pressure versus time plot, (3) understand engineering properties of soil and well design,
and (4) troubleshoot operational problems related to pump pressure and slurry volume. The assistants
will monitor instrumentation and install pneumatic piezometers to measure the performance of
fractures. If ground deformation measurements are to be taken, an individual familiar with the use of
a surveyor's level and/or laser surveying equipment should be added to the crew.
Personnel working at a hazardous waste site should have an Occupational Safety and Health
Administration (OSHA) 40-hour health and safety training, and take an annual 8-hour refresher
course. Specific health and safety requirements will vary depending on the type of site
contamination. Therefore, a site-specific health and safety plan should be prepared.
2.6 POTENTIAL REGULATORY REQUIREMENTS
Hydraulic fracturing can be used to enhance remediation at hazardous waste sites using SVE, in situ
bioremediation, and pump-and-treat systems. The regulations that apply to a particular remediation
activity will depend on the type of remediation site (Superfund or RCRA) and the type of waste being
treated. Because hydraulic fracturing technology is an enhancement technique for other remediation
activities used at the site, regulatory requirements for hydraulic fracturing are not distinct from those
that apply to remediation being conducted at the site.
2-7
-------
Hydraulic fracturing entails the injection of material into the subsurface. The permit requirements for
using hydraulic fracturing at a site will depend on state and local regulations and may involve
describing the process of creating fractures and assuring regulators that the gel is biodegradable and
will break down after the sand is placed in the fracture. Local regulations relating to noise and hours
of operation may also have to be complied with.
2.7 APPROPRIATE WASTE AND SITE CONDITIONS
Hydraulic fracturing can be used to enhance the permeability of any site contaminated with organic
compounds. It has been demonstrated for sites contaminated up to 40 ft bgs. The suitability of the
hydraulic fracturing technology for a hazardous waste site depends on certain site-specific
characteristics. Any in situ treatment technology that can be applied to such a contaminated site can
be enhanced by a thorough assessment of the following site conditions:
Evaluating if the vertical stress is less than the horizontal stress, that is, if the soil is
overconsolidated
Evaluating if vapor extraction is the treatment technology applicable to the site and if
the contaminated soil is unsaturated to permit the flow of vapor through the fractured
soil
Determining if any sand or soft clay lenses in the contaminated horizon tend to
produce steeply dipping fractures that vent to the ground surface
2-8
-------
3.0 ECONOMIC ANALYSIS
The primary purpose of this economic analysis is to estimate costs of utilizing hydraulic fracturing to
enhance remediation in low permeability soils and rock. Site-specific factors affecting cost, the basis
of the economic analysis, cost categories, and costs per fracture are described below. Costs have
been divided into seven categories that are applicable to this technology. These categories include the
following:
Site preparation
Permitting and regulatory
Capital equipment
Labor
Supplies and consumables
Analytical and monitoring costs
Demobilization
Table 3-1 presents the estimated costs for creating four to six fractures in two boreholes located about
100 ft apart. The costs presented in this analysis are order of magnitude estimates, with costs ranging
from -30 to +50 percent.
The five cost categories out of the 12 typically associated with cleanup activities at Superfund and
RCRA-corrective action sites that are not applicable to the hydraulic fracturing technology include the
following:
startup costs
Utility costs
Effluent treatment and disposal
Residuals and waste shipping and handling
Equipment maintenance and modifications
3-1
-------
Table 3-1
ESTIMATED COSTS ASSOCIATED WITH HYDRAULIC FRACTURING
Estimated Daily Cost
Cost Category (1993 Dollars)
1. Site Preparation 1,000
2. Permitting and Regulatory a 5,000
3. Capital Equipment Rental b 1,000
4. startup 0
5. Labor 2,000
6. Supply and Consumables 1,000
7. Utilities 0
8. Effluent Treatment and Disposal 0
9. Residual and Waste Shipping and Handling 0
10. Analytical and Monitoring 700
11. Maintenance and Modifications 0
12. Demobilization" 400
Total One-Time Costs 5,400
Total Daily Costs 5,700
Estimated Cost per Fracture c $950 to $1,425
Notes:
One time costs
Capital equipment includes:
Equipment trailer Notching pump and accessories
. Slurry mixer and pump Pressure transducer and display
Mixing pumps, tanks, hose Uplift survey equipment
Fracturing lance, wellhead assembly Scale
Miscellaneous tools and hardware
Rental cost is based on 30 rentals per year, and depreciation of the $92,900 capital cost over
3 years.
Total daily costs (excluding one-time costs) divided by 4 or 6 fractures per day
3-2
-------
Hydraulic fracturing is an enhancement technology, not a treatment technology that reduces waste
toxicity. The equipment used for creating hydraulic fracturing at contaminated sites is mounted on a
mobile trailer and can be started up at minimal cost. Six to ten fractures can be created at a site in 2
to 3 days. Therefore, a site owner or operator will most probably rent the equipment and crew to
create the fractures and will incur minimal startup costs. Equipment maintenance and modification
costs would be incurred by the technology vendor and would be included in the rental fee.
Hydraulic fracturing uses diesel or gasoline engine powered pumps, and the cost of diesel fuel is
included in the supply and consumable costs. Hence, no utility cost is incurred. The technology does
not treat wastes; therefore, no cost is associated with effluent treatment and disposal and residuals and
waste shipping and handling.
3 1 SITESPECIFIC FACTORS AFFECTING COST
A number of factors affect the estimated costs of creating hydraulic fractures at a site. These factors
include (1) physical site conditions such as site accessibility and degree of soil consolidation; (2)
degree of soil saturation; and (3) geographical location, which affects availability of services and
supplies. The first two factors also affect the effectiveness of hydraulic fracturing.
The costs presented in this analysis are based on conditions found at the Xerox Oak Brook site. A
full-scale demonstration was not conducted for this technology. Because operating costs were not
independently monitored during the pilot-scale demonstrations at the Xerox Oak Brook and Dayton
sites, all costs presented in this section were provided by Xerox and UC Center Hill.
3 2 BASIS OF ECONOMIC ANALYSIS
The hydraulic fracturing technique can be used to enhance the treatment effectiveness of SVE,
bioremediation, and pump-and-treat systems. For the purpose of this economic analysis, a SVE site
is considered because this type of treatment system is commonly used for soil remediation and
because cost information for this method is available from the pilot demonstration conducted at the
Xerox Oak Brook site.
3-3
-------
The following assumptions were made for this economic analysis:
The site is located in the midwest.
Suitable access roads are available.
Boreholes have already been drilled.
The GEMS is available to monitor real time ground uplift.
Four to six fractures are created per day.
3.3 COST CATEGORIES
A discussion of the seven cost categories applicable to the hydraulic fracturing technology and the
elements associated with each category is provided below.
3.3.1 Site Preparation Costs
The costs associated with site preparation include system design (including design of fracture depths
and installation of ground uplift monitoring points), and mobilization of the hydraulic fracturing
equipment.
Sites that require clearing of vegetation and access roads will have significantly increased site
preparation costs. For this analysis, site preparation costs for a 7,500-square-foot site are estimated
to be approximately $1,000. Costs included are for mobilization of the equipment from Cincinnati,
Ohio, to Chicago, Illinois (350 miles) and for rental of a bobcat to move material.
3.3.2 Permitting and Regulatory Costs
These costs are dependent on the type of wastes being treated and the remediation method being used
at Superfund or RCRA corrective action sites. Superfund regulations require that the remedial action
be consistent with applicable or relevant and appropriate requirements (ARAR), including
environmental laws, regulations, and ordinances of federal, state, and local jurisdictions. In general,
ARARs must be determined on a site-specific basis.
3-4
-------
Because hydraulic fracturing is an enhancement technique that results in the injection of guar gum gel
and sand into the subsurface, the permits required are those needed for the remediation method used
at the site and state or local injection permit to demonstrate that the material introduced during
fracturing will not adversely impact soil or ground water.
Permitting and regulatory costs are estimated to be approximately $5,000 based on costs incurred at
the Xerox Oak Brook site.
3.3.3 Capital Equipment Costs
Capital equipment costs include the cost of the hydraulic fracturing equipment and the ground uplift
monitoring system. Based on a trailer mounted fracturing setup assembled by the UC, the capital cost
of the system is $80,100. If real-time uplift monitoring is desired, the GEMS laser surveying system
developed by UC could be acquired for an additional capital cost of $12,800. Because hydraulic
fracturing at a site can be completed in only a few days, it would not be cost-effective to purchase the
equipment and GEMS. Accordingly, this economic analysis assumes that the equipment would be
rented on a daily basis. It is further assumed that the equipment would be in use about 30 times per
year, and that the total capital cost would be recovered in about 3 years. Based on these assumptions,
the rental cost is about $1,000 per day.
334 Labor Costs
Labor costs include the cost of personnel to operate the hydraulic fracturing equipment and the
ground uplift monitoring system, and per diem expenses for the crew. Per diem expenses are
included because the fracturing crew travels to a site for a few days to finish fracturing and then
leaves. Four to five persons can operate the fracturing and monitoring equipment. One person will
operate the slurry mixer and pumps, and three to four persons will handle the fracturing lance and the
uplift monitoring system. Labor costs are estimated to be $2,000 per day, and include the costs
associated with the annual health and safety training.
3-5
-------
3.3,5 Costs
Supplies and consumables costs Include the cost of materials and fuel. Materials Include sand, giiar
gel, and enzyme, Diesel and gasoline for running the pumps is included in the supply cost. The cost
of supplies and consumables is to be $1,000 per day for creating 4 to 6 fractures,
3.3.6 Analytical Monitoring Costs
If subsurface suction are monitored during SVE, pneumatic can be
the wells, A two-person crew can. install about 10 to 15 piezometers per day at a cost of $700.
3.3.7 Demobilization Costs
Demobilization costs are estimated to be about. $400 to move the equipment from Chicago, Illinois, to
Cincinnati, Ohio, Decontamination water will be collected with other SVE-related wastes and
later. Therefore, decontamination costs are assumed to be negligible.
3.4 PER
on. the daily cost of $5,700 and the estimate of 4 to 6 fractures created per day in two
boreholes, the cost per fracture is estimated to vary from $950 to $1,425, The total one-time cost for
obtaining the permits and demobilization is estimated to be $5,400.
3-6
-------
4.0 REFERENCES
Hubbert, M.K., andD.G. Willis. 1957. Mechanics of Hydraulic Fracturing. Petroleum
Transactions, American Institute of Mining, Metallurgical, and Petroleum Engineers (AIME).
Volume 210, pp. 153-168.
University of Cincinnati (UC). 1991. Work Plan for Hydraulic Fracturing at the Xerox PR & S
facility, Oak Brook, Illinois. May 27.
Wolf, A., and L.C. Murdoch. 1992. The Effect of Sand-Filled Hydraulic Fractures on Subsurface
Air Flow: Summary of SVE Field Tests Conducted at the Center Hill Research Facility, UC
Center Hill Research Facility Unpublished Report.
4-1
-------
APPENDIX A
VENDOR'S CLAIMS REGARDING HYDRAULIC FRACTURING
(Three Sheets)
-------
APPENDIX A
VENDOR' S CLAIMS REGARDING HYDRAULIC FRACTURING
Applicability
Hydraulic fracturing is a method of creating layers of granular material in soil or rock. When filled
with sand, these layers increase fluid flow through soils of low permeability and enhance the
performance of in situ remedial technologies, such as soil vapor extraction, soil washing, bioventing,
bioremediation, and pump and treat. Sand-filled fractures can also act as pathways for the delivery of
steam to the subsurface to mobilize contaminants for recovery. Other granular materials may be
placed in the fractures to serve as reservoirs of remediating compounds. These include granular
nutrients and time-release oxygen compounds to enhance biodegradation of organic compounds.
Fractures can also be filled with conductive materials to induce electroosmosis, electromigration, and
electrophoresis. Fractures filled with electrically resistive materials can be used to generate heat in
the subsurface to increase microorganism populations and metabolic activity, or to facilitate
volatilization of organic compounds.
Waste Types
Hydraulic fracturing unto itself is not a method of remediation, but instead is a means of enhancing
the performance of existing in situ remedial technologies. Thus, it is applicable to contaminated soils
that are treatable by in situ methods. Wastes commonly treated in situ include petroleum
hydrocarbons, volatile organic compounds, and other organic contaminants.
Favorable Conditions for Hydraulic Fracturing
Hydraulic fracturing is particularly suited to sites underlain by soils where the lateral component of
stress exceeds the vertical stress applied by the weight of the overburden (these soils are termed
overconsolidated). Fractures created in overconsolidated soils tend to propagate in a horizontal to
subhorizontal plane, allowing the fractures to reach maximum dimension without intersecting the
ground surface. This geometry, in most cases, will be the most favorable for in situ technologies
that utilize vertical wells. Glacial drift of the Midwest and Northeast, swelling clays of the Gulf
coast, and similar soils are frequently overconsolidated and suitable for hydraulic fracturing.
A-l
-------
Fractures created in normally consolidated soils tend to propagate in a vertical direction. This
fracture geometry may beneficial when utilizing directional recovery wells.
Advantages of Hydraulic Fracturing
Hydraulic fracturing will increase fluid flow through the subsurface and will facilitate in situ
remediation of fine-grained soils. Advantages include:
Hydraulic fracturing facilitates use of in situ remediation in soils of low permeability,
typically less than Ifr7 centimeters per second. Without fractures, many of these soils
would not be considered candidates for in situ remediation.
Wells containing sand-filled fractures have been demonstrated to have a greater area
of influence than conventional wells. This reduces drilling and well completion costs
by increasing the spacing of the wells.
The increase in subsurface fluid flow associated with sand-filled hydraulic fractures
may decrease the time required for remediation.
Sand can be placed in hydraulic fractures to create highly permeable pathways for
delivery or recovery of remedial fluids. Filling the fractures with a sand proppant
allows fractures to remain open at depths and in formations where unpropped
fractures may close.
Hydraulic fractures can be filled with a variety of compounds to enhance remediation.
Hydraulic fractures filled with granular nutrients and time-release oxygen compounds
can act as subsurface reservoirs of materials needed for bioremediation. Electrically
conductive materials placed in the fractures offer the potential to induce
electroosmosis, electrophoresis, or electromigration of contaminants. Moreover,
hydraulic fractures may be used as resistive heaters to increase temperature and
volatilize contaminants or to increase bioactivity.
Hydraulic Fracturing Project Schedule
The following schedule is based on a hydraulic fracturing project located less than 500 miles away, a
field crew of five, and installation of 4 recovery wells containing 3 hydraulic fractures each. It
assumes fractures will be created between depths of 5 and 15 feet, will be 20 to 30 feet in diameter,
0.5 to 1 inch in thickness, and contain 600 to 1,400 pounds of sand.
A-2
-------
Description Days
1. Site assessment/fracture design 1
2. Mobilization 1
3. Hydraulic fracturing 3
4. Well completion/monitor installation/decontamination 2
5. Demobilization
Total 8
Cost Information
Hydraulic fracturing capital equipment, including the cost of the Ground Elevation Measurement
System (GEMS), is estimated to be $92,900. Based on renting this equipment about 30 times per
year, and a depreciation period of 3 years, the rental cost per day is about $1,000. Costs for site
preparation, labor, supplies and consumables (sand, guar gum gel, enzyme, and diesel fuel), and
pneumatic piezometer installation for monitoring the fracture performance are estimated to be $4,700
per day. Assuming that 4 to 6 fractures are created per day, the cost per fracture is estimated to be
$950 to $1,425.
A-3
-------
APPENDIX B
SITE DEMONSTRATION RESULTS
-------
APPENDIX B
SITE DEMONSTRATION RESULTS
The hydraulic fracturing technology was demonstrated at a Xerox Corporation (Xerox) SVE site in
Oak Brook, Illinois (the Xerox Oak Brook site), and at a bioremediation site near Dayton, Ohio (the
Dayton site). The Superfund Innovative Technology Evaluation (SITE) demonstration activities and
results are summarized in this appendix. More detailed information about the site demonstration
results is presented in the technology evaluation report (TER).
XEROX OAK BROOK SITE
At the Xerox Oak Brook site, contaminants consisting of trichloroethene (TCE); 1, 1 , 1-trichloroethane
(TCA); 1,1-dichloroethane (DCA); perchloroethane (PCA); ethylbenzene; toluene; and xylene are
present in silty clay till to depths of 20 feet (ft) below ground surface (bgs) Xerox investigated the
site in 1987. In 1991, a two-phase soil vapor extraction (SVE) system was installed. The layout of
the vapor extraction system is shown in Figures B-l and B-2. Vacuum is applied to the wells by a
pump, and the water and vapor in the soils surrounding the SVE wells is withdrawn. An inlet
separator removes the water and a discharge separator removes the moisture in the vapor. The vapor
is treated in a carbon adsorption unit, and clean air is vented out of the treatment building. The water
is passed through a carbon adsorption unit and discharged to the sewer system. The discharge water
is sampled to ensure that it meets the sewer permit requirements.
Hydraulic conductivity at the site varies from 10"7 to 10"8 centimeters per second (cm/sec) This low
permeability hampers the rate of vapor extraction. To enhance vapor extraction, fractures were
created at the site during the week of July 15, 1991. A work plan prepared by the University of
Cincinnati (UC) Center Hill Research Facility (Center Hill) describes the pilot-scale study (UC,
199la). The pilot-scale demonstration consisted of creating six hydraulic fractures at two locations.
Figure B-3 presents piezometers and extraction well locations. RW1 and RW2 are recovery wells in
unfractured ground, and RW3 and RW4 are recovery wells in fractured ground. Before fracturing,
soil samples were obtained in the vicinity of the four wells to a depth of 15 ft bgs. Soil moisture
content was measured every foot bgs, and two samples from each borehole were analyzed for volatile
organic compounds (VOC). The work was performed in accordance with the Quality Assurance Plan
prepared by Xerox' s subcontractor, Woodward-Clyde Consultants (Woodward-Clyde Consultants,
1991).
B-l
-------
RW2
fie
Q
XEROX PR It S FACILITY
TO VAPOR
EXTRACTION
SYSTEM
3-INCH-DIAMETER PIPING
2-INCH-DIAMETER PIPING (TYP.)
LOW SHRUBBERY TO
OBSCURE PIPING
WINDSOR
Source: Modified from Xerox, 1992
figure B1. Xerox Oak Brook Site Piping System Plan for Vapor Extraction
B-2
-------
VAPOR
VAPOR
TREATMENT
i
VAPOR
VAPOR
AND
LIQUID
VACUUM
PUMP
VAPOR
AND
LIQUID
DISCHARGE
SEPARATOR
LIQUID
LIQUID
TREATMENT
WELL
SYSTEM
LIQUID
Source: Modified from Xerox. 1992
Figure B-2. Schematic Diagram of Xerox Oak Brook Site Vapor Extraction System
B-3
-------
w
I
.b.
XEROX PR & S FACILITY
WINDSOR DRIVE
LEGEND
PNEUMATIC PIEZOMETER LOCATION
0 EXTRACTION WELL LOCATION AND NUMBER
Source: Modified from Xerox. 1992
Figure B-3. Extraction well and Piezometer Locations
NOT TO SCALE
B-4
-------
Fracturing Activities
Fractures were created at Wells No. RW3 and RW4 at depths of 6, 10, and 15 ft bgs. However, the
fracture at Well No. RW4 at 6 ft bgs vented to the surface. Ground surface uplift measurements of
up to 1.04 inches were measured at 11.5 and 16.4 ft from the fracturing hole. A week after the
fractures were created, recovery wells and monitoring boreholes were drilled. Multilevel recovery
wells consisting of separate screens and risers for each fracture were installed to make individual
access to each fracture possible. Multilevel monitoring boreholes containing as many as six
pneumatic piezometers were installed at 5, 10, 15, and 20 ft from each recovery well. Cased
boreholes designed to serve as neutron probe access holes were installed near each monitoring
borehole to measure soil moisture content.
The six fractures in contaminated ground were created on the same day, and each fracture required
1.5 to 2 hours to complete. Essential characteristics of the fractures are summarized in Table B-l.
The details include the depth bgs at the point where the fracture was created, the bulk volume of sand
pumped into the fracture, the volume of gel in the fracture, the maximum pressure at the point of
injection, the pressure at the end of pumping, the maximum uplift (typically not at the point of
injection), and the approximate radius of the uplifted area over the fracture. The radius of each
fracture is dependent on the amount of slurry pumped into the fracture.
Xerox monitored the following parameters for the two fractured wells (Wells No. RW3 and RW4)
and the two unfractured wells (Wells No. RW1 and RW2):
Water discharge from the system
Soil moisture content at depths of 4, 8, and 12 ft bgs and at lateral distances of 10,
15, and 20 ft north of the wells
Soil vacuum at recovery wells and monitoring points
Vapor flow rates from recovery wells
On-line gas chromatography (GC) analysis of 1,1-dichloroethane (DCA); 1,1,1-TCA;
TCE; toluene; ethylbenzene; perchloroethane (PCA); and xylenes
B-5
-------
Table B-l. Fracture Characteristics at Xerox Oak Brook Site
Fracture
Designation
OXP1F1
OXP1F2
OXP1F3
OXP2F1
OXP2F2
OXP2F3
Depth
(ft bgs)
6.0
10.0
15.0
6.0
10.0
15.0
Sand
(ft3)
NA*
12
13
6
12
14
Gel
(gallons)
20
130
150
100
140
150
Maximum
Pressure
(psi)
22
38
55
25
45
72
End
Pressure
(psi)
20
8
34
8
10
35
Maximum
Uplift
(inches)
0.12
0.8
0.96
1.04
0.75
1.2
Radius
(ft)
NA*
13.1
16.4
11.5
13.1
14.8
Comment
Vented to
surface
Recovery
Well No.
RW4
Recovery
Well No.
RW4
Recovery
Well No.
RW3
Recovery
Well No.
RW3
Recovery
Well No.
RW3
*Not Applicable
B-6
-------
Fracturing Results
Well discharge was measured using vortex shedding electronic flow meters from December
1991 until December 1992. These flow meters are very sensitive to the presence of water in the
vapor and hence, did not provide reliable data for the flow from each well. Also, the range setting in
these meters did not allow small readings (less than 10 cfm) to be accurately measured. Therefore,
from June 1992 until December 1992, variable area flow meters (rotometers) were used to measure
the flow from each riser connected to a fractured zone (or screened zone in RW2) in a well. A
demister pot was utilized to remove any liquid from the vapor stream before it entered the rotometer,
minimizing the effect of two-phase flow on the accuracy of the readings. A table summarizing the
rotameter discharge is given below.
Table B-2. Summary of Well Discharge Readings
Well ID
RW2
RW3
RW4*
RW4
Discharge
range (acfm)
0.1-4.6
2.2-22.0
27.9-42.7
17.1-29.7
Discharge avg
(acfm)
1.1
14.3
34.2
22.6
Discharge %
6 ft zone
46.3
61.2
36.0
not applicable
Discharge %
10 ft zone
27.3
8.4
41.0
not available
Discharge %
15 ft zone
23.2
30.4
23.0
not available
* The six-foot-deep fracture at RW4 vented to the surface. This data includes discharge when suction
is applied to all three of the fractures. The row below is well discharge when suction is applied to
the 10- and 15-foot-deep fractures only.
The amount of contaminant removed from each well was calculated by using the following equation:
where
Pounds/hour = Concentration x Flow x Molecular weight x 1.53xl010
Concentration is in parts per billion measured by the GC
Flow is in cfm
Molecular weight of compound is in grams per mole
(B-l)
B-7
-------
Vapor flow rates and contaminants removed from each well are presented in Figures B-4 and B-S,
respectively. Data from Well No. RW1 is not presented because it had a leak in its annulus between
the riser and the borehole wall, which allows air from the surface to flow into the well. The
discharge from the fractured Well Nos. RW3 and RW4 is 15 to 20 times greater than the discharge
from the unfractured Well No. RW2. The amount of contaminants removed from fractured Well
Nos. RW3 and RW4 is 7 to 14 times greater than from the unfractured Well No. RW2.
The vapor flow rates decreased during periods of precipitation in spring and early summer 1992,
primarily because of water occupying the pore spaces in the soil. Xerox will cover the surface area
in which the wells are screened with an impermeable membrane to prevent direct infiltration of
rainfall into the contaminated soils.
The soil vacuum readings obtained from pneumatic piezometers in the vicinity of the recovery wells
showed that the zone of pneumatic control around fractured Well Nos. RW3 and RW4 is about 25 ft
from the well compared to less than 1 ft from the unfractured Well No. RW2, demonstrating that
significantly fewer fractured wells are required to remediate a contaminated site using SVE.
Conclusions
Hydraulic fracturing is an effective method to enhance the permeability of silty clays and thereby
increase vapor flow rate by about one order of magnitude. The number of wells required to
remediate the site is reduced significantly, and the rate of contaminant removal is increased by 7 to 14
times.
DAYTON SITE
At the Dayton site, six underground storage tanks (UST) were removed in December 1989.
Three tanks contained gasoline, one tank contained No. 2 fuel oil, and two tanks contained kerosene.
Laboratory analyses of soil samples collected from the UST excavations indicate that benzene
B-8
-------
%$
o
I
ce
o
u,
a:
o
a.
\o
50
40
30
20 4
10
0
23,
50
100
150
TIME
Source: of
B-4. Flow fn end RW4
B-9
-------
w
I
>>
o
w
o
c
3
O
o.
UJ
a:
o
p
10
0
0 20
JUNE 23. 1992
60 80 100 120 140 160
TIME (days)
Source:Modified from University of Cincinnati, 1993
Figure B-5. Contaminants Removed from Wells RW2. RW3. and RW4
-------
concentrations ranged from not detected (ND) to 622 microgram per kilogram G*g/kg). Ethylbenzene
concentrations ranged from ND to 3,800 jug/kg; toluene concentrations from ND to 10,400 /ig/kg;
and xylene concentrations from ND to 41,900 /ig/kg. Total petroleum hydrocarbon (TPH)
compounds ranged in concentrations from 32 to 8,550 /ig/kg; and total lead concentrations from 21 to
150 /ig/kg.
A remedial action contractor investigated the extent of contamination at the site in 1990. The
investigation revealed the following site characteristics:
The site is underlain by stiff, sandy to silty clay with traces of gravel.
The bedrock is shallow, at depths ranging from 15.5 to 17.0 ft bgs, and consists of
claystone and limestone.
The horizontal extent of hydrocarbons is limited to the tank excavation area and the
area east of the former tanks.
The vertical extent of hydrocarbons appears limited to the upper 6.5 to 16.0 ft bgs in
soils.
Fracturing Activities
The remedial action contractor initiated bioremediation activities at the site in 1991. In July 1991, the
UC Center Hill proposed an investigation to determine the extent to which creating sand-filled
hydraulic fractures would enhance bioremediation of the site. A Quality Assurance Project Plan was
prepared by the UC (UC, 1991b). The delivery of water containing hydrogen peroxide and nutrients
to sustain microorganisms through fractured wells was compared to the delivery of similar water
through conventional unfractured wells.
Field tests were conducted from August 16 through 21, 1991. The tests consisted of creating seven
fractures at two locations in contaminated ground near Wells No. SAD2 and SAD3. SAD4 is a
conventional injection well in contaminated ground. Figure B-6 shows these well locations. Essential
characteristics of these fractures are summarized in Table B-3 and include the depth bgs at the point
where the fracture was created, the bulk volume of the sand, the volume of gel, the maximum
pressure at the point of injection, the pressure at the end of pumping, the maximum uplift (typically
B-ll
-------
Table B-3. Fracture Characteristics at Dayton Site
Fracture
Designation
SAD2-7
SAD2-8
SAD2-10
SAD2-12
SAD3-5
SAD3-7
SAD3-9
Depth
(ftbgs)
7
8
10
12
5
7
9
Sand
Volume
(ft3)
6
6
9
9
5
8
9
Gel
Volume
(gallons)
110
100
110
125
85
100
115
Maximum
Pressure
(psi)
42
17
37
42
19
43
39
End
Pressure
(psi)
7 to 11
7 to 15
10 to 20
18 to 26
3 to 7
7to 10
12 to 17
Maximum
Uplift
(in)
0.88
0.8
0.68
0.48
0.72
0.68
0.52
Radius
(ft)
15.1
14.8
16.4
23.0
14.8
15.4
23.0
Comment
Contaminated
soil fractures
No
contamination
detected
No
contamination
detected
No
contamination
detected
B-12
-------
not at the point of injection), and the approximate radius of the uplifted area over the fracture.
During the first week of September 1991, injection Wells No. SAD2 and SAD3 were installed. Soil
samples were obtained using a 2-inch split-spoon sampler and were analyzed for moisture content,
BTEX, and TPH. Samples were collected 5, 10, and 15 ft north of Wells No. SAD2 and SAD4 (the
unfractured well) and 10 ft south of Well No. SAD2. Well No. SAD3 was found to contain no
contamination. Clusters of piezometers were installed at 5, 10, and 15 ft north of Well No. SAD2
and 10 ft south of Well No. SAD2 at depths corresponding to the depths of individual fractures.
Piezometers were installed 5, 10, and 15 ft north of Well No. SAD4.
Water containing hydrogen peroxide and nutrients was introduced into Wells No. SAD2 and SAD4 in
December 1991. The unfractured well, Well No. SAD4, was filled with sand and the water was
gravity fed by a 0.5-inch-diameter pipe grouted into place for delivery at 5 ft. The water was gravity
fed into the fractured well at depths of 7, 8, 10, and 12 ft bgs. A system of capture trenches and a
return pump were installed (see Figure B-6).
Fracturing Results
The impact of hydraulic fracturing at the Dayton site was measured by monitoring the rate of flow of
water in the vicinity of the fractured well No. SAD2 and the unfractured well No. SAD4. Also, soil
samples were obtained at 6 and 8 ft bgs from the vicinity of the wells after 1 and 6 months of
bioremediation. These soil samples were analyzed for moisture content, number of colony forming
units (CPU), microbial metabolic activity, pH, TPHs, and benzene, toluene, ethylbenzene, and
xylene.
The flow rates in the fractured and unfractured wells are presented in Figure B-7. The flow rate in
the fractured well was 25 to 40 times higher than in the unfractured well. Contaminant removal
percentages in the vicinity of the fractured and unfractured wells are presented in Table B-4. This
table shows that benzene, ethylbenzene, and TPHs were significantly remediated in the vicinity of the
fractured well. The variability in removal percentages observed in the second and third rounds of
sampling (after 1 and 6 months of bioremediation) resulted from the system not being run
continuously or at optimal conditions.
B-13
-------
ROAD
«
o
INJECTION Wilt
FRACTURED
INJECTION
PVC PIPE
SOIL LOCATION
NOT TO SCALE
Source: Modified from University of Cincinnati, 1991b
Figure 88, General Dayton Site Plan
B-14
-------
03
©
^^
3
C
CL
CO
C
_o
"o
01
lil
d
o
u.
o:
UJ
0.15
0.05
L
20
30
40
+ NO,
I NO,
Time (days)
Soyrcg: 1992
8-7, of In No, and
-------
Table 1-4. Contaminants Removed at the Dayton Site
Treatment
Fractured Well
No.
Unfractured
Well No SMH
Fractured Wtil
No. SAH2
Unfractured
Well No SAD4
Fiactured Well
No,
I Jnfractured
Well No.
Fiactured Well
No.
Location
From Well
5 ft
5 ft north
10 it north
10 ft north
15 ft north
15 ft north
10 ft
Sampling
Round
Second
Third
Second
Third
Second
Third
Second
Third
Second
Third
Second
Third
Second
Third
Benzene, Etbylbenzene,
Toluene Removal
(Percent compared to first round)
Benzene
MI*
80
NI
NI
46,7
11.7
NI
NI
64,3
37.8
NI
NI
NI
N!
Ethyl-
benzene
97
60
7,9
37,0
78.5
NI
71.4
90.5
72,7
56.2
NI
NI
NI
NI
Toluene
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
TPH
Removed
(Percent
compared to
first round)
77%
71%
0%
55%
58%
54%
27%
67%
519?
681
IK
2"»%
5->«"o
80%
No impact
B-16
-------
Conclusion
The bioremediation activities were conducted by an independent contractor, and UC had no control
over the operating parameters. The erratic nature of the results obtained during the second and third
round of sampling indicate that the system was not run continuously or at optimal conditions.
However, the sampling results obtained during a 1-year period indicate that fractured wells result in a
significant increase in contaminant removal over unfractured wells.
Rates of fluid flow in the fractured well was 25 to 40 times higher than in the unfractured well. On
certain days, water flow rate in the unfractured well was minimal, but significant flow passed through
the soil around the fractured well (see Figure B-7). Fluid flow increased the moisture content around
the fractured well twofold and near the fracture almost fourfold.
REFERENCES
University of Cincinnati (UC). 199la. Work Plan for Hydraulic Fracturing at the Xerox
Corporation (Xerox) PR&S Facility. Oak Brook, Illinois. May 27.
UC. 1991b. Quality Assurance Project Plan, In Situ Biodegradation and Delivery of Oxygen and
Nutrients to Subsurface Microorganisms. Prepared August 15.
UC. 1993. SITE Field Demonstrations of Hydraulic Fracturing. UC Center Hill Unpublished
Report.
Vesper, S. 1992. Report on Bioremediation Studies at the Dayton Site. UC Center Hill Research
Facility (Center Hill) Unpublished Report.
Woodward-Clyde Consultants. 1991. Hydraulic Fracturing Pilot Study, Quality Assurance/Quality
Control Plan, Xerox PR&S Facility. Oak Brook, Illinois. June 1991.
Xerox Corporation (Xerox), 1992, Figures Provided by Mr. Elliott Duffney. Environmental
Engineering Department, Webster, New York.
B-17
-------
APPENDIX C
DESCRIPTION OF FRACTURING TESTS
-------
APPENDIX C
DESCRIPTION OF FRACTURING TESTS
Hydraulic fracturing was conducted at sites in Addison, Illinois, and Grand Ledge, Michigan, in 1991
and 1992, respectively. Important data on the applicability and effectiveness of hydraulic fracturing
was obtained during these tests, which are described below along with references used to prepare this
appendix.
ADDISON, ILLINOIS
The test site is an inactive gasoline retail facility located in Addison, Illinois. The gasoline
underground storage tanks (UST) and pump dispensers have been removed. About 3,900 cubic yards
of soil are contaminated with benzene, toluene, ethylbenzene, and xylene (BTEX). The subsurface
soil consists of silty clay till. Groundwater is present at depths ranging from 5 to 8 feet (ft) below
ground surface (bgs).
Fracturing Activities
Three sand-filled hydraulic fractures were created at depths of 6.5, 9, and 11.75 ft bgs at each of two
locations on site (see Figure C-l). After fracturing, soil vapor extraction (SVE) wells were installed
so that each well screen intersected one of the fractures. Four nested pressure monitoring probes
were installed within the fractured till so that soil pore pressure could be measured at various depths
near of the fractures. Two SVE wells were installed in unfractured till to provide data to compare
with data from the fractured wells. Seven nested pressure monitoring probes were located at lateral
distances of 5, 7.5, 9, 15, 20, 25, and 30 ft from the SVE wells.
Each extraction well was connected to a vacuum blower with a 2-inch-diameter hose. A suction head
of up to 10 inches mercury was applied to the wells. Vacuum gauges measured applied suction at the
wells, and air flow rotometers were used to measure SVE rates. A manifold at the top of each SVE
well allowed both soil vapors and accumulated ground water to be removed from the well. Water
collection rates were measured periodically using a 40-gallon air-water separator.
C-1
-------
USEMll
MAR-2
CONCRETE PAD
PUMP ISLAND
FRACTURE PRESSURE PROBE
FRACTURED SOIL VAPOR EXTRACTION WELL
CONVENTIONAL PRESSURE PROBE
CONVENTIONAL SOIL VAPOR EXTRACTION WELL
APPROXIMATE EXTENT OF SURFACE
UPLIFT DUE TO FRACTURING
EW-1A
EW-1B
EW-2
NOTES:
1. ALL TEST WELL AND HYDRAULIC FRACTURE
LOCATIONS ARE APPROXIMATE
2, FRACTURE PROPAGATION DEFINED BY SURFACE
UPLIFT OF APPROXIMATELY 3 TO 5 MILLIMETERS.
Sourcs: Modified from K*mp«r and Oth*r», 1992
Flgyr* C~1, Hydraulic Fracture and Test We!) Locotlona, Addlson, Illinois
C-2
-------
Pilot-scale tests were conducted for 4 to 14 days. The applied suction head and soil vapor flow rate
was monitored at 4- to 12-hour intervals at the extraction wells and blower unit. Suction heads were
also measured at each of the pressure probes.
Fracturing Results
The performance of the fractured and unfractured wells is closely related to the presence of ground
water and water used to create fractures and complete monitoring probes. In unfractured wells, the
well was pumped dry and suction was applied. The yield was 1 cfm, and the suction head was 8
inches of mercury. Over several hours, the yield decreased to negligible values and suction head
increased as water filled up the well. Vapor extraction from the fracture at the 9 ft bgs interval was
highest after dewatering the well and decreased to 0 to 0.3 cfm at the end of 10 hours.
Suction at the fractured well was greater in magnitude and extended to greater depths than at the
unfractured well. The radius of influence of the fracture created at a depth of 9 ft bgs was
approximately 20 feet. The fluctuations in suction head were erratic and may be related to the
movement of water, which was recovered throughout the test.
Conclusions
The SVE technique is not effective in moist to saturated silty clay because the pore water reduces air
permeability to negligible values. Fracturing the clay increases the water and air permeability;
however, vapor recovery is not significant until the moisture content of the soil has diminished to the
point that vapor flow can be predominant.
GRAND LEDGE, MICHIGAN
Waldo' s Auto Sales, a business located in the vicinity of Lansing, Michigan, is attempting to clean up
on-site hydrocarbon contamination. Removal of gasoline and waste oil from the site revealed
elevated concentrations of BTEX. The site is underlain by sandy clay with occasional sand and silt
seams. The boring logs indicate that soils at the site are underconsolidated and softer than at sites
where hydraulic fracturing has been successfully completed.
C-3
-------
Fracturing Activities
A total of five fractures were created at two separate boreholes in the ^contaminated section of the
site (see Figure C-2). The soft nature of the subsurface soil required the use of a 2.25inch outside
diameter (OD) lance instead of the 1.82-inch OD lance for greater durability and seal surface area.
Guar gum gel was also used instead of water to cut the notch because of the soft nature of the soil.
Fractures were created at depths of 22 and 27 ft bgs in the first borehole and at depths of 18, 21, and
30 ft bgs in the second borehole.
Fracturing Results
The pressure log (see Figure C-3) reveals that the fractures were discontinuous because of stress
gradients in the soil that pinched off propagating fractures. After the fracture pinches off and the
effective length of the fracture decreases, an increase in the injection pressure is required to propagate
the fracture. This increase results in the pressure versus time plot having several peaks and valleys.
Conclusions
Conducting hydraulic fracturing in clays that are underconsolidated (that is, where vertical stress is
greater than horizontal stress) results in steeply dipping fractures that are not as effective in improving
vapor phase permeability as gently dipping or horizontal fractures.
The hydraulic fracturing method is not effective in appreciably enhancing the permeability of
underconsolidated soft clays. The in situ stresses in the clay result in steeply-dipping fractures that
may be discontinuous and that will be less effective than fractures that are gently dipping and
continuous.
REFERENCES
Kemper, M, Others. 1992. Hydraulic Fracturing to Facilitate Remediation, Field Demonstrations,
1992. Interim Report Submitted to the U.S. Environmental Protection Agency (EPA),
Contract No. 68-C9-0031 Work Assignment No. 4.
C-4
-------
SAGINAW (M-4« HIGHWAY
LEGEND
O Hydraulically fractured recovery well
O Conventional recovery well
A Monitoring borehole
NOT TO SCALE
Source: Modified from Kemper and Others, 1992
C-2. Location of Proposed Wells and Monitoring Probes, Grand Ledge. Michigan
C-5
-------
3SV3UOMI NOUVU1NI3NO3 QN₯S
3
C
LU
2
p
o
en
o
o>
o
I
o
o
2
"S.
o
%-
o
£
o
V
N b
o> f~^
I I
o
-2 3
fe
Q. >
I I
r- W
t u
§
3
I
o
C-6
en
(I
* U.S. GOVERNMENT PRINTING OFFICER 993-750-002/80284
------- |