&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
Air
National Dioxin
Study Tier 4 —
Combustion
Sources
Project Plan
EPA-450/4-84-OUa
February 7985
-------
' i
:1
I
....... S
:!"...
rip 'n.iii •.:'
ii
ii m
i|
I
iNI
it
as
T! i
-------
EPA-450/4-84-014a
National Dioxin Study Tier 4
Combustion Sources
Project Plan
By
Air Management Technology Branch
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office Of Air And Radiation
Office Of Air Quality Planning And Standards
Research Triangle Park, NC 27711
February 1985
-------
This report has been reviewed by The Office Of Air Quality Planning And Standards, U.S. Environmental
Protection Agency, and has been approved for publication. Mention of trade names or commercial products
is not intended to constitute endorsement or recommendation for use.
EPA-450/4-84-014a
-------
ACKNOWLEDGEMENTS
This project plan was prepared by the Air Management Technology Branch,
Monitoring and Data Analysis Division, Office of Air Quality Planning and
Standards, Office of Air and Radiation with significant input and assistance
from Mr. Andrew Miles and his staff at Radian Corporation. Many helpful
review comments on drafts were provided by the Tier 4 National Dioxin Study
Work Group composed of representatives of the U. S. Environmental Protection
Agency Regional Offices, Office of Research and Development, Office of Pesti-
cides and Toxic Substances, Office of Solid Waste and Emergency Response,
Office of Water and others. Useful comments were also received from members
of the Agency's Science Advisory Board, the National Council for Air and Stream
Improvement (wood products industry), the State and Territorial Air Pollution
Program Administrators/Association of Local Air Pollution Control Officials
(STAPPA/ALAPCO), the Illinois Pollution Control Board, the Wisconsin Depart-
ment of Natural Resources, and faculty of the Louisiana State University, the
University of Arizona and the University of Utah. All participants and reviewers
are acknowledged generally and individually for their efforts in developing
this project plan.
iii
-------
-------
Table Of Contents
Page
List of Tables • viii
List of Figures -•' V111
Section 1 Introduction And Summary Of Tier 4 Project Plan 1
1.1 Introduction •
1.2 Background •> • • • • 1
1.3 Management Of The National Dioxin Strategy • 3
1.4 Project Approach • 3
1.5 Basis For Investigation Of Combustion Sources 5
1.6 Summary Of Tier 4 Program Plan 8
- 1.7 Possible Results- Of The Tier 4 Study 13
1.8 Format Of This Report • 1^
Section 2 Alternate Study Plans Which Were Considered 15
2.1 Alternative 1 - Focus The Tier 4 Effort Primarily
On Municipal Waste Incinerators 15
2.2 Alternative 2 - Investigate Conditions Which
Increase/Decrease Dioxin Emissions • • • • • 1?
2.3 Alternative-3 - Develop A Mechanistic Model To
Estimate 2378-TCDD emissions • 17
2.4 Alternative 4 - Collect Ash Samples From A Large
Number Of Source Categories • 18
-------
Table Of Contents
Section 3 The Tier 4 Project Plan . 21
3.1 Overview 21
3.2 Literature Review And Evaluation 21
3.3 Summary Of Available Data 21
3.3.1 PCDD In Feed ; 23
3.3.2 Precursors In Feed 25
3.3.3 Chlorine In Feed 27
3.3.4 Combustion Conditions 27
3.4 Identification Of Source Categories With
A Potential To Emit Dioxin 31
3.5 Development Of Criteria For Prioritizing
The Candidate Source List 34
3.5.1 Rank D Source Categories 34
3.5.2 Rank C Source Categories 35
3.5.3 Rank B Source Categories 39
3.5.4 Rank A Source Categories 41
3.6 Types Of Samples To Be Collected Under The
Stack Testing Program 42
3.7 Stack Sampling Analysis Priority 46
3.8 Ash Screening Program 47
vi
-------
Table Of Contents
Page
Section 4
4.1
4.2
' 4.3
4.4
4.5
4.6
4.8
4.9
References
Appendix A
Implementation Of The Tier 4 Plan 51
Overview 51
Selection Of Individual Sources For Testing 51
Ash Sampling Program 52
Source Testing Program • 53
Quantitative Exposure And Risk Assessment: 54
Management Of Tier 4 55
Coordination With The EPA Regional Offices 55
Additional Information 57
•; 59
65
vii
-------
-------
LIST OF TABLES
Number'
1-1
1-2
1-3
3-1
3-2
3-3
3-4
3-5
3-7
3-7
3-8
4-1
4-2
Ranked Source Categories List For Stack Testing
Combustion Source Categories Where Ash Samples And
Stack Samples Will Be Collected
Sampling And Analyses Model
Combustion Sources For Which TCDD Test Data Are Available
Characteristics Of Sources Tested For Dioxins . .,
Combustion Sources '
Combustion Source Categories Believed To Have The Greatest
Ranked Source Category List For Stack Testing
Summary Of Tier 4 Stack Sampling Procedures
Sampling And Analyses Model
-Combustion Source Categories Where Ash Samples And
Stack Samples Will Be Collected
Tier 4 Work Group •
Regional Office Contacts - Tier 4 •>
Page
7
9
,. . 11
, 22
, . . 24
. . . 32
. . . 33
. . . 36
, 43
... 45
48
. .. 56
.. . 58
Number
1-1
3-1
A-l
LIST OF FIGURES
Schedule: National Dioxin Study - Tier 4 - Combustion
Sources
Preliminary Source List (from Table 3-4)
Determination Of Sample Size For Municipal Waste Incinerators
Page
12
35
67
viii
-------
-------
Section 1
Introduction And Summary Of Tier 4 Project Plan
1.1 Introduction
This report presents the U. S. Environmental Protection Agency's (EPA's)
plans for identifying likely sources and assessing the magnitude of dioxin"
emissions from combustion sources under Tier 4 of the National Dioxin Study.1
The report describes the information EPA plans to collect to improve the Agency's
current understanding of combustion sources which may emit dioxin to the ambient
air. This report also describes the rationale used in developing the project
plan, as well as alternative approaches that were considered. While source
sampling and analytical methods that will be used during the study are mentioned,
•this report is not intended to describe these methods in detail. References to
other documents which describe methods and procedures more fully are identified.
1.2 Background
There are 75 different chlorinated dioxin isomers, divided into eight
homologues, each with different physical and chemical properties.2 One of the
22 isomers with four chlorine atoms is 2,3,7,8-tetrachlorodibenzo-p-dioxin
(2378-TCDD).* There are three reasons why this isoraer is the principal focus
of the National Dioxin Strategy:
1. It is believed to be the most toxic of the chlorinated dioxins,
2. It is the isomer most often associated with exposure and potential
health risks to humans, and
throughout this project plan, the terms dioxin and dioxins are used generically
to indicate chlorinated dioxin compounds, as distinguished from 2378-TCDD, the
specific isomer of major concern.
-1-
-------
3. There is sufficient associated health and exposure information
available to allow a targeted study to be developed.
On December 15, 1983, EPA released a Dioxin Strategy! which provides a
framework under which EPA will:
1. Study the nature and extent of contamination of 2378-TCDD and the
associated risks to humans and the environment;
2k Implement or compel necessary cleanup actions at contaminated
sites; and
3. Further evaluate regulatory alternatives to prevent future con-
tamination, as well as disposal alternatives to alleviate current
problems.
To implement the strategy, EPA has formed seven study tiers, ordered by the
decreasing potential for 2378-TCDD contamination:
Tier 1 - 2,4,5-trichlorophenol* (245-TCP) production sites and
associated waste disposal sites;
Tier 2 - Sites (and associated waste disposal sites) where 245-TCP
was used as a precursor to make pesticidal products;
Tier 3 - Sites (and associated waste disposal sites) where 245-TCP
and its derivatives were formulated into pesticidal products;
Tier 4 - Combustion sources;
Tier 5 - Sites where pesticides derived from 245-TCP have been and
are being used on a commercial basis;
Tier 6 - Certain organic chemical and pesticide manufacturing
facilities where improper quality control on certain production
*2378-TCDD is a known contamination of 2,4,5 trichlorophenol, hence the
focus on sites producing and/or handling 245-TCP.
-2-
-------
processes could have resulted in the formation of a 2378-TCDD contam-,
inated product waste stream; and
'Tier 7 - Control sites where 'contamination from 2378-TCDD is not
suspected.
1.3 Management Of The National Dioxin Strategy
Overall responsibility for the management of the Strategy has been assigned
to the Assistant Administrator for the Office of Solid Waste and Emergency
Response (OSWER). OSWER is also managing the investigations for sites in Tier 1
and 2 of the study. The Office of Water Regulations and Standards (OWRS) is man-
aging Tiers 3,5,6 and 7, while the Office of Air and Radiation (OAR) is respon-
sible for Tier 4. The Office of Research and Development (ORD) is responsible
for sampling and analytical guidance. ORD is also providing analytical support
for Tiers 3-7 through EPA laboratories, collectively known as the Troika." As
described later, EPA's Regional Offices are responsible for implementing various
aspects of the National Dioxin Strategy, including portions of Tier 4.
1.4 Project Approach
There are many data gaps in the available information concerning emissions
of dioxins from combustion sources.3 As such, a number of different study
approaches could be taken to fill these data gaps. After considering various
alternatives, the Agency has decided to focus the Tier 4 effort on the collection
of samples from a number of combustion source categories which are believed to
have the greatest potential to emit 2378-TCDD to the ambient air. The primary
objective of the Tier 4 project thus, is one of problem definition, i.e., what
*Environmental Research Laboratory (ERL), Duluth, MN; Environmental Monitoring
and Support Laboratory (EMSL), Research Triangle Park, NC; and the Office of
Pesticides and Toxic Substances, Environmental Chemistry Laboratory (ECL),
Bay St. Louis, MS.
-3-
-------
source categories emit dioxin and at what concentrations? Extensive stack and
ash samples from selected source categories will be collected and analyzed for
'2378-TCDD and the higher dioxin and furan homologues (i.e., tetra, penta, hexa,
hepta, and octa). Such efforts are needed to broaden our understanding of what
combustion source categories emit dioxins and furans to the ambient air.
This approach is believed necessary for a number of reasons. Only a limited
number of combustion source categories have previously been tested to determine
if they emit dioxins '(including 2378-TCDD) to the ambient air. While dioxLn and
furan compounds have been found in stack emissions and fly ash from some source
categories, they have not been detected at all of the source categories which
have been tested. Ranges of dioxin concentrations have been shown to be very
broad (orders of magnitudes). Some of the variation in the ranges may be due,
to differences in process operations within a source category [e. g., mass burn
versus refuse derived fuel (RDF) as subcategories of municipal incinerators; dry
process versus wet process as subcategories of RDF, etc.] but data are generally
insufficient to allow such conclusions to be drawn. Also measurement methods
have varied considerably within the data base now available. These and other
factors, not surprisingly, have led to conflicting theories on how and when
dioxins are formed. In summary, currently available data do not clearly define
the magnitude and scope of dioxin (and furan) air emissions from combustion
sources.3
The scope of the Tier 4 study must be limited, for several practical rea-
4»
sons. While the Tier 4 budget of over $1.3 million is a considerable sum, it,
does not provide the funds for the Agency to implement a comprehensive study
program that can examine all aspects of dioxin emissions from combustion sources.
For example, Tier 4 will not adequately answer the question of how to minimize
dioxin emissions. An important factor which limits the scope of Tier 4 is the
—4—
-------
high cost of stack sampling which is estimated to cost approximately $90,000
for each source tested. This cost does not include similarly expensive analy-
tical work. Because of these high costs and the limited resources available,
the scope of the study must be sharply focused.
Also of considerable impact is the fact that the National Dioxin Strategy is
a two year effort scheduled to end in December 1985.* At that time, a report to
Congress is planned which will summarize the results of the study. This schedule
limits to some degree the types of activities which can be initiated and com-
pleted by that date. Certain research studies, such as developing and validat-
ing a mechanistic model of dioxin formation in combustion sources, which would
likely take years to design and complete, would not be practical for this study.
Because all aspects of the dioxin issue cannot be included in Tier 4,
followup studies may be.needed. The results of the Tier 4 study should provide a
reasonable basis on which to determine any need for additional studies.
1.5 Basis For Investigation Of Combustion Sources
There are several unproven hypotheses concerning dioxin emissions from
combustion processes. Dow Chemical's "Chemistries of Fire" theory proposes that
dioxins are a natural byproduct of fire and will be formed at some quantities in
all combustion processes.5 However, experimental results by Buser and Rappe and
an evaluation of data from the literature3 suggest that dioxins are emitted only
under limited conditions. The most prevalent theories, including Esposito's
formation mechanism,2 involve the incomplete combustion of polychlorinated
.dibenzo dioxin (PCDD's) or PCDD precursors. Neither the Dow hypothesis nor the
" *5 Q
precursor hypothesis have been conclusively supported or refuted. J»-) Recent
*$90,000 is an estimate based on a typical source test, and does not include
any laboratory analysis for dioxins, homologues or isomers. Actual costs will
vary depending on source configuration, specific analytical work done., etc.
Each work plan for individual sites will have individual estimates and costs
will be optimized. See Reference 4.
-5-' - •!
-------
studies involving pyrolysis of wood with and without chlorination, in conjunc-
tion with the studies of pyrolysis of chlorinated coal, suggest that inorganic
chlorine at high levels with any organic material may very well lead to PCDD
formation.
Although there is some disagreement, PCDD precursors are defined for the
purpose of this plan as chlorinated aromatics that can produce PCDD's through
biraolecular reactions and thermal rearrangements during incomplete combustion.
Examples include chlorinated phenols and chlorinated benzenes. When PCDD's are
formed in a combustion source, or are already present in fuel, they can escape
in the exhaust gas stream with the fine particulate especially when the thermal
destruction efficiency of the combustion process is low.3
The Agency has recently performed a qualitative analysis of available data
considering precursors and combustion conditions believed to contribute to
dioxin emissions.3 Combustion source categories determined to have the great-
est potential to emit 2378-TCDD are listed in Table 1-1. Source categories.
rated as "Rank A Sources" are those for which limited data are available and
are believed to have the greatest potential to emit dioxin. Rank A sources
will receive the highest priority for testing. Some B and C ranked Source
categories will also be tested.
Since combustion sources are ubiquitous and are fairly concentrated in
urban areas, it is conceivable that a large portion of the United States popu-
lation may be exposed to dioxins including 2378-TCDD in the ambient air. More
data are needed to determine the magnitude and extent of exposure to dioxins
from combustion sources. While current evidence does not suggest that an unrea-
sonable risk exists, the Dioxin Study provides an opportunity to perform a more
orderly and thorough investigation of the issue. For each stack test by Tier 4
where dioxins are determined to be emitted to the air, the risks of dioxin
emissions will be estimated.
-6-
-------
Table 1-1. Ranked Source Category List For Stack Testing1
Rank A Source Categories:2
Sewage Sludge Incinerators
Black Liquor Recovery Boilers
Rank B Source Categories:3
Industrial Incinerators (including hospital incinerators)
Boilers (Firing PGP treated Wood)*
PCP Sludge Incinerators^
Carbon Regeneration (industrial)
Metal Reclamation
Rank C Source Categories:5
Wood Stoves
Charcoal Manufacturing
Mobile Sources
Small Spreader-Stoker Coal Boilers
Chlorinated Hazardous Waste Incinerators
Lime/Cement/Aggregate Kilns Co-fired with Chlorinated
Organic Wastes
Commercial Boilers Firing Fuels Contaminated with Chlorinated
Organic Wastes % '
Open Burning
Apartment House Flue Fed Incinerators
Rank D Source Categories :*>
Municipal Solid Waste (MSW) Incineration
Industrial Boilers Co-firing Wastes • • . -
list of combustion source categories under investigation by Tier 4 continues
to be revised as new information is received and is therefore subject to change.
2Rank A are large source categories (greater than 1 million tons of fuel and/or
waste burned annually) with elevated dioxin precursor contamination of feed/
fuel. These categories are judged to have a high potential to emit TCDD.
3Rank B are small source categories (less than 1 million tons of fuel and/or
waste burned annually) or source categories with limited dioxin precursor con-
tamination of feed/fuel. These categories have some potential to emit TCDD.
is an abbreviation for pentachlorophenols.
C are source categories less likely to emit TCDD.
D are source categories which have already been tested three or more times.
-7-
-------
1.6 Summary Of Tier 4 Program Plan
Two levels of source sampling are planned in Tier 4. On one level,
relatively inexpensive -ash samples (screening) will be- collected from combus-
tion source categories listed in Table 1-2. As explained later, these source
categories are believed to have a likelihood of emitting dioxins. Ash samples
will be collected from approximately three different sources in each source cate-
gory. These ash samples will generally be collected from control devices, or
at some other point in the device duct work downstream from the combustion zone.
"Bottom" ash samples will only be taken where downstream samples are not pos-
sible. These samples will be analyzed to determine the concentration of 2378-
TCDD and the higher dioxin and furan homologues present in the ash. These data
will be used to provide a qualitative indication of the presence of dioxins in
the flue gas emissions. Due to the uncertainties demonstrated in previous ash
and stack sample comparisons, ash samples can neither be used with certainty to
quantify dioxin and furan air emissions nor to'conclude their absence from stack
emissions.3 However, dioxins measured in ash will identify those source cate-
gories which probably should be further tested. Tier 4 ash data will also add
to the available information on the ash and stack gas relationships of di'oxin.
In addition to the ash sampling program, a number of combustion source
categories have been selected for detailed source testing. Because of the cost
involved with stack testing, only 10 - 12 different sources can be tested from
the source categories listed in Table 1-2. These stack tests will be performed
using EPA's modified Method 5 and source testing procedures consistent with those
described in the recently developed American Society of Mechanical Engineers
(ASME) sampling protocol for waste-to-energy facilities.6
The samples to be collected during the stack testing program are discussed
in Section 3 and are briefly summarized here. They include total stack emission
samples before and after controls, ash or slurry samples from the control device
-8-
-------
TABLE 1-2. COMBUSTION SOURCE' CATEGORIES WHERE ASH SAMPLES AND
STACK'SAMPLES WILL.BE COLLECTED1
Source Categories
Sewage Sludge Incinerators
Black Liquor Recovery Boilers
Industrial Incinerators (including hospital incinerators!)
Metals Reclamation (including wire reclamation
incinerators and secondary copper smelters)
Chemical Sludge Incinerators
Industrial Carbon Regeneration Units
Charcoal Manufacturing Ovens
Wood Stoves
PGP Treated And Salt Laden Wood Combustion In Boilers
Small Spreader-Stoker Coal Fired Boilers
Kilns And Commercial Boilers Burning Hazardous Wastes
Open Burning (including forest fires and agricultural
burning)
Apartment House Flue Fed Incinerators
Mobile Sources
Samples To Be
Collected -
'Ash
X '
X
X
X
X
X
X
X
X
X
X
X
X
X2
S tack
X
X
X
X
X
X
-
X2
X
X
-
—
-
-
list of Tier 4 categories continues to be revised as new information is
received and is_ therefore subject to change.
2Includes some samples and results being supplied to Tier 4 by other programs.
-9-
-------
and firebox, samples of the feed or fuel, and samples of nearby soils. Precom-
bustion air samples will also be collected at those sources where precursors are
suspected to be present in significant concentration in the air from other nearby
sources. A summary of the samples to be collected during stack testing and the
analyses to be performed under Tier 4 are outlined in Table 1-3.' Although not
noted in the table, an additional number of samples will be collected and
analyzed for quality assurance/quality dontrol purposes.
Stack, ash, feed, fuel, soil and precombustion air samples for Tier 4 will
be analyzed for the 2378-TCDD isomer, and for each of the higher homologues of
dioxin and furans (i.e., tetra, pemta, hexa, hepta, and octa). Precombustion
air and feed samples will also be analyzed for selected suspect dioxin precur-
sors (e.g., chlorophenols and chlorobenzenes). Stack samples will be collected
primarily by EPA contractors who will ship them to the designated Troika labora-
tory for analysis. Procedures for the collection and shipping of samples, as
well as the analytical and quality control procedures to be followed, are
discussed elsewhere.''
At the conclusion of the Tier 4 study, the Agency will have collected either
qualitative screening samples or detailed stack samples from a number of combus-
tion source categories. These data should provide a better understanding of the
source categories which emit dioxins and furans to the atmosphere. The detailed
stack test data will allow the Agency to determine the magnitude of the dioxin
and furan emissions from these sources and calculate the risk to which the people
in the vicinity of such sources are exposed. Figure 1-1 presents the anticipated
schedule for-the Tier 4 project.
The information collected during the study will be summarized in a final
report. The report will also summarize the available data from other studies.
For example, stack test data being collected from various muncipal incinerators
by New York State and others will be included, if available. Any preliminary
-10-
-------
TABLE 1-3. SAMPLING AND ANALYSES MODEL
Sample
Inputs
Precombustion Air
Haste Feed and/or Fuels
Outputs
Stack Before Control
Stack After Control
Bottom Ash
Ash From Control Device
Other
Soils (in vicinity)
Sampling Train Blanks
Method
XAD-2
(Ambient)*
Grabs
MUST*
MM5Th
Grabs
Grabs
Borings
MUST*
Samples
0-1«
Dally Composite
Condenser Rinse
Adsorbent Resin
Fllter(s) Catch
Probe Rinse
Daily Composite
Dally Composite
1 Composite
Resin, Fil-
ter(s), Rlnsec
, Analyses*
2378-TCDD,
4-8 CDD/CDFf, TOCL8, Cl-phMM5T la a modified EPA "Method 5" train an defined in the ASiME protocol. ,,«,.„ ,., i. . ,
iFor blank sampling trains, one is designated priority oae and the second will be analyzed only if the first blank train
has detectable background levels of dioxin/furans.
JMultiple analyses for each sample will have a multiplier effect on the analytical costs also. Total does not include
quality assurance samples (approx. 20Z).
-11-
-------
•j
X
en
K
C5
Oi
a
2
o
CQ
I
sr
a£
u
M
H
I
%
o
U)
03
O>
** "J
b.
"5
o
CO
I
o
X
CM
(a
•
r-l
i-4
a)
o
s
e o
•* i
b, l
I
•u O
2 i
0!
o
fa
01
Q-
o
u
41
O
fa
a.
e
I1
i-
<*-*
Q.
3
a
fa
U
fa
3
-------
results from the three million dollar, three year effort being initiated by
Environment Canada to examine the significance of various operating conditions
on dioxin emissions from municipal incinerators will also be included. The
Tier 4 teport will also include a list of recommendations for future testing
and study.
1.7 Possible Results Of The Tier 4 Study
The primary objective of this study is problem definition. The degree to
which Tier 4 can satisfy this objective depends in large measure on the magni-
tude and pervasiveness of dioxin and furan emissions found during the study.
For examples—stack test data and subsequent risk assessments may indicate, at
. least for the sources or source categories tested, that dioxin and furan emis-
sions are not likely to result in an unreasonable risk to the public. Assuming
this is the result, some general conclusions about the scdpe of dioxin and furan
emissions from combustion sources can be drawn. Such conclusions will obviously
need to be stated carefully, acknowledging the possiblity that some untested
sources may exist which may emit elevated levels of 2378-TCDD or other dioxins
and furans.
A more likely outcome of the study is that the ash sampling program will
identify a number of source categories with a sufficient probability of having
levels of dioxins and/or furans present in the ash which merit more detailed
testing. Similarly, the stack test program which can be conducted under Tier 4
may identify relatively high emission rates from some of the sources which are
tested, thereby indicating a need for additional testing at similar sources.
As a result, some followup studies would be needed to define more fully the
magnitude of the problem.
-13-
-------
1.8 Format Of This Report
This section has provided an overview of the scope and purpose of Che
Tier 4 program. Section 2 discusses various alternatives which were considered.
Section 3 provides further detail on how the source categories to be tested under
Tier 4 were identified, the number and types of samples to be collected and the
analyses to be performed. Section 4 describes the implementation aspects, of the
program, such as how individual sources were selected for testing and how the
various samples will be collected.
-14-
-------
Section 2
Alternative Study Plans Which Were Considered
The previous section presented a brief overview of the scope and purpose
of the Tier 4 study. In designing the study, the Agency considered -various
alternatives. The most recommended alternatives are described below, along
with the reasons why they were not included in the Tier 4 plan.
2.1 Alternative 1; Focus The Tier 4 Effort Primarily on Municipal
Waste Incinerators
The most frequently suggested alternative was to focus the Tier 4 study,
in whole or in part, on determining the magnitude of dioxin and furan emissions
from various municipal waste incinerators.' There were several reasons for this.
First, of the combustion source categories which have been tested, dioxin
emissions are greater for this source category than for any others. This source
category is, therefore, viewed with continued interest and concern. Second, a
number of urban areas are presently considering the permitting of: new municipal
incinerators, primarily as resource recovery units, due to the limitations asso-
ciated with land disposal. Solid waste program representatives and other gov-
ernment officials, representatives of industry, and the public are concerned
about the potential exposure and risk from dioxin and furan emissions from these
proposed facilities. As a result, it has been recommended that a number of
municipal waste incinerators be tested in order to gain a better understanding
.of their dioxin and furan emissions.
After careful consideration, the Agency decided to focus the Tier 4 stack
test program primarily on combustion source categories other than municipal
waste incinerators for the following reasons:
-15-
-------
(1) Six of the approximately 40 large conventional (non-modular)
municipal waste incinerators in the United States have already been
tested.3 As such, this source category is by far the most 'tested of
all combustion source categories.
(2) Based on the results of previous tests, the Agency has tenta-
tively concluded that the emissions from well operated facilities do
not pose a significant public health problem to those persons living
in the vicinity of these sources. This position was initially con-
tained in a report entitled "Interim Evaluation Of Health Risks
Associated With Emissions Of Tetrachlorinated Dioxin From Municipal
Waste Resource Recovery Facilities,"8 and was reaffirmed by the
Agency in December 1983.9
(3) It is possible that other source categories may emit similar or
larger quantities of dioxins and furans or have a greater exposure and
risk associated with them than municipal waste incinerators. This
cannot be determined unless resources are devoted to testing other
source categories, and
(4) There appears to be considerable interest in testing specific
municipal waste incinerators, so that a number of municipal inciner-
ators will be tested, even if they are not tested under Tier 4. At
present, approximately ten municipal waste incinerators have tests
planned or are being tested, primarily by Region III, NASA, individual
States, and Environment Canada.
For these reasons, the Agency has decided not to focus the Tier 4 stack
testing effort on municipal incinerators. However, available data from tests
o'f municipal incinerators by other agencies will be obtained and included in
the final Tier 4 report.
-16-
-------
2.2 Alternative 2: Investigate Conditions Which Increase/Decrease
Dioxin And Furan Emissions
Another alternative considered was to include a field study designed to
examine the significance of various combustion device operating parameters on
the magnitude of dioxin emissions. Under this option, detailed stack sampling
would be performed at one or two facilities. Various operating conditions (such
as waste load and excess air) would be carefully monitored over a range of con-
ditions and emissions measured for correlation.
It is recognized that such work is useful and necessary to understand how
dioxin emissions can be minimized. However, such a study has not been included
in Tier 4. The primary reasons a-re—the high cost to conduct such a field study
and that similar studies are planned by others.
For example, Environment Canada, has recently initiated a three year, $3
million study designed to investigate the significance of changes of various
operating conditions upon emissions at three municipal waste incinerators.
Also, the ASME plans a similar series of tests at a United States incinerator.
For these reasons, a program to collect additional field data to characterize
the significance of operating parameters has not been included. However, the
results of the studies previously mentioned will be obtained and summarized in
the final Tier 4 report.
2.3 Alternative 3: Develop A Mechanistic Model To Estimate
Dioxin And Furan Emissions
Since resources to conduct detailed stack tests for each combustion source
*
category are not possible, it has been suggested that an empirical -or ^mecha-
nistic model be developed which could be used to estimate emissions of dioxins
and furans from source categories which could not be stack tested. This alter-
native was not included in the final Tier 4 plan for the following reasons:
-17-
-------
(1) A mechanistic or empirical model data may require a substantial
•amount of time to develop and validate. Given the current schedule of
the National Dioxin Strategy, there is probably not sufficient time to
(a) design such a model, (b) collect the necessary source test data,
(c) validate the model, and (d) estimate dioxin emissions for many
source categories.
(2) It is not entirely clear, if a single model could be developed
which would be applicable to all combustion source categories, or
whether a number of models would need to be developed for various
types or groups of combustion source categories. Certain theoretical
models now exist, but the practical field application of these models
is believed limited at this time.
(3) The development and validation of such models may not be possible
or successful,, given the limited amount of test data which can be
collected under Tier 4.
For these reasons, the Agency has decided not to include the development of
mechanistic or empirical models as part of the Tier 4 program. However, it is
anticipated that a followup project to further examine the feasibility of devel-
oping a practical empirical or mechanistic model(s) will likely be initiated at
a later date. Such efforts could use the data collected under Tier 4 and other
ongoing projects as a framework to design such a model development and validation
program.
*
2.4 Alternative 4; Collect Ash Samples From A Large Number Of
Source Categories
In order to determine which source categories emit dioxins and furans, it
has been suggested that ash samples be collected from a large number of combus-
tion source categories. Under this alternative, ash sampling would the primary
activity of Tier 4.
-18-
-------
A comprehensive ash sampling program in lieu of stack testing has not been
included primarily because ash samples have a limited .value in identifying
sources which may emit dioxins.3 A review of available data suggests that con-
centrations of dioxins measured in ash do not correlate sufficiently with dioxin
flue gas concentrations for ash data to be used to estimate stack emissions.
Although the data available tend to suggest a trend, ash sampling has limited
value, even to identify which source categories emit dioxins.
Ash samples collected for Tier 4 stack sampling program will be examined
for any correlation of ash data with stack emissions. It is hoped that with
these additional data and studies by others, more definite statements can be
made regarding the value of ash samping data.
-19-
-------
-------
Section 3
The Tier 4 Project Plan
3.1 Overview
This section describes the process which was used to identify those
combustion source categories which are likely to have t:he greatest potential
to emit dioxin to the ambient air. These source categories are candidates for
source sampling under Tier 4. The process initially involved a literature
review of information related to dioxin emissions from combustion sources.3 The
information was reviewed to determine (a) which combustio.n_-Sjjiirce categories had
already been tested for dioxin and (b) which factors appear to be most important
with respect to dioxin being emitted from combustion sources. Criteria were then
applied to prioritize those source categories believed to have the greatest
potential to emit dioxin.
3.2 Literature Review And Evaluation
An initial literature search for dioxin emission information was made in
1983 for OAR/EPA.!0 This literature search was used as the starting point for
a more specific review of previous studies concerning dioxin emissions from
combustion sources. Contacts were made to identify recently performed or on-
going studies which could be used to supplement the data. base. While the full
results of the literature review are reported elsewhere, the following para-
graphs provide a summary of tHe key findings of the search.3
3.3 Summary Of Available Data
A review of the literature^ has identified 12 combustion source categories
which have been tested for dioxins and for which data are available. Table 3-1
lists these source categories and the concentrations which have been measured.
-21-
-------
I
1
I
8
erf
s
en
O
CO
1
»
i
u
co
m
*J
o
b
2
«
I
g
5
c
s
.5
in
t V
** TJ
4J V
•4 **
f* •
«4 «
0 H
•f
Eli
&
V
•U
5
2
m
i
.
*"* 2
^•^ •*» M
1« M w a
BO B
o «> Sin A o o
.
.tf .* .tf .M J* .* J< Jtf
o o u u u u u u
•2, -5, -2,
*o cn w o ••* P» * »o
1
b ft*
• 0 •
b u b O
O •> O **
u b u •
1 5-1 S . S
• .e u c b
o m e «H « •
o M u ••* b
u e *P* e
2 2-S ^ S 3 :
: s 2: % - £
* a s5 i s -
•H O
• < b u e ca •«•«
s. oa 3 e • x u
- =,« -- a s s '
'BOO o o 5 5 t -S
2 £ £ 3 £
^
a
jj
a
i
o
a
"*.
a
e
0
1
O
en
0
vO
O
o
4
«a
e
u-l
O
•S
09
^
|
b
|
1 Activated Carbon le
«n -»
. .
u
Q*
O.
f * t 1
N
W fib
eu dx
o. o | ,
<5 «
M
£ 1 5
r* i as
ft. O Q
•^ w tn
1 !i
Ck ** ft.
Ot Ot Ot
01 °" <• M
Pj 0 «n
-*
•l •
oe oa
C W C
filii
81 M " w
^ -s.
««r *^
^ (J» ^ .4
s 1
b O
•= IS
5 -S3
tealdentlal Coibuat
Habile SourceaJ
Wire Reclamation In
Li«e/C««ent Kllna (
CO
.
M
"i
o.
en
w-i
O
O
a
c.
o.
»
«M
a
sc
•^
I
:
rM
!
3
U
"3
Accidental Electric
Flreak
a
f
y
c
•J5
o
7
|"
O
M
|
1
O
Is
b -O
« O
H X
• • a
0 £
8 3,
s-^s
w
a
4)
U
1
0
_!
>» t>
I
o
t-
4)
0.
e
«
01
41
A
U
i
a
4)
3 u
S'S
i .
5 S
.a
3 3
B Z
•^
41
••4
(Q
a
1
s
o
e
S
•3
a
e
o
i
4)
S
O
o
oa
c
00
-T
1
•8
I
b
B
3
b
e
4)
U
S
u
a
8
GU
X
e
o
e
u
*
8
£•
reported.
j*
3
u
?
e
41
4)
X
w
§
4>
1
w
3
41
(K
00
^
44
JS
00
4t
a
X
e
o
«
b
b
4)
Q.
b
Q
(
a
a.
9
to
01
t,
=
c
1
2
3
M
-
a
a
•a
a
a
41
•o
a
u
e
a
4)
3
Ct.
41
b
4»
ca
u
a
4)
u
4)
S
e
•
4)
C
O
ca
5
^
4)
rj
Q
41
|
4)
a
Cb
CD
4)
O.
a
CO
u
9
3
1
41
§
«M
§
b
4)
i-4
•
b
41
b
u
0
1
e
tor batteries.
«4
u
(8
a.
CJ
1
ca
b
i
b
O
ca
e
m
u
ea
V
4)
a
u
e
u
00
1
X
.a
e
0
^
"H
b
b
au
i
e.
^
-22-
-------
General operating characteristics of the sources which were tested are sum-
marized in Table 3-2 and are used below to identify combustion related parameters
which may affect the likely presence arid magnitude of dioxin emissions from
combustion sources. •
Dow Chemical's "Chemistries of Fire" hypothesis proposes that polychlori-
nated dibenzo dioxins (PCDD's) are a common and natural byproduct of fire.5
However, an evaluation of data from the literature suggests that PCDD's are
emitted during the incomplete combustion of either (a) dioxins or (b) dioxin
precursors. Although uncertainty remains, dioxin precursors are defined for.
purposes of thi_s_project plan as chlorinated aromatics which can produce dioxins
through bimolecular reactions and thermal rearrangements. Examples include
chlorinated phenols and chlorinated benzenes. Based on the literature review,3
the following factors appear to affect dioxin emissions:
°PCDD in feed,
"Precursors in feed,
°Chlorine in feed,
"Combustion temperature,
"Residence time,
"Oxygen availability,
"Feed processing, • ,
"Supplemental fuel.
3.3.1 PCDD In Feed3
2378-TCDD is a "side reaction" impurity that can result from the manu-
facture of trichlorophenol, which is used to make the herbicide 2,4,5-trichloro-
phenoxy acetic acid (245-T). Pentachlorophenol (PCP) production can also result
in a PCDD contaminant, primarily octachlorodibenzo-p-dioxin (OCDD). The primary
end use for PCP is as a wood preservative. It is anticipated that some dioxin
-23-
-------
t
a
01 a
1-4 01
a a M
-1 fj W
a
0) O O
fa o •
a So
fa f-4 1-4
01 1
Bf O
S S
H
Q.
H fa
S
3 S
fa rt OT
S3 ^H t4
Ch a 41
? "O
C a
• 01 01
U fa
a Of
* CO
fa
§a a
u u
fa fl «-4
9 U U
u a a
oi a B
fa oo
CU fa fa
a a
u u
a
fa
a £
O 0
. 1 i
u u
fa *0
3 0 01
O W fH >
cn • a -4
a B fa
3 o oi
-4 a
n u
« .BO
fi> 0) a
•H > 3
U B fHf-4
VfaS^> S^4f-4
— i a, N u a u fa
fa ^«4CnE-«. >, 4J
tl -4 •— T3 fa J= faB
> *4 3f4T>410 aO
f-i a o- 3 o u 4-1 4j a
oi • c1
B a a ea*-t
« O 01 S X
>• 'H B -4 -4
a " oi x a
5 1 3 S a i «{=•
fa ^4»4a o ja oj oj
a a fafaUfiiB-f4
O 1 OO> Bf-40IU-
rf 3 S 3 ^ a u "325
" UU»fa*«4
>>U« faOl W B 41 0141 B •• -O O
** ts -H --. a -a a u
>-i ia sa -H -4 -< oi oia •-< 41 « •«
*•* 99 *na -u a A fa 93 u U3a>*
« 0^ B^» 0 4) 0 -! 5vi 0 X 0 - •<
9u M •< a« £ S fj •< u J j S
a -a
-24-
-------
contamination may also occur during the manufacturing of other similar chlori-
nated aroraatics, particularly if" the manufacturing efficiency is less than
optimal. Therefore, dioxins are thought to enter the"environment as a contam-
inant of some commerical products, such as wood preservatives and pesticides.
The widespread use of these products increases the possibility of
finding dioxins in the feed of a combustion process. For example, PGP treated
wood may be used to fire industrial boilers such as those at papermills. Runoff
may carry pesticides containing dioxins to water treatment facilities where the
organics are incorporated into the sludge, which may then be incinerated. Like-
wise, contaminated waste materials from some manufacturing processes may be com-
busted as an energy recovery procedure. Two examples are PCP sludge incinerators
used at some wood preserving facilities and black liquor recovery boilers used
at paper mills.
Studies of emissions from boilers included in the literature review3
demonstrate that if PCDD's are detected in the feed of an inefficient or poorly
controlled combustion process, it is very likely that they will be emitted to
the atmosphere.
3.3.2 Precursors In Feed
Most of the studies reviewed in the literature search3 focused on the
formation of dioxins and furans from precursors. Espossito et al. presented
detailed descriptions of these formation mechanisms for chlorinated dioxins.39
This work organizes dioxin precursors into three classes:
Class I - Polyhalogenated phenols, primarily with a halogen ortho to
the hydroxyl group, with a high probability of dioxin formation;
Class II - Ortho-halophenols and ortho-halophenyl esters where the
substituted groups are a mixture of halogens and nonhalogens; and
-25-
-------
Class III - Other chemicals having the possibility, but less likeli-
hood, of dioxin formation, including chlorinated aromatic compounds.
The majority of experimental work to date has centered on three classes
oE precursors: chlorinated phenols, chlorinated benzenes, and PCB's. Dioxin
formation from the combustion of chlorinated phenols has been tested extensively
by Rappe^O jansson^l, and Ahling42,43. Dechlor.ination of highly chlorinated
homologues can result in formation of the more toxic 2378-TCDD isomer. Chlori-
nated phenols are used as wood preservatives, herbicides, and for sap stain
control. Wood or vegetation sprayed with chlorophenols may be disposed of by
incineration or used as a supplemental fuel in boilers. In addition, chloro-
phenol wastes are often disposed of in sludge incinerators and industrial
boilers. The identified references above provide a reasonable basis for con-
sidering .these combustion sources as likely sources of dioxin emissions and
therefore, candidates for the Tier 4 testing program.
Buser investigated the formation of PCDD's and PCDF's from the
pyrolysis of chlorobenzenes.44 The formation mechanism included a chlorophenol
and a polychlorinated diphenyl ester (PCDPE). intermediate. Chlorobenzenes are
used in solvents, dyes, Pharmaceuticals, and rubber production. Such products
make up much of the organic chlorine found in municipal waste incinerator feed.
The associated waste products may also be disposed of in other incinerators or
boilers.
Buser also investigated the formation of PCDF's from the pyrolysis of
PCB's.45-47 NO experimental work has been" identified .on PCDD formation from
PCB's. However, several studies have been identified that found PCDD's emitted
from PCS fires.23,37,48 in addition, PCB's are frequently in solution with
hexachlorobenzenes which have been shown to form PCDD's. PCB's have been used
-26-
-------
as dielectric fluids, in transformers and capacitors hydraulic fluids, plastic-
izers, and dyes. The incineration of PCB's at' waste disposal facilities or in
boilers may result in PCDD and PCDF emissions.
3.3.3 Chlorine In Feed
The chlorine content of fuel is obviously an important parameter
affecting the formation of PCDD's or PCDF's. Shih et al. developed a ranked
priority list of conventional combustion systems emitting polycyclic organic
matter including PCDD's and PCDF's.49 The rationale presented for source
ranking is based on fuel characteristics and combustion conditions. Shih's
work places great emphasis on both the chlorine content: of the feed and the
concentration of aromatics in the feed.
Various authors have demonstrated the effect of chlorine on dioxin
emissions. Mahle et 'al. demonstrated that PCDD's were emitted from coal combus-
tion only when chlorine was added.50 Tiernan et al. .found PCDD formation during
the combustion of pine in the presence of HC1, but no PCDD's were detected during
the combustion of pine alone.51 Liberti studied the combustion of vegetables.52
When inorganic chlorine or PVC were added, PCDD's and PCDF's were detected in the
feed. While there is general acceptance that some chlorinated compounds are con-
clusive to dioxin formation, these inorganic chlorine studies demonstrate that
the specific mechanisms involved in PCDD formation are complex and not well
understood. However, it can be generally stated that chlorine must be present
for the formation of PCDD, and general trends suggest that increased chlorine
concentrations in the feed may improve the possibilities of PCDD emissions.
3.3.4 Combustion Conditions
The remaining factors that are believed to significantly affect
PCDD emissions are combustion conditions. These are combustion temperature,
-27-
-------
residence time, mixing, use of supplemental fuel, and fuel processing. The
overall combustion efficiency of a combustion process is a function of all of
these factors. In order to destroy PCDD's or prevent their formation, the com-
bustion efficiency must be high. This requires a combination of high tempera-
tures, available oxygen, high heat content fuel, and long residence times. Each
factor is discussed separately below.
3.3.4.1 Combustion Temperature
Experimental evidence suggests that temperatures of 500-800°C pro-
mote PCDD formation, while temperatures greater than about 800°C destroy
PCDD's.1,43,47 Buser et al. showed that PCB pyrolysis at 550-650°C forms
PCDF.47 However, pyrolsis at temperature greater than 700°C cause 99 percent
destruction of PCB's and no PCDF formation.38 Ahling et al. produced similar
results for both PCDD's and PCDF's during the combustion of chlorophenols.^2
•Combustion temperature is a function of the BTU- (British Thermal Unit)
content of the fuel or supplemental fuel, the available air, and the degree of
fuel processing. Municipal waste incinerators are considered a significant com-
bustion source of PCDD's.14-19 These large mass burn units are characterized by
low combustion temperatures, due in part, to the high moisture, low BTU content
of the fuel; poor air/feed mixing due to the lack of feed processing; and lack
of supplemental fuel. In comparison, many hazardous waste incinerators and high
efficiency boilers are designed for very efficient combustion. These units burn
high BTU fuels and/or add high BTU supplemental fuels; and, even if the air/fuel
ratio is low, the air/fuel mixing is efficient. The fuels are processed (e.g.,
drying, shredding, combustibles separation, etc.) to decrease mositure and
improve mixing. In many cases, high temperature afterburners are used for the
combustion of offgases. Several studies have identified the effects of high com-
bustion temperatures on PCDD's and PCDD precursors.22,23,26 por example, no
-28-
-------
PCDD's were detected in the emissions of Che Vulcanus incinerator ship during the
combustion of PCDD contaminated Herbicide Orange.22 xhe combustion temperature
during this project was 1600°C.
3.3.4.2 Residence Time
The residence time necessary to destroy PCDD's and the combustion
temperature are inversely related. The higher the combustion temperature, the
shorter the required residence time for PCDD destruction. Likewise, a low
temperature will require a long residence time for destruction of PCDD's.
Sachdev et al. showed that an increase in both temperature and residence time
decreased the formation of PCDD's from chlorophenol combustion.53 Similar
results have been found at hazardous waste incinerators that run with 1.5-2.0
second residence times. Combustion sources with longer residence times and
high temperatures are less likely to form products of incomplete combustion,
such as dioxins. ' '
3.3.4.3 Oxygen Availability
Oxygen availability is a function of both the a.ir/fuel ratio and air/
fuel mixing efficiency, both of which are of concern when burning solid fuels.
Solid fuels and high viscosity liquid fuels (waste tars, etc.) burn as partic-
ulates or large droplets, and therefore portions of the fuel are burned in low
oxygen or pyrolysis conditions. A low air/fuel ratio or poor air/fuel mixing
will promote poor combustion conditions and the formation of PCDD's. Jansson
demonstrates that an insufficient air supply increases! PCDD emissions from
chlorophenol combustion.^ Muncipal waste combustors are usually fired with
excess air. However, large mass burn units have poor air/fuel mixing due to
the lack of fuel processing. Activated carbon regeneration and wire reclamation
incinerators are both designed with limited air (some wire reclamation is done,
-29-
-------
in poorly designed nearly open chambers which are conducive to low temperatures
and mixing). All of these cases have been shown to emit dioxins.14,32,35
Sources with a low air/fuel ratio or poor air/fuel mixing will be considered
potential sources of PCDD emissions.
3.3.4.4 Feed Processing
The feed material for a combustion source may be a liquid, a solid, or
a gas. Both liquid and gaseous fuels can be easily mixed with air resulting in
a high combustion efficiency; solid feeds usually require some processing to
improve combustion. Often solid feeds require drying, shredding, or separation
to improve combustion. Similarly, high viscosity fuels, for example waste
tars, require some preparations including preheating and atomization prior to
combustion.
. Feed processing will determine, in part, both oxygen availability and
residence time. Fi-ne, homogeneous • feed particles will improve air/fuel mixing
and combustion. Larger particles will require longer residence times and may
result in local oxygen deficiencies due to poor mixing. High moisture will also
decrease combustion efficiency. Therefore, highly processed homogeneous feeds
are less likely to emit products of incomplete combustion, such as PCDD's.
3.3.4.5 Supplemental Fuel
When burning a low BTU fuel, the addition of supplemental fuel will
increase the combustion temperature and improve combustion. Haile et al. tested
a boiler cofiring RDF with coal.12 xhe boiler temperature was 1200°C, and no
PCDD's were detected. Dow Chemical tested an industrial incinerator burning
waste tars without supplemental fuel and found parts per billion (ppb) levels of
TCDD's in the fly ash.5 After the addition of a supplemental fuel, no TCDD's
were detected. '
-30-
-------
When selecting sources for the Tier 4 testing program, sources fired
with high BTU supplemental fuel will be considered less likely to emit PCDD's.
Sources burning low BTU wastes without supplemental fuels will be given higher
priority for testing.
3.4 Identification Of Source Categories With A Potential To Emit Dioxin
The previous paragraphs have identified a number of: parameters which are
likely to affect the presence and/or magnitude of dioxins emitted from combustion
sources. In the analysis described below, various combustion source categories
are reviewed to determine which exhibit characteristics more closely associated
with these parameters. Source categories so identified are given a, higher
priority for source sampling under Tier 4.
Abroad list of combustion source categories is provided in Table 3-3.
These source categories were examined in a somewhat judgmental manner with the
combustion related parameters previously described. This process resulted in a
listing of combustion source categories (see Table 3-4) which are considered to
be more closely associated with these parameters. Table 3-4 also includes those
source categories where dioxin has previously been found to be emitted. Source
categories in Table 3-4 represent those which are likely candidates for source
sampling under Tier 4.
A number of source categories initially listed in Table 3-3 do not appear
in Table 3-4. These have been excluded for a variety of reasons. For example,
fossil fuel-fired boilers, process heaters, and gas turbines have been excluded
because of their use of fuel with relatively low chlorine content. For the same
reason, industrial processes using low chlorine fossil i:uels, such as the pro-
duction of carbon black and coal gasification, have not been included.
Other sources have not been included in Table 3-4 because they are rela-
tively small sources burning a low annual tonnage of fuels. Examples included
-31-
-------
en
•z.
O
1
en
Ed
J
I
4)
S
DM
(8
4J
4)
^
u
O
4)
a
a
u
2
o.
*4
4
iw4
b
u
a
3
•o
e
1-4
Q
•H
^J
??
01
*
f-4
o
O*
O
V
a
(9
3
a
g
Q
b
O4
-
a a
b u
41 B
-4 M
O B
n H
^4 b
e o
a u
O 01
* b
' I
o
a
B
a
b
H
1-4
a
U
b
u
a
en
S3
•H
•8
2
u
u
•H
Catalyt
•J
b
O
w
31
c
CJ
B
,•4
a
•p4
b
^j
a
3
^
B
H4
S
01
1-4
•H
O
a
a
b
PU
1-4
a
b
3
u
2
OT
00
12
facturli
3
C
a
X
Charcoa
m
b
01
t-4
0
sa
01
ta
a
a
a
a
•o
41 *Q
i! 2
•44 -< 1
a M
3
1
u
i-4
a
'a,
a
a
b
O
^
b
o
B
01
a
a
3
0
^
b
a
H
a
X
a
a
00
*4
a
b
3
1 1
I |
oo
E
u
a
a
Coffee
,3
01
a
a
1
*4
a
oo
a
B
tM
U
b
1
a
3
O
S
01
.*
a)
a
u
X
Oi
•H
Jl
a
S
4J
a
b
01
B
C
i
a
0)
B
.a
b
3
H
b
>< 01
b a.
qi a
> 0.
o
u w
et "a
b i
a* a
^4 b
^ *H
u o
a a
r-4
00
^
b
O
b
01
G
u a
B b
01
v4
W
U
a
0)
x-x Of
a
•H e
"4 O
14 A
0 S
. u
b
•M
c«.
a
0) b
u o
a 4
B
b
S3
•4 b
B 3
b w
3 i-4
pa 3
U
B ^4
oi b
O <
e
01
u
a
a
0)
b
»4 a o o
s o a o »
2! ! ! ! !
• i t i i
O III!
88
-32-
-------
Table 3-4. COMBUSTION SOURCE CATEGORIES BELIEVED TO HAVE THE
GREATEST POTENTIAL TO EMIT DIOXINS
Source
•-'. Rationale
Hazardous Waste Incinerators
Municipal Waste Incinerators
RDF2 Boilers
Commercial Waste Oil Boilers
Industrial Boilers Co-firing Wastes
Wire Reclamation Incinerators
Activated Carbon Regeneration
PCP3 Sludge Incinerators
Sewage Sludge Incinerators
Mobile Sources
Wood Stoves/Fireplaces
Industrial Incinerators
Wood/Bark Boilers
Charcoal Manufacturing
Bla'ck Liquor Boilers
Cement/Lime Kilns Co-firing Wastes
Small Spreader-Stoker Coal Boilers
TCDD1 Detected
TCDD Detected
TCDD Detected
TCDD Detected
TCDD Detected
TCDD Detected
TCDD Detected
TCDD Detected
PCDD^ Detected
TCDD Detected
TCDD Detected .
TCDD Detected from municipal
incinerators
Experimental results with PCP
treated wood
Experimental results with PCP
treated wood, combustion conditions
High POM5 effluent, PCPs in feed
Precursors present
Favorable combustion conditions for
dioxin formation.
refers to the tetra homologues as a group. Available analyses are mixed,
with some researches reporting "total tetras" and others reporting 2378-TCDD
or both. The presence of TCDD's generally indicates some likelihood of
2378-TCDD being present.
2RDF = Refuse Derived Fuel.
3PCP » Pentachlorophenol.
^PCDD = Total of all dioxin homolgues or Polychlorinated dibenzo dioxins. While
dection of PCDD's does not necessarily indicate presence of TCDD or 2378-TCDD,
there are sufficient data to infer such in this case.
5POM = Polycyclic Organic Matter.
-33-
-------
coffee roasting and meat smoke houses. Similarly, intermittent, small sources,
such as structural fires have been excluded. While some of these sources may
possibly emit 2378-TCDD, these, source categories are less likely to pose an
exposure problem than larger sources.
3.5 Development Of Criteria For Prioritizing The Candidate Source List
Since it will not be possible due to resource constraints to stack test
each of the candidate combustion source categories listed Ln Table 3-4, a
method for prioritizing the candidate source categories was developed. The
method is admittedly subjective, since no realistic assignment of weighting
factors could be made for the ranking criteria. Criteria and the "decision
tree" used to prioritize the candidate source categories are illustrated Ln
Figure 3-1.
Based on these criteria, candidate source categories were placed into one
of the four groups listed in Table 3-5. .Table 3-5 also presents the grouped
source category list. The source categories within a particular group are not
listed in any particular order. The Rank A group represents source categories
that have -not yet been extensively tested and which are expected to be the most
likely to emit dioxins. Rank B and C groups also have a fairly high likelihood
of emitting dioxins but are hypothesized to emit less than Rank A sources. Rank
D source categories have been tested three or more times.
3.5.1 Rank D Source Categories
Rank D source categories include municipal waste incinerators and
industrial boilers co-firing wastes. Six municipal waste incinerators have been
*As described in Appendix A, it is believed that data from three different
sources provides a reasonable amount of data to characterize a particular source
category for this effort. Appendix A also demonstrates that it will not be pos-
sible to obtain statistically significant results under Tier 4, since it is not
possible to test enough sources under this program to obtain a rigorously
representative sample to be able to develop meaningful confidence limits and
account for variabilities that exist among sources.
-34-
-------
Preliminary Source List (From Table 3-4)
\
3 Or More
Valid Tests
Yes
No
Potential To Emit 2378-TCDD
1. TCDD1 Detected
2. Precursor Level
3. Combustion Conditions
Low _
High
Size3 Of
Source Category
I
Small
Large
Rank A
Sewage Sludge Incineration
Black Liquor Recovery Boilers
Rank D
Municipal Waste Incineration
Industrial Boilers Cofiring
Wastes
Rank C
Mobile Sources
Wood Stoveii
Small Spreader Stoker Boilers
Chlorinated Waste Incineration
Aggregate/l.ime/Cement Kilns and
Commercial and Boilers Firing
Chlorinated Wastes2
Charcoal Manufacturing
Open Burning (Including Forest
Fires and Agricultural Burns)
Apartment House Flue-Fed
Incinerators
Rank B
Wire Reclamation
Industrial Carbon Regeneration
PCP4 Sludge Incineration
PCP Treated or Salt Laden
Wood Fired Boiler*
Industrial Incinerators
(Including Hospital
Incinerators)
.n™ reported onl? the total for the tetra hooologue. The
presence of TCDD is assumed for this ranking procedure to be adequate to
indicate the potential to eait 2378-TCDD.
chlorin*ted »«««•«» «:<"»" be « higher ranked category.
C bMCd on "••*«* on Industrial Boiler, firing
t0 *"™" chlorinated waste dis-
on«, tak",into consideration the number of sources, their emis-
sion volume and general location with respect to exposed populations .
"Contacts with sources identified as incinerating PCP sludges in boilers
indicate that these sludges are now generally landfilled.
Figurfe 3-1: Ranking Criteria and Decision Tree
-35-
-------
Table 3-5. Ranked Source Category List For Stack Testing1
Rank A Source Categories: 2
Sewage Sludge Incinerators
Black Liquor Recovery Boilers
Rank B Source Categories ;3
Industrial Incinerators (including hospital incinerators)
Boilers (Firing PGP treated Wood)*
PGP Sludge Incinerators
Carbon Regeneration (industrial)
Metal Reclamation
Rank C Source Categories:5
Wood Stoves
Charcoal Manufacturing
Mobile Sources
Small Spreader-Stoker Coal Boilers
Chlorinated Hazardous Waste Incinerators
Lime/Cement/Aggregate Kilns Co-fired with Chlorinated
Organic Wastes
Commercial Boilers Firing Fuels Contaminated, with Chlorinated
Organic Wastes
Open Burning
Apartment House Flue Fed Incinerators
Rank D Source Categories :*>
Municipal Solid Waste (MSW) Incineration
Industrial Boilers Co-firing Wastes
list of combustion source categories under investigation by Tier 4 continues
to be revised as new information is received and is therefore subject to change.
A are large source categories (greater than 1 million tons of fuel and/or
waste burned annually) with elevated dioxin precursor contamination of feed/
fuel. These categories are judged to have a high potential to emit TCDD.
B are small source categories (less than 1 million tons of fuel and/or
waste burned annually) or source categories with limited dioxin precursor con-
tamination of feed/fuel. These categories have some potential to emit TCDD.
is an abbreviation for pentachlorophenols .
C are source categories less likely to emit TCDD.
D are source categories which have already been tested three or more times.
-36-
-------
tested for dioxin in the United States.11"15 By December 1985, it is expected
that seven additional dioxin tests will have been performed on municipal waste
incinerators in industrial boiler tests have been analyzed. Because more than
three stack tests the U. S. Several fly. ash and stack tests have also been
conducted in Canada and Europe.16"20 Similarly, archived samples from six waste
fired, have been conducted for both of these source categories, they were not
considered for additional testing under Tier 4.
3.5.2 Rank C Source Categories
Rank C source' categories are those which have been judged to be less
likely to emit dioxins when compared to the other candidate source categories.
This is based on the following factors: (a) the level of tetrachlorinated dioxins
detected in previous tests, (b) the anticipated level of precursors in the fuel/
feed, and (c) the likelihood of conditions favorable to incomplete combustion.
Rank C source categories include . mobile sources, wood stoves, wood boilers,
small spreader-stoker coal boilers, chlorinated hazardous waste incinerators,
lime/cement/aggregate kilns and commercial boilers burning chlorinated wastes,
charcoal manufacturing, open burning and apartment house flue-fed incinerators.
Mobile sources have been tested in two earlier studies.5,34 in both
cases, 2378-TCDD was detected at the part per trillion (ppl:) level in particulate
matter scraped from exhaust systems or in filtered exhaust samples. Chlorine
content of the gasoline and diesel fuels are also low (0-100 ppm). Based on
those earlier tests and* chlorine content, mobile sources have been ranked C.
Wood stoves have been tested by Dow Chemical.5,33 2378-TCDD was
detected in chimney scrapings up to 100 ppt. The chlorine content of wood is
less than 100 ppm. Although the combustion conditions found in wood stoves
favor the formation of dioxins, the low chlorine concentrations and the low
-37-
-------
levels of dioxins detected in earlier studies indicate that this category may
not emit high levels of 2378-TCDD. Nevertheless, 48 million tons of wood are
burned annually, 'so the potential exposure could be considerable. Wood-fired
boilers burning untreated wood, which are more efficient combustion systems, are
also ranked C.
Small, spreader-stoker coal boilers have also been placed in the
Rank C category. This source category was ranked in this manner because of
combustion conditions which favor dioxin formation (e.g., relatively low com-
bustion temperatures and poor air/fuel mixing due to large feed size). It should
be noted that tests on high temperature, coal-fired .utility boilers have not
detected dioxins in stack emissions,.
Charcoal manufacturing is a source category with combustion condi-
tions which favor dioxin formation. Furnace temperatures are kept low (500°C),
and both air/fuel ratio and air/fuel mixing are poor.54 While source category
surveys indicate that some charcoal manufacturers may use scrap wood from saw
mills, most use forest scraps. Approximately three to four million tons of
wood are used annually in the production of charcoal.54 The majority of char-
coal manufacturing facilities are located in rural settings. For these reasons,
this source category was ranked C.
Another Rank C source category is commercial boilers co-firing
wastes. Approximately 6 million tons of Resource Conservation and Recovery Act
(RCRA) wastes, many of which are chlorinated aromatics, may be disposed of in
commmercial boilers annually.55 However, studies on industrial boilers have
detected only low concentrations of dioxins. In addition, combustion of chlori-
nated wastes in commercial boilers will most likely be prohibited in the future.
For these reasons, the category is ranked C.
-38-
-------
Both, hazardous waste incinerators and kilns co-firing wastes have
been ranked C. Elevated levels of chlorinated dioxin precursors can be expected
in the feed of these source categories. However, previous tests have shown that
relatively low levels of dioxin are emitted.23-26,36 This is due to combustion
conditions which are not favorable to dioxin formation. Combustion temperatures
for these sources are usually greater than 1000°C, and the residence time for
fuel ranges from one to two seconds. The use of high BTU supplemental fuel and
afterburners also decreases the likelihood of dioxin emissions from these
sources.
Agricultural burning, forest: fires and flue-fed apartment house
incinerators are possible dioxin sources recommended for testing by Tier 4.
Burning of harvest residues and forest fires where phenoxy herbicides have been
applied may be dioxin sources. Since dioxins are emitted from municipal waste
combustion, smaller, less efficient apartment house incinerators may also emit
dioxins. There are many of these units in urban areas, especially in the
Northern U. S., with approximately 2000 in the New York City area alone. The
flue-fed incinerators have the greatest potential of poor combustion conditions.
3.5.3 Rank B Source Categories
Sources with high potential to emit are ranked A or B. Rank B
source categories are smaller source categories (less than one million tons of
fuel burned per year) or are categories with lower concentrations of precursors
in the feed. These include PGP sludge incineration, carbon regeneration, indus-
trial incineration," wire reclamation and high chlorine oir PGP treated wood fuel
boilers.
PGP sludge incineration has been evaluated in two studies.30>56 jn
both cases, PCDDs were detected in fly ash samples. Through experimental
studies, PCPs have been shown to form octachlorodibenzo-p-dioxins (OCDD) during
-39-
-------
combustion.41 Dechlorination of OCDD can produce lower chlorinated dioxins and
thus 2378-TCDD.29 Annually, 19,000 tons of .PGP are used to preserve wood.57
Only a small percentage of this is incorporated into sludge which must be dis-
posed. Historically, the sludge is incinerated at the wood preserving, plant,
along with scrap wood. Contacts with facilities thought to incinerate PGP
sludges indicate a recent trend towards landfilling of the sludge. Where sludges
are incinerated, it is exptected that a lack of feed processing and a high mois-
ture content sludge can result in poor combustion conditions. Therefore, PGP
sludge incineration is a potential source of dioxin emissions.
Industrial activated carbon regenerations where activated carbons is
heated to drive off captured organics, was also been ranked B. In order to
prevent the combustion of the carbon during the carbon regeneration process, the
air/fuel ratio and the combustion temperature (800°C) are kept low.58 These con-
ditions are conducive to incomplete combustion. The thermal regeneration of
activated carbon has been studied at one source, where the spent carbon had been
used to treat a water supply system.32 Ppt levels of TCDDs were detected in the
fly ash and stack emissions. Precursor levels in the municipal water treatment
carbon are low in comparison to what may be expected in carbon used to treat an
industrial waste stream.
Industrial incinerators, including hospital incinerators, are also
ranked B. There are approximately 3000 of these units in the United States.59
These incinerators have many applications including disposal of high organic
chlorine wastes such as industrial waste plastics and used medical supplies.
One industrial incinerator burning municipal solid waste has been tested and
low levels of 2378-TCDD were detected.14 An industrial incinerator burning a
more chlorinated organic waste, suck as chlorinated plastics, is a possible
dioxin source.
-40-
-------
Another Rank B source category is wire reclamation incinerators where
plastics are burned off wire and other electrical equipment to reclaim metals,
especially copper. In previous tests, ppt levels of TCDDs were detected in ash
samples.35 As with carbon regeneration and charcoal manufacturing, both the
combustion temperature and available oxygen are kept low in this process.60 wire
insulation incinerated during the process often contains PCBs and polyvinyl
chloride.
The final Rank B source category is boilers burning salt laden wood or wood
that has been treated with PGP. This source category was ranked on the large
volume of wood burned in the U. S. and the results of laboratory studies where
dioxins were formed in the pyrolysis of wood in the presence of HCL. Burning
salt laden wood, that has been stored in salt water, can result in a flyash with a
chlorine salt content of 65 percent.61 Burning PGP treated wood is of similar
concern as PGP sludge combustion with wood.
3.5.4 Rank A Source Categories
The remaining source categories were ranked A. These are sewage
sludge incinerators and black liquor boilers.
Approximately 1.5 million tons of sewage sludge are incinerated
annually.62 The sludge is most commonly burned in a series of hearths to reduce
bulk and recover energy from volatiles in the sludge. Fluid bed combustion is
also used. The high moisture content of the feed can result in poor combustion
conditions. In addition, sewage from an industrial area may contain elevated
levels of PCBs and other chlorinated aromatics. In an earlier Canadian study,
PCDDs were detected in stack emissions.27 Many newer units are operated on a
fuel efficient system. In order to save energy, both the combustion temperature
and the available oxygen are held low, conditions which, increase the potential
for dioxin emissions.
-41-
-------
The other Rank A category is black liquor recovery boilers. Approx-
imately 13 percent of all fuel used in large industrial boilers is pulping
liquor.63 Some 42 million tons of pulping liquor are recovered by burning in
boilers of .which black liquor boilers are a subcategory.6^ Sources with greatest
probability to emit dioxin are those recovering liquor with a higher chlorine
(1-3 percent) than most black liquor boilers (less than 1 percent chlorine in the
liquor). In order to prevent smelt oxidation, the temperature is often kept be-
low 800°C in these boilers. In addition, earlier studies have detected POM
emissions from these boilers.65
3.6 Types Of Samples To Be Collected Under The Stack Testing Program
The stack testing program is designed to obtain a quantitative measure of
dioxin and furan emissions discharged to the atmosphere from a given source.
Gaseous and particulate matter stack samples will be collected, both before and
after existing control devices in most cases. .Stack emissions will be sampled
using methods similar to those described in the newly developed draft ASME samp-
ling protocol.6*66 This protocol is based on use of an EPA modified Method 5
train with an XAD-2 sorbent.
A summary of the stack testing requirements contained in the protocol is
outlined in Table 3-6. A detailed summary can be found in National Dioxin Study
Tier 4 - Combustion Sources; Sampling Procedures, EPA-450/4-84-014c.66 Three
independent sampling runs will be made at each site over a period of about one
week. The analyses of stack samples will include the tetra and higher dioxin and
furan homologues (i.e., penta, hexa, hepta, and octa), and 2378-TCDD. Samples
will be analyzed to ppt level of detection (approximately 40 ppt).
Ash, water, soil, and precombustion air samples will be collected in
addition to stack samples. Ash samples will be periodically collected during
-42-
-------
Table 3-6. SUMMARY OF TIER 4 STACK SAMPLING PROCEDURES^
"Sampling procedures will be based on the draft ASME protocol requirements.
- Three separate test runs.
- Sample quantities will allow for initial analysis archiving.b
- Stack tests includes the following samples:
— Stack Emissions
— Air Pollution Control Device Inlet Streamc
— Grab Samples of Feed/Fuel
— Bottom Ash Grab Samples
— Ash or Slurry from Control Devices
— Precombustion Air^
— Quench Water Effluent
-- Soil
"Standard EPA Method 5 train, modified to incorporate a sorbent module containing
XAD-2.
"Emissions Characterization:
- Stack Temperature
- Stack Gas Flow
- Moisture Content
- C02, CO, 02, NOX, THC (continuous monitoring)
- HC1, SOX Opacity (where appropriate)
aData requirements set forth in the ASME protocol will be collected as appro-
priate. Since the ASME protocol is directed toward municipal waste incinerator
tests, actual data collection for Tier 4 sites may vary. For example, opacity
continous instack monitoring data after a wet scrubber may not be very meaning-
ful but an HC1 sampling train may provide information on available chloride in
flue gases. See Reference 66.
^Except most stack samples.
cExcept uncontrolled sources or where data is not meaningful.
^Only at site where dioxin or precursors may be present in intake air.
-43-
-------
each o£ the stack test runs and composited daily. As described later, compari-
sons of the magnitude of dioxin in the ash with that in the flue gas are planned.
Analysis of ash samples may identify potential disposal problems which may exist
if high levels of dioxins and furans exist in the ash. Procedures for collection
of ash samples are also described in Sampling Procedures66.
Limited soil samples will be collected at each of the sources, primarily
around ash handling areas. Soil samples will be collected using sampling proce-
dures described in Sampling Procedures66. Composite soil samples from the
various sites will be analyzed for 2378-TCDD, and dioxin and furan homologues.
Precombustion air will be sampled at any source where there are "hits" and
any significant dioxin precursors are suspected to be in the air as a result of
site contamination. Some heavily industrialized areas for example, could have
relatively high (ppb to ppm levels) concentrations of precursors of dioxin in
the ambient air. The decision to test precombustion air will be made source-by-
souirce from information obtained during the pretest survey. The "criteria for
this decision is outlined in the pretest survey protocol described in Section 4.
Precombustion air samples will be collected with an ambient sampler described in
Tier 4 Sampling Procedures66. These samples will be analyzed for 2378-TCDD, and
the higher dioxin and furan homologues, PCBs, chlorinated phenols, chlorinated
benzenes, and other chlorinated organics.
During the stack test, fuel/feed samples will be periodically collected and
composited. The fuel/feed samples will be analyzed for PCBs, total chlorine,
chlorinated phenols, and chlorinated benzenes, since these compounds may be pre-
cursors of dioxin. If dioxins are detected in the stack emissions, the fuel
samples will also be analyzed for 2378-TCDD, and the higher homologues of
dioxins and furans. Table 3-7 summarizes the analyses to be performed on each
sample collected during the stack test. In addition to the number of samples
-44-
-------
TABLE 3-7. SAMPLING AND ANALYSES MODEL
Simple
Input!
Precombuation Air
Waste Feed and/or Fuels
Outputs
Stack Before Control
Stack After Control
Bottom Ash
Ash From Control Device
Other
Soils (in vicinity)
Sampling Train Blanks
Method
XAD-2
(Ambient)*
Grabs
MMSTh
HMSTh
Grab*
Grabs
Borings
MUST5
Saaples
0-ie
Dally Coaposlee
Condenser Rinse
Adsorbent Resin
Filter(s) Catch
Probe Rinse
Daily Composite
Dally Composite
1 Composite
Resin, Fil-
ter^), Rlnseo
*
Analyses4
2378-TCDD,
4-8 CDD/CDFf, TOCLS, Cl-phenols,
Cl-benzenes
2378-TCDD,
4-8 CDD/CDFf Cl-phenols,
Cl-benzenes, PCB, TOCLS
2378-TCDD,
4-8 CDD/CDF*
2378-TCDD,
4-8 CDD/CDFf
2378-TCDD,
4-8 CDD/CDF*
I
2378-TCDD,
4-8 CDD/CDFf
2378-TCDD, -
4-8 CDD/CDFf
2378-TCDD,
4-8 CDD/CDFf
TOTAL
Total
Samples'1
for
Analysis
1«
3*
3
3
3
3
1
2
19J
Analysis0
Priority
2
2
2
1
1
1
3
l/2i
bBased on 3 sampling days. Numbers shown are for dloxin/furan samples only.
cpriority 1 indicates analyses should be performed for 2378-TCDD and higher (tetra-oeta) homologues of dioxins
and furans. If there is a detectable concentration, Priority 2 samples are to be analyzed. Priority 3 samples may be
analyzed on a case-by-ease basis if Priority 2 samples have a positive response.
"XAD-2 (ambient) denotes a modified high volume air sampler for ambient organic compounds.
•Sample to be collected If there is e significant source of suspected precursors in vicinity of sampling site.
£4-8 CDD/CDF - Analyses done on higher (teera through octa) chlorine number dloxlns and furans with results reported as
totals for each group only.
8TOCL • Total organic chlorine.
hMM3T is a modified E?A "Method 3" train as defined in the ASME protocol.
iFor blank sampling trains, one is designated priority one and the second will be analyzed only if the first blank train
has detectable background levels of dlosln/furans.
^Multiple analyses for each sample will have a multiplier effect on the analytical costs also. Total does not include
quality assurance samples (appro*. 20Z).
-45-
-------
listed in the table, additional analyses will be required for quality assurance/
quality control purposes (blanks, duplicates, blinds etc.)*
3.7 Stack Sampling Analysis Priority
Tier 4 has been allocated approximately 700 of the 3600 analyses by Troika
for the National Dioxin Study. Estimates of Tier 4 analyses for 10-12 stack and
45-60 ash sampling sites indicate that up to 900 analyses would be required. In
order to meet the analyses allocation and to utilize expensive Troika analyses
most efficiently, an analyses decision tree was developed. A decision tree
reduces the number of analyses performed on samples collected in the field,
thereby reducing analytical costs. The guiding principle of the decision tree
is to analyze sufficient samples to determine whether or not dioxin enters the
environment. Internal process streams would be analyzed only in the event of
detection of dioxin in the effluent streams. By using the decision tree, total
analytical resource constraints can be met without affecting the Tier 4 goal to
assess the risk of dioxin emissions from combustion.
The decision tree will be used to divide field 'samples into three prior-
ities. Each sample will be identified by the sampling organization prior to
submission to the analytical organization on the following priority basis.
Priority 1;
All Priority 1 samples will be analyzed. These samples are identi-
fied, such that if the results are negative, a very high probability will exist
that Priority 2 and 3 samples will also be negative.
Priority 2;
If 2378-TCDD is detected above the selected criteria value for
Priority 1 samples, Priority 2 samples will be analyzed.
-46-
-------
Because no minimum level of concern can be initially designated for
2378-TCDD, any 2378-TCDD detected above the determined analytical minimum detec-
tion limit will be the criteria for analyzing Priority "2 samples unless higher
thresholds are designated by risk assessments.
Priority 3:
These samples will be collected in the field but are not planned for
analysis unless a "major" dioxin emission is determined. Therefore, they will
be archived by the sampling organization. Analysis of these samples may occur
on a case-by-case basis.
To minimize possible sample interferences, aqueous phase samples should be
extracted within approximately fourteen days of collection. All aqueous phase
samples that require1 extraction will be identified before submission to Troika.
3.8 Ash Screening Program
The Tier 4 plan includes a program to collect ash screening samples from a
number of combustion source categories. The ash sampling program is relatively
inexpensive and will provide dioxin related information for several combustion
sources. Table 3-8 lists the ranked source categories which have been recom-
mended for ash screening. Additional source categories that have a reasonable
liklihood of emitting dioxins may be added to the screening list, based upon
recommendations from the EPA Regional Offices and others.
Unfortunately, at this time, collection and analysis of ash samples is not
a demonstrated, reliable screening technique to provide conclusive estimates of
dioxin emissions from combustion sources. A considerable fraction of dioxins
are hypothesized to be emitted as a gas or absorbed upon the fine particulate.
If this is true,-it would generally not be concentrated in bottom ash or the
flyash collected in an air pollution control device. Data are available which
-47-
-------
TABLE 3-8. COMBUSTION SOURCE CATEGORIES WHERE ASH
SAMPLES WILL BE COLLECTED
Source Categories^
Sewage Sludge Incinerators
Black Liquor Recovery Boilers
Industrial Incinerators (Including Hospital Incinerators)
Metals Reclamation (Including Wire
Incinerators and Secondary Copper Smelters)
Chemical Sludge Incinerators
Industrial Carbon Regeneration Units
Charcoal Manufacturing Ovens
Wood Stoves
PGP Treated And Salt Laden Wood Combustion In Boilers
Small Spreader-Stoker Coal Fired Boilers
Lime/Cement/Aggregate Kiln And Commercial
Boilers Burning
Open Burining (Including Forest Fires And Agricultural
Burning)
Apartment House Flue Fed Incinerators
•Includes some sources which have been identified by State,
Regional, and other EPA Program Offices with reasonable likeli-
hood of emitting 2378-TCDD.
The list of Tier 4 categories continues to be revised as new
information is received and is therefore subject to change.
. -48-
-------
indicate there is likely to be selective partitioning of homologues between
those in the flue gas and those in the fly ash.2^ These data imply that,
tetra chlorinated dioxins concentrate in the flue gas, while the higher dioxin
homologues concentrate in the flyash. As a result, ash samples can only provide
a general indication of the presence of dioxin at a, source. For example, if
2378-TCDD is found in the ash, it can o.nly be assumed that 2378-TCDD was pro-
bably emitted in' the flue gas. When 2378-TCDD is not found in the ash, it can-
not be assumed that 2378-TCDD is not present in the flue gas. At least at one
source, where both stack and ash data were collected from the same source, tetra
chlorinated dioxins were found in the flue gas, even though no tetra-octa chlo-
rinated dioxins were found in the fly ash.17 Ash samples? can probably now best
be used as a screening tool to identify those source categories which have a
likelihood of emitting dioxin and which probably should be stack tested;.
Despite the limitations associated with these samples, ash sampling is
expected to be useful. Ash sample data collected by Tier 4 may provide addi-
tional data to qualitatively indicate if Dioxin and furan are emitted from source
categories not stack tested. Results of stack and ash programs will be augmented
by other studies outside of Tier 4 in the development ol: any hypotheses on the
scope and magnitude of dioxin emissions from combustion sources as well as the
combustion conditions which are conducive to such emissions.
-49-
-------
-------
Section 4
Implementation Of The Tier 4 Plan
4.1 Overview
The previous section identified the sspecific source categories to be
sampled under Tier 4. This section focuses on how individual sources are being
selected for testing and who is responsible for various activities under Tier 4.
This section also discusses the management: and coordination aspects of Tier 4.
4.2 Selection Of Individual Sources For Testing
Ash and/or stack samples will be collected from approximately three sources
within each of the source categories which has been selected for testing. The
process of selecting individual sources for testing within each of the source
categories will be coordinated with the EPA Regional Offices. The EPA Regional
staff have been asked to recommend certain sources from amongst these sources
meeting model criteria for ash screening or stack testing. In selecting these
sites, the Regions and Tier 4 management must consider a number of practical
constraints.
For example, accessibility to a site will be an important factor. While
Section 114 authority of the Clean Air Act: will be used to gain access to par-
ticular facilities, if necessary, it is more practical to select sites where
accessibility is not a problem. The support and cooperation of State/local
*
agencies is also being incorporated into the study since much of the ash sampling
program will be conducted with EPA Regional Offices for sources located within
their jurisdiction. Selecting sites in those States which provide assistance
and cooperation facilitates sample collection.
-51-
-------
4.3 Ash Sampling Program
As individual sites are approved for the ash sampling, the appropriate
Regional and/or State Office(s) are contacted to arrange for the collection of
the ash samples. Arrangements include obtaining access to the site and providing
specifics of the facility ash handling system and other information regarding
the site(s) to Tier 4 ash program management. This information determines what
sampling kits and guidance will be sent to the Regional and/or State Office to
collect the samples.
Specific guidance' on the types of information to be collected and how the
ash samples are to be collected has been developed. The ash sampling plans and _
procedures are described in two Tier 4 documents: Sampling Procedures, EPA-450/
4-84-014C66 and Ash Sampling Program, EPA-450/4-84-014d.67 Ash samples are to
be collected from particulate matter control devices, or from the ash pit, in the
event no control device is used. Samples from open burning will be collected
from ground surfaces and will not include soil substrate. Multiple samples
collected at a site will be combined into a single composite for each site. A
portion of this composite sample will be stored for possible follow-up analyses.
Equipment and detailed procedures needed to collect, package and ship
samples to the Troika for analysis are provided in the Ash Sample Program docu-
ment.67 Briefly, each sample is assigned an episode number by the Sample Control
Center which has been established to manage the influx of samples to the Troika
under the National Dioxin Study. Episode numbers are provided to Tier 4 manage-
ment which assigns sample numbers. Episode and sample numbers are then for-
warded to the sampling group (Region or State). Once samples are analyzed, the
results will be validated by the laboratory before being released to Tier 4.
Tier 4 will authorize the validated data to be made available to the Regional
Offices.
-52-
-------
4.4 Stack Testing Program
Similar procedures are used to identify individual sources for the
detailed s.tack testing program. Tier 4 has identified general source charac-
teristics believed to be representative of the source categories to be stacked
tested. This information has been sent to the EPA Regions, who have recommended
some sources for testing. The Tier 4 sampling contractor selects possible samp-
ling sites from these recommendations as well as sites identified from industry
contacts and EPA reports. Once selected, the contractor initiates the process
of gaining access in coordination with the Regional/State/local agencies. After
access to a possible testing site has been agreed upon, the Tier 4 contractor
and, for some sites, an EPA representative visit the source to make a pretest
survey to obtain the necessary information needed to make a final testing deci-
sion and plan the source test. When a site is finally selected, a site specific
test plan is prepared by the contractor and submitted to Tier 4, the Region,
and a quality assurance contractor prior to the actual testing. This plan will
include:
(a) reference to descriptions of the sampling procedures equipment to
be used;
(b) the number of individual source tests to be performed;
(c) the number and types of samples (e.g., before or after controls,
feed/fuel, ash, soil) which will be collected;
(d) the location at the source where each sample will be collected;
(e) information on the various operating parameters which will be
recorded during source testing; and
(f) procedures to be followed concerning the labeling, storage and
shipment of the samples to the Troika for dioxin and furan analysis and to
the contractors' laboratory for precursor analyses.
-53-
-------
Once the plan is approved, the contractor will make arrangements to proceed
with the stack test. EPA personnel will oversee most of ' the stack sampling,
which includes three separate test runs over 5-7 days.
In addition, an independent quality assurance contractor will conduct on
site performance and systems audits. These types of audits will be conducted on
i
20 percent to 30 percent of the stack tests and will evaluate the sampling con-
tractors adherence to the Sampling Procedures document*^ and the Quality Assur-
ance Project Plan, EPA-450/4-84-014e.68 From these quality assurance audits
will come an independent evaluation of the stack sampling program.
After the stack test has been completed, a report will be prepared for each
source tested. A guidance manual, has been prepared for dioxin and furan source
test reports for testing data to be used in Tier 4. The guidance document will
assure consistency and comparability for Tier 4 data. The report is entitled
National Dioxin Study Tier 4 - Combustion Sources; Dioxin Source Test Report-
ing Format Guide, EPA-450/9-84-014f.69
4.5 Quantitative Exposure And Risk Assessment
In the event that 2378-TCDD, and higher dioxin and furan homologues are
detected in the stack emissions at any of the sources tested, a quantitative
exposure and risk assessment for atmospheric concentrations will be performed
for that site. Evaluation of inferrences that can be made to other sources
within this source category will also be performed. A revised exposure and
risk methodology is currently under development by EPA. The Strategies and Air
Standards Division (SASD), OAQPS will perform exposure and risk analyses for
Tier 4 stack testing results.
-54-
-------
4.6 Management Of Tier 4
The responsibility for managing Tier 4 has been assigned to the Office of
Air and Radiation (OAR). Within OAR, the Tier 4 program manager is Edward J.
Lillis, Chief, Air Management Technology Branch (AMTB). AMTB is part of OAR's
Office of Air Quality Planning and Standards (OAQPS), Monitoring and Data
Analysis Division (MDAD), Research Triangle Park, North Carolina. The Tier 4
program staff consists of James H. Southerland, William H. Lamason, William B.
Kuykendal and Edward J. Hanks within the Air Management Technology Branch.
Technical support for stack sampling methods and f.est program implemen-
tation is being provided to Tier 4 from the Office of Research and Development
(ORD), by Donald Oberacker of the Hazardous Wastes Engineering Research Labora-
tory (HWERL) and Larry Johnson of the Air and Energy Engineering Research Labo-
ratory (AEERL). Tier 4 risk assessments are performed by David Cleverly and
Mike Dusetzina of the Strategies and Air Standards Division, OAQPS.
An advisory Tier 4 Work Group, comprised of representatives of various EPA
program offices (Table 4-1), has been formed to:
(a) assist OAR in the design and implementation of Tier 4;
(b) seek technical expertise and guidance from other EPA represent-
atives on issues which arise during the project;
(c) provide suggestions on how to accomplish the study objectives;
(d) review and comment on the adequacy of protocols, sampling stra-
tegies, source test reports, and other material which may be prepared;
and
(e) keep other EPA officials appraised of the status of the project.
4.8 Coordination With The EPA Regional Offices
EPA's Regional Offices are expected to play a key role in the implementation
of the Tier 4 program. Regional contacts have been identified in each of the
-55-
-------
TABLE 4-1. - TIER 4 WORK' GROUP
Kane
Don Barnes
David Cleverly
Jim Cummings
Mike Dellarco
Paul des Rosiera
Herbert Jauoraki
Larry Johnson
Steven Kroner
William H. Laaason
Edward J. Mills
Israel Hilner
Donald Oberacker
Garrett Smith
Janes H. Southerland
Dave Sussoan
Boniface Thayil
Organization
Chlorinated Dioxin Work Group, Washington, DC
Office of Air Quality Planning and Standards, Strategies
and Air Standards Division (OAQPS/SASD) , RTP, NC
Dioxin Manageaent Coordination Staff, Washington, DC
Office of Research Development (ORD), Washington, DC
Office of Research Development (ORD), Washington, DC
Environmental Research Laboratory, ORD, Duluth, MN
Air and Energy Engineering Research Laboratory
(AEERL), RTF, NC
Office of Water- Regulations and Standards, Office of
Water (OWRS/OW), Washington, DC
Office of Air Quality Planning and Standards, Monitoring
and Data Analysis Division (OAQPS/MDAD) , RTP, NC
Office of Air Quality Planning and Standards, Monitoring
and Data Analysis Division (OAQPS/MDAD), RTP, NC
Air Programs Branch, Region III, Philadelphia, PA
Hazardous Waste Engineering Research Laboratory (HWERL),
Cincinnati, OH
Solid Waste Branch, Region II, New Torkk, NY
Office of Air Quality Planning and Standards Monitoring
and Data Analysis Division (OAQPS/MDAD), RTP, NC
Office of Solid Waste, Office of Solid Waste and
and Emergency Response (OSW/OSWER), Washington, DC
Air and Radiation Branch, Region V, Chicago, IL
Mail Code
TS-788
MD-12
WH-562B
RD-675
RD-681
-
MD-62
WH-553
MD-14
MD-14
3AM12
MS179
2AWM-SW
MD-14
WH-565A
5AR26
Telephone
Number
(202) 382-2897
(FTS) 629-5645
(919) 541-5645
(202) 382-4686
(202) 382-5794
(202) 382-2722
(FTS) 783-9550
(218) 727-6692
(FTS) 629-7943
(919) 541-7943
(202) 382-7027
(FTS) 629-5575
(919) 541-5575
(FTS) 629-5586
(919) 541-5586
(215) 597-9090
(513) 684-7696
(212) 264-3407
(FTS) 629-5575
(919) 541-5575
(202) 382-7927
(312) 886-6054
-56-
-------
Region, .to establish and maintain lines of communication. A curreat list of
Regional Office contacts is provided in Table 4-2. Briefly, their role is to:
(a) coordinate with Tier 4 on Regional and State plans for dioxin
sampling of combustion sources;
(b) recommend combustion source categories for dioxin sampling;
(c) identify potential sampling sites for combustion source
categories;
(d) assist Tier 4 in obtaining access to sites selected for ash and
stack sampling;
(e) collection of ash samples at screening sites, either directly or
by State and local programs;
(f) appraise other Regional Office staff on the scope, objectives and
status of Tier 4; and
(g) coordinate with other Regional Office dioxin staff and the
designated public affairs liason on community relations issues of
Tier 4.
4.9 Additional Information
For additional information on Tier 4, please contact James Southerland
or William Lamason at (919) 541-5575 or FTS 629-5575-.
-57-
-------
Table 4-2
Regional Office Contacts - Tier 4
Region Kane
I Margaret McDonough
XI Garrett Smith
III Israel Hllner
IV Brian Beals
V . Boniface Thayil
71 Jill Lyons
VII Bill Fairless
VIII Nora Huey
IX Kent Kitch'ingaan
X Dana Davoli
Organization
Air Programs Branch (RM 2313)
U.S. EPA, John F. Kennedy Bldg.
Boston, HA 02203
Solid Waste Branch (2AHM-SW)
U.S. EPA, 26 Federal Plaza
New York, NT 10278
Air Branch (3AM12)
U.S. EPA, 841 Chestnut Street
Philadelphia, PA 19107
Air Management Branch
U.S. EPA, 345 Courtland St., NE
Atlanta, GA 30365
Air and Radiation Branch
U.S. EPA, 230 S. Dearborn St.
Chicago, IL 60604
Environmental Services Division
U.S. EPA, 1201 Elm Street
Dallas, TX 75270
Surveillance and Analysis
U.S. EPA, Central Regional Lab.
25 Funston Road
Kansas City, MO 66115
Air Programs Branch
U.S. EPA, 1860 Lincoln Street
Denver, CO 80295
Air Toxics Unit, AMD
U.S. EPA, 215 Freaont Street
San Francisco, CA 94105
Air Programs Branch (532)
U.S. EPA, 1200 Sixth Avenue
Seattle, WA 98101
Telephone Number
(617) 223-4870
(FTS) 223-4870
(212) 264-3407
(FTS) 264-3407
(215) 597-9090
(FTS) 597-9090
(404) 881-3067
(FTS) 257-3067
(312) 886-6054
(FTS) 886-6054
(214) 767-9832
(FTS) 729-9832
(913) 236-3884
(FTS) 926-3884
(303) 837-3763
(FTS) 327-3763
(415) 974-8381
(FTS) 454-8381
(206) 442-1495
(FTS) 399-1495
-58-
-------
References .
1. Dioxin Strategy, Office Of Water Regulations And Standards and the Office
Of Solid Waste And Emergency Response, U. S. Environmental Protection
Agency, Washington, DC, November 28, 1983.
2. Dioxins, EPA-600/2-80-197, U. S. Environmental Protection Agency,
Cincinnati, OH, November 1980.
3. National Dioxin Study Tier 4 - Combustion Sources; Initial Literature
Review And -Testing Options, EPA-450/4-84-014b, Monitoring and Data
Analysis Division, U. S. Environmental Protection Agency, Research
Triangle Park, NC, October 1984.
4. "Dioxin Test Costs", Memorandum from Bob Parks, Radian Corp., Research
Triangle Park, NC, to Bill Lamason, Office Of Air Quality Planning And
Standards, U.S. Environmental Protection Agency, Research Triangle Park,
NC, April 3, 1984.
5. The Trace Chemistries Of Fire - A Source Of And Routess For The Entry Of
Chlorinated Dioxins Into The Environment, Chlorinated Dioxin Task Force,
Michigan DivisionofDowChemical, U. S. A., Midland, MI, 1978.
6. Environmental Standards Workshop, American Society of Mechanical Engineers,
McLean, VA, January 23-25, 1984.
7. Analytical Procedures And Quality Assurance Plan For The National Dioxin
Study (Draft), Office Of Research And Development, U. S. Environmental
Protection Agency, Washington, DC, January 1984.
8. "Interim Evaluation Of Health Risks Associated With Emissions Of Tetra-
chlorinated Dioxins From Municipal Waste Recovery Facilities", EPA
Environmental News, U. S. Environmental Protection Agency, Washington,
DC, November 19, 1981.
9. "TCDD Emissions From Municipal Waste Combustors", Memorandum from
Michael B. Cook, U. S. Environmental Protection Agency, Washington, DC,
to Division Directors, Regions I-X, U. S. ' Environmental Protection
Agency, December 16, 1983.
10. Summary Of A Literature Search To Develop Informatioia On Sources Of
Chlorinated Dioxin And Furan Air Emissions, EPA Contract Number 68-02-
3513, Radian Corporation, Research Triangle Park, NC, September 1983.
11. C. L. Haile, et al., Assessment Of Emissions Of Specific Compounds From
A Resource Recovery Municipal Refuse Incinerator, EPA Contract Number
68-015915, Midwest Research Institute, Kansas City, MO, November 1983.
12. Comprehensive Assessment Of The Specific Compounds Present In Combustion
Processes, Volume 1, Pilot Study of Combustion Emissions Variability,
EPA-560/5-83-004, Office Of Toxic Substances, U. S. Environmental
Protection Agency, Washington, DC, June 1983.
-59-
-------
13. Comprehensive Assessment Of The Specific Compounds Present In Combustion
Processes, Volume 3, National Survey Of Organic Emissions From Coal
Fired Utility Boiler Plants, EPA-560/5-83-006, Office Of Toxic Sub-
stances, U. S. Environmental Protection Agency, Washington, DC, September
1983.
14. G. H. Higgins, An Evaluation Of Trace Organic Emissions From Refuse
Thermal Processing Facilities, EPA Contract Number 68-01-6071, Systech
Corporation, Xenia, OH, July 1982.
15. T. 0. Tiernan, et al.-, "Characterization Of Toxic Components In The
Effluent From A Refuse-fired Incinerator", Resources And Conservation,
9_:343-354, 1982.
16. A. Cavallaro, et al., "Sampling, Occurrence And Evaluation Of PCDDs And
PCDFs From Incinerated Solid Urban Waste", Chemosphere, 9(10):611-621,
1980.
17. A. Cavallaro, et al., "Summary Of Results Of PCDD Analyses From Incin-
erator Effluents", Chemosphere, U_(9):859-868, 1982.
18. F. Gizzi, et al., "Polychlorinated Dibenzo-p-dioxins (PCDD) And Polychlo-
rinated Dibenzofurans (PCDF) In Emissions From An Urban Incinerator - 1,
Average And Peak Values", Chemosphere, 11(6):577-583, 1982.
19. J. Janssens, et al., "Qualitative And Quantitative Analysis Of Emissions
Of A Municipal Incineration Installation", presented at CEC Physicochem-
ical Behavior Of Atmospheric Pollutants, 2nd Symposium, Varese, Italy,
September 29 - October 1,1981.
20. K. Olie et al., "Polychlorinated Dibenzo-p-dioxins And Related Compounds
In Incinerator Effluents", Pergamon Series On Environmental Science.
21. D. G. Ackerman, et al., At-sea Incineration Of PCS Containing Wastes
Onboard The "M/T Vulcanus", EPA-600/7-83-024, U. S. Environmental
Protection Agency, Research Triangle Park, NC, April 1983.
22. D. G. Ackerman, Destruction Efficiencies For TCDD During At-sea Inciner-
ation Of Herbicide Orange, EPA Contract Number 68-02-2660, Radian
Corporation, Research Triangle Park,,NC, March 1979.
23. PCS Disposal By Thermal Destruction, PB82-241860, National Technical
Information Service, Springfield, VA, Research Triangle Park, NC,
June 1981.
24. H. 0. Rghei and G. A. Eiceman, "Adsorption And Thermal Reactions Of
1,2,3,4-Tetrachlorodibenzo-p-dioxin On Fly Ash From A Municipal Inciner-
ator", Chemosphere, U_(6):569-676, 1982.
25. "Results of Silvex Test Burn DREs", letter from K. H. Jones, PhD.,
Roy F. Weston Asscociates, Westchester, PA, to Morris Hunt, Rollins
Environmental Services, Inc., Bridgeport, NJ, September 9, 1983.
-60-
-------
26. Performance Evaluation Of Full Scale Hazardous Waste Incinerators (Draft
Final Report), prepared for Incineration Research Division, U. S.
Environmental Protection Agency, Cincinnati, OH, NC January 1984.
27. Sewage Sludge Task Force Report, Office of Water, U.S. Environmental
Protection Agency, Washington, DC. Unpublished.
28. P. F. Fennelly, et al., Environmental Characterization Of Disposal Of
Waste Oils In Small Combustors (Draft), EPA Contract Number 68-02-3168,
GCA Technology, Bedford, MA, August 1983.
29. C. Castaldini, et al., Emissions Testing Of Industrial Boilers Co-firing
Hazardous Wastes - Sites A, D. E, G, EPA Contract Number 68-02-3176,
Acurex Corporation, Mountain View, CA, 1983.
30. B. DaRos, et al., Measured Multi-media Emissions From The Wood Pre-
serving Industry, EPA Contract Number 68-03-2567, Acurex Corporation,
Mountain View, CA, March 1981.
31. J. Hall, et al., Evaluation Of PCS Destruction Efficiency In An Indus-
trial Boiler, GCA-TR-81-82-G, GCA Corporation, Bedford, MA, April 1981.
32. J. E. Howes, et al., Determination Of Dioxin Levels In Carbon Reactiva-
tion Process Effluent Streams, EPA Contract Number 68-02-3487, Battelle,
Columbus, OH.
33. T, J. Nestrick, et al., Methodology And Preliminary Results For The
Isomer-specific Determination Of TCDDs And Higher Chlorinated Dibenzo-
p-dioxins In Chimney Particulates firom Wood-fueled Domestic Furnaces
Located In Eastern, Central And Western Regions Of The United States,
presented at the International Symposium On Chlorinated Dioxins And
Related Compounds, Arlington, VA, October 25-29, 1981.
34. Dioxins In Mobile Sources Particles, memorandum from John E. Sigsby,
"Chief, to Karl H. Hellman,Control Technology and Application Branch,
U. S. Environmental Protection Agency, Ann Arbor, MI, February 23, 1982.
35. D. 0. Hryhorczuk, et al.., "A Wire Reclamation Incinerator As A Source Of
Environmental Contamination With Tetrachlorodibenzo-p-dioxins And Tetra-
chlorodibenzofurans", Archives Of Environmental Health, 36(5): 228-234,
1981.
36. J. A. Peters, Evaluation Of Hazardous Waste Incineration In Cement Kilns
At San Juan Cement Company, Monsanto Research Corporation, EPA Contract
Number 68-03-3025, Dayton, OH, January. 1983.
37. R. L. Harless, High Resolution Mass Spectrometry Methods Of Analysis For
Chlorinated Dibenzo-p-dioxins And Dibenzofurans. Human And Environmental
Risks Of Chlorinated Dioxins And Related Compounds, 1983.
38. Polychlorinated Biphenyls (PCBs); Manufacture. Processing, Distribution
in Commerce and Use Prohibitions; Use in Electrical Transformers, 49 FR
11070, March 23, 1984.
-61-
-------
39 M P. Esposito, et al.. Dioxins: Volume I - Sources, Exposure, Tran-
,;Q,t. And ControirEPA^OO/2-80-156. U. S. Environmental Protection
Agency, Cincinnati, OH, June 1980.
An r Ranoe et al., "Formation Of Polychlorinated Dibenzo-p-dioxins (PCDDs)
L Diben^^ns (PCDF) By Burning Or Heating Chlorophenates" , Cherno^
sphere, 7(3): 269-281, 1978.
Al B Jansson et al., "Formation Of Polychlorinated Dibenzo-p-dioxins
During CombusSolTor Chlorophenol Formulations", The Science Of The
Total Environment, 10(3 ): 209-217 , 1978.
42. B. Ahling and A. Lindskog, "Emission Of Chlorinated Organic Substances
From Combustion", Pergamon Series «n Environmental Science, 1982, pp.
215-225.
Al * Ahline et al., "Formation Of Polychlorinated Dibenzo-p-dioxins And
4 Dlbe^ofu8ranT^ing Combustion Of A 2,4,5-T Formulation", Chemosphere,
£(8): 461-468, 1977.
44 H. R. Buser, "Formation Of Polychlorinated Dibenzo-p-dioxins (PCDDs)
And Dibenzofurans (PCDFs) From The Pyrolysis of Chlorobenzenes ,
Chemosphere, £(6): 415-424, 1979.
45. H. R. Buser, et_al^, "Identification Of M^^*
Isomers In Fly Ash And PCB Pyrolyses", Chemosphere, 1(5). 419
46. H. R. Buser, et al., "Formation Of .
From The Pyrolysis Of PCBs", Chemosphere, 7j
47. H. R. Buser and C. Rappe, "Formation Of Polychlorinated Dibenzofurans
From The Pyrolysis Of Individual PCB Isomers", Chemosphere, 8(3).157 174,
1979.
48. C. Rappe, et al., "Polychlorinated Dioxins Dibenzofurans, And Other
Polynuclear-Ar^mTtics Formed During PCB Fires", Chemica Scnpta,
20_: 56-61, 1982.
49 C. Shih, etal., POM Emissions From Stationary Conventional Combustion
ny. u. onin, ei. _ ; _ .—^ „_,„_u-i ,rina(:e» T3 r> ir*\jr XTf1
Number 68-02-3138, Radian Corporation, Research Triangle Park, NC,
September 1979.
50. N. Mahle and L. Whiting, "The Formation Of Crhlor.°dlrb^
Air Oxidation And Chlorination Of Bituminous Coal , Chemosphere, 1(11).
693-699, 1980.
51 M. L. Tiernan, et al., "Chlorodibenzodioxins , Chlorodibenzof urans , And
Related Compounds In The Effluent .From Combustion Processes ,
Chemosphere, 12_( 4/5 ): 565-606, 1983.
-62-
-------
52. A. Libert!, ' et al., "PCDD And PCDF Formation In The Combustion Of
Vegetable Wastes", Chemosphere, 12_ (4/5);661-663, 1983.
53. A. K. Sachdev, Thermal Destruction Of Chlorophenol Residues, Project
Number KE 111-1-0008, Dearborn Environmental Services, Mississauga,
Ontario, Canada, July 1983.
54. P. B. Hulman, et al., Screening Study On Feasibility Of Standards Of
Performance For Wood Charcoal Manufacturing, EPA Contract Number 68-
. 02-2608, Radian Corporation, Research Triangle Park, NC, August 1978.
55. Impact Of Burning Hazardous Waste In Boilers, SCA Chemical Services,
Inc., Boston, MA, August 1982.
56. C. Rappe and S. Marklund, "Thermal Degradation Of Pesticides And
Xenobiotics: Formation Of Polychlorinated Dioxins And Dibenzofurans,
Pesticide Chemistry: Human Welfare And Enviroment'", Procedures Of The
5th International Congress On Pesticide Chemistry, 3_:317-322, 1983.
57. K. R. Rao (editor), Pentachlorophenol; Chemistry, Pharmacology, And
Environmental Toxicology, Plenum Press, New York, 1978.
58. S. Kulkarni, "Pollution Aspects Of The Thermal Regeneration Of Spent
Activated Carbon", Radian Corporation, .Research Triangle Park, NC,
1983. Unpublished.
59. Source Category Survey; Industrial Incinerators, EPA-450/3-80-013,
Emission Standards And Engineering Division, U. S. Environmental
Protection Agency, Research Triangle Park, NC, May 1980.
60. B. G. Liptak, Environmental Engineers Handbook. Volume 3, Chilton Book
Company, Ontario, Canada, 1974, p. 283.
61. "Summary Of Recent Tier 4 Pretest Survey Visits," EPA Contract Number
68-03-3148, Radian Corporation, Research Triangle Park, NC, 27709,
November 1984.
62. R. M. Dykes, A Review Of Standards Of Performance For New Stationary
Sources - Sewage Sludge Incinerators, EPA Contract Number 68-02-3816,
Radian Corporation, Research Triangle Park, NC, January 1984.
63. Report On The 1980 Manufacturing Industries Energy Consumption Study And
Survey Of Large Combustors, DOE/EIAI-0358, U. S. Department of Energy,
Washington, DC, January 1983.
64. Kraft Pulping: Control Of TRS Emissions From Existing Mills, EPA-450/2-
78-003b, U. S. Environmental Protection Agency, Research Triangle Park,
NC, March 1979. '
65. Application Of Combustion Modifications To Industrial Combustion Equip-
ment, EPA-600/7-79-015a, U. S. Environmental Protection Agency, Research
Triangle Park, NC, January 1979.
-63-
-------
66. National Dioxin Study Tier 4 - Combustion Sources; Sampling Procedures,
EPA-450/4-84-014c, Monitoring And Data Analysis Division, U. S. Environ-
mental Protection Agency, Research Triangle Park, NC, October 1984.
'57. National Dioxin Study Tier 4 - Combustion Sources; Ash Sampling Program,
EPA-450/4-84-014d, Monitoring And Data Analysis Division, U. S. Environ-
mental Protection Agency, Research Triangle Park, NC, 27711, January
1984.
68. National Dioxin Study Tier 4 - Combustion Sources; Quality Assurance
Project Plan, EPA-450/4-84-014e, Monitoring And Data Analysis Division,
U. S. Environmental Protection Agency, Research Triangle Park, NC, 27711,
February 1984.
69. National Dioxin Study Tier 4 - Combustion Sources: Dioxin Source Test
Reporting Format Guide, EPA-450/4-84-014f, Monitoring And Data Analysis
Division, U. S. Environmental Protection Agency, Research Triangle Park,
NC, 27711, February 1984.
-64-
-------
Appendix A
Number Of Sources To Be Tested In A Source Category
The" number of individual sources to be tested within each source category
depends upon several factors, 1« uding the purpose of the testing, the number
of sources .in the source category which have been tested previously, and the
available resources. Calculations have been made to estimate the number of
sources within a particular category which should be tested in order to
characterize statistically the emissions from the source category.
In order to determine the number of tests needed, information concerning
the variability of emissions from the source categories is needed. Since emis-
sion data are not available for many source categories,, the municipal waste
incinerator source category was chosen as an example in Figure A-l to estimate
the number of sources within this category which -should be tested. Based on
these data, it has been determined that to characterize this source category in
a statistically significant manner (i.e., at a 95 percent: confidence), between
14-39 tests would be required (see attached). More specifically, 39 tests would
be required if the results of each of the six previous tests are used In the
analysis* Fourteen tests would be required if certain source tests believed to
be nonrepresentative are not considered in the analysis. Assuming a cost of
$90,000 per source test (including non-dioxin analyses), it is clear that the
entire Tier 4 budget could be spent testing emissions from only one source
category for statistically representative results.
While it will not be possible under Tier 4 to sample enough sources to
obtain rigorous and statistically significant results, a reasonable number of
stack test and ash samplings should be performed in each source category. The
-65-
-------
confidence intervals around these mean emissions should be fairly narrow to allow
comparison among source categories. The confidence interval around a mean is
given by:
X + t S
/~
The size of the confidence interval, therefore, varies with the size of the
sample (n) and the value of the t statistic. The size of the t value varies
with the degrees of freedom (number of samples minus one) and the confidence
level chosen. Some example values of the t statistic are shown below.
t Statistic
Number of Samples
2
3
4
5
Inspe tion of the above shows that, beyond three samples, there are diminishing
returns. Similarly, the effects of 1/n diminish beyond n=3. For this reason,
and for practical considerations (i.e., resources), a sample size of three tests
will be considered optimal for the purposes of this study. This approach applies
to the number of sources where ash samples or stack samples would be collected.
95% C.I.
12.706
.4.303
3.182
2.776
90% C.I.
6.314
2.920
2.353
2.132
t.05/ /n
8.99
2.48
1.59
1.24
-66-
-------
Figure A-l. Determination of Sample Size For Municipal Waste Incinerators
PROBLEM: Determine nunrber of municipal waste combustors that should be tested
in order to estimate -the average dioxin emission-rate within a 5 ng/m3 interval
with 95 percent confidence.
Municipal Waste Combustors:: Total population = 40
Site TCDD (n.g/m3)
A 240
B • 29.7
C 6.3
D 3.15
E 1.2
F ND
Assume: Sampling from a finite population without replacement
n = N z2 SD2
d2 (N-l) + Z2 SD2
i
n = sample size
N = population size = 40
Z = reliability coeff = 2 (95% confidence)
SD = standard deviation
d = confidence interval = 5 ng/m3
Case I: (all sites listed)
x - 46.7
SD - 87.0
n • 40 (2)2 (87)2 =39
52 (40-1) + (87)2
Case II: (drop sites with highest and lowest average TCDD levels)
x » 10.1
SD » 11.46
n = 40 (2)2 (11.46)2 - 14
52 (40-1) + 22 (11.46)2
SOLUTION: In order to estimate average dioxin emission rate for municipal waste
combustions, 14-39 sources should be tested. Based on 93% confidence, results
would be within 5 ng/m3 of true mean.
-67-
-------
-------
TECHNICAL REPORT DATA .
(Please read Instructions on the reverse before comp ting}
1 _——— i —
EPA-450/4-84-014a
J.. TITLE AND SUBTITLE
Tier 4 - Combustion Sources
Karional Dixoin Stud
Prolect Plan
7. AUTHOH(S)
Edward Lillis, et. al.
9 PERFORMING ORGANIZATION NAME AND
Air Management Technology Branch (MD-14)
Monitoring and Data Analysis Division
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711 ;
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
OAQPS, MDAD, AUTB, FIS
Research Triangle Park, NC 27711
RilciPIENT'S ACCESSION-NO.
. REPORT DATE
February 1985
. PERFORMING ORGANIZATION CODE
. PERFORMING ORGANIZATION REPORT
10. PROGRAM ELEMENT NO.
B53B2F
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Final . .
TZTlSPONSORlNG AGENCY CODE
*****
ORD and Regions II, III and V) .
a»d SPA Office, (OPTS, OSWER,
g , . ___ _ . --- _ - -
assessing dioxin emissions from combustion sources ^J£e fche t.
Dioxin Study. The primary objectives of Tier 4 are ou concentrati6ns
These are to determine what - "«« crib^s the rationale used
approaches that were
considered.
methods and procedures are identified.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Toxic Studie
Dioxins
Furans Combustion Sources
2378 Tetrachlorodibenzo-p-dioxin
__ . /
TCDD
PCDD
Tier 4
Waste Combustion
Wood Stoves
Mobile Sources
Nati<
l i) toxi
tud
13. DISTRIBUTION STATEMENT
Unlimited
.IDENTIf
I ENDED TERMS
on Studies
I) (This Report)
Unclassified
2O. SECURITY CLASS! (This page)
Unclassified
. COSATI Field/Group
21. NO. OF PAGES
26
22. PRICE
EPA Form 2220-1 (9-73)
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2, LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall cany a date indicating at least month and year. Indicate the basis on which it was selected fe z date of issue date of
approval, date of preparation, etc.). ' ' '
8. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe. J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared. '
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13, TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
' ««f«nce of,
16. ABSTRACT
17. KEY WORDS AND DOCUMENT ANALYSIS
£S™£^^^^^ «— *•' «•"«*
-------
-------
?§ 3
51.2.
•?2.
o?
Is
TJ
O
522
w r:r;
s s
las
Si
------- |