Great Lakes Binational Toxics Strategy

IMPLEMENTING THE BINATIONAL TOXICS STRATEGY
Implementation

Stakeholders Forum
November 16-17, 1998 - Chicago, Illinois

CGLI's First Year Report

Acknowledgments

This report was produced by the Council of Great Lakes Industries (CGLI), supported by a grant from U.S. EPA (project number GL98229), and the greatly appreciated hard work of many people in industry who voluntarily provided their time and effort toward the success of the Binational Toxics Strategy. The information contained, regarding the status of Level I Binational Toxics Strategy substance releases, was acquired through a multi-Industry/multi-company participative effort requiring hundreds of conversations, meetings with many industrial groups and organizations, and responses to information requests and questionnaires. The report was prepared by Dale K. Phenicie (Environmental Affairs Consulting), with assistance from Evelyn Strader (Strader and Company), and contributions from members of CGLI along with many others within industry.

Requests for copies of this report, or questions, should be directed to Mr. George Kuper, President and CEO, Council of Great Lakes Industries, PO Box 134006, Ann Arbor, MI 48113-4006, (734) 663-1944, FAX (734) 663-2424, e-mail:ghk@cgli.org.

This report and additional information regarding the Great Lakes Binational Toxics Strategy can be found on the CGLI Website at http://www.cgli.org .

IMPLEMENTING THE GREAT LAKES BINATIONAL TOXICS STRATEGY

A Report by the Council of Great Lakes Industries
Regarding Progress on Project GL98229

Mobilizing/Coordinating Industry Support of the
Great Lakes Binational Toxics Strategy

Introduction

Through U.S. EPA project no. GL98229, Mobilizing/Coordinating Industry Support of the Great Lakes Binational Toxics Strategy, the Council of Great Lakes Industries (CGLI) has undertaken an effort to assist the Agency in the implementation of the Great Lakes Binational Toxics Strategy (BNTS or the Strategy).

CGLI regards the BNTS as an important voluntary vehicle for achieving the persistent toxic substances virtual elimination objectives of the Great Lakes Water Quality Agreement. Through this implementation effort, CGLI has sought to increase awareness of the voluntary program opportunities provided by the Strategy within industrial organizations and collect information regarding the successes already achieved and the future commitments made by industries in eliminating or reducing persistent toxic substance discharges. This report describes the results of these efforts.

The Council of Great Lakes Industries is a non-profit organization that represents the common interests of U.S. and Canadian industrial organizations from the manufacturing, utilities, transportation, communications, financial services and trade sectors that have investments in the Great Lakes Basin. The Council works to ensure that industry is a substantive partner in the Great Lakes region's public policy development process. The Council is a partner organization with the World Business Council for Sustainable Development, Geneva, Switzerland.

Findings and Conclusions

CGLI's findings and conclusions regarding Strategy implementation can be summarized as follows.

• In spite of the limitations described in the conclusions that follow, substantial progress has been made towards achieving BNTS goals.
• A significant number of industrial organizations contacted by CGLI have been either unaware of the existence of the Strategy or are unfamiliar with its challenge goals and voluntary program opportunities. These contacts have substantially increased industrial community awareness of the Strategy.
• New commitments to address Strategy Level I substances are now being made by area industries. A very recent example is the voluntary agreement regarding mercury use and release signed by three Indiana steel mills, the Lake Michigan Forum, U.S. EPA and the Indiana Department of Environmental Management.
  Another example of these new commitments is the action of the Dow Chemical Company which has set a goal for the company to reduce air and water emissions of hexachlorobenzene and mercury compounds by 75 percent by 2005.
• Smaller industries exhibited less Strategy awareness than larger organizations. CGLI continues to seek effective ways to communicate with this sector.
• The significant overlaps between Strategy elements and other regulatory programs are of concern to industrial organizations. Strategy chemical release targets do not necessarily match those set by other regulatory programs. The characterization of chemical substances regulated by other programs such as Clean Air Act MACT standards often differs from the substances included on the Strategy Level I list. (e.g., total PAHs vs. B(a)P.) This difference in compound focus makes it less likely that release and control capability data are available for Level I listed substances.
• Industrial organizations value voluntary programs such as the Binational Toxics Strategy. However, the fact that substance lists and challenge goals were fixed within the Strategy document at the time participants were asked to "volunteer," decreases the flexibility available to organizations wishing to comply with these fixed elements. Industries suggested that a broader "volunteer" effort, which allows participation in chemical selection and reduction targets, would have more appeal.
• Limitations on resources available for industries to meet a wide range of regulatory mandates deter, to some extent, interest in "volunteering" for Strategy implementation efforts. At the least, industries must approach participation in the BNTS on a risk based prioritized basis. Those substances present at levels which represent the most significant threat to ecosystem health and which can be addressed on a cost effective basis must be tackled first. In addition, "volunteering" to address these higher priority substances will have to be weighed against the need to commit resources for other, non-optional, compliance mandates. A Project XL style trading program where Strategy substances are addressed in lieu of other mandated programs may provide an avenue of increased participation.
• As pointed out above, in spite of barriers encountered, CGLI found many examples of progress and/or commitment towards achieving the goals of the Great Lakes Binational Toxics Strategy. Examples of the substantial progress made regarding individual Level I substances include:
  • Mercury
    • Battery manufacturers have reduced the use of mercury in their products by 99.95%. In 1984/85 this sector was the largest mercury user (55% ). In 1995, the U.S. Department of Interior reports that they used less than 0.5 tons of the metal per year.
    • Electric lamp producers have reduced the use of mercury (necessary for the production of fluorescent bulbs) from 48 to 23 mg (average use basis) per four foot lamp.
    • Electric Power Research Institute (EPRI) tests show that ESP/fabric filer and ESP/FGE systems reduce mercury emissions from coal fired power plants, but results are variable.
    • Thermostat manufacturers have implemented recycling programs for decommissioned devices.
    • The American Home Appliance Manufacturers Association has produced training material urging dealers and the public to recycle mercury-containing devices found in decommissioned equipment.
    • The Chlorine Institute has committed to a 50 percent reduction of mercury (80 metric tons annually), used by the chlor-alkali sector by 2005.
    • The voluntary agreement signed by three Indiana steel mills, referenced above, will provide an inventory of mercury uses and releases at these plants and lead to reductions.
    • Consumers Energy through an ongoing Mercury Pollution Prevention Initiative begun in 1996 has taken action to identify mercury sources within the company, estimate the total quantity of mercury used, review existing disposition practices and investigated future management options and costs. The program has heightened the awareness of mercury concerns in the company and identified options for use of non-mercury containing equipment. It has reduced the use of equipment containing mercury and associated stock inventory. 231 pounds of the substance were removed from use in 1996. 171 additional pounds in 1997.
  • Octachlorostyrene
    • Graphitic electrode processes, suggested as a source of OCS contamination during chlorine and caustic soda production, are not used in the Great Lakes Basin. No release of OCS from these sources is thought to occur.
    • Pulp and paper processes have not been found to generate or release OCS. The Canadian test data on which an association between pulp mills and OCS generation, failed QA/QC standards and has been removed from the MISA database. In addition a review of pulp and paper chemistry has shown that formation of highly chlorinated compounds, like OCS, are not favored.
  • PCBs
    • Many industrial facilities are working to, or have already, phased out PCB containing equipment.
    • In U.S. industry, storage of used or decommissioned PCB equipment is not a widespread practice.
  • Dioxin
    • Improvements in combustion processes have lowered, and are lowering dioxin emissions levels from sources in all sectors. MACT standards compliance programs will produce further reductions.
    • Virtual elimination of dioxins has been accomplished by the pulp and paper and chemical process industries.
    • The Vinyl Institute (VI), a business unit of the Society of the Plastics Industry, Inc., characterization of dioxin/furan releases from ethylene dichloride, vinyl chloride monomer, and polyvinylchloride production processes show that, of the 3,000 grams of dioxins which U.S. EPA estimated were released to air, land, and water in 1995, only 11.9 to 38.5 grams (TEQ basis) came from the above named processes.
    • The Chlorine Chemistry Council (CCC), a business council of the Chemical Manufacturers Association, survey of dioxin/furan releases from chlor-alkali processes shows that, of the same 3,000 grams, wastewater releases from this sector amount to 6.4 to 8.8 grams annually (TEQ basis). Approximately one percent of the total dioxin/furan release reported by U.S. EPA comes from sources tested through the VI and CCC programs.
  • Benzo(a)Pyrene
    • Relative to 1992, significant reductions in steel industry coke oven emissions have resulted in substantially reduced releases of B(a)P.
  • Hexachlorobenzene
    • Agricultural uses of hexachlorobenzene (HCB) have ceased in North America. Use of the material as a fungicide was terminated in Mexico in 1991.
    • HCB contaminate levels in products such as pesticides and solvents have decreased dramatically. When measurable, they are substantially (orders of magnitude) below regulatory levels.
    • Few, if any, aluminum smelters use the hexachloroethane degassing process, limiting this potential source of HCB release. Pulp and paper processes have not been found to generate or release HCB.
  • Alkyl-Lead
    • Use of alkyl-lead in consumer automotive fuels ceased on or before December 31, 1995.
    • Quantities of leaded fuel, not yet identified by CLGI, continue to be used in piston type
      aviation engines. Much smaller quantities, perhaps 20 percent of the current total, continues to be used in auto racing engines.
    • Future trends for these remaining leaded fuel uses are uncertain, but evidence suggests that these remaining uses may be insignificant as lead levels near U.S. roadways decreased 97 percent between 1976 and 1995. CGLI continues to pursue use quantity information for these applications.
    • The use of alkyl-lead as/or in chemical intermediates or organo-metallic compounds is not wide spread. Little information is available regarding this suggested use.
  • Pesticides
    • There are no industrial sources of the Level I pesticides.
    • Velsicol Chemical Corporation ceased production of export market Chlordane in mid 1997. All remaining stocks were to have been sold by the end of 1997. Velsicol pledged not to make its proprietary technology, for the production of Chlordane, available to any other company.
      

Project Approach

Awareness contacts and awareness materials:
In addition to CGLI Members, Binational Toxics Strategy (BNTS) awareness contacts were made (via telephone, FAX, mail, and personal communications) with many industrial organizations. They included individual companies, manufacturing sector trade and technical associations, and associations which represent more than one manufacturing sector. Letters providing information on the BNTS were sent to the Chief Executives Officers of 46 major corporations in the Great Lakes Region. More than 80 associations were contacted regarding the strategy and were given a newsletter article about the BNTS for use in their own association newsletters. CGLI president met with the Great Lakes Commissioners regarding the strategy and CGLI's chairman sent letters to each of the Great Lakes governors and premiers. CGLI devoted much of its own newsletter to the BNTS issues. In all, nearly 200 organizations were contacted directly. The associations included among these reach additional organizations through their memberships. A listing of the industry sectors represented by these contacts appears in Appendix 1.

Awareness materials were prepared and used for these organization contacts. They included:

• a "backgrounder" on the Strategy,
• a success story reporting form, and
• a substance baseline summary.

The "backgrounder" summarized the BNTS, the chemicals involved, and the challenges. Emphasis was placed on the voluntary aspect of the program and the opportunities, which the program provides for industries. The success story reporting form provided summary information regarding the BNTS and requested information from industries regarding the status of Level I listed substances along with steps taken towards release reductions. The baseline summary was produced to respond to questions CGLI received regarding the challenge baselines. It provides the information contained within the BNTS program document and cites supporting references or programs in a one-page table format. Copies of these materials are included as Appendix 2 to this report.

Awareness presentations:
In addition to the individual contacts made and described above, several presentations were made to industry trade association and ad hoc assembled groups of industrial personnel. Within these presentations, emphasis was placed on: fundamentals regarding the BNTS, the voluntary nature of the program, the opportunities presented, and need for industry data regarding substance release status.

Materials used during the presentation and as leave-behinds or handouts included: copies of overheads used during the presentations, the CGLI BNTS backgrounder, the CGLI BNTS success story reporting form and, the one page baseline summary.

Copies of the presentation overheads can be found in Appendix 3.

Website use:
CGLI has made copies of the awareness campaign backgrounder and success story request forms available on its website. The site at http://www.cgli.org includes a separate section on the Binational Toxics Strategy that includes the documents. Success stories that have been collected are also on the website. Copies of the home page and the web site version of the documents are included in Appendix 4.

Project Findings

Binational Toxics Strategy Awareness:
Outside of member companies, CGLI found a lower than anticipated level of awareness regarding the BNTS. Many of the organizations contacted had heard of the Strategy, but were unaware of the details, especially the challenge goals. The level of unawareness was greater for smaller companies than larger ones. CGLI is seeking additional ways to communicate with the small company sector.

Binational Toxics Strategy Interest:
Once the program was explained to those contacted, companies were interested. Many companies or organizations have already put programs in place which have met, exceeded or will exceed Strategy challenge goals (for their particular contribution). These have been described in the substance by substance section of this report. Many were included in CGLI's earlier submission to EPA's BNTS coordinating contractor, made in response to the recent request for "success stories."

Some of the organizations contacted were unsure of the applicability of the Strategy to their operations. Organizations operating on the border of the Great Lakes Basin, or those which do not have a record of release or association with Level I substances within the Basin, were uncertain of any role which they may play in BNTS implementation. CGLI's contact with these organizations has increased their interest level and is prompting participation where there was previously no or little interest.

All industrial organizations are busy complying with a myriad of environmental discharge and emissions control regulations, most involving significant capital expenditure requirements. Those rule or permit conditions that require actions that also advance or compliment Strategy goals produce the most interest in participation. Other actions for which additional programs may be needed to address a specific BNTS goal are often viewed as a lesser priority because of limited resources.

Without hesitation, industries contacted like the voluntary compliance aspect of the program. Some questioned the chemical listing and substance release challenge setting process. They have suggested that the BNTS is a "voluntary compliance program" rather than a totally voluntary program since the reduction targets and the substances have been predetermined.

A continuing need for a prioritized approach:
A repeated point made by those contacted was that while the Strategy is important because it seeks to address the virtual elimination challenge voluntarily, it must still be implemented on a risk based, prioritized basis. This means that the most significant sources and the sources that can be most readily addressed must come first. Organizations, which must respond to many regulatory challenges and environmental needs as identified above and may also wish to participate in the BNTS, will have to do so on a prioritized schedule. A compliance requirement trading program, similar to the Project XL concept, where Strategy participation could be substituted for other regulatory requirements would provide an important incentive for Strategy implementation.

Environmental management systems in place include voluntary programs:
The prioritized, continuous improvement approach is consistent with the pollution prevention and sound environmental management principles that have been adopted by all CGLI members, and most of the additional organizations contacted. Each has a formal environmental management program in place. Each has stated a commitment to respond to existing regulatory programs, and all have participated in other voluntary programs. Examples of these include:

• EPA's 33/50 program,
• CMA's Responsible Care( program,
• the American Automobile Manufacturer's pollution prevention program,
• the Wisconsin Paper Council's P3 program,
• Michigan's Mercury Pollution Prevention Task Force program,
• the American Forest & Paper Association's Environmental Principles and Goals,
• the Great Printers Program.

The requirements of other programs vs. the BNTS:
As mentioned above, industries reported that the specific requirments of other environmental obligations and programs impact their ability to participate in the Binational Toxics Strategy.

For example, many are participating in the several Maximum Available Control Technology (MACT) standards setting processes now underway. In these, technology standards for controlling air emissions from production equipment and processes are established through a source testing, equipment evaluation and public participation process. Most often, surrogate pollutants are used to represent mixed gas emissions produced by the sources to be regulated. Such is the case with coke oven emission regulation. One of the Strategy Level I substances is benzo(a)pyrene (B(a)P). During coke oven MACT evaluations B(a)P emission levels are not measured, total PAHs (polycyclic aromatic hydrocarbons) are. Because coke oven emissions monitoring work has focused on total PAH, rather than B(a)P, operators do not have the information necessary to easily respond to the B(a)P challenge. A solid connection between Strategy challenge goals and implementation of the MACT process is needed in order to make it more convenient for coke oven operators to make BNTS commitments.

The BNTS Level I Substances

Source review and status:
CGLI has sought to confirm suggested sources of Level I substances as compiled by the EPA/Environment Canada Workgroups. Parties contacted during the awareness campaign were asked to verify sources and update release status information. In large part, this process is still on-going. For many sectors, release information regarding the Level I materials is not readily available. Individual companies and sector technical organizations are searching for information.

In some cases the suggestion that a particular industry sector or manufacturing process may produce one or more of the listed substances has been based on anecdotal or inaccurate information. Example cases include:

• Octachlorostyrene releases from a number of chemical processes and,
• Octachlorostyrene/hexachlorobenzene releases from pulp and paper processes.

The OCS releases described in the Battelle OCS report are attributed to chemical processes cited in obscure references. Through some effort, these reference materials have been located and are in the process of being reviewed.

The suggested pulp and paper process effluent releases of OCS and HCB were originally raised in a MISA report. No other reference, in all of the available literature related to organochlorine releases from pulp and paper processes, has reported releases of these materials. Most significantly, the authors of the original MISA report concluded that the data upon which the release quantities were reported failed to meet QA/QC standards. Split samples tested by two of three laboratories involved in the sampling effort did not show the presence of OCB or HCB in any of the pulp mill samples. Based on this information, a review of process chemistry, and the lack of reported "hits" from any other analysis performed on effluent from this sector, pulp and paper mills are not sources of these Level I substances.

In other cases, programs are underway which have produced or will produced dramatic reductions in use and/or release of Level I substances. Details appear in the substance by substance discussion that follows, but examples include:

• Elimination of the use of mercury for the manufacture of household batteries. This industry, which was previously the largest user of elemental mercury, is no longer listed among sectors which use mercury for manufacturing.
• The chlor-alkali industry, which represents the largest industrial process mercury use in the U.S., has committed to a 50 percent reduction in mercury use by year 2005.
• Other mercury use reductions programs have been accomplished or are underway in the electric lamp, electrical switch, auto and telecommunications industries.
• Many manufacturing organizations have implemented programs to reduce or eliminate the use of mercury in process instrumentation.
• The use of PCB containing electrical equipment is being phased out at many industrial facilities. Electric power generation, auto, vinyl, telecommunications, and rubber manufacturing are examples of the wide range of industrial sectors where these use reductions have been accomplished or are underway.
• The pulp and paper industry has virtually eliminated dioxin discharges from pulp mill operations through pollution prevention programs which substitute chlorine dioxide for elemental chlorine in the bleaching process.
  

Substance by Substance Release Status Review

Details regarding CGLI's findings on substance releases appear in the individual chemical discussions follow.

Mercury

Sources
CGLI brought together key industry representatives to discuss mercury uses, releases, and progress made towards reductions in both. From the information obtained during this session, it was clear that what is known regarding uses and emission inventories for mercury is at a much more mature than for the other substances addressed by the Great Lakes Binational Toxics Strategy. This maturity is, no doubt, due to past focus on mercury and its compounds as a contaminant in the Great Lakes watershed. This focus is manifested by many fish consumption advisories issued regarding mercury content of fish and by the inclusion of mercury on the list of bioaccumulative chemicals of concern in the recent US Great Lakes specific regulation referred to as the Great Lakes Water Quality Guidance. Great Lakes states, notably Minnesota, have developed extensive regulatory programs regarding mercury, that surpass the requirements of federal regulations.

USEPA recently issued the comprehensive Mercury Study Report to Congress (EPA-452/R-97-003, December 1997) which reviews the Agency's current state of knowledge regarding mercury air emissions and uses. Mercury emissions to the atmosphere from various sources and processes are estimated, but it is acknowledged that many of these estimates are highly uncertain and that the source list itself may be inaccurate and incomplete. Nevertheless, this inventory represents a starting point for the evaluation of where to begin in developing prioritized action plans to reduce mercury emissions. An ongoing challenge continues to be the confirmation of the magnitude of these emissions from various processes and sources and the ability of technology to control them. It is important to identify all significant sources and refine emission estimates so that priority action can be focused on the most significant, and most readily controllable, sources if the goal of a 50% reduction is to be achieved by 2006.

The Mercury Study Report to Congress also identifies various uses of mercury in processes and products. There is somewhat less uncertainty in the manufacturing use of mercury due to an annual estimate prepared by the US Geological Survey, the Minerals Yearbook. This report contains consumption estimates for major users of mercury but many smaller uses are grouped as "other." In addition, several of the major consumption categories (e.g., wiring devices and switches) can lead to subsequent uses in many different industries and products.

The process of mercury emission to the atmosphere on a regional and global basis, with subsequent re-deposition to watersheds, is believed to be the major source of mercury in fish tissue. Human health impacts can arise from direct exposure (inhalation, dermal) to mercury itself or through ingestion of fish contaminated with mercury. Reducing exposures through these two pathways should be the ultimate goal of any mercury control strategy. These pathways should be kept in mind when devising action plans for mercury controls, either voluntary or regulatory.

A source of mercury to the Great Lakes that is not well quantified is the amount contained in wastewater discharges from both municipal and industrial point sources. EPA estimates are in the range of several hundred kilograms per year of mercury, or well below one ton per year. As such, these sources are considered minor compared to air deposition.

Of the Level I Binational Toxics Strategy substances, A significant issue that is unique to mercury is the fact that there are very significant natural sources which redistribute mercury within the environment. They include degassing of soils and rocks, volcanoes, geothermal vents and forest fires. These sources have been releasing mercury to the environment for, in many cases, millions of years. For example, the mercury contained in coal being used today came from mercury circulating in the atmosphere or found in water, hundreds of thousands or millions of years ago. There is much uncertainty regarding estimates of these natural sources, but on a global basis, their may be as much emitted from natural sources as from anthropogenic (human) activities.

In some isolated regions, there are cases of mercury concentrations of concern in fishes attributable to local geological conditions. The extent of the contribution of these "natural emissions" to mercury concentrations in fish vary widely, depending upon the influence of carious geological factors. However, in certain regions, it is likely that there would be fish advisories even if anthropogenic sources of mercury could be virtually eliminated.

Nationwide Air Emissions
The Technical Support Document attached to the Binational Strategy states that the baseline used for mercury emissions should be based on the most current inventory of sources at the time of the signing of the Binational Strategy in April 1997. The most recent inventory available in 1997 was described in the June 1996 review draft of the Mercury Study Report to Congress, conducted during the period 1990 to 1993. Alexis Cain, USEPA Mercury Project Manager, states in both a January 15, 1998 memorandum and in an August 4, 1998 draft update memorandum that the 1990-93 baseline is 221 tons. This emission inventory was recently corrected for errors in medical waste incinerator and municipal waste combustor emission estimates. Making use of these corrected figures, therefore, set the 2006 emission reduction goal at 110 tons. The Binational Strategy estimates that controls to be implemented on medical waste incinerators and municipal waste combustors will result in a reduction of 70 tons of emissions by 2002. By 1994/95, 59 of these 70 tons had already been realized, leaving an additional 11 tons to occur by 2002. An additional 40 tons reduction will be needed from other sources by 2006 to meet the 50 percent reduction goal. USEPA is developing rules for hazardous waste incinerators and cement kilns that burn hazardous waste. The largest remaining mercury emission sources, according to the Mercury Study Report to Congress, are utility boilers at about 51 short tons/year and commercial/industrial boilers at about 29 short tons/year.

Nationwide Deliberate Use of Mercury
The baseline for the mercury use reduction challenge is the most recent mercury use inventory available at the time the Binational Strategy was signed in April 1997. This was the USGS estimate of mercury consumption for calendar year 1995; 436 metric tons. The goal for 2006 is thus a mercury use of 218 metric tons. The 1997 estimate is 346 metric tons, a drop of 90 metric tons (mainly from the categories "measuring and control instruments" and "other uses"). With the chlorine industry commitment to reduce use by 80 tons by 2005, there remains an additional reduction of 48 metric tons needed by 2006 to reach the 50% goal (assuming there are no actual increases from other categories). Chlorine industry use is still expected to be the largest category at about 80 metric tons in 2006. Wiring devices and switches, and dental equipment and supplies are the remaining significant categories, with 1997 use levels of 57 and 40 metric tons, respectively.

Additional commitments regarding mercury use include:

• A new voluntary agreement regarding mercury use and release signed by three Indiana steel mills, the Lake Michigan Forum, U.S. EPA and the Indiana Department of Environmental Management. it will provide an inventory of mercury uses and releases at these plants and lead to reductions.
• Consumers Energy through an ongoing Mercury Pollution Prevention Initiative begun in 1996 has taken action to identify mercury sources within the company, estimate the total quantity of mercury used, review existing disposition practices and investigated future management options and costs. The program has heightened the awareness of mercury concerns in the company and identified options for use of non-mercury containing equipment. It has reduced the use of equipment containing mercury and associated stock inventory. 231 pounds of the substance were removed from use in 1996. 171 additional pounds in 1997.
• In support of the Binational Toxics Strategy, the Dow Chemical Company has set a goal for the company to reduce air and water emissions of mercury compounds by 75 percent by 2005.

Releases to Water in the Great Lakes Basin
In Alexis Cain's August 4, 1998 draft update memorandum, the estimated releases of mercury from point source discharges for 1993/94 was 269 kilograms (592 lbs. or about 0.3 ton). This is a very small input to the Great Lakes compared to the national air release estimate of 158 tons in 1994/951. Therefore, this input has not received direct attention from the Binational Strategy Mercury Workgroup. It should be noted that there are several State and/or POTW initiatives to reduce mercury discharges, some of which are related to existing NPDES permit conditions and some in anticipation of new permits to be issued when the Great Lakes Water Quality Guidance is fully implemented. The water release goal will not be further discussed in this report due to low relative loading and probable significant reduction under water discharge regulations currently being implemented.

Industrial Sector Review
In this section of the report, specific industrial use and emission sectors will be reviewed. Not all potential sectors have provided information for inclusion in this report. Opportunities to discuss use and emission reduction plans with other sectors are identified under the Future Directions section of this report.

Battery Manufacturing
In 1984/85, battery manufacturing accounted for about 55% of mercury consumption in the USA, by far the largest user of this material. By 1994, this sector had reduced the use of mercury in batteries by over 99.95%, from over 1000 metric tons/year to less than 0.5 ton/year in 1995. In the 1996 USGS report on mercury consumption, battery manufacture was not even mentioned as a separate category, since use had dropped to a small fraction of 1 ton. These reductions were accomplished through the impetus of legislative mandates spurred by the need to reduce mercury in municipal solid waste streams destined for incineration and the development of advanced battery technology by manufacturers.

Mercury batteries are still produced for specialized non-household applications in the medical, military and emergency response fields. Reductions in use are expected to continue for some period of time as the specialized equipment that requires mercury batteries reaches the end of its useful life and is replaced. During that time, battery manufacturers will provide users with information on the proper recycling and disposal of these batteries to prevent them from entering waste streams destined for incineration.

Mercury containing batteries will continue to make up an ever decreasing proportion of in the municipal waste streams as they are replaced by non-mercury containing batteries. This industry has certainly provided an excellent example of how new technology can provide an answer to a problematic environmental issue.

Fluorescent Lamp Manufacturing
Fluorescent lamps and other specialty lighting require the use of mercury as an essential ingredient for producing light. These lamps are much more energy efficient than traditional incandescent lighting and replacement with fluorescent lighting can actually decrease mercury emissions by avoiding the fossil fuel power generation that would otherwise be necessary. Lamp manufacturers have been aggressive in developing lamp technologies to reduce the amount of mercury needed per lamp from about 48 mg in 1985 to about 23 mg in 1994. The lamp industry expects to further reduce this value to below 12 mg per lamp by 2000.

Disposal of fluorescent lamps can lead to air releases if uncontrolled thermal treatment methods are employed. These lamps are being disposed at an increasing rate by reclamation (recycling) and by landfilling. The National Electrical Manufacturers Association (NEMA) has performed an environmental risk analysis that demonstrates the environmental soundness of these two waste disposal options. Only negligible amounts of mercury are being released to the environment. In addition, regulatory controls on mercury emissions have been finalized for municipal waste incinerators.

In summary, it appears that lamp manufacturers have taken steps to reduce mercury emissions from their products and have reduced mercury usage by over 50%, and are expected to provide another 50% reduction by 2000. These actions represent a proactive approach that fully meets the intent and requirements of the Binational Strategy. Remaining issues include how to reduce the number of spent lamps still being incinerated by encouraging recycling or proper disposal in well designed landfills.

Coal Fired Electric Power Industry
Because coal contains trace amounts of mercury and is used in large quantity as a fuel to generate electric power in many parts of the nation, air emissions from these generating plants constitute the single largest anthropogenic release of mercury to the atmosphere in the USA. EPA's Mercury Study Report to Congress estimates that about 52 tons of mercury per year are emitted by coal fired utility plants, accounting for about 1/3 of the total anthropogenic emissions of mercury in 1994/95 from USA sources (158 tons).

The Electric Power Research Institute (EPRI) has done extensive testing and determined that ESP/fabric filter and ESP/FGD systems provide variable but demonstrable coal fired boiler mercury emission reductions. In December, 1996, EPRI reported emission levels at 39 and 41 tons per year from the coal fired boiler sector (23% less than the EPA estimate) utilizing two different testing protocols.2

EPA has also recently released the Utility Air Toxics Report to Congress which concludes that mercury is the air toxic of greatest public health concern but that EPA has not been able to identify any currently demonstrated, feasible and commercially available technology for reducing mercury emissions from coal-fired utilities. Regulations have been deferred until further research can be conducted. EPA agreed that further research should include efforts in the areas of mercury fate and transport modeling, and assessment methods for health and environmental impacts, in addition to the assessment of new mercury control technologies.

Thermostat and Switches Manufacturing
Thermostats utilizing a mercury switch have proven to be very reliable and are in wide use in industry and households (e.g., furnace thermostats). The industry has begun a used mercury switch thermostat recycling program in several states, including six Great Lakes states, to encourage heating ventilating and air conditioning (HVAC) contractors to recycle.

The American Home Appliance Manufacturers Association has surveyed manufacturers and issued a bulletin #8 regarding known mercury uses, i.e. (1) gas pilot-light ranges/ovens, (2) chest freezers with lid lights, (3) fluorescent bulbs used in kitchen ranges and laundry appliances and (4) a few models of clothes washers made prior to 1972. AHMA will continue to gather information on the use of mercury in small appliances as well.

Mercury Cell Chlor-Alkali Manufacturing
The mercury cell segment of the chlor-alkali manufacturing industry is currently the single largest industrial user of mercury in the USA at 154 metric tons of consumption in 1995, or about 35 percent of the total utilized. Mercury is a necessary part of the manufacturing process, serving as a cathode in the electrolytic cell where salt (sodium or potassium chloride) solution is converted to chlorine gas, hydrogen gas and caustic (soda or potash, respectively). This industry is also the largest manufacturing (in contrast to sources from fossil fuel combustion) air emission source of mercury, according to the EPA Mercury Report to Congress, at about 7.1 tons per year in 1995.

The industry, through the Chlorine Institute, was one of the first to commit to a 50 percent reduction in use of mercury under the Binational Strategy. The baseline used was an average consumption of 160 metric tons for the period 1990-1995, so the reduction commitment is 80 tons reduction by the year 2005. Technical committees have been established in several areas to develop the procedures and technology necessary to meet this goal (see the first annual report of the Chlorine Institute on the Binational Strategy web site).

Portland Cement Manufacturing
This industry has been contacted about their interest in participating in the Binational Strategy program.

Secondary Mercury Production (Recycling)
Most of the mercury "production" in the USA is currently done through secondary mercury recovery and recycling. Several companies provide this reclamation service for mercury containing devices, lamps, etc. and the recovered mercury is purified and re-sold, thus closing the recycling loop. Reclamation and recycling is an important alternative to mercury disposal methods, such as incineration, that may lead to releases to the environment. However, mercury releases may also occur during these reclamation operations. For instance, NEMA estimates that about 3 percent of mercury destined for reclamation is emitted to the atmosphere, mostly through breakage of lamps during collection, storage and transport.

Future Direction and Opportunities

Priorities
To assure attainment of the 50 percent reduction goal by 2006, priorities must focus on those industrial sectors that may have the greatest opportunity for reductions. EPA's emission inventory does not capture all the potential sources of emission nor does it accurately estimate the releases from known sources. Nevertheless, it provides a good starting point for prioritization and investigation, even if it leads to new estimates and new priorities. If we assume the emission inventory from the Mercury Report to Congress as a starting point, and the 1990/93 inventory from the draft Mercury Study Report to Congress as the baseline, then there is a need to identify 48 tons per year in additional air emission reductions to meet the goal of 50 percent reduction (221 tons in 1992/93 to 110 tons in 2006, with a 1994/95 estimate of 158 tons). As noted in this report's mercury section introduction, a total of 70 tons reduction is expected from emission controls on medical waste incinerators and municipal waste combustors, leaving an additional 40 tons reduction needed from other sources. The sources identified in EPA's inventory with more than one ton emission in 1992/93 for which CGLI has not confirmed and reduction information has not yet been obtained are:

Assuming EPA rules will result in a 90% reduction of mercury emissions from hazardous waste incinerators, a 6 ton reduction will accrue, leaving about 34 tons of additional reduction required. All of the rest of the sources listed above add up to a total of 46 tons, so a 34 ton reduction from these sources would mean about a 74% reduction requirement. This does not seem feasible at this time.

Achievement of the 50% emission reduction goal will be a significant challenge. Of course, new sources may be identified and the actual emission estimates from some sectors may be adjusted as more information is acquired.

A more accurate characterization of the emissions from utility and commercial/industrial boilers would probably result in about a 35 ton reduction from the estimates used in the 1990/1993 baseline. There is an 11 ton difference in utility emissions as characterized by EPRI and the value used by EPA. This is due to the appropriate credit taken by EPRI for the removal of mercury from the emissions stream by scrubbers and electrostatic precipitators. There is probably also a 24 ton difference between actual emissions from industrial/commercial boilers and EPA's figure. These boilers sue 10 percent or less of the Nation's coal and generally utilize emissions control equipment similar to that used by utility boilers. Consequently, emissions from these units should be closer to about 10 percent of the utility boiler emissions, or about 5 tons.

It is important to note that this 35 ton reduction from EPA's 1990/1993 baseline does not represent an actual reduction that has occurred. (Although some small reduction from the industrial/commercial sector probably actually has occurred since these smaller units are more likely to have been replaced by natural gas boilers or other energy sources.) The 35 ton difference simply underscores the need to reevaluate the baseline, the goal and how the goal relates to Great Lakes Water Quality Agreement objectives. These type of discrepancies and questions regarding sources, baselines, and reduction goals would appear to be a priority item for resolution by the BTS team.

Priorities for Use Reductions
To assure the attainment of the 50% use reduction goal by 2006, priorities must focus on the sources that have the greatest opportunities. According to the USGS report, the 1997 uses of mercury are (in metric tons):

These totals compare to the 2006 goal of 218 metric tons (50% reduction). There appears to have been a reduction of 90 metric tons already, leaving an additional 128 tons to reach the target.

The chlor-alkali industry has pledged a reduction of 80 tons by 2005, thus leaving a need for 48 more tons to be identified and achieved. Since a 57 ton reduction (61%) has already been achieved for "other", it is unlikely that significant additional contribution from these 'unknown' sources can be relied upon. Electric lighting has achieved significant reductions in the recent past and the industry expects to reduce mercury content per light by about 50% from 1994 levels by the year 2000. There has been a 32% drop in the mercury used for wiring devices and switches, and it may be reasonable to assume that that sector will achieve a 50% or greater reduction by 2006 due to the emphasis placed on this use by federal and state governments. Assuming a possible 75% reduction over baseline for switches, then we can expect perhaps 35 tons more reduction, leaving about 13 tons to be identified. Measuring instruments have achieved about a 44% reduction to date, while dental use has held steady due to the superior properties of dental amalgam fillings.

Opportunities
Other opportunities exist in both emission reduction and use reduction. They need to be further explored and projects developed, if they are likely to provide measurable improvements by 2006. Potential projects identified by industrial representatives are discussed briefly below:

Orphan or unused mercury metal and compounds can be identified and recovered from almost any industrial/commercial/laboratory establishment, especially in older equipment. Detroit Edison did a complete inventory of their facilities and found that the mercury discovered far exceeded their stack emissions. This mercury inventory in place in many older facilities may be a source of uncontrolled emissions if improperly disposed, spilled or otherwise released to the environment. A project could be developed to make industries aware of the problem and encourage them to do their own mercury inventories. Case studies could be developed, and technical guidance on what type of equipment may utilize mercury would be needed. One industry representative identified the fact that mercury was the working fluid in filter rate controllers commonly used in drinking water plants built in the 1950s, for instance. Surplus or unusable quantities identified and earmarked as surplus could then be collected and recycled.

Barriers
Two possible barriers to further mercury emission and use reductions are identified below. These are meant to be a starting point for further discussion:

Octachlorostyrene

Sources
Unlike mercury, information regarding octachlorostyrene (OCS) releases is scarce. EPA recognized this situation and commissioned a work plan document, produced by Battelle Memorial Institute. This draft document suggested several OCS sources, but investigators could not confirm releases, quantify suggested releases, or establish ecosystem risks associated with OCS presence.

The CGLI Work Plan
CGLI has used the Canada-Ontario Agreement (COA) report and the draft Battelle work plan has a starting point for its source investigations. Information from the work plan and other references have been assembled into a "Work Sheet."

This Work Sheet is being used to guide our efforts to find information about the suggested sources. Under the column "lead organization," we have indicated whether CGLI or another organization is best suited to pursue information searches. In the few cases where we have been unable to determine what the best source of information is, "????" indicate that we are continuing to search for the information source.

The column headed "Data (yes/no)" provides and indication of the existence of known data. This data is being reviewed to determine suggested source status and significance. Other columns provide space for tracking and summarizing the information as it is collected.

The column headed "Responsible Party" provides an opportunity to track personnel responsibilities. This information is intended for internal use only and not reproduced in this report.

Literature References
CGLI has located all of the references which were relied on by Battelle to derive their list of OCS sources. These are being reviewed to determine relevance and accuracy and locate additional sources of information or knowledgeable persons. This review is to be completed by October 31, 1998.

Release Status Conclusions
CGLI has been able to reach conclusions regarding two OCS sources.

Graphitic electrode process chlor-alkali plants:
CGLI has contacted chlor-alkali producers to determine if graphitic electrode process plants are in operation within the Great Lakes Basin. None are. Additionally, only one such plant may be in operation within the U.S. The status of the Southwest region plant is being determined. Graphite electrode processes are "old technology." In addition, the process is more costly. This cost penalty for producing chlorine and caustic soda from this process will serve as a strong disincentive for its use.

Pulp and Paper Mills:
CGLI has contacted representatives of the pulp and paper industry regarding the suggested release of OCS [and Hexachlorobenzene (HCB)] from pulp and paper making processes. Pulp and paper mills are NOT thought to produce OCS or HCB.

Details are provided in the attached letter3 (Appendix 5) It explains that the only indication of OCS or HCB release was contained in a single report from the Ontario Ministry of Energy and Environment, which reported MISA monitoring data. Upon investigation, NCASI learned that ìthe preliminary data upon which the estimates are based were rejected from the MISA database after QA/QC review found sufficient irregularity in them to make it very doubtful that the mills were actually emitting the compounds."

In addition, NCASI consulted with pulp and paper chemistry specialist Dr. Douglas Reeve of the University of Toronto. Dr. Reeve examined the chemical reactants and conditions necessary to produce OCS and HCB and compared them with pulp and paper making process conditions. He concluded that pulp and paper making processes "have never favored formation of highly chlorinated substances. The use of high or complete substitution of elemental chlorine with chlorine dioxide reduces the tendency even further." OCS and HCB are, of course, very highly chlorinated materials. Conditions do not favor their formation in pulp mills.

PCBs

Sources:
CGLI has confirmed that PCBs in the U.S. industrial sector are largely present as in-service electrical equipment. Storage of used, or decommissioned PCB equipment is not a widespread practice.

Equipment Phase Out
Many industrial facilities are working to, or have already, phased out PCB containing equipment. Examples include:

• Goodyear Tire and Rubber Company, which produces tires, synthetic rubber and ground rubber products, began a project in 1993 to eliminate PCB transformers. To date, 165 transformers have been eliminated in the United States including 27 in the Great Lakes. As a result of the project, 15 plants in the U.S. and two plants in Canada are PCB-free. PCBs are being removed from additional plants based on risk evaluation.
• A PCB elimination program was undertaken at Chrysler Corporation's North American U.S., Canadian and Mexican facilities. The company has eliminated all 500 PCB transformers and all but 50 of 10,000 capacitors to date. All Chrysler facilities will be PCB free by the end of 1998.
• Ford Motor Company's global phase-out of PCB containing transformers continues and has been targeted for completion by 2010.
• Niagara Mohawk Power Corporation, an electric and gas utility, began replacing or retrofilling all high level PCB equipment in 1986. The company has reduced the number of PCB transformers from approximately 649 to 3. The remaining three will be addressed by 1999. In addition, the company has reduced the number of PCB capacitors from approximately 29,700 to 0. Through its action, Niagara Mohawk has surpassed the Binational Toxics Strategy challenge of a 90 percent reduction of high level PCBs used in electrical equipment. Niagara Mohawk Power Corporation is committed to the virtual elimination of the use of PCBs in its service territory and has achieved the U.S. PCB challenge for Level I substances.
• The Detroit Edison Co., a wholly owned subsidiary of DTE Energy, removed and replaced all PCB capacitors to which the public had access by 1984. This effort began in 1978, was completed 4 years before legislation required it and included the removal of more than 19,000 capacitors. Going beyond publicly accessible equipment, Detroit Edison agreed to eliminate all of its PCB capacitors in a 10 year Voluntary Phasedown Program that will remove and replace PCB capacitors at its substations. There were 18,000 capacitors identified for removal at the beginning of this 10 year voluntary program in 1995. Today, there are 8000 substation capacitors remaining to be removed by the targeted completion date of year 2005.

Other organizations are tracking equipment status and retrofilling units to remove contaminated fluids. For example:

• Bell Atlantic, a provider of telecommunications services, has PCBs that are contained in electrical transformers bought from others. 56 electrical transformers at company facilities were found to have PCB concentrations above regulatory action levels. The transformers were previously retrofilled with a silicon-based dielectric fluid that became contaminated with PCBs that leaked from the windings. Bell Atlantic contracted with a company to perform a process whereby they replace the contaminated dielectric fluid containing PCBs with a new fluid. The process enables the company to replace the dielectric fluid, not the transformer units. Over the past three years, PCB levels in the transformer have been significantly reduced. PCBs are safely collected and incinerated at a hazardous waste incinerator. The project is expected to be completed in 1999.

Canadian examples show how similar activities outside of the U.S. will benefit the Great Lakes Basin.

• Inco Limited, a mining company, has recently completed its PCB clean up at its Port Colborne, Ontario, nickel refinery. The refinery was started in 1919. The clean up, started in the mid-1980's.
  • Progress includes:
    • Destroyed 45,280 L of low level liquid PCB's 50-1900 ppm (sodium/salt process).
    • Destroyed 16,600L of low level liquid PCB's 2-115 ppm (sodium/salt process).
    • Destroyed 6356 L of low level liquid PCB's 115 ppm (sodium/salt process).
  • Since 1995:
    • Capacitors, ballasts and PCB clean-up debris, with a net weight of 11,900 kg PCB were destroyed in January 1996.
• Inco's Port Colborne Refinery PCB storage site was declared an historical site by the Ontario Ministry of Environment.
• Ontario Hydro, headquartered in Toronto, Canada, is one of the largest utilities in North America in terms of installed generating capacity. Using 1994 as a baseline, Ontario Hydro had a total of approximately 7,700 metric tonnes of both in-service and in-storage PCB materials. To date, approximately 1,900 metric tonnes of PCB wastes or 24.7% of the total inventory has been destroyed. Destruction numbers have been limited by delays at the destruction site. The company target is to destroy approximately 81% of the total PCB inventory by the end of 2005 and to be PCB free by the end of 2015.

On a larger industry sector scale, CGLI is working with several trade associations to obtain more detailed information regarding PCB equipment use and status.

Dioxin
CGLI has contacted industry sectors most familiar with dioxin release issues. The current status of releases from these sectors is summarized below.

Sources:

Combustion sources
The majority of suggested sources of dioxin in the industrial sector involve combustion processes. Understanding of dioxin formation dynamics in combustion, gained over the last several years, has led to very significant release reductions. In addition, the specific dioxin congeners produced are typically those exhibiting the least toxicity, relative to TCDD.

Additional controls are on the way. Many, if not most, of these sources have, or are undergoing, MACT emissions standards setting processes. The result will be further, significant, reductions.

Manufacturing processes
Some manufacturing processes have been associated with the production of dioxins as unwanted by-products. Process changes have led to very dramatic reductions in and even virtual elimination of dioxins from these sources.

Pulp and Paper Industry
Since 1988, the pulp and paper industry has dramatically reduced the generation and release of 2,3,7,8 TCDD/TCDF from pulp bleaching operations. 1996 data show that none of the five U.S. mills located within the Great Lakes Basin reported effluent 2,3,7,8 TCDD or 2,3,7,8 TCDF concentrations above the U.S. EPA minimum level of 10 parts per quadrillion (ppq); i.e., these mills have virtually eliminated these Binational Toxics Strategy Level 1 compounds from their effluents.

2,3,7,8 TCDD/TCDF releases from the other process vectors (wastewater treatment plant sludges and product pulp) have been similarly reduced. At Basin mills, 2,3,7,8 TCDD levels in both sludge and pulp are not measurable at EPA minimum levels of 1 part per trillion (ppt). 2,3,7,8 TCDF was detected in sludge at two of the Basin mills; but, sludge test results calculated on a combined 2,3,7,8 TCDD/ TCDF TEQ basis, are still below 1 ppt, a value equal to the 2,3,7,8 TCDD minimum level. 2,3,7,8 TCDF levels in pulp were measurable at only one Basin mill. Again, when calculated on the 2,3,7,8 TCDD/TCDF TEQ basis, the pulp values are below 1 ppt.

Mills within the Basin match or exceed percent release reductions made nationally by the industry's more than 100 mills. The magnitude of these reductions are demonstrated by the figures shown in the table below.

It must be noted that these reductions are calculated by assuming that, when results below minimum levels occur (as was the case for all but a very few of the 1996 data points), 2,3,7,8 TCDD and 2,3,78 TCDF are present at quantities equal to one-half the test detection level. In fact, the actual level could be much lower.

Chemical Industry
A Canadian example shows how chemical industry process changes have produced similar results in the chemical industry. The Vinyl Chloride Monomer Plant at Fort Saskatchewan developed and successfully operated a pilot plant for 95 per cent removal of trace levels of dioxins and furans. Full scale implementation is schedule in1998.

Vinyl Industry
In 1994, the Vinyl Institute (VI), a business unit of The Society of the Plastics Industry, Inc. committed to cooperate with the U.S. Environmental Protection Agency (EPA) in the Agency's efforts to better characterize the ethylene dichloride (EDC), vinyl chloride monomer (VCM), and polyvinylchloride (PVC) manufacturing industry as a potential source of releases of polychlorinated dibenzo-p-dioxins (PCDD) and polychlorinated dibenzofurans (PCDF). That year, as part of this commitment, the VI launched a voluntary comprehensive multi-year program intended to quantify potential PCDD/PCDF releases to the environment from the EDC/VCM and PVC production processes.

The VI program involves identifying point sources to be characterized, developing sampling and testing protocols for the accurate measurement of PCDD/F at possible release points, obtaining representative samples and testing, and evaluation and reporting of data. Materials or environmental releases having the potential to contain PCDD/F are tested and quantified at the point at which they leave the control of EDC/VCM/PVC manufacturing sites. This includes PVC resins, EDC, wastewater, intermediate streams (e.g., commercial HCl) that are sold as chemical feedstocks, and other streams and products leaving the EDC/VCM/PVC manufacturing process. "In-process" streams and streams that are otherwise regulated are outside the scope of the program. To aid in the review of the methods and assumptions used in this program, the VI enlisted an independent third-party technical expert panel comprised of Dr. Heidelore Fiedler (University of Bayreuth), Gary Amendola (Amendola Engineering, Inc.), Patrick Dyke (PD Consulting), and Professor Michael Gross (Washington University).

To date, the VI has sampled and analyzed pipe, bottle, packaging, and dispersion PVC resins; sales EDC; treated wastewater effluent; wastewater treatment plant solids; and releases to air from stack gas emissions and emissions from PVC dryers. Data show that the estimated annual releases from these streams in the U.S. are approximately 24 to 66 grams (worst case) of PCDD/F (TEQ). Since nearly all of the 12.1 - 27.5 gms of wastewater treatment solids are disposed in landfills and therefore not considered released into the environment, total environmental estimated releases of PCDD/F (TEQ) are 11.9 - 38.5 gms.

In "Phase 2" of its dioxin characterization program, the VI plans to characterize potential PCDD/F contributions from aqueous hydrochloric acid, spent catalyst, fugitive emissions, chlorinated solvents, maintenance waste, furnace coke, incinerator ash, and acid brick.

Chlor-alkali Industry
In 1995, the Chlorine Chemistry Council (CCC), a business council of the Chemical Manufacturers Association, initiated a similar program to collect samples and to examine potential PCDD/F wastewater releases from chlor-alkali processes. The program first focused on characterization of wastewater from U.S. and Canadian chloralkali manufacturing facilities for the presence of PCDD/F. Under this phase of the program, six companies submitted 15 wastewater samples from 12 chloralkali manufacturing sites. Eleven of the sites were located in the U.S.; one was located in Canada. In terms of chlorine production, the 11 U.S. sites manufacture approximately 4.4 million of the approximately 13 million short tons per year of chlorine manufactured in the U.S. annually. CCC estimated that the U.S. chloralkali manufacturing industry releases approximately 6.4 grams or 8.8 grams PCDD/F from wastewater per year on a TEQ basis, assuming ND=0 and ND=DL/2, respectively. CCC's program is now examining potential PCDD/F emissions from stacks, wastewater solids, and cell renewal solids.

EPA estimates that in 1995, U.S. dioxin releases to air, land, and water from all known sources totaled approximately 3,000 grams.4 Approximately one percent of this total is from streams sampled to date under the VI and CCC programs.

Benzo(a)pyrene

Sources:
A challenge regarding benzo(a)pyrene [B(a)P] releases from industrial sources is their measurement. Few data points exist. Measurements typically involve determinations of total PAHs or other surrogates. Many efforts have gone forward to control and reduce PAH materials.

Relative to 1992 emission rates, coke oven emissions from steel producers have been substantially reduced. CGLI contacted representatives of industries using these and other combustion processes, suggested as B(a)P sources. We will continue to seek information regarding the nature and status of emissions from these sources.

Hexachlorobenzene

Sources:
In researching hexachlorobenzene sources, CGLI found that a significant amount of work had been done on global sources.5 Robert Bailey, author of the Global HCB emissions report, is currently characterizing these relative to the Great Lakes Region. His conclusions, so far, are included in the following discussion.

Long range transport
HCB is one of those chemicals which has been observed to migrate to all parts of the earth in spite of its use and release primarily in the northern hemisphere. Atmospheric concentrations of HCB are relatively uniform in the northern hemisphere at 60 to 200 pg/m3. Its atmospheric half-life of about a year is long enough, and its vapor pressure is in the right range, to make this possible. Thus, emissions of HCB anywhere in the northern hemisphere will affect the concentrations of HCB over the Great Lakes, and thus its concentration in the water and biota. Modeling suggests that the majority of the HCB in the Great Lakes atmosphere comes from outside the region6.

Environmental analyses around the Great Lakes over the past 30 years have shown substantial and continuing decreases of HCB concentrations. The recent publication of results from the Integrated Atmospheric Deposition Network indicate decreasing atmospheric HCB concentrations at sites on the Great Lakes with half-lives of 3.3 to 12.4 years in the mid 1990s. Measurements of HCB concentrations in the water of the Great Lakes shows the upper lakes are in approximate equilibrium with the atmospheric HCB concentrations. Thus, the water concentrations (and resulting biota concentrations) are not expected to decrease appreciably until the atmospheric concentration decreases.

Product use
Hexachlorobenzene (HCB) was used as a seed fungicide in the USA and Canada through the 1950s and 1960s. The last registrations were canceled in 1985. However, HCB was used as a fungicide in Mexico up until 1991. Currently, other uses of HCB, military pyrotechnics and use as a chemical intermediate, are believed to be relatively minor in terms of quantity and release to the North American environment.

Process releases
HCB can be produced as a trace byproduct in many reactions where chlorine and carbon are present. Thus, it has been produced as a waste product from the production of many chlorinated chemicals. In the USA and Canada, this waste is either incinerated or placed in secure underground facilities. None of the facilities listed in the US Toxics Release Inventory which release HCB are located in the Great Lakes region.

HCB has been present as a microcontaminant in some chlorinated pesticides. Manufacturers have made process and raw material changes which have dramatically reduced or eliminated HCB concentrations. EPA regulates the maximum concentration of HCB possibly present in the pesticides as shown in Table 1. Current manufacturing methods produce materials with average HCB concentrations well under those listed in Table 1.

For example, atrazine producers confirm that HCB levels in their product are nearly always less than detection limits. In the attached letter,7 (Appendix 5) Novartis Crop Protection Senior Regulatory Manager Tom Parshley, explains that ìthe average HCB level for a typical production campaign is < 1 ppm.î ì[T]he maximum hexachlorobenzene content found in atrazine technical was 9.7 ppmÖ.î Therefore, regulatory levels should not be used to calculate HCB release quantities from pesticide use.

HCB has also been potentially present as a microcontaminant at 2 ppb or less in some chlorinated solvents. About 300,000,000 kg of solvents were used in 1992 in North America. Thus less than 1 kg of HCB could be released annually from the use of chlorinated solvents in North America.

Auxiliary activity releases
Small amounts of HCB are released from many processes where both carbon and chlorine are present at high temperature. Thus, nearly all combustion processes and some metal processes emit small amounts of HCB. The measurement of very low emissions is difficult and emissions vary widely depending on many factors so that it is not possible to develop precise emission estimates at present. Production of these small quantities of HCB is not easily avoided since both carbon and chlorine are natural components of the environment.

Metals
The degassing of molten aluminum prior to casting by addition of hexachloroethane has been reported to release small amounts of HCB. Few aluminum furnaces currently use this process in the Great Lakes region. The smelting and refining of secondary copper has also been reported to possibly release HCB. CGLI is seeking information on the status of these releases.

Combustion processes
The emission of HCB from municipal waste incineration has been reported. The emission factors determined in laboratory experiments and by stack monitoring vary widely. HCB has also been detected in emissions from the combustion of coal and hazardous waste.

Pulp and Paper Mills
CGLI has contacted representatives of the pulp and paper industry regarding the suggested release of HCB (and OCS) from pulp and paper making processes. Pulp and paper mills are NOT thought to produce HCB or OCS.

Details are provided in the attached letter8 (Appendix 5) It explains that the only indication of HCB or OCS release was contained in a single report from the Ontario Ministry of Energy and Environment, which reported MISA monitoring data. Upon investigation, NCASI learned that ìthe preliminary data upon which the estimates are based were rejected from the MISA database after QA/QC review found sufficient irregularity in them to make it very doubtful that the mills were actually emitting the compounds."

In addition, NCASI consulted with pulp and paper chemistry specialist Dr. Douglas Reeve of the University of Toronto. Dr. Reeve examined the chemical reactants and conditions necessary to produce HCB and OCS and compared them with pulp and paper making process conditions. He concluded that pulp and paper making processes "have never favored formation of highly chlorinated substances. The use of high or complete substitution of elemental chlorine with chlorine dioxide reduces the tendency even further." HCB and OCS are, of course, very highly chlorinated materials. Conditions do not favor their formation in pulp mills. In addition, the Cohen report5 concluded that, "[t]here are no reliable data that indicate that hexachlorobenzene is generated at and/or emitted from pulp and paper mills."

Reductions, by Sector

Progress So Far
HCB is no longer used in agriculture in North America. This has removed the largest single source of HCB in the environment. Over the past years, the concentrations of HCB in pesticides have dropped steadily. For example, the maximum HCB concentration in pentachloronitrobenzene (PCNB) was reported to be 27,000 ppm in a 1971 study, and 10,000 ppm in 1976. The HCB concentration had dropped to 5000 ppm by 1983, 1000 ppm by 1988 and is presently less than 500 ppm in PCNB.

The past improvements in combustion practices, designed to reduce the emission of all the incomplete combustion products have also reduced HCB emissions because of similarities in formation mechanisms. However, the database of HCB emission factors is not complete enough to quantify the emission reductions.

As explained in the previous discussion regarding OCS sources, pulp and paper processes do not release HCB as has been suggested by one data source. That information, produced as part of a Canadian MISA monitoring report, was found to not meet QA/QC requirements and has been removed from the MISA data base. (Appendix 5)

Additional initiatives have been initiated in the chemical industry. The Dow Chemical Company has set a goal for the Company to reduce air and water emissions of hexachlorobenzene by 75 percent by 2005.

Barriers to Further Reductions
Long range air transport of HCB is the largest source of HCB into the Great Lakes Basin. This fact potentially diminishes industry's incentive to promote regional initiatives, including new processes and other changes which may pose other risks to the environment or financial risks to the company.

Alkyl-lead

Sources:
CGLI has contacted representatives of the chemical, petroleum, automobile, and lead industries to determine use of alkyl-lead compounds.

Internal combustion engine fuels
Of course, lead is no longer used in consumer automotive fuels. This use ceased in the U.S. in 1995. Remaining use are:

• Aviation gasoline, and
• NASCAR racing fuel.

The prohibition against the use of lead or lead additives in automotive fuels was published in the Federal Register on February 2, 1996. 40 CFR 80.22 (a) prohibits lead based additives as of December 31, 1995. See letter and enclosures attached 9 (Appendix 5).

Leaded gasoline has not been banned for aircraft use because of the need to provide sufficient anti-knock properties to the high compression/high performance engines needed to balance the weight vs. power demands in this application. The use of alternative, no-lead, fuels continues to be explored by aircraft industry and government personnel. CGLI is attempting to gather use information from fuel producers.

Racing fuels are a specialty application for which international standards must apply. Use estimates suggest that emissions of lead from this application are about 20 percent of those generated by piston powered aviation engines.

Remaining uses of leaded gasoline may be insignificant since EPA has determined that "[l]ead air pollution levels measured near the Nation's roadways have decreased 97 percent between 1976 and 1995….10î

Chemical manufacture
It has been suggested that some chemical process intermediates, and organo-metal production processes produce or make use of alkyl-lead compounds. CGLI has vigorously pursued information from chemical producers but has not been able to confirm this suggestion. This leads to the conclusion that the practice, or use/release of alkyl-lead in this fashion, is not common place. We are continuing to search for contacts within this specialty sector to obtain answers to these questions.

Pesticides

Sources:
CGLI has contacted the American Crop Protection Association regarding the Level I pesticides. This organization is not aware of industrial sources for the banned substances.

Chlordane had been produced, for export, by Velsicol Chemical Corporation in Memphis TN. On May 23, 1997 Velsicol announced that it would cease production of the material and sell all remaining stock by the end of 1997 (See Appendix 5).

CGLI continues to search for information regarding the export, storage, or use as chemical intermediates of Level I pesticide materials.

CGLI - Implementing the BNTS 9/30/98