TAG reports Volume 4, Number 2;
August, 1995
Recent 009 Treatabilty Studies Raise Doubts About Stabilization
Overview
Recent studies raise new questions regarding the usefulness of in situ stabilization at
the 009 toxaphene landfill in Glynn County. The EPA selected an experimental stabilization
process as preferential to soil extraction, mainly due to cost considerations. In situ
stabilization involves pumping and mixing concrete underground to solidify waste. In
making the decision to use stabilization EPA argued that the nature of the landfill
suggested stabilization might be an affordable methodology for treatment of the 009
dumpsite. However, recent studies on the landfill indicate deeper contamination, more
complex waste distribution, and a variety of buried interfering debris that could hinder
waste immobilization. Further, benchtop treatability studies show that site soils mixed
with concrete can release toxaphene contaminated fine particles at unacceptably high
rates. At present, field trials are underway to determine if underground soil waste
immobilization is practical at this site.
Background of Site Remediation Studies
The Hercules 009 Superfund site is a former roadbed "borrow pit." After
construction of Georgia 25 the pit was partially filled with "stumpdirt"
produced from manufacturing operations at the Hercules plant. Manufacturing sludge,
containing the banned pesticide toxaphene, was placed in the landfill as layers or bands
and covered with additional stumpdirt. The landfill was closed in 1980 after finding
toxaphene waste eroding beyond the permitted borders of the landfill. Toxaphene, known to
cause mutations and cancer in laboratory studies, represents a threat to exposed humans.
Prior to this survey the only information on landfill construction and contents came from
historical records and, sometimes conflicting, eye witness accounts. Earlier tests during
the Remedial Investigation (RI) comprised only ground penetrating radar and sonic imaging.
These highly ambiguous results had little scientific merit. The EPA ordered two sampling
cores through the sludge cells in 1992. One of these tests indicated deep migration of
toxaphene into undisturbed soils.
Laboratory studies suggest toxaphene moves slowly in environmental processes. However,
actual field studies show toxaphene migration hundreds of miles from known areas of use.
The process of toxaphene movement is not readily predicted or understood.
Preliminary studies of in situ stabilization, the selected remedy for the 009 dump, are
performed to better understand how the stabilization process may inhibit further toxaphene
migration. The dumpsite is purported to be a deeply excavated pit filled above the water
table with "clean" waste-free stumpdirt followed by a layer of bentonite clay
for excluding water. Above the clay liner manufacturing sludge of "1% toxaphene
sludge" was buried. The borrow pit was reportedly divided into six storage cells,
each cell separated by a berm of clean stumpdirt. Presumably, due to toxaphene's poor
ability to migrate in soil, all of the toxaphene would be contained within the bentonite
lined cells. In situ stabilization of this site could fail for any number of reasons, such
as the presence of intractable debris that resists mixing with concrete, migration of
toxaphene below the areas to be remediated, site soils that resist formation of a concrete
block, concrete formulations too porous for groundwater exclusion, and/or the presence of
toxaphene on particles too fine to be trapped in the concrete. The recent EPA managed
study represents a first attempt to determine by direct observation the nature of
immobilization conditions at the 009 landfill.
Methodology
Distance measurements for buried objects and layers are given by two different criteria.
The distance "above mean sea level" (MSL), is a standard engineering term for
height measurements. For example, the dumpsite surface level varies from between 10 and 20
feet above mean (average) sea level. Estimates of the burial depth is given as "below
ground surface" or the distance the sampling auger traveled from the surface. It is
easier to visualize the water table and toxaphene levels by considering their distance
from the mean sea level mark. Thirty-two soil borings, about 4 from each cell, were taken.
The deepest boring was 26 feet below ground surface. The auger returned samples from each
1-foot level for visual examination and chemical analysis. Also, shafts of material were
excavated as test pits at 22 points around the expected edge of the landfill. The material
was sampled for toxaphene by both GC (gas chromatography) and immunoassay. The various
geological and waste levels were categorized, and the soils were examined for presence of
"foreign" objects. Additionally, a magnetic screening technique was used to
determine the distribution of metallic debris at the site.
Observations on the Storage Cells
General
Very little historical agreement occurred between these direct observations and reports of
cell construction described in the RI. No clay bentonite liner was found. Sludge occurred
in multiple layers in all cells, except the northernmost Cell 1. Toxaphene occurred
outside of the sludge layers in all cells. No true berms were found separating the cells;
instead, toxaphene contamination was spread between, above and beneath sludge layers.
Toxaphene levels exceeded by as much as 60 times the amounts expected from deposition of a
" 1% sludge." Toxaphene is migrating into undisturbed native soils beneath the
original bottom of the borrow pit. Toxaphene contamination in several areas already
extends to the Mean Sea Level datum. Thus, toxaphene migration greatly exceeds models
predicting its immobility. Far more debris was discovered in the landfill than historical
documents indicated; including drums, concrete, glass, metal pipes and wood. The following
is a condensed version of the studies findings.
Cell 1. Bentonite layer: None observed. Average thickness of sludge band: 2 feet thick.
Lower depth of sludge: 5.5 feet. Lower level of toxaphene: Found at mean sea level, about
20 feet below ground level. Concentrations of toxaphene to 230 ppm (parts-per-million)
about 2.5 feet above mean sea level. Debris: Drums, fiberglass, wood, metal, plastic,
metal pipes, concrete block. Observations: Water level reported at 6-7 feet below ground
level, indicating toxaphene contamination within the surficial aquifer.
Cell 2. Bentonite layer: None observed. Average thickness of sludge bands: two primary
layers varying from 0.5 to 4 feet thick. Lower depth of sludge: about 10 feet below ground
level. Lower level of toxaphene: Concentrations of toxaphene to 160 ppm about 10 feet
above mean sea level. Debris: Drum parts, wood, metal, metal pipes, rocks, steel cable and
steel bands. Observations: toxaphene concentrations as high as 1,900 ppm were found 10
feet below ground level.
Cell 3. Bentonite layer: Observed in only one sample forming a discontinuous layer about 1
foot thick. Average thickness of sludge bands: two primary layers varying from 1 to 4 feet
thick. Lower depth of sludge: about 15 feet below ground level. Lower level of toxaphene:
Concentrations of toxaphene to about 40 ppm at 5 feet above mean sea level. Debris: None
reported.
Cell 4. Bentonite layer: Observed in one sample as a 4 inch band between two sludge
layers. Average thickness of sludge bands: As many as four bands of varying thickness.
Lower depth of sludge: about 15 feet below the surface. Lower level of toxaphene:
Concentrations of toxaphene to 40 ppm at mean sea level. Debris: None reported.
Cell 5. Bentonite layer: One sample reported at 2-3 inches. Average thickness of sludge
bands: two primary layers varying from 0.5 to 4 feet thick. Lower depth of sludge: about
10 feet below ground level. Lower level of toxaphene: Concentrations of toxaphene to 1,400
ppm about 10 feet above mean sea level. Debris: Rocks and bits of metal. Observations:
Toxaphene concentrations as high as 4,900 ppm were found at 15 feet and 6,400 ppm at 10
feet below ground level.
Cell 6. Bentonite layer: None observed. Average thickness of sludge bands: two primary
layers varying from 0.5 to 4 feet thick. Lower depth of sludge: about 15 feet below
surface level. Lower level of toxaphene: Concentrations of toxaphene to 245 ppm about 8
feet above mean sea level. Debris: wood, scrap metal, plastic and wood.
These observations indicate that containment cells were constructed with no, or only
token, bentonite clay liners thus leaving the sludge in direct contact with site soils.
Additionally, toxaphene materials of either raw product or high percentage solutions were
dumped at this site. As a result the ground beneath the sludge is highly contaminated and
there is toxaphene in direct contact with the surficial aquifer. Further, the cells
contain a variety of large solid objects that can interfere with the mixing processes of
in situ stabilization.
Concrete Leaching Tests
Laboratory tests on the ability of concrete to immobilize sludge in site soils were also
performed. The basic idea of in situ stabilization involves drilling deep into the
landfill with large hollow augers, penetrating to depth, and then withdrawing the auger
while pumping concrete to blend with waste and soil. The large concentrations of sludge
become diluted with site soil and concrete to reduce the overall concentration of
toxaphene. After the concrete solidifies it forms a block that inhibits erosion.
The site survey studies indicate the depths needed for immobilization and offer insight
into the types of soil conditions likely to be encountered by the auger. Information is
also needed on the best cement formulation to use with sludge and site soil.
The cement formulation was studied in the laboratory using landfill samples. These samples
were mixed under laboratory conditions with different concrete formulations and tested
first for hardness of the formed blocks. Concrete formulations that passed the hardness
test were subjected to leachability testing.
Leaching is the environmental process that degrades cement and carries toxaphene into the
water table. Water moving through or around a block dissolves a small amount of cement and
frees toxaphene for erosion. Over a period of decades the entire concrete block is
expected to slowly dissolve and release toxaphene and toxaphene byproducts in a controlled
fashion. The environment can better handle this slow release.
Unfortunately, for this project, the concrete leaching tests were plagued with problems.
Material leaches out of the toxaphene blocks at a high enough rate to interfere with
toxaphene measurements. It has been difficult for the laboratory to measure concentrations
of toxaphene below 5 parts per million. Accordingly, during much of the past year's
research, the process could not be readily validated at the 0.060 ppm target
concentration. In recent tests the cement was treated to reduce the nonspecific leaching
problem and field trials are now scheduled for testing at the landfill.
Conclusions
It is very unclear at this time if the bench studies on concrete formulations are a
true indicator of on-site performance of stabilization. Cement block manufacturing in the
laboratory under idealized conditions has generated apparent release of toxaphene. Under
field conditions where there is little control over the process it may be difficult to
reproduce the laboratory work. Also, immobilization on-site cannot simply be redone to get
it right. So far, the cement formulation work is not encouraging.
It is unclear how the numerous buried objects can be retrieved prior to stabilization.
Magnetic anomaly maps were generated that can be used for metal; but that technique would
not accurately detect cement blocks, plastic pipes, rock artifacts, or the glass and wood
debris.
Assuming the cement can be perfected to work in the environment of the landfill only cells
3, 4, and 5 appear to be good candidates for the in situ stabilization process. Finally,
it is clear that stabilization, if used, must extend much deeper than the levels estimated
in the Feasibility Study.
LINKS:
Jump to top of page
visit TAG Advisor R. K. Pegg's Homepage
Return to TAG table of Contents