EPA542-R-03-013
August 2003
Treatment Technologies for Historical Ponds
Containing Elemental Phosphorus -
Summary and Evaluation
FINAL REPORT
U.S. Environmental Protection Agency
Office of Solid Waste and Emergency Response
Washington, DC 20460
-------
DISCLAIMER
This report is intended solely as general information to the public. It is not intended, nor can it be relied
upon, to create any rights, substantive or procedural, enforceable by any party in litigation with the
United States. Furthermore, information in this report is solely based on the specific characteristics at the
EMF site and may not apply to a particular situation based upon the circumstances. The U.S.
Environmental Protection Agency reserves the right to change this report at any time without public
notice. Mention of trade names or commercial products in this report does not constitute endorsement or
recommendation for their use.
NOTICE
Preparation of this report has been funded wholly or in part by the U.S. Environmental Protection Agency
(EPA) under Contract Numbers 68-W-99-003 and 68-W-02-034. This report may be obtained from
EPA's web site at . For more information regarding this report please contact Mr. Kelly
Madalinski, EPA Office of Solid Waste and Emergency Response, at (703) 603-9901 or
ACKNOWLEDGMENT
Special acknowledgment is given to the federal and state staff and other remediation professionals for
providing information on individual sites and for sharing their expertise on treatment technologies to
support the preparation of this report.
August 2003
-------
FOREWORD
The U.S. Environmental Protection Agency (EPA) administers the Superfund program in cooperation
with individual states and tribal governments to clean up the nation's uncontrolled hazardous waste sites.
One aspect of the Superfund program is conducting long-term remedial response actions at hazardous
waste sites listed on the National Priorities List (NPL). These sites are considered to be the most
contaminated uncontrolled hazardous waste sites nationwide with the highest priority for investigation
and remediation.
The historical operations at the FMC plant area of the Eastern Michaud Flats Superfund site (EMF Site),
located near Pocatello, Idaho, managed wastes generated during the manufacturing of elemental white
phosphorus (WP) in ponds at the site. The site was added to the NPL in August 1990 and a Record of
Decision was signed in 1998 that specified capping 16 historical ponds that contained waste contaminated
with WP, metals, and radionuclides. WP is a highly toxic and reactive contaminant and limited
information is available about the treatment of WP-containing wastes. Following concerns raised by local
stakeholders about the capping remedy, EPA committed to re-evaluate treatment technologies potentially
applicable at the EMF Superfund site.
This report includes information about six technologies that may be applicable for the treatment of
contaminated soil and sludge in the historical ponds at the EMF site. Information presented in this report
is specific to the EMF site; therefore, it may not be applicable to other WP-manufacturing sites.
Furthermore, the report is not intended to recommend any specific remedial approach for the EMF site or
other sites. It does not consider many other factors, such as risk reduction to human health and the
environment, that are critical in choosing a remedial approach for a site.
Because limited information was available about the historical ponds at the EMF site, assumptions using
best engineering judgment were made about the characteristics of the waste. Information provided in this
report about the applicability of specific technologies to treat the soil and sludge in the historical ponds is
based on those assumptions. Costs presented in this report are estimates only, and are also based on those
engineering assumptions.
This report is made available by EPA to provide additional information on potentially applicable
technologies for the treatment of soil and sludge contaminated with WP, metals, and radionuclides.
ii August 2003
-------
TABLE OF CONTENTS
Section
1.0 INTRODUCTION
2.0 PROPERTIES OF ELEMENTAL PHOSPHORUS 12
3.0 OVERVIEW OF ELEMENTAL PHOSPHORUS ELECTRIC ARC FURNACE
MANUFACTURING SITES 15
4.0 OVERVIEW OF HISTORICAL PONDS AT EMF 27
5.0 TREATMENT TECHNOLOGIES FOR SOIL AND SLUDGE IN HISTORICAL PONDS . . 42
6.0 SUMMARY AND DISCUSSION OF FINDINGS 72
7.0 REFERENCES 77
ADDITIONAL SOURCES OF INFORMATION 80
TABLES
Table Page
1-1 Eastern Michaud Flats Superfund Site - Characteristics of Historical Ponds 4
2-1 Physical Properties of Elemental White Phosphorus 12
2-2 Factors Affecting Oxidation Rate 13
3-1 Summary of Information About WP Electric Arc Furnace Manufacturing Sites 16
3-2 Summary of Information About Selected Military Facilities Using Elemental Phosphorus ... 24
4-1 Soil Boring Results for Surface (less than 20 ft) of Unlined Ponds 35
4-2 Summary of Soil Characteristics Noted in Soil Boring Activities 40
4-3 Average Range of Concentrations for Selected Constituent in Historical Ponds 41
5-1 Comparison of Contaminants in Pond Materials 53
5-2 Analytical Results from HSAD Test of Astaris Sludge 62
A-l Contaminants of Concern and Corresponding Concentrations at SMC (U.S. EPA, 1999) . . . A-2
A-2 Past Studies at Tarpon Springs Site (U.S. EPA, 1999) A-5
FIGURES
Figure Page
1-1 Location of Historical Ponds at Astaris Facility (FMC, 1996) 3
iii August 2003
-------
APPENDICES
Appendix
Appendix A Summary of Stauffer Chemical Company, Tarpon Springs, FL, Planned Remediation of
Unlined Ponds Using in Situ Solidification/Stabilization
iv August 2003
-------
, Executive Summary
EXECUTIVE SUMMARY
This report provides a summary and evaluation of available information about technologies that have the
potential to treat soil and sludge in historical ponds contaminated with elemental white phosphorus (WP),
heavy metals, and radionuclides at the Eastern Michaud Flats (EMF) Superfund site located near
Pocatello, Idaho. These ponds were used by Astaris Idaho LLC (previously FMC Corporation) during the
manufacture of WP from phosphate ore, and received the following wastes: phossy water, precipitator
dust slurry, phossy solids, slag pit water and solids, and residuals from reclaiming WP in other ponds.
The site was added to the National Priorities List in August 1990 and a Record of Decision (ROD) was
signed in June 1998 that included capping 16 historical ponds at the site. Based on concerns raised about
the decision to cap the historical ponds, the Technology Innovation Office (now the Technology
Innovation Program) of the U.S. Environmental Protection Agency (EPA), Office of Solid Waste and
Emergency Response, was requested to evaluate technologies potentially applicable for the treatment of
soil and sludge in the historical ponds that contain WP, heavy metals, and radionuclides.
The scope of this report is limited to a summary and review of available information from the technical
literature and previous studies about the following:
The 16 historical ponds identified in the EMF ROD - Ponds 1S-7S, 9S, 10S, and 1E-7E
Soils and sludges in the historical ponds that are not physically covered by ponds under authority
of the Resource Conservation and Recovery Act
Treatment technologies that have been used or show promise for treatment of WP based on
available information
Containment technologies (such as capping) or use of off-site treatment, storage, or disposal facilities
(TSDFs) were not examined for this report, although they are discussed as they relate to treatment
technologies. The report does not provide recommendations regarding selection of technologies or
evaluate non-treatment options (e.g., institutional controls), nor does it compare the costs and benefits for
use of any treatment technology or remedial option. That is, the report does not evaluate the treatment
technologies using the nine criteria from the Superfund program which includes all factors (e.g.,
applicable requirements, stakeholder acceptance) necessary to evaluate a remedy. As such, this report is
not a feasibility study for remediation of the historical ponds at the EMF site.
ES-1 August, 2003
-------
, Executive Summary
Treatment technologies were identified and evaluated based on (1) available information in the technical
literature, including previous studies, and work conducted at similar sites; (2) technology performance
data; (3) technology application considerations, including technical judgment using information from
previous studies of the EMF site and experience in waste site remediation; and (4) cost to implement the
technology.
Overall Findings
Six technologies were identified as potentially applicable: solidification/stabilization (S/S); caustic
hydrolysis; chemical oxidation; mechanical aeration; incineration; and thermal desorption. The following
is a summary and discussion of the findings in this report.
No technology has been used at full-scale to treat waste material similar to that found at EMF.
Only limited information is available in the literature covering the remediation of WP, with fewer than 10
studies identified which discuss the potential use of treatment technologies for WP. This primarily
includes studies performed at WP manufacturing facilities and at military facilities. While technologies
have been used to treat WP in bench- and pilot-scale studies, no technologies were identified as having
been used for full-scale treatment.
No new treatment technologies have emerged as potentially applicable since the EMF Feasibility
Study. The six technologies in this report were identified in the Feasibility Study report prepared in
1996.
Other WP manufacturing facilities primarily used capping as the remedy for similar waste. Eight
other WP manufacturing sites were identified that have similar contaminated historical ponds as the EMF
site. Six of the eight sites have installed or plan to install caps. For the two remaining sites, one (Rhodia,
Silver Bow, Montana) indicated that the ponds are not under corrective action, and the other (Stauffer,
Tarpon Springs, Florida) is evaluating a remedy of in situ S/S (see discussion below).
Minimal performance data currently exist for use of the six technologies to treat similar waste
material as found at EMF. Performance data were identified for treatment of WP using chemical
oxidation, mechanical aeration, and incineration. However, these data are not for treatment of wastes in
historical ponds at a WP manufacturing facility. Performance data for chemical oxidation and mechanical
ES-2 August, 2003
-------
, Executive Summary
aeration are for work at bench- and pilot-scale, while data for incineration are for ordnance wastes with a
higher percentage of WP than found in the historical ponds.
Thermal desorption was used for the treatment of WP in contaminated soil. Recent attempts to obtain
specific information on the project revealed that the technology vendor was sold to another company.
Personnel at this company were not familiar with the current availability of the technology. Therefore, it
is unknown if the WP-contaminated soil was similar to waste material as found at EMF and no specific
performance data were available for review.
No performance data were identified for treatment of WP using S/S and caustic hydrolysis. Both
technologies have been considered for the treatment of similar waste material at WP manufacturing sites.
The Stauffer site in Tarpon Springs, Florida is planning to test in situ S/S in 2003. Although the
information from this test program could be used to evaluate the effectiveness of S/S at EMF, the test
program at Tarpon Springs may not be an accurate predictor of performance on the EMF waste. Reasons
for this include: the type of phosphate ore used at Tarpon Springs is different from that used at EMF,
elemental phosphorus is not a primary contaminant at Tarpon Springs, and the scale of the Tarpon
Springs site is smaller. Caustic hydrolysis was considered for use at the Rhodia site in Silver Bow,
Montana. In addition, caustic hydrolysis was identified as the Land Disposal Restrictions treatment
technology for process wastes at EMF, but construction of this technology was halted with the plant
shutdown.
Additional testing would be necessary to assess whether treatment technologies could perform
adequately across a range of contaminant concentrations and properties of the waste material as
found at EMF. Limited site characterization data are available and the historical ponds are assumed to
be heterogeneous in physical and chemical composition. Therefore, extensive site assessment and
treatability testing would be needed to verify the potential for any technology to treat the soil and sludge
at EMF. Treatability tests would include evaluating how the technology would perform for the specific
matrices in the different ponds, and the variations in performance across the range of concentrations and
physical properties.
Although the six technologies are at various stages of commercial development, the technologies would
all require testing to establish that they could perform reliably for the waste material in the historical
ponds. S/S, chemical oxidation, incineration, and thermal desorption have been applied commercially at
ES-3 August, 2003
-------
, Executive Summary
full-scale for site remediation, but have not been used to treat WP pond material. Caustic hydrolysis and
mechanical aeration have not been used extensively for site remediation, and significant developmental
testing would be entailed for scale-up along with treatability testing. Developmental and treatability
testing for the six technologies would require additional time and resources to undertake.
A series of technologies may be necessary to collectively treat all the types of contaminants. The soil
and sludge in the historical ponds contain multiple types of contaminants which all may require treatment.
For example, incineration, thermal desorption, mechanical aeration, and chemical oxidation show
potential to treat WP, but would not be able to treat heavy metals or radionuclides. In these cases, an
additional treatment process would likely be needed, such as S/S, using what is often referred to as a
"treatment train".
Of the six technologies, only S/S and caustic hydrolysis have the potential to be effective for
treatment of heavy metals and radionuclides. S/S is applied frequently at full-scale to reduce the
mobility of heavy metals (radionuclides are expected to behave in a manner similar to heavy metals) at
contaminated sites. Caustic hydrolysis would convert heavy metals to metal oxides and hydroxides,
which generally are less soluble than the metal compounds, and could be removed by filtration or settling
processes.
Residuals from treatment, such as solid, liquid, or gaseous materials, would require further
management. Residual management may include characterizing and transporting these residuals to a
storage or disposal facility (on- or off-site), or performing further treatment (such as for off gases) prior to
release to the environment.
Soil and sludge may require preprocessing to homogenize the material for use by treatment
technologies. Preprocessing may include crushing, grinding, or milling, to break up large masses of soil
and sludge. For both ex situ and in situ technologies, preprocessing may be necessary depending on the
distribution of contaminants in the ponds and the methods used to implement the treatment technologies.
The estimated volume of waste material to treat (500,000 cubic yards) would entail a large
remediation project, including significant engineering issues. The physical layout of the site, where
historical ponds are located near RCRA ponds, structures, and slag piles, may impact the implementation
ES-4 August, 2003
-------
, Executive Summary
of a treatment technology, including the need for space to stage equipment or to store material before or
after treatment.
Site workers would need to follow stringent health and safety precautions about handling soil or
sludge containing WP. WP is an inorganic compound that ignites spontaneously in warm air. It is toxic
by ingestion and inhalation and skin contact with WP causes burns. Site workers would likely need to
use Level C personal protective equipment (respiratory and skin contact protection) when conducting
work on the soil or sludge in the historical ponds. In addition, health and safety precautions related to
metals and radionuclides would also have to be considered.
The cost to implement any of the six treatment technologies would be high, based on the criteria
used to identify high cost projects by EPA's National Remedy Review Board (NRRB). The NRPJ3
identifies high cost remedial actions as those that cost more than $30 million or more than $10 million
and 50 percent greater in cost than the least costly cleanup alternative.
Although the technology cost estimates in this report could be above or below the actual costs, the actual
total treatment costs are likely to be higher. Specifically, the technology costs estimates do not include
costs for associated project components, such as excavation (if required), preprocessing of waste material,
health and safety (such as ambient gas control), and residual management, which could be integral parts
of a remediation project at the EMF site using any of the six technologies.
Important Considerations
For this report, several assumptions were made about pond characteristics based on what is known about
the ponds and experience with site cleanups. Therefore, the evaluations put forth for each technology are
dependent on these assumptions and limited available characterization data. Should the information on
the characteristics of the historical ponds change, the analysis of the treatment technologies provided in
this report may also need to be modified.
Quantitative cleanup levels have not been identified for WP, heavy metals, or radionuclides in the soil
and sludge in historical ponds at the EMF site. Therefore, treatment technologies could not be evaluated
against their capability to reach specified cleanup levels at the site. This report discusses technology
ES-5 August, 2003
-------
, Executive Summary
performance in general terms, focusing on the technology's ability to reduce the concentration, volume,
or mobility of a mixture of contaminants in a heterogeneous soil and sludge matrix.
The level of uncertainty for the cost estimates provided in this report to implement the six technologies at
the EMF site is high. The estimates presented in this report represent a range of possible costs for the
treatment of the EMF waste. This range is highly dependent on the assumptions used to characterize the
historical ponds and the specific design and operating conditions used for implementing the technology.
It should be noted that historical cost data of technology applications indicate costs are highly variable,
impacted by many factors, and that those factors are site-specific.
ES-6 August, 2003
-------
. Introduction
1.0 INTRODUCTION
The Technology Innovation Office (now the Technology Innovation Program) of the U.S. Environmental
Protection Agency (EPA), Office of Solid Waste and Emergency Response (OSWER), has prepared this
report to summarize and evaluate information about the potential use of existing or new technologies to
treat soil and sludge in historical ponds contaminated with elemental white phosphorus (WP) at the
Eastern Michaud Flats (EMF) Superfund site located near Pocatello, Idaho.
The intended audience for this report is the stakeholders for the EMF site, including EPA regional, state,
Shoshone-Bannock tribal representatives, and other interested parties. The report addresses issues
specific to treatment of historical ponds containing WP, while providing relatively little overall
background information about hazardous site remediation.
Background
The EMF Superfund site (EPA ID no. IDD98466610) is located in southeastern Idaho approximately 2.5
miles northwest of Pocatello, Idaho. The site includes two adjacent phosphate ore processing plants - the
J.R. Simplot Company Don Plant (Simplot) and the Astaris Idaho LLC (now owned by FMC, LLC Idaho)
Elemental Phosphorus Plant (Astaris facility). The Astaris facility processes phosphate ore into WP, and
the historical ponds at this facility are the subject of this report. Production of WP at the Astaris facility
has recently shut down permanently.
The Astaris facility began operation in the 1940s. From 1954 to 1981, phossy water, precipitator dust
slurry, and other wastes from WP manufacturing were disposed of in ponds at the site. The soil and
sludge in these ponds is contaminated with WP; heavy metals, such as antimony, arsenic, beryllium,
boron, cadmium, lead, manganese, mercury, nickel, selenium, silver, thallium, vanadium, and zinc; and
radionuclides that contribute to gross alpha and gross beta contamination. Groundwater at the site was
found to have elevated concentrations of major ions and arsenic, lithium, manganese, total phosphorus,
selenium, boron, barium, cobalt, and fluoride. No significant concentrations of organic compounds were
detected in soil or groundwater at the site. (FMC, FS, 1996)
The EMF site was added to the National Priorities List (NPL) in August 1990 and the Remedial
Investigation and Feasibility Study (RI/FS) were performed between 1991 and 1996. A Record of
1 August, 2003
-------
. Introduction
Decision (ROD) was signed in June 1998 that included capping the 16 historical ponds with a soil cover
of at least 12 inches.
In addition, the ROD provided an overall summary of site risks in terms of human health and ecological
receptors, including potential release mechanisms and routes of exposure. The release mechanisms were
identified as infiltration/percolation and exposure to media such as groundwater and homegrown produce,
meat, and dairy products. Routes of exposure were identified as ingestion or dermal contact to nearby
residents. For two of the historical ponds, air emissions were identified as a potential release mechanism,
with inhalation as the route of exposure. (EPA, 1998) The risk to human health and the environment
posed by WP is due to either direct contact with the waste or exposure of the waste to oxygen or water,
which could result in the generation of hazardous by-products that could result in direct exposure or
inhalation. There has been no indication that WP is migrating from the surface impoundments or the
historic pond areas into groundwater. (EPA, 2002)
The 16 historical ponds, shown in Figure 1-1, consist of Ponds 1S-7S, 9S, 10S, and 1E-7E. All 16 ponds
are being addressed under the Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLA). Table 1-1 summarizes available information about these ponds, documented in the RI report.
(FMC, 1995), FS report (FMC, 1996), and ROD (EPA, 1998) This table includes a description of the size
of the pond, the years of active operation, types of materials disposed, cover materials, estimated volume
of "phossy" waste, and other information.
The historical ponds range in size from 0.5 to 10.4 acres per pond, with atotal areal extent of 52.3 acres.
Fifteen of the ponds are unlined; one pond, Pond 10S is lined. As shown in Figure 1-1, nine of the
historical ponds have been partially or fully buried by the Resource Conservation and Recovery Act
(RCRA)-regulated lined ponds or ferrophos storage areas. These include Ponds 3S, 6S, 7S, IE, 2E, 3E,
5E, 6E, and 7E. The estimated area of these historical ponds that have been covered is 34 acres (about 70
percent of the original total surface area of the historical ponds).
The stratigraphy of the site is generally described as discontinuous layers of unconsolidated sediments
deposited on an erosional surface that was incised in volcanic bedrock. Deposits of windblown silt
(loess) and a colluvial silt mantle of variable thickness covers the study area. The loess layer ranges from
2 to more than 100 feet (ft) thick and is calcareous. In the vicinity of the ponds the thickness of the loess
is approximately 30 ft or more. The native soils are generally alkaline (pH of 7 or greater) because of
2 August, 2003
-------
. Introduction
Figure 1-1. Location of Historical Ponds at Astaris Facility (FMC, 1996)
. = Scrubber
, Calciners Overflow
Chemical Lab
Drainpil —-
Legend
FMC Property Line
County Boundary Line
Fort Hall Indian Reservation
Boundary
Union Pacific Railroad
Plant Structures
Storage Areas
Presently in Use
Storage Areas
Capped or Covered
Inactive Storage Area
Wastewater Treatment Plant
industrial Wastewater
1 ooo aooo ft
gH
4QO
Caiciner
Pond Solids
(soil cover)
EASTERN MICHAUD FLATS
POCATELLO, IDAHO
FEASIBILITY STUDY
FMC FACILITY PLAN
EXISTING AND FORMER FACILITIES
Kinport
Substation
Site I
August 2003
-------
. Introduction
Table 1-1. Eastern Michaud Flats Superfund Site - Characteristics of Historical Ponds
Pond
IS
2S
3S
4S
5S
6S
Size
(acres)
0.5
0.8
1.2
0.8
1.0
2.3
Active Life
1954 to Oct.
1961
1955 to Oct.
1961
Nov. 1961 to
June 1965
Apr. 1966 to
Mar. 1967
July 1965 to
Mar. 1967
Apr. 1967 to
Feb. 1969
When Pond
Dried/
Backfilled
1972
1972
Dec. 1976
July 1976
Mar. 1976
July 1976
Material Disposed
in Pond
Phossy water and
phossy solids
Phossy water and
phossy solids
Precipitator dust slurry;
slag pit water and
solids; phossy water and
phossy solids; residuals
from P4 reclaim
operation on ponds 1 S
and 2S and east end of
3S
Precipitator dust slurry
Phossy water and
phossy solids
Precipitator dust slurry;
some phossy water and
phossy solids in NE
corner
Cover Material(s)
Slag, soil
Slag, soil
Capped with 3 ft of soil,
then covered with
crushed slag
Capped with 3 to 6 ft of
soil
Capped with baghouse
dust; precipitator dust
slurry; fluid bed drier
product prills and dust;
slag; final soil cap on
top
Capped with soil; south
end partially filled with
slag and paved with
asphalt for use as a new
slag haul road
"Phossy" Waste
Volume
(cubic yards)
2,400
875
8,800
6,500
8,500
24,600
Other Notes
Initially hauled in slurry truck;
pipeline installed in 1957. P4 was
reclaimed to plant twice per year from
1966-1972.
Assumed not covered.
P4 was reclaimed to plant twice a year
until September 1965. P4 continued to
be reclaimed to plant twice per year
from 1966-1972.
Assumed not covered.
Settled solids were routinely dug out
twice a year until 1965. P4 in east end
was reclaimed in 1972-1976;
approximately 100 ft of east end was
filled with slag after reclaiming; this
area is not capped as is the rest of the
former pond.
Assumed 50% covered.
Assumed not covered.
Very difficult to dry because of
pyrophoric contents; fine solids would
not support cover weight.
Assumed not covered.
New slag haul road over south end.
Assumed 50% covered.
August 2003
-------
. Introduction
Table 1-1. Eastern Michaud Flats Superfund Site - Characteristics of Historical Ponds (continued)
Pond
7S
9S
10S
IE
2E
3E
Size
(acres)
3.6
4.0
1.0
1.9
3.3
10.4
Active Life
Mar. 1969 to
Sept. 1970
1971 to 1974
(?)
Not
identified
Apr. 1965 to
Fall 1982
Apr. 1965 to
Oct. 1967
May 1967 to
Sept. 1970
When Pond
Dried/
Backfilled
Jan. 1980
Nov. 1980
Not
identified
Oct. 1980
1977
1980
Material Disposed
in Pond
Precipitator dust slurry
with phossy hot spots
Precipitator dust slurry;
slag pit water and
solids. Material dried
and sold
Fluid bed dryer slurry
Phossy water and
carryover fine solids
from upstream ponds;
precipitator dust slurry
and dried slurry;
Material dried and sold
Phossy water and
carryover fine solids
from upstream ponds
Phossy water and
carryover fine solids
from upstream ponds
Cover Material(s)
Two high - P4 areas
capped with cement;
entire area capped with
6 to 10 ft of pit-run slag,
then 3 ft of soil
Not capped
Capped
Not capped
Site is beneath current
Phase IV ponds (8E)
Site is beneath current
Phase IV ponds (11 S-
14S)
"Phossy" Waste
Volume
(cubic yards)
18,200
0(3)
Not identified
9,000
0(1)
0(1)
Other Notes
New slag haul road over south end;
this site is not byproduct
ferrophosphorus stockpile,
approximately 25 ft high.
Assumed 50% covered.
Contents were dried in place and about
20 to 25 ft dug out for outside sales;
small quantity remains in place.
Assumed not covered.
Lined pond.
Assumed not covered.
Filled with dredged precipitator dust
slurry from fluid bed drier surge pond
in fall of 1982.
Assumed 90% covered.
Site was used for storage of
precipitator slurry fluid bed drier
product, then dug out for lined pond
8E construction in 1984; residual
precipitator dust sent to 4E site. Some
material was removed and sold.
Assumed 90% covered.
Contents dug out for construction of
new lined ponds in 1980; this site now
occupied by lined ponds US, 12S,
13S, and 14S.
Assumed 100% covered.
August 2003
-------
Table 1-1. Eastern Michaud Flats Superfund Site - Characteristics of Historical Ponds (continued)
. Introduction
Pond
4E
5E
6E
7E
Size
(acres)
1.8
6.6
6.7
4.3
Active Life
May 1967 to
1980
Apr. 1968 to
1972-73(7)
Nov. 1968 to
1980-81
Dec. 1969 to
1980-81
When Pond
Dried/
Backfilled
Oct. 1980
1981
1981
1981
Material Disposed
in Pond
Phossy water and
carryover fine solids
from upstream ponds;
precipitator dust slurry
overflow
Phossy water and very
minor carryover fine
solids from upstream
ponds
Same as 5E
Received phossy water
only a few seasons; no
solids observed in 7E
Cover Material(s)
Not capped
Site is beneath current
Pond 15S
Same as 5E
Not capped
"Phossy" Waste
Volume
(cubic yards)
29,000
0(1)
0(1)
0(2)
Other Notes
Received precipitator slurry from fluid
bed drier slurry pond in fall of 1982.
Some material removed and sold.
Assumed not covered.
Dried gray settled soil (4" to 6")
placed in area just south of new 15 S
lined pond. New lined pond 15S was
built on this site in 1982.
Assumed 95% covered.
Same as 5E.
Assumed 95% covered.
Eastern ± 150 ft used for construction
of lined pond 15S (1982) and 9E
(1986).
Assumed 50% covered.
(1) Material was removed and placed in another pond or sold. New ponds were constructed over all or part of the area occupied by this pond.
(2) Most of the pond was removed during construction of another pond (Pond 15S).
(3) Pond 9S was excavated and used as a storage area for precipitator slurry solids.
Sources: FMC RI; FMC FS (Table 4.1-1), 1996; ROD (Table 1) 1998.
August 2003
-------
. Introduction
their calcareous nature, and have little structure. Depths to groundwater in the shallow aquifer range from
170 ft below ground surface (bgs) in the Bannock Range area to 55 ft bgs in the Michaud Flats area, and
less in the area near EMF. The site is located in a semi-arid region, with approximately 11 inches of total
precipitation a year. (FMC, FS, 1996)
Scope
Representatives of the Shoshone-Bannock Tribes, one of the stakeholders for the EMF site, have raised
concerns about the decision to cap the historical ponds. Following a meeting with the Tribes and a letter
by EPA's OSWER Assistant Administrator, the Technology Innovation Program committed to work with
EPA Region 10 and the Tribal Business Council to evaluate technologies potentially applicable for the
treatment of historical ponds that contain WP.
The scope of this report is limited to a summary and review of available information from the technical
literature and previous studies about the following:
The 16 historical ponds identified in the EMF ROD - Ponds 1S-7S, 9S, 10S, and 1E-7E
• Soils and sludges in the 16 historical ponds that are not physically covered by RCRA ponds
Treatment technologies (commercially-available and not commercially-available) that have been
used or have shown promise for treatment of WP
For each treatment technology, this report provides an evaluation of how it could potentially be applied
for on-site treatment of WP as well as for heavy metals and radionuclides in the historical ponds, and
discusses its strengths and limitations, performance, costs, preprocessing requirements, material removal
and handling (if necessary), residual management, status of development, and prior use on WP. No
testing of treatment technologies (e.g., in a laboratory or at a pilot-scale) was performed as part of the
preparation of this report.
This report does not address the following:
Containment technologies (e.g., capping) or other non-treatment remedies (e.g., off site disposal)
Reclamation or reuse of pond materials, such as with Astaris production facilities
Using a combination of technologies to collectively treat all the types of contaminants
Recommendations regarding selection of technologies (i.e., which technologies would be most
appropriate for the EMF site)
Analysis of risk to human health and the environment posed by the historical ponds and how a
technology or series of technologies would reduce risk
7 August 2003
-------
. Introduction
As these items are necessary components to the complete evaluation of the effectiveness of any site-
specific remedy, this report is not intended to be a feasability study for remediation of the historical ponds
at the EMF site.
Overview of Methodology
Information about the properties of WP, heavy metals, and radionuclides, the types of activities
performed at other WP manufacturing plants, and the characteristics of the historical ponds at the EMF
site was reviewed. Based on this information, technologies potentially applicable to the EMF site were
identified and evaluated.
Because the properties of WP are not widely understood, this report includes a brief summary of WP
physical and chemical properties, as well as health and safety concerns with human exposure. The
properties of heavy metals and radionuclides are widely known and well documented in technical
literature (for example, see fact sheets provided by EPA's Office of Groundwater and Drinking Water at
, EPA's Technology Screening Guide for Radioactively
Contaminated Sites at ,
and the Radiation Effects Research Foundation at ), and
therefore specific considerations related to their treatment are not described in this report.
At least ten manufacturing facilities across the U.S. that produce WP have been identified, and eight of
those sites have used unlined ponds for disposal of phossy water and other wastes. Information about
these sites provide background about historical ponds at WP manufacturing sites. While the basic process
for manufacturing WP from phosphate ore did not vary substantially among facilities, the characteristics
of the phosphate ore as well as end-of-pipe treatment varied among regions and sites, resulting in
differences in contaminants and concentrations in ponds between them. For example, the ores in Idaho
are different from those in Tennessee or Florida, yielding different types and concentrations of WP, heavy
metals, and radionuclides.
To identify treatment technologies for the historical ponds at the EMF site, it was necessary to develop a
conceptual understanding of the characteristics of soil and sludge in those ponds, as well as the fate and
transport of WP in the ponds, including transformation processes such as oxidation and hydrolysis.
Available data were combined with information available for the RCRA ponds, as well as assumptions
8 August 2003
-------
. Introduction
about the physical and chemical characteristics of the ponds, to develop a conceptual understanding of the
historical ponds.
Data Sources
Sources of information used for this report included files from EPA Region 10, including the ROD and
the RI/FS for EMF, and correspondence with key staff in EPA regional offices and states, and file
reviews. An open literature search of the Online Computer Library Center Inc. (OCLC) was conducted in
October 2001, using the following key words: white phosphorus, elemental phosphorus, and remediation.
A complete list of references used in preparation of this report is provided in Section 7.0.
Fewer than 10 studies were identified concerning the use of treatment technologies for WP in soil and
sludge, either at other WP manufacturing facilities or in other applications. Some of this work was done
in support of WP manufacturing facilities and some in support of military operations (WP is used in
certain types of military ordnance). Much of that work was limited in scope, and follow-on work/studies
were not identified. No studies were identified from international sources, although an exhaustive search
was not conducted.
In the mid-1990s, the U.S. Army Corps of Engineers (USAGE) prepared a report, "Summary and
Evaluation for White Phosphorus Remediation: A Literature Review", Technical Report IRRP-96-7, that
summarized available information about remediation of WP. (Rivera, 1996) The report included bench-
and pilot-scale data for six treatment technologies and two non-treatment remedial technologies, primarily
for use on WP in military operations. At the time of the report, the U.S. Army had identified WP as a
contaminant of concern in soil and water at military training and munitions production facilities.
However, the Army since has down-graded its level of concern with WP, and the Army researchers who
prepared the USAGE report did not perform any follow-up actions to further evaluate treatment
technologies for WP.
A search of available documents (e.g., reports, articles, and conference proceedings), including the results
of the search performed using the OCLC, showed that most of the studies were performed in the early
1990s. The only recent work (since 1996) was that performed by the U.S. Army at a Superfund site in
Alaska (Eagle River Flats) where there were waterfowl deaths because of WP contamination in pond
August 2003
-------
. Introduction
sediments. At that site, the Army evaluated use of pond draining as a means to reduce the concentration
of WP in sediments.
A search for technology vendors who have demonstrated treatment for WP was conducted in February
2002 using EPA's REACHIT online database. EPA REACHIT, available at ,
provides information about more than 1,200 technologies and 1,600 sites where treatment technologies
have been used. The search revealed two vendors who claimed to have treated WP. However, after
reviewing the information, the two vendors likely performed treatment of phosphate compounds instead
ofWP.
Important Considerations
Available data provide limited information about the nature and extent of contamination and of the
physical characteristics of the historical ponds at the EMF site. Data are available from 14 soil borings in
and near 7 disposal ponds. Therefore, several assumptions were made about the physical and chemical
characteristics of the historical ponds based on what is known at the EMF site and experience with site
remediation. Should further work change the information on the pond characteristics, the assumptions
and evaluations of treatment technologies provided in this report would also likely need to be modified.
Quantitative cleanup levels have not been identified for WP, heavy metals, or radionuclides in the soil
and sludge in historical ponds at the EMF site. Therefore, treatment technologies could not be evaluated
against their capability to reach specified cleanup levels at the site. Since cleanup levels have not been
identified, this report discusses technology performance in general terms, focusing on the technology's
ability to reduce the concentration, volume, or mobility of a mixture of contaminants in a heterogeneous
soil and sludge matrix.
As presented in this report, the uncertainty associated with cost estimates discussed in this report for the
six technologies is high. The cost estimates include only treatment costs, not associated project
components, such as excavation, preprocessing of waste material, health and safety, and residual
management, which likely would significantly increase total project costs. It should be noted that
historical cost data of technology applications indicate costs are highly variable, impacted by many
factors, and that those factors are site-specific.
10 August 2003
-------
. Introduction
Report Organization
The remainder of this report is organized as follows:
Section 2 provides a brief summary of the properties of WP, including its physical and chemical
properties, and health and safety concerns
Section 3 discusses activities at WP manufacturing sites other than the EMF site
Section 4 presents a conceptual description of the historical ponds at the EMF site, including the
fate and transport of WP in the historical ponds
Section 5 describes the types of technologies that may potentially be applicable to treatment of
historical ponds, including an overview of attributes necessary for any applicable treatment
technology
Section 6 provides an overall summary and discussion of the findings in this report
Section 7 lists the references used in the report preparation
11 August 2003
-------
Properties of Elemental Phosphorus
2.0 PROPERTIES OF ELEMENTAL PHOSPHORUS
This section provides an overview of the physical and chemical properties of elemental phosphorus and
provides a brief summary of its effects on human health and safety.
Physical Properties
Elemental phosphorus exists in three distinct configurations called allotropes, all with the same molecular
weight (123.89), but each differing significantly from the other allotropes in physical and chemical
characteristics. The chemical formula for all allotropes of elemental phosphorus is P4; however, they
have different names according to their respective colors, including black, red, violet, and white
(sometimes called yellow due to impurities). The EMF facility was involved exclusively in the
production of the white (or yellow) allotrope of phosphorus; therefore the remainder of this section will
not discuss the other allotropes. WP is a waxy solid that may be colorless, white, or yellow, and has a
garlic-like odor. Table 2-1 provides a summary of some of the physical properties of WP, and shows that
it is relatively insoluble in water, with a solubility of 3 mg/L at 15 °C.
Table 2-1. Physical Properties of Elemental White Phosphorus
Physical Property
Chemical Formula
Appearance
Boiling Point
Density
Flash Point
Melting Point
Molecular Weight
Solubility
Vapor Pressure
Volatility
Characteristic
P4
White (sometimes colorless or yellow) waxy solid
280.5°C
1.82g/cm3at20°C
Spontaneously ignites in air above 30°C
(moist air)
44. FC
123.89
Water -3 mg/L at 15°C
0.026 mm Hg at 20°C
May be ignited by heat, sparks, or flame
Sources: Rivera, 1996; Van Wazer, 1972.
Chemical Properties
The primary processes for chemical transformation of WP are oxidation and hydrolysis. In a solid phase
such as soil, WP oxidizes spontaneously with oxygen in air (when the concentration of WP is • 4,000
12
August 2003
-------
Properties of Elemental Phosphorus
mg/kg) to form phosphorus pentoxide (P4O10, commonly expressed as P2O5), which exists as a participate
at ambient conditions. Phosphorus pentoxide has a strong affinity for water and will react with water
(hydrolyze), including moisture from the atmosphere, to form various phosphorus acids. The primary by-
product from oxidation of WP is orthophosphoric acid. (FMC, Attachment A, Not Dated)
In pure water, dissolved WP is rapidly oxidized by dissolved oxygen (DO) to form various forms of
soluble phosphorus acids, including H2PO4", HPO4"2, and PO4"3. In water with other dissolved ions, and
depending on environmental conditions such as pH and Eh, these acids may be further converted to a
solid metal phosphate compound such as calcium phosphate. The rate of phosphorus oxidation in water is
governed by the form of the phosphorus (dissolved or suspended), DO concentration, salt concentration,
metal ion concentration, pH, and temperature. As shown in Table 2-2, the oxidation rate increases with
increasing concentrations of dissolved phosphorus, DO, metal ion, pH, and temperature.
Table 2-2. Factors Affecting Oxidation Rate
Factor
Increased proportion of suspended elemental phosphorus
Increased proportion of dissolved elemental phosphorus
Increased dissolved oxygen concentration
Increased salt concentration
Increased metal ion concentration
Increased pH Value (>6)
Increased temperature
Affect on Rate of Oxidation
Decreased
Increased
Increased
Decreased
Increased
Increased
Increased
Source: Rivera, 1996.
WP also is hydrolyzed in water to form phosphine (PH3) and lesser amounts of phosphorus acids.
Phosphine is a toxic gas that has a low solubility, and thus is expected to migrate from the water to the
air; the portion of phosphine that dissolves is generally oxidized to form the above-mentioned forms of
phosphorus acids. The rate of hydrolysis of WP is enhanced by an increase in the pH of the water
reacting with the WP. (USAGE, 1996)
Health and Safety Concerns
WP is a toxic inorganic substance, and when ignited, gives off an irritating smoke. A fatal dose of WP
ranges from 1 to 16 mg of WP per kg body weight, depending on how the phosphorus is ingested. The
permissible exposure limit for a workplace is 0.1 mg/m3. When inhaled, it is not known whether the
13 August 2003
-------
Properties of Elemental Phosphorus
contaminant will enter directly into the blood stream. Ingestion of WP may cause abdominal pain,
jaundice, liver damage, kidney damage, eye damage, shock, coma, and death. The most common effect
of chronic WP poisoning is necrosis of the mandible (deformation of the lower jaw) commonly referred
to as "phossy jaw" followed by damage to the teeth. Chronic poisoning from long-term absorption,
particularly through the lungs and through the gastrointestinal tracts, leads to a generalized form of
weakness accompanied by anemia, loss of appetite, diarrhea, and pallor. Seriously affected bones may
become brittle, leading to spontaneous fractures. WP also is very damaging to the eyes. Some of the
symptoms that have been reported following inhalation of WP are photophobia with myosis, dilation of
pupils, retinal hemorrhage, congestion of blood vessels, and, in rare instances, optic neuritis (a lesion of a
nerve). (U.S. Army Center for Health Promotion and Preventive Medicine, 1965)
Because WP ignites spontaneously in warm air and could cause severe damage if in contact with the skin,
burns are the most common hazard. Workers commonly maintain a water blanket or slurry form of
phosphorus-containing materials to limit its exposure to air. WP reacts with carbon dioxide at elevated
temperatures to produce phosphorus pentoxide and carbon monoxide; thus, carbon dioxide extinguishers
should not be used to extinguish WP fires.
Additional information about the toxicological and health effects of WP are provided in Toxicological
Profile for White Phosphorus. (ATSDR, 1997)
14 August 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
3.0 OVERVIEW OF ELEMENTAL PHOSPHORUS ELECTRIC ARC FURNACE
MANUFACTURING SITES
This section provides a summary of information about ten WP electric arc furnace (EAF) manufacturing
sites and remedies for historical ponds containing WP at those sites. Table 3-1 provides a summary of
information about these facilities (EMF is discussed in Section 4.0), including site name and location, the
type of contaminants and characteristics of the historical (unlined) ponds, remedy selection, other
remedies considered, health and safety issues, current status, regulatory background, points of contact,
and sources of information.
In addition, information is provided about seven military facilities where WP was identified to provide
additional context (WP is used in certain types of military ordnance, however these are not EAF
manufacturing sites). Table 3-2 provides a summary of information for the military facilities, including
site name and location, site characteristics, point of contact, and source of information.
In the EAF process, WP is produced from phosphate ore. The ore is formed into briquettes and calcined
to drive off moisture, remove organic matter, and hardened for further processing. The calcined
briquettes, also called nodules, are blended with coke and silica to form a mixture that is fed into electric
arc furnaces. Within the furnaces, the ore is reduced to produce WP in vapor form. WP is recovered by
water spray condensers, collected as a liquid in sumps, and pumped to a product storage area for shipment
or into tanks for interim storage. (FMC, 1996).
Of the ten WP EAF manufacturing sites, eight have unlined ponds and two did not provide information
about whether they have unlined ponds. The following eight sites that manufactured WP using EAF have
unlined ponds:
Monsanto Chemical Company (Solutia), Soda Springs, Idaho
• Stauffer Chemical Company (Rhodia, Inc.), Silver Bow, Montana
• Stauffer Chemical Company, Tarpon Springs, Florida
• Exxon Mobil ElectroPhos Division, Mulberry, Florida
• Agrifos Nichols Plant, Nichols, Florida
Stauffer Chemical Company (Rhone-Poulenc), Mt. Pleasant, Tennessee
Monsanto Chemical Company, Columbia, Tennessee
• Albright and Wilson America Limited Phosphorus Plant, Long Harbor, Newfoundland, Canada
15 August 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Monsanto Chemical Company (Solutia, Inc.), Soda Springs Idaho
WP, radionuclides
such as radium-
226, lead-210, and
uranium-238,
arsenic, beryllium,
selenium, and
zinc.
Phossy water was disposed of in
unlined ponds at this site. The
manufacturing process used a
distilling process to capture
elemental phosphorus, thereby
resulting in less phosphorus in the
waste. The phossy water was
expected to have very little
sediment. Additional
characterization data are not
available.
The remedy selected in the
ROD was either institutional
controls or excavation of
contaminated soils and
replacement with clean soil,
and disposal of the
contaminated soils within
the plant. The ponds
containing WP have been
capped and the site is
currently undergoing
groundwater monitoring.
1 . No further action
2. Groundwater monitoring
3. Land use and access
restrictions, and groundwater
monitoring
4. In situ biological treatment
(phytoremediation), land use and
access restrictions, and
groundwater monitoring
5. Soil excavation/replacement/
containment on-site, and
groundwater monitoring
Information not
provided
The unlined ponds
have been capped
and the site is
currently
undergoing
groundwater
monitoring.
CERCLA
Wallace Reid,
EPA
(206)553-1728
ROD, April 30, 1997
Telephone
conversation with
Wallace Reid, EPA,
Jan. 2002
16
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Stauffer Chemical Company (Rhodia Inc.), Silver Bow (Butte) Montana
WP, arsenic,
barium, cadmium,
chromium, lead,
and mercury.
Phossy water was disposed of in
unlined ponds at this site. In
addition, sludge was disposed of
in a concrete clarifier. The
clarifier is approximately 100 ft in
diameter, and is an open-topped,
in-ground unit that is constructed
of reinforced concrete, and
contains 400,000 - 500,000
gallons of sludge, with an
estimated 20 to 40 percent WP.
The clarifier may have a bottom,
but it is believed to have leaked,
and the exact integrity of the
clarifier is not known. Rhodia
does not want to remove and treat
the wastes from the clarifier
because they believe it would be
too risky for workers and too
expensive. Corrective action
alternatives for the clarifier are
being reviewed under a RCRA §
7003 order, and corrective action
is expected to be conducted under
the order.
Average background gamma
radiation levels at the site ranged
from 19 to 170 microRoentgens
(• R) per hour. Gamma radiation
closest to the clarifier ranged from
65 to 90 • R/hr. No data were
provided that the clarifier material
would have a gamma radiation
level any higher than 95 • R/hr.
The unlined ponds at this
site are not currently under
any form of corrective
action.
Remedies under consideration for
contents of the clarifier include:
1. Soil cap
2. Enhanced cap, with multi-layer,
multi-material cover
3. Off-site incineration
4. On-site phosphorus recovery
(roaster process*)
5. On-site incineration
6. Zimpro reactor process with
caustic hydrolysis of phosphorus
wastes at elevated temperatures
and pressure
7. Recovering phosphorus at
Solutia facility in Columbia,
Tennessee
8. Recovering elemental
phosphorus as phosphoric acid at
the Rhodia Phosphoric Acid
production facility
9. Recovering elemental
phosphorus at the Glen Springs
Holding Company phosphorus
facility in Columbia, Tennessee
10. Recovering phosphorus as
phosphoric acid at the Samancor
Phosphoric Acid production
facility located in Meyerton,
Gauteng Province, South Africa
* A roaster process was used to
treat waste while the plant was in
operation. The plant currently
does not use a roaster, as it was
dismantled when the plant ceased
operations.
Information not
provided, except
general reference to
elevated risk to
workers concerning
removal and
treatment of clarifier
materials.
Not an operating
facility. Rhodia,
prepared a waste
plan, dated
November 16, 2001,
as required under
RCRA § 7003
Order, Docket No.
RCRA-8-2000-07,
for management and
disposition of the
clarifier contents.
Site-wide corrective
action is not
currently being
addressed.
For clarifier waste,
EPA has conducted
a study on options
for treating this
material. In
negotiations with
the potentially
responsible party
(PRP) about
possible treatment.
RCRA
EPA Handler
ID
MTD05755854
6
Rosemary
Rowe,
EPA
(406) 457-5020
Rosemary Rowe,
EPA, Oct. 29,2001.
Waste Plan, Rhodia
Silver Bow Plant,
Butte, Montana,
November 16, 2001.
E-mail from
Rosemary Rowe,
February 27, 2002.
17
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Stauffer Chemical Company Operable Unit 1, Tarpon Springs Florida
Soil and pond
material is
contaminated with
arsenic, antimony,
beryllium,
cadmium,
chromium, lead,
fluoride, WP,
thallium, radium-
226, radon-222,
polonium-210,
and PAHs (RPM
stated that
radionuclides were
of particular
concern -
specifically
radium 226
because of
community
concerns.)
Phossy water was disposed in
unlined ponds at this site. Site
characterization data are available
in the Stauffer Management
Company Final RI/FS Work Plan.
The maximum detected
concentrations in soil/waste for
arsenic are 127 mg/kg, antimony -
32.3 mg/kg, beryllium - 6 mg/kg,
WP - 0.854 mg/kg, thallium - 13.4
mg/kg, and radium 226 - 73.8
pCi/g.
The remedy selected in the
ROD is in situ
solidification/stabilization
(S/S) of pond material and
contaminated soil below the
water table in the on-site
consolidation areas. In situ
S/S would be performed by
injecting and mixing binding
agents into the saturated
pond material to form a
solid, low permeability mix.
Treatment technologies
considered after the
screening process in addition
to the S/S process were:
1. Conversion to phosphoric
acid
2. Incineration
3. Aqueous oxidation
4. Low temperature air
oxidation.
1 . No action with continued
monitoring
2. Institutional controls
3. Consolidation and cover of
radiological and chemical waste
material on site
4. Consolidation and cover with
additional remediation of
radiologically and chemically
contaminated soil for future
residential use.
5. Consolidation and capping
radiological and chemical waste
material on site
6. Consolidation and capping with
additional remediation of
radiologically and chemically
contaminated soil for future
residential use
7. Consolidation, capping, and
saturated zone source control for
commercial use
8. Consolidation, capping, and
saturated zone source control for
residential use
9. Consolidation, stabilization,
and cover for commercial use
10. Consolidation, stabilization,
and cover for residential use.
Health and Safety
Plan prepared that
included worker
safety procedures for
well installation and
soil sampling.
Not an operating
facility. Currently
undergoing
additional studies to
address community
concerns (including
treatability studies
to evaluate the
performance of the
stabilized material).
CERCLA
EPA CERCLIS
ID
FLD01059601
3
Nestor Young,
EPA (404) 562-
8812
ROD, July 2, 1998.
Final RI/FS Work
Plan, Roy F.
Weston, Inc.,
February 1, 1993.
Telephone
conversations with
Nestor Young, EPA,
July and Oct. 2,
2001.
S/S Treatability
Studies Work Plan,
December 2000.
E-mail from Nestor
Young, February 28,
2002.
18
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Exxon Mobil ElectroPhos Division, Mulberry, Florida
Arsenic -0.0 13
ppm, cyanide -
0.36 ppm, fluoride
- 7.9 ppm, ortho-
phosphate - 5.96
ppm, and total
phosphorus - 6. 83
ppm.
Phossy water was discharged from
the plant to sedimentation ponds,
then to chemical neutralization
ponds and additional
sedimentation ponds, and finally
to clear-water ponds (old mine
cuts) for water reuse. The clear
water ponds are upgradient of the
phossy water ponds and presently
discharge through a NPDES
permitted outfall during high
rainfall events. There have been
no recent water quality violations
from this outfall. Total calculated
area is approximately 30 acres.
According to Florida DEP,
the phossy water ponds were
backfilled with overburden
(500,000 to 550,000 cubic
yards) to a depth of
approximately 5 ft over the
phosphorus sludges and the
water was transferred
through a recirculation
system to a liming station
for treatment. Closure plans
were submitted in May
1986. Surficial and
intermediate aquifer monitor
wells were installed circa
1976 surrounding the phossy
water ponds.
Information not provided.
Levels in surficial
aquifer wells down-
gradient of the ponds
exceeded water
quality standards for
gross alpha, radium,
fluoride, pH, and
TDS. Detection of
elemental phosphorus
impacts in
groundwater and
cyanide exceedances
in surface monitor
stations were
discovered.
Not an operating
facility.
Enforcement
commenced by the
Florida DEP in
1995, and the
facility recently
agreed to install a
low permeability
soil-bentonite slurry
wall around the
perimeter of the
phossy water ponds
and down to the
confining unit, a
HOPE liner over the
top of the ponds,
and to continue the
monitoring
program.
Information not
provided
EPA Handler
ID
FLD07086428
5, NPDES
Permit No.
FL0002666
David Clowes,
Florida DEP
(813)744-6100
x!39
EPA report, June 4,
1997.
CERCLIS 3
Wasteland Database
for Region 4
(queried November
6, 2001).
Conversation with
David Clowes,
FDEP on November
13, 2001. Memo
from Florida DEP to
Sunitha Ravi, Tetra
Tech EM Inc.,
January 28, 2002.
19
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Agrifos Nichols Plant, Nichols, Florida
Arsenic -0.0 13
ppm, cyanide -
0.36 ppm, fluoride
-23.75to 111.11
ppm, ortho-
phosphate - 5.96
ppm, and total
phosphorus - 6. 83
ppm.
Phossy water was disposed in
three unlined ponds covering an
area of 18.6 acres at this site.
Additionally, wastewater from a
sulfuric and phosphoric acid
fertilizer manufacturing plant was
also deposited in the ponds.
According to Florida DEP,
one of the phossy water
ponds was covered with an
earthen cap in the late
1950's. The other two ponds
were used until 1978 and
then closed. The closure
involved backfilling the
individual phossy ponds
with overburden
(approximately 200,000
cubic yards) to a depth of
about 2 ft over the elemental
phosphorus sludges, and
transferring the water
through the recirculation
system to a liming station
for treatment. Additionally,
drums of sludge were
apparently buried adjacent to
the ponds. Closure plans
were submitted in November
1984. Past and/or present
elevated levels of arsenic,
sodium, fluoride, TDS, gross
alpha, and radium have been
identified in surficial and
intermediate aquifer
monitoring wells.
Information not provided.
Mining by Agrifos in
the area has been
curtailed due to
groundwater
contamination.
Ajoint
contamination
assessment of the
facilities was
conducted although
further assessment
may be necessary
due to the location
of the ponds within
the areas of
contamination from
adjacent phosphate
fertilizer
manufacturing
facility.
Information not
provided.
EPA Handler
ID
FLD00410640
7, NPDES
Pemit No.
FLA132454
David Clowes,
Florida DEP
(813)744-6100
x!39
EPA report, June 4,
1997.
CERCLIS 3
Wasteland Database
for Region 4
(queried November
6, 2001).
Conversation with
David Clowes,
FDEP on November
13, 2001. Memo
from Florida DEP to
Sunitha Ravi, Tetra
Tech EM Inc.,
January 28, 2002.
20
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Stauffer Chemical Company (Rhone Poulenc) Mt. Pleasant, Tennessee
Information not
provided.
Phossy water was disposed in
unlined ponds.
The ponds were capped per
a 1998 Amended Closure
Plan.
Information not provided.
Information not
provided.
Not an operating
facility.
RCRA
EPA Handler
ID
TND00061039
4
Brad Martin,
Tennessee DEC
(617)687-7113
Dennis
Lampley,
Tennessee DEC
(931)840-4162
EPA report, June 4,
1997.
CERCLIS 3
Wasteland Database
for Region 4
(queried November
6, 2001).
Monsanto Chemical Company, Columbia, Tennessee
WP and fluoride.
Phossy water was disposed in an
unlined pond at this site. Water
was drained from the pond and
treated at an on-site wastewater
treatment plant.
Areas at the site
contaminated with elemental
phosphorus were capped.
Information not provided.
Information not
provided.
The elemental
phosphorus
manufacturing
facility was shut
down in October,
1986.
RCRA
EPA Handler
ID
TND00404810
4
Robert Gibbs,
Tennessee DEC
(931)840-4165
Grady Clark,
Solutia Inc.
(931)380-9329
EPA report, June 4,
1997.
CERCLIS 3
Wasteland Database
for Region 4
(queried on
November 6, 2000).
Conversation with
Mr. Clark of Solutia,
Inc. on November
13,2001 and April
30, 2002.
Conversation with
Ms. Apple,
Tennessee DEC on
November 14, 2001.
21
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for TJnlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
Albright and Wilson America Limited Phosphorus Plant (Rhodia, Inc.), Long Harbor, Newfoundland, Canada
WP, cyanide,
arsenic, fluoride,
cadmium, and
ammonia.
Dust-contaminated phosphorus,
known as "mud" was stored in
several mud holes on site.
Elemental phosphorus from
phosphorus containing mud
was recovered on site. Mud
holes were decommissioned
using in situ containment
(capped). Twenty
monitoring wells were
installed around the ponds
for groundwater monitoring.
The two other remedies that were
considered were mud still
treatment and landfilling.
Treatment technologies were
considered but determined to be
non-feasible, including burning in
tanks or pits, rotary kiln, high
pressure oxidation, steam
distillation, hypochlorite washing,
oxidation/aeration under water,
solvent extraction, sludge-to-
fertilizer extraction, in situ
chemical oxidation, and in situ
vitrification.
Potential hazards
identified for in situ
containment were
those associated with
worker exposure to
WP mud, WP sludge,
debris, and phossy
water, as well as
groundwater
contaminated by
leachate.
The plant ceased
operations in 1989.
Decommissioning
began in 1996, with
the majority of
activities completed
by 2000.
Geological
modeling performed
over a 100 year
period predicted the
WP would not
migrate significant
distances (i.e., less
than 1 meter).
Canada -
Environmental
Assessment
Act, EIS
completed
January 1996
Dexter Pittman,
Department of
Environment,
Government of
Newfoundland
and Labrador at
(709) 729-6771
Review of Pond 8S
Closure Plan,
Volume 1 and 2,
EMC Corporation
(Actual title
unavailable) (20
pages), January 23,
1998, EPA. Fact
Sheet No. 20,
Albright and Wilson
America Limited;
available on the
Internet
. Conversation
with Mr. Pittman on
February 6, 2002.
E-mail from Dexter
Pittman, March 12,
2002
22
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-1. Summary of Information About WP Electric Arc Furnace Manufacturing Sites (continued)
Types of
Contaminants in
the Historical
(Unlined) Ponds
Characterization of the Unlined
Ponds or of the Site
Remedy Selection
for Unlined Ponds
Other Remedies Considered
for Unlined Ponds
Health & Safety
Issues
Current Status
Regulatory
Background
Point
of Contact
Source of
Information
U.S. Tennessee Valley Authority (TVA) National Fertilizer Development Center, Muscle Shoals, Alabama
Information not
provided.
130,000 liters of phosphorus
sludge was stored in rail cars and
tanks and the remaining sludges
were buried as entombments at the
site.
The entombments consists
of phosphorus sludge with
one foot of crushed
limestone placed on the top.
A six-inch thick concrete
cap was constructed on the
top of the limestone layer.
Information not provided.
Information not
provided.
Not an operating
facility. Site
preliminary
assessment
completed on July
28, 1988 No further
remedial action
planned (NFRAP) at
site.
Information not
provided. EPA
ID#
AL3640090004
Ann Godfrey,
EPA
(404)562-8919
Doug Murphy,
Crane Army
Ammunition
Activity
(256) 386-2268
Review of Pond 8S
Closure Plan,
Volume 1 and 2,
EMC Corporation
(Actual title
unavailable) (20
pages), January 23,
1998, EPA.
Oldbury Electrochemical Company, Niagara Falls, New York
Information not
provided.
Information not provided,
including whether phossy water
was disposed in unlined ponds
and characteristics of such ponds.
Information not provided.
Information not provided.
Information not
provided.
Not an operating
facility.
Oldbury
Electrochemical
Company was
acquired by Hooker
Electrochemical
Company in 1956.
Hooker was
acquired by
Occidental
Chemical
Corporation in
1968. Information
about current status
not provided.
Information not
provided.
Michael
Negrelli, EPA
(212)637-4278
Kirk-Othmer
Encyclopedia of
Chemical
Technology, 2ntl
Edition, Vol. 15,
1972.
23
August, 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-2. Summary of Information About Selected Military Facilities Using Elemental Phosphorus
Site Name, Location
Eagle River Flats, Fort Richardson,
Anchorage, Alaska
Crane Army Ammunition Plant,
Indiana
Site Characterization
Sediments were identified as
contaminated with elemental
phosphorus from its use in ordnance.
Elevated waterfowl mortality was
reported in the 1980s which resulted
from the ingestion of elemental
phosphorus particles in the
sediments.
Soil and sludge were not reported as
contaminated with WP; rather site
stored munitions containing WP.
Point of Contact
Michael R. Walsh, U.S. Army Corps
of Engineers (US ACE) Cold Regions
Research and Engineering
Laboratory (CRREL)
(603) 646-4363
Randall W. Burcham, Crane Army
Ammunition Activity, Crane, Indiana
(812) 854-1353
Source of Information
Collins, 1999
Burcham, 1991
Military Installations with the Presence of Elemental Phosphorus1
Fort McCoy, Madison, Wisconsin
Yakima Training Center, Yakima,
Washington
Fort Bragg, North Carolina
Number of samples where elemental
phosphorus was detected per total
number of samples evaluated: 1 1/45
Highest reported concentration of
elemental phosphorus: 58 mg/kg
Number of samples where elemental
phosphorus was detected per total
number of samples evaluated: 3/27
Highest reported concentration of
elemental phosphorus: 430 mg/kg
Number of samples where elemental
phosphorus was detected per total
number of samples evaluated: 7/90
Highest reported concentration of
elemental phosphorus: 0.1 mg/kg
Information not provided.
Information not provided.
Information not provided.
Rivera, October 1996.
Rivera, October 1996.
Rivera, October 1996.
24
August 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Table 3-2. Summary of Information About Selected Military Facilities Using Elemental Phosphorus (continued)
Site Name, Location
Site Characterization
Point of Contact
Source of Information
Fort Riley, Fort Riley, Kansas
Number of samples where elemental
phosphorus was detected per total
number of samples evaluated: 3/24
Highest reported concentration of
elemental phosphorus: 0.24 mg/kg
Information not provided.
Rivera, October 1996.
Fort Drum, Fort Drum, New York
Number of samples where elemental
phosphorus was detected per total
number of samples evaluated: 11/45
Highest reported concentration of
elemental phosphorus: 0.023 mg/kg
Information not provided.
Rivera, October 1996.
1. USAGE evaluated military installations where elemental phosphorus might be present and identified these five sites as those where the risk of
contamination from elemental phosphorus was high.
25
August 2003
-------
. Overview of Elemental Phosphorus Electric Arc Furnace Manufacturing Sites
Information was not provided about whether the following two EAF sites have unlined ponds:
U.S. TVA National Fertilizer Development Center, Muscle Shoals, Alabama
• Oldbury Electrochemical Company, Niagara Falls, New York
Of these ten sites, only Monsanto, Soda Springs currently is operating as a manufacturer of WP.
Of the eight sites with unlined ponds, six have installed or plan to install caps for the unlined ponds. Of
the remaining two sites, at one site (in Silver Bow Montana), the ponds are not under corrective action,
and at another site (in Tarpon Springs Florida), solidification/stabilization in combination with the
consolidation and capping of contaminated material has been selected as a remedy for the ponds. The
following is a brief summary of key information provided in Table 3-1 for sites with unlined ponds.
At the Monsanto Chemical Company (Solutia), Soda Springs, Idaho site, regulated under CERCLA, the
manufacturing process used a distillation step to recover WP, and, therefore, the unlined ponds are
expected to contain less WP than found at EMF. The unlined ponds at the Soda Springs site have been
capped and the site currently is undergoing groundwater monitoring. At the Stauffer Chemical Company
(Rhodia, Inc.), Silver Bow, Montana site, regulated under RCRA, the unlined ponds are not under
corrective action. However, a concrete clarifier used to contain WP-contaminated sludge is being studied
for possible corrective action. At the Stauffer Chemical Company, Tarpon Springs, Florida site, regulated
under CERCLA, the remedy selected for the unlined ponds is in situ solidification and stabilization (see
further discussion in Section 5.0 and in Appendix A). At the Exxon Mobil ElectroPhos Division,
Mulberry, Florida site, regulated under CERCLA, the facility recently agreed to install a high-density
polyethylene (HOPE) liner over the top of the ponds and a low-permeability slurry wall around the
perimeter of the ponds, and to continue groundwater monitoring. At the Agrifos Nichols Plant, Nichols,
Florida site (regulatory background not provided), one unlined pond was covered with an earthen cap in
the 1950s, and two other unlined ponds were backfilled with overburden after 1978 to a depth of two feet.
These ponds are located within an area that is contaminated by a phosphate fertilizer manufacturing
facility. At the Stauffer Chemical Company (Rhone-Poulenc), Mt. Pleasant, Tennessee site, the unlined
ponds were capped under a 1998 Amended Closure Plan. At the Monsanto Chemical Company,
Columbia, Tennessee site, an unlined pond was drained and capped. At the Albright and Wilson,
Newfoundland, Canada site, unlined ponds (referred to as "mud holes") were capped. Use of treatment
technologies was considered for this site but determined not to be feasible.
Two of the seven military facilities performed treatment for WP (Eagle River Flats and Crane Army
Ammunition plant), and these are discussed further in Section 5.0.
26 August 2003
-------
, Overview of Historical Ponds at EMF
4.0 OVERVIEW OF HISTORICAL PONDS AT EMF
This section provides an overview of the physical and chemical characteristics of the historical ponds at
the EMF site, and the fate and transport of WP in the ponds. Limited information was available to
characterize the historical ponds at the EMF site. The PJ/FS and ROD for EMF provide only general
descriptions for many of the ponds and do not identify the specific physical and chemical characteristics
of each pond, such as concentration and distribution of specific contaminants or moisture content within
and across the 16 historical ponds (although data for samples from several of the ponds was presented, as
discussed below). For the purposes of this report, several assumptions were made about pond
characteristics using technical judgment, based on what is known about the ponds and experience with
site cleanups.
Sixteen historical ponds are addressed in this report - 15 unlined ponds (Ponds IS through 7S, 9S, and IE
through 7E) and one lined pond (Pond 10S), all of which are being addressed under CERCLA.
One unlined pond at EMF, Pond 8S, is not included in this report because it is currently being addressed
under RCRA. However, available data about the characteristics of this pond, which handled wastes
similar to that of the historical ponds, provides a better understanding of the characteristics of the
historical ponds at EMF. In 1993-1994, a time-critical removal action under CERCLA was performed by
FMC to remove the hydraulic head and perform interim capping of Pond 8S. Under this action, Pond 8S
was dewatered, filled with sand and slag, and covered. This pond became subject to regulation under
RCRA in March 1990. The Pond 8S RCRA closure plan was approved be EPA in August 1998 and FMC
completed and certified the closure in 1999. Pond 8S is currently being monitoring under a RCRA post-
closure plan.
Physical Characteristics
The historical ponds were used during manufacture of WP from phosphate ore from the 1950s to 1981.
During this time, the ponds received the following wastes: phossy water, precipitator dust slurry, phossy
solids, slag pit water and solids, and residuals from reclaiming WP in other ponds. Phossy water is an
aqueous stream that contacted WP during its manufacture and contains elevated levels of WP.
Precipitator dust slurry is the result of the furnace process electrostatic precipitators dry dust that consists
of unreacted (unmelted) ore and WP that condenses in the precipitators prior to WP product recovery in
27 August 2003
-------
, Overview of Historical Ponds at EMF
the wet condensers. The dry precipitator dust was slurried with water to prevent WP oxidation and to
convey the dust to the ponds.
WP was reclaimed from six of the ponds as part of operating the pond, with some ponds reclaimed on a
regular basis (e.g., twice each year for Ponds IS, 2S, and 3S) or on a one-time basis (e.g., Pond 9S to a
depth of 25 ft bgs, 2E, and 4E). Reclamation consisted of excavating pond materials or oxidizing WP in
the pond. Information was not provided about how the WP was oxidized at that time. The reclaimed
areas of Pond 3S were filled in with slag. Portions of other ponds were covered by stockpiled materials,
(e.g., Pond 7S has a ferrophosphorus stockpile approximately 25 ft high), or were reused as precipitator
dust slurry sedimentation ponds.
In the 1970s, free liquid was removed (i.e., pond dried) from seven of the ponds that were no longer in
use, with free liquids removed from eight other ponds between 1980 and 1981 (information was not
provided about the year that Pond 10S had liquids removed). After drying, eight ponds were covered
with a variety of materials including soil, slag, crushed slag, baghouse dust, fluid bed drier product prills,
asphalt paving, and cement. Examples of covers include capping with 3 to 6 ft of soil, slag, a paved
asphalt road, cement, and 6 to 10 ft of pit-run slag. In addition, nine ponds were covered by RCRA lined
ponds or ferrophos storage. Most of the ponds were regraded to direct surface precipitation away from
the ponds.
Assumption About Depth of Contaminated Soil and Sludge in Ponds: The assumed maximum depth of
contaminated soil and sludge in the historical ponds is 20 ft. It is assumed that the ponds were originally
constructed to a depth of 10 ft, but that there was downward migration of contaminants in ponds that had
a sustained hydraulic head, or where materials were integrated in fill, causing contamination of soil to a
depth of as much as 20 ft. The EMF RI report notes the depth of fill, the maximum depth of impacted
soil encountered, and the predominant constituents exceeding representative levels at this maximum depth
for each area or investigation within the FMC area. The maximum depth of impacted soil associated with
the historic FMC ponds (at which a sustained hydraulic head was present during operation) was 180 feet
bgs (Pond 6S). Reference to a maximum depth of impact of 20 ft bgs in this report refers to an estimated
general thickness of fill (for example, phossy pond sediments) and fill that had been mechanically mixed
with the underlying native soils. Because of the physical activities performed during pond operation,
such as excavation/reclamation and oxidation, as well as filling with a variety of materials, the actual
depths of contaminated soil and sludge in the ponds may be less than 20 ft. Further, it is likely that
depths vary within a given pond and from one pond to another. For example, one pond may contain a
28 August 2003
-------
, Overview of Historical Ponds at EMF
substantial amount of fill such as slag that would not contain elevated levels of contaminants; in this
example, the depth may be less than 20 ft.
Basis: The soil boring data show that most of the contamination is in the top 10 ft of the ponds. Further,
these data identify no phossy wastes at depths greater than 10 ft. However, data also show that there is
some contamination at depths from 10 to 20 ft, in the native soil (for example, see Table 4-1 for soil
borings F034B in Pond 9S, F033B in Pond IE, and F024B in Pond 4E, all with elevated levels of WP at
depths of approximately 20 ft). Soil boring data show substantially little contamination below 20 ft.
Assumption About Volume of Contaminated Soil and Sludge in Ponds: The assumed total volume of
contaminated soil and sludge in the historical ponds is approximately 500,000 cubic yards (CY), based on
a depth of 20 ft and an areal extent of 16 acres. It is likely that some of the ponds only have
contamination in the upper 10 ft; therefore, this estimate may be high. In addition, information in the FS
shows that the total volume of contaminated soil and sludge in the 16 historical ponds is approximately
120,000 CY. The FS estimate is based on a volume per pond "which contain constituents of potential
concern at levels which conceivably could require remedial action". No further information was provided
about how the volume per pond was identified in the FS.
Basis: The maximum contamination depth was assumed to be 20 ft, and the areal extent of ponds that
have not been covered by RCRA ponds or ferrophos storage was calculated as 16 acres.
Assumption About Heterogeneity of Ponds: The contents of the ponds are assumed to have substantial
physical and chemical variation (heterogeneity) throughout the vertical profile and across the areal extent,
both within and across ponds.
Basis: Pond contents consist of phossy wastes, slag, fill, dust, solids, gravel, and a variety of types of
silty materials. These materials were disposed of in ponds as they were available. In addition, soil
borings (Table 4-2) show heterogeneity within the vertical profile and across the areal extent of selected
ponds.
Assumption About Matrix Characteristics of Ponds: The moisture content of the ponds is assumed to be
"medium", with the ponds not fully-saturated or completely dry. Due to the pond heterogeneity, it is
likely that the moisture content varies within and among ponds. The density of pond materials is assumed
to be 1.5tons/CY.
29 August 2003
-------
, Overview of Historical Ponds at EMF
Basis: All free liquids have been removed from the ponds. Because the site is in a semi-arid area, with
average precipitation of 11 inches per year, there is relatively little moisture that enters the ponds and it is
likely that some moisture in the ponds is removed by evaporation or runoff. The density of silty soils is
reported to range from approximately 95 to 125 pounds per cubic foot (lbs/ft3) (Lindenburg, 1986).
Using an average density of 110 lbs/ft3 corresponds to 1.5 tons/CY.
Chemical Characteristics
Table 4-1 provides a summary of soil boring results for samples collected during the RI in or adjacent to 7
ponds to a depth up to 20 ft. Soil samples were collected from three borings located near Pond 7S, three
borings located near Pond 8S, two borings in Pond 9S, two borings located within the boundary of former
Pond IE, one boring in former Pond 4E, two borings in former Pond 5E, and two borings in former Pond
6E. While Pond 8S is not addressed by this report, the data from soil borings collected near the pond help
to describe the characteristics of soil located near the historical ponds. The samples collected from soil
borings were analyzed for total phosphorus, other constituents (metals and nonmetals), and radiological
parameters (gross alpha and gross beta). Table 4-1 shows elevated concentrations of total phosphorus, as
well as other constituents including cadmium, fluoride, and zinc. Maximum concentrations shown in
Table 4-1 include total phosphorus at 86,000 mg/kg (8.6%), cadmium at 5,610 mg/kg, chromium at 278
mg/kg, copper at 221 mg/kg, fluoride at 44,800 mg/kg, and lead at 1,300 mg/kg. Gross alpha levels were
measured as high as 1,530 picocuries per gram (pCi/g) and gross beta as high as 1,070 pCi/g. In addition,
the RCRA ponds at the site have been found to emit hydrogen cyanide and phosphine in areas of low or
high pH. The cyanide resulted from the reaction of nitrogen with the carbon in the reduction furnace.
The results show the concentration of total phosphorus, but do not identify the fraction that is WP. Total
phosphorus results include WP, phosphate, and other phosphorus-containing compounds. For other
constituents, only those constituents that are reported by the RI as greater than background (shown in the
RI as the representative range) are summarized in Table 4-1. Results for individual radionuclides were
not reported for these samples.
Soil characteristics were noted during description of the soil borings, and are summarized in Table 4-2.
The descriptions of materials encountered during borehole drilling included "pure precipitator dust,"
"phossy solids," slag, and fill. According to the RPM, there is likely to be a heterogeneous mixture of
materials present in the soil/sludge within these ponds.
30 August 2003
-------
, Overview of Historical Ponds at EMF
Outside of the ponds, the contaminants of potential concern in the soil, as identified in the ROD, include
antimony, arsenic, beryllium, boron, cadmium, fluoride, lead-210, manganese, mercury, nickel,
polonium-210, potassium-40, selenium, silver, thallium, uranium-238, vanadium, and zinc. In addition,
gross alpha and gross beta contamination were identified as being of concern.
Although individual radionuclides were not analyzed for in the soil, the RI identified specific
radionuclides based on an analysis of the data for gross alpha and beta. Radon also was identified for its
potential to infiltrate buildings under alternate future commercial or industrial uses of the site. With
radionuclides, the primary alpha emitters at the site were identified as uranium-238, uranium-234,
thorium-230, radium-226, and polonium-210. The primary beta emitters were identified as lead-210 and
potassium-40. The activity of radium-226 ranged from 4 to 24% of the gross alpha measurements.
Assumption About Source of Contamination: The only source for chemical contaminants in the soil and
sludge in the unlined ponds (WP, heavy metals, and radionuclides) is assumed to be the constituents
found in the phosphate ore processed at the site and in the by-products/wastes from manufacturing WP,
and no contamination is associated with the relatively small amounts of reagents, catalysts, or fuels used
to manufacture elemental phosphorus.
Basis: Soil boring data show that the contaminants found in the unlined ponds are the constituents
expected to be present based on the content of phosphate ore.
Assumption About Percentage of Phosphorus in Selected By-products/Wastes: Among the by-
products/wastes from the manufacture of WP, it is assumed that phossy water and precipitator dust had a
relatively higher percentage of phosphorus than other wastes such as slag pit water and solids, and
residuals from reclaiming WP in other ponds.
Basis: Soil boring data showed relatively higher concentrations of phosphorus in ponds where phossy
water and precipitator dust had been disposed. Data for Pond IE (sample at 0.5 ft) showed a total
phosphorus concentration of 8.6 percent, and soil boring logs show the material that was sampled
consisted of various materials including precipitator dust and phossy water solids. This sample had the
highest concentration of phosphorus measured in any of the available soil borings.
Assumption About Contaminant Concentrations in Historical Ponds: Table 4-3 shows a summary of the
average contaminant concentrations of soil and sludge in historical ponds, assuming that the
31 August 2003
-------
, Overview of Historical Ponds at EMF
concentrations are the same as the average contaminant concentrations in phossy sediment generated
during production of WP. As this table shows, the ponds contain heavy metals such as arsenic, cadmium,
chromium, vanadium, and zinc at levels as high as 26,000 mg/kg, fluoride as high as 17,000 mg/kg, total
phosphorus as high as 28,000 mg/kg (2.8%), and radionuclides as high as 780 pCi/g. These
concentrations range as high as 1,000 times higher than soil background concentrations. For example,
total phosphorus ranges as high as 500 times higher than soil background. As mentioned above, the
ponds are heterogeneous, and the concentrations of contaminants are non-uniform throughout the pond,
with some portions of a pond exhibiting relatively higher concentrations (such as areas where wastes were
stored or where a sustained hydraulic head was present), while other portions of a pond would have
relatively lower concentrations (such as where fill was added to a pond).
Basis: The contaminant concentrations in phossy sediments disposed of in RCRA ponds likely are
similar to those disposed of in historical ponds. Further, soil borings were performed only in a limited
number of historical ponds and data from these borings likely are not as accurate as RCRA data at
capturing the expected average of concentrations for wastes that were disposed in historical ponds.
Assumption About Distribution of Total Phosphorus in Ponds: Only a relatively small fraction of the
total phosphorus measured is assumed to be in the elemental form.
Basis: The range of total phosphorus concentrations of 21,300 to 28,000 mg/kg estimated to be present in
waste materials in the ponds consists of WP, as well as other forms of phosphorus such as phosphate. As
discussed above, there were a number of actions taken to reclaim phosphorus from some of the ponds.
WP is the main product from the facility and efforts likely were made to reduce product losses to the
waste ponds and to reclaim the product from the ponds.
Fate and Transport of Elemental Phosphorus
The following discussion about fate and transport of WP is based on technical judgment using
information presented in Section 2.0, what is known about the ponds, and experience with site cleanups.
(FMC FS, 1996, and FMC, 1997a). In the historical ponds at EMF, WP likely would not exhibit
substantial oxidation or hydrolysis, nor would it likely be transported from the ponds, leaving it to remain
in the soil and sludge as WP. The oxidation and hydrolysis of WP would be limited primarily because of
its relatively low solubility in water (3 mg/L), as well as the limited amount of oxygen expected to be
present. Because the ponds contain no free liquids and are covered, the only available moisture likely
32 August 2003
-------
, Overview of Historical Ponds at EMF
will be due to the moisture content of the soil and sludge prior to being covered, or the relatively small
amount that enters the pond from percolation of rainwater and snowmelt (annual precipitation of 11
inches).
In this environment, WP that is dissolved in the soil and sludge moisture would be hydrolyzed to form
phosphine and lesser amounts of phosphoric acid. However, the hydrolysis of WP is a slow process,
especially under limited moisture conditions. The limited amount of DO present in the soil and sludge
would react with phosphorus to form various types of soluble phosphorus acid ions, and, depending on
environmental conditions, these ions may be further converted to a solid metal phosphate compound such
as calcium phosphate. The rate of biological transformation of WP, which results in the same soluble
forms of phosphorus acids as are produced through hydrolysis and oxidation, is even slower than with the
chemical transformation processes.
For these reasons, the expected fate of WP is to remain in the ponds as WP or as a solid metal phosphate.
This conclusion is further supported by research that demonstrates that chemical and biological
transformation of WP in soils and sediments, where partial pressures of oxygen are low, will be very slow
(Spanggord et al, 1983).
Oxidation of WP through contact with atmospheric oxygen would occur in portions of ponds that have
not been covered, or as oxygen diffuses through the soil or slag cover and contacts phosphorus particles.
Given that pond covers may be several feet thick, may have other oxygen-absorbing compounds in the
cover, soil, and sludge, and phosphorus particles may be as deep as 20 ft below ground surface in the
ponds, it is unlikely there will be substantial oxidation of WP in the ponds at this time through contact
with atmospheric oxygen. In addition, the phosphorus particles may be present in dispersed locations
throughout the ponds, and, even if there were an area of the pond with a more direct channel for
delivering oxygen to the soil and sludge, it might not be in the area where the phosphorus particles are
present. It is likely that any phosphorus that is in readily-accessible locations (e.g., near the surface, near
sources of oxygen) was oxidized earlier in that period of time and would no longer be available as WP.
The direct transport of WP from the ponds to groundwater would be limited primarily because of its
relatively low solubility in water (3 mg/L), as well as the limited amount of available moisture, even
during the active life of the ponds. The by-products from oxidation, hydrolysis, and biological
transformation of WP, including phosphorus acids, are highly soluble in water, but these compounds are
not expected to be generated at high enough concentrations or at high enough rates in the sludges and
33 August 2003
-------
, Overview of Historical Ponds at EMF
soils to cause large-scale transport of phosphorus acids to the groundwater, except where a substantial
hydraulic head within an unlined pond was present during operation.
The expected low rate of phosphorus migration to the groundwater, relative to the amount of phosphorus
that was originally pumped to the ponds, is demonstrated by the low concentrations of total phosphorus
(generally between 10 and 100 mg/L) in the groundwater that were presented in the FS. The FS showed
the maximum concentration of total phosphorus as follows: central FMC area - 34.25 mg/L, eastern FMC
area - 96.8 mg/L, northern FMC properties (Michaud) - 7.61 mg/L, northern FMC properties (Bannock) -
29.63 mg/L, southwest FMC area (Michaud) - 52.6 mg/L, and southwest FMC area (Bannock) - 665
mg/L. Within historical ponds, the concentrations of total phosphorus in the soils and sludge after closure
ranged from 1,640 to 59,700 mg/kg.
The total phosphorus in the groundwater has been found, by concurrent analyses of groundwater samples
for both total phosphorus and orthophosphate, to be almost entirely orthophosphate Orthophosphate is
less toxic to humans, plants, and animals than WP. However, orthophosphate can stimulate plant growth,
and thus has the potential to cause algal blooms in surface water bodies that are hydraulically connected
to groundwater.
WP contained in ponds would not be released directly to the air as its elemental form, but may be released
as phosphorus pentoxide or phosphine gases from the surface of the ponds. However, the slow rate of the
hydrolysis reactions, combined with the effects of dilution of the gas in the outside atmosphere, are
expected to result in low, environmentally insignificant concentrations of these gases. The rate of gas
generation and the emission rate of the gas after it is generated are expected to be even lower in filled
ponds than in those same ponds before they were filled, primarily because of the limited availability of
soil moisture. In addition, any gas that is be generated would be in soil-pore spaces where it may undergo
oxidation and hydrolysis to phosphoric acid. (FMC, 1997a)
34 August 2003
-------
, Overview of Historical Ponds at EMF
Table 4-1. Soil Boring Results for Surface (less than 20 ft) of Unlined Ponds (1)
Pond
7S
7S
7S
Soil
Boring
(2)
F055B
F056B
F057B
Sample Depth
(ft below ground
surface)
3
5
5
7
5
7
Total
Phosphorus
(mg/kg)
8,240
8,940
9,530
51,400
2,230
12,700
Other Constituents
(mg/kg)
Cadmium 154
Chromium 139
Copper 27.9
Fluoride 4,720
Lead 30.8
Nickel 31.4
Silver 17.2
Vanadium 153
Zinc 2,940
Cadmium 32.2
Zinc 801
Cadmium 33.7
Chromium 1 14
Copper 21.3
Fluoride 2,630
Nickel 31.6
Vanadium 148
Zinc 437
Cadmium 5,610
Chromium 242
Cobalt 19.3
Copper 221
Fluoride 44,800
Lead 1300
Manganese 1,240
Molybdenum 26.9
Nickel 38.1
Selenium 24.4
Silver 786
Vanadium 149
Zinc 499,000
Cadmium 27.1
Chromium 33.7
Fluoride 890
Zinc 485
Cadmium 235
Chromium 109
Copper 28.3
Fluoride 3,330
Lead 56.9
Molybdenum 4.7
Nickel 29.4
Selenium 9.1
Silver 30
Vanadium 117
Zinc 3,860
Radiological
Parameters
(pCi/g)
Gross Alpha: 16.6
Gross Beta: 22.6
Gross Alpha: 46
Gross Beta: 42
Gross Alpha: 25.4
Gross Beta: 32.5
Gross Alpha: 1,530
Gross Beta: 1,070
Gross Alpha: 31.4
Gross Beta: 28.1
Gross Alpha: 47.4
Gross Beta: 34.9
August 2003
-------
i Overview of Historical Ponds at EMF
Table 4-1. Soil Boring Results for Surface (less than 20 ft) of Unlined Ponds (1) (Continued)
Pond
8S
8S
8S
8S
Soil
Boring
(2)
B12
B12
B13
B150
Sample Depth
(ft below ground
surface)
0
10
20
0
10
20
1
2
10
20
Total
Phosphorus
(mg/kg)
774
9,980
1,610
699
3,250
992
954
612
4,080
1,630
Other Constituents
(mg/kg)
None different from
background
Cadmium 535
Chromium 154
Copper 40.8
Fluoride 7,030
Lead 137
Nickel 35.7
Selenium 6.7
Silver 56
Vanadium 138
Zinc 10,800
Fluoride 784
None different from
background
Cadmium 15.2
Zinc 637
Zinc 52.7
None different from
background
None different from
background
Cadmium 13.5
Chromium 178
Fluoride 11,400
Vanadium 157
Zinc 239
Zinc 64
Radiological
Parameters
(pCi/g)
Not measured
Gross Alpha: 143
Gross Beta: 117
Gross Alpha: 8.3
Gross Beta: 20
Not measured
Gross Alpha: 6.9
Gross Beta: 14.8
Gross Alpha: 11.3
Gross Beta: 20.2
Not measured
Not measured
Gross Alpha: 172
Gross Beta: 84
Gross Alpha: 8.7
Gross Beta: 18.7
36
August 2003
-------
i Overview of Historical Ponds at EMF
Table 4-1. Soil Boring Results for Surface (less than 20 ft) of Unlined Ponds (1) (Continued)
Pond
9S
9S
IE
IE
Soil
Boring
(2)
B-2
F034B
F033B
B-l
Sample Depth
(ft below ground
surface)
0.5
6.5
11.5
5
10
15
20
4.5
9.5
14.5
19.5
0.5
6.5
11.5
Total
Phosphorus
(mg/kg)
74,100
1,970
1,810
2,730
1,870
767
477
6,410
2,380
1,900
1,630
86,000
5,770
2,820
Other Constituents
(mg/kg)
Cadmium 4,150
Chromium 216
Copper 141
Lead 574
Selenium 73.4
Silver 69.9
Vanadium 183
Zinc 52,600
Zinc 180
Zinc 67.2
Cadmium 25.1
Fluoride 1,980
Zinc 366
Fluoride 2,120
Zinc 146
Zinc 210
Orthophosphate 53.1
Boron 50.6
Chromium 51.8
Copper 22.9
Fluoride 4,130
Lead 79.6
Zinc 7,320
Fluoride 1,460
Zinc 295
Cadmium 3.9
Fluoride 1,560
Nickel 17.6
Zinc 174
Fluoride 1,210
Zinc 75.7
Cadmium 5,110
Chromium 278
Copper 155
Lead 683
Selenium 77.6
Silver 79.6
Vanadium 234
Zinc 69,800
Zinc 1,590
Zinc 1,070
Radiological
Parameters
(pCi/g)
Not measured
Not measured
Not measured
Gross Alpha: 32.1
Gross Beta: 47.3
Gross Alpha: 11.6
Gross Beta: 25
Gross Alpha: 31.8
Gross Beta: 42.9
Gross Alpha: 7.72
Gross Beta: 18
Gross Alpha: 23.9
Gross Beta: 34.6
Gross Alpha: 19.8
Gross Beta: 37.5
Gross Alpha: 1 14
Gross Beta: 152
Gross Alpha: 20
Gross Beta: 30.9
Not measured
Not measured
Not measured
37
August 2003
-------
i Overview of Historical Ponds at EMF
Table 4-1. Soil Boring Results for Surface (less than 20 ft) of Unlined Ponds (1) (Continued)
Pond
4E
5E
Soil
Boring
(2)
F024B
B-114
Sample Depth
(ft below ground
surface)
0
10
20
5.5
Total
Phosphorus
(mg/kg)
25,700
1,640
1,160
3,440
Other Constituents
(mg/kg)
Boron 54.6
Cadmium 918
Chromium 55.9
Copper 41.4
Fluoride 8,030
Lead 157
Molybdenum 6.5
Potassium 15,700
Selenium 11.3
Silver 87.1
Vanadium 55
Zinc 15,200
Fluoride 1,510
Potassium 8,000
Zinc 85.3
Fluoride 810
Potassium 5,900
Cadmium 21.7
Potassium 6, 170
Silver 2.5
Zinc 1,720
Radiological
Parameters
(pCi/g)
Gross Alpha: 95.3
Gross Beta: 113
Gross Alpha: 16.1
Gross Beta: 25
Gross Alpha: 18.5
Gross Beta: 26.2
Not measured
38
August 2003
-------
i Overview of Historical Ponds at EMF
Table 4-1. Soil Boring Results for Surface (less than 20 ft) of Unlined Ponds (1) (Continued)
Pond
5E
6E
6E
Soil
Boring
(2)
F025B
B-130
F026B
Sample Depth
(ft below ground
surface)
5
10
20
16
0
10
Total
Phosphorus
(mg/kg)
12,300
33,600
1,630
1,700
1,400
1,640
Other Constituents
(mg/kg)
Boron 28.4
Cadmium 532
Chromium 48.2
Copper 25
Fluoride 7,670
Lead 106
Potassium 11,900
Selenium 5.6
Sodium 1,050
Silver 61
Zinc 11,600
Boron 67.9
Cadmium 2,080
Chromium 108
Copper 57.3
Fluoride 35,000
Lead 360
Nickel 24.9
Potassium 23,700
Selenium 18.6
Silver 152
Sodium 2,710
Vanadium 78
Zinc 200,000
Fluoride 1,120
Potassium 5,790
Potassium 6,250
Zinc 66.5
Fluoride 1,610
Fluoride 1,570
Potassium 4,430
Radiological
Parameters
(pCi/g)
Gross Alpha: 111.1
Gross Beta: 113.8
Gross Alpha: 303
Gross Beta: 265
Gross Alpha: 26.4
Gross Beta: 27.7
Not measured
Gross Alpha: 22.3
Gross Beta: 21.4
Gross Alpha: 16.8
Gross Beta: 23.3
Source: FMCRI
(1) Results show the concentration of total phosphorus, but do not identify the fraction that is WP. Total phosphorus
results would include elemental, phosphate, and other phosphorus-containing compounds that might be present. For
other constituents, only those constituents that are reported by the RI as greater than background (shown in the RI as
the representative range) are summarized in Table 4-1. Results for individual radionuclides were not reported.
(2) Soil borings for Ponds 7S and 8S were performed near the ponds but not within the pond boundaries.
39
August 2003
-------
, Overview of Historical Ponds at EMF
Table 4-2. Summary of Soil Characteristics Noted in Soil Boring Activities
Pond
7S
8S
9S
IE
4E
5E
6E
7E
Borehole (1)
F055B
F056B
F057B
B12
B13
B150
B-2
F034B
F033B
Bl
F024B
B114
F025B
B130
F026B
F162B
Depth (ft)
o
J
5
5
7
5
7
0
10
20
0
10
20
1
2
10
20
0.5
6.5
11.5
5
10
15
20
4.5
9.5
14.5
19.5
0.5
6.5
11.5
0
10
20
0
5.5
21
5
10
20
16
0
10
16
Description
Brown silty material
Brown silty material
Tan-brown silt
Tan to black silt, precipitator dust
Tan silt
Tan sand/silt, gray clay with precipitator dust
Black clayey fill with phossy water solids or precipitator dust
Black clayey fill with phossy water solids or precipitator dust
Brown silt
Slag gravel fill
Dark brown sandy silt fill
Brown silt
Slag
Slag
Slag
No description
Dark gray to black slag fill
Yellowish brown silt, just below fill/soil interface
Yellowish brown silt
Pale brown silt
Pale brown silt
Reddish, yellow sandy gravel
Reddish, yellow sandy gravel
Brownish yellow silt, slag fill
Brownish yellow silt
Brownish yellow silt
Brownish yellow silt
Dark gray silty sand fill, slag with precipitator dust or phossy water
solids to 4.5 ft
Very pale orange silt
Very pale orange silt
Slag
Brown silt fill with precipitator dust or phossy water solids fill
Brown silt
Slag, gravel fill
Grayish brown silt slag with phossy solids
Dark yellowish brown silt
Dark olive gray silt with phossy water solids
Dark olive gray silt with phossy water solids
Yellowish brown silt
Yellowish brown silt
Light yellowish brown silt
Light yellowish brown silt
Pale, yellowish brown gravelly sand
Source: FMC RI, 1996
(1) Soil borings for Ponds 7S and 8S were performed near the ponds but not within the pond boundaries.
40
August 2003
-------
, Overview of Historical Ponds at EMF
Table 4-3. Average Range of Concentrations for Selected Constituent in Historical Ponds
Constituent
Arsenic
Cadmium
Chromium
Fluoride
Phosphorus (total)
Vanadium
Zinc
Gross Alpha
Gross Beta
Lead-210
Potassium-40
Uranium-238
Units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
pCi/g
pCi/g
pCi/g
pCi/g
pCi/g
Soil Background Level
7.7
1.9
27.5
600
672
45.4
52.8
NC
NC
6.36
20.5
3.88
Estimated
Concentration in
Historical Ponds
20.4 - 256
1,100-2,040
71.6-133
8,600 - 17,100
21,300-28,000
42.9 - 93.4
10,400 - 26,600
71.1 -289
254 - 783
204 - 465
13.1-27.4
ND(5)
NC - Not Calculated
ND - Not Detected (detection limit)
Source: FMCFS, Table 2.2-1, 1996 (data for Ponds 8S, US, 12S, and 15S)
41
August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
5.0 TREATMENT TECHNOLOGIES FOR SOIL AND SLUDGE IN HISTORICAL PONDS
Section 4.0 provided an overview of the historical ponds at EMF, based on available data and several
assumptions, and showed that the portions of the historical ponds of interest for this report:
• Cover a total area of approximately 16 acres and extend to a depth of 20 ft
Contain approximately 500,000 cubic yards (CY) of soil and sludge
• Are contaminated with WP, at concentrations as high as 2.8 percent, as well as heavy metals and
radionuclides
Are heterogeneous in physical and chemical composition
• Are covered with 3 to 10 ft of soil or slag
This section describes the types of technologies that may be applicable for treatment of soil and sludge in
historical ponds at the EMF site. It includes an overview of the methodology used in identifying and
evaluating potential treatment technologies, discusses several attributes that are important for
implementing any treatment technology (referred to as "cross-technology considerations"), and provides a
description of specific remedial technologies. First, excavation is described, followed by six treatment
technologies where data have been identified concerning their potential for use at historical ponds at a
WP-manufacturing site.
Methodology for Identifying and Evaluating Potential Treatment Technologies
Commercially-available treatment technologies were identified that have been used to reduce the
concentration, volume, or mobility of WP, heavy metals, and radionuclides in soil and sludge. In
addition, technologies were identified that are not commercially available but have shown promise for
treating WP. Containment technologies were not considered.
Both ex situ and in situ technologies may be applicable for treatment of soil and sludge in unlined ponds
containing WP. Ex situ technologies require excavation of soil and sludge followed by either disposal or
use of an above-ground treatment technology. When using ex situ technologies, there are additional
resources involved for performing excavation and disposal of treated residues than are required when
using in situ technologies. In situ technologies require addition of physical, chemical, or biological
agents directly to the soil and sludge without removal from the ponds. Some technologies, such as
solidification/stabilization or chemical oxidation, can be performed on either an ex situ or in situ basis.
Ex situ technologies require that a portion of the site be used to implement the technology, including
42 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
possibly staging the soil and sludge prior to and after treatment. Technologies that involve in situ
treatment would need to consider the effects of treatment on the physical and chemical characteristics of
the ponds.
The six specific treatment technologies discussed in this report include processes where the mechanism is
based on chemical fixation (1), hydrolysis (1), oxidation (2), and thermal treatment (2).
Solidification/Stabilization (S/S) - S/S is a commercially-available chemical fixation technology
commonly used to reduce the mobility of heavy metals and radionuclides in matrices such as soil
and sludge, and was identified in the EMF FS as a technology retained for further consideration
for treatment of soils and solids at the Astaris (previously FMC) Subarea. In addition, it was
selected for further testing on unlined ponds at a WP manufacturing facility in Tarpon Springs,
Florida. While typically performed ex situ, S/S also may be performed by injecting and mixing
stabilizing agents into unexcavated soil or sludge (i.e., in situ).
Caustic Hydrolysis - Caustic hydrolysis was selected for treatment of process waste streams
containing WP at the EMF site and hydrolysis was identified in the EMF FS as a technology
retained for further consideration for treatment of soils and solids at the Astaris Subarea. The
construction of a caustic hydrolysis system was approximately 95 percent complete when Astaris
stopped production of WP.
Chemical Oxidation - Chemical oxidation was identified in the EMF FS as a technology retained
for further consideration for treatment of soils and solids at the Astaris Subarea. The Tennessee
Valley Authority (TVA) reported that sludge containing WP would be oxidized when reacted
with nitric and sulfuric acids. Researchers at the University of Alabama developed and tested a
high speed air dispersion process to oxidize WP in a batch, stirred-tank reactor, and the USAGE
reported about several types of oxidants that can oxidize WP.
Mechanical Aeration - A mechanical aeration system has been suggested by the USAGE that
could be used to increase the exposure of WP particles to oxygen and increases its rate of
oxidation. The USAGE described several studies that were conducted to evaluate the degradation
of WP in solutions with varying amounts of DO and mixing intensities.
Incineration - Along with open burning, incineration was identified by the United Nations as a
recommended method for disposal of wastes containing WP (ATSDR, 1997). Incineration is a
commercially-available thermal treatment technology that could be used to oxidize WP and
reduce the volume of material requiring further treatment or disposal. Since 1990, the Crane
Army Ammunition Activity has used a rotary kiln furnace to oxidize WP from ordnance, and a
hydrator to convert the residual stream to phosphoric acid.
Thermal Desorption - Thermal desorption is a commercially-available technology that involves
heating the soil and sludge (directly or indirectly) to cause contaminants to volatilize and separate
from the solid matrices without combustion. The USAGE reported that 300 tons of WP-
contaminated soil were treated using thermal desorption at a site in Ogden, Utah. Thermal
desorption was reported as capable of removing WP, though the mechanism was not identified.
43 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
In addition to the six technologies listed above, pond pumping has been used to treat sediments
contaminated with WP. Pond pumping involves removal of free liquids from a pond and exposing pond
solids to ambient air and increase its rate of oxidation. At the Eagle River Flats military impact area
contaminated with WP, the U.S. Army Cold Regions Research & Engineering Laboratory (CRREL)
conducted a two-year feasibility study of a pond pumping system. However, at the EMF site, free liquids
were removed more than 20 years ago, and it is not clear how additional pumping would be conducted at
this time.
The six technologies were evaluated primarily based on how they would be applied for treatment of WP
in the historical ponds at EMF. Therefore, in some cases, additional technologies may need to be
considered for the treatment of heavy metals and radionuclides. The specific factors that were used to
evaluate each technology are (1) the mechanism by which the technology would treat WP as well as
heavy metals and radionuclides; (2) available data about where the technology previously has been used;
(3) engineering considerations, such as material handling, pretreatment, residual management, and health
and safety; (4) technology performance data; and (5) cost to implement the technology.
No testing of technologies was performed as part of this evaluation. Additional treatability testing would
be needed to verify the potential for a given technology to treat the historical ponds at EMF. Treatability
tests include evaluating how a particular treatment technology would perform for the specific matrices in
the historical ponds, and the variations in performance across the range of concentrations and properties
of the ponds. Results from such testing also could be used during the detailed design of a full-scale
system, and for estimating the cost of such a system.
Performance. As discussed in Section 1.0, limited data are available in the technical literature
concerning the performance of specific technologies for treating WP in soil and sludge. No full-scale
remediation projects are known to have been implemented to treat waste material similar to that found at
EMF. For this report, technical judgement was required for a qualitative assessment of expected
performance. When data were not available, the technology was evaluated based on its operational
principles and how those principles would affect WP, based on the physical and chemical properties of
WP.
Radionuclides and heavy metals have differences in their physical and chemical properties and the
treatment of radionuclides may require specific engineering considerations. However, radionuclides are
44 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
expected to behave similarly to heavy metals during treatment, and these two types of contaminants are
referred to as heavy metals in the remainder of this section, except as noted. In addition, all heavy metals
present were considered as one group of contaminants in this evaluation, recognizing that there are
differences in physical and chemical properties among specific heavy metals (e.g., arsenic and selenium
behave differently than cadmium and zinc). These types of differences are discussed in a variety of
reference materials, such as:
EPA Technology Screening Guide for Radioactively Contaminated Sites (EPA 402-R-96-017)
(Available on-line at )
• FRTR Screening Matrix and Reference Guide (December 4, 2001)
(Available on line at
Also as discussed in Section 1.0, no quantitative contaminant cleanup levels have been specified for soil
and sludge in the historical ponds. Therefore, this report discusses technology performance in general
terms, and does not draw conclusions about specific technologies or groups of technologies that may be
appropriate for use in treating historical ponds at EMF.
Cost. As mentioned in Section 1 .0, cost information provided for each treatment technology has a great
deal of uncertainty, given the limited amount of available data about the ponds, about use of technologies
to treat historical ponds at WP manufacturing sites, and use of technologies to treat soil and sludge
containing WP. Given these uncertainties, this report provides cost estimates as a range of costs, and
provides an analysis of the sensitivity of the cost estimates. Technology cost was estimated for treating
all 500,000 CY of contaminated soil and sludge estimated to be in the historical ponds at EMF. For
compatibility on unit costs (cost per unit of measure) among technologies, calculated costs assume a
density for pond materials of 1.5 tons per CY (refer to discussion of density in Section 4.0).
Cost data for specific technologies are based on information in commonly-used databases of remediation
market cost information and in other publicly-available sources. These include publications such as RS
Means Environmental Remediation Cost Data - Unit Price, 8th Annual Edition; 2002; published by
Environmental Cost Handling Options and Solutions (ECHOS, 2002), RSMeans Heavy Construction Cost
Data, 16th Annual Edition, published by RSMeans Company, Inc., (RSMeans, 2002), and Remediation
Technology Cost Compendium-Year 2000 (EPA, 2001). The unit cost information from ECHOS is based
45 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
on the costs for use of a technology at a typical remediation site, with minimal health and safety concerns
(referred to as Level E, which ECHOS calls "normal conditions"). ECHOS reports that performing work
at Safety Level C (which includes respiratory protection and other more significant activities) would
increase labor costs by 80% and equipment costs by 33%, while working at Level B (which includes all
the requirement of Level C plus supplied air respiratory protection) would increase labor costs by 140%
and equipment costs by 67%.
The base unit costs in ECHOS do not include costs for design, testing, oversight, contingency, profit, or
more stringent safety precautions (use of more stringent safety precautions would entail higher costs for
use of a technology). The application of a specific technology at a remediation site typically includes
costs for all of these items, however the estimates provided in this section are only for use of the treatment
technology and do not refer to the entire site cleanup.
Also as mentioned in Section 1.0, the estimates in this report focus specifically on the cost of treatment
and, in many cases, do not include costs for associated project components, such as excavation,
preprocessing of waste material, health and safety (such as ambient gas control), and residual
management (to treat heavy metals, if not treated by the primary treatment technology or disposal of
treated solids). The magnitude of the costs of these associated project components could vary
significantly depending on the specific characteristics of the material treated and treatment goals for the
site.
An analysis was conducted to generally compare the magnitude of these associated costs to the cost of
treatment under assumed site conditions and project requirements. A rough estimate showed that, while
the treatment cost likely would make up a bulk of the overall project cost, these costs of associated project
components likely would add 15 to 75 percent to the total project cost. This estimate for the associated
costs is based on specific assumptions, including that the waste material will be able to be handled as a
bulk solid rather than as a slurry and that the residual material and treated soil generated during treatment
will be able to be managed on-site at the EMF site rather than requiring off-site disposal. If these
assumptions are determined not to be reasonable after further site characterization or treatability testing,
estimates of total project costs could increase significantly. For example, material handling costs could
be increased by a factor of 10 if the waste material must be handled as slurries or disposal costs of
residuals could increase substantially if off-site disposal is required. In addition to these cost factors, off-
46 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
site disposal of waste or residuals from treatment at the EMF site may be challenging due to the relatively
few disposal sites that likely would take this waste.
Sources of Information. In addition to the ECHOS database and EPA's Cost Compendium, sources of
information used for this evaluation include remedial efforts conducted previously or considered for use
at the EMF site and other WP manufacturing sites, including those described in the FS for the EMF and
Tarpon Springs sites, reports about WP remediation prepared by the USAGE Waterways Experiment
Station (WES) and Cold Regions Research and Engineering Laboratory (CRREL), published articles
about WP treatment, the Remediation Technology Screening Matrix, and Cost and Performance Case
Studies, both provided by the Federal Remediation Technologies Roundtable (available on line at
) as well as information provided by specific technology vendors. Additional
information about technology vendors was obtained from a search of EPA Remediation and
Characterization Innovative Technologies (REACH IT ), a comprehensive
database of remedial technology information available on the internet, for treatment of phosphorus
(search conducted February 2002).
Cross-Technology Considerations
The implementation of any applicable treatment technology would include the following cross-
technology considerations:
Limited Information About Pond Characteristics - Because of the uncertainties about the physical and
chemical characteristics of each pond, a treatment technology would need to be sufficiently robust and
capable of treating soil and sludge that may be substantially different from that identified in Section 4.0 of
this report. The characteristics of the soil and sludge likely vary within a pond and from one pond to the
next, and the treatment technology would need to be able to perform adequately across a range of
conditions.
Limited Information About Technology Performance and Cost - WP manufacturing facilities have closed
unlined ponds in the past; however, no information could be located in the technical literature concerning
any treatment that may have been used in closing the ponds. In addition, technology vendors have
provided minimal information about the performance or cost for using specific technologies to treat soil
and sludge containing WP. The technologies identified in this report for treating this material would need
47 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
to be tested in both laboratory and field settings for interested parties to have an adequate degree of
confidence that they would be effective when implemented on a broad scale.
Health and Safety - WP is reactive in air and needs to be stored under water or in an oxygen-deficient
environment. Treatment technologies that require excavation or other movement of soil or sludge, or that
generate dust, may cause WP to become exposed to air. Plant workers and treatment technology
operators would need to follow stringent health and safety precautions about handling soil or sludge
containing WP. In addition, treatment technologies considered for the soil and sludge must be able to
accommodate the presence of WP, even if the technology is being used to treat other contaminants such
as heavy metals and radionuclides. Health and safety issues related to the treatment of radionuclides and
heavy metals would also have to be considered.
Multiple Contaminant Types - The soil and sludge contains multiple types of contaminants, such as WP,
heavy metals, and radionuclides, all of which may require treatment. It may be necessary to use a series
of technologies that collectively would be able to treat all the types of contaminants. The use of a series
of technologies is often referred to as a "treatment train".
Logistical Concerns - The historical ponds cover a physically-wide area and the facility has a variety of
other ponds, structures, waste piles, and production operations near the ponds. In addition, in some cases
there are newer, lined ponds adjacent to parts of the ponds which likely have engineered berms and piping
or other material transfer equipment nearby. Technologies that require excavation would need to consider
the possible changes in structural stability of remaining ponds, as well as issues associated with accessing
the soil and sludge in the ponds. Other relevant logistical issues include the need for storage facilities,
construction management, operation support, and control of ambient gas generated during treatment (for
example, the potential need to construct a tent over excavation or treatment areas to collect and control
ambient emissions).
Excavation
Excavation is the removal of contaminated soil or sludge by use of mechanical equipment. Excavation
equipment ranges widely from hand tools, such as pick axes and shovels, to backhoes, front end loaders,
clamshells, draglines, and hydraulic dredging equipment, depending on the amount of material to be
48 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
excavated, the total depth of the excavation, moisture content, and the space allowed at the site for staging
of excavated material.
Mechanism. Excavation physically removes contaminated soil and sludge from historical ponds before
treatment using ex situ technologies or off-site disposal, and is a common approach for performing site
cleanups. Excavation reduces the amount of contaminated soil and sludge in the ponds, and hence the
potential for migration of WP and heavy metals from the ponds. Confirmation sampling is used to verify
that contaminated materials have been removed. Excavation does not reduce the concentration of the
contaminants in the soil and sludge. Therefore, the excavated material requires additional treatment prior
to disposal. As discussed in Section 1.0, off-site disposal is not included within the scope of this report.
Available data. Ponds at the EMF site were partially excavated when active (i.e., more than 20 years
ago). No information was available about how or where the excavation process was performed at EMF.
Engineering considerations. WP in the soils and sludges may degrade into phosphorus pentoxide or
phosphine gases, or ignite spontaneously when soil and sludge is exposed to air. These gases and
spontaneous ignition may pose risks to workers during the excavation process. To minimize these
potential risks, safety training; special personal protective equipment (PPE), including respirators; and
emissions control equipment or structures would be needed. To reduce the potential for uncontrolled
burning, water is generally added and mechanically mixed with the soil and sludge prior to removal,
resulting in an aqueous slurry that is less prone to spontaneous ignition. The slurry would then be
pumped from the pond, using equipment such as centrifugal pumps or hydraulic dredges or excavators. It
may be necessary to perform pretreatment by crushing, grinding, or milling to break up large masses of
soil or slag prior to slurrying.
The cover material on the ponds may need to be partially or fully removed to excavate the contaminated
soil and sludge. In addition, ponds located around or near aboveground and below-ground structures may
require the use of underpinning or sheet piling to stabilize the structure prior to excavation, or exploratory
excavation and hand-digging. Shoring or sloping may be required in sandy soil to maintain trench wall
stability.
During the excavation or slurrying operations, air quality monitoring may be required. Use of mechanical
equipment to physically move soil and sludge may expose the waste material to air and water. As
49 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
discussed in Section 2.0, this exposure may generate fugitive air emissions, such as phosphine gas, and
these emissions may lead to elevated concentrations of dust or specific chemicals at down-gradient
receptor points. If these concentrations exceed allowable standards, it may be necessary to use
engineering controls on the excavation and slurrying operations, or to limit or reduce the quantity of soil
that can be excavated daily. It may be necessary to conduct these activities within a containment structure
(e.g., a tent) that incorporates air scrubber or carbon adsorption treatment equipment to control ambient
emissions and protect workers.
Performance. Excavation would remove contaminated soil and sludge from the ponds, but would not
reduce the mass of contaminants. By adding water to the materials, the concentration of contaminants
would be decreased based on the amount of water added.
Cost. The estimated unit cost reported by ECHOS for hydraulic excavation is $6/CY for a 3 CY trench
excavator, loose rock, 14 to 20 ft deep, and 104 CY/hour. If applied to the 500,000 CY of soil and sludge
at EMF, the total cost for excavation would be approximately $3 million. ECHOS reports costs for trench
hydraulic excavators ranging from $1 to 6 per CY, depending on type of material to be excavated, and
rate and depth of excavation. As discussed above, if the ponds materials are slurried for health and safety
reasons, the material may need to be removed using a dredging technique, such as hydraulic dredging.
The average estimated unit cost reported by RSMeans of $9.08/CY for hydraulic dredging is S7.30/CY
minimum and $10.85/CY maximum. If applied to the 500,000 CY of soil and sludge at EMF, the total
cost for excavation would be approximately $4.5 million. RSMeans assumes that 2,000 gallons of water
per CY of solid material is required during hydraulic dredging, increasing the volume of the material that
must be treated by approximately tenfold.
The costs for excavation and dredging identified in ECHOS assume minimal precautions taken to address
health and safety concerns (Level E). Because of the presence of WP in the pond materials, a higher level
of safety would need to be used, and these would result in higher costs for excavation. Excavation or
dredging using a higher level of safety equipment (say Level C) also would take longer to perform than
work done under Level E, because of the time needed for activities such as decontamination and
monitoring, thus increasing the cost for operating labor. Additional factors that affect excavation and
dredging costs include extent of underground infrastructure and need for shoring. Nearby structures and
extensive infrastructure, along with increasing depth of contamination tends to raise the costs. Therefore,
50 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
the actual cost for excavation or dredging at the EMF site would likely be considerably more than the cost
estimate of $3 to 4.5 million.
Solidification/Stabilization
Solidification/Stabilization (S/S) involves mixing a waste with binding agents, such as Portland cement,
to create a slurry, paste, or other semi-liquid state, and then providing time and conditions (e.g.,
temperature, humidity) for the mixture to set or cure into a solid form. The S/S process may also include
the addition of pH adjustment agents or other additives such as sulfur compounds. Key operating
parameters for an S/S process include the type and concentration of binding agents and other additives,
the proportion of stabilizing agents to waste, and the amount of time and conditions for curing.
Mechanism. S/S reduces the mobility of contaminants in a matrix through both physical and chemical
means. The S/S process physically binds or encloses contaminants within a stabilized mass and
chemically converts the contaminants into less soluble, mobile, or toxic forms. S/S is an established
technology for treatment of heavy metals and radionuclides (for example, see the Solidification/
Stabilization Resource Guide; EPA, 1999, and Solidification/Stabilization Use at Superfund Sites; EPA
2000). Depending on the type and concentration of binding agents and other additives, S/S may be able
to reduce the mobility of WP, heavy metals, and radionuclides in pond material.
Available data. No data have been identified on the effectiveness of S/S for treatment of WP in ponds,
and treatability testing likely would need to be performed to identify agents and additives that would be
effective. For the Stauffer Management Company (SMC) Superfund site, a WP manufacturing site in
Tarpon Springs, Florida, EPA selected in situ S/S as the remedy for pond material and contaminated soil
containing WP, heavy metals, and radionuclides. While the Tarpon Springs site is similar to the EMF site
in that it also was used to manufacture WP, there are many differences between these two sites;
specifically, the scale of the Tarpon Springs site is much smaller and elemental phosphorus is not a
primary contaminant at the Tarpon Springs site. Therefore, a direct comparison between these two sites
may not be appropriate. A summary of the planned remedial activities at the Tarpon Springs site is
provided in Appendix A.
At the Tarpon Springs site, a two-step process is being planned to identify a mix design for use of in situ
S/S for pond materials (identified as CaF sludge, gray dust, and black silty sand). In the first step, S/S
51 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
agents will be identified based on available literature, past experience, material availability, and potential
applicability. In the second step, laboratory testing will be performed to evaluate the effectiveness of the
S/S agents. The system will be designed to operate with reagents injected into the pond materials. The
Explanation of Significant Differences (BSD) for Tarpon Springs (EPA, 1999) required that S/S material
achieve a minimum unconfined compressive strength (UCS) of 50 pounds per square inch (psi) and a
maximum permeability of 1 x 10~5 cm/sec.
The consultant at Tarpon Springs reported that the agents to be tested might include coal combustion fly
ash, Portland cement, lime products (such as quicklime, hydrated lime, and lime kiln dust), cement kiln
dust, and soluble silicates. They suggested that Portland cement used in combination with silicate likely
would be the "preferred means" to implement in situ S/S, although it had not yet been tested at that site.
They reported that Portland cement is likely to be more readily available and of higher quality than other
reagents at the site.
The S/S treatment planned for Tarpon Springs includes control of fugitive dust emissions and extensive
ambient air monitoring. To reduce fugitive dust, the injection process likely will use water-slurried S/S
reagents and subsurface mixing. (O'Brien & Gere, 2000)
The S/S test program at Tarpon Springs is expected to take place in 2003; however, a specific schedule
has not yet been identified. Project managers are working with the local community to plan the testing
for this site. (Madalinski, 2002a) It should be noted that the type of phosphate ore at Tarpon Springs is
different from that used at EMF, with a relatively higher concentration of radium-226, and successful
results from the test program at Tarpon Springs would not necessarily mean that a similar test of S/S
would be successful at EMF. Table 5-1 contains a summary comparing the contaminants in the pond
materials at the Tarpons Springs site to the EMF site.
52 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Table 5-1. Comparison of Contaminants in Pond Materials
Tarpon Springs Site
EMF Site
Primary Contaminants
Metals/Inorganics
Radionuclides
Other
As, Sb, Be, Tl
Ra-226
PAHs
As, Cd, Cr, Fl, V, Zn
Pb-210,K-40,U-238, Gross
Alpha, Gross Beta
None
Maximum Phosphorus Concentrations
Total
Elemental
Not provided
0.854 mg/kg
86,000 mg/kg
Not provided
Note: Most Tarpon Springs pond material does not contain WP.
Sources: EPA Region 4, 1998; EPA 1998
The ROD for Tarpon Springs identified the total present worth cost for the selected remedy as
$9,356,000. It is important to note that this cost includes the in situ S/S process, as well as excavation,
consolidation, and capping of selected contaminated soil, and institutional controls. The FS for Tarpon
Springs shows a total volume of 14,795 cubic yards to be treated with S/S, at depths as great as 19 ft, with
a corresponding capital cost of $750,000 and no O&M costs. (Weston, 1996) The capital cost of
$750,000 corresponds to a calculated unit cost of $50/CY.
Engineering considerations. At the EMF site, S/S might be applied ex situ or in situ, and for some or all
of the ponds. For areas to be treated that contain WP, special precautions would be needed to address
safety concerns. Pre-processing of soil and sludge may be necessary to break up larger masses, such as
agglomerated WP. If used in situ, it may be necessary to inject chemicals through existing covers. If
used ex situ, soil and sludge would need to be excavated, and the solid residual from treatment (stabilized
mass) would need to be disposed or backfilled.
If S/S were conducted ex situ, the engineering considerations associated with excavation would need to be
addressed. Some portion of the excavated soil would likely require storage or staging prior to S/S. The
storage structure or staging area would need to include features that protect the health and safety of on-
site workers and prevent the release of soil or potentially hazardous off-gasses. S/S equipment would
require space for mixing and setting equipment. Soil and sludge may require crushing, grinding, milling,
or other pretreatment to break up large masses of soil or slag and to homogenize the soil.
53
August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
In situ S/S would reduce the health and safety hazards associated with excavation. While performing in
situ S/S on surface soils, safety precautions similar to those used for excavation may be required.
However, in situ S/S at greater depths would present less of a hazard to site workers because workers
would not come into direct contact with contaminated soil. However, large masses of slag or other
subsurface obstructions may impede the distribution of S/S binders and reagents and prevent uniform
mixing of the soil.
Bench- and pilot-scale treatability testing would be necessary to identify a mixture of S/S binding agents
and additives, time, and environmental conditions, that would be effective for treating the contaminants.
For in situ S/S, the immobilized soil would be left in place. For ex situ S/S, the immobilized soil would
need to be backfilled or disposed. Metals are generally less likely to leach when disposed of in a dry,
high pH environment like the one found at EMF. However, some metals such as arsenic may be more
likely to leach in a high pH environment.
Performance. S/S would reduce the mobility of heavy metals, thus reducing the potential for these
contaminants to migrate to the groundwater. However, S/S has not been demonstrated on WP and it is
difficult to estimate the anticipated level of effectiveness for pond materials without performing
treatability studies. S/S changes the pH of the soil and sludge, and this could lead to an increase in the
amount of phosphine gas generated. Depending on environmental conditions, there may also be concerns
about the long-term performance. In addition, the heterogeneity of the pond materials may make it
difficult to achieve uniform distribution of stabilizing agents throughout the ponds, due to the
heterogeneous nature of the ponds. This may lead to a reduced overall effectiveness for use of S/S,
especially when performed in situ. Data in the technical literature show that S/S can reduce the
leachability of heavy metals to below the levels identified for the RCRA toxicity characteristic (TC).
(EPA, 2000)
Cost. The estimated unit cost of in situ S/S at Tarpon Springs ($50/CY), if applied to all of the 500,000
CY of soil and sludge at EMF, would result in a total treatment cost of $25 million. This assumes that all
pond materials are treated with in situ S/S using a process similar to that considered for use at Tarpon
Springs. Performing ex situ S/S likely would result in a higher cost because of the additional efforts
needed for material handling. In addition, ex situ S/S would involve costs for excavation of the ponds and
disposal of residuals. S/S involves increasing material volume, which may lead to higher costs for
54 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
disposal. Further, should there be elevated levels of radionuclides present, the cost for landfilling
treatment residues would be relatively higher.
A commonly-used source of remediation market cost information (ECHOS) reports a unit cost for ex situ
S/S using Portland cement (Type I, bulk) of $78/ton. For other stabilizing agents, ECHOS reports unit
costs ranging from less than $10/ton (e.g., bottom ash at $2.1 I/ton, fly ash [Class F] at $9.00/ton, cement
kiln dust at $6.50/ton, and lime at $7.50/ton) to more than $100/ton (e.g., Portland cement Type K at
$105/ton, sodium silicate at $280/ton). Assuming aunit cost of $78/ton and a density of 1.5 tons/CY
(corresponding to $117/CY), the total cost for treatment of 500,000 CY of soil and sludge would be
approximately $60 million.
Caustic Hydrolysis
Caustic hydrolysis is a chemical process where WP reacts with lime and water at elevated temperature
and pressure to form various phosphite and phosphate compounds, as well as phosphine. This technology
uses complex processing equipment operated by personnel with a relatively higher level of training. Key
operating parameters include the amount of lime and water added, reaction pH, as well as the reactor
temperature and pressure. The process is based on the chemical transformation of WP by hydrolysis, as
discussed in Section 2.0. Caustic hydrolysis would reduce the concentration of WP, but depending on the
pH used, may result in the generation of significant amounts of phosphine gas as a by-product.
Mechanism. Caustic hydrolysis would convert WP in soil and sludge to various phosphite and phosphate
compounds, and heavy metals to metal oxides and hydroxides. The metal oxides and hydroxides
generally are less soluble than the metal compounds, and could be removed by filtration or settling.
Available data. No data have been identified on the effectiveness of caustic hydrolysis for treatment of
WP in ponds. However, before deciding to stop the manufacture of WP, EMF had partially constructed a
Land Disposal Restrictions (LDR) waste treatment system that is based on use of caustic hydrolysis for
treatment of hazardous wastes containing WP. The system was designed to respond to a Consent Decree
for the RCRA LDR, and to provide for plant water storage and recycle requirements. In addition to
hazardous waste streams, accumulated solids in Pond 18 (not a historical pond) had been planned to be
dredged and sent to the LDR treatment facility for clarification and treatment. (Meyer, 2002)
55 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
The LDR treatment facility includes size reduction equipment, reactor feed tanks, high-pressure waste
and lime feed pumps, static mixers, reactor vessels, flash separators, flash vapor condensers, treated slurry
coolers, solids blowdown tanks, and pumps for blowdown and treated slurry transfer.
The LDR system consists of three reactor feed tanks operated in parallel. The feed tanks maintain a
slurry temperature of 150 °F, a WP concentration of 2.15 percent, a suspended solids concentration of 18
percent, a retention time of 12 hours, and a throughput of 50 gpm. The reactor has a limitation on the
maximum suspended solid diameter present in its feed, and size reduction of solids to 100 microns
(pretreatment) may be needed.
The reactors are designed to operate at temperatures of up to 464 °F and pressures of up to 600 psig.
Waste slurry is pressurized and fed to the reactors at a design rate of 25 gpm. Lime (20 percent slurry) is
injected into the reactor feed line upstream of a static in-line mixer. High-pressure steam is injected into
the lower portion of the reactor to bring the slurry temperature up to the target level. Several chemical
reactions are expected to occur in the reactors, including reaction of WP with lime and water to form
calcium biphosphite, calcium hypophosphite, calcium phosphate, phosphine gas, and hydrogen gas, and
reaction of cyanide compounds with water to form gases such as ammonia.
Slurry material from the reactors is treated using lime for pH adjustment, and then filtered using a press.
Aqueous effluent from the press is treated with hydrogen peroxide and discharged or reused, while solid
residue (filter cake) is mixed with cement kiln dust to stabilize heavy metals prior to land disposal.
Gases generated in the reactors are treated using a thermal oxidizer to form a nominal 40 percent
phosphoric acid product stream that is intended to be distributed as a commercial product. Additional
equipment used to control air emissions includes a quench tank, quench vapor-liquid separator, particulate
scrubber, and polishing filter.
According to staff in EPA Region 10, the LDR system that is partially constructed at EMF has many
operational requirements (e.g., injection rates, particulate size cutoffs) that potentially could make
treatment of the waste from the historic ponds infeasible. For example, the process waste planned to be
treated by the LDR system would be more uniform than the waste in the historical ponds, and therefore
would require little or no pre-processing prior to treatment. As of December 2001, construction of the
56 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
caustic hydrolysis system at the site was 95 percent complete, but has been halted due to the plant shut-
down. (Madalinski, 2001)
Engineering considerations. Caustic hydrolysis is an ex situ treatment technology and engineering
considerations associated with excavation would need to be addressed, including storage or staging
structures that include features to protect the health and safety of on-site workers and prevent the release
of soil or potentially hazardous off-gasses.
Modifications (specific activities not identified) to the existing, partially-completed caustic hydrolysis
system would be required prior to it being used for treatment of historical pond material. If the existing
caustic hydrolysis unit could not be used, a new unit would need to be designed and constructed, and
space would need to be identified on site for the unit and associated equipment. Some of the likely
modifications that would be required to the existing system in order to treat the waste from the historic
ponds are described below.
Pond material may need to be pretreated, such as crushed, ground, or milled, prior to caustic hydrolysis to
break up larger masses of soil or slag and homogenize the material. The caustic hydrolysis treatment
process may also result in the evolution of hazardous off-gasses, such as phosphine, and off-gasses from
the treatment system will likely require collection and treatment. The existing on-site caustic hydrolysis
unit is designed to produce a marketable phosphoric acid product from the off-gas. Additional
characterization of the historical pond materials may be necessary to determine whether a marketable
product could be generated from the off-gas from treatment of these materials.
The caustic hydrolysis process would also generate a slurry of treated soil and water. The slurry would
need to be dewatered, and the resulting liquid and solid waste streams further managed. The liquid waste
stream would likely require pH adjustment, and may require additional treatment depending on
concentrations of metals or other contaminants present. The solids from dewatering may need to be
treated by S/S prior to disposal to immobilize metal contaminants.
Performance. Caustic hydrolysis is a robust technology that would be capable of treating soil and sludge
that may have substantial variations in characteristics (e.g., concentrations of WP and other contaminants,
moisture content), to reduce the concentration of WP and heavy metals. Given that the ponds are not well
characterized, use of a robust technology like caustic hydrolysis would lead to a higher likelihood that
57 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
technology performance would be more uniform across pond materials. However, the existing system at
the site was designed to handle a specific and mostly homogeneous waste stream, rather than the more
complex and heterogeneous material from the historic ponds. Therefore, additional testing and analysis
would be required to determine the modifications to the existing system that would be necessary to
adequately treat the pond materials.
Cost. According to EPA Region 10, approximately $120 million already has been invested in the LDR
treatment plant. An additional $13.5 million would be needed to complete plant construction, and $12 to
$15 million would be needed for startup and testing, according to FMC. This plant was designed to
receive a slurry material and to treat material from a RCRA pond (Pond 18) at the site. The cost to
retrofit the plant to treat material from the historic (CERCLA) ponds would be $30 to $35 million - not
including additional O&M costs to treat the pond material. This estimate was provided by FMC and
made prior to the closing of the manufacturing facility. The constituents from the non-CERCLA and
CERCLA ponds are similar although the properties of the materials likely would be different in part
because the CERCLA ponds are older than the slurry materials the plant was designed to receive.
(Meyer, 2002)
Chemical Oxidation
Chemical oxidation involves the addition of chemical agents to react with contaminants in the soil and
sludge to form oxidized by-products. Two types of ex situ chemical oxidation technologies have been
identified to treat WP: (1) nitric/sulfuric acid, and (2) oxygen, using a high speed air dispersion (HSAD)
process. The acid oxidation technology was developed and tested by researchers at TVA on sludge
containing WP. The effluent from this process was then reacted with ammonia to produce a nitrogen- and
phosphate-containing plant nutrient. The HSAD technology was developed and tested by researchers at
the University of Alabama on sludge from Astaris (previously FMC), TVA, and Occidental Chemical.
These processes are based on the chemical transformation of WP by oxidation, as discussed in Section
2.0.
In addition, others have suggested the possibility of performing chemical oxidation in situ, using a
flushing process (i.e., adding chemical oxidants to the pond materials, either through injection or an
alternative method for distributing the oxidants). In situ flushing has been demonstrated at remediation
58 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
sites contaminated with a variety of chemicals, however no information was identified about its possible
use for treatment of WP, and it is not discussed further in this report. (FRTR, 2002)
Mechanism. Chemical oxidation would convert WP to various phosphate compounds, however it would
not reduce the concentration of heavy metals.
Available data. Available data are presented separately for the nitric/sulfuric acid and HSAD processes.
Nitric/Sulfuric Acid Oxidation: TVA tested the nitric acid reaction process in a laboratory from 1991 to
1992, and constructed a pilot plant in 1993. For the laboratory test, a 5 5-gallon drum of WP sludge that
contains 7.5-7.9 percent WP was collected. Approximately 1 liter of sludge per batch was mixed with
acid, antifoam agent, and air in a glass vessel 90 mm ID by 122 cm height. The reactor contents were
heated to between 102-110 °C, agitated with a magnetic stirrer-heater, and held at the reaction
temperature for 2 hours. In the first series of tests, sludge was reacted with nitric acid (57-70 percent).
These reactions reduced WP to 5.9-9.9 mg/kg, and produced orthophosphate (47-57 percent of total
phosphorus), orthophosphite (43-53 percent), and large amounts of nitrogen oxide gases. The relatively
large fraction of orthophosphite produced was identified as problematic, since these compounds are toxic
to plants at high concentrations.
In a second series of tests, sludge was reacted with a mixture of sulfuric and nitric acids (weight ratio of
sulfuric to nitric acids ranged from 1:3 to 3:2). These reactions reduced WP to 0.8-3.2 mg/kg, and
produced orthophosphate (50-98 percent of the total phosphorus). These results showed that the fraction
of total WP converted to orthophosphate was more than 90 percent when the sulfuric to nitric acid ratio
was 1:1 or higher.
In 1992, a bench-scale facility designed to process approximately 20 liters of sludge per batch was
constructed and operated. This facility had a 316 stainless steel reactor that was 66 cm diameter by 100
cm high, and was operated with an acid ratio of 1:1. The results from the bench-scale tests were similar
to those from the laboratory test with WP in the reactor product ranging from 0.3-3 mg/kg, and
orthophosphate making up 97-99 percent of the total phosphorus. (Edwards, 1995)
A pilot plant was constructed in 1993 that was capable of processing 380 liters of sludge per batch. TVA
reported that operation of the pilot plant was successful with approximately 30,000 Ibs of WP sludge
converted to phosphate fertilizer. Operational concerns included sludge granulation, and the generation
59 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
of relatively large amounts of fertilizer and nitrogen oxide gases that would require further management
or disposal. They reported that operation on a larger scale may not be manageable because of these
concerns. TVA already had access to granulation facilities, but this could be a major concern if there
were none at a site because the granulation process is difficult to carry out. Results from the pilot study
were not published, and in-house data are not readily accessible. (Sherrill, 2002) Information was not
provided about the cost for construction or operation of the pilot plant or current status of the plant.
HSAD: Researchers at the University of Alabama developed and tested a HSAD technology to treat
sludges generated during production of WP. In this process, sludge is first treated to remove particles
greater than 48 mesh. Underflow material is treated in the HSAD wet oxidation process (reactor) by
sparging in air and oxygen, and agitating the sludge. Key operating parameters include oxygen and
sludge concentrations, and agitation intensity. The researchers reported that as long as the sludge
temperature was higher than 43°C, temperature was not a significant operating parameter. In the reactor,
WP is rapidly oxidized, with the test samples generally treated in less than one hour.
Tests were conducted using 180 gram samples of filtered sludge (source of sludge not identified), oxygen
concentrations of 20, 50, and 80 percent, sludge concentrations of 3.75, 7.5, and 15 percent, and agitation
intensities of 4,000, 6,000, and 8,000 rpm. The sludge sample contained WP at 44,000 mg/kg (4.4
percent). Table 5-2 provides a summary of the analytical data obtained during the test of the Astaris
sludge.
These results show that the concentration of WP was reduced substantially, but increased for heavy
metals. The toxicity characteristic leaching procedure (TCLP) results show that treatment residuals
exceed the TCLP limits for cadmium, chromium, and lead. The researchers indicated that treated solids
and liquids may have to be further treated for heavy metals before release to the environment (Jefcoat,
1995). Information was not provided about the projected cost for construction or operation of a full-scale
facility using this technology.
The USAGE reported that other oxidants also may be effective for treatment of WP, including ozone,
sodium hypochlorite, potassium permanganate, silver oxide, zinc metal, and potassium hydroxide. The
USAGE stated that these oxidants have not been field tested, and that the feasibility of applying these to
sediments would need additional research and experimental work. (USAGE, 1996)
60 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Engineering considerations. If chemical oxidation were conducted ex situ, the engineering
considerations associated with excavation would need to be addressed. Some portion of the excavated
soil would likely need storage or staging prior to treatment, with the storage structure or staging area
having features that protect the health and safety of on-site workers and prevent the release of soil or
potentially hazardous off-gasses.
At the EMF site, chemical oxidation might be applied by construction of a chemical reactor and treatment
using either sulfuric/nitric acid or the HSAD process. Construction of an above-ground processing
system, with associated design and operational considerations, would be needed for chemical oxidation.
Extensive developmental testing in a bench- or pilot-scale facility would be needed prior to construction
of a full-scale treatment facility. On-site treatment would require space on site for the chemical oxidation
unit and associated equipment. The nitric/sulfuric acid process involves the storage and use of strong
acids, which generally need special equipment and handling procedures. Preprocessing of pond material
using crushing, grinding, or milling prior to treatment to break up larger masses of soil or slag and
homogenize the material may be needed for both the nitric/sulfuric acid oxidation and HSAD processes.
61 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Table 5-2. Analytical Results from HSAD Sludge Test
Constituent
Concentration
Before Treatment
Concentration
After Treatment
Total Waste Analysis Results - mg/kg
WP - from solid residue
WP - from supernatant solution
Cadmium
Calcium
Chloride
Chromium
Copper
Fluoride
Lead
Magnesium
Nitrate
Nitrite
Phosphate
Phosphite
Sulfate
Zinc
44,000
160
0.018
8.54
59.5
0.006
0.013
112
0.312
3.99
ND
ND
1,116
1,645
52.9
1.064
37
0.08
95
2,290
174
10.0
0.34
1,750
18
310
ND
ND
6,440
2,400
150
2,275
TCLP Results - mg/L
Reacted Solids
Cadmium
Calcium
Chromium
Copper
Lead
Magnesium
Zinc
NR
NR
NR
NR
NR
NR
NR
47.0
377
3.0
9.0
8.0
9.0
43.7
Reacted Liquids
Cadmium
Calcium
Chromium
Copper
Lead
Magnesium
Zinc
NR
NR
NR
NR
NR
NR
NR
67.0
77.0
12.0
7.0
30.0
185
430
ND - not detected
NR - not reported
Source: Jefcoat, 1995
62
August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Both processes also are likely to generate hazardous off-gasses, such as phosphine, that would need to be
collected and treated prior to release to the atmosphere. EPA Region 10 has also noted that the pond
material also has the potential to release hydrogen cyanide at the lower pH conditions that are associated
with chemical oxidation. Both processes generate a sludge residue, which would need to be dewatered,
and the resulting liquid and solid waste streams further managed. The liquid waste stream from the
nitric/sulfuric acid process would likely need pH adjustment and additional treatment to remove metals
dissolved by the strong acids used in the process. Solid residues from the nitric/sulfuric acid process may
be able to be sold for their commercial value as a fertilizer, or, if residual metal concentrations are too
high, disposed of as a by-product. For both processes, S/S of solids from dewatering may be needed prior
to disposal, to further treat the heavy metals present. Although there may commercial outlets for the
liquid and solid waste streams generated by this treatment, it is also possible that the waste streams will
be deemed to be hazardous waste and will require transport and disposal at a permitted hazardous waste
disposal facility.
Performance. Analytical data available from the two chemical oxidation processes discussed above
indicate that the concentration of WP in the reactor product ranged from 0.08 to 37 mg/kg, depending on
process used and initial concentration. The HSAD process showed an increase in concentrations of heavy
metals, with some results exceeding the TCLP, while data were not provided about the concentration of
heavy metals in the reactor product from the acid treatment process. It is likely that acid treatment would
also increase the concentrations of heavy metals in the reactor product due to the solubility of heavy
metals in acidic conditions. Further treatment of reactor product for heavy metals likely would be
necessary for using either technology.
Cost. Information was not provided about the cost for using either of these technologies on a full-scale
basis to treat 500,000 CY of pond material. The costs for use of this technology would likely be higher
than ex situ S/S (estimated as $60 million) because of the need for additional capital equipment. Both S/S
and chemical oxidation technologies require chemical reaction equipment (e.g., reactor, piping, mixing)
and the chemical oxidation processes likely would have additional costs for pretreatment (e.g., size
reduction), treatment of heavy metals, emission controls, and possibly for disposal of solid residues
containing radionuclides.
63 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Mechanical Aeration
Mechanical aeration is the use of physical equipment, such as a mixer, to agitate a solid or slurry material
and expose the material to ambient conditions, where ambient oxygen would react with and oxidize WP.
The USAGE indicated that a mechanical aeration system could be used in situ or in a vessel (ex situ) to
expose WP particles to oxygen. In addition to mixing rate, operational parameters that would affect
performance would include the concentration of dissolved oxygen (DO), matrix effects, depth of
treatment, temperature, and pH. The USAGE stated that mechanical aeration has not been field tested,
and that the feasibility of applying this to WP-contaminated sediments would need additional research
and experimental work. They did not suggest a design for an in situ system. (Rivera, 1996)
Mechanism. Mechanical aeration would convert WP to various phosphate compounds, however it would
not reduce the concentration of heavy metals. Aeration will also convert elemental phosphorus to
phosphine or to phosphorus pentoxide.
Available data. The USAGE provided a summary of research performed by Lai, and showed that use of
mechanical aeration would reduce the concentration of WP in an aqueous solution from near 1,000 ug/L
to less than 10 ug/L in 60 days. They also provided results from experiments that showed that mixing at
30 and 60 rpm released WP 4 and 6 times faster than without mixing, respectively. (Rivera, 1996)
Engineering considerations. At the EMF site, mechanical aeration might be applied in situ or ex situ to
treat the ponds. When applied in situ, a mechanical aeration system likely would need some type of
hydraulic or pneumatic injection and mixing equipment. When applied ex situ, excavation of pond
materials, and construction and operation of a reactor vessel would be needed.
If mechanical aeration were conducted ex situ, the engineering considerations associated with excavation
would need to be addressed. Some portion of the excavated soil would likely need storage or staging
prior to treatment, with the storage structure or staging area having features that protect the health and
safety of on-site workers and prevent the release of soil or potentially hazardous off-gasses. Aeration
equipment would need space for processing equipment. Pretreatment of soil and sludge may include
crushing, grinding, or milling to break up large masses of soil or slag and to homogenize the soil. Both in
situ and ex situ aeration would likely generate hazardous off gasses, such as phosphine, during treatment
that would need to be collected and treated. Emissions during in situ aeration likely would be more
64 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
difficult to control than during ex situ aeration because the treatment reaction would not be conducted in a
controlled, engineered environment.
In situ mechanical aeration would reduce the health and safety hazards associated with excavation.
However, large masses of slag or other subsurface obstructions may impede the distribution of air and
prevent uniform mixing of the soil.
Bench- and pilot-scale treatability testing would be necessary to identify the processing equipment, time,
and environmental conditions that would be effective for treating the contaminants.
For in situ treatment, the aerated soil would be left in place. For ex situ treatment, the aerated soil would
need to be backfilled or disposed.
Performance. Information was not provided about the effectiveness of using mechanical aeration to treat
soil and sludge containing WP in situ or in a vessel. Mechanical aeration would likely reduce the
concentrations of WP particles that are exposed to oxygen, based on the results provided by the USAGE.
However, it is not known whether aeration would be able to reduce the concentrations of WP to below
cleanup goals (which have not been established) or how well the oxygen would be distributed throughout
the pond materials, given the heterogeneous nature of the ponds. Mechanical aeration would not reduce
the concentrations of heavy metals in the ponds, and another treatment method, such as S/S, would be
needed to treat heavy metals.
Cost. Information was not provided about the cost for using mechanical aeration technology to treat
500,000 CY of pond material. The cost for use of this technology would likely be at least as much as for
use of S/S ($25 million for in situ, $60 million for ex situ). Both technologies require mechanical
injection equipment (for chemicals or oxygen) and the mechanical aeration process likely would have
additional costs for treatment of heavy metals. If used ex situ, there might be additional costs for disposal
of solid residues containing radionuclides.
65 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Incineration
Incineration is a thermal process where soil, sludge, or other wastes are treated at elevated temperatures
(1,400 to 2,200°F) to volatilize and combust contaminants. Incinerators typically are constructed using
primary and secondary combustion chambers, and use a variety of off-gas treatment equipment, including
filtration, wet-scrubbing, electrostatic precipitator, baghouse, vapor-phase carbon adsorption, and thermal
oxidation.
The technology is typically used for treatment of organic compounds such as hydrocarbons and to reduce
waste volume, and its use for that purpose is well documented (for example, see Cost and Performance
Remediation Case Studies [FRTR, 2001]). Incineration of hazardous wastes has regulatory requirements
for destruction and removal efficiency (DRE) and stack emissions.
Mechanism. Incineration of excavated pond material would reduce the concentration of WP and volume
of pond material. WP would be oxidized and natural organic matter in the soil and sludge would be
converted to carbon dioxide and water. Oxidized WP would be captured by off-gas treatment systems
such as scrubbers and treated further or discharged as a by-product. Heavy metals in the pond materials
would not be oxidized, and mainly would be concentrated in the residual ash from the incinerator. More
volatile heavy metals, such as arsenic, likely would be volatilized.
Available data. The U.S. Army has constructed and operated a facility to convert WP in ordnance to
phosphoric acid using a rotary kiln furnace. The facility, referred to as WP to Phosphoric Acid
Conversion (WP-PAC), is located at the Crane Army Ammunition Activity (CAAA), in Crane, Indiana.
A pilot plant for this facility was designed by the Army at the Tooele Army Depot and operated over a 4-
month period in the early 1980s, producing 2.2 million pounds of acid. The facility was subsequently
moved to CAAA with the transition completed by 1990.
At the WP-PAC facility, a hole is punched in ordnance using a 115 ton hydraulic press at ambient
temperature, exposing the WP in the ordnance to oxygen. The ordnance items are then pushed into the
first section of a rotating kiln furnace. The heat in the furnace melts the WP, which then flows out of the
punched hole. The WP is then burned in the oxygen rich furnace to form phosphorus pentoxide. The
Army reported that there is complete burning of the WP in the furnace. The final section of the furnace
66 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
melts and oxidizes any remaining WP. No data were provided to identify the concentrations of WP in
residual materials.
Off-gases from the furnace, containing phosphorus pentoxide, are pulled through water-jacketed
ductwork into a co-flow/counterflow hydrator, at a temperature of approximately 650°C. Seventy to
eighty percent of the phosphoric acid is produced in the hydrator. Product acid is filtered to remove
suspended solids greater than or equal to 1 micron, and the acid is transferred to storage tanks. Exhaust
gases from the hydrator flow to a variable throat venturi scrubber unit, at a temperature of approximately
140°C. The venturi has two tangential nozzles that spray dilute acid into the gas stream in the throat area,
and hydrate residual phosphorus pentoxide.
As of August 1991, the facility had operated for more than 7,000 hours, and 1.5 million rounds containing
4.3 million pounds of WP had been processed to form 18 million pounds of phosphoric acid.
Environmental permitting was limited to notifying the Indiana Department of Environmental
Management of an air pollution source. The Army reported that no permit was required under RCRA
because the facility converts hazardous waste into a usable product (Burcham, 1991). Information was
not provided about the current status of the WP-PAC facility and whether it might be available for use by
commercial companies such as WP manufacturers. In addition, information was not provided about the
cost for construction or operation of the WP-PAC facility.
Engineering considerations. At the EMF site, incineration might be applied to treat pond materials
contaminated with WP. Construction of an on-site incinerator would involve a potentially wide range of
requirements associated with siting, permitting, and operation. Use of an off-site incinerator (e.g., at a
commercial treatment, storage, or disposal facility [TSDF]) would involve transportation of the pond
materials to the off-site location, raising additional safety issues concerning the potential exposure of
populations other than plant workers to WP. However, a discussion of off-site treatment is not within the
scope of this report.
If incineration were performed at EMF, the engineering considerations associated with excavation would
need to be addressed. Some portion of the excavated soil would likely need to be stored or staged prior to
treatment, using structures/areas including features that protect the health and safety of on-site workers
and prevent the release of soil or potentially hazardous off-gasses. Space also would be needed for
incineration and off-gas treatment equipment at the site. Soil and sludge may need to be pretreated by
67 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
crushing, grinding, or milling to break up large masses of soil or slag and to homogenize the soil,
however this type of pretreatment likely would be less than for other, less robust technologies.
Incineration would likely generate hazardous off gasses, such as phosphine, during treatment which
would need to be collected and treated.
Incineration typically requires performance of a trial burn prior to use at full-scale. A trial burn is used to
test the DRE and stack emissions during treatment of one or more difficult-to-treat compounds, and is
used to help determine incinerator operating conditions. Additional operational concerns include quantity
of fuel needed to sustain combustion and the potential for slagging. The soil and sludge in the historical
ponds have relatively low BTU values, and, when slurried during excavation, would likely have a
relatively high moisture content (see discussion on excavation). Therefore, the amount of fuel required to
incinerate soil with a low BTU value would need to be considered. The elevated concentrations of
inorganic contaminants and minerals in the sludge and soil may result in slagging, which may be retained
in the incinerator ash and reduce throughput. (EPA, 1998a)
The high temperatures used in incineration will likely cause some of the metal contaminants in the soil,
such as arsenic, cadmium, polonium, and lead to volatilize or fume, resulting in metals in the incinerator
off-gas. The incinerator will need to be equipped with appropriate air pollution control devices to control
emissions of these toxic metals. Metal contaminants are likely also to be retained in the incinerator ash,
and additional treatment of the ash, such as with S/S, may be required prior to disposal.
Performance. Incineration is a robust technology that would be capable of treating soil and sludge that
may have substantial variations in characteristics (e.g., concentrations of WP and other contaminants,
moisture content) to reduce the concentration of WP and heavy metals. Given that the ponds are not well
characterized, use of a robust technology like incineration would lead to a higher likelihood that
technology performance would be more uniform across pond materials. Data from the CAAA showed
that incineration is effective in reducing the concentrations of WP in ordnance, and is likely to be
similarly effective for oxidizing WP particles in sludge and soil.
Incineration may be a more likely technology for pond materials that have relatively high concentrations
of WP, but low concentrations of heavy metals. Data available in the technical literature shows that
incineration concentrates the mass of heavy metals, including radionuclides, in the residual ash from an
incinerator. (EPA, 1998a)
68 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Cost. The estimated unit cost for on-site incineration reported by ECHOS is $310/ton. If applied to the
500,000 CY of soil and sludge at EMF, the total treatment cost would be approximately $230 million.
EPA's Remediation Technology Cost Compendium-Year 2000 reported calculated unit costs for
incineration of soil and sludge as ranging from $160/ton to $l,400/ton. Typically, there are economies-
of-scale when treating relatively larger quantities and the actual cost for treating a relatively large volume
may be closer to the lower end of this range. Assuming a unit cost of $ 160/ton, the total cost for
treatment of 500,000 CY of soil and sludge would be approximately $120 million. As with other ex situ
technologies, this cost would be in addition to that for performing excavation.
Thermal Desorption
Thermal desorption is used to treat soil and sludge by heating (directly or indirectly) to volatilize
contaminants and separate them from the solid matrices without combustion. The temperatures used in a
thermal desorber are lower than in an incinerator, generally on the order of 200°F to 600°F. The
volatilized contaminants (vapors) are collected and generally are treated by one or more off-gas treatment
technologies. Types of off-gas treatment include filtration, wet-scrubbing, electrostatic precipitator,
baghouse, vapor-phase carbon adsorption, and thermal oxidation. Common configurations include rotary
kiln, thermal screw, and infrared (with infrared treatment, soil and sludge to be treated typically are
placed in a 5 CY tray [rectangular box]). The rotary kiln and thermal screw typically operate on a
continuous basis, with heat applied by combustion of hydrocarbon sources, while the infrared system
operates on a batch basis.
The technology is typically used for treatment of organic compounds such as hydrocarbons and to reduce
waste volume, and its use for that purpose is well documented (for example, see Cost and Performance
Remediation Case Studies.) (FRTR, 2001)
Mechanism. Thermal desorption would reduce the concentrations of WP through volatilization, and WP
vapors could be recovered by treatment of the off gases. Thermal desorption would not reduce the
concentrations of heavy metals in the ponds.
Available data. The USAGE reported that a patented infrared system, operating on a batch basis, was
used to treat 300 tons of WP-contaminated soil in Ogden, Utah at a cost of $80,000 ($267/ton). The
treatment system was provided by McLaren-Hart Environmental Engineering Corporation; however, the
69 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
source of the WP-contaminated soil and the period of operation for the treatment was not provided.
Operational parameters included a soil treatment temperature of 260°F, an average treatment time of 1 hr
40 minutes, treatment of 107 soil batches (nominally 5 CY each), and a total treatment time of 4 days.
The USAGE reported that one instance of auto-ignition of WP was detected inside the desorber and that
the condition was controlled (control method not identified). Information was not provided about the
concentrations of WP or other contaminants in the soil before or after treatment, or about the disposition
of treatment residuals. (USAGE, 1996)
Recent attempts to contact McLaren-Hart revealed that the company was sold to ENSR. ENSR personnel
were not familiar with the current availability of the infrared desorption unit. (Stewart, 2001)
Engineering considerations. Thermal desorption is an ex situ technology and the engineering
considerations associated with excavation would need to be addressed. Some portion of the excavated
soil would likely need to be stored or staged prior to treatment with the structure/area needing to include
features that protect the health and safety of on-site workers and prevent the release of soil or potentially
hazardous off-gasses. Space for thermal desorption and off-gas treatment equipment would be needed at
the site. Pretreatment of soil and sludge including crushing, grinding, or milling may be needed to break
up large masses of soil or slag and to homogenize the material.
Operational concerns include quantity of fuel needed to sustain desorption and possible fire or explosion
hazards. The soil and sludge in the historical ponds have relatively low BTU values, and, when slurried
during excavation, would likely have a relatively high moisture content (see discussion on excavation).
Therefore, thermal desorption would likely require a significant consumption of fuel in order to reach and
maintain treatment temperatures. Hot spots of WP contamination may present a fire or explosion hazard
if it spontaneously ignites during the thermal desorption process.
Thermal desorption temperatures generally do not cause metals to volatilize or fume. Therefore, metal
contamination in the off-gas will likely not occur. However, thermal desorption off-gasses may contain
hazardous chemicals, such as phosphine, and require collection and treatment. Thermal desorption does
not treat metals, and these will be retained in the solid residue from desorption. S/S or other treatment of
solid residues may be needed prior to disposal.
70 August 2003
-------
, Treatment Technologies for Soil and Sludge in Historical Ponds
Performance. Information was not provided about the effectiveness of using thermal desorption to treat
WP. This technology would likely reduce the concentrations of WP through volatilization, and that WP
vapors could then be recovered through off gas treatment. Thermal desorption generally does not reduce
the concentrations of heavy metals in soil or sludge (at higher temperatures it may reduce the
concentration of some more volatile metals).
Cost. The estimated unit cost reported by the USAGE is $267/ton. If applied to the 500,000 CY of soil
and sludge at EMF, the total treatment cost would be approximately $200 million. ECHOS reports that
the cost for thermal desorption ranges from $24 to $97 per ton treated, corresponding to a projected total
cost ranging from $18 to $73 million for treatment of 500,000 CY of soil and sludge. EPA's Remediation
Technology Cost Compendium-Year 2000 reported calculated unit costs for treatment of more than
100,000 tons of soil as less than $50/ton, within the range provided by ECHOS. Using a unit cost of
$50/ton for the 500,000 CY of soil and sludge at the EMF site, the total treatment cost would be
approximately $38 million. These cost estimates are only for the use of thermal desorption and do not
include the costs for excavation of soil and sludge, or for disposal of treated materials.
71 August 2003
-------
. Summary and Discussion of Findings
6.0 SUMMARY AND DISCUSSION OF FINDINGS
A summary and review of available information from the technical literature and previous studies was
conducted to identify potentially applicable treatment technologies for soil and sludge in the historical
ponds contaminated with WP, heavy metals, and radionuclides at the EMF Superfund site. No testing of
treatment technologies (e.g., in a laboratory or at a pilot-scale) was performed for this report.
Limited information was available to characterize the historical ponds. For this report, several
assumptions were made about pond characteristics based on what is known about the ponds and
experience with site cleanups. The portions of the historical ponds not covered by RCRA ponds cover a
total area of approximately 16 acres and extend to a depth of 20 ft; contain approximately 500,000 cubic
yards of soil and sludge; are contaminated with WP, heavy metals, and radionuclides (radionuclides are
expected to behave in a manner similar to heavy metals relative to treatment); and are heterogeneous in
physical and chemical composition. WP in the historical ponds at EMF would likely remain in the pond
soil and sludge as WP or as a solid metal phosphate. WP would not exhibit substantial oxidation or
hydrolysis, which are the major degradation processes for WP, nor likely be transported from the ponds.
Specific factors used to evaluate each technology were based on (1) the mechanism by which the
technology would treat WP, heavy metals, and radionuclides; (2) available data about where the
technology previously has been used or work conducted at similar sites; (3) engineering considerations,
such as material handling, pretreatment, residual management, and health and safety; (4) technology
performance data, in terms of reducing the concentration, volume, or mobility of WP, heavy metals, and
radionuclides; and (5) cost to implement the technology.
The following technologies were identified as potentially applicable:
S/S
Caustic Hydrolysis
Chemical Oxidation
Mechanical Aeration
Incineration
Thermal Desorption
72 August 2003
-------
. Summary and Discussion of Findings
None of the six technologies has been used at full-scale to treat waste material similar to that found at
EMF. Only limited information about remediation of WP is provided in the available literature with
fewer than 10 studies identified about the use of treatment technologies forWP. This primarily includes
studies performed at WP manufacturing facilities and at military facilities. The six technologies were
identified in the EMF Feasibility Study (FS) report prepared in 1996. That is, no new treatment
technologies have emerged as potentially applicable to treat the historical ponds since the FS report.
The remedy used for similar waste at other WP manufacturing facilities was primarily capping. Eight
other WP manufacturing sites were identified that have similar contaminated unlined ponds as the EMF
site. Six of the eight sites have installed or plan to install caps. Of the remaining two sites, one site
(Rhodia, Silver Bow, Montana) indicated that the ponds are not under corrective action, and another site
(Stauffer Management Company, Tarpon Springs, Florida) is evaluating a remedy of in situ S/S.
From information currently available, minimal performance data exists for using the six technologies to
treat similar waste material as found at EMF. Performance data were identified for treatment of WP using
chemical oxidation, mechanical aeration, and incineration. However, these data are not for treatment of
wastes in historical ponds at a WP manufacturing facility. Performance data for chemical oxidation and
mechanical aeration are for work at bench- and pilot-scale, while data for incineration are for ordnance
wastes with a higher percentage of WP than found in the historical ponds.
Thermal desorption was used for the treatment of WP in contaminated soil. Recent attempts to obtain
specific information on the project revealed that the technology vendor was sold to another company.
Personnel at this company were not familiar with the current availability of the technology. Therefore, it
is unknown if the WP-contaminated soil was similar to waste material as found at EMF and no specific
performance data were available for review.
No performance data were identified for treatment of WP using S/S and caustic hydrolysis. Both
technologies have been under consideration to treat similar waste material at WP manufacturing sites.
The Stauffer site in Tarpon Springs, Florida is planning to test in situ S/S in 2003. The test program
involves identifying S/S agents and evaluating the effectiveness of the S/S agents through laboratory
testing. Although the information gained from this test program could be used to evaluate the
effectiveness of S/S at EMF, the test program at Tarpon Springs may not be an accurate predictor. The
type of phosphate ore used at Tarpon Springs is different from that used at EMF (e.g., concentration of
73 August 2003
-------
. Summary and Discussion of Findings
the contaminant radium-226 being relatively higher and concentrations of WP being relatively lower at
the Tarpon Springs site). Therefore, the results from the test program at Tarpon Springs would not
necessarily mean similar results at EMF. A summary of the Stauffer site is included as Appendix A.
Caustic hydrolysis was considered for use at the Rhodia site in Silver Bow (Butte), Montana. In addition,
caustic hydrolysis was identified as the Land Disposal Restrictions treatment technology for process
wastes at EMF, but construction of this technology was halted with the plant shutdown.
Additional testing would be necessary to assess whether treatment technologies could perform adequately
across a range of contaminant concentrations and properties of the waste material. Extensive site
assessment and treatability testing would be needed to verify the potential for any technology to treat the
substantial physical and chemical heterogeneity of the ponds at EMF. Treatability tests would include
evaluating how the technology would perform for the specific matrices in the different ponds, and the
variations in performance across the range of concentrations and physical properties.
Although the six technologies are at various stages of commercial development, all of these technologies
would require testing to establish that they could perform reliably on the waste material in the historical
ponds. S/S, chemical oxidation, incineration, and thermal desorption have been applied commercially at
full-scale for site remediation, but have not been used to treat WP pond material. Caustic hydrolysis and
mechanical aeration have not been used extensively for site remediation, and significant developmental
testing would be entailed for scale-up along with the treatability testing identified above. Developmental
and treatability testing for the six technologies would require additional time and resources to undertake.
The soil and sludge in the historical ponds contain multiple types of contaminants which all may require
treatment. Therefore, a series of technologies may be necessary to collectively treat all the types of
contaminants. However, incineration, thermal desorption, mechanical aeration, and chemical oxidation
would not be able to treat heavy metals or radionuclides. In these cases, an additional treatment process
would likely be needed, such as S/S, using what is often referred to as a "treatment train".
Of the six technologies, only S/S and caustic hydrolysis have the potential to be effective for treatment of
heavy metals and radionuclides. S/S is applied frequently at full-scale to reduce the mobility of heavy
metals at contaminated sites. Caustic hydrolysis would convert heavy metals to metal oxides and
hydroxides, which generally are less soluble than the metal compounds, and could be removed by
filtration or settling processes.
74 August 2003
-------
. Summary and Discussion of Findings
In applying any of the six technologies to treat the historical ponds at EMF, there are several cross-
technology considerations. Soil and sludge may require preprocessing such as crushing, grinding, or
milling, to break up large masses of soil and sludge and to homogenize the soil. Preprocessing may be
necessary for both ex situ and in situ technologies, depending on the distribution of contaminants in the
ponds and the methods used to implement the treatment technologies. Residuals from treatment, such as
solid, liquid, or gaseous residues, would require further management. Residuals management may
include characterizing and transporting these residuals to a storage or disposal facility (on- or off-site), or
performing further treatment (such as for off gases) prior to release to the environment, which could
significantly increase overall treatment costs.
The estimated volume of waste material to treat (500,000 cubic yards) would entail a large construction
project, including significant engineering issues. The physical layout of the site, where historical ponds
are located near RCRA ponds, structures, and slag piles, may impact the implementation of a treatment
technology, including the need for space to stage equipment or to store material before or after treatment.
WP is an inorganic compound that ignites spontaneously in warm air. It is toxic by ingestion and
inhalation and skin contact with WP causes burns. Therefore, site workers would need to follow stringent
health and safety precautions about handling soil or sludge containing WP. Level C personal protective
equipment (respiratory and skin contact protection) would likely need to be used when conducting work
on the soil or sludge in the historical ponds. In addition, health and safety precautions related to metals
and radionuclides would also have to be considered.
Based on the criteria used to identify high cost projects by EPA's National Remedy Review Board
(NRRB), the specific costs to implement any of the six treatment technologies would be high. The NRRB
identifies high cost remedial actions as those that cost more than $30 million or more than $10 million
and 50 percent greater in cost than the least costly cleanup alternative. The estimated treatment costs for
the six technologies identified range from $25 to $230 million per technology, depending on type of
technology and whether it is performed ex situ or in situ, as well as the assumptions presented in
Section 4.
The level of uncertainty for the cost estimates provided in this report to implement the six technologies at
the EMF site is high. The estimates presented in this report represent a range of possible costs for the
treatment of the EMF waste. This range is highly dependent on the assumptions used to characterize the
75 August 2003
-------
. Summary and Discussion of Findings
historical ponds and the specific design and operating conditions used for implementing the technology.
It should be noted that the review of historical cost data of technology applications indicate costs are
highly variable, impacted by many factors, and that those factors are site-specific.
Although the technology cost estimates in this report could be above or below the actual costs, the actual
total treatment costs are likely to be higher. Specifically, the technology costs estimates do not include
costs for associated project components, such as excavation (if required), preprocessing of waste material,
health and safety (such as ambient gas control), and residual management, which could be integral parts
of a remediation project at the EMF site using any of the six technologies. For example, all the ex situ
technologies would incur costs for excavation, preprocessing of contaminated material, and disposal of
treatment residuals. Costs for associated components, such as excavation, preprocessing of waste
material, health and safety, and residual management likely would add 15 to 75 percent to the cost of
treatment using any of the six technologies. This estimate for the associated costs is based on specific
assumptions, including that the waste material will be able to be handled as a bulk solid rather than as a
slurry and that the residual material and treated soil generated during treatment will be able to be
managed on-site at the EMF site rather than requiring off-site disposal. If these assumptions are
determined not to be reasonable after further site characterization or treatability testing, estimates of total
project costs could increase significantly. For example, material handling costs could be increased by a
factor of 10 if the waste material must be handled as slurries or disposal costs of residuals could increase
substantially if off-site disposal is required. In addition to these cost factors, off-site disposal of waste or
residuals from treatment at the EMF site may be challenging due to the relatively few disposal sites that
likely would take this waste.
76 August 2003
-------
References
7.0 REFERENCES
ATSDR. 1997. Toxicological Profile For White Phosphorus. Prepared by Sciences International, Inc.
under subcontract to Research Triangle Institute, Under Contract No. 205-93-0606. September.
Brannen, John V., Organic Solutions, 2001. Fax to Susan Stewart, Tetra Tech EM Inc. Information
About Remediation Plus. October 8.
Burcham, Randall W., Crane Army Ammunition Plant. 1991. White Phosphorus to Phosphoric Acid
Conversion Facility, Presented at the International Seminar on Demilitarization Technology for
Explosives and Explosives Ordnance. November.
Collins, Charles M., USACE-CRREL. 1999. Remediation of a White Phosphorus Contaminated Salt
Marsh, Eagle River Flats, Alaska, Alaska Section, American Water Resources Association, Northern
Abstract for Brown Bag. February.
Edwards, Ronald E., Jack M. Sullivan, and Oscar E. Moore, TVA. 1995. Recovery of Phosphates from
Elemental Phosphorus-Bearing Wastes. American Chemical Society. Environmental Research Center,
Tennessee Valley Authority.
Environmental Cost Handling Options and Solutions (ECHOS). 2002. RS Means Environmental
Remediation Cost Data - Unit Price, 8th Annual Edition.
EPA. 2002. Memorandum from Michael F. Gearheard, Director, Office of Environmental Cleanup,
EPA, Region 10 to Walter W. Kovalick, Jr., Ph.D., Director, Technology Innovation Office, EPA Office
of Solid Waste and Emergency Response, regarding Region 10 Comments on the Draft Report of
4/30/2002 to Evaluate "Treatment Technologies for Historical Ponds Containing Elemental Phosphorus -
Summary and Evaluation". July 2.
EPA. 2001. Remediation Technology Cost Compendium - Year 2000. EPA 542-R-01-009. September.
EPA. 1998. Superfund Record of Decision for EMF Contamination, Pocatello, Idaho. EPA541-R-
98-034. June.
EPA. 1998a. On-Site Incineration: Overview of Superfund Operating Experience. EPA 542-R-97-012.
March.
EPA REACH IT. 2002. EPA Remediation and Characterization Innovative Technologies Web Site.
www.epareachit.org. February.
FMC. 2001. LDR Waste Treatment System Engineering Package Update. September.
FMC. 1996. Remedial Investigation for EMF Site. August.
FMC. 1996. Feasibility Study for EMF Site. August.
FMC. 1997a. FMC letter to EPA, Information Responding to EPA and Tribal Comments on Pond 8S
Closure Plan, Attachment A - Fate of Phosphorus in Phossy Waste Pond Solids, FMC, Pocatello.
77 August 2003
-------
References
FMC. 1997b. FMC letter to EPA, Information Responding to EPA and Tribal Comments on Pond 8S
Closure Plan, Attachment B - Summary of Evaluation of Alternatives for Soils/Solids Excavation, Ex-
Situ Treatment, and In-Situ Treatment.
FRTR. 2001. Screening Matrix and Reference Guide. 3rd Edition. Federal Remediation Technologies
Roundtable. Available on line at www.frtr.gov/matrix2/. December 4.
FRTR. 200la. Cost and Performance Remediation Case Studies and Related Information. Second
Edition. EPA 542-C-01-003. May.
Hunter, Craig, Tetra Tech EM Inc., 2001. Summary of Meeting with Wallace Reid, U.S. .EPA Region 10.
October 28.
Jefcoat, Irvin A., and Sundeep Potluri, Univ. of Alabama. 1995. "Removal of Elemental Phosphorus
from Electric Furnace Sludges of Various Origins." Environmental Progress, Vol. 14, No. 2. May.
Lindeburg, Michael R., P.E., 1986. Civil Engineering Reference Manual, Fourth Edition. Professional
Publications, Inc., San Carlos CA.
Madalinski, Kelly, EPA/TIO. 2002. Personal Communications with Wallace Reid and Linda Meyer,
EPA/Region 10. January.
Madalinski, Kelly, EPA/TIO. 2002a. Personal Communications with Nestor Young, EPA/Region 4.
January.
Madalinski, Kelly, EPA/TIO. 2001. Personal Communications with Wallace Reid, Linda Meyer, Nick
Ceto, EPA/Region 10. December
Madalinski, Kelly, EPA/TIO. 200la. Personal Communications with Trudy Olin and Steve Pranger,
USACE-WES. July.
Madalinski, Kelly, EPA/TIO. 2001b. Personal Communications with Dennis Teefy and Wayne Sisk,
USAEC. July.
Madalinski, Kelly, EPA/TIO. 200Ic. Personal Communications with John Austin, Bill Kline, and Steve
Hoffman, EPA/OSW. September-October.
Mathews, Joseph B., Olin Corp., and Irvin A. Jefcoat, Univ. of Alabama. 1997. "Isothermal Oxidation of
White Phosphorus Dispersed in Water in a Stirred-Tank Reactor". Journal of the Air and Waste
Management Association, Vol. 47:1103-1110. October.
Meyer, Linda, EPA Region 10. 2002. E-mail to Kelly Madalinski, EPA/TIO, Regarding LDR System
Questions. March 22.
Meyer, Linda, EPA Region 10. 1998. Review of Pond 8S Closure Plan, Volume 1 and 2, FMC
Corporation. January.
O'Brien & Gere Engineers, Inc. 2000. Solidification/Stabilization Treatability Studies Work Plan,
Stauffer Management Company, Tarpon Springs, Florida. December.
78 August 2003
-------
References
Rivera, Yilda B., Trudy Olin, and Mark R. Brika, USAGE, Waterways Experiment Station. 1996.
Summary and Evaluation for White Phosphorus Remediation: A Literature Review, Technical Report
IRRP-96-7. October.
RSMeans. 2002. Heavy Construction Cost Data, 16th Edition.
Sherrill, Mary, Tetra Tech EM Inc. 2002. Telecon with Ronald E. Edwards, TVA, Information About
Oxidation Process Used at TVA. February 8.
Spanggord, Ronald J., Rewick, Robert, U.S. Army, et al.1985. Environmental Fate of White
Phosphorus/Felt and Red Phosphorus/Butyl Rubber Military Screening Smokes.
Stewart, Susan, Tetra Tech EM Inc. 2001. Telecon with Hassan Armeni, McLaren Hart/ENSR,
Information About Infrared LTTD. October 18.
Stewart, Susan, Tetra Tech EM Inc. 200la. Telecon with John V. Brannen, Organic Solutions.
Information About Remediation Plus. September 27.
U.S. Army Center for Health Promotion and Preventive Medicine. 2001. Detailed Facts About White
Phosphorus. Available online atwww.apgea.army.mil/dts/docs/detwp.pdf. November
U.S. Army Material Command Regulation 385-103. 1965. Safety Guide for the Processing, Handling,
and Decontamination of White Phosphorus. December.
U.S. Department of Defense, Virtual Naval Hospital. 2001. "Chemical Burns and White Phosphorus
Injury". Emergency War Surgery NATO Handbook, Chapter III: Burn Injury.
http://www.vnh.org/EWSurg/ch03/03ChemicalBurns.html. Searched December.
Walsh, Michael R., Marianne E. Walsh, and Charles M. Collins. 2000. "Method for Attenuation of
White Phosphorus Contamination in Wetlands". Journal of Environmental Engineering. November.
Walsh, Michael R., Walsh, Marianne E., Collins, Charles M., USACE-CRREL. 1999. "Enhanced
Natural Remediation of White-Phosphorus Contaminated Wetlands Through Controlled Pond Draining"
CRREL Report 99-10. November.
Walsh, Michael R., Walsh, Marianne E., Collins, Charles M., USACE-CRREL. 1999. "Remediation
Methods for White Phosphorus Contamination in a Coastal Salt Marsh". Environmental Conservation.
26(2): 112-124
79 August 2003
-------
Additional Sources of Information
ADDITIONAL SOURCES OF INFORMATION
ATSDR. 1997. White Phosphorus, CAS #7723-14-0, Fact Sheet. September.
ATSDR. 1997. Public Health Statement for White Phosphorus. September.
Bersanti, Dan, Plant Manager, Rhodia, Inc. 2001. Waste Plan - Volumes I and II. Rhodia Silver Bow
Plant, Butte, Montana, November.
Chem Expo. 1997. "Phosphorus- Chemical Profile."
Darling, Marie, USACE-CRREL. 2001. CRREL, Army Clean Up Contaminated Alaskan Marsh,
Engineer Update, Army Internet Site. Searched October.
EPA. 2000. Solidification/Stabilization Use at Superfund Sites. EPA 542-R-00-010. September.
EPA. 1999. Solidification/Stabilization Resource Guide. EPA 542-B-99-002.
EPA. 1998. Record of Decision for Stauffer Chemical Tarpon Springs Site, Tarpon Springs, Pinellas
County, Florida. July.
EPA. 1997. Elemental Phosphorus Facilities Operating in the U.S. in 1980 Using Electric Arc Furnace
Method.
EPA. 1996. Technology Screening Guide for Radioactively Contaminated Sites. EPA402-R-96-017.
November, http://www.epa.gov/superfund/resources/radiation/pdf/techguide.pdf
Environmental Yellow Pages 2001 - Phosphorus. October.
Henderson, Scott. 1997. Telecon with Ed Greutert about Albright-Wilson Phosphorus Facility Closure.
November.
Nam, Sae-Im, Walsh, Michael R., Collins, Charles M., USACE-CRREL. 1999. Eagle River Flats
Remediation Project, Comprehensive Bibliography-1950-1998, Special Report 99-13. August.
Racine, Charles H., Walsh, Marianne E., et al. 1992. Waterfowl Mortality in Eagle River Flats, Alaska.
The Role of Munitions Residues, CRREL Report, 92-5. May.
Racine, Charles H., Walsh, Marianne E., USAGE, CRREL, et al. 1993. White Phosphorus
Contamination of Salt Marsh Pond Sediments at Eagle River Flats, Alaska, CRREL Report.
Roy F. Weston, Inc. 1996. Stauffer Management Company, Tarpon Springs, Florida Site, Final
Feasibility Study Report. January.
Van Wazer, John R., Monsanto Company. 1972. "Phosphorus and the Phosphides". Kirk-Othmer
Encyclopedia of Chemical Technology, 2nd Edition, Volume 15.
80 August 2003
-------
Additional Sources of Information
Walsh, Michael R. and Collins, Charles M., USACE-CRREL. 1998. Dredging as Remediation for
White-Phosphorus Contamination at Eagle River Flats, Alaska, CRREL Report 98-5. August.
Walsh, Michael R., Walsh, Marianne E., Collins, Charles M. 1997. "Method for Attenuation of White
Phosphorus Contamination in Wetlands", Journal of Environmental Engineering.
Walsh, Michael R., Chamberlain, Edwin J., Henry, Karen S., Garfield, Donald E., Sorenson, Ed, USACE-
CRREL. 1996. "Dredging in an Active Artillery Impact Area, Eagle River Flats, Alaska" Special Report,
96-22.
Walsh, Marianne E., Collins, Charles M., Racine, Charles H., USACE-CRREL. 1996. "Persistence of
White Phosphorus Particles in Salt Marsh Sediments". Environmental Toxicology and Chemistry.
Volume 15, No. 6.
Walsh, Marianne E., Taylor, Susan. 1996. Development of an Analytical Method for White Phosphorus
(P4) in Water and Sediment Using Solid Phase Microextraction, Special Report 96-16.
Walsh, Marianne E., Collins, Charles M. 1995. Persistence of White Phosphorus Particles in Sediment,
CRREL Report 95-23.
Walsh, Marianne E., USAGE, CRREL. 1995. Analytical Method for White Phosphorus in Water,
Environmental Contamination and Toxicology.
Walsh, Marianne E., Taylor, Susan. 1993. "Analytical Method for White Phosphorus Residues in
Munitions-Contaminated Sediments". Analytica, Chimica Acta, 282.
Walsh, Michael R., USACE-CRREL. Not dated. Dredging Contaminated Sediments at an Active Impact
Range: an Ordnance Avoidance Success. Army Internet Site.
http://hnd.usace.army.mil/oew/news/dredge.html. Searched October 2001.
81 August 2003
-------
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_ Appendix A
APPENDIX A
SUMMARY OF STAUFFER CHEMICAL COMPANY, TARPON SPRING, FL
Planned Remediation of Unlined Ponds Using in situ Solidification/Stabilization
Site Background
The 130 acre Stauffer Chemical Company plant in Tarpon Springs, Pinellas County, Florida, produced
elemental phosphorus using an electric arc furnace process from 1947 until it shut down in 1981. Stauffer
Chemical Company purchased the plant from Victor Chemical Company in 1960. (EPA, 1999) and the
plant was decommissioned and dismantled in 1983. The divestiture of the Stauffer Chemical Company in
1987 resulted in the formation of the Stauffer Management Company (SMC). The SMC site was listed
on the National Priorities List (NPL) in 1994. (Weston, 1996)
The plant is located on the Anclote River which flows into the Gulf of Mexico. The plant site layout
consists of a former phosphate processing area, former elemental phosphorus production facilities, and
administrative buildings. Land use in the surrounding locality includes light industry, commercial
establishments, and residential properties along with some undeveloped land. (Weston, 1996)
Approximately 10,000 people live within one mile of SMC. (EPA, 1999)
Approximately 300,000 cubic yards of soil, pond sediments, and slag at the SMC plant is contaminated
with radionuclides, metals, and elemental phosphorus. (EPA, 1999) Subsurface soil samples collected
during the remedial investigation detected arsenic, lead, fluoride, and total phosphorus as well as the
radiological components of gross alpha, gross beta, radium-226, radon-222, and polonium-210 in the
subsurface soil. The results from surface soil sampling indicated that the surface soil is primarily
contaminated with radionuclides found in the uranium decay chain, such as radium-226. Surface soil
sampling also detected other chemical contaminants including arsenic, antimony, beryllium, cadmium,
chromium, thallium, PAHs, and fluoride. Elemental phosphorus has not been found in any locations
other than in pond material on the site.
Groundwater sampling identified elevated levels of inorganics and radionuclides in the surficial aquifer
and deeper Floridian aquifer. Samples taken from private wells in the Tarpon Springs area detected
contaminants at concentrations below their maximum contaminant levels (MCLs). Surface water and
A-l August 2003
-------
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_ Appendix A
sediment samples taken from the Anclote River and the Myers Cove area located directly adjacent to the
site did not contain site-related contamination above background levels.
Seventeen unlined waste ponds containing process wastes are located in the phosphate processing area.
These ponds no longer contain water but still contain waste material that was disposed after elemental
phosphorus production. In some locations, the waste material extends to depths below the groundwater
table. Results from approximately 150 borings indicate the extent of contamination in the pond material
extends as deep as 19 ft below ground surface . Elemental phosphorus contamination was found in the
ponds at depths up to 15 feet below ground surface. Radiological levels detected in the ponds exceed
residential and commercial use standards. Table A-l shows the maximum concentrations of the
contaminants that were detected in samples of the pond materials, as well as the site cleanup standard for
each contaminant. (EPA, 1999)
Table A-l. Contaminants of Concern and Corresponding
Concentrations at SMC (EPA, 1999)
Contaminant
Arsenic
Antimony
Beryllium
Elemental Phosphorus
Thallium
Radium-226
Radium-226 Dose
Maximum
Detected
Concentration
205 mg/L
52.0 mg/L
2.0 mg/L
0.854 mg/L
37.0 mg/L
34.0 pCi/g
35uR/hr
Cleanup
Standard
21.1 mg/L
28.1 mg/L
120 mg/L
1.4 mg/L
1.4 mg/L
5pCi/g
20uR/hr
A baseline risk assessment for the SMC site indicates that the primary exposure pathways for chemical
contaminants at the site are inhalation, ingestion, and dermal contact with contaminated soil and slag.
The exposure pathways for radiological contaminants on the site are incidental ingestion of soil, ingestion
of vegetation grown on contaminated soil, direct irradiation by contaminated soil, inhalation of radon,
incidental ingestion of sediment, ingestion of groundwater, or irradiation by roadbed material.
Results from the risk assessment show that on-site maintenance workers could potentially be exposed to
site-related contaminants in surface soil or fugitive dust emissions during landscaping, mowing, and other
outdoor activities. The risk assessment also modeled the air pathway from the site to determine if nearby
residents are at risk for exposure to contaminated air particulates. The results show that fugitive dust is
not an exposure pathway for particulate emissions to the surrounding community.
A-2
August 2003
-------
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_ Appendix A
The risk assessment determined that current off-site residents may potentially be exposed to site-related
contaminants present in surface water and sediment via fishing and/or swimming in the Anclote river.
The river is classified as suitable for fishing and swimming and is the receptor of groundwater from the
site. The risk assessment also determined that exposure pathways for future on-site residents are
incidental ingestion and dermal contact of contaminated soil as well as ingestion of drinking water from
local groundwater wells.
Environmental Setting
Hydrogeology
Two aquifers, a surficial aquifer and the Floridian aquifer, lie beneath Pinellas County and SMC. Overall
groundwater flow is to the southwest toward the Anclote River. The depth to the surficial aquifer is
approximately 8 feet and is separated from the Floridian aquifer by a semi-confining bed of clay. (EPA,
1999) The water table rises and falls within the surficial aquifer with influence from precipitation, tidal
changes, and changes in atmospheric pressure. (EPA, 1998). The Floridian aquifer is composed of a
thick sequence of carbonate rocks which are connected to one another. (Weston, 1996) The depth to the
Floridian aquifer ranges from 17 to 37 feet below the surface. (EPA, 1999) This aquifer provides most of
the public water supply for Pinellas County. (Weston, 1996)
Geology
SMC is located on the Gulf Coastal Lowland Physiographic Province. This province is made up of
unconsolidated fine sand with interbeds of clay and marl, fossilferous limestone and dolomite, and
gypsiferous limestone and dolomite. The carbonate rocks that underlie Pinellas County range in thickness
from 10,000 to 12,000 feet and form a Peninsular Arch which trends in the northwest direction. This arch
is the dominant surface substructure in southwest Florida. (EPA, 1998)
Surficial sand and the Hawthorn Formation are the two distinct stratigraphic units that are exposed in
Pinellas County. The surficial sand consists of a thin veneer of fine sand with clay, marl, and phosphorite
interbeds. The Hawthorne Formation is thicker and consists of variable strata of a calcareous sand and a
sandy clay containing black phosphate nodules and chert. (EPA, 1998)
A-3 August 2003
-------
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_ Appendix A
Remedy Selection and Status of Unlined Ponds
Over the past 15 years, numerous studies have been conducted at SMC in efforts to delineate the nature
and extent of contamination at the site. Table 2 provides a chronology of these studies with a brief
description of each study, year the work was completed, and the firm that completed the work. (EPA,
1999)
In 1998, U.S. EPA signed a Record of Decision (ROD) that outlines remediation efforts for the pond
sediments, contaminated soil below the water table, and in the on-site consolidation areas at SMC. These
efforts include in situ solidification/stabilization (S/S) of contaminated soils below the water table,
excavation of contaminated soils, consolidation and capping of contaminated soils over the material
treated by in situ S/S, and enforcement of institutional controls to prohibit residential use of the property.
In situ S/S would be performed on the unlined pond material at SMC by injecting and mixing binding
agents into the saturated pond material to form a solid, low permeability mix. The Final Feasibility Study
Report indicates a total volume of 15,000 cubic yards of pond material would be treated by S/S at depths
as great as 19 feet. In December 2000, a S/S work plan was prepared that summarizes the methodology
that will be used to determine the best approach to conducting S/S. An Explanation of Significant
Differences (ESD) incorporated into the 1998 ROD requires that S/S material achieve a minimum
unconfined compressive strength (UCS) of 50 pounds per square inch and a maximum permeability of 1 x
10"5 cm/sec. Coal combustion fly ash, Portland cement, lime products (such as quicklime, hydrated lime,
and lime kiln dust), cement kiln dust, and soluble silicates will be evaluated to determine which is the best
S/S agent. In addition to the use of S/S for pond sediments and contaminated soil, extensive ambient air
monitoring will also be implemented. To reduce fugitive dust, the injection process likely will use water-
slurried S/S reagents and subsurface mixing. (O'Brien & Gere, 2000) The S/S test program at SMC is
expected to take place in 2003; however, a specific schedule has not yet been identified. (Madalinski,
2002)
Remedial Cost
The ROD (EPA, 1998) for the SMC site identified that the total present worth cost for the selected
remedy is $9,356,000. This cost includes the S/S process, as well as excavation, consolidation, and
capping of selected contaminated soil, and institutional controls. The treatment of the pond material was
A-4 August 2003
-------
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_^^^^^^^^^^^^^^^^^^^^^^^^^^^^^_ Appendix A
estimated to cost $750,000 for 15,000 cubic yards, corresponding to a calculated unit cost of $50 per
cubic yard. (Weston, 1996) Further remedial cost estimates for the site are currently being developed by
the Corps of Engineers.
Table A-2. Past Studies at Tarpon Springs Site
(EPA, 1999)
Type of Study
Hydrogeologic Assessment
Final Expanded Site
Investigation
Interim Final Listing Site
Inspection
Radiological Studies
Site Sampling Program
Environmental Sampling
Program
Sediment Sampling
Program
Elemental Phosphorus
Borings Program
Treatability Study - Bench
Scale Oxidation Study
Chemical Oxidation of
Phosphorus Study
Phosphorus Recovery,
Bench Scale, and
Treatability Testing Studies
Treatability Study-
Solidification/Stabilization
Study of Pond Wastes
Soil Depth Cover Study
Soil/Slag Leachability
Study
Asbestos Sampling
Conducted by
Seaburn and
Robertson, Inc.
NUS, Inc.
NUS, Inc.
PBS&J
Weston
Weston
Weston
Weston
Weston
unknown
unknown
Weston
Weston
Parsons Engineering
Parsons Engineering
Year
1987
1989
1991
1990
1989-90
1991
1991
1991
1991
unknown
unknown
1991
1994
1997
1998
Description
Study the hydrogeologic characteristics of the
surficial and Floridian aquifers at the site
Evaluate surface soils, groundwater, subsurface
soils, groundwater in surficial and Floridian
aquifers, surface water in the Anclote River
Evaluate surface soils at school nearby, re-sample
groundwater, collect sediment samples from
Anclote River
Conducted external gamma radiation surveys of
roadways, ponds
Soil and pond sampling
Characterize soil and disposal pond materials
Collected 13 sediment samples from the Anclote
River
Collected samples from 47 phosphorus soil borings
Evaluate performance of various oxidizers in
treating elemental phosphorus
Determine feasibility of oxidizing elemental
phosphorus clarifier using Nitric Acid
Determine if elemental phosphorus can be
recovered, separated from other materials and
effectively treated on-site. Excavation and
treatment difficult.
Evaluated various stabilization mixes to see which
would meet strength and economic feasibility
requirements
Recommended 24 inch soil cover to minimize
gamma radiation dose
Determined that contaminants leaching from soil
and slag in the slag field are minimal
Asbestos Sampling-on-site and air
A-5
August 2003
-------
Appendix A
Type of Study
Final Remedial
Investigation Report
Elemental Phosphorus Soil
and Groundwater Sampling
Program
Background soil sampling
for arsenic and beryllium
Conducted by
Weston
unknown
unknown
Year
1993
1994
1999
Description
Evaluated all media
Distinguish between elemental and total phosphorus
Background levels below State of Florida cleanup
levels
References
O'Brien & Gere Engineers, Inc. 2000. Solidification/Stabilization Treatability Studies Work Plan,
Stauffer Management Company, Tarpon Springs, Florida. December.
Roy F. Weston, Inc. 1996. Stauffer Management Company Tarpon Springs, Florida Site Final Feasibility
Study Report. January.
U.S. Environmental Protection Agency (EPA). 1999. Site Briefing for Stauffer Chemical Company
Superfund Site, Tarpon Springs, Florida. November 29.
EPA. 1998. Record of Decision, The Decision Summary, Operable Unit 1, Stauffer Chemical Tarpon
Springs, Pine lias County, Florida. July 2.
Madalinski, Kelly, EPA/TIO. 2002. Personal Communications with Nestor Young, EPA. December.
A-6
August 2003
------- |