A Sophisticated Approach for Trial Burn Plan Development to Support a Post-Trial Burn Risk Assessment[1]

 

Jerry Drake and Elizabeth Drake

Compliance Strategies & Solutions, Inc.

Houston, TX  77058

 

John Jones, P.E.

Reilly Industries, Inc.

Indianapolis, IN  46241

 

ABSTRACT


Over the past few years, the United States Environmental Protection Agency has been increasing the use of Risk Assessments in the permitting process.  With this regulatory approach, facilities throughout the United States are being faced with the challenge of integrating Risk Assessment and Trial Burn results in order to obtain a Part B Permit.  Thus, the development of a Trial Burn Plan is more challenging and complicated owing to the fact that it now must support a post-trial burn Risk Assessment.  A goal for the Trial Burn Plan is to yield favorable operating limits while demonstrating acceptable risks and hazards for the facility.

 

Typically, a Trial Burn Plan includes a worst-case waste stream determination.  This determination can be quite simple for one combustion unit that burns one homogeneous waste stream from one continuous process, and quite complex for several combustion units that burn an accumulated waste stream generated from several production processes.  Due to the recent Risk Assessment requirements, the worst-case waste stream determination now includes risk considerations.  The worst-case waste stream determination is based on a facility’s production processes, a defined universe of compounds, and risk considerations.  For one facility, the risks associated with the universe of compounds was ranked based on current toxicity values and persistence in the environment.  The ranked listing was developed for five different toxicity categories in order to assign an overall risk to each compound.  The results were then used to determine the worst-case waste stream to be burned during the Trial Burn to generate the worst-case emissions.

 

In addition to the worst-case waste stream determination, various Agencies have developed their own specific requirements for the Risk Assessment that must be incorporated into the target compound list for the Trial Burn Plan.  The target compound list is the list of compounds that should be measured in the stack emissions from the combustion unit and it includes the chemicals of greatest concern for the Risk Assessment.  The development of a target compound list requires assessing the waste stream and the available analytical methods for each chemical of concern for the Risk Assessment.  The Trial Burn Plan and Quality Assurance Project Plan must include the appropriate sampling and analytical methodologies to ensure that scientifically defensible data is collected during the actual Trial Burn for use in the post-trial burn Risk Assessment.

 

A sophisticated approach for Trial Burn Plan development to support a post-trial burn Risk Assessment was developed for Reilly Industries, Inc.’s boilers located in EPA Region V. The worst-case waste stream was determined, a universe of compounds list was developed, and the compounds were ranked based on risks.  A specific target compound list was developed utilizing the list of chemicals, generator knowledge, and available sampling and analytical methodologies.  Both the worst-case waste stream determination and the target compound list were then integrated into the Trial Burn Plan, the Quality Assurance Project Plan, and the Risk Assessment Protocol.

 

INTRODUCTION


EPA Region V requested a RCRA Part B permit modification to include three hazardous waste burning boilers located at the Reilly Industries, Inc. facility in Indianapolis, Indiana in February 1998.  The three boilers associated with the permit modification are referred to as Boiler 28K, Boiler 30K, and Boiler 70K.  As a portion of the Part B permit modification, Trial Burn Plans and Quality Assurance Project Plans (QAPPs) were developed and submitted for EPA approval.  A Demonstration of Similarity was submitted and approved for Boiler 28K (1).  The EPA required the facility to establish limitations on operations pursuant to the requirements of 40 CFR Part 266 (Boiler and Industrial Furnace Regulations) and Part 270 (Part B Permit Regulations), and to demonstrate acceptable risk using their omnibus authority.

 

The production processes at the facility generate several hazardous waste streams, which are blended in tanks prior to being burned in the boilers.  Due to this practice, the Agency required the facility to develop a worst-case waste stream composition for use during the testing.  The testing combined the traditional Trial Burn with the data collection for a site-specific Human Health and Ecological Risk Assessment (hereafter, Risk Assessment).  Therefore, the composition of the worst-case waste stream was required to satisfy the needs of a Trial Burn and a Risk Assessment.  A methodology used to develop a worst-case waste stream that satisfied these diverse needs is presented in the following sections.

 

The EPA Region V presented the facility with a list of compounds that they requested to be incorporated into the Risk Assessment.  This list of compounds was used as a starting point for the development of a target compound list specific to the testing of the boilers.  The methodology used to develop the final target compound list is also discussed in the following sections.

 

Having documented the worst case waste stream, the facility and the EPA can demonstrate to the public that the hazardous waste used during the Trial Burn testing is representative of the highest risk constituents that the facility may burn.  This adds a level of conservatism to the Risk Assessment results, which should be viewed positively by the EPA and the public.  Coupled with a target compound list that is developed from the Agency’s list, the facility can produce a successful, technically sound, and sophisticated approach for developing the Trial Burn Plan.

 

Developing a Worst-Case Waste Stream


The specifications for a worst-case waste stream were developed in order to justify the waste stream composition used for the Trial Burn.  The goal of this determination was to justify to EPA Region V that the waste stream composition represented worst-case conditions for the Trial Burn.  Since the emission results from the Trial Burn were to be used in the Risk Assessment, this was a critical determination that needed to error on the side of conservatism in order to be protective of human health and the environment.

 

The Trial Burn consisted of the traditional Trial Burn testing parameters (i.e., destruction and removal efficiency, particulate matter emissions, etc.) and the Risk Burn testing parameters for conducting a Risk Assessment (i.e., dioxin/furans, volatile organics, etc.).  Therefore, the worst-case waste stream determination was required to simultaneously address both the Trial Burn parameters and the Risk Burn parameters.

 

The addition of the Risk Assessment to the testing objectives complicated the determination process because of the increase in the number of compounds that had to be considered.  In order to develop a worst-case waste stream meeting the Trial Burn and Risk Burn criteria, a process was developed using a universe of compounds, a ranking method for the compounds, information on the waste streams, and generator knowledge.

 

Universe of Compounds


First, a universe of compounds was determined in order to justify the composition of the worst-case waste stream.  Four sources were used to develop this universe of compounds as follows:


 

These sources were used to develop the universe of compounds for several reasons.  The Appendix VIII compounds relate directly to the universe of hazardous wastes, and are the constituents used to define most hazardous waste codes.  The Human Health Risk Assessment Protocol (HHRAP) is the most recent EPA guidance for conducting a Risk Assessment.  The Risk Assessment Emissions Guidance is the most recent EPA guidance used for determining the methods to employ in sampling stack emissions for a Risk Assessment.  The EPA Region V list is the list of compounds specified by the Agency as the target compounds for the Trial Burn and associated Risk Assessment.  The combination of these sources resulted in a universe of 779 compounds for evaluation.

 

Ranked List of Compounds


The universe of compounds were then ranked based on a scheme developed by the International Center for Toxicology and Medicine in conjunction with Compliance Strategies & Solutions, Inc. and Reilly Industries, Inc.  The ranking scheme was developed as a mechanism for assigning a relative degree of risk to each compound based on the compound’s toxicity (i.e., reference doses/concentrations, cancer slope factors, and unit risk factors) and persistence in the environment (i.e., octanol-water partition).  The results of this ranking method would then provide a mechanism to determine which components of the waste streams generated at the facility would be the worst-case compounds for the Risk Assessment.

 

For each of the compounds on the universe of compounds list, the following values were obtained from toxicology databases, including EPA’s Integrated Risk Information System (IRIS) (5) and the EPA’s Health Effects Assessment Summary Tables (HEAST) (6):


 

For the non-carcinogenic compounds, the ratio between the measure of persistence (Kow) and toxicity (RfD and RfC) was used to rank the compounds.  For the carcinogenic compounds, the multiplication of the measure of persistence (Kow) and toxicity (Oral CSF, Inhalation CSF, and Inhalation URF) was used to rank the compounds.  This mechanism resulted in five lists of ranked compounds based on the five categories of toxicity.

 

The toxicological databases reviewed did not include values for each of the parameters specified above.  In fact, it was a rare occurrence that all of the values were obtained for a single compound.  For those compounds that were missing data on a specific parameter, a ranking within that category could not be determined.  Since these compounds were not ranked, they could not be used in the determination of the worst-case waste stream.  This was considered valid because it is believed that those compounds that present the largest risk to human health and the environment have the largest amount of data available on their risk.  Due to the unavailability of data, the universe of compounds was significantly reduced.  The non-carcinogenic ranking of the RfD resulted in a list of 131 ranked compounds.  The non-carcinogenic ranking of the RfC resulted in a list of 128 ranked compounds.  The carcinogenic ranking of the Oral CSF resulted in a list of 73 compounds.  The carcinogenic ranking of the Inhalation CSF resulted in a list of 70 compounds.  The carcinogenic ranking of the Inhalation URF resulted in a list of 71 compounds.

 

The assessment of the universe of 779 compounds may seem excessive for simple well defined hazardous waste streams.  However, it does serve to demonstrate which compounds may present a relative risk whether they are found in the hazardous waste feed stream or in the emissions as products of combustion.  The information may prove beneficial for the facility in discussions concerning the uncertainty of the Risk Assessment and results from total organic emissions testing.  Starting with a complete and thorough evaluation also helps the facility to justify other criteria (i.e. the target compound list) to the EPA and the public.

 

Facility Waste Streams


Once the universe of compounds had been identified and evaluated based on regulatory and scientific information, the list can be narrowed to a facility specific basis.  The Reilly Industries, Inc. facility is divided into several production plants, of which six produce hazardous waste streams that are burned for energy recovery.  Each of these six production plants was evaluated to determine the compounds present in the production processes.  This evaluation included those compounds that are used as raw materials, produced as production intermediates, or generated as products and by-products from the production processes.  A summary list of compounds for the six plants was generated that represented the compounds that would be expected to be present in the waste streams.

 

Each of the compounds expected to be present in the waste streams was then assigned the ranking value obtained from each of the five risk based categories.  Those compounds that did not contain a ranking for any of the five ranking categories were eliminated from the worst-case waste stream determination since there simply was not any data available for making a fair comparison.  Table I presents the results of this determination.

 

Table I

Facility Compound Rankings

 

Compound

Generating

Plant(s)

Kow/RfD

Rank

Kow/RfC

Rank

Oral CSF

*Kow Rank

Inhal. CSF

*Kow Rank

Inhal. URF

*Kow Rank

Acetaldehyde

27

114

113

73

70

71

Acetonitrile

27

117

117

 

 

 

Benzene

27, 40, 41

83

84

56

53

56

Ethylbenzene

26, 40, 47, 48

95

 

 

 

 

Formaldehyde

27 and 40

122

123

68

64

65

Methanol

26 and 48

128

 

 

 

 

Pyridine

26, 27, 41, 47, 48

89

91

 

 

 

Toluene

26, 27, 40, 47, 48

93

93

 

 

 

o-,m-,p-Xylene

40

103/104/106

101/102/104

 

 

 

 

The generation rate of the hazardous waste streams from each of the six production plants was also evaluated for a two-year period.  This was done in order to determine if any constituents were more or less likely to be present in the waste stream.  Table II summarizes the generation rates for each of the plants.

 

Table II

Waste Generation Rate Summary

Generating

Plant

Gallons

Generated

Percentage

Of Total

26

326,659

6.6

27

3,141,226

63.6

40

655,448

13.3

41

78,409

1.6

47

499,881

10.1

48

235,041

4.8

Total

4,936,664

100.0

 

Using the list of compounds presented in Table I, the plants associated with each compound, and the waste generation rate for each of the generating plants from Table II, a worst-case waste stream composition was formulated for the Trial Burn.  Table III presents the targeted composition for the worst-case waste stream and the actual composition used for the Trial Burn. 

 

Table III

Targeted Composition

Plant

Number

Target

Composition

Actual

Composition

26

1 – 5%

0.0%

27

65 – 75%

72.5%

40

15 – 25%

14.1%

41

0 – 5%

3.2%

47

0 – 5%

0.0%

48

0 – 5%

10.1%

 

The targeted worst-case waste stream for the Trial Burn was essentially satisfied with the possible exceptions of Plants 26 and 48. However, a review of Table I shows that the compounds associated with Plant 26 are the same as those associated with Plant 48.  Therefore, the primary objective of the worst-case waste stream determination, to include all of the constituents that could significantly affect the emissions of compounds with the greatest risk, was satisfied.  Therefore, the results of the Trial Burn were obtained using a conservative and defensible approach.

 

Developing a Target Compound List – EPA Region V


In addition to determining the worst-case waste stream for the Trial Burn and Risk Burn, some agencies in different regions and states have developed their own specific list of compounds for use in the Post-Trial Burn Risk Assessment.  These target compound lists include the compounds of greatest concern for the Risk Assessment of the stack emissions from the combustion unit.  The development of the target compound list for a Trial Burn and associated Risk Assessment in EPA Region V is discussed in the following sections.

 

EPA Region V List of Compounds


Early in the discussion process for the development of the Trial Burn Plans, QAPPs, and Risk Assessment Protocols, EPA Region V provided Reilly Industries, Inc. with a list of compounds (4) for inclusion in the Risk Assessment.  This list of compounds went beyond the list in the HHRAP (2), Risk Assessment Emissions Guidance (3), or the North Carolina Protocol (7). 

 

The EPA Region V list of compounds was used as a starting point in developing a final target compound list for the Trial Burn.  The target compound list was applied to the analyses of both the stack gas and waste stream samples.  EPA Region V requested the addition of two compounds to the list of compounds based on their presence in the waste stream.  These additional compounds were pyridine and xylene.  These compounds were added to the list and were assessed with all of the compounds in the following sections.

 

Availability of Toxicity Data


Each of the compounds on the EPA Region V list was compared to the toxicological databases to determine if toxicity data was available for each.  In order to conduct the Risk Assessment, toxicity data has to be available for the compound.  If toxicity data is not available for a given compound and thus not available for the Risk Assessment, there was no compelling reason on the part of the facility to obtain emissions data on that compound.  These compounds are included in the uncertainty section of the Risk Assessment as they are measured by the total organic emissions as unspeciated compounds.  The most current databases were used to determine whether toxicity data was available in order to capture as many compounds as possible.

 

Several compounds were identified on the EPA Region V list that did not have available toxicity data and were thus dropped from the target compound list.  These compounds are identified as follows:


 

A special situation arose concerning four compounds on the EPA Region V list.  These compounds are cis-1,4-dichloro-2-butene, trans-1,4-dichloro-2-butene, cis-1,3-dichloropropene, and trans-1,3-dichloropropene.  These four compounds were specifically listed on the EPA Region V list, although toxicity data did not exist for each.  However, there was toxicity data available for 1,4-dichloro-2-butene and 1,3-dichloropropene, which EPA Region V requested be applied to each of the isomers of the compounds.  Further investigations by the Agency indicated that the toxicity data for 1,3-dichloropropene was obtained from a study where the mixture tested contained 49.5% cis-1,3-dichloropropene and 42.6% trans-1,3-dichloropropene with the balance consisting of epoxidized soybean oil (8).  The Agency assumed that the same situation would apply to the 1,4-dichloro-2-butene although this was not confered  with HEAST.  Although the  practice of assigning toxicity data obtained from a mixture of compounds to the individual isomers may not be scientifically defensible, the Agency and Reilly agreed to use this information in the Risk Assessment.  However, the laboratory was able to provide results for the individual isomers of 1,3-dichloropropene, but was only able to provide results for 1,4-dichloro-2-butene as the total of the cis- and trans- isomers.

 

Availability of Sampling, Preparation, and Analytical Methods


Following an assessment of those compounds that could be used in the Risk Assessment, the compounds on the EPA Region V list were evaluated to determine if emissions sampling, preparation, and analytical methods were available for them.  In order to conduct the Risk Assessment, emissions data had to be obtained for each of  the compounds using readily available, approved, standard methods. The EPA’s SW-846 (9) was reviewed to determine whether sampling, preparation, and analytical methods were readily available for each of the compounds.

 

Only a few compounds were identified on the list that did not have methods available for sampling, preparation, and/or analysis.  These compounds are identified as follows:


 

Since readily available methods were not available for them, they were dropped from the target compound list.  An interesting observation was made at this point that, with the exception of benzidine, this list is identical to the list of compounds that do not have available toxicity data.

 

Three (3) compounds were identified in SW-846 (9) as having available analytical methods, but an adequate response for the preparation method had not been determined by the Agency.  This situation brings to question the applicability of the analytical method when a specific preparation method has not been assessed as being acceptable by the Agency.  Discussions with the laboratory indicated that these compounds could be analyzed using standard methods.  With the EPAs approval, the following compounds and the identified analytical method remained on the target compound list:


 

Two (2) additional compounds were identified as having an available analytical method, but another method was selected for the Trial Burn.  Both of these compounds, n-hexane and 1,1,2-trichloro-1,2,2-trifluoromethane, were shown in SW-846 (9) as using Method 0040.  The analytical laboratory indicated that satisfactory results could be obtained using Method 8260B.  Both of these compounds had boiling points between 30°C and 100°C, which indicated that Method 0030 was acceptable as the sampling method. The Agency agreed to this substitution and n-hexane and 1,1,2-trichloro-1,2,2-trifluoromethane remained on the target compound list.

 

Limitations of Sampling, Preparation, and Analytical Methods


Each of the sampling, preparation, and analytical methods proposed for the Trial Burn were evaluated to determine if any limitations existed that might affect the results.  In order to conduct the Risk Assessment, valid and defensible emissions data had to be obtained for each of the compounds.

 

The limitations discovered during the method review pertained to the volatile organic and semivolatile organic sampling and analytical methods.  Each of these is discussed in the following sections.

 

The semivolatile organic compounds were sampled using Method 0010, prepared using Method 3542, and analyzed using Method 8270C.  Pyridine was identified as having a serious problem with the preparation and analysis methods.  Pyridine is miscible with water (10) and would tend to be lost during the extraction process of Method 3542.  In addition, Section 1.4.7 of Method 8270C (9) indicates that degradation of pyridine occurs due to the injection port temperatures used in the method.  A solution to this problem would be to lower the injection port temperature.  However, the method warns that the performance of the other analytes could be adversely affected.  Both of these problems would cause the analytical result to be lower than expected.  These problems were presented to the Agency as concern for including pyridine in the target compound list.  Because Reilly Industries, Inc. is a manufacturer of pyridine and derivatives of pyridine, the Agency acknowledged these concerns and emissions data for pyridine was collected.

The volatile organic compounds were sampled using Method 0030, prepared using Method 5041A, and analyzed using Method 8260B.  The primary limitation encountered was in relation to the boiling point range and the sampling and preparation methods.  Section 1.1.1 of Method 0030 (9) indicates that the sampling method is applicable for compounds with boiling points between 30°C and 100°C.  Section 1.2 of Method 5041A (9) indicates that the preparation method is most successfully applied to organic compounds with boiling points between 30°C and 100°C.  Section 1.3 of Method 5041A (9) states that compounds with boiling points between 100°C and 120°C may be acceptable based on a case-by-case assessment of such information as the method collection efficiency, tube desorption efficiency, and analytical precision and bias.  Organic compounds with boiling points below 30°C would tend to break through the sampling media and be lost during the emissions sampling, resulting in low recovery rates.  Organic compounds with boiling points above 100°C would tend to adhere to the sampling media and not be analyzed, resulting in low recovery rates.

 

Again, the boiling point concerns were presented to the Agency for these compounds.  The Agency acknowledged these concerns and emissions data for these compounds was collected.

 

Laboratory Standard Analyte Lists


For purposes of the Trial Burn, the EPA Region V list of compounds was refined by removing compounds that did not have available toxicity data or readily available, approved, standard sampling, preparation, and/or analytical methods.  This list was then compared to the standard analyte lists provided by the analytical laboratories.  When the comparison was performed, it was noted that several compounds were not included in the standard analyte list of the laboratories.  If results for these compounds were required for the Risk Assessment, then additional developmental work would be required of the laboratory in order to provide this data.  This would be rather time consuming and the cost of this developmental work would be passed to the facility.

 

Therefore, each of these compounds were reviewed on a case-by-case basis to determine if the compound could be eliminated from the Risk Assessment for valid, defensible reasons.  The following compounds were evaluated:


 

Each of these compounds were reviewed to determine if they were raw materials, process intermediates, or products of the manufacturing process, and whether they could be present in the waste streams burned in the boilers.  The sampling, preparation, and analytical methods were then reviewed to determine if any problems were discovered.  The physical properties of the compounds were also reviewed to determine if they were water soluble.

 

The generator knowledge review related to the raw materials, process intermediates, products and by-products, and waste streams.  This review indicated that all 13 of the compounds could be eliminated from the target compound list based on generator knowledge.  The review of the sampling, preparation, and analytical methods also indicated that 3 of the compounds had significant problems and could be eliminated from the target compound list.  The water solubility problem affected 5 of the compounds.  Therefore, based on each of the above evaluations, the 13 compounds were proposed for elimination from the target compound list.  The EPA Region V subsequently agreed with the proposal and allowed the modification to the target compound list, knowing that these compounds would be considered in the uncertainty section of the Risk Assessment Report by being captured in the total organic emissions analysis as an unspeciated organic.  This evaluation process resulted in significant time and cost savings for the facility by avoiding developmental work by the laboratory to analyze these compounds.

 

Final Target Compound List


The development of a target compound list for the Trial Burn was driven by the requirements of the Risk Assessment.  EPA Region V requested a list of compounds for inclusion in the Risk Assessment.  This list was refined by discounting compounds having no currently available toxicity data and those having no readily available, approved, standard sampling, preparation, and analytical methods.  Two compounds were added to the original compound list based on their presence in the waste stream and the availability of sampling and analysis methods.  The target compound list was finalized by removing those compounds that were not included on the laboratory’s standard analyte list based upon valid and defensible information related to generator knowledge, method limitations, and physical constraints.  Table IV presents the final target compound list that was used for the Trial Burn of Boilers 30K and 70K located at the Reilly Industries, Inc. facility in Indianapolis, Indiana.  In order to account for any risk based contributions associated with compounds that were ultimately discounted during this assessment, a total organic emissions analysis was performed and the uncertainty section of the Risk Assessment Report is used to address these findings.

 

Table IV

Final Target Compound List

Acetonitrile

3,3’-Dichlorobenzidine

Methyl isobutyl ketone

Benzene

1,4-Dichloro-2-butene

Methyl-tert-butyl ether

Benzo(a)anthracene

     (cis and trans)

Naphthalene

Benzo(b)fluoranthene

Dichlorodifluoromethane

2-Nitroaniline

Benzo(k)fluoranthene

1,1-Dichloroethane

Nitrobenzene

Benzo(a)pyrene

1,2-Dichloroethane

Pentachlorophenol

Bis(2-chloroethyl)ether

1,1-Dichloroethene

Polychlorinated biphenyls

Bis(2-ethylhexyl)phthalate

1,2-Dichloropropane

2,3,7,8-PCDDs

Bromoform

cis-1,3-Dichloropropene

2,3,7,8-PCDFs

Bromomethane

trans-1,3-Dichloropropene

Pyridine

2-Butanone

2,6-Dinitrotoluene

Styrene

Carbon disulfide

Di-n-octylphthalate

Toluene

Carbon tetrachloride

Ethylbenzene

1,2,4-Trichlorobenzene

Chlorobenzene

Ethylene dibromide

1,1,2-Trichloroethane

Chloroethane

Hexachlorobenzene

Trichloroethene

Chloroform

Hexachlorobutadiene

Trichlorofluoromethane

Chloromethane

Hexachlorocyclopentadiene

2,4,6-Trichlorophenol

Chrysene

Hexachloroethane

1,1,2-Trichloro-1,2,2-

Cumene

Hexane

     Trifluoromethane

Dibenzo(a,h)anthracene

Indeno(1,2,3-cd)pyrene

Vinyl chloride

1,2-Dichlorobenzene

Methyl chloroform

m-/p-Xylene

1,4-Dichlorobenzene

Methylene chloride

o-Xylene

 

Summary


A mechanism was developed and approved for determining a worst-case waste stream for a Trial Burn that satisfied the needs of the Trial Burn and a Risk Assessment.  This mechanism included the development of a universe of compounds list, refining such to address the relative toxicity of each compound, assessing the potential of each to be present in the waste stream, and the generation rate of each.  In addition, a method was developed to generate a target compound list for a facility located in EPA Region V.  This process included using the original compound list provided by the Agency and refining such based on the availability of toxicity data, whether sampling, preparation, and analytical methods were readily available and did not impose limitations on the results, and the ability of the laboratory to provide defensible results for each.  The worst-case waste stream and target compound list methodologies can be applied to facilities located in other states and EPA regions with minor modifications.

 

REFERENCES


1.         J. Drake, L. Douglas, and J. Jones, “Reducing the Burdens of Compliance Testing Under Increasing Regulatory Requirements”, 2000 International Conference on Incineration and Thermal Treatment Technologies, Portland, Oregon, May 8-12, 2000.

2.         U.S. EPA, “Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities”, Peer Review Draft, EPA530-D-98-001A, July 1998.

3.         U.S. EPA, “Guidance on Collection of Emissions Data to Support Site-Specific Risk Assessments at Hazardous Waste Combustion Facilities”, Peer Review Draft, EPA530-D-98-002, August 1998.

4.         U.S. EPA Region V, “Chemicals of Greatest Concern for Risk Assessment of Stack Emissions from Hazardous Waste Combustion Units”, Internal Memorandum from Mario M. Mangino to Combustion Facility Permit Writers, August 28, 1997.

5.         U.S. EPA, Integrated Risk Information System (IRIS), http://www.epa.gov/iris/index.html.

6.         U.S. EPA, Health Effects Assessment Summary Tables (HEAST),

http://www.epa.gov/radiation/heast.

7.         N.C. DEHNR, “North Carolina Protocol for Performing Indirect Exposure Risk Assessments for Hazardous Waste Combustion Units”, January 1997.

8.         U.S. EPA Region V, “Reilly Trial Burn – Target Compound List”, Internal email from George Bollweg to Jae Lee, October 8, 1999.

9.         U.S. EPA, “Test Methods for Evaluating Solid Waste Physical/Chemical Methods”, SW-846, Version 2.0, Update III, December 1997.

10.       N.I. Sax and J. Lewis, Sr., “Hazardous Chemicals Desk Reference”, 1987.

 

[1] As presented at the 2000 International Conference on Incineration and Thermal Treatment Technologies, Portland, Oregon, May 8-12, 2000.