PROCEEDINGS OF THE EXPERT PANEL WORKSHOP
TO EVALUATE THE PUBLIC HEALTH IMPLICATIONS
FOR THE TREATMENT AND DISPOSAL OF POLYCHLORINATED BIPHENYLS-
CONTAMINATED WASTE

Chapter 3 - Expert Panel Report
PCB Incineration Panel

Richard S. Magee, Sc.D., P.E., DEE - CHAIR
Harvey W. Rogers, M.S. - RAPPORTEUR
Betty C. Willis, M.S. - CO-CHAIR

Chapter 3 Table of Contents

Executive Summary
Introduction
Panel Discussions
I. Incinerator Operations and Performance
A. Waste Feeds
B. Combustion Conditions
C. Stack Emissions
D. Residuals Management
II. Facility Issues
A. Siting of Incinerators
B. Transportation of Wastes and Residuals
C. Fugitive Emissions
D. System Safeguards
E. Testing and Monitoring
F. Maintaining Performance
G. Training of Operators
H. Public Input
III. Factors Affecting Public Health
A. Identification of Parameters Critical to Analysis
B. Air Pathway
C. Soil Pathway
D. Water Pathway
E. Food Chain Pathway
F. Identification of Data Gaps
G. Dealing with Uncertainties
IV. Summary Statement of Each Panelist
Adel Sarofin
William Farland
Don Oberacker
Kathryn Kelly
Pat Costner
Curtis Travis
Robert Ginsburg
Andrew Trenholm
Harvey Rogers
Richard Magee
Betty Willis
V. Recommendations
A. Recommendations to ATSDR on Health Issues
B. Recommendations Regarding Data Needs
Appendix A Incineration Panel Biosketches
Appendix B Abbreviations
Appendix C Discussion Document

EXECUTIVE SUMMARY

SUMMARY OF OBSERVATIONS

Many of the Incineration Panel's discussions focused on what is known and not known about incineration and the implications of this incomplete database. The majority of the panelists felt that enough is known to conduct a scientifically defensible public health assessment of this technology. However, some panelists felt that not enough is known about incinerator emissions and the health effects of the emissions to determine the health effects of incineration. Almost all panelists agreed that, at each site, alternative technologies for waste treatment should be reviewed and should be used in place of incineration when an alternative technology could better treat the waste or when the public health impact could be shown to be less. Other suggestions were these: 1) always choose an alternative treatment technology; 2) consider capping and leaving the PCB contamination in place.

The Panel had widely varying views on the reuse of incinerator residuals. Some panelists believed the current database of knowledge does not support the conclusion that incinerator residuals can be safely used in manufactured products. Other panelists supported the maximum reuse of residuals, provided they pass the requirements of the Environmental Protection Agency (EPA) land disposal restrictions, leachability tests, and other tests to determine if the waste-derived product is safe and appropriate for the proposed use.

Regarding incineration, the major points determined by the panel were these:

1. Careful selection and design of facility components (combustor and air pollution control equipment [APCE]) and operator training are critical to ensure proper operation and to minimize environmental and health impacts.

2. A homogeneous waste feed is desired for optimum system operation--physical homogeneity may be more important than chemical homogeneity. Preprocessing and blending of wastes may be needed.

3. Metals, particularly mercury, should be kept out of incinerator waste feeds to the extent possible. One panelist felt that all metals should be kept out of all incinerator waste feeds.

4. The Panel was not aware of any examples of incinerators burning PCB-contaminated soils with municipal solid wastes and sewage sludge. Most Panel members were skeptical (to varying degrees) about the appropriateness of burning that particular combination of waste feeds because of the engineering control problems that this heterogeneous waste feed would cause. Most panelists felt that if incineration of such a combination of wastes is considered, the combination should be tested in a pilot- scale incinerator of similar design, or, if possible, a full-size unit before construction of the proposed facility. The panelists were doubtful that stable operating conditions and adequate PCB destruction could be achieved with this combination of waste feeds.

5. Testing (trial burns and periodic retesting) could be better targeted to provide more relevant information for health impact assessments. If special prepared feeds are used to test worst-case incineration conditions during a trial burn, the facility should also be tested during incineration of actual waste feeds.

6. Fugitive emissions are not frequently measured or accounted for when considering the health impacts of a facility. Fugitive emissions may be independent of the treatment technology used because they generally come from the waste-handling and processing areas.

7. Operator training and inspection and maintenance of the entire facility, as well as calibration of monitoring equipment, are critical to maintaining good incinerator performance.

8. The public should have input into the technology selection and oversight of the facility operation. Some panelists also felt that the affected community should have the right to refuse any technology.

9. The ultimate tests for determining the potential for public health impacts of a facility are biomonitoring and environmental sampling. Most panelists believed that better identification and quantification of incinerator emissions, as well as biomonitoring and environmental sampling in communities around incinerators, needs to be done to determine if incinerators cause adverse health effects.

10. Some panelists felt the best available pollution control technologies should be used to reduce emissions to the maximum extent practicable, even if it is not required by current EPA regulations.

11. Organic products of incomplete combustion (PICs) are inherent to all combustion devices and are not necessarily related to the waste or fuel being burned. Total organic emissions from hazardous waste and PCB incinerators are generally less than 20 parts per million (ppm) of the stack gases, and metal emissions are also in the low ppm or micrograms per cubic meter (µg/m³) range. However, another Panel member commented that some of the constituents of greatest concern that have been identified, such as polyhalogenated dioxin and dioxin-like chemicals, are present in the parts per trillion (ppt) range in stack gases, and, even though they are present in stack gases at extremely low concentrations, an incinerator's cumulative loadings to the environment of these and other persistent, bioaccumulative PICs may have both subtle and long-term impacts on public health and the environment.

12. Polyhalogenated dioxins and furans are believed to be formed when halogens are present in the waste feed; however, these compounds are not always detected in stack gases with current analytical techniques. The quantity and isomers formed are more a function of incinerator design and operation than waste feed; however, the availability of trace quantities of catalytic metals (e.g., copper) and chlorine in the waste feed also affect their production.

13. Emissions of PICs and metals and public exposure to incinerator emissions should not be allowed to be greater for short-duration incinerator operations, such as when mobile incinerators are used at sites being remediated.

14. Quality control is important throughout all aspects of incinerator operation, from waste feed analysis and stack emission monitoring to calibration of process monitors and recording devices.

INTRODUCTION

The expert Panel on PCB incineration was assembled by ATSDR to provide the current scientific knowledge on the incineration of PCB-contaminated wastes currently found in the United States. The Panel was asked to discuss the application of incineration to a variety of waste streams that may be contaminated with PCBs, such as sewage sludge; municipal solid waste; contaminated debris; and liquids, soils, and hazardous wastes that may be found at Superfund sites. The Panel members were not asked to evaluate a particular site or situation, but rather to share their knowledge of the efficacy and public health and environmental impacts of the incineration of similar wastes around the world. The purpose was to provide input that ATSDR can use in its evaluation of the public health implications of incineration facilities (fixed and mobile) across the nation. The Panel discussed incineration of a combination of waste streams, as well as incineration of a single waste stream.

The 11-member Panel consisted of two academicians, one representative of environmental groups, one consulting engineer, three consultants on risk and health assessments, two EPA employees (a combustion engineer and a risk assessment specialist), an incineration specialist from the Centers for Disease Control and Prevention (CDC), and ATSDR's incineration specialist. Because of the diversity of the backgrounds of the panelists, there was no attempt to reach consensus or to identify if consensus was reached. However, there was a general level of agreement on a variety of the issues discussed. The statements in this report represent issues that were discussed and opinions that were expressed by Panel members. The list of panelists and a brief biosketch of each member are provided in Appendix A. Appendix B is a list of abbreviations used in this report.

Before the panel meetings began, a document was developed (see Appendix C) that described the topics for discussion and listed a series of issues or questions under each topic to help focus discussions. The Panel met for seven sessions, each about 2 hours long, on September 13 and 14, 1993, in Bloomington, Indiana. The first two sessions covered incinerator operation and performance issues, such as waste feeds, combustion conditions, stack emissions, and residuals. The next two sessions covered incinerator facility issues, such as siting, transportation, fugitive emissions, system safeguards, testing, monitoring, maintaining performance, training, and public input. The fifth and sixth sessions covered topics related to evaluation of the health impacts of incineration, such as the air, soil, water, and food chain pathways of public exposure, and to identification of data gaps and needs. During the final session, the panel discussed approaches for dealing with uncertainties; then each panel member was given 5 minutes to present a summary statement.

This Incineration Panel Report follows the same format as the discussion document (see Appendix C) used by the Panel. Each section begins with the issues described in the discussion document (open bullets). Whereas the original document listed the questions for discussion, this report summarizes the key points (closed bullets) made by the panelists. Several panelists requested that minor clarifications be made to some of the "open bullet" items copied from the original discussion document. Those changes were made in this report but not in the original discussion document. Section IV is a transcription of each panelist's summary statement. Section V of this report summarizes the major conclusions and observations of the panel. Section VI lists the Incineration Panel's recommendations.

PANEL DISCUSSIONS

This Panel report was prepared to reflect the full spectrum of viewpoints presented by summarizing the panelists' key points, observations, concerns, and recommendations in a topical, bullet format. It is important to point out that the panel meetings were not a consensus seeking process, and ATSDR did not seek to obtain consensus on any issue. Likewise, the points listed in this Panel report are not necessarily the opinion of all panelists. When there was disagreement on a major issue, we have tried to point that out. The listing of discussion points in this report is not intended in any way to imply that all, or even a majority, of the panelists endorsed the points. The full Incineration Panel was given the opportunity to review two drafts of this Panel report. The panelists' comments were addressed before the report was made final.

Discussions in a particular session sometimes covered a wide range of topics, and some points were made that were more relevant to the topic of another session. When that happened, the points made were summarized under the heading that was most appropriate, rather than under the heading corresponding to the session in which the point was made. Consequently, this report does not always reflect the order in which comments were made, i.e., directly correlate with the sequence of discussions in the videotapes of the Panel sessions.

I. INCINERATOR OPERATIONS AND PERFORMANCE

A. WASTE FEEDS

• PCB contaminated wastes exist in a variety of physical and chemical forms, and may contain many other chemical and physical constituents in addition to the PCB compounds.

DISCUSSION

• Incinerators can handle a variety of waste feed types; however, waste feed characteristics (e.g., sludges, fumes, liquids, solids, chlorine, metals, moisture, etc.) drive the actual design of the incinerator combustion system and APCE.

• Homogeneity is desired for optimum system operation--blending and preprocessing of waste may be necessary.

• Physical homogeneity may be more important than chemical homogeneity.

• The Panel was not aware of any examples of incinerators burning PCB-contaminated soils with municipal solid wastes and sewage sludge. Most Panel members were skeptical (to varying degrees) about the appropriateness of burning that particular combination of waste feeds because of the engineering control problems that this heterogeneous waste feed would cause. Opinions ranged from the comment that such a combination should not be allowed at all, to reservations that a single incinerator could effectively treat that combination of wastes. Most panelists felt that if such a combination of wastes is considered, the combination should be tested in a pilot-scale incinerator of similar design, or, preferably, a full-size unit, if its permits would allow it, to determine if stable operating conditions and adequate PCB destruction could be achieved, before construction of such a proposed facility.

• A mix of organic and inorganic compounds in the waste may pose conflicting challenges for an incinerator system.

• Generally, PICs don't correlate well with the constituents in the waste feed. That is not true, however, for some constituents, such as metals and chlorinated materials, and for target compounds with high concentrations that are native or added to the waste for determining the destruction and removal efficiency (DRE). For example, benzene shows up as a PIC no matter what is burned, e.g., tobacco, gasoline, or hazardous waste; if chlorinated materials are present in the waste, dioxin and furan congeners are usually produced.

• Metals, particularly mercury, should be kept out of incinerator waste feeds to the extent possible.

• Waste feed analysis is important to ensure that the incinerator is being operated within its permit limits for waste constituents and characteristics, and that emissions do not exceed regulatory limits. Even though hazardous waste and PCB incineration facilities are required to analyze some of their waste feeds and residues, all wastes received and residues generated are not fully characterized, i.e., a full detailed analysis is not conducted. Generally, a detailed analysis is conducted on the initial shipment of the waste and periodically thereafter. Screening analyses are conducted on all other shipments. Residuals analyses are generally dictated by the EPA "Land Disposal Restrictions" analytical requirements. Most panelists believed that there is a general need for more testing, but they did not believe a detailed analysis of ALL waste feeds and residuals is necessary to properly manage these materials. Others believe the existing regulations are inadequate, and that ALL waste feeds and residuals should be fully characterized.

• If the waste feed proposed to be incinerated is unique and no prior data exist for the incineration of this waste feed, the Panel recommended that pilot-scale tests (or full- scale tests, if a unit is available) be conducted to determine the feasibility of burning this waste and to optimize design and operating parameters.

B. COMBUSTION CONDITIONS

• Extensive test data have shown that combustion conditions, e.g., the three "T's" of combustion: temperature, residence time, and turbulence (mixing), have a significant impact on organic and inorganic emissions.

DISCUSSION

• The relationships of time and temperature to incinerator stack emissions are fairly well understood, but mixing and wall effects (physical parameters) are less understood and are believed to be the major cause of PICs found in incinerator stack emissions.

• Laboratory-scale and pilot-scale studies should be used for evaluation of new waste streams to set operational controls, and for preliminary health effects estimates; however, data from full-scale operating units are preferred for site-specific health effects determinations.

C. STACK EMISSIONS

• EPA regulations require performance testing of all PCB incinerators to ensure that they meet the requirements of 99.9999% PCB destruction and removal efficiency and 99.9% combustion efficiency. For liquid PCB incinerators, the regulations require a residence time of 2 seconds at 1200ºC ± 100ºC at 3% excess oxygen (O₂) or 1.5 seconds at 1600ºC ± 100ºC at 2% excess O₂. Wastes that contain 250 ppm or more of PCBs must be incinerated. Wastes that contain PCBs in the 50-250 ppm range can be incinerated, buried in a chemical waste landfill, or burned in a qualified industrial boiler. EPA regulations also require that PCB incinerators be tested for the following emissions: hydrogen chloride (HCl), particulate, O₂, carbon monoxide (CO), and carbon dioxide (CO₂). For more information on EPA's PCB regulations, see 40 CFR 761.70.

DISCUSSION

• Organic PICs are emitted from all combustion devices and are not necessarily related to the waste being burned, e.g., polycyclic aromatic hydrocarbons (PAHs), dioxins, benzene, etc., are commonly found in emissions from wood-burning stoves, automobiles, cigarettes, coal-fired power plants, municipal solid waste incinerators, and medical waste incinerators. PIC emissions from these combustion sources are generally greater than the PIC emissions from hazardous waste or PCB incinerators.

• Total organic emissions from hazardous waste and PCB incinerators are generally less than 20 ppm of the stack gases. Metal emissions are also in the low ppm or µg/m³ range. The balance of incinerator stack gases are known and are not usually of public health concern. During the panel meetings, statements were made that implied that during incinerator test burns, up to 70% of the components of the small concentration of organic PICs are identified. The following clarification was provided after the panel meetings by one of the panelists involved in the EPA PIC studies.

  In recent hazardous waste and PCB incinerator permitting trial burns, the stack emissions are typically analyzed for 10 metals, the principal organic hazardous constituents (POHCs), and either designated PICs (e.g., benzene, dioxins, furans, etc.) or the top 10 PICs, based on mass of PICs present. One EPA PIC study conducted in December 1986 at a Dow Chemical Company hazardous waste incinerator in Plaquemine, Louisiana, was designed to identify all the organic compounds present in the stack emissions that were analytically possible from a hazardous waste incinerator under normal operating conditions and under upset conditions (Trenholm AR and R Thurnau, Total Mass Emissions from a Hazardous Waste Incinerator, Proceedings of the Thirteenth Annual Research Symposium, EPA/600/9-87/015. Cincinatti, Ohio, July 1987). Upsets were forced to occur in the incinerator every 30 minutes during the "upset test runs." The laboratory identified 70% of the mass of the PICs present.

  In another PIC study conducted in July and August 1988 at the Mobay Chemical Company hazardous waste incinerator in Kansas City, Missouri, stack emissions were analyzed for classes of organic compounds under normal and upset conditions. The induced upset conditions during this test were even more dramatic then during the Dow PIC study. In the Mobay tests, approximately 80% of the PICs were identified as compounds with 1 to 7 carbons that were either gaseous or condensable hydrocarbons. Earlier data from the Dow test indicate that most of the low-molecular-weight compounds are oxygenated aliphatic hydrocarbons (e.g., alcohols, aldehydes, ketones, and organic acids). Compounds in these classes are generally considered to be low in toxicity and not persistent in the environment.

  Eight additional PIC tests were conducted for EPA in which the laboratory identified all the volatile and semivolatile organic compounds that could be identified when the stack samples were analyzed by gas chromatograph/mass spectrometer (GC/MS). These methods identify polyhalogenated dioxins and furans, PAHs, and other known persistent or bioaccumulative compounds. In the reports for these eight tests, the low-toxicity, short-chain aliphatic hydrocarbons, such as methane, were not reported, even though those constituents likely made up the bulk of the mass of the PICs. Approximately 30 individual volatile and semivolatile organic compounds were identified. During the eight test burns, stack emissions were continuously monitored using a hydrocarbon monitor commonly referred to as a "total hydrocarbon" (THC) monitor. The hydrocarbon readings were always less than 10 ppm. One can conclude from these PIC studies that the aggregate concentration of all "unknown" constituents in incinerator stack emissions is generally in the low ppm range, and that individual "unknown" constituents likely have concentrations in the parts per billion (ppb) to ppt range or lower.

• Some panelists pointed out that even though the unknown portion of the stack emissions may be in the "low" ppm range, the cumulative loadings to the environment of polyhalogenated dioxins and other persistent, bioaccumulative PICs may have subtle and long-term impacts on public health and the environment.

• Dioxin and furan emissions are more a function of incinerator design and operation (i.e., combustion efficiency, temperature profile through the equipment, etc.) than waste feed composition; however, the concentration and availability of chlorinated materials and trace catalytic metals, such as copper, also affect their production.

• Dioxin emissions from incinerators contribute to the global dioxin pollution level. However, some panelists believe that municipal waste incinerators, hazardous waste, and PCB incinerators are minor contributors to that global background level.

• Metals emissions are a function of waste feed composition, temperature profile, fly-ash/bottom-ash partitioning, facility design, APCE, etc.

• Mercury emissions have been difficult to control, but carbon injection may be emerging as an effective technology. Nevertheless, the panel recommended that mercury be kept out of an incinerator's waste feed to the extent possible.

• The speciation of trivalent chromium (Cr⁺³) and hexavalent chromium (Cr⁺⁶) in stack emissions is highly variable; therefore, some panelists recommended assuming that 100% is hexavalent if the speciation is not specifically determined in the stack emissions.

• Irritation from acid gases is a common complaint from communities when chlorinated compounds are fed to an incinerator that does not have an acid gas removal system. HCl emissions are easily scrubbed. Sulfur dioxide is not typically found in PCB incinerator emissions; it is usually only a concern if coal or high-sulfur oil are burned.

• The best available control technologies for particulate matter can remove particulates to 0.01 grains per dry standard cubic foot or fewer; therefore, several panelists recommended that all incinerators be required to meet that level, even if it is below the level required by applicable regulations. Those panelists felt that particulate control was critical to controlling metal emissions and organic contaminated particulates, such as dioxins, to the maximum extent practicable, to reduce the public's exposure.

• Stack emissions should be screened for persistent, bioaccumulative compounds.

• Several panelists pointed out that the EPA hazardous waste and PCB incinerator regulations do not require that metals or other persistent pollutants be analyzed for in stack emissions, or that permit conditions address these chemicals. However, other panelists pointed out that, in recent years, some EPA permit writers have been using their "omnibus authority" to try to address these pollutants. The panelists acknowledged that in trial burns a full detailed analysis of the stack emissions is not done, and that in recent years, the facility typically analyzes for predesignated organic compounds (principal organic hazardous constituents [POHCs]) and metals, plus the next 10 highest volume organic compounds in the stack gases.

D. RESIDUALS MANAGEMENT

• Incineration facilities generate a number of waste streams, e.g., residuals such as bottom-ash, fly-ash, and scrubber residues, that require proper management to ensure protection of public health.

DISCUSSION

• Incinerators create residual wastes that require management. The panel recommended that the residuals, i.e., fly-ash, bottom-ash, scrubber residue, be kept separate until they are characterized and disposal options are evaluated.

• Generally, the potential for leachability and bioavailability of heavy metals is the major concern with residuals; however, leachability and bioavailability of organic compounds also needs to be considered.

• The Panel had widely varying views on the reuse of incinerator residuals. Some panelists believe that the current database does not support the conclusion that incinerator residuals can be safely used in manufactured products, e.g., aggregate for road beds or building blocks, concrete, etc. Other panelists supported the maximum reuse of residuals, provided the products pass leachability and other tests to demonstrate environmental and public health suitability.

• Releases of wet scrubber effluents to publicly owned treatment works (POTWs) may contribute contaminants, e.g., heavy metals, to POTW sludges.

II. FACILITY ISSUES

A. SITING OF INCINERATORS

• In the United States, the federal government has regulatory responsibility for permitting hazardous waste and PCB incinerators. The siting regulations pertain to siting of these incinerators when flood plains and active seismic areas are potential sites. Local governments have the authority to determine which land in their jurisdiction will be zoned for industrial use (such as incinerator sites) and residential use. Most states have developed additional siting criteria for hazardous waste incinerators.

DISCUSSION

• Siting is a local issue; at a minimum, the local community should have input into the decision as to whether or not incinerator facilities are located in their community. Several panelists believe that the local community should have the authority to decide whether an incinerator is built in the community.

• Exposure should not be allowed to be higher (i.e., emission limits should not be compromised) for short-duration incinerator operations, such as when mobile incinerators are used.

• Site containment and isolation (capping) may have less public health impact than excavation and remediation, and should be evaluated in addition to other proposed treatment and remediation technologies, such as incineration.

• The Panel recommended that ATSDR catalog the siting criteria used by states. Approximately 40 states are believed to have hazardous waste incinerator siting criteria.

B. TRANSPORTATION OF WASTES AND RESIDUALS

• The U.S. Department of Transportation (DOT) regulates companies and vehicles that transport hazardous materials in interstate commerce. The transport of hazardous wastes, which is regulated as a subset of hazardous materials, is regulated stringently. Most state DOTs have similar regulations for companies that haul materials only within the state (intrastate commerce). There are limited regulations on routing of waste shipments.

DISCUSSION

• The accident database for shipments to incinerators is sparse, but may be quantifiable. Accidents involving hazardous materials or hazardous waste releases must be reported to the U.S. Department of Transportation and EPA; however, the database may not include information on the shipment destination (i.e., shipments to incinerators).

• The risk to the public depends on the physical and chemical characteristics of the wastes being shipped, the transportation route, vehicle design and maintenance, driver training, etc.

• Transportation accidents can be a significant contributor to human health impacts.

• Public health impacts from transportation accidents can be reduced by rapid spill response or more on-site treatment of wastes.

C. FUGITIVE EMISSIONS

• Fugitive emissions at the incineration facility can occur during the transfer, processing, and storage of wastes. These emissions may exceed the emissions from the incinerator stack.

DISCUSSION

• Fugitive emissions depend on facility design and operation. Waste transfer and processing should be done in an enclosure under negative pressure.

• Fugitive emissions are inherent to all technologies, not just incinerators. Because fugitive emissions are typically released near ground level, they may be more of a health concern than stack emissions for workers and nearby residents.

• During excavation of Superfund sites, fugitive emissions are potentially a major health concern and may require special design and operating considerations if residences are close to the site.

• Ambient air monitoring at the fenceline and on-site monitoring in work areas is recommended to ensure that workers and the community are protected from fugitive emissions.

• Several panelists recommended medical monitoring of incinerator workers, including establishing background values before workers begin employment at the incinerator.

  (NOTE: Regulations by OSHA and EPA already require safety training and medical monitoring for all employees who work at Superfund or hazardous waste sites or respond to emergencies. See 29 CFR 1920, 29 CFR 1926, 29 CFR 1960, SARA Section 126 [f], and 40 CFR 311.)

D. SYSTEM SAFEGUARDS

• Permits for PCB and hazardous waste incinerators normally require that a number of incineration operating parameters be continuously monitored and connected to an automatic waste feed cut-off system that will shut off the waste feed to the incinerator if operating conditions are out of the permit range.

• Most incinerators that treat wastes have emergency relief vents or "dump stacks" immediately after the combustion chamber(s) through which the hot combustion gases can be diverted if downstream equipment malfunctions. Emergency relief vents are necessary to prevent equipment fires and venting of the hot combustion gases at ground level, which could be a real hazard to facility workers and nearby residents. When the emergency vent allows the flue gases to bypass the APCE, the public could be exposed to higher concentrations of metals, particulates, acid gases, and organic chemicals.

DISCUSSION

• Incinerators are analogous to industrial operations and are subject to the same potential for upsets, e.g., spills, fires, explosions, etc.

• Impact on the community from incinerator upsets needs to be evaluated; few data are available for evaluation.

• Process conditions, e.g., kiln seal leaks and dumpstack releases, need to be considered when evaluating the public health impacts of incinerator facilities.

• Having process operating conditions interlocked with the waste feed is important to protect equipment and the public.

• One panelist was unaware of any incidents in which there were off-site impacts from facility upset conditions or failure of system safeguards; therefore, that panelist concluded that the systems appear to work as intended. Other panelists pointed out that the fact that no health studies have been conducted at incineration facilities does not mean that there are no off-site impacts. Several panelists named three or four incinerator facilities where there have been explosions, i.e., where "the system" did not appear to work as intended. It was reported that during the explosion at the Chemical Waste Management incinerator facility in Chicago a cloud of smoke migrated off-site which indicates that there may have been off-site impacts as a result of that incident.

E. TESTING AND MONITORING

• EPA requires trial burn tests for all PCB and hazardous waste incinerators when they first begin incinerating PCB or hazardous wastes, and periodically thereafter, to ensure compliance with EPA regulations. Although PCB and hazardous waste regulations do not require it, EPA has also required measurement of other stack emissions such as metals, PICs, dioxins, and furans at many of these incinerators using its omnibus authority.

• Continuous emissions monitors (CEMs) are available for O₂, CO, CO₂, hydrocarbons, NO_x, HCl, SO₂, and opacity, but not all of the monitors are required on every incinerator.

• EPA has conducted a number of research tests on various types of incinerators to more fully characterize the emissions than is normally done during trial burns.

• Not yet available is a CEM that can analyze stack emissions to determine all of the potential chemicals and their concentrations in the stack gases of incinerators.

DISCUSSION

• A number of CEMs are now available for stack emissions monitoring for certain constituents; they should be included in incinerator designs where appropriate.

• A number of new CEMs are under development for metals and specific organic species.

• Continuous monitors for process parameters should be used.

• Quality assurance of all monitoring instrumentation should be conducted by an independent auditor. Facility personnel should routinely calibrate monitors.

• Testing of hazardous waste incinerator stacks during trial burns and research tests has shown that the total concentration of organic compounds in the stack is generally less than 20 ppm; metal emissions are also in the low ppm range. The balance of incinerator stack gases are known and are not usually of public health concern. During incinerator test burns, up to 70% of the components of the small concentration (<20 ppm) of organic PICs have been identified. The aggregate concentration of all "unknown" constituents in incinerator stack emissions is in the low ppm range. Individual "unknown" constituents likely have concentrations in the ppb range or lower. Some panelists pointed out that even though the unknown portion of the stack emissions may be in the "low" ppm range (a subjective term), when looking at long-term health effects, it is the total loading of a particular chemical from an incinerator that is important, especially for chemicals such as dioxins and metals that either persist or bioaccumulate.

• Polynuclear aromatic hydrocarbons (PAHs) are more significant contributors to risk from municipal waste and sewage sludge incinerators than from hazardous waste or PCB incinerators. The total national loading of PAH emissions comes predominantly from automobiles, home heating furnaces, and wood stoves.

• Trial burns are conducted to demonstrate the "worst-case" scenario or range of operating conditions, e.g., ash content, chlorine content, lowest temperature, etc., that the facility wants in its permit. Facilities generally blend a special waste feed for the trial burn. There was considerable discussion as to whether trial burns are conducted under "worst-case" or "best-case" conditions because facility staff and equipment are prepared in advance for the testing and are probably therefore operating at peak performance. This may offset the worst-case operating and waste feed conditions required during the trial burn. During a trial burn, the conditions are varied from the optimum conditions to provide a reasonable operating window in which the emissions are acceptably low.

• Several panelists believe that during trial burns, all facilities should also have to burn real wastes.

• A number of panelists believe that, in addition to trial burns, more incinerator testing is needed on a periodic basis during normal operations. One suggestion was to monitor stack emissions once a day for selected PICs and metals.

• Trial burns typically do not reflect incinerator upset conditions. Testing is needed during upset conditions to determine the impact these events may have on the public.

• More ambient air monitoring in the community is needed at the perimeter of the facility or in the community, or both, during operation to assess the public health impacts of the facility.

• Emissions and facility operating data should be available by remote telemetry to the public and regulators, so they can follow operating conditions and stack emissions on a continuing basis.

• The Panel recommended that risk or health assessors collaborate with trial burn designers to ensure that the compounds targeted to be measured during the trial burn are those needed for assessing the impact of incineration on public health.

• Stack emissions should be screened during trial burns for persistent and bioaccumulative compounds, for compounds of greatest inherent toxicity, and for compounds emitted in the largest volume.

• The Panel recommended that environmental sampling and biomonitoring be undertaken around one or two incineration facilities to better characterize the impacts of deposition of process and fugitive emissions.

• The Panel recommended that ATSDR conduct biomonitoring or body-burden studies of workers and the public (1) in a few communities where existing incinerators are located, and (2) in one or two communities before and after new incinerators are built to determine the facilities' impact on public health.

F. MAINTAINING PERFORMANCE

• A frequent criticism of incineration is that compliance testing (trial burns) occurs when the facility is operating optimally.

DISCUSSION

• Routine maintenance and inspections, as well as calibration of monitoring equipment, are critical to maintaining and ensuring good performance.

G. TRAINING OF OPERATORS

• Essential to the proper operation of a well-designed and well-maintained incineration facility is having well-trained operators on every shift.

DISCUSSION

• Incinerator operators should receive training, testing, and certification by an independent body.

• Incinerator operators should receive training in (1) combustion theory and parameters critical for good combustion, (2) regulations that apply to their facility, (3) potential health impacts on themselves and the community when the incinerator is operated poorly, and (4) site-specific operation and on-the-job-training by an experienced operator.

H. PUBLIC INPUT

• A frequent criticism of hazardous waste facilities in general is that the affected public has little, if any, input into the monitoring and operation of the facility.

DISCUSSION

• The public should have input into the technology selected to remediate a site.

• The public should have input into the location of incinerators.

• The public should be allowed to comment on the trial burn plan before it is final.

• The public should have access to the facility to ensure permit compliance.

• Funds should be made available to educate the public on facility oversight and to allow for meaningful participation.

• The Panel recommended that ATSDR assemble case studies of well-conceived and well-executed public involvement.

III. FACTORS AFFECTING PUBLIC HEALTH

A. IDENTIFICATION OF PARAMETERS CRITICAL TO ANALYSIS

• The major concern to most citizens is the potential for short- and long-term public health impact posed by incineration facilities.

DISCUSSION

• Adequate data are essential to determine the impacts of incineration facilities on public health. In general, there are data gaps in all three major areas of interest:

  • emission data (identification and quantification of emitted compounds)
  • toxicological data (dose/response relationships)
  • human exposure data (dispersion modeling of contaminants or actual environmental data)

• The Panel suggested that more interaction was needed between combustion engineers and health scientists to identify key compounds to target for stack testing.

• When a proposed incinerator is evaluated, the Panel recommended using data from similarly designed, full-scale incinerators. Because such a comprehensive database does not exist, the Panel recommended that ATSDR or EPA develop an incinerators database.

• Although actual environmental sampling data are favored over predictions from models, the panelists generally felt that modeling also was necessary because of difficulties in analytical detection limits in some environmental samples.

• Some panelists felt that risk and health assessments based on worst-case scenarios (i.e., the most exposed individual living for whole life in one spot and growing all food at that spot) are overly conservative and therefore mislead the public; they believe risk assessments should be reasonable, i.e., based on a "real individual" and a mix of worst-case and average assumptions. That is the guidance given in EPA's 1992 Exposure Assessment Guidelines. Two panelists believe that current risk assessment methodologies are inadequate and that risk assessments should not be conducted because they mislead the public.

B. AIR PATHWAY

• Inhalation of incinerator stack emissions and fugitive emissions that are dispersed from the facility is the most direct route of exposure of the public to hazardous constituents from the facility.

• EPA established air dispersion models that require use of the 5 most recently available years of meteorological data from the closest national weather station, or 1 year of meteorological data from an on-site weather station, and land use and elevations around the facility are commonly used to predict maximum ground-level concentrations of stack emissions. Various outputs can be obtained from the models, i.e., average hourly concentration, maximum 15-minute concentration, etc.

• Ambient air is not normally monitored in communities around incinerator facilities. CERCLA sites are more likely than permanent incinerator facilities to conduct fenceline monitoring.

• In an urban setting, other industrial and combustion sources, such as automobiles and power plants, contribute significant concentrations of many of the same chemicals as incineration facilities. Ambient air monitors detect the combined influence of all sources in the area.

DISCUSSION

• The Panel had confidence in dispersion modeling in flat terrains, but less confidence in the technique in complex terrains. Several panelists recommended that additional validation testing is needed for some of the models currently in use.

• The EPA dispersion models were recommended because they tend to be more readily accepted by the scientific, local, and regulatory communities.

• There was strong disagreement among the panelists about risk assessments of persistent and bioaccumulative chemicals, such as dioxins, and whether the air pathway from an incinerator was a minor contributor to the overall risk of the community. From models, but not actual data, the greatest risk to the community can be the increase in background concentrations of such chemicals already in the food chain.

C. SOIL PATHWAY

• Off-site soils can become contaminated by an incineration facility and other industrial and combustion sources in the area via airborne contaminants that are transported and deposited off site, surface water or leachate drainage off site, or spills. The public can be exposed to contaminated soil via dermal contact, direct ingestion of the soil, and inhalation of airborne dust or chemicals volatilized from the soil.

• Children typically ingest more soil while playing outside than adults do during activities such as sports, yard work, or gardening. Pica children (children who have a craving or habit of eating things that are not normal food items, such as dirt) may ingest as much as 5,000 milligrams per day (mg/day) of soil.

DISCUSSION

• Site-specific risk assessments have shown that, at some sites, contaminants in soil do not contribute much risk from direct ingestion; however, soil contamination is a mechanism for contaminants to enter the food chain. Soil contamination is also important because it can continue to be a source of long-term exposure to contaminants after air emissions are stopped.

• It was noted that background levels of some chemicals (both naturally occurring or caused by contamination in a particular geographic area) sometimes are at levels of health concern. However, it was also pointed out that the presence of "manmade chemicals" or "contaminants" in the environment does not automatically mean that adverse public health effects will result. The concentrations of contamination may be unlikely to pose a public health concern.

• Rather than depending on dispersion modeling to estimate deposition of contaminants, more environmental soil sampling is needed in communities around incinerators to verify actual concentrations. These data could be used to validate deposition models.

D. WATER PATHWAY

• Bodies of surface water near an incineration facility can become contaminated by deposition of airborne contaminants, surface water drainage, or direct discharge of liquid process effluents from incinerators and other industry in the area.

• The public can be exposed to contaminated bodies of water via ingestion of the water, dermal or ocular contact with the water, or inhalation of the water or the contaminants in the water.

DISCUSSION

• The major concern usually associated with contaminants in the water is their bioaccumulation in the food chain, e.g., fish, shellfish, etc.

• The impact of discharges from industrial facilities, such as incinerators, on POTW sludges needs to be evaluated.

E. FOOD CHAIN PATHWAY

• Airborne contaminants can be deposited on the foliage of vegetation near an incineration facility or contaminants can be taken up from contaminated soil by the root system. If portions of the plant are edible, the public may be exposed to contaminants via ingestion of the edible portions.

• Other industrial and combustion sources in the community may also emit many of the same chemicals as the incinerator, and in larger quantities.

• The public can also be exposed to contaminants by ingestion of animals or animal products that have been contaminated because the animals ate contaminated plants or soil or inhaled airborne emissions.

• Dermal contact with contaminated plants, animals, or products can be another route of public exposure.

• Some aquatic species bioaccumulate or bioconcentrate some contaminants in the water, such that they become more contaminated than the water. If these species are a part of the food chain, then the public can be exposed to higher concentrations of contaminants than would be expected based on analysis of the water.

• Risk assessments of hazardous waste incinerators conducted by several authors using modeling data found that secondary exposure to dioxins and furans through the food chain can be higher than exposure by the inhalation route. However, this exposure has not been verified through actual measurements.

• Studies of mother's milk have shown that infants can be exposed to low levels of PCBs, dioxins, and furans during breastfeeding. The expert Panel assembled by the World Health Organization to look at this issue concluded that the levels of PCBs, dioxins, and furans in mothers' milk in the general population are not high enough to outweigh the benefits of breastfeeding. They concluded that breastfeeding should be encouraged and promoted.

DISCUSSION

• People are exposed to emissions from all combustion sources worldwide via the food chain. This is a national and global issue. It is also usually a local issue, especially when there is farming in the area. Subsistence farming could be a major local concern.

• When conducting risk assessments of incinerators using computer models, the food chain typically accounts for more than 95% of the risk; however, that finding has not been validated by environmental and biological sampling. Because the higher risks of food chain exposure have not been verified through actual measurements, they may be an artifact of uncertainty in modeling rather than actual risk.

F. IDENTIFICATION OF DATA GAPS

• Because of extensive testing and research for more than a decade, more is known about incineration than any alternative technology. However, many questions have not been fully answered.

• It is not possible to know on a continuing basis the exact chemicals (and their concentrations) that are being emitted by any combustion source, including incinerators.

• Even during a trial burn, it is not possible to identify all of the chemicals (and their concentrations) that are being emitted by an incinerator.

• Fugitive emissions are not routinely identified and quantified at incineration facilities.

• The deposition rates to soil and water of all potential incinerator stack emissions are not well known.

• Data regarding the bioconcentration or bioaccumulation in animal and vegetable species of all potential incinerator stack emissions are limited.

• The health effects caused by exposures to various concentrations (i.e., toxicity) of all potential incinerator stack emissions are not well known.

DISCUSSION

In addition to the data gaps listed in the previous bullets and in other sections of this document, the panel identified the following data gaps:

• Emissions during incinerator process upsets are not identified and quantified.

• When stack emissions are analyzed for metals, the specific metal compounds or species present are not identified.

• Concentrations of contaminants in environmental samples taken from media around incinerator facilities, e.g., soil, water, and ambient air, are typically not measured.

• Biological monitoring studies in communities around incinerator sites have not been conducted. The Panel recommended that ATSDR conduct these types of studies at both well-operated, state-of-the-art facilities, and at poorly operated and non-state-of-the-art facilities.

  (NOTE: ATSDR has two such studies in progress--the Vertac site study in Jacksonville, Arkansas, and the University of North Carolina study of three North Carolina incinerators are in the data collection phase; the Caldwell Systems study in Lenoir, North Carolina has completed data collection.)

G. DEALING WITH UNCERTAINTIES

• Because there are a number of gaps in our full understanding of the impacts on public health of any remediation technology, including incineration, some people argue that we should not employ these technologies, and the wastes should be stored in concrete bunkers until a safe technology that won't affect public health is developed.

DISCUSSION

• Most of the Panel members felt enough was known to make scientifically defensible health risk evaluations of incinerators; however, the Panel felt more research is needed to determine whether the projections of long-term health effects or no health effects associated with the incineration of various waste streams are accurate. Some panelists recommended that new incineration facilities not be allowed until such studies are completed.

• Some panelists discussed the use of techniques, such as Monte Carlo probability analysis or sensitivity analysis, to help determine the relative importance of what we know about incinerator emissions and what we don't know, so research can be directed towards filling the data gaps that have the largest potential impact on public health.

• The ultimate tests for determining the potential for public health impacts of a facility are biomonitoring and environmental sampling. Most panelists believe that better identification and quantification of incinerator emissions, as well as biomonitoring and environmental sampling in communities around incinerators, needs to be conducted to determine if incinerators cause adverse health effects.

  (NOTE: ATSDR has completed a cross sectional symptom and disease prevalence study of residents living near an allegedly poorly designed and operated incinerator: Caldwell Systems, Inc., in Lenoir, North Carolina. The study showed residents living near the incinerator had a higher prevalence of respiratory and neurologic symptoms than residents who lived further away. ATSDR has several other studies in progress in which biological markers are being analyzed in communities living near incinerators.)

Next Section          Table of Contents