EPA/600/A-97/030
Submitted for the Proceedings of the '97 Battelle Conference

   SITE CHARACTERIZATION METHODS FOR THE DESIGN OF
         IN-SITU ELECTRON DONOR DELIVERY SYSTEMS

Steven D. Acree (US-EPA, NRMRL-Ada, OK), Mike Hightower (Sandia National
Laboratory, Albuquerque, NM), Randall R. Ross (US-EPA, NRMRL-Ada, OK),
Guy  W.  Sewell  (US-EPA, NRMRL-Ada, OK), Brent Weesner (Lockheed
Martin Specialty Components, Largo, FL)

ABSTRACT: The Department of Energy and the U.S. Environmental Protection
Agency have been involved in designing and evaluating a pilot field demonstration
of reductive anaerobic biological in-situ treatment technologies (RABITT) for use as
a standard remedial technology for chloroethene contamination. Innovative site
characterization techniques have been utilized to identify the hydraulics of the site
and in particular the vertical distribution of relative hydraulic conductivities. Direct
extraction  of intact frozen cores has been utilized to determine the  vertical
distribution of contaminants in the pore spaces and on the solid matrix of site
material.   The  combination  of these techniques  along  with standard  site
characterization methods has been used to develop a three-dimensional picture of the
site with vertical resolutions down to 0.5 ft (15 cm). This information has then been
used to evaluate different scenarios for nutrient/electron donor delivery at the site,
and when used with appropriate  transport and flow codes was used to  exclude
designs which did not allow for significant mixing of donor and contaminants, or
which did not efficiently deliver nutrients/donors to all contaminated zones.  It is felt
that the use of site characterization data in this manner is critical to the effective and
appropriate design and implementation of RABITT and other in-situ treatment
technologies.

INTRODUCTION
       Reductive anaerobic biological in-situ treatment technologies for the
remediation of ground water contaminated with chloroethenes is a  promising
approach to an all too common  environmental problem.  The design of these
treatment systems is a complex environmental engineering challenge requiring a
clear understanding of the contaminant distribution, the hydrogeologic setting and
the geochemistry. To an even greater extent than with pump-and-treat systems or
aerobic-catabolic-bioremediation, the application of RABITT requires that this site
conceptual  model be a detailed three-dimensional representation,  incorporating
flow/time dynamics to ensure interaction of electron donor, contaminants and active
microorganisms under appropriate conditions.
       The Department of Energy, the U.S. Environmental Protection Agency, The
State of Florida, and Industry Partners have been involved in the design  and
implementation of a pilot-scale field demonstration of RABITT to support the
DOE's Innovative Treatment Remediation Demonstration (ITRD) Program.  The
study involves creation of an in-situ circulation cell and the injection of electron
donor/nutrients to stimulate biological transformation processes. Constraints on
system design included a potentially short half-life of the injected solutions and the
need for controlled mixing of donor and contaminants.  Travel time through the
contaminated media was required to be no longer than approximately 100 days.
Design of an effective and efficient system under such constraints requires three-
dimensional characterization of contaminant  distribution,  hydrology,  and
geochemistry.
       Detailed, three-dimensional site characterization is seldom performed due, in
part, to a lack of appreciation of the potential effects of heterogeneity on remedial

-------
design and effectiveness. Estimates of bulk or "average" parameters obtained from
traditional monitoring wells and aquifer tests generally have been used in design.
Techniques for obtaining detailed hydraulic information,  such as extensive
laboratory permeameter testing and multi-level slug tests (Molz and others, 1990),
have been available but are  often costly,  difficult to  apply,  and may not be
representative at the field scale. Sensitive borehole flowmeters suitable for detailed
characterization have  only recently become commercially available.  Despite
availability, there still exists a general lack of recognition regarding uses  for these
tools and the  value  of detailed  data in defining contaminant transport  and fate
processes/rates and in remedial  design. The following case study illustrates the
potential value of detailed hydraulic parameter and contaminant distribution data in
cost-effective designs.

BACKGROUND
       The site of the pilot study is located in central Florida and was used in the
1960's for disposal of drums of waste  and construction debris.  Subsurface
contamination, as indicated by contaminant concentrations in ground-water samples
from monitoring wells,  is heterogeneously distributed in the  shallow  aquifer.
Geology of the upper 30 ft (9.1 m) of the saturated zone is predominantly fine
sands with varying fractions of silt and clay.  Fill material, construction debris, and
lagoon sediments are present in the upper few feet of the subsurface.  The top of the
Hawthorn Formation is encountered at a depth of about 30 ft (9.1 m) in  this area
and consists of clay and limestone.
       The water table at the site is located at depths of less than  10 ft (3 m) below
land surface and varies  seasonally.  Ground-water  flow at the site is  strongly
influenced by a ground-water extraction system that is currently in operation.  Bulk
hydraulic conductivity of aquifer materials was estimated using data from a 72-hour
multi-well pumping test. Estimates of horizontal hydraulic conductivity clustered in
a relatively narrow range from approximately 1 ft/d to 3 ft/d (0.3-0.9 m/d).
Estimates of vertical hydraulic conductivity  were less certain.  The most reliable data
set indicated that a horizontal to vertical  anisotropy ratio of about 10:1 may be
representative of bulk conditions in the shallow saturated zone of interest. Based on
the potential for significant heterogeneity in hydraulic conductivity and contaminant
distribution, detailed characterization of site conditions was undertaken,

MATERIALS AND METHODS

Borehole  Flowmeter   Description  and  Methodology.  A sensitive
electromagnetic borehole flowmeter was used to define the relative hydraulic
conductivity distribution of aquifer materials screened by a test well.  The study
consisted of measuring the vertical component of ground-water flow at several
depths in the well under undisturbed (ambient) conditions and during constant-rate
ground-water extraction. Measurements made during constant-rate  ground-water
extraction  or injection indicate  the distribution of flow to the well  and allow
interpretation of the relative hydraulic conductivity distribution of materials within
the screened interval (Molz and others, 1994).
       Test well NEBIOTW-1  was  installed in the vicinity  of  the proposed
bioremediation pilot study site.  The borehole was drilled using a wash rotary
technique whereby a casing is driven in advance of the rotary bit and materials are
washed from the hole. The objective of this technique was to minimize formation
damage during  drilling  so  that more representative data regarding hydraulic
properties could be obtained. The well was installed using an approximately 1-inch
(0.4  cm) thick artificial sand pack and screened from approximately 5 ft (1.5 m)
below the  water table (i.e., 9 ft or 0.7 m below land surface)  to the top of the

-------
Hawthorn Formation at approximately 30 ft (9. 1 m) below land surface.  Casing
and screen were standard 2-inch Schedule 40 PVC materials. The investigation was
performed using the following general protocol:

       Ambient vertical flowrates (undisturbed conditions) were measured from
       total depth to the top of the well screen at 1 ft intervals using the 0.5-inch
       (0.2 cm) ID probe.

       A peristaltic pump was used to stress the aquifer and establish a stable flow
       field under pumping conditions.  Total flowrates were measured using
       graduated cylinders and a stop watch at routine intervals. Tests were
       performed at two different extraction rates (i.e., approximately 2.7 1/min and
       4.8 1/min).

       After conditions in the well stabilized, the flowmeter was used with the 1 .fl-
       inch  (1 .54 cm) ED probe to measure vertical flowrates at each of the
       elevations occupied during the ambient  flow profile.  Measurements were
       also repeated at different times following the start of extraction to ensure a
       stable flow distribution was maintained.

       Measurements of flowrates under ambient and constant-rate  pumping
conditions were analyzed using methods described by Molz and others (1994).
Flow to the well from each interval is assumed to be horizontal and proportional to
the transmissivity of the formation after an initial stabilization period. The relative
hydraulic conductivity profile was estimated using Equation 1 developed by  Molz
and others (1990), which relates the dimensionless ratio Kj/K to the net induced
flow from each interval, interval thickness, total flow from the well, and aquifer
thickness influenced during the test.

               K
              K ~      Qp/b       '   -''--

where:
   K,  = horizontal hydraulic conductivity of interval i,
   K   = average hydraulic conductivity of screened materials,
   Qi  = induced flow from interval i,
   qi   = ambient flow from interval i,
   Zj   = interval i thickness,
   QP  = total extraction rate, and
   b   = aquifer thickness influenced by the test.

Determination  of   Contaminant  Distribution.   Three-dimensional
characterization of contaminant distribution was also performed. Chemical analysis
of monitoring well samples was used to identify the areal extent of contamination at
the site and to identify an area for the pilot demonstration.  However, greater
vertical resolution of contaminant distribution was needed to refine the design of the
delivery system to ensure appropriate mixing.  Hollow-stem augers in combination
with an impact driven core barrel were used to collect continuous sleeved cores of
aquifer materials. The sleeved cores were sectioned, sealed, and frozen in the field
for transport to the laboratory.  Subsections of the frozen sleeved cores  were
obtained and used for various characterization studies. Subsections representing the
combined pore water and soil matrix were analyzed by GC/Mass Spectrometry for

-------
types and relative concentrations of contaminants. Vertical resolutions of as little as
15 cm (0,5 ft) were achieved.

RESULTS  AND DISCUSSION

Flowmeter Results. The majority of the induced flow entered the well in two
zones located near the middle and bottom of the screened interval. Approximately
25% of the flow entered the well within the bottom 0.5 foot (15 cm) of the well
screen.  This indicates that the top of the Hawthorn Formation is much  more
conductive in this area than originally conceptualized, resulting in a significant flow
contribution from depths below this interval.  Therefore, the value of hydraulic
conductivity estimated for this zone is considered relatively uncertain.
       The hydraulic conductivity profile (Figure 1), estimated from this study and
the results of the previous multiwell pumping test, indicates that a stratified
hydraulic structure exists. A zone of relatively high conductivity exists between
approximately 15 ft and 22 ft (4.6-6.7 m) below TOC. This interval is bounded by
zones of lower hydraulic conductivity. Aquifer materials in this zone are as  much
as approximately  one-half order of magnitude more conductive  than the bulk
hydraulic conductivity in the screened interval. Hydraulic conductivity estimated for
aquifer materials  near the well  screen ranged from less  than  0.1  ft/d to
approximately 9 ft/d (0.03-30 m/d), spanning two orders of magnitude.
                                 10
15
20
           Concentration (mg/Kg), Hydraulic Conductivity (ft/day)

FIGURE 1.  Absolute hydraulic  conductivity of each measured  flow
interval  estimated  using  a  bulk  hydraulic  conductivity  of  2  ft/d
obtained   near   well  NEBIOTW-1,  and  organic   contaminant
concentrations  in soil/pore water samples  from composite site cores
(PS1  and PS2),  vs depth.  Depth is relative to top of casing which is
approximately land surface.

-------
Contaminant Distribution.  Results of analyses from initial cores (Figure 1)
indicate contaminants at these locations were heterogeneously distributed and
predominantly associated with materials of lower hydraulic conductivity identified
in the flowmeter survey.  The profile of the contaminants detected in the core
material suggests that reductive biotransformations are occurring in-situ and that the
potential for augmenting those transformations, with the addition of exogenous
electron donor, is high. It is also felt that recoveries of vinyl chloride in the mg/kg
range  argue that the sample collection and handling procedures  are robust and
appropriate for the characterization activities.

System Design. Detailed, three-dimensional characterization data combined with
more conventional information  were used to define the conceptual model  for
contaminant distribution and transport at this site.  A site specific three-dimensional
ground-water flow model was developed with these data and used to screen various
injection and extraction scenarios.  Scenarios were evaluated with respect to well
configurations for delivery of these solutions and potential travel times of injected
solutions through  contaminated zones.  Potential designs that did not provide
sufficient transport of injected solutions through target zones,  which included
materials with relatively low hydraulic conductivity, in approximately a 100 day
time frame were excluded from consideration.
       Detailed design considerations such as these would not have been possible
without the three-dimensional characterization data to identify contaminated zones
and the hydraulic conductivity distribution. The design chosen for the pilot study
(Figure 2) incorporates infiltration galleries and horizontal wells for fluid circulation
which may be scaled up in a cost-effective manner.

ACKNOWLEDGMENTS
       We would like to thank the other ITRD members who have contributed to
this effort, in particular we would like to  acknowledge efforts of Todd McAlary
(BEAK  Consultants  Limited), Hal Koechlein  (Lockheed  Martin Specialty
Components), Frank Beck and Mike Cook (US-EPA, NRMRL-Ada).
       Although the research described in this paper is supported in part by the US-
Environmental Protection Agency through an in-house research program, it has  not
been subjected to Agency review and therefore does not necessarily reflect  the
views of the Agency, and no official endorsement should be inferred.

REFERENCES

Molz, F.J., O.K. Boman, S.C. Young, and W,R. Waldrop, 1994.  Borehole
flowmeters: field application and data analysis, J. Hydrology. 163:347-371.

Molz, F.J., O. Guven, J.G. Mellville, I. Javandel, A.E. Hess, and F.L. Paillet,
1990. A new approach and methodologies for characterizing the hydrogeologic
properties of aquifers, U.S. Environmental  Protection Agency, EPA/600/2-90/002,
205 pp.

-------
                                   Plan View
        Horizontal wells
                                              Infiltration trench
                                              .      I
                                                           Cluster well

                                                 Fully Screened MW
                                                1 cm = 5 ft
                                Cross Section

                      Infiltration trench

                                 \L
                                                            Cluster well
 5 •-

10-


15--

20-


25 ••

30--
                       I
                         Horizontal well     Fully Screened MW
                                                             /
FIGURE 2.   Schematic of  pilot  biotreatment system design  using
infiltration  galleries and horizontal wells for extraction/injection,  and
monitoring system.

-------
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-97/030
4. TITLE AND SUBTITLE
Site Characterization Methods for
Donor Delivery Systems
2.
the Design of In-situ Electron
7. AUTHOR (S)
Steven D. Acree, Randall R, Ross, and Guy W, Sewell - US-EPA, NRMRL-
Ada, Oklahoma
Mike Hightower - Sandia National Laboratory, Albuquerque, New Mexico
Brent Weesner - Lockheed Martin Specialty Components, largo, Florida
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. EPA, National Risk Management Laboratory
Subsurface Protection and Remediation Division
P.O. Box 1198
Ada, Oklahoma
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. EPA, NRMRL, SPRD
P.O. Box 1198
Ada, Oklahoma
3. .
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/ GRANT NO.
In-House RPGS3
13. TYPE OF REPORT AND PERIOD COVERED
Symposium Paper
14. SPONSORING AGENCY CODE
EPA/600/15
15, SUPPLEMENTARY NOTES
To be published in proceedings for the Fourth International Symposium On In Situ and On-Site Bioremediation.
16. ABSTRACT
The Department of Energy and the U.S. Environmental Protection Agency have been involved in designing and
evaluating a pilot field demonstration of reductive anaerobic biological in-situ treatment technologies
{RABITT) for use as a standard remedial technology for chloroethene contamination. Innovative site
characterization techniques have been utilized to identify the hydraulics of the site and in particular the
vertical distribution of relative hydraulic conductivities. Direct extraction of intact frozen cores has been
utilized to determine the vertical distributions of contaminants in the pore spaces and on the solid matrix of
site material. The combination of these techniques along with standard site characterization methods has been
used to develop a three-dimensional picture of the site with vertical resolutions down to 0.5 ft (15 cm). This
information has then been used to evaluate different scenarios for nutrient/electron donor delivery at the
site, and when used with appropriate transport and flow codes was used to exclude designs which did not allow
for significant mixing of donor and contaminants, or which did not effic^ntly deliver nutrients/donor to all
contaminated zones. It is felt that the use of site characterization daja in this manner is critical to the
effective and appropriate design and implementation of RABITT and other in-situ treatment technologies.
17.
A. DESCRIPTORS

18. DISTRIBUTION STATEMENT
RELEASE TO THE PUBLIC
KEY WORDS AND DOCUMENT ANALYSIS
B, IDENTIFIERS /OPEN ENDED TERMS

19. SECURITY CLASS (THIS REPORT)
UNCLASSIFIED
20. SECURITY CLASS (THIS PAGE)
UNCLASSIFIED
C. COSATI FIELD, GROUP

21. NO. OF PAGES
6
22. PRICE
EPA FORM 2220-1 (REV.4-77)
PREVIOUS EDITION IS OBSOLETE

-------