?/EPA
                            United States
                            Environmental Protection
                            Agency
EPA/540/SR-93/505
October 1993
                            SUPERFUND  INNOVATIVE
                            TECHNOLOGY EVALUATION
                            Technology Demonstration
                            Summary

                            Hydraulic Fracturing  Technology
                              Two pilot-scale demonstrations of the
                            hydraulic fracturing technology for en-
                            hancing the permeability of contami-
                            nated  silty clays have been evaluated
                            under  the Superfund Innovative Tech-
                            nology Evaluation (SITE) Program. The
                            technology was jointly developed by
                            the University of Cincinnati  (UC) and
                            the U. S. Environmental Protection
                            Agency's (EPA's) Risk Reduction Engi-
                            neering Laboratory (RREL). Tests were
                            also conducted at UC Center Hill Solid
                            and Hazardous Waste Research Facil-
                            ity (Center Hill Facility) by UC. These
                            tests were conducted to determine the
                            factors affecting soil vapor flow through
                            sand-filled hydraulic fractures.
                              The hydraulic fracturing technology
                            was demonstrated in 1991 and 1992 at
                            a Xerox Corporation soil vapor extrac-
                            tion (SVE) site in Oak Brook, IL (the
                            Xerox Oak Brook site), and at a biore-
                            mediation site near  Dayton, OH (the
                            Dayton site). The sites were contami-
                            nated  with volatile organic compounds,
                            and the in-situ permeability of onsite
                            soils  was less than 10'7 centimeters
                            per second. The technology created
                            multiple, sand-filled hydraulic fractures
                            at depths of up to 15 feet below ground
surface (ft bgs). At the Xerox Oak Brook
site, these fractures increased vapor
flow by one order of magnitude in an
area 30 times greater than the area
affected by wells in unfractured soil. At
the Dayton site,  the groundwater flow
rate was 25 to 40 times higher in a well
screened in the  fractured soil than in
one screened in unfractured soil.
  This  Summary was developed by
EPA's Risk Reduction Engineering
Laboratory, Cincinnati, OH, to announce
key findings of the SITE demonstration
that is fully documented in two reports
under one cover (see ordering informa-
tion at back).

Introduction
  In response to  the Superfund Amend-
ments  and Reauthorization Act of  1986,
EPA established a formal program to ac-
celerate the development, demonstration,
and use of new or innovative technologies
that offer permanent, long-term cleanup
solutions at Superfund sites. This SITE
program is administered by EPA's Office
of Research and Development, RREL. One
component of the SITE program is the
demonstration program, which develops
reliable performance and cost information

          ^9  Printed on Recycled Paper

-------
on innovative technologies so that poten-
tial users can assess a technology's suit-
ability for specific site cleanups.  During
the demonstrations, hydraulic fractures
were created to increase  the tn-situ per-
meability of contaminated soil. This per-
meability enhancement significantly  im-
proved the rates of vapor and water flow
through the soil near the fractured wells.

Technology Description
  The hydraulic fracturing technology is
designed to create sand-filled fractures up
to 1-in. thick and about 20 ft in radius in
tow permeability  soils. These  fractures,
when created at several depths from 5 to
40 ft bgs,  increase the  permeability of
contaminated soil. This increased perme-
ability promotes the flow  of vapors and
liquid through the  soil and enhances  the
effectiveness of SVE, bioremediation, and
pump-and-treat remediation techniques.
  The hydraulic fracturing  equipment con-
sists of a fracturing lance,  a tool to create
the initial notch, a continuous slurry mixer,
and  a  positive  displacement  pump
mounted on a trailer. A typical sequence
of operations for creating hydraulic frac-
tures is shown in Figure 1, and the slurry
             mixer and trailer-mounted pump assembly
             is shown in Figure 2.
                A borehole is drilled using 6  or  8-in.
             outside diameter, hollow-stem augers. In-
             dividual 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 then 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).
             Steel tubing with a narrow orifice at one
             end (the notching tool) 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, is
             rotated within the borehole and produces
             a disc-shaped notch extending 4 to  6 in.
             from the borehole (see Figure 1).
                Sand  slurry is then pumped  into the
             notch to create a hydraulic fracture. Sand
             slurry  is produced by mixing one part of
             granular solid (coarse sand proppant) with
             two parts of viscous fluid in a continuous
             mixer.  The viscous fluid consists of a mix-
             ture of guar gum gel, water, and an en-
             zyme that breaks down the gel after the
                         proppant has been deposited into the frac-
                         ture. A hydraulic fracture is created  by
                         pumping a predetermined volume of slurry
                         at rates of  10 to 25  gals/min into the
                         notch.  Lateral  pressure 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 radially up to about 20 ft
                         from the borehole wall.
                           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 pos-
                         sibility exists of the fracture venting to the
                         surface. In cases where the fracture propa-
                         gates horizontally, the sand content  is in-
                         creased  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
                         gal of gel and 14 cu ft of sand are used.
                           The direction and distance of propaga-
                         tion of the fracture is measured by moni-
                         toring the uplift of  the ground surface.
                         Several stakes are  placed  along radial
                       1
             Casingr—
Cap
                         •>"Rod
                           'Lance tip
  Extension
rf"'  rod
                                      /-Steel tubing.
                                    ,/- Cutting jet.
                                                                                                          Removal
                                                                                                          of lance
                                                                                              Sand slurry
                                                                              ;     Pressure from soil
                                                                              	  seals casing
                                                 .:,- Notch
Flgura 1. Sequence of operations for creating hydraulic fractures.

-------
 Figure 2.  Slurry mixing and pumping equipment mounted on trailers.
 directions around the borehole before frac-
 turing. 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
 was developed by DC  (patent pending) to
 measure real-time uplift data during  hy-
 draulic fracturing. The system uses a la-
 ser and  an array of sensors to track  the
 displacement of  each  point  in the array
 with time.

 Factors Affecting Technology
 Effectiveness
   Several factors affect the  performance
 of the hydraulic fracturing  technology.
 These include various site characteristics
 and the  communication between horizon-
 tal fractures. These factors, as applied to
 the SITE demonstrations,  are discussed
 below.

 Site Characteristics
   Hydraulic  fracturing is  a permeability-
 enhancement technique used in conjunc-
 tion with other soil remediation methods.
 Therefore, overconsolidated silty clays that
 have  low in-situ permeabilities  are best
 suited for the use of hydraulic fracturing.
 In overconsolidated clays, the horizontal
, stress is greater than  the vertical stress,
 and this stress condition permits fractures
 to propagate in a horizontal direction. Frac-
 tures that remain horizontal  can  grow to
significant lengths, thereby enhancing flow
in the subsurface.
  Hydraulic fracturing is ineffective in nor-
mally consolidated clays  where the hori-
zontal stress is less than the vertical stress.
Demonstrations of hydraulic fracturing in
such  clays  created fractures  that were
steeply dipping and vented to the surface.
Also,  the presence of water  decreases
the efficiency of SVE; hence, the use of
hydraulic fracturing to enhance SVE should
be limited to unsaturated clays.

Communication Between
Horizontal Fractures
  Horizontal fractures increase the per-
meability of soil in their immediate vicinity.
However, the permeability of soil between
the fractures is not affected unless verti-
cal or inclined fractures are created  be-
tween the horizontal fractures. Further work
at UC is being planned to  study this. In
SVE applications, changes in soil vacuum
(suction  head) applied to horizontal frac-
tures may induce communication between
the fractures.

Overview of SITE
Demonstrations
  The SITE demonstrations had the fol-
lowing 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 unfractured
    soils,
  • to compare the water  flow rates and
    moisture content  in  fractured and
    unfractured soils, and
  • to develop information  required to es-
    timate the operating costs of the tech-
    nology.
  The  Xerox Oak Brook  site  contained
silty clays contaminated with ethylbenzene;
1,1 -dichloroethane;  trichloroethene;
tetrachloroethene; 1,1,1 -trichloroethane;
toluene; and xylene. Two out of four wells
used for two-phase SVE were fractured at
depths of 6,  10,  and 15 ft bgs. Over a
period  of 1 yr, the soil vapor flow rates,
suction  head, and contaminant removal
rates were measured and  compared for
the fractured and unfractured wells. The
soil vapor flow was  measured with vari-
able area flow meters; the suction head in
the vicinity of the wells  was  measured
using pneumatic piezometers; and the con-
taminant concentration in  the  soil vapor
was measured with a gas chromatograph.
  Dayton site contamination included ben-
zene, 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 grav-
ity fed into these wells intermittently for
about 6  mos. The site operator was re-
sponsible for this activity,  and UC was
responsible for monitoring the progress of
bioremediation near the  fractured and
unfractured  wells. Two rounds of sam-
pling 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 mos. Soil samples
were obtained and analyzed for moisture
content, microbial metabolic activity, num-
ber of colony forming units,  BTEX, arid
petroleum hydrocarbons.
  The Center Hill Facility tests were con-
ducted in uncontaminated ground during
the winter and summer of 1992.  Perfor-
mance  of three  fractured  wells  (CHF1,
CHF2, and  CHF3) was  compared with
that of two conventional vapor extraction
wells (CHC1  and CHC2). Well No. CHF1
intersects hydraulic fractures created at
depths of 5 and  10 ft bgs, and Well Nos.
CHF2  and CHF3 intersect fractures at a
depth of 5 ft bgs. At Well  No.  CHF2, the
fracture  intersected  the ground surface
(vented). Air flow from these  wells was
measured with variable area flow meters,
suction head  near the wells  was mea-

-------
sured with pneumatic piezometers,  and
tha intensity of rainfall was measured with
a rain gauge.

Demonstration Results
  The pifot-scale demonstration at the
Dayton site and tests at the Center Hill
Facility were conducted under the Quality
Assurance Project Plan (QAPP), Project
Category IV, developed by UC Center Hill.
Testing at the Xerox Oak Brook site was
conducted  by Woodward-Clyde Consult-
ants (WCC) under a QAPP developed by
WCC. Center Hill Facility test results and
Xerox Oak Brook and  Dayton site demon-
stration results are discussed below.

Center Hill Facility Tests
  Parameters measured  during the Cen-
ter  Hill  Facility tests  included  air yield,
suction head, and rainfall intensity. Air yield
was obtained as a function of time and
rainfall intensity for three fractured wells
and was compared with air yield data from
two unfractured wells. Suction head was
measured  near  the  fractured  and
unfractured wells. The average and maxi-
mum yields (in cubic feet per minute [cfm])
and average zone of pneumatic control
for the five wells are summarized in Table
1.
  The results demonstrate that  air yield
from fractured wells is one order of mag-
nitude higher than that  from  unfractured
wells. The zone of pneumatic control of
the fractured well was  15 to 30 times
greater  than that  for unfractured wells.
Rainfall intensity affected the performance
of fractured wells, decreasing the yield
and increasing the suction head.

Xerox Oak Brook Site
Demonstration Results
  Vapor discharge  from, and  suction
heads in, two fractured wells  (Nos. RW3
and RW4) and one unfractured well (No.
RW2) were obtained over a period of 6
mos.  The vapor discharge readings  (in
actual cfm [acfm]) from these wells are
summarized in Table 2.
  The vapor discharge from fractured wells
was about 15  to 30 times greater than
that from the unfractured well. The water
recovery rate from the wells fluctuated
widely,  ranging from 20 to  as much  as
500 gallons per day. Water recovery rates
and vapor discharge rates were inversely
related. The cumulative mass of contami-
nants recovered from fractured wells was
consistently one  order of  magnitude
greater than that for the unfractured well.
The suction  head near the fractured well
extended radially about 25 ft, compared
with  less than  1 ft from the unfractured
well. These  data correlated well with the
radius of fractures  in Well Nos. RW3 and
RW4.

Dayton Site Demonstration
Results
  Water flow rates and  soil  sampling  re-
sults  near a fractured  well  (SAD2) with
fractures at  7, 8, 10, and 12 ft bgs were
compared with those from an unfractured
well  (SAD4). The flow rate of water en-
riched with hydrogen peroxide and nutri-
ents was about 25 to 40 times greater in
the fractured well than  in the unfractured
Table 1. Performance of Wells at the Center Hill Facility
Wall 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 25
15 to 20
Less than 1
Less than 1
'Zone in which tha pressure distribution can be controlled by varying the applied suction head.
Teblo 2. Summary of Well Discharge Readings at Xerox Oak Brook Site


Week No.
RW2
RW3
RW4f
RW4*


Range of Rates
0.1-4.6
2.2-22.0
27.9-42.7
17.1-29.7
Average
Discharge
(acfm)
1.1
14.3
34.2
22.6
Discharge
Percentage in
6 ft bgs Zone
46.3
61.2
36.0
not applicable
Discharge
Percentage in
10 ft bgs Zone
27.3
8.4
41.0
not available
Discharge
Percentage in
15 ft bgs Zone
23.2
30.4
23.0
not available
'Thae-tt-daepfractureat WeHNo. RW4 ventedto the surface. Data for this well include the discharge average
 when suction was applied to all three of the fractures.
f Wet discharge average when suction was applied to the 10- and 15-ft-deep fractures only. Hence, well
 discharge was smaller than when suction was applied to all three of the fractures.

-------
 well  (see  Figure 3).  This increased flow
 resulted in doubling the moisture content
 near the fractured well. The  demonstra-
 tion did not provide  reliable  quantitative
 data on contaminant  removal  rates at the
 site.

 Costs
   Economic data obtained from the SITE
 demonstrations were analyzed to estimate
 the cost of using this hydraulic fracturing
 technology at a  hazardous  waste site
 remediation. Twelve cost categories were
 examined, including  capital costs, labor
 costs,  and  supplies and consumables
 costs.
   The  capital equipment  costs include
 costs for an  equipment trailer on  which
 the slurry mixer, pumps, tanks, and hoses
 are mounted; a fracturing lance with well-
 head assembly; pressure transducer and
 display; and uplift survey instruments. The
 capital  equipment cost was $92,900. A
 capital  equipment daily rental cost,  as-
 suming 30 rentals/yr and a depreciation
 life of 3 yrs, was $1,000. A crew of four to
five can operate the fracturing and moni-
toring equipment. Labor costs were  esti-
mated to  be about $2,000/day. Supplies
and consumables include sand, guargum
gel,  enzyme, and diesel or gasoline for
operating  the  pumps. The cost  for  sup-
plies and consumables was about $1,0007
day. The daily cost for creating about four
to six fractures was about $5,700  and
included site preparation ($1,000) and cost
of installing pneumatic piezometers ($700)
near the fractures.  Hence, the cost per
fracture can vary from about $950 to
$1,425.

Conclusions
  The tests conducted at the Center Hill
Facility and the pilot-scale demonstrations
completed at the Xerox Oak Brook  and
Dayton sites lead to the conclusions  pre-
sented below.
  1. Sand-filled hydraulic fractures, up to
     1 in. thick and 25 ft in radius, can be
    created in overconsolidated, low per-
    meability silty clays.
   2. The fractures remained effective for
     a period of more than 1 yr. Rainfall
     decreased vapor yield and increased
     suction head of fractured wells.
     Unfractured wells were not affected
     by rainfall.
   3. Fractured wells yielded vapor flow
     rates 15 to 30 times greater than did
     unfractured wells. This increased flow
     was obtained from a distance of 25 ft
     away from the fractured well.
   4. The water flow was about 25 to 40
     times greater in a fractured well than
     in an unfractured well.
   5. The cost of creating a fracture can
     vary from about $950 to $1,425.
   The technology demonstrations  have
established the increased permeability in
silty clays  resulting  from  hydraulic frac-
tures and the radius  of influence of these
fractures. Further enhancements of in-situ
permeability can be achieved if communi-
cation can be created between horizontal
fractures.
            0.4
            0.3
      .65
      I   0.2
      CC


      iŁ
      I
           0.1
                                              20
                                   *~r-
                                    40
                                                                                                             60
                                                       Time (days)

                               ~^- Well No. SAD2 (fractured)       —•— Well No. SAD4 (unfractured)


Figure 3. Flow volumes of injected water in wells no. SAD2 and SAD4.
                                                                     &V.S. GOVERNMENT PRINTING OFFICE: 1993 - 7«MW«0079

-------


-------

-------
 The EPA Project Manager, Naomi Berkley (see below), is with the Risk
    Reduction Engineering Laboratory, Cincinnati, OH 45268.
 The complete report, entitled "Hydraulic Fracturing Technology—Technology
    Evaluation and Applications Analysis Reports," (Order No. PB94 -100 161'/
    AS^ Cost: $27.00, subject to change), will be available from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield,  VA 22161
        Telephone: 703-487-4650
 The EPA Project Manager can be contacted at:
        Risk Reduction Engineering Laboratory
        U.S. Environmental Protection Agency
        Cincinnati, OH 45268
United States
Environmental Protection Agency
Center for Environmental Research Information
Cincinnati, OH 45268
     BULK RATE
POSTAGES FEES PAID
         EPA
   PERMIT No. G-35
Official Business
PenaKy for Private Use
$300
EPA/540/SR-93/505

-------