Treatment Technologies
For Mercury in Soil,
Waste, and Water
-------
Cover: Photograph of a drop of mercury taken by Adam Hartley.
Exposure: 1/10 of one millisecond. Used by permission.
-------
August 2007
Treatment Technologies for Mercury in Soil,
Waste, and Water
U.S. Environmental Protection Agency
Office of Superfund Remediation and Technology Innovation
Washington, DC 20460
-------
This page is intentionally left blank.
-------
TABLE OF CONTENTS
Section Page
ACRONYMS. ABBREVIATIONS, AND SYMBOLS iii
FOREWORD v
NOTICE AND DISCLAIMER vi
ACKNOWLEDGMENTS vi
EXECUTIVE SUMMARY ES-1
PART I OVERVIEW
1.0 BACKGROUND 1-1
1.1 Introduction and Purpose of Report 1-1
.2 Sources of Information forthe Report 1-1
.3 Methodology Used to Compile Technology-Specific Information 1-2
.4 Background 1-2
.5 Environmental Occurrence 1-3
.6 Chemistry and Behavior 1-5
.7 Regulatory Considerations 1-6
.8 Mercury Removal and Emergency Response 1-8
.9 Detection and Site Characterization 1-8
2.0 COMPARISON OF TREATMENT TECHNOLOGIES FOR MERCURY 2-1
2.1 Types of Technologies Used 2-1
2.2 Frequency of Technology Use 2-2
PART II MERCURY TREATMENT TECHNOLOGY SUMMARIES
PART IIA TREATMENT TECHNOLOGIES APPLICABLE TO SOIL AND WASTE
3.0 SOLIDIFICATION AND STABILIZATION (INCLUDING AMALGAMATION) 3-1
4.0 SOIL WASHING AND ACTD EXTRACTION 4-1
5.0 THERMAL TREATMENT 5-1
6.0 VITRIFICATION 6-1
PART IIB TREATMENT TECHNOLOGIES APPLICABLE TO WATER
7.0 PRECIPITATION/COPRECIPITATION 7-1
8.0 ADSORPTION TREATMENT 8-1
9.0 MEMBRANE FILTRATION 9-1
10.0 BIOLOGICAL TREATMENT 10-1
PART III ONGOING RESEARCH ON MERCURY TREATMENT
11.0 ONGOING RESEARCH ON MERCURY TREATMENT 11-1
11.1 Nanotechnology 11-1
11.2 Phytoremediation 11-2
11.3 Air Stripping 11-3
11.4 In Situ Thermal Desorption (ISTD) 11-3
August 2007
-------
TABLES
Table Page
ES-1 Overview of Selected Mercury Treatment Technologies ES-4
ES-2 Summary of Technology Use and Availability of Technology Performance Data for Mercury
Treatment ES-5
1.1 Number of Superfund Sites with Mercury as a Contaminant of Concern by Media Type 1-4
1.2 Number of Superfund Sites with Mercury as a Contaminant of Concern by Site Type 1-4
1.3 Physical and Chemical Properties of Some Mercury Compounds 1-6
1.4 RCRA Listed Hazardous Wastes for Mercury 1-7
1.5 Land Disposal Restrictions for Low Mercury Wastes, High Mercury Wastes, and Elemental
Mercury 1-7
2.1 Applicability of Mercury Treatment Technologies 2-5
2.2 Mercury Treatment Technologies Screening Matrix 2-6
2.3 Available Mercury Treatment Cost Data for Various Technologies 2-8
3.1 Solidification/Stabilization Treatment Performance Data for Mercury 3-7
3.2 Available Mercury Treatment Cost Data 3-10
4.1 Soil Washing and Acid Extraction Performance Data for Mercury 4-7
4.2 Available Mercury Treatment Cost Data for Soil Washing and Acid Extraction 4-10
5.1 Thermal Treatment Performance Data for Mercury 5-5
5.2 Available Mercury Treatment Cost Data for Thermal Treatment 5-7
6.1 Vitrification Treatment Performance Data for Mercury 6-6
6.2 Available Mercury Treatment Cost Data for Vitrification 6-7
7.1 Precipitation/Coprecipitation Treatment Performance Data for Mercury 7-5
7.2 Available Mercury Treatment Cost Data for Precipitation/Coprecipitation 7-8
8.1 Adsorption Treatment Performance Data for Mercury 8-5
8.2 Available Mercury Treatment Cost Data for Adsorption 8-7
10.1 Biological Treatment Performance Data for Mercury 10-5
FIGURES
Figure Page
2.1 Technologies Applicable to Treatment of Mercury-Contaminated Media 2-1
2.2 Number of Identified Applications of Mercury Treatment Technologies for Soil and Waste 2-2
2.3 Number of Identified Applications of Mercury Treatment Technologies for Water 2-2
3.1 Model of a Solidification/Stabilization System 3-1
4.1 Model of a Soil Washing System 4-1
5.1 Model of a Thermal Desorption or Retort System 5-2
6.1 Model of a Vitrification System 6-1
6.2 Typical Ex Situ Vitrification System 6-2
7.1 Model of a Precipitation/Coprecipitation System 7-1
9.1 Model of aMembrane Filtration System 9-2
10.1 Model of a Biological Treatment System 10-1
Appendix A
Information on Select Mercury Vapor Detection Instruments
August 2007
-------
ACRONYMS, ABBREVIATIONS, AND SYMBOLS
l-ig/kg
Hg/L
(ig/m3
urn
APC
APGEN
ARAR
BOAT
BNL
CAA
CBPC
CERCLA
CERCLIS
CFM
CWA
cm
cy
DM
DoD
DTC
EPA
ERT
FATE
FRTR
g/cm3
GFT
g/L
gpm
HEPA
HTTD
IA
ISTD
L/min
LANL
LDR
LTTD
MACT
MCL
Micrograms per Idlogram
Micrograms per liter
Micrograms per cubic meter
Micron
Air pollution control
Applied PhytoGenetics, Inc.
Applicable or relevant and appropriate requirement
Best demonstrated available technology
Brookhaven National Laboratory
Clean Air Act
Chemically bonded phosphate ceramics
Comprehensive Environmental Response, Compensation, and Liability Act
Comprehensive Environmental Response. Compensation, and Liability Information
System
Chemische Fabrik Marktredwitz
Clean Water Act
Centimeters
Cubic yards
Deutsche Mark
Department of Defense
Dithiocarbamate
U.S. Environmental Protection Agency
Emergency Response Team
Field Analytic Technologies Encyclopedia
Federal Remediation Technologies Roundtable
Grams per cubic centimeter
Glass furnace technology
Grams per liter
Gallons per minute
Fligh efficiency particulate air
Fligh-temperature thermal desorption
Immunoassay
In situ thermal desorption
Liters per minute
Las Alamos National Laboratory
Land disposal restrictions
Low temperature thermal desorption
Maximum achievable control technology
Maximum contaminant level
111
August 2007
-------
ACRONYMS, ABBREVIATIONS, AND SYMBOLS (Continued)
mg/L
mg/kg
mg/mj
NaDTC
ND
ng/L
NPL
NRMRL
O&M
OU
P&T
PAH
PCB
PNNL
POTW
ppb
ppm
PRO
psi
RCRA
RPM
S/S
SDWA
SITE
SNAP
SPC
SPSS
SVOC
TCLP
Thiol-SAMMS
TLV
TOC
TWA
uv
voc
WQC
XRF
Milligrams per liter
Milligrams per kilogram
Milligrams per cubic meter
Sodium dithiocarbamate
Non-detect
Nanograms/liter
National Priorities List
National Risk Management Research Laboratory
Operation and maintenance
Operable unit
Pump and treat
Poly cyclic aromatic hydrocarbon
Polychlorinated biphenyl
Pacific Northwest National Laboratory
Publicly owned treatment works
Parts per billion
Parts per million
Preliminary remediation goal
Pounds per square inch
Resource Conservation and Recovery Act
Remedial project manager
Solidification/Stabilization
Safe Drinking Water Act
Superfund Innovative Technology Evaluation
Superfund NPL Assessment Program
Sulfur polymer cement
Sulfur Polymer Stabilization/Solidification
Semivolatile organic compound
Toxicity Characteristic Leaching Procedure
Thiol Self-Assembled Monolayers on Mesoporous Silica
Threshold limit value
Total organic carbon
Total Waste Analysis
Ultraviolet
Volatile organic compound
Water quality criteria
X-rav fluorescence
IV
August 2007
-------
FOREWORD
This report provides a synopsis of the availability, performance, and cost of eight technologies for
treatment of mercury in soil, waste, and water. The report also describes research under way on
innovative methods to treat mercury contamination. The report's intended audience includes hazardous
waste site managers; those generating or treating mercury-contaminated waste and wastewater;
regulators; and the interested public.
There is a need for cost-effective mercury treatment. Historical use of mercury has resulted in soil and
groundwater contamination that may require remediation. Mercury contamination is difficult to treat and
may pose a risk to human health and the environment. In addition, some industrial wastes and
wastewaters currently being produced require treatment to remove or immobilize mercury.
This report is intended to be used as a screening tool for mercury treatment technologies. It describes the
theory, design, and operation of the technologies; provides information on commercial availability and
use; and includes data on performance and cost, where available. As a technology overview document.
the information can serve as a starting point to identify options for mercury treatment. The feasibility of
particular technologies will depend heavily on site-specific factors, and final treatment and remedy
decisions will require further analysis, expertise, and possibly, treatability studies.
August 2007
-------
NOTICE AND DISCLAIMER
Preparation of this report has been funded by the U.S. Environmental Protection Agency (EPA) under
Contract Number 68-W-02-034. Information in this report is derived from numerous sources (including
personal communications with experts in the field). Not all the source documents have been peer-
reviewed. This study has undergone EPA and external review by subject-matter experts. Mention of
trade names or commercial products does not constitute endorsement or recommendation for use.
A PDF version of Treatment Technologies for Mercury in Soil, Waste, and Water is available for viewing
or downloading from the Hazardous Waste Cleanup Information (CLU-IN) system Web site at
http://clu-in.org/542R07003. A limited number of printed copies are available free of charge and may be
ordered via the Web site, by mail, or by facsimile from:
U.S. EPA/National Service Center for Environmental Publications (NSCEP)
P.O. Box 42419
Cincinnati, OH 45242-2419
Telephone: (800)490-9198
Fax: (301)604-3408
ACKNOWLEDGMENTS
Special acknowledgement is given to the federal and state staff and other remediation professionals who
provided information for this document. Their willingness to share their expertise encourages the
successful application of mercury treatment technologies at other sites. Contributors to the report
included U.S. EPA Office of Superfund Remediation and Technology Innovation; U.S. EPA Office of
Research and Development; U.S. EPA Regional Offices; U.S. EPA Office of Water; U.S. Army Corps of
Engineers; Glen Fryxell, Ph.D., of the Pacific Northwest National Laboratory; and Jonathan McBurney of
Lockheed Martin Corporation. Special acknowledgement also is given to Adam Hartley, Photographer,
who provided the cover photograph.
vi August 2007
-------
Executive Summary
EXECUTIVE SUMMARY
This report contains information on the treatment of soil, waste, and water that contain mercury, a
contaminant that is difficult to treat and may cause a variety of adverse health effects in humans. A total
of 57 projects were identified, of which 50 provide performance data. Twenty-three of the 57 projects
were conducted at pilot scale and 34 at full scale. This information can help managers at sites with
mercury-contaminated media and generators of mercury-contaminated waste and wastewater to:
• Identify proven and effective mercury treatment technologies;
• Screen technologies based on application-specific goals, characteristics, and cost; and
• Apply experiences from sites with similar treatment challenges.
This report identifies eight technologies used to treat mercury in soil, waste, and water. These
technologies were selected based on the availability of project-specific information for mercury treatment.
Other technologies (for example, ion exchange) may be reported as being able to treat mercury
contamination but are not covered in this report because project data were not found in the references
used. Table ES-1 identifies and briefly describes each of these technologies. Mercury removal and
emergency response are also covered briefly in the report, with a link to a guidebook for more detailed
information. Treatment of air emissions that contain mercury is not covered in this report; however,
technologies used to treat mercury-containing off-gas are mentioned where the information was available.
Part II of this report contains more detailed information about each technology, including project-specific
data. Case studies for some projects are provided based on the availability of data and the uniqueness of
the application. Table ES-2 summarizes the technology applications and performance data identified for
each technology and medium. Technologies that are applicable to one type of soil or waste are typically
applicable to other types. For example, the media treated in the projects identified for
solidification/stabilization include soil, concrete fines, sludge, liquid waste, mine tailings, and elemental
mercury. Similarly, technologies that are applicable to one type of water are generally applicable to other
types. For example, both groundwater and various types of wastewater are treated in the projects
identified for precipitation/coprecipitation.
Mercury occurs naturally in the environment and can be found in elemental (metallic), inorganic, and
organic forms. Modern uses for mercury include chemical manufacturing, thermometers, and lighting
(mercury vapor and fluorescent lamps). The chemical and allied products industry group is responsible
for the largest quantity of mercury used in the United States.
Mercury, particularly the organic methylmercury form, is a potent neurotoxin capable of impairing
neurological development in fetuses and young children and of damaging the central nervous system of
adults. Mercury regulations span multiple federal and state environmental statutes, as well as multiple
agency j urisdiction s.
Soil and Waste Treatment Technologies
Among the projects identified, solidification/stabilization (S/S) is the most frequently used technology to
treat soil and waste contaminated with mercury. The data show that this technology has been used to
meet regulatory cleanup levels, is commercially available to treat both soil and waste, and generates a
residual that typically does not require further treatment before disposal. However, the data sources used
for this report did not contain any information about the long-term stability of mercury -containing soil
and waste treated using S/S.
ES-1 August 2007
-------
Executive Summary
Other technologies for soil and waste, including soil washing, vitrification and thermal treatment, are
typically considered for specific applications or soil types, and therefore are not used as often as
solidification/stabilization.
Water Treatment Technologies
Among the mercury treatment projects identified, precipitation/coprecipitation is the most commonly
used process to treat mercury-contaminated water. The effectiveness of this technology is less likely to
be affected by characteristics of the media and contaminants compared with other water treatment
technologies.
Alternative mercury treatment technologies include adsorption and membrane filtration; however, these
technologies are used less frequently because both are more likely to be affected by characteristics of the
media and non-mercury contaminants than is precipitation/coprecipitation. Adsorption tends to be used
more often when mercury is the only contaminant to be treated, for relatively smaller systems, and as a
polishing technology for effluent from larger systems. Membrane filtration is used less frequently
because it tends to produce a larger volume of residuals than do other mercury treatment technologies.
Bioremediation has been used to treat mercury in two pilot-scale studies identified for this report. In one
study, mercury ions were converted to elemental mercury by bacteria and then adsorbed onto a carbon
unit. In the other study, a series of aerobic and anaerobic biological treatment steps was used to remove
mercury from wastewater.
Innovative Approaches for Treatment of Mercury
Innovative approaches for treatment of mercury have been applied at the bench and pilot scale and show
potential for deployment at full scale. These technologies include nanotechnology, phytoremediation, air
stripping, and in situ thermal desorption. In a pilot-scale test of the nanotechnology, thiol-SAMMS (Self-
assembled Monolayers on Mesoporous Silica), 97.4 percent of mercury dissolved in an aqueous waste
stream was removed after the first treatment round, and 99.4 percent was removed after two additional
rounds.
Phytoremediation is also being evaluated for its effectiveness in removal of mercury from sediments and
other media. In a bench-scale study of rice genetically engineered for mercury resistance, the
concentration of elemental mercury was higher in the genetically-modified germinates than in the wild-
type germinates, indicating enhanced mercury-reducing activities of the genetically engineered plants.
Eastern cottonwood trees have also been tested in the field for their ability to remediate mercury;
however, results from the study were not available when research was conducted for this report.
Air stripping is another technology being evaluated for its ability to remove mercury from water. In a
bench-scale study on contaminated groundwater at the Savannah River Site, chemical reduction using
stannous chloride followed by collection of the elemental mercury from the water was found to remove
low levels of mercury. Stannous chloride at doses greater than 0.011 milligrams per liter (mg/L) resulted
in more than 94 percent mercury removal, with the residual total mercury reduced to levels below 10
nanograms per liter (ng/L). However, low doses of stannous chloride showed little removal of mercury.
ES-2 August 2007
-------
Executive Summary
Mercury Treatment Cost Data
A limited amount of cost data on mercury was found from the sources reviewed. In many cases, only
partial cost data were available, such as capital only, or operations and maintenance (O&M) only, or total
costs without a breakdown. No mercury-specific cost data were identified for some technologies. A
summary of cost data is provided in Table 2.3, with more detailed information presented in Section 3.0
through 10.0. A summary or interpretation of the costs is not provided, however, because mercury
treatment is waste- and site-specific and because of the variation in the type and quality of information
available.
ES-3 August 2007
-------
Executive Summary
Table ES-1
Overview of Selected Mercury Treatment Technologies
Technology
Description
Technologies for Soil and Waste Treatment
Solidification/
Stabilization
Soil Washing/
Acid Extraction
Thermal
Desorption/
Retorting
Vitrification
Physically binds or encloses contaminants within a stabilized mass and chemically
reduces the hazard potential of a waste by converting the contaminants into less
soluble, mobile, or toxic forms.
Uses the principle that some contaminants preferentially adsorb onto the fines
fraction of soil. The soil is suspended in a wash solution and the fines are
separated from the suspension, thereby reducing the contaminant concentrations in
the remaining soil. Acid extraction uses an extracting chemical, such as
hydrochloric acid or sulfuric acid.
Application of heat and reduced pressure to volatilize mercury from the
contaminated medium, followed by conversion of the mercury vapors into liquid
elemental mercury by condensation. Off-gases may require further treatment
through additional air pollution control devices such as carbon units.
High -temperature treatment that reduces the mobility of metals by incorporating
them into a chemically durable, leach-resistant, vitreous mass. The process also
may cause contaminants to volatilize, thereby reducing their concentration in the
soil and waste.
Technologies for Water Treatment
Precipitation/
Coprecipitation
Adsorption
Membrane
Filtration
Biological
Treatment
Uses chemical additives to: (a) transform dissolved contaminants into an insoluble
solid, or (b) form insoluble solids onto which dissolved contaminants are adsorbed.
The insoluble solids are then removed from the liquid phase by clarification or
filtration.
Concentrates solutes at the surface of a sorbent, thereby reducing their
concentration in the bulk liquid phase. The adsorption media is usually packed into
a column. Contaminants are adsorbed as contaminated water is passed through the
column.
Separates contaminants from water by passing the water through a semi-permeable
barrier or membrane. The membrane allows some constituents to pass, while it
blocks others.
Involves the use of microorganisms that act directly on contaminant species or
create ambient conditions that cause the contaminant to leach from soil or
precipitate/coprecipitate from water.
ES-4
August 2007
-------
Executive Summarv
Table ES-2
Summary of Technology Use and Availability of Technology Performance Data for Mercury Treatment
Technology
Solidification/Stabilization
Soil Washing
Thermal Treatment
Vitrification
Precipitation
Adsorption
Membrane Filtration
Bioremediation
Total
Media Treated
Soil and
Waste
+
•
+
•
-
-
-
-
-
Water
-
-
-
-
•
+
•
+
-
Number of Projects Identified a (Number with
Performance Data)
Pilot Scale
6(6)
6(6)
5(5)
2(2)
0
2(2)
0
2(2)
23 (23)
Full Scale
12(6)
2(2)
3(3)
1(1)
11(11)
4(3)
1(1)
0
34 (27)
Total
18(12)
8(8)
8(8)
3(3)
11(11)
6(5)
1(1)
2(2)
57 (50)
a Projects were identified through previously compiled sources and information gathered from readily available data. The applications
include only those identified during the preparation of this report and are not comprehensive.
- Not applicable
Source: Adapted from data in Sections 3.0 to 10.0.
ES-5
August 2007
-------
This page is intentionally left blank.
ES-6
-------
PARTI
OVERVIEW
-------
This page is intentionally left blank.
-------
Section 1 - Introduction
1.0 BACKGROUND
1.1 Introduction and Purpose of Report
This report provides information on the current state of treating mercury in environmental media such as
soil, groundwater, and waste. Elemental mercury has a high vapor pressure which causes it to evaporate
to the atmosphere easily and potentially become an inhalation hazard. Organic mercury, particularly
methylmercury, is a potent neurotoxin capable of impairing neurological development in fetuses and
young children and damaging the central nervous system of adults. In addition to the toxicity, mercury
behaves in a complex manner depending on its form, oxidation state, and environmental conditions, such
as pH. These factors contribute to making the treatment of mercury challenging. Project-specific
information on treatment technologies for mercury contamination in environmental media and waste is
provided in this report and can be used by remediation site managers, hazardous waste generators, and the
public to:
• Identify proven and effective mercury treatment technologies;
• Screen technologies based on application-specific goals and characteristics and cost; and
• Apply experience from sites with similar treatment challenges.
Air emissions are not covered in this report; however, more detailed information on air pollution control
(APC) technologies can be found on the U.S. Environmental Protection Agency's (EPA's) mercury Web
page at http://ww\\ .epa.gov/mercun/control emissions/technologv.htm.
The term "soil" as used in this report includes soil, debris, sludge, sediments, and other solid-phase
environmental media. "Waste" includes nonhazardous and hazardous solid waste generated by industry.
"Water" includes groundwater, drinking water, nonhazardous and hazardous industrial wastewater,
surface water, mine drainage, and leachate. Most sources used for this report contained information about
treatment of mercury in soil, sludge, sediments, waste, groundwater, and wastewater. Limited
information was identified on treatment of drinking water.
Part I of this report, Overview and Findings, contains an executive summary, an introduction, and a
comparison of mercury treatment technologies. The introduction describes the purpose of the report,
presents background information, discusses ongoing research in mercury treatment, and summarizes the
methodology used to gamer and analyze data. The Comparison of Technologies section (Section 2.0)
analyzes and compares the technologies used for mercury treatment.
Part II of this report contains eight sections, each summarizing information for a specific type of mercury
treatment technology. Each summary includes a brief description of the technology, information about
how it is used to treat mercury, its status and scale, and data on cost and performance. Case studies for
some projects are provided based on the availability of data and the uniqueness of the application. The
technology summaries are organized as follows: the technologies typically used to treat soil and waste
appear first, in the order of their number of full-scale applications, followed by those typically used for
water, in the same order.
1.2 Sources of Information for the Report
This report is based on previously compiled sources on mercury treatment and information gathered from
readily available data sources, including:
1-1 August 2007
-------
Section 1 - Introduction
• Documents and databases prepared by EPA and other federal agencies;
• Technical literature;
• Information supplied by vendors of treatment technologies;
• Internet sites; and
• Information from technology experts.
1.3 Methodology Used to Compile Technology-Specific Information
This report provides information on the eight technologies listed in Table ES-1 in the executive summary.
These technologies have been used at the full or pilot scale to treat mercury in soil, waste, and water.
Full-scale projects include technologies used commercially to treat industrial waste and to remediate an
entire area of contamination. Pilot-scale projects are usually conducted in the field to test the
effectiveness of a technology on a specific soil, waste, and water or to obtain information for scaling a
treatment system up to full-scale.
The information gathered for this report includes many projects that used treatment trains. Treatment
trains consist of two or more technologies used together, either integrated into a single process or
operated as a series of processes in sequence. The technologies in a train may treat the same contaminant.
A common treatment train used for mercury in water includes pH adjustment as a pretreatment, followed
by precipitation, sedimentation, and filtration.
Some treatment trains are employed when one technology alone is not capable of treating all of the
contaminants. For example, an aboveground system consisting of electrochemical precipitation, followed
by air stripping and granular activated carbon adsorption was used to treat groundwater contaminated
with volatile organic compounds (VOC) and heavy metals, including mercury, at the King of Prussia
Technical Corporation Superfund Site in New Jersey (Table 8.1). Electrochemical precipitation involves
using electricity to convert contaminants to an insoluble form, which can then be removed. The
electrochemical treatment and air stripping steps were used to remove most of the heavy metals and
VOCs, while the adsorption treatment was a polishing step for the heavy metals, including mercury.
In many cases, the information did not specify the technologies within the train that were intended to treat
mercury. Influent and effluent concentrations often were provided for the entire train and not for the
individual components. In such cases, engineering judgment was used to identify the technology that
treated mercury. For example, a treatment train consisting of precipitation followed by carbon adsorption
and pH adjustment was used at the Olin Corporation Site in Alabama (Table 8.1) to treat groundwater
contaminated with VOCs and mercury. The precipitation step was assumed to remove most of the
mercury, and the carbon unit was assumed to serve as a polishing step.
When a treatment train included more than one potential mercury treatment technology, all mercury
treatment technologies were assumed to contribute to mercury treatment unless information indicated
otherwise. Information about these projects is presented in all applicable technology sections.
1.4 Background
Mercury occurs naturally in the environment and can be found in elemental (metallic), inorganic, and
organic forms. The most common natural fonns of mercury are elemental mercury, mercuric sulfide
(cinnabar ore), mercuric chloride, and methylmercury. Historically, mercury and its compounds have
been used for industrial, medicinal, and cosmetic purposes. Modern uses for mercury include production
of chlor-alkali, in wiring devices and switches, measuring and control devices, lighting, and dental work.
In 2004. the largest use of mercury was in wiring devices and switches (63 tons), followed by dental work
(35 tons), and chlor-alkali production (14 tons). Based on data provided by the chlor-alkali, lamp, and
dental industries, it appears that use of mercury declined by more than 50 percent between 1995 and 2004
(Ref. 1.19)
1-2 August 2007
-------
Section 1 - Introduction
Mercury, and particularly the organic methylmercury form, is a potent neurotoxin capable of impairing
neurological development in fetuses and young children and of damaging the central nervous system of
adults (Ref. 1.2, 1.21). High exposures to inorganic mercury may damage the gastrointestinal tract, the
nervous system, and the kidneys. Both inorganic and organic mercury compounds are absorbed through
the gastrointestinal tract and affect oilier systems via this route. However, organic mercury compounds
are more readily absorbed via ingestion than are inorganic mercury compounds (Ref. 1.4).
People are most likely to be exposed to harmful quantities of mercury through consumption offish
contaminated with methylmercury (Ref. 1.3). Elemental mercury causes adverse health effects when it is
breathed as a vapor and absorbed through the lungs. These exposures can occur when elemental mercury
is spilled or products that contain elemental mercury break and expose the substance to the air,
particularly in warm or poorly ventilated indoor spaces (Ref. 1.4). Exposure to inorganic mercury can
also occur from drinking contaminated water and touching contaminated water and soil, though these
exposures are generally not thought to be harmful at typical ambient levels (Ref. 1.2).
1.5 Environmental Occurrence
Mercury can be found in air, soil, and water. The most prevalent ore form of mercury is cinnabar
(mercury sulfide). In the U.S., large naturally occurring cinnabar deposits exist in California, Nevada,
Utah, Oregon, Arkansas, Idaho, and Texas. Inorganic mercury also occurs in small amounts in many rock
types (for example, granite and shale) (Ref. 1.5). Atmospheric levels of mercury tend to be higher in the
vicinity of active volcanoes and significant anthropogenic sources than in other areas. Coal-burning
power plants emit more mercury to the air than other anthropogenic sources in the U.S., accounting for
about 45 percent of all domestic mercury emissions (Ref. 1.19). Once in the atmosphere, mercury is
widely disseminated and can circulate for years, accounting for its wide-spread distribution. Most natural
waters, including groundwater and surface waters, also contain naturally occurring mercury in the low
micrograms per liter ((.ig/L) range (depending on the types of rock the water flows through), with
freshwater concentrations reported as high as 70 ^g/L (Ref. 1.5). Inorganic mercury in the atmosphere
that is deposited in water has the potential to be converted to methylmercury, which tends to
bioaccumulate in fish and animals (Ref. 1.1).
Of the estimated 111 tons of mercury emitted annually into the atmosphere by anthropogenic sources in
the U.S. as of 2002, approximately 45 percent is from utility coal boilers; 10 percent from electric arc
furnaces; 9 percent from industrial/commercial/institutional boilers and process heaters; 6 percent from
gold mining; and 5 percent from chlorine production. Total estimated emissions decreased 47 percent
between 1990 and 2002. However, the actual reduction in emissions has likely been greater than this
estimate because two of the biggest 2002 emissions source categories — electric arc furnaces and gold
mining — are not included in the 1990 inventory (Ref. 1.19).
At waste sites, mercury occurs in various media, including soil, groundwater, sediment, sludge, and
leachate. Thimerosal, an organic form of mercury (C9H9HgO2SNa), has been used as a preservative in a
number of medicines and can be found in waste streams from hospitals, clinical laboratories, and
pharmaceutical industries. Waste sites fall under several cleanup programs, including Superfund
responses, Resource Conservation and Recovery Act (RCRA) corrective actions, and state cleanups.
Information is available on the occurrence and treatment of mercury at Superfund National Priorities List
(NPL) sites. Based on records of decision and other site data, 290 NPL sites include mercury as a
contaminant of concern. A contaminant of concern is any compound that is expected to be present at a
site based upon past and current land uses and associated releases. Table 1.1 lists these sites by the
medium that contains mercury. Soil and groundwater are the most common media, with mercury-
contaminated soil at 173 sites and mercury-contaminated groundwater at!44 sites. The table shows four
sites with air emissions of mercury; however, this actual number is likely larger than four because fugitive
1-3 August 2007
-------
Section 1 - Introduction
air emissions of mercury may not be incorporated with the emissions inventory estimations technique
used. The number of sites in Table 1.1 exceeds the total number of sites with mercury (290) because
more than one type of media contaminated with mercury may be present at some sites. Mercury may also
be present at sites not listed on the NPL; however, this information was not captured in Table 1.1.
Table 1.1
Number of Superfund Sites with Mercury as a Contaminant of Concern by Media Type
(Total Number of NPL Sites = 290)
Media Type
Soil
Ground water
Sediment
Surface Water
Debris
Sludge
Solid Waste
Leachate
Other
Liquid Waste
Air
Residuals
Number of Sites"
173
144
92
43
35
24
18
16
11
4
4
1
Source: Ref. 1.7 (data through 2000), 1.8, 1.9
a Some sites contain more than one type of media contaminated with mercury.
Table 1.2 lists the number of Superfund sites with mercury as a contaminant of concern by site type. The
most common site types are landfills, and chemicals and allied products facilities.
Table 1.2
Number of Superfund Sites with Mercury as a Contaminant of Concern by Site Type
Site Type
Landfills
Chemicals and Allied Products
Groundwater Plume
Metals Fabrication and Finishing
Military
Batteries and Scrap Metal
Transportation Equipment
Primary Metals Processing
Ordnance Production
Mining
Electrical Equipment
Chemicals and Chemical Waste
Research and Development
Other
Number of Sites
102
15
14
12
11
9
9
7
6
5
5
5
5
85a
Source: Ref. 1.7 (data through 2000), 1.8, 1.9
a Includes site types with fewer than five sites, sites where site types were identified as
"other" or "multiple," and unspecified waste management.
1-4
August 2007
-------
Section 1 - Introduction
Information for Tables 1.1 and 1.2 was compiled from the Comprehensive Environmental Response,
Compensation, and Liability Information System (CERCLIS) database (Ref. 1.6), the Superfund NPL
Assessment Program (SNAP) database (Ref. 1.7). and the database supporting the document "Treatment
Technologies for Site Cleanup: Annual Status Report (Eleventh Edition)*' (Ref. 1.8).
1.6 Chemistry and Behavior
Elemental mercury is a silver-white, heavy liquid at room temperature. Because of its high surface
tension, it forms small compact spherical droplets when it is released in the environment. Although the
droplets themselves are stable, the high vapor pressure of mercury compared with other metals causes it
to evaporate. Therefore, released or uncovered mercury can rapidly become an inhalation hazard (Refs.
1.1 and 1.5).
The most prevalent oxidation states for mercury are Hg+1 (mercurous) or Hg+2 (mercuric). The properties
and chemical behavior of mercury strongly depend on its oxidation state. Mercury can form a variety of
inorganic salts and is unusual among metals because, unlike the reactive alkali and alkaline earth metals,
it can form covalent bonds with organic radicals. Most of the mercury encountered in water, soil,
sediments, and biota and other environmental media (except the atmosphere) is in the form of inorganic
mercuric salts and organomercurics. Organomercurics are defined by the presence of a covalent C-Hg
bond. The presence of a covalent C-Hg bond differentiates organomercurics from inorganic mercury
compounds that merely associate with the organic material in the environment but do not have the C-Hg
bond.
The following mercury compounds are most frequently found under environmental conditions: mercuric
salts HgS. HgCl, Hg(OH) and HgS; the methylmercury ion (HgCH3+) and its compounds methylmercuric
chloride (CH3HgCl) and methylmercuric hydroxide (CH3HgOH); and, in small fractions, other
organomercurics (for example, dimethylmercury and phenylmercury) (Refs. 1.5 and 1.10).
Methylmercury — which can cause nervous system damage, especially in fetuses and infants — is
formed when mercury enters soil or sediments and is acted on by anaerobic microorganisms (Ref. 1.4).
The solubilities of mercury compounds vary, ranging from negligible (HgCl, HgS) to very soluble
(HgCl2) (Ref. 1.5). Table 1.3 provides the physical and chemical properties of elemental mercury and
some of its compounds (Ref. 1.14).
Ionic forms of mercury are strongly adsorbed by soils and sediments and are desorbed slowly. Clay
minerals optimally adsorb mercury ions at pH 6. Iron oxides also adsorb mercury ions in neutral soils.
Most mercury ions are adsorbed by organic matter (mainly fulvic and humic acids) in acidic soils. When
organic matter is not present, mercury becomes relatively more mobile in acid soils and can evaporate to
the atmosphere or leach to groundwater (Ref. 1.5).
1-5 August 2007
-------
Section 1 - Introduction
Table 1.3
Physical and Chemical Properties of Some Mercury Compounds
Compound Name
Molecular formula
Molecular weight
Solubility
Density
Boiling Point (°C)
Melting Point (°C)
Oxidation State
Elemental
Mercury
Hg"
200.59
5.6 X 10"
g/L at 25°C
13.534
g/cmj at
25°C
357
-39
0
Mercuric
Chloride
HgCl2
271.52
69 g/L at
20°C
5 .4 g/cm3 at
25°C
302
277
+2
Mercurous
Chloride
Hg2Cl2
472.09
2.0 X 10'J g/L
at 25°C
7.15g/cm"at
19°C
384
302
+1
Methyl-
mercuric
Chloride
CH3HgCl
251.1
0.100 g/L at
21°C
4.06 g/cm" at
20°C
Not available
170
+2
Dimethyl
Mercury
C2H6Hg
230.66
lg/Lat21°C
3.1874g/cmJ
at 20°C
93
-43
+2
Source: Ref. 1.20, 1.18
Notes:
g/L = Grams per liter
1.7 Regulatory Considerations
An improved understanding of the toxic health effects of mercury and its bioaccumulative properties has
led to greater regulatory control. Mercury regulations span multiple federal and state environmental
statutes, as well as multiple agency jurisdictions. EPA has developed regulations to control mercury
emissions to air through the Clean Air Act (CAA); to water through the Clean Water Act (CWA) and Safe
Drinking Water Act (SDWA); and from wastes and products through RCRA (Refs. 1.2 and 1.3). For
example, EPA regulates emissions of mercury to the atmosphere from stationary sources under various
regulations promulgated under the CAA and RCRA. Likewise, under the SDWA, EPA has established a
maximum contaminant level (MCL) for mercury of 2 (.ig/L in drinking water. Under the CWA, EPA has
established ambient water quality criteria (WQC) for mercury and has issued technology-based standards
for specific industries to control mercury discharges into rivers, streams, lakes, and wetlands (Ref. 1.2).
Under RCRA, EPA has identified certain specific wastes that contain mercury as "listed hazardous
wastes" and also has provided criteria to determine whether mercury concentrations in nonlisted wastes
may meet the definition of a RCRA hazardous waste based on the "toxicity characteristic". Table 1.4 lists
the waste codes and descriptions for the RCRA listed wastes that contain mercury.
Nonlisted wastes are defined as hazardous wastes if the concentration of mercury in the waste extract
exceeds 0.2 milligrams per liter (mg/L) using the Toxicity Characteristic Leaching Procedure (TCLP)
under the RCRA toxicity criteria. Media cleanup standards for releases of mercury under the
Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) are based on
applicable or relevant and appropriate requirements (ARAR) in federal and state laws and regulations (for
example, MCLs under the SDWA) or site-specific levels, such as preliminary remediation goals (PRO),
established by evaluating potential receptors and exposure pathways.
1-6
August 2007
-------
Section 1 - Introduction
Table 1.4
RCRA Listed Hazardous Wastes for Mercury
Waste Code
K071
K106
K175
P065
P092
U151
Description
Brine purification muds from
which separately pre-purified
Wastewater treatment sludge
the mercury cell process in chlorine production, in
brine is not used
from the mercury cell process in chlorine production
Wastewater treatment sludge from the production of vinyl
mercuric chloride catalyst in an acetylene-based process
chloride monomer using
Mercury fulminate
Phenylmercury acetate
Mercury
Source: Ref 1.2
Cleanup goals for excavated soils at mercury-contaminated sites are often based on the RCRA land
disposal restrictions (LDR), which prohibit land disposal of hazardous wastes and contaminated soils mat
contain mercury unless the waste or soil is treated to meet specific standards or stringent criteria
associated with the regulator}' variances that are associated with the LDRs standards. EPA has
established different sets of LDR standards for mercury-containing hazardous wastes versus mercury-
contaminated soils. The LDR standards for wastes are based on the physical form of the waste that
requires treatment and the total mercury content. The LDRs categorize mercury wastes as low mercury
wastes, high mercury wastes, or elemental mercury wastes. Table 1.5 describes the LDRs for each of
these wastes.
Table 1.5
Land Disposal Restrictions for Low Mercury Wastes, High Mercury Wastes, and Elemental
Mercury
Type of Waste
Low mercury waste
(contain less than 260 mg/kg of total mercury)
High mercury waste
(contain greater than 260 mg/kg total mercury)
Elemental mercury waste
(with radioactive contamination)
Land Disposal Restrictions
If retorted, 0.2 mg/L TCLP
If other technologies are used - 0.025 mg/L TCLP
(solidification/stabilization often used to meet this
level)
Required to be roasted or retorted until waste
becomes a low mercury waste.
Residuals then required to meet 0.2 mg/L TCLP
Required be treated using amalgamation
Source: Ref. 1.10
Notes:
mg/L = Milligrams per liter
mg/kg = Milligrams per kilogram
EPA also has set a LDR standard of 0.15 mg/L for the treatment of wastewater. The LDRs for mercury
contamination in soils require treatment to reduce by 90 percent the original concentration of mercury in
the soil: however, treatment would not be required if the 90 percent reduction would result in a
concentration less than 10 times the appropriate TCLP concentration shown above in Table 1.5. Instead,
treatment would be required only to meet 10 times the appropriate TCLP concentration shown in Table
1.5. The treatment standard for contaminated soil is particularly relevant during remedial responses and
cleanup because it is less stringent than other LDR standards. The regulations at 40 Code of Federal
1-7
August 2007
-------
Section 1 - Introduction
Regulations Part 268.9 discuss additional rules about how to identify applicable LDR treatment standards
for characteristic hazardous wastes, including hazardous wastes that are listed and also exhibit one or
more RCRA characteristics.
1.8 Mercury Removal and Emergency Response
Various steps need to be taken to safely clean up and contain the contamination when an indoor spill of
mercury occurs. U.S. EPA Region 5 recommends that the following six "R's" of a response be followed:
• Referral - notifying the appropriate agencies about the spill, for example, the local health agency;
• Reconnaissance - assessing the extent and level of contamination;
• Relocation - determining the need for relocation, notifying residents, and securing alternate
accommodations;
• Removal - conducting decontamination, such as bagging contaminated items, cleaning
contaminated surfaces with appropriate chemicals, and ventilating the area;
• Replacement - appraising the removed items and restoring the area after decontamination is
complete; and
• Reoccupation - notifying residents when buildings are ready for reoccupation and assisting in the
reoccupation process.
These six factors are described in detail in the "Mercury Response Guidebook," developed by EPA's
Emergency Response Team (ERT) and EPA Region 5 (Ref 1.13).
The guidebook is designed to assist emergency and remedial professionals to coordinate and clean up
indoor mercury spills. States and other EPA regions also may have specific requirements during an
emergency response.
1.9 Detection and Site Characterization
Samples may be analyzed to measure the mercury concentration in a controlled laboratory environment or
in the field using site characterization techniques. In a laboratory, soil and water samples can be analyzed
using several different methods. The most common analytical methods use cold-vapor atomic absorption
spectroscopy. For aqueous samples, these methods include 245.1 (manual) (Ref. 1.15), 245.2 (automated)
(Ref. 1.15), and SW-846 Method 7470A (Ref. 1.20). For solid samples, these methods include 245.5
(sediment), 245.6 (tissues), and SW-846 Methods 7471A and 7473 (Refs. 1.15, 1.16).
A number of additional analytical procedures are available in laboratories that have the necessary
instrumentation. Organomercurial compounds may be analyzed by high-performance liquid
chromatography with an electron capture detector using Method 245.8. Total mercury may be measured
by atomic fluorescence spectrometry using Method 245.7 (Ref. 1.15) and SW-846 Method 7474 (Ref.
1.21). Mercury may also be analyzed by inductively coupled plasma/mass spectrometry using SW-846
Method 6020A (Ref. 1.15); by anodic stripping voltammetry using SW-846 Method 7472 (Ref. 1.15); and
by thermal decomposition, amalgamation, and atomic absorption spectrophotometry using SW-846
Method 7473 (Ref. 1.15). Some of the SW-846 methods require a digestion step (selected from SW-846
Methods 3005A through 3052) before analysis for nonaqueous samples or for all samples (Refs. 1.15,
1.16).
1-8 August 2007
-------
Section 1 - Introduction
Detection and site characterization of mercury using field-based analytical methods allows for ease of on-
site application and quicker decision-making. The technologies are portable and provide continuous, real-
time data. Some methods can be used outdoors with hand-held equipment; other, more rigorous methods
require the controlled environments of a mobile laboratory.
Field-based analytical methods, with systematic project planning and dynamic work strategies, form the
basis of the Triad, an innovative approach to decision-making for hazardous waste site characterization
and remediation. The three-pronged approach of the Triad aims to identify and manage decision
uncertainties in site cleanup (http://wwAv.triadcentral.org). Systematic project planning ensures that the
goals for the project are clearly identified and involves charting the most resource-effective course to
reach the desired outcomes. Dynamic work strategies, often in the form of a decision tree approved by
the regulatory agency, guide the Triad project teams in making on-site decisions. Using this strategy,
future site activities progress based on previous findings in the field. Real-time measurements generated
using field-based analytical techniques, rapid sampling platforms (for example, direct-push technologies),
geophysical tools, and on-site data management and display software make real-time decision-making
possible. Application of the Triad can condense a project's overall budget and lifetime, while increasing
the likelihood that the gathered data will guide better, more transparent decisions (Ref. 1.23).
The most commonly used site characterization techniques for mercury are immunoassay (IA) test kits,
field-based X-ray fluorescence (XRF) analyzers, and atomic absorption spectrometry instruments. lAs
employ antibodies that bind with a target compound or class of compounds. The process is colorimetric,
with the change in color indicating the approximate concentrations of the target compound. The presence
of the target compound is identified by comparing the color developed by a sample of unknown
concentration with the color formed by a standard that contains the analyte at a known concentration. The
concentration of the analyte is determined by the intensity of color in the sample. One type of
commercially available immunoassay test kit for inorganic contaminants is the BiMelyze immunoassay
for mercury in soil and water, manufactured by BioNebraska. The kit measures inorganic mercuric (+2)
ions in soil and water and has a detection limit of 0.5 milligrams per kilogram (mg/kg) for soil and 0.25
|ig/L for water (Refs .1.11,1.12).
XRF is used to characterize the mercury concentration in solid media such as soil and debris. It operates
on the concept of energy dispersive X-ray fluorescence spectrometry and responds to all chemical forms
of mercury. Samples are bombarded with X-rays produced by radioisotopes. When the X-rays are
absorbed by metals in the sample, the atoms emit X-rays of a specific wavelength. Each metal gives off
X-rays of different energy levels. The specific type or energy of the emitted X-ray is unique to a metal
and is called a '"characteristic" X-ray. It is possible to identify and sometimes quantify the composition of
metals in a sample by measuring the different energies of X-rays emitted by a sample exposed to an X-ray
source. A qualitative analysis of the samples can be made by observing the characteristic X-rays
produced. The quantity or intensity of the energy of the X-rays emitted is proportional to the
concentration of the target analytes (Refs. 1.11, 1.12).
Field instalments typically use atomic absorption spectrometry for detecting mercury vapor. Four
instruments that are representative of available technologies for mercury vapor detection are shown in
Appendix A. One of the technologies described in Appendix A is from the Ohio Lumex Company
(http://www.ohiolumex.com). This instrument is based on the principle of the Zeeman effect, which
involves splitting the spectrum by an applied magnetic field to achieve very low detection levels.
Further information on field-based analytical methods is available on EPA's Dynamic Field Activities
web page at http://w:ww.epa.gov/siiperfund/programs/dfa/fldmeth.htm. and through EPA's Field Analytic
Technologies Encyclopedia (FATE) Web site at http://fate.clu-m.org.
1-9 August 2007
-------
Section 1 — Introduction
References
1.1 U.S. Geological Survey (USGS). 2000. Circular 1197. The Materials Flow of Mercury in the
Economies of the United States and the World. http://pubs.usgs.gOv/circ/c 11977.
1.2 U.S. Environmental Protection Agency (EPA) and Environment Canada. 1999. Update:
Binational Toxics Strategy - Mercury Sources and Regulations.
http: //wwvv. epa.gov/bns/mercu rv/stephg .html.
1.3 U.S. EPA. 2005. Great Lakes Pollution Prevention and Toxics Strategy. Background
Information on Mercury Sources and Regulations.
http:/Avww.epa.gov/grtlakes/bnsdocs/mercsrce/merc_srce.htnil.
1.4 U.S. EPA. Mercury, http://www.epa.gov/mercury/. Accessed February 2005.
1.5 Massachusetts Department of Environmental Protect (MADEP). 1996. Mercury: Forms. Fate &
Effects (Chapter 2). http://www.mass.gov/dep/toxics/stypes/ligch2.htm.
1.6 U.S. EPA. 2004. Office of Emergency and Remedial Response. Comprehensive Environmental
Response Compensation and Liability Information System database (CERCLIS).
1.7 U.S. EPA. 2004. SuperfundNPL Assessment Program (SNAP) database.
1.8 U.S. EPA. 2004. Treatment Technologies for Site Cleanup: Annual Status Report (Eleventh
Edition). Office of Solid Waste and Emergency Response. EPA-542-R-03-009. February.
http ://clu -in .org/download/remed/asr/11/asr. pdf.
1.9 U.S. EPA. Mercury Laws and Regulations: Wastes and Products.
http://www.epa.gov/mercury/regs.htm#wastes. Accessed August 2006.
1.10 U.S. EPA. Office of Air Quality Planning & Standards and Office of Research and Development.
1997. Mercury Study Report to Congress Volume III: Fate and Transport of Mercury in the
Environment. EPA 452-R-97-005. http://www.epa.gov/ttn/oarpg/t3/repoits/volume3.pdf.
1.11 California Department of Toxic Substances Control. Final Decision to Certify Hazardous Waste
Environmental Technologies: BioNebraska, Inc., BiMelyze Field Screening Assay.
http://www.dtsc.ca.gov/Teclinolog\'Developnient/TechCert/bionebraska-bmfsa-techcert.cfm.
Accessed August 2006.
1.12 U.S. EPA. Field Sampling and Analysis Technologies Matrix: Sample Analysis and Tools Web
site, http://www.fitr.gov/site/analvsismatrix.html. Accessed August 2006.
1.13 U.S. EPA Region 5 Emergency Response Team (ERT). 2001. Mercury Response Guidebook.
March.
http://www.epa.gov/epaoswer/liaz^raste/merciiiy/pdf/cover-iiitro.pdf
http://www'.epa.gov/epaoswer/hazwaste/niercun/pdf/chapterl.pdf.
http://www.epa.gov/epaoswer/hazwaste/mercury/pdf/chapter2.pdf.
http: //ww w .epa. gov/epaoswer/hazw aste/iiierciin7pdf/chapter3. pdf.
http://vxww.epa.gov/epaoswer/liazwaste/inercurv/pdf/chapter4.pdf
http://www.epa. gov/epaoswrer/liazwaste/mercun-7pdf/cliapter5.pdf.
http://www. epa.gov/epaoswer/liazw aste/mercurv/pdf/chapter6.pdf.
http://www.epa.gov/epaoswer/hazwaste/mercurv/pdf/chapter7.pdf
1-10 August 2007
-------
Section 1 — Introduction
1.14 Committee on the Toxicological Effects of Methylmercury, Board on Environmental Studies and
Toxicology, National Research Council. 2000. Toxicological Effects of Methylmercury.
National Academy Press. Washington, B.C.
1.15 U.S. EPA New England Region 1 Library. April 2003 Revised Edition. Index to EPA Test
Methods, http://www.epa.gov/ne/info/testmethods/index.html. Accessed August 2006.
1.16 U.S. EPA. 2003. Office of Solid Waste and Emergency Response (OSWER). Test Methods-
SW-846. http://wvvw.epa.gov/epaoswer/hazwaste/test/sw846.htm.
1.17 U.S. EPA. 2004. Fact sheet: Improving Sampling, Analysis, and Data Management for Site
Investigation and Cleanup. April, http://clu-in.org/download/char/2004triadfactsheeta.pdf.
1.18 The Physical and Theoretical Chemistry Laboratory, Oxford University. Chemical and Other
Safety Information, http://ph\schem.ox.ac.uk/msds/. Accessed August 2006.
1.19 U.S. EPA. 2006. Binational Toxics Strategy Mercury Progress Report. From Alexis Cain, EPA
Region 5, to the Binational Toxics Strategy Mercury Workgroup.
http://www.epa.gov/region5/mercun/progress06.pdf. March 16.
1.20 U.S. EPA. 1994. Method 7470A - Mercury in Liquid Waste (Manual Cold-Vapor Technique)
http://www.epa.gov/svv-846/pdfs/7470a.pdf. September.
1.21 U.S. EPA. 2006. EPA's Roadmap for Mercury. http://vvww.epa.gov/mercury/roadmap.htm.
July.
1-11 August 2007
-------
This page is intentionally left blank.
1-12
-------
Section 2 - Comparison of Treatment Technologies
2.0
COMPARISON OF TREATMENT TECHNOLOGIES FOR MERCURY
The following section describes the types of technologies used to treat mercury and the application of the
technologies to treat soil, waste, and water.
2.1 Types of Technologies Used
Eight technologies have been identified to treat mercury-contaminated soil, waste, and water at full or
pilot scale. Figure 2.1 lists these technologies and categorizes them into two general groups: soil and
waste, and water. The technologies are discussed as follows in the remainder of this report: the
technologies typically used to treat soil and waste appear first, in the order of their number of full-scale
applications, followed by those typically used for water, in the same order. Only one of the eight
technologies, solidification/stabilization, has been used to treat elemental mercury; the remainder have
been used to treat mercury compounds.
Figure 2.1
Technologies Applicable to Treatment of Mercury-Contaminated Media
Mercury Treatment Technologies
Soil and Waste Treatment
• Solidification/Stabilization
• Soil Washing
• Thermal Treatment
• Vitrification
Water Treatment
• Precipitation/Coprecipitation
• Adsorption
• Membrane Filtration
• Bioremediation
Table 2.1 summarizes the applicability of the technologies to mercury-contaminated media. Technologies
that are applicable to one type of soil or waste are typically applicable to other types. For example, the
media treated in the projects identified for solidification/stabilization include soil, concrete fines, sludge,
liquid waste, mine tailings, and elemental mercury. Similarly, technologies applicable to one type of
water are generally applicable to other types. For example, both groundwater and various types of
wastewater are treated in the projects identified for precipitation/coprecipitation.
Regardless of the technology used and the type of media treated, an industrial hygiene review should be
conducted during the design phase for the treatment process and an industrial hygiene inspection should
be conducted during the initial startup phases for any mercury treatment project. These reviews are
particularly important for technologies that concentrate mercury from environmental media, and are
necessary to ensure that the treatment processes are designed and operated so that plant personnel will not
be exposed at concentrations in excess of the threshold limit value (TLV) for mercury, both elemental
(0.025 micrograms per cubic meter [mg/mj]) and organic (0.01 mg/nv1) (Ref. 2.1).
2-1
August 2007
-------
Section 2 - Comparison of Treatment Technologies
2.2 Frequency of Technology Use
Figures 2.2 and 2.3 show the number of mercury treatment projects that could be found in the literature
for each technology identified. Figure 2.2 shows the number of projects for soil and waste treatment
technologies, and Figure 2.3 shows the number of projects for water treatment technologies. The soil and
waste treatment technologies listed in Figure 2.2 are described below.
Figure 2.2
Number of Identified Applications of Mercury Treatment Technologies for Soil and Waste
Solidification/Stabilization Soil Washing Thermal Treatment
Technology
Source: Projects listed in Sections 3.0 to 6.0.
Figure 2.3
Number of Identified Applications of Mercury Treatment Technologies for Water
I
•5
Precipitation/
Coprecipitation
Adsorption Membrane Filtration Bioremediation
Technology
Source: Projects listed in Sections 7.0 to 10.0.
2-2
August 2007
-------
Section 2 - Comparison of Treatment Technologies
• Solidification/stabilization (in situ or ex situ) is the most frequently used technology for soil and
waste contamination in the projects identified. The data show mat this technology has been used
to meet regulatory cleanup levels, is commercially available to treat both soil and waste, and
generates a residual that typically does not require further treatment prior to disposal. Oilier
technologies for soil and waste are typically used for specific soil types.
• Soil washing/acid extraction (ex situ) is used primarily to treat soils with relatively low clay
content because these soils tend to be separable into a highly contaminated fines fraction and a
less contaminated sand fraction. It is also less effective for soils with high organic content
because organic compounds tend to interfere with contaminant desorption.
• Thermal treatment (ex situ), such as thermal desorption or retorting, is routinely used to treat
industrial and medical wastes mat contain mercury, but is also generally not suitable for soils with
high clay or organic content and typically requires an APC unit to treat mercury off-gas.
• Vitrification (in situ or ex situ) may be used when a combination of contaminants is present that
cannot be treated using only solidification/stabilization. It has also been used for wastes with a
high organic content because combustion of the organic content liberates heat, reducing the
external energy requirements.
The water treatment technologies listed in Figure 2.3 are described below.
• Precipitation/Coprecipitation is the most frequently used technology for water contamination in
the projects identified. The effectiveness of this technology is less likely to be reduced by
characteristics or contaminants that may affect other technologies, such as hardness or other
heavy metals. Systems that use this technology generally require skilled operators; therefore,
precipitation/coprecipitation is more cost-effective at a large scale where labor costs can be
spread over a larger amount of treated water produced.
• Adsorption for mercury treatment is more likely to be affected by media characteristics and
contaminants other than mercury when compared with precipitation/coprecipitation. Small-
capacity systems using these technologies tend to have lower operating and maintenance costs
and require less operator expertise. Adsorption tends to be used more often when mercury is the
only contaminant to be treated, for relatively smaller systems, and as a polishing technology for
the effluent from larger systems.
• Membrane filtration is effective for the treatment of mercury but is used less frequently because
its costs tend to be higher and it produces a larger volume of residuals than other mercury
treatment technologies. In addition, it is sensitive to a variety of contaminants and characteristics
in the untreated water. Suspended solids, organic compounds, colloids, and other contaminants
can cause membrane fouling.
• Bioremediation (in situ or ex situ) has been shown to be effective in several pilot-scale studies
identified in the research conducted for this report. The mechanisms that enable bioremediation
to reduce the concentration of mercury are not fully understood at this time. Mechanisms that
have been suggested include converting mercury to species that are retained in the biomass or
converting it to species that are more easily removed from water by another technology, such as
precipitation or adsorption. Bench-scale and additional pilot-scale studies are being conducted to
assess the effectiveness of bioremediation technologies for mercury at full scale.
2-3 August 2007
-------
Section 2 - Comparison of Treatment Technologies
Table 2.2 is a screening matrix for mercury treatment technologies. It can assist decision-makers in
evaluating candidate treatment technologies by providing information on relative availability, cost, and
other factors for each technology. The matrix is based on the Federal Remediation Technologies
Roundtable (FRTR) Treatment Technologies Screening Matrix (Ref. 2.2). However, it is limited to the
technologies addressed in this report, and the technology evaluation criteria have been reassessed to
compare the technologies only with those listed in Table 2.2, rather than all treatment technologies. In
addition, the table includes information about the factors that affect technology cost and performance.
Table 2.2 includes the following information:
• Development Status - The scale at which the technology has been applied. All technologies to
treat mercury have been applied at full scale, with the exception of bioremediation, which is
currently at pilot scale.
• Treatment Trains - ''Yes'' indicates that the technology is typically used in combination with
other technologies, such as pretreatment or treatment of residuals (excluding offgas). "No"
indicates that the technology is typically used independently.
• Residuals Produced - The residuals typically produced that may require additional management.
All of the technologies generate a solid residual, with the exception of membrane filtration, which
generates only liquid residuals. Vitrification and thermal treatment produce a vapor residual.
• O&Mor Capital Intensive - This category indicates the main cost-intensive parts of the system.
"O&M" indicates that the operation and maintenance costs tend to be high in comparison to other
technologies. "Cap" indicates that capital costs tend to be high in comparison to other
technologies.
• Factors that May Affect Performance or Cost - These include matrix characteristics and
operating parameters for each technology that may affect mercury treatment performance or cost.
These factors are described in more detail in Sections 3.0 through 10.0.
A limited amount of cost data on mercury treatment was found. Table 2.3 summarizes this information.
Considerations about the Performance and Cost Data Included in this Report
The information in this report is based on previously-compiled sources on mercury treatment and
information gathered from readily-available data sources. The data were collected over a period of time
(from December 2004 to January 2007), and the level of detail of the performance data varied among
projects. Not all the source documents have been peer-reviewed, and most data were not independently
verified. The remedial project managers (RPMs) for some Superfund sites were called, however, to
verify existing data and obtain additional project information.
Some projects treated numerous contaminants, in addition to mercury. Therefore, mercury may not have
been the target contaminant for these projects. In many cases, the cost information was incomplete. For
example, some projects included data for O&M costs only, and did not specify the associated capital
costs. In other cases, total costs were provided, but a breakdown of these costs was not included. No
mercury-specific cost data were identified for some technologies.
The cost data were taken from many types of sources, including EPA, Department of Defense (DoD),
other government sources, and information from technology vendors. The quality of these data varied,
with some sources providing detailed information about the items included in the costs, while other
2-4 August 2007
-------
Section 2 - Comparison of Treatment Technologies
sources provided little detail about the basis for costs. In most cases, the specific year for the costs was
not provided. The costs in Table 2.3 are reported in the identified references and are not adjusted for
inflation. This report does not provide a summary or interpretation of the costs in Table 2.3 because of
the variation in the type and quality of information.
In general. Table 2.3 includes only costs specifically for treatment of mercury'. General technology cost
estimates are unlikely to accurately predict actual treatment costs because mercury treatment is waste- and
site-specific.
Other Treatment Technologies for Mercury
This report identifies eight technologies used to treat mercury in soil, waste, and water. These
technologies were selected based on the availability of project-specific information for mercury treatment.
Other technologies (for example, ion exchange) may be reported as being able to treat mercury
contamination but are not covered in this report because project data were not found in the references
used.
Table 2.1
Applicability of Mercury Treatment Technologies"
Technology
Solidification/Stabilization
Soil Washing and Acid
Extraction
Thermal Treatment
Vitrification
Precipitation/Coprecipitation
Adsorption
Membrane Filtration
Biological Treatment
Soil"
•
•
•
•
Wastec
•
•
+
+
Water
Groundwater and
Surface Water d
•
•
Wastewater e
•
•
•
•
Source: Sections 3.0 to 10.0 of this report.
Notes:
+ Indicates treatment has been conducted at full or pilot scale.
a Media indicated here only if project-specific data are available. Some technologies may be applicable
to more than one type of media.
b Soil includes soil, debris, sludge, sediments, and other solid-phase environmental media.
c Waste includes nonhazardous and hazardous solid waste generated by industry.
d Groundwater and surface water also includes mine drainage.
e Wastewater includes nonhazardous and hazardous industrial wastewater and leachate.
2-5
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.2
Mercury Treatment Technologies Screening Matrix
Technology
Development
Status
Treatment
Train
(excludes off-
gas treatment)
Residuals
Produced
O&Mor
Capital
Intensive
Factors That May Affect Performance or Cost
Matrix Characteristics
Operating Parameters
Soil and Waste
Solidification/
Stabilization
Soil Washing and
Acid Extraction
Thermal Treatment
Vitrification
Full scale
Full scale
Full scale
Full scale
No
Yes
No
No
Solid
Solid,
Liquid
Solid,
Liquid,
Vapor
Solid,
Vapor
Cap
Cap&
O&M
Cap&
O&M
Cap&
O&M
• pH of media
• Presence of organic
compounds
• Particle size
• Moisture content
• Oxidation state of
mercury
• Soil homogeneity
• Presence of organic
compounds
• Particle size
• pH of media
• Moisture content
• Presence of organic
compounds
• Particle size
• Moisture content
• Lack of glass-forming
materials
• Particle size
• Moisture content
• Subsurface air pockets
• Presence of organic
compounds
• Type of binder and
reagent
• Mixing of waste and
binder
• Temperature
• Residence time
• System throughput
• Temperature and
pressure
• Temperature
2-6
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.2
Mercury Treatment Technologies Screening Matrix (continued)
Technology
Development
Status
Treatment
Train
(excludes off-
gas treatment)
Residuals
Produced
O&Mor
Capital
Intensive
Factors That May Affect Performance or Cost
Matrix Characteristics
Operating Parameters
Water
Precipitation/
Coprecipitation
Adsorption
Membrane Filtration
Biological Treatment
Full scale
Full scale
Full scale
Pilot scale
Yes
Yes
Yes
Yes
Solid
Solid
Liquid
Solid,
Liquid
Cap
O&M
Cap&
O&M
Cap
• pH of media
• Presence of other
contaminants
• Presence of other
contaminants
• pH of media
• Molecular weight of
contaminants
• Temperature
• Presence of other
contaminants
• pH of media
• Presence of other
contaminants
• Chemical dosage
• Fouling of adsorption
media
• Flow rate
• Type of filter
• Pressure
• Temperature
• Available nutrients
• Temperature
Source: Adapted from FRTR Screening Matrix (Ref 2.1) and Sections 3.0 to 10.0 of this report.
Cap Capital
O&M Operation and maintenance
2-7
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.3
Available Mercury Treatment Cost Data for Various Technologies
Site Name,
Location
Scale
Media,
Amount
Treated,
Date
Capital Cost
Annual
O&M Cost
Unit Cost
Total Cost
Cost Explanation/
Comments
Source
Solidification/Stabilization
Bunker Hill
Mining and
Metallurgical
Complex. Idaho
Rocky Mountain
Arsenal, Colorado
DOE Facility,
Portsmouth. Ohio
Full
Full
Pilot
Soil, 30,000
cy, 1997
Soil, 26,000
cy
Liquid waste
(ion
exchange
resin waste).
99 kg
NA
$2.8 million
$30,000
NA
NA
NA
NA
NA
NA
$230,000
$2.8 million
NA
Cost year not specified.
Components of total cost
not specified.
The implementation cost
is $2.8 million, and there
were no direct associated
O&M costs. Cost year
not specified.
Components of total cost
not specified.
Includes disposal costs.
Cost year not specified.
Components of total cost
not specified.
Section
3.0
Ref 3.19
Section
3.0
Ref. 3.21
Section
3.0
Ref. 3.13
Soil Washing
King of Prussia
Superfund Site.
New Jersey
Full
Soil, 13,570
cy, October
1993
NA
NA
NA
$7. 7 million
Includes all off -site
disposal costs for the
sludge cake. Cost year
not specified.
Components of total cost
not specified.
Section
4.0
Refs. 4.5.
4.17
2-8
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.3
Available Mercury Treatment Cost Data (continued)
Site Name,
Location
Georgia-Pacific
Oiler-alkali Plant,
Washington
Harbauer Facility,
State of Bavaria,
Germany
Scale
Full
Pilot
Media,
Amount
Treated,
Date
Sludge, 3,300
Ibs/day
Soil, 62 tons,
November
1994
Capital Cost
$1.25 million
$30 million
Annual
O&M Cost
$20,000
$3. 2 million
(Total O&M
costs)
Unit Cost
NA
NA
Total Cost
NA
NA
Cost Explanation/
Comments
Capital cost based on
treatment capacity of
2.000 to 3, 000 pounds per
day. Annual operating
costs based on high
automation of process
and occasional oversight.
Cost year not specified.
Costs are estimated and
are for a treatment train
of soil washing followed
by thermal desorption.
Costs are based on a 1994
conversion rate of 1 .5
DMto$l.
Source
Section
4.0
Ref 4.8
Section
4.0
Ref. 4.13
Thermal Treatment
Lipari Landfill,
OU3. New Jersey
Full
Soil and
sediment,
80,000 tons,
September
1995
$430,000
$5,019,292
(Total O&M
costs)
$68 per ton
$5,449,292
$632,737 of O&M costs
were for response to
baghouse fire, caustic
consumption, and
equipment modifications.
These costs were not
including in the unit cost.
Cost year is 1995.
Section
5.0
Ref. 5.8
2-9
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.3
Available Mercury Treatment Cost Data (continued)
Site Name,
Location
Site B (site name
and location
confidential),
Western U.S.
Harbauer Facility,
State of Bavaria,
Germany
Scale
Full
Pilot
Media,
Amount
Treated,
Date
Soil, 26,000
tons, August
1995
Soil, 62 tons,
November
1994
Capital Cost
$429,561
$30 million
Annual
O&M Cost
$2,830,956
(Total O&M
costs)
$3. 2 million
(Total O&M
costs)
Unit Cost
$125 per
ton
NA
Total Cost
$3,601,377
NA
Cost Explanation/
Comments
$285,000 of O&M costs
were for compliance
testing and analysis, and
$55,860 were for
excavation and disposal.
These costs were not
including in the unit cost.
Cost vear is 1995.
Costs are estimated and
are for a treatment train
of soil washing followed
by thermal desorption.
Costs are based on a 1994
conversion rate of 1.5
DMto$l.
Source
Section
5.0
Ref. 5.9
Section
5.0
Ref. 5.5
Vitrification
Parsons
Chemical/ETM
Enterprises
Superfund Site,
Mchigan
Lower Fox River,
Wisconsin
Full
Pilot
Soil and
sediment,
3,000 cy,
1994
Sediment,
27,000 Ibs,
2001
NA
$12 per ton
NA
NA
$267 per cy
NA
$800,000
NA
NA
Costs are estimated for a
full-scale facility treating
600 tons of sediments per
day, operating 350 days a
year for 15 years. Actual
costs for pilot test were
not available.
Section
6.0
Ref. 6.3
Section
6.0
Ref. 6.4
2-10
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.3
Available Mercury Treatment Cost Data (continued)
Site Name,
Location
Science and
Technology
Application
Research (STAR)
Center, Idaho
Scale
Pilot
Media,
Amount
Treated,
Date
Mixed waste.
NA, 1998
Capital Cost
$50,000,000
to
$86,200.000
Annual
O&M Cost
$48,000,000
to
$62,000,000
Unit Cost
NA
Total Cost
$124,000,000
to
$184,000,000
Cost Explanation/
Comments
Costs are estimated for a
full-scale facility treating
23.235 cy of waste over a
5 -year period. Actual
costs for pilot test were
not available.
Source
Section
6.0
Ref. 6.5
Precipitation/Coprecipitation
Marine Corps
Base Camp
Lejeune, North
Carolina
Squamish
Remediation Site.
Squamish, Canada
Full
Full
Ground-water,
41,0001bsof
contaminants
removed
Groundwater,
25 million
gallons
$4.66 Million
NA
$1 Million
NA
$49 per Ib
of
contaminant
removed
$2 per
gallon of
treated
ground-
water
NA
NA
Costs in 2000 dollars.
The costs are for the
entire treatment, which
consists of oxidation, pH
adjustment, precipitation,
air stripping, and
adsorption.
The unit cost is for the
reagent only. Costs in
2004 dollars.
Section
7.0
Ref. 7.8
Section
7.0
Ref. 7.6
2-11
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.3
Available Mercury Treatment Cost Data (continued)
Site Name,
Location
Scale
Media,
Amount
Treated,
Date
Capital Cost
Annual
O&M Cost
Unit Cost
Total Cost
Cost Explanation/
Comments
Source
Adsorption
King of Prussia
Superfund Site,
New Jersey
Civil and
Environmental
Engineering
Department,
Villanova
University.
Pennsylvania
Confidential
testing laboratory,
Massachusetts
Full
Pilot
Pilot
and Full
Ground-water,
225 gpm,
December
1997
Waste water,
2,000 gallons
Wastewater.
1,800 gallons
per day
$2.03 million
$7,700
$60.000
$449,000
(total for
performance
period is
$785,000)
$6,188
$91,980
NA
$0.107
capital and
annual
O&M
NA
$2.816
million
$69,580
NA
Costs in 1997 dollars.
The costs are for the
entire treatment system
(chemical precipitation
followed by air stripping
and granulated activated
carbon). The treatment
system treated multiple
contaminants. Total cost
is for the performance
period April 1995
through December 1997.
Costs in 2001 dollars.
Costs are projections
based on the results of the
pilot study. Costs were
based on a design life of
10 years. Design volume
is 64,980 gallons per
year.
Costs in 1992 dollars.
Treatment is ongoing
thereby affecting the
capital cost per gallon.
Costs for the pilot study
and full-scale operation
were combined in the
reference.
Section
8.0
Ref 8.1
Section
8.0
Ref. 8.6
Section
8.0
Ref. 8.7
2-12
August 2007
-------
Section 2 - Comparison of Mercury Treatment Technologies
Table 2.3
Available Mercury Treatment Cost Data (continued)
Site Name,
Location
Scale
Media,
Amount
Treated,
Date
Capital Cost
Annual
O&M Cost
Unit Cost
Total Cost
Cost Explanation/
Comments
Source
Membrane Filtration - No cost data identified
Bioremediation - No cost data identified
Notes:
NA
cy
kg
Ib
DM
Data not provided
Cubic yards
Kilograms
Pound
Deutsch Mark
O&M Operation and maintenance
2-13
August 2007
-------
Section 2 — Comparison of Mercury Treatment Technologies
References
2.1 Occupational Safety and Health Guideline for Mercury Vapor. U.S. Department of Labor.
Occupational Safety and Health Administration. Accessed January 2007.
http://wvvvv.osha.gov/SLTC/liealtliguidelines/mercurwapor/recognition.htnil.
2.2 Federal Remediation Technologies Reference Guide and Screening Manual, Version 4.0. 2004.
Federal Remediation Technologies Roundtable. June.
http://myw.frtr.gov/matrix2/section3/table3 2.html.
2-14 August 2007
-------
PART II
MERCURY TREATMENT TECHNOLOGY SUMMARIES
-------
This page is intentionally left blank.
-------
PART IIA
TREATMENT TECHNOLOGIES APPLICABLE TO SOIL AND WASTE
-------
This page is intentionally left blank.
-------
Section 3 - Solidification and Stabilization
3.0 SOLIDIFICATION AND STABILIZATION (INCLUDING AMALGAMATION)
Summary
Solidification and stabilization (S/S) is used to treat elemental mercury and mercury-contaminated soil
and sludge. This technology has been implemented at full scale and pilot scale. S/S reduces the
mobility of contaminants in the media by physically binding them within a stabilized mass or inducing
chemical reactions. Amalgamation, the dissolution of mercury in other metals and solidification to
form a non-liquid, semi-solid alloy called an amalgam, is often used for elemental mercury.
Technology Description and Principles
S/S is a process that has been used at full scale to treat mercury-contaminated soil and waste. Many pilot-
and bench-scale studies also have been conducted to identify' binders and reagents that are more effective
than conventionally used methods such as cement. S/S involves physically binding or enclosing
contaminants within a stabilized mass (solidification) or inducing chemical reactions between the
stabilizing agent and the contaminants to reduce their mobility (stabilization) (Ref. 3.1). Figure 3.1 shows
a model of an S/S system. Solidification is used to encapsulate or absorb the waste, forming a solid
material, when free liquids other than elemental mercury are present in the waste. Waste can be
encapsulated in two ways: microencapsulation and macroencapsulation. Microencapsulation is the
process of mixing the waste with the encasing material before solidification occurs. Macroencapsulation
refers to the process of pouring the encasing material over and around the waste mass, thus enclosing it in
a solid block (Ref. 3.2).
Figure 3.1
Model of a Solidification/Stabilization System
Dry
Reagents
Water
(If Required)
Liquid
Reagents
Mixer
T
Waste
Material
Stabilized Waste
3-1
August 2007
-------
Section 3 - Solidification and Stabilization
The stabilization process involves mixing soil or waste with binders such as Portland cement, sulfur
polymer cement (SPC), sulfide and phosphate binders, cement kiln dust, polyester resins, or polysiloxane
compounds to create a slurry, paste, or other semi-liquid state, which is allowed time to cure into a solid
form. For ex situ S/S processes that involve polyester resins, the resin and the waste are mixed at a low
rate for 5 to 10 minutes until a homogeneous mixture is formed. A catalyst is then added to mis mixture
to initiate a polymerization reaction and mixing continues at a high rate for another 2 to 5 minutes until a
rise in temperature signals the inception of curing. This waste-resin mixture is then allowed to cool down
to harden (Ref. 3.16).
EPA has identified amalgamation as the best demonstrated available technology (BOAT) for treatment of
liquid elemental mercury contaminated with radioactive materials (Ref. 3.3). Amalgamation is the
dissolution and solidification of mercury in other metals such as copper, nickel, zinc and tin, resulting in a
solid, nonvolatile product. It is a subset of solidification technologies, and it does not involve a chemical
reaction. Two generic processes are used for amalgamating mercury in w-astes: aqueous and non-aqueous
replacement. The aqueous process involves mixing a finely divided base metal such as zinc or copper
into a wastewater that contains dissolved mercury salts; the base metal reduces mercuric and mercurous
salts to elemental mercury, which dissolves in the metal to form a solid mercury-based metal alloy called
amalgam. The non-aqueous process involves mixing finely divided metal powders into waste liquid
mercury, forming a solidified amalgam. The aqueous replacement process is applicable to both mercury
salts and elemental mercury, while the non-aqueous process is applicable only to elemental mercury (Ref.
3.4). However, mercury in the resultant amalgam is susceptible to volatilization or hydrolysis. Therefore,
amalgamation is typically used in combination with an encapsulation technology (Ref. 3.5). The resulting
material must be disposed and must meet LDRs.
The S/S process may also include addition of pH adjustment agents, phosphates, or sulfur reagents to
reduce the setting or curing time, increase the compressive strength, or reduce the leachability of
contaminants (Refs. 3.6, 3.7). Information gathered for this report included 18 projects mat treated soil or
waste containing mercury using S/S. Twelve projects are full-scale applications, and six are at pilot scale.
Technology Description: S/S reduces the mobility of hazardous substances and contaminants in the
environment through both physical and chemical means. It physically binds or encloses contaminants
within a stabilized mass and chemically reduces the hazard potential of a waste by converting the
contaminants into less soluble, mobile, or toxic forms. Amalgamation is typically used to immobilize
elemental mercury by dissolving the mercury in another metal to form a semisolid alloy known as an
amalgam. The process is a physical immobilization and is often combined with encapsulation to
prevent volatization of mercury from the amalgam.
Media Treated:
• Soil
• Sludge
• Other solids
• Liquid wastes
• Industrial waste
• Elemental (liquid) mercury
3-2 August 2007
-------
Section 3 - Solidification and Stabilization
Technology Description (continued)
Binders and Reagents Used in S/S of Mercury:
• Cement
• Calcium polysulfide
• Chemically bonded phosphate ceramics (CBPC)
• Phosphate
• Platinum
• Polyester resins
• Polymer beads
• Polysiloxane compounds (silicon hydride and silicon hydroxide)
• pH adjustment agents
• Sodium dithiocarbamate
• Sodium metasilicate
• Sodium sulfide
• Sulfur polymer cement (SPC)
Binders and Reagents used in Amalgamation of Mercury:
• Copper
• Tin
• Nickel
• Zinc
Applicability, Advantages, and Limitations
S/S renders mercury more stable and less leachable. However, S/S does not reduce the total mercury
content of the waste. Instead, it reduces the leachability of mercury, yielding a product that still may
require disposal in a landfill. S/S often increased the volume of contaminated materials (Ref. 3.12).
The applicability of S/S depends on the mobility of mercury, which in turn depends on its oxidation state,
the pH of the waste disposal environment, and the specific mercury compound contained in the waste
(Ref. 3.6). This mobility is usually measured by testing the leachability of mercury under acidic
conditions. Typically, the leachability of mercury increases with a decrease in pH. Therefore, acidic
environments may increase the mobility of mercury in stabilized waste. Some studies also suggest that
soluble compounds of mercury, such as mercurous sulfate and mercuric sulfate, may form at a higher pH
(Ref. 3.2). This complex behavior of mercury in various disposal environments makes it a challenging
contaminant to treat. Some S/S processes involve pretreatment of the waste with reagents such as sodium
sulfide to convert mercury to a less soluble form (in this case, mercuric sulfide) before stabilization (Refs.
3.6, 3.8, 3.15). Other additives include pH buffers, catalysts such as platinum and sodium sulfide, and
other proprietary chemical agents. These processes may render the waste less mobile under a variety of
disposal conditions, but also may be more costly by adding more treatment steps.
3-3 August 2007
-------
Section 3 - Solidification and Stabilization
Factors that Affect S/S Performance and Cost
General factors:
• pH and redox potential: The pH and oxidizing or reducing properties of the waste and waste
disposal environment may affect the leachability of the treated material because these factors
affect the solubility of mercury and its leachability (Refs. 3.2, 3.11).
• Waste characteristics: Certain non-mercury compounds in the waste may interact with the
chemical reagents used in S/S, thus affecting the performance of the stabilization process. For
example, high concentrations of chloride in the waste may render phosphate additive
ineffective in stabilizing mercury (Ref. 3.10). Stabilization of dry wastes may be easier and
less expensive when compared with S/S of liquid wastes (Ref. 3.13).
• Particle size distribution: Fine particulate matter coats the waste particles and weakens the
bond between the waste and the binder (Ref. 3.18). Large aggregates in the waste could affect
operation of the mixer (Ref. 3.3).
• Mixing: Thorough mixing is necessary to ensure that the waste particles are coated with the
binder.
• Type of binder and reagent: The effectiveness of S/S depends in part on using the right type
of binder and reagent. The use of proprietary binders or reagents may be more expensive than
non-proprietary binders.
• Moisture content: For certain binders to be effective, the waste to be stabilized needs to
have a specific moisture content. Therefore, the waste may need to be pretreated to remove
the moisture (Ref. 3.2)
• Equipment scale: Application of S/S at large scale may reduce the unit costs when compared
with a small-scale application.
Factors specific to S/S of mercury-contaminated media:
• Oxidation state: The specific mercury compound or oxidation state of mercury may affect
the leachability of the treated material because these factors affect the solubility of mercury.
In addition, the presence of more than one species of mercury may complicate the treatment
process.
• Amount of mercury in waste: A higher concentration of mercury in waste may result in a
higher concentration of mercury in the leachate.
Type, Number, and Scale of Identified Projects
S/S of soil and waste that contain mercury has been applied at the full scale and tested at the pilot and
bench scales. Data sources used for this report included information about 12 full-scale and six pilot-
scale applications of S/S to treat mercury. All of the full-scale applications identified at sites in the U.S.
are at Superfund sites.
Summary of Performance and Cost Data
Table 3.1 summarizes information for the 12 full-scale and six pilot-scale projects. The performance of
S/S treatment is usually measured by leach testing a sample of the stabilized mass. Among the full-scale
projects, the media treated include soil and sludge. The amount of media treated ranged between 5,000
and 83,000 cubic yards. Site types include metal mining and smelting; landfills; chemical or radioactive
products manufacturing, and wood preserving. Performance data were available for six of the 12 full-
scale projects. Of the six projects, three were conducted in the United States (Projects 1, 2, and 6): the
other three were conducted in Canada (Projects 3, 4, and 5). The final leachable mercury concentrations
3-4 August 2007
-------
Section 3 - Solidification and Stabilization
were for 0.05 mg/L for Projects 1. 0.024 mg/L for Project 2, and 0.2 mg/L for Project 6. The final
leachable mercury concentrations were 0.01 to 0.04 mg/L for Project 3, 0.05 to 0.105 mg/L for Project 4,
and 0.0172 mg/L (first trial) and 0.0728 mg/L (second trial) for Project 6. The regulatory standard in
Canada specifies a treatment goal of 0.2 mg/L mercury TCLP before it can be disposed in a nonhazardous
landfill. All three projects met the standard.
Performance data were available for all of the pilot-scale projects. The media treated in the projects
include soil, sludge, ion exchange resin waste, mine tailings, and elemental mercury. The quantity of
media treated ranged from 0.45 kilograms to 330 kilograms. Various types of binders and stabilizing
agents were used, including Portland cement, sodium dithiocarbamate, sodium metasilicate, and
proprietary stabilizing agents. Performance data show that the final concentration of mercury in the
leachate was reduced to below 0.025 mg/L for all six projects. The data sources used for this report did
not contain any information about the long-term stability of mercury-containing soil and waste treated
using S/S.
Table 3.2 provides cost data for three projects, two full scale and one pilot scale.
Case Study: Allied Technology Group Mercury Stabilization Process
The S/S process was applied at the pilot scale on soils excavated from a Brookhaven National
Laboratory (BNL) facility (Ref. 3.9). These soils contained approximately 4,000 mg/kg of mercury.
The initial concentration of mercury in the TCLP leachate was 0.282 mg/L. The soil was split into
two parts, and each part was treated with a different stabilizing agent in a 7 cubic foot Essick mortar
mixer. One portion of the soil was treated with sodium dithiocarbamate (DTC), which reduced the
mercury levels in soil to 0.0139 mg/L (TCLP). The other portion of soil was treated with a liquid
sulfide formulation, which resulted in a final concentration of 0.002 mg/L of mercury in the TCLP
leachate (Project 13, Table 3.1). Both DTC and liquid sulfide additive reduced mercury to below the
regulatory limit (less than 0.025 mg/L in TCLP leachate). A total of 200 kilograms of contaminated
soil was treated by the end of this project, in November 2000.
Case Study: Brookhaven Chemical Holes
BNL conducted pilot-scale studies to demonstrate treatment of liquid elemental mercury and soils
contaminated with mercury using S/S (Ref. 3.3). The treatment process, known as Sulfur Polymer
Stabilization/Solidification (SPSS), involved use of SPC as the binding agent. The mercury-
contaminated soil was mixed along with the binding and stabilizing agents in a 1 cubic foot pilot-scale
vertical cone blender/dryer. SPC was added to the blender first in order to "lubricate" the walls of die
mixer, followed by the soil, and another layer of SPC. The soil/SPC mixture was mixed for
approximately 3 hours at 40±5 °C. The temperature of the cone blender was then increased to 100 °C
to reduce moisture in the soil. After the soil had dried, SPC chips were added to the blender, and the
temperature was increased to 135±5 °C to melt the SPC and encapsulate the chemically stabilized
mercury waste.
3-5 August 2007
-------
Section 3 - Solidification and Stabilization
Case Study: Brookhaven Chemical Holes (continued)
There was a slight variation in the treatment process for elemental mercury. First, SPC was added to
the blender, followed by liquid mercury in small fractions. This mixture was mixed for approximately
4 hours until the color of the powder changed from bright yellow to dark grey or black, indicating that
mercuric sulfide had formed. The vessel temperature was raised to the melting point of SPC (135±5
°C) to encapsulate the mercury. Off-gas generated in this process was passed through an off-gas
treatment unit composed of a heat exchanger, a liquid nitrogen cryogenic trap, and an activated carbon
filter, before it was vented to the atmosphere. Initial concentrations of leachable mercury in soil
ranged from 0.2 to 0.91 mg/L. The concentrations of leachable mercury in the stabilized soil ranged
from 0.0005 to 0.016 mg/L (Project 17, Table 3.1). Initial concentrations of leachable mercury were
not available for the samples of liquid elementary mercury. Final leachable concentrations for the
SPSS-treated elemental mercury ranged from less than 0.0004 to 0.004 mg/L (Project 18, Table 3.1).
A total of 330 kilograms of contaminated soil and approximately 62 kilograms of elemental mercury
were treated during this project.
3-6 August 2007
-------
Section 3 - Solidification and Stabilization
Table 3.1
Solidification/Stabilization Treatment Performance Data for Mercury
Project
Number
1
2
3
4
Industry and
Site Type
Metal Mining and
Smelting
Chemical and
Allied Products
Manufacturing
Former Chlor-
Alkali
Manufacturing
Plant
Former Chlor-
Alkali
Manufacturing
Plant
Waste or Media
Soil (ex situ)',
30,000 cy
Soil, 26,000 cy
Soil, 7,410 kg
(approx. 6.4 cy
assuming a soil
density of 1.5
g/cm3) (average
of two full-scale
trials)
Concrete fines
(from demolition
ofachlor-alkali
plant)
Scale
Full
Full
Full
Full
Site Name, Location,
and Project
Completion Date
Bunker Hill Mining
and Metallurgical
Complex,
Smelterville, Idaho;
1997
Rocky Mountain
Arsenal, Colorado
Site name unknown,
British Columbia
Province, Canada
Site name unknown,
British Columbia
Province, Canada
Initial Mercury-
Concentration
(mg/kg) or
Leachability (mg/L)
[Test method]
0.2 - 65 mg/L [TCLP]
965 j.ig/L [TCLP]
4,300 mg/kg [TWA]
245 ng/L (average of
two full-scale trials)
168 ng/L [TCLP]
Final Mercury
Concentration
(mg/kg) or
Leachability (mg/L)
[Test method]
0.05 -0.1 mg/L
[TCLP]
0.024 mg/L [TCLP]
430 -1,075 mg/kg
[TWA]
0.01 -0.04 mg/L
[TCLP] (combined
range for two full-
scale trials)
0.05 -0.105 mg/L
[TCLP]
Binder or
Stabilization Process
Proprietary stabilizing
agents.
6% cement slurry
Ferric sludge
(generated from
precipitation process
with ferric chloride) -
a mixed waste solid
consisting of water,
humic substances,
ferric hydroxides, and
small amounts of
other metals.
Ferric sludge
(generated from
precipitation process
with ferric chloride) -
a mixed waste solid
consisting of water,
humic substances,
ferric hydroxides, and
small amounts of
other metals.
Source
3.19
3.21
3.22
3.22
3-7
August 2007
-------
Section 3 - Solidification and Stabilization
Table 3.1
Solidification/Stabilization Treatment Performance Data for Mercury (continued)
Project
Number
5
6
7
8
9
10
11
12
Industry and
Site Type
Former Chlor-
Alkali
Manufacturing
Plant
Industrial and
Municipal
Landfills
Industrial
Landfill
Industrial
Landfill
Transportation
Industrial
Landfill
Radioactive
Products
Manufacturing
Lumber and
Wood Products
Manufacturing/
Wood Preserving
Waste or Media
Brine purification
sludge, 3,600 kg
(approx 3.12 cy
assuming a soil
density of 1.5
g/cm3) (first full-
scale trial); 1,800
kg (approx. 1.56
cy assuming a
soil density of 1.5
g/cm3) (second
full-scale trial)
Soil (ex situ)
Soil (ex situ),
17,140 cy
Sludge, 82,100 cy
Soil, 20,000 cy;
Sludge, 21, 347 cy
Soil
Soil, 4,667 cy
Soil
Scale
Full
Full
Full
Full
Full
Full
Full
Full
Site Name, Location,
and Project
Completion Date
Site name unknown,
British Columbia
Province, Canada
Cleve Reber,
Sorrento, Louisiana;
October 1995
Auto Ion Chemicals,
Michigan.
Davie Landfill,
Florida
Gulf Coast Vacuum
Services, Louisiana
Naval Surface
Warfare Center,
Dahlgren, Site 9
(1400 Area Landfill),
Virginia
Savannah River
(USDOE),OU16,
South Carolina
Wyckoff/Eagle
Harbor, Washington
Initial Mercury
Concentration
(mg/kg) or
Leachability (mg/L)
[Test method]
188 ug/L [TCLP]
(first full-scale trial)
252 ug/L [TCLP]
(second full-scale
trial)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Final Mercury
Concentration
(nig/kg) or
Leachability (mg/L)
[Test method]
0.0172 mg/L [TCLP]
(first trial, after 7 days
of treatment)
0.0728 mg/L [TCLP]
(second trial, after 8
days of treatment)
0.2 mg/L [TCLP]
N/A
N/A
N/A
N/A
N/A
N/A
Binder or
Stabilization Process
Ferric-lignin
derivatives and
Portland cement.
pH control
N/A
Cement
N/A
N/A
N/A
Portland cement
(Type II)
Source
3.23
3.19
3.19
3.19
3.19
3.19
3.19
3.19
3-8
August 2007
-------
Section 3 - Solidification and Stabilization
Table 3.1
Solidification/Stabilization Treatment Performance Data for Mercury (continued)
Project
Number
13
14
15
16
17
18
Industry and
Site Type
Research
Laboratory
DOE Facility
Research
Laboratory
Metal Mining
(Gold Mine)
Research
Laboratory
Research
Laboratory
Waste or Media
Soil; 200 kg
Liquid waste (ion
exchange resin
waste); 99 kg
Sludge from
LANL; 0.45 kg
Mine tailings;
10.35 kg
Soil; 330 kg
Elemental
mercury; 62 kg
Scale
Pilot
Pilot
Pilot
Pilot
Pilot
Pilot
Site Name, Location,
and Project
Completion Date
BNL, New York;
November 2000
DOE Facility,
Portsmouth, Ohio
LANL, New Mexico;
September 1998
Newmont Mining
Corporation,
Yanacocha, Peni
BNL - Chemical
Holes, New York
BNL - Chemical
Holes, New York
Initial Mercury
Concentration
(mg/kg) or
Leachability (mg/L)
[Test method]
0.282 mg/L [TCLP]
1,060 mg/L [TCLP]
0.125 mg/L [TCLP]
N/A
0.20 - 0.91 mg/L
[TCLP]
4,1 90 -5.570 mg/kg
[TWA]
N/A
Final Mercury
Concentration
(nig/kg) or
Leachability (mg/L)
[Test method]
NaDTC: 0.0139
mg/L [TCLP]
Sulfide: 0.002 mg/L
[TCLP]
0.0012 -0.0169 mg/L
[TCLP]
0.0012 -0.0169 mg/L
[TCLP]
0.009 - 0.039 mg/L
[TCLP]
0.0005 -0.016 mg/L
[TCLP]
O.0004 - 0.004 mg/L
[TCLP]
Binder or
Stabilization Process
Portland cement with
NaDTC or liquid
sulfide.
DTC with polymer
beads for absorption.
Cement with sodium
metasilicate.
SPC with sodium
sulfide as catalyst.
SPC with proprietary
stabilizing agent.
SPC with proprietary
stabilizing agent.
Source
3.9
3.13
3.15
3.16
3.5
3.5
Note:
BNL: Brookhaven National Laboratory
cy: cubic yards
DTC: Dithiocarbamate
kg: Kilograms
LANL: Las Alamos National Laboratory
mg/kg: Milligrams per kilogram
mg/L: Milligrams per liter
Hg/L: Micrograms per liter
N/A: Not Available
NaDTC: Sodium dithiocarbamate
SPC: Sulfur polymer cement
TCLP: Toxicity Characteristic Leaching Procedure
TWA: Total Waste Analysis
3-9
August 2007
-------
Section 3 - Solidification and Stabilization
Table 3.2
Available Mercury Treatment Cost Data**
Site
Bunker Hill
Mining and
Metallurgical
Complex,
Idaho
Rocky
Mountain
Arsenal.
Colorado
DOE Facility,
Portsmouth,
Ohio
Scale
Full
Full
Pilot
Amount
Treated
30,000 cy
26,000 cy
99kg
Capital
Cost
N/A
$2.8 million
$30,000
Annual O&M
Cost
N/A
Not applicable
N/A
Unit Cost
N/A
N/A
N/A
Total Cost
$230,000
$2.8 million
N/A
Cost
Explanation
Cost year not
specified.
Components of
total cost not
specified.
The
implementation
cost is $2.8
million and mere
were no direct
associated O&M
costs. Cost year
not specified.
Components of
total cost not
specified.
Includes disposal
costs.
Source
3.19
3.21
3.13
Project
Number in
Table 3.1
1
2
14
** All costs provided are estimated costs. Cost year is not available for any of the data provided.
Cy: Cubic yards
Ib: Pound
kg: Kilogram
N/A: Not available
O&M: Operation and maintenance
3-10
August 2007
-------
Section 3 - Solidification and Stabilization
References
3.1 Federal Remediation Technologies Reference Guide and Screening Manual, Version 4.0. 2001.
Federal Remediation Technologies Roundtable. September 5.
http://\vwvv.frtr.aov/matrix2/top_page.html.
3.2 Chattopadhyay, S., and W.E. Condit. 2002. Advances in Encapsulation Technologies for the
Management of Mercury-Contaminated Hazardous Wastes. August.
3.3 Kalb, P.O., J.W. Adams, and L.W. Milian. 2001. Sulfur Polymer Stabilization/Solidification
(SPSS) Treatment of Mixed-Waste Mercury Recovered from Environmental Restoration
Activities at BNL. BNL-52614. January.
3.4 U.S. EPA. 1998. Technologies for Immobilizing High Mercury Subcategory Wastes. July.
3.5 Spence, R. Date unknown. MASS: Mercury Amalgamation Stabilization/Solidification.
3.6 Bishop, P., R.A. Rauche, L.A. Reiser, M.T. Suidan. and J. Zhang. 2002. Stabilization and
Testing of Mercury Containing Wastes: Borden Sludge. April.
3.7 Osborne-Lee, I.W., T.B. Conley, G.A. Hulet, and M.I. Morris. 1999. Demonstration Results on
the Effects of Mercury Speciation on the Stabilization of Wastes. ORNL/TM-1999/120. August.
3.8 Mattus, C.H.. 2003. Measurements of Mercury Released from Solidified/Stabilized Waste
Forms-FY2002. ORNL/TM-2002/283. February.
3.9 Hulet, G.A.,V.C. Miao, M.I. Morris, J. Lewis, P. Randall, and L. Rieser. 2001. Demonstrations
to Support Change to the >260 ppm Mercury Treatment Regulations. March. (Presented at the
WM'Ol Conference, February 25-March 1,2001)
3.10 Rieser, L.A., P. Bishop, M.T. Suidan, H. Piao, R.A. Fauche, and J. Zhang. 2001. Stabilization
and Testing of Mercury Containing Wastes: Borden Catalyst. EPA/600/R-02/019. September.
3.11 Chattopadhyay, S. 2003. Evaluation of Chemically Bonded Phosphate Ceramics for Mercury
Stabilization of a Mixed Synthetic Waste. EP A/600/R-03/113. March.
3.12 Innovative Technology Summary Report: Mercury Contamination - Amalgamate (contract with
NFS and ADA) Demonstration of DeHg Process. 1999. DOE/EM-0471. September.
3.13 Innovative Technology Summary Report: Stabilize High Salt Content Waste Using Polysiloxane
Stabilization. 1999. DOE/EM-0474. August.
3.14 Innovative Technology Summary Report: Demonstration of GTS Duratek Process for Stabilizing
Mercury Contaminated (<260 ppm) Mixed Wastes. 1999. DOE/EM-0487. September.
3.15 Bowerman, B., J. Adams, P. Kalb, R-Y. Wan, and M. LeVier. 2003. Using the Sulfur Polymer
Stabilization/Solidification Process to Treat Residual Mercury Wastes from Gold Mining
Operations. February. (Presented at the Society of Mining Engineers Conference, Cincinnati,
OH, February 24-26," 2003)
3-11 August 2007
-------
Section 3 - Solidification and Stabilization
3.16 Innovative Technology Summary Report: Mixed Waste Encapsulation in Polyester Resins -
Treatment for Mixed Wastes Containing Salts. 1999. DOE/EM-0480. September.
3.17 U.S EPA Office of Solid Waste and Emergency Response. 2002. Arsenic Treatment
Technologies for Soil Waste, and Water. EPA-542-R-02-004. September.
http://www.cluin.org/download/remed/542r02004/arsenic_report.pdf.
3.18 U.S EPA Office of Solid Waste and Emergency Response. 2004. Annual Status Report Search
System. March, http://cfpiib2.epa.gov/asr.
3.19 U.S. EPA. 2004. E-mail from Larry Kimmel, U.S EPA to Danielle Wohler, Tetra Tech EM Inc.:
regarding S/S of mercury at the Rocky Mountain Arsenal site. November 18.
3.20 Zhuang, J.M., T. Walsh, T. Lam, and D. Boulter. 2003. Application of Ferric Sludge to
Immobilize Leachable Mercury in Soils and Concrete. Environmental Technology, Vol. 24, pp
1445 _1453.
3.21 Zhuang, J.M., T. Lo, T. Walsh, and T. Lam. 2004. Stabilization of high mercury contaminated
brine purification sludge. Journal of Hazardous Materials, Vol. B113, pp 157- 164.
3-12 August 2007
-------
Section 4 - Soil Washing and Add Extraction
4.0
SOIL WASHING AND ACID EXTRACTION
Summary
Soil washing and acid extraction are used for ex situ treatment of mercury-contaminated soil and
sediment. These technologies have been implemented at the full scale and pilot scale. In soil
washing, the contaminants are preferentially adsorbed onto the fines fraction of the soil or sediment.
The separated fines must be further treated to remove or immobilize the contaminant. In acid
extraction, the process depends on the solubility of the contaminants in the acidic extraction solution.
Spent extraction solution may require further treatment.
Technology Description and Principles
Soil washing is a water-based process that uses a combination of physical particle size separation and
aqueous-based chemical separation to reduce contaminant concentrations in soil (Refs. 4.7, 4.12). This
process is based on the concept that most contaminants tend to bind to the finer soil particles (clay and
silt) rather than the larger particles (sand and gravel). Physical methods can be used to separate the
relatively clean larger particles from the finer particles because the finer particles are attached to larger
particles through physical processes (compaction and adhesion). This process thus concentrates the
contamination bound to the finer particles for further treatment (Ref. 4.7).
Figure 4.1
Model of a Soil Washing System (Ref. 4.7)
Cont
Soil
.
animated
$
— », Ho mag
Sere
•"
Volatiles
J L
Makeup Water Recycl
Extracting Agent(s)
(Surfactants, etc.) | Lnem'
II 1
oil Prepared Soil SoN Washing ^Jf
enizing/ Process
ening
-Washing
-Size Separation
-Gravity
Separation
-Attrition Scrubbing
+
i r
Treat edflr
Emission Emissions
Contact *"
ed Water
j L
oals
wdown T Treated
ter Wastewater Water
* Ireatrnent i *
Contaminated
Sludges /Fines
Clean Scil
Oversized Rejects
Soil is first screened to remove oversized particles and then homogenized. The soil is then mixed with a
wash solution of water or water enhanced with chemical additives such as leaching agents, surfactants,
acids, or chelating agents to help remove organic compounds and heavy metals. Particles are separated
by size (cyclone or gravity separation, depending on the type of contaminants in the soil and the particle
size), concentrating the contaminants with the fines. The resulting concentrated fines or sludge usually
require further treatment because the soil washing process removes and concentrates the contaminants but
4-1
August 2007
-------
Section 4 - Soil Washing and Acid Extraction
does not destroy them. The coarser-grained soil is generally relatively "clean," requiring no additional
treatment. Wash water from the process is treated and either reused in the process or disposed (Ref. 4.7).
Commonly used methods for treating the wastewater include ion exchange and solvent extraction (Ref.
4.14)
Two of the projects in Table 4.1 use the BioGenesis Sediment Decontamination Technology. This
technology is designed to decontaminate fine-grained material contaminated with heavy metals, including
mercury (Refs. 4.2 and 4.18). The dredged material is first screened to remove oversized debris and then
is pre-treated using high-pressure water and proprietary biodegradable surfactants, specialty chemicals,
and chelators (Refs. 4.2 and 4.4). In this step, the solid particles physically separate from each other, and
naturally occurring organic material breaks up. After pre-treatment, an aeration step introduces air into a
chamber of contaminated material, and the resulting floatable organic material and free-phase organic
contaminants are then skimmed off the top of the chamber.
A second washing step uses collision impact forces to strip the sorbed contaminants and complex organic
coatings from the solid particles (Ref. 4.2). The slurry of solids and stripped organics is then treated in an
oxidation and cavitation unit to destroy the organic constituents. Oxidants used include hydrogen
peroxide and sodium percarbonate. Then a liquid/solid separation step mechanically removes the
decontaminated particles from the liquid phase. The resulting cake, containing cleaned sand, silt, and
clay, is mixed with amendments to create beneficial soil products such as topsoil. The contaminated
liquid is treated to precipitate metals and remove other contaminants and then is reused in the process
(Refs. 4.2, 4.4). This process was used in a pilot application to treat contaminated sediment from the New
York/New Jersey harbor. (See case study and Project 3 in Table 4.1.)
Acid extraction is an ex situ technology that uses an extracting chemical such as hydrochloric acid or
sulfuric acid to extract contaminants from a solid matrix by dissolving them in the acid (Ref. 4.6). It is
also known as chemical leaching. Contaminated material is first screened to remove coarse solids. Then
acid is introduced into the extraction unit. The characteristics of the treated material, contaminant type,
and contaminant concentration determine the residence time in the extraction unit, which usually varies
from 10 to 40 minutes (Ref. 4.6). The solid and liquid phases are then separated using hydroclones, and
the solids are transferred to a rinse system, where they are rinsed with water to remove entrained acid and
contaminants. They are then dewatered and mixed with lime and fertilizer to neutralize any residual acid
(Ref. 4.6). The acid extraction fluid and rinse waters are mixed with commercially available precipitants,
such as sodium hydroxide, lime, or other proprietary formulations, and a flocculant to remove the heavy
metals (Ref. 4.6). The precipitated solids may require additional treatment or may be disposed in a
landfill if they meet a TCLP mercury level of less than 0.025 mg/L (Ref. 4.10). Elemental mercury can
be recovered from the residual liquid from the precipitation step (Ref. 4.10).
Two of the projects in Table 4.1 employed the REMERC process, a patented acid extraction process that
includes a low-temperature extraction technique for recovering mercury and other heavy metals from
industrial wastes and contaminated soils (Refs. 4.10, 4.11). This process has been used to treat caustic
wastewater sludge that contains mercury from chlor-alkali plants, designated as hazardous waste
classification K106 by EPA (Ref. 4.8). The K106 waste consists of mercuric and mercurous chlorides,
mercuric sulfide, elemental mercury, and species of mercury absorbed on activated carbon (Ref. 4.10).
REMERC can be operated in both batch and continuous modes. This process involves leaching in two
stages (at pH 2 and 6), with each leaching step followed by a simultaneous washing and thickening step
using a patented thickener (Ref. 4.11). Chlor-alkali plant waste products such as sodium hypochlorite,
spent drying sulfuric acid, and depleted waste brine are used as additives to the leaching solution. The
mercury-containing solution then flows to a reactor, where triple-distilled quality elemental mercury is
recovered using aqueous-phase electrolysis or cementation on iron (Refs. 4.8 and 4.11). This process was
used in three projects (Projects 2, 6, and 7) described in Table 4.1.
4-2 August 2007
-------
Section 4 - Soil Washing and Add Extraction
Technology Description: Soil washing is an ex situ technology that takes advantage of the behavior
of some contaminants to preferentially adsorb onto the fines fraction. The contaminated soil is
suspended in a wash solution and the fines are separated from the suspension, thereby reducing the
contaminant concentration in the remaining soil. The contaminated water generated from soil washing
is treated with a technology suitable for the contaminants.
Acid extraction is an ex situ technology that uses an extracting chemical such as hydrochloric acid or
sulfuric acid to extract contaminants from a solid matrix by dissolving them in the acid. The metal
contaminants are recovered from the acid leaching solution using techniques such as aqueous-phase
electrolysis.
Media Treated:
• Soil (ex situ)
• Sediment (ex situ)
Agents Used in Soil Washing and Acid Extraction:
• Leaching agents
• Surfactants
• Acids - Hydrochloric acid, sulfuric acid
• Chelating agents
• Sodium chloride
Applicability, Advantages, and Limitations
Soil washing is suitable for use on soils contaminated with semivolatile organic compounds (SVOCs),
fuels, pesticides, and some VOCs (Refs. 4.7, 4.12). Soil washing has been used to treat soils
contaminated with heavy metals, including mercury. The technology can also be used to recover metals
(Ref. 4.7). This technology is best suited to treat contaminated soil composed primarily of larger-grained
particles such as sand and gravel, rather than fine-grained particles such as clay and silt (Ref. 4.15). It is
also best suited to treat contaminants that preferentially adsorb onto the fines fraction (Ref. 4.14).
An advantage of soil washing is that it can be used to reduce the volume of material that will require
further treatment, which potentially lowers the cost of cleanup and disposal of the contaminated material.
Soil washing may not be cost-effective for small quantities of contaminated material (Ref. 4.15). Soil
washing is usually performed at the site of the contamination, avoiding the risks associated with
transporting the contaminated soil off-site to a treatment facility. Chemicals are seldom released from the
soil washing process to the air. When the procedure is properly designed and operated, soil washing is
relatively safe (Ref. 4.15).
Soil washing is effective on homogenous, relatively simple contaminant mixtures (Ref. 4.14). Its
applicability and effectiveness may be limited for complex waste mixtures (such as metals mixed with
organic compounds) that make it difficult to formulate the appropriate washing solution. Soils with high
humic content may require pre-treatment to prevent interference with contaminant desorption. This
process may also have difficulty treating soils with a high (more than 40 percent) silt and clay fraction
(Refs. 4.7, 4.12).
Acid extraction is suitable for use on sediments, sludges, soils, and solid wastes contaminated with heavy
metals, including mercury (Ref. 4.6). The principal advantage of acid extraction is that hazardous
contaminants are separated from soils and sediments, thereby reducing the volume of hazardous waste to
be treated (Ref. 4.6).
4-3 August 2007
-------
Section 4 - Soil Washing and Add Extraction
The effectiveness of acid extraction may be limited in soils with higher clay content, which may reduce
the extraction efficiency and require longer contact time (Ref. 4.6). Certain health and safety issues
should be considered in using acid extraction, such as emissions from extracting agents (for example.
surfactants and concentrated acids) or the corrosivity of strong acids. Exposure to the extracting agents
and to the wastewater treatment chemicals may be controlled if the chemicals are added to the system
under closed or properly ventilated conditions. Additional respiratory protection (for example, an air-
purifying respirator with organic vapor cartridges) may also be used to control inhalation exposures (Ref.
4.6)."
Factors Affecting Soil Washing and Acid Extraction Performance and Cost
• Soil homogeneity: Soils that vary widely and frequently in characteristics such as soil type,
contaminant type and concentration, and where blending for homogeneity is not feasible, may
not be suitable for soil washing (Ref. 4.12).
• Soil type and soil particle size distribution: A larger amount of treatment residual may
require disposal for soils with a high proportion of fines (Ref. 4.14). Soils with high clay
content, cation exchange capacity, or mimic acid content tend to interfere with contaminant
desorption and limit the effectiveness of soil washing and acid extraction (Refs. 4.6, 4.12).
Carbonaceous soils tend to neutralize extracting acids and interfere with the acid extraction
process (Ref. 4.12).
• Total Organic Carbon: Total organic carbon (TOC) affects the desorption of contaminants
from the contaminated material and therefore the efficiency of the soil washing process (Ref.
4.20).
• pH: pH affects the operation of the soil washing process since the contaminants are extracted
from a matrix at specified pH ranges on the basis of the solubility of the contaminant at a
particular pH (Ref. 4.20).
• Multiple contaminants: Complex, heterogeneous contaminant compositions can make it
difficult to formulate a simple washing solution, requiring use of multiple, sequential washing
processes to remove contaminants (Ref. 4.12).
• Moisture content: The moisture content of the soil may make it more difficult to handle.
Moisture content may be controlled by covering the excavation, storage, and treatment areas
to reduce the amount of moisture in the soil (Ref. 4.14).
• Temperature: Cold temperature can freeze the washing solution and can affect leaching rates
(Ref. 4.14).
Type, Number, and Scale of Identified Projects
A total of eight projects were found. Of these eight, five were identified where soil washing was used to
treat organic mercury. Of these, one project was conducted at full scale at a Superfund site. Four projects
were conducted at pilot scale. Three projects were identified where acid extraction was used to treat
inorganic and elemental mercury. Of these, one was conducted at full scale, while two were performed at
the pilot scale.
Summary of Performance and Cost Data
Table 4.1 summarizes information for the two full-scale and six pilot-scale projects. The performance of
a soil washing or acid extraction treatment is typically measured by leach testing a sample of the treated
residue.
4-4 August 2007
-------
Section 4 - Soil Washing and Add Extraction
The two full-scale projects that used soil washing and acid extraction for treatment of mercury included a
waste processing facility and a chlor-alkali plant. The full-scale soil washing project treated 13,570 cubic
yards of contaminated soil, with the final concentration of inorganic mercury reduced from 100 mg/kg to
1 mg/kg, a reduction of 99 percent. This project is described in the case study and on Table 4.1 (Project
1). The full-scale project using acid extraction treated up to 3,300 pounds of sludge per day.
Performance data indicated a reduction in the concentration of mercury from 60,000 mg/L to 150 mg/L in
K106 wastewater sludge, and a TCLP value of less than 0.025 mg/L. This project is described in Table
4.1 (Project 2).
Among the six pilot-scale projects, four used soil washing and two used acid extraction. The first pilot-
scale soil washing project treated 700 cubic yards of contaminated sediment from the New York/New
Jersey harbor, with the final concentrations of inorganic mercury reduced from 3.1 mg/kg to 0.3 mg/kg
(Refs. 4.2 and 4.4). This project is described in the case study and on Table 4.1 (Project 3). The second
pilot-scale soil washing project treated 800 liters of homogenized sediment from the Kai Tak Approach
Channel in Hong Kong (Table 4.1, Project 4). Performance data indicate a reduction in concentrations of
inorganic mercury from 0.73 mg/kg to nearly 0.08 mg/kg (Ref. 4.3). Soil washing was also used at the
pilot scale to treat mercury-contaminated soil at a facility in Germany (Table 4.1, Project 5), where
inorganic mercury was reduced from 875 mg/kg and a TCLP of 0.082 mg/L to a final concentration of
less than 20 mg/kg and a TCLP of 0.006 mg/L. The BioGenesis Sediment Decontamination Technology
was used at a pilot scale (See Table 4.1, Project 8) and reduced mercury concentrations from 9.5 mg/kg to
1.49 mg/kg.
The two pilot-scale acid extraction projects were conducted at chlor-alkali plants, and both used the
patented REMERC process described previously to treat K106 wastewater sludge. Before treatment,
concentration of mercury in the wastewater sludge ranged from 55,000 mg/kg to 110,000 mg/kg.
Performance data for both projects indicated nearly 99 percent reduction in the concentration of mercury
and a TCLP of less than 0.025 mg/L for the treated material. These projects are described in Table 4.1
(Projects 6 and 7).
Case Study: Soil Washing of Mercury-Contaminated Soil at King of Prussia Superfund Site
The King of Prussia Superfund Site in Winslow Township, New Jersey, is a former waste processing
and recycling facility. Soils were contaminated with arsenic, beryllium, cadmium, chromium, copper,
lead, mercury, nickel, selenium, silver, and zinc from improperly disposed wastes (Table 4.1, Project
1). Approximately 13,570 cubic yards of mercury-contaminated soil, sludge, and sediment was
treated using soil washing in 1993. This project was EPA's first full-scale application of soil washing
to remediate a Superfund site. The system consisted of a series of hydroclones, conditioners, and froth
floatation cells. Soil washing additives included a polymer and a surfactant. The treatment reduced
concentrations of inorganic mercury from 100 mg/kg to the cleanup goal of 1 mg/kg, a reduction of 99
percent. Residual sludges were disposed off site as nonhazardous waste, and the treated soil was used
as backfill at the site.
4-5 August 2007
-------
Section 4 - Soil Washing and Add Extraction
Case Study: Soil Washing of Sediment from the New York/New Jersey Harbor
Sediments and soils around the New York/New Jersey harbor have been contaminated by various
industrial activities from colonial times to the present (Ref. 4.9). Because of the shallow natural
depth, the harbor must be periodically dredged to maintain safe passage for vessels, generating
millions of cubic yards of contaminated sediment annually that require some form of decontamination
for disposal (Ref. 4.1). The harbor sediment is contaminated with polycyclic aromatic hydrocarbons
(PAHs), pesticides, dioxins and furans, and heavy metals, including mercury. The sediment in the
New York/New Jersey harbor contains 52 percent silt, 42 percent clay, 4 percent sand, and a small
fraction of larger grained particles. The total organic content ranges from 3 to 10 percent (Refs. 4.2,
4.9).
The pilot demonstration was conducted from November 1998 to March 1999 and treated 700 cubic
yards of a representative sample of dredged material from near a petroleum facility (Table 4.1. Project
3) (Ref. 4.2). The project evaluated the effectiveness of the BioGenesis Sediment Washing
Technology, which is designed to decontaminate fine-grained sediment material (Ref. 4.2).
Concentrations of inorganic mercury in this sample were found to be about 3.1 mg/kg. which is within
the typical range of mercury in New York/New Jersey harbor sediment of 0.2-13.6 mg/kg (Ref. 4.2).
This process resulted in a 92 percent removal of mercury, to a final concentration of 0.3 mg/kg. This
concentration was below the mercury cleanup criteria of 14 mg/kg (New Jersey Residential Direct
Contact Soil Cleanup Criteria) (Refs. 4.2 and 4.4).
4-6 August 2007
-------
Section 4 - Soil Washing and Acid Extraction
Table 4.1
Soil Washing and Acid Extraction Performance Data for Mercury
Project
Number
1
2
3
Industry and Site
Type
Waste processing
facility
Chlor-alkali plant
Petroleum facility
Waste or
Media
(Quantity)
Soil (13,570 cy)
K 106 Waste
water sludge
(3,300 Ibs/day)
Sediment (700
cy)
Scale
Full
Full
Pilot
Site Name,
Location, and
Project Completion
Date
King of Pnissia
Superfund Site,
Winslow Township,
New Jersey; October
1993
Georgia-Pacific
Chlor-alkali plant,
Bellingham,
Washington
New York/ New
Jersey Flarbor
Sediment Treatment
March 1999
Initial Mercury
Concentration
(mg/kg) and/or
Leachability (mg/L)
[Test Method]
100 mg/kg
60,000 mg/kg
3.1 mg/kg
Final Mercury
Concentration
(mg/kg) and/or
Leachability (mg/L)
[Test Method]
1 mg/kg
150 mg/kg; <0. 025
mg/L [TCLP]
0.3 mg/kg
Soil Washing or Acid
Extraction Process
Soil washing consisting
of screening, separation,
and froth floatation.
Acid Extraction
(REMERC Process)
involving leaching with
5 to 27 percent sodium
chloride at pH 6,
followed by solid/liquid
separation and solids
washing, followed by a
second leaching with 5
to 27 percent sodium
chloride atpH2,
followed by a second
solid/liquid separation
and solids washing and
finally mercury
recovery on liquid
residual using
cementation on iron;
operated continuously.
Biogenesis Sediment
Decontamination
Technology. Soil
washing consisting of
pre-processing,
aeration, sediment
washing, and cavitation
and oxidation.
Source
4.5,4.17
4.8,4.10,
4.11
4.2,4.4
4-7
August 2007
-------
Section 4 - Soil Washing and Acid Extraction
Table 4.1
Soil Washing and Acid Extraction Performance Data for Mercury (continued)
Project
Number
4
5
6
Industry and Site
Type
N/A
N/A
Chlor-alkali plant
Waste or
Media
(Quantity)
Homogenized
sediment (800
liters)
Soil (62 tons)
K 106 Waste
water sludge
(volume
unknown)
Scale
Pilot
Pilot
Pilot
Site Name,
Location, and
Project Completion
Date
Kai Tak Approach
Channel. Hong Kong
Harbauer Facility,
State of Bavaria,
Germany; November
1994
BF Goodrich, Calvert
City, Kentucky
Initial Mercury
Concentration
(mg/kg) and/or
Leachability (mg/L)
[Test Methodl
0.73 mg/kg
875 mg/kg; 0.082
mg/L [TCLP]
11 0,000 mg/kg
Final Mercury
Concentration
(mg/kg) and/or
Leachability (mg/L)
[Test Method]
0.08 to 0.25 mg/kg
<20 mg/kg; 0.006
mg/L [TCLP]
220 mg/kg; <0.025
mg/L [TCLP]
Soil Washing or Acid
Extraction Process
Soil washing consisting
of pre-processing.
aeration and sediment
washing using chelating
agents and surfactants.
Treatment train
consisting of soil
washing followed by
vacuum-distillation
process to treat fines.
Acid Extraction
(REMERC Process)
involving leaching with
5 to 27 percent sodium
chloride at pH 6,
followed by solid/liquid
separation and solids
washing, followed by a
second leaching with 5
to 27 percent sodium
chloride at pLI2,
followed by a second
solid/liquid separation
and solids washing and
finally mercury
recovery on liquid
residual using mercury
recovery aqueous phase
electrolysis; operated in
batch treatment mode.
Source
4.3
4.13
4.8,4.11
4-8
August 2007
-------
Section 4 - Soil Washing and Acid Extraction
Table 4.1
Soil Washing and Acid Extraction Performance Data for Mercury (continued)
Project
Number
7
8
Industry and Site
Type
Chlor-alkali plant
Chemical
Industry contamina
tion of Shipping
Canals
Waste or
Media
(Quantity)
K 106 Waste
water sludge
(volume
unknown)
Sediment (330
in3)
Scale
Pilot
Pilot
Site Name,
Location, and
Project Completion
Date
Pioneer Chlor-Alkali,
St. Gabriel, Louisiana
Venice Port
Authority, Venice,
Italy,
Initial Mercury
Concentration
(mg/kg) and/or
Leachability (mg/L)
[Test Methodl
55,000 mg/kg
9. 5 mg/kg
Final Mercury
Concentration
(mg/kg) and/or
Leachability (mg/L)
[Test Method]
50 mg/kg; O.025
mg/L [TCLP]
1.49 mg/kg
Soil Washing or Acid
Extraction Process
Acid Extraction
(REMERC Process)
involving leaching with
5-27 percent sodium
chloride at pH 6,
followed by solid/liquid
separation and solids
washing, followed by a
second leaching with 5-
27 percent sodium
chloride at pH2,
followed by a second
solid/liquid separation
and solids washing and
finally mercury
recovery on liquid
residual using
cementation on iron;
operated continuously.
Biogenesis Sediment
Decontamination
Technology
Source
4.8,4.11
Note:
cy: cubic yards
Ibs: pounds
mg/L: milligrams per liter
mg/kg: milligrams per kilogram
N/A: Not Available
TCLP: Toxicity Characteristic Leaching Procedure
4-9
August 2007
-------
Section 4 - Soil Washing and Acid Extraction
Table 4.2
Available Mercury Treatment Cost Data for Soil Washing and Acid Extraction
Site
King of Prussia
Superfund Site,
Winslow
Township, New
Jersey
Georgia-Pacific
Chlor-alkali plant,
Bellingham,
Washington
Harbauer Facility,
State of Bavaria,
Germany* *
Scale
Full
Full
Pilot
Amount
Treated
13,570 cy
3,300
Ibs/day
Soil, 62
tons. Nor
1994
Capital
Cost
N/A
$1.25
million
$30
million
Annual
O&M Cost
N/A
$20,000
$3.2 million
Total Cost
$7.7 million
N/A
N/A
Cost Explanation
Includes all off-site
disposal costs for the
sludge cake. Cost year
not specified.
Components of total cost
not specified.
Capital cost based on
treatment capacity of
2,000 to 3.000 Ibs/day;
Annual operating costs
based on high automation
of process and occasional
oversight. Cost year is
not available for any of
the data provided.
Costs are estimated and
are for a treatment train
of soil washing followed
by thermal desorption.
Costs are based on a 1994
conversion rate of 1.5
DMto$l.
Source
4.5,4.17
4.8
4.13
Project
Number in
Table 4.1
1
2
5
** All costs and amounts treated are estimated.
cy: Cubic yards
Ib: Pound
kg: Kilogram
DM: DeutschMark
N/A: Not available
O& M: Operation and maintenance
4-10
August 2007
-------
Section 4 - Soil Washing and Acid Extraction
References
4.1 BioGenesis™ Sediment Washing. New York/New Jersey Harbor. Web site accessed on October
18,2004. http://www.bio.geiiesis.com/2pager~2.html.
4.2 BioGenesis Enterprises, Inc. and Roy F. Weston, Inc. 1999. "BioGenesisSM Sediment Washing
Technology". Full-Scale, 40 Cy/Hr, Sediment Decontamination Facility for the NY/NJ Harbor.
Final Report on the Pilot Demonstration Project. December.
http://www.bnl.gov/wrdadcon/publications/reports/BioGenesis BNL.PDF.
4.3 DeDen, M., K. Johnson, C. Stevens, M. Amiran, and C. Wilde. 2003. Ex-Situ Remediation of
Hong Kong Marine Sediments Using BioGenesisSM Washing Technology. Proceedings of the
Second International Conference on Remediation of Contaminated Sediments, Venice, Italy,
September 30 - October 3, 2003. http:/Avww.biogenesis.com/venicedocs/ 10TechPap03.pdf.
4.4 Jones, K.W., H. Feng, E.A. Stern, J. Lodge, and N.L. Clesceri. 2001. Dredged Material
Decontamination Demonstration for the Port of New York/New Jersey. Journal of Hazardous
Materials. 85: 127-143. July, http://www.elsevier.com/locate/iliazmat.
4.5 Federal Remediation Technologies Roundtable (FRTR). 1995. Technology Cost and
Performance - Soil Washing at the King of Prussia Technical Corporation Superfund Site.
http://costperformance.org/profile.cfm?ID=125&CaseID=125.
4.6 FRTR. 2001. Chemical Extraction. Federal Remediation Technologies Reference Guide and
Screening Manual, Version 4.0. Septembers, http://wvvw.fitr.gov/matrix2/section4/4-15.html.
4.7 FRTR. 2001. Soil Washing. Federal Remediation Technologies Reference Guide and Screening
Manual, Version 4.0. Septembers. http://www.frtr.gov7inatrix2/section4/4-l 9.html.
4.8 Selby. J., and L.G. Twidwell. The Recovery and Recycle of Mercury from Chlor-Alkali Plant
Wastewater Sludge. Web site accessed on December 8, 2004.
http://wwvv.mtech.edii/metallurgv/Facultv%20Info/Twidwell/Mercun/Microsoft%20Word%20-
%20JOM%20HG%20final%20\vord%201%202001 .pdf.
4.9 Stern. E,, K.W. Jones, K. Donato. J.D. Pauling, J.G. Sontag, N.L.Clesceri, M.C. Mensinger, and
C.L. Wilde. 1998. Maintaining Access to America's Intermodal Ports/Technologies for
Decontamination of Dredged Sediment: New York/ New Jersey Harbor. May.
http://www. bnl.gov/wrdadcon/publications/articles/Same-1998. pdf.
4.10 Universal Dynamics. 1994. The "REMERC" Process for Treatment of K106 Mercury Mud.
March, http://www.udl.com/pdf/p35.pdf.
4.11 Universal Dynamics. The REMERC™ Process. Web site Accessed on October 20, 2004.
http ://udl .com/s\stems/re me re x .html.
4.12 U.S. Environmental Protection Agency (EPA) Office of Emergency and Remedial Response.
1997. Engineering Bulletin: Technology Alternatives for Remediation of Soils Contaminated
with As, Cd, Cr, Hg, and Pb. EPA-540-S-97-500. August.
http://www.epa.gov/tio/download/remedAdtclialt.pdf.
4-11 August 2007
-------
Section 4 - Soil Washing and Acid Extraction
4.13 EPA Office of Research and Development. 1996. Harbauer Soil Washing/ Vacuum-Distillation
System, Innovative Technology Evaluation Report. August.
http://www.epa.gov/ORD/SiTE/Bilateral/harbauer.pdf.
4.14 EPA Office of Solid Waste and Emergency Response. 2002. Arsenic Treatment Technologies
for Soil, Waste, and Water. EPA-542-R-02-004. September.
http://wivw.cliiin.org/download/remed/542r02004/arsenic report.pdf.
4.15 EPA Office of Solid Waste and Emergency Response. 2001. A Citizen's Guide to Soil Washing.
EPA 542-F-01-008. May. http://www.epa.gov/tio/dovvnload/remed/soilwash.pdf.
4.16 EPA Office of Solid Waste and Emergency Response. 2004. Treatment Technologies for Site
Cleanup: Annual Status Report (Eleventh Edition). March.
http://clu-org/dow-nload/remed/asr/11 /asr.pdf.
4.17 FRTR 1998. Guide to Documenting and Managing Cost and Performance Information for
Remediation Projects. EPA 542-B-98-007. October, http://www.frtr.gov/pdf/guide.pdf.
4.18 BioGenesis Italia, LLC, MWH Americas, Inc., and Jan de Nul, N.V. April 2005. 'Technical
Report, Pilot Scale Demonstration Project of the BioGenesis Sediment Decontamination Process,
Autorita Portuale di Venezia, Porto Marghera, Venice, Italy". Web site accessed on June 7, 2007.
http://biogenesis.com/VenicePilotFinalReport20050412-l.pdf.
4-12 August 2007
-------
Section 5 - Thermal Treatment
5.0 THERMAL TREATMENT
Summary
Thermal desorption and retorting are two common ex situ methods of thermal treatment for mercury.
These technologies have been used to treat mercury-contaminated soil, sediments, and wastes at full
scale. Thermal treatment usually involves application of heat and reduced pressure to volatilize
mercury from the contaminated medium, followed by conversion of the mercury vapors into liquid
elemental mercury by condensation. The liquid elemental mercury collected from the condenser units
can be reused or further treated, typically by amalgamation, for disposal. Off-gasses also may require
further treatment. Thermal treatment is also used at mercury recycling facilities to treat wastes that
contain mercury.
Technology Description and Principles
Thermal treatment has been used at the full scale to treat mercury-contaminated soil and waste. It uses
heat to volatilize mercury from the contaminated medium. Thermal treatment processes commonly used
to treat mercury-contaminated wastes are thermal desorption and retorting or roasting (Refs. 5.1, 5.2, 5.3).
The main components of a thermal desorption system are a pre-treatment and material handling unit, a
desorption unit, and a post-treatment unit for treatment of off-gas and processed medium (soil or waste).
Pre-treatment of contaminated media involves removing extraneous matter such as plastic or rubber and
dewatering to achieve suitable moisture content (Refs. 5.1, 5.2). The pre-treated waste is then placed in
the thermal desorber, where it is heated to volatilize mercury in the waste. Types of thermal desorbers
include direct-fired rotary kilns and indirectly heated screw or auger systems. The motion of the rotary
drum or auger agitates the waste, promoting mixing and more uniform heating. A typical thermal
desorption unit for mercury removal operates at temperatures ranging from 320 to 700 °C (600 to 1,300
°F) (Refs. 5.2, 5.3, 5.5, and 5.10).
The boiling point of elemental mercury at 1 atmosphere pressure is 350°C (662°F), which renders it
suitable for removal by thermal processes (Refs. 5.3, 5.4). The high temperatures used in thermal
desorption units convert mercury into the gaseous or vapor phase, which is collected and further treated,
removing mercury from the contaminated medium.
Off-gas generated by the desorption unit is passed through wet scrubbers or fabric filters to remove
particulate matter. The off-gas stream free from particulates is then condensed to collect liquid elemental
mercury. The remaining off-gas vapors generated in the process are passed through additional APC
devices such as sulfur-impregnated carbon to capture any residual mercury. Mercury reacts with sulfur in
the activated carbon to form a more stable compound, mercuric sulfide. Organic contaminants in the off-
gas are treated in a secondary combustion chamber or a catalytic oxidizer and then released into the
atmosphere (Refs. 5.2, 5.3, and 5.4). Any APC should undergo periodic performance testing to ensure
that mercury is not leaking.
A retorting or roasting process is similar to thermal desorption except that it does not involve agitation
(Ref 5.4). The main components of a mercury retort system include a retort oven, condensers, a mercury
trap, a sulfonated carbon adsorber, and a vacuum pump. Mercury-contaminated material is placed in pans
that are stacked in the retort oven. Retorts can be heated either electrically or with fuel burners. In
electric retorts, the heating elements are in direct contact with the contaminated medium, while in fuel-
fired retorts the burner heats the air surrounding the retort (Ref. 5.3). The retort chamber typically
operates at temperatures of 425 to 540°C (797 to 1004°F) under a vacuum to facilitate volatilization of
5-1 August 2007
-------
Section 5 - Thermal Treatment
mercury. The resulting off-gas is passed through condensers to collect liquid elemental mercury. The
off-gas is then passed through sulfur-impregnated carbon to capture any residual mercury (Ref 5.3).
Figure 5.1 presents the flow diagram of a thermal treatment system.
Figure 5.1
Model of a Thermal Desorption or Retort System
Particulate Removal
(Wet Scrubber or
Fabric Filter)
Contaminated
Soil/Waste
Thermal Desorber/
Vacuum Retort
Disposal of
Treated Material
Technology Description: Thermal treatment processes are physical methods to remove mercury from
the contaminated medium. Heat is supplied under reduced pressure to the contaminated soil or waste,
volatilizing mercury. The off-gas is treated by condensation to generate liquid elemental mercury.
The treated medium may be used as fill material or disposed.
Media Treated:
• Soil
• Sludge
• Sediment
• Other solids
Types of Thermal Treatment Systems:
• Rotary kiln - combustion
• Heated screw or auger - hot oil or steam
• Retort - conductive electrical heating or fuel-fired
Applicability, Advantages, and Limitations
Thermal desorption and retorts are used to treat soil and wastes that contain mercury. Thermal desorption
treatment systems are more commonly used to treat volatile and semivolatile organic contaminants such
as polychlorinated biphenyls (PCBs), PAHs, and pesticides. Because of the high volatility of mercury,
they are also applicable in treatment of mercury-contaminated waste (Ref. 5.4). Thermal desorption and
retorting are both usually performed at reduced pressures to lower the boiling point of mercury, in turn
allowing the unit to be operated at a lower temperature, which reduces energy costs. Operating under a
5-2
August 2007
-------
Section 5 - Thermal Treatment
vacuum also reduces emissions of process gases. In some cases, the treated soil can be used as fill
material (Ref. 5.4). In the presence of high concentrations of mercury, the treated wastes may be passed
through the system more than once or the residence times may be increased to achieve treatment goals for
mercury (Ref. 5.4). APC devices must be used to control emissions of mercury vapors. The vapor
density of mercury is seven times higher than of air; therefore, mercury vapors in an air environment tend
to settle rather than rise. Because of this property of the metal, the gas outlet and the fugitive emission
collection points should be located at the bottom of the thermal desorption or retort chamber. The
presence of large quantities of organic materials in a mercury retort system that contain small amounts of
oxygen may result in pyrolysis, which can create combustible off-gasses that present an explosion hazard
(Ref. 5.3). The presence of larger particles in the contaminated medium may impair heat transfer between
the heating elements or the combustion gas and the medium. Conversely, smaller particles may increase
the particulate content in the off gas (Ref. 5.6).
Factors Affecting Thermal Treatment Performance and Cost
General factors:
• Soil type: Sandy soils are more suitable for thermal treatment than are silty and clay soils.
Finer soils and soils with high humic content require longer processing time, which results in
increased unit costs (Refs. 5.2, 5.7).
• Organic content: The presence of organic materials may present an explosion hazard if the
materials pyrolyze (Ref. 5.3). In one project, the presence of elevated levels of sulfur pyrite
sparked a fire when it was exposed to high temperatures. This fire destroyed the air emission
control equipment (Ref. 5.8).
• Particle size: The presence of large particles in the contaminated medium may impair heat
transfer between the heating elements or the combustion gas and the medium (Ref. 5.6).
• Moisture content: Wastes with moisture content higher than 20 to 25 percent have to undergo
pre-treatment (dewatering or mixing with dry materials) before they are loaded into the
thermal desorption or retort unit. Pre-treatment can increase the overall processing time and
may result in higher treatment costs (Ref. 5.7).
• Residence time: A higher concentration of mercury in waste may require a longer residence
time to meet treatment goals. This longer residence time can increase the treatment costs (Ref.
5.4).
• System throughput: Wastes with moisture content higher than 20 to 25 percent usually must
be treated in smaller volumes. This smaller volume may result in reduced throughput and
higher treatment costs (Ref. 5.8).
Factors specific to thermal treatment of mercury-contaminated media:
• Amount of mercury in waste: A higher concentration of mercury in waste may require the
waste to be passed more than once through the treatment system (Ref. 5.4).
• Operating Temperature and Pressure: The boiling point of elemental mercury at 1
atmosphere (1,014 kilopascals) is 350 °C. Exposure of mercury-contaminated wastes to
higher temperature (380 °C) at a reduced pressure, ranging from 100 to 200 kilopascals,
increases the volatility of mercury in the waste (Ref. 5.5). A thermal desorber is exposed to
temperatures ranging from 320 to 700 °C (600 to 1,300 °F) to facilitate mercury removal
(Refs. 5.2, 5.3, 5.5, and 5.10).
5-3 August 2007
-------
Section 5 - Thermal Treatment
Type, Number, and Scale of Identified Projects
Thermal treatment of soil and waste that contain mercury has been applied at full scale and tested at the
pilot scale. Data sources used for this report include information about three full-scale and five pilot-
scale applications of thermal technologies to treat mercury. One of the full-scale applications identified is
at a Superfund site (Project 1, Table 5.1).
Summary of Performance and Cost Data
Table 5.1 summarizes information for three full-scale and five pilot-scale projects. The performance of
thermal treatment is usually measured by leach testing a sample of the treated mass. Among the full-scale
projects, the media treated include soil, sludge, and sediment. The amount of media treated during
Projects 1 and 2 (Table 5.1) were 80,000 and 26,000 tons. The other full-scale project (Project 8, Table
5.1) involves an ongoing, continuous treatment process that treats approximately 7 tons of mercury-
contaminated media per day. Site types include an industrial landfill, a pesticide manufacturing site, and
a mercury-recycling site. The initial concentration of mercury at one site was 0.16 mg/kg (Project 2,
Table 5.1); the initial concentrations for the other two projects were not available. Final leachable
mercury concentrations as measured by the TCLP were less than 0.2 mg/L for Projects 1 and 8 (Table
5.1). The concentration of total mercury was less than 1 mg/kg for Proj ect 1, and it was less than 0.12
mg/kg for Project 2.
Among the pilot-scale projects, the media treated include soil and sludge. The quantity of media treated
ranged from 0.5 tons to 62 tons. Performance data show that, for one of the five projects, the final
mercury concentration in the leachate was reduced to below 0.025 mg/L (Project 3, Table 5.1). Final
concentrations of mercury in the leachate were not available for the other four projects (Projects 4 through
7, Table 5.1). The final concentrations of total mercury for these four projects ranged from 0.01 mg/kg to
18.7 mg/kg.
Table 5.2 provides total actual costs for one pilot-scale and two full-scale projects. Thermal desorption
was part of a treatment train, applied after soil washing, for the pilot-scale project. The costs provided for
this project (Table 5.2) include soil washing as well. Data sources used for this report did not provide
unit costs for any of these projects.
Case Study: Brookhaven Chemical Holes
BNL conducted pilot-scale studies to demonstrate the treatment of mercury-contaminated soil using
high-temperature thermal desorption (HTTD) under a high vacuum (Ref. 5.10). The contaminated soil
was excavated from the BNL Chemical Holes during a cleanup effort at the research laboratory. The
treatment process involved application of heat at temperatures as high as 700 °C (1,300 °F). The
mercury-contaminated soil was shredded before it was loaded into the thermal desorption unit. The
HI ID unit was then sealed and a vacuum of approximately 25 inches of mercury was applied. Fuel-
fired heat was supplied indirectly to the treatment system. As the temperature reached 700 °C (1,300
°F), mercury and other volatile substances in the waste vaporized and were drawn out of the desorber
by vacuum. The initial total concentration of mercury in the waste was as high as 5,510 mg/kg, while
the leachable concentrations of mercury ranged from 0.2 to 1.4 mg/L (TCLP). The final concentration
of total mercury was reduced to less than 10 mg/kg, while the leachable concentrations of mercury
ranged from non-detect levels to 0.0084 mg/L (TCLP) (Project 3, Table 5.1). The concentration of
mercury in the air emissions ranged from 1 to 29 micrograms per cubic meter (ug/m3), which is below
the maximum achievable control technology (MACT) standard of 40 (.ig/m3. A total of approximately
3,000 pounds of contaminated soil and sludge were treated during this project (Ref. 5.10).
5-4 August 2007
-------
Section 5 - Thermal Treatment
Table 5.1
Thermal Treatment Performance Data for Mercury
Project
Number
Industry and
Site Type
Waste or
Media
Scale
Site Name,
Location, and
Project
Completion
Date
Initial Mercury-
Concentration
(mg/kg) or
Leachability
(mg/L) [Test
method]**
Final Mercury
Concentration
(mg/kg) or
Leachability
(mg/L) [Test
method]**
Process
Source
Thermal Desorption
1
2
3
4
5
Industrial
Landfill
Pesticide
Manufacturing/
Use
Research
Laboratory-
Disposal Pit
Pesticide and
Herbicide
Manufacturing/
Use
Chlor-alkali
Manufacturing
Soil and
Sediment
80,000 tons
Soil; 26,000
tons
Waste
Sludge;
1.3 tons
Soil; 62 tons
Soil and
Sludge;
1 ton
Full
Full
Pilot
Pilot
Pilot
Lipari Landfill,
OU3,
Pitman, New-
Jersey;
September 1995
Site B (site name
and location
confidential),
Western U.S.;
August 1995
BNL Chemical
Holes, Upton,
New York 2001
CFM Site,
Bavaria,
Germany;
November 1994
Chlor-alkali
Plant,
Houston, Texas
N/A
0.16mg//kg
8 to 5,5 10
mg/kg
0.2 to 1.4 mg/L
[TCLP]
780 to 1,080
mg/kga
53.5 to 97.3
ug/L [TCLP]
1 to 350 mg/kg
<0.2 mg/L
[TCLP]
<1 mg/kg
<0. 12 mg/kg
<10 mg/kg
non-detect to
0.008 mg/L
[TCLP]
<1 to 2 9 ug/nr
[air emissions]
16.2 to 18.7
mg/kg
0.6 to 14.6 ug/L
[TCLP]
0.01 to 0.7
mg/kg
Low temperature thermal desorption
(LTTD) system with the following
components: a direct-heated rotary kiln;
feed metering unit; baghouse; thermal
oxidizer; wet quench; acid gas absorber;
and control unit that housed the controls,
data logger, and analyzers.
LTTD system with the following main
units: feed system, rotary dryer, baghouse,
thermal oxidizer, wet quench, and acid gas
scrubber. The rotary dryer (desorber) unit
consisted of a direct-heated rotary kiln, feed
belt, a propane-fired burner, and discharge
screw conveyor.
High vacuum rotary thermal desorption
system with the following components:
materials handling unit; indirectly heated,
high temperature, high vacuum rotary
retort; condensers; high efficiency
participate air (HEPA) filters; and carbon
adsorber.
Treatment train consisting of soil washing
and high vacuum thermal desorption.
Components of the thermal desorption
system include rotary drum dryer; indirectly
fired vacuum distillation chamber;
condensers; and off-gas treatment unit.
High vacuum rotary thermal desorption
system with the following components:
materials handling unit; indirectly heated,
high temperature high vacuum rotary retort.
condensers; HEPA filters; and carbon
adsorber.
5.8
5.9
5.10
5.5
5.11
5-5
August 2007
-------
Section 5 - Thermal Treatment
Table 5.1
Thermal Treatment Performance Data for Mercury (continued)
Project
Number
6
7
Industry and
Site Type
Metals
Recycling
Abovegroimd
Storage Tank
Waste or
Media
Soil and
Sludge;
0.5 tons
Soil; 0.5 tons
Scale
Pilot
Pilot
Site Name,
Location, and
Project
Completion
Date
Metals
Recycling Site,
Claypool,
Arizona
Pipeline
Metering Site,
Louisiana
Initial Mercury
Concentration
(mg/kg) or
Leachability
(mg/L) [Test
method]**
500 to 1,260
mg/kg
5 to 5,770
mg/kg
Final Mercury
Concentration
(mg/kg) or
Leachability
(mg/L) [Test
method]**
0.07 to 5.9
mg/kg
0.02 to 1 mg/kg
Process
High vacuum rotary thermal desorption
system with the following components:
materials handling unit; indirectly heated,
high temperature high vacuum rotary retort.
condensers; HEPA filters; and carbon
adsorber.
High vacuum rotary thermal desorption
system with the following components:
materials handling unit; indirectly heated,
high temperature high vacuum rotary retort;
condensers; HEPA filters; and carbon
adsorber.
Source
5.10
5.10
Retorting
8
Metals
Recycling
Soil, Sludge,
and Debris;
7 tons per day
Full
Mercury
Recycling site,
Pennsylvania
N/A
<20 ug/L
[TCLP]
Advanced programmable high vacuum
mercury retorts. Information about system
components is not available in the sources
used for this report.
5.11,
5.12
Note:
" Initial concentration is before soil washing. Information on the concentration of mercury between soil washing and thermal desorption steps in the treatment train was not
available in the documents used for this report.
BNL: Brookhaven National Laboratory
CFM: Chemische Fabrik Marktredwitz
Ib: Pound
mg/kg: Milligrams per kilogram
mg/L: Milligrams per liter
ug/L: Micrograms per liter
N/A: Not Available
OU: Operable unit
TCLP: Toxicity Characteristic Leaching Procedure
TWA: Total Waste Analysis
**A11 data for mercury concentration are TWA values unless specified otherwise
5-6
August 2007
-------
Section 5 - Thermal Treatment
Table 5.2
Available Mercury Treatment Cost Data for Thermal Treatment
Site
Lipari
Landfill,
OU3,
Pitman.
New Jersey
Site B (site
name and
location
confidential)
, Western
U.S.
Harbauer
Facility,
State of
Bavaria,
Germany
Seal
e
Full
Full
Pilot
Amount
Treated
80,000
tons
26,000
tons
62 tons
Capital
Cost
$430,000
$429,561
$30,000,
000
Total O&M
Cost
$5,019,292
$2,830,956
$3,200,000
Total Cost
$5.449,292
$3,601,377
N/A
Unit Costs
$68 per
ton
$125 per
ton
N/A
Additional Information
Other project costs,
including response to
baghouse fire, caustic
consumption, and
equipment modifications,
totaled $632,737.
However, these costs were
not included in calculating
the unit cost. Cost year is
1995.
Other costs include
$285,000 for compliance
testing and analysis, and
$55,860 for excavation and
disposal. However, these
costs were not included in
calculating the unit cost.
Cost year is 1995.
** Thermal desorption was
part of a treatment train,
applied after soil washing.
Costs are based on a 1 994
conversion rate of 1.5 DM
to$l.
Source
5.8
5.9
5.5
Project
Number in
Table 5.1
1
2
4
** Cost data for this project are estimated and include costs for soil washing and thermal desorption.
DM: Deutsch Mark
N/A: Not available
O&M: Operation and maintenance
OU: Operable Unit
5-7
August 2007
-------
Section 5 - Thermal Treatment
References
5.1 P. Kulakow. A Guide to Thermal Desorption (adapted from EPA Technology Fact Sheet: EPA
542-F-96-005). Date unknown. http://www.eiigg.ksii.edu/HSRC/Tosc/desorp.pdf.
5.2 Federal Remediation Technologies Reference Guide and Screening Manual, Version 4.0. 2001.
Federal Remediation Technologies Roundtable. September 5.
http://www.frtr.go\7matrix2/top_page.html.
5.3 Washburn, Charles, and Eldan Hill. 2003. Mercury Retorts for the Processing of Precious
Metals and Hazardous Wastes. The Member Journal of The Minerals, Metals & Materials
Society (JOM). April.
5.4 The Interstate Technology & Regulatory Council. 1998. Technical Guidelines for On-Site
Thermal Desorption of Solid Media and Low Level Mixed Waste Contaminated with Mercury
and/or Hazardous Chlorinated Organics. September 2. Available at:
http ://ww:w. itrcweb .org/u ser/td-3 .pdf.
5.5 U.S. Environmental Protection Agency (EPA) Office of Research and Development. 1996.
Harbauer Soil Washing/Vacuum Distillation System. EPA - BMBF Bilateral SITE
Demonstration. Innovative Technology Evaluation Report. August.
http://www.epa.gov/ORD/SlTE/Bilateral/liarbauer.pdf.
5.6 U.S EPA Office of Solid Waste and Emergency Response. 2002. Arsenic Treatment
Technologies for Soil, Waste, and Water. EPA-542-R-02-004. September.
http://www.cluin.org/download/remed/542r02004/arsenic __report.pdf.
5.7 Blanchard, John, and Robert Stamnes. 1997. U.S. Environmental Protection Agency Office of
Solid Waste and Emergency Response. Engineering Forum Issue Paper: Thermal Desorption
Implementation Issues. EPA 540/F-95/031. January.
http://www.epa.gov/swertiol/tsp/dow~nload/tdissiie.pdf.
5.8 U.S. EPA. 2002. Cost and Performance Summary Report: Thermal Desorption at the Lipari
Landfill, Operable Unit 3, Pitman, New Jersey. March.
http://costpeiformance.org/profile.cfni?TD=137&CaseID=137.
5.9 U.S. EPA. 2003. Cost and Performance Summary Report: Thermal Desorption at Site B,
Western United States. February, http://costperfonnance.org/profile.cfm?ID=333&CaseID=333.
5.10 U.S. Department of Energy Office of Science and Technology. 2002. Innovative Technology
Summary Report: The SepraDyne™-Raduce System for Recovery of Mercury from Mixed
Waste. September. DOE/EM-0633. http://apps.em.doe.gov/OST/pubs/itsrs/itsr2380.pdf.
5.11 EPA Remediation and Characterization Innovative Technologies (EPA REACH IT) online
database. SepraDyne Corporation - Vacuum Desorption. http ://ww w.epareacliit.org. Accessed
January 2006.
5.12 Bethlehem Apparatus Company, Inc. http: //w w w. bethlehemapparatus. com. Accessed January
2006.
5.13 EPA EPA REACH IT - online database. Bethlehem Apparatus Company, Inc. - Advanced
Programmable High Vacuum Retort, http://www.epareachit.org. Accessed January 2006.
5-8 August 2007
-------
Section 6 - Vitrification
6.0
VITRIFICATION
Summary
Vitrification is used to treat mercury-contaminated soil and sediments in situ and ex situ. This
technology has been implemented at full scale and pilot scale. Vitrification is a high-temperature
treatment designed to immobilize contaminants by incorporating them in the vitrified end product,
which is chemically durable and leach resistant. This technology typically requires a large amount of
energy to achieve vitrification temperatures. Off-gases may require further treatment.
Technology Description and Principles
Vitrification is a high-temperature treatment technology designed to immobilize contaminants by
incorporating them into the vitrified end product, which is chemically durable and leach resistant (Ref.
6.1). Figure 6.1 shows atypical vitrification system (Ref. 6.7). Vitrification uses electrical current to
heat (melt) and vitrify the treatment material in place. Electric current is passed through soil by an array
of electrodes inserted vertically into the surface of the contaminated zone. As soil is non-conductive, a
starter pattern of electrically conductive glass frit mat contains graphite is placed on the soil in the paths
of the electrodes (Ref. 6.1). When power is fed to the electrodes, the graphite and glass frit conduct
electricity through the soil, heating the surrounding area and directly melting the adjacent soil. Once
molten, the soil becomes conductive. The melt grows outward and downward as power is gradually-
increased to a full constant operating level. The temperature of the contaminated soil could reach
between 1,600 and 2,000 °C. A single melt can treat a region up to 1,000 tons. Vitrification is used to
treat wastes to a depth of 20 feet (Ref. 6.1). Large contaminated areas are treated in multiple blocks that
fuse together to form one large treated zone. Specific processes used to treat environmental media and
industrial wastes containing mercury are described in Table 6.1.
Figure 6.1
Model of a Vitrification System
Electrodes
Off-Gas
to Treatment
Off-Gas
Collection
Hood
Source: Ref. 6.7
6-1
August 2007
-------
Section 6 - Vitrification
Ex situ vitrification involves heating contaminated material in a melter or furnace. The heat is typically
generated by combustion of fossil fuel (coal, natural gas, and oil) or input electrical energy by direct joule
heat. arcs, plasma torches, or microwaves (7.1). The melter or furnace is lined with refractory material,
which resists chemical and physical abrasion, has a high melting point, and provides a high degree of
insulating value to the process. Vitrification systems that have been used to treat mercury-contaminated
soil and sediment include the Glass Furnace Technology (GFT) (Ref 6.4) and the Plasma Hearth Process
(Ref 6.5).
The patented GFT process developed by Minergy Corporation uses a dryer to reduce the moisture content
in the waste feed to below 10 percent (Ref. 6.4). The dried feed is mixed with a flux material to control
melting temperatures and improve the physical properties of the glass aggregate product. It is then fed to
the furnace. Oxygen and natural gas are combusted in the furnace to raise its internal temperature to
about 1,600 °C. PCBs and organic contaminants are destroyed or volatilized at this temperature. The
feed melts and metals are encapsulated within a glass matrix. The matrix flows out of the furnace as
molten glass. The molten material is then quickly cooled in a water-quench system to form a glass
aggregate product. The off-gas from the process enters the air quality control system, which consists of a
wet scrubber, a fabric filter, and a carbon filter. The wet scrubber eliminates sulfur dioxide and hydrogen
chloride, the fabric filter removes particulate matter (dust), and the carbon filter captures mercury.
The Plasma Hearth Process uses an electric arc to melt non-combustible, inorganic material and volatilize
and oxidize organic materials (Ref. 6.5). Inorganic material is collected in the fixed hearth, where it
separates by gravity into layers of slag and molten metal. The organic fraction is partially combusted and
pyrolyzed, and then is ducted to a secondary combustion chamber where it is subjected to a high
temperature in the presence of excess air and a methane flame to complete combustion. After it exits the
secondary chamber, the off-gas is quenched to approximately 204°C in an evaporative cooler. The off-
gas then enters a baghouse, followed by a carbon filter and a High Efficiency Particulate Air (HEPA)
filter bank. The baghouse removes large participates, the carbon filter removes the volatilized mercury,
and the HEPA filter bank removes fine particulates. Figure 6.2 shows atypical ex situ vitrification
system.
Figure 6.2
Typical Ex Situ Vitrification System
Wa
Dr
ste
Dried Wa
Waste =,nH
i
Off -gas
Melter
ste Furnace
i M, Material f"""
Flux Oxygen
Material
Molten
Glass
Wet
Scrubber
J^ Fabric , ^ Carbon
1 Filter Filter
Water
Quench
System
wwmiipmmm MateTJal Lwiwmwmwmw
Natural
Gas
JL Glass
| Aggregate
wft
•^ Exhaust
Source: Adapted from Ref. 6.4
6-2
August 2007
-------
Section 6 - Vitrification
Technology Description: Vitrification is a high-temperature treatment designed to immobilize
contaminants by incorporating them in the vitrified end product, which is chemically durable and
leach resistant. The primary residual generated by this technology is typically glass cullet or
aggregate. Secondary residuals generated are air emissions, scrubber liquor, carbon filters, and used
hood panels. This process may also cause contaminants to volatilize or undergo thermal destruction,
thereby reducing their concentration in the soil or waste.
Media Treated:
• Soil
• Sediments
Energy Sources Used for Vitrification:
• Fossil fuels
• Direct joule heat
Energy Delivery Mechanisms Used for Vitrification:
• Arcs
• Plasma torches
• Microwaves
• Electrodes (in situ)
In Situ Application Depth:
• Maximum demonstrated depth is 20 feet
• Very shallow depths or depths greater than 20 feet may require innovative techniques
Applicability, Advantages, and Limitations
The presence of chlorides, fluorides, sulfides, and sulfates may interfere with the process, resulting in
higher mobility of mercury in the vitrified product. Feeding additional slag-forming materials such as
sand to the process may compensate for the presence of chlorides, fluorides, sulfides, and sulfates (Ref.
6.1). Chlorides in excess of 0.5 weight percent will typically fume off and enter the off-gas. If chlorides
are excessively concentrated, salts of alkali, alkaline earth, and heavy metals may accumulate in the solid
residues collected by off-gas treatment. Separation of the chloride salts from the residue may be
necessary, therefore, if the residue is returned to the process for treatment. Dioxins and fiirans may also
form when excess chlorides are present and enter the off-gas treatment system (Ref. 6.1). The presence of
these constituents may also lead to the formation of volatile metal species or corrosive acids in the off-gas
(Ref. 6.1).
During vitrification, combustion of the organic content of the waste liberates heat, reducing the external
energy requirements. Therefore, this process may be advantageous in the treatment of wastes that contain
a combination of mercury and organic contaminants or for the treatment of organo-mercury compounds.
However, high concentrations of organic materials and moisture may result in high volumes of off-gas as
the organic constituents volatilize and combust and water rums to steam, which may impair the function
of the emissions control systems.
Vitrification can also increase the density of treated material, thereby reducing its volume. In some cases,
the vitrified product can be reused or sold (Ref. 6.4).
6-3 August 2007
-------
Section 6 - Vitrification
Excavation of soil is not required for in situ vitrification. In situ vitrification has been used to a depth of
20 feet (Ref. 6.3). Contamination at very shallow depths or depths greater than 20 feet may require
innovative application techniques.
Ex situ treatment processes typically are not transportable, and so contaminated materials must be
delivered to the facility for treatment. If sediments are treated using this process, the materials must be
dewatered before treatment. In cold climates, indoor storage of feed materials may be required to avoid
freezing (Ref. 6.4).
High concentrations of mercury in soil or sediment may limit performance of this process. Metals
retained in the melt must be dissolved to minimize formation of crystalline phases that can decrease the
leach resistance of the vitrified product. Mercury may be difficult to treat because of its high volatility
and low solubility in glass (less man 0.1 percent), but may be effectively treated at low concentrations
(Ref. 6.1).
Factors Affecting Vitrification Performance and Cost
• Lack of glass-forming materials: If insufficient glass-forming materials (SiO2 at more than
30 percent by weight) and combined alkali (Na + K at more than 1.4 percent by weight) are
present in the waste, the vitrified product may be less durable. The addition of frit or flux
additives may compensate for the lack of glass-forming and alkali materials (Ref. 6.2).
• Particle size: Some vitrification units require that the particle size of the feed be controlled.
Size reduction may be required to achieve acceptable throughputs and a homogeneous melt
for wastes containing refractory compounds that melt above the unit's nominal processing
temperature, such as quartz and alumina. High-temperature processes, such as arcing and
plasma processes, may not require size reduction of the feed (Ref. 6.1).
• Presence of ground-water: Groundwater control may be required for in situ vitrification if
contamination is below the water table and hydraulic conductivity of the soil is more than 10"4
centimeters per second (cm/sec) (Ref. 6.2).
• Moisture content: Wastes that contain more than 25 percent moisture content may require
excessive fuel or dewatering before treatment (Ref. 6.1).
• Subsurface air pockets: For in situ vitrification, subsurface air pockets, such as those that
may be associated with buried drums, can cause molten material to bubble and splatter,
resulting in a safety hazard (Ref. 6.2).
• Presence of halogenated organic compounds: The combustion of halogenated organic
compounds may cause incomplete combustion and deposition of chlorides, which can result in
higher mobility of mercury in the vitrified product (Ref. 6.1).
• Metals content: For in situ vitrification, high metals content may result in pooling of molten
metals at the bottom of the melt, causing electrical short-circuiting. This technology is not
recommended for sites with metals content greater than 25 percent by weight (Ref. 6.6).
• Organic content: For in situ vitrification, high organic content may cause excessive heating
of the melt, damaging the treatment equipment. High concentrations of organic compounds
may also create large volumes of off-gas as the organic constituents volatilize and combust
and may overwhelm air emissions control systems. This technology is not recommended for
sites that contain organic content of more than 7 to 10 percent by weight (Ref. 6.6).
• Characteristics of treated waste: Depending on the qualities of the vitrified waste, the
treated soil and waste may be able to be reused or sold (Ref. 6.4).
6-4 August 2007
-------
Section 6 - Vitrification
Type, Number, and Scale of Identified Projects
Vitrification of soil and sediments that contain mercury has been applied at the full scale and
demonstrated in the field. Data sources used for this report include information about one full-scale and
two pilot-scale applications of this technology to treat mercury-contaminated soil and sediments.
Summary of Performance and Cost Data
Table 6.1 summarizes information for one full-scale and two pilot-scale projects. The performance of
vitrification treatment is usually measured by leach testing a sample of the final vitrified material.
In the full-scale project, the media treated included soil and sediments. The amount of media treated was
3.000 cubic yards (5.400 tons). The site was an agricultural chemical manufacturing plant. Confirmation
coring samples indicated mat vitrified materials met cleanup requirements for mercury in soil. Stack gas
emissions met off-gas emissions requirements. The cleanup contractor's cost at the site was $1,763,000,
of which $800,000 was for vitrification and corresponded to $267 per cubic yard of soil treated (Ref 6.3).
The pilot-scale project using the GFT was used to treat river sediments contaminated with PCBs and
metals. The river sediments were obtained from the Lower Fox River in Green Bay, Wisconsin.
Approximately 27.000 pounds of dried sediments were treated using this process. Analytical results for
the sediments before treatment indicated that the mercury concentration was about 0.001 micrograms per
kilogram (fig/kg) (Ref. 6.4). Confirmation sampling results for the glass aggregate indicated that the
mercury concentration was less than 0.0025 ug/kg. The concentration of leachable mercury in the glass
aggregate was less than 0.000002 ug/L, as measured by the Synthetic Precipitate Leaching Procedure
(SPLP). The cost to treat dewatered sediments was estimated to be $38.74 per ton (see Table 6.1, Project
2).
Performance data for the pilot-scale project using the Plasma Hearth Process were unavailable. Table 6.2
provides cost data for the projects described above.
Case Study: Parsons Chemical Superfund Site
The Parsons Chemical Superfund Site in Grand Ledge, Michigan, was an agricultural chemical
mixing, manufacturing, and packaging facility. Full-scale vitrification on site using the Geosafe
process was implemented to treat 3,000 cubic yards of soil and sediments contaminated with
pesticides and heavy metals, including mercury and dioxin. The technology is capable of treating soil
in situ; however, in this case, the contaminant depth was too shallow for the electrodes to be used
effectively. Therefore, the technology was adapted by excavating the contaminated soil, placing it in a
cell, and treating it in a trench on site. The contaminated area consisted of nine melt cells: eight
separate melts were conducted at the site. The duration of each melt was 10 to 19.5 days, and melts
required about a year to cool sufficiently to sample. Typical initial concentrations of mercury ranged
from 2,220 to 4,760 ug/kg. Results for confirmation coring samples of the final vitrified soil indicated
the concentration of mercury to be less than 40 ug/kg. The concentration of leachable mercury in the
vitrified soils ranged from less than 0.2 to 0.23 ug/L, as measured by the TCLP. Concentrations of
mercury in the off-gas emissions ranged from 12.9 to 17.7 ug/nr3. or 0.0000989 to 0.000125 pounds
per hour (see Table 6.1, Project 1).
6-5 August 2007
-------
Section 6 - Vitrification
Table 6.1
Vitrification Treatment Performance Data for Mercury
Project
Number
1
2
3
Industry and
Site Type
Agricultural
chemicals
manufacturing.
formulating,
and packaging
River
sediments
N/A
Waste or
Media
Soil and
Sediments
(in situ)',
3,000 cy
Sediments
(ex situ).
27,000 Ib
Mixed
waste
Scale
Full
Pilot
Pilot
Site Name, Location,
and Project
Completion Date
Parsons
Chemical/ETM
Enterprises Superfund
Site, Grand Ledge,
Michigan
1994
Lower Fox River,
Wisconsin Hazen
Research Center,
Golden, CO and the
Minergy GlassPack
Test Center,
Winneconne,
Wisconsin
2001
Science and
Technology
Applications Research
(STAR) Center, Idaho
Falls, Idaho
1998
Initial Mercury
Concentration
(ug/kg) or
Leachability
(ug/L) [Test
method]
2,220 to 4,760
(ig/kg (Total)
Not analyzed
(TCLP)
< .001 (iig/kg
(Total)
Not analyzed
(SPLP)
N/A
Vitrified Product
and Final Mercury
Concentration
(ug/kg) or
Leachability (ug /L)
[Test method]
< 40 ng/kg (Total)
0.2 to 0.23 (.ig/L
(TCLP)
glass aggregate;
< 0.0025 (.ig/kg
(Total)
< 0.000002 ng/L
(SPLP)
N/A
Vitrification
Process
Vitrification
(on site)
Ex situ Glass
Furnace
Technology
Plasma
Hearth
Process
Source
6.3
6.4
6.5
Note:
cy: cubic yards
N/A: Not Available
SPLP: Synthetic Precipitate Leaching Procedure
TCLP: Toxicity Characteristic Leaching Procedure
Hg/kg: Micrograins per kilogram
Ug/L: Micrograins per liter
6-6
August 2007
-------
Section 6 - Vitrification
Table 6.2
Available Mercury Treatment Cost Data for Vitrification*
Site/
Cost year
Parsons
Chemical/ETM
Enterprises
Superfund Site,
Grand Ledge.
Michigan
Lower Fox
River, Green
Bay, Wisconsin
Scale
Full
Pilot
Amount
Treated
3,000 cy
27,000
Ibs
Capital Cost
N/A
$12 per ton
Annual
O&M Cost
N/A
N/A
Unit
Cost
$267 per
cubic
yard
N/A
Total Cost
$800,000
N/A
Cost Explanation
N/A
Costs assume that
the facility would
treat 600 tons of
sediments per day
and that the facility
would operate for
350 days per year
for 15 years.
Costs are estimated
for full-scale
remediation.
Actual cost for
pilot test not
available.
Source
6.3
6.4
Project
Number in
Table 6.1
1
2
6-7
August 2007
-------
Section 6 - Vitrification
Table 6.2
Available Mercury Treatment Cost Data** (continued)
Site/
Cost year
Science and
Technology
Application
research
(STAR)
Center, Idaho
Falls, Idaho
Scale
Pilot
Amount
Treated
N/A
Capital Cost
$50,000,000
to
$86,200,000
Annual
O&M Cost
$48,000,000
to
$62,000,000
Unit
Cost
N/A
Total Cost
$124,000,000
to
$184,000,000
Cost Explanation
Cost assumes that
the facility would
process 23.235 cy
of waste over 5
years of operation.
Costs are estimated
for full-scale
remediation.
Actual cost for
pilot test not
available.
Source
6.5
Project
Number in
Table 6.1
3
** All costs and amounts treated are estimates.
cy: cubic yard
Ib: Pound
kg: Kilogram
N/A: Not available
O&M: Operation and maintenance
6-8
August 2007
-------
Section 6 - Vitrification
References
6.1 U.S. Environmental Protection Agency (EPA) Office of Research and Development. 1997.
Engineering Bulletin, Technology Alternatives for the Remediation of Soils Contaminated with
Arsenic. Cadmium. Chromium. Mercury, and Lead. Cincinnati, OH. EPA-540-S-97-500.
March. http://www.epa.gov/claritoii/clhtml/pubtitleOSWER.html.
6.2 U.S. EPA. 1995. Contaminants and Remedial Options at Selected Metal-Contaminated Sites.
Office of Research and Development. EPA-540-R-95-512. July.
http://www.epa.gov/clariton/clhtiiil/pubtitleOSWER.litnil.
6.3 Federal Remediation Technologies Roundtable (FRTR). Cost and Performance Report. Parsons
Chemical/ETM Enterprises Superfund Site Grand Ledge. Michigan.
http://costperfomiance.org/pdf/parsons.pdf. Date unknown.
6.4 U.S. EPA Office of Research and Development. 2004. Minergy Corporation Glass Furnace
Technology Evaluation Report. EPA/540/R-03/500. March.
http://costpeiformance.org/pdf/20Q407()2_353.pdf.
6.5 U.S. Department of Energy. 1998. Innovative Technology Summary Report. Plasma Hearth
Process at the Science and Technology Research (STAR) Center, Idaho Falls, Idaho. November.
http://costperformance.org/pdf/itsr26.pdf.
6.6 U.S.EPA. 1995. Superfund Innovative Technology Evaluation (SITE) Technology Capsule,
Geosafe Corporation hi Situ Vitrification Technology. Office of Research and Development.
EPA/540/R-94/520. March. http://www.eDa.gov/ORD/SlTE/reports/540 r94 520.pdf.
6.7 U.S EPA Office of Solid Waste and Emergency Response. 2002. Arsenic Treatment
Technologies for Soil, Waste, and Water. EPA-542-R-02-004. September.
http://www.cliun.org/dowiiload/renied/542r02004/arsenic report.pdf.
6-9 August 2007
-------
This page is intentionally left blank.
6-10
-------
PART IIB
TREATMENT TECHNOLOGIES APPLICABLE TO WATER
-------
This page is intentionally left blank.
-------
Section 7 - Precipitation/Coprecipitation
7.0
PRECIPITATION/COPRECIPITATION
Summary
Precipitation/coprecipitation is a full-scale technology used to treat mercury-contaminated
groundwater and wastewater. Based on the information collected to prepare this report, this
technology typically can reduce mercury concentrations to less than 2 |Jg/L. However, some of the
processes used multiple precipitation steps and additional treatment with other technologies such as
activated carbon to reach this level.
Figure 7.1
Model of a Precipitation/Coprecipitation System
Groundwater
Reagent
Polymer
pH Adjustment and
Reagent Addition
oundwater
Solids to -4-
Disposal
Effluent
Clarification
Filtrate
Sludge
Thickening
Source: Ref. 7.2
Technology Description and Principles
Technologies were considered precipitation/coprecipitation if they involved the following steps:
• Mixing of treatment chemicals into water;
• Formation of a solid matrix through precipitation, coprecipitation, or a combination of these
processes; and
• Separation of the solid matrix from the water.
Precipitation/coprecipitation usually involves pH adjustment and addition of a chemical precipitant or
coagulant to transform soluble metals and inorganic contaminants into insoluble metals and inorganic
salts (Ref. 7.2). Mercuiy removal usually includes changing the pH of the water to be treated because
removal is maximized at the pH where the precipitated species is least soluble. The optimal pH range for
precipitation/coprecipitation depends on the waste treated and the specific treatment process. The
precipitated/coprecipitated solid is typically removed by clarification or filtration.
7-1
August 2007
-------
Section 7 - Precipitation/Coprecipitation
The most common precipitation method used to remove inorganic mercury from wastewater is sulfide
precipitation (Ref. 7.1). In this process, the adjusted pH range is between 7 and 9, and a sulfide
precipitant (such as sodium sulfide) is added to the wastewater stream. The sulfide precipitant converts
dissolved mercury to the relatively insoluble mercury sulfide form.
One coprecipitation process for mercury uses lignin derivatives to form a lignin-mercury colloid.
Precipitated solids can then be removed by gravity settling in a clarifier. This process can be enhanced by
the addition of a chemical coagulant or settling aid such as ferric chloride, and then flocculation and
settling.
Technology Description: Precipitation uses chemicals to transform dissolved contaminants into an
insoluble solid (Ref. 7.2). In coprecipitation, the target contaminant may be in a dissolved, colloidal,
or suspended form. Dissolved contaminants do not precipitate, but are adsorbed onto another species
that is precipitated. Colloidal or suspended contaminants become enmeshed with other precipitated
species or are removed through processes such as coagulation and flocculation. Processes to remove
mercury from water can include a combination of precipitation and coprecipitation. The precipitated/
coprecipitated solid is then removed from the liquid phase by clarification or filtration. Mercury
precipitation/coprecipitation can use combinations of the chemicals and methods listed below.
Media Treated:
• Groundwater
• Wastewater
Chemicals and Methods Used for Mercury Precipitation/Coprecipitation:
• Ferric salts (for example, ferric chloride), ferric sulfate, or ferric hydroxide
• Alum
• pH adjustment
• Lime softening, limestone, and calcium hydroxide
• Sulfide
• Lignin derivatives
Applicability, Advantages, and Limitations
Precipitation/coprecipitation is an active ex situ treatment technology designed to function with routine
chemical addition and sludge removal. It usually generates a sludge residual, which typically requires
treatment such as dewatering and subsequent disposal. Some sludge from the precipitation/
coprecipitation of mercury can be a hazardous waste and require additional treatment such as
solidification/stabilization for disposal as a solid waste or could require disposal as a hazardous waste.
Excess use of the chemical sulfide precipitants can form soluble mercury sulfide species. Mercury can
resolubilize from sulfide sludges under conditions mat exist in landfills (Ref. 7.4), which could lead to
mercury contamination of leachate and potential groundwater pollution.
The effluent from mercury precipitation may also require further treatment, such as pH adjustment, before
discharge or reuse. Sulfide precipitation of mercury may generate residual sulfide in the effluent.
Treatment to remove residual sulfide may be needed before discharge (Ref. 7.1).
7-2 August 2007
-------
Section 7 - Precipitation/Coprecipitation
Factors Affecting Precipitation/Coprecipitation Performance and Cost
• pH: In general, mercury removal will be maximized at the pH where the precipitated species
is least soluble. The optimal pH range for precipitation/coprecipitation depends on the waste
treated and the specific treatment process. The most effective precipitation of mercury for the
sulfide precipitation process occurs within a pH range of 7 to 9 (Ref. 7.1). The most effective
precipitation of mercury for the hydroxide precipitation process occurs within a pH range of 7
to 11 (Ref. 7.9).
• Presence of other compounds: The presence of other metals or contaminants may affect the
effectiveness of precipitation/coprecipitation.
• Chemical dosage: The cost generally increases with increased chemical addition. Larger
amounts of chemicals added usually results in a larger amount of sludge that requires
additional treatment or disposal. Excess use of sulfide precipitant can form soluble mercury
sulfide species (Ref. 7.1).
• Treatment goal: For some applications, a single precipitation step or using precipitation
alone may not achieve the treatment goals. Multiple precipitation steps or the use of
additional technologies may be needed to meet stringent cleanup goals, effluent guidelines, or
disposal standards.
• Sludge disposal: Sludge produced from the precipitation/coprecipitation process could be
considered a hazardous waste and require additional treatment before disposal as a solid waste
or disposal as hazardous waste (Ref. 7.2).
Type, Number, and Scale of Identified Projects Treating Groundwater and Wastewater Containing
Mercury
Precipitation has been applied at full scale to treat groundwater and wastewater contaminated with
mercury. Data sources used for this report include information about three full-scale applications for
treatment of mercury contaminated groundwater and eight full-scale applications for treating wastewater.
Summary of Performance and Cost Data
Table 7.1 presents the available performance data for full-scale precipitation/coprecipitation treatment of
mercury. Table 7.2 provides cost data for two pump-and-treat systems that treated mercury-contaminated
groundwater with precipitation/coprecipitation. Three full-scale projects treated contaminated
groundwater and eight full-scale projects treated wastewater. The three projects mat treated groundwater
reduced the mercury concentration to less than 2 |lg/L. A total of 41,000 pounds of contaminants were
removed from January 1997 to March 1999 at Marine Corps Base Camp Lejeune in North Carolina. The
treatment system removed volatile organic compounds and metals, including mercury. The pump and
treat (P&T) system included oxidation of ferrous iron to ferric iron, pH adjustment, precipitation, air
stripping, and granular activated carbon adsorption. The cost of the P&T system was approximately $49
per pound of contaminant removed. The cost for the precipitation/coprecipitation process alone is not
identified or discussed in the references used for this report.
At the Squamish Mercury Remediation Project in British Columbia, Canada (see Table 7.1, Project 2), 25
million gallons of mercury-contaminated groundwater were treated using a coprecipitation process
followed by addition of ferric chloride. The treatment reduced the mercury concentration from 15 |lg/L to
1 |Jg/L. The unit cost for the coprecipitating chemical was $2.00 per gallon of water treated. Additional
information on the cost is not identified or discussed in the references used for this report.
7-3 August 2007
-------
Section 7 - Precipitation/Coprecipitation
Precipitation, carbon adsorption, and pH adjustment were used at the Olin Corporation Mclntosh Plant
Site in Washington County, Alabama, to treat mercury-contaminated groundwater. This treatment
reduced the average influent concentration of mercury from 44 |j,g/L to 0.3 |ig/L (Ref. 7.6). No cost data
were identified or discussed for this project in the references used for this report.
Of the eight projects that treated mercury-contaminated wastewater, five treated wastewater at chlor-
alkali plants. The influent mercury concentration ranged from 362 |ig/L to 0.07 |ig/L, and the effluent
concentration was reduced to less than 2 |ig/L in all projects. However, some of the processes used
multiple precipitation steps and additional treatment with other technologies such as activated carbon to
reach this level. No cost data were identified or discussed for these projects in the references used for this
report (Ref. 7.3), except for projects treating wastewater from the centralized waste treatment industry
(Projects 10 and 11). Chapter 11 of the reference used for these projects (Ref. 7.10) provides detailed
cost data and information on how the costs were derived.
Case Study: Olin Chemical Site
Olin Corporation operated a mercury cell chlor-alkali plant from 1952 to 1982 at the Olin Corporation
Mclntosh Plant Site in Washington County, Alabama. In 1978. Olin began operating a diaphragm cell
caustic soda/chlorine plant, which is still in operation. The plant on site produces chlorine, caustic
soda, sodium hypochlorite. and sodium chloride. Releases of mercury and other organic chemicals
have contaminated the soil and the shallow groundwater beneath the site.
A P&T remedy is being used for groundwater at this site. The treatment system consists of
precipitation, carbon adsorption, and pH adjustment before discharge to the Mobile River. The
cleanup level in groundwater is 2 |ig/L. The treatment system has reduced the concentration of
mercury in the pumped groundwater from 44 |ig/L to 0.3 |ig/L. The P&T system began operating in
1986 and treats an average of 350 to 400 gallons per minute (gpm) (Ref. 7.6).
7-4 August 2007
-------
Section 7 - Precipitation/Coprecipitation
Table 7.1
Precipitation/Coprecipitation Treatment Performance Data for Mercury
Project
Number
1
2
3
4
5
Industry and
Site Type
Marine Corps
Base
Mercury-Cell
Electrolysis
process for the
production of
chlorine and
caustic soda
Chlorinated
organic
pesticides,
chlorine, sodium
hypochlorite, and
caustic soda
manufacturing
plant
Natural gas
production
Commercial
hazardous waste
combustor
Waste or
Media
Groundwater
Groundwater
Groundwater
Wastewater
Wastewater
from quench
chamber run-
Scale
Full
Full
Full
Full
Full
Site Name,
Location, and
Project
Completion
Date
Marine Corps
Base Camp
Lejeune,
North Carolina
Squamish
Remediation
Site,
Squamish.
British
Columbia,
Canada
Olin Corp,
Mclntosh,
Alabama
Confidential
Site, Gulf of
Thailand
N/A
Initial
Mercury
Concentration
(HS/L)
N/A
15
40
9,600
59.2
21.4
Final Mercury
Concentration
(HS/L)
0.025
1
0.3
0.035
21.4
2.6
Precipitating Agent
or Process
Oxidation of ferrous
iron to ferric iron, pH
adjustment,
precipitation, air
stripping, and
granular activated
carbon adsorption.
Coprecipitation
followed by addition
of ferric chloride.
Precipitation, carbon
adsorption, and pH
adjustment.
Thiol-based chelating
agent.
Sodium hydroxide
precipitation
Ferric chloride
Source
7.7
7.5
7.6
7.8
7.3
7-5
August 2007
-------
Section 7 - Precipitation/Coprecipitation
Table 7.1
Precipitation/Coprecipitation Treatment Performance Data for Mercury (continued)
Project
Number
6
7
8
Industry and
Site Type
Commercial
hazardous waste
combustor
Commercial
hazardous waste
combustor
Commercial
hazardous waste
combustor
Waste or
Media
down and
packed tower
Wastewater
from quench
tank, packed
tower, and a
venturi
scrubber
Wastewater
from quench
tank and a wet
scrubber
Wastewater
from
circulation
loop of
incinerator
Scale
Full
Full
Full
Site Name,
Location, and
Project
Completion
Date
N/A
N/A
N/A
Initial
Mercury
Concentration
(HB/L)
2.6
0.7
0.4
3.3
3.04
Final Mercury
Concentration
(HS/L)
ND*
0.4
ND*
ND*
0.2
Precipitating Agent
or Process
Sand filtration
Primary precipitation
system: Sulfide
precipitation followed
by hydroxide
precipitation
Secondary
precipitation system:
Hydroxide
precipitation followed
by sedimentation and
ultrafiltration
Sulfide precipitation
followed by filtration
followed by carbon
adsorption
Primary precipitation
system: Hydroxide
precipitation followed
by sedimentation;
Secondary
precipitation system:
ferric chloride
precipitation followed
by sedimentation
Source
7.3
7.3
7.3
7-6
August 2007
-------
Section 7 - Precipitation/Coprecipitation
Table 7.1
Precipitation/Coprecipitation Treatment Performance Data for Mercury (continued)
Project
Number
9
10
11
Industry and
Site Type
Commercial
hazardous waste
combustor
Centralized
Waste Treatment
Facility* *
Centralized
Waste Treatment
Facility**
Waste or
Media
Wastewater
from APC
system
Wastewater
from metal
bearing liquids
and solids
Wastewater
from metal
bearing liquids
Scale
Full
Full
Full
Site Name,
Location, and
Project
Completion
Date
N/A
Confidential
Confidential
Initial
Mercury
Concentration
(ng/L)
219
115
362
102.57
Final Mercury
Concentration
(HS/L)
0.48
0.20
1.67
0.51
Precipitating Agent
or Process
Hydroxide
precipitation and
ferric chloride
precipitation followed
by sedimentation and
sand filtration
Selective metals
precipitation, liquid
solid separation,
secondary
precipitation, liquid-
solid separation,
tertiary precipitation.
and clarification
Primary precipitation,
liquid-solid
separation, secondary
precipitation and sand
filtration
Source
7.3
7.10
7.10
Note:
* : Detection limits not provided
* *: Influent and effluent values provided represent two sets of data collected for the facility
APC: Air pollution control
N/A: Not available
ND: Non-detect
ug/L: Micrograms per liter
7-7
August 2007
-------
Section 7 - Precipitation/Coprecipitation
Table 7.2
Available Mercury Treatment Cost Data for Precipitation/Coprecipitation
Site/Cost year
Marine Corps
Base Camp
Lejeune, North
Carolina
Squamish
Remediation Site,
Squamish, British
Columbia,
Canada
Scale
Full
Full
Amount
Treated
41,000
pounds of
contaminants
removed
from 1997 to
1999
25 million
gallons
Capital Cost
$ 4.66 Million
N/A
Annual
O&M
Cost
$ 1 Million
N/A
Unit Cost
$49 per
pound of
contaminant
removed
$2 per gallon
oftreated
groundwater
Total
Cost
N/A
N/A
Cost
Explanation
Costs in 2000
dollars. The
costs are for the
entire treatment
system.
The unit cost is
for the reagent
only. Costs in
2004 dollars.
Source
7.7
7.5
Project
Number in
Table 7.1
1
2
N/A: Not available
O&M: Operation and maintenance
7-8
August 2007
-------
Section 7 - Precipitation/Coprecipitation
References
7.1 U.S. Environmental Protection Agency (EPA). 1997. Office of Research and Development.
Capsule Report, Aqueous Mercury Treatment. EPA-625-R-97-004. July.
http://wwvv.epa.gov/ORD/NRMRL/pubs/625r97004/625r97004.pdf.
7.2 U.S. EPA. 2002. Office of Solid Waste and Emergency Response. Arsenic Treatment
Technologies for Soil Water, and Water. EPA-542-R-02-004. September.
http://clu-in.org/dow'iiload/remed/542r02004/arsenic report.pdf,
7.3 U.S. EPA Office of Water. 2000. Development Document for Final Effluent Limitation
Guidelines and Standards for Commercial Hazardous Waste Combustors. EPA-821-R-99-020.
January, http://epa.gov/waterscience/chwc/fmal/technical.html.
7.4 U.S. EPA. 1992. Biological and Physio-Chemical Remediation of Mercury-Contaminated
Hazardous Waste. Conly L Hansen and Daid K. Stevens. EPA-600-R-92-105. PP 121-125.
7.5 NORAM Engineering and Construction Ltd. 2004. Report on Remedial Goals and Solutions at
the Squamish Mercury Remediation Project. November. Vancouver, BC, Canada.
7.6 Faxed data from Mike Arnet regarding the Olin Corporation Mclntosh Plant Site. 2004. U.S.
Environmental Protection Agency, to Danielle Wohler, Tetra Tech EM Inc. December 16.
7.7 U.S. Navy and Marine Corps. 2000. Site Specific Report: Marine Corps Base Camp Lejeune,
Operable Unit 1 and 2. SSR-2539-ENV. January.
7.8 E-mail from Carl Hensman, Frontier Geosciences Inc., to Younus Burhan, Tetra Tech EM Inc.,
regarding the use of thiol-based chelating agent at a confidential site in Thailand. December 30,
2004.
7.9 West General Inc. 2005. Mercury solubility-mercury hydroxide curve. January.
http://www.westgeneral.com/outofthebox/compounds/hgsol.html.
7.10 U.S. EPA. 2000. Office of Water. Development Document for Effluent Limitations Guidelines
and Standards for the Centralized Waste Treatment Industry - Final.EPA-821-R-00-020. August.
http://www.epa.gov/\\aterscience/guide/cwt/final/develop/index.html.
7-9 August 2007
-------
This page is intentionally left blank.
7-10
-------
Section 8 - Adsorption Treatment
8.0 ADSORPTION TREATMENT
Summary
Adsorption has been used to remove inorganic mercury from groundwater and wastewater. Based on
the information collected for this report, this technology can reduce concentrations of inorganic
mercury to less than 2 ug/L. Its effectiveness is sensitive to a variety of contaminants and
characteristics of the untreated water. It can be a primary treatment method but is often used as a
polishing step (removal of mercury left in the waste stream after a primary treatment process) for other
water treatment processes.
Technology Description and Principles
This section describes mercury removal processes that pass water through a fixed bed of media. Some of
the processes described in this section rely on a combination of adsorption, precipitation/coprecipitation,
pH adjustment, or filtration. Adsorption is used in these processes either as the primary removal
mechanism or as a polishing step (Refs. 8.1 and 8.2) to remove residual contamination that was not
removed by a primary treatment step. All of the treatment applications identified include adsorption as
part of a treatment train rather than as a stand-alone treatment unit. A treatment train is a series of
treatments combined in sequence to provide the necessary or required treatment result.
Adsorption of molecules can be represented as a physical reaction:
A + B <-» A-B
where
= The adsorbate (contaminant)
"B" = The adsorbent
"A»B" = The adsorbed compounds
Adsorbates are held on the surface by various types of chemical forces such as hydrogen bonds, dipole-
dipole interactions, and van der Waals forces. If the reaction is reversible, as it is for many compounds
adsorbed to activated carbon, molecules continue to accumulate on the surface until the rate of the
forward action (adsorption) equals the rate of the reverse action (desorption). When this condition exists,
the adsorbent is saturated, and no further accumulation will occur (Ref. 8.7). One of the most important
characteristics of an adsorbent is the quantity of adsorbate it can accumulate. Important characteristics
that affect adsorbate capacity by an adsorbent are surface area, pore size distribution, and surface
chemistry (Ref. 8.7). Water is passed through a bed or adsorbent in adsorption treatment for mercury.
Mercury or mercury compounds are adsorbed as the contaminated water passes through the adsorbent,
removing them from the water.
The most common adsorbent used for mercury is granular activated carbon (Refs. 8.2, 8.5, and 8.6). One
vendor, Selective Adsorption Associates, Inc., also markets sulfur-impregnated activated carbon for
mercury adsorption (Ref. 8.8). Another adsorption process, the Lancy Sorption Filter System, uses a
patented process to remove mercury from water. This technology is typically used as a polishing step
after other treatment processes have been applied. A soluble sulfide is added to the water in a reaction
tank in the first stage of the Lancy filtration process, converting most of the heavy metals to sulfides.
From the sulfide reaction tank, the solution is passed through the Lancy sorption filter media.
Precipitated metal sulfides and other suspended solids are filtered out, and the remaining soluble metals
8-1 August 2007
-------
Section 8 - Adsorption Treatment
and sulfides are absorbed by the media. The system generates either a slum' or a solid cake of spent
media (Ref. 8.15).
The effectiveness of adsorption is sensitive to a variety of water characteristics. Pretreatment is used in
many adsorption systems to alter the characteristics of the untreated water and ensure effective
adsorption. For example, adsorption may be preceded by filtration to remove solids that may plug the
adsorption bed. Ultraviolet (UV) light treatment may be used to control bacterial growth, which can also
plug the adsorption bed and reduce its adsorption capacity. Equalization tanks are often used to ensure a
constant flow rate through the bed. This constant flow helps ensure that the residence time is sufficient to
remove the contaminants. In addition, high flow rates can cause channelization and erosion of the
adsorption bed. which can reduce effectiveness. Adjustment of pH to the range where adsorption is most
effective is also a common pretreatment step. After adsorption treatment, the pH may require further
adjustment to acceptable criteria of the receiving waters or Publicly Owned Treatment Works (POTW).
Once the absorbent bed is saturated, the bed may be regenerated or disposed. The data sources used for
this report did not contain information on the management of spent adsorbent used for mercury treatment.
Technology Description: In adsorption, solutes (contaminants) concentrate at the surface of a
sorbent, thereby reducing their concentration in the bulk liquid phase. The adsorbent is usually packed
into a column. Contaminants are adsorbed as contaminated water is passed through the column. The
column must be regenerated or disposed and replaced with new media when adsorption sites become
filled.
Media Treated:
• Groundwater
• Drinking water
• Wastewater
Types of Sorbent used to Treat Mercury:
• Granular activated carbon
• Sulfur-impregnated activated carbon
• Lancv filtration
Applicability, Advantages and Limitations
The effectiveness of adsorption is sensitive to a variety of contaminants and characteristics in the
untreated water. Suspended solids, organic compounds, and biological growth can cause fouling and
plugging. Pretreatment with flocculation, settling, filtration, or oil-water separation may be used to
reduce suspended solids and organic compounds. UV light treatment has been used to reduce biological
growth before the process is used to adsorb mercury. Spent adsorption media must be regenerated or
disposed. Competition for adsorption sites could reduce the effectiveness of adsorption because other
constituents may be preferentially adsorbed, resulting in a need for more frequent bed regeneration or
replacement.
Regeneration of granular activated carbon usually involves heating to desorb contaminants, which could
release volatile mercury compounds. Use of APC equipment may therefore be necessary to remove
mercury from the off-gas produced (Ref. 8.11). Further treatment may be necessary before spent
adsorption media from mercury treatment systems can be disposed.
8-2 August 2007
-------
Section 8 - Adsorption Treatment
Factors Affecting Adsorption Performance and Cost
Contaminant concentration: Competition for adsorption sites can reduce the effectiveness of
adsorption if competing contaminants are adsorbed instead of or in addition to mercury, resulting in a
need for more frequent bed regeneration or replacement (Ref. 8.12). In addition, the capacity of the
adsorption media decreases with increasing contaminant concentration (Refs. 8.13, 8.14). High
contaminant concentrations can exhaust the adsorption media quickly, requiring frequent regeneration
or replacement.
Fouling: The presence of suspended solids, organic compounds, and solids can cause fouling of
adsorption media (Ref. 8.7).
Flow Rate: Increasing the flow rate through the adsorption media can decrease adsorption of
contaminants and erode the adsorbent bed (Refs. 8.6 and 8.10).
Wastewater pH: The optimal pH to maximize adsorption of mercury by granular activated carbon is
acidic (pH 4-5). Therefore. pH adjustment may be needed both before and after adsorption treatment
(Refs. 8.3, 8.4, 8.7 and 8.11).
Spent Media - Spent media that can no longer be regenerated would require treatment or disposal
(Ref. 8.9).
Type, Number, and Scale of Identified Projects Treating Water Containing Mercury
Adsorption technologies to treat mercury-contaminated water are commercially available. Data sources
used for this report included information about four full- and two pilot-scale applications of adsorption to
treat mercury. Two of the projects (Projects 1 and 2) were implemented at Superfund sites.
Summary of Performance and Cost Data
Table 1 summarizes the performance information for the four full- and two pilot-scale applications. The
effectiveness of adsorption treatment can be evaluated by comparing influent and effluent contaminant
concentrations. Performance data were available for three of the four full-scale processes. Influent
concentrations ranged from 3.3 to 60 (ig/L in these treatments (Projects 2, 3, and 4). The effluent
mercury concentration was below 2 (ig/L for all three of these projects. Of the two pilot-scale
applications, one (Project 5) reported a mercury influent concentration of 2,500 (ig/L and a removal rate
of 98 percent. Data on the effluent concentrations were not available for this project. Project 6 reported a
mercury influent concentration of 60 (.ig/L and an effluent concentration below the detection limit. The
detection limit was not provided, however.
Table 2 provides the available cost data for mercury adsorption treatment. In one pilot study, the cost of
removing thimerosal, a mercury salicylate salt that is used as a bacteriostat/fungistat in many test kits,
using granulated activated carbon was $0.107 per gallon of water treated (Ref. 8.5). The capital cost was
$0.012 per gallon treated and the O&M cost was $0.095 per gallon treated. Another full-scale treatment
system developed for treating thimerosal (Ref. 8.7) reports capital costs of $60,000 and monthly operating
costs of $7,665 for a plant designed to treat 1,800 gallons of thimerosal-contaminated wastewater per day.
August 2007
-------
Section 8 - Adsorption Treatment
Reported costs for systems to treat ground-water include $49 per pound of contaminant removed (Ref. 8.2)
and $520 per pound of contaminant ($19 per 1,000 gallons of groundwater extracted) removed (Ref. 8.1).
However, the costs include removal of other contaminants in addition to mercury. The information
sources used for this report did not contain cost data on the granular activated carbon or Lancy filter
treatment step alone.
Case Study: Reducing Mercury Discharge at a Testing Laboratory
A testing laboratory in Massachusetts generates wastewater containing thimerosal concentrations
averaging 60 |ig/L before treatment. In 1992, the laboratory conducted pilot testing of carbon
adsorption and constructed a full-scale system. The full-scale system includes a 15-micron bag filter,
UV light, an equalization tank with pH adjustment to the 4 to 5 range, granular activated carbon
filters, a mixing tank with pH adjustment to 5.5 to 9.5, and a neutralization tank. The micron bag filter
was necessary to remove solids that tended to plug the filters. UV light was introduced to control
bacterial growth in the carbon adsorption system. Flow equalization was provided to establish
constant detention time and avoid channelization of the beds caused by increasing flows experienced
when equalization was not included. The pH was adjusted to the range where the adsorptive capacity
of the carbon filter is greatest (4 to 5). After the carbon filter, the pHis readjusted to the 5.5 to 10.5
range before the wastewater flows into the neutralization tank. Treated water is finally neutralized to
die range accepted by the receiving POTW. This acceptable range was not reported, however. This
system reduced the concentration of thimerosal in the wastewater from 60 (ig/L to below7 the goal of 1
Costs for the full-scale treatment system (1,800 gallons per da}') include capital costs of $60,000 and
monthly operating costs of $7,665 .
8-4 August 2007
-------
Section 8 - Adsorption Treatment
Table 8.1
Adsorption Treatment Performance Data for Mercury
Project
Number
1
2
3
4
Industry or
Site Type
Waste disposal
and recycling
Chlorinated
organic
pesticides,
chlorine, sodium
hypochlorite,
and caustic soda
manufacturing
plant
Hazardous
waste
combustion
Testing
laboratory
Waste or
Media,
Quantity
Groundwater,
225 gprn
Groundwater
Wastewater
stream from flue
gas quench and
ash/slag quench
Wastewater
discharge from
testing
laboratory,
1,800 gallons
per day
Scale
Full
Full
Full
Full
Site Name,
Location, and
Project Completion
Date
King of Prussia
Technical
Corporation
Superfund Site,
Winslow Township,
New Jersey,
December 1997
(Interim Results)
Olin Corp, Mclntosh,
Alabama
N/A
N/A
Initial Mercury
Concentration
(HB/L)
N/A
44 (average)
3.3 (first stage
Lancy filter) NDC
(second-stage
carbon
adsorption)
60 (source of
mercury-
thimerosold)
Final Mercury
Concentration
(Hg/L)
N/A
0.3 (average)
NDC (first Stage
Lancy filter) 0.4
(second-stage
carbon
adsorption)
1
Adsorption Process
or Media"
Electrochemical
precipitation
followed by air
stripping and
granular activated
carbonb
Precipitation, carbon
adsorption, and pH
adjustment
Lancy filter followed
by carbon bed
Granulated activated
carbon bed
Source
8.1
8.3
8.4
8.7
8-5
August 2007
-------
Section 8 - Adsorption Treatment
Table 8.1
Adsorption Treatment Performance Data for Mercury (continued)
Project
Number
5
6
Industry or
Site Type
Pharmaceutical
industry,
hospitals, and
clinical
laboratories
Testing
laboratory
Waste or
Media,
Quantity
Pharmaceutical
wastewater,
2,000 gallons
Wastewater
discharge from
testing
laboratory,
1,800 gallons
per day
Scale
Pilot
Pilot
Site Name,
Location, and
Project Completion
Date
Civil and
Environmental
Engineering
Department,
Villanova University,
Villanova,
Pennsylvania
N/A
Initial Mercury
Concentration
(ng/L)
2,500 (organic
mercury d)
1,300 (Hg(IT))
60 (source of
mercury-
thimerosold)
Final Mercury
Concentration
(HS/L)
99.8 percent
mercury
removal'
NDC
Adsorption Process
or Media"
Granulated activated
carbon bed
Granulated activated
carbon bed
Source
8.6
8.7
a Some processes employ a combination of adsorption, precipitation/coprecipitation, pH adjustment and readjustment, and filtration to remove
mercury from water.
b According to the text, heavy metals were removed by an electrochemical system and granular activated carbon was used to remove organic
constituents. The GAC unit will serve as a polishing step for mercury.
0 Detection limit not provided
d Thimerosal: (C9H9HgO2SNa)
e Final mercury concentration not provided
N/A: Not available
gpm: Gallons per minute
iig/L: Micrograms per liter
8-6
August 2007
-------
Section 8 - Adsorption Treatment
Table 8.2
Available Mercury Treatment Cost Data for Adsorption
Site/Cost year
King of Prussia
Technical
Corporation
Superfund Site,
Winslow
Township, New
Jersey,
December 1997
(Interim Results)
Civil and
Environmental
Engineering
Department,
Villanova
University,
Villanova,
Pennsylvania
Confidential
testing
laboratory in
Massachusetts
Scale
Full
Pilot
Pilot
and
Full
Amount
Treated
225 gpm
2,000
gallons
(pilot
study)
1,800
gallons
per day
Capital
Cost
$ 2.03
Million
$7,700
$60,000
Annual
O&M Cost
$449,000 (total
for
performance
period is
$785,000)
$6,188
$91,980
Unit
Cost
Not
available
$0.107
capital
and
annual
O&M
Not
available
Total Cost
$2.816
Million
$69,580
Not
available
Cost Explanation
Costs in 1997 dollars. The
costs are for the entire
treatment system
(electrochemical
precipitation followed by air
stripping and granulated
activated carbon). The
treatment system treated
multiple contaminants.
Total cost is for the
performance period April
1995 through December
1997.
Costs in 2001 dollars. Costs
are projections based on the
results of the pilot study.
Costs were based on a
design life of 10 years.
Design volume is 64, 980
gallons per year.
Costs in 1992 dollars.
Treatment is ongoing.
thereby affecting the capital
cost per gallon. Costs for
the pilot study and full-scale
operation were combined in
the reference.
Source
8.1
8.6
8.7
Project
Number in
Table 8.1
1
5
4 and 6
gpm
O&M
Gallons per minute
Operation and maintenance
8-7
August 2007
-------
Section 8-Adsorption Treatment
References
8.1 U.S. Environmental Protection Agency (EPA). 1998. Pump and Treat of Contaminated
Groundwater at the King of Prussia Technical Corporation Superfund Site, Winslow Township,
New Jersey. Federal Remediation Technologies Roundtable.
http://costperformance .org/profile.cfm?TD= 126&CaseID= 126.
8.2 Arnet, Mike. 2004. Faxed data regarding pump and treat at Olin Corporation. From Mike Arnet,
U.S. Environmental Protection Agency, to Danielle Wohler, Tetra Tech EM Inc. December.
8.3 U.S. EPA. 2000. Effluent Guidelines, Commercial Hazardous Waste Combustor Subcategory of
the Waste Combustors Point Source Category, Final Technical Development Document. January.
http://www.epa.gov/waterscience/chwc/final/technical.html.
8.4 Business Communications Company, hie. DMM95 The Microporous Materials Industry Review.
http://www.buscom.com/membrane/DMM95.html. Accessed January 18, 2005.
8.5 Cyr, Patric J., P.S. Suri Reminder, and Edward D. Helmig. 2002. A Pilot-Scale Evaluation of
Removal of Mercury from Pharmaceutical Wastewater Using Granular Activated Carbon. Water
Research.
8.6 Commonwealth of Massachusetts. 1995. Toxic Use Reduction Case Study: Reducing Mercury
Discharge at a Testing Laboratory. Executive Office of Environmental Affairs, Office of
Technical Assistance. December.
8.7 American Water Works Association. 1990. Water Quality and Treatment, A Handbook of
Community Water Supplies. Frederick W. Pontius, Technical Editor. McGraw-Hill Inc.
8.8 Selective Adsorption Associates, Inc. Product sheet for PERSORB® LW mercury adsorbent.
http://www.selectiveadsorption.com/mersorlw.htm. Accessed January 2006.
8.9 Corbitt, Robert A. 1998. Standard Handbook of Environmental Engineering, Second Edition.
McGraw-Hill Books.
8.10 Metcalf &Eddy. 1991. Wastewater Engineering: Treatment, Disposal, and Reuse. Third
Edition. McGraw-Hill Publishing Company.
8.11 U.S. EPA. 1990. Final Best Demonstrated Available Technology (BOAT) Background
Document for K031, K084, K101, K102, Characteristic Arsenic Wastes (D004), Characteristic
Selenium Wastes (DO 10), and P and U Wastes Containing Arsenic and Selenium Listing
Constituents. Office of Solid Waste. May.
8.12 Federal Remediation Technologies Roundtable. 2001. Reference Guide and Screening Manual.
Version 3.0. March 30. http://www.frtr.gov/matrix2/top_page.html.
8.13 U.S. EPA. 2000. Regulations on the Disposal of Arsenic Residuals from Drinking Water
Treatment Plants. Office of Research and Development. EPA/600/R-00/0025. May.
http://www, epa.gov/ORD/WebPubs/residuals/index.htm.
August 2007
-------
Section 8 - Adsorption Treatment
.14 Twidwell, L.G., and others. 1999. Technologies and Potential technologies for Removing
Arsenic from Process and Mine Wastewater. Presented at "REWAS"99." San Sebastian, Spain.
September.
.15 U.S. EPA. 2000. Effluent Guidelines, Centralized Waste Treatment Development Document:
Final Rule. December.
8-9 August 2007
-------
This page is intentionally left blank.
8-10
-------
Section 9 - Membrane Filtration
9.0 MEMBRANE FILTRATION
Summary
Membrane filtration can remove a wide range of contaminants from water. This technology has been
used in a limited number of full-scale applications to treat wastewater contaminated with mercury.
Before membrane filtration, a pretreatment step may be used to cause mercury to form precipitates or
coprecipitates that can be more effectively removed by this technology. Membrane filtration can
reduce concentrations of mercury to less than 2 |lg/L.
Technology Description and Principles
Membrane filtration passes water through a semi-permeable, microporous membrane to concentrate
contaminants into a smaller volume of water. This technology separates the influent into two effluent
streams:
1. The permeate, or effluent stream, is the fluid that passes through the membrane. It usually
contains reduced levels of contaminants. This stream may or may not contain contaminants at
concentrations below the desired levels. If a reduction in concentration is still required, this
stream may be sent to additional treatment units. In the case study described later, the penneate
is neutralized and sent to an equalization and disposal unit.
2. The concentrate, or reject stream, contains water and contaminants that have not passed through
the membrane. The reject may be recycled back through the membrane filtration system to
further concentrate the contaminant and reduce the volume of reject. If the concentrate is not
recycled, further treatment may include processing in a filter press, as in the case study described
later (Ref. 9.5), or dewatering in solar evaporation ponds (Ref. 9.2).
Technology Description: Membrane filtration separates contaminants from water by passing it
through a semi-permeable barrier or membrane. The membrane allows some of the constituents to
pass through while blocking others (Ref. 9.4).
Media Treated:
• Drinking water
• Groundwater
• Surface water
• Industrial wastewater
Types of Membrane Filtration Processes:
• Microfiltration
• Ultrafiltration
• Nanofiltration
• Reverse osmosis
9-1 August 2007
-------
Section 9 - Membrane Filtration
Figure 9.1
Model of a Membrane Filtration System
Contaminated
Water
Membrane filtration processes vary based on the pore size of the membrane. The pore size is selected
based on the molecular weight or size of the target contaminant and the pressure required to move
wastewater through the filter (Ref 9.3). The four types of membrane filtration processes, from largest to
smallest filter pore size, are (Ref. 9.1):
1. Microfiltration
2. Ultrafiltration (UF)
3. Nanofiltration
4. Reverse osmosis
UF has been used in a treatment train with precipitation/coprecipitation to treat wastewater that contains
mercury. The sources used for this report did not contain information on the use of microfiltration,
nanofiltration, or reverse osmosis to remove mercury. Therefore, these technologies are not discussed in
this technology summary.
UF units are used to remove oils, suspended particles, and biological solids (Ref. 9.1). UF requires a
pressure of 5 to 100 pounds per square inch (psi) to move the fluid through an ultrafiltration filter (Ref.
9.6). UF filters can filter out contaminants with a molecular weight greater than 1000 g/mole (Ref 9.3).
Applicability, Advantages and Limitations
UF is primarily used to remove high-molecular weight contaminants and solids. This technology is
usually preceded by precipitation/coprecipitation to cause the mercury species to form or adsorb onto a
suspended solid because dissolved mercury species are typically too small to be effectively removed by
UF. The energy required to operate membrane filtration units is related to the pressure requirements (Ref.
9.3). This type of treatment may be run in either batch or continuous mode. This technology's
effectiveness is sensitive to a variety of contaminants and characteristics in the untreated water.
Suspended solids, organic compounds, colloids, and other contaminants can cause membrane fouling.
9-2
August 2007
-------
Section 9 - Membrane Filtration
Type, Number and Scale of Identified Projects Treating Water Containing Mercury
The references used for this report included information about one full-scale application of UF to treat
mercury-contaminated wastewater.
Case Study: Episode #4671, Hazardous Waste Combustor
Ultrafiltration was included as part of a treatment train used to treat a variety of contaminants in
wastewater generated by the APC equipment of a hazardous waste combustor. The wastewater
treatment system included a primary and secondary treatment loop. The secondary treatment loop
contained a stage for precipitation with sodium hydroxide followed by sedimentation and
ultrafiltration. Analysis of samples collected at the influent and effluent of this treatment loop showed
that the mercury concentration was reduced from 0.4 |ig/L to below the detection limit of 0.2 |ig/L
(Ref. 9.5).
Summary of Performance Data
UF was used as part of a treatment train consisting of a primary and secondary loop in one application of
membrane filtration to mercury-contaminated water. The secondary' loop achieved a 54.5 percent
reduction in the concentration of mercury. More detailed information about this site is presented in the
box labeled "Case Study: Episode #4671". Cost data were not provided for this application.
9-3 August 2007
-------
Section 9 - Membrane Filtration
References
9.1 America Water Works Association. 2005. Factsheet: Membrane Filtration. American Water
Works Association. http://w\\'w.aw\va.org/Advocacv/pressroom/MembraneFiltration.cfiti.
9.2 American Water Works Association. 1990. Water Quality and Treatment, a Handbook of
Community Water Supplies, 4th Ed. McGraw-Hill, Inc.
9.3 General Electric Office of Water & Process Technologies. 1997-2003. Treating Industrial Water
with Membrane Technology. http://\vww.gewater.com/libran/tp/707 Treating Industrial.jsp.
9.4 U.S. EPA Office of Solid Waste. 1996. BOAT Background Document for Spent Potliners from
Primary Aluminum Reduction - K088. IPA 530-R-96-015. February.
http: //www .epa.gov/ncepi/Catalog/EPA5 3 OR96015 .html.
9.5 U.S. EPA Office of Water. 2000. Development Document for Final Effluent Limitations
Guidelines and Standards for Commercial Hazardous Waste Combustors. EPA 821-R-99-020.
January. http://epa.gov/waterscience/chwc/final/chwcfd_a.pdf.
9.6 U.S. EPA Office of Water. 2000. Technologies and Cost for Removal of Arsenic from Drinking
Water. EPA 815-R-00-0028. December.
http://www.epa.gov/safewater/ars/treatments and costs.pdf.
9-4 August 2007
-------
Section 10- Biological Treatment
10.0 BIOLOGICAL TREATMENT
Summary
Biological treatment has been tested at pilot scale for ex situ treatment of mercury-contaminated
wastewater in a limited number of projects. Biological treatment may convert mercury to species that
are retained in the biomass or are more easily removed from water by another technology, such as
adsorption or precipitation.
Technology Description and Principles
Although biological treatment has usually been applied to degrade organic contaminants, it is also
applicable to treat mercury-contaminated wastewater. Figure 11.1 shows a simplified model of a
biological treatment system to remove mercury from water (Ref. 11.5).
Figure 10.1
Model of a Biological Treatment System (Ref. 10.5)
Influent
Packed
media and
microbes
Effluent
Aerobic Biotreatment Process for Mercury
One process that has been applied at a pilot-scale project uses aerobic biological treatment to convert
soluble ionic mercury (Hgi+) into elemental or metallic mercury (Hg"). This reaction is catalyzed by
enzymes such as mercuric reductase, which is generated naturally in the cytoplasm of certain bacterial
species (Refs. 10.1. 10.2. and 10.3). For example, certain strains ofPsendomonas are capable of reducing
mercuric ion to elemental mercury. The less soluble elemental mercury collects in the microbial mass as
small droplets, which must be subsequently extracted using another technology (Ref. 10.3).
Biological treatment technologies typically require amendments to create optimal growth conditions and
provide nutrients to the microbial population. In this process, the treatment system includes pretreatment
to adjust the pHto the optimal range of 6.5 to 7.5 using sodium hydroxide and phosphoric acid. An
adequate amount of nutrient is required to support microbial growth. Nutrient additives used in this
process include sucrose, sodium chloride (NaCl), and yeast extract (Ref. 10.3).
10-1
August 2007
-------
Section 10- Biological Treatment
The contaminated water is passed through a packed bed bioreactor. The reactor bed is usually made of
materials suitable for microbial growth, such as silica (SiO2) and alumina (A12O3). Before treatment
begins, the microbial inoculum is distributed throughout the bioreactor bed through a series of steps that
involves pumping combinations of nutrients, microbes, and wastewater through the bed (Ref. 10.3). After
biological treatment, the bioreactor effluent is usually treated by activated carbon to remove residual
mercury (Ref. 10.3). The biological treatment process results in solid residuals consisting of dead
bacteria and elemental mercury. The sources used for this report do not contain information about the
disposal methods for the solid residuals generated by mis process.
Aqueous Biocyanide Process
Another process that has been applied at a pilot scale uses a series of aerobic and anaerobic biological
treatment steps to remove mercury from wastewater (Refs. 10.1, 10.2). This process is designed primarily
to treat cyanide in mining wastewaters, but also removes heavy metals, including mercury. This process
uses a fixed-bed reactor with a combination of microbes, including proprietary microbial cultures and
microorganisms isolated from the contaminated wastewater. The equipment used in this process includes
an aerobic unit and an anaerobic unit, each consisting of a microbial culture tank and a bioreactor vessel.
The treatment process involves alternating cycles of treatment and inoculation with periodic flushing
between the two cycles. Contaminated wastewater is passed through the aerobic fixed-bed reactor system
made of porous ceramic medium. The effluent from the aerobic reactor is then passed through the
anaerobic reactor. This biological process converts soluble fonns of mercury into less soluble forms,
such as sulfides and other mineral phases. However, the references used for this report did not contain
more specific information about the mechanism of mercury removal in this process (Refs. 10.1, 10.2).
Technology Description: Biological treatment of mercury-contaminated wastes is catalyzed by
microbial enzymes. In one process, the soluble, ionic form of mercury is aerobically converted to
insoluble elemental mercury by an enzyme called mercury reductase. The less soluble elemental
mercury must be extracted using another technology (Ref. 10.3). In another process, a combination of
aerobic and anaerobic treatment methods is used to convert soluble forms of mercury into insoluble
mineral phases, such as sulfides (Refs 10.1 and 10.2). The effluent from the biological treatment
system is normally subjected to further treatment by an activated carbon bed or precipitation before
disposal (Refs. 10.1, 10.2, 10.3, 10.4, and 10.5).
Media Treated:
• Wastewater
Microbes Used:
• Mercury-tolerant strains ofPseudomonas spp.
• Proprietary microbial cultures
Amendments Used:
• Sucrose
• Yeast extract
• NaCl
• pH control reagents, such as NaOH and H3PO4
• H9S
10-2 August 2007
-------
Section 10- Biological Treatment
Technology Description (continued):
Technology Types Used:
• Fixed-bed aerobic bioreactor
• Fixed-bed bioreactors, with series of aerobic and anaerobic treatment steps
Media for Bed Packing:
• A12O3 and SiO2
• Porous ceramic medium
Applicability, Advantages, and Limitations
Biological treatment of mercury involves conversion of soluble mercury into a less soluble elemental
form or into insoluble mercuric sulfide (Refs. 10.1, 10.3). Fiigh concentrations of contaminants such as
mercury or chlorine can inhibit microbial activity. Nutrients, pH, and temperature must be maintained at
levels that optimize biological activity and growth (Refs 10.1, 10.2, and 10.3). Pretreatment with pH
amendment agents such as NaOH or H3PO4is essential to maintain an optimal pH range. Nutrient
additives such as sucrose, yeast, and NaCl may be required to support the growth of microbes. The
bioreactor effluent typically requires further treatment by methods such as activated carbon adsorption or
precipitation to ensure residual mercury is removed before disposal (Refs. 10.3 and 10.4).
Factors Affecting Biological Treatment Performance and Cost
• pH: neutral pH of 6.5 to 7.5 is optimal for aerobic biotreatment processes (Ref. 10.3).
• Contaminant concentration: Fiigh concentrations of mercury may be toxic to
microorganisms used in biological treatment (Ref. 10.4).
• Available nutrients: The presence of sufficient amount of nutrients, such as sucrose and
yeast extract, is crucial to the performance of a biological system because nutrients are
essential for the growth of microorganisms (Refs. 10.3 and 10.4).
• Temperature: Lower temperatures (35 to 45 °F) decrease biological reaction rates. Heating
may be required to maintain biological activity (Ref. 10.4). Temperature higher man 85 °F
may be harmful to the microorganisms (Refs. 10.1, 10.3).
• Chloride concentration: The presence of chloride at concentrations greater than 0.5 mg/L
may inhibit microbial growth (Ref. 10.3).
Type, Number, and Scale of Identified Projects Treating Wastes Containing Mercury
Data sources used for this report included information about two pilot-scale applications of biological
treatment to wastewater contaminated with mercury.
Summary of Performance and Cost Data
The information collected to prepare this report contained data on two applications of biological treatment
for mercury. Table 10.1 summarizes the performance information for one pilot-scale application for
electrolysis wastewater from a chlor-alkali manufacturing plant and one pilot-scale application for metal
mining wastewater. Information about the amount of media treated was not available. For Project 1
(Table 10.1), the initial concentration of mercury ranged from 2,000 to 5,000 ug/L. The concentration of
mercury in the effluent ranged from 30.7 to 40.7 ug/L. For Project 2, initial concentration of mercury
10-3 August 2007
-------
Section 10- Biological Treatment
ranged from 151 to 164 (.ig/L. The concentration of mercury in the treated effluent ranged from 3 to 11
(.ig/L. Although significant mercury reductions were achieved, this technology did not reduce the
concentration to less than 2 [ig/L in either of these applications. However, information on the treatment
goals for these applications was not available, and these applications may not have been designed to treat
mercury to less than 2 |lg/L.
The data sources used for this report did not provide information about the cost of these projects.
Case Study: Echo Bay/McCoy Cove Mine Site, Nevada
An innovative bioremediation process was tested at pilot scale to treat mercury in wastewater from the
Echo Bay/McCoy Cove Mine (Table 10.1, Project 2). This project was evaluated by the U.S. EPA's
Superfund Innovative Technology Evaluation (SITE) program (Ref. 10.1, 10.2). The treatment
method used is a proprietary process known as the Aqueous Biocyanide Process. This process
consists of aerobic and anaerobic fixed-bed bioreactor units. This application of the Aqueous
Biocyanide Process used microorganisms isolated from the mine stream in combination with
proprietary microbial cultures. A biofilm of the microbial mass was formed on the reactor bed. which
was made of a porous ceramic medium. This biofilm converted the soluble ionic form of mercury
(Hg2+) into more stable mineral phases, primarily mercuric sulfide (HgS). The treatment process in
this pilot test included aerobic and anaerobic pathways with the following three steps: (1) inoculation
of the culture tanks with microbial cultures; (2) treatment of contaminated media fed into the
inoculated tanks; and (3) flushing to remove process wastes from the culture tanks. Influent
concentrations of mercury ranged from 151 to 177 (.ig/L. Final concentrations of mercury ranged from
3 to 11 (ig/L. The information sources used for this report did not provide data about the total amount
of waste treated during this project. The sources also did not provide information about the treatment
goals for this project and the disposal methods for the flush water and treated effluent.
10-4 August 2007
-------
Section 10 - Biological Treatment
Table 10.1
Biological Treatment Performance Data for Mercury
Project
Number
1
2
Industry and
Site Type
Chlor-alkali
manufacturing
Metal ore
mining
Waste or
Media
Electrolysis
wastewater
Mine
wastewater
Scale
Pilot
Pilot
Site Name,
Location, and
Project
Completion
Date
Chlor-alkali
Manufacturing
site; Completed
Echo
Bay/McCoy
Cove Mine
site, Nevada;
Completed;
August 1997
Initial
Mercury
Concentration
(US/L)
2.000 to 5,600
Mg/L
151 to 164
ug/L
Final Mercury
Concentration
(US/L)
30.7 to 40.7
l-ig/L
3 to 1 1 ng/L
Process
Fixed-bed aerobic bioreactor
packed with granular A12O3 and
SiO2 and inoculated with
mercury-resistant strains of
Pseudomonas species.
Treatment train consisted of pH
amendment by NaOH or
H3PO4, followed by addition of
nutrients including sucrose,
yeast extract, and NaCl,
treatment in an aerobic
bioreactor, and activated carbon
adsorption.
Aerobic and anaerobic fixed-
bed bioreactors packed with
porous ceramic medium and
inoculated with a consortium of
site-specific and proprietary
microbial cultures.
Source
10.3
10.1,
10.2
Note:
(.ig/L: Micrograms per liter
A12O3: Alumina
SiO2: Silica
NaOH: Sodium hydroxide
H3PO4: Phosphoric acid
10-5
August 2007
-------
Section 10- Biological Treatment
References
10.1 U.S. Environmental Protection Agency (EPA). 2000. SITE Technology Capsule Pintail Systems
Inc.'s Aqueous Biocyanide Process. EPA-540/R-00/501a. May.
http://wwvv.p2pavs.Org/ref/0 7/0666 7 .pdf.
10.2 Pintail Systems, Inc. - Aqueous Biocyanide Process, http://www.epareachit.org. Accessed
February 2006.
10.3 Wagner-Dobler, Irene, Harald Von Canstein. Ying Li, Kenneth N. Timmis, and Wolf-Dieter
Deckwer. 2000. Removal of Mercury from Chemical Wastewater by Microorganisms in
Technical Scale. Environmental Science and Technology, 34.
10.4 Federal Remediation Technologies Roundtable. 2001. Reference Guide and Screening Manual,
Version 4.0. September 5. http://www.frtr.gov/matrix2/topj3age.html.
10.5 U.S. EPA Office of Solid Waste and Emergency Response. 2002. Arsenic Treatment
Technologies for Soil, Waste, and Water. EPA-542-R-02-004. September.
http://www.cluin.org/download/remed/542r02004/arsenic report.pdf.
10-6 August 2007
-------
PART III
ONGOING RESEARCH ON MERCURY TREATMENT
-------
This page is intentionally left blank.
-------
Section 11 - Ongoing Research on Mercury Treatment
11.0 ONGOING RESEARCH ON MERCURY TREATMENT
Sections 3.0 to 10.0 of this report provide examples of full- and pilot-scale projects of technologies that
apply to treatment of mercury. This section summarizes selected bench-scale projects that involve the
evaluation of innovative technologies for their ability to treat mercury. The technologies covered include
nanoteclinology, phytoremediation, and air stripping. Research is under way on reactive capping
materials such as bauxite for mercury in sediments (Ref. 11.8). These innovative technologies have the
potential to provide more cost-effective and reliable alternatives for mercury treatment. Based on
information from the limited number of applications of these technologies that have been identified, they
may be used to treat mercury more frequently in the future. However, additional data are needed to
obtain a full understanding of their applicability and effectiveness.
11.1 Nanotechnology
An example of research on use of nanotechnology for mercury in an aqueous waste stream is the Tliiol
Self-Assembled Monolayers on Mesoporous Silica (Thiol-SAMMS). This novel adsorbent was
developed by the staff of Pacific Northwest National Laboratory (PNNL). It consists of a nanoporous
ceramic substrate with a high surface area made functional by a monolayer of thiol groups. The substrate
is typically synthesized through a self-assembly process that uses sol-gel precursors and surfactant
molecules. Subsequent calcination to 500°C removes the surfactant templates and leaves nanoporous
ceramic substrate with a high surface area, which can be used for self-assembled monolayers of
adsorptive functional groups (Ref. 11.1). Thiol functional groups are known to have a high affinity for
various heavy metals, including mercury. The thiol groups are embedded in the ceramic substrate on one
end and bind with mercury on the oilier end (Ref. 11.2).
Thiol-SAMMS is capable of selectively binding with mercury and can achieve a mercury loading
capacity as high as 635 milligrams per gram. In addition, test data indicate that the mercury adsorption of
SAMMS is not significantly affected by the presence of other cations and complexing anions in waste
solutions (Ref. 11.2).
A pilot-scale treatability test was conducted to assess the ability of thiol-SAMMS to remove soluble
mercury from an aqueous melter condensate waste stream. The treatment was conducted on 160 liters of
waste using a module consisting of a batch reactor equipped with a mixer, a filtration unit, a pump, and a
holding drum for treated effluent. The waste was filtered into the batch reactor using a 1.0 micron (um)
filtration unit. The initial treatment was conducted by adding 195 grams of thiol-SAMMS (3.5
nanometers pore size). The SAMMS material was stirred into the waste by turning on the mixer. The
adsorption reaction was allowed to proceed for 12 hours, when the spent sorbent was separated by
pumping the mixture through the filtration unit equipped with a 50-urn filter. An aliquot of treated
effluent was removed using a syringe filter and analyzed for residual mercury concentration (Ref. 11.5).
Two more sequential treatment steps were conducted on the treated effluent to achieve higher levels of
mercury reduction. In the second step, 248 grams of thiol-SAMMS was added to the effluent from the
first treatment. In the third step, approximately 190 grams of thiol-SAMMS material was added to the
effluent from the second treatment (Ref. 11.5).
The results showed that the first treatment removed about 97.4 percent of dissolved mercury from the
waste with a residual concentration of 0.28 mg/L. The second treatment reduced the residual
concentration of mercury to 0.18 mg/L. The final treatment removed 99.4 percent of the dissolved
mercury originally present in the untreated waste, resulting in a residual concentration of 0.06 mg/L.
These data demonstrate that dissolved mercury in the melter waste stream can be effectively scavenged
11-1 August 2007
-------
Section 11 - Ongoing Research on Mercury Treatment
using thiol-SAMMS sorbent (Ref. 11.5). The success of these tests, among others, has led PNNL to
begin commercializing thiol-SAMMS (Ref. 11.2).
11.2 Phytoremediation
Phytoremediation is another technology that is being evaluated for its effectiveness in removing mercury
from sediments and other media. Phytoremediation uses plants to remove, transfer, stabilize, or destroy
contaminants in soil, sediment, and groundwater. Phytoremediation applies to all biological, chemical.
and physical processes that are influenced by plants (including the rhizosphere) and that aid in cleanup of
the contaminated substances. Plants can be used in site remediation, both through mineralization of toxic
organic compounds and through accumulation and concentration of heavy metals and other inorganic
compounds from soil into aboveground shoots. Phytoremediation may be applied in situ or ex situ to
soils, sludges, sediments, other solids, or groundwater (Ref. 11.7).
Plants can be genetically engineered to enhance their ability to detoxify mercury. An example of the
development of a transgenic plant is the modification of the rice plant (Oryza sativd) to remove mercury
from aquatic sediments. The plant is injected with a gene (merA) that produces an enzyme, mercuric
reductase (merA) capable of converting ionic mercury to elemental mercury, which then volatilizes from
the sediments (Ref. 11.3). Various studies have been conducted on rice and other plants to evaluate their
ability to remediate mercury. Examples of two such studies are provided below. Although both examples
show that phytoremediation of mercury is possible, further research and pilot-scale studies will be needed
to assess the effectiveness of the technology at full scale. In addition, full-scale implementation will
require consideration of issues, such as disposal of contaminated plants and the impacts of volatilized
mercury on other ecosystems (Ref. 11.3).
A bench-scale study was conducted to evaluate the ability of transgenic rice to detoxify mercury-polluted
aquatic sediments. A culture of rice embryos was injected with merA, and the plants were then grown
from the embryo culture on different types of media spiked with Hg+2 ions. Mercury vapor assays were
conducted to study the extent of mercury resistance among the plants and their ability to remove mercury
from the substrate. Three 1-week-old germinants of merA and the wild type (not genetically engineered)
were incubated in 2 millilters of assay medium containing 250 micro molar HgCl2. Incubation was
carried out in a specialized reaction tube, and the headspace from the reaction tube was evacuated into a
Jerome 431-X mercury vapor analyzer immediately after each seedling was placed in the medium. The
headspace was then resampled each minute for 10 minutes. The assays showed that the concentration of
elemental mercury was higher in the headspace of the merA germinates than in the wild-type germinates,
indicating enhanced mercury-reducing activities of the genetically engineered plants. Both sets of plants
were able to remove Hg+i ions when grown in a mercury-spiked hydroponics medium. However, the
lower concentration of Hg+2 in the tissues of the merA rice compared with the wild type indicates that
more mercury was volatilized from the merA plants. These results suggest that genetically engineered
rice can remove mercury from its substrate (Ref. 11.3).
In addition to rice plants, cottonwood trees have been evaluated for their ability to remediate mercury.
Eastern cottonwood trees (Populus deltoides) grow rapidly in a variety of conditions, including riverbanks
and floodplains. They have been engineered to express the merA gene, convert methylmercury to ionic
mercury (merB gene), and hyperaccumulate mercury. A field study using these trees was conducted by-
Applied PhytoGenetics, Inc. (APGEN) in 2003 at a Brownfields site in Danbury, Connecticut, that was
formerly used for hat manufacturing. Results of the study, however, were not available when research
was conducted for this report (Ref. 11.6).
11-2 August 2007
-------
Section 11 - Ongoing Research on Mercury Treatment
11.3 Air Stripping
Air stripping is another technology that is being evaluated for its ability to remove mercury from water.
Air stripping generally has not been used to remove inorganic compounds such as mercury. However, a
bench-scale study was conducted at the Savannah River Site in South Carolina to evaluate whether
chemical reduction followed by collection of elemental mercury from the headspace air can remove low
levels of mercury from groundwater (Ref. 11.4). The groundwater was obtained from the feed and
effluent of an existing groundwater treatment system installed to remediate a chlorinated solvents plume.
The technology tested used stannous (Sn+2) chloride to reduce Hg+2 to Hg", which is volatile (vapor
pressure of 0.0027 milligrams per mercury per liter of air), followed by collection of elemental mercury
from the headspace air to remove the elemental mercury from water. Mercury concentrations in the
extracted groundwater ranged between 120 and 150 nanograms per liter (ng/L), and more than 95 percent
of the mercury was Hg+2. Stannous chloride was added to groundwater samples in dosages ranging from
0 to 766 mg/L. After stannous chloride had been added, groundwater samples were sparged with air at an
air-water ratio of approximately 30 to 1. Stannous chloride doses greater than 0.011 mg/L resulted in
more than 94 percent mercury removal, with the residual total mercury reduced to levels below 10 ng/L.
However, low doses of stannous chloride (less than 0.00023 mg/L) showed little removal of mercury.
This study indicates that chemical reduction coupled with air stripping is effective for treating mercury.
The technology does not produce any liquid or solid secondary wastes, and off-gas treatment may not be
required for the expected air concentrations and mass release. If necessary, however, off-gas treatment
could be incorporated using a low-temperature treatment system for gas-phase elemental mercury. As a
result, it is a promising, low-cost option for treating mercury and is expected to cost less than
conventional metal treatment technologies. Before it is implemented in the field, however, the
environmental effects of introducing stannous chloride should be evaluated. Successful development of a
chemical-reduction-based treatment system requires additional data on the key scientific questions (for
example, stoichiometry and robustness) and engineering evaluation of whether reliable long-term
operation can be achieved (Ref. 11.4).
11.4 la Situ Thermal Desorption (ISTD)
ISTD is being evaluated for its ability to treat mercury. ISTD is a soil remediation process that applies
both heat and a vacuum to the subsurface to extract and degrade contaminants.
Laboratory soil column experiments have been conducted that show that ISTD is capable of treating
mercury-contaminated soils. In one such experiment, approximately 15.03 grams of mercury were
injected into a column packed with Ottawa sand. Experiments were performed at 244°C and an airflow
rate of 76 mL/minute. Analysis of the soil after remediation showed that only 11.1 mg of mercury
remained in the soil after the experiment, corresponding to a removal efficiency of 99.9% (Ref. 11.9).
This study indicates that ISTD has the potential to remove mercury from contaminated soil. Further
research activities are ongoing to determine the effectiveness of the technology in the field.
11-3 August 2007
-------
Section 11 - Ongoing Research on Mercury Treatment
References
11.1 Mattigod, S.V.. G. Fryxell, R. Skaggs, and K. Parker. Mercury Removal from Melter Condensate
Waste Using Thiol-Samms™ Sorbent. Date unknown.
11.2 Mattigod, S.V., G. Fryxell, R. Skaggs K. Parker, J. Liu, and X Feng. 2003. Mercury Removal
from Waste Streams Using a Novel Synthetic Nanoporous Sorbent. Industrial Water Conference.
Las Vegas, Nevada. December.
11.3 Heaton, A.C.P., C.L. Rugh, T. Kim, N.J. Wang, and R. B Meagher. 2003. Toward Detoxifying
Mercury-Polluted Aquatic Sediments with Rice Genetically Engineered for Mercury Resistance.
Environmental Toxicology and Chemistry. Vol. 22, No. 12, pp. 2940-2947. March.
11.4 Looney, B.B. and others. 2001. Ultralow Concentration Mercury Treatment Using Chemical
Reduction and Air Stripping. WSRC-MS-2001-00388. April 24.
11.5 Mattigod, S.V., G. Fryxell, R. Skaggs K. Parker, and S. Juracic. Mercury Removal from High-
Efficiency Mist Eliminator Waste Using Thiol-Samms™ Sorbent. Pilot-Scale Treatability Test 2.
Date unknown.
11.6 Applied PhytoGenetics (APGEN). 2003. Letter from David Glass, APGEN, to Walter Kovalick,
EPA Technology Innovation Office (TIO), regarding the field study of Eastern cottonwood trees
to treat mercury contaminated soil at a Superfund site in Danbury, Connecticut.
11.7 Federal Remediation Technologies Reference Guide and Screening Manual, Version 4.0. 2004.
In Situ Biological Treatment - Phytoremediation. Federal Remediation Technologies
Roundtable. June. http://\v\vw.frtr.gov/matrix2/section4/4-3.html.
11.8 Personal communication between Paul Randall, EPA National Risk Management Research
Laboratory (NRMRL). and Marti Otto, EPA OSRTI. April 21, 2006.
11.9 Siebert, J. 2005. An Examination of Using In-Situ Thermal Desorption to Remediate Mercury
Contaminated Soils Through Laboratory Experiments and Numerical Modeling. Masters Thesis.
University of Texas at Austin. May.
11-4 August 2007
-------
Appendix A - Information on Select Mercury Vapor Detection Instruments
Appendix A
Information on Select Mercury Vapor Detection Instruments
Manufacturer
Distributor
Units
Range
Methodology
Accuracy,
percent
Jerome
431
Arizona
Instrument,
LLC, AZ
-
mg/nr
.003 to
0.999
mg/nr1
Change in
resistance
of gold
after
mercury
absorption
±5% at
100 ug/m3
Jerome
J405
Arizona
Instrument,
LLC,AZ
-
i.ig/mj
0.5 to 999
(ig/m3
Change in
resistance of
gold film
after
mercury
absorption
±10% at 1
iig/m3
Jerome 471
Arizona
Instrument,
LLC,AZ
-
iig/m3
0.030 to 250
Hg/m3
UV Atomic
Absorption
Spectrometry
±10% at 0.5
L.ig/m3
Mercury/
Emp-1
Nippon
Instruments
Corporation.
Japan
Brandt
Instruments
Inc., LA
mg/mj
Zero to 0.999
1 to 5.00
Ultraviolet
absorption
"cold vapor
measuring
technique"
5
Vm-3000
Mercury
Instalments
GmbH.
Germany
ST2 Service
Technologies,
Inc. CO
L.ig/m3
Zero to 100
Zero to 1000
Zero to 2000
Ultraviolet
absorption
"cold vapor
measuring
technique"
-
Lumex RA-
915+
Lumex,
Russia (final
assembly-
Twinsburg,
Ohio)
OhioLumex
Co., OH
ng/m3
Zero to
500,000
Zeeman
atomic
absorption
spectrometry-
high-
frequency
modulation
of light
polarization
+/-10%
Lumex RA-
915+/RP-
91C
Lumex,
Russia (final
assembly-
Twinsburg,
Ohio)
OhioLumex
Co., OH
ppb
lOppbto
lOOOppm
Zeeman
atomic
absorption
spectrometry-
high-
frequency
modulation
of light
polarization
Direct, (no
sample
preparation)
solid/ water
analysis in 2
minutes
+/-10%
Lumex RA-
915+/Pyro
915
Lumex,
Russia (final
assembly-
Twinsburg,
Ohio)
OhioLumex
Co., OH
ppb
0.5 ppb-
lOOppm
Zeeman
atomic
absorption
spectrometry-
high-
frequency
modulation
of light
polarization
Direct, (no
sample
preparation)
solid/ water
analysis in 2
minutes
+/-10%
A-l
August 2007
-------
Appendix A - Information on Select Mercury Vapor Detection Instruments
Sensitivity
Response
Time
Wavelength
UV source
Stabilization
Optical cell
Heating of
cell
Pump
Filter
Jerome
431
0.003
mg/nr"
12 seconds
- sample
mode
4 seconds
- survey
mode
Not
applicable
Not
applicable
Not
applicable
Not
applicable
Not
applicable
0.75 L/min
Yes
Jerome
J405
0.013
|ig/m3
12 seconds-
sample
mode
2 seconds-
survey
mode
Not
applicable
Not
applicable
Not
applicable
Not
applicable
Not
applicable
0.75 L/min
Yes
Jerome 471
0.001 (.ig/m"
40-55
seconds until
1st reading,
then 1
reading
everv second
253. 6 nm
Mercury
Vapor Lamp
20 minutes
Dual-
reference and
sample cell
Not
applicable
5 L/min
Yes
Mercury/
Emp-1
0.001 (low
range)
0.01 (high
range)
Instantaneous
and 5 -minute
averages
254 nm
Low-pressure
mercury
discharge lamp
Reference
beam
—
1.5 L/min
Glass wool
Vm-3000
0.1
1 second
253. 7 nm
Electrodeless
Hg low
pressure lamp
Reference
beam and
thermal
Fused silica,
25
centimeters
long
70°C
Membrane, 2
L/min
PTFE, 1 n,
47-50 mm
diameter
Lumex RA-
915+
0.5 ng/mj
1 second
254 nm
Glow
discharge
mercury
lamp
5 minutes
Multi-path
cell, 1 0 meter
Not required
10 L/min
Yes
Lumex RA-
915+/RP-
91C
lOppb
90 seconds
254 nm
Glow
discharge
mercury
lamp
1 hour
Multi-path
cell, 12 cm.
Heated,
700°C
4 L/pm
Yes
Lumex RA-
915+/Pyro
915
500 ppt
90 seconds
254 nm
Glow
discharge
mercury
lamp
1 hour
Multi-path
cell, 24 cm
Heated,
700°C
4 L/pm
Yes
A-2
August 2007
-------
Appendix A - Information on Select Mercury Vapor Detection Instruments
Calibration
Power
Battery
Weight
Dimensions
(W x H x D)
RS232
standard
(interface)
Jerome
431
Factory
115 V or
230V
Internal
Ni-Cd
batteries,
(5 hours
capacity)
3.18kg
15.6x33.8
x 10.4 cm
None
Jerome
J405
Factory
100-240V
AC
Internal
rechargeable
NiMH (24
hours
capacity)
2.27 kg
16 x 16.5 x
28cm
USB (host
and slave)
Jerome 471
Factory
100-240V
AC
Rechargeable
NiMH (16
hours
capacity)
8.2kg
15.5 x 20.5 x
48.5 cm
USB slave
Mercury/
Emp-1
Automatic,
using an
absorber
None
Integrated
rechargeable
battery (11
hours capacity)
4.2 kg
11.3x23. 8x
25.6 cm
None
Vm-3000
Factory
230 V / 50/60
Hertz or
110-120 V/
50/60 Hertz
optional
Integrated 12
V batteries,
(6 hours
capacity)
Approx. 7 kg
45 x 15 x35
cm
Yes
Lumex RA-
915+
Daily by
built in test
cell/factory
calibration
once per year
220 V, 50
Hertz or 110
V, 60 Hertz
Built-in 6 V,
rechargeable
lead acid
Z>
46x21x11
cm(18x8x
4.3 inches)
Yes-
dataloager
OO
Lumex RA-
915+/RP-
91C
Daily
220V, 50
Hertz or 110
V, 60 Hertz
Built-in 6 V,
rechargeable
lead acid
1 1 kg (24
Lbs)
46x21x26
cm(18x8x
12 inches)
Computer
Lumex RA-
915+/Pyro
915
Daily
220 V, 50
Hertz or 110
V, 60 Hertz
Built-in 6 V,
rechargeable
lead acid
llkg (24
Lbs)
46x21x26
cm(18x8x
12 inches)
Computer
Sources:
Personal communication between Raj Singhvi. EPA Emergency Response Team (ERT) and Marti Otto, EPA Office of Superfund Remediation
and Technology Innovation (OSRTI). March 6, 2006.
E-mail from Joseph Siperstein, Ohio Lumex Co., to Martha Otto, EPA, regarding the Lumex RA-915+, Lumex RA-915+/RP-91C, and Lumex
RA-915+/Pvro 915 instruments. March 26. 2007.
A-3
August 2007
-------
Appendix A — Information on Select Mercury Vapor Detection Instruments
cm Centimeter
— Information not available
kg Kilogram
L/min Liters per minute
(ig/mj Micrograms per cubic meter
mg/mj Milligrams per cubic meter
ng/m" Nanograms per cubic meter
nm Nanometer
ppb Parts per billion
ppm Parts per million
V Volts
August 2007
-------
Treatment Technologies For Office of Solid Waste and EPA-542-R-07-003
Mercury in Soil, Waste, and Water Emergency Response August 2007
z www.epa.gov/tio
http://clu-in.org
United States
Environmental Protection Agency
(5203P)
Washington D.C. 20460
Official Business
Penalty for Private Use $300
------- |