Background Paper for Stakeholder Panel to Address
Options for Managing U.S. Non-Federal Supplies of
Commodity-Grade Mercury
March 14, 2007
U.S. Environmental Protection Agency
Office of Prevention, Pesticides and Toxic Substances
1200 Pennsylvania Avenue, NW
Washington, DC 20460
This paper was developed by the U.S. Environmental Protection Agency and its Federal
partners from:
Council on Environmental Quality
Department of Commerce
Department of Defense (Defense Logistics Agency)
Department of Energy (National Nuclear Security Administration)
Department of the Interior (U.S. Geological Survey)
Department of State
Office of Management and Budget
Office of Science and Technology Policy
Office of the U.S. Trade Representative
1
-------
Table of Contents
Introduction 3
Issue of U.S. Non-Federal Supplies 3
International Context 5
Overview 5
International Supplies 5
Global Demand 6
Summary 6
U.S. Mercury Supply 6
Overview 6
U.S. Government Mercury Stockpiles 8
Chlor-alkali Manufacturing Plants 9
By-Product Mercury from Metals Mining 11
Product Recycling 12
Industrial Waste Recovery 13
U.S. Imports and Exports 14
Overview of Firms in the U.S. Mercury Supply Sector 15
U.S. Mercury Use 16
Mercury Management 17
Mercury Treatment/Storage Requirements Under RCRA 17
Mercury Storage Methods and Protocols 17
Flask Management 18
Storage Facility Management 18
Key References 19
Appendix 20
2
-------
INTRODUCTION
In EPA 's Roadmap for Mercury,1 the U. S. Environmental Protection Agency
(EPA) committed to work with other Federal agencies to initiate a process with technical
experts and interested parties to assess options for managing domestic elemental,
commodity-grade mercury surpluses. In order to meet this commitment, a stakeholder
panel has been established to provide the United States Government (USG) with a
reasonable range of options and an assessment of these options for managing non-federal
supplies of mercury. The panel will present a robust discussion of the pros and cons of
these options for managing non-federal supplies.
There are different sources of mercury, including mercury recovered from the
conversion or closure of chlor-alkali plants, mercury recovered as a by-product from gold
mining, mercury recovered from product collection programs, and other recycled
mercury. How each of these sources are and should be addressed in the short and long-
terms may differ depending upon the sources. For example, the options for mercury
recovered from the closure or conversion of chlor-alkali plants may focus primarily on
storage, while the options for mercury recovered as a by-product from gold mining may
focus on storage only in the longer-term, because in the short-term it may be seen as a
preferred source of mercury to meet on-going demand. Thus, an analysis of each of the
different sources listed above may result in a different set of options for each, particularly
given that, although decreasing, there are still domestic and global demands for mercury.
All options that involve storage of excess mercury would require an assessment of
a number of issues, including: 1) which entities (e.g., a state or private company) could
be allowed to store mercury; 2) who should pay for costs of initial and ongoing storage;
3) who would be responsible and liable for security costs; 4) what should the technical
standards be for safe long-term storage; 5) where storage should be allowed, both
geographically (e.g. multiple sites or a single site) and types of storage (e.g. abandoned
mines, warehouses, etc); 6) what is the legal authority for the storage of the mercury; 7)
who will legally own the stocks and be liable for environmentally safe long-term storage;
8) what is the applicability of RCRA and other existing domestic statutes; and 9) what
legislative/regulatory changes may be needed. All management options that involve
marketing of non-federal mercury supplies should include an examination of likely
commodity mercury transactions given trends in the global price of mercury.
This background paper provides a basic overview of information on mercury
supplies and stocks (both globally and domestically); demand; and mercury management
technology. The paper also provides references to more detailed reports on these topics.
ISSUE OF U.S. NON-FEDERAL SUPPLIES
Mercury is a naturally-occurring metal that is processed and sold as a commodity
in its elemental, liquid form, and subsequently used in manufacturing and industrial
processes and in industrial and consumer products such as thermometers and thermostats.
1 EPA. EPA's Roadmap for Mercury, July 2006.
3
-------
While there are potential health concerns with the inhalation of elemental mercury, the
greater potential concern is exposure through the ingestion of fish and marine mammals
that contain mercury. Mercury that is emitted from a stack or is otherwise released to air
or water during disposal or recycling of mercury-containing products and wastes can be
the source of much of that mercury. Once emitted and transported in the atmosphere, it is
deposited, such that it finds its way into waterbodies, e.g., lakes and oceans, where it can
be biotransformed by bacteria into an organic form of mercury—methylmercury—that
accumulates in plants, then fish, then in humans and wildlife that eat the fish.
EPA expects that excess supplies of elemental, commodity-grade mercury could
emerge on the world market over the coming 10 to 30 years, as various global sources of
mercury-especially surpluses resulting from the shift away from mercury use by chlor-
alkali manufacturing plants—exceed a shrinking demand for mercury-containing
products and industrial use of mercury, particularly in the developed world. However,
demand for mercury use in artisanal mining, a major source of mercury emissions to the
environment, is expected to increase during this time frame. As a result, there is likely to
be an increasing need to ensure that programs are in place to safely manage mercury
supplies for the long-term.
On a global basis, there is currently a net flow of mercury from developed
countries to developing countries. According to a report prepared for the European
Commission (EC) in 2006, almost one-third of the global mercury supply is used for
small-scale gold mining (mostly in Africa, Asia, and South America), much of which is
lost to the environment.2 There is increasing concern that as mercury from developed
countries continues to be readily available, this supply will contribute to a continuing
reliance on mercury in developing countries, particularly in small-scale gold mining, thus
facilitating human exposures and mercury releases to the global environment. However,
the United States is a fairly small player in the global mercury market. In 2001, for
example, the United States mercury demand was about 274 metric tons, which
represented approximately 8 percent of annual global consumption.
Stakeholders have urged the Federal government to develop a coordinated
position to address government surpluses and large private sector stocks of mercury.
These stakeholders foresee an increasing need for a coordinated approach to safely
manage mercury supplies over the long term and are looking to the Federal government
to address this issue.
The Federal government has now adopted a policy of continuing to store, rather
than sell, all mercury stockpiles being managed by the Department of Defense (DOD)
and the Department of Energy (DOE). However, the Federal government also sees a
need to consider policies for addressing non-federal domestic mercury supplies
2 Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
4
-------
INTERNATIONAL CONTEXT
Overview
Domestic policies regarding mercury must also be considered in a global context.
While this paper examines domestic mercury issues, domestic management of non-
federal mercury stocks may affect international policies and trends as they apply to the
global management of commodity grade mercury.
International discussions have been underway on global issues of mercury supply,
demand and flow in the world market. The United States has been a major player in
supporting the commitments of the 2005 United Nations Environment Program (UNEP)
Governing Council mercury decision and to agreeing upon a number of strategic global
steps to further protect human health and the environment from mercury pollution. To
that end, the U.S. has supported a number of partnerships aimed at reducing mercury use
and exposure. The issue of managing global stocks to prevent increased mercury use and
releases was a major focus of the February 2007 Governing Council Meeting in Nairobi.
Another recent international consideration is the October 26, 2006, European
Commission announcement that it had introduced legislation to ban all mercury exports
outside the EU by 2011. If this ban goes into effect, it will have significant impacts on
the global market, since Europe is currently the largest exporter of mercury. In
conjunction with a potential ban, European chlorine producers are working toward a
declared goal of closing all of their mercury cell chlor-alkali plants by 2020.
The proposed European Commission export ban raises the obvious question of
how such a ban will affect the global price of mercury and whether it will lead to
increased supply of mercury, including primary mining of commodity-grade mercury in
other countries. While most policy makers agree with efforts to seek a solution to the
global mercury supply issue, this is not an issue that this panel should attempt to address.
The economic issues associated with a European export ban could have unintended,
serious, long term consequences, most of which would be felt in the developing world.
International Supplies
Mercury is currently mined only in Kyrgyzstan and China. Kyrgyzstan exports
almost all the mercury it mines; China currently mines mercury primarily to meet its
domestic demand.3 The remaining mercury supplies come from secondary sources, such
as industrial wastes and scrap products, as byproduct from gold and other metal mining,
natural gas manufacture and from closing mercury-cell chlor-alkali plants. The European
Commission estimates that in 2005 the amount of mercury available in commerce
globally was 3,690 tons.4
3 Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
4.Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
5
-------
Global Demand
The European Commission estimates that current world demand for mercury is
approximately 3,439 metric tons per year.5 While global demand remained fairly
constant from 2000 to 2005, the demand distribution changed. With the increase in the
price of gold, the use of mercury in small-scale gold mining increased by 54% becoming
the single largest use of mercury. Vinyl chloride monomer manufacture increased
significantly from very little in 2000 to 700 metric tons, becoming the second largest use
of mercury in 2005.6 Concurrently, the use of mercury in products decreased from 57%
to 32%) and chlor-alkali manufacture from 24% to 18%.7 However, the global price per
metric ton of mercury has increased over this time period from approximately $4,000 per
metric ton in 2000 to more than $16,000 per metric ton in 2007. It is expected that the
price would continue to increase following an export ban by the European Union.
Further, given the high price of gold, it is anticipated that supplies of mercury on the
global marked will continue to flow to developing countries where it will be used in
small-scale gold mining. The use of mercury in small-scale gold mining is inefficient
with much of the mercury being released. A substantial quantity contributes to the global
pool of mercury.
Summary
The issue of global mercury supply and demand is multi-faceted. An examination
of the global mercury cycle includes international consideration and knowledge of trade
flows, storage, primary mining, by-product generation and other issues. For purposes of
this panel, the international context should not be ignored; rather, it is one of many
considerations impacting a domestic policy on the long-term management of commodity
grade mercury in the U.S.
U.S. MERCURY SUPPLY
Overview
Mercury supply refers to the movement of mercury into the market over a given
time frame (i.e., the "flow" of the commodity as it is mined, recovered as by-product, or
recovered from waste) and also to certain static "stocks" or inventories of mercury.
Mercury stocks in the U.S. include military stockpiles and reservoirs of mercury
contained in products and in active mercury cells at chlor-alkali plants. Stocks are an
important consideration because they represent potential sources of supply. In the United
States, the vast majority of stocks are found in Federal government stockpiles and in
5 Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
6 Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
7 Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
6
-------
eight mercury-cell chlor-alkali manufacturing plants. Currently known U.S. mercury
stocks total at least 8,010 metric tons, of which 70% are owned and stored by DOD and
DOE, and the remaining 30% are owned and used by the chor-alkali industry. In
addition, unknown (but assumed to be relatively small) quantities of mercury are stored
for use by laboratories and individuals. Finally, mercury contained in consumer products
can be considered a "reservoir" that is a continuing source of supply over time as these
products are recycled. (See "Summary of Current U.S. Mercury Supply and Use" in the
Appendix.)
Mercury supplies produced annually in the United States are estimated to be, on
average, about 255 metric tons (see Appendix), which is fairly small compared to
existing domestic stocks, and also small relative to global mercury supplies produced
annually. Mercury production is also highly variable from year to year. U.S. sources of
mercury supplies fall into two categories:
• By-Product Production: In the U.S., by-product production consists of recovery
of mercury as a by-product from mining other metals, primarily gold. There has
been no primary mercury mining as a principal product in the United States since
the McDermitt Mine in Nevada closed in 1990.
• Secondary (Recovered) Mercury: Secondary mercury includes mercury that is
recovered via a retorting (thermal) process from mercury-containing industrial
process wastes, mining or industrial site remediation wastes, scrap consumer
products, and decommissioned mercury cells at chlor-alkali plants.
Domestic annual production of mercury supplies is highly variable.
Unfortunately there have been no publicly available data on U.S. by-product or secondary
mercury production since 1997, because as domestic demand for mercury has declined,
the number of domestic firms supplying mercury has fallen below the minimum number
needed to allow the U.S. Geological Survey (USGS) to report basic production without
revealing proprietary information.
In 2002, an industry source estimated that the total amount of U.S. by-product
mercury from mining was in the range of 70-100 metric tons per year.8 The amount of
by-product mercury appears to be increasing somewhat over time as mining facilities
strive to capture more mercury in order to decrease the amount released to the air during
purification activities, and also as gold mining activities increase in response to
increasing gold prices.
Based on USGS estimates, U.S. mercury supplies during the six-year period from
1991 to 1997 were predominantly from secondary (i.e., recovered) sources rather than
from by-product mining. In 1997, for example, secondary production was about 389
metric tons compared with 72 metric tons for by-product mining. This secondary
mercury was thought to originate from three sources: closure of mercury cell chlor-alkali
plants; retorting of scrap mercury from discarded mercury-containing products and
8 Lawrence, Bruce. Bethlehem Apparatus Company, Inc. Personal communication to EPA, July 2002.
7
-------
devices; and recovery of elemental mercury from contaminated soil and debris.9 In 2002,
a recycling industry source estimated that 75% or more of secondary mercury production
was from dismantling of chlor-alkali plants, with minor amounts from lamp recycling and
retorting facilities.10 This 2002 estimate no doubt reflected several recent closures of
chlor-alkali plants: one in 2000 in Maine, and two in 2002 in Kentucky and Texas.
U.S. secondary (recovered) mercury is produced primarily by two major metals
retorting facilities in the U.S. These companies remove the mercury from a variety of
secondary sources, and then heat the mercury in a closed vessel called a retort. The
mercury is turned into a gas, and then condensed back into its natural, liquid-metal form.
This purification process continues until the metal is brought to the stage where it meets
industry standards for reuse. It is then sold on the world commodity market, either
directly or through metals brokers.
Any prediction of the amount of secondary supply produced from recycling in the
U.S. is complicated by the option for industry to export some mercury-containing waste
to Canada under the 1986 Canada-USA Agreement on the Transboundary Movement of
Hazardous Waste. It appears that some portion of mercury-containing waste is exported
to Canada for disposal in a regulated landfill instead of being retorted and recovered in
the U.S., due to differences in government restrictions on landfilling mercury.11
Once mercury is sold, it is very difficult to track the end use. Many experts
believe, however, that mercury is sold mostly to customers in developing nations, and is
used primarily for small-scale and artisanal gold mining, followed by use for chlor-alkali
manufacturing and various mercury-containing products.12 In 2004, the U.S. estimates it
exported about 278 metric tons of mercury to countries such as Mexico, Vietnam, Peru
and Brazil.13
Attached is a table that summarizes the amount of mercury in current U.S.
inventories and estimated amounts of mercury produced and consumed annually in the
U.S. (See Table 1 in the Appendix.)
U.S. Government Mercury Stockpiles
The U.S. Government currently holds approximately 5,642 metric tons of
mercury in stockpiles that were used in the 1950s and 1960s in the production of enriched
lithium, a product used in the atomic weapons program. These military stockpiles
account for 70% of all currently known mercury stocks in the U.S. During the early and
mid-1990s, both DOD and DOE authorized the sale of mercury from their stockpiles.
9 EPA. Mercury Market Background Report, May 2005.
10 Lawrence, Bruce. Bethlehem Apparatus Company, Inc. Presentation at Breaking the Mercury Cycle
Conference, Boston, MA. May 1-3, 2002.
11 EPA. Mercury Market Background Report, May 2005.
12 Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August 2006. Report for the
European Commission.
13 U.S. International Trade Commission.
8
-------
Sales of DOD stockpile mercury were suspended in 1994 in response to environmental
concerns.
DOD has 4,436 metric tons of mercury stored in its strategic stockpiles. In 2004,
DOD decided to continue to store its mercury for the next 40 years, and to consolidate its
stockpiles in above-ground storage at one site in Nevada. Under this management
strategy, DOD mercury supplies will not enter the market and will remain as reserves
until 2044.14 Should a treatment technology become available, another decision may be
considered.
DOE has approximately 1,206 metric tons of stored mercury, which the
department decided in late 2006 to continue to store rather than sell.15 The DOE
stockpile is currently stored at its National Nuclear Security Administration (NNSA) Y-
12 facility in Oak Ridge, Tennessee. DOE has no future plans to sell its mercury
stockpile, and will continue to store it at the Y-12 facility while investigating its options
for alternative long-term storage. In evaluating these options, DOE will take into account
a number of factors, including the annual storage costs of about one million dollars.
Chlor-alkali Manufacturing Plants
In the U.S., the chlor-alkali industry is currently the largest private-sector source
of stored and in-use mercury, and therefore the largest private-sector source of potential
new supplies as a result of future closures or conversions of mercury cell chlor-alkali
equipment or plants.
There are currently eight chlor-alkali plants still using mercury cell technology.
A plant in Louisiana is expected to convert to non-mercury technology in 2007, and an
Alabama plant is expected to close in 2008. A third plant in Wisconsin has indicated it is
seeking a favorable electrical contract that would make conversion a viable option, but
negotiations with the state are still on-going. The remaining five plants have not
announced any plans to discontinue use of mercury. The eight plants are located in seven
states in the South and Midwest:
1) PPG, Lake Charles, LA (is expected to convert to non-mercury process by mid-
2007)
2) Occidental Chemicals, Muscle Shoals, AL (is expected to close in 2008)
3) ERCO Worldwide (USA) Inc., Port Edwards, WI (considering conversion)
4) Ashta Chemical, Ashtabula, OH
5) PPG, New Martinsville, WV
6) Olin, Charleston, TN
7) Olin, Augusta, GA
8) Pioneer, St. Gabriel, LA
14 DOD. Defense Logistics Agency, Defense National Stockpile Center, Final Mercury Management
Environmental Impact Statement, March 2004. 69 Federal Register 23733, 4/30/04.
15 Letter from DOE to Senator Obama, December 2006.
9
-------
The most recent Chlorine Institute report16 indicates that at the end of 2005 these
eight plants together contained a total inventory of 2,368 metric tons (2,605 short tons) in
mercury cell process equipment and on-site storage, of which 1,814 are considered to be
recoverable as new supplies. When these operating plants reach the end of their useful
lives, the mercury remaining in process equipment or storage will require recycling.
In recent decades, the chlor-alkali industry has been a major contributor to U.S.
mercury supplies, as the industry has transitioned to using mercury-free processes. This
trend is expected to continue as firms still using mercury cell technology slowly phase
out the use of this process. Alternative chlor-alkali processes have similar or slightly
lower costs than mercury cell production, and they do not face the environmental
concerns associated with the mercury cell process.17 No new mercury cell plant has been
constructed in the U.S. since 1970.
Mercury cell plants typically have a working life of 40 to 60 years, and currently
all U.S. mercury-cell plants are over 30 years old. Hence, it is reasonable to expect that
mercury cell operations will be shut down sometime over the next three decades as the
remaining plants reach the end of their useful lives.18 Since 1980, available information
suggests that 17 U.S. mercury cell chlor-alkali plants have closed.
The rate at which the remaining mercury cell plants will close is uncertain. Under
the current regulatory regime, individual plant economics and chlorine industry market
conditions will probably be the major factors driving the closure rate. Prospective
investments in mercury control equipment needed for compliance with new air emissions
regulations may be partially responsible for some recent decisions to shut or convert
mercury cell plants. Moreover, higher electricity costs could promote shutdowns or
conversions, since membrane cell plants have lower electricity requirements than
mercury cell plants. Reasonable assumptions for the disappearance of mercury cells in
the U.S. range from 10 to 30 years, given the current age of plants and recent industry
trends. However, companies may choose to run their plants at the upper end of this range
because they have identified ways to minimize mercury losses. By adding more mercury
to the cells in a one-time upgrade, the cells can be operated at lower temperatures, which
reduce energy consumption and fugitive mercury emissions.19 Planned mercury cell
upgrades were expected to be completed by 2005.20
Historically, mercury from chlor-alkali closures has re-entered mercury
commodity markets, although in recent years the chlor-alkali industry has reused its own
mercury to a great extent. This mercury is then either re-sold to users of comparable
grade mercury, such as other chlor-alkali facilities, or dealers may distill the recovered
mercury to produce a higher-grade product. Sale of mercury back into commodity
markets reduces the costs of dismantling or converting chlor-alkali plants.21
16 Chlorine Institute, The. Ninth Annual Report to EPA, May 15, 2006.
17 Chlorine Institute, The. Written communication to EPA, October 7, 2002.
18 EPA. Mercury Market Background Report, May 2005.
19 EPA. Mercury Market Background Report, May 2005.
20 Chlorine Institute, The. Sixth Annual Report to EPA. May 12, 2003.
21 EPA. Mercury Market Background Report, May 2005.
10
-------
While there is some local political pressure for these plants to convert to non-
mercury technology or to close because of the plants' mercury emissions, there are no
requirements for existing U.S. plants to close or convert to non-mercury processes. U.S.
regulations limit mercury emissions from chlor-alkali plants, and effectively prohibit the
new construction of mercury-cell chlor-alkali plants.
By-Product Mercury from Metals Mining
Mercury is recovered as a by-product from mining other metals, primarily gold.
Mercury occurs naturally as a constituent of gold ore, and is an impurity during the
smelting process for purifying the gold. By-product mercury is recovered from gold-
processing precipitates and from the calomel22 collected from pollution control devices at
gold smelters, mainly in Nevada. Mercury present in the ore of these other minerals that
is not recovered is generally emitted to the air, water, or land. This industrial gold
mining process is different from artisanal and small-scale gold mining activities, where
elemental mercury is added to the process to aid in the gold recovery.
While by-product mercury production in the United States is most commonly
associated with gold mining, some by-product mercury is also generated from other
metals mining, including copper, zinc, lead and silver mining.
Production of by-product mercury appears to have been relatively stable from
1990 through the early 2000s. Estimates derived by calculating by-product mercury as a
function of gold production range from 59 to 73 metric tons of mercury per year from
1990 to 2002.23 An industry source estimated in 2002 that between 70 and 100 metric
tons of by-product mercury are produced annually.24 However these estimates remain
uncertain, given the lack of recent published data on the amount of by-product mercury
mined in the U.S. Moreover, by-product recovery may be increasing because Nevada
gold mines have recently begun to use new mercury control technology for
environmental reasons, resulting in recovery of more mercury.
The USGS reported in January 2003 that there were still a small number of gold
mines located in western states that continue to produce limited quantities of mercury by-
product.25 The most recent USGS Minerals Yearbook reports that in 2005, by-product
mercury and calomel were produced at several gold and silver mines in Nevada. In
addition, by-product mercury was imported to the U.S. from gold mines in Chile and
Peru, then further refined and sold for domestic use or export. Calomel was also captured
by pollution-control devices at ore roasters at domestic smelters to recover mercury for
resale.26
22 Calomel is the ancient name, often still used, for mercurous chloride (Hg2Cl2). It is a heavy, white or
yellowish-white, tasteless and odorless powder which in the past was used medicinally as a cathartic.
23 EPA. Mercury Market Background Report, May 2005
24 Lawrence, Bruce. Bethlehem Apparatus Company, Inc. Personal communication to EPA, July 2002.
25 USGS. Mineral Commodity Summaries: Mercury, 2003. Prepared by W.E. Brooks.
26 USGS. Minerals Yearbook: Mercury, 2005. Prepared by W.E. Brooks.
11
-------
Because mercury recovery represents a small fraction of income from gold
mining, future domestic production of by-product mercury will depend primarily on
environmental management strategies of mining firms and on levels of U.S. gold
production. Gold prices in recent years have been hundreds of times higher than mercury
prices and gold is produced at approximately five times the quantity of by-product
mercury. The difference in production rates and prices for these two commodities is so
great that gold ultimately will drive any decision about by-product production levels from
industrial scale mines.27
The amount of by-product mercury recovered from mining is expected to increase
over time, as more mining facilities strive to decrease the amount of mercury pollution
released to the air during mining activities, and as the amount of gold mining increases
due to increasing world gold prices.
Product Recycling
The recovery of mercury from industrial and consumer products represents
another source of secondary mercury supply. Mercury can be drained directly from
products and also recovered via retorting. Historically, this source of supply has been
less significant than the chlor-alkali sector, in part because the transaction costs of
collecting products and the costs of retorting are higher.
Estimates in 2002 indicated that mercury recovered from products in the United
States totaled about 35 metric tons per year, with most of this mercury recovered from
measuring devices such as industrial gauges.28 Roughly 3.5 metric tons was estimated to
come from recycling of mercury in fluorescent bulbs.29 Other products such as
thermometers and laboratory instruments contributed a modest quantity of mercury.30
These estimates of product recycling address only a portion of the total mercury
discarded in products each year. EPA's Toxics Release Inventory (TRI) data suggest that
almost 42 metric tons of mercury were disposed in 2002.31 In addition, EPA estimated
that the reservoir (i.e., stock) of mercury in products still in use is about 2,000 metric
tons.32 Many of these products have long working lives, and therefore only a relatively
small portion of this reservoir might be expected to enter the waste stream in any given
year. Assuming a 10 to 20 year period for the current reservoir of about 2,000 metric
27 EPA. Mercury Market Background Report, May 2005.
28 Lawrence, Bruce. Bethlehem Apparatus Company. Personal communication to EPA, July 2002.
29 NEMA (National Electrical Manufacturers Association. Written communication to EPA, September 17,
2002.
30 EPA. Mercury Market Background Report, May 2005.
31 EPA. Toxics Release Inventory (TRI) Program, 2005. The TRI data do not distinguish between mercury
derived from product recycling or waste recovery. Consequently, TRI estimates reported in the "Product
Recycling" category also represent mercury supplies from the "Industrial Waste Recovery" category.
32 EPA. Mercury Market Background Report, May 2005.
12
-------
tons to flow into the waste stream, approximately 100 to 200 metric tons of material
potentially would be available for recovery annually.33
Many state and local governments have for several years promoted public and
private collection programs for both bulk elemental mercury (e.g., from schools,
laboratories and dentists) and discarded mercury-containing products such as
thermometers. Some businesses are also collecting unwanted mercury or mercury-
containing products, such as thermostats and, more recently, fluorescent light bulbs.
The National Vehicle Mercury Switch Recovery Program, which was launched in
2006, is a new voluntary program to remove and collect mercury-containing switches
from scrap (or retired) vehicles manufactured prior to 2003. This national program is the
result of a two-year collaboration involving EPA, States, environmental organizations,
and several industry sectors. The program complements existing state mercury switch
reduction efforts, and will help to reduce up to 75 tons of mercury emissions over the
34
next 15 years.
It is unclear how much mercury is being collected through these various voluntary
collection programs, but it is likely that supplies of recovered mercury may increase over
time due to the increasing number of collection programs.
Industrial Waste Recovery
Information on mercury recovered from hazardous industrial wastes and from
contaminated soil and debris is limited. Lawrence35 estimates that treatment of
contaminated soil and debris produces approximately 35 metric tons of mercury per
year.36 This quantity, however, is likely to be quite variable from one year to the next.
Treatment of hazardous industrial process wastes is also a potential source of secondary
mercury, although it does not appear to be significant in quantity. Analysis of EPA's
Biennial Reporting System for hazardous wastes does not provide data on quantities of
mercury recovered, but it does indicate that the majority of waste that is recovered goes
to large retorting operations for treatment and is thus captured in Lawrence's estimates of
total mercury recovery.
Another potential source of secondary mercury is cleanup of mine sites. Many
western U.S. States have abandoned mercury mines and gold and silver mines
contaminated with mercury. California alone has more than 300 such mines. EPA is
undertaking efforts to characterize abandoned mine sites, and to the extent that a sizeable
number of cleanups occur, the amount of mercury from soil and debris could increase.
33 Note that the mercury in products includes mercury that is imported into the U.S. as products (e.g.,
lights) as well as commodity mercury used to manufacture products domestically.
34 EPA. Fact Sheet: National Vehicle Mercury Switch Recovery Program, August 2006. EPA's Mercury
Website: www.epa.gov/mercury/switchfs.htm
35 Lawrence, Bruce. Bethlehem Apparatus Company. Personal communication to EPA, July 2002.
36 Mercury recovered from waste is separate from and in addition to quantities recycled from mercury-
containing products.
13
-------
In addition, many States are in the process of regulating indirect dischargers to
surface waters (e.g., dental offices). In the future, these efforts may result in recoverable
elemental mercury, especially from discarded dental amalgams. According to a recent
study commissioned by the American Dental Association, dental offices currently release
approximately 6.5 metric tons of mercury each year. Of this quantity, approximately 0.3
metric tons are emitted annually to surface waters after processing at wastewater
treatment facilities; the remaining mercury is transferred to grit solids or bio-solids at the
treatment facility.37
Mercury may also be recovered from contaminated soil in proximity to natural
gas pipelines. These pipelines use mercury manometers for metering flow. Over time,
routine maintenance activities on these meters have released mercury into the soil, and
contamination is now prevalent at many of these sites.
U.S. Imports and Exports
Imports of elemental mercury represent a potential source of supply for domestic
mercury needs. The U.S. imported almost 2,000 metric tons of mercury from 1989 to
2002. In the same period, however, exports totaled over 4,733 metric tons of elemental
mercury.38 Moreover, exports exceeded imports in the majority (nine of 14) of the years
reported. The United Nation Environment Program estimated that in 2004 the U.S.
exported more than 400 metric tons of mercury.39
The overall pattern of imports and exports suggests that a considerable portion of
U.S. trade in mercury is not driven by domestic use or production, but is instead
purchased by distillers or dealers for subsequent sale to other markets. The wide variety
of nations of origin and quantities that comprise the mercury imports to the U.S. are also
indications that trade is not a major source of supply to domestic users.40 If U.S. imports
were specifically targeting domestic demand, one would expect a pattern of more
consistent imports over time reflecting consistent domestic demand. In addition, recent
secondary production estimates indicate that the U.S. has been able to meet, and in some
cases exceed, its own demand.41
While imports do not appear to be a primary source of mercury for domestic
producers, they may perform a necessary smoothing role at times when other sources of
supply are unavailable to meet demand. For example, a recent lull in chlor-alkali plant
closures has reduced secondary production since 2003. Non-domestic trade
considerations (i.e., brokering) also drive imports, especially when enough secondary
mercury is produced to meet demand in the U.S. This suggests that future imports will
37 ADA (American Dental Association). Assessment of Mercury in the Form of Amalgam in Dental
Wastewater in the United States, 2003. Prepared by ENVIRON International Corporation.
38 For years 1989 through 2001, U.S. Customs data as compiled by ITC; for 2002, data provided by USGS,
Mineral Commodity Summaries: Mercury, 2003. Prepared by W.E. Brooks.
39 UNEP; Summary of Supply, Trade and Demand Information on Mercury 2006.
40 U.S. Customs data show eight different "top" suppliers to the U.S. over the 14 year period from 1989 to
2002: Canada, Spain, Germany, Australia, United Kingdom, Russia, Kyrgyzstan, and Chile.
41 EPA. Mercury Market Background Report, May 2005.
14
-------
continue to be driven primarily by the need to smooth out unevenness in domestic
supplies from secondary and by-product sources, or by non-domestic trade
considerations.42 43
Overview of Firms in the U.S. Mercury Supply Sector
The U.S. supply sector comprises two major players—Bethlehem Apparatus
Company, Inc. and D.F. Goldsmith Chemical and Metal Corporation—who supply
virtually all commodity mercury to U.S. consumers. Bethlehem Apparatus operates 29
advanced high vacuum mercury waste retorts, two continuous feed fluorescent lamp glass
retorts, and eight quadruple distillation systems in continuous operation. The company
also provides mercury waste management services. Bethlehem estimated in 2002 that it
supplied about 70 percent of U.S. mercury demand. Bethlehem is the major purchaser of
mercury by-products from U.S. mines, and estimates that it has 40 percent of the U.S.
mercury recycling and recovery market share. Bethlehem Apparatus has a relatively
small group of large-volume customers who typically purchase mercury under six to
twelve-month supply contracts. Approximately 80 percent of the company's sales come
from these long-standing customers.44
D.F. Goldsmith likely accounts for virtually all remaining mercury sales to end-
users in the U.S. These sales represent approximately one-third of Goldsmith's total sales
volume. Goldsmith has a long-standing relationship with the chlor-alkali industry and
appears to be the major buyer of mercury from closed chlor-alkali plants in the United
States.45 Goldsmith is not permitted to treat hazardous wastes, and therefore acts as a
broker of mercury recovered by others (e.g., Mercury Waste Solutions, Inc.).
Other firms that might be counted as part of the mercury supply sector in the U.S.
are recyclers, such as Mercury Waste Solutions (MWS) of Mankato, Minnesota, Onyx
Special Services of Fond du Lac, Wisconsin (a division of Veolia), and AERC Recycling
Solutions, Inc. of Allentown, Pennsylvania. MWS does not have high-level purification
equipment, and therefore sells very little mercury directly to customers. Instead, MWS
42 In addition to trade in elemental mercury, the U.S. imports a variety of mercury-based compounds. At
present, imports of these compounds have fallen to relatively low levels due to the phase-out of various
uses of mercury in pesticides, paint, and other products. The exceptions are mercuric chloride and organo-
mercury compounds. Mercury chlorides (HTS Code 28273920) are occasionally imported by mercury
distillers in the form of calomel (also written as mercury (1) chloride), for recovery and resale of elemental
mercury. From 1989 to 2001, the U.S. imported a total of 411 metric tons of mercury chlorides, potentially
representing as much 330 metric tons of distilled mercury (if all imports are calomel). In 2001, the
imports of mercury chlorides totaled 22 metric tons. Imports of organo-mercury compounds (used in
fungicides, bactericides, and pharmaceuticals) were 37 metric tons in 2001. Mercury compounds are
generally included in production data (either produced in mines or, in the case of calomel, as a secondary
source). However, the fact that these compounds are traded in quantity is an important consideration in
developing any policy that would track or restrict imports or exports of mercury.
43 EPA. Mercury Market Background Report, May 2005.
44 Lawrence, Bruce. Bethlehem Apparatus Company. Personal communication to EPA, July 2002.
45 Lawrence, Bruce. Bethlehem Apparatus Company. Personal communication to EPA, July 2002.
15
-------
sells mercury to either Bethlehem Apparatus or D.F. Goldsmith46 for additional
refinement and resale.47
U.S. MERCURY USE
Use of elemental mercury for products and processes has declined significantly
over the last several decades, in the U.S. and globally. These declines are attributable
both to the phase-out of certain mercury uses and to reductions in the quantity of mercury
used in individual products.
In 1980, the three largest U.S. industrial uses of mercury were in batteries (1,052
metric tons), the chlor-alkali manufacturing process (358 metric tons), and paint (326
metric tons).48 Between 1980 and 2001, there was a dramatic 83 percent drop in annual
mercury use by industries, from 2,225 metric tons to 274 metric tons.49 This reduction in
use was due in large part to state and congressional limits placed on mercury use in
batteries, EPA's cancellation of pesticide registrations for using mercury as a fungicide in
paint, closure of some mercury-cell chlor-alkali manufacturing plants, and voluntary
reductions made under the United States-Canada Great Lakes Binational Toxics Strategy.
By 2001, mercury use in batteries had decreased significantly, use in paint had
ended, and annual use by the chlor-alkali industry had decreased to 38 metric tons or 12
percent of overall mercury use by U.S. industry. Since then, further progress has been
made by the chlor-alkali industry. As a result of a voluntary commitment to mercury
reduction made by the U.S. Chlorine Institute under the Great Lakes Binational Toxics
Strategy, the chlor-alkali industry reduced its annual use of mercury by 91 percent
between 1995 and 2005, after adjusting for shut down facilities. The chlor-alkali industry
reported its annual usage in 2005 to be 9 metric tons.50
Estimated mercury use in products in 2001 was 245 metric tons. The dominant
use was in switches and wiring devices at 42 percent (103 metric tons), followed by
measuring and control devices at 28 percent (69 metric tons), dental amalgam at 14
percent (34 metric tons), and electrical lighting at 9 percent (21 metric tons). All other
51
smaller uses accounted for 7 percent (17 metric tons).
Over time, the distribution of mercury used across various sectors has changed
significantly as environmental regulations have limited some uses (e.g., paint) and new
technology has increased mercury use in other sectors (e.g., high intensity discharge
46 Cornwell, J. Mercury Waste Solutions, Inc. Personal communication to EPA, June 2001.
47 Information on mercury recycling firms viewed at the following websites: ;
; ;
; .
48 Jasinski, S.M. The Materials Flow of Mercury in the United States. U.S. Bureau of Mines, Information
Circular 9412, 1994.
49 For 1980 to 1997: USGS Minerals Yearbook: Mercury, 1994-2001. For 2001: Lawrence Bruce, 2001
and The Chlorine Institute, 2006.
50 Chlorine Institute, The. Ninth Annual Report to EPA, May 15, 2006.
51 Lawrence, Bruce. Bethlehem Apparatus Company. Personal communication to EPA, June 2001.
16
-------
lamps in the lighting sector). These shifts are projected to continue as chlor-alkali use
decreases and alternatives to mercury-containing devices become more prevalent. Most
of these uses are expected to decline in the future. However, future demand for mercury
in switches and relays is uncertain given the lack of viable alternatives for certain
applications. In addition, mercury use is expected to continue in the lighting sector as
new applications are evolving, although the amount of mercury used per unit may
continue to decline. Although alternatives exist, dental amalgam remains the most
widely used and least costly option available for tooth fillings. The current expectation is
that use of mercury in dental amalgams will decline, but more slowly than some other
uses.52
MERCURY MANAGEMENT
Mercury Treatment/Storage Requirements Under RCRA
There is currently no cost-effective or proven environmentally safe treatment and
disposal method for elemental mercury,53 and current regulations under the Resource
Conservation and Recovery Act (RCRA) require high concentration mercury wastes be
retorted for mercury recovery and reuse rather than be disposed of in a landfill.54
How elemental mercury is characterized (i.e., as a waste or as a commodity) is
relevant to discussions of domestic mercury management. Industry has raised the issue
of limitations and liability that RCRA laws place on the storage of mercury.
Under RCRA, elemental mercury that is not declared a waste (i.e., is a
commercial chemical product), or that results from retorting of wastes, is not considered
a RCRA waste if it is put back into commerce or is intended to be put back in commerce.
However, when there is no longer any intention of using or selling the mercury, it is
considered to be a waste, and therefore the RCRA Land Disposal Restriction (LDR)
requirements apply. The LDR standard for "high mercury wastes" (greater than 260
mg/kg total mercury) is retorting, which is obviously not necessary for mercury that is
already in the pure, elemental state. A long-term disposal facility for elemental mercury
that has been declared a waste must have a Subtitle C, hazardous waste RCRA disposal
permit. The mercury placed in such a facility will have met the LDR requirements if it is
in the pure, elemental state.
Mercury Storage Methods and Protocols
DOD's Defense National Stockpile Center (DNSC) provides one of the best
examples of a successful approach to safe storage of elemental mercury. This section
provides a brief summary of methods and protocols that have been used by the DNSC.
52 EPA. Mercury Market Background Report, May 2005.
53 EPA. Economic and Environmental Analysis of Technologies to Treat Mercury and Dispose in a Waste
Containment Facility, 2005.
54 EPA. Mercury Laws and Regulations. For information on how mercury is regulated under RCRA, see:
www.epa.gov/epaoswer/hazwaste/mercury/regs.htm.
17
-------
Flask Management
Storage of mercury may require large amounts of space. For example, at DOD's
Somerville, New Jersey depot, 2,617 tons of mercury has been stored in a warehouse in
80,000 square feet of space.55
There are various technical options for safely storing elemental mercury. For
example, mercury managed by the DNSC is stored in 76-pound flasks, which are in turn
sealed in airtight 30-gallon drums. There are six flasks per drum and five drums per
pallet. Inside the drums, the flasks are individually sealed in plastic bags, separated by
dividers, and placed on an absorbent mat that doubles as cushioning material. The drums
rest on catch trays on wooden pallets on sealed floors. The pallets are not stacked in
order to facilitate inspection and air monitoring. As a result, leakage of mercury in an
amount sufficient to escape the warehouse is unlikely.56
While DNSC mercury is stored in 76-pound flasks, there are other storage options
in use by other entities. Some mercury is stored in metric ton flasks, and some in plastic
bottles.
Storage Facility Management
DNSC has safely stored mercury for over 50 years. Periodic inspections ensure
that mercury storage containers are in good condition and leak free. Any defects in the
packaging are quickly corrected. Inspections are conducted by appropriately trained
DNSC or contract personnel. Warehouses are locked except for inspections and other
periodic maintenance work. Additionally, perimeter fencing and controlled access
(taking into consideration the potential for the most unlikely scenarios) are handled by
appropriately trained personnel.57
55 Quicksilver Caucus. Stewardship of Mercury: Storage of Mercury, October 2003.
56 DOD. Defense Logistics Agency, Defense National Stockpile Center, Final Mercury Management
Environmental Impact Statement, March 2004.
57 DOD. Defense Logistics Agency, Defense National Stockpile Center, Final Mercury Management
Environmental Impact Statement, March 2004.
18
-------
KEY REFERENCES
1. Chlorine Institute, The. Ninth Annual Report to EPA for the Year 2005.
2. DOD. Defense Logistics Agency, Defense National Stockpile Center, Final
Mercury Management Environmental Impact Statement, March 2004. 69 Federal
Register 23733, 4/30/04.
3. EPA. Mercury Laws and Regulations. For information on how mercury is
regulated under RCRA, see:
www.epa.gov/epaoswer/hazwaste/mercury/reg_stand.htm.
4. EPA. Economic and Environmental Analysis of Technologies to Treat Mercury
and Dispose in a Waste Containment Facility, 2005.
5. EPA. Mercury Market Background Report, May 2005
6. EPA. EPA's Roadmap for Mercury, July 2006.
7. Maxson, Peter. Mercury Flows Report: Mercury Flows in Europe and the World,
The Impact of Decommissioned Chlor-alkali Plants, 2004. Report for the
European Commission.
8. Maxson, Peter. Mercury Flows and Safe Storage of Surplus Mercury, August
2006. Report for the European Commission.
9. Maxson, Peter. Summary of Supply, Trade and Demand Information on Mercury,
November 2006. Report for the United Nations Environment Programme
(UNEP).
10. Quicksilver Caucus. Stewardship of Mercury: Storage of Mercury, October 2003.
11. USGS. Minerals Yearbook: Mercury, 2005. Prepared by W.E. Brooks.
12. UNEP. Global Mercury Assessment, 2002. Report of the United National
Environment Programme.
19
-------
APPENDIX
Table 1: Summary of Current U.S. Mercury Supply and Use
Current U.S. Mercury Inventories
Total Metric Tons
DOD stockpile (excess supply)
4,436
DOE stockpile (excess supply)
1,206
8 mercury-cell chlor-alkali plants (in
process equipment and on-site storage)
2,368 (2,605 tons)*
Laboratories (schools, universities,
No data available. Assumed to be a
research & commercial)
relatively small amount.
Individuals (dentists, scientists)
No data available. Assumed to be a
relatively small amount.
Total
At least 8,010
Annual U.S. Mercury Supply
(Domestic Production)
Metric Tons Per Year
By-product production: metals mining
(especially from gold mining)
Variable (est. range: 70-100)**
Recovered from chlor-alkali plants
either closed or converted to non-
Variable (avg. plant: 300)* Depends on
how many plants close/convert in any year
mercury process.
Recovered through retorting from waste,
scrap, soil, debris, product collection
Variable (est.: 70 or more)**
programs
Total
Highly variable
Annual U.S. Use (Domestic
Metric Tons Per Year
Consumption)
Hg-cell chlor-alkali plants (8)
9 (10 tons)*
Products manufacturing (lights,
switches, measuring/control devices,
dental materials, etc.)
245**
Laboratories (schools, universities,
No data available. Amount assumed to be
research & commercial)
relatively small.
Total
Approx. 250 or more
References:
*The Chlorine Institute, Inc., 2006. Ninth Annual Report to EPA for the Year 2005.
Accessible at: www.epa.gov/region5/air/mercurv/7thcl2report.pdf.
** Lawrence, Bruce, 2002. Bethlehem Apparatus Company, Inc. Personal
communications to EPA in July 2002.
20
------- |