NC0003468_10. DRSS_CAP Part 2 Appx G_FINAL_20160210

Appendix G
Evaluation of Potential
Groundwater Remedial
Alternatives

This page intentionally left blank
TECHNICAL MEMORANDUM
EN1009151049DEN 1
Evaluation of Potential Groundwater Remedial
Alternatives for the Dan River Steam Station Ash Basin
Site
PREPARED FOR: HDR Engineering
PREPARED BY: CH2M HILL Engineers, Inc. (CH2M)
DATE: February 8, 2016
Introduction
This technical memorandum summarizes a remedial technology screening evaluation for groundwater
at Duke Energy’s Dan River Steam Station (DRSS) site near Eden, North Carolina. Analysis of samples
collected by HDR Engineering from monitoring wells constructed in the shallow groundwater at the
DRRS site show varying levels of ash-related constituents, some of which exceed Title 15A North
Carolina Administrative Code 02L .0202 standards (2L Standards). Surface water quality data from
samples collected in the nearby Dan River, however, indicate that ash-related constituents do not
exceed the applicable North Carolina surface water quality (2B) Standards (HDR, 2015a).
Duke Energy has agreed to remove the ash in the primary and secondary ash cells and ash storage areas.
Approximately 1.2 million tons of ash will be transported to an offsite lined landfill, and it is anticipated
that the remainder will be placed into a lined landfill that will be constructed at the DRSS site. However,
impacted groundwater has already migrated beyond the basin boundary. Additional groundwater
monitoring is required, since some of the monitoring wells have only been sampled twice; so far,
antimony, arsenic, boron, chromium, cobalt, iron, manganese, selenium, sulfate, thallium, vanadium,
and total dissolved solids have been identified as constituents of interest (COI). The University of North
Carolina at Charlotte (UNCC) attempted to model the COI fate and transport component of the
groundwater by using conservative assumptions to account for data limitations. The results of their
modeling efforts indicate that there is a potential for some of the COIs—namely, cobalt, thallium, and
vanadium—to exceed the State’s 2L Standards at the DRSS site compliance boundary for years into the
future (Langley and Kim, 2015). As discussed below, removal of the COIs may have been
underestimated, since the model did not consider the effects of adsorption by a precipitated COI (that
is, iron hydroxide).
This memorandum discusses potential remedial alternatives to address the groundwater COIs that
appear to be related to historical ash deposition and are predicted to have the potential to exceed the
standards at the compliance boundary.
Background
The DRSS Comprehensive Site Assessment (CSA) (HDR, 2015a) report indicates that no imminent hazard
to human health or the environment is present at the DRSS site as a result of groundwater migration
from the ash basin or ash storage areas. This is largely because groundwater in the surficial aquifer—
both shallow and deep—flows under the DRSS ash basin and beneath the ash storage areas to the
southeast and discharges to the Dan River. There are no water supply wells located between the
Primary and Secondary ash cells and the Dan River, which serves as a downgradient hydrologic boundary
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
2 EN1009151049DEN
for regional groundwater in the DRSS site vicinity. Bedrock appears to bound the impacted groundwater
vertically (Langley and Kim, 2015), limiting offsite impacts and, based on monitoring to date, there are
no known standard exceedances beyond the current DRSS site boundary (HDR, 2015a).
Background groundwater quality was compared to the 2L Standards or Interim Maximum Allowable
Concentration (IMAC) (based on availability) as part of the CSA. Four COIs were identified in background
wells at concentrations exceeding the 2L Standards: cobalt, iron, manganese, and vanadium.
The available groundwater data from within the DRSS site potentially impacted by ash storage and basin
areas was also compared to the 2L or IMAC. Eleven COIs were identified that exceeded the applicable
standard: antimony, arsenic, boron, chromium, cobalt, iron, manganese, selenium, thallium, vanadium,
and sulfate. Of these, cobalt, iron, manganese, and vanadium exceeded the 2L Standard at a high
frequency. While these four COIs all occur naturally above the 2L Standard, the concentrations observed
in DRSS site groundwater samples within or downgradient of the ash basins were 1–2 orders of
magnitude higher than concentrations observed in the background wells, indicating that the observed
levels in the DRSS site wells are likely related to the ash.
As noted above, UNCC modeled the COI fate and transport component of the groundwater using
conservative assumptions to account for data limitations. Their model indicates that antimony, cobalt,
thallium, and vanadium will exceed the 2L Standards at the compliance boundary in the future and that
some of these COIs may already be exceeding standards at the compliance boundary (Langley and Kim,
2015). However, iron and manganese were not modeled in that exercise, and the effects of COI
adsorption onto precipitated COIs (principally iron hydroxide) were not considered. As indicated by
Miller (2015), iron hydroxide has the potential capacity to lower many of the COIs to below 2L
Standards.
Geochemical modeling (HDR, 2015b, 2016) supports Miller’s findings, showing that the iron hydroxide
potentially provides a large amount of adsorptive surface area, though actual adsorption depends on
pH. According to the Eh-pH diagrams (HDR, 2016), the iron should be present as iron hydroxide in most
of the groundwater monitoring wells, but based on the observed water quality in these wells, much of
the iron has not precipitated, most likely due to the limited amount of oxygen at depth.
Concentrations of aluminum, arsenic, chromium, cobalt, copper, lead, thallium, and zinc were measured
above the State’s 2B Standards in a surface water sample collected from the unnamed tributary in the
eastern portion of the DRSS site. In general, these COIs are similar to the groundwater COIs identified in
monitoring wells located between the eastern unnamed tributary and the secondary cell of the ash
basin. Analysis of surface water collected from the Dan River near the confluence with the tributary
stream indicate that the Dan River has not been adversely affected by water quality in the unnamed
tributary (HDR, 2015a).
Development of Remedial Alternatives
As noted above, Duke Energy has agreed to remove the ash in the primary and secondary ash cells and
ash storage areas. Approximately 1.2 million tons of ash will be transported to an offsite lined landfill,
and it is anticipated that the remainder will be placed into a lined landfill that will be constructed at the
DRSS site. Potentially suitable remedial measures that could be used as part of a comprehensive site
remedy or as a stand-alone remedy to address the residual COIs in groundwater were evaluated. The
section below provides a brief description of potentially applicable remedial measures identified for
potential use. Potential correction action alternatives for the specific site conditions of the DRSS site
groundwater are also presented. Table 1 provides an overview of anticipated effectiveness,
implementability, and associated uncertainties with each alternative.
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
EN1009151049DEN 3
Technology Screening
Potentially applicable measures are summarized below. The purpose of this section is to briefly define
the technology and any general qualifying remarks related to the site. This section identifies whether
the technology is a feasible measure to apply to DRSS. This screening was used to develop the site
remedial alternatives.
Source Controls
Groundwater quality is improved by restricting ash contact with groundwater and surface water.
Removal of source material will be implemented at the DRSS site. The plan is to remove the ash from
one of the ash storage basins and take it to an offsite landfill, and then construct a new lined landfill on
site where that ash was previously located. The rest of the ash onsite will then be moved into that
landfill. The objective of removal is to prevent any future impact to groundwater from historical ash
management practices.
Where removal of the ash is not technically or economically feasible, recharge through the ash can be
reduced by placing an impermeable cap or cover over it. Impermeable caps help shed surface water,
preventing infiltration into the ash, and can be designed to direct this rain water to specific locations.
If the ash basin intersects the water table and some of the ash is submerged, remedial options to control
the leaching of COIs from submerged ash can include the installation of an upgradient cutoff wall to
divert groundwater around the ash or ash solidification.
Cutoff walls can be constructed with soil-bentonite slurry, cement grout, or geosynthetic materials.
Grout curtains are thin, vertical grout walls installed in the ground that are constructed by pressure-
injecting grout directly into the soil at closely spaced intervals. The spacing is selected so that the grout
forms a continuous wall or curtain. Grout curtains are similar to slurry walls but typically do not require
extensive trenching.
Geosynthetic material similar to a sheet pile can be vibrated into the ground, provided the overburdens
soils do not have too many obstructions that would complicate construction. Site-specific aspects, such
as the required depth, the anticipated groundwater pressure, and the nature of the subsurface,
determine which approach is appropriate at a specific site.
In situ solidification/stabilization (ISS) involves mixing the ash and contaminated soils with
approximately 8–12 percent by weight of pozzolanic materials, such as portland cement or blast furnace
slag, to reduce or eliminate leaching of COIs from the source zones. Blending portland cement or other
pozzolans with ash or impacted soil can also reduce COI mobility, as the matrix either solidifies or
chemically binds the COI. The net impact of applying ISS to the site is generally to improve ash strength,
reduce the leachability of COI, and reduce hydraulic conductivity, which reduces groundwater contact
with the COIs. Adding a pozzolan can change the local redox conditions or pH, so the overall chemical
stability of the pozzolan addition should be explored at a bench-top scale.
Since the intent is to remove or contain all of the ash from the DRSS site, these alternative source
control measures are not included among the remedial alternatives for this site.
Groundwater Remediation
Monitored Natural Attenuation
Description. While model predictions can simulate long-term natural attenuation using a soil-water
partitioning coefficient to estimate attenuation, natural conditions will dictate local sorption of COIs.
Natural attenuation mechanisms include adsorption of COIs onto soil particles and mineral precipitates,
ion exchange, the formation of precipitated minerals that contain the COIs, and dilution from recharge.
A key aspect of the monitored natural attenuation (MNA) approach is long-term groundwater
monitoring to evaluate naturally occurring adsorption over time. Real-time data are the best indicator of
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
4 EN1009151049DEN
natural attenuation mechanisms. The monitoring results will verify the degree to which natural
attenuation is occurring, and verify that the footprint of site-related impacts is not increasing.
Applicability to DRSS. Given that the ash will be removed and placed in a lined landfill, it is reasonable to
assume that the concentrations of COIs remaining in groundwater would subsequently decrease over
time as recharge flushes nonimpacted water through the aquifer. The results of the groundwater model
(Langley and Kim, 2015) indicate a reduction, though the extent to which concentrations are reduced
vary among the different COIs, which is in part due to attenuation factors. Miller (2015) found that
arsenic and vanadium were clearly attenuating. A conceptual model for COI attenuation involving
reversible and irreversible interaction with clay minerals, metal oxides, and organic matter was
proposed for these COIs and for antimony, boron, chromium, cobalt, selenium, and thallium. The most
significant finding from this analysis and from subsequent laboratory tests was that COIs were
concentrated in samples that had been exposed to groundwater containing higher concentrations of
COIs, most likely indicating that the precipitating iron, and possibly also some aluminum and
manganese, was removing other COIs through coprecipitation and adsorption, thus confirming that
attenuation was occurring (Miller, 2015). Collectively, this attenuation, sorption, and precipitation are all
already occurring without any active controls. However, more data and some follow-up tests are needed
to verify where this natural attenuation is happening and if it is occurring at a sufficient rate to mitigate
COIs.
The groundwater model performed did not allow for removal of COI via coprecipitation with iron oxides,
which likely resulted in an overprediction of COI transport; thus, the model predicts that some of the
COIs will exceed the 2L Standards at the compliance boundary while this is actually unlikely to occur.
Additional testing is required to fully assess the magnitude of this attenuation (Miller, 2015), but it is
feasible to consider that MNA alone will remediate most of the DRSS site. Additional groundwater
monitoring data from existing and new wells can be used to further assess the MNA alternative.
Enhanced Recharge/Flushing
Description. It is possible to increase the rate of groundwater quality improvement by increasing
infiltration of uncontaminated water into portions of a site that are beneath or downgradient of ash
deposits, thereby flushing, diluting, and attenuating the remnant concentrations of COIs. There are
various ways to do this, ranging from short-term or temporary methods (for example, surface irrigation
using mechanical sprayers) to the creation of groundwater infiltration galleries or ponds or wetlands
with a permeable bottom, which could be temporary or permanent. Where a continuing source is
present, permanent infiltration galleries would be needed.
Applicability to DRSS. The groundwater models used a recharge rate of 6.5 inches per year in the area
outside of the ash basins, and 13.14 inches per year within the ash basins (Langley and Kim, 2015) based
on anticipated factors influencing recharge. CH2M evaluated the DRSS site to determine if there existed
a potential location to enhance recharge using an infiltration basin, after the ash is removed. It was
concluded that the primary and secondary basins are too close to the river to allow this options since
enhanced recharge in these areas would likely cause increased downgradient seepage and, potentially,
slope instability.
After the ash in Ash Storage Basin 1 to the east is removed, its contents will be shipped offsite, and the
basin will be contoured and converted into a lined landfill. The content of Ash Storage Basin 2 as well as
the primary and secondary basins will be placed in the lined landfill. Ash Storage Basin 2, to the west,
will no longer contain ash and so it would appear to be an appropriate potential location for this
approach. However, CH2M understands that there is a sensitive wetland between the two basins, and
increasing recharge in that location may drive contaminants toward that wetland. Consequently, it was
decided that there was no practical place to use this technology at this site.
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
EN1009151049DEN 5
Chemical Fixation
Description. Various measures can be taken to enhance adsorptive removal of COIs by blending
materials that have a high adsorptive capacity, such as clays, peat moss, and zeolites, into the
contaminated soils. Contaminated groundwater can also be treated in situ using chemical fixation by
adjusting the pH and/or redox state of the groundwater by, for example, enhancing the precipitation of
iron and manganese oxide and hydroxide minerals in the groundwater. Enhanced formation of these
minerals does more than remove iron and manganese from the groundwater because these minerals
effectively coprecipitate and adsorb other COIs. Redox conditions can be adjusted, either by adding one
or more reagents (in situ chemical oxidation, or ISCO) or through air sparging. Bench-scale treatability
testing and/or pilot-scale tests are usually required to verify the effectiveness of this technology at a
specific site prior to full-scale application, and to select the most appropriate reagent and dosage. It
should be noted, though, that attenuation of boron will be limited and that TDS and chloride
concentrations will probably not be diminished at all using this approach.
Applicability to DRSS. Preliminary geochemical modeling suggests that iron and manganese are already
precipitating beneath and downgradient of the ash basin and storage areas and limiting the mobility of
other COIs in the process (HDR, 2015b, 2016). Reagent addition (or air sparging) could further encourage
the precipitation of COIs by changing the redox conditions over specifically targeted areas. Natural
attenuation could be enhanced using either ISCO or air sparging. While ISCO involves the injection of a
chemical oxidant, such as potassium permanganate, air sparging simply involves pumping air into the
targeted saturated zone. Although it would appear that air sparging would be less expensive since there
are no chemical costs, lifetime costs may be comparable (air sparging has operations and maintenance
costs that may not outweigh the cost of chemicals and possible reinjection events associated with ISCO).
Therefore, before either technology is used, it is recommended that both approaches be tested onsite
during a pilot-scale test to see which approach would be more cost effective, since a myriad of variables
affects their comparable performance.
Due to the large vertical distribution of COIs (that is, their presence in groundwater in the shallow, deep,
and bedrock zones—such as to a depth of 60 feet below the ground surface in the northeast portion of
the DRSS site at GWA-5BR), applying ISCO or air sparging is most likely to be cost effective for key
limited areas of concern, rather than an extensive area.
However, within a smaller targeted area, such as between the secondary cell and the eastern unnamed
tributary to the Dan River, a barrier-style application would likely be cost-effective. The depth to
bedrock is also much less there. Still, periodic reinjection (or air sparging) would be necessary to
maintain the desired redox conditions as upgradient groundwater enters the treatment area until
monitoring demonstrates that the groundwater has been improved by the removal of the ash from the
secondary cell. Until that occurs, monitoring would be required to determine the frequency of
reinjection. In addition, geochemical modeling to evaluate whether the precipitation of COIs is likely to
be permanent under the site-specific conditions is recommended.
Permeable Reactive Barrier
Description. Permeable reactive barriers (PRBs) are a passive form of in situ water treatment that
removes COIs in a subsurface zone using media that react with COIs to remove them from groundwater
as it flows through the barrier. PRBs are typically constructed by excavating a trench that fully
penetrates the saturated zone of the unconsolidated aquifer and places material in the trench to treat
the groundwater. Depending on the required depth, specialized equipment can be used to trench and
place media simultaneously. There are multiple types of media that are used for in situ treatment, and
they are selected based on the contaminants required for removal. Some media are difficult to deploy at
depth, and installation depth may be determined by the media required to treat the COI.
A funnel-and-gate system can be used to channel the contaminant plume into a gate that contains the
media that will treat the COIs. The simplest design of this system consists of a single permeable area
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
6 EN1009151049DEN
(the gate) containing appropriate media to remove the COIs, with non-permeable walls extending from
both sides (the funnel). The main advantage of the funnel-and-gate system is that a smaller reactive
region can be used to treat the plume, which can reduce costs. In addition, if the treatment media need
to be replaced, it is easier because there is less material to replace.
There have been many successful implementations of PRBs at sites with a wide range of constituents,
but only limited testing with water containing the constituents contained in ash leachates (EPRI, 2006;
ITRC, 2005). The principle problem is that conventionally constructed PRBs are not designed to
significantly reduce boron concentrations. However, some materials, such as iron hydroxide and rice
husks, have been shown to absorb boron in the laboratory (Goldberg et al., 1993; Man et al., 2012).
Where boron is present at concentrations that exceed the 2L Standards, the laboratory tests would have
to be scaled up to a pilot-scale level to confirm how well these materials would perform in a PRB and
whether additives would be necessary to maintain the required level of permeability.
Depending on whether the removal mechanisms are compatible (for example, adsorption and ion
exchange) or not (for example, an oxidizing zone to precipitate metals as oxides and a reducing zone
containing organics to precipitate metals as sulfides), the media might be mixed and placed together to
create a single treatment zone or emplaced sequentially to create multiple treatment zones. PRBs
generally have a limited lifespan, depending on the sorption characteristics and flow rate, as the reactive
media are consumed or become less effective. Long-term remediation effectiveness may require the
periodic replacement of the PRB’s media.
Removal of inorganics has been accomplished using a range of materials, including apatite, zero-valent
iron, carbon, and other media. Site-specific media should be evaluated with a range of adsorbents to
best determine the type and blend ratio to effectively remove COI while maintaining hydraulic
conductivity.
Applicability to DRSS. There have been many successful PRB remedies at sites with a wide range of
constituents, but only limited testing with water containing the constituents in ash leachates (EPRI,
2006). However, there are areas, such as the eastern section of the DRSS site, where boron does not
exceed the 2L Standard and a conventional PRB design could be considered. Based on a review of the
data, it appears that a suitable PRB for this portion of the DRSS site could comprise a combination of
limestone aggregate (to provide PRB stability, transmissivity, and pH buffering) and organic materials
(mulch, wood chips, etc.) to promote the reduction of sulfate to sulfide and precipitation of the
inorganics, and potentially zero valent iron to help promote and sustain the reducing conditions.
Should monitoring after the ash is removed and placed in the lined landfill indicate that the Dan River is
being or is likely to be adversely affected by COIs in the groundwater, construction of a PRB could be
considered along the riverfront, just upgradient of the boundary with the river. Installation at this
location would allow the barrier to intercept and remove most of the COIs as the groundwater
discharges to the river. A review of data suggests that both shallow and deep aquifer zones might have
to be treated in such a scenario, which would require that the PRB be installed to approximately 30–35
feet below the ground surface. The current location of the steep berms that form the boundary of the
ash cells limits the feasibility of trenching equipment to maneuver within the available space; however,
with the berms removed, there should be sufficient space.
Near the unnamed tributary east of the ash basin cells, where the depth to bedrock is shallower, a
funnel-and-gate PRB design could be used if treatment warranted. Sheet pile and excavation, which
were routinely used for the installation of early PRBs, could be used to provide the needed support near
the river, should placement there be required.
PRB construction is a relatively expensive remedial approach compared to both natural attenuation and
in situ treatment. Therefore, it should be considered a fallback provision at the DRSS site that could be
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
EN1009151049DEN 7
implemented if water quality monitoring of the Dan River reveals ash-related exceedances of the 2B
Standards following the completion of source control measures (removal of the ash to a lined landfill).
Ex Situ Groundwater Treatment
Description. As an alternative to in situ groundwater treatment methods, as discussed above, impacted
groundwater can be extracted from the flow layers and treated above grade. Impacted groundwater
would be pumped to the surface (pump-and-treat) or captured at the surface in order to provide
hydraulic containment and prevent COI migration to sensitive receptors. Following treatment, the water
may be discharged directly to a surface water body or reinjected underground, depending on the site
conditions and permitting requirements. Water treatment can be active (requiring the continual
addition of chemicals and typically, electrical power) or passive (systems that take advantage of
reactions that occur in nature, such as constructed wetlands or limestone beds to provide
neutralization). The use of passive systems is generally restricted to smaller flows because the approach
typically requires much larger land area than active systems, but has the advantages of less maintenance
and lower operating costs. Passive treatment systems, however, can be ineffective at removing select
COIs, such as boron and total dissolved solids. Active treatment systems are generally costly to construct
and operate but can be designed to effectively lower the concentrations of all of COIs associated with
the DRSS site.
Applicability to the DRSS Site. The area and depth of impacted groundwater across the DRSS site is
widespread. While a zone of depression could be created, thereby minimizing offsite transfer of COIs, it
is anticipated that the system would have to operate a considerable time into the future until
groundwater concentrations dropped to standards. It is important to note though that such a system
could not be placed too close to the Dan River or the unnamed tributary without potentially pumping
water from these water bodies, and these areas would likely be where COIs would cause the greatest
concern and would be the targeted areas for pumping wells. Consequently, this technology was not
considered further.
Assumptions
The following assumptions were used when developing potential remedial alternatives for the DRSS:
• All of the ash will be successfully removed from the DRSS site; berms will be removed or graded.
• The COI concentrations observed to date in the monitoring wells are representative of the site-wide
groundwater quality (initially).
• The recommended remedial alternatives are based on the UNCC groundwater modeling (Langley
and Kim, 2015) of the DRSS site, which has projected the nature of future groundwater quality
downgradient of the ash cells and basins. However, given the areal extent of the ash basin and the
anticipated uniformity of chemical constituents in the ash, the model’s predictions of localized
contaminated areas may be biased by the limited available data or the limitations of the model.
Actual conditions may differ, warranting further review of remedial alternatives.
• Further evaluation of remedial alternatives may be necessary if any of these conditions/assumptions
change. CH2M recommends that the modeling be reviewed to determine whether additional
calibration is needed before it is used for remedial design.
• Compliance boundaries are as identified in figures in part 2 of the corrective action plan report
(HDR, 2016). Where boundaries occur upland of a surface water body, the compliance standard is 2L
or IMAC; where boundaries occur within a surface water body, the compliance standard is 2B and
applied to observed concentrations (if available) or mixing zone modeled concentrations.
• No site inspection or other engineering assessment has been performed regarding the
implementability of any option; therefore, concepts presented will need to undergo a
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
8 EN1009151049DEN
constructability assessment (that is, an evaluation of site conditions following the removal action to
determine whether any site-specific conditions would preclude the construction of the technology
in the proposed location).
Evaluation of Remedial Alternatives for the DRSS
The screening above identified MNA, ISCO, and PRBs as potentially applicable technologies. These
technologies were evaluated with respect to feasibility, anticipated benefits, uncertainties, and cost
effectiveness. PRBs were determined to be much less cost-effective than ISCO, and were not included as
a recommended alternative. The remaining remedial alternatives for addressing COIs in groundwater
following ash removal determined to be potentially applicable to the DRSS are discussed below.
Alternative 1—No Further Action
The purpose of including No Further Action is to provide a baseline for comparison to other measures.
With this approach, there would be no further remedial actions conducted at the site to control or
remove the source of the COIs other than the removal of the ash, and no further remedial action would
be taken. This measure does not include long-term monitoring or institutional controls.
Alternative 2—Monitored Natural Attenuation
MNA involves regularly monitoring select parameters to ensure that concentrations of COIs in the
groundwater are decreasing. Monitoring would be maintained until water quality meets 2L Standards or
IMAC at the compliance boundary (groundwater). It is anticipated that water quality improvement will
occur over time due to natural processes once the COI source material has been removed. The
monitoring framework would be selected based on modeling and historical results. As several COIs are
also measured in background wells, a network of background well locations would also be identified so
that temporal changes that occur naturally can be monitored, and progress toward attainment of
standards can be assessed. Most MNA programs require monitoring at least twice annually.
Based on the groundwater model and its projections, it will take many years to attain standards at the
compliance boundary. However, MNA was examined in more detail (Miller, 2015), and it appears that
precipitation, coprecipitation, and adsorption will greatly reduce the concentrations of at least some of
the COIs. For assessment purposes, it was assumed that this alternative would include five additional
monitoring well couplets to be constructed along the Dan River compliance boundary and that the series
of existing, new, and background wells—a total of 15 wells—would be monitored at least twice annually
for COIs. While attenuation timeframes were not projected, the lifespan of this and all alternatives was
fixed at a maximum of 30 years. As it is anticipated that CAMA CAP requirements will require monitoring
over the near term, this alternative assumes that new wells will be monitored as part of this alternative
for the next 5 years, and a larger network of wells will be monitored for the following 25 years (16
downgradient; 9 upgradient).The data would be compiled and reviewed for MNA annually, and a report
would be issued.
Alternative 3—Monitored Natural Attenuation and Enhanced Natural Adsorption by In Situ
Sorption or Chemical Fixation in the Northeast Corner of the Site
Enhancing adsorptive removal of COIs in groundwater by either adding materials to the saturated zone
that have a high adsorptive capacity or by adjusting the pH and/or redox state of the groundwater, or a
combination of these measures, would enhance adsorption of COIs in targeted areas. As noted in the
technology screening, it is unlikely that ISCO would be cost-effective to apply to large areas of the DRSS
site. Since preliminary modeling has indicated that natural attenuation processes are occurring at the
DRSS site, this alternative was developed to focus ISCO on the area of COI exceedances at the eastern
portion of the DRSS site, to prevent the continued migration of COIs towards the unnamed tributary to
the Dan River. In addition, it is assumed that MNA would be monitored with the same well network as
with Alternative 2.
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
EN1009151049DEN 9
Modeling and bench/pilot tests should be undertaken to confirm the potential cost effectiveness of the
various treatment options, the quantities that would be required, and where the material would
optimally be injected within the targeted area. However, based on the COIs of potential concern at this
site, injecting an oxidant or air to enhance oxidation and precipitation of the iron already present in the
water should add enough additional sorbent capacity to achieve the desired objectives (Evanko and
Dzombak, 1997; Ilavský, 2008; Twidwell and Williams-Beam, 2002). Antimony requires a high iron-to-
contaminant ratio (Pawlak et al., 2002), but considering the amount of iron in the groundwater (HDR,
2015a) and the much lower antimony concentrations, the existing ratio appears appropriate. It is also
possible that an alkaline agent (for example, sodium carbonate) would also have to be injected, since
the precipitation of iron hydroxide will lower groundwater pH. Comparative field trials of air sparging
and ISCO, as previously discussed, would show whether either or both approaches lowers groundwater
pH enough to adversely affect COI adsorption and require such a measure. For assessment purposes,
the following key assumptions were made, but it should be recognized that the actual layout and
operation of the wells will be influenced by site-specific conditions and cannot be determined until after
the pilot-scale tests are conducted:
• The targeted area is approximately 400 feet by 700 feet at and upgradient of the unnamed tributary
on the eastern portion of the DRSS site. A barrier approach to injection, consisting of three
transects, one 400 feet long and two 500 feet long, would be used.
• One hundred twenty-nine two-tiered injection wells, each with a 12-foot radius of influence, would
be installed to form the three transects, with sets of 16 manifolded together to allow simultaneous
injection.
• The selected reagent would consist of a 3 percent solution of potassium permanganate injected at 5
gallons per minute.
• Five injection events would be required, at 2-year intervals.
It should be noted that since the ash is being removed and placed in lined areas, this approach would
only have to continue until the COI concentrations in the groundwater are lowered sufficiently so that
natural adsorption can be relied on to satisfactorily address criteria compliance. For assessment
purposes, this has been assumed to require a total of 10 injection events based on the low groundwater
velocity in this area.
Recommendations
Recognizing that the planned excavation and removal of the ash and drainage of the ash basins will
affect recharge and that the subsequent surface rehabilitation can be designed to direct surface water
and, to a lesser extent, groundwater flow in a desired direction, it is prudent to continue groundwater
monitoring during and after this site remediation. While this is going on, efforts should be made to
enhance the groundwater model so that it can fully consider the effects of transitioning recharge as well
as the potential effects of site geochemistry, such as COI removal by adsorption onto precipitated COIs.
Evidence supports Alternative 2 (MNA) as a viable remediation approach for most of this site; however,
further data and analyses are needed to demonstrate this. MNA should be further investigated with
respect to site-specific conditions to demonstrate where there is adequate attenuation due to naturally
occurring reactions.
There are two areas onsite where impacted groundwater appears to have migrated from beyond where
the ash was stored and to be approaching the compliance boundary: to the southeast of the secondary
ash basin and to the northwest of Ash Storage Basin 2. Given that water quality in the unnamed
tributary to the Dan River has apparently been adversely affected already, appropriate laboratory tests
and modeling should be initiated to define optimal additives and injection locations, and thus determine
how best to enhance natural attenuation in the area between the ash basin cells and the unnamed
EVALUATION OF POTENTIAL GROUNDWATER REMEDIAL ALTERNATIVES FOR THE DAN RIVER STEAM STATION ASH BASIN SITE
10 EN1009151049DEN
tributary (Alternative 3). This would be followed by field tests of air sparging and ISCO to determine
which would be most cost-effective in this area. The same approach could be undertaken beneath
where the ash basins used to be after the ash is removed, if water quality in the monitoring wells
indicates that this is needed, but this possibility was not included in the assessment.
References
EPRI (Electric Power Research Institute). 2006. Groundwater Remediation of Inorganic Constituents at
Coal Combustion Product Management Sites: Overview of Technologies, Focusing on Permeable Reactive
Barriers. EPRI: Palo Alto, CA.
Evanko, C.R., and D.A. Dzombak. 1997. Remediation of Metals-contaminated Soils and Groundwater.
Technology Evaluation Report 97-01. Ground-water Remediation Technologies Evaluation Center,
Pittsburgh, PA.
Goldberg, S., H. S. Forster, and E. L. Heick. 1993. Boron adsorption mechanisms on oxides, clay minerals,
and soils inferred from ionic strength effects. Soil Science Society of America Journal, vol. 57, pp. 704-
708.
HDR. 2015a. Comprehensive Site Assessment (CSA) Report for the Dan River Steam Station Ash Basin.
HDR Engineering Inc.
HDR. 2015b. Corrective Action Plan (CAP) Part 1 Dan River Steam Station Ash Basin. HDR Engineering Inc.
HDR. 2016. Corrective Action Plan (CAP) Part 2 Dan River Steam Station Ash Basin (draft). HDR
Engineering Inc.
Ilavský, J. 2008. “Removal of Antimony from Water by Sorption Materials.” Slovak Journal of Civil
Engineering. pp. 1–6. http://www.svf.stuba.sk/docs/sjce/2008/2008_2/file3.pdf.
ITRC (Interstate Technology and Regulatory Council). 2005. Permeable Reactive Barriers: Lessons
Learned/New Directions. Interstate Technology and Regulatory Council, Permeable Reactive Barriers
Team, PRB-4, Washington, D.C., available on the Internet at www.itrcweb.org.
Langley, W.G., and D. Kim. 2015. Groundwater Flow and Transport Model Dan River Steam Station,
Rockingham County, NC, University of North Carolina, Charlotte, NC, prepared for HDR Engineering Inc.
Man, H.C., W. H. Chin, M. R. Zadeh, and M. R. M. Yusof. 2012. “Adsorption Potential of Unmodified Rice
Husk for Boron Removal.” BioResources. Vol. 7, no. 3. pp. 3810–3822.
Miller, G.P. 2015. Technical memorandum to Mark Filardi, HDR Engineering, Inc.
Pawlak, Z., P.S. Cartwright, A. Oloyede, and E. Bayraktar. 2002. “Removal of Toxic Arsenic and Antimony
from Groundwater Spiro Tunnel Bulkhead in Park City Utah Using Colloidal Iron Hydroxide: Comparison
with Reverse Osmosis.” Advanced Materials Research. Vols. 83–86, pp. 553–562.
Twidwell, L. G., and C. Williams-Beam. 2002. “Potential Technologies for Removing Thallium from Mine
and Process Wastewater: An Abbreviated Annotation of the Literature.” European Journal of Mineral
Processing and Environmental Protection. Vol. 2, no. 1. pp. 1–10.
EV
A
L
U
A
T
I
O
N
 OF
 PO
T
E
N
T
I
A
L
 GR
O
U
N
D
W
A
T
E
R
 RE
M
E
D
I
A
L
 AL
T
E
R
N
A
T
I
V
E
S
 FO
R
 TH
E
 DA
N
 RIVER  STEAM  STATION  ASH  BASIN  SITE  
EN
1
0
0
9
1
5
1
0
4
9
D
E
N

11  
Ta
b
l
e
 1.
 Re
m
e
d
i
a
l
 Al
t
e
r
n
a
t
i
v
e
s

Re
m
e
d
i
a
l

Al
t
e
r
n
a
t
i
v
e
s

De
s
c
r
i
p
t
i
o
n

An
t
i
c
i
p
a
t
e
d
 Ef
f
e
c
t
i
v
e
n
e
s
s

Im
p
l
e
m
e
n
t
a
b
i
l
i
t
y

Uncertainties   
1.
 No
 Fu
r
t
h
e
r

Ac
t
i
o
n

No
 re
m
e
d
i
a
l
 me
a
s
u
r
e
s

im
p
l
e
m
e
n
t
e
d
.

Th
i
s
 al
t
e
r
n
a
t
i
v
e
 pr
o
v
i
d
e
s
 a  ba
s
e
l
i
n
e

fo
r
 co
m
p
a
r
i
s
o
n
 to
 ot
h
e
r
 me
a
s
u
r
e
s
.

No
 ac
t
i
o
n
 wo
u
l
d
 be
 ta
k
e
n
 to
 ad
d
r
e
s
s

gr
o
u
n
d
w
a
t
e
r
 CO
I
s
.

No
t
 ap
p
l
i
c
a
b
l
e
.

2.
 MN
A

Mo
n
i
t
o
r
i
n
g
 of
 su
r
f
a
c
e
 wa
t
e
r
 an
d

gr
o
u
n
d
w
a
t
e
r
 af
t
e
r
 as
h
 re
m
o
v
a
l
.

At
 pr
e
s
e
n
t
,
 th
e
 si
t
e
 is
 no
t
 ca
u
s
i
n
g

an
y
 do
w
n
g
r
a
d
i
e
n
t
 en
v
i
r
o
n
m
e
n
t
a
l

pr
o
b
l
e
m
s
 ex
c
e
p
t
 at
 on
e
 sa
m
p
l
i
n
g

lo
c
a
t
i
o
n
 in
 th
e
 un
n
a
m
e
d
 tr
i
b
u
t
a
r
y
.

MN
A
 is
 pr
o
t
e
c
t
i
v
e
 of
 hu
m
a
n
 he
a
l
t
h

an
d
 th
e
 en
v
i
r
o
n
m
e
n
t
 be
c
a
u
s
e
 si
t
e
‐
re
l
a
t
e
d
 CO
I
s
 wi
l
l
 na
t
u
r
a
l
l
y
 at
t
e
n
u
a
t
e

ov
e
r
 ti
m
e
.

MN
A
 ha
s
 b ee
n
 de
m
o
n
s
t
r
a
t
e
d
 to
 be

ef
f
e
c
t
i
v
e
 at
 ma
n
y
 si
t
e
s
 an
d
 is
 ea
s
y
 to

im
p
l
e
m
e
n
t
 us
i
n
g
 ex
i
s
t
i
n
g
 mo
n
i
t
o
r
i
n
g
 we
l
l
s

an
d
 sa
m
p
l
i
n
g
 si
t
e
s
.

As
h
 re
m
o
v
a
l
 should  eliminate  the  source  of  
th
e
 si
t
e
‐re
l
a
t
e
d
 water  contamination, 
th
o
u
g
h
 it
 may  take  time  for  groundwater  
qu
a
l
i
t
y
 to
 meet  remedial  objectives. 
Ev
i
d
e
n
c
e
 of  MNA  mechanisms  attenuating  
CO
I
 ha
s
 be
en  demonstrated; adsorption  
ca
p
a
c
i
t
y
 an
d
 other  site ‐specific  factors  
sh
o
u
l
d
 be
 further  evaluated. 
3.
 MN
A
 an
d

En
h
a
n
c
e
d

Na
t
u
r
a
l

Ad
s
o
r
p
t
i
o
n
 an
d

In
 Si
t
u
 Ch
e
m
i
c
a
l

Fi
x
a
t
i
o
n

Ad
d
i
n
g
 mi
t
i
g
a
t
i
v
e
 ag
e
n
t
s
 to
 th
e

gr
o
u
n
d
w
a
t
e
r
 du
r
i
n
g
 or
 af
t
e
r
 as
h

re
m
o
v
a
l
 an
d
 si
t
e
 re
m
e
d
i
a
t
i
o
n
 to

re
m
o
v
e
 re
m
n
a
n
t
 CO
I
s
 by

ad
s
o
r
p
t
i
o
n
 an
d
/
o
r
 io
n
 ex
c
h
a
n
g
e
.

Th
i
s
 al
t
e
r
n
a
t
i
v
e
 in
c
l
u
d
e
s

lo
c
a
l
i
z
e
d
 in
j
e
c
t
i
o
n
s
 an
d
 MN
A

ac
r
o
s
s
 th
e
 re
m
a
i
n
d
e
r
 of
 th
e

ar
e
a
.

Co
s
t
 ef
f
e
c
t
i
v
e
n
e
s
s
 ca
n
n
o
t
 be

ac
c
u
r
a
t
e
ly
 as
s
e
s
s
e
d
 at
 th
i
s
 ti
m
e
.

La
b
o
r
a
t
o
r
y
 te
s
t
s
,
 mo
d
e
l
i
n
g
,
 an
d
 fi
e
l
d

te
s
t
s
 ar
e
 re
c
o
m
m
e
n
d
e
d
.

Sh
o
u
l
d
 be
 ea
s
y
 to
 im
p
l
e
m
e
n
t
,
 bu
t
 th
e

fe
a
s
i
b
i
l
i
t
y
 of
 re
a
g
e
n
t
 de
l
i
v
e
r
y
 wi
l
l
 de
p
e
n
d

on
 th
e
 re
a
g
e
n
t
 ty
p
e
 se
l
e
c
t
e
d
 an
d
 re
s
u
l
t
s
.

As
s
e
s
s
m
e
n
t
 as
s
u
m
e
s
 ti
m
e
d
 re
l
e
a
s
e

re
a
g
e
n
t
s
 ar
e
 us
e
d
 an
d
 mu
l
t
i
p
l
e
 in
j
e
c
t
i
o
n
s

ar
e
 re
q
u
i
r
e
d
.
 La
b
o
r
a
t
o
r
y
 an
d
 pi
l
o
t
‐sc
a
l
e

te
s
t
s
 co
u
l
d
 in
d
i
c
a
t
e
 ho
w
 to
 si
g
ni
f
i
c
a
n
t
l
y

lo
w
e
r
 co
s
t
.

Th
e
 in
s
t
a
l
l
a
t
i
o
n
 and  O&M  costs  of  potential  
op
t
i
o
n
s
 sh
o
u
l
d
 be  assessed. Added  reagents  
co
u
l
d
 po
s
s
i
b
l
y
 affect  aquifer  characteristics  
an
d
 mo
d
i
f
y
 flow  patterns. Modeling  and  field  
te
s
t
s
 ar
e
 ne
e
d
e
d
 to  determine  whether  this  
mi
g
h
t
 oc
c
u
r
 and  potential  impacts.